CiXIV42

CiXIV42 Provided by University Library of Tübingen Transcribed with Tesseract

To the best of our knowledge this work is free of known copyrights or related property rights (public domain).

A comparative grammar of the Dravidian or South-Indian family of languages Caldwell, Robert Ry e, o Tt e e e, oo e, e Fange By e i .H-M e b Lol bt . Sate, vl P S P TG o £ 2 P i Wibieg wyd e 0. G Pibing, iy g Ay, Sl ol By, Sy, i g Ty Wi, e i, 25 T & 5 w4 s Fibi, Myt i, ) iy b g " o g A, o 04 Kg Qs el % e argh, iy » i o S B4, v ot At W e, g, s i i — W g iToees xh“?x e iy b el - MR, s g ity i ey T s Uik et s o 5 i, VM b Y ™ o Ut e, s .@«!}1 il oy ", pn &s kg R ety 5. ey I~ i g i v o, i T i i e, Rt i A i e Ko Uitiiyy .fxn:ﬁ. i, I g R, b 2 ol Hlip e e Vi Mo iy A B, S ol ik N T i vt i Ay r.t P g g g, Wy g i Ea.&ft e, w ..ig C‘ilgmr ety Pt el J.fr\: g o it Ay g et s ity i ity g Si.v&n i Wiigia ey bt O ML P St Yer gy Rk ﬁ,a\m &P bz, x, o) e gy il by P, e Eiod ! A e Fﬁ‘s WL Ly s, b iy ey iy 2 s mxb..eﬁ et 2t e i o iy o [dcigz it g . i i i AN, ﬂ‘nuw o g S . Wiy N g gy AL ol e, —A.N e PRy gy PR, gﬁ@% Ty A el Le:f e s mv &ﬂvﬁmm, £ W\? L iy, e w?fh. i&b}!r Teba, i s N i kG, .. Hi o e vy o e :.; P ;fw&. L‘% e g iy g i it gﬂ. it 4 pesng ity IR b ¢ o o, Mgy ..r*&._.b A i 1905 Ao e . g e T, o o s S, s iy n et iy, o2 i i i T il i kﬁé Sy, g Jo A T b il Fia %&xu& iy e Fi, ba e i <l gy ity i fore o Sl iy il ek, e b P il s bl .@a.b?t leg o g 1 e Pta ) - f Wi & Vi it oy, e, | 6494/ it Histyy e Ay s P S 3§ e, e b .iw, 55 ) no s o s g{% o g Hp 5 g, e A b o giﬁ. optie, Sty ey ....cf. et e H\y i 5 aﬁt“w et A beit g, i Ehtag g 4.? .@mﬁﬁ.@e e g el i Ty, kit il e i o i, i e §g&f E&Gk& i W, lu"ﬁ&t Gt P g Ty, N o R‘EN\ - etsp, Al g 3 LY i ‘wg o i, Nobd s ok 1, i 7 B s e rriramf. oy e %_lgwerx Jvtin,rzfrix it s L S Bomadi ity o 2 w% x&ffﬁ Fana, a4 B 5 g 2o fony L R, it o 't D sy St L e g g by i o i i i i i i i 1593 Co &.‘ Pyt ] ik s s 5 S o e Hisiay e & i e P SO0 o, b s i orrecnrs % e 443y etz gl g b v, s O o s oty o iy o i vﬁ% ¥ g, i g £ o o P g3t R0y e o ot RRe T Py, St A S i 2 G i i 7 b 3 bl e ey, iyl o 1.\.».# L, oLl I - /. S i i i ] \n& Sige . 3 . w&«.x«;ﬂ 50 i3 5054 g b ..mw: e ol o i S i b @ H~ ik o 2 SR et i 55 e b 3! Ll Ty, F o e S R S e x s = o e o Wk o i 2 e v S 7 s oy 0 3 s N A féwuwlnv s i 2, e o o in? = (A. Tk s Al e o o, Shre g, % ST A, R T8 T e e o M.u R o s PR o &t e XK, T & 5 e N Ty i S s 2 s & o # g o S G e o S e S It o =5 e o g mw; 2 TR Tk 2 b S5 e, fovs Sy ] s e ey o ¥ o T e W'x s . TR 5 e . e 5 Foa Rk My o R o R 0 e Sl PN e v R 2 = o 0 o S o B i o S o o SR A SR o, o e o £ o el o W i 5 o m i G = e e e v 5 e T A, 5, A e e, oo Sk 5 o S s m e > o, w S Sl oy o e e T e Ay S 7 T e Dy A S e, 2 i .-v@ e A o S e Mw £ e e A s S ¥t e e S e e s A e SR w AR . Bt s i W\&. M ﬁi e = e MVP - L o e it N12<529065791 021 VRN .. e s % ” - w et e 1 = — = e ey T — R S0 Mgy, A COMPARATIVE GRAMMAR OF THE DRAVIDIAN OR SOUTH-INDIAN FAMILY OF LANGUAGES. BY Tue Rev. R. CALDWELL, B.A., MISSIONARY OF THE SOCIETY FOR THE PROPAGATION OF THE GOSPEL IN FOREIGN PARTS, AT EDEYENKOODY, TINNEVELLY, SOUTHERN INDIA. LONDON : e R RIS ON, 9.9, P Ak B MALL y 1856. PRIN % . - iy N ) K, 9 ghd Y M — TED BY HARRISON ST. MARTIN’S LANE v \ e ~ AND SONS, S s - oy 4 R AR My = s e PREFACE. It is many years since I felt convinced that much light might be thrown on the grammar of the Tamil language, by comparing it with that of the Telugu, the Canarese, and other sister idioms; and on proceeding to*make this comparison, I found, not only that my supposition was verified by the result, but that the Tamil imparts still more light than it receives, and also, that none of the South-Indian languages can be thoroughly understood or appreciated without some study of the others. Probably many other students of the South-Indian languages have been led to the same conclusion; but as the mission of the English in India is one which admits of little or no literary leisure,—as the old East, after the sleep of centuries, has begun to wake up and to clamour for the supply of its many material and moral wants, and as the majority of Anglo-Indians, whether they are engaged in the work of government, or in educational and Missionary labours, find that they have a world of work to do, and but little time or strength for doing it, this department of comparative philology, though peculiarly promising, has hitherto lain almost entirely uncultivated. Much, it is true, has been done towards the elucidation of some of the South-Indian languages taken separately, especially the Tamil and the Telugu. Beschi’s Grammar of the Shen- Tamil, and Mr. C. P. Brown’s Telugu Grammar, rise far above the level of the ordinary Grammars of the Indian vernaculars. But the study of those languages, viewed as a whole—the inter- S 1v PREFACE. comparison of their grammars—is still in its infancy; and 1t is only when philology becomes comparative, that it becomes scientific and progressive. The first to break ground in the field was Mr. Ellis, a Madras Civilian, who was profoundly versed in the Tamil lan- guage and literature, and whose interesting but very brief com- parison, not of the grammatical forms, but only of some of the vocables of three Drévidian dialects, is contained in his Intro- duction to Campbell’s Telugu Grammar. The next attempt that was made in this direction, was by the Rev. Dr. Stevenson, of Bombay, in some interesting papers on the languages of the Dekhan, which appeared in the Journal of the Bombay Asiatic Society. The main object which Dr. Ste- venson appeared to have in view, was that of establishing the identity of the Un-Sanscrit element which is contained in the North-Indian vernaculars, with the grammar and vocabulary of the Southern idioms. He failed, as it appeared to me, to establish that point; but many of his remarks on the characte- ristic features of Dravidian Grammar, and on the essential unity of the Drévidian dialects were perfectly correct; and though his papers were of too sketchy a character to be of much permanent philological value, they were decidedly in advance of everything which had hitherto been published on this subject. I was not aware of the existence of Ellis’s or Stevenson’s contributions to Dréividian comparative philology, when my own attention was directed to this department of study; and when at length I made their acquaintance, I felt no less desirous than before of going forward, for though I had lost the satisfaction of supposing myself to be the discoverer of a new field, yet it now appeared to be certain that the greater part of the field still lay not only uncolonized, but unexplored. I have not referred to Mr. Brian Hodgson’s numerous and learned papers on the ¢ Tamulian ’ languages of India, though I have long been acquainted with them, because I regard them as a misnomer. Those valuable papers treat of the Sub-Hima- layan dialects, which are styled ‘Tamulian’ by Mr. Hodgson, PREFACE. v but which might as properly, or improperly, have been styled by any other foreign name; and though they throw much light on the languages, the physiology, and the inter-relationship of the aborigines of the north-eastern frontier of India, they leave the Dravidian or Tamilian languages, properly so called, wholly untouched. From the commencement of my Tamil studies I felt inte- rested also in another question—that of the ulterior relationship of the Drividian family of languages; and before I was aware of the opinion which Professor Rask, of Copenhagen, was the first to express, and which has generally been adopted, I arrived by a somewhat similar process, at the same conclusion, viz., that the Dravidian languages are to be affiliated, not with the Indo-European, but with the Scythian group of tongues, and that the Scythian family to which they appear to be most closely allied is the Finnish or Ugrian. General statements of the Scythian relationship of the Dré- vidian languages, with a few grammatical illustrations, occupy a place in Prichard’s valuable ¢ Researches,” and have been repeated in more recent works on Comparative Philology; but as Prichard himself did not feel satisfied with general statements, impressions, and probabilites, and wished to see the problem solved, and as I was convinced that it never could be definitively solved without previously ascertaining, by a careful inter-com- parison of dialects, what were the most ancient grammatical forms and the most essential characteristics of the Dravidian languages, I found myself under the necessity of working out the entire subject for myself. It was not till T had finished this work, and commenced to prepare to carry it through the press, that I became acquainted with Professor Max Miiller’s treatise, ‘On the present state of our knowledge of the Turanian languages,” which is included in Bunsen’s ¢ Outlines of the Philosophy of Universal History. That treatise is the most comprehensive, lucid, and scholarly investigation of the general question which I have yet read; and I have been gratified to find not only that many of the Y Vi1 PREFACE. conclusions at which the author of that treatise has arrived, but that many of his proofs and illustrations also, arve identical with my own. Notwithstanding our general agreement with respect to the Drévidian grammatical system, and especially with respect to its Ugrian affinities, I have not found the following work fore- stalled by the Professor His work is generic, mine specific His is an admirable survey of the entire field; but he does not profess to cultivate thoroughly any one portion of the field, or even to prepare it for cultivation. He does not occupy himself in clearing away the stones, breaking up the fallow ground, pulverising and analysing the soil, and turning up the sub-soil to the light Occasionally it is true, he enters into details but thoueh his conclusions are always correct, it is too evident that in dealing with details he furnishes an illustration of the principle for which, as he observes, Boehtlingk stands up, viz., that “it is dangerous to write on languages of which we do not possess the most accurate knowledge Whilst the principal features of the Scythian relationship of the Dravidian tongues are strongly marked, and whilst their grammatical principles and syntactic arrangement are of too peculiar a nature to be w7y ‘rnl* mistaken, there is much in the phonics tem of these languages 1 their declen- in their dialectic interchanges and displacements, sional and conjugational forms, and especially in the nature uses, and changes of their formative particles, which cannot be understood without special study. - Dréavidian philology has recently attracted the attention of another writer, Dr. Logan, whose elaborate contributions to the ethnology of Eastern Asia, and of the eastern islands, form the most valuable papers in the Journal of the Indian Archipelago. That writer’s ethnological learning and philological acuteness are very great, and some of his conjectures are remarkably happy; but he is too fond of speculation, and not a few of his oeneralisations and speculations respecting the forms and pho- nology of the Dréavidian languages, are far a-head of his facts. Notwithstandin therefore, the intrinsic general value of PREFACE. vil the researches of Drs. Stevenson, Max Miller, and Logan, a work like the following still appears to be required. The Dra- vidian languages still require to be compared and their relation- ship to other languages investigated by some one who has made them his special study for an adequate space of time. Though I trust that the following work will help to supply this desideratum, yet it only professes to be a contribution I have towards the accomplishment of the object in view. laboured to be accurate throughout—and in a work of this kind accuracy cannot be attained without immense labour ; but not- withstanding my endeavours to be accurate, I am conscious of the existence of many defects. I trust, however, it will be remembered that this is the only systematic treatise on this subject which has yet appeared ; that in a first work on a new subject errors are almost unavoidable; and that, whatever be the defects of this work, it has at all events smoothed the way for those who may hereafter be disposed to investigate the sub- ject more deeply. During the period of my residence in India the work in which I was engaged as a Missionary was of too important a nature to allow me to spend much of my time in book-making. It was necessary for me to be content with jotting down occa- sionally a few notes and illustrations, and working out conclu- sions in my mind. Since my return to this country for a season, I have taken the opportunity of putting together the notes which I had collected, and moulding them iito a syste- matic shape; and the result is now published, in the hope that this work will help to supply a want which I had long felt myself, and which must, I conceived, have been felt by many others. I trust it will be found to contribute to a more enlarged and scientific study of each of the Dravidian languages, to a more accurate knowledge of their structure and vital spirit, and to a higher estimate of their phonic beauty, their philoso- phical organization, and their unequalled regularity. During the period which has elapsed since the commence- ment of this work, a period of a year and ten months, it has viil PREFACE. been my lot to visit two hundred and fifty different parishes in various parts of England, and to deliver about three hundred and fifty lectures and addresses on India and Indian Missions. It may therefore be concluded that the composition and prepa- ration for the press of a work of so laborious a nature, in addi- tion to the duties of a ¢ deputation,” have not left much scope for the relaxation and rest which form the usual adjuncts of a ‘ furlough ;” nevertheless, if the ulterior object which I have had in view should in any measure be accomplished—if in facilitating a more comprehensive study of the Dréividian languages by those Missionaries and East India Company’s Civilians who make use of them as instruments of thought or as vehicles of expression, the welfare of the Drévidian people should in any manner, however indirectly, or in any degree, however small, be pro- moted—I shall have my reward. I beg leave thankfully to acknowledge the facilities which have been afforded for the publication of this work by the kind- ness of the Honourable Court of Directors of the East India Company, in subscribing for a hundred copies, and of the Madras, Ceylon, and Bombay Governments, in subscribing for a hundred and twelve. R. CALDWELL. Office of the Society for the Propagation of the Gospel, 79, Pall Mall, London, June 2nd, 1856. — —— e e P AR LT SR SRS g e SR =y = - DRAVIDIAN COMPARATIVE GRAMMAR. INTRODUCTION. It is the object of th‘e following work to examine and compare the grammatical principles and forms of the various Dravidian lan- guages, in the hope of contributing to a more thorough knowledge of their primitive structure and distinctive character. In pursuing this object, it will be the writer’s endeavour to ascertain the relation which this family of languages bears to the principal families or groups into which the languages of Europe and Asia have been divided. Whilst the grammatical structure of each Dravidian language and dialect will be investigated and illustrated in a greater or less degree, in proportion to its importance and to the writer’s acquaintance with it, it will be his special and constant aim to throw light upon the structure of the Tamil—a language which he has for seventeen years studied and used in the prosecution of his missionary labours, and which is undoubtedly the oldest, richest, and most highly organized, of the Dravidian langnages,—in many respects the representative lan- guage of the family. The idioms which are included in this work under the general term ¢ Dravidian, constitute the vernacular speech of the great majority of the inhabitants of Southern India. With the exception of Orissa and those districts of Western India and the Dekhan in which the Gujarithi and the Marithi are spoken, the whole of the peninsular portion of India;, from the Vindhya mountains and the river Nerbudda (Narmada) to Cape Comorin, is peopled, and from B = S — D ——— i’t 7. = INTRODUCTION. the earliest period appears to have been peopled, by different branches of one and the same race, speaking different dialects of one and the same language—the language to which the term ¢ Drividian’ is here applied; and scattered off-shoots from the same stem may be traced still farther north as far as the Rajmahal hills, and even as far as the mountain fastnesses of Beluchistan. The Gujarathi, the Marathi (with its off-shoot the Konkani), and the Uriya, or language of Orissa, idioms which are derived in the main from the decomposition of the Sanscrit, form the vernacular speech of the Hindu population within their respective limits: besides which, and besides the Dravidian languages, various idioms which cannot be termed indigenous or vernacular are spoken or occasionally used by particular classes resident in Peninsular India. Sanscrit, though it never was the vernacular language of any district of country in the South, is in every district read and to some extent understood by the majority of the Brahmans,—the descendants of those Brahmanical colonists of early times to whom the Dra- vidians are indebted for the higher arts of life and the first elements of literary culture. Such of the Brahmans as not only retain the name, but also discharge the functions of the priesthood, and devote themselves to professional studies, are generally able to converse in Sanserit, though the vernacular language of the district in which they reside is that which they use in their families and with which they are most familiar. They are styled, with reference to the language of their adopted district, Dravida Brahmans, Kérala Brahmans, Karnataka Brahmans, &c.; and the Brahmans of the several langunage- districts have virtually become distinet castes; but they are all undoubtedly descended from one and the same stock ; and Sanscrit, though now regarded only as an accomplishment or as a professional acquirement, is properly their ancestral tongue. Hindféstani is the distinctive, hereditary language of the Mahom- medan portion of the population in the Dekhan and the southern peninsula,—the descendants of those warlike Mahommedans from northern India by whom the Peninsula was overrun some centuries ago. It may be regarded as the vernacular language in some parts of the Hyderabad country; but generally throughout Southern India, the middle and lower classes of the Mahommedans, who constitute the majority, make as much use of the language of the district in which they reside as of their ancestral tongue, and many of them are now unable to put a single sentence together in Hind{istani. Hebrew is used by the small but interesting colony of Jews resi- dent in Cochin and the neighbourhood, in the same manner and for the 3 SOUTH-INDIAN VERNACULARS. same purposes as Sanserit is used by the Brahmans. Gujarathl and MarAthi are spoken by the Gujarathi bankers and the Parsi shop- keepers who reside in the principal towns in the Peninsula: the mixed race of ¢ country-born’ Portuguese are rapidly forgetting (except in the territory of Goa itself) the corrupt Portuguese which their fathers and mothers were accustomed to speak, and learning English instead ; whilst French still retains its place as the language of the French employés and their descendants in the settlements of Pondicherry, Karikél, and Mahe, which still belong to France. Throughout the territories of the East India Company, English is not only the language of the governing race and of its ¢ East-Indian’ or ‘Indo-British’ off-shoot, but is also used to a considerable extent by the natives of the country in the administration of justice and in commerce ; and in the presidency of Madras and the principal towns, it is daily winning its way to the position which was formerly occu- pied by Sanserit, as the vehicle of all higher learning. Neither the English, however, nor any other foreign tongue, has the slightest chance of becoming the vernacular speech of any portion of the inhabitants of Southern India. The indigenous Dravidian languages, which have maintained their ground for more than two thousand years against Sanscrit, the language of a numerous, powerful, and venerated sacerdotal race, may be expected successfully to resist the encroachments of every other tongue.” - * T admit with Sir Erskine Perry (see his valuable paper in the Journal of the Royal Asiatic Society), that English, the language of the governing race, should be employed as the language of public business in every part of British India; and I am certain that this end could be attained in a very short time by simply requiring every candidate for government employment, from the highest to the lowest, to pass an examination in English. The natives would everywhere adapt themselves to this arrangement, not only without reluctance, but with alacrity and pleasure ; and English schools and other facilities for the acquisition of English would multiply apace, as soon as it was found that the new rule eould not be evaded. I do not think, however, that English can ever become the vernacular language of any class of the Hindus, nor even that it is likely to be used to any considerable extent as a lingua franca beyond the circle of govern- Before we can reasonably anticipate the employment of English ment employés. as a conventional language, like Latin in the middle ages, or French in the more modern period, the number of the English resident in India should bear a much larger proportion to the mass of the inhabitants. That proportion is at present infinitesimally small : e. g. the population of the two Collectorates, or provinces, in Southern India with which I am best acquainted—Tinnevelly and Madura, amounts in round numbers to three millions: the number of Englishmen (and Americans) resident in those two provinces is considerably under a hundred and fifty ! and that number includes the judges and magistrates who administer justice in those provinces, the officers of a single regiment of sepoys, the men belonging to a small detachment of foot artillery, a few cotton planters and merchants, and Including women and the missionaries belonging to three missionary societies ! children, the number is about three hundred, with which handful of English people we have to contrast three millions of Hindus ! B2 3 | INTRODUCTION. ENUMERATION OF DRAVIDIAN LANGUAGES. | i The idioms which 1 designate as ¢ Dravidian,” are nine in number, exclusive of the Rajm ah il al, the Uraon, and the Brahui. They are as follows : — 1. The Tamil, by the earlier Europeans erroneously termed ¢the Malabar.”* The proper spelling of the name is ¢ Tamir; but through the dialectic changes of r into 1, it is commonly pronounced Tamil, This and is often erroneously written ‘Tamul’ by FKuropeans. language being the earliest cultivated of all the Dravidian idioms, the most copious, and that which contains the largest portion and the richest variety of indubitably ancient forms, it is deservedly placed at the head of the list. It includes two dialects, the classical and the colloquial, or the ancient and the modern, called respectively the Shen-Tamil* and the ¢ Kodun-Tamil,” which differ one from the other so widely that they might almost be regarded as different langnages. The Tamil language is spoken throughout the vast plain of the Carnatic, or country below the Ghauts, from Pulicat to Cape Como- rin, and from the Ghauts, or central mountain range of Southern India, to the Bay of Bengal. It is also spoken in the southern part of the Travancore country on the western side of the Ghauts, from Cape Comorin to the neighbourhood of Trivandrum; and in the northern and north-western parts of Ceylon, where Tamilians com- menced to form settlements prior even to the Christian era, and from whence they have gradually thrust out the Singhalese. All throughout Ceylon the coolies in the coffee plantations are Tamilians; the majority of the money-making classes even in Colombo are Tamilians ; and ere long the Tamilians will have excluded the Singhalese from The almost every office of profit and trust in their own island. majority of the domestic servants of Europeans and of the camp- followers in every part of the presidency of Madras being Tamil people, Tamil is the prevailing language in all military cantonments in Southern India, whatever be the vernacular language of the district. Hence, at Cannanore in the Malayila country, at Bangalore in the Canarese country, at Bellary in the Telugu country, and at Secunder- abad, where Hind{istin] may be considered as the vernacular, the * Tt is singular that so able and accurate a scholar as Dr. Max Miiller should have supposed the Malabar to be a different language from the Tamil: nor did he confound it, as would have been natural enough, with the Malayilam, for he gives a distinct place (especially in his ‘list of pronouns’) to each of the Dravidian dialects which actually exist, including the MaliayAlam, and thereto he adds the Malabar, on the authority, I presume, of some grammar of the last century, in which the Tamil was called by that name. 5 NUMERATION OF DRAVIDIAN LANGUAGES. Janguage which most frequently meets the ear in the bazaars is the Tamil The majority of the Klings (‘ Kalingas®), or Hindus, who are found in Pegu, Penang, Singapore, and other places in the further east, are Tamilians the coolies who have emigrated in such numbers to the Mauritius and to the West Indian colonies are mostly Tamilians 1B short, wherever money is to be made, wherever a more apathetic or a more aristocratic people is waiting to be pushed aside, there swarm the Tamilians, the Greeks or Scotch of the east, the least scrupulous and superstitious, and the most enterprising and persevering race of Hindus Including Tamilians resident in military stations and distant colonies, and the Tamilian inhabitants of South Travancore, and Northern Ceylon, and excluding not only Mahommedans, &e., but also Brahmans and people of Telugu origin who are resident in the Tamil country, and who form at least ten per cent. of the whole population, the people who speak the Tamil language may be esti- mated at about ten millions 2. The Telugu, in respect of antiquity of culture and glossarial copiousness, ranks next to the Tamil in the list of Dravidian idioms; but in point of euphonic sweetness it claims to occupy the first place. The Telatru called also the Telingu, or Telungu (nasalised from Telugu), is the ¢ Andhra’ of Sanscrit writers, a name mentioned by the Greek geographers as the name of a nation dwelling on or near the Ganges. This language was sometimes called by the Europeans of the last generation the ° Gentoo,” from the Portuguese word for heathens, or ¢ Gentiles The Telugu is spoken all along the eastern coast of the Peninsula, from the neighbourhood of Pulicat, where it supersedes the Tamil, to Chicacole, where it begins to yield to the Uriya ; and inland it prevails as far as the eastern boundary of the Maratha country and the Mysore; including within its range the ‘Ceded districts’ and Kurnool, the greater part of the territories of the Nizam, or the Hyderabad country, and a portion of the Nagpore country and Gond- wana Formerly Telugu appears to have been spoken as far north as the mouths of the Ganges This appears both from the geographical limits which are assigned by the Greeks to the territory of the Andhras, 4 or northern Telugus, and from many of the names of places mentioned by Ptolemy as far as the mouths of the Ganges being found to be [ id Telugu The Telugu people, though not the most enterprising or 16 migratory, are undoubtedly the most numerous branch of the Dravidian 3 e race Including the Naiks or Naidoos (‘ Nayakas’), Reddies, and other Telugu tribes settled in the Tamil country, who are chiefly the 6 INTRODUCTION. descendants of those soldiers of fortune by whom the Péandiya and Chola kingdoms were subverted, and who number not much less than a million of souls; and including also the Telugu settlers in Mysore and the indigenous Telugu inhabitants of the Nizam’s territory and other native states, the people who speak the Telugu language may be estimated as amounting to at least fourteen millions. 3. The next place is occupied by the Canarese, properly the Kannadi, or Karnataka,* which is spoken throughout the plateau of Mysore and in some of the western districts of the Nizam’s territory, as far north as Beder : it is spoken, also, (together with the Malayalam, the Tuluva, and the Konkani, but more extensively than any of them) in the district of Canara, on the Malabar coast, a distriet which originally coustituted the Tuluva country, but which was subjected for centuries to the rule of Canarese princes, and hence acquired the name by which it is at present known. Under the denomination of Canarese many include the Coorg or Kodaga, an idiom which is spoken by the inhabitants of the small principality of Coorg on the Western Ghauts, and which has generally been considered rather as an ancient and uncultivated dialect of the Canarese, modified by the Tulu, than as a distinet language. The Rev. Mr. Migling, a German missionary, who has recently settled amongst the Coorgs, and who is our only reliable anthority on the subject of their language, now states that Coorg is more closely allied to the Tamil and Malayéla than to the Canarese. The speech of the Badagars (‘people from the north’), * <Karpftaka’ is not a Drévidian, but a Sanserit word, and is properly a It is defined to mean primarily ‘a generic name for both Telugu and Canarese. species of dramatic musie,” or ‘comedy :’ it is used secondarily in Telugu as an adjective to signify ‘native,’ ‘aboriginal,’ e.gy. ‘Karnitaka mélam,” Tel. ‘native music:’ it then became the common designation of the Telugu and Canarese, or ‘native’ languages: and, finally, was restricted still further, and became the distinctive appellation of the Canarese alone. I should not have nsed the word ¢ finally, for ¢ Karnataka’ has now got into the hands of foreigners, who have given it a new and more erroneous application. When the Mahommedans arrived in southern India, they found that part of it with which they become first acquainted—the country above the Ghauts, including Mysore and part of Telingdna — called ‘ the Karnitaka country.’ In course of time, by a misapplication of terms, they applied the same name, ‘the Karnatak,’ or ‘ Carnatie,” to designate the country below the Ghaats, as well as that which was above. The English have carried the misapplication a step further and restricted the name to the country below the Ghauts, which has no right to it whatever. Hence the Mysore country, which is properly the Carnatic, is no longer called by that name by the English, and what is now geographically termed ‘the Carnatic’ ig exclusively the country below the Ghauts, on the Coromandel coast, including the whole of the Tamil country and the district of Nellore in the Telngu country. The word °Karnitaka’ was further corrupted by the Canarese people themselves into ‘Kannada’ or ‘Kannara;’ from which the language is styled ¢ Kannadi, and by the English, ‘Canarese.” A province on the Malabar coast is called ¢ Canara,” properly ‘Kannadiyam,” in consequence of having long heen subjected to the government of Karnitaka princes, — — ENUMERATION OF DRAVIDIAN LANGUAGES. 7 commonly called Burghers, the most numerous class of people inhabiting the Nilgherry hills, is undoubtedly an ancient Canarese dialect. The Canarese, properly so called, includes, like the Tamil, two cultivated dialects, the ancient and the modern ; of which the former differs from the latter, not—as classical Telugu and Malayalam differ from the colloquial dialects of those languages—by containing a larger infusion of Sanscrit derivatives, but by the use of different inflexional termina- tions. The dialect called ¢ Ancient Canarese’ is not to be confounded with the character which is denoted by that name, and which is found in many very ancient inscriptions in the Maratha country as well as in Mysore. The language of all really ancient inscriptions in the ‘ Hala Kannada,” or Ancient Canarese character, is Sanscrit, not Canarese. The people that speak the Canarese language, including the Coorgs, &c.. may be estimated at five millions : but, in the case of both the Canarese and the Telugu, the absence of a trustworthy census of the inhabitants of native states, requires all such estimates to be con- sidered as mere approximations. In the Nizam’s territory four languages—the Canarese, the Marathi, the Telugu, and the Hindustani —are spoken by different classes or in different districts; but it is impossible to ascertain the proportionate prevalence of each with any degree of certainty. This 4. The Malayalam, or ‘Malayarma, ranks next in order. language is spoken along the Malabar coast, on the western side of the Ghauts, or ¢ Malaya’ range of mountains, from the vicinity of Man- galore, where it supersedes the Canarese and the Tulu, to Trivandrum, where it begins to be superseded by the Tamil. The people by whom this language is spoken in the native states of Travancore and Cochin, and in the East India Company’s districts of Malabar and Canara, may be estimated at two and a half millions. All along the Malabar coast Tamil is rapidly gaining upon the Malayalam. Though that coast was for many ages more frequented by foreigners than any other part of India; though Phenicians, Greeks, Jews, Syrian christians, and Arabs, traded in succession to the various ports along the coast; and though permanent settlements were formed by the three last classes ; yet the Malayéla people continue to be of all Drividians the most exclusive and superstitious, and shrink most sensitively from contact with Hence ‘the lines and centres of communication’ have foreigners. been seized, and the greater part of the commerce and public business of the Malabar states has been monopolized, by the less scrupulous and more adroit Tamilians, whose language bids fair to supersede the MalayAlam, or at least to confine it within the limits of the hill-country and the jungles. o e —— INTRODUCTION: 5. Last in the list of cultivated Drividian tongues is the Tuiu, or Tuluva; an idiom holding a position midway between the Canarese and the Malayé&lam, but more nearly resembling the Canarese. This language was once generally prevalent in the district of Canara, but is now spoken only in a small tract of country in the vicinity of Mangalore, by not more than a hundred or a hundred and fifty thousand souls. It has been broken in upon by many other languages, and is likely soon to disappear. The four languages which follow differ from those that have been mentioned in that they are entirely uncultivated, destitute of written eharacters, and comparatively little known. 6. The Toda, properly the Tuda, or Tudava ; the language of the Tudavars, a primitive and peculiarly interesting tribe inhabiting the Nilgherry hills, practising quasi-Druidical rites, and commonly believed to be the aboriginal inhabitants of those hills. Their number could not at any time have exceeded a few thousands ; and at present, through opium-eating and polyandria, and through the prevalence amongst them at a former period of female infanticide, they do not, it is estimated, number more than from three to five hundred souls. 7. The Kota ; the language of the Kotars, a small tribe of Helot —— craftsmen inhabiting the Nilgherry hills, and numbering about a thousand souls. The Tuda language may be considered as the indi- genous speech of the Nilgherries; the Kota as a very old and very rude dialect of the Canarese, which was carried thither by a persecuted low-caste tribe at some very remote period. Besides those two, two other languages are vernacular on the Nilgherry hills; viz., the dialect which is spoken by the Burghers or Badagars, an ancient but organized dialect of the Canurese, and the rude Tamil which is spoken by the Trulars (¢ people of the darkness’) and Curbs or Curubars (Tam. “ Kurumbar,” ‘nomade shepherds’), who are occasionally stumbled upon by adventurous sportsmen in the denser, deeper jungles, and the smoke of whose fires may occasionally he seen rising {rom the lower gorges of the hills. 8. The Gond or Goand ; the language of the indigenous inhabitants of the northern and western parts of the extensive hill country of Gondwana, of the northern portion of Nagpore and of the greater part of the Saugor and Nerbudda territories. 9. The Khond, Kund, or more properly the Ku ; the language of the people who are commonly called Khonds, but who call themselves Kus—a primitive race, who are supposed to be allied to the Gonds; who inhabit the eastern parts of Gondwana, Goomsur, and the hilly P ranges of Orissa ; and who have acquired a bad notoriety through their ENUMERATION OF DRAVIDIAN LANGUAGES 9 horrid practice of stealing the children of their neighbours and offering them up in sacrifice I am unable to form a probable estimate of the numbers of the people by whom the Gond and the Ku are spoken, I think, however, that they cannot safely be estimated under half a million of souls The proportionate numbers of the several races by whom the languages and dialects mentioned above are spoken appear to be as follows :— Tamil ° 10,000,000 Telucu 14,000,000 5,000,000 Canarese 2,500,000 Malayalam Tulu 150,000 Tuda T O Ov i W N = Kota 500,000 o] Gond © K —— 32,150,000 According to this estimate the Dravidian race numbers upwards of thirty-two millions of souls There cannot be any doubt of their numbers amounting to at least thirty millions ; of whom about twenty millions are British subjects, and the remainder belong to the native the states of Hyderabad, Nagpore,* Mysore, Travancore, and Cochin gross population of which, including all races, is estimated at twenty millions, but is probably much greater In this enumeration of the Dravidian languages I have not included the idioms of the Ramusies, the Korawars, the Lambédies, the Vedars, the Male-arasars, and various other wandering, predatory or forest tribes The Lambédies, the Gipsies of the Peninsula, speak a dialect of the Hindustani the Ramisies and the majority of the Korawars, a patois of the Telugu the tribes inhabiting the hills and forests, corrupted dialects of the languages of the contiguous plains None of these dialects is found to differ essentially from the speech of The the more cultivated classes residing in the same neighbourhood. Male-arasars, ¢hill-kings’ (in Malay&lam, Mala-araans), the hill tribe inhabiting the southern Ghauts, speak corrupt Mala,ydldm in the northern purt of the range, where the MalayAlam is the prevailing language, and corrupt Tamil in the southern, in the vicinity of Tamil-speaking districts * Since the above was written Nagpore has been incorporated with the British territories 10 NTRODUCTION In the above list of Dravidian languages I have not included the idioms of certain rude tribes of Central India and the north-eastern frontier, which have sometimes of late been included under the general term ¢ Tamulian I refer to the languages of the Koéls and Suras the neighbours of the Gonds and Kunds towards the north, which might naturally be supposed to be allied to the Gond or the Ku, and consequently of Dravidian origin ; but which, though they contain a few Dravidian words, belong to a totally different family of languages Without the evidence of similarity in grammatical structure, the discovery of a few similar words proves only local proximity, or the existence of mutual intercourse at an earlier or later period,—not the original relationship either of races or of languages I leave also out of account the languages of the north-eastern frontier of India, which are spoken by the Bddos, Dhiméls, and other tribes inhabiting the mountains and forests between Kumaon and Assam These are styled ¢ Tamulian’ by Mr. Hodgson, of Nepaul on the supposition that all the aborigines of India, as distinguished from the Aryans, or Sanscrit-speaking race and its offshoots, belong to one and the same stock; and that of this aboriginal race, the Tamilians of Southern India are to be considered as the best representatives But as the relationship of those north-eastern idioms to the languages of the Dravidian family is a supposition which is unsupported by the evidence either of similarity in grammatical structure or of a similax vocabulary, and is founded only on such general grammatical ana- logies as are common to the whole range of the Scythian group of languages, it seems to me as improper to designate those dialects ¢ Tamulian or ¢ Dravidian as it would be to designate them ¢ Turkish’ or ‘Tungusian Possibly they form a link of connection between the Indo-Chinese, or Tibetan family of tongues, and the Kol but even this is at present a hazardous assumption. Prof. Max Miiller proposes to call all the Non-Aryan languages of India, including the Sub-Himalayan, the K6l and the Tamilian families, ¢ Nishada-lan- guages,” the ancient aborigines being often termed ¢ Nishadas’ in the Vedalec writings Philologically I think the wuse of this common term is to be deprecated, inasmuch as the Dravidian languages are radically different from the others, as the Professor himself appears to have perceived. For the present I have no doubt that the safest common appellation is the negative one, ‘Non- Aryan or ¢ Un- Sanseritie.’ The brief vocabulary of the tribe inhabiting the Rajmahal hills in Central India, contained in the Asiatic Researches, vol. v., and the fuller list of words belonging to the language of the same people, DRAVIDIAN IDIOMS NOT MERELY PROVINCIAL DIALECTS. 11 contained in Mr. Hodgson’s collections, prove that the Rajmahal idiom is in the main Dravidian. The proof of this fact will be exhibited in the sequel. This language is not to be confounded with its neighbour, the speech of the Santls, a branch of the extensive Kol family inhabiting the Bhaugalpore range of hills, whose language belongs to a totally different stock. The Brahui, the language of the mountaineers in the khanship of Kelat in Beluchistan, contains, not only some Dravidian words, but a considerable infusion of unquestionably Dravidian forms and idioms ; in consequence of which this language has a much better claim to be regarded as Dravidian or Tamulian than any of the languages of the Nipal and Bhutin frontier, which had been styled ‘Tamulian’ by Mr. Hodgson. I have not included, however, the Brahui, or the Rajmahal and Uraon, in the list of Drividian languages which are to be subjected to systematic comparison (though I shall give some account of them in the sequel, and shall refer to them occasionally for illustration), because the Dravidian element contained in those lan- guages bears but a small proportion to the rest of their component elements. Tae DRAVIDIAN IDIOMS NOT MERELY PROVINCIAL DIALECTS OF THE SAME LANGUAGE. Though I have described the nine vernacular idioms mentioned in the foregoing list as dialects of one and the same original Dra- vidian language, it would be erroneous to consider them as dialects’ in the popular sense of the term,—viz., as provincial peculiarities or varieties of speech. Of all those idiomg no two are so nearly related to each other that persons who speak them can be mutually under- stood. The most nearly related are the Tamil and the Malayalam ; and yet it is only the simplest and most direct sentences in the one language that are intelligible to those who speak only the other. Involved sentences in either language, abounding in verbal and nominal inflections, or in conditions and reasons, will be found by The those who speak only the other language to be unintelligible. Tamil, the Malayalam, the Telugu, and the Canarese, have each a distinet and independent literary culture ; and each of the three former—the Tamil, the Malayilam, and the Telugu—has a system The Canarese character has of witten characters peculiar to itself. been borrowed from that of the Telugu, and differs bnt slightly from I it ; but the Canarese language differs even more widely from the Telugu than it does from the Tamil ; and the ancient Canarese character is totally unconnected with the character of the Telugu. 9 INTRODUCTION. The Malayalam being, as I conceive, an ancient offshoot of the Tamil, differing from it chiefly by the disuse of the personal termina- tions of the verbs,* it might, perhaps, be regarded rather as a very ancient dialect of the Tamil than as a distinct language. Its separa- ? tior from Tamil evidently took place at a very early period, before the Tamil was cultivated and refined. Through the predowinance of Brahmanical influence in the Malayala country, the Malayalam has not been cultivated ab intra to any considerable extent ; and the infu- sion into it of a large proportiou of Sanscrit words is almost the only refinement which it has received. The proportion of Sanscrit words which has been adopted by the Drividian languages is least in Tamil, most in Malayélam ; and the modern Malayala character has been borrowed with but little alteration from the Grantham— the character in which Sanscrit is written in the Tamil country, and which corresponds to the Déva-nagari of Northern India. In consequence of these things, the difference between the Tamil and the Malayélam, though originally slight, has progressively increased ; and hence the claim of the Malay&lam to be considered, not merely as a dialect of the Tamil, but as a sister language, or at least as a very ancient and much altered offshoot, cannot now be called in question. The Tulu has been represented by Mr. Ellis as a dialect of the Malayalam ; but although Malayéla characters are ordinarily employed in writing Tulu, in consequence of the prevalence of Malayilam in the vicinity, and the literary inferiority of the Tulus, it appears to me capable of the clearest proof that the relation of the Tulu to the It differs Canarese 1s nearer than its relation to the Malayalam. * The derivation of the Malaydlam from the Tamil is well illustrated by the word which is used by it to signify < Bast.’ It is ‘Kirakka,” meaning ‘beneath,’ or ‘downwards,’ which is properly a Tamil word, and corresponds to that which isused to denote ¢ West,—viz., ‘Merku,” ‘above’ or ‘upwards; both of which words necessarily originated in the Tamil country, or the country on the eastern side of the Ghauts; where a lofty range of mountains rises everywhere to the westward, and where, consequently to go westward is to go ‘upwards;’ whilst to the eastward the country slopes ‘downwards’ to the sea. The configuration of the Malayila country is directly and strikingly the reverse of this,— the moun- tain range being to the eastward, and the sea to the westward. Notwithstanding this, the Malayila word for ‘East’ is ¢ Kirakka, ** downwards,” identical with the Tamil word ‘Kirakku,-—a clear proof that the Malayilam is an offshoot from the Tamil, and that the people by whom it is spoken were originally a colony of Tamilians. It is evident that they entered the Malayala country through the Paul ghaut Gap, and from thence spread themselves along the coast, northward to Mangalore, and southward to Trivandrum. Throughout the Malayala gram- mar I have noticed only two forms which are not contained either in the collo- quial or in the high dialect of the Tamil: those two forms are'a dative in ‘a,’ which is used in some instances after ‘n,’ instead of the more usual sign of the m dative, ‘kka; and a plural suffix of the second person of the imperative,—viz. in; which is peculiar to this language,—except, indeed, it is derived from the high Tamil ‘min.’ RUDER DRAVIDIAN TONGUES. 13 widely and essentially from the Tamil; and hence the tradition which is mentioned by Mr. Taylor, that the ancient * Kurumbars,” or ‘ nomadie shepherds,” in the neighbourhood of Madras were expelled and their lands seized upon by Vellalars from Tuluva, appears to be highly improbable. The colloquial Tamil of the neighbourhood of Madras is characterized by an infusion of the peculiarities, not of the Tulu, but of the Telugu Of the five cultivated Drividian dialects mentioned above—the Tamil, the Telugu, the Canarese, the Malayala, the Tulu, —the farthest removed from each other are the Tamil and the Telugu. The great majority of the roots in both languages are, it is true, identical ; but they are often so disguised in composition by peculiarities of inflexion and dialectic changes, that not one entire sentence in the one language is intelligible to those who are acquainted only with the other. The various Drividian idioms, though sprung from a common origin, are therefore, to be considered not as mere provincial dialects of the same They are as speech, but as distinet though affiliated languages. distinet one from the other as the Spanish from the Portugu:se, the Irish from the Welsh, the Hebrew from the Aramaic, the Hindi from the Bengéli. If the cultivated Dravidian idioms differ so materially from each other, it will naturally be supposed that the uncaltivated idioms— the Tuda, the Kota, the Gond, and the Ku—must differ still more widely both from one another and from the cultivated languages. This supposition is in accordance with facts. So many and great are the differences and peculiarities which are observable amongst these rude dialects, that it bas seemed to me to be necessary to prove, not that they differ, but that they belong, notwithstanding their differ- <o ences, to the same stock as the more cultivated tongues, and that they have an equal right to be termed ¢Dravidian.’ Evipence tHAT THE TupA, Kora, Gonp, AND KU, ARE REALLY DrAvIDIAN ToNGUES. It is unnecessary to state in this general introduction, the parti- culars in which the cultivated Dravidian idioms agree with one another, Sy e AR T P e eogdes T and the evidences of their essential unity even in minor matters and of their common origin : but the Tuda, Kota, Gond, and Ku being rude uncultivated dialects, and little known, it appears to be desirable D u — SRS - at the outset to furnish the reader with proofs of the assertion that =~ < those languages belong to the same Dravidian stock as the Tamil and = the Telugu. Their Dravidian character and connections will appear = from the following statement of particulars, which I have ascertained concerning each of them respectively. 14 INTRODUCTION. (1.) Tuba.—Tt is a favourite opinion with many persons in India that the language of the Tudars is altogether sui generis, or at least that it is unconnected with any of the languages of the Drividian races of the neighbouring plains. In adopting the conclusion that the Tuda language belongs to the Drividian stock, and justly claims to be regarded as a Drividian dialect, the evidence on which I place most reliance is that of a list of words and short sentences which was kindly communicated to me by the Rev. Mr. Metz, German missionary at Kaity, on the Nilgherry hills. Mr. Metz’s acquaintance with the language of the Tudars, is believed to be more accurate than that which has been acquired by any other European ; and though his knowledge is confessedly defective in the department of verbal modifications and syntax, his list of vocables may be fully depended upon : and inasmuch as 4is knowledge of the Tuda has been acquired through the medium of the language of the Badagars, a language with which he is inti- mately acquainted, it cannot be supposed in his case (as was supposed by some persons with respect to the Rev. Dr. Schmid’s inquiries), that he may have accepted Badaga words for Tuda, through ignorance of the dialectic peculiarities of the old Canarese idiom which is spoken by the Badagars. The following words—* prerogative instances, as they would be called by Abel Remusat—prove the Drividian character of the Tuda language. Tupa. Hicua Tamrn, &ec. an* yan ; 4n, Ancient Can. we om yam, 6m thou ni ni you nima nim, Ancient Can. my en en thy nin nin he (proximate) ivan ivan they (ditto) ivar ivar he (remote) avan avan avar they (ditto) avar three mitdu mdu, Tel. four nanku ningn SIX an Aru ten poth’ pattu hundred nfir nfiru a person al al father af)pam abpan mother avva avva, ‘grand-mother’ in Tel. son mach mag-an hand kai kei foot kal kal eye kannu kan, kannu mouth bai val ear kevi kevi, Can. moon tiggalu tingal sky bén vin-am day nal ill nil. night irul, ¢ darkness’ in Tamil water nir nir £ fire nebbu il nippu, Tel. * In the Tuda language 4 is pronounced broad like aw in fawn.” iUDER DRAVIDIAN TONGUES. 15 The Tuda words given above scarcely at all differ from their Tamil, Canarese, i» and Telugu equivalents. In many cases, however, the word, though undoubtedly Dravidian, can scarcely be recognised in its Tuda shape. The following are examples of this:— Tupa Tawmin, &ec. tooth pal parsh tiger puli, pili pirsh veyil birsh sun, or sunlight finger viral bolh belly vayaru bir fruit param vom In the above examples the regular change of I into rsh is especially deserving of notice. In some cases the Tuda words correspond to the Telugu rather than the Tamil, e. g. TAMIL. TELUGU TuDA. tree maram méinu maeéna fire neruppu nippu nebbu Sometimes the Tuda corresponds with the Canarese, rather than with either the Telugu or the Tamil, e. g. TAMIL. TELUGU. CANARESE. TupA. small s'inna chinna kinna kin ear sevi chevi kevi kevi The Tuda generally agrees more exactly with the Tamil than with the Telugu, the Canarese, or any other Drividian idiom. In many particulars so close is this agreement that the Tuda might be considered as merely a corrupt patois of the Tamil, were it not that in a still larger number of instances it differs, not only from the Tamil, but also from every other Dravidian dialect, pursuing a course of its own with a vocabulary of its own; in consequence of which it must be regarded as a On an examination of the Tuda words contained distinet member of the family. in the lists in my possession, exclusive of pronouns and numerals (which are throughout Drividian), forty per cent. are found to be allied to Drividian words belonging to the languages of the adjacent plains, whilst sixty per cent. appear to be either independent of those languages or to be so greatly corrupted and disguised that their relationship cannot now be ascertained. The following comparison of the forms of the present and future tenses of the substantive verb in Tuda with those of the Tamil will illustrate the verbal The root of the substantive verb in Tamil and in inflexions of this language. Canarese is ‘ir In Tuda the corresponding root is ‘ers,” ‘etars,” or ‘esh.’ TAMIL. TupA. I am yan irukkirén 4n eshken thou art ni irukkirai ni etersbi he is avan irukkirin avan etarji we are Am etarsbimi yam irukkirdm ye are nir irukkirir nima etarshi avar irukkirfr avar etarshi they are 16 INTRODUCTION. TAMIL TuDA. I shall be yan iruppén an ersbini I - thou wilt be nf iruppai ni ersbi he will be avan iruppan avan ersje we shall be Am ershimi yam irupp'Gm ye will be nir iruppir nima ersshi avar ersshi they will be avar iruppir It is evident that the third person singular and plural of each of the Tuda tenses is destitute of personal terminations. This is in accordance in part with the usage of the Canarese, and still more with that of the Telugu. i The Tuda language contains exceedingly few words of Sanscrit origin, and those few have evidently been derived from the intercourse of the Tudars with their neighbours the Badagars, colonists from the Canarese country,—e.g., ‘der, God, from the Canarese ‘dévar, and the Sanserit ‘déva;’ and ‘budi, wisdom, from ‘buddhi,’ the Sanscrit word for wisdom adopted in the Canarese. The substantial agreement of the Tuda with the other Drividian languages in its pronouns, its numerals, the first and second persons of its verbal inflections, [ and in forty per cent. of its stock of ordinary vocables, proves beyond all doubt [4 the propriety of considering it as a Drividian dialect; and it seems scarcely less il certain that of all the Drividian idioms the Tamil is that to which it is most i, fh nearly allied. Sixty per cent. of its words appear, indeed, to be unconnected la with the vocabularies of the other dialects; but those words are chiefly such as T P VB i g e 1 A are remote from the daily business of life, and which are therefore of all words il the most apt to become corrupted by a barbarous, isolated tribe. It is also to be I | remembered that each of the cultivated Drividian languages contains a consider- al able number of roots of this secondary, fleeting class, which are not found in any 10 other dialect of the family. Such words do not necessarily belong to an Extra- Dravidian source ; for no one language of any fam'ly whatever is in possession of all the roots which originally belonged to the parent stem. Rach dialect of the I Sanserit, of the Classical, and of the Germanic families of tongues is found to have retained a certain number of roots which the other dialects have suffered to [{ 80 become obso'ete. Whilst the language and customs of the Tudars have always Ig; (2.) Kora. 10 been regarded with peculiar interest, the Kotars (a tribe of craftsmen, residing %&j tho from an unknown antiquity on the Nilgherry hills), being exceedingly filthy in {hy their habits, and addicted Leyond all other low caste tribes to the eating of iy carrion, have generally becn shunned by Europeans ; and, in consequence, their language is less known than that of the Tudars. Notwithstanding this, the it} following paradigm of the Kota pronouns and of the present tense of its verb, which was furnished me by the Rev. Mr. Biihler, of Kaity, will show that the Iy language of this degraded tribe is essentially Drividian :— i} I Kora. AncrENT CANARESE. Igo ane hogabe 4n pdgdapén fin ir 0ft thou goest ni hdgabi nin pogdapi ¢ he goes awana hogako avam pogdapam adg we go name hogabemme am pdgdapévu ye go niye hégabirri nim pogdapir emp they go awane hégako avar pogdapar deriy i E — T RUDER DRAVIDIAN TONGUES. 17 In this paradigm the first person plural, both of the pronoun and of the verb, and the second person plural of the verb, accord most with the Tamil ; the other forms agree most with the Ancient Caranese, particularly the formative suffix of the present tense of the verb, which is ‘dap’ in Ancient Caranese, and ‘ab’ in the Kota. In the use of ‘h’ instead of ‘p’ (‘hdgu,’ to go, instead of ‘pdgu’), the Kota accords with the modern Caranese. The third person of the Kota verb, which is formed, both in the singular and the plural, by the suffix ‘ko, seems If we at first sight entirely unconnected with all other Drividian forms. consider it, however, not as a verb properly so called, but as an abstract verbal ;) noun, which acquires the force of a verb from juxta-position with a pronoun, like b the third person in the Persian verb, it may easily be brought within the range of Drividian analogies; for many such verbal nouns in the other dialects end in I Sk ‘kel’ “ka’ ‘ge’ &e. The sign of the genitive case in Kéta is ‘a,’ of h the dative, ‘ke,” of the locative, ‘olge,’—all which forms correspond with those Js which are found in the other dialects. The preterite is formed by changing ly ‘ka’ or ‘ga’ into ‘ji’ or ‘di;y—e.g., ‘hégako,” he goes; ‘hdjiko, he went: ‘tinkabe,” I eat; ‘tindibe, 7 ate. In this also we see a family resemblance to the n manner in which the other dialects, especially the Telugu, form their preterites. : X, The Kota forms its infinitive by the addition of ‘alik’ to the root,—e. g., ¢tin, bt eat ; ‘tinalik,” to eat. The. infinitives of the corresponding verb in Canarese o8 are ‘tinna,” ‘tinnalu,” ¢tinnalike. On the whole, though certain analogies with b the Tamil and also with the Tulu may be observed in the Kota, I regard this ol language as more nearly allied to the Canarese than to any other Drividian idiom. a8 3. Tae GOND.—The very complete grammar and vocabulary of the Mahadeo 8 dialect of the GoOnd language, which was compiled by the Rev. Mr. Driberg, a be late missionary of the Society for the Propagation of the Gospel in the Saugor el and Nerbudda country, and which was published at Bishop’s College, Calcutta, ny in 1849, together with Dr. Manger’s interesting paper on the dialect of the Saonee 14 (Seoni) Gonds, including ¢ The Song of Sandsumjee,” in the Journal of the Bengal of Asiatic Society,—contain so many proofs of the close affinity of the Gond the language to the Tamil, the Telugu, and the Canarese, that it seems quite unne- 5 It is not Ve cessary to prove in detail that it is a member of the Drividian family. to 80 easy to determine to which of the cultivated Drividian dialects it is most nearly allied. In many respects it accords most with the Telugu, its neighbour ays to the South and East; but on the whole, it is more closely allied to the Tamil, ing though locally of all Drividian dialects the farthest removed from it—a proof i ‘that the claim of the Tamil to be considered as the best representative of the rOf primitive condition of these languages is not destitute of foundation. el The chief particulars in which the Gond agrees with the Telugu, rather than the with the Tamil or with the Canarese, are as follows :— Clb; (1.) The pronouns of the first and second persons, especially the second {he person plural, have most resemblance to the Telugu. Compare ‘mik,’ Gond, to you, Telugu, ‘miku,’ with the Tamil ‘umakku’ and the Canarese ‘ nimage.’ (2.) Another point of resemblance to the Telugu consists in the absence of a feminine form of the pronoun of the third person singular and of the third person of the verb, and the use of the neuter singular for the feminine singular. (3.) The Gond preterite verbal participle is formed, like the Telugu, by the addition of ‘si’ or ‘ji’ to the root, instead of the ‘du’ which is so largely employed by the Tamil and Canarese. (4.) A certain number of roots of secondary importance and a few Sanserit derivatives seem to have been borrowed by the Génd from the Telugu ;- e. g., C 18 INTRODUCTION. ‘nattur, blood, from the Telugu, ‘netturu, a corrupt derivative from the Sanscrit, ¢ ractam.’ In some instances again the Gond agrees remarkably with the Canarese; e. g., the G6nd infinitive is in ‘4lle,” or ‘ille.’ In Telugu and Tamil the infinitive is invariably in ¢a:’ the Tamil has a verbal noun ending in “al,” of which the dative is used as a supine ; and the High Tamil occasionally, but the Canarese ordinarily The Gond also like the Canarese some- uges this very form al’ as an infinitive. times prefers ¢k’ where the Telugu has ‘ch’ and the Tamel ‘s;’ e. g., the ear, is in Tamil, ‘Sevi; Telugu, ‘chevi; Canarese, ‘kevi; in Gond also, ‘ kaovi’ To do, is in Tamil, ‘dey; Telugu, ‘chéj; Canarese, ‘géy’ (g hard); Gond, ‘ki.’ Such agreements of the Gond with the Canarese are rare; but the particulars in which the Go6nd agrees with the Tamil, though the Telugu country lies between it and the country in which the Tamil is spoken, are very numerous and important. The following are specimens of this agreement. (1.) The Telugu has but one form for the plural of nouns substantive, the suffix ‘lu; the Tamil has two, ‘ar’ and ‘kal’ the former epicene, the latter neuter : the Gond also has two, ‘ar’ and ‘k. (2.) The instrumental case of the Telugu is formed by the addition of ‘ chéta:’ the Gond like the Tamil uses ¢al.’ (8.) The Go6nd differs from the Telugu, and accords with the Tamil in retaining unaltered the initial vowel of its pronouns in the oblique cases. Thus, from ‘adi, Telugu, ¢, comes ‘déni,” of #¢; Tamil, ¢adin,’ of i¢; GOnd, ‘adena. (4.) The Telugu negative particles are ‘1édu,’” there is not, and ‘kédu,” ¢ @s ﬂ not ; the corresponding particles in Tamil are ‘illei” and alla;’ in GOnd, * hille’ and ¢ halle.’ (5.) The Telugu systematically uses ‘d” instead of the Tamil vocalic ‘r: the Gond retains the ‘r’ of the Tamil; e. g., ‘édu’ or “adalu,” Telugu, fo weep ; Tamil ‘ara,” Gond ‘ara. So also compare ‘édu, Telugu, seven, with Tamil “éru’ and Gond ¢éro. A considerable number of Go6nd roots denoting objects of primary importance correspond with the Tamil rather than the Telugu; e. g. TELUGU. TAMIL. GOND. one okati ondru und three mindru mund ! mfidu i hand chéy kei kai tree manu maram marri f] great pedda peru, paru paror to come vachcha (vatssa) vara wara al [h\ In a large number of instances the Go6nd, though retaining the same roots as il the other Drividian dialects, modifies those roots after a fashion peculiar to itself. o This will appear on comparing the following Tamil and Gond words. i TAMIL. GOND. I of flower ph pungar belly vayaru pir i il boy peiyal perdgal to fall vira ara P nira niha b to fill light velicham verchi hy walle an many pala Lo RUDER DRAVIDIAN TONGUES. 1D 16 TAMIL. GOND. district Jy nadu nér (a village) dew pani pini (cold) 18 break urreha Ve udei Iy Notwithstanding the affinities between the Génd and the other Drividian 17 dialects which have now been mentioned ana illustrated, the Gond possesses 18 a large number of roots which are not found elsewhere, and exhibits peculiarities [o of grammatical structure of such a nature as amply to justify our regarding it as 4! a distinet dialect. The difference existing between the Tamil and the Telugu n sinks into insignificance when compared with the difference between the Gond e and every other dialect of the Dravidian family. In the list of Génd words given nf by Mr. Driberg, I have been able to identify only thirty-four per cent. as words contained in or allied to those that are found in the other dialects, which is a the smaller proportion than that which is contained even in the Tuda. ter The principal particulars in which the grammatical siructure of the Génd differs from that of the other dialects are as follows :— £} (1.) Like the idioms of Northern India, the Génd evinces a tendency to con- found the dative with the accusative, though in possession of both forms. mg (2.) It has a passive voice formed, as in some of those Northern idioms, by di, prefixing the past participle of the active voice to the substantive verb. (3.) The remote and proximate demonstratives (¢ille, ¢hic,’) which in 18 Tamil are ‘avar,” ‘ivar;’ in Telugu ‘varu, ‘viru; are in Gond corrupted into ‘wor’ and ‘yer.’ Ile’ (4.) The base of the interrogative pronouns in all the other dialects is ‘y4 ; in this it is ‘bd, or ‘bo. the oy (5.) Instead of the regularly formed negative voice of the other dialects, the Gond forms its negative verbs by simply prefixing the negative particles ¢hills,’ il or ‘halle,” to the verb. For example, thou art not, or thou becomest mot (in Tamil ¢4ghy,’ in Telugu, ‘kivu’), is in Gdénd ‘halle aivi. A similar use ance of the negative particle is found in the Kota language. The only thing in the other dialects which at all corresponds to this, is the occasional formation in poetical Tamil of a negative verb by the insertion of the negative particle ‘al’ between the root of the verb and the pronominal suffix ; e. g., not, for ‘pés-én.’ ‘pésal-én, 1 speak4 (6.) The chief difference, however, in point of grammatical structure between the Gond and the other Drividian dialects, consists in its peculiarly elaborate and complete conjugational system. The Tamil, the Malayilam, the Canarese, and the Tulu possess only a present, an indefinite past, and a future tense,—the future more or less aoristic. fs 88 The Telugu, in addition to these tenseg, has a regularly formed aorist. The indicative and the imperative are the only moods which these tgelf dialects possess, and they are destitute of a passive voice properly so called. All modifications of mood and tense are formed by means either of auxiliary verbs or of suffixed particles. Whilst the more cultivated Dravidian idioms are so simple in structure, the speech of the rude Mahadeo Génds boasts in a system of verbal modifications and inflexions almost as elaborate as that of the Turkish. It has a passive voice: in addition to the indicative and the imperative moods, it possesses a potential: in the indicative mood, where the Tamil has only three tenses, it has a present, an imperfect definite, an indefinite past, a perfect, a conditional, and a future, each of which is regularly inflected : like the other idioms, it has a causal verb, but it stands alone in having also an inceptive. In these particulars €2 - INTRODUCTION. 20 the Gond grammar has acquired a development peculiar to itself, perhaps in some degree through the influence of the highly inflected Santdl, its Kol neighbour to the northward. (4.) Tae Ku.* The Kond, Khond, or Ku language, undoubtedly a Drévidian idiom, has generally been considered as identical with the Gond. It was stated long ago by Captain Blunt in the Asiatic Researches, vol, vii., on the authority of a native Jaghiredar, that the Goénds and Khunds are totally distinct races : notwithstanding this, I have not met with any account of their languages in which they have been regarded as different, though in truth their differences are numerous and essential. In many particulars the Ku accords more closely than the Gond with the Tamil, the Telugu apd the other Drividian tongues; in some things less so. For example :— (1.) The Gond forms its infinitive in ‘alle, or ¢ille;’ the Ku, like the Telugu, the Tamil, and the modern Canarese, forms its infinitive by suffixing ¢a,’ sometimes ¢va, or ‘pa’ Thus, fo become, is in Gond ‘aialle; in Telugu, ) ¢kay in Canarese, ‘4gal’ or ‘iga; in Tamil, ‘4ga; in Ku, ‘4va (2.) The Ku retains the simplicity of the conjugational system of the other Dravidian dialects, in contradistinetion to the elaborateness of the Gond. (3.) The Goénd forms its negatives by prefixing to the indicative aorist the In this point the Ku differs geparate negative particles “hille,” or “ halle.” from the Godnd, and agrees with the other dialects. Thus, 7 do not, is in Gond “hille kion; in Tamil ‘seyyén; in Telugu ‘chéyanu; in Canarese ¢ géyenu,’ in Ku “giénu.’ In the following instances the Ku accords more closely with the Tamil and 1'11] Canarese, though locally very remote, than with its nearer neighbour the t Telugu. I (1.) The Telugu forms its plurals by the use of lu’ alone, except in some of the oblique forms of the ‘rational’ demonstratives. The Ku, like the Tamil, makes a difference between the plurals of nouns which denote rational beings, The Tamil suffix of the first class of and those of nouns of the inferior class. plurals is ¢ar,’ of the second class ‘kal:’ the corresponding suffixes in Ku are “4ru’ or ‘ru, and ‘ka. (2.) The Telugu forms its masculine singular by means of the suffix ‘du?’ § the Canarese and Tamil by ‘anu’ and ‘an.’ The Ku by means of the suffix I ‘Afiju’ or ‘4nyu.’ Thus, compare ‘vadu,” Telugu, he, with the Tamil ‘avan,’ 1 Canarese ‘avanu, Ku ‘avéfju.’ U (3.) The Ku pronouns bear a closer resemblance to the Tamil and Canarese i than to the Telugu and Gond, as will appear from the following comparative view :— TELUGU. GOND. T AMIL. CANARESE. Ku ! I nénu and yan (ancient 4n (ancient) anu we mému amab ydm (do.) am (do.) amu thou nivu 1ma ni ninu fnu § miru imat nir nivue iru ye avanu he, remote vadu wor avan avafiju he, proximate vidu yer ivan ivanu ivadiju * See a 1-ucid1y arranged grammar of this language prepared by Lingam T ———— Letchmajee, Deputy Translator to the Ganjam Agency, and published in Uriya I have characters in the Calcutta Christian Observer for May and June, 1853. — not seen any notice in any scientific work or periodical of this valuable contribu- tion to our knowledge of the Indian languages. 21 RUDER DRAVIDIAN TONGUES. 16 0 (d.) In the Dravidian languages contingency is expressed by the addition of a particle to any verbal tense, person or number. A This subjunctive suffix is in Telugu ¢ éni’ or ‘e in Canarese ‘re, ‘rQl) or ‘agyu.’ One of the suffixes il employed in the Tamil is ‘kkal,’ which in the speech of the vulgar becomes i <kki s and this very particle ‘kkA’ added, as in Tamil, to the preterite, is the ) suffix by which the Ku also forms conditional or contingent verbs: e.g., If 1 do, is in Telugu ‘nénu chéyudunéni; in Canarese ‘ninu géyidare; in colloquial ch Tamil this is ‘nin cheyddkkd; in Ku also, (from the root gi,’ to do), it is ‘4nu 16 gitekki.’ an On the other hand, in the following particulars the Gond agrees more closely 16 with the Telugu than with the Tamil or Canarese. (1) It uses the neuter singular to denote the feminine singular. he (2.) The oblique cases or “inflexions” of the pronouns of the first and ng second persons, singular and plural, are identical with those of the Telugu. Uy (8.) The case terminations of the Ku are nearly in accordance with those of the Telugu. er (4.) The pronominal signs suffixed to the Ku verbs accord on the whole better with the Telugu than with any other dialect : e.g., in Tamil the second and third the persons plural end differently, the one ‘ir, the other ‘ar; in Telugu they end oS alike—both generally in ‘aru; in Ku also both these persons end alike in ¢ eru.’ ind (5.) In Canarese all relative participles, including that of the negative verb, end in ‘a; in Tamil all relative participles, with the exception of that of the future, Dl have the same ending: in Telugu the relative participle of the indefinite or and aoristic tense ends in ‘edi,” or ‘eti; and in the Ku also the relative past par- the ticiple exhibits this ending. Thus, ¢4na,’ Tamil, that became; in Canarese, ‘4da;’ in Telugu (indefinite tense), ‘ ayyéti; in Ku the same form is ‘4ti. The various particularsand illustrations which have now been mentioned prove me 111117 the Ku to be totally distinct from the Gond ; and though it is allied to it, it is allied only in the same manner as to the other Drévidian languages. In some ng* g of points this language differs from all the other dialects of the family; for example, it forms its past verbal participles not by means of the suffixes ‘du, ‘i, or ‘si, e the only suffixes known in the other dialects, but by suffixing to the root ‘4, fu! sometimes ‘sd’ or ‘ji, after the manner of some of the languages of Northern India. In the other dialects of this family the negative verb possesses only one iffis (an, tense, an aorist; the Ku, in addition to this negative aorist, has also a negative preterite,—a decided advantage over the other dialects. The Ku suffixes of the rese present verbal participles are also different from those which are found in the other Drividian dialects. The formative suffix of the present verbal participle is in gtive ) Telugu chu’ or ¢tu; in the Canarese ‘ta’ or ‘te; in the Ku it is i’ or ‘pi. The four dialects referred to above—the Tuda, Ko6ta, Génd and Ku—though rude and uncultivated, are undoubtedly to be regarded as distinctively and essentially Dravidian dialects, equally with the Tamil and Telugu. In addition to these, there are two uncultivated idioms of Central India, the Uréon and the R4jmahal, which contain so many Dréavidian roots of primary importance that they may claim g to be considered as originally members of the same family, though Tripp hat they contain also a large admixture of roots and forms belonging b to the Kol dialects. The Urdon is considered by Mr. Hodgson as a 22 INTRODUCTION. connecting link between the Kol and the Rajmahal ; and the Rajma- il hal as a connecting link between the K6l and the Tamulian families, (1 The Rajmahal is more distinctively Dravidian than the ﬁrﬁon, though { the Males or aborigines of the Rdjmahal or D{Iale range by whom it is 8l spoken, are locally more remote than the Urdons from the present fi seats of the DrAvidian race. The Kols intervene between the Males and the Dravidians ; but whilst the Male is substantially a Dravidian 0 tongue, the Kol belongs to a totally different family. It In the list of R4jmahal or Male words given by Mr. Hodgson, sixteen per cent, are purely Drividian ; in the older list, given in the Asiatic Researches, the Dri- vidian roots form only ten per cent. In the Urfon list the proportion of Drividian roots is fourteen per cent. The principal and most essential analogies which I have noticed are as follows :— § RASMAHAL, URAON. Tamin, &e. eye kiné khin kan I eye-lash kan-meer kan-mayir hand sésht kﬁheiziﬁh chéyi (Tel.), kyé (Canar.) ear 0of kkhétway kadu, kél, to hear th mouth bai vay tooth pall pall pal 1 hair tali talei, head § father abi appa il mother ayi aya ayyo- D tree man man mén-u (Tel.) flower plp phip pl; plppu, a flowering min min fish i dove plirah purd tilah 61 scorpion pain nogi nogu, to pain above méché méyajl; mél, mélé below e kiyah kir, kiré within flé fila ul, ullé 1 en enan nin, én 1o thou nin nien nin (Can.) he, she, it ith asan ata (Can.), he, adu, ¢¢; adum, By (Tuda), ke, she, it of we nam, om en nim, dm ye nina (nim in nimki, nim (Can.) a0y Teo Yyours) it they awir, asabar avar this th i that ah & here ino inge there ino ange one art, orb ﬁnt,émlll. or, ondru two en-otan (Sanserit) rendu (Tel.), ranu (Gond.) three (do.) man-otan mfinna (Malayal). our (do.) nakh-otan nilku, nanku the 1) B— — 23 THE BRAHUIL Unfortunately the inflexions of the Rdjmahal and Urdon nouns and verbs are not given in any of the lists, so that, with the exception of a few incidental parti- culars, the grammatical counstruction of these languages remains unknown. In the particulars that follow they accord with the Dravidian grammatical rules. The Rijmahal expresses the prepositions to, in, on, &c., by suffixes. Its dative suffix is ‘kus ‘m’ is the sign of the plural number of the pronouns of the first and second persons, replacing ‘n,’ the corresponding sign of the singular; “ar’ is the sign of the plural of epicene pronouns of the third person. The sign of the possessive pronouns is ki, or in the Urfon ‘ghi,’ corresponding appa- rently to the ‘ké, ki, of the Hindi, and more remotely to the *yoka’' of the Telugu. The existence of a distinetively Dravidian element in these abori- ginal dialects of Central India being established, the Dravidian race can now be traced as far North as the banks of the Ganges ; and the supposition (which was deduced from other considerations) that this race was diffused at an early period throughout India is confirmed. The Brahui, the language of the Belichi mountaineers in the khanship of Kelat enables us to trace the Dravidian race beyond the Indus to the southern confines of Central Asia. The Brahui language, con- sidered as a whole, is derived from the same source as the Panjabi and Sindhi ; but it unquestionably ¢ontains a Dravidian element, an ele- ment which has probably been derived from a remnant of the ancient Dravidian race incorporated with the Brahuis. The discovery of this Dravidian element in a language spoken beyond the Indus proves that the Dravidians, like the Aryans, the Graeco-Scythians and the Turco- Mongolians, entered India by the north-western route. The following is an outline of the particulars in which the Brahui is fognd to be allied to the Drévidian tongues. (1.) In Brahui, as in the Drividian dialects, the cases of nouns are denoted by post-positions. (2.) The gender of nouns is expressed not by their inflexions, but by prefixed separate words. (3.) The number of nouns is ordinarily denoted by the use of separate particles ly of pluralisation, such as many, several, &e. When a noun stands alone without any such sign of plurality, its number is considered to be indefinite, and it is then regarded as singular or plural according to the context, or the number of the verb with which it agrees. This rule is remarkably in accordance with the Tamil. (4.) Adjectives are destitute of comparatives and snperlatives. (5.) Pronouns form their genitives in ‘mna’ or ‘aj eg., ‘kana,” of me; ‘nana,’ of us. Compare Tamilian ‘nama,’ of us, and the Gond genitive suffix ‘na’ or ‘a.’ (6.) The Brahui dative-accusative is in ‘e’ Compare the Malayéila accusative €&’ Tamil ‘ei’ (7.) The Brahui pronoun of the second person singular is ‘ni, thou, precisely the same as in all the Drividian tongues. The analogy of the plural of this pro- noun, viz., ‘num, you, ‘numa, of you, is also wonderfully in accordance with IRt A LY, T a1t AN e s Z e 24 INTRODUCTION. i classical Dravidian forms The Canarese is ‘nim,” you; the old Tamil possessive ]:; is *num-a,’ your (derived from an obsolete nominative in ‘ntim’ or ‘num’), and the ordinary base of the oblique cases of this pronoun in colloquial Tamil is ¢ um’ (the initial ‘n’ being lost), which is also the termination of the plural of the gecond person imperative. (8.) Whilst ‘nim’ or ‘nfim’ is to be considered as the most classical form of the plural of the Dravidian pronoun of the second person, ‘nir’ is the form ordi- narily used in a separate shape in Tamil, ‘miru’ in Telugu; and in consequence of this plural termination in ‘r,’ in nearly all the Drividian idioms the second person plural of the verb in the indicative mood ends, not in ‘im’ or ‘um, but in ‘ir, ‘eru Y. i A aru, ’ iri; &e. The same peculiarity reappears in the Brahui. Whilst the separate pronoun ends in ‘m,” ‘r’ is the pronominal sign of the second person of the verb; e.g., ‘areri) ye are, ‘aver, they are; with which compare the Canarese ‘iru(tt)iri, ye are, ‘iru(tt)ire,’ they are. (9.) The root of the substantive verb in Brahui is ‘ar, in Canarese and Tamil ¢ir. (10.) A remarkable analogy between the Brahui and the Drividian languages is apparent in the reflexive pronoun self; ‘se.’ In the Drividian languages this pronoun is universally ‘tdn’ or ‘tan :’ in the Brahui ‘ten. (11.) Bopp remarks that the three lowest numerals could never be introduced d into any country by foreigners. The truth of this remark is illustrated by a cir- d cumstance of which Bopp could scarcely have been aware. From jfour upwards, b the Brahui numerals are of Indo-European origin (e.g., ¢char, four, ‘panj,’ five, “shash,’ siz); and in the compound numerals twenty-one and twenty-two, the (¢ _ _ I words for one and fwo are also Indo-European, but the separate numerals one, two, U three, are totally unconnected with the Sanscrit family, and two of them are iden- th tical with Drividian numerals. In Brahui, two is ‘irat; compare Canarese, W ‘ erad-u,’ two; Tamil, irat(tu),” éwofold or double. In Brahui three is 0l ‘ musit;’ compare Canarese ‘mfir-u; Telugu, ‘mtd-u.’ The Drividian bases of these numerals are ‘ir,” two, ‘mf,’ three; and if we notice the terminations I of the Brahui numerals (1, “asit;’ 2, ‘irat;’ 3, ‘musit;’) it is obvious that the ¢ second syllable of each of these words, ‘it,” or ‘at’ is merely a neuter formative, [ like that which we find in the Drividian languages (e.g., compare ‘ir,’ the base be and numeral adjective fwo, with ‘iradu,” the abstract neuter noun #wo): conse- quently the agreement of the Brahui with the Drividian numerals, both in the base and in the formative, is complete. If we remember the interchangeable 10y relation of ‘s’ and ‘r,’ and if we regard the Canarese ‘mfr,’ three, and the et Brahui ‘mus,’” as an instance of this interchange, as I think we may safely do, 0f 1 we may also venture to connect the Drividian numeral base, ‘or,” one, with the i\ Brahui ‘as.”” This connection, however, is doubtful, whereas there cannot be any W doubt respecting fewo and three. Loy (12.) In the class of auxiliary words (prepositions, conjunctions, &c.) compare I the Brahui, ‘moni opposite, with the Tamil, ‘munné,’ before; and also the copulative conjunction ‘Q,” and, with the corresponding Canarese ‘1.’ A In the limited vocabulary of the Brahui language, which is given in the Jousr- W nal of the Bengal Asiatic Society, I notice a few Dravidian roots. In the following list I include also a few Dravidian words, which are found in the Laghmani, an Afghan dialect, containing an element allied to the Brahui. THE BRAHUI 25 sive Branur, &ec Tamir, &ec. and water dir nir um eye khan kan the ear khaff kad-u do., Laghmani and Cashgari kad | om mother Jummé amma ordi- do., Laghmani ae fyi ence belly pid pir (Gond), bir (Tuda) stone khall kal ond um, bow billa billu (Can.) the mud or earth men man sign bedstead kat kat-(til) with hare muru muyal karatik ass (female) Lagh. karudei, ass and cat, Laghmani pusha plsei to come barak bar-u (Can.) 68 i3 to go, Laghmani pak pog-u - this It is true that the great majority of the words in the Brahui language uced are altogether unconnected with Dravidian roots ; but it must be evi- g (il dent from the analogies in structure, as well as in the vocabulary, which ands, have now been exhibited, that this language contains many grammati- fivey cal forms essentially and distinctly Dravidian, together with a small ) the 0, proportion of impdrtant Dravidian vocables. The Brahuis state that their forefathers came from Haleb (Aleppo) ; but even if this tradition iden- 1686, were to be regarded as a credible one, it would apply to the secondary o 18 or conquering race of Indo-European origin, not to the aboriginal, hases indigenous Dravidians. The previous existence of the latter race jtions seems to have been forgotten, and the only evidence that they ever 11 the itive, existed is that which is furnished by the Dravidian element which hag base been discovered in the language of their conquerors. onse: n the The analogies between the Brahui and the Dravidian languages which have now been pointed out, are incomparably closer than any analogy which- subsists ;eable d the between the Drividian languages and the Bodo, the Dhimal, and the languages of the other tribes on the north-eastern frontier of India which have been termed 51:,' do, “ Tamulian ” by Mr. Hodgson. Those analogies appear to me to be as remote b the e any as those of the Tibetan family; and are not only less numerous, but also of a less essential character and less distinctive than the analogies which are discover- able between the Finnish tongues and the Dréividian. np&re Compare the following list of Drévidian words of primary importance with o the analogous words in the Brahui, and with the words in the Bodo and Dhimal which correspond in signification : — Jour owig DRAVIDIAN. Branur, Bobo. DHIMAL, thou ni ni né nni) o nang you num num nangchir nyél we nim nan jong kyel self tan ten goui tal 26 INTRODUCTION. DRAVIDIAN, Branui. Bobo. DyiMAL. abori A one or as-it ché (& them two irad-u irat gné gne on | three mir-u mus-it thim stim whicl eye kan khan mogon mi ear kad-u kad khom# nihithong Tami Madu water nir dir doi chi stone kal khall onthai finthr celebr itself It seems unnecessary to give a larger number of instances; for whilst the Brahui does to a certain extent contain Drividian forms and words, the Bodo and being the Dhimal, and to them may be added the other dialects of the north-eastern forests, present no special analogies whatever; and contain only a few vague adop structural affinities, which they have in common not only with the Drividian, but guag with the Tibetan, and with every other language and family of languages of the word Scythian group. Tami] Il Use or THE ComMoN TErRM ¢ DRAVIDIAN. Pand denom I have designated all the languages now subjected to comparison By th by a common term, ¢ Drévidian,” because of the essential and distine- tive grammatical characteristics which they all possess in common, Popula and in virtue of which, joined to the possession in commnion of a large Bengél ciently number of roots of primary importance, they justly claim to be con- sidered as springing from a common origin, and as forming a distinet the Uy family of tongues. the 8 lay | This family which T style ¢ Dravidian’ has been styled ‘Tamu- lian’ by some recent writers ; but though the Tamil is the oldest and the Ca most highly cultivated member of the family, and that which contains ing j I the largest proportion of the family property of forms and roots, yet as it is but one dialect out of many, and does not claim to be the original Teh'uga speech from which the other dialects have been derived ; as 1t is also or Tam desirable to reserve the terms ¢ Tamil’ and ¢Tamilian’ (or as it is Ity iy generally but erroneously written ¢ Tamulian’) to denote the Tamil language itself and the people by whom it is spoken, I have preferred i(S lQCall to designate this entire family by a term which is capable of a wider {he qu The word which I have chosen is ¢ Drividian,” a word application. kl‘llguag which has already been used as the generic appellation of this family Sy of tongues by the Sanscrit geographers. Properly speaking, the term Of thl I \ W ]de ly ¢ Dravida’ denotes the Tamil country aloune (including Malayalam), and Tamil Brahmans are usually styled ¢ Drvida Brahmans.’ that ¢ h ¢ Dravida’ means the ‘country of the Drividas; and a Drivida is The defined in the Sanscrit lexicons to be “a man of an outcast tribe, ff)ﬂrr g, descended from a degraded Kshatriya.” This name was doubtless K applied by the Brahmanical inhabitants of Northern India to the thg Dy USE OF THE COMMON TERM ‘ DRAVIDIAN. 27 aborigines bf the extreme South prior to the introduction amongst them of Brahmanical civilisation, and is an evidence of the low estima- tion in which they were originally held. In the Maha-Bharata, in which the Drividas are distinguished from the Cholas, or Tanjore 0 Tamilians, the term is still further restricted to the Pandiyas of Madura, doubtless on account of the advanced civilisation and early celebrity of the Pandiya kingdom. The term ¢Drévidian’ is thus in itself as restricted as that of ¢ Tamilian,” but it has the advantage of 6 the do and being remoter’ from ordinary usage, and somewhat more vague, and eastern the further and more special advantage of being the term already vague adopted by Sanscrit writers to designate the southern family of lan- an, bt guages. Consequently, by the adoption of this more generic term, the of the word ¢ Tamilian” has been left to signify that which is distinctively Tamil. The colloquial languages of India are divided by the Sanserit Pandits into two classes, each containing five dialects. These are denominated respectively ‘the five Gauras' and ‘the five DrAviras.’ arison Istine- By the Gauda or Gaura languages are meant the ¢bhashis,’ or popular dialects, of Northern India, at the head of which stands the Mot Bengali, the GéAura proper. Some of the ¢bhaAshas’ or Priecrits an- y large e 0l ciently enumerated have ceased to bespoken. At present the Bengali, the Uriya, the Hindi with its daughter the Hind{istini, the Panjabs, igtinet the Sindhi, the Gujarathi, and the Marathi are the languages which may be regarded as forming the Gaura class ; to which I would add Tami the Cashmirian and the language of Nipal, thus reckoning in this class st and nine idioms instead of five. ontaids The five Dravidas or Draviras, according to the Pandits, are “the | yet a8 Telinga, the Karntaka, the Maratha, the Gfirjara, and the Dravira,” rginal is also or Tamil proper. The Maratha and Girjara are erroneously included in this enumeration. It is true that the Marithi contains a small s it 18 admixture of Drividian roots and idioms, as might be expected from ; T&mil its local proximity to the Telugu and the Canarese ; and both it and efonted the Giirjara, or Gujarithi, possess certain features of resemblance to the , wider languages of the South, which are possibly derived from the same or a , word similar source ; but, notwithstanding the existence of a few analogies fumily of this nature, those two languages differ from the Dravidian family so e term widely and radically, and are so closely allied to the northern group, yfﬂﬂ»ﬂl)’ that there cannot be any hesitation in transferring them to that class, hmg,]ls., The three languages that remain in the classification of Drividian il tongues which is contained in the Sanscrit geographical lists, viz., the 4 tribe Karnataka, Kannada or Canarese, the Telinga, Telungu or Telugu, and bl the Dravida proper or Tamil, are certainly the principal members of ] 28 INTRODUCTION. It will be observed that the the . the southern or Dravidian family smal Malayﬁlam and the Tulu are not contained in the Sanscrit enumeration. they The first was considered to be a dialect of the Tamil, and was included in the denomination of the Dravida proper ; the second was probably of th the I considered as a dialect of the Canarese. The uncultivated dialects— past the Tuda, Kota, Gond, and Kn—appear to have been unknown to the Pandits ; and even had they been known, probably they would not know to, ] have been deemed worthy of notice. No term belonging to the Dravidian languages themselves has ever idion been used to designate all the members of this family, nor are the aoqus acqu native Tamil or Telugu grammarians, though deeply skilled in the grammar of their own tongues, sufficiently acquainted with comparative has ¢ grammar to have arrived at the conclusion that all these idioms have a langu common origin and require to be designated by a common term. Some that ¢ an European scholars who have confined their attention to the study of some one Drividian idiom to the neglect of the others, have fallen into being | the same error of supposing these languages independent one of ever, Th another. The Sanscrit Pandits had a clearer perception of grammatical affinities and differences than the Dravidian grammarians ; and, though betipeg their generalisation was not perfectly correct, it has furnished us with those | the 8y the only common terms which we possess for denoting the northern and southern families of languages respectively. languag from 4 Tae DrRAVIDIAN LANGUAGES INDEPENDENT OF THE SANSCRIT. Sangey Somg | It was supposed by the Sanscrit Pandits (by whom everything Which with which they were acquainted was referred to a Brahmanical origin), 0oy and too hastily taken for granted by the earlier European scholars, that family) the Dravidian languages, though differing in many particulars from 00ty an the North-Indian idioms, were equally with them derived from the Precede, Sanscrit. They could not but see that each of the Dréividian from t languages to which their attention had been drawn, contained a 10 anot) certain proportion of Sanscrit words, some of which were quite | thvig unchanged, though some were so much altered as to be recognized with a\’ing ( difficulty ; and though they observed clearly enough that each language {idging contained also many Un-Sanserit words and forms, they did not kg observe that those words and forms constituted the bulk of the Uy, language, or that it was in them that the living spirit of the language resided. Consequently they contented themselves with ascribing the liy Un-Sanscrit portion of these languages to an admixture of a foreign element of unknown origin. According to this view there was no port; on essential difference between ¢the Drividas’ and ‘the Gaudas ; for DRAVIDIAN TONGUES INDEPENDENT OF SANSCRIT. 29 the the Bengali and the other languages of the Gaura group contain also a 08 small proportion of Un-Sanscrit words and forms, whilst in the main ded they are corruptions of the Sanscrit. This representation fell far short ably of the real state of the case, and the supposition of the derivation of s — the Dravidian languages from the Sanscrit, though entertained in the the past generation by a Colebrooke, a Carey, and a Wilkins, is now 10t known to be entirely destitute of foundation. The orientalists referred to, though deeply learned in Sanscrit and well acquainted with the ever idioms of Northern India, were unacquainted, or but very slightly o the acquainted, with the Drividian languages. No person who has any | the acquaintance with the principles of comparative philology and who atie has carefully studied the grammars and vocabularies of the Dravidian aved languages, and compared them with those of the Sanscrit, can suppose Some that the grammatical structure and inflexional forms of those languages and the greater number of their more important roots are capable of 0y of 1 0t being derived from the Sanscrit by any process of corruption whatso- ever. e of atical The hypothesis of the existence of a remote original affinity hough between the Dravidian languages and the Sanscrit, or rather between s with those languages and the Indo-European family of tongues, inclusive of rthern the Sanscrit, of such a nature as to allow us to give the Dravidian languages a place in the Indo-European group, is altogether different from the notion of the direct derivation of those languages from the RIT Sanserit. The hypothesis of a remote original aflinity is favoured by some interesting analogies both in the grammarand in the vocabulary, rything which will be noticed in their place. Some of those analogies are best accounted for by the supposition of the retention by the Dravidian ')l'l’giﬂ)’ 5, that family, as by the Finnish and the Turkish, of a certain number of g from roots and forms belonging to the Pre-Sanscrit period, the period which preceded the final separation of the Indo-European group of tongues on ‘(he from the Scythian. I think I shall also be able to prove, with respect to another portion of the analogies referred to, that instead of the ined & Dravidian languages having borrowed them from the Sanserit, or both 0 qﬂite having derived them from a common source, the Sanscrit has not ol with disdained to borrow them from its Dravidian neighbours. Whatever guag? i not probabilities may be in favour of the hypothesis now mentioned, the of the older supposition of the direct derivation of the Dravidian languages from the Sanscrit, in the same manner as the Hindi, the Bengali, and I&nguage ping 12 erroneous. the other Gaura dialects are directly derived from it, was certainly (1.) It overlooked the circumstance that the Un-Sanscrit , oregh portion of the Drividian languages was nearly as much in excess of s 10 the Sanscrit, as in the North-Indian idioms the Sanserit was in excess for (38 ; it s ki i 2 J 30 INTRODUCTION. 10 of the barbarian or Un-Sansecrit element. (2.) It overlooked the still more material circumstance that the pronouns and numerals of the the er Dravidian languages, their verbal and nominal inflexions, and the Sl syntactic arrangement of their words—everything, in short, which fing constitutes the living spirit of a language—were originally and Tel radically different from the Sanserit. (8.) The Orientalists who held the opinion of the derivation of the Drividian languages from the 1o the Sanscrit relied mainly on the circumstance that all dictionaries of Dravidian langnages contained a large number of Sanscrit words l'leh scarcely at all altered, and a still larger number which, though much —— ind it altered, were unquestionably Sanscrit derivatives. They were not, however, aware that such words are never regarded by native scholars if1 as of Drividian origin, but are known and acknowledged to be derived (Qb from the Sansecrit, and that they are arranged in classes, according to the degree in which they have been corrupted, or with reference to has from the medium through which they have been derived. They were also unaware that true Dravidian words, which form the great majority of sedu the words in the southern vocabularies, are placed by native gram- deny marians in a different class from the above-mentioned derivatives from Soun the Sanscrit, and honoured with the epithets ¢national words’ and perv: that. ¢ pure words.’ The Telugu grammarians specify even the time when Sanserit derivatives were first introduced into Telugu; by which we good are doubtless to understand the time when the Brahmans established to th themselves in the Telugu country. They say,— The adherents of S0mg King Andhra-riya, who then resided on the banks of the Godavery, The coun spoke Sanscrit derivatives, many of which words in course of time became corrupted. That other class of words consisting of nouns, diale verbals, and verbs, which were created by the god Brahma before the in th time of this king, are called ‘pure (Telugu) words.’ The date of the condit reign of this King Andhra-raya, or King of the Andhras (a division of Positi the ancient Telugus), who is now worshipped at Chieacole as a deity, lt’arne‘ is unknown, but was probably several centuries anterior to the ad Chrigtian era, Ihose 3 In general no difficulty is felt in distinguishing Sanscrit derivatives Ugeth from the ancient Dravidian roots. There are a few cases only in U ( § which it may be doubtful whether particular words are Sanserit or gy Dravidian, e. g., ‘nir, water, and ‘min,” fish, are claimed as com- Pr Ollzq ponent parts of both languages; though I believe that both are of Sleay Dravidian origin. (4.) The Orientalists who supposed the Dravidian languages to be Whl'ck derived from the Sanscrit were not aware of the existence of unculti- (P 1q y vated languages of the Dravidian family, in which Sanserit words are the Pas o DRAVIDIAN TONGUES INDEPENDENT OF SANSCRIT. 31 still not at all, or but Very rarely, employed; and they were also not aware the that some of the Dravidian languages which make use of Sanserit | the derivatives, are able to dispense with those derivatives altogether, rhich such derivatives being considered rather as luxuries or articles of and finery than as necessaries. It is true it would now be difficult for the held Telugu to dispense with its Sanserit: more so for the Canarese; and | the most of all for the Malayilam :—those languages having borrowed from ey of the Sansecrit so largely, and being so habituated to look up to it for yords help, that it would be scarcely possible for them now to assert their much independence. The Tamil, however, the most highly cultivated «b ) 110, intra of all Dravidian idioms, can dispense with its Sanserit altogether, olars if need be, and not only stand alone but flourish without its aid. rived The ancient or classical dialect of the Tamil language, called the ng to ‘Shen-Tamil’ or correct Tamil, in which nearly all the literature ce to has been written, contains exceedingly little Sanscrit; and differs y also from the colloquial dialect, or the language of prose, chiefly in the ity of sedulous and jealous care with which it has rejected the use of Sanserit g[’ftlll- derivatives and characters, and restricted itself to pure Dravidian s from sounds, forms, and roots. So completely has this jealousy of Sanscrit ¢ and pervaded the minds of the educated classes amongst the Tamilians, when that a Tamil composition is regarded as refined, in accordance with h we good taste, and worthy of being called classical, not in proportion to the amount of Sanscrit which it contains, as would be the case in lished nts Of some other dialects, but in proportion to its freedom from Sanscrit ! averys The speech of the very lowest classes of the people in the retired f time country districts accords to a considerable extent with the eclassical dialect in dispensing with Sanserit derivatives. 10UDS, In every country it is mn the poetry and in the speech of the peasantry that the ancient e T,Ile ()f the condition of the language is best studied. It is in Tamil prose com- positions and in the ordinary speech of the Brahmans and the more qon of , deity, learned Tamilians that the largest infusion of Sanscrit is contained - {0 the and the words that have been borrowed from the Sanserit are chiefly those which express abstract ideas of philosophy, science, and religion, tives together with the technical terms of the more elegant arts. Even in Prose compositions on religious subjects, in which a larger amount of et OF Sanserit is employed than in any other department of literature, the proportion of Sanserit which has found its way into Tamil is not § CoM- are of greater than the amount of Latin contained in corresponding composi- tions in English. Let us, for example, compare the amount of Sanscrit which is contained in the Tamil translation of the Ten Commandments S to h'e (Prayer Book version) with the amount of Latin which is contained in oiltl the English version of the same formula, and’ which has found its way s 878 3 - 2 INTRODUCTION. borr into it, either directly, from Ecclesiastical Latin, or indirectly, through then the medium of the Norman-French. Of forty-three nouns and adjec- avk tives in the English version twenty-nine are Anglo-Saxon, fourteen Latin : of fifty-three nouns and adjectives in the Tamil (the difference han any in idiom causes this difference in the number) thirty-two are Dravidian, mak twenty-one Sanscrit. Of twenty verbs in the English, thirteen are the Anglo-Saxon, seven” Latin : of thirty-four verbs in the Tamil, twenty- Of the five numerals of e seven are Dravidian, and only seven Sanserit. dimi which are found in the English, either in their cardinal or their Tefir ordinal shape, all are Anglo-Saxon: of the six numerals found in the Tamil, five are Dravidian, one (¢ thousand’) is probably Sanserit. 0f the Putting all these numbers together, for the purpose of ascertaining and the percentage, I find that in the department of nouns, numerals and Wor verbs, the amount of the foreign element is in both instances the same, In both instances, also, to Ui viz., as nearly as possible forty-five per cent. info all the pronouns, prepositions, adverbs, and conjunctions, and all the religi inflexional forms and connecting particles are the property of the native tongue. of th of x¢ Trench's expressions respecting the character of the contributions which our mother-English has received from Anglo-Saxon and from Sans whic] Latin respectively, are exactly applicable to the relation and propor- tion which the native Dravidian element bears to the Sanscrit con- Wond tained in the Tamil. Doeti oy « All its joints, its whole articulation, its sinews and its ligaments, the great body of articles, pronouns, conjunctions, prepositions, nume- Sang rals, auxiliary verbs, all smaller words which serve to knit together, posi and bind the larger into sentences, these, not to speak of the gramma- I tical structure of the language, are exclusively Anglo-Saxon (Drévi- (!Omp dian). The Latin (Sanscrit) may contribute its tale of bricks, yea of is eml goodly and polished hewn stones, to the spiritual building, but the Aoqui of th( mortar, with all that holds and binds these together, and constitutes them into a house is Anglo-Saxon (Dréavidian) throughout.” Braby Though the proportion of Sanscrit which we find to be contained Togt ( in the Tamil version of the Ten Commandments happens to correspond “flloh so exactly to the proportion of Latin which is contained in the English Sareq version, it would be an error to conclude that the Tamil language is lme“ as deeply indebted to the Sanscrit as the English is to the Latin. the } The Tamil can readily dispense with the greater part or the whole bfanr:’ of its Sanscrit, and by dispensing with it rises to a purer and more By} refined style ; whereas the English cannot abandon its Latin without g Hag\ abandoning perspicuity. Such is the poverty of the Anglo-Saxon that Scg it has no synonymes of its own for many of the words which it has litera —— — DRAVIDIAN TONGUES INDEPENDENT OF THE SANSCRIT. 33 borrowed from the Latin ; so that if it were obliged to dispense with ugh [ec- them, it would, in most cases, be under the necessity of using a very teen awkward periphrasis instead of a single word. The Tamil, on the other enCe hand, is peculiarly rich in synonymes ; and generally it is not through ian, any real necessity, but from choice and the fashion of the age, that it are makes use of Sanscrit. If the Ten Commandments were expressed in nty the speech of the lower classes of the Tamil people, or in the language of every-day life, the proportion of Sanscrit would be very greatly rals helr diminished ; and if we wished to raise the style of the translation to a d refined and classical pitch, Sanscrit would almost entirely disappear. Of the entire nnmber of words which are contained in this formula it ning there is only one which could not be expressed with faultless propriety and poetic elegance in equivalents of pure Dravidian origin: that and word is ¢ graven image’ or ‘idol’! Both word and thing are foreign ale, also, to primitive Tamil usages and habits of thought; and were introduced into the Tamil country by the Brahmans, with the Purdnic system of | the f the religion and the worship of idols. Through the predominant influence of the religion of the Brahmans, the majority of the words expressive of religions ideas which are in actual use in modern Tamil are of Hons from Sanscrit origin ; and though there are equivalent Dravidian words which are equally appropriate, and in some instances more so, such con words have gradually become obsolete, and are now confined to the poetical dialeet ; so that the use of them in prose compositions would sound affected and pedantiec. This is the real and only reason why nents, Sanscrit derivatives are so generally used in Tamil religious com- nume- positions. . efhe, In the other Dravidian languages, whatever be the nature of the mmé- composition or subject-matter treated of, the amount of Sanscrit which )révi- is employed is considerably larger than in Tamil ; and the use of it has (28 of it the acquired more of the character of a necessity. This is in consequence of the literature of those languages having chiefly been cultivated by Htutes Brahmans. Even in Telugu the principal grammatical writers and the most celebrated poets have been Brahmans. There is only one work tained of note in that language which was not composed by a member of the aspoﬂd sacred caste ; and indeed the Telugu Stdras, who constitute par excel- }ngﬁsh lence, the‘Telugu people, seem almost entirely to have abandoned to the Brahmans the culture of their own language, with every other L branch of literature and science. In Tamil, on the contrary, few yhole The lan- Brahmans have written anything worthy of preservation. | more guage has been cultivated and developed with immense zeal and ﬂthvﬂ'& success by native Tamilian Stdras ; and the highest rank in Tamil on th ab literature which has been reached by a Brahman is that of a commen- it o8 D ko it oaicsi T 24 INTRODUCTION. tator. The commentary of Parimélaragar on the Kural of Tiru- valluvar (a Pariar! but the acknowledged and deified prince of Tamil authors) is the most classical production which has been written in Tamil by a Brahman. Professor Wilson observes that the spoken languages of the South were cultivated in imitation and rivalry of the Sanscrit, and but par- tially aspired to an independent literature ; that the principal compo- l sitions in Tamil, Telugu, Canarese, and Malayalam are translations or a8 paraphrases from Sanscrit works ; and that they largely borrow the of phraseology of their originals. This representation is not perfectly & correct, in so far as the Tamil is concerned ; for the compositions that l are universally admitted to be the ablest and finest in the language, P viz., the Cural and the Chintamani, are perfectly independent of the 4] Sanserit, and original in design as well as in execution ; and though it is true that Tamil writers have imitated—I cannot say translated — to the RamAayana, the Maha-bhérata, and similar works, they boast that of the Tamil Ramayana of their own Kamban is greatly superior to the daf Sanserit original of Valmiki. fal (5.) Of all evidences of identity or diversity of languages the i most conclusive are those which are furnished by a comparison of their grammatical structure ; and by such a comparison the independence of fua the Dravidian languages of the Sanscrit will satisfactorily and conelu- lse sively be established. By the same comparison (at the risk of antici- sep pating a question which will be discussed more fully in the body of adg the work), the propriety of placing these languages in the Scythian or group, rather than in the Indo-European, will be indicated. The most prominent and essential differences in point of grammati- dee cal structure between the Dravidian languages and the Sanscrit, are as follows : — ﬂlej Iﬁn(r (i.) In the Dravidian languages all nouns denoting inanimate The dis- Wi substances and irrational beings are of the neuter gender. tinction of male and female gppears only in the pronouns of the third Ot to g person ; in the adjectives (properly appellative nouns) which denote rational beings, and are formed by suffixing the pronominal terminations; ﬂ”yJ f and in the third person of the verb, which, being formed by suffixing the same pronominal terminations, has three forms in the singular and qaah two in the plural, to distinguish the several genders, and in accordance In all other cases where it is lianl with the pronouns of the third person. dig required to mark the distinction of gender, separate words signifying Wh( ‘male’ and ‘female’ are prefixed ; but, even in such cases, though the object denoted be the male or female of an animal, the noun which Ofv cal] denotes it does not cease to be considered nenter, and neuter forms of Qu] DRAVIDIAN TONGUES INDEPENDENT OF THE SANSCRIT. 35 I the pronoun and verb are required to be conjoined with it. This rule il presents a marked contrast to the rules respecting gender which we 10 find in the vivid and highly imaginative Sanscrit, and in the other Indo-European languages, but it accords with the usage of all the ith languages of the Scythian group. al (ii.) Dravidian nouns are inflected, not by means of case-termina- PO tions, but by means of suffixed post-positions and separable particles, 3 O as in the Scythian tongues. The only difference between the declension the of the plural and that of the singular, is that the inflexional signs are otly annexed in the singular to the base, in the plural to the sign of that plurality, exactly as in the Scythian languages. After the pluralising e particle has been added to the base, all nouns, irrespective of number the and gender, are declined in the same manner as in the singular. ugh (iii.) The Dravidian dative ‘ku,” ki, or ‘ge,” bears no analogy ] — to any dative case-termination which is found in the Sanscrit or in any that of the Indo-European languages ; but it perfectly corresponds to the y the dative of the Oriental Turkish, to that of the language of the Scythian tablets of Behistun, and to that of several of the languages of the s the Finnish family. their (iv.) Wherever prepositions are used in the Indo-European lan- y0e of guages, the Dravidian languages, with those of the Scythian group, neli- use post-positions instead,—which post-positions do not constitute a el separate part of speech, but are real nouns of relation or quality, dy of adopted as auxiliaries. All adverbs are either nouns or the gerunds or infinitives of verbs. ythliaﬂ (v.) In Sanscrit and all the Indo-European tongues, adjectives are matl- declined like substantives, and agree with the substantives to which they are conjoined in gender, number, and case. In the Dravidian t, are languages, as in the Scythian, adjectives are incapable of declension. When used separately as abstract nouns of quality, which is the nmate e dis- original and natural character of Dravidian adjectives, they are subject to all the affections of substantives ; but when they are used adjectiv- ) third Jenote ally, i.c., to qualify other substantives, they do not admit of any 110055 inflexional change, but are simply prefixed to ‘the nouns which they qualify. fixing al and (vi.) Tt is also a characteristic of these languages, as of the Mongo- lian, the Manchu, and several other Scythian languages, in contra- rdanct ro it 1 distinetion to the languages of the Indo- European family, that, wherever it is practicable, they use as adjectives the relative participles lmfymg ghovgl of verbs, in preference to nouns of quality, or adjectives properly so ) Which called ; and that in consequence of this tendency, when nouns of quality are used, the formative termination of the relative participle orms 0 ' D 2 A A A A 36 INTRODUCTION. is generally qu{ﬁ\ed to them, through which suffix they partake of the character both of nouns and of verbs. (vii.) The existence of two pronouns of the first person plural, one of which includes, the other excludes the party addressed, is a peculi- arity of the Dravidian dialects, as of many of the Scythian languages; but is unknown to the Sanscrit and the languages of the Indo- European family (viii.) The situation of the governing word is characteristic of each of these families of languages In Sanscrit and the Indo-KEuropean family it usually precedes the word governed: in the Drividian and in all the Scythian languages, it is invariably placed after it; in con- sequence of which the principal verb always occupies the last place in the adverb the sentence The adjective precedes the substantive the substantive which is governed by a verb precedes the verb together with every word that depends upon it or qualifies it, precedes the verb by which it is governed : the relative participle precedes the 1 noun on which it depends: the negative branch of a sentence precedes the affirmative: the noun in the genitive case precedes that which governs it: the pre-position changes places with the noun and becomes I‘ a post-position in virtue of its governing a case and finally the sentence is concluded by the one, all-governing, finite verb In each of these important and highly characteristic peculiarities of syntax the fl'\ Dravidian languages and the Scythian are thoroughly agreed fq (ix.) The Dravidian languages like the Scythian but unlike the Indo-European, prefer the use of continuative participles to conjunc- t tions (x.) The existence of a negative as well as an affirmatjve voice I in the verbal system of these languages, constitutes another essential I point of difference between them and the languages of the Indo- European family: it equally constitutes a point of agreement between Im by them and the Scythian tongues. Oy, (xi.) It is a marked peculiarity of these languages as of the I8 Mongolian and the Manchu, and in a modified degree of many other it Scythian languages, that they make use of relative participles instead of of relative pronouns There is not a trace of the existence of a relative pronoun in any Dravidian language The place of such pronouns is supplied, as in the Secythian tongues mentioned above, by relative ly i participles, which are formed from the present, preterite, and future 1 participles of the verb by the addition of a formative suffix which B suffix is in general identical with the sign of the possessive case literally, the of Thus, the person. who came, is in Tamil vand-a &l who-came person vand’ the preterite werbal participle signifying La the thal A e RO SR e UN-SANSCRIT ELEMENT IN NORTHERN VERNACULARS. a7 he having come, being converted into a relative participle, equivalent to the-who-came, by the addition of the old possessive and adjectival ne suffix ‘a.’ li Many other differences in grammatical structure will be pointed 5, out hereafter, in the course of the grammatical analysis: but in the do important particulars which are mentioned above, the Dravidian languages evidently differ so considerably from the languages of the ach Indo-European family, and in particular from the Sanscrit (notwith- ean standing the predominance for so many ages of the social and religious and influence of the Sanscrit-speaking race), that it cannot be doubted that (0N they belong to a totally different family of tongues. They are neither ¢ il derived from the Sanserit, nor are capable of being affiliated with it: et and it cannot have escaped the notice of the student of comparative eth, philology, that in every one of those particulars in which the gram- edes matical structure of the Dravidian languages differs from the Sanscrit, the it agrees with the structure of the Scythian languages, or the languages odes of Central and Northern Asia. hich omes Is THE Unx-SANSCRIT ELEMENT CONTAINED IN THE VERNACULAR the LANGUAGES OF NORTHERN INDIA DRAvIDIAN ? each The hypothesis of the direct derivation of the Dravidian tongues ¢ the from the Sanscrit, with the admixture of a proportion of words and forms from an unknown source, being now no longer entertained, some o the oriental scholars have adopted an opposite hypothesis, and attributed ajunc- to the influence of the Drividian languages that corruption of the Sanscrit out of which the vernaculars of Northern India have arisen. yolce It has been supposed by the Rev. Dr. Stevenson, of Bombay, eutiﬂl Mr. Hodgson, of Nipaul, and some other orientalists, (1) that the North- Indo- Indian vernaculars have been derived from the Sanscrit, not so much tweell by the natural process of corruption and disintegration, as through the over-mastering, re-moulding power of the Un-Sanscrit element which of the is contained in them ; and (2) that this Un-Sansecrit element is identical / other with the Drividian speech, which they suppose to have been the speech insteﬁd of the ancient NishAdas, and other aborigines of India. »e]ative The first part of this hypothesis appears to rest upon a better foun- ouns 18 dation than the second: but even the first part appears to me to be too elafi®® strongly expressed, and to require considerable modification; for in some future important particulars the corruption of the Sanscrit into the Hindi, the ; whiCh Bengali, &c. has been shown to have arisen from that natural process of change which we see exemplified in Europe, in the corruption of the p 0050 111)"; e Latin into the Italian’ and the French. Nevertheless, on comparing ;nifyiﬂg the grammatical structure and essential character of the Sanscrit, with that of the vernaculars of Northern India, I feel persuaded that those | 38 INTRODUCTION vernaculars bave to a considerable extent been corrupted in a Scythian direction, and through the operation of Scythian influences f f The modifications which the grammar of the North-Indian languages have received, being generally of one and the same character, and in \ J one and the same direction, it is obvious that there must have been a common modifying cause ; and as the barbarian or Un-Sanscrit portion of those languages, which Professor Wilson styles ‘a portion of a id m primitive, unpolished, and scanty speech, the relics of a period prior to civilization,” is generally calculated to amount to one-tenth of the 01 whole, and in Marathi, to a fifth, it seems reasonable to infer that it was from that extraneous element that the modifying influences pro- el ceeded la It is admitted that before the arrival of the Aryans, or Sanscrit- al speaking colony of Brahmans, Kshatriyas, nd Vaisyas, the greater by part of Northern India was peopled by rude aboriginal tribes, called g by Sanscrit writers, Mléchchas, Dasyus, Nishadas, &e. ; and 1t is the 01 Bra received opmlon that those aborlglnal tribes were of Scythlan, or at least of Non-Aryan origin. On the irruption of the Aryans it would Ra naturally happen that the copious and expressive Sanscrit of the con- quering race would almost overwhelm the vocabulary of the rude eve Never- Scythian tongue which was spoken by the aboriginal tribes poi theless, as the grammatical structure of the Scythlan tongues possesses 1l 0 peculiar stability and persistency ; and as the Pre- -Aryan tribes, who of t were probably more numerous than the Aryans, were not annihilated, ph but only reduced to a dependent position, and eventually, in most by instances, incorporated in the Aryan community, the large Sanscrit ) addition which the Seythian vernaculars received, would not necessarily an alter their essential structure, or deprive them of the power of influencing U, and assimilating the speech of the conquering race According to this Lt th theory, the grammatical structure of the spoken idioms of Northern amo) India was from the first, and always continued to be, in the main, lay 0l Scythian ; and the change which took place when Sanscrit acquired Wty the predominance, as the Aryans gradually extended their conquests \(sytl and their colonies, was rather a change of vocabulary than of grammar, heS —a change not so much in arrangement and vital spirit as in the Ty matériel of the language Sy This hypothesis seems to have the merit of according better than liegy any other with existing phenomena Seeing that the northern verna- | culars possess, with the words of the Sanscrit, a grammatical structure Ing; which in the main appears to be Scythian, it seems more correct to fol] represent those languages as having a Scythian basis, with a large and Paxg; almost overwhelming Sanserit addition, than as having a Sanscrit basis, o with a small admixture of a Scythian element ————ene ——— — UN-SANSCRIT ELEMENT IN NORTHERN VERNACULARS 39 Wl Whichever proposition be adopted, there is not much room for dif- ference of opinion respecting the fucts that are involved in the dispute; {4 the existence of a Scythian element in the colloquial dialects of n Northern India having been pointed out many years ago by Sir W | & Jones, and never since called in question. 00 The second part of the hypothesis of Dr. Stevenson, viz., the ) identity of the Un-Sanscrit or Scythian element which is contained to in those languages with the languages of the Dravidian family, rests the on a different foundation, and appears to me to be less defensible 16 According to the supposition in question, the Scythian or Dravidian IO element is substantially one and the same in all the vernacular languages of India, whether northern or southern, but is smallest in ot amount in those districts of Northern India which were first conquered ater by the Aryans; greater in the remoter districts of the Dekhan, Telin lled gana, and Mysore ; and greatest of all in the Tamil country, at the the southern extremity of the peninsula, to which the aggressions of the Brahmanical race had not extended in the age of Manu and the or ab ould Réméayana con This hypothesis is certainly in accordance with the current of rude events in the ancient history of India : but whatever relationship, in Vel point of blood and race, may originally have subsisted between the northern aborigines and the southern—whatever ethnological evidences 08568 who of their identity may be supposed to exist,—when we view the question phalologically, and with veference to the evidence which is furnishe ated. by their languages alone, the hypothesis of their identity does not mOat psenb appear to me to have been established It may be true that various analogies in point of grammatical structure appear to connect the Un-Sanserit element which is contained in the North-Indian idioms neIng with the Scythian or Tartar tongues This connection, however, o this amounts only to a general relationship to the entire group of Scythian hert mai languages ; and no special relationship to the Dravidian languages, in contra-distinction to those of the Turkish, the Finnish, or any other qmred Indeed I conceive that Scythian family, has yet been proved to exist quea the Scythian substratum of the North-Indian idioms presents a greater mmaly number of points of agreement with the Oriental Turkish, or with that m the Scythian tongue or family of tongues by which the New Persian has been modified, than with any of the Dravidian languages T thﬁn The principal particulars in which the grammar of the North- pInd Indian idioms accords with that of the Dravidian languages are as pctie follows :—(1), the inflexion of nouns by means of separate post-fixed ect 1 particles; (2), the inflexion of the plural by annexing to the unvary- 190 ol ing sign of plurality the same suffixes of case as those by which the it b”" K 40 INTRODUCTION. —— singular is inflected ; (3), the use of a dative or dative-accusative in th ha ‘ko’ dr ‘ku :’ (4), the use in several of the northern idioms of two pro- nouns of the first person plural, the one including, the other excluding an the party addressed ; (5), the use of post-positions, instead of preposi- nf tions ; (6), the formation of verbal tenses by means of participles ; pr (7), the situation of the governing word after the word governed. In g the particulars above-mentioned the grammar of the North-Indian 01 idioms nndoubtedly resembles that of the Drividian family : but the argument founded upon this general agreement is to a considerable extent neutralised by the circumstance that those idioms accord in the very same particulars, and to the very same extent, with the Turkish and several other families of the Scythian group. Not one of those particulars in which the Dravidian languages differ from the Turkish or the Mongolian (and there are many such points of diflerence) has as yet been discovered in the North-Indian idioms. For instance, those idioms contain no trace of the relative participle which is used in all the Dravidian tongues instead of a relative pronoun ; they are destitute of the regularly inflected negative verb of the Drividian languages ; and they contain not one of the Drividian pronouns or numerals—not even those which we find in the Scythic tablets of Behistun, and which still survive even in the languages of the Ostiaks and Lapps. If the Un-Sanscrit element contained in the northern vernaculars had been Dravidianwe might also expect to find in their vocabularies a few primary Dréividian roots—such as the words for ‘head,” ‘hand,” foot, ‘eye, ‘ear, &c.; but I have not been able to discover any reliable analogy in words belonging to this class. The only resemblances which have been pointed out are those which Dr. Stevenson has traced in a few words remote from ordinary use, and on which, in the absence of analogy in primary roots, and espe- cially in grammatical structure, it is impossible to place any depen- dence.* The difference between the Dravidian vocabulary and that of the languages of Northern India with respect to primary roots together with the essential agreement of all the Dravidian vocabularies one with another, will appear from the following comparative view of ) * In many instances Dr. Stevenson’s lexical analogies are illusory, and dis- appear altogether on a little investigation. Thus, he supposes the North Indian Plog ‘pet,’ the belly, the womb, to be allied to the first word in the Tamil compound ‘petta pillei, own child. by That word should have been written ‘pettra’ in English, to accord with the pronunciation of the Tamil word : the Tamil spelling Wepy of it, however, is ‘perra.’ It is the preterite relative participle of ‘ per-u, to Iy bear, to obtain, signifying that was borne. ¢ Per-u,” to obtain, has mno con- nexion with any word which signifies #he womb, and its derivative noun e ‘ pér-u,” means a thing obtained, @ birth, a favour. The affinities of this root Doy will be inquired into in the Comparative Vocabulary. 4ppe; UN-SANSCRIT ELEMENT IN NORTHERN VERNACULARS. 41 em It sometimes the pronouns of the first and second persons singular. happens that where one form of the pronoun is used in the nominative, pro- ling another survives in the oblique cases, and a third in the verbal posi- inflexions : it also sometimes happens that the ancient form of the Where such is the case I have les; pronoun differs from the modern. In given all extant forms a place in the list, for the purpose of facilitating dian comparison. the Pronoun of the first person singular :— able 1 the Norra-INDIAN IDIOMS. DrAvIDIAN IDIOMS, rkish (Sanscrit primary form ‘aham ; ’ Tamil, nin, yan, én, en secondary forms, ‘ma, HAN mis’ S Canarese, dn, ninu, en, éne those Turkish primary form, ‘man.’) Tuly, yéan, en, e rkish Hindi, main Malayalam, fijin, én, en, in Bengali, mii ) has Telugu, nénu, nf ance, Marathi, mi Tuda, On, 4n, en, ini Gujarathi, hun Kota, ine, en, e used Sindhi, man Gond dna, 4n y are Ku dnu, nd, énu, e :'idi:m en Rajamahal, n§ or Uraon, enan ts of Pronoun of the second person singular :— tiaks thern Norre-INDIAN IDIOMS. DrAvipiaN Iptoms, thelr (Sanserit primary forms tvam,’ Tamil, ni, nin, nei, i, ‘tav,” ‘te; secondary form, si, Js for ’ Canarese, nin, ninu, 1, i ‘8 Turkish primary form, ¢gen.’) Tulu, i, ni, nin ) able Hindi, tun, tu, te Malayalam, nf, nin, nan class, Bengali, tiii, to Telugu, nivu, ni, nin which Marathi, tln, tu, to Tuda, nf, nin, i y 186, Gujarathi, tln, ta Kota, nf, nin, i Sindhi, tun, to Gond, ima, nf, i esper Ku inu, ni, i Jepen- Uraon nien 1 that Rajamahal nin A 100t Brahui ni Scythic of the Behistun tablets, nt u}fm'es jow of From the striking dissimilarity existing between the North-Indian nd dis pronouns and the Dravidian it is obvious that, whatever may have ] pdian been the nature and origin of the Scythic influences by which they p()llﬂ were modified, those influences do not appear to have been Dravidian. o 1 bpellmg In the pronouns of almost all the North-Indian languages, the Scythian Bl 0 20 08 termination—the obscure ‘%’ which forms the final of most of the pro- nouns—is at once observed : we cannot fail also to notice the entire dis- 0 o bis * appearance of the nominative of the Sanserit pronoun of the first person 42 INTRODUCTION men or man norf singular, and the substitution for it of the Turkish but in no connexion, in no number or case, in no compound or verbal and the inflexion do we see the least trace of the peculiar personal pronouns of nafl the Dravidian family. Possibly, after all, further research may disclose the existence in the northern vernaculars of distinctively Dréavidian lang forms and roots but their existence does not appear to me as yet to g be proved ; for most of Dr. Stevenson’s analogies take too wide a range and and where they are supposed to be distinctively Dravidian, they inva- tical I conclude, therefore, that the We riably disappear on examination Un-RSanserit portion of the northern languages cannot safely be placed prol luc in the same category with the southern, except perhaps in the sense of both being Scythian rather than Indo-European thos the Wite WHAT GRouP oF LANGUAGES ARE THE DRAVIDIAN IpToms TO BE AFFILIATED ? the unde Leaving the idioms of Northern India out of consideration for the hoge present, as extraneous to the object of this work, and restricting our I attention to the Dravidian languages, and the question of their affilia- addy tion, the supposition of their Scythian relationship appears to me to Was be that which is most fully borne out by grammatical analysis and the The comparison of vocabularies. bioey In using the word ‘Scythian,” I use it in the wide general sense and in which it was used by Professor Rask, who first employed it to ) designate that group of tongues which comprises the Finnish, the hab All these lan- Turkish, the Mongolian, and the Tungusian families ]an( guages are formed on one and the same grammatical system, and in fable Their formation of cases accordance with the same general laws Valug Ju moods, and tenses, by the simple agglutination of successive, unchange- o be able suffixes, determine them to be a distinct class of languages— Gmpy a class distinet from the Semitic, which inflects dissyllabic roots by the Ufthe variations of internal vowels, and also from the Indo-European idioms, the J hich make so extensive a use of technical case-signs and other in- Seyth flexions, of euphonic modification, and of composition. These languages [artie have been termed by some the Tatar or Tartar family of tongues, by famﬂy others the Finnish, the Ural-Altaic, the Mongolian, or the Turanian ; benm but as these terms have often been appropriated to designate one or the [ two families, to the exclusion of the rest, they seem to be too narrow follo“ and too liable to misapprehension to be safely employed as common ( designations of the entire group. The term Scythian’ having already dlan ( been used in the Classies in a vague, undefined sense, to denote 1 generally the barbarous tribes of unknown origin that inhabited the Pergy AFFILIATION OF DRAVIDIAN LANGUAGES. 43 dll northern parts of Asia and Europe, it seems to be the most appropriate bal and convenient word which is available. Professor Rask, who was 5 of the first by whom this word was employed as a common generic desig- lose nation, was also the first to suggest that the Dravidian or Tamilian 1an languages were probably Scythian. He has the merit of having f0 suggested this relationship; but the evidence of it was left both by him, 1ge, and by succeeding writers, in a very defective state. In the gramma- [V tical analysis and comparison of the Drividian languages on which the we are about to enter I hope to help forward the solution of a wced problem which has often been stated, and which has been ingeniously e of elucidated up to a certain point, but which has never yet been thoroughly investigated. The various particulars which were recently adduced to prove that the Dravidian family is essentially different from and independent of S the Sanserit (each of which will be more fully considered in the sequel, under its appropriate head) may also be regarded as proving that the those ldnguages are intimately related to the Scythian group. In this introductory part of the work, I shall content myself with : our filla adducing in proof of their Scythian relationship the evidence which was recently furnished by the translation of the Behistun tablets. e to [ the The inscriptions discovered at Behistun record the political auto- biography of Darius Hystaspes in the Old Persian, in the Babylonian, 2058 and also in the language of the Scythians of the Medo-Persian it to cmpire ; and the translation of the Scythian portion of those inseriptions has thrown new light on the propriety of giving the Dravidian j the ) Jan- languages a place in the Scythian group. The language of the Scythic nd iﬂ tablets, at first supposed to be Median, has been shown in Mr. Norris’s (AS63 valuable paper (in the Journal of the Royal Asiatic Society, vol. XV.) to be distinctively Scythian. Consequently we are now enabled to ange- 0eS— compare the Dravidian idioms with a fully developed, copious laﬁguage of the Scythian family, as spoken in the fifth century, B.c.: and whilst )y the Jioms, the language of the tablets has been shown to belong generally to the Scythian group, it has been found to bear a special relationship to a ol in- particular family included in that group—the Ugro-Finnish—that very 1588 family to which the Dravidian dialects have long appeared to me to 5, DY pian ; be most nearly allied. The principal points of resemblance between the Dravidian dialects and the language of the tablets are as follows :— lﬁm‘()“!' (1.) The language of the tablets appears to accord with the Dravi- ol dian tongue in the use of consonants of the cerebral class, ¢t,” ¢d,” and k] Iready N These sounds exist also in the Sanscrit, but I have long been deuote persuaded that the Sanscrit borrowed them from the indigenous Dra- o 00 44 INTRODUCTION. {10 vidian languages (vide the section on “Sounds”); and I find that pr Mr. Norris has expressed the same opinion. fic (2.) The language of the tablets agrees with the Tamil in regard- fie ing the same consonant as a surd in the beginning of a word, and as a sonant in the middle, and in pronouncing the same consonant as a gul Wl sonant when single, and as a surd when doubled. (See in the section on “Sounds” illustrations of the Tamil rule.) Wi (8.) The genitive case of the language of the tablets is formed by the 3 suffixing the syllables ‘na,” ‘ni-na,” or ‘inna.’ The analogous forms of the Dravidian languages are ‘ni,’ in the Telugu, ‘na,’” or ‘a, in 1§ the Gond and Brahui, and “ni,” in the Tamil. There are (4.) The dative of the tablets is ¢ikki’ or ¢ikka.’ analogies to this both in the Tartar-Turkish and in the Ugrian fol families ; but the form which is most perfectly in accordance with it is that of the Dravidian dative suffix ‘ku,” ‘ki,’ ¢ ka,” &ec., preceded, togg as the suffix generally is in Tamil and Malayalam, by an euphonic ‘u’ tom or ‘i, and a consequent doubling of the ‘k.’ Compare ‘ni-ikka,” fo to g thee, in the language of the tablets, with the corresponding Telugu a ki ‘ni-ku,” and the Malayéla ‘nan-i-kka.’ (5.) The pronouns of the language of the tablets form their accu- sative by suffixing ‘un,” “in,” or ‘n.’ Compare the Telugu accusative Wit} inflexion ‘nu’ or ‘mni,” and the Canarese ‘am,” ‘ann-u,’” &e. (6.) The only numeral which is written in letters in the Seythian Ugy Whi tablets is ‘kir,” one, with which appears to be connected the numeral adjective, or indefinite article, ‘ra’ or ¢irra.’ In Telugu, one is ‘ oka,’ ticl ? and in Tamil, ¢or From a comparison of all the shapes which this the numeral has assumed in the various Dravidian dialects and in com- of ¢ pounds, I had long ago come to the conclusion that both the Telugu verh and the Tamil forms were probably derived from a eommon and older fable form, ¢okor’ or ‘kor, which I regarded as identical with the Sam- Prim oyede ‘okur, one. I can now compare it also with the ‘kir’ of the Temg tablets. The Ku numeral adjective one is ‘ra,” corresponding to the £ah) Tamil ¢ oru,” but more closely to the ‘ra’ or ‘irra,” of the tablets. from In the language of the tablets all ordinal numbers end in ¢im,” in 10 the hen@ Tamil in ‘4m,” in Samoyede in ‘im.’ (7.) The pronoun of the second person is exactly the same in the h()rde language of the inscriptions as in the Dravidian languages and the of f Brahui: in all it is ‘ni.’ Unfortunately the plural of this pronoun | is not contained in the tablet,—the singular having been used instead tena of the plural in addressing inferiors. tlog (8.) The language of the tablets, like the Dravidian languages, a4 g makes use of a relative participle. A relative pronoun is used in addi- t()n 0 misna O s AFFILIATION OF DRAVIDIAN LANGUAGES 45 that tion to the relative participle ; but Mr. Norris supposes the use of this pronoun to be owing to the imitation of the Persian original. The par- and- ticular particle which is used in the tablets in forming the relative par- a5 a ticiple differs from that which is generally used in the Drividian lan- a3 & guages ; but the position and force of this particle, and the manner in ction which the participle formed by it is employed, are in perfect harmony with Dravidian usage Perhaps the use of this relative participle is d by the most remarkable and distinctive characteristic of the grammar of every unaltered dialect of the Scythian family. 0rms 3, 11 9.) The negative 1mperat1ve, or prohibitive particle of the tablets 18 “inni, in Gond, ¢ minni.’ (10.) The only verbal roots which appear to be analogous are the ¢ are orlan following : — 1t hit LANGUAGE OF THE TABLETS DRAVIDIAN eded to say, nan an or en 1010 tomake known, uri urei, to explain; aRri, to know [y fo to go pori po el gl a king, ko (the vowel considered kd or kén uncertain) aceu The conjugational system of the language of the tablets accords sative with that of the Magyar, the Mordwin, and other languages of the Ugrian family, but differs considerably from the Dravidian languages, yth\an which form their tenses in a simpler manner, by the addition of par- meral ticles of time to the root, and which form the persons of their verbs by “okay the addition of the ordinary pronominal terminations to the particles ch this of time Notwithstanding this discrepancy in the inflexions of the ] ol gt verbs, the resemblances shown to subsist between the language of the e tablets and the Dravidian idioms, most of which are in particulars of | older primary importance, fully establish the existence of a radical though o Sam remote, connection From the discovery of these analogies, we are of the enabled to conclude that the Dravidian race, though resident in India fo the from a period long prior to the commencement of history, originated i in the central tracts of Asia—the seed plot of nations and that from iy 1 thence, after parting company with the rest of the Ugro-Turanian horde, and leaving a colony in Beluchistan, they entered India by way n the of the Indus nd the Whilst I regard the grammatical structure and prevailing charac- }ronoun teristics of the Dravidian idioms as Scythian, I claim for them a posi- mste"‘d tion in the Scythian group which is independent of its other members as a distinct family or genus, or at least as a distinct sub-genus of 005 i tongues. They belong not to the Turkish family, or to the Ugrian, or n o A s oo B e INTRODUCTION. 46 to the Mongolian, or to the Tungusian (each of which families differs only Gael materially from the others, notwithstanding generic points of resem- blance), but to the group or class in which all these families are com- genel diffe prised. On the whole, the Drividian languages may be regarded as COnD most nearly allied to the Finnish or Ugrian family, with special fam] affinities, as it appears, to the Ostiak ; and this supposition, which I (rael had been led to entertain from the comparison of grammars and voca- bularies alone, derives some confirmation from the fact brought to light (one allieg by the Behistun tablets that the ancient Scythic race, by which the greater part of Central Asia was peopled prior to the irruption of the have coln Medo-Persians, belonged not to the Turkish, or to the Mongolian, but to the Ugrian stock. Taking for granted, at present, the conclusive- the Indo ness of the evidence on which this hypothesis rests, the result at which 1ot we arrive is one of the most remarkable that the study of comparative philology has yet realized. Iangu How remarkable that the closest and most distinet affinities to the each speech of the Dravidians of inter-tropical India should be those that Lithy are discovered in the languages of the Finns and Lapps of Northern gy, Europe, and of the Ostiaks and other Ugrians of Siberia! and, conse- at log quently, that the Pre—Aryan inhabitants of the Dekhan should be mauy beside proved by their language alone, in the silence of history, in the absence of all ordinary probabilities, to be allied to the tribes that appear to Teslify betwe have overspread Europe before the arrival of the Goths and the Pelasgi, and even before the arrival of the Celts ! What a confirmation of the Seem statement that ¢ God hath made of one blood all nations of men, to (‘apa' dwell upon the face of the whole earth !’ to bg In weighing the reasons which may be adduced for affiliating the earl] Drévidian languages with the Scythian group, it should be borne in 1 gy mind that whilst the generic characteristics of the Scythian languages th d] Of Te Jy are very strongly marked and incapable of being mistaken, in a vast variety of minor particulars, and especially in their vocabularies, the §0mp stock languages which are comprised in this family differ from one another more widely than the various idioms of the Indo-European family EHTOP( mutually differ. The Ugrian and the Turkish families can be proved 0f to be cognate almost as certainly as the Gothic and the Sanscrit, or Kl 0] the Zend and the Greek ; yet, apart from the evidence of structure and vital spirit, and looking only at the vocabulary, and the gram- 31011\: | matical materiél, the agreement of any one of the Ugrian dialects with any one of the Turkish is found to be very far inferior even to the By, I agreement of the Sanscrit and the Celtic,—the longest separated and theorv e ———————————— st most widely differing members of the Indo-European family. Thus I’uc]\ whilst in nearly all the Indo-European languages the numerals are not QUPPO: I AFFILIATION OF DRAVIDIAN LANGUAGES. 47 ffers only similar but the same,—(the Sanscrit word for ome and the em: Gaelic word for five are the only real exceptions to the rule of COm general identity),—muot only do the numerals of every Scythian family d 8 differ so widely from those of every other as to present few points of ecial connection, but even the numerals of any two dialects of the same ich I family are found to differ very widely. Whilst the Sanserit and the Y0Ca- Gaelic agree in eight numerals out of ten, and differ in two only light (one and five) ; the Magyar and the Finnish, though as closely h the allied in point of grammatical structure as the Gaelic and the Welsh, »f the have now only the first fonr numerals in common, and perfectly coincide in two numerals only, one and four , but So great indeed is 15Ive- the diversity existing amongst the Scythian tongues, that, whilst the which Indo-European idioms form but one family, the Scythian tongues are rafive not so much a family as a group of families. The Indo-European languages may be regarded as forming but a single genus, of which each language —(Sanscrit, Zend or Persian, Greek, Latin, Gothic, o the o that Lithuanian, Slavonic, Armenian, Celtic)—-forms a species ; whilst the rthern languages of the Scythian group, more prolific in differences, comprise at least five or six authenticated genera, each of which includes as conse- 1ld be many species as are contained in the solitary Indo-European genus ; besides twenty or thirty isolated languages, which have up to this time hsence ear t0 resisted every effort to classify them. This remarkable difference ’Qlﬁfgi! between the Indo-European languages and those of the Scythian stock seems to have arisen partly from the higher mental gifts and higher of the capacity for civilisation, with which the Indo-European tribes appear nen, 10 to have been endowed from the beginning, and still more from the earlier literary culture of their languages, and the better preservation, ing the orme 10 in consequence, of their forms and roots: but, from whatever cause this difference may have arisen, it is obvious that in weighing evidences { of relationship this circumstance must be taken into account ; and that 1 4 \'35[ so minute an agreement of long separated sister dialects of the Scythian res, the stock is not to be expected as in parallel cases amongst the Indo- nother European dialects. 1 fﬁmlh Of late years some inquirers have been inclined to question the ) pr()\"ed relationship of the Drividian languages to the Scythian, either in sertty OF consequence of comparing them with the Tartar or Turkish languages fracture alone, to the exclusion of the more nearly allied Ugrian family, or in p gralr consequence of observing in the DrAvidian languages certain Indo- 3c/[s With European affinities which seemed inconsistent with the Scythian 1 to t»he theory. A friend of mine, who is a good Tamil scholar, was so much ated ol struck with the latter class of analogies that he was led to adopt the - Thoe supposition of the Indo- European relationship of the Dravidian 48 INTRODUCTION. tongues. At the very outset of my own inquiries I observed those and Indo-European analogies myself ; and, rejecting affinities which are in vl unreal and which disappear on investigation—(such as the connection the e of the Tamil numerals ¢ondru or ‘onnu, one; anju,’ five ‘ettu tions, erght ; with ‘un-us,” ‘pancha, and ¢ashta’—a connection which and looks very plausible, but is illusory (see section on ¢ Numerals’),—I I think it capable of satisfactory proof that a small number of the Euro] grammatical forms of the Dravidian languages, and a more considerable I number of their roots, are to be regarded as of cognate origin with Not- ( corresponding forms and roots in the Indo-European languages. withstandine the existence of a few analogies of this character, the ( dem most essential parts of the grammar and vocabulary of the Dravidian idioms are undoubtedly Scythian, and therefore I have no doubt of the propriety of placing those idioms in the Scythian group Though (4 the il the majority of Hebrew roots have been proved to be allied to the d ¢ Sanscrit, yet the Hebrew language does not cease to be regarded as (5 Semitic rather than Indo-European ; so, notwithstanding some (Germ; interesting analogies with the Sanscrit, the Greek, the Gothic, and the modern Persian, which may be discovered on a careful examination of (6 Portior the Dravidian tongues, and which will be pointed out in each of the succeeding sections, the essential characteristics of those tongues are I Th such as to require us to regard them as in the main Scythian In stating that the Dravidian languages contain certain roots and llied Tndo.] forms allied to the Sanscrit, and to the Indo-European languages During the i generally, it is necessary to preclude mlsapprehensmn long period of the residence of the Drévidian and Aryan races in the X Qan\cr same country, the Drividian vocabularies have borrowed largely from { the Sansecrit It is necessary, therefore, to premise that the analogies I ba to which I refer are not founded on the existence in the Dravidian NG tongues of Sanscrit derivatives, but are such as are discoverable in the original structure and primitive vocabulary of those languages. Whilst Iy ‘Il]llh 1 the Dravidian languages have confessedly borrowed much from their more wealthy neighbours, the Sanscrit, in some instances, has not dis- Jith dained to borrow from the Drévidian but in general there is no The ol ]le L) difficulty in distinguishing and eliminating what the one language has borrowed from the other and the statement which I have now made Tafiy | relates not to derivatives, or words which may be supposed to be deri- Sy i vatives, but to radical, deep-seated analogies which cannot be explained e e 1 0§ on any supposition but that of a partial or distant relationship In el Hea[’ext most instances the words and forms in which analogies are discoverable are allied not to the Sanscrit alone, but to the entire Indo-European D 1ay ldl family : in not a few instances analogies are discoverable in the Greek IOPe (] qu AFFILIATION OF THE DRAVIDIAN FAMILY. 49 hose are and Latin, which are not found in the Sanserit; and in many instances tion in which the Sanscrit appears to exhibit the closest analogy, it is not tto, the euphonized, systematised Sanscrit (Samscrita) of written composi- lll.ich tions, but the crude, original Sanscrit, which is discoverable by analysis ~1 and comparison, the ¢ Pre-Sanscrit’ of W. von Humboldt. the I subjoin here a few illustrations of primitive, underived Indo- able Europeanisms, which are discoverable in the Dravidian languages. with I. Analogical grammatical forms. Not (1.) The use of ‘n,’ as in Greek, to prevent Aiatus. . the (2.) The use of ‘d’ or “t as the sign of the neuter singular of demonstrative pronouns. idian bt of (3.) The existence of a neuter plural, as in Latin, in short ¢a.’ ough (4.) The formation of the remote demonstrative from a base in ‘a,’ the proximate from a base in ‘i; as in the New Persian, ‘4n, that, the od a8 and ‘¢in, this. (5.) The formation of most preterites, as in the Persian and the some td the Germanic tongues, by the addition of ‘t’ or “d.’ on of (6.) The formation of some preterites by the reduplication of a portion of the root. f the 5 e II. Analogical vocables. The following are instances of roots which are much more nearly allied to the Greek, the Gothic, or some other western language of the sand Indo-European stock, than to the Sanserit. ;u&ges ‘Ka-y,” to burn: Greek ¢ rai-w ; Sanscrit ‘kam.’ z the in the ‘ki ‘gi,’ or ‘ge’—(Gond and Can.), to do: Old Persic ‘ki; Sanserit “kri. Y from ‘mig-u,’ much ; related words ‘migala’ and ‘mikkili:’ Persian Jogies ‘mih,” English ‘migh-t,” Old High-German ¢ mih-hil,” Norse ‘mikil ;’ ridian Sanserit ¢ mahi. n the ‘mugil, @ cloud : Lithuanian ‘migla, Greek ¢¢-ulx\-y, Gothic . Fhilst ‘milh-ma ;' Sanscrit < mégha.’ 1 fheir ‘pamp-u,’ ¢o send : Greek ¢ méum-w ; no allied word in Sanscrit. lot diS‘ The illustrations which are given above form only a small portion l,iS 10 of the analogous forms and roots which will be adduced in the gram- matical analysis and in the glossarial affinities: they will, however, 7 made suffice to prove that primitive, deep-seated Indo-European analogies 1 d@fi‘ are discoverable in the Dréividian languages. They also serve to Pmined illustrate the statement, that, though the Sanscrit has long been the 0 In nearest neighbour of the Dréividian tongues, there are not a few N erﬁble Dravidian roots which are more nearly allied to the Western Indo- N European idioms than to the Sanseritic or Eastern. Whilst, therefore, I classify the Drayidian family of languages as essentially and in the . ek E ~ INTRODUCTION. J 0 main Scythian, I consider them as of all Scythian tongues those which present the most numerous, ancient, and interesting analogies to the il Indo-European languages. The position which this family occupies, W) Wi if not mid-way between the two groups, is on that side of the Scythian group on which the Indo-European appears to have been severed from ar it, and on which the most distinet traces of the original identity of b the families still remain. If this view be correct (as I think it will (s be shewn to be), the Indo-Europeanisms which are discoverable in the st Dravidian languages carry us back to a period beyond all history, all beyond all mythology, not only prior to the separation of the western th branches of the Indo-European race from the eastern, but prior also to of the separation of the yet undivided Indo-Europeans from the Scythian 11 stoek. B It is a remarkable circumstance, that in the vocabulary of the P Drévidian languages, especially in that of the Tamil, a few Semitic log analogies may also be discovered. In some instances the analogous Tng roots are found in the Indo-European family, as well as in Hebrew, For }’idl though the Hebrew form of the root is more closely analogous. example, though we find in Latin, ‘ave-o,” to desire, and in Sanscrit, II]]p oy ‘ava,” of which to desire is a subordinate meaning; yet the corre- e sponding Tamil words ‘avi,” desire, and ‘aval’ (signifying also desire), ful} a verbal noun from a lost verb ‘4v-u,’ o desire, seem still more directly of ¢ allied to the Hebrew ‘4vah,’ ¢o desire, and the verbal noun ‘avvah,’ desire. tol In addition, however, to such general analogies as pervade the several families of tongues, including the Dravidian, there are roots Dy discoverable both in the Dravidian languages and in the Hebrew, to which I am not aware of the existence of any resemblance in any Ind. langunage of the Indo-European family. The following are illustrations Whj( of such special analogies :— aboy W}]y mMAR-U . to change, or exchange, to sell ; Hebrew ‘mir,” fo Tesey exchange; Syriac ‘mor,” to buy. unqu guvar a wall ; Hebrew, ‘shir,’ a wall. liy | thyy kir a sharp point; Hebrew, ‘kiir,” fo bore, to pierce. * fev-(vei) . equal, level, right; Chaldee, ‘shev-4,” to be equal, level, &c. ; Hebrew, ‘shiv-ah,’ the same. al, 11, 14, 1é no, not; Hebrew, ‘al,” ‘16, not; Chaldee, ‘14,” not. Compare also Chaldee ‘léth,’ i¢ is not, with Telugu “lédu,’ there is not. 01y Seng, ~ AFFILIATION OF THE DRAVIDIAN FAMILY. J 1 h The Semitic analogies observable in the Tamil are neither so 16 numerous nor so important as the Indo-European, nor do they carry )b, with them such convincing evidence; but taking them in connexion il with that more numerous and important class of analogous roots which )i} are found in the Indo-European languages, as well as in the Hebrew, of but of which the Hebrew form is more closely allied to the Dravidian ill (see the Glossarial Affinities), these analogies, such as they are, con- he stitute an additional element of interest in the problem of the origin I, and pre-historical connections of the Dravidian race. I do not adduce eIl these analogies for the purpose of endeavouring to prove the existence ) t0 of any relationship between the Dravidian language and the Hebrew, jan similar to that which subsists between the DraAvidian and the Indo- European languages. Aware of the danger of proving nothing by the proving too much, I content myself with merely stating those ana- itie logies, without attempting to deduce any inference from them. The gOﬂS Indo-Furopean analogies are so intimately connected with the indi- 16w, viduality and vital essence of the Dravidian languages, that it seems For impossible to suppose them to be merely the result of early association, oty however intimate. It is only on the supposition of the existence of a O remote or partial relationship that they appear to be capable of being e, fully explained. In the case of the Semitic analogies, the supposition ectly of a relationship between the two families of tongues does not appear vab, to be necessary. All the analogies that exist can be accounted for on rvade the hypothesis—a very easy and natural one—that the primitive 10018 Dravidian nomades were at some early period before their arrival in W, 10 India, associated with a people speaking a Semitic language. ] any It seems proper here to notice the remarkable general resemblance ations which exists between the Dravidian pronouns and those of the In whatever aboriginal tribes of Southern and Western Australia. way it may be explained or accounted for, the existence of a general resemblance (which was first pointed out by Mr. Norris), seems to be fr, 0 unquestionable; but it has not hitherto been observed that the Austra- lian pronouns of the first person, are more nearly allied to the Tibetan | than to the Dravidian. This will appear from the following compara- tive view of the pronoun of the first person singular. 10 DrAvIDIAN. A USTRALIAN TIBETAN. CHINESE. gqual, ot nin, nd, én nga, ngaii, ngatsa, nga, nge, nged ngo nganya h ﬂ/f Whilst the base of this pronoun seems to be closely allied to the Telugh corresponding pronoun in Tibetan, and in the Indo-Chinese family generally, the manner in which it is pluralised in the Australian dia- 2 B u S MR *&mé’” AN 5 2 -] INTRODUCTION. lects bears a marked resemblance to- the Dravidian, and epecially to the Telugu. The Telugu forms its plurals by suffixing ¢lu’ to the singular; the Australian dialects by a similar addition of ¢lu, ‘li, ¢dlu, dli; &ec. In this particular some of the dialects of the North- Eastern frontier of India exhibit also an agreement with the Telugu : e. g., compare Dhimal ¢ n&, thou, with ‘nyel, you. In the Australian dialects T find the following plurals and duals of the pronoun of the § first person—uwe, or we two, ngalu,” ¢ ngadlu,” ¢ ngadli, ‘ngalata,’ &e. Compare this with the manner in whieh the Telugu forms its plural; e.g. ‘vad’-u,’ he; ‘vAdlu, they ; and even with the colloquial Tamil plural of the pronoun of the first person; e.g., ¢ nan, 7 ; ‘nanggal,’ we The resemblance between the Australian pronouns of the second person, both singular and plural, and those of the Dravidian languages l is more distinet and special ; and is apparent, not only in the suffixes, but in the pronominal base itself. The normal forms of these pronouns I in the Dravidian languages are—singular, ‘nin,’ plural, ‘nim.” The personality resides in the crude root ¢ ni, thow ; which is the same in — both numbers, with the addition of a singular formative ‘n’ (e. g. ¢ ni-n,’ thow), and a pluralising formative ‘m’ (e. g., ‘ni-m,” thous, or you). Tn some cases the pluralising particle ‘m’ has been displaced, and ‘r,’ i which I regard as properly the sign of the epicene plural of the third 6 person, has been substituted for it; e. g., ‘nir, you (in Telugu ‘mir-u’). This abnormal form ‘nir is most used in a separate form : the older and more regular ‘nim’ retains its place in compounds, and in the [ imperative of the verb. Whilst ¢i’ is the vowel which is almost in- 01 variably found in the singular of the pronoun of the second person, in the plural, ‘i’ often gives place to ‘u,’ as in the classical Tamil It is to be noticed also Ql <numa, your, and the Brahui ¢ num,” you. that the modern Canarese has softened ‘nim’ into ‘niva’ or ¢ niwu,’ £X [ in the nominative. qu It is singular, in whatever way it may be accounted for, that in each and all of the particulars now mentioned the Australian dialects 86e Under (y resemble the Dravidian. See the following comparate view. W the Australian head I class the dual together with the plural, as being substantially the same. Wh b DRAVIDIAN. A USTRALIAN. fh thou, nin, nin ninna, nginne, ngintoa, ningte €0 you, nim, nim, nir, num, niwa nimedoo, nura, niwa, ngurle al th Compare also the accusative of the first person singular in Tamil, €l ¢ ennei,’ me, with the Australian accusative ¢ emmo.’ THE REPRESENTATIVE DIALECT. 53 The grammatical structure of the Australian dialects exhibits a general agreement with the languages of the Scythian group. In the use of post-positions instead of prepositions; in the use of two forms of the first person plural, one inclusive of the party addressed, the other exclusive ; in the formation of inceptive, causative, and reflective verbs by the addition of certain syllables to the root ; and, generally, in the n agglutinative structure of words and in the position of wordsin a i sentence, the dialects of Australia resemble the Dravidian—as also the Turkish, the Mongolian, and other Scythian languages: and in the ki same particulars, with one or two exceptions, they differ essentially il from the dialects which are called Polynesian. A The brief vocabularies of the Australian dialects which have been compiled do not appear to give additional confirmation to the resem- »lld blances pointed out above: but it is difficult to suppose those resem- 168 blances to be unreal, or merely accidental ; and it is obvious that the il Australian dialects demand (and probably will reward) further exami- 108 nation. WuaT DIALECT BEST REPRESENTS THE PRIMITIVE CONDITION OF ; 1 THE DRAVIDIAN ToNGUES ? 1) Before entering upon the grammatical comparison of the Dravidian b1 dialects, it seems desirable to ascertain where we should look for their hird earliest characteristics. Many have been of opinion that the Shen- u). Tamil, or high dialect of the Tamil language, is to be regarded as the dder best representative of the primitive Dravidian speech. Without under- the estimating the great value of the Shen-Tamil, I am convinced that no {10 one dialect can be implicitly received as a mirror of Dravidian antiquity. 101, A comparison of all the dialects that exist will be found our best and fanﬂl safest guide to a knowledge of the primitive tongue from which the qlso existing dialects have diverged ; and not only the Shen-Tamil, but il every existing dialect, even the rudest, will be found to contribute its quota of help towards this end. The Tamil pronouns of the first and at 1 second person cannot be understood without a knowledge of the Ancient ﬂects Canarese : and the Ku, one of the rudest dialects, and the grammar of Tnder which was reduced to writing only a few years ago, is the only dialect heing which throws light on the masculine and feminine terminations of the Still it is unquestionable Dravidian pronouns of the third person. that the largest amount of assistance towards ascertaining the primitive condition of the Dravidian languages will be afforded by the Tamil, ngte and in particular by the Shen-Tamil ; and this naturally follows from e the circumstance that of all the Dravidian idioms the Tamil was the earliest cultivated. Tamﬂ, INTRODUCTION. 1 PrioriTy oF THE LITERARY CULTIVATION OF THE TAMIL. ¢ ( 4 The relatively high antiquity of the literary cultivation of the Tamil being a matter of interest considered in itself, irrespective of its W by important bearings on the question of Dravidian comparative grammar, I shall here adduee a few of the evidences on which this conclusion I rests. ¥ W 1. The Shen-Tamil, which is the language of the poetry and of the ancient inscriptions, and which not only contains all the refinements I 0 which the Tamil has received, but also exhibits to a great extent the primitive condition of the language, differs more from the colloquial a Tamil than the poetical dialect of any other Dravidian idiom differs 0 from its ordinary dialect. It differs from the colloquial Tamil so con- siderably that it might almost be considered as a distinet language : T th for not only is classical Tamil poetry as unintelligible to the unlearned Tamilian as the Aneid of Virgil to a modern Italian peasant, but even fic prose compositions written in the eclassical dialect might be read for hours in the hearing of a person acquainted only with the collo- I0( quial idiom, without his understanding a single sentence. Notwith- th Ty standing this, High Tamil contains less Sanserit, not more, than the colloquial dialect. It affects purism and national independence ; and pln its refinements are all ad nira. As the words and forms of the Shen- an Tamil cannot have been invented by the poeets, but must at some e period have been in actual use, the degree in which the colloquial Tamil de has diverged from the poetical dialect, notwithstanding the slowness it with which language, like every thing else, changes in the East, is a hi proof of the high antiquity of the literary cultivation of the Tamil. T 2. Another evidence consists in the extraordinary copiousness of o the Tamil vocabulary, and the number and variety of the grammatical T&i forms of the Shen-Tamil. The Shen-Tamil grammar is a crowded | H]\i museum of obsolete forms, cast-off inflexions, and curious anomalies. ey A school lexicon of the Tamil language, published by the American toy missionaries at Jaffna, contains no less than 58,500 words; and it Wi} would be necessary to add several thousands of technical terms, besides {la provincialisms, in order to render the list complete. Nothing strikesa 1) Tamil scholar more, on examining the dictionaries of the other Dravi- T4 dian dialects, than the paucity of their lists of synonyms in comparison Wa with those of the Tamil. The Tamil vocabulary contains not only those Jos words which may be regarded as appropriate to the language, inasmuch ill( as they are used by the Tamil alone, but also those which may be con- Ty sidered as the property of the Telugu, the Canarese, &ec. Thus, the i ANTIQUITY OF THE TAMIL. 5 J word used for house in ordinary Tamil is ¢ vidu ;’ but the vocabulary contains also, and often uses, the word appropriate to the Telugu, €il (Telugu, ‘illu’), and the distinctive Canarese word, ‘manei’ (Canarese, he ‘mana) ; besides another synonym, ‘kudi, which it has in common Its with the whole of the Finnish languages. The grammar and voca- ar, bulary of the Tamil are thus to a considerable extent the common 0 repository of Dravidian forms and roots: and as the grammars and vocabularies of the other dialects contain only the words and forms he which are now peculiar to themselves, we may conclude that the lite- nts rary cultivation of the Tamil dates from a period prior to that of the the other idioms, and prior to the final breaking up of the language of the ancient Dravidians inte lialects. lial 8. Another evidence of the antiquity and purity of the Tamil o1 01 consists in the agreement of the Ancient Canarese, the Malayalam, the 76 Tulu, and also the Tuda, Gond, and Ku, with the Tamil, in many of ned the particulars in which the modern Canarese and the Telugu differ yen from it. for 4. The fact that in many instances the forms of the Telugu 1lo roots and inflexions have evidently been corrupted from the forms of ith the Tamil, is a strong confirmation of the higher antiquity of the Tamilian forms. Instances of this will be given in the section on the the and phonetic system of these languages. It will suffice now to adduce, as en- an illustration of what is meant, the transposition of vowels which we find in the Telugu demonstrative pronouns. The true Dravidian ome il demonstrative bases are ‘a,” remote, and ‘i, proximate; to which are suffixed the formatives of the genders, with ‘v’ euphonic, to prevent 1688 hiatus. The Tamil demonstratives are €avan, ¢lle, and ‘ivan,’ Aic. 154 The Telugu masculine formative answering to the Tamil ‘an,” is “ du,’ or ‘adu ; and hence the demonstratives in Telugu, answering to the a of Tamil ‘avan,” ¢ivan,” might be expected to be ‘avadu’ and ‘ivadu tieal instead of which we find ¢vadd,” 4lle, and ‘vidu,’ FZic. Here the yded demonstrative bases ‘a’ and ‘i, have shifted from their natural posi- pliess tion at the beginning of the word to the middle ; whilst by coalescing rical with the vowel of the formative, or as a compensation for its loss, their d it quantity has been increased. The altered, abnormal form of the gsideﬂ Telugu is evidently the later one ; but as even the high dialect of the kesd Telugu contains no other form, the period when the Telugu grammar )rﬁ\"l' was rendered permanent by written rules and the aid of written com- u‘iSOD positions, must have been subsequent to the origin of the corruption those in question, and therefore subsequent to the literary cultivation °f the uch Tamil. ) (Ol 5. Another evidence of antiquity consists in the great cor- the oAk LA ’; 56 INTRODUCTION. fo ruption of many of the Sanscrit derivatives that are found in the Tamil. I¢ ? The Sanscrit contained in Tamil may be divided into three portions 00 | d: of different dates, introduced by three different parties. ‘t (1.) The most recent portion was introduced by the school of S X Sankara Acharya, the apostle of Advaita, or Vedantic Saivism, and by de its chief rival, the school of Sri Vaishnava, founded by Raménuja ch ﬁ Acharya. The period of the greatest activity and influence of those sects extended from about the tenth century, A.p., to the fifteenth ;* i 1; and the Sanscrit derivatives introduced by the adherents of these it W | systems (with the exception of a few points wherein change was un- avoidable) are pure, unchanged Sanscrit. Jiki (2.) The school of writers, partly preceding the above and partly T] contemporaneous with them, by which the largest portion of the or i & Sanscrit derivatives that are found in Tamil were introduced, was that of the Jainas, which flourished from about the eighth century, a.p., to g the twelfth or thirteenth. The period of the predominance of the i Jainas (a predominance in intellect and learning—rarely a predomi- i nance in political power) was the Augustan age of Tamil literature, oo the period when the Madura College, a celebrated literary association, Tng] flourished, and when the Cural, the ChintAmani, and the classical voca- Tar bularies and grammars were written. Through the intense Tamilie of | nationalism of the adherents of this school, and their jealousy of bah Brahmanical influence, the Sansecrit derivatives which are employed in oth their writings are very counsiderably altered, so as to accord with Tamil the euphonic rules. Thus ‘loka,” Sanscrit, the world, is changed into Sar ‘ulagu 7 ‘raja,’ a king, into ‘arasu; and ‘ri,’ night (an abbreviation Te of ‘ratri’), into ‘iravu.’ the Nearly the whole of the Sanscrit derivatives that are found in Ien Telugu, Canarese, and Malayalam belong to the periods now men- Jeri tioned, or at least they accord on the whole with the derivatives Jaiy Jaiy * Sankara Acharya is supposed by Professor Wilson to have lived in the eighth or ninth century, A.p.: but the statement which I have here made relates intg the i: not to Sankara Acharya personally, but to the school of theology and philosophy which was founded by him. This school did not reach the acme of its influence thy in the Carnatic till the tenth or cleventh century, when it appears probable that the great temples of the Carnatic were erected. Those temples, the most stupen- The dous works of the kind in the Bast, owe their existence to the enthusiasm and zeal of the adherents of the system of Sankara Acharya. I have not yet been gy able to ascertain the exact date when any of the more celebrated temples was S— erected ; but from inscriptions in my possession recording donations and endow- [ ments made to them, I am able to state that the greater number of the Saiva g temples were in existence in the twelfth century, many in the eleventh, and a few in the tenth. I have not ascertained the existence of any Vaishnava temple in ()]de the South before the twelfth century. only ANTIQUITY OF THE TAMIL, 57 the found in the Tamil of those two periods, especially the former or more recent. They are divided, according to the degree of permutation or l0ng corruption to which they have been subjected, into the two classes of ‘tat-sama,’ the wery same, .., words which are identical with | of Sanserit, and ‘tad-bhava,’ the same nature, i.e., words which are d by derived from a Sanscrit origin, but have been slightly corrupted or I changed by local influences. those The former class, or ‘tat-sama’ words, are scarcely at all altered, ih and generally look like words which have been used only by Brah- these mans, or which had been introduced into the vernaculars at a period 8 U~ when the Sanscrit alphabetical and phonetic systems had become naturalised, through the predominance of the later forms of Hinduism. partly Those Sanscrit derivatives which have been altered more considerably, f the or ‘tad-bhava’ words, do not appear to have been borrowed directly g th&t from the Sanscrit, but are represented by Telugu and Canarese gram- D, 10 marians themselves as words that have been borrowed from the Pra- of the crits, or colloquial dialects of the Sanscrit, which were formerly spoken odomi- in the contiguous Gaura provinces. tature, (8.) In addition to the Sanscrit derivatives of the two pericds now jation, mentioned—the Jaina and the modern Vedantic Saiva periods—the y0ea- Tamil contains many derivatives belonging to the very earliest period E amilio of the literary culture of the language,—derivatives which are pro- bably of an earlier date than the introduction of Sanscrit into the sy of other dialects. gyed in The derivatives of this class were not borrowed from the northern Pracrits (though much more corrupted than even the , Tamil ed 1nto Sanscrit which was borrowed from those Praecrits by the Canarese and Telugu), but appear to have been derived from oral intercourse with 3\'i:lti0ﬂ the first Brahmanical priests, scholars, and astrologers ; and probably und 10 remained unwritten for a considerable time. The Sanscrit of this w men- period is not only greatly more corrupted than that of the period of the Jainas, but its corruptions are of an entirely different character. The vatives Jainas altered the Sanscrit which they borrowed in order to bring it into accordance with Tamil euphonic rules; whereas in the Sanscrit of 4 the Je relates the period which is now under consideration—the earliest period— losophy the changes that have been introduced are in utter defiance of rule. influence The following are instances of derivatives of this class : e that stuped: (a.) The Sanscrit ‘sri, sacred, was altered into °tiru ;> whilst a fastn A0 more recent alteration of the Sanscrit word is into ¢stri)’ yeb heen v 168 a8 (0.) The Sanscrit ¢ karmam,” ¢ work, is in the Tamil of the more od endov modern periods altered into ‘karumam’ and ‘kanmam ; but in the the ST older Tamil it was corrupted into ‘kam, a word which is now found alld g fem fomple 2 only in the old compound, ¢ kam(m)-alan,” an artificer. 58 INTRODUCTION. still (c.) Several of the names of the Tamil months supply us with illustrations of early corruptions of Sanserit. The Tamil months, and | ancle though now solar-siderial, are named from the old lunar asterisms ; Chris the names of which asterisms, and still more the names of the months borrowed from them, are greatly corrupted. Z.g., the asterism ‘plirva- point 4shadam, is changed into ¢plraddam :* ‘ashidam,’ also is changed into also t lon adam,” from which is formed ¢ 4di,’ the Tamil name of the mouth July —August. The name of the asterism ‘aswini’ has been corrupted into T ¢ eippasi,” which is the Tamil name of the month October —November. and t with The change of ¢pfirva bhadra-pada,’ the Sanscrit name of one of the asterisms, into ‘ purattisi’ is still more extraordinary. ¢ Plirva- ‘nir, bhadra-pada’ was first changed into ¢ plrattadi,’ the name of the cor- 1s is tlons responding asterism in Tamil; and this, again, by the shortening of the first syllable and the change of ¢di’ into ¢si,” became ¢ plirattasi,’ the Tamil month September —October. found The corresponding names of the asterisms and months in Telugu, the Canarese, &c., are pure, unchanged Sanscrit ; and hence the greater a e antiquity of the introduction of those words into Tamil, or at least the fur} any] greater antiquity of their use in Tamil written compositions, may safely In be concluded. tablets 6. The higher antiquity of the literary cultivation of the Tamil the Re In Carnataka and may also be inferred from Tamil inseriptions. Tomj] idion Telingina, every inscription of an early date, and the majority even of modern inscriptions, are written in Sanscrit. Even when the knowy characters employed are the Ancient Canarese or the Telugu (characters earliep which have been arranged to express the peculiar sounds of the chrong Sanserit), it is invariably found that Sansecrit is the language in which are ey In the Tamil the inscription is written, if it is one of any antiquity. the Sey country, on the contrary, all inscriptions belonging to an early period are written in Tamil ; and I have not met with, or heard of, a single the inger Sanscrit inscription in the Tamil country which appears to be older Lenfioy than the fourteenth century, A.p., though I have obtained fac-similes °haracter ﬂﬂll]e, wh of all the inscriptions that I could hear of in Tinnevelly and South ha}'e begy Travancore—integral portions of the ancient Pandiyan kingdom. The o U g number of inscriptions that I have obtained is about a hundred and fifty. WhiCh e They were found on the walls and floors of temples, and on rocks and ﬁOIlS of i coh({lll‘i[ pillars. The latest are written in Grantham, or the character in which 6 Ohl Sanserit is written by the Dravida Brahmans ; those of an earlier age * Wy in an old form of the existing Tamil character ; and the earliest in a Dﬂ‘]v pro Elghth 0 0 hayy Tdjy n * T hope at some future period to make public the items of historical informa- tion whieh are contained in those inscriptions; not one of whieh is included in Plceg | ANTIQUITY OF THE TAMIL. 59 vith still older character, which appears to have been common to the Tamil ths, and the ancient Malayéala countries, and is the character in which the s ancient Sasanas in the possession of the Jews at Cochin and of the Syrian Christians in Travancore are written. nths This character presents some IVa points of resemblance to the modern Telugu-Canarese character, and into also to the character in which some undeciphered inscriptions in Cey- lon and the Eastern Islands are written. July nto The language of all the more ancient of these inscriptions is Tamil ; \bet., and the style in which they are written is that of the classical dialect, 6 of without any of those double plurals (e.g., ‘ninggal,” yous, instead of va- ‘nir, you), and other unauthorized novelties by which modern Tamil ) 001 is disfigured ; but it is free also from the affected brevity and involu- 19 of tions of the poetical style. fisy As no inscription of any antiquity in Telinghna or Carnétaca is found to be written in the Canarese or the Telugu language, whatever be lug, the character that is employed, the priority of Tamil literary culture, reater as well as its national independence to a considerable extent, may st the fairly be concluded. safely I may here remark that the Cochin and Travancore ¢shsanas’ or tablets which are referred to above, and which have been translated by Tamil the Rev. Dr. Gundert, prove conclusively, not only the priority of a and Tamil to Malayala literature, but also the derivation of the Malayala idiom from the Tamil. Y evel The date of those documents is not certainly known, but is probably not later than the ninth century, A.p., nor N the racters earlier than the seventh ; for the technical terms of solar-siderial chronology (derived from the Sirya-Siddhanta of Arya—bhatt}a) which of tjhe are employed in these inscriptions were not generally introduced till which Tamil the seventh century. The ‘shsanas’ were written at a time when the pel‘iod the inseriptions belonging to the Mackenzie collection of MSS. , single I may, however, e 01(101' mention here the following interesting items.—(1.) The generally fictitious character of the long lists of kings of Madura, each with a high-sounding Sanserit iiles name, which are contained in the local ¢ purnas’ and other legends, and which have been published by Professor Wilson in his Historial Sketch of the Pandiyan 1 South Kingdom, and by Mr. Taylor in his Oriental Historical MSS. (2.) The veracity 5, The and accuracy of most of the references to the PAndiya and Chodla dynasties od fiftye which are contained in the Mahi-wanso and other historical records and compila- tions of the Singhalese Buddhists. (3.) The fact, or proof of the fact, of the ,cks alld conquest of the whole of the Pdndiya country, including South Travancore, by the Chélas in the eleventh century. (4.) The probable identification of Sundara 1 Which Pandiyan, by whom the Jainas (sometimes erroneously termed Buddhists) were Jier a9 finally expelled from Madura, and whom Professor Wilson has placed in the - st 10 eighth or ninth century a.n. with the ¢ Sender Bandi,’ who is said by Marco Pole AT 33 e Ll PR SN B e et TR to have been reigning in the southern part of the peninsula during his visit to India in the middle of the thirteenth century. The same Sundara Pindiyan is e placed by native Hindu authorities some millions of years before the Christian | for era ! cluded r i3 e i 60 INTRODUCTION. Chérd or Kérala dynasty was still predominant on the Malabat coast :* iti] but though words and forms which are peculiar to the modern Iep Malayéla language may be detected in them, the general style of the language in which they are written is Tamil ; the inflexions of the Jar furt nouns and verbs are Tamil, and the idiom is mostly Tamil ; and we are therefore left to infer that at that period Tamil was the language at lontt TESL least of the court and of the educated classes in the Malayéla country, and that what is now called Malayalam, if it then existed at all, was B probably nothing more than a rustic patois that was current amongst the inhabitants of the hills and jungles in the interior. The fact that the ‘shsanas’ which were given by the ancient Malayala kings to the Ma Jews and Syrian Christians, are in the Tamil language, instead of the £0 | Malayalam, cannot be accounted for from the circumstance of the of temporary conquest of any part of the Malayala country by the 4D ancient kings of Madura ; for the kings in question were Kérala, not Gan Pandiya, kings, with Kérala names, titles, and insignia; and it is evi- dent from the Greek geographers themselves, from whom alone we | know anything of this conquest, that it was only a few isolated places, o (i on or near the Malabar coast, that were really under the rule of the of T Pandiyas. The only part of the Malayéla country which at’ that (rha period could be regarded as belonging bond fide to the Pandiyas, was hill the southern part of ¢ Paralia, 4.e., South Travancore, a district which has always been inhabited chiefly by Pandiyas, and where to the Dr present day the language of the entire people is Tamil, not Malay- aAlam. SR Su— of ¢ From the various particulars mentioned above it appears certain Drj W hll that the Tamil language was of all the Dravidian idioms the earliest cultivated: it also appears highly probable, that in the endeavour to Stry the 11 ascertain the characteristics of the primitive Dravidian speech, from whic which the various existing dialects have been derived, most assistance will be furnished by the Tamil. The amount and value of this assistance lient Whig will appear in almost every portion of the grammatical comparison on which we are about to enter. It must, however, be borne in mind, as LI has already been intimated, that neither the Tamil nor any other of W G * One of them is dated ‘in the seventh year of King Ravi Varma, opposite upl the second year.’ By this vexed expression, ‘opposite the second year, Mr. Whish supposed that a reference was made to the ©second cycle of a thousand & fy years from the building of Quilon, a calculation according to which the present by 1856, would be the thirty-first of the third cycle; but the same expression year, is excee dingly common in the ancient Tamil inscriptions (e.g. ¢the seventh year an | of King Kulaskhara, opposite the fifteenth year ’); and it denotes, I conceive, the year of ‘the cycle of sixty’ (which was formerly the prevailing calculation all Gng over India and the East) to which the year o f the king’s reign stands ¢ opposite,’ or answers. smﬁ'[ I EARLIEST® TRACES OF THE DRAVIDIAN LANGUAGES. 61 R dert single dialect, ancient or modern, can be implicitly adopted asa faithful the representative of the primitive Dravidian tongue. A careful com- [ the parison of the peculiarities of all the dialects will carry us up still (I further, probably up to the period of their mutual divergence, a period long anterior to that of grammars and vocabularies ; and it is upon the 26 b ntry, result of such a comparison that most dependence is to be placed. , was Jongst EArLIEST EXTANT WRITTEN RELICS OF THE DRAVIDIAN LANGUAGES. t that to the The Dréavidian words which are contained in the RamAyana, the of the Mahé-bharata, and other Sanserit poems of undoubted antiquity, are of the so few that they throw no light whatever upon the ancient condition )y tbe of the Dravidian languages, prior to the eighth or ninth centuries A.D., the earliest date to which any extant Tamil compositions la, not Is evi- can safely be attributed. The name ‘Pandiya’ being probably of Sanscrit origin, the only Dravidian names which are contained in the ne W poems referred to, are ‘Chola,’ corrupted from the Tamilic ‘Séra’ place% : Of the (commonly pronounced ‘Chéla’), the collective name of the Tamilians of Tanjore, and ‘ Malaya,’ the name of a mountain range, the Western b that Ghauts, which is probably derived from the DraAvidian ‘mala,’ « 8 Was X8 hall. , which It is 2 remarkable circumstance, that the largest stock of primitive to the Malay- Dravidian words which is contained in any authentic written document of ancient times—the earliest extant traces of the existente of the certain Dravidian languages, as distinguished from the Sanscrit—are those which are contained in the notices of the Greek geographers, Ptolemy, eatliest Strabo, and the author of the Periplus Maris Erythrae; including also your 10 i the Natural History of Pliny. Many of the names and places and tribes p, from @ which are recorded by those geographers, not long after the commence- gistance ment of the Christian era, are identical, letter for letter, with the names lstance which are now in use. Several of those names have become obsolete, or cannot now be identified : but the signification of the compound words [ﬂmdi (& of which they consist cannot be mistaken ; and in several of them we 'ny Other can detect the operation of some interesting dialectic peculiarity or euphonic rule which is still characteristic of these languages. I subjoin , ot )ﬂd yeal a few examples of Dravidian words of this class which are recorded thou“‘m by the Greeks. pe et (1.) ‘6 Mavdiwy’'—* oi Havéidves, ‘Pandiya,’ is probably a word of ¢x 165:10 ey el th yel! Sanscrit origin, but the masculine termination which is given by the )Ilbelve’ I Greeks is unmistakeably Tamil. The Tamilic sign of the masculine [ation & oppoclfe: singular is ‘an ; consequently o Tlavéiwy’ (and still better the plural 62 INTRODUCTION. Tam form of the word, ‘Havéidves, which is applied to the subjects of the (iree Pandiya monarchy), faithfully represents the Tamil nominative sin- gular ‘Pandiyan.’ ( supp The form of the masculine singular in Ancient Canarese which aceor corresponds to the Tamil ‘an,’ is ‘am:” in Telugu it is ‘udu,” so that midy ‘Pandiyudu’ in Telugu, answers to ‘Pandiyan’ in Tamil. Consequently, D) we learn that, as early as the Christian era, the Tamil differed dialec- tically from the other Dravidian idioms, and that its mode of forming the | indi¢ the masculine singular was then the same as it is now. ‘Pandiya’ was not the name of any one king, but the titular name of the dynasty of 10t but 1 Madura (Mo’ﬁov/)a ﬁam'/\ewv Hav&.o'vw). The race were ‘Pandis,” or part ‘PAndiyas’ (mavdwves); the king, the ‘Pandiyan’ (¢ Iavdlwv), or the Ao ‘Pandiya Déva. It is a proof of the advanced social position which Sever was occupied by the Pandiyas, that after the termination of the political relations which subsisted between the Greeks of Alexander’s COr2g iale time, and the princes of the Punjaub, the Pandiyas were the only Indian princes who perceived the advantages of an Kuropean alliance. dre Two embassies were sent by the Pandiyan king to Augustus : the first & Jun Telug (which is mentioned in the Eusebian fragments) was received by Augustus at Tarragona; the second is mentioned by Strabo. The P Moz friendship of the Romans was sought by only one other Hindu prince, 6 knpofBddpos, the King of Chéra (or Kérala), who was also a Dravidian, ‘Ségﬂ; and probably a Tamilian. * (2.) ¢ Korrwapa. This is the name of a place in the country of the (St ¢Aii, or ‘Paralia’ (identical with South Travancore), which is called Undg from ‘Kottiara Metropolis’ by Ptolemy, ‘Cottora’ by Pliny. Undoubtedly the town referred to is ‘Kottira,” or, as it is ordinarily spelled by ag g Vordg Europeans, ‘Kotaur,” the principal town in South Travancore, and now, Dartigy as in the time of the Greeks, distinguished for its commerce. The T — i iy name of the place is derived from ‘Kéd-u,” Tam., a line of circumval- Sén e, L gy lation, a fortification, and ‘4r4," a riwer. It is a rule in the Tamil and the Malayalam, that when a word like ‘K6d’ is the first member of a SOX‘{“I toit {l compound, the final ‘d’ must be doubled for the purpose of giving the ~¥ith word the force of an adjective: it is another rule that sonants when ‘bUP&H A theﬂlee] doubled become surds. Consequently the compound ‘kod’- aRa& > becomes P S A 303 i i) lk el ave, { by rule ‘kott-ara.’ It is interesting to perceive that in the time of the \\Qrd ( Greeks the same peculiar phonetic rules existed which are now in opera- St tion. ufch It is also worth noticing that the Greek writers represent the last syllable of the name of the town, not as ‘aru,” but as ‘ara.’ The Wit Waop, Tamil has ‘4ru,” the Malayilam ‘4ra.’ At Kotaur, the dialectic pecu- Wagg il liarities of the Malayala language begin to supersede those of the il [he o EARLIEST TRACES OF THE DRAVIDIAN LANGUAGES. 63 of the Tamil; and this appears to have been the case even in the time of the e §in- Greeks. (3-) “’Aprarov Baci\ewor. The place referred to by this name was which supposed by one of the editors of Ptolemy to be Bijnagar; which would o that accord well enough, it is true, with the position which Ptolemy,gives it, midway between the sources of the CAvéri and Gondwana: but the enly, dialec- resemblance of the name to that of Arcot, and the circumstance that mming the place is represented as the capital of ‘the nomadic Séras’ (Zdpar), indicate the propriety of identifying it with Arcot in the Carnatic: for 9 WS not only was Arcot included in the ancient Séra or Chéla Kingdom, ity of dis, or but there is a distinct, uniform tradition, that the inhabitants of that or the part of the Carnatic which lies between Madras and the Ghauts, including Arcot, were ‘Kurumbars,” or wandering shepherds—nomades—for which of the several centuries after the Christian era. If this identification is ander’s correct, we have another instance of the antiquity of the existing dialectic peculiarities of the Tamil ; for the second syllable of the name o ouly Areot, (properly ‘4ru-kidu,’ the Jungle ow the riwer), viz., ‘kad-u,’ [liance. he first a jungle, is peculiar to the Tamil,— the corresponding word used in Telugu being ‘atavi’ or ‘adavi.’ ved by Ptolemy gives the name of the people of the neighbouring country ), The more accurately than the Sanserit writers. They are called in Tamil | prince, yidian, ‘Soras;* Cholas in Sanscrit; but Sérae, and also Sorigi, or Sorigéti * I am doubtful whether the eastern coast of India derived from this word y of the (‘Sora’) the name of the ‘Coromandel’ coast, by which it is styled by Europeans. Undoubtedly Fra Paolo & St. Bartolomaeo was wrong in supposing it to be derived § called ubtedly from ‘chéla-mandalam,” the millet country. “Chélam’ is not millet, but maize ; and compounds of indigenous Drividian words like ‘cholam’ and Sanserit Jled by words like ‘mandalam’ are ordinarily inadmissible; and this compound in particular is quite unknown. ¢So6ra-mandalam,’ the country of the Séras, who nd 1oV, are called ‘ Chélas’ in Sanscrit, is a compound which is in actual use, like ¢ Pandi- mandalam,’ the country of the Pdndiyas, and ¢ Séra- 2, T]Ie Séras, or Kérala : and doubtless this is the w mandalam, the country of the il ord with which Paolo’s informants had supplied him. This derivation of the word °Coromandel, viz. from il and S Sl e e s “Sora-mandalam,’” has generally been accepted ; but there is this serious objection to it, that the name of that part of the eastern coast—from Cuddalore to Madras ferof & —with which Europeans first became acquainted, is ‘Tonda-mandalam,” not ¢ Séra-mandalam in addition to which, these terms are rarely used by the natives tg when themselves : their use is restricted to classical compositions, and it is extremely unlikely that the first European mariners and factors ever heard of them. We becomes have, therefore, to seek for some more trite, easy, and natural derivation of the o of the sand), the name of a small village on the eastern coast n word ‘ Coromandel; and this I think we find in ¢ Karu-manal’ (literally black (= ear Pulicat (the first iﬂ 01)61’21‘ o Dutch settlement), written ¢ Coromand which is invariably up to the present day pronounced and i the s = el’ by the Europeans who are resident in Madras; some of — it yeings R e e whom annually tak e refuge in ‘ Karumanal’ or ‘Coromandel’ during the hot land i The s winds. Coromandel is often the first point which is sighted by ships from Europe ot bound to Madras; and the o i peet e bjects on which my own eyes first rested on approach- f the ing the coast in January, 1838, were the cocoa-nut trees of Coromandel and the g 0 distant Nagari hills. ok e S o S TR i 64 INTRODUCTION. (). The two last names must have been applied by by the Greeks. K Ptolemy to the Soras of the Tanjore delta; for the Cavéri flowed 18 through the country of the Sorigi, and “Xafnpos, the emporium at the 1 mouth of the CAvéri (which he calls Xdfpnpis,”), belonged to them. The Soras ate sometimes in poetical Tamil called ‘Soragas’ or ‘Soriyas, de and their country ‘Séragam,—¢g’ being optionally added to many roots qll as an euphonie. The ‘r’ of the Tamil word ¢Séra,’ is a peculiar sound, it not contained in any of the other Drévidian dialects; in which it is (ﬂ al generally represented by ‘I’ or ‘d;’ in Sanserit and in the Pali of the ¢ Maha-wanso’ by ‘1. The more accurate spelling of this word given W] by the Greeks shews that then, as now, the use of this peculiar vocalic [V ‘r’ was a dialectic characteristic of the Tamil. th (4.) Modogalingum. Pliny observes, «Insula in Gange est .... g Modogalingum nomine.” The same island, country, or city (for the o description of it is somewhat obscure) is called by Ptolemy, Triglyphum all( or Trilingum. Though the place referred to is said to be “on the Ta Ganges,” it may have been considerably to the south : for the Godavery has always been considered by Hindus as a branch of the Ganges, or in] as mythologically identical with it; and the Greeks would most 0 probably be taught to regard it in the same light. At all events, from O M A AR 5 b 14 the the circumstance that the Andhras and Calingas (the two ancient divi- the sions of the Telugu people) are represented by the Greeks as Gangetic pations, and as living in or near Triglyphum, it may be considered as the certain that Triglyphum, Trilingum, or Modogalingum, was identical 1a with Telingéna, or Trilingam, °the country of the three lingas;’ cly from which word, indeed, the modern term ¢ Telinga’ is ordinarily the derived by native grammarians. The derivation of ¢ Telugu,’ oby, ¢ Telungu,” or ¢ Telinga’, from ¢ Trilinga’ is repudiated by Mr. C. P. §ent Brown ; who also states that the name ‘ Trilinga’ is not contained in any R —— This statement is probably 161 of the ancient Sanscrit lists of countries. (k ka correct : nevertheless, the ancient use of the appellation ¢ Trilingam,’ and and the identity of the names Trilingam and Modogalingum, are proved s ARt Caste by the evidence of Ptolemy and Pliny, as conclusively as if they had Peap been mentioned by Sanscrit writers. This being the case, the Telugu name and language are fixed near the mouths of the Ganges, or at gz least between the Ganges and the Godavery, about the commencement the I Cay of the Christian era : and not only so, but the existence of the dialectic gy peculiarities of the Telugu, as early as the time of the Greek geogra- ‘(i phers, may safely be inferred ; inasmuch as ¢ modoga,” the word used Gl by Pliny, is the ancient word for three (‘ moda,” or ¢ modoga’), answer- lam ing to the Canarese ¢miru, the Tamil ‘minru’ ( pronounced miin- ki dru), and the modern Telugu midu.’ The word used by Pliny being — e e e EARLIEST TRACES OF THE DRAVIDIAN LANGUAGES. 65 exclusively a Telugu word, we may conclude that at that early by period the dialectic peculiarities of the Telugu, one of which wed is the use of ‘d’ where the other dialects have ‘r, were already ; the in existence. The (5.) Kapovpa ﬂao‘/)\ewv xypofolpov. The place referred to is evi- yas, dently Kartr, a town in the Coimbatoor country, which was formerly 10018 the capital of the Chéra dynasty. ¢ Cerobothrus’ is given as the yund, titular name of the king of the country (ordinarily called by Tamilians 1t 18 ¢the Chéran’), whose rule extended over Coimbatoor, part of Mysore, f t he and a portion of the Malabar coast. Probably ‘Cerobothrus’ is identical gl\ el with ¢ Chéra-putra,” son of Chéra. The Greek spelling of the word ocalic knpo confirms the supposition of the identity of the Dravidian title of the dynasty, Séra or Chéra, with the Sanscrit ‘Kérala, and the greater antiquity of the latter mode of spelling. or § lie The name Kapodpa, in Tamil ¢ Karfir, is derived from ¢ kar,’ dlack, phuny and ‘qar 9 2 a town. The exact agreement of the Greek word with the n the Tamil is remarkable. avery It is deserving of notice that in Ptolemy’s lists of names of places ogs, O in India the termination ovp or ovpe, equivalent to the Dravidian ¢fr,’ - mosh a town, (Anglice, ‘oor’ or ¢ore’), is frequently met with, not only in g from the southern part of the peninsula, but as far north as the mouths of 11 divi- the Ganges. gjngetriﬁ (6.) 6t Kapéor. The Carei of Ptolemy were a people who inhabited red 88 the southern part of Tinnevelly, in whose country part of the ¢ Para- entical lia’ of the author of the Periplus seems at one time to have been in- ingas; cluded. ¢Kare’ or ‘karei’ is the Tamil word for coast or shore (from Jinanlly the verbal root ¢ karei,’ to be melted down,—to be washed away), and is Felugu:’ obviously identical in meaning with the Greek ITapakia. Up to the pre- P sent time part of the Tinnevelly coast—that part where I have myself | inany resided and laboured for thirteen years—is called by the same name )fObﬁbly (¢ karel,” the shore) by which the whole southern coast of Tinuevelly [ngay and Travancore appears to have been known to the Greeks; and a o 10v@ caste of fishermen found farther north are called ‘karei-(y)-ar,’ coast- hey had people, This is one of the names given by Ptolemy to Telug (7.) Kal\iywov. 65, or ab %%% the promontory of Kupv. This promontory is supposed by some to be ool Cape Comorin ; but as it is said to be situated opposite the most northern point of Ceylon, and to form the boundary between the “Gangetic Gulf, or Bay of Bengal, and the ¢ Orgalic’ or ¢ Agaric Gulf,’ the Gulf of Manaar, it is evidently Point Calimere. The Tamil ;ol'd 1se % name of this point, from which ¢ Calimere’ has been corrupted, is f ﬁnswer‘ ‘kalli-médu,’ the cactus eminence,; and it is evident that the first part ol DI F iy being g 7 A g e 66 INTRODUCTION. of the Greek name Ka\\iyucor is identical with the Tamil ¢kalli; cactus, the first part of the name by which the place is now called. il (8.) Amongst many words of less importance of which the Tamil signification can be easily recognised, I subjoin the following :—Ila\otpa ] (obviously from ¢ pal’ milk, and ‘ftr,” a town), a place in the Bay of } Bengal, possibly at the mouth of the ¢Palar,” Milk-river, a river 1¢ th which flows into the Bay of Bengal a little to the south of Madras: Tevviyopa (from the Tamil ‘ten,” south, and the Sanserit ‘nagara,’ Hl a city), a town in the Sora country : also the word épv{a, 7ice, which th is obviously derived from the Tamil ¢ariéi,” »ice deprived of the husk; . this being the state in which rice was then, as now, bought up in India a for exportation to Europe.* of (9.) During the period in which the Greeks traded with India, the i names of places and tribes recorded by them, and various circum- (¢ stances which they have related, prove that the Brahmans had then pl established themselves in the Carnatic, and given names to some of af the principal places. ¢Mosdovpa’ (Madura) is a Sanscrit word, signi- £X fying the sweet city ; the name of the Cavéri, “’XdaBnpds,’ the yellow river, is claimed by the Sanscrit, though possibly Tamil ; and 1eg * The Hebrew word for pea-fowl, which is ¢ thuki’ in the Book of Kings, ) o ¢ thfiki’ in Chronicles, is certainly Drvidian. The pea-fowl is an Indian bird. It was probably on the Malabar or Western coast of India that the pea-fowl was of procured by (or for) Solomon’s servants; and the old classical name of the fowl in Tamil is ¢ tdkei, dialectically pronounced °tdgei. In modern Tamil ¢tokei’ Wit generally signifies only the peacock’s tail, or any similar tail-feathers; but in an old classical Tamil it signifies also the peacock itself. If this identification is be correct, the Hebrew word referred to is the oldest specimen of the Drividian languages which is extant in any written document. The Arabic word for the lan peacock, ‘tawas,’ and the Armenian ‘taus,’ are probably derived from the same Jeq source ; as also the Greek ©radic,” with which, by the insertion of the digamma, some connect the Latin ¢ pavo.’ I cannot connect the Tamil t6g-ei,” as Max T Tnfe Miiller does, with the Sans. ¢sikhin; for it is regarded by Dravidian lexico- graphers as a pure Tamilian word; and the Tamil corruption of ¢sikhin’ is at find “sigi,” @ peacock, which is a recognised Sans. derivative. ‘tdg-ei’ is not in Canarese or Telugu. by gt u Huge old specimeéns of the Baobab, or Adansonia Digitata, an African tree, of which the Hindus do not know even the name, may still be seen in or near P various sites of foreign commerce in the extreme south of the Indian peninsula : i am e. g., in Kottar, near Cape Comorin, and near Tutocorin in Tinnevelly—possibly on the site of the ancient Kolkhi. By what race of foreign merchants were those !:mg trees planted ? The great age to which they are known to grow (they are called by Humboldt ‘some of the oldest specimens of organic life on the globe’) will 0 p‘ fr(m- admit of the supposition that they were brought from the mouth of the Red Sea by the Grecian navigators, or even by the Phenicians and ‘the servants of Solomon ’ themselves. May it not have been by the same people that the Hebrew word ‘sak, a sack (in Tamil ‘sikk-u,” in Malayalam ¢ chikka’), was introduced into Southern % tem India? 0l This word, though so long naturalised that it is considered by native of I scholars to be indigenous, is unknown to the Telugu and Canarese, as well as to the Sanserit. It is found only where the Baobab is found, and where the Hebrew 0f name for the peacock had its origin. i inqh Ty Al E ; EARLIEST TRACES OF THE DRAVIDIAN LANGUAGES. 67 ly Koudpua drpov’ (Cape Comorin) is certainly derived from the Sanserit This word is com- ‘kuméri. @ virgin, a name of the goddess Durga. il monly pronounced in Tamil ¢ kumiri ; and in the vulgar dialect of the pu people residing in the neighbourhood of the Cape, a wirgin is not of It is ‘kumari,’ or ‘kumdiri, but ©kumar,” pronounced ¢komar. er remarkable that this vulgar corruption of the Sanserit is identical with the name which is given to Cape Comorin by the author of the Periplus. 1, He says, ¢ After this, there is another place called ‘ Kouap,” where ich there is a fort and harbour, where also people come to bathe and sk, purify themselves: . for it is related that a goddess was once dia accustomed to bathe there monthly.” This monthly bathing in honour of the goddess Durgd or Parvati, is still continued at Cape Comorin, the but is not practised to the same extent as in ancient times. Cape 11} Comorin formerly ranked as one of the five renowned sacred bathing hen places” (a representation which accords with the statement of the g of author of the Periplus), but the number of visitors to it now is gni- extremely small. Jlow Though the Greek geographers have not given us any information and respecting the languages of India, beyond what is furnished by the names of places contained in their works, the information derived ings, from those lists is exceedingly interesting. The earliest extant traces bird. ] was of the Drividian languages which possess reliable authority, are those 5 fO\\'l with which we have been furnished by the ancient Greeks ; and from kel it in an examination of the words which they have recorded, we seem to jon 8 be justified in drawing the conclusion, not only that the Dréavidian .'idian langnages have remained almost unaltered for the last two thousand i the e years, but also that the principal dialects that now prevail had a sepa- meﬁx rate existence at the commencement of the Christian era, and prevailed J Max exic0- at that period in the very same districts of country in which we now in' 18 find them. The art of writing had probably been introduced, the NALEse grammar of the Dravidian languages had been arranged, and some 166 of progress made in the art of composition, several centuries before the p near arrival of the Greek merchants ;* and the fixity with which those psula: gssibly languages appear to have been characterised ever since that period is o hose in perfect accordance with the history of all other Asiatic languages, ; callgd ) vl from the date of the commencement of their literary cultivation. o Seb s 0 * The arrival in India of those Grecian merchants, appears to have been con- temporaneous with the conquest of Egypt by the Romans. The earliest Roman f Sak; [} coins found in India are those of the reign of Augustus. A very large number putherd of Roman imperial ‘aurei’ were lately found on the Malabar coast; upwards ! naﬁYe e of thirty types of which, commencing with the earlier coins of Augustus, and including many of Nero, were described by me in a pamphlet published at Hebl'ew Trivandrum in 1851 by the Rajah of Travancore, to whom the coins belonged. F 2 68 INTRODUCTION. If the Dravidian family of languages is allied, as I believe it to be, to the Scythian families, it may justly claim to be considered as one of the oldest members of the group. With the exception of the language of the Behistun tablets, no words belonging to any other Scythian language can be traced up to the Christian era. Mr. Norris says, ‘I know of nothing written in the Magyar language earlier than the fifteenth century, and of the other Ugrian languages we have nothing above fifty or sixty years old The great Finnish heroic poem, ‘the Kalevala,” may be of any age, but as it appears to have been brought down to us only by word of mouth, it has naturally varied, like all traditional poetry, with the varying forms of the lan- guage.’ The Uigurs, or Oriental Turks, acquired the art of writing from the Nestorian Christians, the Mongolians from the Uigurs; so that the literary cultivation of neither of those languages is to be s B s i X S S S s R b compared in point of antiquity with that of the Dravidian. Amongst the carliest records of Seythian tongues that have been discovered, is a brief list of words which are recorded by the Chinese as peculiar to the old Turks of the Altai; and of eight words contained in this list, all of which are found in the modern dialects of the Turkish, probably three, certainly two, are Dravidian. Those words, as given by the Chinese, are :— TURKISH OF THE ALTAL MopErN TURKISH. T AMIEL. black, koro quard kar-u old, kori gori kira chieftain, kin khin kon or kd I am strongly inclined to consider the last Tamil word, ‘kén’ or ¢ ko, to be identical with the ¢ kén,” ¢ khan,” or ‘khégan’ of the Turco- Mongolian languages. The Ostiak, an Ugrian dialect, has ‘khon ; and the word signifying king, which is found in the Scythic version of the Behistun tablets, and which certainly commences with ‘k,” or ‘kh,’ In the old Tamil in- is conjecturally written by Mr. Norris ‘ko.’ seriptions I have invariably found ko or ‘kén, instead of the San- scrit ¢ Raja :’ but the word has become obsolete in modern Tamil, except in compounds, and in the honorific title ‘kon,” which is assumed by shepherds. This conjunction of meanings (king and shepherd) is very interesting, and reminds one of the Homeric description of kings ag ¢ 7rm/ue've9 Nabv. The Tamil literature now extant enables us to ascend, in studying the history of the language, only to the eighth or ninth century, A.p.: the Dravidian words handed down to us by the Greeks carry s up, as RELATION OF DRAVIDIANS TO NORTH-INDIANS. 69 we have seen, to the Christian era Beyond that period, the compa- rison of existing dialects is our only available guide to a knowledge of the primitive condition of the Dravidian language The civilization of the Tamil people, together with the literary cultivation of their lan- guage, commenced probably about the sixth or seventh century, B.C., but the separation of the primitive Drividian speech into dialects must have taken place shortly after the arrival of the Dravidians in the districts which they at present inhabit—an event of unknown, but certainly of very great antiquity The Irish and the Welsh dialects of the Celti the Old High and the Old Low dialects of the Teutonic, and the Finnish and Magyar dialects of the Ugrian, had probably become separate and distinct idioms before the tribes by which those dialects are spoken settled in but the various Dravidian dialects which their present habitations % are now spoken appear to have acquired a separate existence subse- quently to the settlement of the Drévidians in the localities in which A we now find them. Supposing that their final settlement in their > 3 ~ present abodes in Southern India took place shortly after the Alyan irruption (though I think it probable that it took place before), every Z grammatical form and root which the various dialects possess in common, may be regarded as at least coeval with the century subse- Every form and root which quent to the arrival of the Aryans. the Brahui possesses in common with the Drividian tongues may be The Brahuic analogies enable regarded as many centuries older still. us to ascend to a period anterior to the arrival in India of the Aryans (which cannot safely be placed later than 1600, B.c.); and they furnish us with the means of ascertaining, in some degree, the condition of the I Dréavidian language before the Dravidians had finally abandoned their original abodes in the central tracts of Asia of POLITICAL AND SoCIAL RELATION oF THE PRIMITIVE DRAVIDIANS TO THE ARYAN AND PRE-ARYAN INHABITANTS OF NORTHERN INDIA ly 1 The arrival of the Dravidians in India was undoubtedly anterior n- to the arrival of the Aryans, but there is some difficulty in determining 1, ol whether the Dravidians were identical with the Scythian aborigines whom the Aryans found in possession of the northern provinces, and 1§ g5 to whom the vernacular languages o Northern India are indebted for their Un-Sanserit element, or whether they were a distinet and more anclent race The question may be put thus :— Were the Dravidians g and ¢ Mléchchas, by whom the progress identical with the ¢ Dasyus of the Aryans was disputed, and who were finally subdued and incor- P} 70 INTRODUCTION. porated with the Aryan race as their ‘Slidras,” or serfs and depen- dents? or were they a race unknown to the Aryans of the first age, and which had already been expelled from Northern India, and duven southwards towards the extremity of the Peninsula before the Aryans it arrived ? This question of the relation of the Dravidians to the primi- tive Stdras, or Aryanised Mléchchas, of Northern India is confessedly involved in obscurity, and can be settled only by a more thorough —_— investigation than any that has yet been made of the relation of the Dravidian languages to the Un-Sanscrit element contained in the northern vernaculars We may, indeed, confidently regard the Dra- vidians as the earliest inhabitants of India, or at least as the earliest race that entered from the North-West, or crossed the Indus; but it is St not so easy to determine whether they were the people whom the Aryans found in possession, or whether they had already been expel- led from the northern provinces by the pre-historic irruption of another Scythian race Some recent inquirers hold the identity of the Dré- vidians with the primitive Stdras; and much may be said in support of this hypothesis I am not competent to pronounce a decided opinion point which lies so far beyond my ewn province, but the diffe- rences which appear to exist between the Dravidian languages and the é Seythian under-stratum of the northern vernaculars induce me to incline to the supposition that the Dravidian idioms belong to an older period of the Scythian speech—the period of the predominance of the Ugro-Finnish languages in Central and Higher Asia, anterior to the westward migration of the Turks and Mongolians If this supposition is correct, it seems to follow that the progenitors of the Seythian por- tion of the Stidras and mixed classes now inhabiting the northern and western provinces must have made their way into India subsequently to the Dravidians, and alse that they must have thrust out the Dra- vidians from the greater part of Northern India, before they were in their turn subdued by a new race of invaders By whoemsoever the Dravidians were expelled from Northern India, and through what causes soever they were induced to migrate southward, I feel persuaded that it was not by the Aryans that they were expelled. Neither the bllbjugatlon of the Cholas, Pandiyas, and other Dravidians by the nor the expulsion from Northern India of the races who after- Ar yans, wards became celebrated in the South, as Pandiyas, Cholas, Kéralas, Calingas, Andhras, &e., is recognised by any Sanscrit authority, or any Dravidian tradition. Looking at the question from a purely Dravidian point of view, I am convmccd that the Driavidians never had any rela- tions with the primitive Alyans but those of a peaceable and friendly character ; and that if they were expelled from Northern India, and i i RELATION OF DRAVIDIANS TO NORTH-INDIANS 71 forced to take 1efuge in Gondwana and Danda-Kéaranya, the great Drévidian forest, prior to the dawn of their civilisation, the tribes that subdued and thrust them southwards must have been Pre—Aryans Those Pre—Aryan Scythians, by whom I have been supposing the Dravidians to have been expelled from the northern provinces, are not and other to be confounded with the Kéles, Sontals, Bhills, Déms aboriginal tribes of the North. Possibly these tribes had fled into the forests from the Dravidians prior to the Pre- Ary&n invasion, just as the British had taken refuge in Wales before the Norman conquest. It is also possible that the tribes referred to had never crossed the Indus at all, or occupied Northern India, but had entered it, like the Bhutan tribes, by the North-East, and had passed from the jungles and swamps of Lower Bengal to their present abodes,—taking care always to keep on the outside of the boundary line of civilisation At all events, w cannot suppose that it was through an irruption of those forest tribes that the Dravidians were driven southwards ; nor does the Un-Sanserit element which is contained in the northern vernaculars appear to accord in any degree with the peculiar structure of the Kole languages. The tribes of Northern India whom the Arya.ns gradually incorporated in their community, as Stidras, whosoever they were, must have been an They were probably identical with organized and formidable race the ¢ Althiopians from the East, who, according to Herodotus, were T e brigaded with other Indians in the army of Xerxes, and who differed from other Athiopians in being ¢ straight-haired. 4 0 I admit that there is a difficulty in supposing that the Dravidians, who have proved themselves greatly superior to the Aryamsed Stdras of Northern India in mental power, independence, and patriotic feeling, should have been expelled from their original possessions by an irrup- tion of the ancestors of those very Stdras It is to be remembered :i however, that the lapse of time may have effected a great change in the warlike, hungry, Scythian hordes that rushed down upon the first 0 Dravidian settlements. It is also to be remembered that the dependent 1 and almost servile position to which this secondary race of Scythians d was early reduced by the Aryans, whilst the more distant Dravidians 6 10 were enjoying freedom and independence, may have materially altered their orlglnal character It is not therefore so improbable as it might at first sight appear, that after the Dravidians had been driven across ) the Vindhyas into the Dekhan by a newer race of Scythians, this new y race, conquered in its turn by the Aryans and reduced to a dependent il position, soon sank beneath the level of the tribes which it had ex- pelled whilst the Dravidians, retaining their independence in the y southern forests into which they were driven, and submitting eventually d — 72 INTRODUCTION. to the Arya.ns not as conquerors, but as colonists and instructors, gradually rose in the social scale, and formed communities and states in the Dekhan, rivalling those of the Aryans in the north.* Mr. Curzon (Journal of Royal Asiatic Society, vol. 16) recently attempted to meet the difficulty which I havg stated by supposing that the Tamilians were never in possession of Arya-vartta, or Northern India, at all ; but tflat they were eonnected with the Malay race, and came to Southern India by sea, from the opposite coast of the Bay of ii Bengal, or from Ceylon. This theory seems, however, perfectly gra- tuitous ; for it has been proved that the languages of the Goénds and Kus are Dravidian, equally with the Tamil itself ; that the Rajmahal is also substantially Drividian ; and that the Brahui partakes so largely of the same character (not to speak of the language of the Scythic tablets of Behistun), as to establish a connection between the Drévidians and the ancient races we:st of the Indus. It has also been shewn that in the time of Ptolemy, when every part of India had long ago been settled and civilised, the Dravidians were in quiet possession, not only of the south-eastern coast, but of the whole of the Peninsula, up nearly to the mouths of the Ganges. It is undeniable that immigrations from Ceylon to the southern districts of India have occasionally taken place. The Teers (properly ¢ Tivar, islanders) and the Ilavars, ‘Singhalese,” (from ‘Ilam,’ Ceylon, T s a word which has been corrupted from the Sanserit ¢ Simhalam,” or rather from the Pali ‘Sihalam, by the omission of the initial f.al); B both of them Travancore castes, are certainly immigrants from Ceylon ; but these and similar immigrants are not to be considered as Singha- lese, in the proper sense of the term, but as offshoots from the Tamilian S s e population of the northern part of the island. They were the partial reflux of the tide which peopled the northern and western parts of Ceylon with Tamilians. Bands of marauding Tamilians (¢ Sollies, ¢Pandis,” and other ¢Damilos,” <. e, Cholas, Pandiyas, and other * ‘Dekhan’ is a corruption of the Sanserit ¢ dakshina,’ the south, literally the :; reght (‘dexter’) , an appellation which took its rise from the circumstance that the Brahman in determining the position of objects, looked towards the East, which he called ‘plirva,’ the opposite region, when whatever lay to the southward was to the right. The South, as the region of freedom, safety, and peace, was to the primitive Dravidian what the East was to the Brahman. He called it ‘ten,” of whieh one meaning in Tamil is opposite, another, sweet : whenee also *tennei’ is the Tamil name of the cocoa-nut, literally the sweet nut > whilst the North was ‘vada’ (the north-wind ‘videi), which is probably connected with ‘vad-u,’ to wither,—the north wind being regarded by Tamilians with as much dread as the south wind (mythologically the car of Kima, the Indian Cupid) was associated with the idea of everything that was agreeable. Referring to the physical configuration of the Carnatic, the Drividians called the East downward 5 the West, the region of the Ghauts, upward. USE OF THE TERM ¢ SUDRA. 73 Tamilians) frequently invaded Ceylon, as we are informed by the o) Maha-wanso, both before and subsequently to the Christian era On several occasions they acquired the supreme power, and at length y they permanently occupied the northern provinces of the island M There is no relation, however, between the Singhalese language—the language of the Singhalese, properly so called, who were Buddhists and colonists from Magadhd or Behir—and the language of the f Tamilians ; nor is there any reason for supposing that the natural course of emigration (viz., from the mainland to the island) was ever d inverted to such a degree as to justify the supposition that the whole )| mass of the Dravidians entered India from Ceylon £ OR1GINAL UsE AND PROGRESSIVE EXTENSION oF THE TERM ‘SODRA 6 I The term ‘Stdra,” which is now the common appellation of the 8 mass of the inhabitants of India, whether Gaudians or Dréividians, l seems originally to have been the name of a tribe dwelline near the J Indus Lassen recognises their name in that of the town =dépos on the Lower Indus; and especially in that of the nations of the S¢4&poc in Northern Arachosia. He supposes them to have been, with the Abhiras and Nishadas, a black, lonrr-hzured race of aborigines, not b originally a component part of the Alyzm race, but brought under its ¥ influence by conquest; and that it was in consequence of the Stdras having been the first tribe that was reduced by the Aryans to a dependent condition, that the name ‘Stdra’ was afterwards, on the conquest of the aboriginesin the interior part of the country, extended to all the servile classes Whatever may have been the origin of the 4 name ‘Stdra,” it cannot be doubted that it was extended in course of f time to all who occupied or were reduced to a dependent condition ; whilst the name ¢ Mléchcha’ continued to be the appellation of the unsubdued, Un- Ar} anised tribes. 1 Most writers on this subject, including Lassen and Max Miiller, 4 suppose that the whole of the Sidras, or primitive, servile classes of 0 Northern India, to whom this name was progressively applied, 0 belonged to a different race from their Alyan conquerors. Whilst T G assent to every other part of the supposition, I am unable to assent to f this. It seems to me to be probable that a considerable proportion of the slaves, servants, dependents, or followers, of the high-caste Aryans 0 belonged to the Aryzm race. As the Slavonian serfs are Slavonians, 4 and the Magyar serfs Magyars, there 1s no improbability in the sup- I position, that a large number of the Aryan serfs or Sidras, perhaps the majority, were Aryans, and I cannot on any other supposition 74 INTRODUCTION. Il account for the fact that nine-tenths of the component materials of the northern vernaculars are Sanscrit f The supposition of the Aryan origin of a large number of the Stidras, seems also most in accordance with the very old mythological statement of the oricin of the Stdras from Brahm feet; for though 4 the Brahmans, Kshatriyas, and Vaisyas, the twice-born classes, are represented as springing from more honourable parts of Brahmé’s body i? | yet the Stdras are represented to have sprung from the same divinity, though from "an ignoble part; whereas the Nishadas, or barbarian aborigines, are not represented to have sprung from Brahmé at all It E but formed a ‘fifth class,’ totally unconnected with the others. ol i appears from this mythological tradition that the Sudms were supposed 1 in the first ages to differ from the ‘twice born’ Aryas in rank only, not in blood I regard as confirmatory of this view the statement of Manu that €all who become outecasts are called Dasyus Whother they > for in the speak the language of the Mléchchas or that of Ar yas: same manner, all who enjoyed the protection of the Aryas, as their dependents and servants, would naturally receive a common appella- tion, probably that of Stidras,—whether, as aborigines, they spoke ‘the lan guage of Mléchchas,’ the Scythian vernacular, or Whether as Aryas of an inferior rank in life, they spoke ‘the language of Ary‘Ls, e T—— a colloquial dialeet of the Sanscrit. It is true, as Professor Max % Miiller says, that the three twice-born castes alone are called Alyas by the Satapatha-Brahmana of the Rigveda: but as ‘the four castes,’ including the Stidras, but excluding the Dasyus and le}mda,s, are E distinctly referred to in the most ancient hymns; as outcaste Aryas are styled Dasyus by Manu; and as the higher classes of the Tamilians monopolize the national name in this very manner, and pretend that the lower classes of their race are not Tamilians, I think that we may safely attribute the statement in question to the pride of ‘the twice-born. Even the Vratyas, who are distinguished from the Stdras and are regarded as an inferior class, did not dlfﬂ)l‘ from the Brahmans in language, and must, therefore, have been Alyas The aboriginal Scythian inhabitants of India seem to have been subdued, and transformed from Mléchchas into Stdras, by slow degrees At ACRPANT AR N R PR s S T R o sy B In the age of Manu, they retained their independence and the appel- lation of Mléchchas in Bengal, Orissa, and the Dekhan; but in the earlier period which is referred to in the historic legends of the Maha-bharata, we find the Mlechchas and Dasyus disputing the pos- gession of Upper India itself with the Aryas Sagara, the thirty- fifth king of the Solar dynasty, is related to have laboured in vain to subdue the heterodox aborigines residing on or near his frontier: and a :.)f USE OF THE TERM ¢ SUDRA. 75 lie in the reign preceding his, in conjunction with certain tribes connected with the Lunar line, those aborigines had succeeded in overrunning his he territories.® cal The introduction of the Drividians within the pale of Hindaism, th and the consequent change of their appellation from Mléchchas to that e of Slidras appears to have originated, not in conquest, but in the ly peaceable process of colonisation and progressive cwlhsatlon. There i, is no tradition extant of a warlike irruption of the Alyas into Southern Al India, or of the forcible subjugation of the DrAvidians; though if such ll, an event ever took place, it must have been subsequent to the era It of Manu and the Raméiyana, and therefore some remembrance of it ad would probably have survived. All existing traditions, and the names ly, by which the Brahmanical race is distingunished in Tamil, viz., ‘Eiyar of instructors, fathers, and ‘Parppar,” overseers, (probably the émwkomor of ey Arrian), tend to show that the Brahmans acquired their ascendency by their intellicence and their administrative skill Lie The most adventurous immigrations from Northern India to the elr la Dekhan were those of the offshoots of the Lunar dynasty, a dynasty he which originated from the Solar, and whose chief city was Ayodhya, % a8 Oude, the traditional starting point of most of their immigrations 15 The Pandiya kings of Madura were feigned to have sprang from the ax Lunar line The title ‘Pandiya,” is supposed to be derived from the lj '{U * SAgara, finding himself unable to extirpate or enslave those heterodox b8, E tribes, entered into a compromise with them, by imposing upon them various distinguishing marks; by which, I think, we may understand their obstinate AT persistence in the use of the distinguishing marks to which they had been accus- tomed. Omne of those marks is worthy of notice in an inquiry into the relations yaa ‘The PéAradas s of the early Dravidians it is recorded, ¢ wore their hair long % in obedience to his commands Professor Wilson observes, with reference to this ab statement (in his notes on the Vishnu Purana), ¢ What Oriental people wore their hair long, except at the back of the head, is questionable ; and the usage would be we characteristic rather of the Teutonic and Gothic nations.’ The usage referred to U is equally characteristic of the Drividians, and it is even possible that the Piradas may have been a Drividian tribe. Up to the present day the custom of wear- the ing the hair long, and twisted into a knot at the back of the head, is charac- he teristic of all the inferior castes in the southern provinces of the Tamil country, and also of the shepherds and Maravars In ancient times this mode of wearing the hair was in use amongst all Drividian soldiers; and sculptured represen- el tations prove that at a still earlier period it was the general Drividian custom The Kotas of the Nilgherry Hills wear their hair in the same manner. The Tudas 08 ol wear their hair long, but without confining it in a knot. Probably it was from the é: Dravidian settlers in Ceylon that the Singhalese adopted the same usage; for as the early as the third century A.n., Agathemerus, a Greek geographer, describing Ceylon, says, ‘ the natives cherish their hair as women among us, and twist it round the their heads.” The wearing of the hair long appears to. ha,ve been regarded by the 0 early Drividians as a dlstmctwe sign of national independence : whllst the shaving of the hair of the head, with the (xccptmn of the ‘kudumi,” or lock at the back ty of the head, couespondmw to the tail of the Chinese, was considered as a sign of [tO A1 yamwtmn or submission to Aryan customs, and admission within the palc of Aryan protection. nd 76 INTRODUCTION. A name of the ‘Pandavas’ of Northern India, the celebrated combatants in the great war of the Mahé-bhérata, to whom every Cyclopean work of unknown antiquity is traditionally aseribed. Probably this deriva- tion of the name of ‘Pandiyas’ is correct ; but there is no reason what- ever to suppose that the kings of Madura, by whom this name was assumed, sprang from any of the royal dynasties of Northern India. The Aryan immigrants to the South appear to have been Brahmanical priests and instructors, not Kshatriya soldiers; and the kings of the Pandiyas, Cholas, Calingas, and other Dravidians, appear to have been simply Dravidian chieftains, or ‘Poligars,” whom their Brahmanical preceptors and spiritual directors dignified with Aryan titles, and taught to imitate and emulate the grandeur and cultivated tastes of the Solar, Lunar, and Agni-kula races of kings.* In our own times we may see the progress of a similar process in Gondwana, where barbarous Gond chieftains have learned from their Brahman preceptors, not only * I find that a similar opinion respecting the rclation that subsisted between the Aryans and the early Drividians, has been expressed by Professor Max Miiller (Report of British Association for 1847). He says, ¢ Wholly different from the manner in which the Brahmanieal pcople overcame the north of India, was the way they adopted of taking possession of and settling in the country south of the Vindhya. They did not enter there in crushing masses with the destroying | force of arms, but in the more peaceful way of extensive colonisation, under the protection and countenance of the powerful empires in the north. ‘Though sometimes engaged in wars with their neighbouring tribes, these colonies generally have not tuken an offensive but only a defensive part; and it appears that, after having introduced Brahmanical institutions, laws, and religion, especially along the two coasts of the sea, they did not pretend to impose their language upon the much more numerous inhabitants of the Dekhan, but that they followed the wiser policy of adopting themselves the language of the aboriginal people, and of conveying through its medium their knowledge and instruetion to the minds of uncivilised tribes. In this way they refined the rude language of the earlier inhabitants, and brought it to a perfection which rivals even the Sanserit. By these mutual concessions, a much more favourable assimi- lation took place between the Arian and aboriginal race; and the south of India became afterwards the last refuge of Brahmanical science, when it was banished from the north by the intolerant Mahommedans. It is interesting and important to observe how the beneficial influence of a higher civilisation may be effectually exercised, without forcing the people to give up their own language and to adopt that of their foreign conquerors, a result by which, if successful, every vital principle of an independent and natural development is necessarily destroyed.’ I cannot see how this statement of the Professor can be reconciled with his identification of the old Drividians with the Nish&das of Northern India, In hig more recent Results of Turanian Researches, he interprets Siva’s triumph over Tripura, and the Garuda’s devouring of the Kirfitas, as traditions of the conquest of Nishdda races by the Aryans; and represents the same Nishddas as retirirg before the Aryas to the south of the Vindhyas, broken and scattered in the centre, and violently pressed together even in the south. If the Nishédas who were thus dealt with had been Drividians, I think we should find some distinct notice of this in the Mahd-bhérata, in which the peaceable, polished Cholas, Kalingas, Pandiyas, and other Drévidians of the South, are carefully distin- ——————— guished from the Nishddas of various races, whom the old Solar and Lunar kings are represented as subduing in Northern India. e PR T ———— 77 PRE-AYRIAN CIV;LIS ATION. =29 to style themselves Rajahs, but even to assume the sacred thread of = the ‘twice-born’ Kshatriyas. The only Dravidian kings who are commonly believed to have been really Kshatriyas (though with what truth it is now impossible to say), were the kings of the Kérala dynasty on the Malabar coast; from [; whom the modern Rajahs of Cochin claim to be descended. I It is proper to notice here that the title ‘Stdra’ conveys a higher 3 meaning in Southern than in Northern India. The primitive ‘Stdras’ of Northern India were slaves to the Aryans, or in a condition but 1 little superior to that of slaves. They had no property of their own, and no civil rights. In Southern India, on the contrary, it was upon f the middle and higher classes of the Dravidians that the title of 0 ¢Stdra’ was imposed ; and the classes that appeared to be analogous 1§ to the servile Stdras of Northern India, were not called ¢Stdras,” but ¢ Pallas,” ¢ Parias,” &c., names which they still retain. The application of the term ‘Stdra,’ to the ancient Dravidian chieftains, soldiers, and cultivators may prove that the Brahmans, whilst pretending to do them ! an honour, treated them with contempt ; but it does not prove- that 10 16 they had ever been reduced by the Brahmans to a dependent position, of or that they ever were slaves, like the Northern Siidras, to any class 4 16 of Aryans. The Brahmans, who came in ¢ peaceably, and obtained the kingdom by flatteries,” may probably have persuaded the Dravidians, that in calling them Stidras, they were conferring upon them a title of 16 honour. If so, their policy was perfectly successful ; for the title of 1, Ir ¢Sldra’ has invariably been regarded by Dravidians in this light : i and hence, whilst in Northern India the Stdra is a low caste man, in 10 d Southern India he ranks next to the Brahman, and the place which he e ls occupies in the social scale is immeasurably superior, not only to that I of the Pariars, or agricultural slaves, but also to that of the un- 1 d enslaved low castes, such as the fishermen, and the cultivators of the cocoa-nut and palmyra palms. b [y b PRE-;&RYAN CIVILISATION OF THE DRrAvVIDIANS., al 1§ Though the Dravidians were destitute of letters, and unacquainted 18 ol with the higher arts of life, prior to the arrival of the Brahmans, they do not appear to have been so barbarous and degraded a people as the Puranic legends represent. They are represented to us by the Brah- 16 10 mans as tncouth ‘rakshasas,’ or giants; as monkeys (by an interesting Ch anticipation of the theory of the author of the Vestiges of the 5 Natural History of the Creation, who regards the monkeys of the ) Dekhan as the progenitors of the human race) ; or as vile sinners, who B 78 INTRODUCTION. ate raw meat and human flesh, and disturbed the contemplations of holy Rishis. Even Hanuméan, their king, and Rama’s most useful ally, is half-praised, half-ridiculed, as a monkey-god. This picture may in some few particulars have correctly enough represented the condition of the barbarous Kole, or Gond tribes who inhabited the Vindyha forests; but it cannot be doubted that the Dra- vidians, properly so called, had acquired, at least, the elements of civilisation prior to the arrival amongst them of the Brahmans. If we eliminate from the Tamil language the whole of its Sanserit derivatives, the primitive Dravidian words that remain will furnish us with a faithful picture of the simple, yet not savage, life of the Un- Aryanised Dravidians. Mr. Curzon holds that there is nothing in the shape of a record of the Tamil mind which can recall to us anything independent of an obvious Sanscrit origin ; and that if the contrary supposition were tenable, we ought to find the remains of a literature embodying some record of a religion different from Hinduism. Unequivocal traces of the existence amongst the Un-Aryanised Dravi- ; dians, hoth ancient and modern, of a religion different from Hinduism, will be pointed out in the Appendix. At present I will merely adduce those records of the primitive Tamil mind, manners, and religion RO AT S which the ancient vocabularies of the language, when freed from the admixture of Sanserit, will be found to furnish. § From the evidence of the words in use amongst the early Tamilians, we learn the following items of information. They had ‘kings,” who dwelt in “fortified houses,” and ruled over small ¢ distriets of country :’ they were without ¢books,” and probably ignorant of written alphabetical characters, but they had ¢ minstrels,” who recited ‘songs’ at ‘festivals :’ they were without hereditary ‘priests’ and :‘g“ “idols,” and appear to have had no idea of ¢ heaven’ or ‘hell, of the “soul” or ‘sin ;" but they acknowledged the existence of God, whom they styled ko, or king—a realistic title which is unknown to f orthodox Hinduism. They erected to his honour a ‘temple,’ which i they called ‘Ko6-il,” God’'s-house ; but 1 cannot find any trace of the { “worship’ which they offered to him. The chief, if not the only ! actual worship which they appear to have practised was thut of ‘devils, which they worshipped systematically by ¢giving to the devil, .e. offering bloody sacrifices, and by the performance of frantic ‘devil dances.’ They were acquainted with all the ordinary metals, with the exception of ‘tin’ and ‘zine; with the planets which were ordinarily known to the ancients, with the exception of f ¢ Mercury’ and ‘Saturn. They had numerals up to a ¢hundred,’— some of them to a ‘thousand ; but were ignorant of the higher deno- 0 ] DATE OF DRAVIDIAN CIVILISATION. { 9 of minations, a ‘lakh’ and a ¢crore.” They had ¢medicines, b.ut no ul ‘medical science,” and no ‘doctors ;) ‘hamlets’ and ‘towns,” but no “cities ;' ‘canoes,” ‘boats,’ and even ‘ships’ (small ¢decked’ coasting h vessels), but no foreign ‘ commerce ;' no acquaintance with any people o beyond sea, except in Ceylon, which was then accessible on foot at low water ; and no word expressive of the geographical idea of of ‘island’ or ¢continent.’ They were well acquainted with ¢agricul- ture,” and delighted in ¢war.’ All the ordinary or necessary arts of ! life, including ‘cotton weaving’ and ¢ dyeing,” existed amongst them, s but none of the arts of the higher class. They had no acquaintance ) with ¢painting,” ¢sculpture,’ or ‘architecture ; with ¢astronomy,” or ho even ‘astrology;’ and were ignorant, not only of every branch of ng ¢ philosophy,” but even of ¢ grammar.’ Their unecnltivated intellectual Iy condition is especially apparent in words that relate to the operations 16 of the mind. Their only words for the ‘ mind’ were the ¢diaphragm’ Il (the ®pjr’ of the early Greeks), and ‘the inner parts’ or ‘interior.” VI They had a word for ‘thought; but no word distinet from this for I, ‘memory,’” ‘judgment,’ or ¢ conscience, and no word for ¢ will.’ To express ‘the will’ they would have been obliged to describe it as, ‘that 00 on which in the inner parts says, ‘I am going to do so and so0.” he This brief illustration, from the primitive Tamil vocabulary, of the social condition of the Dravidians, prior to the arrival of the Brah- ly mans, will suffice to prove that the elements of civilisation already existed amongst them. They had not acquired much more than the ad (fs elements ; and in many things were centuries behind the Brahmans of whom they revered as ‘instructors,” and obeyed as ¢ overseers :’ but if they had been left altogether to themselves, it is open to dispute ol nd whether they would not now be in a much better condition, at least in point of morals, than they are. he il The mental culture and the higher civilisation which they derived to from the Brahmans, have, I fear, been more than counterbalanced by the fossilising caste rules, the unpractical, pantheistic philosophy, and ch the cumbersome routine of inane ceremonies, which were introduced he amongst them by the guides of their new social state. y of TrE ProBABLE DATE OF THE he ARYAN CIVILISATION OF THE DRAVIDIANS. (vf ! W7 It would appear, from the unanimous voice of ancient legends, that o8 the earliest Dravidian civilisation was that of the Tamilians of the of Pandiya kingdom, near the southern extremity of the Peninsula. This civilisation is traditionally, and with much show of reason, attributed 10 80 INTRODUCTION. to the influence of successive colonies of Brahmans from Upper India, who were probably attracted to the South by the report of the ferti- lity of the rich alluvial plains that were watered by the Cavéri, and other Peninsular rivers; or as the legends relate, by the fame of Rama- chandra’s exploits, and the sacred celebrity of the emblem of Siva, which Rama discovered and worshipped at Ramisseram, or Ram-is- varam, a holy place in the island of Paumben, between the mainland 5} The leader of the first, or most influential colony, is and Ceylon. traditionally said to have been Agastya, a personage who is celebrated in Northern India as a holy ¢rishi,’ or kermat, but who is venerated in the South with greater reason as the first teacher of science and litera- ture to the primitive Dravidian tribes. It is very doubtful whether :li: Agastya (if there ever were such a person) was really the leader of t the Brahman immigration : more probably he is to be considered as its sbasta e R i mythologica] embodiment. ¢ The Vindhya mountains,” it is said, 8 ¢ prostrated themselves before Agastya; by which I understand that fl they presented no obstacle to his resolute, southward progress; for he N is said to have penetrated as far south as Cape Comorin. He is called Il by way of eminence the ¢ Tamir muni,’ or Zamilian sage; and is cele- a0 brated for the influence which he acquired at the ¢court’ of Kula- il stkhara, according to tradition the first Pandiyan king, and for the fo numerous elementary treatises which he composed for the enlighten- Ul ment of his royal disciple ; amongst which his arrangement of the pr grammatical principles of the language has naturally acquired most €0 renowin. He is mythologically represented as identical with the star 1§ Canopus, the brightest star in the extreme southern sky in India, and ki is worshipped near Cape Comorin as Agast-isvara. By the majority of Iy orthodox Hindus he is believed to be still alive, though invisible to my ordinary eyes, and to reside somewhere on the fine conical mountain, Cer commonly called ¢ Agastya’s hill,” from which the ¢ Porunei’ or ‘Tamra- wh parni,” the sacred river of Tinnevelly, takes its rise. any The age of Agastya, and the date of the commencement of the whi Brahmanical civilisation of the Tamilians cannot now be determined Drig with certainty ; but data exist for making an approximate estimate. It was certainly prior to the Christian era: for then the whole country isly appears to have been already Brahmanised, and the Pandiya dynasty It was as certainly Ag of kings had become known even in Kurope. lite subsequent to the era of the RaméAyana and Manu: for then the whole * The proper name of this mountain is ‘Podeiyam’ or ‘Podeiya-mi-malei,’ the great common mountain, which has received this name from the circumstance that it is equally conspicuous on the Pandiya or Tinnevelly side of the Ghauts, ! and on the Chéra or '1‘1'av§,ncore side. quity DATE OF DRAVIDIAN CIiVILISATION 81 of the Coromandel coast was still inhabited by ¢ Mléchchas,” who ‘ate human flesh,” ¢ consorted with demons,” and ¢disturbed the contempla- tions of holy hermits The age of Agastya is undoubtedly to be placed between those two eras If we be could be sure that the J references to the Cholas, Dravidas, Knntalas, Kéralas, Mushicas, and Karnatakas, which are contained in the present text of the Maha- bharata, formed originally part of that poem, the era of the commence- 1S ment of Tamilian civilisation, and the date of the Agastyan colony d from which it proceeded, might be brought within a still narrower I compass, and placed between the age of Manu and that of the Maha- bharata The gennineness of those references being as yet doubtful er and the era of Manu (in deference to an allusion to the Chinese, under of the name of ‘Chinas,” which, like similar allusions to the ‘Chinas’ and 1ts ‘ Yavanas’ in the Mahé-bharata, is probably an interpolation) being id, generally placed I think too low, I am inclined to look to Ceylon for b the best means of arriving at an approximate date. The immigration he into Ceylon of the colony of Aryans from Magadha, probably took ed place about B.c. 550, or, at least, some time m the course of that century: |l and I think we may safely argue that the Aryas or Sanscrit-speaking la inhabitants of Northern India, must have become acquainted with, and the formed establishments in, the Dekhan and the Coromandel coast. and el must have taken some steps towards clearing the Danda-karanya, or the primitive forest of the Peninsula, before they thought of founding a 106 colony in Ceylon Wijeya, the leader of the expedition into Ceylon, star is related in the Maha-wanso to have married the daughter of the and king of Pandi; and though it may be doubtful enough whether he yOf really did so (for on the same authority we must believe that he etO married also the queen of the Singhalese demons) this at least is il certain, that it was the persuasion of the earliest Singhalese writers, ) who were, on the whole, the most truthful and accurate of oriental annalists, that the Pandi kingdom of Madura (the first kingdom the which was established on Aryan principles in the Peninsula) existed uued prior to the establishment of the Magadhi rule in the neighbouring L island otry Probably, therefore, we shall not greatly err in placing the era of 1aSty Agastya, or that of the commencement of Tamilian civilisation and iy literature, in the seventh, or at least in the sixth century, B ¢ vh()l RELATIVE ANTIQUITY OF DRAVIDIAN LITERATURE mﬂlel, lcmﬂce Notwithstanding the antiquity of Dravidian civilisation, the anti- hauté quity of the oldest Dravidian literature extant is much inferior to that G T — 82 INTRODUCTION. of the Sanscrit. Indeed it is questionable whether the word ¢anti- quity > is a suitable one to use respecting the literature of any of the T E—————_— Dravidian languages. The earliest writer on Telugu grammar is said to have been a sage called ¢Kanva,” who lived at the court of Andhra—réya,, the king in whose reign Sanscrit was first introduced into the Telugu country, according to the tradition which was formerly mentioned. For this tradition there is probably a historical groundwork, the introduction of Sanserit derivatives being necessarily contemporaneous with the immigration of the Brahmans ; and the statement that the first attempt to reduce the grammatical principles of the language to writing proceeded from a Brahman residing at the court of a Telugu prince, is a very reasonable one. Kanva’s work, if it ever existed, is now lost; and the oldest extant work on Telugu grammar (which is composed,like all Telugu grammars, in Sanscrit) was written by a Brahman, called Nanniah Bhatta, or Nannappa, who was also the author of the greater part of the Telugu version of the Maha-bharata, which is the oldest extant composition of any extent in Telugu. Nannappa lived in the reign of Vishnu Vard- hana, a king of the Calinga branch of the Chalukya family, who reigned at Rajamundry. The reign of this king is placed by Mr. A. D. Campbell about the commencement of the Christian era ; but Mr. C. P. Brown, in his Cyclic tables, places it, on better authority, in %’% the beginning of the 12th century, A.p. ¢ With the exception of a few other works, which were composed towards the end of the 12th century, nearly all the Telugu works that are now extant were written in the fourteenth and subsequent centuries, P after the establishment of the kingdom of Bijnagar, or Vijaya-nagara ; 0 and many of them were written in comparatively recent times. il Though the Telugu literature which is now extant cannot boast of a high antiquity, the language must have been cultivated and polished, and many poems that are now lost must have been written in it, prior to the twelfth century—the date of Nannappa’s translation of the Maha-bharata : for as this translation is considered ‘the great standard of Telugu poetry,” it cannot be supposed to have sprung into existence all at once, without the preparation of a previous literary culture. Tamil literature is undoubtedly older than Telugu, though the high antiquity which is ascribed to some portions of it by the Tamilian my literati cannot be admitted. W The sage Agastya occupies in Tamil literature a place of still el greater eminence and importance than that of Kanva ia Telugu. lfa Not only is the formation of the Tamil alphabet attributed to ANTIQUITY OF DRAVIDIAN LITERATURE. 83 Agastya, and the first treatise upon Tamil grammar, together with the original settlement of the grammatical principles of the language; but he is also said to have taught the Tamilians the first principles of medicine, of chymistry or alchymy, of architecture, astronomy, and law ; and some of the most ancient and admired treatises on all these sciences, as well as many modern ones, are attributed to his pen. It is admitted by Tamilians that his grammar does not now exist; but they suppose him to have been the anthor of most of the extant treatises on medicine and other sciences which bear his name. Though the literary cultivation of the Tamil language may have commenced, as the Tamilians believe, in the age of Agastya (premising however, that it is undecided whether he was a real personage, or is only to be regarded as the mythological representative of a class or period), I feel quite certain that none of the works which are com- monly ascribed to Agastya, were written at so early an age. Probably there is not any one of them older than the tenth century, A.p. Of the works which are attributed to him, those which advocate f the system of the ‘Siddhas’ (Tamulice ‘Sittar’), or the ‘Siddbhantam,’ a mystical compound of alchymy and quietism, with a tinge of Christianity, were certainly written after the arrival of Europeans in India: and Agastya's name appears to have been used by the writers, as had been done by many successions of authors before, for the pur- 1t i pose of gaining the ear of the people for whose use the books were composed. We cannot doubt that the substance of the following stanza, od at which is contained in the ¢ Njina ntru,’” or centum of wisdom, a small poem attributed to Agastya, has been borrowed from statements of P8y Christianity, notwithstanding that Christianity is not directly named &) in it, or in any other work of this class :— of ¢ Worship thou the Light of the Universe; who is One; el, Who made the world in a moment, and placed good men in it ; 101 Who afterwards himgelf dawned upon the earth as a Guru; Who, without wife or family, as a hermit performed austerities; the ard Who appointing loving sages (siddhas) to succeed him, Departed again into heaven :—worship Him.’ 700 It is a striking illustration of the uncritical structure of the Hindu the mind, that this stanza is supposed, even by Tamil literati, to have been Jian Heathens written by Agastya himself many thousands of years ago. endeavour to give it a heathen meaning, and Hindu Christians regard stin it as a kind of prophecy. Though there is not a single archaism in it; though it is written d t0 G 2 INTRODUCTION. 84 not only in the modern dialect, but in a vulgar, colloquial idiom, abounding in solecisms; veither party entertains any doubt of its antiquity. Leaving out of account various isolated stanzas, of high but unknown antiquity, which are quoted as examples in the grammatical and rhetorical works, the oldest Tamil works now extant are those which were written, or are claimed to have been written, by the Jainas, or which date from the era of the literary activity of the Jaina sect. The Jainas of the old Pandiya country were animated by a national and anti-Brahmanical feeling of peculiar strength ; and it is chiefly to them that Tamil is indebted for its high culture and its com- The Saiva and Vaishnava parative independence of the Sanscrit. writers of a later period, especially the Saivas, imbibed much of the enthusiasm for Tamilic purity and literary independence, by which the Jainas were distinguished : in consequence of which, though Tamil literature, as a whole, will not bear a comparison with Sanscrit lite- rature, as a whole, it is the only vernacular literature in India which has not been contented with imitating the Sanserit, but has honourably attempted to emulate and outshine it. In one department, at least, that of ethical epigrams, it is generally maintained, and I think must be admitted, that the Sanscrit has been outdone by the Tamil. The Jaina period extended probably from the eighth or ninth century, A.D., to the twelfth or thirteenth. In the reign of Sundara Pandiya, which appears to synchronize with Marco Polo’s visit to India, the adherents of the religious system of the Jainas, were finally expelled from the Péndiya country : consequently, all Tamil works which advocate or avow that system must have been written before the middle of the thirteenth century, A.p., and probably before the f decadence of Jaina influence in the twelfth. It seems reasonable to conclude that the period in which Jaina 1 literature chiefly flourished was that which preceded the enthusiastic 0 propagation of the Vedantic doctrines of Sankara Achérya. If this conclusion is correct, the most celebrated poem which was written by d an avowedly Jaina author—the ¢Chintémani,’ a brilliant romantic epic, containing 15,000 lines—cannot be placed later than the tenth century. 0 The ¢ Nan-ntl,” a High Tamil grammar of great excellence, and the poetical vocabularies, which were all written by Jaina scholars, h P must be placed a little later than the ‘ChintAmani; but yet anterior to the Chola conquest of the Pandiya country, which took place in the § th eleventh century. m The ¢Tol-kippiyam, or ancient composition, the oldest extant % 1n; 86 ANTIQUITY OF DRAVIDIAN LITERATURE. Tamil grammar, is probably to be placed at the very commencement of the Jaina period. Though written by a Saiva, its Saivism is not that of the mystical school of Sankara ; and in the chapters which are extant (for much of it has been lost), native grammarians have noticed the existence of various grammatical forms which are considered to be archaic It is traditionally asserted that the author of this treatise, who is styled technically ¢ TolkAppiyanar was a disciple of Agastya bimself, and that he embodied in his work the substance of Agastya’s grammatical elements This tradition is on a par with that which ascribes so many anonymous works to Agastya nevertheless, if any relics of poems of the first age of Tamil literature still survive, they are to be found amongst the poetical quotations which are contained in this and similar works, and in commentaries which have been written upon them Some of those quotations are probably the oldest specimens of the poetical style that are now extant L, TR Ty e Y L o, I The <Kural’ of Tiruvalluvar, a work consisting of 1330 distichs, or poetical aphorisms, on almost every subject connected with morals and political economy, and which is regarded by all Tamilians (and per- haps justly) as the finest composition of which the Tamil can boast ) appears to be not only the best but the oldest Tamil work of any extent which is now in existence I think we should not be warranted in placing the date of the % Kural later than the ninth century, A.p The reasons which iuduce me to assign to it so high an antiquity i are as follows :— y (1.) The Kural contains no trace of the distinctive doctrines of ks Sankara Achérya It teaches the old Sankhya philosophy, but ignores 16 Sankara’s additions and developments ; and would therefore appear he to have been written before the school of Sankara had risen to notice, if not before Sankara himself, who lived not later than the ninth 0% century £10 his (2.) There is no trace in the Kural of the mysticism of the modern Puranic system ; of Bhaktt, or exclusive, enthusiastic faith in any one by deity of the Hindu Pantheon of exclusive attachment to any of the ti sects into which Hinduism has been divided since the era of Sankara ; ith or even of acquaintance with the existence of any such sects The work appears to have been written before Saivism and Vaishnavism ud had been transformed from rival schools into rival sects ; before the Mt Purénas, as they now stand, had become the text books of Hindu to theology ; and whilst the theosophy of the early Vedinta and the the mythology of the Mahé-bbéarata GOHIPI'lbed the entire creed of the majority of Hindus. ant 86 INTRODUCTION. (8.) The author of the Kural is claimed with nearly equal reason by Saivas, Vaishnavas, and Jainas. On the whole the arguments of | the Jainas appear to me to preponderate, especially that which appeals to the Jaina fone that pervades the ethical part of the work :—e.g., scrupulous abstinence from the destruction of life is frequently declared to be the chiefest excellence of the true ascetic. Nevertheless, from the indistinctness and undeveloped character of the Jaina element which is contained in it, it seems probable that in Tiruvalluvar’s age Jainism was rather an esoteric ethical school, than an independent objective system of religion, and was only in the process of development out of the older Hinduism. This would carry back the date of the Kural to the eighth or ninth century. (4.) Tt is the concurrent voice of various traditions that Tiru- valluvar lived before the dissolution of the Madura Sangam ; i. ., the college of literati, or board of literary examiners, at Madura. It is asserted that the Kural was the very last work which was presented for the approval of that body ; and that it was in consequence of their rejection of the Kural (on account of the low caste of its author) that the college ceased to exist. If any weight is to be attached to this tradition, which has the appearance of verisimilitude, the Kural must be the oldest Tamil composition of any extent that is now extant : for every composition which is attributed (with any show of reason) to the literati who constituted that college, who were in any way connected with it, or who lived prior to the abolition of it (some of whom were the traditional fathers of Tamil literature), has long ago perished. (5.) The Kural is referred to and quoted in grammars and prosodies which were probably written in the tenth century. For these reasons I think the Kural should be placed in the eighth or ninth century at least. It is admitted, however, as in almost every similar inquiry pertaining to Indian literature, that the reasons for this conclusion are rather negative than positive. 15 Certain poetical compositions are attributed to Auveiyar, ¢the Matron, a reputed sister of Tiruvalluvar, of which some, at least, do & not belong to so early a period. § It is a remarkable circumstance that the author of the Kural is 1t represented to have been a Pariar. A later legend represents him to ol have been the offspring of a Brahman father by a Pariar mother. tt His real name is unknown. The Valluvars are the priestly division (s of the Pariars, and the author of the Kural is known only as ¢ Tiru- W valluvar, the sacred Valluvan, or Pariar priest. It is a still more Do remarkable circumstance that the poetical compositions which are now ey 87 ANTIQUITY OF DRAVIDIAN LITERATURE. referred to (small works of universal use and popularity in the Tamil country, and of considerable merit) are ascribed to a sister of Tiru- valluvar, a Pariar woman ! Auveyar’s real name, like that of her brother, is unknown,—* Auvei,” or ¢ Auveiyér,’ signifying a mother, o venerable matron. | The brief verses (éach commencing with a consecutive letter of the Z Tamil alphabet) which are ascribed to Auveiyar, appear to be of con- siderable antiquity : but the Advaita work which is called ‘ Auveiyar’s Z Kural’ was written subsequently to the arrival of the Mahommedans in Southern India ; and the collection of moral epigrams (most of them possessed of real poetic merit) which is called the ¢ Mudurei,” or pro- verbial wisdom, was written after the arrival of Europeans, perhaps after the arrival even of the English. The proof of the modern origin of the Mtdurei is contained in the 16 18 following simile :—¢As the turkey that had seen the forest peacock 3 dance, fancied himself also to be a peacock, and spread his ugly wings T and strutted, so is the poetry which is recited by a conceited dunce.’ ab As it is certain that the turkey is an American bird, which was brought to Europe from America, and introduced into India from 118 86 Europe, there cannot be any doubt of the late origin of the Midurei, if this stanza was always an integral portion of it, as it is represented o1 t0 to have been. When I have mentioned this anachronism to native scholars, and have called their attention to the circumstance that the 2y Tamil word for ¢ turkey ’ (like the words denoting ¢ tobacco,” ¢ potato,’ me ng &ec.), is not an original root, but a descriptive compound, signifying ‘the heavenly fowl’ ¢.e., ‘the great fowl’ they have courageously os maintained that the turkey was always found in India. The date which is commonly attributed to the Tamil translation, hth or rather the Tamil imitation, of the Ramayana, a highly finished and very popular work, is considerably too high. In a stanza which is ery prefixed to the work, and which is always believed to have been written for by the author himself, it is related that it was finished in the year of the Salivihana era corresponding to A.D. 733. This date has been (the accepted as genuine, not only by natives, but by those few European do scholars who have turned their attention to matters of this kind. If it were genuine, the Tamil version of the Raméyana would be the J 18 oldest Tamil composition which is now extant—a supposition to which n 0 the internal evidence of style is opposed ; and the author, Kamban her. (so called from ¢ Kamba nadu,’” a district in the Tanjore country, to glon which he belonged), would claim to be regarded as the father of Tamil [0 poetry. ot N This date, though it is the only one with which T am “acquainted o 88 INTRODUCTION. in the whole range of Tamil literature, is I fear a surreptitious addi- tion to Kamban’s poem, which was prefixed to it by some admiring editor, for the purpose of giving it a higher antiguity than it can justly claim, It is generally stated that Kamban finished his poem in the reign of Kulotunga Choéla : and as certain poetical riddles, purporting to have been given him by Kulétunga Chéla to selve, have come down to the present time, there seems to be no reason to doubt the propriety of placing him in the reign of that king. Mr. Taylor, in his analysis of the MacKenzie MSS., mentions a tradition that Kamban presented his poem to Rajéndra Chéla. As Réajéndra, Kul6tunga’s father, was the Anugustus of the Chola line, it may be supposed that the more celebrated name crept into the story, instead of the less celebrated. = Mr. Taylor 1 represents Rajéndra as Kulétunga's father, not his som: but in an inscription in my possession procured from Kottar, in South Tra- vaneore, and which was written during the period of the occupation of the Pandiya country by the Chélas, it is stated that the temple on which the inscription is cut was ‘erected by Kulotunga Chéla to the honour of the divinity of Rajéndra Chol-isvara,’ 7.e., to Réijendra T g b 3 P R Chola, ‘deified,” or considered as identified with Siva, after his death. I therefore conclude that Kulétunga was Réjéndra’s son, not his father. It makes little difference, however, whether he were father or son : for Kamban may be supposed to have lived in both reigns, and a single reign is of no importance to my present argument. The other premiss of my argument is founded upon the evidence of an inseription it whieh is found on the walls of an old temple at Cape Comorin. That inscription is dated in the reign of Réjéndra Chola, and celebrates a victory gained by Réjéndra over Abhava Malla (a Jaina king, of the The date of the Chalukya race), on the banks of the Tunga-bhadra. inscription is in the two hundredth year of the Quilon era (a popular local era), answering to 1025, A. . Mr. Walter Elliot’s inscriptions, found in the old Chalukya country, place Ahava Malla’s battle with D 1 Rajéndra Chola a little later than this, but in the same century; and IS they also claim the victory not for the Chéla, but for the Chalukya § king. This discrepancy, however, is not of any importance : for it is clear, from both sets of inscriptions, that Ré&jéndra Chéla lived about | the beginning of the eleventh century, and Kulotunga Chéla about the middle of it; and, in consequence, it appears to be certain that the publication of Kamban’s Ramayana, which professes to have been in — A.D. 733, has intentionally and mendaciously been ante-dated three hundred years. W This is not the proper place for attempting to furnish the reader e 0 h Iy E ESTIMATE OF DRAVIDIAN LITERATURE. 89 | with an estimate of the intrinsic value of Dravidian poetry. Whilst L an elevated thought, a natural, expressive description, a pithy, sen- ly tentious maxim, or a striking comparison, may sometimes be met with, unfortunately elegance of style, or an affected, obscure brevity, has 1 always been preferred to strength and truthfulness, and poetic fire has fo been quenched in an ocean of conceits. 4! Nothing can exceed the refined elegance and ¢linked sweet- by ness’ of many Telugu and Tamil poems ; but a lack of heart and 51§ purpose, and a substitution of sound for sense, more or less characterise ed them all : and hence, whilst an anthology composed of well-selected he extracts would please and surprise the English reader, every attempt to translate any Tamil or Telugu poem in extenso into English, has lor proved to be a failure. an To these causes of inferiority must be added a slavery to custom b and precedent at least equal to what we meet with in the later San- of scrit. Literature could never flourish where the following distich 0n (contained in the ‘Nan-nil,” or classical Tamil grammar) was accepted he as a settled principle :— Ira ¢ On whatsoever subjects, in whatsoever expressions, with whatsoever arrangement, th Classical writers have written, so to write is denoted propriety of style.’* hi her For the last hundred and fifty years the Dravidian mind appears to nd have sunk into a state of lethargy,—partly in consequence of the dis- her couraging effect of foreign domination, but chiefly through the natural 101 tendency to decay and death which is inherent in a system of slavery hat to the authority of great names. 8 8 With the exception of a small ethical poem, called the ‘Niti-neri- the vilakkam,” the only Tamil poems or treatises of any real value which have been written within the period mentioned, have been composed the lar * It is deserving of notice that alliteration is of the essence of Drividian )i ) poetry, as of Welsh ; and that the Drividians have as just a claim as the Welsh to the credit of the invention of rhyme. ith The rhyme of modern European poetry is supposed to have had a Welsh or Celtic origin; but Drividian rhyme was and necessarily invented by Drividians. The chief peculiarity of Drividian rhyme kya consists in its seat being, not at the end of the line, but at the beginning —a natural result of its origin in a love of alliteration. The rule in each Drividian f 18 dialect is that the consonant which intervenes between the first two vowels in a )ollt line is the seat of rhyme. A gingle Tamil illustration must suffice :— the sirei (t)tédil, érei (t)tédu.—AvveL the ¢ If you seek for prosperity, 1 Seek for a plough.’ 760 The agreement of those two consonants constitutes the minimum of rhyme which is admissible : but often the entire first foot of one line rhymes with the same foot in the second ; sometimes the second feet in each line also rhyme ; and u]el the rhyme is sometimes taken up again further on in the verse, according to fixed laws in each variety of metre. 90 INTRODUCTION. by European missionaries. At the head of compositions of this class, and high in the list of Tamil classics, stands the ¢ Tém-ba-vani,’ of Father Beschi. This long and highly elaborated scriptural epic possesses great poetical merit, and exhibits an astonishing command of the resources of the language : but unfortunately it is tinged with the fault of too close an adherence to the manner and style of ¢the ancients,” and is still more seriously marred by the error of endeavour- to Hinduize the facts and narratives of Holy Scripture, and even Scripture geography, for the purpose of pleasing the Hindu taste. Now that native education has commenced to make real progress, and the advantages of European knowledge, European civilisation, and European Christianity are becoming known and felt by so many of the Hindus themselves, it may be expected that the Dravidian mind will ere long be roused from its lethargy, and stimulated to enter upon a new and brighter career. If the national mind and heart were stirred to so great a degree a thousand years ago by the diffusion of Jainism, and some centuries later by the dissemination of the Saiva and Vaishnava doctrines, it is reasonable to expect still more important results from the propagation of the grand and soul-stirring truths of Christianity, and from the contact of the minds of the youth with the ever-progressive literature and science of the Christian nations of the West. It is a great and peculiar advantage of the English and vernacular education which so many Hindus are now receiving from European missionaries and from Government teachers, that it is communicated to all who wish to receive it, without distinction of caste. In former ages the education of the lower castes and classes was either prohibited or seduously discouraged, and female education was generally regarded as disgraceful ; but now the youth of the lower classes, of both sexes, are generally admitted to the same educational advantages as those The hitherto uncultivated that are enjoyed by the higher castes. minds of the lower and far most numerous classes of the Hindu community, are now for the first time in history brought within the range of humanising and elevating influences. A virgin soil is now for the first time being ploughed, turned up to the air and light, and sown with the seed of life ; and in process of time we may reasonably expect to reap a rich crop of intellectual and moral results. In the Appendix I have endeavoured to answer the question, ‘are the Pariahs and the Tudas Dravidians ¥’ I have also subjoined some S remarks ‘on the Dravidian physical type,” and ‘on the religion of the ancient Dravidian tribes. of ic of he he I en 5, nd of nd [} » 168 1§ on he COMPARATIVE GRAMMAR, e a1 al f0 ;es or a8 %) 56 ol du he W nd ly 10 1117 he NOTE. All foreign words, to whatever family of languages they may belong, are represented in this work in the Roman character, for the double purpose of preventing unnecessary expense and trouble, and of facilitating comparison. Long vowels are invariably marked thus,—‘4:" when no such accent is placed over a vowel, it is intended that it should be pro- nounced short. Bl (0 All vowels are pronounced in the Continental manner. il di The ‘cerebral’ consonants are denoted by a subscribed dot, e.g., Bl ‘t, d, n:’ the peculiar vocalic ‘r, and the surd ‘1’ of the South- to Indian languages are denoted in a similar manner, e.g., ‘r, 1:’ the ’ 9 obscure, inorganic nasal ‘n, or ‘m; is italicized, ¢g., ‘m, or ‘m. ? and the hard, rough ‘r, is represented by a capital ‘Rr.’ th A\t The dental ‘d,” in Tamil, and the corresponding ‘t,’ in Malayélam, are pronounced in the middle of a word, or between two vowels, like Vi the English ‘th,’ in tkan ; and in Telugu, ¢j’ and ‘ch,” when fol- ﬂh lowed by certain vowels, are pronounced like ‘dz’ and ‘ts: but as 01 these are merely peculiarities of pronunciation, and one consonant is S — (if not exchanged for another, no change has been made in the characters 1 by which those sounds are represented. 1LY In colloquial Telugu, a ‘y’ euphonic is generally written, as well ny QX] as pronounced, before ‘i’ and ‘e; and a similar ‘v’ before ‘o :’ but as this is merely a colloquial corruption, and one which tends to hinder far E thy e e comparison with other dialects, all such words will be written without the ‘y’ or ‘v, and it will be left to the reader to pronounce them as usage requires. This is the rule in Tamil, in which ‘evan,’ who ? is Y1z always pronounced, but never written, ‘yevan.’ 'the j; 1 dep ear) | P whi 12y SECTION 1. the d of uch SOUNDS. ro It will be my endeavour in this section to elucidate the laws of sound by which the Dravidian languages are characterized, and which contribute to determine the question of their affiliation. Special notice will be taken of those regular interchanges of sound in the different dialects which enable us to identify words under the various 6.0 shapes that they assume, and to which it will frequently be necessary the to allude in the subsequent sections of this work. ‘ th m DrAvipiaN ArLpHABETS.—Before entering on the examination of the Dravidian sounds, it is desirable to make some preliminary obser- vations on the alphabets of the Dravidian languages. Jam, There are three different Dravidian alphabets at present in use, fike viz., the Tamil, the Malayalam, and the Telugu-Canarese. I class fol- the Telugu and the Canarese characters together, as constituting but f a8 one alphabet; for though there are differences between them, those 1t 18 differences are few and unimportant. The Tulu is ordinarily written oferd in the Malayéla character: the Ku grammar of which I have made use, is written in the characters of the Uriya—characters which are much less appropriate than those of the Telugu would have been, for The other uncultivated dialects of this well expressing the Ku sounds. ' hub family have hitherto been content to have their sounds expressed in the Roman character. pder ot The three Dravidian alphabets which have been mentioned above, m a8 viz., the Tamil, the Malayalam, and the Telugu-Canarese, together with 1Bt} their older but now obsolete shapes, and the ‘Grantham,” or character in which Sanserit is written in the Tamil country, have all been derived, I conceive, from the early Déva-nagari, or from the still earlier characters that are contained in the cave inscriptions—characters which have been altered and disguised by natural and local in- 94 SOUNDS. al fluences, and especially by the custom, universal in the Dekhan, of of writing on the leaf of the palmyra palm with an iron stylus. It was supposed by Mr. Ellis, and the supposition has gained Wi currency, that before the immigration of the Brahmans into the Tamil al country, the ancient Tamilians were acquainted with the art of writing ; that the Brahmans recombined the Tamil characters which la 0 they found in use, adding a few which were necessary for the expres- sion of sounds peculiar to the Sanscrit; and that from this amalgama- ch tion, which they called ‘Grantham, or the book, the existing Tamil af characters have been derived. There cannot be any doubt of the by derivation of the Tamil character from the Grantham: for some characters are evidently identical with Grantham letters which are ! still in use; others with more ancient forms of the Grantham: but the other part of the hypothesis, viz., the existence of a Pre-Sanscrit 0! Tamil character out of which the Grantham itself was developed, is U8 a8 very doubtful ; and though it is true that there is a native Tamil word which signifies ‘a letter; yet there is no tradition extant of the eixstence of Tamil characters older than those which the first Brah- eq man immigrants introduced. The Indian characters referred to by ag] Iambulus, as quoted by Prinsep, evidently differed widely from the Vo ab Tamil, and appear to have been identical with, or allied to, ‘the cave character; and the character called Hala Kannada, or Old Canarese, al and the various characters in which Tamil is found to be written wh U§ in old inscriptions, are plainly founded on the basis of an alpha- betical system which was originally intended for the use of the Sanscrit. thy The modern Telugu-Canarese differs considerably from the modern va Tamil, and departs more widely than the Tamil from the Déva-nagari der type; but there is a marked resemblance between many of the Telugu-Canarese characters and the corresponding characters that are tior found in early Tamil inseriptions, such as the ‘Sasanas,’ or royal rep1 The modern | grauts, in the possession of the Jews of Cochin.* Malayéla character is manifestly derived from the Tamilian Grantham. Tel Thus, there is reason to conclude that all the alphabetical characters gt which are used or known in Southern India have a common origin ; Sys Sar * The Cochin inscriptions have been published and interpreted by the Rev. Dr. Gundert, in the Journal of the Madras Literary Society. They are the written in the Tamil language, though in an idiom which is tinged with the af peculiarities of the Malayalam. The character in which they are written, was once supposed to be peculiar to the Malaydla country : bul I have in my posses- o) sion many fac similes of inscriptions in the same character, which were obtained g in various districts of the Southern Tamil country, or Pindiyan kingdom ; and it would appear to have been the character which was most generally used at an carly period all over the South. DRAVIDIAN ALPHABETS. 95 of and that their origin is the same as that of all the existing alphabets of Northern India, namely, the system of characters in which Sauscrit ned was written by the ancient Brahmans. i The difference between the northern and the southern alphabets of arises from the antiquity of the literary cultivation of the southern lich languages, as compared with the northern. The southern languages 168 commenced to be cultivated in that early period when the cave ma character was used: the northern vernaculars were not cultivated till il after the cave character had become obsolete, and had been superseded the by the later Déva-nagari. ome The Telugu and the Canarese alphabets correspond to the Déva- are nagari in power and arrangement. The only difference is that a short the ‘e’ and ‘o, and a hard ‘r’ which is unknown to the Sanscrit, are serb contained in those alphabets, together with a surd ‘1, which is not used in the modern Sanscrit, but is found in the Sanscrit of the Védas, l, 18 ord as well as in the Dravidian languages. th 16 In other respects the characters of those alphabets are convertible 1ah equivalents of the Déva-nagari. The Malayéla alphabet generally ) by agrees with the Telugu-Canarese: it differs from them in having the vocalic ‘r,” of the Tamil, in addition to the other characters mentioned the cave above; and in having only one character for long and short ‘e, and another for long and short ‘o.” The aspirated letters and sibilants 1686 itten which all those alphabets have borrowed from the Sanscrit, are seldom lpha used except in pronouncing and writing Sanscrit derivatives. Those letters are not really required for native Dravidian purposes; ' the though, through the prevalence of Sanscrit influences, they have acquired a place in the pronunciation of a few words which are not pdern agarl derived from the Sanscrit. the The letters ‘ch’ and ¢j,” are pronounced in Telugu in certain situa- tions ‘ts’ and ‘dj:’ but no additional characters are employed to \ are royﬁl represent those sounds. The Tamil alphabet differs more widely than the Malayalam, or the Odern than Telugu-Canarese, from the arrangement of the Déva-nagari. The grammar of the Tamil language having, to a considerable degree, been aefers systematised and refined independently of Sanscrit influences, and ngils Sanscrit modes of pronunciation being almost unknown to Tamilians, hv ﬂ]e the phonetic system of the Tamil demanded, and has secured for itself, ey 10 a faithful expression in the Tamil alphabet. The materials of that ih the alphabet are wholly, or in the main, Old Sanserit ; but the use which is o], “ﬁb 0386 made of those materials is Tamilian. pained a0 ] The following are the principal peculiarities of the Tamil alphabet, b In common with the Telugu and Canarese alphabets, the Tamil i 96 SOUNDS. alphabet possesses separate characters for long and short ‘e, and for long and short ‘0.” Formerly it had but one character for the long and the short sounds of those vowels; and it is believed that the marks by which the long are now distinguished from the short were first introduced by Beschi. The Tamil has no characters corresponding to the liquid semi-vowels ‘ri® and ‘li,” which are classed amongst vowels by Sanscrit grammarians; and it has not adopted the ‘anusvéra,’ or 1l obscure nasal of the Sansecrit. Much use is made of nasals in Tamil: but those nasals are firm, decided sounds, not ‘echoes,’ and are classed amongst consonants by native grammarians. ‘m’ is the natural sound of the Tamil nasal, and this sound is uniformly retained at the end of words and before labials: when followed by a guttural, ‘m’ is changed into ‘ng,” the nasal of the guttural row of consonants; and it is changed 1t in a similar manner into ‘@ij,” ‘n,” or ‘n,” according as it is followed by % a palatal, a cerebral, or a dental. The Tamil alphabet has nothing to Bl correspond with the ‘half anuswara’ of the Telugu—a character and sound which is peculiar to that language: nevertheless, the tendency 00, th to euphonize hard consonants by prefixing and combining nasals, from {hy which the ‘half anuswAra’ has arisen, is in full operation in Tamil. The Tamil makes no use whatever of aspirates, and has not th alz borrowed any of the aspirated consonants of the Sanscrit, nor even the isolated aspirate ‘h. of I\ In arranging the consonants, the Tamil alphabet follows the Déva- b nigari in respect of the ¢vargas,” or rows, in which the Sanserit con- sonants are classified and arranged. Tt adopts, however, only the 1 d first and the last consonant of each row, omitting altogether the inter- mediate letters. In the first or guttural row, the Tamil alphabet Iy adopts ‘k,” and its corresponding nasal ‘ng,” omitting ‘kh,’ ‘g’ and ‘gh:’ in the second or palatal row it adopts ‘ch,” and its corresponding ofh nasal ‘nj,” omitting ‘chh,” ‘j,’ and ‘jh: in the third or cerebral row it adopts ‘t,” and its nasal ‘n,’ omitting ‘th,” ‘d,’ and ‘dh:’ in the fourth § or dental row it adopts ‘t,’ and its nasal ‘n,” omitting ‘th,’ ‘d,” and *dh?’ B in the fifth or labial row it adopts ‘p,” and its nasal ‘m,” omitting . “Bh,; *b..and ‘bh. Thus, the Tamil alphabet omits not only all the aspirated con- sonants of the Déva-nagari, but also all its soft or sonant letters. The sounds which are represented by the sonants of the Déva-nagari, are as commonly used in Tamil as in Sanscrit: but in accordance with a peculiar law of sound (to be explained hereafter) which requires the same letter to be pronounced as a surd in one position and as a sonant in another, the Tamil uses one and the same character for representing both sounds; and the character which has been adopted for this pur- DRAVIDIAN ALPHABETS. 97 or pose by the Tamil alphabet, is that which corresponds to the first consonant, viz., the tenuis or surd, in each of the Déva-nigari ‘vargas.’ ng ks In the ‘varga’ of the semi-vowels the Tamil follows the Déva- nagari ; but it subjoins to that ‘varga’ a row of four letters which are I8 to not contained in the Déva-nagari. Those letters are a deep liquid ‘r,’ ela which will always be represented in this work as ‘r; a harsh, rough or ‘r,” which will be represented as ‘r; ‘1, a peculiar surd ‘1, with a il mixture of ‘r;’ and ‘n,’ a letter to which it is unnecessary to affix any distinctive mark, the difference between it and the ‘n’ of the dental \ed nd ¢ varga’ being one of form rather than of sound. This peculiar ‘n’ is | of that which is invariably used as a final ; and it is also much used in combination with ‘r,’ to represent the peculiar Tamil sound of ¢ ndr.’ ool opd The Tamil alphabet is not only destitute of aspirated consonants, but it is also without the separate aspirate ‘h,” which hasa place in the alpha- by r 10 bets of so many other languages. It is destitute also of the Sanscrit sibilants ‘8, ‘sh,” and ‘s.’ The second and third of these sibilants are and occasionally used in pronouncing and writing Sanscrit derivatives; but ney rom these letters are never found in the ancient dialect of the Tamil, or in the classics, nor have they a place in the Tamil alphabet : when used, 108 they are borrowed from the Grantham, from which a few other letters also are occasionally borrowed to express Sanscrit sounds. The first the (,)f the three Sanscrit characters referred to above, namely, ¢the § of v iva,” is never used at all in pure Tamil : the Tamil palatal or semi- sibilant which corresponds to the Sanscrit ‘ch,” and which is pro- con: nounced as a soft ¢ S 7 the > or ‘sh,) when single, and as ‘ch,” when doubled, is the letter which is used instead. nter- The following comparative view of the ¢Déva-nigari’ and the abeb and Tamil alphabets exhibits the relation which the one bears to the other. ding VowELS. W lt Sanserit a, & ol u, risrl It : —8 :af : —06 :alt: n : ah )ul‘t»h Tamil — a, & i :—— e, & ef : o, 0 all 1 —: — bt fing CONSONANTS. Gutturals, Sans. k, kh gh Ditto, Tamil k, con Sans. chh The Palatals, ch, jh Ditto, Tamil ch, e Cerebrals, Sans. t, th dh ith 8 Tamil Ditto, o8 the Dental, Sans. th dh Ditto, Tamil onalt Labials, Sans. bh nting p, ph P e e Bt B Ditto, Tamil p, BEERBEBEEEE Pur 98 SOUNDS. CONSONANTS. Semi-vowels, Sans. NE T, 1, Ditto, Tamil Y T, 1, T, L, Sibilants and aspirate, Sans. g, sh, 8, Tamil DrAvIDIAN SysTEM OF Sounps.—We now proceed to inquire into the sounds of the Dravidian letters, and the laws of sound, or phonetic system, of this family of languages ; and in doing so, it will be found advantageous to adhere to the order and arrangement of the ¢ Déva- nagari’ alphabet. It is not my object to explain in detail the pro- nunciation of each letter: but such observations will be made on each vowel and consonant in succession as seem likely to throw light on I the principles and distinctive character of the Drividian system of sounds. Tamil grammarians designate vowels by a beautiful meta- phor, as ‘uyir,’ or the life of a word; consonants as ‘ mey,” or the body ; and the junction of a vowel and consonant as ¢ uyir mey,” or an animated body. a v 3 and ‘4.’ The sound of these vowels in the th I. Vowels.—(1.) ¢ DrAvidian languages corresponds to their sound in Sanscrit. In Tamil, 80 4 a is the heaviest of all the simple vowels, and therefore the most U liable to change, especially at the end of words. In the other dialects 8 it maintains its place more firmly ; but even in them it is ordinarily i strengthened at the end of words by the addition of the euphonic syllable ¢vu,’ consisting of the enunciative vowel ‘u,’ and the euphonic formative ¢ v ¢4 has almost entirely disappeared from the end of This rule nouns in Tamil, and has been succeeded by ‘u’ or ‘ei. 05 holds universally with respect to nouns singular. ‘When the Greeks ey visited India, ‘iru,’” @ fown, appears to have been invariably pronounced i ¢ Gira :’ it has now become in Telugu and Tamil either ¢{ru,’ or ‘dr,’ Te but remains ‘{ra, in Malayalam. Where final “a’ changes into ¢ei,’ il in Tamil, it generally changes into ‘e, in Canarese, or else it is tlo propped up by the addition of ¢vu. In Telugu, and especially in 1§ Malayalam, this vowel is less subject to change. Neuter plurals of 81 appellatives and pronouns, which originally ended in ‘a’ in all the Ca dialects, and which still end in ‘a, in MalayAlam, now end in most [4 2 ad instances in €l b in eolloquial Tamil, in ¢i,” in Telugu, and in ‘u,” in | Canarese. Thus, ‘ava,’ those (things), has become ‘avei,’ in Tamil; in; ¢avi, in Telugu ; ‘avu,” in Canarese : in Malayalam alone, it is still 2 of ‘ava In the same manner, the long final ‘4’ of Sanscrit feminine foif( VOWELS. 99 abstracts, becomes in Tamil ¢ei,—e.g., CALA 2 Sans., desire, Tamil, 4 A/ . asa, A2 asel ;> ¢Chitrd,” Sans., April—May, Tamil Sittirei.’ The same ¢ a becomes ‘e,’ in Canarese, e.g., ‘Gangh,’ the Ganges, is in Canarese ‘ Gange’ or ¢ Gange-yu.’ The diphthong into which final ‘a’ and ‘4’ are weakened in Tamil, is represented more properly as ‘ei’ than as ¢ 9 al The origination of the Tamil ‘ei’ from ‘a,” and the analogy of to the Sanscrit diphthong ¢ ai,” which is equivalent to ¢ 4i,’ might lead us e to regard the Tamil diphthong as “ai,” rather than ‘ei.’ It is curious, however, that though it originated from ¢a,” every trace of the sound nd Va of “a’ has disappeared. It is represented in Grantham by a double ‘e, and in Telugu-Canarese by a character which is compounded of 10 ‘e’ and ‘i it accords in sound also very nearly with the sound of 1ch CA? e or ‘ey, in Turkey. on It is also to be observed that the Tamil ‘ei, ¢A? is the equivalent of the ) of the Malayala accusative, and is the of ' ) ta ordinary representative of the final €, of Canarese substantives and verbal nouns. I conclude, therefore, that it is best represented by the the ¢ A2 diphthong ‘ei,” which corresponds to the €L of the Greeks. or 8.) 51 mnd “80 These vowels call for no remark. (39 ‘uand ¢4} In the Indo-European languages, and also in the the Semitic, the vowels ‘i’ and ‘04, are very decided, inflexible mil sounds, which admit of little or no interchange with other vowels, nost or euphonic softening. In the Dravidian languages, long i’ is leCtv sufficiently persistent ; bat short ‘u’ is of all vowels the weakest and iy lightest, and is largely used, especially at the end of words, for 10010 euphonic purposes, or as a help to enunciation. ol In grammatical, written Telugu, every word without exception d of must end in a vowel ; and if it has not naturally a vowel ending of its rule own, ‘u is to be suffixed to the last consonant. This rule applies eeks even to Sanscrit derivatives ; and the neuter abstracts ending in ‘m,’ 1nced which have been borrowed from the Sansecrit, must end in ‘m-u,” in “ury Telugu. Though this ‘u’ is always written, it is often dropped in pro- nuneciation. fel In modern Canarese a similar rule holds, with this addi- 15 tional development, that ‘u’ (or, with the euphonic copula ‘v,” ‘vu’) 1t I v is suffixed even to words that end in ‘a:’ e.g., compare the Tamil ls of “sila,’ few (things), and ¢ pala,” many (things), with the corresponding | the Can. ‘kela-vu’ and ¢pala-vu.’ The Tamil rule, with regard to the most addition of ‘u’ to words which end in a consonant, accords with the rule of the ancient Canarese. That rule is, that in words which end 1, 1 g il in any hard or surd consonant, »iz., in ‘k,’ ‘ch,’ ¢t t, or ¢p,’ (each 5 il of which is the leading consonant of a ‘varga’), or in the hard, rough ‘R,” which is peculiar to these languages, the hard consonant shall be qiaine followed by ‘u’ (as ‘q,’ by ‘sh’v,’ in Hebrew), in consequence of its H 2 100 SOUNDS. being impossible for Tamilian organs of speech to pronounce those letters without the help of a succeeding vowel. In most instances this [ enunciative ‘u’ is not merely short, but so very short that its quantity d is determined by grammarians to be equal only to a fourth of the Vi quantity of a long vowel. 2 The Malayalam uses invariably a short fi (a” in those connexions and for those purposes for which ‘u’ is used el in the other dialects. 8 It often happens (though it is not an invariable rule) that the final surd, to which enunciative ‘u’ or ‘a’ has been appended, is doubled, & apparently for the purpose of furnishing a fulerum for the support of the appended vowel. Thus, the Samscrit ¢ vak, speech, becomes in 1I Tamil ‘vak(k)-u; °ap,” water, becomes ‘ap(p)-u;’ and so in all similar cases. The rule is further extended in Tamil so as to apply to the If a ( final consonants of syllables, as well as to those of words. syllable, though in the middle of a word, terminates in one of the hard Sl consonants above-mentioned, and if the initial consonant of the suc- I§ ceeding syllable is one which cannot be assimilated to it, the final consonant is doubled, and ‘u’ is affixed. Thus, ‘advaita,” Sans., ! @ in-duality, becomes in Tamil ‘attuveida.’ The rule by which “d,’ when thus doubled, becomes ‘t,” will be explained hereafter. 2 1t In modern collogquial Tamil, ‘u is suffixed to almost every final bey consonant,—to the semi-vowels and nasals, as well as the surds; and even in the ancient or classical Tamil it is sometimes suffixed to final U ¢ ‘1) e.g., ‘Sol(1)-u,” speak, instead of simply ol.” The employment of ‘u,’ in the manner and for the purposes now mentioned, is obviously quite foreign to Indo-European usages. It is i f not derived from the Sansecrit, and is directly opposed to Sanscrit laws of sound. It will be termed the ¢enunciative u,’ and will generally €l o be separated off by a hyphen. il ‘e H ‘e A AT (€] A9 n (4.) 5 of The Dravidian languages possess, and largely employ the short sounds of the vowels ‘e’ and ‘o’ (epsilon and omicron), and have different characters for those sounds, for fror i.)ﬁ the purpose of distinguishing them from the corresponding long vowels. dlp] The Sanskrit is destitute of short ‘e’ and o. The entire absence a of those sounds from a language which attends so nicely as the San- ll(‘q serit to the minutest gradations of sound, cannot be the result of 1B accident ; and the important place which they occupy in the Dravidian Q(‘,u system of sounds, shows that the Drividian languages are independent of the Sanscrit. \In In a few cases, both in Telugu and in Tamil, particularly in the oy instance of the interrogative base ‘e,” the short vowel has sometimes ot Wit} VOWELS. 101 been corrupted into a long one, or lengthened by becoming the seat of emphasis ; but such cases are rare and exceptional, and in general the difference between short ‘e’ and ‘o, and the corresponding long vowels, is a difference which pertains, not to the euphony or inflexional form, but to the bases or roots of words, and is essential to the differ- ence in the signification. Z.g., in Tamil, ‘té&l,” means clear, and tél,’ scorpron ; ‘kil,) stone, and “kal,’ foot. (5.) ‘el It has already been mentioned that ‘ei,” unlike the I Sanscrit diphthong ¢ai,’ is derived from ‘e’ and ‘i, not from ‘a’ and 2 b The primitive Dravidian ‘a’ changes into ‘e,” and this again into ‘el ’ \r Thus, the head, is ‘tala,” in Telugu and Malayilam; tale,” in 16 Canarese ; and ‘talei,” in Tamil. rd When ‘ei’ is succeeded in Tamil by another ‘ei,” with only a single consonant between them, the first ‘ei, though naturally long, 0 is considered short by position, and is pronounced short accordingly ; al In such e.g., ‘udeimei,” property, is regarded in prosody as ‘udeimei.’ bk cases ‘ei’ is an equivalent to its original ‘4’ or ‘& dw (6.) ‘aun.’ This diphthong has a place in the Tamil alphabet; but al it is not really a part of any of the Dravidian languages, and it has been placed in the alphabets solely in imitation of the Sanscrit. It is md used only in the pronunciation of Sanscrit derivatives; and when such nal derivatives are used in Tamil, they are more commonly pronounced without the aid of this diphthong. Ordinarilythe diphthong is separated 10\’V into its component elements : that is, the simple vowels “a’ and ‘u,’ ) from which it is derived, are pronounced separately, with the usual s euphonic ‘v’ of the Tamil between them to prevent hiatus. Z.g., the lly Sanserit noun ‘saukhyam,” health, is ordinarily pronounced and written in Tamil, ‘savukkiyam.’ and It is a peculiarity of the Tamil system of sounds, as distinguished lon from that of the other languages of the family, that the vowels ‘i, i W By 1 9 H for ‘e fe and ‘u, acquire before certain consonants a compound, Jong diphthongal sound, which is different from the sound which they have as simple vowels. Thub, 1" before 4, ‘n, ‘x’ fr' Cn Ol ale] .) 6/\, e acquires something of the sound of ‘e . 1 before the same conso- San- nants, with the exception of the first ‘r’ and the first <1, takes a It of sound resembling ¢t " ‘Qi” remains always unchanged; but ‘i, not idian only before the above-mentioned seven consonants, but before all 1dellﬁ single consonants, when it is not succeeded by “i,” ‘u 2 or ‘e, is pro- nounced nearly like ‘o ; and in Telugu, ‘o’ is generally used in ) fhe writing those words. ‘e, before the consonants above mentioned, fmes with the exception of the semi-vowels, loses its peculiarly slender 102 SOUNDS. sound, and is pronounced nearly as it would be if the succeeding con- bo th sonant were doubled. ¢¢, with the same exceptions, acquires a sound similar to 0. n The circumstance which is most worthy of notice, in connection D with these changes, is that each of the short vowels ‘1, ¢ u,” and ‘e, retains its natural sound, if it is succeeded by another ¢i,’ ‘u,” or e. de Thus, ura, Tamil, infinitive, o have, to be, is pronounced ‘oRa,” but Ta the imperative ‘uru’ is pronounced as it is written. ] This rule discloses a law of sound which is unlike anything that is aly thy discoverable in Sanserit. So far as it goes, it is evidently connected 00 with the Scythian law of harmonic sequences, which will be referred to hereafter. “d C01 The vowel ‘3, occurring in the last syllable of a word ending in wh ‘p) ‘m’ ‘r’ ‘r) ‘1, or ‘1’ acquires a slender sound resembling that of ¢; eg., “avar, Tamil, they, (honorifically, /e) is pronounced ¢ aver.’ lie wh This change corresponds to the weakening of the sound of heavy Con vowels, in the ultimate or penultimate syllables of words, which is sometimes observed in the Sanserit family of tongues. a00( i I1. Consonants.—Tamil grammarians divide all consonants into iy three classes :—(1.) Surds, which they call ‘vallinam,” or the hard class, (00§ viz., ‘k, ‘ch, ‘t’ ‘t’ ‘p, ‘R’ (2.) Nasals, which they call ¢ melli- dept nam, or the soft class, viz., ¢ ng, &), fm, fn,) ‘m, with final S8 L and (3.) semi-vowels, which they call ¢ ideiyinam,” or the medial class, fo gy et v, St fo In this enumeration, as I have already observed, the sonant equi- ke valents of the surd consonants (viz., ‘g’ the sonant of ‘k’; ¢§,’ the the sonant of “ch ; ¢d,” the sonant of ‘t; “d,’ the sonant of ‘t;’ and Fhy tog, the sonant of ¢p’) are omitted. In the other Dravidian dialects the (ase difference between surds and sonants is generally expressed by the use ﬂenUr of different characters for each sound, in imitation of the system of the g | Déva-négari ; but in Tamil, and in part in Malayalam, in accordance be B with the peculiar Dravidian law of the convertibility of surds and the sonants, one set of consonants serves for both purposes, and the differ- ence between them is expressed in the pronunciation alone. the i It is desirable before proceeding further to enquire into this law, viz.: ) The Convertibility of Surds and Sonants.—We have seen that the an| Tamil alphabet adopts the first and last of each of the DevA-nagari ang ‘yargas, OT TOWS of consonants, viz., the un-aspirated surd and the nasal 1t 8 of each ¢varga; we have also seen that the Tamil has not separate Ingty, characters for surds and sonants, but uses one and the same character— Operg that which, properly speaking, represents the surd only—to express Int he CONSONANTS. 103 both. This rule does not apply merely to the written characters of the language, but is the expression of a law of sound which is inherent in the language itself. There are distinct traces of the existence of this law in all the Dravidian dialects ; but it is found most systematically and most fully developed in Tamil, next in Malayilam. The law, as apparent in the Tamil system of sounds, is as follows : ‘k, “t,” < t,” ‘p,” the first, un- aspirated consonants of the first, third, fourth, and fifth ¢ vargas, are always pronounced as tenues or surds (i.e., as ‘k,) ‘t’ ¢t ‘p,’) at the beginning of words, and whenever they are doubled. The same | consonants are always pronounced as medials or sonants (i.e., as ‘g 9 €d, ¢d, ‘b,’) when single, in the middle of words. A sonant cannot 1 commence a word, neither is a surd admissible in the middle, except b when doubled ; and so imperative is this law, and so strietly is it ad- hered to in Tamil, that when words are borrowed from languages in which a different principle prevails, as the Sanscrit or the English, the y 15 consonants of those words change from sonants to surds, or vice versd, according to their position : e.g., ¢ dantam,” Sanserit, a tooth, becomes in Tamil, ¢ tandam ;> ‘bhigyam, Sanscrit, Aappiness, becomes phk- 0 kiyam.” This rule applies also to the case of compounds. The first S consonant of the second word, though it was a surd when it stood in- i dependent, is regarded as a sonant when it becomes a medial letter in a compounded word. This difference is marked in Telugu by a dif- %, ference in the character which is employed ; e.g., ‘anna-dammulu’ (for ¢ anna-tammulu’), elder and younger brothers ; ¢ kotta-badu’ (for l “kotta padu’), to be beaten ; but in Tamil, and generally in Malayélam, he the difference appears in the pronunciation alone. This rule applies b to all compounds in Telugu: but in Tamil, when the words stand in a [ he case relation to one another, or when the first is governed by the 186 second, the initial surd of the second word is not softened, but doubled the and hardened, in token of its activity; e.g., instead of ¢ kotta-badu,’ to 106 be beaten, it prefers to say, ¢ kotta-(p)padu.’ In ‘dwanda’ compounds nd the Tamil agrees with the Telugu. fer- A similar rule applies to the pronunciation of ¢ ch’ (the Tamil ¢ g’), the first consonant of the second ¢varga.’ When single it is pro- /1L nounced as a soft, weak sibilant, with a sound midway between ¢sh’ he and ‘ch” This pronunciation is unchanged in the middle of words, and in all cases in which the letter is single ; but when it is doubled it is pronounced exactly like ‘ch.’ The principle involved in this sl rate instance is the same as in the cases previously mentioned ; but the operation of the rule is in some degree different. The difference consists g in the pronunciation of this consonant in the beginning of a word, as p1'655 104 L4 SOUNDS. Zey well as in the middle, as a sonant, ¢.e., as ¢ 8 By theory it should be pronounced as ‘ch’ at the beginning of a word,—and it is worthy of notice that it always receives this pronunciation at the beginning of a word in vulgar colloquial Tamil; and in Telugu it is written as well I as pronounced ‘ch.’ A somewhat similar rule prevails with respect to the rough ‘®’ of the Tamil, which is pronounced as ‘®r’ when f single, and like ‘ttr’ when doubled. Iy The Tamilian rule which requires the same consonant to be pro- nounced as ¢ k’ in one position and as ¢ g’ in another—as ‘t,’ ‘t, ‘p,’ 3 in one position, and as ‘d,” ‘d,” ‘b,” in another—is not a mere dialectic peculiarity, the gradual result of circumstances, or a modern refine- ment invented by grammarians ; but is essentially inherent in the lan- guage, and has been a characteristic principle of it from the beginning. b The Tamil characters were borrowed from the earlier Sanscrit, and ta the language of the Tamilians was committed to writing on, or soon it dg after, the arrival of the first colony of Brahmans, probably more than six centuries before the Christian era. Yet even at that early period the Tamil alphabet was arranged, not in accordance with Sanscrit laws n of sound, but in such a manner as to embody the peculiar Drividian law of the convertibility of surds and sonants. The Tamil alphabet (oL systematically passed by the sonants of the Sanscrit, and adopted the surds alone, considering one character as sufficient for the expression Sar el of both classes of sounds. This circumstance clearly proves that ab Its initio the Dravidian phonetic system, as represented in the Tamil, its most ancient exponent, differed essentially from that of the €X 1 Sanscrit, * 1t ) In none of the Indo-European languages do we find surds and sonants convertible ; though Hebrew scholars will remember the y existence in Hebrew of a rule which is somewhat similar to the Ta- The Hebrew milian respecting ‘k,” ‘t,’ ‘p,” and their equivalents. that consonants composing the memorial words, ¢begad kephath,” are pro- o nounced in two different ways, according to their position. When any of T those consonants begins a word, or in certain cases a syllable, it is to be 0r g pronounced hard, that is, as asurd or tenuis; and if it be an aspirated Inf letter, it is then deprived of the aspirate which it naturally possesses. It i To denote this, such consonants have a point, called a ¢ dagesh,” insecri- 0ry bed in them. When those consonants are found in any other position frop ])Va * See also the evidence which is furnished in the Introduction respecting the existence of this law of the convertibility of surds and sonants in the names of places in Southern India that are recorded by the Greek geographers; e.g., Prog Cottora (Kottar), where the ‘d’ of ‘kdd,” the first part of the compound, being doubled, has become ‘t. ]HP ] ]n “’I CONSONANTS 105 be they are pronounced as sonants, and two of them ‘ph’ and ‘th, as of aspirates. fa This rule resembles the Tamilian in some particnlars ; but the vell resemblance which will be found to exist between the Tamilian rule 6ct and the law of sounds which prevails in some of the languages of hen the Scythian family, amounts to identity In the Finnish and Lappish there is a clearly marked distinction )10 between surds and sonants a sonant mever commences a word or tp, syllable in either tongue. But in the oldest specimen of any Scythian ofie language which is extant—the Scythic version of the inscription at ne- Behistun—Mr. Norris has ascertained (Journal of the Royal Asiatic Jan Society for 1853) the existence of a law of convertibility of sonants and ng surds which is absolutely identical with the Tamilian He has ascer- and tained that in that language, in the middle of a word, the same conso- S00L nant was pronounced as a sonant when single and as a surd when doubled thian riod We now enter upon an examination of the Drividian consonants Jaws in detail dian (1) The guttural ‘varga : ‘k, ¢ g, and their nasal, ‘ng.’—These ahet cousonants are pronounced in the Drividian language precisely as in ] the Sanserit. 85101 ‘g, the sonant of ‘k,’ which is expressed by the same cha- racter in Tamil, is pronounced in Tamil in a peculiarly soft manner it 00 Tts sound resembles that of the Irish ‘gh,’ and is commonly used to il express the ‘h of other languages Thus, the Sanscrit adjective ' the ‘mahd,” great, is written in Tamil ¢magi but so soft is the ¢g’ that it may be considered as an equivalent to ‘h,” pronounced with less ) and roughness than is usual with that aspirate he o Ta (2.) The palatal ‘varga: ¢cl J, and ‘fij’—It has been observed that the Tamil rejects the Sanscrit sibilants ¢ S, 2 sh,” and ‘g The ohrew consonant which it adopts instead is ch,” which is pronounced in p 1)10 v of Tamil in a manner somewhat similar to the soft aspirated ‘§ of Siva, s tobe or as a very soft ‘sh,” with as little sibilation or aspiration as possible In fact, it should be regarded as a palatal, not asa sibilant and when M%Lt@d 38565 it is doubled it takes precisely the sound of the Sanscrit palatal ¢ch,’ or its English equivalent in ¢ which.’ To distinguish the Tamil letter HPCH f00 from the sibilant ¢s’ of the Sanscrit, it will be denoted, when single, by an accent, thus—¢§.’ In Telugu the sound of ‘ch’ is that with which this consonant is g the pronounced, not only when doubled, but also when single ; and a simi- poés lar pronunciation prevails in the lowest colloquial dialect of the Tamil g; 0 e 13 in which sey to do, is prounounced chey,” asin Telugu 106 SOUNDS. ¢j,” the second un-aspirated consonant of this row, is not used in 99 ‘I'T mads correct Tamil ; but in Telugu it is both written and pronounced : in The same inste vulgar Tamil also ‘ch’ is sometimes pronounced like ¢j.’ with sound of ¢j’ is sometimes admitted in the use of those Sanscrit deriva- ( nan tives in which the letter ‘j’ is found in Sanscrit ; but ordinarily the Tamil sound of ‘ch,” or ‘&’ is used instead. Saus, Tam; ‘1ij,’ the nasal of this row of consonants, is pronounced as in elrenl Sanserit, in all the Dravidian languages. It is frequently used in 2 Malayalam as an initial where the Tamil uses ‘n, ey, Cijan, diom instead of the Tamil ‘nin. Junct It is necessary here to notice the existence in Telugu of a pecu- 18 oy Jiang. liarly soft pronunciation of ‘ch’ and ¢j,” with their aspirates, which is I unknown in Sanscrit and the northern vernaculars, and is found only in Telugu and in Marathi. ¢ch’ is pronounced as ¢ ts,” and ¢’ as ‘dz,’ Sonan g N e 1 Before these ex- the ¢ before all vowels except ‘i, e, ‘&’ and ‘el.’ cepted vowels, the ordinary sounds of ‘ch’ and ¢j’ are retained. Whe- the D ther the Telugu borrowed these sounds from the Marathi, or the Marathi (4 from the Telugu, I will not venture to express an opinion ; but this is ‘varga not the only particular in which those languages are found to agree. The o in Ty (8.) The cerebral ‘warga :* “t, ¢d, ‘n.’—The pronunciation of the 10t g consonants of the cerebral ¢ varga’ in the Drividian languages does not differ from their pronunciation in Sanscrit. In expressing these con- oy y hove sonants, with their aspirates, in Roman characters in this work, a dot i ) will be placed under each, to distinguish them from the ¢t,” ¢d,” and 2 QHS}’ ] (n’ of the dental row. Though “t’ is the surd consonant of the cerebrals, it is not pro- the }f nounced at the beginning of any word in Tamil, like the other surds. t0 e ¢ Its sound is too hard and rough to admit of its use as an initial ; and therefore, in those few Sanserit derivatives which commence with this 0 Whe fd.! ( letter, ‘¢’ is preceded in Tamil by the vowel ‘i, as a help to enunciation, and a3 When ‘t’ is thus proceeded by a vowel, it is no longer an initial, and of fhe therefore no longer a surd ; and hence it becomes ‘d’ by rule: so that ‘da’ €y the sound of ‘t’ is never heard in Tamil, except when ‘d’ is doubled. illitia]’ In the other Dravidian dialects ‘t’ is sometimes pronourced singly, as in Sansecrit. (5 The Tamil differs from the other dialects in refusing to combine nd it ‘t’ with ‘n,” and changing it into ‘d” when ‘n’ is combined with it. ‘Hl’ in This peculiarity is founded upon a general Tamilian law of sound, ' which is that nasals will not combine with surds, but coalesce with alata) sonants alone. In consequence of this peculiar law, such combinations “augy, Wun y, CONSONANTS. 107 | in as ‘nt,” ‘nt,” and ‘mp,” which are admissible in Telugu and Caf)arese, are 1 inadmissible in Tamil; in which ‘nd,” ‘nd,” and ‘mb,” must be used Ame instead. This rule applies also to ‘K’ and “ch;’ which, when combined 1V2: with the nasals corresponding to them, become ‘g’ and ¢ 3’ Thus, the ‘mantapam,” Sans., @ porch, becomes in Tamil ¢ mandabam ;* ‘antam,’ Sans., end, becomes ‘andam.’ Probably the difference between the B Tamil and the other Dravidian languages in this point, arises from the | m circumstance that the Tamil has remained so much freer than its sister 'l idioms from Sanscrit influences. A similar rule respecting the con- junction of nasals with sonants alone, is found in the Finnish; and ecl- is possibly owing to that delicacy of ear which both Finns and Tami- chis lians appear to possess. I reserve to the close of this examination of the Dravidian con- only ( (lZ,’ sonants, some observations on the circumstance that the consonants of j ex-‘ the cerebral class are found in Sanserit as well as in the languages of Vhe the Dravidian family. fthi (4.) The dental ‘varga ? “t) ¢d,” ‘n.’—The letters of the dental 18 18 ‘varga’ have the same sound in the Dravidian languages as in Sanscrit. ce. The only exception consists in the peculiarly soft pronunciation of ¢t,’ in Tamil and Malayalam, when used as a sonant: it is then pronounced »f the not as ‘d,” but with the sound of the soft English ¢th,” in ¢that.” Tt is s 1100 only when it is combined with a nasal (as in the word which was cited ({11} above ‘andam, end,) that the sonant of ‘t’ is pronounced in Tamil as 2 dot ‘d; the sound of ‘d’ being, in such a conjunction, more natural and ' and easy than that of ‘th.’ As this peculiar sound of ‘th’ is found only in Tamil, and in the Malayalam, a daughter of the Tamil, it is doubtful whether ¢ th’ is i pro- ard o} to be considered as the original sound of the sonant equivalent of ¢ t,’ ; and or whether it is to be regarded as a corruption or further softening of j this ad On the whole the latter supposition seems the more probable: atiOﬂ. and as the ‘th’ of the Tamil corresponds to the ‘d’ of the Telugu and of the other dialects, in position and power, I will always write it as , and ) that ‘d,’ even when quoting Tamil words, except where it is used as an initial, and is therefore a surd, when it will be written as ‘t.’ [Ibledo ;ly’ a8 (5.) The labial ‘varga: ‘p, ‘b, ‘m.—The pronunciation of s and its sonant ‘b,’ requires no remark. With regard to the use of mbill(’/ ‘m’ in combination, I have only to observe that though it changes into 10 ‘ng,” ‘Qij,’ ‘n, or ‘n,’ when immediately succeeded by a guttural, a souﬂd) palatal, a cerebral, or a dental, it is not to be confounded with the ‘anusvara’ of the Sanserit alphabet. The true ‘anusvéra, i.c., the qtions sound which ‘m’ takes in Sanscrit before the semi-vowels, the sibilants, 108 SOUNDS. 11SES and the letter ‘h,’” is unknown to the Drividian languages. A cha- subs racter called by the name of ‘anusvara,’ but of a different power from at § the ‘anusvéra’ of the Sanscrit, is in use in Telugu and Canarese; but ‘kir it is used merely as the equivalent of the consonantal ‘m,” in euphonic We ¢ combinations, and even as a final. The Telugu has also a vocalic the nasal, the half ‘anusvara,” which, though it is used merely for euphony, softe bears a close resemblance to the true ‘anusvara’ of the Sanserit. ‘Kt There is nothing in any of the Dravidian languages which corresponds ( to the use of the obscure nasal ‘anuswéra’ as a final, in Hindi and in the the other northern vernaculars. styl The euphonic use of ‘m’ and its modifications, and also the use of Sour ‘n’ and its equivalents, to prevent Aiatus, will be cousidered at the Sonar close of this section. T Sane (6.) The ‘varga’ of the semi-vowels: ‘y, ‘x,; €1} ‘v ‘)’ €1} ‘R’— Itis: In classical Tamil neither ‘r nor ‘1’ can commence a word: each of it; b them requires to be preceded by an euphonic auxiliary vowel; ‘1’ by Iny “toand by n, Thus, the Sanscrit ¢ra,’ night, abbreviated from evider ¢rAtri, is written and pronounced ¢ ird; and this again is softened into ‘irdvu.’ In like manner ¢ l6kam,” Sans., the world, becomes ulogam,’ thr(JUg and by a further corruption ¢uligu’ The same rule applies to the xegpt ever, second set of semi-vowels, ‘r,” ‘L’ ‘r,’ which are the exclusive property 2Ppr of the Dravidian languages, and none of w:hich can be pronounced the 5 without the help of preceding vowels. emblg Of these distinctively Dravidian semi-vowels, ‘1’ is found in the I Tamil alone. Its sound resembles that of the English ‘x’ after a long which vowel, as in the word ‘farm; but it is pronounced farther back in the mouth, and in a still more liquid manner. It is sometimes expressed in ( (i Rp’ English books as ‘zh,” or ‘rzh; but this is merely a local pronuncia- tion of the letter, which is peculiar to the northern district of the T of ¢ Tamil country: it is at variance with its affinities and its inter- Wﬂ'y 1l Wt changes, and is likely to mislead the learner. ‘r’ is the only Dra- vidian consonant which is pronounced differently in different districts. Ei&ﬁon souydy In the southern districts of the Tamil country, it is pronounced by the i, mass of the people, exactly in the same manner as ‘1’ which is the letter invariably used instead of ‘r’ in Canarese. Between Tanjore (i) Tam) and Pondicherry, it is softened into ‘rzh,” or ‘zh; and in Madras and the neighbourhood, this softening process has been carried to such a Prefiy, length, that in the speech of the wvulgar, ‘r’ has become a silent COmpou letter. ike { K@ h( [l The Telugu, which commences to be spoken about two days’ 8 Of fol) OWeq journey north of Madras, has lost this letter altogether. Gen(?rally it CONSONANTS, 109 cha uses ‘d’ instead, as the Canarese uses ‘1; but sometimes it uses no substitute, after the manner of the vulgar Tamil of Madras. Looking from but at such Telugu words as ‘kinda,’ below, answering to the Tamil 00i¢. ‘kirnda,’ and ‘vingu,’ #o swallow, answering to the Tamil ‘ virungu,’ we cannot but suppose, that the Telugn had this letter originally, like calie 00}, the Tamil, and that it lost it gradually through the operation of that softening process which, in the colloquial Tamil of Madras, converts serit. AA Y ‘kiré,” below, to ¢ kié yonds id ‘I’ is a peculiar heavy ‘1, with a mixture of ¢ r, which is found in the Védic Sanscrit, as well as in the Dravidian languages. Tt may be 56 0f styled the cerebral ¢1; and it is probably derived from the same source, whatever that source may be, from which the cerebral con- it the sonants ‘t,’ ‘d,” and ‘n,” have proceeded, The hard rough ‘r’ of the Dravidian languages is not found in Sanserit, and is not employed in pronouncing Sanscrit derivatives. p— ch of It is found in Telugu poetry, and the grammarians insist upon using it; but in the modern dialect of the Telugu it has fallen into disuse. ( 1,’ by In Canarese also, the use of this letter is confined to the poets. It is from d into evident that it was originally contained in all the dialects; though, through the influence of the Sanscrit, it has now ceased to be used fany to the except in the Tamil and MalayAlam, in which it has as firm footing as ever. In some of the older Tamil alphabets I have found this letter operty appropriately expressed by a double ‘r; and to distinguish it from punced the softer letter, it will be represented in this work by a capital ‘wr,’ emblematical of its greater strength, in the In the use of this hard ‘®’ in Tamil, there are two peculiarities 2 Jong which are worthy of notice. § iﬂ the (1) ‘®, when doubled, is pronounced as ‘ttr, though written g:‘SGd in ‘RR The ‘t’ of this compound sound differs both from the soft dental uncit- ‘t" of the fourth ‘varga,” and from the cerebral ‘t, and corresponds of the very mnearly to the emphatic final “t of our English interrogative g inter- ‘what? This sound of ‘t” is not expressed in writing, but in pronun- 1}! Dl’é‘ ciation it is never omitted; and it is one of those peculiar Dravidian iricts: sounds which are not derived from the Sanscrit, and are not found in it. by the ) 18 the (ii.) The letter ‘n’ (not the dental ‘n,’ but the final ‘n’ of the Tap]0re Tamil), a letter which is not found in the Telugu or Canarese, is often I g8 Al prefixed in Tamil to the rough ‘R’ for the sake of euphony ; when the 0 ach @ compound ‘nr’ acquires the sound of ‘ndr'—a sound of which the Tamil, Jent g 8 like the language of Madagascar, is exceedingly fond. In another class of words, the “n’ which is prefixed to »’ is radical, and should be 7o days followed by ¢d,” according to rule (e.g., in the preterites of verbs whose Jly ) e 110 SOUNDS 2 root ends-in ‘n’); but ‘R’ is suffixed to ‘n instead of ‘d,” in eon- The Wi sequence of which the sound of ‘ndr’ is substituted for that of ‘nd. The ‘R’ is radical, and the ‘n’ euphonically prefixed, in ‘minru’ (mlndru), Tam., three, (for ‘miru,” Can., the more ancient form of the §0n4 othe word), and in ¢ onru,’ (ondru), Tam., one, (for ‘oru.’) The ‘n’ is radical (or an euphonised form of the radical), and the ‘R’ is used euphonically Larl instead of ¢d, in the following examples; ‘enru’ (endru), having spoken, instead of ‘endu; ‘$enru’ (Sendru), kawving gone, for ¢ Sendu,’ (which is instead of the less euphonic ‘&eldu.”) In the speech of the and Dré vulgar in the Tamil country, and in the Malayalam, this compound ‘ndr,” is further altered into ‘nn’ or ‘nn.’ In Telugu and Canarese (d}’ mazj ‘nd’ is always found instead of ‘ndr.’ 1ot €& (7.) The sibilants and the aspirate S ‘sh, g’ ‘h”’—It has 0bser The fami] already been mentioned that the Tamil is destitute of sibilants. other DrAvidian idioms freely use the sibilants and aspirates of the the T Sanscrit, in writing and pronouncing Sanscrit derivatives, and to some I extent, through the prevalence of Sanscrit influences, in the pronun- 0ng ¢ / ciation even of pure Drividian words. In Tamil ‘the S of Siva,’ the of and occurring in Sanserit derivatives, is represented by the peculiar palatal which answers to the ¢ ch’ of the Sanserit, and the sound of which, when Tl ¢ &2 S. single, closely resembles that of The other sibilants, ‘sh,” and *s, Indig are altogether excluded from pure classical Tamil. In later Tamil from books, and in the speech and letters of the better educated Tamilians have beey of the present age, those sibilants are freely employed in writing and pronouncing words which have been borrowed from the Sanserit; and in guﬂgt such cases, the characters which are used to express them are taken from bomy the Grantham. By the mass of the people, however, those letters are artiy rarely pronounced aright; and in the remoter districts the vulgar Iead i substitute for them, in accordance with the genius of the language, (L those letters which the ancient grammars enjoin, and the use of which laggg is exemplified in the Sanscrit derivatives that are employed in the E8Pecg Tamil classics. The substitutions are as follows :—¢gh,’ the cerebral WhETE‘&: sibilant of the Sanscrit is represented in general by the cerebral ‘d; instead sometimes by the liquid ‘r; sometimes even by the dental ‘t’ or ¢d.’ Of the | ¢s,” the sharp sibilant of the Sanscrit, is sometimes represented by ‘t; @) sometimes it is omitted altogether; sometimes it is changed into the in any Tamil ¢ch,’ the equivalent of ‘8. When this sibilant stands at the Th & ¢ beginning of a Sanscrit derivative, and when it is desired in aceordance with modern usage, to pronounce it with the unmodified Sanserit sound, Celtic, 2 they ax it is preceded (at least in pronunciation) by the vowel ‘i, without Magey which it cannot be enunciated, in that connexion, by Tamil organs. a‘IldS 0) CONSONANTS. 38 ¢ 1 €01~ Thus, strl) Sans., @ woman, is always pronounced and generally nd, written ¢istiri.’ iRy’ The Tamil is destitute of the sound of ¢h,” and of aspirated con- of the sonants, as well as of sibilants. Aspirates are plentifully used in the adical other dialects of the Drividian family ; and in Canarese, ‘h’ is regu- > nically P larly used as a substitute for ¢ p iy endu, OriciN oF THE CEREBRAL CoNsoNANTs.—In all the languages of the and dialects of India, whether they belong to the Sanscrit or to the npound Dravidian families, much use is made of a series of consonants—¢ 1, \arese ¢d,” with their aspirates, and ‘n’—which are called by Hindu gram- marians ¢ cerebrals,” because they are pronounced far back in the mouth, with a hard, ringing sound. I have reserved to this place some Tt has observations on the existence of this peculiar class of sounds in two ) The families of tongues which are so widely different from one another MI&S of the the Dravidian and the Sanscrit. ) Some It seems natural to suppose, and it will readily be admitted, that pl‘()}lllﬁ' one of those families must have borrowed the sounds in question from f SiVﬂ‘a‘ the other ; but it remains to be determined which was the borrower, ' palatal and which was the original proprietor. The Hindi, the Bengali, and the other vernaculars of Northern h, when and S:’ India may be conceived to have borrowed the cerebral consonants r Tamil from the Sanscrit, from the decomposition of which those languages qmilians have mainly arisen: but it is very difficult to suppose that they have fing ad been borrowed in this manner from the Sanscrit by the Dravidian lan- guages. On the contrary, I have long been persuaded that they were lt, ﬁl]dill ken from borrowed from the Drividian languages by the Sanserit, after the arrival of the Sanscrit-speaking race in India. The reasons which thers 4t ) Vlllgfl‘f lead me to adopt this view are these :— anguigd (J.) The cerebral consonants are essential component elements of a large number of primitive Drividian roots, and are often necessary, Of whlch 0 n the especially in Tamil, for the discrimination of one root from another g whereas in most cases in Sanscrit, the use of cerebral consonants cerehnal instead of dentals, and especially the use of the cerebral ¢ n,” instead 0 ohral dj of the dental ‘n,’ is merely euphonic. ( t o | d.i (2.) None of the cerebral consonants has ever been discovered | into B¢ in any of the primitive languages which are related to the Sanserit, They are not found in the Classical languages, the Gothic, or the ds at ﬂle Celtic, in the Lithuanian, the Slavonian, or the modern Persian: 00 Ol'dﬂ'uce they are not found in the Cuneiform Persian, or the Zend—those lan- Cl'it goundy nuages, or rather sister dialects, with which the Sanserit finally shook ; gifhont hands on crossing the Indus and settling in Arya-vartta. On the other il o 112 SOUNDS. Dré hand, the DréAvidian languages, which claim to have had an origin independent of the Sanserit, and which appear to have been spoken here throughout India prior to the arrival of the Brahmans, possess the o tl cerebral sounds in question, and for aught that appears, were in posses- lang sion of them always. They are found even in the Brahui. There is fran no trace of these sounds in the Aryan family of tongues, west of the adop Indus: but no sooner does a member of that family cross the Indus, fhe § and obtain a lodgment in the ancient seats of the Dravidians and other Teig Seythians in India, than the cerebral sounds make their appearance in who fall their language. It is worthy of notice also, that the Pracrit, a local dialect or vernacular of the Sanscrit, makes a larger use of the cere- { brals than the Sanscrit itself.® heen (3.) Those consonants which the Tamil has borrowed from the Sans- treat crit within the period of the existence of Dravidian literature, have been fact 1 greatly modified to accord with the Tamilian laws of sound and delicacy ra}ly of ear. Thus, the Tamil omits the aspirates even of Sanscrit deri- the 3 vatives, and omits or changes all the sibilants. It systematically them softens down all harsh sounds. Even the Sanscrit cerebral-sibilant the d¢ ‘sh’ cannot be pronounced by Tamil organs. Hence it seems impro- indug bable that a series of harsh, ringing sounds, like the cerebral ¢t <d; him, and ‘n, should have been borrowed by the Tamil from the Sauscrit knowy without change, and used in the pronunciaticn, not only of Sanscrit Gram derivatives, but also of a large number of the most essential Dravi- { t; th dian roots. formg (4.) Though the Telugu has been more exposed to Sanscrit influ- of ¢, ences than the Tamil, yet larger use is made of those sounds in Tamil oceur than in Telugu,—a circumstance which is incompatible with the suppo- I\ sition of the derivation of those sounds from the Sansecrit. Dy Putting all these considerations together, it appears probable that of the instead of the Dravidian languages having borrowed the cerebral con- It sonants from the Sanscrit, the Sanserit has borrowed them from the lext |, Dréavidian languages ; and it will, I think, be demonstrated in the midd]y ¢ Glossarial Affinities,” that the Sanscrit has not disdained to borrow lafe from the Dravidian languages words as well as sounds. betwee. After the foregoing observations were written, I met with strengt Mr. Norris’s paper on the language of the Scythic tablets’ of Behistun, and found a similar opinion expressed therein respecting the Dy % The Védic Sanserit possesses a peculiar ‘1’—the cerebral ‘1’ of the Drévi- intgpg Ventgg dian languages—which has disappeared from the more modern Sanserit. This “1° f the Dravidian languages, especially of is one of the most distinctive features o illterch : the Canarese and the Tamil; and its origin is probably the same as that of the other cerebrals. It has nearly disappeared from the Telugu, apparently through Dy the influence of the more modern Sanscrit. ﬂlI‘OW h ::i \E — DIALECTIC INTERCHANGES. 113 i Dravidian origin of the Sanscrit cerebrals. Mr. Norris says, ‘I will en here express my conviction that the sounds called cerebral are peculiar ghe to the Tartar or Finnish class of languages ; that the really Indian 69 languages are all of Tartar origin, or, at least, that their phonetic and 1§ grammatical affinities are Tartar; and that the writers of Sanscrit the adopted the sound from their Indian neighbours, in the same way that s, the Scandinavians appear to have adopted a similar sound from their hfn' neighbours, the Lapps, who are undoubtedly Tartars ; the Icelanders 11 who retain the old Scandinavian language, pronouncing the words el ‘falla’ and ‘fullr,” as though written ¢fadla’ and ¢ fudir.’ 16 “Tt is certainly the case that this peculiar articulation has not been noticed as cerebral, so far as I know, by the writers who have M1 treated of those languages; but this may be accounted for from the eel fact that Tartars have had few, if any, native grammarians ; that gene- a0y rally speaking, their languages are unwritten ; and that, where written, leri the alphabet, not having been adopted by themselves, but given to ally them by nations more civilized than themselves, the difference between lant the dentals and cerebrals was not striking enough to a foreigner to tpl‘() induce him to invent new characters to designate the sounds new to o him. But the existence of a ‘t’ or *d,” convertible into ‘1, is well- sentt known to Finnish philologers. Castrén, a Finnlander, in his Osteak sont G'rammar, uses distinct characters for the cerebral and dental ‘d’ and ravl ‘t,’ though not giving them these denominations, and directs that the former should be pronounced somewhat aspirated, with the addition nflu of ‘1, as “dhl’ or ‘dl,’” and ¢ thl’ or ¢ t] ;’ observing that similar sounds aml occur in the Lappish and Finnish tongues.” These observations undoubtedly strengthen the supposition of the Ippe Dravidian origin of the cerebral consonants of the Sanscrit, as well as of the Scythian relationship of the Dravidian languages. that oo It is remarkable that the Drividian ‘1’ (as will be seen under the i tll@ next head) is interchangeable with the cerebral ¢d,” through their ’ i the middle point, the vocalic ‘r All these letters appear to have a cog- 110w nate origin ; and the supposition of the existence of a remote connection between the Dravidian and the Ugrian families evidently grows in kil 1th strength as we proceed. ts’ Of Diarecric INTERCHANGE oF ConsoNaANTs.—Under this head I intend to consider, not the euphonic refinements which have been in- Dratt vented by grammarians, but those natural, unintentional mutations and s} interchanges which are brought to view by a comparison of the various jally Dravidian dialects. These dialectic interchanges will be found to of 6 g throw much light on the Drividian laws of sound, whilst they enable I — = - 114 SOUNDS. us to identify many words and inflexional forms contained im the various dialects which appear at first sight to be unconnected, but which are in reality the same. I Following as before the order of the Déva-nigari alphabet, I pro- ceed to point out the dialectic changes to which each Dravidian conso- nant appears to be liable. I omit the aspirated consonants, as not really Dravidian. J{ 11 1. The gutturals.—‘k,’ ‘g, ‘ng.’ ‘g’ being merely the sonant of ‘k,’ in the changes now to be [t It enquired into, ‘k’ and ‘g’ will be regarded as identical. (i.) ‘k,” when used as a sonant, that is, as ‘g, changes into ‘v Where we have ‘g, in Tamil, we sometimes find ‘v, in Telugu; m e.g., “ agu,’ Tam., to become; avu,’ Tel. In ¢ k4, the infinitive of this verb in Telugu, which corresponds to the Tamil ‘4ga’ ‘k’ (or ‘g’) ill reappears. It is especially in the middle of words that this consonant evinces a tendency to be changed into ‘v.’ This tendency constantly thy appears in the spoken language of the lower classes of the Tamil people in the southern provinces ; and has found a place even in the all poets ; e.g., ‘mova,’ to be pained, instead of the more common I | ¢ noga.’ @ Tn Telugu ‘v’ is often not only pronounced, but written, instead fro 9 ot g e.g., ¢ pagadamu,’ coral, corrupted into ¢ pavadamu.’ th Compare with this the change of the Sanserit ‘laghu,’ light, into €y sometimes the Latin ¢levis.’ It will be seen that, per contra, ‘v for becomes ‘g’ in Telugu. .l (ii.) ¢k’ changes into ‘ch’ or IS As the Tamil ‘&’ becomes ¢ ch,” when doubled, and is represented in the alphabet by the equiva- (ti lent of the Déva-nigari ‘ch,” the change of ‘k’ into ‘ch, is identical with that of ‘k’ into ¢ § The former change appears in the Telugu, the latter in the Tamil. Compare the change of the Greek and Latin ¢k, into the Sanserit ‘§; e.g., ‘ééca’ and ‘decem, softened into beiy “daa, ten. The Canarese retains ‘k,” the older pronunciation of this consonant, of | and where ‘k’ is found in the Canarese, we generally find ‘ch’ in lege Telugu, and ‘§’ in Tamil ; e.g., ¢ kinna,’ Can., small ; ¢ chinna,” Tel.; Can «§inna, Tam. ‘kevi, Can , the car ; ‘chevi, Tel.; ¢&evi,” Tam. intg «Geéy, Can.,to do; ‘chéy,” Tel ; ‘Sey, Tam. Sometimes the older fl'()n] ¢k’ is retained by the Tamil as well as by the Canarese, and the Comy RNV softening appears in the Telugu only ; e.g., ‘kei,” Tamil, the hand ; 14§ ‘kyé, or ‘keiyyi, Can.; ‘chéy,’ Tel. ‘Kedu,” Tam. and Can., fo I spoil ; Tel.,, ‘chedu,’” or ¢ cheru.’ illf() ¢ DIALECTIC INTERCHANGES 115 A similar change of ‘k’ into ¢ ch,’ appears even in Sanscrit; e.g., ‘vich-as,” of speech, from the ecrude nominative ¢ vak, speech. (iii.) ‘kk’ change systematically into ‘ch.’ This change may be regarded as the rule of the pronunciation of the lower classes of the Tamil people in the southern districts. Further north, and in gram- b matical Tamil, it is rarely met with, but in the Telugu country the rule re-appears ; and in a large class of words, especially in the for- matives of verbs, the double ‘k’ of the Tamil is replaced regularly by ¢ch’ in Telugu. The following instances of this change are con- tained even in grammatical Tamil: ‘kajchu, o beil, for the more regular ‘kajkku,” and ¢ phychu,’ to wrrigate, for ¢ paykku.’ A single illustration will suffice to illustrate the perfect conformity I, in this point between the vulgar pronunciation of Tamil in the extreme his south and the regular, grammatical use of ‘ch’ for ‘kk’ in Telugu. /) “veikka,” Tamil, to place (infinitive), is pronouneed °wveicha,” by the b illiterate in the southern Tamil districts ; and in grammatical Telugu fly the same word is both written and pronounced ¢veicha.’ mil (iv.) ‘k’ appears sometimes to have changed into ‘t.’ I cannot the adduce a good instance of this change in the Dravidian languages ; but 101 I suspect that the ‘t’ of some inflexional terminations in Goénd (e.g., the nominative plural of the personal pronouns) has been derived ead from the Tamil ‘k.’ Compare also ¢ vakili, a doorway, Telugu, with the Malayala form of the same word, ‘vatal. I am doubtful, how- nto ever, whether this illustration can be depended upon, because the Tamil mes form of the same word is ¢véafal,’ classically ¢ vayil,” apparently from ¢ vay-il,” literally mouth-house. mes In other families of languages the interchange between ‘k’ and d Vi t’ is not uncommon ; e.g., Doric ¢ 7avos,’” ke, instead of ¢ e-xetvos.’ fieal 2. The palatals :— ch’ or “§’ ¢}, ‘nj. hlgu; atin I class the changes of ‘ch, ‘§’ and ¢j, together, those letters nto being in reality but one in the Drividian languages. The only change to which this letter, ‘&’ or ¢j,” is liable is that of being softened into ‘y.” On comparing the Tamil with the Cana- ,nal‘.t, )’ D rese, many instanees of this process are brought to light ; e. g., ¢ hefar,’ Can., @ name (ancient Canarese, ¢ pefar’), has been softened in Tamil Tel, into ‘peyar,’ ‘pejr, or ¢pér. In words borrowed by the Tamil i older from the Sanserit, ¢y’ is optionally used instead of ‘§’ and very commonly instead of ¢j. Thus ‘rija,” Sans., @ kirng (in Tamil 4 the Wld ) ‘rasa,” and with the masculine formative, ¢ rdé-an’), becomes ¢ ray-an.’ In the southern provinces of the Tamil country this change of <&’ Dy 0 into ‘y’ has become a characteristic of the pronunciation of the lower 1.2 116 SOUNDS. s o i g classes. In those provinces in all words in which this letter occurs, whether Sanscrit or Tamil, the ¢§’ is changed into ‘y; e. g., they say ‘ariyi,” rice, instead of ‘ariéi.’ 8. The cerebrals :—°t,’ ¢d, ‘n’ (i.) The cerebral ¢t when used as a sonant and pronounced ? in Tamil : €. 1. .‘ as ‘d, is sometimes changed into the vocalic ‘r nAdi, Sans., @ measure, is commonly written and pronounced in Tamil “néri; and this is colloquially pronounced ‘nali’ in the The counter- southern districts, by a further change of ‘1’ into ‘1’ part of this change, wiz., the change of ‘r’ into d,” is much more common in the Dravidian languages. (See ‘r’) In Telugu there are some instances of the change of ¢d’ into the hard rough ‘R, e. g., “chedu, ?o spoil (Tamil and Canarese, ¢ kedu’), should have for its transitive form ¢cheduchu,” answering to the Tamil ‘kedukku f whereas ¢ cheruchu’ is used instead. - (ii.) “n’ This cerebral nasal is frequently softened in Telugu into ¢n, the nasal of the dental row. The Tamil, the most correct repre- E;‘ sentative of the ancient speech of the Drividians, makes much use of i 01 ) as well as of the other cerebrals ; and the colloquial Tamil and the Malayalam go beyond the grammatical Tamil in preferring ¢n’ 1 to ‘n.” The Telugu, on the other hand, whilst it uses the other cere- brals freely enough, often prefers ‘n’ to ‘n Thus, it softens the Tamil (and old Drividian) words ¢kan, eye, ‘vin,’ lkeaven, ¢man, earth, into ‘kannu, vinnu, and ‘mannun. It softens even some Sanscrit words in a similar manner ; e. g., ¢ guna,’ quality, instead of Sometimes, both in Tamil and in the other idioms, ‘n’ is ¢ guna.’ first enphonized into ‘nd,’ and then converted into ‘d,” which when doubled becomes ‘t; e. g., ‘en,’ etght, has first become in Canarese e — ¢entu, and then in Tamil ‘ettu’’ ‘pen,’ a female, has become f ¢pendu; and in the equivalent Tamil word, ¢pedei, a fen, the ‘n’ i has disappeared and left no substitute. | i 4. The dentals :—t, d, “n. i (i.) ‘t, or its sonant equivalent ‘d,’ changes into ‘1’ in Tamil. ) In the interchange of the cerebral ‘d’ and ‘r) ‘r sometimes ap- pears to have been the original sound, and “d’ the corruption ; but in the change which is now referred to, it is ‘d’ that is the original sound, and which is changed into ‘r. This change may arise from C0) the circumstance that the ‘r’ into which ‘d’ is altered is pronounced Px very like a dental, and bears a considerable resemblance to ‘d.’ In th; the southern districts of the Tamil country the change of ‘d’ (when of preceded and followed by a vowel) into ‘r’ is exceedingly common in ]JLp DIALECTIC INTERCHANGES. I8 the pronunciation of the lower classes: but the same change has in some instances found its way into the written language ; e. g., ¢virei, seed, or to sow, instead of the more correct ¢ videi.’ In Canarese ¢ ad,’ the inflexional increment, or basis of most of the oblique cases of certain l singular nouns, changes in some instances into ‘ar; e. g., compare “id-ar-a,’ of this, from ¢id-u,’ this, with ‘mar-ad-a,’ of @ tree, from ¢ mara,’ a tree. I In this instance the change from “d’ to ‘r,” or some equivalent (4 change, was obviously required by euphony : ‘id-ad-a’ would have been [« intolerably monotonous, and ¢ mar-ar-a’ not less so. This change of ‘&’ (i into 0’ is not unknown to the North Indian languages; and in that 16 family it is often followed up by a further change of ‘r’ into ‘L.’ [ Bopp has pointed out some instances in the Hindustani and Bengali ; 18 e.g., ‘des, ten, becomes ‘rel’ in the compound numbers, as ¢ ba-reh,’ twelve. ) An instance of the change of ‘1’ into ‘I’ is furnished by ano- ther compound numeral, sizteen, which is not ¢s6-reh,’” but ¢sb-leh.’ 1to The Précrit also changed ¢ d into ‘r,” as is seen in the instance of the (g word ‘raha, ten, which has superseded ¢daha,” a softened form of the Sanserit ¢ daSa,” and which is used instead of ¢daha’ at the end of of e compound numerals. Ln\l It seems to me not improbable that in these cases, and also in the ‘0 re- use in Bengali and Marathi of ‘1’ instead of ¢d’ or ‘t,” as a sign of the preterite and passive participle, we see an evidence of the ancient t 16 prevalence of Dravidian influences in Northern India. A1, It may be noticed here that the Umbrian also regularly changed me ‘d’ into ‘r; e. g. ‘sedes’ was written ‘seres.’ As in Tamil, how- | of ever, this change took place only when ¢d’ came between two vowels. ] 13 ¢ ? (ii.) ¢d’ sometimes changes into S hen This change appears in Tamil in the optional use of &’ in the 150 formatives of nouns instead of ¢d. Thus, ‘vayadu,” age, becomes ome ‘vayasu ;° and ‘periSu,’ large, or that which is large, is commonly used Bl instead of ¢ peridu,’ the more correct form. In Telugu, ‘d’ is still more frequently subject to this change. We have a remarkable in- stance of the softening of ‘d’ into ‘s, of ‘s’ into ‘y,” and finally of bl the obliteration of the ‘y itself in the Dravidian word signifying amﬂ. a name. This in Tulu is ¢ pudar,” in ancient Canarese ¢pefar, in P bub classical Tamil ¢ peyar ; and finally in modern Tamil ¢ pér.’ (iii.) ‘nd’ changes in Tamil into ¢0j. In this change ¢j’ must be Lginﬁl (s BIAE) from considered as identical with 8, being the sound which ¢§’ takes when preceded by a nasal; and it is always expressed by ‘&’ in Tamil. In uu(}ed this conjunction the dental ‘n’ changes into ‘fi,’ which is the dental & of the palatal row. The change of ‘nd’ into ‘nj’ especially takes o place after the vowels ‘i’ or ‘ei’ In general it is heard in the pro- 118 SOUNDS. nunciation of the lower classes only ; but in a few instances it has found its way into grammatical compositions ; e.g. ‘eindu,’ five, has changed into ¢ einju,’ and this again into ¢ anju,” a form which is found even in the Tamil classies. (iv.) “tt’ change into ‘ch’ in Tamil after the vowels i’ and ei. ¢ 229 The change to which I refer appears to be one of dd’ into 88, if the form of the Tamil letters is regarded : but it has already been ex- plained that sonants become surds when doubled ; and hence ‘dd’ must ¢ L2 be expressed as ¢tt, and Ss as ¢ch,” this being their pronunciation when in juxtaposition. The corruption of the double, soft dentals ¢ tt° ¢ L% into the palatals 88, which are represented by ¢ ch,’ is peculiarly easy and natural. This ¢ch’ which arises out of ¢tt, though almost uni- versally characteristic of the pronunciation of the mass of the Tamil people, as distinguished from the literati, is rarely found in grammatical compositions, except in the formatives of derivative nouns, after the semi-vowel ‘r; e g., ‘unar-chi,) semsation, knowledge, instead of ¢ unar-tti, which is more in accordance with analogy. In Malayalam this change not only appears in the pronunciation of the vulgar, but is the rule of the language after the vowels ¢i and ‘e; and ‘ch’ is written as well as pronounced : e. ¢., compare ‘siricha,’ that laughed, with the corresponding Tamil ¢giritta.’ (v.) ‘n, the nasal of the dental ‘varga,’ changes or is softened % 9 into ‘y This change rarely occurs ; but we have an indubitable in- stance of it in the change of ‘nu,’ the Telugu copulative conjunction and, into ¢ yu.” ‘yu’ has been still further softened in Canarese into Y We have also an instance of this in the softening in classical Tamil of na,’” the termination of certain preterite relative participles Hto " ya sk e/, “folli-ya,” that said, instead of the more regular ¢ §olli-na.’ ! (vi.) “n’ also changes, though still more rarely, into ‘m? e.g, ¢miru, you, in Telugu, must have been altered from ¢ niru,’ the form which answers to the Tamil ¢nir,” and which Telugu analogies would lead us to expect. See the section on The Pronoun. 5. The labials :—*p,” ‘b, ‘m. ! This remarkable rule (i.) ¢p’ changes in Canarese into ‘h.’ applies to the initial ‘p’ of nearly all words in modern Canarese, whether they are pure Dravidian words or Sanscrit derivatives ; ¢.g. ‘pattu,” Tam., ten (‘padi,’ Tel.), is in Canarese ¢hattu.’ In like manner, ‘pana,’ money, a Sanscrit derivative, is in modern Canarese ‘hana.” This change of ‘p’ into ‘h’ has taken place in comparatively d recent times ; for in the old Canarese, and in the dialect of the Bada- t _— DIALECTIC INTERCHANGES. 119 gars of the Nilgherries, ¢ p’ maintains its ground. A change similar to this is occasionally apparent in the Marathi, the neighbour of the Canarese on the north : the Sansc. participle ¢bhita-s,” one who kas been, being altered in Marathi to “hoté ; e.g., ¢ hoto-n,” I was. Com- pare also the Prakrit ¢ hé-mi, I was, from bhita-smi.’ A gsimilar change of ‘p’ into ‘h’ appears in Armenian; e.g., foot is in Armenian ‘het’ (for ‘pet’), and father, ‘hayr’ (for ¢payr’). )t (ii.) ‘b, the sonant of ‘p,’ sometimes changes into ‘m ; eg., ¢ padi, Tel., fen, becomes ‘midi’ in ‘tom-midi,” nine, a compound which the analogy of both the Tamil and the Telugu would require to be ‘tom-badi : ¢ enbar,” they will say, is often in poetical Tamil ¢ enmar.’ ‘b’ is also euphonically added to ‘m’ in vulgar Tamil. I do not refer 1l to such words as ¢ pAmbu,” Tam., a snake, as compared with ‘pamu,’ il Telugu ; for in those instances the ‘m’ itself is euphonic, and bu’ 16 (in Can., ‘vn’) is the real formative. Cases in which the ‘m’ is radi- cal and the ‘b’ euphonic occur plentifully in colloquial Tamil ; e.g., of ¢ kddumei,” wheat, commonly pronounced ‘kédumbei.’ 18 (iii.) ¢ b’ is often softened into ¢ v’ in Tamil. Most transitive verbs 18 in Tamil form their future tense by means of ‘p’ or ‘pp ; and in the 0 corresponding intransitives we should expect to find the future formed by ¢b,” the sonant of ‘p ) Where the root ends in a nasal consonant, ed this ‘b’ appears ; but where it ends in a vowel, ‘b’ is ordinarily See the section on The Verb. In some instances in I changed into ‘v. on the Tamil poets this ‘b’ of the future is changed, not into ‘v’ but into 1t0 ‘m, according to the previous rule. (iv.) ‘m’ changes into ‘n This change is often apparent in the al les nominatives of neuter nouns in Tamil, the natural termination of many of which is ‘m,” but which optionally terminate in ‘n:’ e.g., ‘pala-n,’ Jar profit, a derivative from ¢phala.’ Sans., is more commonly used than ¢ pala-m.’ In Telugu ‘kola-nu, a tank, answers to the Tamil g ‘kunla-m.’ In the same manner ‘um, the Tamil aoristic future for- 1 ald mative, has become ‘nu’ in Telugu; and ‘um, the Tamil copulative particle, has in Telugu been changed into ‘nu.’ (v.) ‘m’ changes into ‘v; e.g., ‘nam, we, and ‘ném, you, in ancient Canarese are softened in the modern dialect to ¢nav-u’ and ‘ niv-u.’ le 0S6y 6. The semi-vowels :—y, ‘r} <1} ‘v ‘r) 1’ ‘R ves ; (i.) ‘y’ changes into ‘8§ It has been shown that ‘ch,’ ‘8§, and ik ‘)’ are softened into ¢y’ in Tamil Notwithstanding this, and in rese direct opposition to it, we find in the colloquial Tamil, especially in ol ? that of the southern districts, a tendency also to harden ¢y’ into ‘8§ ad& £ e ; 120 SOUNDS. < S 7/ Through some peculiar perversity, where ought to be, it is pro- e i T S Sy nounced as ¢y, and where ¢y’ ought to be, it is pronounced as ‘5§ t e. g., ‘ padl,” hunger, is mispronounced by the vulgar ‘payi; whilst ‘ vayaru, the belly, is transformed into ¢ vasaru’ This change of ? ¥ into ‘&’ is not confined to the South, though it is more frequently met with there. Even in Madras, ‘payangal,’ boys, is pronounced ‘pafangal,’ and ‘ayal’ mear, is not only pronounced but written g ‘asal. C.29 In Telugu ¢y’ is invariably converted into 8, after the par- ticipial ‘1; e.g., ‘chéyi, having done, becomes ¢ chési’ When ‘y’ is used euphonically to prevent hiatus, it invariably retains its proper sound. z (ii.) ‘r’ changes into ‘d. A change of ‘d’ into ‘r’ has already been mentioned. This is sometimes met by a counter-change of ‘r’ into ‘d; e.g., ¢ per-u, or ‘per-iya,” Tam., large, becomes in Telugu ¢ pedd-a.’ P T T T, (iii ) ‘r’ changes into ‘1. “r’ and ‘I are found to be interchange- able in many families of languages ; and in the Dravidian family this G Sometimes ‘1’ is cor- interchange is one of very common occurrence. rupted into ‘r ;’ but in a larger number of cases 1’ appears to be the ¢l original, and ‘1’ the corruption. In the case of the distinetively Dra- i E}% vidian ‘r’ and ‘1, the change is uniformly of the latter nature ; and 1l the change of the ordinary semi-vowel ‘1’ into the corresponding ‘1, U though not uniform, is an exceedingly common one ; and one which may be regarded as a characteristic of colloquial Tamil. It is espe- cially at the beginning of words that this change occurs, and it takes place as frequently in the case of derivatives from the Sanscrit, as in 1I the case of Dravidian roots ; e.g., ¢ rakshi,’ o save (‘raksh-a,” Sans.), is pronounced by the vulgar ¢lakshi,’ or ‘latchi.’ In the middle of words ‘r’ is less frequently changed into ¢I; e nevertheless where the Tamil uses ‘r’ we sometimes find 416 C; the Telugu; e.g., ‘teri] to appear, in Tamil, becomes ¢teli-yn’ in il Telugu. ) Seeing that a ‘tendency to change ‘r’ into ‘T still exists and IOS operates in the Dravidian languages, especially in Tamil, it may be concluded that in those ancient roots which are the common property of several families of langnage, and in which an interchange appears Ve to exist between ‘1’ and ‘1l ‘r’ was the original, and ‘I’ the altered 1 sound : e. g., if the Dravidian ‘kar-u,’ or ‘kar, black, is connected, 8 as it probably is, with the Sanscrit ‘ kal-a,’ black, it may be concluded sl that the Sanscrit form of the root is less ancient than the Dravidian ; and this supposition is confirmed by the existence of this root ‘kar,’ of black, in many of the Scythian languages. 1ng g DIALECTIC INTERCHANGES. i - 2 1 The fact of the frequency of the interchange between ‘1’ and ‘1,’ (irrespective of the question of priority), would lead us to suspect a remote connection between several sets of DrAvidian roots, which are ¢ SIR, now considered to be independent of each other ; e. g., compare Tam., small, with ¢sil, few ;” and * par’ (another form of ¢ per’), large, with ¢pal) many. (iv.) ‘I’ changes into ‘r Whilst the ordinary change is that of ‘r’into ‘1’ the change of ‘I’ into ‘1’ is occasionally met with, and forms one of the peculiarities of the Tulu. The Tulu generally changes the final ‘1’ of the other Drividian languages into ‘r; e.g., ¢vil, In this instance Tam., a bow, (* billu,” Can.), becomes in Tulu ¢bir.’ it cannot be doubted that ‘1’ was the original termination of the word ; for we find the same root west of the Indus in the Brahui ¢billa,” @ bow. A similar interchange between ‘I’ and ¢ 1’ takes place in Central Asia. The ‘1’ of the Manchu is converted into ‘r’ in the Mongolian. In Zend and Old Persian ‘1’ was unknown, and ‘r’ was systemati- cally used instead. In Telugu, ‘lu, the pluralising suffix of nouns, is sometimes changed into ‘ru.’ This change, however, of ‘1’ into ‘r’ is mnot systematic as in the Tulu, but exceptional. In Tamil, ‘1’ is eupho- nically changed, not into ‘r,” but into ‘R’ before ‘p; eg., ‘palpala,’ various, becomes in written compositions ¢ parpala.’ This proves that ) a change of ‘1’ into ‘r is not contrary to Tamil laws of sound. (v.) ‘1’ changes in the language of the Kus to ‘d.’ The change S of “d’ into ‘1’ is common enough ; but the regular change of ‘1’ into ‘d’ is peculiar to this idiom ; e.g., ‘palu,’ Telugu, milk, is in Ku ¢ padu;’ ‘illu, house, is ¢iddu.’ (vi.) v’ is generally hardened in Canarese into ‘b’ in the beginning of a word ; eg., var, Tamil, to fourish, becomes in W here 347 1 Canarese ‘bal. is not changed into ‘b,” viz., in the middle of words, the Canarese generally softens it into ‘w. The i same softening is sometimes observed in the pronunciation of the lower classes of Tamilians. il In Malayalam, ‘v’ is always ‘w (vii.) The ¢v’ euphonic of the Tamil is sometimes changed into e ‘g’ in Telugu. Both ¢y’ and ‘v’ are used euphonically to pre- i vent Aiatus in Tamil ; so in Telugu ‘g’ is sometimes used not only ) ol instead of ‘v,” but also instead of ‘y. Compare Tam. ¢ 4rRu-(v)-ar,’ sux persons, with the Tel. ‘aru-(g)-uru,” and the Tam. honorific d, singular ‘tandei-(y)-ar,” father, with the corresponding Tel. ¢ tandri- el 19 (g)-ar-u. This will, perhaps, explain the occasional use of ‘g’ instead of ‘v’ as the sign of the future tense in High Tamil; e.¢., ¢ Seygén,’ Iy instead of ‘éeyyén,’ I will do. 122 SOUNDS. (viii.) ‘r’ (the pecnliar vocalic ‘r’ of the Tamil) interchanges with three different consonants. Sometimes it becomes ‘n; e.g., ‘mirugu, Tamil, fo sink, is changed in Telugu to ‘munagu; and ¢kuri/ Tam., @ hole, becomes in Canarese ‘ kuni.’ Ordinarily ‘r’ is changed in Telugu into ¢d.’ Neither the Telugu nor the Canarese possesses the Tamil ‘r.’ In a very few instances the Telugu uses ‘n’ or ‘1’ instead: sometimes it omits the consonant altogether, without using a substitute ; but in a vast majority of instances it converts ‘r’ into “d. ‘r’ is ordinarily converted in Canarese into ‘1: the same change characterises the pronunciation of the mass of the Tamil people in the southern districts of the country, and prevails in the Malayalam also. This change of ‘r’ into ‘1’ and the previous one of ‘r’ into ‘d, That law is that form the constituents of an important dialectic law. the same consonant which is ‘r’ in Tamil, is generally ¢d’ in Telugu, and always ‘1’ in Canarese. Thus, fo caress, is ‘tar-u’ in Tamil; ‘tad-u,’” in Telugu; and ¢tal-u, in Canarese. The numeral seven 1s ¢ér-u, in Tamil; ‘éd-u,” in Telugu; and ¢él-u,’ in Canarese. In the 1l compound numeral ¢eélnfiru,’ seven hundred, the Telugu ‘éd-u’ is found to change, like the Canarese, into ¢ él-u. The word signifying time, which is included in the adverbial nouns then and now (literally D that time and ths time), is in Tamil *poru-du,” in Telugu ¢ pud-u,’ or a ¢ podd-u,” and in Malayélam ¢pol.’ In this instance the Canarese uses a different word. It thus appears that ‘1’ and ¢d’ are as intimately allied as ‘d’ and ‘y. This is a point of some importance in the affiliation of languages; for an interchange of ‘d’ and 1’ is b characteristic of the Ugrian family of languages, as well as of the Dréavidian family and the North-Indian vernaculars. The same word th is written with “t’ or ‘d’ in the Ostiak, and with ‘1’ in the Magyar Ie and Finnish. al A corresponding interchange is occasionally observed even in the of Indo-European languages; e.g., compare °8&axpvua, a tear, with lay lackryma : but in those languages it is rarely met with, whereas it is a characteristic dialectic sign of several families of tongues belonging a to the Scythian group. (ix.) It may be added that ‘1’ changes, though rarely, into ‘r; off e.g., ‘kammélan,” Tamil, an artificer, from ‘kam,” work, and ‘Al Wl to exercise, becomes in Canarese ‘kamméran-u,’ though ¢ali) a suffix equivalent to “alan,” is used in Canarese as well as in Tamil. (' (x.) ‘R’ (the strong, rough r of the Tamil), is frequently changed b in Tulu into ¢j; e.g., ‘muru’ (the original form of ‘mindru’), Tam. fan three, becomes ‘muji; ¢ aru, siz, ¢4ji. o 2 DIALECTIC INTERCHANGES. 3 ¢ L ) This change of ‘®’ into ‘j,’ the equivalent of is directly the converse of the change of ‘s’ into ‘r’ which is so common in the Indo-European tongues. Having now finished the consideration of the dialectic changes which pure Drividian consonants undergo, it remains to point out the [; changes which take place in the Sanscrit sibilants, when words in which they oceur are borrowed from the Sanscrit by the Tamil. (1.) “sh.’ The hard, cerebral sibilant of the Sanscrit is unknown to the classical Tamil. Sometimes it is changed into ‘8§, a change which ordinarily takes place at the present day in the pronunciation of the lower classes in the southern districts: sometimes, though more rarely, it is changed into ‘r;’ but most commonly it is converted into X ¢ d.’ This ‘d’ is sometimes softened down into the dental ‘d. Thus, 1 ‘ manushya,” Sans., man, becomes in classical Tamil ¢ménida-n;’ I, and this by a further change becomes ‘manida-n.’ A very old ¥ example of the change of the Sanserit ‘sh’ into ‘d’ in Tamil, can 1§ be adduced. The month ¢ Ashada,’ Sans., July—A ugust, has become 16 in Tamil ‘Adi? and this change dates probably from the earliest 18 period of the cultivation of the Tamil language. In ¢Teigha, ‘D g January—February, the hard ¢gh,’ instead of being changed, has ly been discarded altogether: the Tamil name of this month, as far back or as the literature reaches, has been ¢ Tei.’ 2 ‘s 9 A The hissing sibilant of the Sanscrit, answering to our a8 English ¢s,’ is ordinarily in Tamil converted into ¢ d,’” the sonant of 10 ¢t, which is pronounced as ‘th’ in that; e.g., ‘méasam,” Sans., @ month, 1§ becomes in classical Tamil ¢ madam ;’ and ¢ manas,” the mind, becomes In this conversion of the Sansecrit ¢s he ¢ manad-u.’ into ‘d’ in Tamil, there is a change from the sibilant to the dental, which is exactly the d at reverse of that change from the dental to the semi-sibilant which has already been described. It may be compared with the weakening the of : g into ‘h’ which we find in several of the Indo-European ith languages. When ‘s’ happens to be the first consonant of a Sanscrit deri- L8 ing vative, it is sometimes omitted in Tamil altogether; e.g., ¢ sthAnam,’ a place, becomes ‘tanam.’ More commonly in modern Tamil, an ) ‘14) effort is made to pronounce this ‘s with the help of the vowel ‘i, which is prefixed to it in order to assist enunciation; e.g., ¢istiri’ iy fix (‘str, Sans.), @ woman. The Sauscrit sibilant never changes into ‘r 9 in Tamil, This change, though very common in languages of the Indo-European ;ged i family, rarely, if ever, appears in the Dravidian. The ounly instances in which it may be conjectured to have taken 124 SOUNDS. place, are the following. The Tamil-Canarese root ‘ir, to be, in Brahui ‘ar,” may be allied to the Indo-European substantive verb, as represented by the Sanscrit as the Canarese ‘miur-u,” three, is identical with the Brahui ‘mus-it,” and the Tulu ‘mdj-i:’ the Tamil plural of ratiecnal beings ¢ar,’ resembles the Sanscrit epicene plural “as? and perhaps, thongh more doubtfully still, the Tamil ‘iru,” romn, o euphonized into ¢iru-mbu,” may be compared with the Sanserit ‘ayas,’ and the English word ¢iron’—which is allied to ‘ayas,’ through the | change of ‘s’ into ‘r. 'I'he instances, however, which I have now cited, are not by any means decisive; for the only reliable aflinity 1 amongst them is that of ‘mlr-u’ and ‘mus-it; and in that instance ‘r’ was probably the original letter, and ‘s’ or ¢j’ the corruption. 8 EvpnoNic PERMUTATION OF ConsoNANTS.—The permutation of consonants for euphonic reasons, though it throws less light on the 8 laws of sound than dialectic interchange, includes a few points of con- siderable interest. Dravidian grammarians have bestowed more atten- tion and care on euphonic permutation than on any other subject; t and the permutations which the grammar of the Tamil requires or 8 allows, are at least twice as numerous, and more than twice as per- i plexing to beginners, as those of the Sansecrit. On examining the al permutations of consonants prescribed in the grammar of the Tamil, 0 the Telugu, and the Canarese—the three principal languages of this 0] family—it is evident that a considerable proportion of them are f founded upon Sanserit precedents: another class in which Sanserit b rules of euphony have been, not imitated, but emulated and surpassed, 1 may be regarded rather as prosodial than as grammatical changes: but h after these have been eliminated, a certain number of euphonic permutations remain, which are altogether peculiar to these languages, th and which proceed from, and help to illustrate, their laws of sound. I; It will suffice to notice a few of those permutations; for the subject 80) is too wide, and ‘at the same time not ofJsufficient importance, to allow an of our entering on a minute investigation of it. ‘t (1.) In “dwanda’ compounds, ¢.e.,, in nouns which are united do together, not by copulative conjunctions, but by a common sign of the plurality (in the use of which common sign the Dravidian languages Te resemble, and perhaps imitate, the Sanserit), if the second member D) of the compound commences with the first or surd consonant of any ; of the five ‘vargas’ (wiz., ‘k, ‘ch orfgl i) ¢4, “p %), dhe surd Th must be changed into the corresponding sonant or soft letter. In to | those Dréividian languages which have adhered to the alphabetical E i system of the Sanserit, as the Telugu and the Canarese, this conver- i, EUPHONIC PERMUTATION. 125 sion of the surd into the sonant is carried into effect and expressed by 3 the employment of a different character. In Tamil, in which the § same character is used to represent both surds and sonants, a different i character is not employed, but the softening of the first consonant il of the second word is always apparent in the pronunciation. l This peculiar rule evidently proceeds from the DrAvidian law, that ] the same consonant which is a surd at the beginning of a word, should 16 be regarded as a sonant in the middle; for the first consonant of the second word, being placed in the middle of a compound, has become a y medial by position. The existence of this rule in the Telugu and H Canarese, notwithstanding the Sanscrit influences to which they have been subjected, proves that the law of the convertibility of surds and sonants is not confined to the Tamil. of All the Dravidian dialects agree in softening the initial surd of the he second member of ‘dwanda’ compounds: but with respect to com- )it pounds in which the words stand to one another in a case relation, 1l e.g., substantives, of which the first is used adjectivally or to qualify b; the second, or an infinitive and its governing verb, the Telugu pur- sues a different course from the Tamil. 0r The rule of the Telugu is el that when words belonging to the ¢ druta’ class, including all infinitives, he are followed by any word commencing with a surd consonant, such consonant is to be converted (as in ‘dwanda’ compounds) into its soft i, hla or sonant equivalent. The rule of the Telugu on this point resembles are that of the Lappish, and still more the rule of the Welsh; and it has b been observed that the Welsh, possibly through the pre-historic !ed) influence of the Finnish, is the most Scythic of all the Indo-European languages. but nie It is curious that in combinations of words which are similar to geS,’ those referred to above, and uniformly after infinitives in ‘a,’ the Tamil, instead of softening, doubles and hardens the initial surd- nd sonant of the succeeding word. The Tamil also invariably doubles, ject and consequently hardens, the initial surd of the second member of low ‘ tat-purusha’ compounds, <.¢., compounds in which the words stand in a case-relation to each other. In such combinations, the Canarese, ited though it is less careful of euphony than either the Tamil or the uOf Telugu, requires that the initial surd of the second member of the ages compound should be softened: it requires, for instance, that ¢huli her togalu,’ @ tiger’s skin, shall be written and pronounced “huli dogalu.’ any The Tamil, on the contrary, requires the initial surd in all such cases qurd to be hardened and doubled; e.g., the same compound in Tamil, viz., [n ‘¢ puli tol,” a tiger's skin, must be written and pronounced, not ¢ puli fical dol,” but ¢ puli-(t)tol.’ This doubling and hardening of the initial is yer 126 SOUNDS. evidently meant to symbolize the transition of the signification of the first word to the second ; and it will be seen that this expedient has been very generally resorted to by the Tamil. When the first word is used not as a noun or adjective, but as a verb or relative participle, the initial surd of the second word becomes a sonant in Tamil also, as in Telugu; e.g., compare ‘kéj gombu,’ @ withering branch, with ¢kay-(k)kombu,” a branch with fruit. (2.) The Tamil system of assimilating, or euphonically changing, concurrent consonants, is in many particulars almost identical with that of the Sanscrit, and has probably been arranged in imitation of 1t. Nevertheless, there are some exceptions which may be regarded as distinctively Drévidian, and which are founded upon Drévidian laws of sound; e.g., the mutation of ‘1’ into ‘n’ in various unex- pected combinations. Through this tendency to nasalisation, pél-da,’ like, becomes ‘podn-da, or rather ¢ pon-dra; ‘kol-da,” taken, bought, becomes ‘kon-da; and the latter euphonic mutation has found its C way in Telugu into the root itself, which is ‘kon-u,’ 7o buy, instead of the older Tamil ‘kol.’ It does not appear to have been noticed even by Tamil grammarians, that ‘1’ in a few instances, has been converted into ‘n’ before ‘k. Thus ‘nan-ku, or ‘nén-gu,’ four, is 0 derived from ‘nal-ku,’ an older form of the word; and ‘Panguni,’ the Tamil name of the month of March-April, has been altered from the Sanserit ¢ Phalguna. In Telugu a correspondihg tendency &l appears in the change of ‘1’ into ‘n’ before ‘t; e.g., “ilti,” of « house, is softened into ‘inti. In all these cases ‘1’ is undoubtedly the original; and these proofs of the priority of ‘1’ to ‘n, cor- roborate the suspicion that the Latin ¢alius’ is older than its Sanscrit th equivalent ‘anyas.’ €x 0n Evpnoxic NUNNATION, oR NasaLizATioN.— Much use is made in th ‘k the Dravidian languages, especially in the Tamil and Telugu, of the nasals ‘ng,’ “fij,’ ‘n,” ‘n,” and ‘m’ (to which should be added ‘n’ or ‘m,’ Wi the ¢ half anuswéra’ of the Telugu), for the purpose of euphonising the harder consonants of each ¢varga.’ All the nasals referred to, with Ve the exception of the ¢half anuswéra,” which is an inorganic sound, P are regarded by native grammarians as modifications of the sound int of ‘m; the nature of each modification being determined by the the In manner in which ‘m’ is affected by succeeding consonants. Ve Tamil, as in Sanserit, all those modifications are expressed by the the nasal consonants which constitute the final characters of each of the th five ‘vargas.’ In Telugu and in Canarese one and the same cha- racter, which is called ‘anuswara,” but which possesses a greater range Of\ EUPHONIC NUNNATION. 127 of power than the ‘anusvara’ of the Sanscrit, is used to represent the whole of the nasal modifications referred to. The pronunciation of this character, however, varies so as to accord with the succeeding consonant as in Tamil. The ‘nunnation,’ or nasalization, of the Dravidian languages is of three kinds. 1. The first kind of ‘nunnation’ is used to a greater extent in Tamil than in any other dialect. It consists in the insertion of a nasal before the initial consonant of the formative suffix of many of nouns and verbs. The formative syllable or suffix, the nature of d which will be explained more particularly in the succeeding section, 11} is added to the crude root of the verb or noun, and constitutes the inflexional theme, to which the signs of inflexion are annexed. The by nasalised formative is used in Tamil by the intransitive form of the i verb and by the isolated form of the noun. When the verb becomes t§ transitive, and when the noun becomes adjectival, or is placed in a d case-relation to some other noun, the nasal disappears, and the con- od sonant to which it was prefixed—the initial consonant of the formative —1s hardened and doubled. el 1§ The nasal is modified in accordance with the nature of the initial 1, consonant of the formative suffix: it becomes ‘ng’ before ‘k’or ‘g 5 “fij’ before S or ‘ch;’ ‘n’ before ‘t’ or ‘d; ‘n’ before ‘t’ or g om Y and ‘m’ before ‘p’ or ‘b’ The Telugu uses the ‘anuswéira’ to () express all these varieties of sound; and the *half anuswéra’ in certain other cases. lly (I (i.) Of the use of the first nasal, ‘ng,’ to emphasize and euphonize the formative suffix ‘k-u’ or ‘g-u, the Tamil affords innumerable it examples. One verb and noun will suffice; e.g., ‘ada-ngu,’ to refrain oneself, to keep in, is formed from the root ‘ada’ by the addition of Bl the formative, intransitive suffix ¢ gu,” which is euphonized into ‘ngu the ‘ké-nggei,” heat, is from ‘ki’ or ‘ k4§, to burn (in Telugu ‘ka-gu’); with the addition of the suftix ‘gei,” euphonized into ‘nggei.’ 10y the (ii.) Instances of the euphonic use of the nasal of the second ‘varga, ‘1, are more common in Telugu than in Tamil. Thus, jith ‘ pafich-u,” Tel,, to divide, is derived from ¢pag-u, Tamil (changed lﬂd; Lllnd into ‘pach-u,’ and then nasalized into ¢ pafich-u’); and is analogous to the Tamil noun ‘pang-u,’ a share, which is derived from the same the In verbal root: ¢ ret,ti—ﬁchu,’ Tel., to double, is an example of the use of the euphonic nasal by verbs of the tranmsitive class; a class in which the e that nasal is not used by any other dialect but the Telugu. the (iii.) The cerebrals ‘t’ and ‘d’ are not used as formative suffixes chas of verbs, though some verbal roots end in those consonants; but they gt 5 128 SOUNDS. are not unfrequently used as formatives of neuter nouns; e.g., ¢ ira-d-u,’ the original of the Tamil numeral fwo, corresponding to the Canarese ¢era-du,” has been euphonised to ¢“ira-nd-u.’ The Tamil adverbial nouns ‘a-nd-u,’ there, ‘i-nd-u,’ here, ¢ ya-nd-u,’ where, are derived from ¢ 4 ) a and ‘1’ the demonstrative bases, and ¢ ya,’ the interrogative base, with the addition of the usual neuter formative ¢d-u,” euphonised to ¢nd-u’ In Telugu a large number of masculine formatives in ‘d-u’ receive in pronunciation the obscure nasal ‘n;’ e.g., for ‘ vadu-lu’ or ¢ vAd-lu, they, ¢ vind-lu’ is commonly used. On comparing the Tamil ‘karandi,” @ spoon, with ¢ garite,” the Telugu form of the same word, we find that sometimes the nasal is used of one dialect and rejected by another. (iv.) We see an example of the euphonic use of ‘n,’ the nasal of the dental ‘varga,’ in the intransitive verb ¢tiru-nd-u,’ Tamil, ¢o become correct, from ¢ tiru, the radical base, and ‘du,” the formative, euphonised into ‘ndu: the transitive form of the same verb is ¢ tiru-ttu,’ fo correct. We find the same euphonic insertion in the Tamil demonstrative adjectives ¢ anda, ‘inda,’ that, this, which are derived from the demonstrative pronouns ¢ad-u,’ that, ¢id-u, this, by the addition of the adjectival or relative participial ‘a,’ and the inser- An tion of the euphonic nasal before ¢d,’ the neuter formative. example of the nasalisation of a moun of this class is found in ‘maru-ndu, Tamil, medicine, which is derived from ¢maru, fragrant, with the addition of the formative ‘du,’ euphonised to ‘ndu.’ (v.) Many examples of the euphonic insertion of ‘m’ before the suffix in ‘b’ might be adduced; but the following will suffice, ¢ tiru-mbu,” fo twrn (intransitively), of which the root is unques- tionably ¢tiru, as appears from the corresponding Telugu ¢tiru-gu’ and Canarese ¢ tiru-vu.’ The Tamil form of the transitive of the same verb is ‘tiru-pp-u, to furn. An example of a similar insertion of euphonic ‘m’ before the formative ‘b’ of a noun, is seen in ‘eru-mbu,’ Tamil, an ant, when compared with the equivalent Canarese word ¢iru-ve.’ The formatives ‘nd-u’ and ‘ mbu,’ are extremely common terminations of Tamil nouns; and with few if any exceptions, where- ever those terminations appear, they will be found on examination to be euphonized suffixes to the root. 2. The second use to which the euphonic nasal is put is altogether It consists in the insertion of an euphonic ‘n,’ peculiar to the Tumil. between the verbal theme and the ¢d’ which constitutes the sign of f the preterite of a very large number of Tamil verbs. The same ‘d ordinarily forms the preterite in ancient Canarese, and it is not I ., __ — R — —— EUPHONIC NUNNATION. 129 unknown to the Telugu ; but in those ]angqages the nasal, ‘n, is not prefixed to it. | The following are examples of this nasalisation of the sign of the preterite in Tamil: ¢ vAr-nd-én’ (for ¢ var-d-én’), I flourished, from the Il root ‘var in Canarese ‘bal: compare Old Canarese preterite, ) 0 ¢ bil-d-en.’ So also, ¢ virn-nd-u’ (for ¢ viru-d-u’), having fallen, from the root ‘viru’ or ‘vir:’ High Tamil, ‘vir-d-u ;7 Canarese equivalent, Iy ¢ bidd-u.’ The corresponding Malayala ¢ vin-u,’ is an example of the il absorption of the dental in the nasal. d In colloquial or vulgar Tamil, this euphonic insertion of ‘n’ is carried further than the grammatical Tamil allows. Thus, ¢$ejy-d-a, ol done, and ° pey-d-a,” rained, are vulgularly progounced ‘Sey-nj-a’ and of ‘ pey-nj-a.’ 3. A third use of the euphonic nasal, is the insertion, in Tamil, fo of ‘n’ or ‘n, before the final ‘d’ or ‘d, of some verbal roots. Ve It} The same rule sometimes applies to roots and forms that terminate 2 in the rough ‘m, or even in the ordinary semi-vowel ‘r Thus, nil ed ¢kar-u,” Can., @ calf, is ‘kanr-u’ in Tamil (pronounced ¢ kandr-u’); and he ¢mir-u, Can., three, is in Tamil ‘mfnr-u’ (pronounced ‘mind-u’). In the first and second classes of instances in which nunnation is o\ used for purposes of euphony, the Dravidian languages pursue a course An nd of their own, which is different from the usages of the Scythian, as well as of the Syro-Arabian and Indo-European families of languages. I, In the Syro-Arabian languages, especially in Talmudic Hebrew, to euphonic ‘n is always a final, and is often emphatic as well as euphonic. the In the Turkish, ‘n’ is used between the bases of words and their 166 In the North- 16 inflexions, in a manner similar to its use in Sanserit. Indian vernaculars an obscure nasal, ‘n,’ is often used as a final. gll But none of these usages perfectly corresponds to the Dravidian nasa- e lisation referred to under the first and second heads. In the third uof class of instances, the Dravidian usage bears a close resemblance to bo, the Indo-European. In the seventh class of Sanscrit verbal roots, a ord nasal is inserted in the special tenses, so as to coalesce with a final oD dental, e.g., ‘nid,” fo revile, becomes ‘nindati, ke reviles. Compare (4% also the root ‘uda,” water, with its derivative root ‘und,’ to be wet. A 1l fo In Latin we similar nasalisation is found both in Latin and Greek. find the unaltered root in the preterite, and a nasalised form in the ther' present: e.g., compare ¢ scidi’ with ¢ seindo;’ ¢ cubui’ with ‘cumbo ; ‘n, ‘tetigi’ with ‘tango; ‘fregi’ with ‘frango. Compare also the nOf Latin ¢centum, with the Greek °é-xarov.’ In Greek, compare the d roots ‘ ua® and ‘AaB, with the nasalised forms of those roots found | nﬂt K B IR 205 T e 130 SOUNDS. in the present tense, e.g., ¢ pav0-avw.’ to learn, and NappB-dvw, to take. The principle of euphonic nasalisation contained in these Sanserit, Greek, and Latin examples, though not perfectly identical with the Dravidian usage, corresponds to it in a remarkable degree. The dif- ference consists in this, that in the Indo-European languages the inser- tion of ‘n’ is purely euphonic, whereas in Tamil it contributes to gram- matical expression. The consonant to which ‘n’ is prefixed by neuter verbs, is deprived of the ‘n,’ and also hardened and doubled, by transitives. PrEvENTION OF ‘H1ATUS.—An examination of the means employed in the Dravidian langnages to prevent hiatus between concurrent vowels, will bring to light some analogies with the Indo-Kuropean languages, especially with the Greek. In Sanserit, and all other languages in which negation is effected by the use of ¢alpha privative,” when this ‘a’ is followed by a vowel, o) ‘m is added to it to prevent Aiatus, and ‘a’ becomes ‘an, ‘in, or ¢ un ’ In the Latin and Germanic languages, this ‘n,” which was used at first euphonically, has become an inseparable part of the privative particles, ¢in’ or ‘un.’ In the greater number of the Indo-European languages, this is almost the only conjuncture of vowels in which Aiatus In Sansecrit and 'l%; is prevented by the insertion of an euphonic ‘n.’ PAli, ‘n’ is also used for the purpose of preventing hiatus between the ¢ final base-vowels of nouns or pronouns and their case terminations, in order that the vowels of the base may escape elision or corruption, and be preserved pure. In some instances (a probably older) ¢ m’ is used for this purpose, instead of ‘n.’ This usage is unknown in the cog- nate languages, with the exception of the use of ‘n’ between the 11§ vowel of the base and the termination of the genitive plural in the \( Zend and the Old High German. iy It is in Greek that the use of ‘n,” to prevent Aiatus, has been most n fully developed : for whilst in Sanserit contiguous vowels are combined or changed, so that Asatus is unknown, in Greek, in which vowels are tly 2 th more persistent, ‘n is used to prevent Aiatus between contiguous I vowels, and that not only when they belong to the same word, but also, and still more, when they belong to different words. b On turning our attention to the Dravidian languages, we may be chance at first sight to observe nothing which resembles the system now tlo mentioned. In Tamil and Canarese, and generally in the Dravidian lan languages, hiatus between contiguous vowels is prevented by the use bl of ‘v pr oy, Vowels are never combined or changed in the Drévi- hig dian languages, as in Sanserit, except in the case of compounds which lig PREVENTION OF ¢ HIATUS. 1 3 £ 31 have been borrowed directly from the Sanserit itself ; nor are final vowels elided in these languages before words commencing with a vowel, with the exception of some short finals, which are considered as mere vocalisations, In Telugu and Canarese, a few other unimportant vowels are occa- sionally elided. Ordinarily, however, for the sake of ease of pronun- ciation, and in order to the retention of the agglutinative structure which is natural to these languages, all vowels are preserved pure and pronounced separately : but as ‘hiatus’ is dreaded with peculiar intensity, the awkwardness of concurrent vowels is avoided by the interposition of ‘v’ or ¢y, between the final vowel of one word and the initial vowel of the succeeding one. The rule of the Tamil, which in most particulars is the rule of the Canarese also, is that ‘v’ is used after the vowels ‘a,’ ‘u,’ and ‘o, with their long vowels, and ¢au,’ | and that ‘y’ is used after ‘i, ‘e, with their long vowels, and “ei.” Thus in Tamil, ¢ vara illei,” 2ot come, is written and pronounced ¢ vara-(v)- illei, and ¢vari-alla,’ (it is) wot the way, becomes ° vari-(y)-alla.’ i This use of ‘v, in one conjunction of vowels, and of oy another, is doubtless a result of the progressive refinement of the lan- gnage. Originally, we may be sure that one consonant alone was used 1S for this purpose. These euphonic insertions of ‘v’ and ¢ vy’ between i contiguous vowels are observed in the common conversation of I3 Drévidians, as well as in written compositions; and they are found 1n even in the barbarous dialects : e.g., in the Ku, which was reduced to 1 writing only a few years ago, ‘v’ may optionally be used for euphony, ol as in Tamil. Thus in Ku, one may say either ¢4alu, she, or ¢4(v)alu.’ Iig This insertion of ‘v’ or ‘y, takes place, not only when a word termi- he nating with a vowel is followed by a word beginning with another he vowel, but also (as in Sanscrit) between the final vowels of substan- tives and the initial vowels of their case terminations: e.g., ¢ puli-(y)-il,;’ st wn the tamarind, ¢ pila-(v)-il,’ in the jack. The use of ‘alpha priva- ed tive’ to produce negation being unknown to the Dravidian languages, 916 there is nothing in any of them which corresponds to the use of ¢ an, )us “in,’ or “un’ privative, instead of ‘a,’ in the Indo-European languages, 50, before words beginning with a vowel. Hitherto the only analogy which may have appeared to exist 16y between the Drividian usage and the Greek, in respect of the preven- 0W tion of Aiatus, consists in the use of ‘v’ or ‘y,” by the Dréavidian a0 languages as an euphonic copula. use As soon as we enter upon the examination of the means by whick ﬁw. huatus is prevented in Telugu, a real and remarkable analogy comes to el light ; for in many instances, where the Tamil uses ‘v, the Telugu, K 2 132 SOUNDS. like the Greek‘, uses ‘n By one of the two classes into which alk 2 is used to words are arranged in Telugu for euphonic purposes, ‘y prevent Aiatus when the succeeding word begins with a vowel; by the other, a very numerous class, ‘n’ is used, precisely as in Greek. Thus, instead of ‘tinnagd égenu, ¢ went slowly, the Telugu requires us to say ‘tinnaga-(n)-égenu.’ When ‘n’ is used in Telugu to prevent hiatus, it is called ¢druta, and words which admit of this euphonic appendage are called ‘druta prakrits, words of the ‘druta’ class. ‘ Druta’ is used in the sense of extra, and ¢ the druta n’ may be in- terpreted as ‘the extra n, or ‘the n which has no meaning of its own.” The other class of words consists of those which use ¢y’ instead of ‘n, or prevent elision in the Sanscrit manner, by ‘sandhi,” or com- bination. Such words are called the ‘cala’ class, and the rationale of their prefering ‘y’ to ‘n’ was first pointed out by Mr. Brown. When- ever ‘n’ (or its equivalent, ‘ni’ or ‘nu’) could have a meaning of its own, e.g., wherever it could be supposed to represent the copulative conjunction, ‘ni’ or ‘nu,’ or the case sign of the accusative or the loca- tive, there its use is inadmissible, and either ¢y’ or ‘sandhi’ must be used instead. Hence, there is no difference in principle between n’ and ‘y; for the latter is used in certain cases instead of the former, merely for the purpose of preventing misapprehension; and it can scarcely be doubted that both letters were originally identical in origin and in use, like ‘v’ and ‘y,” in Tamil. The Telugu ‘n’ directly cor- responds to the Tamil ‘v Compare the Tel., ‘r4-(n)-é lédu,” (he, she, or it) has not come indeed, with the Tam,, ¢ vara-(v)-é illei.’ Even in Tel., 0] is replaced by ‘v, after the emphatic ‘é e.g., ‘A-ast’-6-(v)-8&, that very property. After ‘e, the Tamil requires < y’ instead of ‘v.’ An euphonic peculiarity of the Telugu may here be noticed. ‘ni’ or ‘nu, the equivalents of ‘n,’ are used euphonically between the final vowel of any word belonging to the ‘druta’ class (the class which uses ‘n’ to prevent hiatus) and the hard, surd initial consonant of the succeeding word—which initial surd is at the same time converted into its corresponding sonant. They may also be optionally used before any initial consonant, provided always that the word terminating in a vowel to which they are affixed, belongs to the class referred to. It is deserving of notice, that in this conjunction ‘ni’ or ‘nu’ may be changed into that form of ‘m’ (the Telugu ¢ anuswara’) which coalesces with the succeeding consonant. I regard ‘n’ as the original form of this euphonic copula of the Telugu, and ‘y,’ as a softening of the same. An undoubted and independent instance of this softening process is seen in the change o of the Telugu copulative particle, nu, and, into ¢ yu,’ in certain s P },’liEVEN’l‘IOl;I OF © HIATUS. 133 t:onjunctic'ms in the higher dialect of the language. This word has been softened still further in Canarese into ¢ u A 0,* In the Sanserit of the Vedas also, “ y’ is often used euphonically instead of ‘n, between base vowels and case terminations. That ‘nu’ was the original of [4 K yu, not conversely ‘yu’ the original of ‘nu,” appears from the con- Another instance nection of ‘nu’ with its Tamil equivalent ‘um. of this interchange of ‘um’ and ‘nu,” has already been pointed out in the identity of the ‘nu’ of the Telugu aorist, and the ‘um’ of the Tamil aoristic future. It has been mentioned that ‘v ) and ‘y’ are the letters which are used in Tamil for preventing hiatus, where ‘n’ and ‘y’ are used by the Telugu. On examining more closely the forms and inflexions of the classical Tamil, we shall find reason for advancing a step farther ; inasmuch as in Tamil also ‘n’ is used instead of v’ in a considerable number of instances, especially in the pronominal terminations of verbs in the classical dialect. Thus, the neuter plural demonstrative being ‘avei’ (for ‘a-(v)-a, from a-a’), we should expect to find the same ‘a-(v)-ei,” or the older ‘a-(v)-a,” in the third person plural neuter of verbs: but we find ‘a-(n)-a’ instead; v.e, we find the hiatus of I‘, ‘a-a’ filled up with ‘n’ instead of ‘v e.g., “ irukkindra(n)a, they are il (neuter), instead of ‘irukkindra(v)a. So also, whilst in the separate demonstratives ‘avan,” ke, and ‘avar, they (epicene), the hiatus is r- filled up with ‘v’ (‘a-(v)-an, ‘a-(v)-ar’), in the pronominal termina- 6 tions of verbs in the classical dialect we find “a-(n)-an’ often used instead of ‘a-(v)-an,’ and ‘a-(n)-ar’ instead of ‘a-(v)-ar:’ e.g., ‘irunda- (n)an,’ ke was, instead of ‘irunda(v)an,” or its ordinary contraction ¢irundan.’ es We sometimes also find the same ‘n”’ in the neuter plural {7 of appellative nouns in the classical dialect ; e.g., porula(n)a,” things that are real, realities, instead of ‘porula(v)a,’ or simply ‘porula.’ We y b find the same use of ‘n’ to prevent hiatus in the preterites and relative al past participles of a large number of Tamil verbs; e.g., ‘kitti(n)én, 7 showed, ‘katti(n)a,” which showed : in which forms the ‘n’ which comes 16 be between the preterite participle ¢ katti’ and the terminations ‘én’ and 9 2 he is clearly used (as ‘v, in ordinary cases) to prevent hiatus. The he euphonic character of the ‘n’ of ‘na,” whatever be its origin (respecting which see the section on Verbs,—Preterite tense), is conformed by the 00 ol circumstance that ‘n’ optionally changes in classical Tamil into ¢ Vit e.g., we may say, ‘katti(y)a, that showed, instead of ¢katti(n)a. the * According to this view of the case, the connection between the Canarese particle of conjunction, ‘i, and the copulative conjunction, ‘u,” which is found d in the Védas, and also the '’ of the Semitic languages, will appear to be ¢ u of the i accidental rather than real; for we have no reason to suppose the ;aill Sanscrit and the ‘0’ of the Hebrew to be softened forms of *um,” ‘mu,’ or ‘nu.’ 134 SOUNDS. Another instance of the use of ‘n’ in Tamil for the prevention of kiatus, is furnished by the numerals. The compound numerals between ‘ten’ and ‘twenty’ are formed by the combination of the word for ‘ten’ with each numeral in rotation. The Tamil word for ten is “pattu;’ but ‘padu’ is used in the numerals above twenty, and ‘padi,’ identical with the Telugu word for ‘ten, is used in the numerals from eleven to eighteen inclusive. Between this ‘padi’ and the units which follow, each of which, with the exception of ‘mfindru,” three, and i ‘nalu,’ four, commences with a vowel, ‘n is inserted for the preven- tion of hiatus, where the modern Tamil would have used ¢v.’ The euphonic character of this Tamil ‘n’ will appear on comparing the Tamil numerals with those of the Telugu, in most of which ‘h’ is used instead of ‘n:’—e.g., TeLvGU. TAMIL. 15 padi-(h)-énu padi-(n)-eindu 16 padi-(h)-dru padi-(n)-4ru 17 padi-(h)-édu padi-(n)-éru In the Tamil compound numeral, ‘padi-(n)-mindru, thirteen, we find the same ‘n’ used as in the previous examples, though there is no f hiatus to be prevented. The Telugu has here ¢ pada-midu ; the Cana- I rese, ‘hadi-muru ;' and as the Canarese uses ‘n,’ like the Tamil, in all h the other compound numbers between ‘eleven’ and ‘eighteen’ inclu- sive, and dispenses with it here, I think it is to be concluded, that in the Tamil ¢ padi(n)mundru,’ the ‘n’ has crept in through the influence of the numerals on each side of it, and in accordance with the euphoniec tendencies of the language in general. 1] We have an indubitable instance of the use of ‘n,” even in common (3 Tamil, to prevent hiatus,in appellative nouns ending in ¢ ei : e.g., when ‘A an appellative noun is formed {rom ‘ilei,” youtk, or young, by annexing 14 d an, the sign of the masc. sing., the compound is not ‘ilei-(y)-an,’ 8] but ¢ilei-(fij)-an’ or even ‘ilei-(n)-an.’ ‘nj’ is merely a more liquid da form of ‘n,” and in Malayalam regularly replaces ‘n’ in the pronoun X of the tirst person. Probably also ‘manér,” the epicene plural of the U future tense of the Tamil verb in some of the poets, is for ‘ma-ar ; e.g., ‘enma-(n)-ar, they will say, for ‘enmér,” and that for ‘enbar, the more common form. Pa the There is thus reason to suppose, that originally the Tamil agreed U with the Telugu in using a nasal instead of a semi-vowel, to keep contiguous vowels separate. It may be objected that ‘n’ evinces no tendency to change into v.’ I admit this; but if we suppose ‘m,” not fur sing ‘n,’ to have been the nasal which was originally employed for this qua . - e ——— PREVENTION OF ¢ HIATUS. 135 purpose, every difficulty will disappear, for ‘m’ readily changes on the one hand to ‘v,” and on the other to ‘n.’ Nor is it a merely gratuitous supposition that the Telugu may have used ‘m’ at a former period instead of ‘n,’ for we have already noticed that ‘ni’ or ‘nu,” the euphonic equivalents of ‘n,” are interchangeable in certain conjunctions with the ‘anuswara’ or assimilating ‘m;’ that in two important instances (the copulative particle and the aorist formative) the ‘n’ of the Telugu replaces an older ‘m’ of the Tamil; and that in Sanscrit also, instead of the ‘n’ which is ordinarily inserted between certain pronominal bases and their case-terminations, an older ‘m’ is some- times employed. It may also be noticed that the ‘ni’ or ‘nu,” which may be considered as the euphonic suffix of the accusative in Telugu, is replaced in Old Canarese by ‘m.’ The reader cannot fail to have observed that whilst the Dravidian languages accord to a certain extent with the Sanscrit in the point which has now been discussed, they accord to a much larger extent, with the Greek, and in one particular (the prevention of hiatus between the contiguous vowels of separate words) with the Greek alone. It is impossible to suppose that the Dravidian languages horrowed this usage from the Sanscrit, seeing that it occupies a much less important place in the Sanscrit than in the Dravidian languages, and has been much less fully developed. It should be mentioned here that the letter ‘r’ is in some instances used to prevent hiatus in each of the Dravidian idioms. In Tamil, ‘ki,’ the imperative singular of the verb fo preserve, becomes in the plural, not ¢ka-(v)-um,” but ‘ka-(r)-um.” The Canarese in certain cases inserts ‘r’ or ‘ar’ between the crude noun and the case terminations, instead of the more common ‘v, ‘n, or ‘d; eg., ¢ karid’-ar-a,” of that which is black. The Telugu inserts ‘r’ in a more distinctively euphonic manner between certain nouns and ‘alu,’ the suffix by which the feminine gender is sometimes denoted; e.g., ‘sun- daru-(r)-alu,” a handsome woman. Compare the latter with the Tamil ¢ soundariya-(v)-al,” in which the same separation is effected by the use of the more common euphonic ‘v.’ The ¢d’ which intervenes between the ‘i’ of the preterite verbal participle and the suffixes of many Canarese verbs (e.g., ‘madi-(d)-a,’ that did), though probably in its origin a sign of the preterite, is now | used simply as an euphonic insertion. ¢This ‘d’ becomes invariably ‘n’ in Telugu and Tamil; and in Tamil it is sometimes softened ’ further into ‘y “t’ is stated to be used in Telugu for a b similar purpose, viz., to prevent hiatus between certain nouns of quality and the nouns which are qualified by them; e.g., ¢ karaku-t- 136 SOUNDS. ammu,” a sharp arrow, but I have no doubt that this ‘t’ is identical with ‘ti and originally an inflexional particle ‘g 18 m sSome instances, used by the Telugu to prevent hiatus, or at least as an J o b euphonic formative, where the Tamil would prefer to use the rational’ plural noun of number, siz persons, may either be he, for vadu 1S aru(g)ur-u, or ‘aru(v)ur- probably ¢kadu another instance of the optional use of ‘g’ for ¢ in Telugu It is used euphonically, instead of the ¢y’ euphonic of the Tamil, in such words as ¢ tandri-(g)-ar-u, Jathers (used honorifically to signify Jather), compared with the Tamil ¢tandei-(y)-ar HarmoNic SEQUENCE oF Vowers.—In all the languages of the Scythian group (Finnish, Turkish, Mongolian, Manchu), but especially in Manchu, a law has been observed, which may be called ¢the law of harmonic sequence The law is, that a given vowel occurring in one syllable of a word, or in the root, requires an analogous vowel i.e., a vowel belonging to the same set (of which sets there are in the 4 Turkish four) in the following syllables of the same word or in the n particles appended to it, which, therefore, alter their vowels accord- al ingly. This rule, of which some traces remain even in the modern Persian, appears to pervade all the Scythian languages; and has been e regarded as a confirmation of the theory that all those langu'mes have by sprung from a common 011g1n fr In Telugu a similar law of attraction, or harmonic sequence, is 0 found to exist The range of its operation is restricted to two vowels “Sitcsand Ea but in principle it appears to be identical with the t Scythian law, u’ being changed into ‘i, and mto ‘u according th to the nature of the preceding vowel Thus the copulative particle is of fona after ‘i 9 {4k X3 1, el; and ‘nu’ after ‘u’ and the other vowels la ¢ ku,’ the sign of the dative case, becomes in like manner ‘ki’ after ig £ LI Y and ‘ei 1, lay In the above mentioned instances it is the vowels of the appended ()'I'a particles which are changed through the attraction of the vowels of the words to which they are suffixed: but in a large number of cases lay the suffixed particles retain their own vowels, and draw the vowels dey of the verb or noun to which they are suffixed, as also the vowels of oth any particles that may be added to them, into harmony with them- eu} selves Thus, the Telugu pluralising termination or suffix being ¢lu €39 the plural of katti) a knife, would naturally be ‘kattilu; but the tey. vowel of the suffix is too powerful for that of the base, and accord- Ing; ingly the plural becomes kattulu.” So also, whilst the singular dative Dra s ¢ katti-ki,” the dative plural i not ¢ kattila-ki but ¢ kattula-ku; B Iy HARMONIC SEQUENCE, 13 = { eal for “la,’ the plural inflexion, has the same power as the pluralising 0me particle ‘lu’ to convert ‘katti’ into ‘kattu, besides being able to Al change ‘ki,’ the dative post-position of the singular, into ¢ku.’ €l In the inflexion of verbs, the most influential particles in Telugu ' be are those which are marks of time, and by suffixing which the tenses 1§ are formed. Through the attraction of those particles, not only the It vowels of the pronominal fragments which are appended to them, but such even the secondary vowels of the verbal root. itself, are altered into nify harmony with the vowel of the particle of time. Thus, from ¢ kalugu,” o be able, ‘du,’ the aorist particle, and ‘nu,’ the abbrevia- tion of the pronoun ‘nénu,’ 7, is formed the aorist first person singular the ‘kalugu-du-nu,” 7 am able. On the other hand, the past verbal par- fally ticiple of ‘kalugu,” is not ‘ kalugi’ but ‘kaligi,” through the attraction law of the final ‘i’—the characteristic of the tense; and the preterite it of the first person singular is not ¢kalugi-ti-nu,” but ¢ kaligi-ti-ni.’ el Thus the verbal root ‘kalu’ becomes ¢ kali; ‘nu,’ the abbreviation of ‘nénu,” becomes ‘ni ;. an K the d both have by these changes been brought the into harmony with ti,’ an intermediate particle, which is probably an cord ancient sign of the preterite. pdern This remarkable law of the Telugu phonstic system evidently been accords with the essential principles of the law of harmonic sequence have by which the Scythian languages are characterised, and differs widely from the prevailing usage of the Indo-European languages. The e, 15 change which is apparent in the pronominal terminations of the various tenses of the Telugu verb (e.g., ‘nu’ in the first person of owels h ¢ he the present tense, ‘ni’ in the preterite), have been compared with the variation in Greek and Latin of the pronominal terminations pdin (;]e 1§ of the verb according to the temse: but the change in Greek and wels Latin arises merely from euphonic corruption; whereas the Dravi- dian change takes place in accordance with a regular fixed phonic af fer law, the operation of which is still apparent in every part of the grammar. onded Though I have directed attention only to the examples of this el of (2865 law which are furnished by the Telugu, in which it is most fully developed, traces of its existence could easily be pointed out in the oW 6]5 other dialects. Js of Thus in the Canarese verbal inflexions, the final jels 0 them euphonic or enuntiative vowel of the personal pronouns is ‘u,’ ‘e, or ‘1" according to the character of the preceding vowel; e.g., ‘madut- ol it the tév-e, we do, ‘maduttir-i, ye dg ¢ madidev-u,’ we did. If in the means employed to prevent hiatus between contiguous vowels, the LCCOTd Dravidian languages appeared to have been influenced by Indo- dative Euaropean usages, still more decided traces of Scythian influences and k0 138 SOUNDS. influene a Scythian relationship may be noticed in the phonetic law now mentioned. but also The rep it must PriNcipLEs oF SyLLABATION.—The chief peculiarity of Dravidian syllabation is its extreme simplicity and dislike of compound or (ven concurrent consonants; and this peculiarity characterizes the Tamil, nting u the most early cultivated member of the family, in a more marked Sanserit degree than any other Dravidian language. nstead ¢ In Telugu, Canarese, and Malayalam, the great majority of primi- rpim:i tive Dravidian words, 4.e., words which have not been derived from the Tan Sanserit or altered through Sanscrit influences, and in Tamil all words (gh)), 0 without exception, including even Sanscrit derivatives, are divided asthe § into syllables on the following plan. Double or treble consonants at “diya, ¢ the beginning of syllables, like ¢str’ in ‘strength,” are altogether inad- Anot missible. At the beginning not only of the first syllable of every of an up word, but also of every succeeding syllable, only one consonant is formeq ; allowed. If in the middle of a word of several syllables, one syllable Consonar ends with a consonant, and the succeeding one commences with issuffixe another consonant, the concurrent consonants must be euphonically ‘aprayj assimilated, or else a vowel must be inserted between them. At the L counclusion of a word, double and treble consonants, like ‘gth’ in 4gluting ‘strength,” are as inadmissible as at the beginning: and every word Very vl must terminate, in Telugu and Canarese, in a vowel ; in Tamil, either hnguage in a vowel or in a single semi-vowel, as ‘1’ or ‘r,’ or in a single simple, nasal, as- ‘a’or'‘m.’ It is obvious that this plan of syllabation is of enup, extremely unlike that of the Sanserit. son, The only double consonants which can stand together in the The | middle of a word in Tamil without an intervening vowel, are as Peﬂmps, | follows. 2 The various nasals, ‘ng,’ “fij,’ ‘n, ‘n, and ‘m, may ('IHSS arp precede the sonant of the ‘varga’ to which they belong; and hence, thO 0ng § ‘1ij-8 ‘n-d, ‘n-d, ‘m-b,” may concur; also ‘ngng,’ ‘njnj,’ Stl'ength a ¢ ng",g;’ S, ‘nn ‘mm,” ‘nm,” and ‘nm: the doubled surds ‘kk, ‘chch,’ Tacterigtj, ‘t.t.,’ ‘tt,” ¢pp, ‘1l ‘RR’ (pronounced ‘ttr’): also ‘tk’ and ‘tp; ¥ Whic} ¢rk, ‘mreh,” and ‘®rp; ‘yy, ‘Il ‘vv; and finally ‘nr, pronounced rfividian ‘ndr The only treble consonants which can coalesce in Tamil under gfﬂppling any circumstances, are the very soft, liquid omes, ‘rnd’ and ¢§nd. Tat}]er th Tamilian laws of sound allow only the above mentioned consonants SO]]{«mts ; to stand together in the middle of words without the intervention of g vowel. All other consonants must, be assimilated, that is, the first of the n must be made the same as the second, or else a vowel must be eﬂ“emmam inserted between them to render each capable of being pronounced by Tamilian organs. In the other Dravidian dialects, through the Wl]ils[ 10p tha:t PRINCIPLES OF SYLLABATION. 139 10w influence of the Sanserit, nasals are combined, not with sonants only, but also with surds; e.g., ¢ pamp-u,” Tel., to send, ¢ ent-u,” Can., eight. The repugnance of the Tamil to this practice is so very decided, that dian it must be concluded to be Un-Dravidian. d or Generally ‘i’ is the vowel which is used for the purpose of sepa-~ aml, rating unassimilable consonants, as appears from the manner in which rked Sanserit derivatives are Tamilised. Sometimes ! is employed instead of ¢i. K Thus the Sanscrit preposition ‘pra’ is changed into rimi- ‘pira’ in the compound derivatives which have been borrowed by from the Tamil; whilst ¢ Krishna’ becomes ¢ Kiruttina-n’ (‘tt’ instead of <) words ‘sh’), or even ¢ Kittina-n. Even such soft conjunctions of consonants vided as the Sanscrit ‘dya,” ‘dva, ‘gya, &c., are separated in Tamil into ts ab ‘diya,’ ‘diva,’ and °giya.’ nad- Another rule of Tamil syllabation is, that when the first consonant every of an unassimilable double consonant is separated from the second and ant 18 formed into a syllable by the intervention of a vowel, every such Jlable consonant (not being a semi-vowel) must be doubled before the vowel with issuffixed. Thus, ‘tatva,’ Sans., nature, becomes in Tamil ‘tat(t)uva ; ically ‘aprayodjana, wunprofitable, ‘ap(p)iraydsana.’ \t the In consequence of these peculiarities of syllabation and the h' in agglutinative structure of its inflexions, the Tamil language appears very verbose and lengthy when compared with the Sanscrit and the word either languages of Europe. Nevertheless, each syllable being exceedingly simple, and the great majority of the syllables being short, rapidity single fion 18 of enunciation is made to compensate for the absence of compres- sion. The mental physiology of the different races may be illustrated, n the a6, 43 perhaps, by their languages. The langvages of the Indo-European class are fond of combining clashing consonants, and welding them Ay into one syllable by sheer force of enunciation ; and it is certain that strength and directness of character and scorn of difficulties are cha- g Ilj Uj)’ ¢cheh, racteristics not only of the Indo-European langunages, but of the races by which those languages are spoken. On the other hand, the ] ‘s Dravidian family of languages prefers softening difficulties away to munced grappling with them; it aims at ease and softness of enunciation | und@l‘ rather than impressiveness: multiplying vowels, separating con- 4 i}nd,, sonants, assimilating differences of sound, and lengthening out its sonants words by successive agglutinations, it illustrates the characteristics o0 of 8 of the races by which it is spoken by the soft, sweet, garrulous ho first effeminancy of its utterances. st be Whilst the syllabation of the Dravidian languages differs widely el by from that of the Indo-European and Semitic families of tongues, it [gh thB 140 SOUNDS. tion o exhibits many points of resemblance to the system of the Scythian anothy group, and especially to that of the Finnish or Ugrian family. The Finnish, the Hungarian, and other languages of the same Tam, a8 stock, allow of only one consonant at the beginning of a syllable. When foreign words which begin with two consonants are pro- differe TECUTT nounced by a Magyar, the consonants are separated by the insertion Where the first con- of a vowel; eg., ‘kral’ becomes ¢ kiraly. sonant is a sibilant, it is formed into a distinet syllable by a prefixed In vowel ; e.g., ‘schola’ becomes ¢iskola.’ How perfectly in accordance with Tamil this is, is known to every European resideut in Southern 1ent which India who has heard the natives speak of establishing, or sending The same peculiarity has when their children to, an English *iskool.” been discovered in the language of the Scythic tablets of Behistun. that ¢ the g In rendering the word ‘Sparta’ into Scythian, the translator is found to have written it with a preceding ‘i; e.g., ¢ Isparta,” precisely as it pensat would be written in the present day in Magyar or in Tamil. dian plura, I do not suppose the Tamilian system of separating contiguous consonants by a vowel to be older than the Indo-European system of of a (i 1ouns ¢ combining them into one syllable. On the contrary, many of the lexical affinities which will be found in the section of ‘Roots’ and in illg to ! the ¢ Glossarial Affinities, appear to me to prove that the Dravi- they he shoulq dian roots were originally monosyllabic, and that the tendency to separate consonants by the insertion of a vowel, was not a charac- TQ\ugu teristic of the older speech, whatever it may have been, from which 2 (:hang the Dravidian family branched off. The inference which I draw is, Second that as a similar phonetic peculiarity appears in the Ugrian family the fy of languages, and is found in the Behistun tablets to have been a disgyl characteristic of the oldest Scythian tongue of which written records If]m’ ¢ 1 y survive, the Dravidian languages probably claim kindred rather with Sllﬂig g the Scythian group than with the Indo-European. d8~cnbe( Minor DiALeEcTiC PECULIARITIES.— ‘d iy’ 1. Euphonic displacement of Consonants. Mag In the Dravidian languages, consonants are sometimes found to €4, “uy change places, through haste or considerations of euphony. OPalu 1 We have an example of this in the Tamil ‘tasei,’ flesh, which by A\ § a displacement of consonants, and a consequent change of the surd brOUU‘ht 1 into the sonant, has become °gsadel. ¢ kudirei,” @ korse, is in this It g o manner often pronounced by the vulgar in the Tamil country ‘kuridei:’ are fmm( and looking at the root-syllable of the Telugu word, ¢ gur-ram,” it is ) 1y hard to decide whether ‘kuridei’ or ‘kudirei’ is to be regarded as ‘lﬂdu P £x the true Dravidian original. In many instances, through the opera- MINOR DIALECTIC PECULIARITIES. 141 eythiat . tion of this displacement, we find one form of a word in Tamil, and another, considerably different, in Telugu or Canarese. Thus, ¢ koppul,’ e same Tam., the navel, is in Telugu ‘pokkili; and ‘padar,” Tam., to spread syllable. as a creeper, is in Canarese ¢parad-u.’ In comparing words in the ate pro- different dialects, it is always necessary to bear in mind the frequent nsertion recurrence of this displacement. st con- 2. Euphonic displacement of Vowels. preﬁxed cordance In Telugu we find many instances of a still more curious displace- ment of vowels. Southern This displacement occurs most commonly in words which consist of three short syllables beginning with a vowel ; and sending rity has when it occurs, we find that the second vowel has disappeared, and Behistun, that the first vowel has migrated from the beginning of the word to i found the second syllable, and at the same time been lengthened to com- sely as I pensate for the vowel that is lost. I take as an example the Dravi- dian demonstrative pronouns, remote and proximate; and I select the plural, rather than the singular, to get rid of the disturbing element ntignous of a difference which exists in the formatives. gitem of In Tamil those pro- nouns are ‘avar,’ they, remote ; and ‘ivar,’ they, proximate, correspond- y of the O qnd 0 ing to “4lli’ and *he’ The Canarese adds ‘u’ to each word, so that they become ‘avaru’ and ‘ivaru.’ e Dr&\l By analogy this is the form we should expect to find in Telugu also; but on examination, we find in Jency 10 g charac- Telugu ‘véaru’ instead of ‘avaru, and ‘virn’ instead of ©ivaru; a change which has evidently been produced by the rejection of the om which second vowel, and the substitution for it of a lengthened form of [ draw iS, the first. 40 family The neuter demonstrative pronouns of the Telugu being dissyllables, there is no displacement in their nominatives (¢adi,’ e bee & that, ¢idi,’ this, corresponding closely to the Tamil ‘adu,’ ‘1dui); 1 records ther with but when they become trisyllables by the addition of the inflexional suffix ‘ni] we find a displacement similar to that which has been described : e.g., “ adini,’ 4 or of it, becomes ¢ dni,’ and ¢idini’ becomes ¢dini. Many ordinary substantives undergo in Telugu a similar change; e.g., ‘ural, Tamil, @ mortar, pronounced ¢ oral,” should by analogy be g fOuﬂd tU ‘oralu’ in Telugu; but instead of ‘oralu’ we find ¢ rolu.’ As soon as this peculiar law of the displacement of vowels is ,7 wllich bV brought to light, a large number of Telugu words and forms which at fhe S‘“‘ld first sight appear to be widely different from the Tamil and Canarese, 1 10 th]s are found to be the same or but slightly altered. Thus ¢kadu,” Tel., ; Lkul'idei : it will not be, or it is not, is found to be the same as the Tamil ¢ Agadu;’ 'I'ﬂm; i s “lédu,’ there is mot, corresponds to the Tamil ‘illadu:’ and by an el extension of a similar rule to monosyllables, we find 15, Tel., within, fhe 0P 142 SOUNDS. As to be identical with ‘ul,” Tam.; and ‘nu,” Tel., the copulative particle, to be identical with ‘um, Tam. be dist vowel A similar rule of displacement appears in the Tulu, though in Intrans a less degree. syllablg 3. Rejection of Radical Consonants. 18 upon The Telugu evinces a tendency to reject or soften away consonants I8 mor in the middle of words, even though such consonants should belong to verh 1 the root, not to the formative. Thus, ¢neruppu,” Tamil, fire, is i, Th softened into ‘nippu; ‘elumbu,’ @ bone, into ‘emmu ; ‘udal’ (pro- of the nounced ‘odal’); body, into ollu ; ¢ porudu,’ ¢ime, into ¢ poddu ;’ ‘erudu,’ second an oz, into ‘eddu; ‘marundu,’ medicine, into ¢ mandu.’ Princip Something similar to this process takes place, but not so systema- the mg tically, in vulgar colloquial Tamil. Tests ¢ In a few instances, on the other hand, the Telugu appears to have henee j retained a radical letter which has disappeared from the Tamil. For tw0 ace example, if we search for the origin of ¢odu,’ with, together with, the The suffix of the Tamil conjunctive case, no trace of its origin is apparent in Tamil. in the On examining the Telugu, we find that the corresponding Tani] suffix is ‘toda.’ It has already been shown that ‘d’ in Telugu cor- 2 (in egeh responds to ‘r in Tamil; and consequently ‘t6da’ would become in “Ody vi, Tamil ¢ tora.’ ‘tora’ (tora-mei) is actvally contained in Tamil, and Bven iy means companionship; and thus by the help of the Telugu we find (13“}' 'm that the Tamil ¢ 6du’ and ‘tbra’ are virtually identical ; that the mean- iy ing of the suffix ¢6du’ exactly accords with its use ; and that there is sty also reason to conclude another pair of similar words to be allied, viz., The “udan,’ with, a suffix of the conjunctive case in itself a noun signify- Tariag ing connection, and ¢ tudar,’ a verbal root, to follow, to join on. enoteg 4, Accent. Teanipo It is generally stated that the Dravidian languages are destitute of L] Hahle, | cA? {53 accent, and that emphasis is conveyed by the addition of the HOHH; g emphatic alone. Though, however, the Dravidian languages are pmﬂ(l!l]]ce destitute of the Indo-Greek system of accents, the use of accent is not e‘{/'a ‘6duv altogether unknown to them; and the position of the Dravidian accent, always an accute one, accords well with the agglutinative structure of Dravidian words. The accent is upon the first syllable of the word, that syllable alone, in most cases, constituting the base, prior to every addition of formatives and inflexional forms, and remaining always unchanged. The first syllable of every word may be regarded as the natural seat of accent; but if the word is compounded, a secondary accent distinguishes the first syllable of the second member of the compound. MINOR DIALECTIC PECULIARITIES. 143 article, As in other languages, so in the Dravidian, accent is carefully to be distinguished from quantity; and in enunciation an accented short ugh in vowel is more emphatic than an unaccented long one. Thus in the intransitive Tamil verb, ‘adangugiradu,’ 4t s contained, the second syllable ‘ang’ is long by position ; yet the only accent is that which sonants is upon the first syllable ¢ad,” which, though shorter than the second, long to is more emphatic. Another example is furnished by the compound verb ¢ udeind’-irukkiradu, it ¢s broken, literally Aaving been broken it fire, 18 (2 Though in this instance the second syllable of the first word |’ (pro- of the compound is long, not only by position, but by nature, and the ‘erudu, second syllable of the auxiliary word is long by position, yet the principal accent rests upon the first syllable of the first word, ¢ud,’ systema- the most emphatic portion of the compound, and the secondary accent 1o have rests upon ‘ir,’ the first syllable and crude base of the auxiliary ; hence it is pronounced ¢ tideindirukkiradu,’ every syllable, except the 1. For two accented ones, being enunciated lightly and with rapidity. pith, the The general rule of the Dravidian languages which fixes the accent pparent in the first or root-syllable, admits of one exception. In poetical ponding Tamil one and the same form is used as the third person of the verb 10 oot (in each tense, number, and gender) and as a participial noun; e.g., oome I ¢6duvan,” means either he will read, or one who reads, v.e. a reader. il aud Even in the colloquial dialect, the third person neuter singular, espe- | W6 ﬁﬂd cially in the future tense, is coustantly used in both senses; e.g., he mean- ¢oduvadu,” means either # will read, or that which will read, or there 18 abstractedly, yet more commonly still, @ reading, or to read. lied) Viz" The same form being thus used in a double sense, Tamil gram- sigﬂifY‘ marians have determined that the difference in signification should be denoted by a difference in accent. Thus when ‘6duvan’ is a verb, meaning e will read, the accent is left in its natural place, on the root syllable, e.g., ¢ 6duvan; but when it is an appellative or participial ginte of the ‘¢’ noun, meaning ke who reads, the pronominal termination is to be pronounced more emphatically, that is, it becomes the seat of accent, ages Me e.g., ‘6duvdn.’ b i5 00 1 acceﬂfv puctur® le Of ﬂ]e A PTiOr t0 vemﬂiﬂing , regﬂ,ﬂlel 1 )u]ldedf & | peaber of th faithi word separ agree famil : conj SECTION IL I Dréy —_—— by i A ROOTS. 003, Berore proceeding to examine and compare the grammatical every forms of the Dravidian languages, it is desirable to examine the 100t characteristics of Dravidian roots, and the nature of the changes the which are effected in them by the addition of the grammatical forms. fo an The manner in which various languages deal with their roots is unkng strongly illustrative of their essential spirit and distinctive character ; and it is chiefly with reference to their differences in this particular, 1 that the languages of Europe and Asia admit of being arranged into ot an classes. in gen Those classes are as follows:—(1.) The monosyllabic, uncom- a the pounded, or isolative languages, in which roots admit of no change or nle, combination, and in which all grammatical relations are expressed eithel either by auxiliary words or phrases, or by the position of words in a perat; sentence. (2.) The Semitic or intro-mutative langnages, in which Jerson grammatical relations are expressed by internal changes in the vowels Which of dis_sylla,bic roots. (8.) The agglutinative languages, in which thoge grammatical relations are expressed by affixes or suffixes added to the teon In the latter class I include both the root or compounded with it. A Indo-European and the Scythian groups of tongues. They differ, verhy indeed, greatly from one another in details, and that not only in their Case VA vocabularies, but also in their grammatical forms; yet I include them the Con both in one class, because they appear to agree, or to have originally Proba] ; agreed, in the principle of expressing grammatical relation by means be the | of the agglutination of auxiliary words. The difference between them is rather in degree than in essence. Agreeing in original additigr construction, they differ considerably in development. In the highly Stareely cultivated languages of the Indo-European family, post-positional gr&mma additions have gradually been melted down into inflexions, and some- a8 tWiIl times even blended with the root; whilst in the less plastic languages f ths, o CLASSIFICATION OF ROOTS. 145 of the Scythian group, the principle of agglutination has been more faithfully retained, and every portion and particle of every compound word has not only maintained its original position, but held fast its separate individuality. In this particular the Dravidian languages agree in general with the Scythian; and hence in each dialect of the family, there is, properly speaking, only one declension and one conjugation. I here proceed to point out the most notable peculiarities of the Dréavidian root-system, and of the manner in which roots are affected by inflexional combinations. ARRANGEMENT oF DRrAvIDIAN Roors inTo Crasses.—Dravidian roots, considered by themselves, apart from formative additions of ‘,ﬁC&l every kind, may be arranged into the three classes of—(1.) verbal the roots, capable of being used also as nouns, which constitute by far nges the most numerous class ; (2.) Nouns which cannot be traced up rms. to any extant verbs; and (3.) Particles of which the origin is s 18 unknown. ofer; lat, 1. Verbal Roots.—The Dravidian languages differ from the Sans- jnto crit and Greek, and accord with the languages of the Scythian group, in generally using the crude root of the verb, without any addition, com- as the imperative of the second person singular. This is the general rule, and the few apparent exceptions that exist are to be regarded 1ge or essed either as corruptions, or as euphonic or honorific forms of the im- perative. In a few instances, both in Tamil and in Telugu, the second smd hich person singular of the imperative has cast off its final consonant, 'owels which is generally in such cases a soft guttural or a liquid ; but in those instances the unchanged verbal theme is found in the less used which to the seeond person plural, or in the infinitive. A considerable proportion of Dravidian roots are used either as th the diﬂ'el‘, verbal themes or as nouns, without addition or alteration in either ) their case; and the class in which they are to be placed, depends solely on the connection. The use of any such root as a noun may be, and o them probably is, derived from its use as a verb, which would appear to gin&“)’ means be the primary condition and use of every word belonging to this class; but as such words, when used as nouns, are used without the oteel pighsl addition of formatives or any other marks of derivation, they can scarcely be regarded as derivatives from verbs; but in respect of highly qiond grammatical form, the verb and the noun must be considered either as twin sisters or as identical. The following will suffice as examples ] some- of this two-fold condition or use of the same root :—¢ karei,” Tam., as g L 146 ROOTS. a verb, means to melt, to be washed away; as a noun, a bank, a shore; “alei; Tam., as a verb, fto wander; as a noun, ¢ wave. In these instances it is evident that the radical meaning of the word is unre- I strained, and free to take either a verbal or a nominal direction. fl Moreover, as the Dravidian adjective is not separate from the noun, I¢ but is generally identical with it, each root may be said to be capable I¢ of a three-fold use; viz., (1) as a noun, (2) as an adjective, and V] (3) as a verb. Thus, in Tamil, ‘kad-u,’ if used as the nominative i of a verb, or followed by case terminations, is a noun, and means pungency or sharpness: if it is placed before another noun for the d purpose of qualifying it, it becomes an adjective; e.g., ¢ kadu nadei,’ a sharp walk; < kadu vay,” the tiger, literally sharp mouth : and when li it is followed by verbal suffixes, it becomes a verb; e.g., ¢ kadu-kkum,’ it v or will be sharp or pungent. With the formative addition ¢gu,’ Iy the same root becomes ‘kadu-gu,’” mustard. &l In these and in all similar instances, the quantity of the root il vowel remains unchanged; whereas in those few instances in which 0r the Sanscrit root is not tied to a single condition, the nominal and W1 verbal forms differ in the quantity of their root vowel; e.g., compare 1101 ‘vach-as’ (for ¢vék-as’), of speech, with ¢ vik-mi,” 7 speak. iy It would appear that originally there was no difference whatever fi0g in any instance between the verbal and the nominal form of the root bes in any Dravidian dialect; gradually, however, as the dialects became he more cultivated, and as logical distinetness was felt to be desirable, a separation commenced to take place. This separation was effected to by modifying the theme by some formative addition, when it was W] desired to restrict it to the one purpose alone, and prevent it from be being used for the other also. whi In many instances the theme is still used in the poetry, in accord- othe ance with ancient usages, indifferently either as a verb or as a noun ; iy but in prose more commonly as a noun only,_ or as a verb only. Jilk} ( Vel (2.) Nouns.—In Sanserit and the languages allied to it, all words, of { with the exception of a few pronouns and particles, are derived by frog native grammarians from verbal roots. In the Dravidian languages the number of nouns which are incapable of being traced up or 419 L resolved into verbs, is more considerable. Still such nouns bear but a Woxg small proportion to the entire number; and not a few which are Iy generally considered to be underived roots, are in reality verbal nouns deri“ or verbal derivatives. Many Dréavidian dissyllabic nouns have for their second syllable “al,” a particle which is a commonly used formative of verbal nouns in ( anq CLASSIFICATION OF ROOTS. 147 t) Tamil, and a sign of the infinitive in Canarese and Gond. All nouns N {7 of this class may safely be concluded to have sprung from verbal )it roots. In some instances their themes are discoverable, in others no trace of the verb from which they have been derived is now appa- n, rent. I cannot doubt that the following Tamil words, generally e regarded as primitives, are derived from roots which are still in use: nd viz., ‘viral,} a finger, from ‘viri) to expand; ‘kadal, the sea, from (% ‘kada,’ to pass beyond; ‘manal,’ sand, from ‘man, earth; ¢kudal, 15 I cannot he @ bowel, and ‘kural,’ @ pipe, from ‘kudei,’ to hollow out. discover the derivation of ‘niral, shade, “séval, @ cock, and a few g, similar nouns; nevertheless, judging of them by analogy, I have 11} little doubt that they also have been derived from verbal themes. 1l There are many nouns denoting primary objects, which in most ‘ru’ languages are primitive words, but which in the Dravidian languages are evidently derived from, or are identical with, extant verbal roots. 00t Thus, “nilam, Tam., the ground, is from ‘nil,’ to stand ; ‘madu,’” an ich oz, 18 from ‘madu,’ Can., to do, to work; ‘4du,” @ sheep, is identical nd with ¢adu,” to frisk; ¢kurangu, a monkey, is from ¢kura,’ to make a are noise; ‘pagal,’ day, as distinguished from night, is from ¢pagu,’ fo divide; ‘kan,’ the eye, is identical with ‘kan, to see; ¢miikku,’ the Vel nose, 1s from ‘migu,’ Can., to smell. Probably also, ‘kei, the hand, root bears the same relation to ¢ gé, Can., to do, which ¢ kara,’ Sans., hand, Ame bears to ‘kri’ or ‘kar, the Sanscrit verb ¢o do. o, Though the greater number of Dravidian nouns are undoubtedly cted to be regarded as verbal derivatives, a certain proportion remain Was which appear to be underived and independent. In this class are to Tom be included the personal pronouns; most of the nouns of relation which are used as post-positions, answering to the prepositions of ord- other languages, such as ‘mel,’ above, ‘kir,’ below; and a considerable un number of common nouns, including names of objects, e.g., ‘kal,’ foot, ‘man,’ earth, ‘vin, the sky, and nouns of quality, e.g., ‘kar, black, “vel, white, ‘se, red, &e. A suspicion may be entertained that some ouds, of the apparently simple nouns belonging to this class are derived | by from verbal roots which have become obsolete. Thus, ‘mun,” defore, ages a noun of relation, appears at first sight to be an underived radical, D or yet it is evident that it is connected with ‘mudal,’ first; and this hut word, being a verbal noun in ‘dal,’ is plainly derived from a verb in ) 1o ‘mu,; now lost; so that after all ‘mun’ itself may be a verbal derivative. juns Jable (8.) Particles.—A large majority of the Dravidian post-positions and adverbs, and several of the particles employed in nominal and ¥ 2 i 148 ROOTS. verbal inflexions are in reality verbs or nouns adapted to especial uses. Every word belonging to the class of adverbs and prepositions in the Dravidian languages is either the infinitive or the participle of a verb, or the nominative case of a noun used in a locative sense; and even of the inflexional particles which are employed in the declen- sion of nounsand in conjugating verbs, several are easily recognized to be derived from nouns. Thus, in Telugu, the signs of the instrumental ablative, ¢ ché’ and ‘chéta, are the nominative and locative of the word I hand; and the same case in Tamil is formed by the addition of ‘al; which is probably a corruption of ‘kél, in the seuse of @ channel. So f also the Tamil ‘locative of rest’ may be formed by the addition of any noun which signifies ¢a place; and the ‘locative of separation,” a case I denoting ‘motion from a place, or rather ‘the place from whence motion commences, is formed by the addition of ‘il’ or ¢in,’ which means a kouse. The same suffix added to the crude acristic form of the verb, con- t stitutes the subjunctive case in Tamil, e.g., ¢var-il,” 7 (he, she, it, or I} they) come, literally i (his or their) coming, that is, n the event of 4 (his or their) coming. fo Whilst all the post-positional adverbs and some of the inflexional 011 particles are certainly derived either from verbs or nouns, there are g several particles in use in the Drévidian languages which do not LK) appear to be connected with any nouns or verbal roots that are now a0 extant, and of which the origin is unknown; e.g., the copulative par- | ticle, ‘um’ in Tamil, ‘nu’ in Telugu, and ‘0’ in Canarese ; the suffixes il of present time, which form the present tense of verbs; viz. ‘gir,’ in Tamil ; ‘dap,’ in ancient, ‘utta,’ in modern Canarese ; and ‘chu’ or ‘tu,’ 0 in Telugu; “d’ or ‘i, the suffix of past time, and ‘v’ or ‘b, the sign (ar of the future. e Of the post-positions, or suffixes, which are used as signs of case, Bif some distinctly retain their original meaning ; in some, the original per meaning shines more or less distinctly through the technical appropri- gl ation; but no trace whatever remains of the original meaning of ‘ ku,’ Tigy ‘ki,’ or ‘ge,’ the sign of the dative, or of ‘ei,” ‘e,” ‘annu,’ or ‘am,’ the 10| sign of the accusative. The Dravidian dative and accusative have, therefore, assumed the (e Cong character of real grammatical cases; and in this particular the Dravi- f()u[ dian languages have been brought into harmony with the genius of the Indo-European grammar by the literary cultivation which they have but a0 gl received. It is impossible, I believe, to identify or connect any of the One above-mentioned particles with any verbal or nominal roots which are now discoverable in the Dravidian languages, as will be shown respect- a8 that ROOTS ORIGINALLY MONOSYLLABIC. 149 I ing each of them in order ; yet it is not only possible but probable S that some of them may have sprung from some such origin. f DrAvipiaN Roors oriGiNALLY MonNosyLLaBIC.—It may appear at first sight scarcely credible that the Dravidian roots were originally 0 monosyllabic, when it is considered that the majority of the words in al every Dravidian sentence are longer than those of (perhaps) any other d language in Asia or Europe (e.g., compare ‘irukkiradu,” Tamil, ¢¢ s, I with the Latin ¢est’), and are inferior in length only to the words of %0 the poly-synthetic languages of America. 1y The great length of Dravidian words arises partly from the sepa- 156 ration of clashing consonants by the insertion of euphonic vowels, but 06 chiefly from the successive agglutination of formative and inflexional ch A considerable number of Dra- particles and pronominal fragments. vidian verbal themes, prior to the addition of inflexional forms, are it trisyllabic ; bat it will generally be found that the first two syllables or have been expanded out of one by the euphonic insertion or addition of of a vowel ; whilst the last syllable of the apparent base is in reality a formative addition, which appears to have been merely euphonic in nal origin, but which now serves to distinguish transitive verbs from in- are transitives. In some instances the first syllable of the verbal theme 00t contains the root, whilst the second is a particle anciently added to it 0W and compounded with it for the purpose of expanding or restricting the Jar signification. The syllables that are added to the inflexional base are 368 those which denote case, tense, person, and number. ' Hence, whatever be the length and complication of Dravidian fu, words, they may invariably be traced up to monosyllabic roots, by a 191 careful removal of successive accretions. Thus, when we analyse ¢ perugugiradu,” Tam., i ncreases, we find that the final ‘adu,” repre- Asey sents the pronoun ‘it; ¢gir,’” is the sign of the present tense; and rll]ﬁ»l ¢perugu,’ is the base or verbal theme. Of this base, the final syllable ‘gu, is only a formative, restricting the verb to an intransitive or )pfl ko, neuter signification ; and by its removal we come to ¢peru,” the real the root, which is used also as an adjective or noun of quality, signifying greatness or great. Nor is even this dissyllable ¢peru,” the ultimate | the condition of the root: it is an euphonized form of per,” which is found in the adjectives ¢ per-iya’ and ¢ per-um,’ great ; and a lengthened AL f the but monosyllabic form of the same is ¢pér. Thus, by successive have agglutinations, a word of six syllables has been found to grow out of f the one In all these forms, and under every shape which the word can ) att assume, the radical element remains unchanged, or is so slightly changed, that it can readily be pointed out by the least experienced scholar. PeCt 1 J » 0 ROOTS. The root always stands out in distinet relief, unobscured, un- 1 absorbed, though surrounded by a large family of auxiliary affixes. ( This distinctness and prominence of the radical element in every word i3 a characteristic feature of all the Scythian tongues (e.g., of the ¥ Turkish and the Hungarian) ; whilst in the Semitic and Indo-Euro- P‘ pean tongues the root is frequently so much altered that it can scarcely be recognised. 11 It is desirable here to explain in detail the manner in which Dra- vidian roots, originally monosyllabic, have been lengthened by the 18 nsertion or addition of euphenic vowels, or by formative additions, or a in both ways. al Y Evpnonic LENeTHENING oF Roors. — There are two modes in which the crude Drévidian root is euphonically lengthened. al First, by the ingertion of an' euphonic vowel between the two th initial consonants of the original base. It has already been shown €X that in the Tamilian or oldest DraAvidian system of sounds, a double 1l consonant cannot stand at the beginning of any word or syllable. A vowel must be inserted. or ome of the consonants must be omitted. 001 This is invariably the rule in Tamil, and generally so in Telugu ; and Jo in the event of a vowel being inserted in the double initial, it is fol obvious that the root, if a monosyllable, will become a dissyllable. ad Thus, ¢viri, Tam., ¢o expand, the origin of ¢viral, « Jinger, was pul probably at first ¢vri. The double consonant ‘vr,” was incapable b of being pronounced by Tamil organs, and was, therefore, converted C . into a dissyllable by the insertion of a vowel. The probability of the change in this instance is strengthened by the circumstance that il where the Tamil has ¢ viral, a finger, the Telugu has first ¢ vrélu,’ and finally ‘vélu” In the same manner, where the Tamil has ‘maram,’ b a tree (Canarese ‘ mara’), the Ku has ‘ mranu,” softened in Telugu into the ‘méanu ;’ and where most of the Dravidian languages have ‘tala,’ head, Car the Ku has ‘tlava; the final ‘vu’ being an euphonic addition to ‘tla The best proof that in the Dravidian languages dissyllables 4 S&] Cay were in this manner lengthened from momnosyllables, is furnished by fou) the circumstance that all Sanserit words and particles which com- able mence with a double consonant, are altered on this very plan when thy they are borrowed by the Tamil; e.g., “tripti, satisfaction, is converted (l\a into ‘tirutti’ and ‘pra,’ the preposition before, into ¢ pira.’ This gl euphonic lengthening out of the crude base by the insertion of an 1 euphonic vowel, is apparent also in those bases which become poly- lent syllabic by the further addition of formatives. Thus, ‘tirumbu,” 7o eup| turn, is compounded of ¢ tiru,’ the original base, and ‘bu’ (euphonized ot EUPHONIC LENGTHENING OF ROOTS. 1 J 1 in the intransitive into ‘mbu’), a formative. ‘tiru’ itself, however (answering to °tiri,’ fo wander, and to several other related words), was doubtless originally a monosyllable, probably tri. We find this very form in the Telugu transitive verb, which is ‘tri-ppu,’ corres- ponding to the Tamil ‘tiru-ppu; with which we may compare the Greek ¢ 7péme,” a word which is almost identical in sound as well as in signification. The second mode in which crude Dravidian roots are lengthened, is by 'the addition of an euphonic vowel to the base. This euphonic ) addition to the final consonant takes place in grammatical Telugu and Canarese in the case of all words ending in a consonant, whatever be the number of syllables they contain. It Vowel additions to roots which contain two syllables and upwards, are made solely for the purpose of helping the enunciation ; but when {1} the additions which have been made to some monosyllabic roouts are il examined, it is found that they are intended not merely for vocalisa- lle tion, but rather for euphonization. When it is desired merely to help the enunciation of a final d consonant, ‘u’ is the vowel that is ordinarily employed for this pur- 2 pose (in Malayalam ‘a’), and this ‘u is uniformly elided when-it is ‘n.’l b 18 followed by another vowel: but ‘u 1s not the only vowel which is le added on to monosyllabic roots, though perhaps it is most frequently [as met with; and in some of the instances under consideration, it ble becomes so intimately blended with the real base, that it will not ted consent to be elided ; e.g., ‘adu,” Tamil, ¢o be near, the final ‘u’ of the which does not admit of elision, though the crude base is probably hat ¢ad.” Next to ‘u,’ the vowel which is' most commonly employed is nd ¢i; then follows ‘a; then ‘e’ or ‘ei.’ Verbal roots borrowed from the Sanscrit, have generally ‘i’ added to the final consonant in all 1, nfo the Dravidian languages; to which the Telugu adds ‘nchu,” and the i, Canarese ‘su, formatives which will be noticed afterwards. Thus, ¢sap, Sans., to curse, is in Tamil < &abi; in Telugu, ¢sabinchu;’ in 10 bles Can., ‘sabisu.’ On comparing the various Dréividian idioms, it is by found that all these auxiliary or enunciative vowels are interchange- able. Thus, of Tamil verbs in ‘a,’” ‘kada,’ to pass, is in Telugu, ‘gada- ik ? hen chu; ‘mara,’ fo forget, is in Canarese ‘ mare: of Tamil verbs in ¢i 2 ‘kadi, to bite, is in Telugu ‘kara-chu; ‘geli,) o win, is in Canarese grted This gilln. Of Tamil verbs in ‘ei) ‘mulei, to sprout, is in Telugu, ‘moluchu.’ These final vowels being thus interchangeable equiva- fan lents, it is evident that they are intended merely for the promotion of olf £ euphony, and as helps to enunciation, that they are not essential parts ied 152 ROOTS. of the themes to which they are suffixed, and do not add anything to their meaning. 1] FormaTivE AvpITioNs 10 Roors.—Formative suffixes are ap- pended to the crude bases of nouns, as well as to those of verbs. They 1l are added not only to verbal derivatives, but to nouns which appear to be primitive; but they are most frequently appended to verbs properly so called, of the inflexional bases of which they form the f last syllable, generally the third. Whatever may have been the origin of these particles, they now serve to distinguish transitive 0 ———— verbs from intransitives, and the adjectival form of nouns from that i In which stands in an isolated position and is used as a nominative. Tamil, in which these formatives are most largely used and most fully il developed, the initial consonant of the formative is single when it [ marks the intransitive or neuter signification of the verb, or that 18 form of the noun which governs verbs or is governed by them: when 1l it marks the transitive or active voice of the verb, or the adjectival al form of the noun, viz., that form of the noun which is assumed by I the first of two nouns that stand in a case relation to one another, the Pa initial consonant of the formative is doubled, and is at the same time to changed from a sonant into a surd. The single consonant, which is be characteristic of the intransitive formative, is often euphonised by prefixing a nasal, without, however, altering its signification or value. The Tamilian formatives are—(1.) ‘gu’ or ‘ngu,’ and its transitive ¢ kku, answering to the Telugu ‘chu’ or ‘nchu; (2.) ‘éu’ and its transitive ‘&u’ or ‘chu; (8.) ‘du’ or ‘ndu,’ and its transitive ‘ttu; 0 and (4.) ‘bu’ or ‘mbu,” with its transitive ‘ppu.’ th Though I call these particles ¢ formatives,” they are not regarded by in this light by native grammarians. They are generally suffixed even to the imperative, which is supposed to be the crude form of the illh verb; they form a portion of the inflexional base, to which all signs al of gender, number, and case, and also of mood and tense, are the appended ; and hence it was natural that native grammarians should Wl regard them as constituent elements of the root. I have no doubt, ad however, of the propriety of representing them as formatives, seeing y i that they contribute nothing to the signification of the root, and that g, it is only by means of a further change, 7.e., by being hardened and Wi doubled, that they express a grammatical relation, viz., the difference Sa n which subsists between the transitive and the intransitive forms of n verbs, and between adjectival and independent nouns. I In this particular perhaps more than in any other, the high finj it FORMATIVE ADDITIONS TO I#OOTS. 153 &1 grammatical cultivation of the Tamil has developed a tendency to imitate the Indo-European tongues by retaining syllables of which it has lost the original distinctive meaning, and combining such syllables B after a time with the radical element of the word, or using them for a hey new purpose. gar I proceed to consider the various formatives more particularly, ths with examples of their use and force. the (1.) ‘gu’ or ‘ngu,’ with its transitive ‘kku.’ Tamil examples ; the ¢ peru-gu,’ intrans., fo become increased, peru-kku,” trans., to cause to tive So also in tnerease; ‘ ada-ngu,’ to be contained, ¢ada-kku,’ to contain. hat the case of dissyllabic roots, e.g., ¢4-gu,’ to become, ¢ a-kku, to make; In ‘ni-ngu,’ to quit, ‘ni-kku, ‘o put away. There is a considerable number of nouns, chiefly trisyllabic, in which the same formative is ully n it employed. In this case, however, there is no difference between the that isolated shape of the noun and the adjectival shape. Whatever par- hen ticle is used, whether ¢gu,” ‘ngu,” or ‘kku,” it retains its position in ival all circumstances unchanged. Examples: ¢ pada-gu,” a boat, ¢ kura- ngu,’ a monkey, ‘sara-kku,” any article of merchandize. From a com- by the parison of the above examples, it is evident that ‘ng’ is equivalent time to ‘g’ and euphonized from it; and that ‘ng,’ equally with ‘g’ h 18 becomes ‘kk’ in a transitive connection. | by In a few instances, ¢ kku, the transitive formative, is altered in colloquial Tamil usage to ¢chu,’ according to a law of interchange lue. itive already noticed ; e.g., ‘ kajkku,” to boil (crude root ¢ kay ’), is generally ] its written and pronounced ‘kaychu.’ This altered form of the sign of the transitive, which is the exception in Tamil, is in Telugu t; the rule of the language, ‘kku’ being regularly replaced in Telugu by ‘chu’ rded ixed In Telugu the intransitive formative ‘gu’ is not- euphonically altered into ‘ngu’ as in Tamil; but an obscure nasal, the half f the ‘anuswara, often precedes the ‘ gu,” and shows that in both languages signs 1o the same tendency to nasalisation exists. It is remarkable, that whilst the Tamil often nasalises the formative of the neuter, and never iould Oﬂb@ admits a nasal into the transitive formative, the Telugu, in a large el number of cases, nasalises the transitive, and generally leaves the neuter in its primitive, un-nasalised condition. Thus in Telugu, that aad whenever the base terminates in ‘i’ (including a large number of Sanscrit derivatives), ‘chu’ is converted into ‘nchu;’ though neither rence in this nor in any case does the ‘kku’ of the Tamil change into ms ol ‘ngku’ FK.g., from ‘ratti, double, the Tamil forms °ratti-kka’ (in- finitive), fo double; whilst the Telugu form of the same is ‘retti-ncha.’ |ﬁh ¢ manni-ncha,’ 7o forgive, in Telugu, corresponds in the same manner to 154 ROOTS. its a the Tamil ‘manni-kka.’ In some cases in Telugu the euphonic nasal and is prefixed to ‘chu,’ not after ‘i’ only, but after other vowels besides. Verby Thus, ¢ perugu,’ o increase, neut., is the same in Tamil and in Telugu. few" but instead of finding ¢peru-chu’ to be the tramsitive or active (corresponding to the Tamil transitive ¢ peru-kku’), we find ¢ penchn,’ “Ding is ey corrupted from ‘peru-chu:’ so also instead of ‘pagu-kku,” Tam., to for 1] divide, we find in Telugu ¢panchu,’ for ¢pagu-nchu.’ The identity of the Tamil ‘k’ and the Telugu ‘ch’ appears also nasal 3 ‘ning from the circumstance that in many cases ‘vu may optionally be used in Telugu instead of ¢chu’ This use of ‘vu’ as the equivalent s'mlp of ‘chu’ points to a time when ‘gu’ was the formative in ordinary ing (1} bO\ use in Telugu as in Tamil ; for ‘ch’ has no tendency to be converted into ‘v,” ‘b, or ‘p, whilst ‘k’ or ‘g, constantly evinces this ten- 5 } alwa dency to change into ‘v, not only in Telugu, but also in colloquial Tamil ; and ‘v’ is regularly interchangeable with ‘b’ and its surd doat ¢ 2 Beco I conclude, therefore, that ‘gu’ was the original shape of this sleep, formative in the Drividian languages; and that its doubled, surd I shape, ‘kku,’ the formative of transitives, was softened in Telugu into tive, ¢chu,” and in Canarese still further softened into ‘Su.’ 100t (2.) “$u, and its transitive ‘$Su,” pronounced ¢chu.—This formative “tar. “nily is very rare in Tamil, and the examples which the Telugu contains, though abundant, are not to the point, inasmuch as they are apparently o § tray altered from the older ‘ku’ and ‘kku,’ by the ordinary softening process by which ‘k’ changes into ‘6, and ‘kk’ ipte ‘ch. . A Tamil e, example of this formative is seen in ¢adei-8u, to lake r¢fuge, of which or' the the transitive is ‘adei-chu,’ to enclose, to twine round. 130) du’ or ‘ndu,” with its trausitive form ‘ttu.’—There appears to trapy be no difference whatever between this formative and the other three, ¢ gu, 2556 % S u,’ or ‘bu, in meaning or grammatical relation; and as ‘gu’ is the euphonized in the intransitive to ¢ ngu, so is ‘du,’ to ‘ndu; whilst in nap the transitive the doubled ¢d’ (and its equivalent ‘nd’) changes by &?,‘ rule into ¢ tt.” The euphonic change of ‘dn’ to ‘ndu,’ has so generally 11y, taken place, that ‘ndu’ is invariably used instead of ‘du’ in the for- but matives of verbs; and it is only in the formatives of nouns that ‘du, 0] the more primitive form, is sometimes found to have survived. The formative ‘gu’ remains unaltered in the adjectival form of fing nouns; but ¢du’ changes into ‘ttu,” when nsed adjectivally, in the Noyy same manner as in the transitive voice of verbs. Tamil examples of ‘kip this formative ;—¢ tiru-ndu,” fo become correct, ©tiru-ttu,’ fo correct; Ing ¢ maru-ndu,’ medicine, adjectival form of the same, ‘maru-ttu,’ e.g., ¢ maruttu-(p)pei, @ medicine bag. The primitive unnasalised ‘du’ and Oll"] o FORMATIVE ADDITIONS TO ROOTS. 15 9 sl its a;djectival ‘ttu, are found in such words as ¢ eru-du,” a bull, an ox, des. and ‘eru-ttu-(p)plittu,” the fastening of an ox’s traces. N early all the 191, verbs which take ‘du,’ or ‘ndu,” as a formative are trisyllabic. Of the tive few dissyllabic verbs of this class in Tamil, the most interesting is hn, ‘nindu,’ ¢o swim, of which I consider ‘ni’ as the crude form. ‘ Nindu, , 10 is evidently an euphonized form of ‘nidu,” (‘du’ changed into ‘ndu’); for the verbal noun derived from it, ‘nittal, swimming, is without the als0 nasal, and the Telugu uses ‘nidu, for the verb itself, instead of ‘nindu.’ 7 be I have little doubt that the ‘du,” or ‘ndu’ of this word, is Jent simply a formative, and that the crude primitive base is ‘ ni, answer- ary ing to the Greek vé-w, the Latin ‘no, ‘nato’ and also to ‘nau,’” Sans. tted a boat, of which the Sanscrit does not contain the root. ten Derivative nouns formed from verbs which have formative suffixes 1uial always prefer as their formative the transitive suffix, or that which qurd doubles and hardens the initial consonant. Thus from °tiru-ndu,’ #o become correct, is formed ¢ tiru-ttam,’ correction; and from ¢ tii-ngu,’ fo this sleep, ¢ ti-kkam, sleep. surd In some instances the crude root of a verb is used as the intransi- 1nto tive, whilst the transitive is formed by the addition of ‘ttu’ to the root. H.g. ‘padu,) to lie down, ‘padu-ttu, to lay; ‘tar, to be low, tive ‘tar-ttu,’ to lower; ‘mil’ (Telugu ‘nilu’), fo stand, niru-ttu’ (for ‘nilu-ttu’), to establish. In such cases the Canarese uses ‘du’ instead ains, ntly of the Tamil ‘ttu,’ e.g., ‘tal-du, fo Jower, instead of ‘tar-ttu.”’ This 1ing transitive formative is sometimes represented as a causal; but it will gl be shown in the section on ¢ The Verb, that ¢vi’ (euphonically ¢bi, hich or ‘ppi’) is the only real causal in the Dravidian languages. In all the cases now mentioned, where ¢ ttu’ is used as the formative of the £s 10 transitive by the Tamil, the Telugu uses chu’ or ‘pu.’ I class under the head of this formative all those nouns in which 1168 the ‘cerebral’ consonants ‘d, ‘nd, and ‘tt,’ are used in the same ;117 15 5t 1 manner and for the same purpose as the dentals ¢d,” ‘nd, and ‘tt A e.g. ‘ kuru-du,’ blindness, adjectival form of the same, ¢ kuru-ttu, blind ; 5 by ‘ira-ndu,’ two, adjectival form, ¢ira-ttu,” double. The Telugu hardens 1ally for but does not double the final d’ of such nouns; e.g., ‘0d-u,’ @ leak, ¢ Otl, leaky. / ‘du” In some instances in Tamil, the hard rough ‘R, when used as a p of final, seems to be equivalent to ‘du,’” or ‘du,’ and is doubled and pro- nounced with a ‘t; eg., ‘kina-ru, o well, ‘ kina-rRRu’ (pronounced ) the ¢ kinattru’), of @ well. On this point, however, see Section on Nouns.— les of Increment ¢ti, or ‘attu.’ n ]Wt ;' 0l (4.) ‘bu’ or ‘mbu,” with its transitive ‘ppu.’—In Canarese, ‘bu,’ the original form of this intransitive suffix, has been softened into ‘vu,’ and 'and 156 ROOTS. il in Tamil ‘bu,” has universally been euphonized into ¢mbu. This Al Tamilian formative ‘mbu,” is in some instances softened in Telugu It nouns into ‘mu.’ The “bu’ or ‘mbu’ of Tamil verbs is superseded by (d} “vu’ or ‘gu’ in Telugu; and the forms answering to the Tamil tran- 0r sitive ¢ ppu’ are ‘pu’ and ‘ mpu,’ rarely ¢ ppu.’ Example of the use of this formative by a verb:—¢nira-mbu,” Tam. n ¢o be full, ‘nira-ppu,’ o fill; of which the crude base ‘nir,” re-appears in the related verb ‘nir-ei, fo be full, or to fill. The Telugu has i the ‘nindu’ instead of ‘nirambu; but the transitive ‘nimpu,” answers very i nearly to the Tamil ‘nirappu.’ Example of a noun in ‘mbu’ and ¢ ppu;—F‘iru-mbu,” Tam., ©ron, adjectival form, ¢iru-ppu, of iron, e.g., qu ‘iruppu-(k)kol,” an wron rod. In Telugu ¢irumbu’ is softened into the ‘inumu,’ adjectival form ¢inupa.’ The Canarese still adheres to the of original form of this suffix, generally softening ‘b’ into ‘v,” but leaving par it always unnasalised ; e.g., Canarese ‘havu, a snake, properly ¢ pavu?’ Tamil ¢ pAmbu,” nasalised from ‘pébu; adjectival form °péappu,’ e.g., gyl‘ tt ¢ pappu-(k)kodi, the serpent bannmer: Telugu, still further altered, ‘pamu. This example clearly illustrates the progress of the formative 1l in question, and confirms the supposition, that it was merely euphonic 0n | in its origin, and that it was by degrees that it acquired the character of a formative. Sllp Tefg It has been mentioned that the Telugu uses ‘pu’ or ‘mpu’ as a formative of transitive verbs, where the Tamil uses ¢ ppu.’ It should Ve Ty be added that even in those cases where the Tamil uses the other formatives previously noticed, viz., ‘kku’ and ‘ttu,” the Telugu often prefers ‘pu’ Compare the following infinitives in Tamil and in brg Telugu, e.g. ‘méykka,” Tam. to feed cattle, ‘mépa,” Tel.; ‘nirutta,” Tam. to establish, ¢ nilupa,” Tel. ko Where ‘kku’ in Tamil, and ‘pu’ in Telugu o b are preceded by ¢1i,” this formative becomes in Telugu either ‘mpu’ or ]iar]; ‘nchu;’ e.g., compare oppuvi-kka,” Tamil, to deliver over, with the corresponding Tel. infinitive, ¢ appavi-mpa,” or ‘ appavi-ncha.’ Pesp fron It appears from the various particulars now mentioned, that tran- sitive verbs, and nouns used adjectivally, must have been regarded by 1oy the primitive Tamilians as possessing some quality in common. The helg COng common feature possessed by each, is doubtless the quality of transition; for it is evident that when nouns are used adjectivally there is a tran- sition of the quality or act denoted by the adjectival noun to the noun Syll:. substantive to which it is prefixed, which corresponds to the transition frop ($¢) of the action denoted by the transitive verb to the accusative which it I governs. It is manifest that the various particles which are used as forma- tives do not essentially differ from one another either in signification, Pring FORMATIVE ADDITIONS TO ROOTS. 157 This in the purpose for which they are used, in the manner in which they are affixed, or in the manner in which they are doubled and hardened. lngu L ] It was euphony only that determined which of the sonants ‘g S ¢d; ¢d; or ‘b, should be suffixed as a formative to any particular verb fran- oI noun. [am Possibly, indeed, the use of these formatives originated altogether in considerations of euphony. The only point in which a grammatical Jears has principle appears to exist, is the doubling of the initial consonant of the formative to denote or correspond with the putting forth of energy very which is inherent in the idea of active or transitive verbs, as distin- and el guished from intransitives. From the statements and examples given above, it may be concluded into ) the that wherever Dravidian verbs or nouns are found to terminate in any of the syllables referred to there is reason to suspect, that the first Aving part of the word alone constitutes or contains the root. The final ave: syllables ¢ gu,’ ‘ngu,’ ‘kku; ‘8u,’ ‘chu; ‘du, ‘ndu, ‘¢tu; ‘du,’ ‘ndu,’ ] €‘g', ‘ttu; ‘bu, ‘mbu,’ ‘mpu,’ ‘pu, ‘ppu; ‘mu,’ ‘ vu; may, asa general ered, rule, be rejected as formative additions. afive This rule will be found on examination to throw unexpected light honic on the derivation and relationship of many nouns which are commonly acter supposed to be primitive and independent, but which, when the syllables referred to above are rejected, are found to be derived from or allied to ' 958 verbal roots which are still in use. I adduce,as examples, the following hould Tamil words:—‘kombu,” @ branch, a twig ; ¢vémbu,’ the Margosa tree ; other ¢ vambu,’ abuse ; ‘pambu,’ a snake. As soon-as the formative final, often ‘mbu,” is rejected, the verbs from which these nouns are derived are pd 0 brought to light. Thus, ‘ko-mbu,’” @ #wig, is plainly derived from Tam ‘ko-y,’ to pluck off, to cut ; ‘vé-mbu,’ the Margosa tree, is from ©veé-y, elugt o be umbrageous, to screen or shade (the shade of this tree being pecu- Pu’ or liarly prized) ; ¢va-mbu,” abuse, is from ¢vei,” properly ‘va-y’ (cor- ih the responding to the Canarese ¢ vayyu’), fo revile ; ¢ pa-mbu,’ @ snake, is from ‘ph-y,” to spring. In these instances, the verbal base which is - fran- now in use ends in ‘y,” a merely euphonic addition, which does not od by belong to the root, and which disappears in the derivatives before the The consonants which are added as formatives. gition; The same principle applied to nouns ending in the other formative ) tran- syllables will be found to yield similar results; e.g., ‘par-andu,’ a hawk, ) ﬂ()ﬂn from ¢para,’ to fly; and ¢ kirangu,’ @ root, from ‘kir, o be beneath, the psition ‘1’ of which, though long in Tamil, is short in the Telugu ‘kinda,’ below. hich it RepurricaTiON oF THE FiNAL CoNsONANT oF THE RooT.—The formd- principle of employing reduplication as a means of producing gramma- catiOllz 158 ROOTS I tical expression is recognized by the Dravidian languages, as well as by those of the Indo-European family ; though the mode in which the man char reduplication is effected and the objects in view are different. It is in to by Tamil that this reduplication is most distinctly apparent, and it should here be borne in mind, that when a Tamil consonant is doubled, it is the The final consonant of a Tamil changed from a sonant into a surd. eveg root is doubled—(1) for the purpose of changing a noun into an thre adjective, showing that it qualifies another noun, or putting it in the their guag genitive case, e.g., from ‘madu,’” an oz, is formed ‘matt-u (t)tol, ox- hide ; (2) for the purpose of converting an intransitive or neuter verb siimy hag into a transitive, e.g., from ¢6d-u,” to rum, is formed °ottu, to drive ; (3) for the purpose of ‘forming the preterite,” e.g., ‘ tag-u,’ to be fit, finst “takk-a,’ that was fit ; and (4) for the purpose of forming derivative 86001 pean nouns from verbal themes, e.g., from ‘erud-u,” fo write, is formed ¢ erutt-u,’ @ letter. [See this subject further elucidated in the sections on €S8er famy ‘ The Noun’ and ‘ The Verb.”] It is remarkable, that whilst the Indo- European tongues mark the perfect tense by the reduplication of the case Jirst syllable, it is by the reduplication of the last letter that the that the { Dravidian languages effect this purpose; and also, that whilst the Tibetan converts a noun into a werd by doubling the last consonant, conty this should be a Dravidian method of converting a verb into a noun. sylla The rationale of the Dravidian reduplication is, that it was felt to be a to thy natural way to express the idea of transition both in the act and in gy the result. Ace Up to this point it has been found that all Dravidian pelysyllabic Hel roots are traceable to a monosyllabic base, lengthened either by Dex euphonic additions and insertions, or by the addition of formative par- exhi ticles. An important class of dissyllabic bases remains, of which the liken, second syllable is neither an euphonic nor a formative addition, but an or e inseparable particle of specialisation, into the nature and use of which It we shall now inquire. T00tg e L4 ParticLEs oF SproranisatioN.—The verbs and nouns belonging lawy to the class of bases which are now under consideration, consist of a Hely, monosyllabic root or stem, containing the generic signification, and a second syllable, perhaps the fragment of a lost root or lost post-posi- Tl tion, by which the generic meaning of the stem is in some manner dl\ 1de modified. The second syllable appears sometimes to expand and some- paly times to restrict the signification, but in some instances, through the 1ntey absence of synonyms, its force cannot now be ascertained. As this be d01 syllable is intended in some manner to specialise the meaning of the l{lt 1e1 root, I call it ‘the particle of specialisation.’ glamy PARTICLES OF SPECIALISATION. 159 ell as The principle which is involved in the use of this partiéle, and the h the manner in which it is carried into effect, remarkably correspond to a A characteristic feature or law of the Semiticlanguages, which it appears hould to be desirable to notice here somewhat particularly. As far back as 1 18 the separate existence of the Semitic family of languages can be traced, [amil every root is found to consist of two syllables, comprising generally 0 an three consonants. When Semitic bi-literal roots are compared with n the their synonyms, or corresponding roots, in the Indo-European lan- o guages, and especially with those which are found in Sanserit, a ' verh simpler and more primitive root-system hasbeen brought to light. It Ire has been ascertained, in a large number of instances, that whilst the be f, first syllable of the Hebrew root corresponds with the Sanscrit, the vative second syllable does not in any manner correspond to any Indo-Euro- omed pean synonym. It is found also, that the second syllable has not any Ons 01 essential connection with the first, and that a considerable number of Indo- families of roots exist in which the first syllable is the same in each f the case, whilst the second continually varies. It is therefore inferred t the that in such cases the first syllable alone (comprising two consonants, ¢ the the initial and the final, together with the vowel used for enunciation) onant, contains the radical base and generic signification, and that the second noun. syllable, perhaps the fragment of an obsolete word, has been appended to the first and afterwards compounded with it, for the purpose of obea ad in giving the generic signification a specific and definite direction. According to this view, which appears to be in the main correct, -Hebrew roots are to be regarded mnot singly and separately, as inde- yHabiC er by pendent monads, but as arranged generically in clusters or groups, exhibiting general resemblances and special differences. The family (; paI‘- ch the likeness resides in the first syllable, the radical base ; the individuality or special peculiarity in the second, ¢the particle of specialisation.’ hut an which It is true that in some instances the second syllable of Semitic roots meets with its counterpart in the Indo-European languages, as well as the first, or even instead of the first ; but the peculiar rule or law now referred to is found to pervade so large a portion of the nging "Hebrew roots, that it justly claims to be considered as a characteristic ¢ of & of the language. and & ;t'pOSi‘ Thus, there is a family of Hebrew roots signifying generally ‘to divide,” ‘to cleave,’ ‘to separate, &c. The members of this family are jannet ‘palah,’ < palah, ¢ palag,’ ‘pala,’ “ palal;’ and also (through the dialectic | some- interchange of ‘1’ with r’), ‘parash,’ ‘paras;’ Chaldee ‘peras.’ It cannot 0 oh the be doubted that in all these instances the first syllable “pal’ or ‘par,’ or Ag this rather ‘p-r,” ‘p-1’ (for the vowel belongs not to the root, but to the of the grammatical relation), expresses merely the general idea of ¢ division it 160 ROOTS. whilst the second syllable (which is in some instances a reduplication of the final consonant of the bi-literal) expresses, or is supposed to express, the particular mode in which the ¢ division’ or ¢ partition’ is effected. an The first syllable, which is the same in all the members of this group of roots, is that which is to be compared with synonyms in other lan- guages, whilst the second syllable is merely modal. In this instance we not only observe a distinct analogy between the Hebrew roots, ¢p-r,’ bl ‘p-1; and the Greek ¢wop-w,” the Latin ¢par-s,” ¢par-tis, and the Sanscrit ¢ phal,’ o divide, but we also discover the existence of a dis- fin tinct and remarkable analogy with the Dravidian languages. Compare 8y with the Hebrew ‘p-r,” ¢p-1” the Tamil ‘piri,’ ¢o divide, and ‘pal, a of part; ‘pila’ and por, fo cleave; as also ‘pagir’ and ‘pagu,’ o See also the ¢ Glossarial Affinities. portion out, to divide. On turning our attention to the root-system of the Dravidian lan- an guages, we are struck with the resemblance which it bears to the ani Semitic root-system referred to above. We find in these languages ane groups of related roots, the first syllables of which are nearly or any ann wholly identical, whilst their second syllables are different in each instance, and in consequence of this difference produce the required degree of diversity in the signification of each member of the group. We also find in these languages, as in Hebrew, that the generic particle or common base, and the added particle of specialisation, are so thaf The specialising par- conjoined as to become one indivisible etymon. {0 ticle, which was probably a separable suffix or post-position at first, 01 has become by degrees a component part of the word,—and this word, ba so compounded, constitutes the base to which all formatives and all 0 inflexional particles are appended. whi This root-system exists in all the languages of the Dravidian S family, but its nature and peculiarities are especially apparent in the Tamil. 1 Out of many such groups of related Tamil roots, I select as Varig illustrations two groups which commence with the first letter of the of gy alphabet. in g Show 1. Roots which radiate from the syllable ¢ad »’ ({7 6 ¢ adu to come near ; also ‘adu,’ transitive, fo unite. It iy ada to joim, to join battle. ﬂlOSe ada-ngu to be contained, to enclose; verbs formed from ‘ada,” the preceding ada-kku } verb, by the addition of the formatives ‘ngu’ and ‘kku. When able adei to attain, to get in, to roost : transitive, to enclose. adeisu to take refuge, from €adei,” with the addition of the formative ‘su; of sp also ‘adeigu’ to be close together, to be crowded. T adar WPpey aduk-(ku) to place one thing upon another, to pile wp. This verb is properly Sf)nan 161 PARTICLES OF SPECIALISATION. of ‘aduk,’ but final ‘k’ in Tamil is always vocalised by the help of ‘u,” and often doubled, as in this instance, before receiving the ‘u ’ d andu (Telugu ar_l':su), to approach. This verb seems to be identical with ‘adu,’ \]]P ! the first in the list, and euphonized from it by the insertion of the nasal. 106 I It is obvious that all these roots are pervaded by a family resem- e blance. All contain the generic notion of ‘nearness,’” expressed by the s first or base syllable ‘ad; whilst each, by means of the second e syllable, or particle of specialisation, denotes some particular species of nearness. 0 /1] 2. Roots which radiate from the base syllable < an 7— all anu to touch. the anl to put on, to adorn. Lges anei o connect, to embrace ; as a noun, @ weir, @ dam. anavu or to cleave to. (‘vu’ is probably an euphonic addition.) ch annu to lean upon. (From this verb is derived ‘annal’ or ‘annan’ an red elder brother, one to lean upon, a derivation as poetical as it is reliable). up. cle The generic idea signified by the base syllable ‘an’ is evidently 80 that of ‘contact;’ and this group differs from the previous one as actual pﬂI‘ ‘contact’ differs from ¢contiguity’ or ‘nearness.’ Probably ‘ani,’ st a nail, a fastening, is derived from the same verb, and it appears pro- ord, bable also that this is the origin of the Sanscrit ‘ani’ or ‘ani, the pin [ all of an axle. At all events it seems a more natural derivation than that which is given by the Sanscrit grammarians, viz., from ‘ana,” fo fian sound. the The illustrations given above prove, that the second syllables of the a8 various verbs now adduced have not been added merely for purposes the of euphony, but have been appended in order to expand, to restrict, or in some manner to modify and specialise the signification. It was shown in a previous part of this section, that the vowels ¢ gl Sy ‘e, and ‘ei’ are sometimes added euphonically to monosyllabie roots. It is obvious, however, that this is not the only purpose for which those vowel additions are used; and it is of importance to know that (i il when they are merely euphonic they are found to be interchange- able with other vowels, whereas when they are used as particles of specialisation they retain their individual character more firmly. ‘Uy The examples already given may suffice to illustrate the use of appended wowels as specialising particles. Syllables ending in con- ‘peﬂ) sonants, especially in ‘1’ and ‘r, are also used very frequently for i M 162 ROOTS. this purpose ; and it seems desirable here to adduce examples of the use of particles of this class. The following examples are mostly The other from the Tamil, in which ‘1’ and ‘r’ may stand as finals. dialects add ‘i’ to the final consonant of each of these particles. The Tamil requires this euphonic addition of ‘u’ when a word ends in the hard, rough ‘R, or.in any consonant besides the nasals and semi- vowels. Each root being considered either as a verb or as a noun according to circumstances, I give examples of nouns as well as of verbs. Some of the following roots, though used as verbs, are more commonly used as nouns, and some, though used as nouns, are more commonly used as verbs. Some of the examples, again, are used either as nouns only or as verbs only :— FinaAL PARTICLES. VERBS. Nouns. ar suvar, @ wall valar, fo grow 1r tulir, fo sprout ugir, a finger nail ur nudur-u, Tel., the forehead ar idar, a flower leaf pugar, to praise i magir, o rejoice tamir, sweetness, Tamil. aR-u idar-u, to trip kinar-u, @ well iR-u muyir-u, the red ant A SN al sural, to wharl iral, the liver il kuyil, to utter a sound tigil, @ fright ul pagel-u, Tel., to break al tuval, to bend tinggal, the moon il madil, e fort wall ul irul, darkness urul, fo roll ui;-(ku) kaduk-(ku), to suffer pain koduk-(ku), @ sting Of all the fourteen specialising particles ending in consonants, of which examples have now been adduced, only one appears occasionally b to be used as an equivalent for a vowel addition. ‘ar’ alternates with fl ‘ei; e.g., ‘amar, Tam., to rest, and ‘amei,” are apparently equivalent. Vi The verb to grow, also, is in Tamil ¢valar, and in Canarese ‘bale,’ f which in Tamil would be ¢valei.’ Similar instances, however, abound in Hebrew, without invalidating the general principle ; and even with L respect to the latter of the two Dravidian illustrations, there is a th marked distinction in Tamil between ¢ valar,’ and a related theme, il ¢vilei, ‘valar’ meaning fo grow ‘upwards, as a man or a tree,’ tl whilst ¢vilei, means fo grow ‘as a crop ; hence as a noun ©vilei’ €) means «a field. I¢ I here subjoin an example of another peculiar and interesting set th of groups of roots which are found in the Dravidian languages, and 3(] INTERNAL CHANGES. 163 which are formed upon a plan differing considerably from that which has now been explained. The roots referred to are dissyllabic, but they contain only one con- sonant, which is preceded and followed by a vowel. This consonant appears to represent the ultimate or radical base, whilst the initial and final vowels alter in accordance with the particular shade of significa- tion which it is desired to convey. When we compare ‘idu,” Tam., to press or crush, ‘odu,’ to squeeze, to bring into a smaller compass, and 4 “idi,’ to bruise, to beat down, as also “adi, to beat ; or ‘odi, to break d in two, and ‘udei’ (pronounced ‘odei’), to break open ; we cannot b} avold the conclusion that the first four roots are closely related )1 members of the same family or groap ; that the last two roots are in like manner mutually related ; and that possibly the whole of them have an ulterior relationship, in virtue of their possessing in common the same nucleus or radical base, the central consonant ‘d,’ and the same generic signification. DrAvipiaN RooTs susTaiN No INTERNAL CHANGE ON RECEIVING ForMATIVE or INFLEXIONAL ADDITIONS, OR IN COMPOSITION.—In general this rule is so strictly adhered to, and the deviations from it are so few and unimportant, that it may be regarded as a characteristic of the Dréividian root-system, and a counterpart of the rigid unchange- ableness which characterizes Scythian roots. The vowels of Dravidian roots belong as essentially to the radical base as the consonants. They neither belong, as in the Semitic lan- guages, to the system of means by which grammatical relations are expressed, nor are they modified, as in the Indo-European languages, by the addition of inflexional forms. of In the Semitic languages the radical base is destitute of vowels, and |3,Hy by itself unpronounceable. The insertion of vowels not only vocalises pith the consonants of the root, but constitutes it a grammatically inflected ent. verb or noun, the signification of which varies with the variation of the interior vowels. )‘(11,97, und In the Indo-European languages grammatical modifications are produced by additions to the root; and though in the earliest period of s‘viﬂl the history of those languages, the root, generally monosyllabie, is sup- 15 & posed to have remained unaltered by additions and combinations, yet 161116,‘ 766 the existence of that rigidity is not capable of direct proof ; for on examining the Sanscrit, Greek, Latin, and German, the most faithful nle] representatives of the early condition of those languages, we find that g ¢ the root vowels of a large proportion of the words are modified by the addition of the suffixes of case and tense ; and in particular, that the : ad M 2 164 ROOTS. reduplication of the root, by which the perfect appears usually to have been formed, is often found either to alter the quantity of the root- vowel, to change one vowel into another, or entirely to expunge it. In the Scythian family of tongues, not only does the vowel belong essentially to the root, but it remains unalterable under all circum- stances. Neither the vowel nor the consonant (or consonants) of which the root is composed, sustains any change or modification on the addition of the signs of gender, number, and case, or of person, tense, and mood ; which are successively agglutinated to the root, not welded into combination with it. This rigidity or persistency is characteristic also of the roots of the Drévidian langnages, with a few exceptions which will shortly be men- tioned. In general, whatever be the length or weight of the additions made to a Dravidian root, and whether it stands alone or is combined with other words in a construct state, it is represented as fully and faithfully in the oblique cases as in the nominative, in the preterite and future as in the present tense or in the imperative. I proceed to point out the principal exceptions to this rule. 1. Buphonic Exceptions. (1.) Some exceptions are purely enunciative, and consist only in such changes as are necessary to enable Dravidian organs to enunciate double consonants. See the portions of this section in which the lengthening of roots by the euphonic insertion or addition of vowels is explained. (2.) A second class of euphonic exceptions is connected with one of the ‘minor dialectic peculiarities > noticed at the end of the section on Sounds. It consists in the occasional omission or mutation of the final consonant of a root when it is followed by a formative or in- flexional particle. Most of the instances which I have noticed, occur in Canarese or Telugu, especially in the latter. They are such as the following, viz., ‘eddu, Tel, an oz, instead of ‘erdu’ (in Tamil I erudu’); ¢ penchu,’ Tel., to increase, instead of ¢ perunchu’ (in Tamil ‘perukku’); ‘biddu,” Canarese, kaving fallen, for ‘bildu’ (Tamil ‘virdu’); and ‘tiddu, Can., o correct, for ‘tirudu’ (in Tamil ‘tiruttu’). This 1 omission, or softening, has no relation to grammatical expression, and appears to have arisen chiefly from haste in pronunciation. A few a I examples of this change are found even in Tamil; e.g. ¢ vandu, Aaving In this case the omitted ‘r’ ol come, instead of ¢ varndu’ or ¢varundu.’ has not a place in the imperative of the second person singular, which i ¢ is ¢va,’ come, not ‘var; and hence it might be doubted whether the ‘1’ veally belongs to the root, or whether it is only an euphonic T fy i b INTERNAL CHANGES. 165 addition. I suspect, however, that this ‘r’ is radical, for the Telugu ¢ pA? imperative singular is 1a, not ‘va,’ asif from ‘vara; and we find in the Rajmahal dialect that fo come is bAri. In Tamil also the imperative of the second person plural is ¢ vAr-um.’ Hence ¢vandu,’ having come, seems really to be a softened form of ¢ varndu.” Another example appears to be furnished by a Tamil verb meaning to give, which is ‘tar,” in the infinitive, the present, and the future; ‘ta,” in the imperative singular; and ti’ in the preterite, e.g. *tandén’ (for ‘tarndén’), 7 gave. The resemblance or identity of the Tamil ‘ta’ and the Sanscrit ‘da,” o give, might lead us to suppose ‘ta’ to be a Sanscrit derivative, in which case the ‘r 3 referred to would be an euphonic addition. It is difficult, however, to suppose that this ‘r’ 1§ has been added euphonically, and the difficulty is increased by the d d circumstance that in every part of this verb, with the exception of the imperative, the form of the root which we find to be used, is not 19 “ta” but ©tar-u. Hence it seems open to conjecture that ‘tar-u’ is not derived from the present shape of the Sanscrit, though related to it, but that it springs from an older source, of which a trace remains in the Greek ‘éip-ov, and possibly also in the Hebrew base, ¢tan.’ (8.) A third class of euphonic exceptions to this rule is connected 1n g te with another of the ‘minor dialectic peculiarities’ referred to. consists in the occasional softening or rejection of the medial con- he sonant of a dissyllabic root or verbal noun, together with the els coalescence of the vowels that preceded and followed it. It has been shown that ‘g’ has a tendeucy to be softened into ‘v’ and then to ne disappear, and that ¢§’ changes in the same manner into ‘y,” when it 01 sometimes become absorbed. When either of these consonants is a he medial, it is apt to be thus softened down and rejected. Thus, 10 “dogal-u,” Canarese, skin, becomes in Tamil ¢tol; ¢pefar, Canarese, a pur name, becomes in Tamil first ‘pejar’ and then ‘pér.’ So in Tamil, he ‘togup-pu,’ @ collection, is softened into ¢tép-pu,” which has the mil restricted meaning of @ collection of trees, a tope. mil (4.) The most important class of euphonic exceptions to the Uy general rule of the unchangeableness of the root appears at first sight s to correspond to a characteristic usage of the Indo-European languages, aml and especially of the Sanserit. In those languages the quantity of the fow root vowel is sometimes altered when the crude or abstract noun is g changed into an adjective. Thus in Sanscrit ¢ Dravida,” a gentile i appellation, becomes ¢Dravida,” pertaining to the Dravidas, the ‘a’ CA D A hich changing into a; and if the vowel is naturally long, as the ¢ e n 1he ‘ Véda,” it becomes a diphthong when the word is changed to an adjec- gnic tive; e.g., ¢ Vaidika,” pertaining to the Védas. In Tamil we discover a 166 ROOTS class of changes which, though in reality they are purely euphonic and unconnected with grammatical relations, appear at first sight to resemble the above-mentioned Indo-KEuropean usage Dravidian roots, though originally monosyllabic, have very generally taken a dis- 1l syllabic form by the insertion or addition of a vowel which is intended to facilitate enunciation In such cases the first syllable, always a short one, represents the crude root, the added vowel constitutes the euphonic suffix; eg., per-u, great ¢ kar-u, bleck ar-u, precrous In Tamil, especially in the old poetical dialect and in the speech of the peasantry, such dissyllabic adjectives, or nouns of quality, are often 1l found to sustain a further change The final euphonic vowel is rejected, and to compensate for its loss, the interior vowel of the root is lengthened Thus per-u becomes per ¢kar-u “Bar, and g ar-u ‘ar In the same manner ¢ or-u,’ one, become or; and ‘ir-u two ¢ A 1T This lengthened monosyllabic form is considered to be peculiarly elegant, and is much used in combinations It is also used il more frequently than the dissyllabic form as a concrete noun of quality Thus ¢kar, black, is much used by itself to denote ‘the 0f or ¢ Coromandel monsoon or ‘the rice orown at that th ralny season, season This euphonic lengthening of the root vowel and rejection al of the final will be found to throw licht in the derivation of some il & nouns of quahty ; eg., ‘par, desolate, a wilderness, is evidently derived from ‘ para, old. When the final consonant of the crude root belongs to the class of hard letters (‘k, <&’ “t,” ‘t, ‘p, ‘®’) it cannot be enunciated by ’% Tamilians without the help of an appended vowel and in such I cases, though the interior vowel of the root is lengthened, the final S remains £ paé-u,’ green, becomes in poetical and vulgar b 2 usage, not ¢ pas,” but pas- This final ‘u,” however, being retained th solely for the sake of enunciation, is considered like the Hebrew br sh’vd,” as only half the length of an ordinary short vowel L1 At first sight the change in the interior vowels of Dravidian to roots now pointed out may appear to resemble the usage of the by Sanserit ; but on further examination 'the resemblance 1s found to dis- appear It is evident that the Dravidian increase of quantity is 118 wholly euphonic, and not, like that of the Sanserit, a means of pro- th ducing grammatical modification for though that form of the Dravi- dian noun of quality, or adjective, in which the root vowel is 1ig lengthened, is more frequently employed as a concrete noun than the st; older dissyllabic form, yet the dissyllabic form is also used as a concrete, ¢ and both forms are used indiscriminately as adjectives; from which it is obvious that the difference between them pertains, not to gram- Tog P — - INTERNAL CHANGES. 167 matical relation, but only to considerations of euphony. Thus, though ‘ar,’ precious, is more often used than ‘ar-u,’ to signify pre- ciousness, or that which s precious, yet ‘ar-u’ also is used by the poets in the same sense ; and either ‘ar-u’ or ‘a4r’ may optionally be used in composition as an adjective. 2. Real Exceptions. It has been stated as a general rule that the internal vowels of f Dravidian roots sustain no internal change on recciving formative or 1 inflexional additions or in composition; it has also been stated that 1S deviations from this rule exist, but that they are few and unim- ot portant. The apparent exceptions mentioned above have been shown i to be merely euphonie. I proceed to notice the few real exceptions which are observed. (1.) In most of the Dravidian languages the quantity of the root- he el vowels of the pronouns of the first and second persons, both singular The nominatives of and plural, is shortened in the oblique cases. 16 of those pronouns are long; e.g., ‘nin,’ Tamil, Z, ‘nam, we; ‘ni,’ thou, ‘nir,’ you. But in Tamil, Canarese, Malayilam, and Tulu, in 20 on all the oblique cases the vowels are shortened before receiving the e suffixed inflexional particles. Thus, in Canarese, to me is not ‘ nin- The el a-ge,” but ‘nin-a-ge; ¢o thee is not ‘nin-a-ge, but ‘nin-a-ge’ Telugu, Gond, and Ku generally retain the quantity of the root- vowel unaltered: e.g., in Telugu we find ‘ni-ku,’ ¢o thee, as well as a5 ‘ni, thou; but in the accusative, ‘ nin-u,” thee, the quantity is altered by in the same manner. 1ch The only other instance of a similar shorten- nal ing of the root-vowel of a Dravidian word is that which is supplied by the numerals. The radical portion of the Tamil numeral ‘miindru,’ gi\f [4 three, is ‘mu; but this becomes mu,” when used as an adjectival ned In prefix, as in ‘ muppattu,’ thirty, and ‘munniru,’ three hundred. Al like manner when ‘ar-u,’ siz, is used adjectivally, it is shortened to ‘ar-u; and ¢ér-u,’ seven, to ‘eru; eg., ‘arubadu,’ sizty, ¢eru- an badu, seventy. The oblique case of a noun or pronoun is identical the with that form which the same noun or pronoun takes when it is ds used adjectivally ; and hence both these classes of instances fall under Vlb the same rule. prO" The shortening of the root-vowel takes place in the personal pro- V) nouns and numerals alone. All other pronominals and nouns sub- ] 18 stantive adhere to the general rule of the Dréividian languages of the preserving the root-vowels unaltered. retes Singularly enough, this exception from the general rigidity of the it root-vowels is a Scythian exception, as well as a Dravidian one. In the il 168 ROOTS. S Seythian .Version of the Behistun tablets, whilst the nominative of the pronoun of the second person is ‘ni, ¢hou, as in the Drividian lan- U guages, the possessive case is ‘ni,” thy, and the accusative ‘nin,” thee, ) o corresponding in quantity to the Dravidian oblique cases; e.g., Telugu and Tulu, ‘nin-u,’ ¢thee, High Tamil ¢ nin,’ thy, and ‘ ninnei,’ thee. le (2.) Another class of exceptions appears in those few instances in i which the Tamil shortens the quantity of the long vowel of the root 001 in the preterite. This shortening is occasionally observed in the gr Canarese, but the best illustrations are those which are furnished by 18 the Tamil: e.g., ‘végu,’ properly ‘vé,” ¢o burn, has for its preterite par- ticiple, not ‘végundu’ or ‘véndu,” but ‘vendu; ‘négu,’ to be in pain, ¢! properly ‘nd,” has in the preterite, not ‘négundu’ or ‘néndu,” but fi ‘nondu ;' and ‘kan,’ fo see, not ‘kandu,’ but ‘kandu.’ The two classes of exceptions mentioned above evidently accord, 1l as far as they go, with a prevalent usage of the Indo-European lan- guages, inasmuch as they are examples of the shortening of the interior I vowels of the root on receiving the addition of the inflexional particles, i to make compensation for the additional weight which is thus 10 imposed on the root-vowel. O (3.) A third class consists of instances in which the quantity of a ah vowel is lengthened when a verbal root is formed, directly and without Ta K any extraneous addition, into a noun. The alteration which the root of If vowel sustains is prior to any inflexional additions being made. 0 any formative particle is added to 2 verbal root to convert it into a noun, the quantity of the root-vowel remains unchanged. The 1y lengthening of the root-vowel to which I refer takes place only in fe (some of) those cases in which the verbal base itself is used as a noun. 1 Thus, the verb ‘ked-u,’ ¢o destroy or to become destroyed, may become a an verbal noun by the addition of the formative ‘di, e.g., ‘kedudi, aly destruction, in which event the root-vowel remains unaltered ; but the ¢hs verbal base may also be used without addition as a verbal noun, iu W which case ‘ked-u’ is lengthened into ¢kéd-u.’ ‘k The following Tamil examples of the lengthening of each of the five primary vowels will suffice to illustrate this usage : thy From ¢pad-u,’ to suffer, is formed ¢ pid-u,’ a suffering ; from ‘min,’ st to shine, ‘min, a star; from ‘Sud-u,” to burn, ¢§td-u,” keat ; from Ya ‘per-u,’ to obtain, ‘pér-u, a benefit obtained ; and from ‘kol,” to recewve, ‘ kol reception. I am not aware of the existence of a sinﬁ]ar rule in any of the Scythian languages, but it is well known to the Sanscrit (e.g., comp. ‘vach,’ to speak, with ‘vach,’ @ word ; ‘mar’ (‘mri’), to die, with ‘méara,’ death) Nevertheless, I can scarcely think it likely that it is from the E | F INTERNAL CHANGES. 169 fthe Sanserit that the Drividian languages have derived a usage which Jan prevails ameng them to so great an extent, and which has every thee, appearance of being an original feature of their own. It may here be luu added, that in two instances in Tamil the root vowel has been lengthened in the imperative of verbs: eg., ¢ tara,’ fo give, is in the 8 11 imperative plural ‘tar-um,” give ye; and ‘vara,’ fo come, ¢ vir-um, 100t come ye. I consider this change as euphonic, not pertaining to the i 16 grammatical expression, for in the parallel forms in Telugu the vowel ' by is short, e.g., ‘ra-(m)mu,’ come ye. par In concluding this section it seems desirable to notice an apparent pain change of interior vowels occurring in Tamil, which has been supposed ' but to accord with the Sanserit change of a short vowel into a long one, and of a naturally long vowel into a diphthong, on a noun being changed cord, into an adjective. It consists in the change of ‘pafum,’ green, in cer- lan- tain conjunctions, into ‘peim ; e.g., ‘peim-pon,” excellent (literally or10r green) gold. It is certain, however, that this is merely an euphonic jcles, change, in no way affecting grammatical relations. ‘pasum,’ green, is thus not derived, as Beschi supposes, from ¢pasumei,’ greenness, by the omission of the final ‘ei; for ‘mei,’ not ¢ei,” is the particle by which of a abstracts are formed, and the ‘m’ is the most essential part of that thout particle. It is derived from ¢pa8,’ green, the crude adjective or noun 100t of quality, with the addition of ‘um,’” the sign of the aorist, commonly ik called ‘the future,’ by which it is made an aoristic relative participle, no & a class of participles which all Scythian tongues delight to use as adjec- tives. It has already been shown that ¢S’ when medial, has a The tendency to be softened into ‘y,” and then to disappear altogether ; and }y i joul. in consequence of this tendency, ‘pasum’ naturally became ‘pajum,’ and this again, by an easy change, and one which in pronunciation is ne & udi, almost imperceptible, ¢ peim.’ We have a parallel instance of this ‘ change in the noun ‘kaSuppu,’ bilterness, which may optionally be 1 the written and pronounced ¢keippu; ‘kaSuppu’ changing first into n, ‘kajuppu,” and then into ‘keippu.’ It should also be observed that ‘peim’ has not superseded ¢ pasum,’ { the though it may optionally be used instead of it, for ¢pasum’ also is still in use ; and this proves that both forms are grammatically equi- i, valent. from | 1 f the 0P néfﬂ)’ | ) the % i onl Sem gen lang ey SECTION IIL fam iy ped for THE NOUN, i gr I~ this section it will be my endeavour to investigate the nature and affections of the Dravidian noun, with the view of ascertaining its P Al method of expressing the relations of gender and number, and the principles on which that method proceeds, together with the charac- wh teristics and origin of its case-system, or system of means for expressing 1ou; It will be shown the relationship of nouns with other parts of speech. L at the close of the section oh ¢ The Verb,” how derivative nouns are are formed from verbal roots; and the various classes of participial nouns ther 10 will then also be investigated. fiye Part I.—Gender and Number. (N lat 1. GENDER. an When the Indo-European laws of gender are compared with those ] mp‘ of the Scythian group of tongues, it will appear that in this point, as ligge in many others, the Dravidian languages accord more closely with the ‘he Scythian than with the Indo-European family. neyt In all the more primitive Indo-European languages, not only are tnly words that denote rational beings and living creatures regarded as Worg masculine or feminine, according to the sex of the objects referred to, h] fjh but also inanimate objects and even abstract ideas have similar sexual add; distinctions attributed to them; so that many nouns which are naturally destitute of gender, and which ought therefore to be regarded ﬁcat‘ as neuters, are treated by the grammars of those Janguages as if the that : objects they denote were males and females, and are fitted not with the neuter, but with masculine or feminine case-terminations, and with Sc),( pronouns of corresponding genders. This peculiar system is a proof the of the highly imaginative and poetical character of the Indo-European ]311 0 mind, by which principles of resemblance were discerned in the ot ¢ GENDER. 171 midst of the greatest differences, and all things that exist were not only animated, but personified. A similar remark applies to the Semitic languages also, in which the same or a similar usage respecting gender prevailed. In the progress of the corruption of the primitive Indo-European languages, a less imaginative but more natural usage gained ground: nevertheless, in a majority of the modern colloquial dialects of this family, both in Europe and in India, the gender of nouns is still an important and difficult section of the grammar, and a standing im- pediment in the way of the idiomatic use of those languages by foreigners. On the other hand, in the Manchu, Mongolian, Turkish, and Finnish families of tongues—the principal families of the Seythian ure group—a law or usage respecting the gender of nouns universally it prevails, which is generically different from that of the Indo-European the and the Semitic idioms. In those families, not only are all things rac which are destitute of reason and life denoted by neuter nouns, but no ng nouns whatever, not even nouns which denote human beings, are regarded as in themselves masculine or feminine. All nouns, as such, oWl are are neuter, or rather are destitute of gender. In those languages s there is no mark of gender inherent in, or inseparably annexed to, the nominative of any noun (the crude root being generally the nomina- tive) ; and in none of the oblique cases, or post-positions used as case-terminations, is the idea of gender at all involved. The unimagi- native Scythians reduced all things, whether rational or irrational, animate or inanimate, to the same dead level, and regarded them all as hose impersonal. They prefixed to common nouns, wherever they found it {, 48 necessary, some word denoting sex, equivalent to ‘male’ or ¢female,’ ‘he’ or ‘she;” but they invariably regarded such nouns as in themselves , the neuters, and generally they supplied them with neuter pronouns. The 10 only exceptions to this rule in the Scythian languages consist in a few words, such as ¢ God,” “man,” ‘ woman,” ¢ husband,” ¢ wife,” which are so d a8 d to, highly instinet with personality that of themselves, and without the axuﬁl addition of any word denoting sex, they necessarily convey the signi- fication of masculine or feminine. - are When our attention is turned to the Dravidian languages we find 1led £ the that whilst their rules respecting gender differ generally from those of the Indo-European group, they are not quite identical with those of the yith Scythian. It seems probable, however, that the particulars in which with the Dréavidian rules respecting gender differ from those of the Scythian Proof Opeﬂll languages, and evince a tendency in the Indo-European direction, are not the result of Sanscrit influences, of which no trace is perceptible ] thp S A A 5 e 172 THE NOUN. all n in this department of Drividian grammar, but have arisen from the Tami progressive mental cultivation of the Dravidians themselves. more Dravidian nouns are divided into two classes, which Tamil gram- word marians denote by the technical terms of ¢high caste’ and ¢ caste-less’ lpon nouns, but which are called by Telugu grammarians ¢ mahat,” majors which and ‘a-mahat,” minors. ¢ High-caste’ nouns, or ‘majors, are those whicl which denote the celestial and infernal deities and human beings,’ comm or, briefly, all things that are endowed with reason; and in all the Ir Dravidian dialects (with a peculiar exception which is found only in word the Telugu and the Gond) nouns of this class are treated in the denty singular as masculines or feminines respectively, and in the plural as gend epicenes, that is, without distinguishing between masculines and the | feminines, but distinguishing both from the neuter. The other class of peasa nouns, called ¢caste-less, or ‘minors,” includes everything which is destitute of reason, whether animate or inanimate. . This classification vidiay Tivers of nouns, though not so imaginative as that of the Indo-European and or ge; Semitic tongues, is decidedly more philosophical; for the difference Tefaln between rational beings and beings or things which are destitute of reason, is more momentous and essential than any difference that Thu, exists between the sexes. The New Persian, which uses one plural- singu] the ising particle for nouns that denote animated beings and another and different one for things that are destitute of life, is the only Un-Dré- 1oup) vidian Janguage in which nouns are classified in a manner which is in Seyt 1 {or any degree similar to the Dravidian system.* The peculiar Dravidian law of gender which has now been described would appear to be a Pén fr 3 result of grammatical cultivation; for the masculine, feminine, and gende epicene suffixes which form the terminations of Dravidian ‘high-caste’ nouns, are properly fragments of pronouns or demonstratives of the the b, third person, as are also some of the neuter formatives. It may, jﬂsepar lnent ¢ indeed, be stated as a general rule that all primitive Dravidian nouns the Drﬁ are destitute of gender, and that every noun or pronoun in which the idea of gender is formally expressed, being a compound word, is A shoulq necessarily of later origin than the uncompounded primitives. The Tgmb technical term by which such nouns are denoted by the grammarians is ¢ pagu-padam,’ divisible words, i.e., compounds. Hence the poetical and fy dialects, which retain many of the primitive land-marks, are fond of gendey arg W]] discarding the ordinary suffixes of gender or rationality, and treating the fen * This is not the only particular in which the Drividion idiom attributes greater importance than the Indo-European, to reason and the mind. We make Ty our bodies the seat of personality. When we are suffering from any bodily Subst, ailment, we say ‘Z am ill;’ whereas the Drividians denote the mind—the con- Ordinal scious self or ‘4tmd’—when they say J, and therefore say, more philosophically, ‘my body is ill. GENDER. 173 n the all nouns, as far as possible, as abstract neuters. Thus in poetical Tamil ¢Dév-u, God, a erude noun destitute of gender, is reckoned more classical than ‘Dév-an,’ the corresponding masculine noun. This fram- Josy word is a Sanscrit derivative, but the same tendency to fall back ' upon the old Scythian rule appears in the case of many other words jors those which are primitive Drividian nouns; e.g., ‘irei, « king, a word ings, which is destitute of gender, is more classical than ¢irei-(v)-an, the commoner form, which possesses the masculine singular termination. | the In the modern Tamil which is spoken by the educated classes, the ly in words which denote ‘sun’ and ‘moon’ (‘sfiriy-an’ and ¢$andir-an, n the ral as derived from the Sanscrit ¢ siirya’ and ¢ chandra,’) are of the masculine s and gender, in accordance with Sanscrit usage and with the principles of the Brahmanical religion; but in the old Tamil of the poets and the ass of ich 18 peasants, ‘fijiyiru,’ the sun, and tinggal,’ the moon, both pure Dra- vidian words, are mneuters. All true Drividian names of towns, cation rivers, &c., are in like manner destitute of every mark of personality nand or gender. orence In some few instances the Malayalam and the Canarese retain the primitive laws of gender more faithfully than the Tamil. it of that Thus, in the Tamil word ¢peiyan,’ a boy, we find the masculine singular termination ‘an ; whereas the Malayilam (with which agrees , Tural- the Canarese,) uses the older word ¢peital,” a word (properly a verbal or and 1-Drd- noun) which is destitute of gender; to which it prefixes in a thoroughly Scythian manner words that signify respectively ‘male’ and ¢female,’ ) isin to form compounds signifying ¢ boy’ and ¢ girl;’ e.g., ‘4n peital,” @ boy, yidian ¢ pen peital,’ a girl. ) be & The nature and origin of the terminations which are used to signify ) and gender in the various Dravidian dialects, will be enquired into under .cﬂ.ste’ the head of ¢ Number,” with the consideration of which this subject is of the inseparably connected. Under this head I restrict myself to a state- ) lﬂaya ment of the general principles respecting gender, which characterize pouns the Drayidian languages. i the A peculiarity of the Telugu, which appears also in the Génd, Dl’ll | 18 should here be mentioned. Whilst those dialects agree with the other The members of the Drividian family in regarding masculines and feminines uﬁriﬂﬂs and both combined as constituting in the plural a common or epicene etical gender ; they differ from the other dialects in this respect, that they jond of are wholly or virtually destitute of a feminine singular, and instead of the feminine singular use the singular of the neuter. This rule includes in its operation pronouns and verbs as well as fributes [ make substantives, and applies to goddesses and queens, as well as to ordinary women. The Telugu possesses, it is true, a few forms which fhe con- are appropriate to the feminine singular, but they are rarely used, and phicaﬂﬁ'> 174 THE NOUN. that only in certain rare combinations and conjunctures. ¢ He’ and ‘it are the only pronouns of the third person singular, which are forn ordinarily made use of by fourteen millions of the Telugu people; iden a Ve and the colloquial dialect does not even possess any provnoun, equivalent to our pronmoun ‘she,” which is capable of being applied mat addi to women of the lower as well as of the higher classes. Ordinarily every woman is spoken of in Telugu as a chattel or a thing, or as we forn are accustomed to speak of very young children (e.g., ¢4 did so and with utll so’), apparently in the supposition either that women are destitute of reason, or that their reason, like that of infants, lies dormant. Whilst leas each woman taken singly is treated by Telugu grammar as a chattel Ppro or as a child, women taken collectively are regarded with as much wh respect as by the other Drividian dialects. In the plural they are g honoured with the same ‘high-caste’ or ‘rational’ suffixes and pronouns that are applied to men and gods. ‘A The Canarese and Malayilam agree in this point with the Tamil, pro and regard women, not in the plural only, but also in the singular, as to g pertaining to the class of ‘rationals:’ accordingly in those languages beco there is a feminine singular pronoun equivalent to ¢she,” which corres- age, ponds in the principle of its formation to the masculine ¢he. With e those languages agrees the Ku, which, though the near neighbour of the Telugu and the Gond, pursues in this respect a politer course than an either, 1 In the idioms of the Tudas and Kotas, the rude aborigines of the of Nilgherry hills, there is no pronoun of the feminine singular; but e instead of the feminine, those dialects appear to use not the neuter but the masculine. This extraordinary usage reminds one of the an employment in the Old Hebrew of the same pronoun, ht,’ to signify any, both ¢ he’ and ¢ she. lhe1 itsel 2. NUMBER. CHSQ; The Dravidian languages recognize only two numbers, the singular laty and the plural. The dual, properly so called, is unknown, and there ‘mg is no trace extant of its use at any previous period. Several of the Moy, languages of this family contain two plurals of the pronoun of the first the person, one of which includes the party addressed as well as the party eup} of the speaker, and which may therefore be considered as a species of 1 I}o dual, whilst the other excludes the party addressed. As, however, this peculiarity is restricted to the personal pronouns, it will be eleg examined in that connexion. Under the head of ¢ Number,” we shall but enquire into the Dravidian mode of forming the masculine, feminine, (any and neuter singular, and the epicene and neuter plural. NUMBER—MASCULINE SINGULAR. 175 "and (L) Masculine Singular.—It has already been intimated that the ) are formatives by which the gender of nouns is occasionally expressed, are ople; identical with the terminations of the demonstrative pronouns. From Doun, a very early period of the history of these languages, particles or for- plied matives of gender were suffixed to the demonstrative bases, by the atily addition of which suffixes demonstrative pronouns were formed. Those 18 We formatives of gender were not originally appended to or combined ] and with substantive nouns; but their use was gradually extended as their ite of utility was perceived, and nouns which included the idea of gender, Vhilst learned to express that idea by suffixing the gender-terminations of the hattel prououns, whereby they became appellative nouns. The manner in much which all these suffixes are added will be sufficiently illustrated by the instance of the masculine singular. Y are nouns The masculine singular suffix of the Tamil is ‘an,” ‘4n,” or ‘6n.’ ‘An,’ the shorter formative, is that which appears in the demonstrative amil, pronoun ‘avan’ (‘a-(v)-an’), ke, and by suffixing any of these formatives to an abstract or neuter noun, the noun ceases to be abstract, and at, 88 1ages becomes a concrete masculine-singular appellative. Thus ‘mipp-u,’ OITES- age, by the addition of ‘an’ becomes ‘mipp-an,’ an elder, literally With age-he, or age-man; and from ¢ Tamir’ comes ¢ Tamir-an,” a ZLamilian, w of a Tamil-man. , than These and similar nouns are called generically ¢ compound or divi- sible words’ by Tamil grammarians. They are obviously compounded of a noun—generally a noun of quality or relation—and a suffix of f the gender, which appears also to have been a noun originally. ; but euter In the instances which have been adduced, the suffix of gender is f the annexed to the nominative or casus rectus; but in many cases it is annexed to the oblique case or inflexional base, viz., to that form of onify the noun to which the case signs are suffixed, and which when used by itself has the meaning of the genitive. When the inflexion, or oblique case, is employed instead of the nominative in compounds of this nature, it generally conveys a genitival or possessive signification: e.g. gular there ‘malelyinan’ (‘ malei-(y)-in-an’), a mountaineer, literally a man of the jf tlle mountavn ; ‘pattinattan’ (‘ pattin’-att’-an’), a citizen, literally a man of the city. Sometimes, however, the genitival ‘in’ is merely added o fist euphonically ; e.g., there is no difference in meaning between ‘villan,’ party 168 of a bowman, and ‘villinan’ (‘vill'-ir-an’), which is considered a more elegant form. revel Words of this description are in some grammars called ¢ adjectives ; 1l be gl but they are never regarded as such by any native grammarians: they ]iﬂi!lfx cannot be simply prefixed for the purpose of qualifying other words ; 176 THE NOUN. and it is evident from their construction that they are merely appella- il tive nouns. il A subdivision of appellatives consists of words in which the suffixes “a su of gender are aunexed to adjectival forms; e.g., ‘kodiya-n,’ a cruel man I regard words of this class as participial nouns, and they will pe be investigated in the section on ¢ The Verb,” under the head of ¢ Appel- In lative Verbs; but whatever be the nature of ‘kodiya’ (the first part of the compound), ¢kodiya-n,” is certainly not an adjective; for before f it can be used adjectivally we must append to it the relative participle Cel ‘ana,’ that is; e.g. ‘kodiyan-ana,’ that is a cruel man, and as the com- 8 pound cruel man, cannot be called an adjective in English, neither is ‘kodiyan’ an adjective in Tamil: it is properly an appellative noun. It may be said that the neuter plural of this word, viz., ¢ kodiya,” may t be prefixed adjectivally to any substantive: but ¢ kodiya,” cruel things, the neuter plural of ‘kodiyan,” is not identical with the adjective 1 ‘ kodiya,” cruel, but totally distinct from it, though so similar in appear- ance. I The ‘a’ of the former word is the neuter suffix of plurality ; whereas the ‘a’ of the latter is that of the possessive case and of the relative participle, as will be shown at the close of this section (see d ¢ Adjectival Formatives’) and in the section on ¢ Verbs.’ ik . Another species of Tamil appellative nouns is said by Beschi to be f.g, formed by annexing suffixes of gender to verbal roots, e.g., ‘oduvéan,’ the a reader, from ¢6du,’ to read ; but this, I believe, is an error. Those €a words are to be regarded as participial nouns, and ‘6duvan,’ is literally T he who will read, t.e., he who is accustomed to read. ¥ In the same manner ‘6dinan,’ is the participial noun of the preterite T tense, and means ke who read or is accustomed to read : ‘6dugindravan,’ 1l the corresponding present participial noun, ke who reads, belongs to the same class; and these forms are not to be confounded with appel- f lative nouns properly so called. On the other hand, such words as den ‘kappan,’ a protector, are true appellatives; but ‘kappan’ is not ST formed from the future tense of the verb (though ¢képpén’ means Ze ik will protect), but from ‘kappu, protection, a derivative noun, of which Cay the final and formative ‘ppu’ is from the same origin as the corresponding n final of ‘mippu,” old age. See the concluding part of the section on or ¢The Verb. bt The suffixes of gender which form the terminal portion of appella- 0N tive nouns vary somewhat in form; but they are one and the same in g origin, and their variations are merely euphonic. It is the vowel only Wi that varies, never the consonant. 1f When a neuter noun ends with a vowel which is essential to it, ay; of ¢ R NUMBER—THE MASCULINE SINGULAR. 177 ollg- and is incapable of elision, and also when a noun happens to be a long monosyllable, ‘an,” or in poetry ¢6n,” is more commonly suffixed than Ixes ‘an b In some cases ‘avan,’ ke, the full demonstrative pronoun, is 1] suffixed instead of its termination only; and this mode is thought will peculiarly elegant. Thus from ¢vill-u, @ bow, we may form ¢ vill-an,’ pel- “vill-an,” and ‘¢vill-6n, an archer, a bowman, and also °vill-avan.’ part Indeed ¢ 4n’ and ¢ 6n,’ have possibly been formed, not from ¢an,” but fore from ‘a-(v)-an,” by the softening of the euphonic ‘v, and the coales- iple cence of the vowels. This corruption of ‘avan’ into ¢4an,’ appears (0= systematically in the third person masculine singular of the colloquial erls Tamil verb; e.g., ‘pon-4n’ (not ¢ pén-avan’), ke went. oun, The Canarese masculine singular suffix ‘anu,’ is identical with may the Tamil ‘an,’ the additiou of ‘u’ being a phonetic necessity of the modern dialect. ings In the older Canarese, the termination which was ofive used was ‘am; a particle which is to be regarded as the equivalent ear- of “an ‘n’ and ‘m’ being interchangeable nasals. The Malayalam ity; is, in this particular, perfectly identical with the Tamil. f the The Telugu masculine singular formative is ‘d-u,” ‘ud-u,” or ‘ad-u; eg., ‘“vadu’ (‘va-adu’), he; and by suffixing the same for- (aee mative to any substantive noun, it becomes a masculine singular; o be e.g., ‘mag-adu,” @ husband, a word which is identical in origin with wén,' the Tamil ‘mag-an,’ ¢ son (the primitive and proper meaning of hose each word being a male). The masculine singular suffix of the rl‘ﬁuy Telugu often takes the shape of ¢ud-u, and in like manner the epicene plural suffix, which is in Tamil ‘ar-u, is often ‘ur-u’ in Telugu; but in these instances ‘a’ changes into ‘u’ through attrac- erite Mll,’ tion. ;rs to As the Tamil forms masculine appellatives by suffixing the demon- ppel- strative pronoun ‘avan,” so does the Telugu sometimes suffix its full demonstrative pronoun ‘vadu; eg., ‘chinna-vadu,’ @ boy (Tamil, ls 4 . 0b ‘8inna-(v)-an’), literally Ze who vs little. It is probable that the Telugu s e masculine singular suffix was originally ‘an’ or ‘an-u,” as in Tamil- Canarese. ‘adu,” ‘ud-u,” or ‘du, is found only in the nominative ,thh in correct Telug;ru, and it is replaced in all the oblique cases by ‘ani’ ding n 0 or ‘ni; and that this ‘ni’ is not merely an inflexional increment, but the representative of an old masculine singular suffix, appears on comparing it with ¢ri,’ the corresponding oblique-case suffix of the 3ella masculine-feminine plural, which is certainly” formed from ‘ar-u.’ 01D only When ¢ vaniki,’ ¢o Aim, i1s compared with its plural ‘variki,’ ¢o them, it is evident that the former corresponds as closely to the Tamil ‘avanukku’ as the latter to ‘avarukku; and consequently the ‘ni’ 0 i of ‘vaniki’ must bg significant of the masculine singular. Probably N s R—— | e i __ — 1 lod { 8 THE NOUN. the same termination survives in the demonstrative, °ayana,’ /e, a form which is more rarely used than ¢vadu. That the Tamil-Canarese masculine suffix ¢an,” and the Telugu ‘ad-u’ or ‘ud-u, were originally one and the same, will, I think, appear when the derivation and connections of both are inquired into. The Ku, though one of the most barbarous of the Dravidian dialects, throws more light than any other upon this point. It forms its demonstrative pronouns in a simple and truly primitive manner by prefixing ¢4, the demonstrative base, to common nouns, which sig- nify man and woman. Those nouns are Aafij-u, @ man, and ‘al-u, a woman ; and ¢ 4afij-u’ (compare Tam., ‘a(v)an’), literally that man, is used to signify 4e, and ¢ 44lu’ (compare Tam. ¢ a(v)al’) that woman, to C A signify she. The Ku anj -u,” @ man, is certainly identical with the Tamil noun ‘a4n,’ @ male: and we see the same root in the Ancient Can. ‘4nma,’ a husband, a ruler, and ‘Anmu,’ to be brave (compare the Tamil abstract noun ‘Aan-mei,’ strength). In the use to which this primitive root is put in the Ku word ¢4-Afij-u,” we cannot but see the origin of ‘an,’ the suffix of the masculine singular in most of the 5 Dravidian dialects. The final Gt n, and probably the entire termina- d tion ‘fiju,’ of the Ku word ‘é&fi-ju,” being merely euphonic, the root ¢ h appears to be ‘4n; and as ‘n and n’ have been shown to be interchangeable, ‘an’ must be regarded as only another form of 4an.’ ‘n,’ again, is not only often euphonised by suffixing ‘du’ (e.g., ‘pen,’ Tam., a female, colloquially and poetically °pend-u’), but it is also sometimes directly changed into ¢d,” of which we have an instance in the classical Tamil ‘ped-ei, a hen, a word which is derived by this process from, and is identical with, ‘pen,’ a female. Hence, the 1 Telugu suffix ‘ad-u,’ might naturally be derived from an older form C in ‘an,’ if it should appear that that form existed; and that it did la exist, appears from the vulgar use to the present day of ‘n’ instead a of ‘n’ in some of the oblique cases (e.g., ¢ vanni, him, instead of Il ‘vani’), and from the ‘half anuswara, or obscure nasal, which pre- n cedes ‘du’ itself in the speech of the vulgar and in the written com- positions of the pedantic; eg., ¢vandu,’ for ¢vadu, Ze. A close 1 connection is thus established between the Tamil-Canarese ‘an’ aud U the Telugu ‘ad-u,’ through the middle point ‘an. The only difficulty in the way of the perfect identification of the Ii ¢ formative all with the Ku, ‘4dj-u, @ man, and with the Tamil T “an, a male, lies in the length of the vowel of the latter words. a( Here again the Ku comses to our assistance; for we find that the i vowel was euphonically shortened in some instances in the very dialect in which the origin of the word itself was discovered., In Ku A NUMBER-—THE FEMININE SINGULAR. 179 a the ‘4 of ‘Afij-u’ is long, both when it is used as an isolated word and in the demonstratives, Aafij-u, ke, and €4al-u,” she; but when l the demonstrative pronoun is appended to, and combined with, the 1k, relative participle of the verb, so as to form with it a participial ed 9 noun, the ‘4 of ¢Afij-u’ is shortened into ‘a,’ and in this shortened kil form the connection of the Ku formative with the Tamil-Canarese is NS seen to be complete. Compare the Ku participial noun ¢ gitdfij-u,’ ke by who did, with the corresponding Canarese ‘géyidin-u; ¢gitar-u, Ku, 10 they who did, with ‘géyidar-u,’ Can., and also ‘gital-u,’ Ku, she who A did, with ¢ géyidal-u,” Can. y 8 10 (2.) Feminine Singular.—Though the Telugu and the Gond gene- the rally use the neuter singular to supply the place of the feminine ent singular, the other Dravidian dialects possess and constantly use a are feminine singular formative which is quite distinet from that of the this neuter. This formative is ‘al, in Tamil and Malayilam, ‘al-u,’ in the Canarese ; and by suffixing the sign of gender to the demonstrative the base, the feminine singular demonstrative pronoun ‘aval’ (a-(v)-al’), it she, is formed—a word which perfectly corresponds to ‘avan’ (a-(v)-an’), 00t he e A numerous class of feminine singular appellative nouns is formed al. by suffixing the same particle to abstract or neuter nouns in their o, crude state ; e.g., compare ‘mag-al, Tam., a daughter, with ¢ mag-an,’ also a son; and (with an euphonic lengthening of the vowel) “ill-al) a e house-wife, a wife, with ‘ill-an,” a husband. by The Telugu, in some few connexions, uses a feminine singular for- the mative which appears to be identical with that of the Tamil- Canarese. orm That formative is ¢&l-u,” which is used by the Ku more did largely than by the Telugu ; and its identity with the Tamil-Canarese end “al,” will be found to furnish us with a clue to the origin and literal | of meaning of the latter. As ¢anj-u,” in Ku, means a man, so ‘al-u, means @ woman : ¢ 4al-u, ske, is literally that woman. The same word p\e 01 ¢ al-u,” means @ woman, @ wife, in poetical and vulgar Telugu also ; and Jose in Gond there is a word which is apparently allied to it, ‘ ar,} a WOMAN. 2 aud Even in Sans. we meet with ali, a womaw’s female friend. It is evident that ¢al-u,” would be shortened into ‘al’ as easily as “anj-u’ into ‘an,’ and the constant occurrence of a cerebral <1’ in the Tamil and Canarese, where the Telugu has the medial 1 fully ﬂﬂﬂl accounts for the change of the one semi-vowel into the other. The Drda the unchanged form of this suffix appears in Telugu in such words as very ¢ manama-(r)-alu, a grand-daughier, compared with ‘manama-dun, a Ku grand-son. The abbreviation of the vowel of the feminine suffix, N2 54 180 THE NOUN. which is characteristic of the Tamil and Canarése, is exemplified in Telugu also, in the words ‘maradal-u,’ @ niece, and ‘kodal-u, @ daughter- an-law;” in which words the feminine suffix ¢al-u,’ is evidentical identical both with the Tamil-Canarese ‘al’ or ‘al-u,’ and also with “al-u,’ the older and more regular form of this suffix, which is capable of being used by itself as a noun. Probably the Tel. ¢ 4d-u,” adj., female, is identical in origin with “al-u,” through the very common interchange of ‘d’ and ‘1; an illus- tration of which we have in ‘kei-(y)-4lu,” Tam., ¢o use, which is con- verted in the colloquial dialect to ‘kei-(y)-Adu.’* The feminine singular suffix, ¢al’ or ‘al-u,” appears in Tamil and Canarese in the terminationsof verbs as well as in those of pronouns. The Telugu, on the other hand, which uses the neuter demonstrative instead of the feminine singular, uses the final fragment of the same demonstrative as the termination of the feminine singnlar of its verb. It may be remarked that in some of the Caucasian dialects, ‘n’ and ‘1" are used as masculine and feminine terminals, exactly as in Tamil: e.g., in Awar, ‘emen,’ is father, ¢ evel,’ is mother. There is another mode of forming the feminine singular of appel- lative nouns, which is much used in all the Dravidian dialects, and which may be regarded as especially characteristic of the Telugu. It consists in suffixing the Telugu neuter singular demonstrative, its ter- mination, or a modification of it, to any abstract or neuter noun. The neuter singular demonstrative being used by the Telugu instead of the feminine singular (¢t for ske), this neuter suffix has naturally in Telugu supplied the place of a feminine suffix ; and though in the other dialects the feminine pronounsare formed by means of feminine suffixes, not by those of the neuter, yet the less respectful Telugu usage has crept into f the department of their appellative nouns. In Tamil, this neuter-feminine suffix is ‘atti’ or tti.’ This will g appear on comparing ‘vellal-atti,” @ woman of the cultivator caste, with * Tt is more doubtful whether the Génd-Telugu, ‘4l-u,” @ woman, is allied to the Tamil common noun ‘4l @ person; and yet the existence of some alliance seems probable. ‘41, means properly a subject person, a servant—male or Jemale, a slave. It is derived from ‘41’ (Tel. “él-u’), to rule, and this seems a ri natural enough origin for a word intended to signify a Hindu woman. The ordinary Tamil word which signifies a woman is ‘pen,” the literal signification of which is desire, from the verbal root ¢ pen,’ to desire; but the word is generally restricted to mean a young woman, a bride. Hence, taking into consideration the subject position of women in India, the word ‘41, one who is subject to rule, a person whose sole duty is to obey, is as natural a derivation for a word signify- D ing @ woman, a female, as ‘pen;’ and perhaps more likely to come into general use as a suffix of the feminine singular. h NUMBER THE NEUTER SINGULAR. 181 I ‘vellal-an,” @ man of the same caste ; ¢ oru-tti, one woman, ¢ una, with “oru-(v)-an, one man, ‘wunus; and ‘vanna-tti, o washerwoman, with “vanna-n,” a washerman. “tt, a portion of this suffix, is erroneously th used in vulgar Tamil as a component element in the masculine appella- le tive noun ‘oruttan, one man, instead of the classical and correct ‘ oruvan.’ With this solitary exception its use is exclusively th feminine. 13- The same suffix is “iti” or ‘ti’ in Canarese, e.g., ¢ arasiti, a queen - (corresponding to the Tamil rasatti’), ¢okkalati, o farmer’s wife. The Telugu uses ‘adi’ or ‘di,’ e.g., ‘ komati-(y)-adi’ or ‘ komati-di,” a 9 d woman of the Komtr caste; ‘méala-di, a« Pariar woman ; ¢chinna-di, IN a girl. Ve It seems to me evident, not only that all these suffixes are identical, 16 but that the Telugu form of the demonstrative neuter singular, viz. iR “adi, ¢, which is used systematically by the Telugu to signify ske, is the root from whence they have all proceeded. 1 Another feminine singular suffix of appellatives which is occa- 1l sionally used in the Dravidian languages, has been derived from the imitation of the Sanscrit. It consists in the addition of ‘i’ to the J. crude or neuter noun ; and it is only in quantity that this ‘i’ differs nd from the long ‘1, which is so much used by the Sanscrit as a feminine It suffix. In the majority of cases it is only in connexion with Sanscrit er- derivatives that this suffix is used ; but it has also come to be appended he to some pure Dravidian nouns; e.g., ‘manei-(v)-i, Tam., a kouse-wife, he from ‘manei, a fouse; and ‘talei-(v)-i; Tam., a lady (compare lgll “talei-(v)-an,” @ lord), from ‘talel,” a head : compare also the Gond ot ¢perdgal,’ a boy, with ‘perdgi,’ a girl. This feminine suffix is not to by be confounded with i, a suflix of agency, which is much used iu the \} formation of nouns of agency and operation, and which is used by all genders indiscriminately. See ¢ Verbal Derivatives,” at the close of ill the section on ¢ The Verb.’ ith (8.) Neuter Singular.—There is but little which is worthy of remark in the singular forms of neuter Dravidian nouns. Every | t0 Dréavidian noun is naturally neuter, or destitute of gender, and it nce or becomes masculine or feminine solely in virtue of the addition of a S & masculine or feminine suffix. When abstract Sanscrit nouns are The jon adopted by the Dravidians, the neuter form of those nouns (ending in ally jon ¢am’) is generally retained ; and there are also some neuater nouns of “lllf} pure Dravidian origin which end in ‘am,” or take ‘am’ as their for- jif mative. The Dravidian termination ‘am’ is not to be regarded, Tk however, as a sign of the neuter, or a neuter suffix, though such is un- 182 THE NOUN. It is merely one of a numerous doubtedly its character in Sanscrit. elass of formatives, of which much use is made by the Dravidian 1] I dialect, and by the addition of which crude verbal roots become deri- vative nouns. é Such formatives are to be regarded as forming a part See ¢ Verbal f of the noun itself, not of the inflexional additions. Derivatives,” at the close of the section on ¢ The Verb.’ 1] All animated beings that are destitute of reason are placed by Drévidian grammarians in the ‘caste-less, or neuter class, and the gr i nouns that denote such animals, both in the singular and in the plural, are uniformly regarded as neuter or destitute of gender, irrespective of fi the animal’s sex. If it happens to be necessary to distinguish the sex of any animal that is included in this class, a separate word, signifying ‘male’ or ‘female,” ‘cock ’ or ‘hen, is prefixed. Even in such cases, however, the pronoun with which the noun stands in agreement is neuter, and notwithstanding the specification of the animal’s sex, the noun itself f remaing in the ‘caste-less’ or neuter class. For this reason, suffixes expressive of the neuter gender, whether th {f singular or plural, were not much required by Dravidian substantive nouns. The only neuter singular suffix of the Dravidian languages, al which is used in the same manner as the masculine ‘an’ or ‘adu,” and ¢ the feminine ¢al’ is that which constitutes the termination of the y neuter singular of demonstrative pronouns and appellative nouns. This pronoun is in Tamil-Canarese ¢ adu,” that, ¢idu,’ this; in Telugu § ‘adi, ‘idi ; in Malayalam ‘ata, ‘ita; in Gond ‘ad, ‘id.’ The same neuter demonstrative, or in some instances its termina- t tion only, is used in the conjugation of Dréividiun verbs as the sign of I the neuter singular of each tense, and in Telugu as the sign of the th feminine singular also. The bases of the Dravidian demonstratives fo being ‘a’ and i’ (‘a’ remote, ‘i’ proximate), that part of each pro- S noun which is found to be annexed to those demonstrative vowels is J I evidently a suffix of number and gender ; and as the final vowels of D ‘ad-u, ‘ad-i, ‘at-a,” ‘id-u,’ ¢id-i, ‘it-a,” are merely euphonic, and to have been added only for the purpose of helping the enunciation, it is fa evident that ‘d’ or ‘t’ alone constitutes the sign of the neuter sin- 80 gular, This view is econfirmed by the circumstance that ¢d’s or 9% never appears in the neuter plural of this demonstrative, but isreplaced ? ¢ v OL ¢ by e, ‘u, i or short ‘a,’ with a preceding euphonic I e.g., compare ‘adu’ (‘a-d-u’), Tam., that, with ‘ava’ (‘a-(v)-a’), 81 It will be shown afterwards that this final “a’ is Malayalam, those. h a sign of the neuter plural. of Appellative nouns which form their mase. singular in Tamil in. N — —— FEE— P— NUMBER PLURALISATION. 183 1§ ¢an,’ and their feminine sing. in ‘al,” form their neuter sing. by annex- 1 ing ‘du, with such euphonic changes as the previous consonant happens to require; e.g., ‘nalla~-du,” @ good thing ; ‘al-du,’ euphonically if ‘andru, a thing that is not; ¢ periya-du’ or ¢peri-du,’ great, a great l thing. This neuter singular suffix ‘d,” is largely used in all the dialects y in the formation of verbal nouns, e.g., ‘pogira-du, Tam., the act of 16 going, ‘pona-du, the having gome, ‘pova-du,’ the being about to go. i This form has been represented by some, but erroneously, as an infini- 01 tive : it is a concrete verbal or participial noun of the neuter gender, which has gradually come to be used as an abstract. al The affinities of the neuter singular suffix in ‘d’ or t, are ex- or clusively Indo-European, and they are found especially in the Indo- Jr‘ European pronouns and pronominals. We may observe this suffix In d the Sanserit “tad’ or *tat,’ that; in ‘tyad, that; in ‘adas,’ a weakened ! form of ¢adat, that; in ¢ étad, this. We find it also in the Latin ¢illud,” ¢id, &ec. (compare the Latin ¢id, with the Tamil ¢i-du,’ T this); and in our English demonstrative neuter ‘it’ (properly ¢hit’), (7 the neuter of “he,’ as also in ¢ what,” the neuter of ¢who.’ Compare g also the Vedic ¢it, an indeclinable pronoun, described as ¢ a petrified m neuter, which combines with the negative particle ‘na’ to form ‘nét,’ he if mot, apparently in the same manner as in Telugu the aoristic IN neuter ¢16du, there is not, is compounded of the negative ‘la’ and the 2 ! sutlix ¢du Thongh the Dravidian languages appear in this point to be allied to the Sanscrit family, it would be unsafe to suppose that they bor- I of rowed this neuter singular suffix from the Sanserit. The analogy of he the Dravidian neuter plural in ‘a, which though Indo-European, is es foreign to the Sanscrit, and that of the remote and proximate demon- strative vowels ‘a’ and ¢i, which though known to the Indo-Euro- 18 pean family, are used more systematically and distinctively by the of Dravidian languages than by any other class of tongues, would lead to the supposition that these particles were inherited by the Dravidian m‘(l 15 family, in common with the Sanserit, from a primitive, Pre-Sanscrit 1D source. t7 TaE PLURAL: PrINCIPLES oF PLURALISATION.—In the primitive ce(g Indo-European tongues, the plural is carefully distinguished from the Iy singular ; and with the exception of a few nouns of quantity which l% have the form of the singular but a plural signification, the ‘number’ 15 of nouns is always clearly denoted by their inflexional terminations. it Nouns whose number is indefinite, like our modern English ‘sheep,’ 0 4 AN (IR — B — 184 THE NOUN. are unknown to the older dialects of this family. In the languages of o the Scythian group a looser principle prevails, and number is generally left indefinite, so that it is the connexion alone which determines il whether a noun is singular or plural. The Manchu restricts the use i of its pluralising particle to words which denote animated beings: all e other words are left destitute of signs of number. Even the Tartar, I or Oriental Turkish, ordinarily pluralises the pronouns alone, and leaves the number of other nouns indeterminate. In the Brahui also, th the number of nouns is generally left undefined; and when it is e desired to attach to any noun the idea of plurality, a word signifying il ‘many’ or ‘several’ is prefixed to it. Notwithstanding this rule, N Brahui verbs are regularly pluralised ; and the number of an inde- of terminate noun may often be ascertained from the number of t he verb with which it agrees. & With respect to principles of pluralisation, the Dravidian tongues differ considerably from the Indo-European family, and accord on the V1 whole with surprising exactness with the languages of the Scythian bl stock. ‘The number of Drévidian nouns, especially of neuter nouns, is th ordinarily indefinite ; and it depends aupon the connexion whether any 10) It is true that when noun is to be regarded as singular or as plural. te) more ‘persons’ than one are referred to, the high-caste’ or ‘rational’ R pronouns that are used are almost invariably plural ; and that even hi neuter nouns themselves are sometimes pluralised, especially in polished 0o prose compositions : but the poets and the peasants, the most faithful 0] guardians of antique forms of speech, rarely pluralise the neuter, and ) are fond of using the singular noun in an indefinite singular-plural Cl sense, without specification of number, except in so far as it is S expressed by the context. This rule is adhered to with especial nei strictness by the Tamil, which in this, as in many other particulars, exhibits most faithfully the primitive condition of the Dravidian lan- or guages. Thus in Tamil, ‘méidu,’ oz, means either an oz or owen, 10y according to the connexion ; and even when a numeral is prefixed 18§ — which necessarily econveys the idea of plurality, idiomatic speakers Yl\h prefer to retain the singular or indefinite form of the noun. Hence 00e they will rather say ‘nélu madu méygiradu,” literally four oz 18 feed- if ing, than ‘nalu madugal méygindrana,’ four oxen are Jeeding, which 0ng would sound clumsy and pedantie. e Even when a neuter noun is pluralised by the addition of a plura- Us¢ lising particle, the verb is rarely pluralised to correspond ; but the gy singular form of verb is still used for the plural,—the number of the Obj( neuter singular being naturally indeterminate. This is invariably the eup practice in the speech of the lower classes ; and the colloquial style hu 185 NUMBER—PLU RALISATI ON. 5 of of even the best educated classes exhibits a similar characteristic. The ly Tamil language contains, it is true, a plural form of the third person Ines neuter of the verb, and the existence of this form is a clear proof of U5¢ the high cultivation of the Tamil ; but the use of the neuter plural all verb is ordinarily restricted to poetry, and even in poetry the singular ttat, number both of neuter nouns and of the verbs that correspond is much and more commonly used than the plural. It should be remarked also, also, that the third person neuter of the Tamil future, or aorist, is altogether t 1 destitute of a plural. In this particular, therefore, the Tamil verb is g more decidedly Scythian in character than the noun itself. Max rule, Miiller supposes that a Dravidian neuter plural noun, with its suffix nde- of plurality, is felt to be a compound (like ¢animal-mass’ for yeth “animals,” or ¢stone-heap’ for ‘stomes’), and that it is on this account that it is followed by a verb in the singular. The explanation which I have given seems to me preferable. The number of all Dra- Zues ) the vidian nouns, whether ¢high-caste’ or ¢caste-less,’ was originally hian indefinite : the singular, the primitive condition of every noun, was 18,18 then the only number which was or could be recognized by verbal or ] any nominal inflexions, and plurality was left to be inferred from the con- rhen text. As civilization made progress, the plural made its formal nal’ appearance, and effected a permanent settlement in the department of even high-caste or masculine-feminine nouns and verbs ; whilst the number shed of caste-less or neuter nouns, whether suffixes of plurality were used thful or not, still remained generally unrecognized by the verb in the Dravi- dian languages. Even where the form exists it is little used. It is , and Jural curious, that in this very point the Greek verb exhibits signs of it 1 Scythian influences, viz., in the use of the singular verb for the plural neuter. ecial las, The Drévidianulanguages ordinarily express the idea of singularity Jan- or oneness, not by the addition of a singular suffix to nouns and pro- nouns, or by the absence of the pluralising particle (by which number (el fixed is still left indeterminate), but by prefixing the numeral adjective ‘one.’ ket Thus, ‘madu,” Tam., oz, does not mean exclusively either an oz or ozen, but admits of either meaning according to circumstances ; and [pnce if we wish distinctly to specify singularity, we must say ‘oru madu,’ fel- one or a certain ox. ich Europeans in speaking the Dravidian dialects use this prefix of singularity too frequently, misled by their habitual use of an indefinite article in their own tongues. They also make too Jurd- b the free a use of the distinctively plural form of neuter nouns, when the objects to which they wish to refer are plural. Occasionally, when f the euphony or usage recommend it, this is done by Dravidians themselves, y fhe but as a general rule the neuter singular is used instead of the neuter siyle 186 THE NOUN. plural, and that not in the Tamil only, or in the Drividian lan- guages only, bat also in almost all the languages of the Scythian group Another important particular in which the Indo-European languages differ from the Seythian is, that in the former the plural has a different set of case-terminations from the singular, by the use of which the idea of plurality is not separately expressed, but is conjoined with that of case-relation ; whilst in the latter family the plural uses the same set of case-terminations as the singular, and plurality is expressed by a sign of plurality common to all the cases, which is inserted between the singular or crude form of the noun and the case-terminations. In the Indo-European languages, each inflexion includes the twofold idea of namber and of case. Thus there is a ‘genitive singular’ and a ‘genitive plural,’ each of which is a complex idea; but there is no inflexion which can be called ¢ genitive,” irrespective of number ; and in many instances (this of the genitive being one) there is no apparent connexion between the case-termination of the singular and that which is used in, and which constitutes, the plural. In those few cases in which the sign of number and the sign of case seem to have been originally distinct, and to have coalesced into T one, the sign of case seems to have preceded that of number : e.g., the Bxpm Gothic plural accusative °ns,” is derived from ‘n’ or ¢m,’ the sign of fhe the accusative singular, and ‘s, the sign of plurality. When the Sa 1 Seythian family of languages is examined, it is found that each of their case-signs is fixed and unalterable. It expresses the idea of bage 3 ‘case’ and nothing more, and is the same in the plural as in the sin- abfa gular, with the exception of those few trivial changes which are direc required by euphony. The sign of plurality also is not only distinct Mm from the case-sign, but is one and the same in all the cases. It is an un- Smey alterable post-position—a fixed quantity ; and it is not post-fixed to of the the case-sign, much less compounded with it, as in the Indo-European by th languages, but is prefixed to it. It 1s attached direct]y to the root Heng itself, and followed by the variable signs of case. Dty In the Dravidian languages a similar simplicity and rigidity of thoge structure characterizes the use of the particles of plurality. They are I added directly to the crude base of the noun (which is equivalent to theg the nominative singular), and are the same in each of the oblique and cases as in the nominative. The signs of case are precisely the same Comyy in the plural as in the singular, the only difference being that in the Modg singular they are suffixed to the crude noun itself, in the plural to the Teta)y pluralising particle, after the addition of that particle to the crude bus noumn. “um NUMBER—PLURALISATION. 187 lan- For example, in Hungarian ‘haz, « kouse, is declined as follows : hian SINGULAR. 4 PrugaL. Nominative, haz Nom hiz-ak ages arent Genitive, hiznak Gen héz-ak-nak hiznak Dat. héiz-ak-nak Dative, 1dea Accusative, héz-at Ace hiz-ak-at at of 6 et In Tamil ‘manei,’ @ kouse, is declined as follows : bya Weel SINGULAR. PLURAL. In ' Nominative, manei Non. manei-gal | 1dea Accusative, manei-(y)-ei Ace manei-gal-ei nd a Instrumental, manei-(y)-al Instr. manei-gal-al s 10 Conjunctive, manei-(y)-6du Conj. manei-gal-6du Dative, manei-kku Dat manei-gal (u)-kku and ) Ablative, manei-(y)-il-irundu Ablat manei-gal-il-irundu arent Gen Genitive, manei-(y)-in manei-gal-in Locative, manei-(y) idatt-il Locat manei-gal-idatt-il vhich Vocative, manei-(y)-é Voe manei-gal-é ! of | into The particular signs which are used to express plurality and as ., the exponents of case, are taken from the resources of each langnage; but g of the manuer in which they are used in both languages is precisely the same. 1 the ch of The neuter of DrAvidian nouns being identical with the crude Jea of base, when the pluralising particle is attached to a neuter noun, it is ' 16 S10- attached to it not as a substitute for any suffix of the singular, but directly and without any change: it is attached to it pure and simple. I are In the case of masculine and feminine nouns, including pronouns, a stinct somewhat different method of pluralisation is necessary. The singular 1 un- of the masculine and feminine is formed, as has already been pointed out, ced to by the addition to the root of particles denoting ‘a male,” or ‘a female.’ gpeal Henee to pluralise those nouns, it is necessary either to add a pluralising » 1000 particle to the masculine and feminine suffixes, or to substitute for those suffixes an epicene pluralising particle. /] ity of In all the Dravidian languages the primitive plan of pluralising ey 418 these two classes of nouns was that of substituting for the masculine ent 10 and feminine singular suffixes a suffix of plurality which applied in blique common to men and women, without distinction of sex. This 1s the » S0 mode which is still used in most of the dialects; but in Telugu it n the retains its place only in connexion with pronouns and verbs, and fo the has disappeared from substantives, which form their plural by means cl’(](le of a neuter suffix. 188 THE NOUN. The classification of Dravidian nouns into ‘rationals’ and firra- are tionals,” has already been explained: it has also been shown that in anal the singular, the masculine of ‘rational’ nouns is distinguished from hoth ] the feminine. In the plural both those genders are combined; the sign high caste particle of plurality, or plural of rational beings, is the same for both genders, and includes men and women, gods and god- thern desses, without distinction of sex. their and ¢Irrational’ or neuter nouns have a particle of plurality different neut from this and peculiar to themselves. Hence the Dravidian Janguages have one form of the plural which may be called ¢ epicene’ or ¢ mascu- Telo line-feminine,” and another which is ordinarily restricted to the neuter ; COIY and by means of these pluralising particles, gender and number are plur plur conjointly expressed in the plural by one and the same termination. The masculine-feminine plural expresses the idea of plurality conjointly T’ with that of rationality; the neuter plural, the idea of plurality con- -l 1383 jointly with that of irrationality. Arrangements of this kind for giving combined expression to eollo gender and number, are very commonly observed in the Indo-European low | family; and even the plan of classing masculines and feminines ant together in the plural, without distinction of sex, is also very common. Thus the Sanscrit plural in ‘as is masculine-feminine; so is the 18 e Latin plural in “es,” and the Greek in © es. epice The chief difference with respect to this point between the Dra- of 1) ‘g vidian system and the Indo-European one lies in this, that in the Dravidian languages, the masculine-feminine particle of plurality is dial carefully restricted to rational beings; whereas in the Indo-European languages irrational and even inanimate objects are often complimented ? with inflexional forms and pluralising particles which imply the hut ¢ existence, not only of vitality, but even of personality, that is, rational he a self-consciousness. dia lec A still closer analogy to the Dravidian system is that which is T exhibited by the New Persian. That dialect possesses two pluralising {h 16 1) particles of which one, ‘4n,’ is suffixed to nouns denoting living beings,* 1ty A the other, ‘h4,” to nouns denoting inanimate objects. The particles doul which are employed by the Persians are different from those which avy but * Bopp derives ‘4n,’ the New Persian plural of animated beings, from the Sanserit ‘4n,’ the masculine-plural accusative. I am inclined with Colonel Raw- One ¢ linson to connect this particle with the Chaldaic and Cuthite plural ‘&n,” allied njp to ‘“im’ and ‘in’ (e.g., ‘anin, Chald., we); the New Persian being undoubtedly tinged with Chaldaco-Assyrian elements, through its connection with the Pehlvi. iy One is tempted to connect with this suffix our English plural suffix ‘en,’ in like brethren; a suffix which is regularly used by the Dutch as a particle of plurality. the g Bopp, however, holds that this ‘en,” is an ancient formative suffix, which was originally used by the singular as well as the plural. lhe (o NUMBER—PLURALISATION. 189 - are used in the Dravidian languages, but the principle is evidently at 1n analogous. The Persians specialise life, the Dravidians reason,; and from both of them class the sexes together indiscriminately in the plural. 5 the In Telugu some confusion has been introduced between the epicene 5 the sign of plurality ¢ar-u,” and the neuter ¢lu.’ The pronouns pluralise o0d their masculines and feminines regularly by substituting ‘ar-u’ for their masculine and feminine singular suffixes, whilst the substantives ferent and some of the appellative nouns append ‘lu,” which is properly the b neuter sign of plurality, instead of the more correct ¢ar-u. Thus the JUAZES NasCll- Telugu demonstrative pronoun ¢ var-u,’ they (the plural of ¢ vadu,’ Ze), euter; corresponding to the Canarese ‘avar-u,” exhibits the regular epicene er are plural ; whilst ‘magadu,” @ husband (in Tamil ‘magan’), takes for its ation, plural not “‘magaru,’ but ¢ magalu; and some nouns of this class add ‘lu’ to the masculine or feminine singular suffix; eg., “alludu,’ a son- ointly in-law, makes in the plural not ¢alluru,” nor even ‘allulu,’ but ‘allundlu,’ y con- nasalised from ‘allud’-lu; and instead of ¢véru, they, ¢ vandlu,’ is colloquially used, a word which is formed on the same plan as the on f0 low Madras Tamil ‘avan-gal,’ they, instead of ‘avargal’ or the higher "OPC‘HH mines and purer ‘avar.’ Perhaps the only case in which the ¢irrational’ pluralising particle nmon, s the is used in the higher dialect of the Tamil instead of the ¢rational’ epicene, is that of ‘makkal’ (mag-gal”), mankind, people. The singular of this word being ‘mag-an,’ the plural ought by rule to have been , Dr- in the ‘mag-ar ; and it is interesting to notice that there is in the higher dialect a rarely used plural, ‘magar, in addition to the ordinary dity i ‘ makkal.’ ropeall nented The Ku rational plural is ‘nga,” which is properly an irrational one, but the pronouns and participial nouns form their rational plural by ]v the the addition of ¢4ru,’ which is identical with the ‘aru’ of the other tional dialects. The modern colloquial Tamil has been influenced in some degree by ich 18 the usage of the Telugu, and has adopted the practice of adding the lising irrational plural to the rational one, thereby systematically forming a eingﬁ{% ) double plural ar-gal,’ instead of the old rational plural ‘a 5% g 1t fcles which ‘avan,’ he, and © aval,’ she, properly take ‘avar,’ they, as their plural ; but the plural which is preferred by the modern Tamil, is the double o the one ‘avar-gal’ So also the plural of the second person is properly ol Rar- 1 gllie “nir;’ but the plural which is most commonly used as ‘ning-gal’ (from ‘nim,” an older form of ‘nir,” and ‘gal’), which is a double plural otedly | Pehm' like ¢ avar-gal.’ Two forms of the epicene plural being thus placed at ¢ en"’ in the disposal of the Tamil people (the classical ‘nir’ and ‘avar, and Josalif ich ¥ the colloquial ‘ning-gal’ and ‘avar-gal’), they have converted the B THE NOUN. 190 the I former, in colloquial usage and in prose compositions, into an honorific singular, and the same practice has been adopted in Canarese. This TESPe whicl usage, though universally prevalent now, was almost unknown to the poets. Yowe) I have not observed in the poets, or in any of the old inscrip- T tions in my possession, any instance of the use of the epicene plural as (for an honorific singular, except in connection with the names and titles of be the divinities, whether those names and titles are applied to the gods Even in those iy themselves, or are conferred honorifically upon kings. plura cases, however, the corresponding pronoun follows the ordinary rule, and is very rarely honorific. In modern Telugu a double plural, whic similar to that of the Tamil, has gained a footing ; e.g., ¢ varu-lu’ (for “may ¢ var-u’), they, and ‘miru-lu’ (for ¢mir-u’), you. allie The Telugu, as has been observed, pluralises masculine and al feminine substantive nouns by the addition, not of the rational, but (I, of the neuter or irrational sign of plurality: by a similar inversion fixed of idiom, the Gond sometimes uses the rational plural to pluralise a ) neuter nouns; e.g., ‘kawalor,” crows. Such usages, however, are suff to th evidently exceptions to the general and more distinctively Dravidian Tami rule, according to which the neuter pluralising particle is restricted to plur ) neuter nouns, and the epicene particle to rational or personal nouns used t.e., masculines and feminines. Ny We shall now consider in detail the pluralising particles them- selves. Tar Tngy Usg 1. Epicene Plurailsing Particle—This particle is virtually one and the same in all the dialects, and the different forms which it has may taken are owing merely to euphonic peculiarities. In Tamil nouns, Pare R I pronouns, and verbs, it assumes the forms of ‘ar,” ‘ar 2 or; 24C ir’, gy 2 SIe in Canarese and Telugu, ‘aru,” ‘uru; ‘are,” ‘éru; ‘ri,) ‘ru? in I§u; “aru ;) in Gound, <6r The Brahui also forms the second person ar [4 plural of its verb in ‘ere, ‘ure, &c., the third person in ur or 0oy ‘ar. I regard “ar’ (not simply ‘r’) as probably the primitive shape Pa ey of this pluralising particle, from which the other forms have been 0y derived by euphonic mutation. It is true that ‘ni,’ thou, forms its deng plural in modern Tamil by simply adding ‘r; but this does not prove of a] that ‘r’ alone was the primitive form of the epicene plural, for an Ingy) older form of ¢nir, you, is ni-(v)-ir’ or ‘ni-(y)-ir,” from which ¢nir’ Ofnw has evidently been derived. It might naturally be supposed that in 2 tiop this case ¢ 1T is used instead of ‘ar, through the attraction of ‘the My ] preceding long vowel, ¢i; but we also find “ir’ used as a pluralising ﬂ]e ( [4 ir, particle in ‘magalir, High Tam., women, and also a longer form, of 4 in ‘magalir:’ consequently ¢ir’ has acquired a position of its own in “h](. NUMBER THE EPICENE PLURAL. 191 onific the language, as well as ‘ar’ All that we can certainly conclude This respecting the original shape of this particle is that the final ‘r, 0 the which is plainly essential, was preceded by a vowel, and that that - vowel was probably b9y al as The Canarese rational plural sufix ¢andar, e.g., ‘avandar-u,’ les of (for ¢avar-u’), “4lls, and ‘ivandar-u’ (for “ivar-u’), ¢ ki, seems to gols be identical with the Tel. indefinite plural ‘andar-u, indar-u, so those many, the final ‘ar’ of which is the ordinary suffix of the epicene ule, plural. lural, The Tamil and Malayalam have another particle of plurality (for which is applicable to rational beings, viz., ‘mér,” or in High Tamil ‘mar,” which has a considerable resemblance to ¢ ar, and is probably allied to it. ‘mar 5 and is used to pluralise rational nouns substantive ], but alone, and is not like ‘ar’ used by pronouns and verbs. It is suf- 15100 fixed to the noun which it qualifies in a different manner also from ¢ ralise ar; for whilst ‘ar’ is substituted for the masculine and feminine ) are suffixes of the singular, not added to them, ¢ mar’ is generally added idian to the singular suffix by idiomatic writers and speakers. Thus in od to Tamil, ¢ purushan’ (a Sans. derivative), a man, a husband, when pluralised by suffixing ‘ar 0Uns, becomes ¢ purushar; but if ‘mar’ is used instead of ‘ar,” it is not substituted for ‘an’ the masculine them- singular suffix, but appended to it, e.g., ¢ purushan-mér,” not ¢ purusha- mar.” ‘Mar, it is true, is sometimes added to ‘ar, e.g., ‘purushar- mar ; but this is considered unidiomatical. ¢ Mar’ is also sometimes y one used as an isolated particle of plurality in a peculiarly Scythian 'it has manner, eg., ‘tay-tagappan-méar, Tam., mothers and Jathers, parents; in which both mother and father are in the singular, and Jouns, “mir’ is separately appended to pluralise both. ;" ‘ir,’ g I Probably there was originally no difference in signification between eIl ‘ar’ and ‘mar,’ whatever difference there may have been in their origin. In modern Tamil, I ; or ‘mar’ is suffixed to nouns signifying shape parents, priests, kings, &e., as a plural of honour, like the Hungarian ‘mek ; but it may be suffixed, if necessary, to any class of nouns - heen ms Its denoting rational beings. . In Malayalam it is used with a wider range of application than in Tamil, and in cases in which an honorific pro\'e meaning cannot be intended, e.g., ¢ kallan-mar,” thieves. The antiquity or an of many of the forms of the Malayala grammar, favour it s the supposi- tion that in ancient Tamil, which was probably identical with ancient b 11 Malayalam, ‘mar’ may generally have been used instead of ‘ar, as of the the ordinary pluralising particle of ‘high caste’ nouns. A few traces 1, of this use of the particle ‘méir’ survive in classical Tamil ; ¢ migng il n which is evidently equivalent to ‘mar, and probably older, being 192 THE NOUN. tend sometimes used in poetry instead of ‘ar,’ e.g., ‘ en-mar’ (from en,’ ¢o count), accountants. ‘1 We have now to inquire whether ‘ar’ and ‘mar, the Dravidian imp! onig plurals of rationality, sustain any relation to the plural terminations, or pluralising suflixes, of other languages. 0 Dré It might at first sight be supposed that the formation of the plural Tely by the addition of ‘r’ to the singular which characterises some of the plur Teutonic tongues, is analogous to the use of ‘r’ or ‘ar’ in the Dravi- dian languages. In the Icelandic the most common plural is that forn Wou which terminates in ¢ r’—sometimes the consonant ‘r’ alone, some- A relic Tan times the syllables ‘ar, ir, ‘ur, e.g., ‘ konungur,’ kings. of this plural may be traced in the vulgar English ¢childer,” for form firgt ¢ children.’ The same plural appears in the Old Latin termination “nar of the masculine plural in ‘or’ which is found in the Eugubian tables, e.g., ‘subator’ for ‘subacti,’ and ¢screhitor’ for ¢scripti.’ of ] Compare also ‘mas,” the termination of the first person plural of Unf verbs in Sanscrit, with ‘mar,” the corresponding termination in Irish, the ] answering to the Doric ues and the ordinary Greek wuev. ill 01 In these cases, however, the resemblance to the Dravidian plural and ‘ar’ is rather apparent than real; for the final ‘r’ of these forms £qoi has been hardened from an older ‘s; whilst there is no evidence of the existence of a tendency in the Drévidian languages to harden i s’ into ‘r, and therefore nothing to warrant the supposition that e the Dréividian epicene ‘ar’ has been derived from, or is connected €0 with, the Sanscrit masculine-feminine ‘as.’ 151 to ¢ It should also be noted that the Irish ‘mar’ is a compound of two forms, ‘ma,” the representative of the singular of the personal > pronoun ¢ 7, and ‘r, the hardened equivalent of the plural suffix s; the ; and that, therefore, it has no real resemblance to the Dravidian Prete ‘mar, which is entirely and exclusively a plural suffix of the third 0, t person. i There is more probability of the Dravidian plural suffixes being ‘el related to the pluralising particles of some of the Scythian languages. ‘ndy The Turkish plural suffix, which is inserted, as in the Dravidian an languages, between the crude noun and each of the case terminations, 1t} is ‘lar or ‘ler, e.g., ‘an-lar,” they. Dr. Logan says, but on what Iy authority does not appear, that ‘nar’ is a plural suffix in Kol. Mon- ay golian nouns which end with a vowel, are pluralised by the addition teny of ¢nar’ or ‘ner, a particle which is evidently related to, or identical Thy with, the Turkish ¢lar’ or ‘ler:’ and the resemblance of this Mongol Pax sufix ‘nar’ to the Dravidian ‘mar, both in the final ‘ar’ and in tha the nasal prefix is remarkable. It is well known that ‘m’ evinceg a ang NUMBER—THE EPICENE PLURAL. 193 tendency to be softened into ‘n’ (witness the change of the Sanserit ‘mama,’ my, into ‘mana’ in Zend); and in this manner it seems not improbable that the Dravidian ‘mar’ may be allied to, or even the origin of, the High Asian ‘nar’ Again, in the Scythian tongues ‘n’ is often elided or dropped, and the same peculiarity characterises the Dravidian languages. Thus, ‘nn,’ the conjunctive particle of the 2 6 Telugu, becomes ‘i in Canarese. In this manner the Dravidian plural suffix ‘ar may have been softened from ‘mar; and if both I forms continued to be occasionally used, ‘mar,” the older of the two, would naturally and regularly acquire an honorific signification. The 6 Tamil ¢ileifijar’ (“ilei-fijar’), young people, a plural appellative noun, )8 formed from ‘ilei, youth, exhibits a form of pluralisation which at I first sight seems very closely to resemble the Mongolian nar.’ Nay, N ‘nar’ is actually used in this very instance instead of fijar’ by some of the poets, and it is certain that ‘fij’ and ‘n’ often change places. of Unfortunately we find this ‘nj’ or ‘n’ in the singular, as well as the plural; which proves it to have been inserted merely for euphony in order to prevent hiatus, and therefore ‘ileifijar’ must be re-divided, | and represented not as ilei-fijar,” but as ‘ilei-(fij)-ar’ or ‘ilei-(n)-ar,’ 1§ equivalent to ¢ilei-(y)-ar.’ (6 Probably the same explanation should be given of ‘manéir, the 7|} epicene plural termination of the future tense in some of the poets, b especially Tolképpiyan, the most ancient Tamil grammarian; e.g., ) ol ¢ enmanar,’ they will say, instead of the more common ¢enbar.’ ‘m Is in this connection used as the sign of futurity, and is equivalent to ‘b, and ‘enmér’ is equivalent to ¢ enmanéar.’ al The insertion of an euphonic ‘n’ between the sign of tense and the pronominal suffix is exceedingly common in the present and Ll preterite ; e.g., ‘nadanda-(n)-em’ (for ‘nadand’-ém’), we walked,; and if 9 S0, there seems no reason why the same ‘n should not make its 1 appearance in the future also, converting ‘enmér’ (for ¢enba-ar’ or “enbar’) into ‘enma-(n)-ar. ¢ 2 If this explanation does not suffice, “par must, in this instance, be regarded as the equivalent of ¢ méar, S, an and therefore as directly allied to ‘nar,’ the Mongolian plural suffix. 15 It is deserving of notice that the Turkish, besides its ordinary plural 9 ‘lar’ or ‘ler,” uses ‘z as a plural suffix of the personal pronouns, 8t ) as may be observed in ‘biz,’ we, and ¢siz, you, and that the Turkish on terminal ‘z’ corresponds to the ‘r’ of some other Scythian languages Thus ¢yaz,’ Turkish, summer, is in Magyar ¢ yar’ or ‘nyar’ (com- al pare the Tamil ‘nyayir, the sun). It would almost appear, therefore, gol n that the Turkisk suffix of plurality has undergone a process of change and comminution which is similar to that of the Tamil, and that the 0 194 THE NOUN. Turkish ¢z’ and the Tamil ‘r’ are remotely connected, as the last remaining representatives or relics of ‘mar, ‘nar, and ‘lar.’ 2. Pluralising Particles of the Neuter.—There are two neuter pluralising particles used by the Dravidian languages.— (1.) The neuter plural suffic ¢gal; with its wvarieties.—It has already been noticed that ‘gal’ is occasionally but improperly used in Tamil and Canarese as the plural suffix of ‘rational’ nouns and pronouns ; and that the corresponding Telugu ¢lu’ is still more systematically used in this manner. Nevertheless, I have no deubt that it was originally and is essentially a suffix of the neuter plural. This suffix is in both dialects of the Tamil ‘gal, eg., ‘kei-gal, hands, with only such changes as are required by Tamilian rules of euphony. In accordance with one of those rules, when ¢g,” the initial consonant of ‘gal,’ is doubled, or preceded without an intermediate vowel by another consonant, ‘gal’ is regularly hardened into ‘kal’ or ‘kkal.’ Thus ¢ kal-gal,” stones, is changed by rule into ‘kar-kal.’ ¢gal’ is occasionally lengthened in Tamil poetry into ¢gal. In Malayéalam this particle is generally ‘kal’ or ¢ kkal’ but sometimes the initial ‘k’ coalesces with a preceding nasal and becomes ‘ng, e.g., ‘ning-ngal,’ you, instead of ‘nim-kal,”—in Tamil ning-gal.’ In The modern Canarese we have gal-u,” in ancient ‘gal,’ as in Tamil. three southern idioms are in perfect agreement with respeet to this par- ticle, but when we advance further north we shall find its shape con- siderably modified. In Telugu, the corresponding neuter plural suffix is ¢lu,” of which the ‘1’ answers, as is usual in Telugu, to the cerebral ¢1’ of the other dialects : ‘1-u,’” therefore, accords with the final syllable of the Canarese ‘gal-u’ The only real difference between the Telugu and the Tamil-Canarese consists in the omission by the former of the initial bl consonant ‘k’ or ‘g Traces, however, exist in Telugu, of the use of a vowel before ¢lu.’ Thus, in ‘gurralu,” Aorses, the long ¢ A a 1S derived from the combination of the short final “a’ of the inflexional base ‘gurra’ and a vowel, evidently ¢a,” which must have preceded £ Jul We thus arrive at ‘al-u,’ as the primitive form of the Telugu plural ; and it is obvious that ‘al-u’ could easily have been softened from ¢gal-u’ Conjecture, however, is scarcely needed, for in some nouns ending in ‘n-u,” of which the Tamil equivalents end in ‘m, the old Dravidian pluralising particle in ¢gal,’ is exhibited in Telugu almost as distinctly as in Tamil. Thus, ‘kolan-u,’ a tank (Tamil ‘kulam’), takes as its plural ¢ kolan-kul-u’ (Tamil ‘kulang-gal’), and 4 S NUMBER—THE NEUTER PLURAL. 195 ‘gon-u,” the name of a species of tree, forms its plural in ‘gon-gul-u.’ When ‘kul-u’ and ‘gul-u’ are compared with the Tamil-Canarese forms ‘kal,’ ‘gal’ and ‘gal-u, it is obvious that they are not only equivalent but identical. An illustration of the manner in which the Telugu ‘lu’ has been softened from ¢gal-u,’” may be taken also from colloquial Tamil, in which ‘avar-gal’ they, is commonly pronounced ¢aval.’ ‘kion &8 b 3 is dropped or elided in a similar manner in many languages of the Scythian family. The Tulu, though locally remote from the Telugu, follows its example in many points, and amongst others in this. It rejects the ‘k’ or ‘g’ of the plural, and uses merely ‘lu’ or <1’ like the Telugu ; rarely ¢kulu. ) The same form of the pluralising particle appears in the languages | of some of the tribes of the north-eastern frontier-—languages which possibly form a link of connection between the Drévid_ian and the Tibetan families. In the Miri or Abor-Miri dialect, ‘nd,” thou, forms its plural in ‘ndlu, yow ; and in the Dhimal, ‘ni,” thou, is pluralised 1 into ‘nyél, you. The pronoun of the Mikir is pluralised by adding N ‘i e g., “na-li; you, whilst substantives have no plural form. In the i Dhimal, substantive-nouns are pluralised by the addition of ¢ galai,’ which is possibly the origin of the pronominal plural ‘1’ though this le particle or word, ¢ galai, is not compounded with, or agglutinated to, the noun, but placed after it separately. Though it is used as a sepa- rate word it does not seem to retain any signification of its own inde- pendent of its use as a post-position. The resemblance of ¢ galai’ to h the Tamil-Canarese ‘gal’ or ‘galu,’ is distinct and remarkable. The 0 pluralising particle of the Naga also is ¢khala.’ It is not an uncommon occurrence to find one portion of a much used prefix or suffix in one language or dialect of a family, and another portion of it in another member of the same family. Seeing, il of therefore, that the Telugu has adopted the latter portion of the par- 1§ ticle ‘kal, ‘gal’ or ‘galu,” and omitted the initial ‘ka,’” ¢ ga,’ or ‘k,’ we may expect to find this ‘k’ used as a pluralising particle in some al d other Dravidian dialect, and the final ‘lu’ or ‘1’ omitted. Accord- ingly in Gond we find that the plural neuter is commonly formed by I d the addition of ‘k’alone, e.g., ‘nai,’ a dog, ‘naik,’ dogs (compare Tamil ‘naykal’ pronounced ‘néjgal’) The Seoni-Génd forms its 16 plural by adding ‘nk,” e.g., ‘neli,” @ field, ‘nelnk,” fields. The Ku h i dialect uses ‘nga,” and also ‘ska,” of all which forms ‘k’ or ‘g’ con- stitutes the basis. i d ‘k’ is sometimes found to interchange with ‘t,’ especially in the 02 196 THE NOUN. languages of High Asia. This interchange appears also in the Gond pluralising particle ; for whilst ¢k’ is the particle in general use, the pronouns of the first and second persons form their plurals, or double plurals, by the addition of ‘t’ to the nominative, e.g., ‘amat, we, ‘imat,” youw. The same interchange between ‘k’ and ‘t’ appears in the Brahui. Though a separate word is usually employed by the Brahui to denote plurality, a suffix in ‘k’ is also sometimes used ; but this ‘k’ is found only in the nominative plural, and is replaced by ‘t’ in the oblique cases. When we turn to the grammatical forms of the Finnish family of languages, we find some tolerably distinct analogies to this Dravidian plural suffix. Compare with the Drividian forms noticed above the Magyar plural in ‘k’ or “ak ; the Lappish in ‘k,’ ‘ch,’ or ‘b also the ‘t” by which ¢k’ is replaced in almost all the other dialects of the Finnish family: and observe the re-appearance of the sound of ‘1’ In Ostiak the dual suffix is ‘kan’ or in the Ostiak plural suffix ¢tl.’ 4 2 gan ;' in Samoied-Ostiak ‘ga’ or ‘ka ; in Kamass ‘gai.’ Castren supposes these suffixes to be derived from the conjunctive particle ‘ka’ or ‘ki,” also,; but their resemblance to the Dravidian signs of plurality renders this derivation doubtful. Even the Armenian forms its plural in ‘k; e.g., “tu, thow, ‘tuk,” you; ‘sirem,” I love, ‘siremk,” we love. In the Turkish also, ‘k’ is the sign of plurality in some forms of the first person plural of the verb, e.g., ‘¥dum,” 7 was, ‘iduk, we were. ‘t,” on the other hand, is the sign of the plural in Mongolian, and in the Calmuck is softened into ¢d.’ Even in Zend, though a language ot a different family, there is a neuter plural in ‘t.’ Thus, for ¢iméani,’ Sans., these things, the Zend has ‘imat.’ In those instances of the interchange of ‘t’ and ‘k,” in which it can be ascertained with tolerable clearness which consonant was the one originally used and which was the corruption, ‘t’ appears to be older than k.’ Thus the Doric ¢ 7fjros’ is in better accordance with related words, and therefore probably older, than the Aolian ¢ kfjros, the origin of ¢e-xéwos.” The Semitic pronoun or pronominal fragment ‘ta, thow (preserved in ‘attd’ and ‘antd’), is also, I doubt not, a more accurate and older form than the equivalent or auxiliary suffix ‘ka In several of the Polynesian dialects, ‘k’ is found instead m of an undoubtedly earlier Sanscrit or Pre-Sanserit ¢t. 1, accordance with these precedents, where ‘k’ and ‘t’ are found to be interchanged, ‘t’ is to be regarded as older than ‘k, it would follow that kal,” the Dravidian plural suffix now under consi- deration, may originally have been *tal’ I cannot think that the Dravidian ‘gal’ has been derived, as Dr. Stevenson supposes, from the NUMBER—THE NEUTER PLURAL. 197 Sanserit ‘sakala’ (in Tamil ¢sagala’), all. ¢kal,” the base of ‘sa-kala,’” has been connected wich ¢ éX-o< ;’ but the root signifying al, which the Dravidians have preferred to retain, viz., “ell,” is connected, not with ¢ o\,” whole, the Hebrew ¢ kol,” &e., but with the Saxon ‘eal,’ English all. This being the case, it is unlikely that they would have preserved the other root also. The Drévidian ‘tal-a’ or dal-a,” a host, a crowd, would give a good meaning ; but even this derivation of “kal’ or ‘tal, is altogether destitute of evidence. The supposititious Dravidian ‘tal’ may be compared with the Ostiak plural suffix €t but in the absence of evidence it is useless to proceed with conjectural analogies. The New Persian neuter plural, or plural of inanimate objects, which corresponds generally to the Dravidian neuter plural, is ha,” a It form which Bopp derives with much probability from the Zend. may here be mentioned, though I do not attach any importance to a resemblance which is certainly accidental, that the Tamil plural ‘gal’ sometimes becomes ‘ha’ in the pronunciation of the peasantry, e.g., ¢ irukkirargal,’ they are, is vulgarly pronounced ¢irukkiraha.’ ' (2.) Neuter Plural Suffiz vn ‘a.’—In addition to the neuter plural in ‘gal’ with its varieties, we find in nearly all the Dravidian lan- guages a neuter plural in short ‘a,” or traces of the use of it at some former period. Il ¢ gal,” though a neater plural suffix, is occasionally used, especially in the modern dialects, as the plural suffix of rationals; but in those dialects in which ¢a’ is used, its use is invariably restricted to neuters, and it seems therefore to be a more essentially neuter form than ¢ gal’ i | itself. | We shall first exa;lline the traces of the existence and use of this 7 suffix which are contained in the Tamil. h g ‘gal’ is invariably used in Tamil as the plural suftix of uncom- pounded neuter nouns; but ‘a’ is preferred in the classical dialect for it pluralising neuter compounds, that is, appellative nouns, or those which are compounded of a base and a suffix of gender, together with demon- strative pronouns, pronominal adjectives, and participial nouns. Even d in the ordinary dialect, ‘a’ is generally used as the suffix of the N e e neuter plural in the conjugation of verbs. d The second line in one of the distichs of Tiruvalluvar's ¢ Kural,’ contains two instances of the use of “a’ as a neuter plural of appella- i tive nouns ; e.g., ‘ Agula nira pira,’” vain shows (are all) other (things). 0 The first of these three words is used adjectivally; and in that case the final ‘a’ is merely that which remains of the neuter termination 0 198 THE NOUN. ‘am,’ after the regular rejection of ‘m ; but the next two words ‘nira’ and ‘pira,’” are undoubted instances of the use of ‘a’ as a suffix of the neuter plural of appellatives. The much used Tamil words ¢ pala,” several, or many (things), and ‘sila,” some, or some (things) (from ‘pal’ and ‘sil’), though commonly considered as adjec- j tives, are in reality neuter plurals; e.g., ‘pini pala,” diseases (are) many ; ‘ pala-(v)-in pal,’ the neuter plural gender, literally, the gender of the many (things). The use of these words adjectivally, and with the signification, not of the collective, but of the distributive plural, has led some persons to overlook their origin and real meaning, but I have no doubt that they are plurals. So also ‘alla,” not, is properly a plural appellative : it is formed from the root “al,’” not, by the addi- tion of ‘a,’ the plural suffix, and literally means things that are not, and the singular that corresponds to ‘alla’ is ‘al-du,’ not, euphonically ‘andru,’ literally a thing that is not. In the higher dialect of the Tamil, all nouns of quality and relation may be, and very frequently are, converted into appellatives and pluralised by the addition of ¢a e.g., ‘ariya,’ (Kural), things that are difficult, ‘dificilia.’ We have some instances in High Tamil of the use of ‘a’ as the plural suffix even of substantive nouns, e.g., ‘porula,’ substances, things that are real, realities (from the singular ‘porul,’ @ thing, @ substance) ; also ¢porulana’ and ¢ porulavei,’—with the addition of ‘ana’ and ‘avei’ (for ‘ava’), the plural neuters of the demonstrative pronouns. The neuter plural of the third person of the Tamil verb, a form which is used occasionally in ordinary prose as well as in the classical ‘ana 2 dialect, ends in ‘amna; eg., ‘irukkindrana,” they (neut.) are. is undoubtedly identical with ‘ava’ (now ‘avei’), the neuter plural — of the demonstrative pronoun, and is probably an older form than A ‘ava.’ It is derived from the demonstrative base ‘a,” with the addi- tion of ‘a’ the neuter plural suffix, and an euphonic consonant (‘n’ or ‘v’) to prevent hiatus; e.g., ‘a-(n)-a’ or ‘a-(v)-a. Sometimes in classical Tamil this ‘a,’ the sign of the neuter plural, is added directly to the temporal suffix of the verb, without the addition of the demon- strative base of the pronoun, e¢.g., ‘minda,’ they (neut.) returned, instead of ‘mindana.’ This final ‘a’ is evidently a sign of the neuter plural and of that alone. Possibly we should also regard as a sign of the neuter plural, the final ‘a’ of the high Tamil possessive adjectives ‘ena, my (things), mea; ‘nama, our (things), nostra. The final ‘a’ of ‘ena’ would, on this supposition, be not only equivalent to the final ‘a’ of the Latin ‘mea,’” but really identical with it. These possessive adjectives are regarded by Tamil grammarians as genitives; and it will be shown 199 NUMBER—THE NEUTER PLURAL. hereafter that ‘a’ is undoubtedly one of the forms of the genitive in Is the Dravidian languages. The real nature of ‘ena’ and ‘nama’ 1l will be discussed when the genitive case-terminations are inquired r’e into. It should be stated, however, under this head, that Tamil 0~ grammarians admit that ‘ena’ and ‘nama,’ though, as they say, genitives, must be followed by nouns in the neuter plural ; e.g., ‘ena ) (o keigal, my hands; and this, so far as it goes, constitutes the principal i arcument in favour of regarding the final ‘a’ of these words, not as l, a genitive, but as the ordinary neuter plural suffix of the high dialect. In Malayalam, the oldest daughter of the Tamil and a faithful & preserver of many old forms, the neuter plurals of the demonstrative pronouns arve ‘ava, those (things), and ‘iva,’ these (things). The Ji- existence, therefore, in Tamil and Malayalam of a neuter plural in 0 [’ short ‘a, answering to a neuter singular in ‘d,” is clearly established. ly The Canarese appears to have originally agreed with the Tamil in ho all the particulars and instances mentioned above: but the neuter ‘]y plural in ‘a’ is now generally hidden in that dialect by the addition Ly of a formative or euphonic syllable. Thus ¢ pira,” Tam., other (things), ne is in Canarese ‘peravu,” of which the final syllable ‘vu’ is undoubtedly of s an euphonic addition—an addition of which the Canarese is very fond. nd The neuter plural of the demonstrative pronoun is not ‘ava’ in he Canarese, as it is in Malayalam, and as it must have been in primitive Tamil, but ‘avu.’ Though, however, the nominative is ¢avu, all the oblique cases in the ancient Canarese reject the final ‘u’ before il receiving the case suffixes, and must have been formed from the base al of an older ‘ava; eg., ‘avara’ (‘ava-ra’), of those things. i) The Telugu plural neuters of the demonstratives are avi,’ those, al “ivi,” these, answering to the singular neuters ‘adi’ and ‘idi’ The all | oblique forms of the same demonstratives, to which the case-termina- li- tions are suffixed, are ¢vA’ remote, and ‘vi’ proximate, which are evidently formed (by that process of displacement which is peculiar to n the Telugu) from the primitive bases ‘ava’ and ‘iva, like ¢véru’ ly from ‘avaru,” and ‘viru’ from ‘ivaru.’ - The neuter plural of the Telugu verb is formed by suffixing ¢ avi i, OFs & Vi .’ el In Gond the singular demonstratives are ‘ad’ and ‘id;’ the cor- responding plurals ‘av’ and ‘iv.’ he If the Telugu and the Gond were the only extant dialects of the /u‘); Dravidian family, we should naturally conclude that as ‘d’ is the on sign of the neuter singular, so ‘v’ is the sign of the neuter plural. fin When the other extant dialects, however (the Tamil, Malayalam, and e Canarese), are examined, we perceive that this ‘v’ is not a sign i 200 THE NOUN. of plurality, nor a sign of anything but of abhorrence of hiatus; and that it is merely an euphonic link between the preceding and succeed- ing vowels. The Telugu and Géond must therefore yield to the over- powering weight of evidence which is adducible in proof of this point from their sister dialects. Nor is there anything opposed to analogy in the supposition that the Telugu has changed the ‘a,” which was the sign of the neuter plural of its pronouns and verbs, into ‘i ) and then adopted to represent the idea of plurality a consonant which was used originally merely to prevent hiatus. In the case of ‘avaru, they, “ille, converted into ¢ varu,” and ‘ivaru,’ they, ‘hi,’ converted into fving " ¥ 2 though only euphonic in its origin, has become an initial and apparently a radical ; aud the old initial and essentially demonstrative vowels ‘a’ and ‘i,” have been thrust into a secondary place. The conversion, therefore, of ‘ava’ into ‘v4,’ and of ‘iva’ into ¢ vi,” the oblique forms of the Telugu plural demonstratives, is directly in accordance with this analogy; and thus the Telugu cannot be considered as opposed to the concurrent testimony of the other dialects, which is to the effect that ‘v’ is merely euphonic, and that ‘a’ is the sign of the neuter plural of the demonstrative prououns. It may here be remarked as a curious irregularity, that in Tulu ‘v’ has become the sign of the neuter singular instead of ‘d,” eg., “avu,’ o The Tulu ‘atu,” corresponding to the Tamil-Canarese ‘adu,’ which should have been used to signify 4, has come to be used for yes. If short ‘a’ be, as it has been shown to be, a sign of the neuter plural which is inkerent in the Dravidian languages, and most used by the oldest dialects, we have now to inquire into the relationship which it evidently sustains to the neuter plural suffix of some of the Indo- European languages. I know of no neuter plural in any of the Scythian tongues with which it can be compared ; and we appear to > be obliged to attribute to it, as well as to ¢d , the suffix of the neuter singular, an origin which is allied to that of the corresponding Indo- European forms. In the use of ‘a’ as a neuter plural suffix, it is evident that the Dravidian family has not imitated, or been influenced by, the Sanscrit, and that it was not through the medium of the Sansecrit that Indo—European influences made their way into this department of the Dravidian languages; for the Dréividian neuter plural <4, differs widely from the Sanscrit neuter plural ¢4ni,” and it 1s as certainly unconnected with the masculine-feminine plural ‘as’ (softened in modern Sansecrit into ‘ah > . It is with the short ‘a which constitutes the neuter plural of the Zend, the Latin, and the Gothic, that the Dravidian neuter plural ¢a’ appears to be allied. Compare also the Old Persian neuter plural <.’ R — — CASE-FORMATION. 201 d Part IT.—Formation of Cases. 1t Principles of Case-Formation.—The Indo-European and the Scythian 4] families of tongues originally agreed in the principle of expressing the 16 reciprocal relations of nouns by means of post-positions or auxiliary i words. The difference between those families with respect to this d point consists chiefly in the degree of faithfulness with which they have retained this principle. In the Scythian tongues, post-positions or appended auxiliary \i words have rigidly held fast their individuality and separate existence. In the Indo-European tonguues, on the contrary, the old post-positions or suffixes have been welded into combination with the roots to which I 1§ they were appended, and converted into mere technical case-signs or h inflexional terminations; whilst in the later corruptions to which those 0 languages have been subjected, most of the case-terminations have b been abandoned altogether, and prepositions, as in the Semitic tongues, T have generally come to be employed instead of the older case-signs. It cannot reasonably be doubted, that the case-terminations of the primitive dialects of the Indo-European family were originally post- positions, which were added on to the root to express relation, and at length blended into an inseparable union with it, through that love of composition by which every member of the family was characterised. In most instances the root and the original signification of those post. 4 positions are now unknown, or they are ascertained with difficulty by h means of analogy and comparison. Both in Greek and in Latin some post-positions are used in a manner 0 which illustrates the conversion of a portion of this class of words into case-endings; e.g., in Latin ‘nobiscum,” and in Greek such words as CaypoOu, inthe country; < dNade,’ to the sea; and ‘dvpavelev,’ from heaven. T The post-positional auxiliary words which are used in those instances are appended to their bases in a truly Scythian manner. If there is 5 any difference between them and the usage of the Scythian post- 4 positions, it consists in this—that in the Scythian tongues, €0’ ¢ ée, ¢ 9ev, would be appended to the nominative; whereas in Greek, they b T are suffixed either to a crude form of the noun differing from the nominative or to the accusative ; and also, that in most of the languages t :;: of the Scythian group they would be written as separate words. s .i & One of the Greek post-positions quoted above, ¢3¢, signifying direction to @ place, has been supposed to be allied to ‘de,” the dative of the Manchu ; and the Greek 6cv’” has been conjectured to be allied to the Tartar ablative ‘din’ or ‘den.’ I am doubtful whether any 202 THE NOUN. such connexion can be established ; but in the manner in which the particles are appended to their bases a distinct analogy may be observed. - On turning our attention to the Dravidian languages, we find that the principle on which they have proceeded in the formation of cases is distinetively Scythian. All case-relations are expressed by means of post-positions, or post-positional suffixes. Most of the post-positions of the Telugu are, in reality, separate words; and in all the Dravidian dialects most of the post-positions retain traces of their original character as auxiliary nouns. Several case-signs, especially in the more cultivated dialects, have lost the faculty of separate existence, together with their original signification, and can only be treated now as case-terminations; but there is no reason to doubt that they were all post-positional nouns origiually. There is another point in which the Scythian principles of case- formation differ materially from the Indo-European. In the Indo- European family the case-endings of the plural differ from those of the singular. It is true, that on comparing the case-terminations of all the members of the family, some traces have been discovered of the existence of an original connexion between the singular and the plural terminations of some of the cases; but in several instances, e.g., in the instrumental case, no such connexion between the singular and the plural has been brought to light by any amount of investigation ; and it may be stated as a general rule, that the languages of this family appear to have acted from the beginning upon the principle of expressing the case-relations of the singular by one set of forms, and the case-relations of the plural by another set. On the other hand, mn all the languages of the Scythian group, the same case-signs are employed both in the singular and in the plural, without alteration, or In the with only such alterations as euphony is supposed to require. singular, the case post-positions are appended directly to the nomina- tive, which is identical with the base: in the plural they are appended, i‘i not to the nominative or base, but to the particle of pluralisation which has been suffixed to the base. In general, this is the only difference between the singular case-signs and those of the plural. The only exception of importance is, that in some of the Scythian tongues, especially in the languages of the Finnish family, the included vowel of the case-sign differs in the two numbers : it is generally ‘a’ ) in the singular and ‘e in the plural—a change which arises from the ¢law of harmonic sequences’ by which those tongues are characterized, and which re-appears, but little modified, in the Telugu. When the Dréavidian languages are examined, it is found that they I | | H% I — NUMBER OF DECLENSIONS. 3 differ from those of the Indo-European family, and are in perfect accordance with the Scythian group, in their use of the same signs of case in the plural as in the singular. The only exception is the truly Seythian one which is apparent in the Telugu, in which the dative case -sign is either ‘ki’ or ‘ku,” according to the nature of the vowel by which it is preceded or influenced ; in consequence of which it is generally ‘ki’ in the singular and ¢ku’ in the plural. This identity of the singular and plural case-endings in the languages of the Scythian group, including those of the Dréividian family, will be found greatly to facilitate the comparison of the case- signs of one language of either of those families with those of the other. Number of Declensions.—There is only one declension, properly so called, in the Dravidian languages, as in the Scythian family gene- rally. Those varieties of inflexional increments which have been called ¢ declensions’ by some European scholars, especially with reference to the Canarese and Telugu, are considered by native grammarians to constitute but one declension ; and in truth they do constitute but one, for there is no difference between one so called declension and another with respect to the signs of case. Those signs are precisely the same in all : the difference which exists relates solely to suffixes of gender, or to the euphonic and inflexional increments which are added on to the bases before the addition of the case-signs. On proceeding to analyse the case-formation of the Dravidian languages, we shall follow the order in which they have been arranged by Dréividian grammarians, which is the same as that of the Sanscrit. The imitation of the Sanscrit in this particular was certainly an error; for whilst in Sanserit there are eight cases only, the number of cases Every post-position annexed to in Tamil, Telugu, &ec., is indefinite. a noun constitutes, properly speaking, a new case; and therefore the kot number of such cases depends upon the requirements of the speaker and the different shades of meaning which he wishes to express. Notwithstanding this, the usage of Drévidian grammarians has restricted the number of cases to eight; and though there are disad- vantages in this arrangement, it will conduce to perspicvity to adhere to the ordinary usage in the analysis on which we are about to enter. Tamil grammarians, in following the order of the Sanscrit cases, have also adopted or imitated the Sanscrii mode of denominating them— not by descriptive appellations, as ¢dative’ or ‘ablative,” but by numbers. They have affixed a number to each case in the same order 204 THE NOUN. as in Sanserit, e.g., ‘first case,” ‘second case,’ &e., to ‘eighth case.’ Though a nominative, or ‘first case,” stands at the head of the Drévidian list of cases, the only cases, properly so called, which are used by these languages are the oblique cases. Instead, therefore, of proceeding to examine the Dravidian nominative, the particular which now falls to be noticed is— The absence of Nominative Case-Terminations.—The Dravidian A nominative singular is simply ¢pejr- 6 * “the noun ilself’—the inflex- ional base of the noun—without addition or alteration; but it neces- sarily includes the formative, if there be one. The nominative plaral differs from the nominative singular only by the addition to it of the pluralising particle. There are three apparent exceptions to this rule, or instances in which the nominative might appear to have terminations peculiar to itself, which it is desirable here to inquire into. (1.) The neuter termination ‘am’ might at first sight be sup- posed to be a nominative case-sign. In Sanscrit ¢ am is the most common sign of the nominative neuter ; and in Tamil also, all nouns ending in ‘am’ (in Telugu ‘am-u’), whether Sanscrit derivatives or pure Dravidian words, are neuter abstracts. In Sanserit the accusative of the neuter is identical with the nominative, but in the other cases ‘am’ disappears. In Tamil, ‘am’ is discarded by all the oblique cases of the singular without exception : every case retains it in the plural, but in the singular it is used by the nominative alone. This comprises the sum total of the reasons for regarding ‘am’ as a termi- nation of the nominative. On the other hand, though ¢am’ disappears in Tamil from the oblique cases in the singular, it retains its place in every one of the cases in the plural. The particle of \plurality is % regularly suffixed to ‘am,’” and the signs of case are then suffixed to the particle of plurality ; which is a clear proof that, whatever ¢ am’ may be, it is not a mere termination or case-sign of the nominative. The ;( Telugu regards ‘am or ‘am-u,” as part of the inflexional base, retains it in each case of botk numbers alike, and suffixes to it in the singular the case-signs, in the plural the particle of plurality. The modern Canarese makes no use whatever of this termination ‘am,” In any case, or in either number. The ancient Canarese uses it, like the Tamil, in the nominative singular, but discards it, not only in the oblique cases of the singular, but in every case of the plural also. In that dialect a ¢re¢ is ‘maram,” as in Tamil, but the plural nominative, rees, is not ‘maranggal’ (‘maram-gal’), but ‘mara-gal.’ Neuter nouns borrowed from the Sanscrit by the Tamil ordinarily I ~ THE. NOMINATIVE. 20 retain (in the nominative alone, in the singular) the ‘am’ of the San- scrit nominative singular : this ‘am’ is used in every one of the cases in the plural ; so that even in Sanscrit derivatives, ‘am’ is regarded in Tamil, not as a case-sign, but as a portion of the inflexional base. Whatever be the origin of the Tamil ‘am, considered (as I think we must consider it) as a formative, it does not appear to have been borrowed from the Sanscrit, in which it is used for so different a purpose ; and 1 question whether it does not spring At all from a source altogether independent of the Sansecrit. events we find it added to many of the purest Dravidian roots, and by the addition of it many verbs of that class are converted into nouns. Thus ¢ kulam, Tam., ¢ tank, is from kuli, o bathe ; and ‘nil-am,’ Tam., the ground, is from nil;" o stand. See ‘Derivative Nouns,” 1 the section on ¢ The Verb. (2.) In Canarese the crude form of the personal pronouns is occa- sionally used instead of the nominative, e.g., ‘nd,” instead of ‘nanu,’ 7, and ‘t4, instead of ¢tAnu,’ self; and hence it might be supposed that the final ‘n’ or ‘nu’ of those pronouns constitutes a nominative termination. This supposition, however, is inadmissible ; for in all the oblique cases, without exception, the final ‘n’ or ‘nu’ retains its place, and it is to it that the signs of case are added. Consequently it is evident that ¢n’ is not a sign of the nominative, but a formative, which has been compounded with the inflexional base, or annexed to it, though it is capable of occasional separation from it. (3.) In all the Dravidian languages, the quantity of the included vowels of the personal pronouns in some of the oblique cases (and in Tamil-Canarese in all the oblique cases), differs from the quantity of the same vowels in the nominative. In the nominative the vowel is invariably long, in the oblique cases generally short : e.g., in Canarese we find ‘nanu, 7, ‘nanna,’ my; ‘ninu,’ thou, ‘ninna,’ thy ; ¢ thnu,’ self, ‘tanna,’ of one’s self. This is the only instance contained in these languages in which there is a difference between the nominative and stk the oblique cases of such a nature as almost to constitute the nomina- tive a case by itself. In this instance, however, the nominative is the true, unchanged, inflexional base, and the shortening of the quantity of the vowel in the oblique cases, prior to the addition of post-positions, The has arisen from the euphonic tendencies of the language. Telugu shortens the root-vowel in the accusative only. In Tamil the shortened form, without any inflexional addition, is often used as a possessive; e.g., ‘nin, thy, from the obsolete ‘nin, thou—a usage which is in accordance with the ordinary Dravidian rule that the 206 THE NOTUN. inflected form of every noun, or the basis of the oblique cases, is to be regarded as of itself a possessive or adjective. Before proceeding to consider the oblique case-signs seriatim, it is necessary to enquire into the changes which the base sustains prior to receiving the suffixes. Inflexional base of the oblique cases.—In the majority of instances that form of the Dravidian noun which constitutes the erude base, and which is used as the nominative, constitutes also the inflexional base. The nominative of this class of nouns and the base of the oblique cases are identical ; and the case-signs are added to the base or nomi- native without any link of connexion, whether inflexional or euphonie, beyond the ordinary ‘v’ or ‘y,” which is inserted to prevent hiatus between concurrent vowels. In a smaller number of instances (a number which constitutes, however, a very large minority), the base or nominative undergoes some alteration before receiving the addition of the terminations, or case-signs of the oblique cases. In the solitary instance of the Tamil-Canarese personal pronouns, as pointed out under the preceding head, the nominative sustains a cur- tailment (viz, by the shortening of the quantity of the included vowel) on becoming the inflexional base, or base of the oblique cases : but in all other instances the alteration which the base sustains consists in an augmentation, which is sometimes optional and sometimes neces- sary ; and it is to this augmented form (augmented by the addition of some inflexional increment) that the case-signs are attached. This Dravidian rule may be illustrated by the Hebrew. In Hebrew the personal and other suffixes of substantives and verbal nouns are attached, not to the base or nominative, but to the ¢construct state,’ v e, the state in which a noun stands when it is qualified by a subse- quent noun. Just so in the Dravidian languages, in that large class of nouns in which the inflexional base of the noun, or its adjectival form, differs from the crude form or nominative, the signs of case are attached not to the crude, natural form of the noun, but to the altered, inflected form, viz., to that form which a Dravidian noun assumes when it % | qualifies or is qualified by asubsequent noun, or when it stands to such noun in the relation of an adjective. This inflected form of the noun is frequently used by itself, without the addition of any case-termina- tion, and when so used it has a possessive or adjectival force. Tamil :E grammarians hold that the ¢ inflexion’ is not a possessive, though they cannot but admit that for every purpose for which the possessive or genitive is used; the ‘oblique case,” or inflected form of the noun may i INFLEXIONAL INCREMENTS. 207 )¢ be used instead. They admit that it is used adjectivaliy: but it appears to me that its use as an adjectival formative is a secondary 1§ one, and that it was originally, like many other adjectival formatives 0 Its use eventually as the in various languages, a sign of the genitive. inflexional basis of all the cases is in perfect harmony with this view of its origin, and testifies to the existence of a period in the history of ) the language when each of the post-positions of case was known and d felt to be a substantive, which required to be united to its base by a sign of possession or adjectival relationship. 16 At present, however, it is our object to seek out and arrange the various increments which are used for forming the inflexional base Gy of the oblique cases, without reference to the other uses to which those 13 increments are put. § © (1.) The inflexional increment “in, with its varicties.—The particle “in’ constitutes the inflexion of certain classes of nouns in Tamil- I Canarese; and the corresponding Telugu particles are ni and ‘na.’ All these particles are, I believe, virtually one and the same. The V] Tamil uses ‘in’ in the singular and in the plural alike; and its original signification has been forgotten to such a degree that it is now often used merely as an euphonic link of connexion between the base and its case-signs. For this reason its use both in Tamil and in Cana- rese is optional. In Telugu the corresponding particles are used only in the singular ; and where they are used, their use is not euphonic f merely, but is intended to constitute the ¢inflexion.’ 8 The Ku, which in this respect is more nearly allied than the Telugu is to the Tamil, and more regular, uses ‘ni’ as the inflexion of the plural as well as of the singular of all classes of nouns. When ¢in’ is used in Tamil as the inflexion of the neut. sing. demonstratives ‘adu, that, ‘idu, this, it is apt to be confounded with ¢an. a termination which those pronouns often take, especially in the 9 Instead of ‘adu’ and ‘idu, we may oblique cases, instead of ‘u say in Tamil ‘adan’ and ‘idan.’ In the nominative these forms are very rarely used ; but the accusative, ‘adan-ei,” is more common, and % the dative, “adarku’ (‘adan-ku’), still more so. “id-in-al,” through i s this, ‘ad-in-al,” through that, and cases similarly formed, must therefore The ¢an’ be carefully distinguished from ¢idan-al’ and “adan-al. of the latter is a formative, which is probably of the same origin as the ‘am’ of many neuter nouns (that ‘am’ being almost always con- vertible into ¢an’); whereas ‘in’ is an inflexional increment, and was originally a case-sign of the genitive. The use of “in’ as an inflexional increment effects no alteration Pe— 208 THE NOUN. whatever in the meaning of the case-sign which is suffixed to it. { ? Where it is not followed by a case-sign, it becomes of itself a mode of expressing the genitive ; but where a case-sign follows, it is merely euphonic, and its use is optional. Thus, we may say either ¢keiyal’ (‘kei-(y)-al’), with the hand, or ‘keiyin-al’ (‘kei-(y)-in-al); either “kalal, with the foot, or ‘kalinal’ (‘kal-in-al’). In the first of these instances (‘kei-(y)-in-al’), ‘y’ is used to keep the initial vowel of ‘in’ pure, in accordance with the ordinary rule of the language ; from the use of which, in this instance, it is evident that ‘in,’ though merely euphonic in its present application, was in its origin something more than a mere euphonic expletive. ‘In’ 1s not only attached as an inflexional increment to the crude base of Tamil nouns, but it is appended also to other inflexional inere- ments, viz., to ‘attu,” and to the doubled final ‘d’ and ‘R’ of certain classes of nouns. Thus, by the addition of ‘attu’ to ‘mara-m,’ @ tree. we form ‘marattu,’ the inflexional base of the oblique cases, by suffix- ing to which ‘al,’ the sign of the instrumental case, we form ‘marattal,’ by @ tree: but we may also attach ‘in’ to ‘attu,’ forming ‘attin’ (‘att-in’), a doubled and euphonized increment, e.g., ‘marattinal’ (‘mara-attu-in-al’). As ‘in’ when standing alone, without the suffix of any case-sign, has the force of the genitive, so also has the double increment, attin ;' e.g., “marattin’ signifies of a éree. In; Tamils Gt is the ‘inflexion’ of all nouns, except those which end in ‘am,” or in ‘d-u’ or ‘rR-u: in Canarese ‘in’ is much more rarely used than in Tamil ; but where it is used, its use is rather euphonic and optional, than inflexional, and it cannot be used by itself to express the force of the genitive. As in Tamil ‘guruvil, in a priest, and ‘¢ guruvinil’ are identical, so we may say in Canarese either ‘guruvalli’ or ¢ guru- vinalli.’ In Telugu the corresponding particles ‘ni’ and ‘na’ constitute the inflexion, or natural genitive of certain classes of nouns, and are also attached as inflexional increments to the base before suffixing the case- signs; e.g., ‘diniki’ (‘di-ni-ki’), ¢o i¢, ‘tammuniki’ (‘ tammu-ni-ki 2140 a younger brother, ‘guru-na-ku, to a spiritual teacher. These incre- ments are attached only to the singular in Telugu : they constitute the singular inflexion,” 4.c., the genitival or adjectival base of the e — noun, and are not merely euphonic ; nor are they to be regarded as the inflexion of masculine nouns and pronouns alone, though they are chiefly used by them, for ¢daniki, to that, ¢diniki,’ to this, are ii neuters, The Telugu ‘ni,” and the Tamil-Canarese ‘in,” are doubtless iden- tical in origin. The change in the position of the vowel is in accord- iz INFLEXIONAL INCREMENTS. 209 A ance with the change of il; Tam., the negative particle, into *18&; in f Telugu, and of ¢ul, Tam., witkin, into €16, in Telugu. It also cor- responds to the change of the position of the vowel which is apparent when ‘¢in,’ the Latin preposition, is compared with the corresponding i\ Sanserit preposition ¢ni.’ (23 The inflewional increments ‘ad’ and ¢ ar’—The particles ¢ Gl f and ‘ar, are extensively used by the Canarese as inflexional incre- ments. Their use exactly resembles that of ‘in’ in the same language, i} though each is restricted to a particular class of words. ‘in’ is used as an increment of the base solely in connexion with nouns which end in ‘u,’ eg., ‘guru, @ priest; and ‘ad’ and ‘ar’ are used in connexion with neuter nouns and demonstratives, and with those alone. In the Canarese genitive case-endings, ‘ara,” ‘ada,’ ‘ina,’ and ‘a, it will be seen that the real and only sign of the genitive is ‘a,’ the final vowel of each; and therefore Dr. Stevenson has erred in comparing ‘ara’ or ‘ra’ (properly ‘ar-a’ or ‘ad-a’) with the New Persian ‘ra.’ ¢ad’and ‘ar’ are prefixed to the signs of case, not by the genitive only, but by three cases besides, viz, by the accusative, the instru- mental, and the locative. Thus we may say not only ‘idara’ (‘id-ar-a’), e of this, and ‘marada’ (‘mar-ad-a’), of a free, but also ‘idaralli’ (‘id-ar-alli’), in this, and ¢maradinda’ (‘mar-ad-inda’), by a tree. Consequently ‘ad’ and ‘ar,” whatever be their origin, are not signs of case, in so far as their use is concerned, but are used merely as incre- ) ments of the base, or inflexional bonds of conjunction between the base Moreover, the Canarese differs f and the case-signs, like ‘in,” “ni,” &e. in its use of these increments from the Telugu and the Tamil in this, that it never suffixes them alone without the addition of the case-signs, and never gives them the signification of genitives or adjectival for- matives. ‘ad’ and ‘ar’ are not only related, but are, I believe, really identical. Both are increments of the neuter alone; and where the ¢d’ and ‘r’ are known to 0 Canarese uses ‘ar,’ the Tulu uses ‘ad. change places dialectically, as in the southern provinces of the Tamil country, in which ‘adu,’ ¢, is pronounced ‘aru ;7 and the Canarese increment ‘ad’ or ‘ar’ is, I have no doubt, identical with that very word, viz., with the Tamil-Canarese demonstrative ‘adu’ or ‘ad,’ . Though the Tamil has not regularly adopted the unchanged form of this demonstrative, ‘adu,” as an inflexional increment of the base in the declension of nouns, it makes use of it occasionally in a manner “% which perfectly illustrates the origin of the Canarese use of it. In classical Tamil the neuter demonstrative may optionally be B 210 THE NOUN. added to any neuter noun in the singular, not for the purpose of alter- ing the signification, but merely for the improvement of the euphony, and for the purpose of meeting the requirements of prosody. ¢adu may thus be added even to the nominative; e.g., we may not only write ‘pon,’ gold, but also poetically ‘ponnadu,” gold, or etymologically gold-that, i.e., that (which is) gold. It is much more common, however, and more in accordance also with the Canarese usage, to use ‘adu’ in the oblique cases ; in which event it is inserted between the base and {i the case-sign, so as to become virtnally (yet without losing its proper ¢ character) an inflexional increment; e.g., instead of ‘ponnei,” the accu- sative of ‘ pon,” gold, we may write ¢ ponnadei’ (‘ponn-ad-ei’). I connect with the Canarese ‘ar,’” and therefore with ‘ad,’ and ultimately with the neuter demonstrative itself, the euphonic consonant b tr, which is used by the Telugu in certain instances to separate between a Sanscrit noun of quality used as an adjective and the 1 feminine suffix ¢4lu,” e.g., ¢ sundaru-r-alu,” a handsome woman. This {1 would be quite in accordance with the peculiar Telugu usage of employing the neuter demonstrative singular in place of the feminine C singular. €4 al (8.) The inflexvonal increment ¢ ti’—In Telugu €41’ or ‘ti’ is the th most common and characteristic inflexional increment of neuter singular nouns, and it is used in Telugu, like the corresponding ‘attu’ in Tamil, fi not merely as an increment of the base, but as the ‘inflexion,” with the signification of the possessive case or of that of an adjective, as the context may require. Two instances of the use of this increment will suffice out of the very numerous class of neuter nouns which form 1l their singular inflexion by the addition of ‘i’ or ‘ti’ (or rather by the substitution of that particle for their last syllable); e.g., ‘vagili,” a door ! b way, inflexion ¢ vagiti; ‘nudaru,’ the forehead, inflexion ‘nuduti.’ In (A these instances of the use of ‘t1’ or ti,” the inflexional inerement is substituted for the last syllable ; but it is certainly to be considered as (r Ig an addition to the word—as a particle appended to it; and the blending of the increment with the base, instead of merely suffixing it, has arisen from the euphonic tendencies of the language. { I have no doubt that the suffixed particle which constitutes the Telugu inflexional increment was originally ¢ti,” not ¢ ti’—the dental, 1l i 1 not the cerebral. This would account for the circumstance that ¢+’ i I alone follows words of which the final consonant is ‘r’ or ‘1;’ for on the addition of the dental ‘t’to ‘r’ or ‘1’ both consonants dialec- tically coalesce and become ‘t;’ the hard cerebral being regarded as I euphonically equivalent to the two soft letters. In no case in Telugu fo INFLEXIONAL INCREMENTS. P ) 11 is there a double ‘t’ in the inflexional increment. It is sometimes, y however, euphonised by prefixing a nasal, e.g., ¢ tolli,’ antiquity, forms its inflexion in ‘tonti’ (instead of °tolti), or ‘tollinti.’ The dental J “ti’ is used instead of the cerebral ‘ti, as the inflexion of nouns y ending in ‘yu’ after a pure vowel, e.g., ‘vayu,’ the mouth, inflexion 3 Cyati.’ This circumstance proves that it was the dental ¢ti’ which was originally used in all cases. The dental ‘t’ on being appended to consonants changes naturally into the cerebral ; whereas the T cerebral rarely, if ever, changes into the dental. If we now conclude, as I think we undoubtedly may, that the Telugu inflexion was originally ¢ti,’ not ¢ti,’ this inflexional increment d may at once be connected with the Telugu neuter demonstrative, ‘adi,’ 1t in the same manner as the Canarese ‘ad,’ and the Tamil ¢attu,’ are f connected with the Tamil-Canarese neuter demonstrative ¢ adu.” This 16 I8} identification is confirmed by the circumstance that ‘atti’ is some- times used for ‘adi’ in Telugu, and ¢itti’ for ¢idi,’ just as ‘attu’ is of sometimes used for ‘adu’ in colloquial Tamil. Though the identifi- 10 cation of the inflexion and the neuter sing. demoustrative could not easily be established from the Telugu alone, or from any one dialect alone, yet the cumulative argument derived from a comparison of all 10 the dialects has irresistible force. An important link of evidence is ar furnished by the inflexion which follows.— l I (4.) The inflexional increment ‘attu’ or ‘attrun.’—All Tamil nouns he which end in ‘am, whether Sanscrit derivatives or pure Tamil roots, ill reject ‘am’ in the oblique cases in the singular, and take ‘att-u’ instead ; and it is to this increment that the various case-signs are I suffixed: e.g., the locative case-sign “il’ is not added to ° aram,” depth, ) but to the inflexional base ‘Ar-attu,’ so that, in the depth is not n “aram-il, but ¢ar-att-il.’ This rule admits of no exception in the 18 ordinary dialect of the Tamil ; but in the poetiéa,l dialect, which a8 represents more or less distinctly an older condition of the language: I ¢attu’ is sometimes left unused, and the case-sign is added directly to 49 the crude base: e.g., instead of ‘kay-attu-kku, to the depth (from ‘kayam,’ depth’), ¢ kaya-kku’ is used in the Chintdmani. When the he increment ‘attu’ is not followed by any sign of case, but by another al, noun, like the other inflexion ‘in’ and like the corresponding Telugu t, inflexion ¢ ti,” it has the force either of the genitive or of an adjective ; or e.g., ‘mar-atta koppu, the branch of a tree, ‘kul-attu min,’ tank fish. id This inflexion, like ‘ad’ and ‘ar’ in Canarese, and ‘ti,’ or ‘ti’ in # Telugu, is used in connexion with the singular alone. ‘am,” the i formative of the base, which is used only by the nominative in the P2 212 THE NOUN. singular, is retained in the plural, not in the nominative only, but in all the oblique cases. To it the sign of plurality is appended, and the case-sign follows the sign of plurality; e.g., ‘maranggalil’ (‘ maram- gal-i1"), in trees. There are in Tamil a few naturally plural (neuter) pronominals and nouns of relation (e.g., ‘avel,’ those (things); ‘sila, few,; ‘pala, many; ‘ella,” all), which receive in their oblique cases the inflexional increment ‘arru,’ pronounced ‘attru.’ Thus, from ‘ellam,” all, which is properly ‘elli-um’ (‘um’ being the conjunctive and intensitive particle even, and elld-um’ or ‘ellam’ signifying even all, all together), the locative which is formed by the Tamil is ¢ellavattrilum, (ella-(v)- attr’-il-um), én all, literally even wn all. So also “avei,’ they (neut.), forms its accusative not by adding ‘el the accusative case-sign, to ‘avei, but by inserting ¢attru, and adding ‘ei’ thereto, e.g., ‘avattrei’ (‘av-attr-ei’), them ; in which instance ‘ei’ (for ¢a’), the sign of the plural, is rejected, and its place is supplied by attru,” the inflexional increment of this class of plurals. E It is evident that the Tamil increments, ‘attu’ and ‘attru,” are virtually identical. The difference in use is slight, and in pronunci- 1 ation still slighter ; and in general ‘attru’ is pronounced exactly like ‘attu’ by the vulgar. We may, therefore, conclude that they are one and the same, and on examining the Telugu we find additional confirmation of their identity. In Telugu, avi, they (neut.), answer- ing to the Tamil ‘avei,” forms its inflexion in ‘vati’ (for ‘avati’) This Telugn (supposititious) ‘avati’ is evidently identical with the Tamil ¢avattru.’ The “ti’ of this inflexion is certainly the same as the ti’ of Telugu nouns substantive : and if there is no difference a in Telugu between the ‘ti’ which forms the inflexional increment of l neuter singular nouns and demonstratives and the plural inflexion ‘ti’ I\( of such words as ¢ vati,’ we may also conclude that there is no real dif- P ference between the singular ‘attu’ and the plural ‘attru’ of the Tamil. i Whence did the ‘r which is included in ‘arru, or ¢attr-u,’ take its rise ? 4 We see its origin clearly enough in Canarese; for in the ancient dialect ‘ar,” or ‘r,’ forms the inflexional increment of I every one of the plural pronominals which take ‘arru’ in Tamil: 0 e.g., ‘avara’ (corresponding Tam. ‘avarru’), of thosc things; ‘ellavara’ (Tam. ‘ellavarru’), of all things,; ‘kelavara’ (Tam. ¢silavarru’ /3 ) R —— of some (things). The Canarese ‘r’ is, as we have seen, derived from, and originally identical with, ¢d’ or ‘t;’ and hence the Tamil in doubling ‘R’ gives it the sound °ttr. Thus, not only the Tamil increment ‘att-u,’ but also ‘arr-u,’ is clearly derived from the same origin as the Canarese ‘ad’ or ‘ar,” and the Telugn ‘ti,’ viz., from INFLEXIONAL INCREMENTS. 213 Both these inflexions have been the neut. sing. demonstrative. 0 formed also by the same process; for ‘ar,” when doubled, becomes - ‘arr-u’ (‘attr-u’), as naturally as ‘ad,’ when doubled, becomes ‘att-u ; and in each case the doubling arises from the adjectival use to which the suffixed pronoun is put. Tt is a recognized rule of the Sw l Tamil that when a noun ending in ‘d-u’ is used adjectivally, the al td:n may either become ‘d-in’ or ¢tt-u; e.g., from ¢erud-u,’ an h ox, 1s formed either ‘erud-in’ or ‘erutt-u,’ of an ozx. So also ¢ ad-u,’ b 6 it, which is now generally inflected by the addition of ‘in 2 seems to ’)7 have been inflected formerly as ¢att-u.’ “adu’ is vulgarly pro- nounced in the oblique cases as ‘attu’ by the bulk of the northern ) Tamilians. The majority of the natives of Madras, for instance, use 1l “atlei’ (‘attu-ei’) as the accusative of “adu,’ that, instead of ‘adei; s and in the neuter singular pronominal suffixes to the verb the same he pronunciation is not only commonly heard, but is often written: he e.g., instead of ‘irukkiradukku,” to its being (the dative of ‘iru- kkir-adu,” it is, the being, or that which is), Madras Tamilians invari- 10 ably write ¢irukkirattukku; in which compound ¢attu’ is evidently Cl It is also used as the neuter demonstrative singular instead of ‘adu.’ &‘{ deserving of notice, that the feminine singular suffix of a large class of appellative nouns, which is ¢di o or ‘adi’ in Telugu, and which has nal been shown to be identical with the neuter demonstrative, is in Tamil (3% “tti’ or ‘atti, e.g., ‘TAsAtti, @ woman of the Rdjd caste, a queen. Even ) in the nominative ¢atti’ is sometimes used in Telugu instead of ‘adi,’ the that, and “itti’ instead of ¢idi,” ¢his. 48 Two instances will suffice to prove the identity of the Tamil ¢attu’ il and the Canarese ‘ad,” and thus supply the only link that is wanting of to the perfect identification of ‘attu’ with the Telugu “ti,’ and of both with ‘adu. il The Tamil ¢ plrv-att-il,) in ancient times, is com- lif- pounded of ¢ plrv-am’ (Sans. deriv.), antiquity, ‘att-u,’ the inflexional il increment, and ‘il,” the sign of the locative. Compare this with the b v_u, corresponding Canarese ¢ plrv-ad-alli, in which it is evident that for “ad”’ is used in the same manner as ‘att-u,” and perfectly agrees with it in signification. Again, the Tamil ‘dyirattondru,” a thousand and of one, is formed from ¢ayiram, a thousand (the inflexion of which is il ; “ayir-attu’), and ‘ondru,’ one. When this is compared with the b corresponding Canarese word saviradondu,” from ¢ savir,” @ thousand 18/] (equivalent to the Tamil ¢Ayir’)— inflexional form sivir-ad’—to 0y 1 | m which ‘ondu,” identical with ¢ondru,’” is appended, it is evident that il the Canarese increment ‘ad’’ and the Tamil ‘att’’ are one and the % g same ; and also that in this instance the Canarese ‘ad’’ is used for 700 precisely the same purpose as the Tamil ‘att’,” viz., as an inflexional increment with an adjectival signification. 214 THE NOUN. (5.) The formation of the inflexion by means of doubling and hardening the final consonant. — Tamil nouns ending in ¢d-u’ and ¢r-u’ form the basis of their oblique cases by doubling the final d’ and ‘R; and the doubled ¢d’ becomes by rule tt,” and the doubled ‘R, ‘ttr’ (though spelled ‘RR’): eg., from ‘kad-u’ @ jungle, is f formed ¢katt-(u)-kku,’ fo a jungle; from ¢ar-u’ a river, ‘attr-il’ (‘4RRr-il’), in a river. I This doubling of the final consonant of such nouns places them in an adjectival relation to the succeeding noun. It is to be regarded as a sign of transition, for when intransitive or neuter verbs ending in ‘d-u’ or ‘r-u’ double their finals, they acquire a transitive sig- nification ; e.g., from ‘6d-u,’ to run, is formed ‘ott-u,’ fo drive; from “tér-u, to become clear, comes ‘téttr-u’ (‘térr-u’), to clarify, to 5 comfort. Properly speaking, therefore, this doubling of the final is an adjectival formative, rather than an inflexion or case-sign basis: but t in this, as in many other cases, the same form is used in both con- | nections, in consequence of the case-sign which is appended to the doubled final having originally been a mnoun, and still retaining in d compounds the force of a noun. In Telugu the final consonant of nouns of this class is hardened, fi but not doubled, to form the inflexion or basis of the oblique cases; ) e.g., the inflexion of ¢ ér-u,’ @ river, is not ¢ éttri’ (‘érri’), but ‘éti,’ of a rwer; and that of ‘nddu, a country, is ‘nati] of a country. In some instances the Telugu corresponds more closely to the Tamil in forming the inflexion of nouns in ‘RR’ by changing that into ‘®t:’ e.g., ‘arr-u,’ the neck; inflexion of the same ‘arti’ If we regarded the Telugu alone, we should consider these examples, not as instances 1 of the doubling of a final ‘d’ or ‘®,” but rather as instances of the f incorporation of ‘ti,” the usual inflexional suffix with those finals ; and we should suppose this view to be confirmed by the circumstance that i the Telugu does not, like the Tamil, double the final ‘d-u’ or ‘r-u’ i of intransitive verbs on converting them into transitives, but adds a 1 formative ‘chu.’ Nevertheless, the Tamil rule is so clear and ex- press and so evidently founded upon grammatical reasons, and the Telugu words in question, ‘nati, &c., so exactly agree with the Tamil, that we cannot but recognise in them the operation of the same principle, though somewhat disguised. In other and parallel instances, though the Telugu hardens, it does not double: e.g., from ¢pad-u,’ Tam. and Tel., to sing, the Tamil forms patt-u, a song, the Telugu ¢ pat-a.’ The final i’ of such Telugu inflexions as ‘néti,’ of a country (from ¢nad-u’), instead of ‘nat-u,” which the Tamil would lead us to expect, is owing, I have no doubt, to the influence of ti, which is b the ordinary suffix of the inflexion of neuter nouns. { E% INFLEXIONAL INCREMENTS. 215 i 7. The inflexional increment ‘i’—The inflexion of the plural of 1 the Telugu epicene demonstrative pronoun consists in ‘i’ e.g., * varu’ d} (from ‘avaru’), those persons; inflexion °vAari, of them, their. ? 2 o The final ‘u’ of ¢var-u of ¢vari’ is merely euphonic, but the ‘i 1§ is certainly an inflexivnal increment; and possibly the final ‘i’ of I’ the singular masculine demonstrative inflexional ‘vAni’ is not to be regarded as a portion of ‘ni,” the ordinary inflexional increment of 1 Telugu masculine nouns, but is identical with the final ‘i’ of ¢véri.’ a8 A small class of Telagu nouns form their singular inflexion also in ng $10 eg., “kal-i, of a foot, ‘iér-i, of @ car What is the origin j0- of this “i? I think we are guided to a true idea of its origin by m comparing it with the possessive pronoun ¢véridi,” Tel., that which s f] theirs, which in Ku also is ¢évaridi. When ¢ varidi’ is compared an with the Tamil possessive ‘avaradu,’ the meaning of which is exactly )b the same, we see that in each language the termination is that ot the it neuter demonstrative pronoun, which is ‘adu’ in Tamil, ‘adi’ in the Telugu; and we also see that the penultimate ‘i’ of ‘varidi’ is mn derived by attraction, according to Telugu usage, from the succeeding I ‘1, which is that of the neuter demonstrative singular ¢adi.’ The final €1’ of ‘vari’ may therefore be regarded as an abbreviation ed, of “adi, or at least as derived from it. £ 70/‘ In (8.) ZTelugu plural inflexional increment in ‘a’—In Telugu ‘a’ 1 constitutes the plural inflexion of most colloquial pronominals, and p of all substantive nouns without exception. ¢l-u,” properly ‘I, is ed the pluralising particle of all neuter nouns in Telugu, and of the jceS majority of ‘rational’ ones: the inflexion is eflected by changing the this ‘I-u’ into ‘la,’” or to speak more correctly, by suffixing ‘a’ to nd ¢1°’—the final vowel of ‘lu’ being merely euphonic; and it is to this hat incremental ‘a,” as to ‘ni’ and ti, the singular inflexions, that all Bl the case-signs are appended, e.g., ¢ kattulu,’ knives; inflexion ¢ kat{rtu]a;' instrumental ‘kattula-chéta,” by knives. lsa I have no doubt that this inflexional increment ¢a 1s identical ex- with ‘a,’ one of the Tamil-Canarese signs of the genitive, and of the the use of which as a genitive, in the singular as well as in the plural, we the have an illustration even in Telugu, in the reflexive pronouns ‘tan-a,’ e of self, ‘tam-a,” of selves. This increment also, therefore, is to be 66 vy regarded as a genitive in origin, though in actual use merely an i, inflexion; and I have no doubt that each of the Dravidian inflexions ugl proceeds from some genitive case-sign. it g0 | Before leaving this subject, I should briefly refer to one which 18 bears some relation to it, viz.:— :; 216 THE NOUN. L Luphonic links of connection between the base and the inﬂexion, 00 the base and the case-signs, or the inflexion and the case-signs. ch In Tamil the dative case-sign ‘ku’ is generally preceded by an a euphonic ¢ u,” and through the influence of this ‘u’ the ‘k’ is doubled. Thus, from ‘avan,” he, is formed not ‘avanku,’ ¢o &im, but ‘avanukku’ th (‘avan-u-kku’). The personal pronouns, both in the singular and in 2 in this connection, instead hg the plural, make use of an euphonic ‘a il of ‘u; eg., from ‘nan’ (or rather from an older ‘én’), 7, is formed the inflexion ‘en; and this takes as its dative not ‘enku’ or 1 la ¢enukku,’ ¢o me, but ‘enakku’ (en-a-kku). tl In MalayAlam the personal pronouns require the insertion of an : euphonic ‘i’ between the inflexion and the case-sign, e.g., ‘inikka,’ to me, ‘nanikka,’ to thee. In some instances in Tamil the euphonic 1S vowel which is made use of in this connection is not ‘n’ or ‘a,” but L “lel.” 1t Thus “nil, a day, forms its dative not in ‘nalku,’ ‘nalukku, [ or ‘mnalakku,” but in *¢naleikku’’ In the higher dialect of the Tamil I the dative case-sign ‘ku’ is often directly attached to the noun, (é especially in those instances in which the noun terminates in a liquid th or semi-vowel ; e.g., we find in that dialect not ‘avarukku’ (‘avar- In ancient Canarese also, the dative di u-kku’), to them, but ‘avarku.’ 1n case-sign was invariably attached in this manner. a Whenever concurrent vowels meet in Tamil ‘v’ and ‘y’ are il used, as has already been shown, to prevent hiatus; and accordingly they are used between the final vowel of nouns and those inflexions or case-signs which begin with vowels; e.g., ‘naduvil’ (‘nadu-(v)-il’), Compare this § (’ wn the middle; ‘variyil’ (‘vari-(y)-il’), in the way. (6 with the use of ‘v’ for a similar purpose in Magyar: e.g., from €16, al a horse, and ¢ at,” the sign of the objective case, is formed not ‘164t ‘v’and ‘y’ are but “1é6vat,” precisely as would be done in Tamil. I used by the Canarese in the same manner as by the Tamil; but in d Telugu, as has already been shown, ‘n’ is used as a preventive of fl hiatus instead of ¢v.’ 0 The way has now been prepared for the investigation of the A Dravidian oblique cases, and of the signs of case properly so called. U The accusative or “second’ case.—In the Indo-European languages t the case-sign of the accusative of neuter nouns is identical with that of the nominative case. This identity has arisen, T conceive, not from the nominative being used as an accusative, but vice versé from I the accusative being used as a nominative. The accusative case-suffix B is a sign of passivity, or of being acted upon; and it was suffixed to masculine and feminine nouns to denote that in that instance they g THE ACCUSATIVE. 21 ( lon, were to be regarded not as agents, but as objects. Subsequently, I conceive, it was adopted, because of this signification, as a general characteristic of the neuter, objective, or dead class of nouns, and so ya ed, came to be used as the nominative, or normal case-ending of nouns of J\H’ that class. 1 n In the Dréividian languages also an accusative case-sign seems to ead have been adopted as a formative termination of abstract neuter med nouns. The Old Canarese accusative case-sign ‘am’ seems to be 0r identical with, and is probably the origin of, the ‘am’ which is so largely used by Dravidian neuters. Notwithstanding this, the use of fan the nominative, or rather of the simple, unformed base, as the accu- sative of neuter nouns, is the ordinary and almost universal colloquial ko, onie usage of the Dravidians, and is often found even in their classical com- but positions. The accusative case-termination may be suffixed whenever ko, it appears to be desirable to do so, either for the sake of euphony or amil to prevent ambiguity; but it is rarely employed except when it is required for those purposes. When this case-termination is used oull, without ilecessity, it sounds stiff and unidiomatic; and this is one of qlli& the peculiarities by which the Tamil of foreigners is marked. Dravi- far dian masculine and feminine nouns and their corresponding pronouns tive invariably take the accusative case-suffix when they are governed by active verbs. This probably proceeds from the principle that it is are more natural for rational beings to act than to be acted upon ; and u«lv hence when they do happen to be acted upon—when the nouns by jos which they are denoted are to be taken objectively—it becomes neces- 1) sary, in order to avoid misapprehension, to suffix to them the objective this case-sign. On the other hand, the difference between the nominative o, and the accusative of neuter nouns is often allowed to pass unnoticed, bty because such nouns, whether they act or are acted upon, are alike ae destitute of personality and inert. Whether the accusative is used as b 10 the nominative, as in the Indo-European languages, or whether, as is 0 Of often the case in the Scythian tongues, the nominative is used for the accusative, the principle involved appears to be one and the same. the The use of the nominative of neuter nouns for the accusative is not ! unknown to the North-Indian vernaculars, and is one of those par- ticulars in which those vernaculars appear to beallied to the Dravidian 1088 family. Ordinarily, however, the North-Indian vernaculars are dis- that tinguished from the southern by their use of the dative case-sign for not the accusative. In the Dravidian family, with the solitary exception rom of the Gond, the dative case-sign is always quite distinct from the accusative; whereas in the Gauda or North-Indian family, there is | f generally little or no difference between those two cases. In most hey 218 THE NOUN. instances, the case-sign which is allied to the Drévidian dative suffix, sofs and which appears to be essentially a dative, is that which is used fern for both cases indiscriminately; and it is the connexion which " determines whether the dative or the accusative is to be understood. ‘e mus 9 EENTD € (1.) Adccusative case-signs © ei, and ‘a.’—The only sign of the inve accusative which the Tamil recognizes is ‘ei,” which is suffixed to both COVE numbers and to all genders; though, as has been mentioned, the accusative of neuter nouns is often identical with the nominative or the base. Examples, ‘avan-ei,’ him, ‘aval-ei,’ her, < ad-ei, it. The accusa- tive case-sign of the MalayAlam is ¢¢; and this is evidently a primi- n ¢ tive form of the Tamil ¢ei.’ The Canarese ordinarily uses either ‘a’ 1o! her or ‘annu,’ as its accusative case-sign; but in some instances (e. g., ‘nanna,’ me, ‘ninna,’ thee,) it appears to make use of ‘na,’ instead of all 4 aQ This ‘a’ seems to be equivalent to the Malayala €&’ and the 1nfy Tar Tamil ¢ei,” into which the Canarese short <a’ is often found to change by rule. 400! ’\)7 The Tamil-Malayila accusative case-sign ¢ e or ‘ei, may be com- Case pared with ‘hé’ or €&’ the dative-accusative of Hindi pronouns; with 2 al the Gujarathi dative-accusative singular ¢é J and with the preponde- addlj rance of the vowel ¢ &’ which is observed in the dative-accusatives of 1on) the Bengali and Sindhi. Compare also the Brahui dative-accusative feen ‘ne’ or ‘e, and the Malay ‘e.’ fror On pushing the comparison amongst the Scythian tongues, not a few of their accusative case-signs are found to resemble the Tamil of accusative. Thus the Wotiak accusative is formed by adding ‘4’ to the liag root, e.g., ‘ton,” thow, ‘ton-a, thee. The Turkish accusative is ‘1’ or fore ‘yi; the Mongolian ‘i’ after a consonant: ¢dji, instead of the Turkish app ¢ yi,” after a vowel. el e The Turkish ¢ 1 is doubtless a softened form of the Oriental to g Turkish accusative case-sign ¢ ni,” from which it has been derived, by the same process by which the Turkish dative case-sign ‘eh’ or ‘yeh,’ arg is undoubtedly derived from the old Oriental Turkish ¢ g4’ or ¢ ghah.’ an( It would therefore appear that the Scythian accusative origina,lly the contained a nasal; and in accordance with this supposition we find in an( the Calmuck pronouns an accusative case-sign corresponding to the fing Oriental Turkish ‘ni’ e.g., ¢bida-ni, ws, from ‘bida,’ we, and also ace ‘ na-mai,’ me, and ¢ dzi-mai,” ¢ee, from the bases ‘na’ and ¢dzi’ With this we may again compare the Brahui dative-accusative ‘ne’ or ‘e’ 10y That the Oriental Turkish ‘ni’ could easily and naturally be softened el ($45%7 1 into ‘yi’ or appears from the Drividian languages themselves; in which, for instance, the personal pronoun ‘nin,” Tam., 7, has been (‘Xp Illay THE ACCUSATIVE. 219 uffix, softened first into ¢ yAn,” and then, in the oblique cases and the verbal used terminations, to ¢ én’ and ¢en.’ ‘ni’ being evidently the basis of the which Turkish and Mongolian sign of the accusative, if the Dravidian ‘ei’ or 0d, €&’ is allied to it, as we have supposed to be probable, this ei’ or ‘&’ must originally have been preceded or followed by a nasal; and in of the investigating the other Dravidian accusative case-signs we shall dis- ) both cover some reasons for concluding this to have been actually the case. , the Ve o (2.) Accusative case-signs ‘am,” ‘annu,’ ‘anna,’ ‘nu,’ &e.— am’ is (CUSa- the characteristic sign of the ancient Canarese accusative, and is used - in connexion with nouns and pronouns alike, e.g., ‘aval-am,” ker. The ) more modern form of the Canarese accusativeis ‘annu,’ e.g., ‘aval-annu,’ 2 (e.9 her; and this ‘annu’ is certainly identical with the older ‘am ad of ‘am’ has in other instances besides this evinced a tendency to change the into ‘an;’ for Ae is ‘avam’ in ancient Canarese, though ‘avan’ in Tamil. ange The change of the old Indo-European ‘m, the sign of the accusative in Latin and Sanserit, into the Greek ‘v’ is also a parallel com case. The ancient Canarese case-sign ‘am’ no sooner changed into with ‘an,’ than it would irresistibly be impelled to euphonise ‘an’ by the nde- addition of ‘nu.’ Even in Tamil vin,’ the sky, is commonly pro- s of nounced ‘vinnu,’” and in Telugu it is vinnu’ by rule. Hence we seem to be quite safe in deriving ‘annu’ directly from ‘an,” and ‘an’ ative from ¢am. Another form of the Canarese accusative case-sign is ‘anna,’ instead 0t & of “annu,” or simply ‘nna’ or *na, eg., ‘na-nna,’ me. The final ¢ u,’ qmil has in this instance been changed into ‘a,” through the attractive o the ]"Or force of the primitive ‘an:’ or rather, perhaps, the entire euphonic kish appendage ‘nu,’ has been rejected, and the original case-sign ‘an’ been softened to ¢a,” whilst the final ‘n’ of the base has been doubled ontal to augment or express the objectivity of the signification. 1, b Y The Tulu accusatives ‘nu’ and ‘n’ (e.g., ‘yanu’ or ¢ yannu,’ me) are evidently identical with the Canarese, and also with the Telugu; yeh) pih, and they are peculiarly valuable as tending to show the connection of the Telugu accusative suffix ‘nu’ or ‘ni,” with the older Canarese ‘an’ ]ally and the still older ‘am.’ The Tuda accusative of the pronoun of the d1mn first person singular ends in ‘ ama,’ e.g., ‘en-ama,’ me : when the Gond the accusative differs from the dative it is denoted by ¢un.’ also In Telugu the neuter accusative is ordinarily the same as the V ith nominative, as in the other Dravidian dialects; but when the noun ‘ev belongs to the class of ‘rationals’ or ‘majors,” the accusative must be )I]Ed expressed by the addition of a sign of case. The accusative case-sign 1 may optionally be suffixed, as in Tamil, to neuter nouns; but whether o0 i 220 THE NOUN. the noun be a ‘major’ or a ‘ minor,” singular or plural, the sign of case questi must be suffixed to the inflexion, genitive, or obliqu’e-ca,se‘basis, not to the the nominative. When the inflexion is the same as the nominative, ‘nm the noun to which the case-sign is attached is still regarded as the “nin- inflexion, so that in theory the rule admits of no exceptions. I6 el ¥ The sign of the accusative in Telugu is ‘nu’ or ‘ni:’ when pre- acous ceded by €i’ it is “ni] e.g., “inti-ni,’ ¢ dom-wm ;> where-it is preceded In o by any other vowel it is ‘nu,’ e.g., ¢ bidda-nu,” ¢ puer-um.’ these A similar “ni’ or ‘na’ is used in Telugu (but not so systematically (ans as the corresponding ‘in’ in Tamil) as an euphonic inflexional incre- ment; and ‘na or ¢ ni 9 ! is also a sign of the locative in Telugu. havi Probably those locative and genitive suffixes were originally, and are still to be regarded, as one and the same; but the sign of the accusa- the tive, though nearly identical in sound, proceeds apparently from a i8h different source. Comparing it with the Canarese and especially with orig the Tulu accusative, we can scarcely avoid the conclasion that, though in sound it is identical with the ordinary inflexional augment, it is to tive be regarded as a relic of the Canarese accusative case-sign ‘annu’ or fanr ‘am. The suffixes of the accusative of the Telugu personal pronouns “nimy can be explained on this supposition alone. The “inflexions’ of those dout pronouns are essentially different from their accusatives,and incapable proc of being confounded with them ; and the accusatives of those pronouns Proc to take of necessity, and not merely for euphony, the nasal suffixes < nu’ or ‘nnu’ in the singular, and ‘mu’ or ‘mmu’ in the plural. Thus, il whilst n&,” of me, is the inflexion of ¢ nénu, 7, its accusative is ‘nanu’ 10f or ‘nannu, me; the accusative of the second person is ‘ninu’ or Gar ‘ninnu,’ thee, and their plurals are ‘mamu’ or * mammu,” us, ‘mimu° 8 d or ‘mimmu,” you, whilst the inflexions of those plurals are ‘mé’ and h 3 “imis Tar When these accusatives are compared with the Canarese and Tulu, especially with ¢ yanu,’ me, and ‘ninu,’ thee, in the latter, their vir- tual identity, and therefore the origin of them all from the ancient dif Canarese ‘am,’ can scarcely be doubted. Tel We may now proceed to compare this acensative case-sign ‘am,’ loc ‘an,’ ‘annu,’ ‘nu,’ or ‘na’ with the Gujarathi dative-accusative ‘ne,’ ;S (tr with the Panjabi ‘nu’ or ¢ num,” and also with the Brahui ‘né’ or ‘8 fle and the Turkish and Mongolian “ni’ or ¢1. In the Finnish tongues the greater number of singular accusatives are formed by suffixing “en,’ gi “an, &e., which are also used as signs of the genitive: in the plural th there is rarely any difference between the nominative and the accusa- 1] tive. Ascending further towards the source of the Scythian tongues, Te we find in the language of the Scythian tablets at Behistun an un- 0 THE INSTRUMENTAL. 221 of ease questionable link of connexion with the Dravidian. The pronoun of 10t to the second person singular in that language is ‘ni,’ thow, of which ative, ‘nin’ is the accusative; and when this is compared with the Tulu s the ‘nin-u,’ thee, we cannot fail to be struck with the closeness of the resemblance. We should also notice the extensive use of ‘m’ or ‘n’ as an | pre- ceded accusative case-sign in the languages of the Indo-European family. In Sanscrit, Latin, and Gothic, ‘m’ predominates, in Greek, ‘n; but 2 of the ancient ieally these consonants are virtually identical, like the ‘m ncre- Canarese, and the ‘n’ of the modern. clug. A similar form of the accusative being extensively prevalent, as we d are have seen, in the Scythian tongues, it would be unreasonable to derive In this instance it CCUSa~ the Dravidian case-sign from the Indo-European. is better to conclude that both families have retained a relic of their om 4 with original oneness. 0ugh It only remains to inquire whether the Tamil-Malayalam accusa- 150 tive case-sign ‘ei’ or ‘e’ cannot be connected with the Canarese ‘am,’ 1 or ¢annu,” and ‘na.’ On comparing the ancient Canarese accusative 0uns ‘ninnam, thee, with the more modern °ninna,” it can scarcely be hose doubted that the latter is derived from the former by the ordinary pable process of the softening away of the final nasal. Through this very 0uns process the final ‘am’ of many substantive nouns has been softened ‘mn’ to ‘a, eg., ‘maram, Ancient Can., a tree, ‘mara’ or ‘mara-vu, modern Can. If then the sign of the accusative in ‘ninna,’ thee, is [hus, not ‘na’ but ‘a’ (instead of ‘am’), as is probably the case, there ant’ l or cannot be any difficulty in deriving from it the Tamil accusative case- sign ¢ei’ for the change of ‘a’ into ¢ ei,” takes place so frequently jmy’ and that it may almost be considered as a dialectic one, e.g., compare Old Tamil ‘¢ila,” not, with the modern Tamil ¢illei.’ ulu, The instrumental, or “third’ case.—Different particles are used by VI different Dravidian dialects as suffixes of the instrumental case. In jent Telugu the most classical instrumental is identical with the inflexional locative, and consists in changing ‘ti’ or °ti, the ¢inflexion,’ into ally ‘ta’ or ‘ta; eg., ‘ri-ta,” with a stone, from ‘ra-yi’ @ stone, the in- 16, flexion of which 1s ‘ré-ti. ) ‘67 This form of the instrumental was probably a locative in its ori- ;UQS. ginal signification, and at all events it is identical with an old form of el the locative ; e.g., ‘inta,” @n @ house, from ‘illu,’ @ house, of which the il inflexion is ‘inti. The more commonly used instrumental of the 184 1 Telugu is formed by the addition to the inflexion of any noun of ‘ché’ 16% or ‘chéta,’” which is itself the instrumental form of ¢ ché-yi’ (Tam., i 222 THE NOUN. in th ‘kei’), the hand, signifying by the hand (of); eg., ‘nibbu-ché’ta,’ by Jire, literally by the hand of fire. (and | se-§ The inflexion, or genitive, without the addition of any special suffix, the ! is also occasionally used in Telugu, as in High Tamil, to denote the instrumental case, as well as the ablative of motion and the locative. the ) with | The particle ‘na’ is also sometimes suffixed to neuter nouns to denote all three ablatives. and w A In Ancient Canarese the instrumental case-sign is ‘im; in the (ase, modern dialect ¢inda, an euphonised, adjectival form of the same suffix. The suffix of the Tulu is ‘d’da.’ Verne ‘to 1 I consider ‘im,” the old Canarese instrumental suffix, to be iden- tical in origin with ‘in,’ the suffix of the Tamil ¢ablative of motion,’ £ or ‘locative. It has already been seen how easily ‘m’ changes into knif 1 and both in Canarese and in Tamil there is so close a connexion hawi between the ablative of motion and the instrumental, that the case- I sign of the former is very often used for the latter, especially by the Soyth poets; e.g., ¢ val-in aya vadu,’ Tam., @ wound inflicted by’ a sword, not after ‘from’ a sword. In Canarese also the ablative of motion is denoted mere: the T¢ more frequently by the suffix of the instrumental than by its own suffix, Through a similar tendency to confound these cases, the case- the o] Word sign of the instrumental has disappeared from the Latin, Greek, &e., and the sign of the ablative has come to be used instead. Even in the v English, ‘by,’ originally a locative (e.g., ‘close by’), has lost this peal 1 meaning altogether, and is used at present to form the ablative, or more properly the instrumental. Dré‘ In Tamil and Malayélam the suffix of the instrumental is ¢4l ; in High Tamil ‘4n’ also. “al’ is the case-sign of the ablative or instru- loup mental in Gond, though in Telugu, which is spoken between the geniy dian Tamil country and the country of the Gonds, a different case-sign is used. This suffix ‘al’ is possibly derived from, or allied to, ¢kl B s i e o g, Tam., a channel, a noun which is contained not only in the Dravidian S&nsm dialects, but also in Bengali. tllyg In some dialects ‘channel’ is a compound word (Tam., ‘kalvéay ; been ingty Te., ‘kalava; Can., ‘kélive’), and the only meaning of ‘kal’ is « Soot. This meaning is contained in the Tamil, but that of a ¢channel,’ ot which the Tamil contains also, suits better the supposed use which is made of ¢kél,’ as a sign of the instrumental case. ‘kal’ may have in’f lost its initial ‘k’ in the same manner as ‘kal,’ the neuter sign of st oy plurality, is known to have done in Telugu and Tulu, in which it has become ‘l-u,’ by corruption from ¢kal-u’ or ¢gal-u.’ €ney In the Indo-European family of languages there are no signs of the instrumental case which at all resemble those that we have noticed by THE INSTRUMENTAL. 223 fa, by in the Drividian family. The only analogies which I have noticed (and probably they are illusory) are those which exist between the case-sign of the Tamil-Malayélam and the corresponding case-signs of suffx, fe the the Finnish tongues. Compare al’ with the instrumental suffix of tive, the Magyar, which is ‘al’ in the singular, ‘el’ in the plural ; and 108 to with ¢alla,” ‘ella,’ &ec., the instrumental suffixes of the Finnish proper, and which are euphonically augmented forms of ¢al’ and ‘el n the A secondary or periphrastic mode of forming the instrumental ) Sane case, which obtains in the Dravidian languages, as also in the northern vernaculars, is by means of the preterite verbal participle of the verb o 1den- ‘to take, and the accusative or abstract nominative of any noun ; otion, e.g., ‘kattiyei (k)kondu,” Tam , with a knife, literally kaving taken a 63 10£0 knife : compare the corresponding Bengali ¢churi diya,” with (i.e., nexion having taken) a knife. This has arisen from the repugnance of the Drividian (as of all the N Scythian) languages to continue to make use of any inflexional form by the after it has ceased to express its original meaning, and has become a yd, 00t mere technical sign. When that has taken place, as in the instance of enoted S the Tamil ‘kal’ or ¢4l those languages are often found to abandon the old form, or let it fall gradually into disuse, and to adopt some p (8e- word or phrase instead which has a distinct meaning of its own, and k, &ty the use of which recommends itself at once to the intelligence of the yen i1 speaker. ¢ this Ei\'e7 o Under this head it is desirable to enquire into the force of the Dréavidian coNJUNCTIVE cASE, aud the svffixes by which it is denoted. Dravidian grammarians have arranged the case system of their i/ nouns in the Sanscrit order, and in doing so have done violence to the stro- genius of their own grammar. It is very doubtful whether the Dravi- i the dian ‘ablative of motion’ and the ¢locative’ are not one and the same ;& ; ign 18 case, though represented as different by grammarians, in deference to ‘el % Sanscrit precedents ; and the Dravidian ‘social ablative,” as some have 1vidiﬂaﬂ | called it, or rather, as it should be termed, the conjunctive case,” has f been omitted in each dialect from the list of cases, or added on to the lvif ) instrumental, simply because it is a case of which the Sanserit knows 55 0 nothing. |allﬂel; The only reason why the case-signs of the conjunctive are classed jich 18 § in Tamil with that of the instrumental is that the fact of their being b y hf{\'e destitute of a proper place of their own is less obvious in that posi- sig o tion than it would be in any other. Notwithstanding this, the differ- it s ence hetween those two cases is considerable. The ‘instrumental is best rendered in English by the preposition ‘by,” ‘by means of ;' the force of the conjunective is that of the prepo- uaticed s 224 THE NOUN. sition'¢with,” in the sense of the Latin ‘cum,’ or together with. Some- the ab times the English preposition ¢with’ is used in either sense; eg., ‘I (@i cut it with a knife’—‘1 went with him: but in the Dravidian lan- tonshiy guages the former ‘with’ would be represented by the sign of the in- (. cong strumental case, the latter by that of the conjunctive; e.g., ‘katti-(y)-al,’ of con Tam., by a knife, ‘avan-6du,’ with him. immed Though the Sanscrit and the Indo-European languages generally and ul are destitute of this case, the Latin evinces a tendency towards it in that a such forms as ‘nobiscum ; whilst most of the Scythian tongues have a regularly formed conjunctive case equally with the Dravidian and | “den,” the conjunctive case-sign of the Calmuck, may even be com- and | pared (though probably the resemblance is accidental) with the Tamil the ¢ conjunctive case-sign, ‘udan.’ only v The Tamil conjunctive case-signs are ‘udan,” ¢ odu,’ and ‘6du ; of and. which the two last have now no meaning of their own, and the first is Dy occasionally used as a noun signifying °conjunction’ or ‘continuity.’ to de It is also capable of being combined with another word as an adjective, 000 e.g., ‘udan &l a fellow servant; and with the addition of the emphatic dialee “¢’ (‘udan-¢’), it is used also as an adverb to signify vmmediately. family The final ‘an’ of ‘udan’ (Tel., ¢ todan-u’), is probably a forma- ane §) tive addition to the root-syllable, for ‘udam’ is another and equiva- T lent Tamil form ; and the first syllable can scarcely be doubted to Kk be allied to ‘odu,” the other sign of the same case in High Tamil. the ‘u’ is always pronounced as ‘o’ before ‘d,” and other cerebrals, 0l whenever the word in which it appears has a second syllable. Hence Ed and ‘udan’ is invariably pronounced ¢odan 5 and in the Canarese post- OM position ‘odane’ (Tel., ¢tdodane’), this pronunciation is written as the o well as heard. ¢0du’ (emphatically “6d-¢"), the third suffix of this haye case in Tamil, and the most common in the colloguial dialect (in 0f] Malayalam “6ta’), is evidently allied to ‘odu, and through it to In ‘udan.’ As neither ‘odu’ nor ‘6du’ has any meaning of its own in Tamil, it is evident that they have undergone some alteration, and it any g tlll‘nir is desirable to trace their connexions in the other dialects. step, On turning to the Telugu, we find that its conjunctive case-signs, 1 which are evidently allied to those of the Tamil, have the consonant With ‘t’ prefixed to each of them, eg., ‘téda’ and °t6. Supposing ‘toda,” Tel., with, to be identical with the Malayalam ¢ o6ta’ and the mly Tamil ¢6du” (and its identity is put beyond a doubt by comparing the coll Tamil adjective ‘ udan’ with the Tel. ‘tédan-u,” and the Tam. adverb in | ¢udané’ with the Tel. ‘todané’), the conjunctive suffixes of the Tamil- Dr Malayalam, which were destitute of meaning by themselves; are now found to acquire a very appropriate meaning ; for the Tamil ¢ téra’ (in Bhy THE DATIVE 22 J Some the abstract ¢téra-mei which is phonetically equivalent to ‘toda eg, ‘1 (‘d’ in Telugu corresponding by rule to ‘r’ in Tamil), means compan- an lan tonship ctodu’ itself also is found, with the related signification of the a congeries, a collection ; and though “udan’ has by itself the meaning ()l of conjunction, or continuity, yet when t’ is prefixed to it, we can immediately detect its relationship to ‘tudar’ or todar,’ fo follow, erally and ultimately to ‘todu’ and ‘tora Thus it appears highly probable 81t n that all these words and forms are virtvally identical have - and The dative or “fourth case —In the North-Indian dialects one )¢ GOl and the same post-position or suffix is used as a sign of case both by Tamil the dative and by the accusative In the Dravidian languages, not only is the difference between the dative and the accusative essential dn! 0 and strongly marked, but there is less discrepancy amongst the various 1018 Dravidian dialects with respect to the particular suffix which is used ity to denote the dative, than with respect to any other case-sign The Jective, accusatives, instrumentals, ablatives, and genitives, of the various up} atic dialects, exhibit material diflerences; but in all the dialects of this ly family—in the rudest as well as in the most polished—there is but formay one suffix of the dative. The dative is formed in Tamil by suffixing ‘ku’ (in construction equl\ il bted ¢kku’); in Malayalam ‘kka’; in Telugu ‘l\u or ‘Kki, according to n nil the nature of the preceding vowel,—properly and naturally ¢ku ehal 0Old Canarese ge or ‘ke;’ in the modern dialect ‘ge (0] ¢ kke Hence and in construction 1ge From a comparison of these forms it is obvious that the guttural ‘k’ or (followed by a vowel) constitutes P()st tten 48 the only essential part of this suffix ; and that, as the vowel seems to of his have been added chiefly for the purpose of helping the enunciation, it Y1 (m R is of little moment what vowel in particular is used for this purpose. b 1 to In the primitive Indo-European tongues we discover no trace of any such dative suffix or case-sign as the Dravidian ‘ku;’ but on own I and 1 turning to the Scythian family, interesting analogies meet us at every step .18, In the vernaculars of Northern India, which are deeply tinged sonaat with Scythian characteristics, we find a suffix which appears to be not only similar to the Dravidian, but the same 3}]0:11]”' ﬂd hg The dative-accusative in the Hindi and Hindustani is ¢ko,” or 5 colloquially ‘ku; 3 in the language of Orissa ‘ku; in Bengali ké g hd in Sindhi ¢ kheé in Singhalese ghai in the Uraon, a Semi- 3(1‘ ¢l b Dravidian Kole dialect, Tami ‘ gai; in the language of the Bodos, a Bhutén hill-tribe, ¢kh6;’ in Tibetan ‘gya.’ e 007 1 (Iﬂ The evident existence of a connexion between these suffixes and Q 226 THE NOUN. Of all the itsel the Dravidian dative case-sign ku’ is very remarkable. analogies between the North-Indian dialects and the Southern, this is 0 the clearest and most important; and it cannot but be regarded as betokening either an original connexion between the Northern and have Southern races, prior to the Brahmanic irruption, or the origination of Jeart both races from one and the same primitive Scythian stock. with The dative-accusatives of the North-Indian vernaculars have byt are | commonly been supposed to be accusatives in their original significa- tion, and datives in a secondary application alone. This is the opinion n TES of Dr. Max Miiller, who attempts to derive ‘ké,’ the Bengali dative- accusative, from the Sanscrit adjectival formative ¢ka.’ I need not the here criticize the Professor’s arguments ; for the extensive use of this daf particle, or its equivalent, as a distinctively and exclusively dative ole suffix in the Dravidian languages, and also, as will be shown, in the Scythian tongues, appears to me to prove, beyond reasonable doubt, 1 that it was a dative, not an objective suffix, originally ; and that its % Jef origin was far earlier and more remote than the late Sanscrit genealogy which is attributed to it by the Professor. Dr. Kay is, I believe, (u right, in holding that the dative has a better claim than the Teg] accusative to the use of ‘ké’ even in the Northern vernaculars, and in directing attention to the parallel use of the Semitic pre- | Scy position ¢la.’ the The suffix of the dative in the various languages of the Turkish the family perfectly corresponds to the Dravidian dative and to the North- Indian dative-accusative. The forms of this suffix which are found in 4 the Oriental Turkish are ‘ke,” ‘ka,” ‘ge,’ ‘ga,” ¢ ghah,’ and also ‘a S The Osmanli Turkish dative is ‘eh’ or ¢ yeh,” the initial ‘k’ or ‘g’ ‘r of the older dialect having been softened into ‘y,” and then discarded. ha The Manchu ‘de’ and the Mongolian “dou’ are possibly allied to (01 % the Tartar ‘ke;’ for it has already been remarked that the change of ~] ¢k’ into ‘t’ or ‘d, or wvice wersd, is not an uncommon one in this ¢ group of tongues, and that even amongst sister dialects belonging to — N the same family or sub-genus, the pluralising particle in one dialect is e “ek, and in another ‘et.’ Perhaps, therefore, we may venture to copnect with ‘ke’ not only the Mongolian ‘de,” but even the Uriya ‘te’ and the Singhalese ‘ta,” which are commonly supposed to have a different origin from ‘ku’, and *ghai.’ In the Finnish family of languages the Turko-Dravidian dative re-appears ;—though the Finnish proper has ¢le,” not ‘ke. tg In the Irtish and Surgutish dialects of the Ostiak the suffix of the dative is ‘ga,” corresponding to the Oriental Turkish ‘ga’ or ‘ge. The ordinary Ostiak has also ‘a,’ softened, as in the Oriental Turkish THE DATIVE. 227 | the itself, from °ga.’ Compare also the Mordwin adessive suffix ¢ va’ his 1 or ‘ga. ed 2 The most interesting and remarkable analogies are those which We - and have been brought to light by the Scythian tablets of Behistun. on of learn from those tablets that a dative suffix, which is almost identical with the Dravidian, and also with the Turkish and Ostiak, was used have by the oldest Scythian dialect of Central Asia of which any remains 1fica- are extant. The dative case-sign or suffix which is most largely used Mr. Norris noticed the Inion in the Scythic tablets is ¢ikki’ or ¢ikka.’ tive- resemblance of this suffix to the Magyar genitive-dative ‘nek’ and 1 not the Telugu genitive post-position ¢ yokka;’ but its resemblance to the this dative suffix of the Telugu and of the other Dravidian dialects is lative closer and more reliable. n the The ¢Tamil ‘ku’ becomes, as we have seen, ‘akku’ or ¢ukku’ Joubt, in construction ; the Canarese ‘ge’ becomes ‘ige,’ and the Malayala at 1§ ‘kka’ becomes ‘ikka; which last form of the suffix is identical, alogy letter for letter, with the Scythian of Behistun. Compare, e.g., the eve, Cuneiform Secythian ¢ni-ikka’ or ‘ni-ikki} ¢o thee, with the cor- the responding Malayala ¢ nani-kka, and the Telugu ‘ni-ku. tlars, It has thus been shown that the principal languages of the e Scythian family accord very exactly with the Dravidian languages in the nse of ‘ka,” ‘ki, ‘ku, or some related particle, as the suffix of ish the dative. It may be noticed also, that in the language of the Malays there is orth- a prefix, ‘ko’ or ‘ka,’ which signifies fo or for, and that there is a nd ir{ )iti8 similar preposition even in Russian. It is difficult to determine whether the Finnish dative suffix ‘le’ g’ rded. has any connexion with ke.’ It certainly seems much more closely connected with the Tibetan, Pushtoo, and Marathi dative suffix €14 ;’ o t0 < 2 o 0f —which ¢la is evidently equivalent to the New Persian ra this [Compare, e.g., the Marathi ¢ tu-14,’ to thee, thee, with the corresponding iff New Persian ¢ to-rd.’] 1w 10 och 18 The Malayalam alone of all the Dravidian dialects appears to e t0 possess two suffixes of the dative, viz., ‘kka, which is the suffix most largely used, and ¢inna,’ ‘na,’ or ‘a,’ which is occasionally used in Jriys have the dative singular only. This ‘inna’ is a compound form; and is The Tamil is fond evidently euphonized and softened from ¢in-ka.’ of adding to the base of nouns which are to be declined the euphonic alive increment ‘in’ (originally a genitive), before suffixing the signs of case. The same practice prevails in Malayélam also. Consequently, f the 10 this exceptional Malayéla dative is not ¢inna’ or ‘mna,’ but is simply ish “a; and the doubled ‘n’ which sometimes precedes it (e.g., ‘awanna,’ Q 2 228 THE NOUN. it has to him) is an euphonic compensation for the loss of the ¢ k/ 9 The ¢k’ or ‘g of ‘ka’ or ‘ga’ has been softened away in some dialects of ghlati locati the Turkish and Ostiak, precisely as in Malayalam. T The ablative of motion or fifth’ case.—This case appears to have (ana Tami been included in the list of cases by Drividian grammarians out of deference to the grammatical principles of the Sanscrit. locati It is true, that if we look at the construction and meaning of a (ana Dravidian sentence, the signification of an ‘ablative of motion’ will form, be found to exist ; and it will be found to be expressed much more clearly even than in Sanscrit: but a distinction is to be drawn between f i.mz' “Im the existence of a case and the existence of a case-sign, or regular, suff technical suffix of case. The Dravidian languages have undoubtedly Ingt an ‘ablative of motion,’ and a great many other ablatives besides ; mot but I doubt whether they have any case-suffix which belongs exclu- hut sively to the ablative of motion. diff On comparing the suffixes of the ablatives of motion (which are also ing used sometimes in an instrumental sense) with those of the locatives in the various dialects of this family, no real difference is apparent between Subs the one class and the other; or at least, no adequate reason appears that for regarding them as distinet and independent suffixes; for whatever ql difference does exist is to be attributed, not to the signs of case, but to the the verbs or verbal participles which are annexed to them. The the object of the ablative of motion is to furnish an answer to the the question, whence ? and this answer is obtained in the Dravidian tongues by suffixing to a noun of place the sign of the locative, and annex- sigl (m ing to that sign a verb of motion. By this means the locative is 181 converted into what is called the ablative, without changing its case- this suffixes, and the idea of change of place is thus naturally and neces- hav sarily educed. Native Tamil grammarians appear to hold that ¢il,’ €0n the ordinary suffix of the ablative, and ¢il,” the most largely used sign Mag of the locative in the colloquial dialect, though written and pronounced. sig alike, are different particles with different significations. I am per- iy suaded, however, that this view is erroneous; and that a natural an system of case classification would determine that the Dravidian Ii languages have no ablative, properly so called, but only a variety of th locative and instrumental suffixes, which are capable of becoming (X4 ablatives by the addition of appropriate verbs. t} In Tamil, the suffixes which are used in forming the fifth * case, or ablatives of motion, are €il’ and ‘in.” €il’ (Tel. ‘illu’) signifies by itself a house, a place, e.g., ‘ko-(v)-il,’ a temple, God’s house; and it is TI i therefore well suited for becoming a sign of the locative. Accordingly of THE ABLATIVE OF MOTION. 229 s it has a place in the list of locative suffixes, as well as in those of the ts of ablative; and in the colloquial dialect it is used as a sign of the locative far more frequently than any other particle. The other suffix, ¢in,’ is identical, I conceive, with ‘im,” the Old have Canarese sign of the instrumental : it is used as an instrumental in 1t of Tamil also; but probably both that ‘im’ and ‘in’ were previously locative suffixes, and were originally suffixes of the genitive. In of & Canarese the proper suffix of the ablative is ¢ attanim,” which is itself wil formed from the demonstrative adverb ‘attana’ (identical with more “attal-u’ or ‘atta, there, or ‘attal, that side), by the addition of ween “im,” the old instrumental suffix, from which inda,’ the more modern ular, suffix, is derived ; and this ¢inda,’ though the ordinary sign of the edly instrumental. is also ordinarily used, with the addition of a verb of des ; motion, as the sign of the ablative. Whilst I think that not only ¢ il Yelu- but also ‘in’ and ¢im, were originally locative suffixes, it is more difficult to determine whether ¢il’ and ‘in’ were originally identical also in sound and signification, as well as in application. 5 11 In every instance in which ‘il’ is used in Tamil, ‘in’ may be o6l substituted for it poetically ; and it is almost exclusively by the poets gars that ¢in’ is used. Moreover,in Telugu, ¢illu,” @ house, identical with ever “il, is euphonized into ¢in,” in the inflexion ¢ inti, of @ house. On 1t 0 the other hand, ¢il> and ¢im’ do not seem to have been regarded by The the Canarese as identical ; for ¢il’ is in that dialect ‘li, the base of the demonstrative local suffixes ‘alli’ and ¢illi,;’ which are used as the signs of the locative exclusively, whilst “in’ is in Canarese ‘im gues (modernized into ¢inda), and though possibly a locative in its origin, nexl- is used not as a locative, but as an instrumental and ablative. Besides 618 a5 this, if we regard ‘in as originally a locative, it will be found to Ces have a far wider range of analogies than ¢il’ and may therefore be concluded to have sprung from a different root. In Finnish and Al éj sigh Magyar we find ‘an,’ ‘en, and still more frequently ‘in,’ used as Even in Sanscrit we find ‘in’ used as a loca- 1ced | signs of the locative. tive case-sign of pronouns of the third person, e.g., tasmin,” % hum pEI' mal and though this ¢in’ may have been euphonized from ‘i, yet in the Godb 9 Latin locative preposition ‘in > and the Greek €V, corresponding to [jan the Sanscrit *ni, we find the existence of a very remarkable analogy. ol g ¢il, on the other hand, has no apparent affinities out of the pale of the Dravidian family. e St A AN It seems probable that ¢in, one of the signs of the locative in , OF Tamil, is identical with ¢in,” a sign of the genitive, or inflexional by increment, in Tamil-Canarese : and if so, a new and very wide range {15 | y of affinities is disclosed, as will be seen when the case-signs of the 230 THE NOUN genitive are inquired into attu,” whieh is often used in Tamil as a locative, is undoubtedly a genitive suffix, and this shows the possi- form bility of the use of ¢in’ as a locative being derived from its use as a yal genitive (1 The Tamil ¢il’ and in’ agree in this, that when they are used the' as suffixes of the ablative, they both require to be followed by verbs the of motion In the spoken dialect of the Tamil, the verb of motion is | 1l preceded by the verbal participles nindru,’ standing, or ¢irundu of a being. 'The use of these participles strengthens the supposition that 1t 18 i1’ and in’ are properly to be regarded as locatives In the i1’ or 6. higher dialect, however, they are ordinarily dispensed with, and m is followed by a verb of motion alone: e.g., ‘malei-(y)-in virum 1t, 10f aruvi, the cataract which falls from the mountain In this expression the the idea of ‘motion from a place is plainly implied in the aoristic 101 relative participle virum, which fualls; and hence in,” whatever it Ca may have been in origin, acquires the force of a sign of the ablative of motion of In Canarese the compound ablative suffixes attanim and inda, the terminal alj adeseyinda are not so commonly used as member of the second compound suffix; and though inda is des- ad (ua cribed to be the sign of the instrumental, I have no doubt that it is identical with im’ and n, and a locative in its origin inda 1 beg to- not only used by itself to form the ablative, but is also added te alli or jul illi,/ the sign of the locative, for the purpose of denoting the ablative Compare the Canarese allinda’ or illinda,” from, with 40 1l-irundu il-nindru IN the corresponding Tamil compound In Telugu the particle ‘na, which corresponds to the Tamil ¢in 0n sta and the Old Canarese ‘im,’ is more distinctively a locative than an ablative of motion This particle is 1 after ¢ 2 and if this is its the normal form it may at once be identified with the Tamil ¢ det ha The Telugu ablative of motion is ordinarily formed by means of the verbal participle nundi or ‘nunchi alone, without the aid of i) na or ¢ 0t Sal éor Sin any such suffix as 2 consequently this ablative has still less of the character of an independent case than in b Tamil and Canarese A locative particle, however, viz 1 which Ty ul’ and the Canarese ol 1s often corresponds to the Tamil ¢il” or suffixed in Telugu to the noun of place (precisely as ‘il or “in’ is in 0] Tamil), before the addition of the verbal participle ¢nundi This i participle is properly ¢ unchi from a verb signifying to place, which has been euphonised by pleﬁxmg to it the nasal ‘n. It corresponds in use, though not in origin, to the participle nindru,’ from ‘nil, to stand, which is used by the Tamil. ¢ — T B 231 THE GENITIVE. a8 & The genitive, or siwth’ case.—The genitive or possessive case is s formed in the Dravidian languages in various ways and by means of ) 888 various suffixes, each of which requires to be examined separately. used (1.) The abbreviated pronominal genitive—The personal pronouns of the Tamil form their inflexion,” or ordinary genitive, by shortening rerhs the included vowel of the root; e.g., ‘ni’ (properly ‘nin’), thou, on 1 This shortened form has the force ‘nin,’ thy,; ‘nAm,” we, ‘ndm, our. ndu, of a genitive in Tamil without any suffix or addition whatever, though that it is often strengthened by the addition of a suffix in the other dialects: ! {he e.g., in the Canarese it requires to have a genitive suffix appended to il or it, and of itself is merely an inflexional basis. It may be worth irum noticing that in the Scythian of the Behistun tablets the nominative of 8101 the pronoun of the second person is long, viz., ‘ni, whilst the inflex- ristic ional form and enclitic possessive ‘ni’ is short, precisely as in Tamil- er 1t Canarese. ve of We shall best understand the origin and force of this peculiar form and of the genitive of personal pronouns, by considering it as a pronominal adjective. Every Dravidian noun of quality or relation becomes an inal adjective on being prefixed to a noun-substantive for the purpose of des qualifying it ; and ordinarily the only changes which it undergoes on 118 becoming an adjective are such petty euphonic changes as are intended 1§ The change to facilitate the combined enunciation of the two words. alli’ in the quantity of the personal pronoun, to which I have now referred, ‘ the appears to have this origin alone: it is simply euphonic, and euphony with is certainly promoted by this conversion of a long vowel into a short one prior to the addition of the case-suffixes, or of the governing sub- ‘I stantive. We find a similar euphonic shortening of the quantity of n an the vowel of the root, on the conversion of the abstract noun into an is 1t6 adjective, in the section on Numerals ; e.g., ‘4ru,’ Tam., sir, ¢ iRu- badu, sizty ; ¢ éru, seven, ‘€rubadu,’ seventy. The principle which is s of involved in both classes of instances is precisely the same. id of (2.) The neuter inflexional genitive.— The neuter inflexions ‘attu,’ this cattru,’ ‘ti, ¢ti, &c., are largely used in forming the genitive in il in Jhich Tamil and Telugu. The various suffixes which are used to form the ‘inflexion’ were often originally, I conceive, signs of the possessive case : but in process of 510 This time they have come to convey either a possessive or an adjectival sig- nification, according to the connexion; and in many cases, as has been fhich shown, they have shrunk into inflexional increments of the base, or i’ "IJ’L become mere euphonic links of connexion between the base and the ; case-suffix. The inflexion which is now under consideration is in S o —— 232 THE NOUN. Tamil ‘attu,” and is used by the singular of neuter nouns alone. fhe b a0 ‘arRru,’ pronounced ‘attru,” is occasionally used by neuter pronominal plurals. The same inflexion, for I believe I have shown it to be the ¢uph suff same, is in Telugu ‘ti’ or ‘ti.’ The inflexional suffixes being, as I conceive, genitive or possessive oblig suffixes in their origin, their adjectival use naturally flowed from their a8 C use in forming possessives. There is little difference in signification between the genitive and the adjective (e.g., ¢ a mountain of gold’ and the ‘a golden mountain’ come to the same thing) ; and in several lan- tree, forr guages besides the Drividian, the adjectival formative either appears to have been derived from the possessive suffix, or to be identical with der 1t Thus, in Tamil, it matters little whether ¢ knlattu min’ (from tre “kulam,’ a tank, and ‘min,’ fisk) be translated adjectivally tank fish, or (as genitivally the fish of the tank; whether we render ‘Attru manal’ for (from ¢4ru, a river, and ‘manal, sand), the sand of the river or We river sand ; or whether ‘mattu-(t)-t6l’ (from ‘madu,’ an oz, and “t0l,” hide’), be translated ox kide or the kide of an ox. The adjectival an rendering is ordinarily the more natural one, but if a few words be I]lO.I added to the compound expression, so as to bring out the full force of sele the inflexional suffixes, it will be evident that those suffixes must have con been signs of case, or genitives, originally, and that their adjectival o use s secondary to their use as signs of the possessive. suf Thus, when we say in Tamil, ‘i-(k)kulattu min perugittru,’ to 81y render the sentence this tank fish has increas-d, would not only be bar- barous, but would partly fail to express the meaning, which is, the fish (! of t'uas tank have increased. In this instance it is evident that the suffix “attu,’ has in itself the force of a sign of the genitive, though {18 capable of acquiring in certain connexions the force of an adjectival 8 formative. So also, ‘marattu (k)koppu, can only be rendered the of branch of a tree, and @ tree-branch would be as barbarous as it is un- €l necessary. Moreover, this same suffix ‘attu,” has sometimes in Tamil (as ‘atta’ in Malayalam) the force of a sign of the locative, like the 8 corresponding inflexional suffixes in Telugu ; and when used as a suffix W of the locative, it is governed by a verb not by a noun ; from which it is absolutely certain that it is a case-suffix in origin, I have already mentioned the connexion which subsists between the inflexional suffix ‘attu’ and ‘adu,’ 4, the neuter singular demon- strative pronoun. It is deserving of notice in this place that ‘adu’ (the very same demonstrative, I doubt not) is one of the recognized suffixes of the possessive case in Tamil, and is occasionally used as a possessive in the other dialects also. Thus we may say in Tamil either “marattu (k)koppu’ (from “maram,” a tree, and ¢ koppu,’ @ branch), oy THE GENITIVE. 233 ilone. the branch of a tree, or ‘marattinadu koppu’ (‘mar-attin-adu’). ‘mar- minal amadu’ may also be used, though not in ordinary use, because in- e the euphonic ; but the possessive case-sign ‘adu’ is quite as frequently suffixed to the crude form of the noun, or the nominative, as to the S51Ve oblique form : e.g., ‘varei-(y)-adu param,’ the fruit of the plaintain, is as common as ‘varei-(y)-in-adu param,’ and is even more elegant. th'eir ation I have no doubt of the identity of the ‘adu’ of ‘vérei-(y)-adu’ and and The old crude base of ¢ maram, a the ‘attu’ of ‘marattu’ in origin. | lag- tree, is ‘mara,” as found in Canarese, the final ‘am’ or ‘m’ being a peats formative; and on ‘adu,’ the sign of the possessive (originally a with demonstrative) being added to ‘ mara,” we shall have ‘maradu,’ of a from \ {ree (in Canarese ‘marada’); of which the ‘d’ has only to be doubled sh, or (as it is colloquially by the Tamil people, many of whom say ‘attu’ anal’ for “adu’), when the word becomes ¢ marattu,” the very form in which er or we now find it. and In Telugu, the inflexional suffixes “ti’ and ‘ti’ are used without fival any additional particle as signs of the possessive or genitive, even 5 be more frequently than in Tamil. The post-position ‘yokka’ is but ¢ of seldom added to it, and needs not ever be added. In Telugu also the have connexion subsisting hetween this suffix and the neuter demonstrative fival pronoun is still more obvious than in Tamil. ‘adi,” ¢, is systematically suffixed in Telugu to nouns and pronouns, to convert them into posses- , 10 sives (e.g., ¢ varidi,’ their or their’s), and the relation subsisting between ‘adi’ (or di, as it is in some instances) and ‘ti’ or ‘ti) is very bar fish close. the In Canarese the corresponding particles ‘ad’ and ‘ar, though used as inflexional increments of the base, prior to the addition of jugh ival several of the signs of case to certain classes of nouns, have not now the of themselves a possessive signification. Their present use is purely euphonic, and does not contribute to grammatical expression. Ul il Nouns in which ‘ad’ and ‘ar’ are introduced form their posses- the % sives in ‘ada’ and ‘ara ; and in these forms the final ‘a’ is that which contains and conveys the possessive signification. ‘ad’ and ifix ich “ar’ have only the same incremental or euphonic force in ‘ad-a’and ¢ar-a, that ¢in’ has in ‘in-a, which is a corresponding Canarese possessive. eel 101 du’ (3.) The neuter-demonstrative genitives.—adu,’ t, and its eupho- nically lengthened equivalent ¢adu,’ are often used, especially in zed classical Tamil, as signs of the possessive, and they are ranked by s ‘adn’ is the neuter het native grammarians amongst genitive case-signs. bl singular demonstrative (derived from ‘a, the remote demonstrative ] f 234 THE NOUN. base, and ¢d,’ the sign of the neuter singular). Its meaning when § [ standing alone is invariably that of a demonstrative pronoun, but by usage it has acquired the signification of a genitive or possessive, when il annexed to any noun as a suffix. ‘avan-adu,’ is literally Ae—that, equ but by usage it means /is. This use of ‘adu,’ as a possessive suffix, is for derived from its use as the formative of nouns of possession. y1 doe By the addition of this demonstrative to any noun or pronoun (generally it is added to the inflexion,—in the case of prononns it is e always to the inflexions that it is added) a compound noun of posses- thy sion or relation is formed, which, like all Dravidian nouns of relation, S 1s capable of being used as an adjective ; and it is the use of nouns 1I( with this termination as possessive adjectives which has led to ‘adu, and its equivalents, being regarded as signs of the possessive case. The al noun to which ‘adu’ is appended may be used, and often is used, f without any addition or modification, as the nominative of a verb or o a sentence. Thus, ‘enadu,” Tam. (from ‘en, my, and ‘adu,” that), i signifies properly that (which is) mine; and this compound possessive may either be used adjectivally, e.g., ‘enadu kei,’ my hand, literally ¢he 0 kand that is mine (in which instance ‘adu’ is called by grammarians U a genitive case-sign) ; or it may be used as a possessive noun, and as W ;; such it becomes the nominative of a verb, e.g., <enadu poyittru,” mene 9 (or my property) is gone. Thus ‘adu, which at first meant that, il became secondly the formative of a possessive noun (‘avan-adu,’ #hat which us hus, literally he—that), thirdly the formative of a possessive adjective (‘avan-adu,’ %is), and lastly a sign of the possessive case gene- rally, signifying of or belonging to. Another reason for regarding the genitive case-sign ‘adu’ as originally and properly the formative of a noun or adjective of possession, is that it cannot be followed indiscri- b minately by any kind ot noun, but by neuter nounsalone, and properly by the neuter singular alone. Thus we may say ‘enadu kei,’ my { hand, but not ‘enadu keigal,’ my hands ; except indeed in the colloquial dialect, in which the singular is used for the plural more frequently than in the higher dialect, or by the poets. The higher dialect would prefer in this instance ‘ena keigal,—¢ena,’ instead of ‘enadu,’ i.e., ‘mea’ instead of ‘meum. ‘adu’ is not only a formative, therefore, but is distinctively a neuter singular formative, employed to give a possessive signification to the noun to which it is suffixed. Like all other nouns, these possessive nouns in ‘adu’ are capable of being used as adjectives, by being prefixed without altera- tion to other nouns ; and when so prefixed, ‘adu’ came to be used and regarded as a possessive case-sign. This explanation seems to account for all the phenomena, and therefore to be the true explanation. THE GENITIVE. [N 3 O vhen A similar use of the neuter singular of the demonstrative as a pos. by sessive suffix obtains in Telugu also ; e.g., ‘ nadi,” mine, literally that vhen (which is) mine, from ‘na,” my, and ‘adi,’ that, a form which is exactly 0t equivalent to the Tamil ¢enadu.’ The Telugu uses a similar suffix to X, I form a plural possessive to correspond with ‘enadu’ or ‘nadi, viz., ‘vi, which bears the same relation to ‘avi’ those (things), which ¢di’ 10un does to ‘adi,” that (thing); e.qg., < varivi) theirs or the (things which it 18 are) theirs. In this respect the Telugu acts more systematically than 8868 the spoken Tamil. It is not so fond, however, of using these posses- tion, sive nouns adjectivally as the Tamil, and therefore ¢ di’ and vi’ have ouns not in Telugu come to be regarded as case-signs of the genitive. The Canarese and the Tamil not only form neuter possessive nouns ad, The and adjectives by adding to them the neuter demonstrative, but they 1sed, form also masculine and feminine possessives, or possessive appellatives, ' or of both numbers, by adding the masculine and feminine formatives to the genitive case or ¢ inflexion’ of nouns and pronouns. fat), 38176 All the Dravidian dialects agree in appending the demonstrative- the possessive suffixes to the inflexion not to the nominative, as a general rule, wherever the nominative differs considerably from the inflexion. 1318 | as When nouns receive in Tamil a double inflexional increment, e.g., e ‘attu’ and ‘in’ (in combination ‘attin’), the possessive suffix is added to this double increment, e.g., ¢ mar-attin-adu koppu,” the Tt branch of @ tree. that 5Ive ene (4.) The possessive suffiz ‘in, and its varieties—"in,’ in Tamil, and ‘ni, in Telugu, and corresponding particles in the other dialects, the are not only used as inflexional augments of the base and euphonic of 8 i bonds of connexion between the base and the case-signs, but also as suffixes of the possessive and as adjectival formatives. I have no ly doubt that ¢in’ and ‘ni,’ of themselves and originally, were genitive i or possessive suflixes, and that every other use to which they have nial | Native been applied grew out of their use as signs of the genitive. ptly Tamil grammarians do not include ‘in’ amongst their genitive suffixes, but describe it as a formative angment or adjectival increment alone : na but on comparing its use in Tamil with its use in the other dialects, I lnly am convinced that it was originally and distinctively a sign of the geni- 1ve, tive, and that it is still to be regarded, notwithstanding its subsidiary f 18 uses, as one of the most characteristic of the genitive suffixes. are In Tamil, of all genitive case-signs, ‘in’ is that which is most fre- o1 quently used. ‘attu’ is used in the neuter singular alone, and ‘arru’ qod (‘attru’) in the neuter plural alone; but ‘in’ is used in connexion Mﬂt with both numbers and with all genders. A similar use of ‘in’ appears 236 THE NOUN. in the Malayéalam In Canarese, on the other hand in’ is used only and One of the so-called n as an inflexional augment, not as a sign of case declensions of the Canarese is said by grammarians to take ‘ina’ as 10-) 1ts genitive case-sign ; but in this instance the final 1s the real sign of the genitive, as it invariably is in Canarese ; and this genitive e ‘a’ is found to be preceded by various euphonic increments—¢in, the @, o S§ i: oty accordlng to mrcumstances Doubtless the ‘in’ of ‘in-a,” was a genitive suffix originally, but it fror has long ceased to contribute to grammatical expression, and therefore 110§ Wi cannot now be regarded as a sign of case In Telugu na’ or ‘ni, the dialectic equivalent of ‘in is used as a possessive suffix, as in In Tamil, though not so frequently. The only difference in principle is Sl Sl that ‘ni’ is used in Telugu in connexion with the singular alone, and might be called a genitive singular case-sign, if the Telugu stood in an th isolated position ; whereas in Tamil it is used in connexion with plural nouns as frequently as with the singular Wi gé In Ku, which has special resemblances to the Telugu, ‘ni’ constitutes the inflexion (in reality be the genitive) of all classes of nouns, whether singular or plural, pre- 2 10 cisely like the Tamil The Gond uses as genitive case-signs ‘na bl of and ‘na ‘da and ¢a’—forms which are probably allied one to as another, as well as to the Brahui na,” and to the Telugu and Gond an ‘ni and the Tamil ‘in bﬂl Though ‘in’ is not regarded by Tamil grammarians as a sign of % the genitive, yet when those particles which are regarded as genitive 18 case-signs are suffixed to any noun, ‘in’ is ordinarily inserted between the noun and those case-signs ; so that all auxiliary or additional par- a ticles are appended to this incremental ¢in, ’ not to the noun itself i e.g., from ‘adu,’ #, is formed not ‘ad-udeiya,” but ¢ad-in-udeiya,” of n from ‘tambi,’ @ younger brother, is formed not ‘tambi-(y)-adu, toJ but more commonly ‘ tambi-(y)-in-adu,” of @ younger brothes and this il rule seems to indicate that ¢in’ is more essentially a genitive case sign il than the genitive particles which have subsequently been suffixed to it The same inference is still more clearly deducible from the circumstance ) that in a large number of instances, both in the singular and in the plural, each of the case-suffixes in succession is appended, not to the crude form of the noun, but to the increment “in These case-suffixes are not mere post-positional fragments, but were, or are still, nouns of relation ; and ‘in,’ the particle by which they are united to the base, serves as a bond of connexion, in virtue, as I conceive, of its own ori- ginal and natural signification as a suffix of the genitive. Thus, in the colloquial Tamil, ‘kallinidattil (‘kal(l)-in-idattil’), i @ stone, ‘idat- til, the local ablative or locative suffix, literally means i the place THE GENITIVE. 237 mly and this suffix evidently requires, cr at least desires, the possessive lled ‘in’ (with the signification of’) to connect it with the base. Hence ‘ kal(l)- a8 in-idattil,’ literally signifies in the place of (or occupied by) a stone. real The adjectival meaning of ‘in,’ though not its only or original ftive meaning, is one which is recognised by native grammarians, and which in, they prove by examples ; e.g., ‘ponnin’ (‘pon(n)-in’) kudam, a golden vessel. This adjectival use of ‘in’ is not only allied to, but is derived it 1t from, its use as a sign of the genitive, and in the illustration which has fore now been adduced it is evident that ‘ponnin kudam,” might be rendered It will be found also in the ‘n with equal propriety, @ vessel of gold. § 1 Indo-European analogies which will presently be adduced, that the e is similarity or identity of the adjectival formative,and the genitive case- and sign which is apparent in this instance, has a wider range than that of 1 all the Dravidian languages. ural There is another particle resembling ¢in,’ and possibly identical with it in origin, viz., ‘am,” which is occasionally used in Tamil for lal lity both those purposes, and, like ‘in,’ it is sometimes appended to the We see this fusion pre- noun itself, and sometimes to the neuter inflexion. it} of the adjectival and the genitive signification of ‘am,” in such forms t0 as ‘4lam (‘al-am’) pu,’ the banyan flower or the flower of the banyan, dnd and ‘attrang karei’ (‘attru,’ the inflexion of ‘a4ru,’ @ river), the river- bank or the bank of the river. Evidently ‘attram’ (before the ‘k’ of 1 of ‘karei’ ‘am’ changes into ‘ang’) is equivalent to ‘4ttr-in’ (a form which is also commonly used), and ‘am’ to ‘in; and as ‘am’ or ‘an’ and tive een ‘in,’ are identical in meaning, though not used with equal frequency, p&l‘ and so nearly alike in sound, I think we may safely regard them all as sprung from one and the same origin. ‘am,” indeed, changes by rule elf; into ‘an,” when it is followed by a dental, e.g., ¢ panan’ (‘panei-am’) f d, toppw,” @ palmyra tope. The same adjectival formative is much used in Malayéla also ; eg., ‘mal-am puli’ (‘mala-am puli’), @ mountain this tiger or a tiger of the mountain, a royal tiger. igh o1t We have now to inquire whether any trace of the genitive case- sign or adjectival formative in ‘in,” ‘an ‘ni,’ or any related form, nee the can be found beyond the circle of the Dravidian dialects. i Of all the North-Indian vernaculars the Gujarathi is the only one the which contains a form of the genitive resembling that which we have e gof been examining. That language has a genitive suffix in ‘n’ (‘né ‘ni,’ ‘nun’), which cannot but be regarded as allied to the Telugu 156y . piy Snay el In the language of the Bodos, a Himalayan tribe, the 11 the pronominal genitive is regularly formed by suffixing ‘ni,’ e.g., ‘ang-ni,’ of me, ‘ nang-ni,’ of thee, ¢ bi-ni,’ of him. Jat: In Sanscrit the ‘n’ which precedes the ‘ah’ or ‘as’ of certaip ft 238 THE NOUN. genitives is undoubtedly euphonic; but both in Sanscrit and in other plu inte members of the Indo-European family, we may observe distinct traces of the adjectival or genitival use of a particle of which the consonant L‘clp [4 n Ed is the most essential element. Compare with the Dravidian plur P particle ¢ an-a,” the Sanscrit adjectival formative, and an,’ the suffix of appellatives; the Greek possessive suffix ¢ wv;’ the adjectival use of &e [4 Ly 2 in Greek words like ¢A/6-w-o0s,” and of ‘en,” in the Germanic “wooden; and also ‘in,’ the Sanscrit suffix of agency, which is pre- fam served in the adjectives of the New Persian. These forms are, it is plu evident, reciprocrally related ; and it also appears probable that there Bel is some ulterior relationship between them and the Tamilian ‘in.” There S are also traces in the Indo-European family of languages of the use of ‘In’as a distinctively genitival suffix. The Celtic forms its genitive systematically by means of ¢n,” ‘an,” ¢en,” &ec.: nor is it the genitive 1§ plural only of the Celtic dialects which uses this case-sign (as in the Bt Sanserit family), but it is employed to form the genitive singular also. 8l It should be noticed too that in the ancient Egyptian ‘n’ (alternating Bl with ‘m’) was used to express all case-relations, but particularly con that of the genitive. Compare also the Sanserit genitive or possessive ‘mama’ (‘ma-ma’) of me, my, with the Zend ‘mana,” the Old Per- as~] i} sian ‘mand,’ and the Gothic ‘meina, mine, °theina, thine, seina,’ Wit} kis; in each of which examples the final ‘na,” or its Sanscrit equiva- 0f¢ lent ‘ma,’ resembles the Dravidian ‘in’ or ‘ni,’ not only in sound, but also in the union of an adjectival signification with that of the 8l possessive or genitive case. i} The Lithunian goes further than any other Indo-European tongue in resemblance to the Tamil in this point, for it not only uses ‘n’ as or a sign of the pronominal possessive (of the first person), but it adopts per this genitival ‘man’ as the inflexional base of all the rest of the oblique cases of the same pronoun. Was In the languages of the Scythian stock we find a large number fon of still more essential analogies with the Drividian genitival suffix 1 fin?” or ¢ni’ for Compare both with the Dravidian and with the Indo-European possessives the Mongolian and Manchu ¢mini’ (‘mi-ni’), of me, my ; gﬂ and the Mongolian ¢tchini’ and the Manchu ‘sini’ (‘si-ni’), of thee, thy fu In the languages of the Finnish family, the prevailing form of the 2 T( genitive is that which corresponds to the Dravidian: it is ‘n,’ ‘an, ak en,” ‘un, &ec., not only in pronominal inflexions, but universally. ch Thus in Mordwin and Cheremiss, the genitive is formed by suffixing 8 ‘n’or ‘en; eg., ‘kudo,’ @ house, ‘kudo-n, of a house. The genitive frg THE GENITIVE. 239 ofher plural of the Mordwin is “mnen,’ possibly a reduplication of ‘n, ra0es intended to symbolise the plural; e.g., ‘kudot-nen,’ of houses. The onant Lappish genitive takes ‘n’ or ‘en’ in the singular, and ‘i’ in the idian plural. ¢¢’ forms the ordinary possessive suffix of the Magyar. The uffix ) Finnish proper forms the genitive by suffixing ‘n,” ‘un,” ‘in ‘an, 156 of &ec., e.g., ‘mind’ (‘min-a’), I, ‘min-un,’ of me, my. nanie The prevailing form of the genitive in the Tartar or High Asian § pré- families, corresponds to ‘nen,” the reduplicated suffix of the Mordwin ; it i plural, and to its equivalent reduplication in the Old Scythian of the there Behistun tablets; but whilst the reduplicated suffix is very frequently There used, it systematically alternates with the simpler suffix ‘un’ or ¢in.’ 15¢ of The Oriental Turkish forms its genitive by suffixing ¢ning’ or nifive fmin,’ or ‘ning’ or ‘nin’ In the Ottoman Turkish the initial nasal litive is only occasionally used: the genitive plural is uniformly ‘un; the n the singular takes ‘un’ or ‘nun,’ according as the noun to which it is also. suffixed ends in a consonant or in a vowel. A9 In the Mongolian, the 9 sign of the genitive is * u after the consonant ‘n; after every other ating Jazly v consonant, ‘in; and after a vowel, ‘yin. The personal pronouns, :ssive as has already been observed, from their possessive by suffixing Per ‘ni, e.g., ‘mi-ni,) my. Compare the Mongolian ‘kél-tn,’ of the jfoot, eing, with the ordinary Tamil genitive of the corresponding noun ¢kéal-in, of o foot. [uim- Olm‘l} The Calmuck dialect of the Mongolian forms its genitive by )f the suffixing ‘i’ to nouns ending in ‘n,’ and ‘yin’ to all other nouns. 2259 The Tibetan postfixes in like manner ° 1 or ¢yin. ngue The Manchu makes much use of a possessive relative suffix ¢ ngge,’ n' 88 or ‘ningge,’ signifying which has; but it also forms genitives, pro- perly so called, by suffixing ‘ni’ or ‘i dopfs [ the In the language of the Scythian tablets of Behistun, the genitive was ordinarily formed by suffixing ‘na:’ the first personal pronoun formed its genitive by suffixing a reduplicated form of this particle, mber quffix ‘ni-na,’ e.g., ‘hu-ni-na,” of me; whilst the genitive plural was generally formed by means of the addition of ‘inna,’ probably softened from fpEana. The nearest direct resemblance to the Behistun—Scythian »pean s genitival ‘na,’ is the Brahui ‘n4,” and the Gond ‘na’ or ¢4’ This interesting record of the speech of the ancient Scythians, ﬁlt’ei furnishes us, I think, with a clue to the origin of ‘nun’ or ‘nin,’ the j:; Tartar genitive suffix. f the In the Tartar tongues ‘nun’ is interchange- able with and equivalent to ‘uxn; and ‘un’ or ‘in’ is also inter- ‘au aHV changeable with ‘ni or ‘nuj in Mongolian, ‘yin’ and ‘un’ are suffixed to substantives, ‘ni’ to the personal pronouns. It appears 1% from the Behistun tablets that ‘na,’ the ordinary genitive suffix, was fire et 9 40 THE NOTUN. sometimes euphonically changed into ‘ni-na,’ and that this again was sign softened into ¢inna.’ I conceive that the Tartar ‘un’ was in this ‘I same manner, by the reduplication of the nasal, converted into ¢ nun;’ bl which in Manchu became ‘ngge’ or ningge Possibly also ‘ni’ or ‘nu’ was nasalised by the addition of a final ‘n’ or ‘ng,’ of the use 1oun of which we have an instance in point in the final euphonic ‘n’ of the CODS( alre first and second personal pronouns in most of the Scythian languages. A parallel instance of the reduplication of a nasal is apparent in the ‘W Telugn itself, in the conjunctive or copulative particle. This particle the 18 mm ) vin v Pamil, 4G in Canarese, and ‘nu’ in Telugu; but this and Telugu ¢ nu becomes euphonically ‘nnu,’ and by reduplication ‘nunnu’ in particular instances. gen (5.) The genitival suffiz ‘a.’—This sign of the genitive or possessive stal claims to be regarded as equally with ‘in’ or ‘ni,” a distinctively ohe Dravidian suffix. It is little used in Tamil, though placed first in the taf list of genitive case-signs by Tamil grammarians; but if we take all Ise the Drividian idioms into consideration, it is perhaps more largely infl used than any other suffix of the genitive—a proof of the accuracy suff of the Tamil classification. GXPD I conceive this suffix to be identical with ‘a,’” the formative of the aldi most frequently used Dravidian relative participles (see the section hase on ‘The Verb’), but totally distinct from ‘a,’ the neuter particle of the pluralisation, which bas already been investigated. hot In Canarese ‘a’ is the only sign of the genitive which is ever the used. It is sometimes preceded by an euphonic consonant, which is of inserted between it and the base, to form a link of connection between nt them, viz., by ‘v’ or ‘y,’ the use of which is purely of an euphonic plet nature, and by ¢in ¢ad’) or ‘ar’, which are inflexional increments Teflg of the base, and old petrified genitives; e.g., ‘ guru-(v)-a, of a priest; by g ‘kuri-(y)-a,’ of a sheep; ‘kis-in-a,” of @ child; ‘ mar-ad-a,’ of a tree; iﬂ C: ‘ad-ar-a,” of that (thing), or of it. When this genitive ‘a’ is added ! to the abbreviated inflexional form of the Canarese personal pronouns, oo the final nasal of those pronouns is doubled, e.g., ‘nanna’ (from ‘nin,’ I), of me; ‘namma’ (from ‘nam,” we), of us. A comparison ity of these forms with the Tamil and Tulu ‘nama, of ws, our, proves Q“k that the doubling of the final nasal arises from an euphonic source. o ¢a’ forms the genitive suffix not only of the singular of Canarese an nouns and pronouns, but also of the plural, whether the noun belongs to the ‘rational’ or to the ‘irrational’ class, e.g., ‘avar-a,’ of them i} (epicene), ‘avugal-a,’ of them (neuter). fo These examples prove that ‘a’ is the true Canarese genitive case- g 241 THE GENITIVE. | Was sign: and it is also to be noted that this case-sign is never used, like {lis ‘in’ in Tamil, as the common fulcrum of the suffixes of all the ;) oblique cases, but is used solely as the case-sign of the genitive. or In Tulu ‘a’ forms the genitive of by far the larger proportion of ) use nouns. In many instances it is preceded by “d’ or ‘t:’ but this f the consonant is merely the equivalent of the Canarese ‘ad,’ which has ges. glready been referred to; and in the genitive of the personal pronouns 1 the ‘a’ is preserved purer in Tulu than in Canarese. Thus, instead of rticle the Canarese ‘nanna,’ of me, the Tulu has ‘yan-a’ (for ‘nan-a’), - this and instead of ‘ninna,’ of thee, it has ‘nin-a.’ ation The language of the Kotas of the Nilgherry hills forms all its genitives by suffixing ‘a .’ In Telugu ‘a’ forms the plural inflexion or genitive of all sub- Ve stantive nouns without exception. “lu,” the pluralising particle, is ively changed into ‘la;’ and as the ‘u’ of ‘lu’ is added merely to facili- n the tate enunciation, and ‘1’ alone coustitutes the suffix of the plural, it ke all is evident that the ‘a’ of ‘la’ is a suffix of case. As the plural oely inflexion, ¢a’ constitutes the fulerum to which the other case-signs, or racy suffixes of the oblique cases, are added; and as the genitive plural, it expresses the signification of the genitive, without any auxiliary or f the additional particle. The Telugu personal pronouns use their crude otion bases adjectivally as their inflexion and genitive. The pronouns of the third person, or the demonstratives, generally form their genitives cle of both iu the singular and in the plural by adding ‘i’ to the root: in gver the singular a few of them suffix ¢ ni,” as is done by the greater number of nouns in the singular. One of the Telugu pronouns uses ‘a’ both ioh 18 ween in the singular and in the plural, as the sign of the genitive, in com- plete accordance with the Canarese and Tulu. The genitive of the honic 1ents reflexive pronouns ¢tan-u,’ self, ‘thm-u,’ selves, is formed in Telugu st by shortening the quantity of the radical vowel and suffixing ‘a,’ as in Canarese ; e.g., ‘tan-a,’ of self, ‘tam-a,” of selves. The adjectival free; 1dd€d ca’ of some Telugu substantives is evidently identical with this genitival ‘a; eg., ‘tUr-a kavi, a village poet or a poet of the village. 0UNS, In Tamil, though “a’ is placed first in the list of genitive suffixes, (from it is less used than any other sign of the genitive, and indeed is used LriSOH only as the classical genitive of the personal and reflexive pronouns ; p0Ves uree: e.g., ‘nam-a,’ our (from ‘ném, we), like the Sanscrit ‘mama,” my, and ‘tava,’ thy. a1est It is difficult, indeed, to determine whether this suffix has retained Jongs in Tamil any genitival signification whatever. Whether it be attached e to a singular or to a plural pronoun, it must be followed by, and be in g agreement with, a neuter plural noun ; and this circumstance would R 4 242 THE NOUN. i lead to the conclusion that in Tamil it is a suffix of plurality, not a othe sign of the genitive. On this supposition, in the words ¢ena keigal,’ my hands, ‘ena’ would signify not ‘mei, of me, but ©mea,’ (the it ! things that are) mine. It would be a pronominal adjective or posses- this sive plural, not a genitive ; and the fact that ‘a’ is largely used in classical Tamil as a sign of the neuter plural (e.g., ‘sila, few, anal Kol literally « few things, ¢ pala’ many, literally many things), proves that this supposition would be a very natural one. poss On the other hand, “a’ is classed with genitive suffixes by Tamil The Grammarians themselves, and those grammarians, who are remarkably res well acquainted with the principles of their own language, are perfectly Bu aware that ‘a’ is also a sign of the plural of ¢irrationals’. Moreover, ch th though it is stated by Tamil grammarians that the genitive in ‘a’ must always be in agreement with a plural noun, yet they admit that the sigl f noun with which it agrees is sometimes singular in form though plural mn g in signification ; e.g., the expression ‘nun-a §ir'adi,” thy small foot ; but oceurs in the Chintamani. They say that foot is here used for feet, and m this is certainly true ; but it does not follow that ‘nun-a’ is determined sig thereby to be a plural, for the use of the singular with a plural signifi- the; cation, yet with the declensional and conjugational forms of the singular, e is a fixed usage of these languages. I think, therefore, that we may con- con fidently regard this ‘ nun-a’ as an illustration of the use of ‘a,’ even in Witk Py Tamil, in connection with the singular. In Tamil, it is true, ‘a 18 (ot ordinarily followed by the neuter plural alone; but in Canarese and Do Telugu it may be followed by any gender or number; and the ‘a’ of oy the Tamil ¢ tan-a,” of self, is evidently identical with that of the corre- 101 sponding Telugu ‘tan-a; whilst the ‘a’ of ‘nam-a, of us, our, is evi- of | dently identical with the Canarese ‘ namm-a.” Hence, as the one ‘a’ is 0rg unquestionably a genitive, so must the other have been originally ; and of ¢ thus we are led to the supposition that the Tamil rule which requires Tent ‘a’ to be followed by the neuter plural is merely a secondary, (Y recent, dialectic peculiarity, which has arisen from the influence of its \ln(} accidental resemblance to the sign of the plural of ‘irrationals.’ This Sing peculiarity of the genitival ‘a’ in Tamil may be compared with the somewhat parallel case of the use in Hindustani of one possessive ?; suffix rather than another, out of the three that exist in it, according the to the gender of the noun which follows and governs that to which it a8 is suffixed. of Though in grammatical Tamil ¢a is always followed by the of plural, yet the vulgar in the rural districts commonly use it without ol discrimination of number, as in Canarese and Telugu. Thus, they will say “nama’ (or more commonly, as in Canarese, ‘namma’) dr, the THE GENITI1VE. 243 not a our village ; and this confirms the supposition that in Tamil, as in the 2 ¢ D eigal, other dialects, the original use of this was simply that of a suffix ' (the of the genitive. 0856S We have now to inquire with what language or family of languages ed in this genitive suffix should be affiliated. There is no direct Scythian for, analogy for it, and the only affinities which I have observed are 3suthat Kole and Indo-European. In the Ho,a Kole dialect, ‘a’ is a common possessive suffix ; and it is also, as in Tamil, an adjectival formative. Tamil The most direct and reliable Indo-European analogy is that which is kably presented by the personal proncuns, which in some of the Indo- rfectly European dialects have a possessive in ‘a,” strongly resembling this BOVEL, Dravidian possessive. If we look only at the Gothic ‘meina, my ; ' must ‘theina,” thy; ¢seina, his or i«ts; we should naturally conclude the at the sign of the possessive in these words to be, not ¢a,’ but ‘na’ (answer- ing to the Old Scythian and Brahui ‘na’ and to the Telugu ‘ni’); plural but on comparing the forms which this sign of the possessive assumes foot in various languages, it appears probable that ‘a’ alone conveys the #,and mined signification of the possessive ; and that the nasal which precedes it in the Sanscrit ¢ mama,” the Zend ‘mana,” and the Gothic ‘meina,” may it merely have been inserted euphonically for the purpose of keeping the gulat, contiguous vowels pure. Compare ‘mama,’ Sans. my, (from ‘ma,’ /), y con- with ¢tava,’ thy, (from ‘tva,’ thou); and especially compare the yen I Gothic ‘meina, ¢theina,” ¢ seina,” with the corresponding Lithuanian 0 8 2 possessives mana-s, ‘ tava-s,’ ¢ sava-s. In these 1nstanees Cv- e and euphonic is used as the equivalent of ‘n.’ The Indo-European pro- '’ of % is exceptional : for the primitive languages CoITe- nominal possessive in ‘a ig evi- of that family evince an almost perfect agreement in the use of ‘as)’ ‘3’fs or some closely related form, as the sign of the genitive singular, and In the later of ‘sAm’ or ‘Am,” as the sign of the genitive plural. i and | Teutonic dialects, however, a genitive case-sign in ‘a’ becomes :quires exceedingly common, and is found in the plural as well as in the ndary, of Its singular. Thus in the Frisian all plural substantives and such This singulars as end in a vowel form their possessive by suffixing ‘a’: in the Icelandic all plurals and all masculine and neuter singulars use th the ‘a’ as their case-sign; and in the Anglo-Saxon all plurals. Though G the oldest Gothic possessives accorded with the ordinary Sanscrit forms, ording ‘as’ and ‘4m; yet the resemblance between the possessives of some hih i of the Teutonic vernaculars and the Dravidian possessive is deserving of notice. The use of ‘a’ as a sign of the possessive by all plural )y [he substantives in Telugu is especially remarkable. Has the Dravidian ifhott s £a ander consideration been softened from ‘as’ (of which, however, L there is not the smallest trace or analogical probability)? or has it IIfy R 2 244 THE NOUN. been softened from ‘na,’ the old Scythian suffix ? The latter suppo- sition, though unsupported by direct evidence, is not an improbable ‘Re one; for we have seen that the Gond ‘na’ alternates with “4; the JeIsC Scythian ‘ni-na’ with ¢ inna; the Turkish ¢ nun’ with ‘un;’ and the ¢ ‘nu, the Telugu particle of conjunction, with ‘i,’ its Canarese and equivalent. (4 Engl (6.) The Malayila genitive singular suffiz, ‘re’ or ¢de.’—In most cases this Malayéla genitive takes the shape of “indre’ or ‘inde,” of ‘me which “in’ is the genitival suffix and inflexional increment, which Canl has already been described. In ‘en-de,’ my, the inflexional base is of Wit itself a genitive, and the addition of “in’ is not required; hence it the appears that ‘de’ or ‘dre’ is an auxiliary genitive suffix, like the mof ‘adu’ which is so often added to ¢in’ in Tamil, and is probably from adj' the same origin. This suffix is written ‘wre; but it is always added to ‘n,” and when it is thus added, the compound is regularly pro- uf nounced, not as ‘nre,” but as ‘ndre’ or ‘nde.’ Neither the Tamil nor den the Malayilam possesses any other method of producing the sound sing which is indicated by these letters (a peculiarly euphonic ‘nd’), but Uses that of conjoining the final ‘n’ of those languages and the hard ‘r;’ bf[ 01 which, when pronounced in combination, have the sound of ‘ndr’ or, e, as some pronounce it, ‘ ndz,’ or more commonly still ‘nd.’ Thus, from The “ en,’ fo say, and ¢ du, the regular formative of the preterite participle, fro the Canarese forms ¢ endu,’ saying or having said; and this in Tamil is Th written ‘enru; but it would be erroneous to suppose ‘Ru’ to be W the sign of the preterite in Tamil instead of ¢du, for ‘enru’ is tiy intended to be, and is pronounced ‘endu’ or ‘endru,’ nearly as in ({0 Canarese. po— be Hence some analogies to the Malayala ‘re’ (in reality ¢de’) bein The which might be suggested, appear at once to be illusory. f uff Malayala ‘re,’ has been connected by Dr. Stevenson with ¢the H Ty Canarese genitive’ ‘ra It has been shown that ‘a, not ‘ra,” is the Tay genitive suffix of the Canarese, and that the ‘R’ which precedes it is Whi properly ¢ar,” an inflexional increment (like ‘ad’ and ¢in’), which is sing inserted between the root and the case-signus of three cases, besides 10 the genitive, of certain classes of nouns. The Malayala ‘re’ (de), on Wl the other hand, is suffixed exclusively to the genitive, and no other Nevertheless, as I connect ¢ de’ suffix of case is ever appended to it. Nl with the Tamil ¢adu,” 2¢; and as with this I connect also the Canarese Vi) T ¢ad’’ and its hardened form ‘ar, it may be admitted that in this modified and remote manner, the Malayala and the Canarese forms a8 g are allied. { THE GENITIVE. 245 Uppo- Still more illusory is the apparent resemblance of this Malayala bable ‘re’ or ‘de’ to the adjectival possessive suffixes of the Hindustani " the and personal pronouns, ‘rA’ and ‘ri’ (e.g., ‘méra,’ meus, ¢ méri, mea) ; to the corresponding New Persian inflexion ‘r,’ (e.g., ‘ to-rd, thy, thee) ; larese and to ¢ ra,’ the Gothic genitive plural suffix of the personal pronouns 9 of our (¢.g9., ‘unsara,’ owr, ‘izvara,’ yowr), from which the final ‘r English ¢ our’ and ‘your’ has been derived. 1 oSt The Hindustani ‘r,’ is supposed by Bopp to be derived from ¢ d;’ de,’ of ¢méra, meus, being derived from the Sanscrit ‘madiya,’ my;” but I which cannot suppose that the Malayala form has any connexion whatever 66 15 of with the Hindustani and the Persian, except on the supposition that once 1t the ¢d’ of the Tamil demounstrative neuter singular, ¢adu is re- ke the motely connected with the formative ‘d’ of the Sanscrit possessive y from adjective. added The Malayila ‘de,’ like the Tamil ‘adu,” is used as a genitive y pr- suffix of the singular alone, a confirmation of the opinion that it is nil ot derived from ‘¢ adu,” which in its original signification is the neuter sound singular of the demonstrative. In the genitive plural, the Malayalam ), but uses ‘ ute,” answering to the colloquial Tamil ‘udeiya’ (from ‘udei’), 1] belonging to, of. Compare the Malayala € enre,” ‘endre,” or ‘ende, of h‘ , 0ry me, with the corresponding Tamil ¢enadu,’ of me, that whick is nine. from The Malayéla possessive noun wmine, or that which is mine is ¢ endeta,’ tiCiP]e7 from ‘en-de, my, and ‘ata,’ i¢, corresponding to the Tamil ¢ enadu.’ amil 18 This latter ‘enadu, however, is not the genitive ‘enadu, sy, with fo he which I have compared ‘en-de,” but a possessive noun in the nomina- Rll, iS tive case; and though I suppose the Malayala ‘de’ to be itself a 7 88 n corruption from ¢ ada’ or ‘ata,” i, yet the demonstrative suffix would be appended a second time, on the origin and true meaning of ‘de’ “de’) being forgotten. We see illustrations of this repetition of an ancient The sufix in many languages; e.g., ‘malei-(y)-in-in,” High Tam. from « ‘the mountain; and this very demonstrative ‘adu,’ i, is twice used in the is the Tamil negative participial noun ‘illadadu,’ the thing which is not; in 5 1t 18 which the first ¢d,” though a representative originally of the necuter singular demonstrative, has lost its proper signification, and become a jich 18 esides mere euphonic link of connection, or technical sign, in consequence of le) ; on which ¢d ’ is required to be repeated, In Tulu the genitive of neuter mouns is commonly formed by - other 1 de’ suffixing <d,’ “da, or ‘ta, eg., ‘katti-da,’ of a knife, ‘kei-ta,” of a hand. nares The “d’ of this “da,” or © ta,” is not, however, as in the Canarese, used as the inflexional base of any other case; but is restricted to the : 1 this foruis genitive alone : hence it bears a close resemblance to the Malayala genitive suffix. 246 THE NOUN. (7.) Auxilvary suffizes of the genitive tn Telugu and Tamal Inea and § (1.) In Telugu yokka,” or ¢ yoka, s sometimes appended to the inflexion, or natural genitive, as an auxiliary suffix of case; e.g., from distin the ordinary possessive ‘na,’ my, is formed optionally, the equivalent I 18 g1V form na-yokka, my, of me This suffix is rarely used, and seems foreign to the idiom of the Asa langnage ; for no other pure Dravidian dialect possesses any suffix look! resembling it. mean A suﬁix somewhat resembling ¢ yokka, is found in the Rijmahal sed and Uron languages, which econtain an overwhelming prepon- s1gn derance of Kol elements, though formed probably upon a Dravidian quia basis Comy The possessive suffix of the Rajmahal is “ki that of the Uraon o pl ghi and If these particles are at all connected with the Telugu ¢ yoka which seems doubtful, we should be warranted in connecting the whole (il with the ordinary possessive or adjectival suffix of the Hindustani, the Prep feminine of which is ‘k (masculine k& ’), and through that suffix Place with the formative ‘ka,” of the Sanscrit possessive adjectives ‘mamaka, distix my, ‘tavaka, thy, asmakam,’ of us, our, &c A closer analogy to suffix yoka, is that of the dative post-fix of the Mikir, which is €yok’ or aotua ayok locat (ii.) In Tamil udeiya ’ is commonly appended to the inflexion of verh nouns and pronouns, as an auxiliary possessive saffix udeiya wit] udei~(y) ) means belonging to, or, literally which is the property of; and is derived from the noun udei property, possession, by the a8 ] addition of “a,” the sign of the relative participle, on the addition of adw which to any noun it is converted into an adjective Thus, enudeiya adye kei,” my hand, means literally the hand which is my property, for <en G a4 of itself signifies my. Through usage, however, there is no difference oy, in signification, or even in emphasis, between ‘en’ and ¢ en-udei-(y)-a and The Malayalam dispenses with “ya’ or “a,’ the sign of the relative suffy < u participle, ‘and uses (in Tamil ‘udei’), the uninflected noun | itself, as its auxiliary suffix of the genitive m udeiya’ is very largely used as an auxiliary genitival suffix in 1nig collogquial Tamil and in some grammars written by foreigners it is classed with the signs of the genitive; but, properly speaking, it is not \"eSe a case-sign, or suffix of case at all, but the relative participle of an 1dgy appellative verb used adjectivally, and it is to be compared not with By our preposition of, but with the phrase belonging to a8 tion the gy Locative or ‘seventh’ case.— Dravidian grammarians state that any g word which signifies ‘a place’ may be used to express the locative 4¥g, R — THE LOCATIVE. 247 In each dialect, however, some words or post-positions are so frequently to the and systematically used for this purpose that they may be regarded as , from distinctively locative suffixes. valent In Tamil, ¢ kan, an eye, which has also the signification of a place, is given in the grammars as the characteristic suffix of the locative. of the As a verbal root ‘kan’ means Zo see : its secondary signification was 7 suffx look! its third there, its fourth @ place; and in consequence of the last meaning it came to be used as a sign of the locative. It is very rarely ijma.hal used; and I have no hesitation in saying that the most distinctive sign of the Tamil locative is ‘il,’ @ place, literally a house. In collo- [)I'Ep()ll- Avidian quial Tamil the most commonly used sign of the locative is ¢ idattil,” a compound suffix, which is derived from ¢idam,” the ordinary word for Urion a place, ‘attu, the inflexion or basis of the oblique cases (‘id-attu’), and ‘il an older, purer word for @ place, which is added to €id-attu’ “yoka, ¢ whole (‘id-att-il), as the real sign of the locative, with the meaning of our ani, the preposition in. The signification of the whole suffix is literally in the { suffix place of or in the place occupied by ; but it is evident that what really maka, distinguishes the locative in this compound is ¢il, sn—the suffix of a gy o suffix ; and that the meaning which the entire compound receives in In the lowest patois of colloquial Tamil the ok’ or actual use is simply un. locative suffix which is most used is ‘Kkitta,” near, the infinitive of a xion of verb.® The higher dialect of the Tamil uses also ‘ul’ and ‘uri, ideiya’ within, amonyg, as signs of the locative. The ancient Canarese used 01, corresponding to the Tamil ¢ul; ey O as its Jocative snffix ; whilst the modern dialect uses “alli’ or ¢illi,’ an by the ition of adverbial form which answers to the Tamil ‘il “alli’ is properly an adverb of place, compounded of ‘li’ and the remote demonstrative nudeiy?f ‘a ; and its fellow is ¢illi,) compounded of the same root and i, the g ‘on proximate demonstrative. These words mean literally that place ference: and this place, or there and here ; and their use as locative (72 suffixes betokens a later state of the language than the use of ‘i1’ and elative ¢ul’ in Tamil, and of ¢6l’ in Canarese. Compare the change of ‘il’ 1 nond Sk in Tamil to ‘li’ in Canarese, with the similar change of ‘in’ in Latin into ‘ni’ in Sansecrit. offix 0 s it 8 * 1 cannot forbear noticing the remarkable (though doubtless accidental) resemblance between the double meaning of ¢il’ in Tamil (which is, perhaps, {is ot identical with ‘in’ the sign of the ablative of motion) and of ‘in’ in Latin. Each , of 4l is used as a locative suffix or affix, with the meaning of iz ; and each is used also of With as a particle of negation. The Latin ‘in’ not only means ¢», but has also the addi- tional meaning of not in such compounds as ‘ indoctus’ (like the Gothic ‘un’ and the Greek and Sanserit ‘a’ or ‘an’ privative); and in like manner the Tamil €il’ means not only e, but also not. Moreover, as the Latin ‘in’ privative is ‘an’ in some other dialects, so the Tamil ‘il,” not, takes also the shape of “al,” with hat 8] a very slight difference, not in the meaning, but only in the application. 248 THE NOUN. In Telugu the sign of the locative, which is most commonly used, the ‘16 1s more whic is ‘162’ another form frequently employed is ‘andu.’ intensely locative in its signification than ‘andu:’ it means within, and Was is obviously identical with the Canarese ‘61’ and the Tamil ‘ul. 0 be ‘andu’ means simply in, and like the Canarese ‘alli’ is properly an adverb ; and is derived in a somewhat similar manner from ¢a,’ the ih an remote demonstrative, with the additiou of a formative ‘d,” whilst i 11§ ‘indu,’ its correlative adverb of place, is derived from ¢i,” the proxi- l-u mate demonstrative. The Canarese also possesses adverbs correspond- cati ing to these, viz., ‘anta’ and ‘inta,” ‘antalu’ and ‘intalu ; but uses R E—— them chiefly to express comparison, like our adverb than. The Telugu blay locative suffix ‘andu’ (meaning ¢z) bears some apparent resemblance [nd to the Sanscrit ¢ antar,” among, but this resemblance is wholly illusory: for “andu’ is derived from ‘a,’ that, by the addition of the neuter formative ‘du,” which becomes euphonically ‘ndu,’ and corresponds not] _— not to the Sanscrit, but rather to ‘anda, that, the demonstrative Yoc; adjective of the Tamil. Sien In Telugu the post-position ¢na,” which becomes ‘ni’ after i ) 18 per used as a locative suffix in connexion with neuter nouns. ‘ni’ (and hence its equivalent ‘na’ also) is evidently identical with ¢in, the “lmp sign of the ablative of motiou in High Tamil, which I have supposed bys to be properly a sign of the locative ; and probably both are identical 9 Lon with “in, the Tamil, and “ni’ and ‘na,’ the Telugu, genitival or ay inflexional suffixes, The locative is more likely to be derived from 4g the genitive, than the genitive from the locative. With this Telugu the locative ‘na,’” we may compare the Ostiak locative ‘na,” ‘ne; and fon the Finnish and Magyar locatives ‘an’ and ‘en.’ 1af] In Telugu, and in the higher dialect of the Tamil, the inflexion or af s s —— basis of the oblique cases, which has naturally the force of a genitive, is sometimes used to denote the locative also. In Tamil the inflexion the- — which is chiefly used in this manner is “attu ; e.g., ‘nilattu, upon the earth. The Malayalam uses ‘atta’ in a similar manner: and in Telugu a tnly fuag corresponding change from ‘ti’ to ‘ta’ converts the inflexion or ot obsolete genitive into a locative ; e.g., ‘inti, of a house, ‘inta,” in a house. The same inflexion in ‘ta’ denotes the instrumental in Telugu, fhe as well as the locative ; e.g., compare ‘chéti,’ of a hand, with ¢chéta,’ Sony by a hand, but this form seems to have been a locative originally. a This fusion of the meaning of the genitive and locative suffixes corres- i ponds to a similar fusion of the signs of those cases which a comparison In of the various Indo-European tongues brings to light. The genitive Da and locative case-signs are often identical in the Finnish family of lan- on| guages also. Bearing this in mind, may we not conclude that ¢in,. ) THE VOCATIVE. 49 y used, the Tamil sign of the ablative of motion, probably a locative, and y more which is identical with ‘im, the Canarese sign of the instrumental, in, and was originally a genitive, and identical with in,’ which we have seen 1l to be so exceedingly common as a genitival suffix ? etly an In all the Dravidian idioms the locative suffixes are used like our 4 the Sometimes the locative alone is used for than, to express comparison. whilst this purpose : oftener the conjunctive particle is added to it; e.g., proxie ¢il-um,” in Tamil, ‘l6-nu,” in Telugu, which compound has the signifi- spond- cation of our even than. b uses None of the Dravidian suffixes of the locative bears any resem- Telugu blance to the locative case-signs of the Sanscrit, of any other of the blance Indo European languages, or of the North-Indian vernaculars. lusory: neuter The wocative or “eighth’ case.— In the Dravidian languages there is 3sponds nothing which properly deserves to be styled a suffix or case-sign of the trative vocative. The vocative is formed merely by aflixing or suffixing some sign of emphasis, or in certain instances by suffixing fragments of the i personal pronouns. " (and \ The most common vocative in Tamil is the emphatic ‘&, which is ), the simply appended to the noun. Sometimes, also, the vocative is formed pposed by substituting ‘4’ for the formative of gender, eg., from ‘kartan, entical Lord, is formed ‘kartd,” O Lord ; by converting the final vowel into yal ot “4y’ (a fragment of the old pronoun of the second person singular), 4 from e.g., from -tangei,’ sister, is formed ‘tangay,” O sister; or by lengthening the vowel of the pluralising particle, e.g., from ¢ pavigal,” sinners, is Telugu ) and formed ¢pavigal,” O sinners. Sometimes, again, especially in poetry, rational plurals are put in the vocative by appending to them ¢ir, on 0 a fragment of ‘nir, you, e.g., ellir, literally °ell-ir,” all ye. nitive, | In the Indo-European languages the nominative is often used for Jexion the vocative, and what appears to be a vocative case-ending is often on the only a weakened form of the final syllable. In the Dravidian lan- Jugud guages, in like manner, the crude-root, deprived of all increments, is often used as the vocative. on OF In Telugu the vocative singular is ordinarily formed by lengthening ' 0 !el llgu) the final vowel of the nominative (and all Telugu words end in ‘ara’ or Chéma‘ some vowel), or by changing the final ‘u’ into ‘a’ or ¢4’ ‘ara, from the same root as the Tamil pronominal fragment ‘ir’ (viz. nally: [4 I1e8° nir,’ ye), is post-fixed as the vocative of masculine-feminine plurals. qpisol In addition to these suffixes, various unimportant vocative particles, or particles of exclamation, are prefixed to nouns; some to one number )niti"e only, some to both. In Canarese the vocative is ordinarily formed by )f Jape L appending ‘4,” by lengthening the final vowel of the nominative, or 250 THE NOUN. by adding ‘e’ or ‘&’ Masculine-feminine plurals form their vocative In 2 with 1 not only by means of ‘e or ‘&’ but also by suffixing ‘ira’ or ¢ira,” from the same source as the Telugu ‘ard,’ viz., the old ‘nir, ye Such being the origin and character of the Dravidian signs of the vocative, it is evident that we cannot expect to find allied forms in any other family of languages. Compound case-signs.—As in the Hungarian and other Scythian tongues, and in some of the languages of the Kastern islands, so in Dravidian, two or more case-signs are occasionally compounded 1 th together into one. We have already noticed the custom of annexing the various signs of the oblique cases to the inflexion or sign of the Thus, genitive ; but other combinations of case-signs are also in use. there is a combination of the dative and locative, e.g., ¢ vittukkul’ (‘vittu-kk’-ul’), colloquial Tam., within the house, in which the locative “ul’ is combined with the dative or directive ‘kku,” for the purpose of intensifying ¢n, and educing the meaning of within. The higher I\ dialect would in this instance prefer ¢vittul,” the simple locative ; but Iangu: The ablative of motion in each of the ) ‘yittukkul’ is also idiomatical. Dravidian dialects is generally a compound case, being formed of the over, locative and a verbal participle, or even of two locatives ; e.g., ‘ mane- appe (y)-ill-inda, Can., out of the house, from €illi’ or ‘alli/ the sign of the locative, and ‘inda, a sign of the instrumental, which is ton used also as a sign of the ablative, but which was, I conceive, a Sty locative originally, and identical with ¢ im,” the Canarese form of the Tamil ¢in.’ The Malayala ‘inikkulla’ (‘in-i-kka’ and ‘ulla’), my, is a com- pound of the dative of the personal pronoun (which is itself a com- pound), and a relative-participial form of ‘ul, within ; in colloquial the § Tamil also a similar form is used as a possessive. 8 v¢ Such compounds may indeed be formed in these languages at pleasure, and almost ad wnfinitum. Arother instance of them in qua‘: Tamil is seen in the addition of the dative to the locative (e.g., of ¢idattir-ku’ or ‘kat-ku’), to constitute the locative-directive, which lay is required to be used in such expressions as / sent to him. g tio Pre Possessive compounds.—The Dravidian languages are destitute of Ing that remarkable and very convenient compound of nouns and prono- dig minal or possessive suffixes which is so characteristic of the Turkish, are Ugrian, and other Scythian families. H.djq 9 51 ADJECTIVES. pative In Hungarian they form the following compounds of “ur,” master, 3 o ' with the pronominal fragments, used as possessives :— “nir, ur-am, my master ur-aim, my masters of the ur-ad, thy master ur-aid, thy masters ur-a, his master ur-ai, his masters m any our master our masters ur-un-k, ur-ain-k, ur-at-ok, your master ur-ait-ok, your masters their master their masters ur-ok, ur-ai-k, ythian § I These compounds are regularly declined like uncompounded nouns ? unded in the usual way, e¢.g.— 1exing of the uramnak (ur-am-nak), to my master Thus, urunknak (ur-un-k-nak), to our master kkal’ 5 uraimnak (ur-aim-nak), fo my masters urainknak (ur-ain-k-nak), {0 our masters ative 0se of The entire absence of compounds of this nature in the Dravidian igher ; hut languages, notwithstanding their agreement with the Seythian group of the in so many other points, is very remarkable: it is the only point, how- ever, in which any structural difference of a generic, or class type, of the appears to exist. mane- In all the Dravidian languages the possessive pronouns are prefixed ) Sigl to nouns, as in the Indo-European tongues, never post-fixed as in the ich 18 Scythian. ive, 8 of the Part ITI.—Adjectives, or Nouns used Adjectivally. , Com- The difference between the Indo-European languages and those of con the Scythian group, with respect to the formation and use of adjectives, )quial is very considerable. The agreement of adjectives with the substantives which they s 06 qualify, in gender, number, and case, forms an invariable characteristic n 1 of the languages of the Indo-European family ; whilst in the Scythian (glvfl,a languages adjectives have neither number, gender, nor case, but are nhiCh mere nouns of relation or quality, which are prefixed without altera- tion to substantive nouns. In this particular the Dravidian languages present no resemblance to the Sanscrit, or to any other member of the te Of Indo-European stock, but are decidedly Scythian in character. Dravi- jponos [} dian adjectives, properly so called, like those of the Scythian tongues, rkiihx are nouns of quality or relation, which acquire the signification of adjectives merely by being prefixed to substantive mouns without 252 L4 THE NOUN. declensional change ; and, i’n virtue of that acquired signification, they IEpug il that th are called by Tamil grammarians ‘uri chol,” qualitative words. Parti- ciples of verbs, and nouns with the addition of participial formatives, the ma used ac are also largely used as adjectives in the Dravidian, as in the Scythian, In family. gven In Such being the simplicity of the construction of Dravidian adjec- form o tives, it will not be necessary to occupy much time in the investiga- tion of this department of grammar. It may suffice to state, seriatim, five n the various modes in which words are used as adjectives, and the many, and a formative or euphonic modifications which they undergo on being ‘nal’ prefixed to the substantives which they qualify: nor will it be neces- “mals sary to state @/l the modifications which are discoverable in each dialect, but only those which appear to be most characteristic, or 2 T which are peculiarly worthy of remark. 1. The majority of adjectives in all the Dravidian dialects, as in to of the dialects of the Scythian group, are nouns of quality or relation qood Tl which become adjectives by position alone, without any structural change whatever, and without ceasing to be, in themselves, nouns of from { quality. Thus, in the Tamil phrases, ‘ pon aridu,” gold (is) scarce, 4 (SOIH[.' and ‘pon mudi, a golden crown, ‘pon, gold, is precisely the same in of the both instances, whether used as a substantive, in the first, or as an Py adjective, in the second. adject In a similar manner in English and the other modern Indo- fion hard European dialects the same word is often used as a noun in one connexion, and as an adjective, without addition or change, in another 0y connexion ; e.q., ‘gold’ is more ductile than silver; a ‘gold’ watch: a8 al suffy but this is contrary to the original genius of languages of this family, and is the result of a process of corruption. Whilst adjectival nouns titery of this class undergo in the Dravidian languages no structural deriy; change, their combination with the nouns to which they are prefixed “atty is facilitated in certain instances by unimportant euphonic changes, ogt, to th such as the assimilation of the final consonant of the adjective ota—— and the initial consonant of the substantive, in accordance with the ¢ ay&, requirements of Dravidian phonetics (e.g., ‘por chilei’ (for ‘pon Or g chilei), @ golden vmage’) ; the softening, hardening, or doubling of the t « initial of the substantive; or the optional lengthening of the included they vowel of the adjectival word, to compensate for the abandonment of Sign the euphonic final ‘u,’ e.g., ‘kar, black, in place of ‘karu,’ or vice Wit} versd. These changes are purely euphonic; they differ in the diflferent Cous dialects; and they contribute to grammatical expression only in so fory far as they serve to indicate the words which are to be construed of {] together as adjective and substantive. It is only on the ground of the the FORMATION OF ADJECTIVES. 253 n, they repugnance of the Dravidian ear to certain classes of concurrent sounds Parti- that the changes referred to are required by Dravidian rules; and in atives, the majority of instances nouns sustain no change whatever on being used adjectivally. ythian, In the poetial dialects, adjectival formatives are less used than 1 adjee- even in the colloquial dialects; and it is generally the crude ultimate form of the noun of quality which performs the functions of the adjec- vestiga- eriatim, tive in classical compositions. Thus, whilst ‘nalla,” good, and ¢pala,’ nd the many, are commonly used in spoken Tamil, the higher idiom prefers, | being and almost invariably uses, the crude nouns of quality and relation ) 116Ces- ‘nal’ and ‘pal;’ eg., ‘nal vari,’ the good way, and ‘pan’ (for ‘pal’) 0 each ‘malar,” many flowers. stic, or 2. Sanserit derivatives (neuter nouns of quality) ending in ‘am’ in Tamil, and in ‘amu’ in Telugu, become adjectives when prefixed to other nouns by rejecting the final ‘m’ or ‘mu;’ eg., ‘subam,’ 5 48 1 elation goodness, and ‘dinam, a day, become ‘suba dinam, a good day. This, however, is a Sanscrit rather than a Drividian rule; and it flows nctural from the circumstance, that when two Sanscrit nouns are formed into s of scaee, a compound, the crude form of the first of the two nouns is used instead me 1 of the nominative,—¢ subha ’ instead of ¢subham. Pure Dravidian nouns ending in ‘am’ or ‘amu’ rarely becomie as an adjectives in this manner; and when they do, it is through imita- tion of Sanecrit derivatives. Indo- In Telugu, final ‘amu’ is sometimes hardened into ‘ampu; e.g.,from ‘andamu,’ beauty, is formed ¢ andapu’ in 00é or ‘andampu,’ beautiful.’ In Tamil, when a noun of this class is used nother as an adjective ; ‘am’ is generally rejected, and ‘attu,” the inflexion, wateh: f&mil ) suffixed instead ; e.g., from ¢ puram, externality, is formed ¢purattu,’ external. Sometimes also the Tamil deals in this manner with Sanscrit n0uDs derivatives, converting them into adjectives by means of the inflexional chural “attu:’ but in all instances of nouns ending in ‘am’ or ‘amu,’ the efived most common method of using them adjectivally is that of appending »ange% to them the relative participle of the verb to become (‘é4na,” Tam., jec»tive ‘ayana,’ Tel., or ¢4da,” Can.), without any change, whether structural th the or euphonie, in the nouns themselves. { pon 3. Many Tamil nouns ending in ‘§-u,” ‘d-u,’” ‘nd-u,’ or ‘®r-u,” double of the their final consonants when they are used as adjectives, or when case- ;Iuded signs are suffixed to them : e.g., compare nad-u,” Tam., the country, 0 of with ‘natt-u varakkam,” the custom of the country, or ¢ natt-il,’ in the or 'l'l'ca’ country. From the corresponding Telugu ‘nid-u,’ the country, is ferent formed ‘nati, of the country. In these instances the final consonant m & of the root is doubled and hardened (or in Telugu hardened only), for i ged the purpose of conveying the signification of an adjective: but in ffhe 2 THE NOUN. another class of instances the root 'remains unchanged, and it is the the pe monn consonant of the formative addition that is doubled. When Tamil nouns ending in the formative ‘mbu’ are used fi tirp (prope adjectivally, ‘mbu’ changes into “ppu;’ eg., from ‘irumbu,’ iron, and ‘kol, a rod, is formed ¢iruppu (k)kol, an iron rod. A similar (f-y: ‘ change sometimes takes place in Telugu, in which ‘inumu,’ ron, J 1nto 1 becomes inupa,’ e.g., ‘inupa pette, an iron bozx. are la Tamil nouns ending in the formative ‘ndu’ and ‘du’ change in the same manner to ‘ttu’ on being used as adjectives. E.g., compare e 18 ‘marundu,’ medicine, and ‘erudu, an oz, with ‘maruttu (p)pei,’ @ High medicine bag, and ‘eruttu (p)podi, an ox load. I of Iy Nouns ending in the formative ‘ngu’ do not, as might have been Both these Insta expected, change into ‘kku’ on becomming adjectives. classes of changes precisely resemble those which neuter or intransitive form verbs ending in ‘d-u’ or ‘r-u’ (or with the formative additions of fo th'( ‘mb-u; ‘ng-u,” ‘nd-u,” &ec.) undergo on becoming active or transitive, Telaf] disle and a similar principle is in each instance apparent in the change; for when nouns of quality are prefixed to other nouns adjectivally, there by the v is a transition of their signification to the nouns which they are fll Tan intended to qualify, which is analogous to the transition of the action feiple of a transitive verb to the object which it governs. (See the Section Idere on ‘Roots,” and also that on ¢ The Verb’). | Qf)‘ Il 4. Each of the inflexional increments, or petrified case-signs of the woul genitive, is used for the conversion of substantives into adjectives. Bt These are ‘in’ in Tamil and ‘ni’ in Telugu, ‘attu’ in Tamil and the In those instances in which ‘in 2 in Tamil I8 1 “ti” or ‘ti’ in Telugu. and ‘ni’ in Telugu are used as adjectival formatives, their use is a §)x | optional ; e.g., in Telugn we can say either ¢ tella,” white, or tella-nij § and in Tamil either ‘miral] skady (literally shade, a noun used adjeg adjectivally), or (but in the poetical dialect only) ¢niral-in.’ So also, | 186q g we may say either ‘mara (k)koppu,” Tam., the branch of a tree, or ey ‘mara-ttu (k)koppu.” In Tamil ‘am, an inflexional increment which b?C(!'/ilf is apparently equivalent to ‘in,’ is often used as an adjectival forma- the f The tive; e.g., ‘panan toppu’ (‘panei-am toppu’), a palmyra tope. Partie same formative is used in Malayalam also; e¢g., ‘malam pAmbu’ }sz’iw; (‘mala~am pambu’), @ rock snake. fhig Tt has been shown that the inflexions or inflexional augments, @ by N “attu’ and “ti,’ are in reality genitive or possessive case-signs ; and { that they are used to convert substantives into adjectives through the the P nltimate relation subsisting between genitives, e.g., of gold, and adjee- Part) tives, e.q., golden. In consequence of the frequency of tteir use in this Pl‘ete connexion, they have come to be appended even to adverbial forms for 1d ) FORMATION OF ADJECTIVES. 255 Is the the purpose of giving to them an adjectival meaning. Thus, from ‘monna,’ Tel., before, is formed the adjective ¢ monna-ti” (e.g., ‘monna- % used ti tirpu,’ the former decision’); and in Tamil from ¢vadakku,” north, , rom, (properly a dative) is formed the adjective ¢ vadakk’-att-u,’ northern similar (e.g9., ‘vadakkattiyan, a northern Tamilian’). ) A0, 5. Relative participles of verbs, and nouns of quality converted into relative participles by the addition of participial formatives, nge i are largely used as adjectives in all the Dravidian languages. Much OMpare use is made of relative participles as adjectives by the languages of High Asia also. el 0 It often happens that the same root is used, or at least is capable e been of being used, both as a verb and as a noun; and hence, in many h these instances of this kind in the Dravidian languages, two methods of nsitive forming adjectives are practicable, viz., either by prefixing the noun 1ons of to the substantive which we wish to qualify, or by using one of the nsifive, relative participles of the related and equivalent verb. The colloquial og; for dialect of the Tamil prefers the latter method: the former is preferred ; there by the poets on account of its greater simplicity and brevity. Thus, in Tamil either ‘uyar,’ height (adjectivally high), or the relative par- ey are action ticiple ¢ uyarnda,” Aigh, literally that was ligh (from ‘uyar’ con- Section sidered as a verb signifying to be high), may be used to expres high or lofty; eg., ‘uyar malei’ or ‘uyarnda malei,” @ loffy Aill: but ¢ uyar’ of the would be preferred in poetical compositions, whilst ¢uyarnda’ is afives better suited to prose and colloquial purposes, and is consequently il and the form which is commonly used by the Tamil people. This usage Tamil is not unknown to the Indo-European tongues also, but it constitutes a special characteristic of the Scythian group. 158 18 iy 6. The past verbal participle of Telugu verbs is sometimes used adjectivally in Telugu; hence when Sans. neuter nouns in ‘am’ are ] used used as adjectives ¢ ayi,” having become (the verbal participle), is often 0 also, peey OF annexed to them instead of ¢ayi-na’ (Tam. ‘4na,’ Can. ‘4da’), that became, that s (the relative participle). It is evident, therefore, that yhich the final ‘i’ of many Telugu adjectives is that by which the past forma- - The participles of verbs are formed; e.g., ‘kindi,’ low, from ‘kinda,’ adverb, below; e.g., ‘kindi illu,’ the lower part of the house. The addition of smba’ this <1 2 converts substantives also into adjectives; e.g., from kin-u, @ hump, is formed ¢ kini,’ hump-backed. :I]]eDtS; 7. A very numerous class of Dravidian adjectives is formed by ) and oh t he the addition to ecrude nouns of quality of the suffixes of the relative participles, more or less modified. adjee ‘uyarnda’ is a perfectly formed 15 preterite relative participle, comprising, in addition to the verbal root, 10 i o ‘nd,’ the sign of the preterite tense, and ¢ a,” the sign of the relative ; s 256 THE NOUN. and though the idea of time is in this connexion practically lost sight parici became of, yet that idea is included and expressed. On the other hand, in the class of words now to be considered, the signs of tense are modified always founs or rejected, to correspond with their use as adjectives, and the idea of time is entirely merged in that of relation. It is words of this class having which are commonly adduced by native grammarians as specimens of SiR qualitative words, or adjectives; and if the name can correctly be used any re at all in the Dréavidian family of tongues, it is to this class that it is aonistl applicable. I am convinced, however, that it is more correct to regard (reat, these words simply as relative participles; and I class them under they g this head, immediately after the investigation of the noun, because in * periy most instances, the root to which the relative signs are suffixed is used aljet by itself not as a verb, but only as a noun of quality or relation, or as and ar an appellative. / of {Gé’l‘ (1.) Many Tamil adjectives of this class are formed by the addition It of “iya’ to the root, e.g., ‘periya,’ great, ¢ siriya,” small. The roots of many t v) those words are ‘per-u’ and ‘sir-u; and as ‘u is merely a help to sive, o enunciation, I do not say that ‘u’ is changed into ‘i,” but prefer to say 9 “ I have no doubt that we shall be able that ‘iya’ is added to the root. of the to explain each part of this addition grammatically, without having of tengg recourse to arbitrary mutations. ‘iya’ (‘i-y-a’), is, I conceive, com- Tel. g pounded of ‘i,” a sign of the preterite tense, and ‘a,’ the sign of the dQTiVet relative participle. It has probably been originally softened from Tamy ‘ida,” the suffix of the preterite relative participle in ancient Canarese, but i to which ¢ina’ corresponds in colloquial Tamil. In Telugu, the past- dil‘ect e participle alone is often used adjectivally without the suffix of the Prege relative, as we have already seen; and the i’ with which that parti- % ciple terminates, explains the ‘i’ which precedes the final a’ of such Which Tamil adjectives as ‘peri-(y)-a.” ‘i’ is the sign of the verbal participle, i and the addition of “a’ or ¢ ya,” transforms it into a relative participle. éimply In classical Tamil compositions ‘iya’ is generally used instead of ‘ina,’ by i as the sign of the preterite relative participle of ordinary verbs; e.g., Tamil, ‘panniya,” instead of ‘pannina,’ that made. When the same suffix is Fl/)fr( added to a noun of quality like ¢per-u’ great, it converts it into a Ty ¢ relative participle, which, with the form of the preterite, contains in it iy f 80 no reference to time, and which may therefore be called an adjective. The suffix ‘iya,” being somewhat archaic, readily loses the idea of L time; whereas that idea is firmly retained by ‘ida,” ‘ina,’ and the Al ¢y other preterite relative suffixes which are in ordinary use. A good SOme { illustration of the adjectival use of ‘iya’ is furnished by the very roots of bej to which we have referred, viz., ¢ peru,’ great, ‘siru,” small. 100t y When these roots are regarded as verbs, their preterite relative Telat) thig ) | FORMATION OF ADJECTIVES. 257 t sight participles are ¢ perutta,’ that was or became great, ¢ sirutta,” that was o 1 the became small; in which participles the ideas of time and change are odified always included: whereas, when ¢ peru’ and ‘siru’ are regarded as e of nouns of quality, they are adapted for general use as adjectives by 18 class having ¢iya’ suffixed to them, eg., ¢periya, ¢siriya,’ (‘per-iya,’ mens of ¢sir’-iya.’) In this shape they mean simply great and small, without be used any reference to time; and in consequence of ¢ iya’ being so purely at 1t 18 aoristie, adjectives of this mode of formation are largely used. ‘periya,’ tegand great, ‘kodiya,” cruel, may properly be styled adjectives, seeing that 1 under they are used as such; but it is a mistake to regard ‘periya-(v)-an,” or alse 1 ‘ periya-n,” a great man, ‘kodiya-n, « cruel man, and similar words as s used adjectives. They are compounds of adjectives and suffixes of gender; 1, OF a8 and are properly appellative nouns, as has been shown under the head of ‘Gender,” and as appears from the manner in which they are used. ddition It is remarkable that ‘a” or ‘ia’ is post-fixed in Kole also to oots of many adjectives; and that the same participle is a sign of the posses- sive, as ‘a’ is in Dravidian. help to ' 0 say (2.) Some adjectives are formed by simply suffixing ¢a,’ the sign > e able of the relative participle, without the preterite ‘i, or any other sign having of tense whatever; e.¢., ‘nalla,” Tam. good ; ‘dodda,’ Can. great; ‘pedda,’ Tel. great. The examples here given may be, and doubtless are, i, coll- of the derived from preterite relative participles (‘nalla’* from the high 4 from Tamil ‘nalgiya’ and ‘dodda’ from the ancient Canarese ¢ doddida *); ATese, but in some instances, ‘a,’ the sign of the relative participle, is appended o past: directly Lo nouns, without borrowing any portion of the sign of the preterite. We have an instance of this even in colloquial Tamil, viz., of the ‘udeiya’ (“udei-(y)-a’), the ordinary colloquial suffix of the genitive, 11 Pﬂ,fﬁ' 3 of such which literally signifies that belongs to, that is the property of, from “udei,” property, to which ‘a,” the sign of the relative participle, is rticipley simply saffixed. This mode of forming adjectives from substantives iiple. of ingy by directly suffixing ‘a’ is very common in the classical dialect of the Tamil, especially in connexion with substantives ending in “ei’ or i 985 64 €.g., from ‘malei,’ « %ill, comes © malei-(y)-a,” adj., killy, or of a hill; auffi 18 o & from ‘sunei,’ a spring, comes ‘sunei-(y)-a,” that relates to a spring. So also from ‘ti, evil, is formed ‘ti-(y)-a,” adj., evil. The circumstance s dn it jectire idea of * “Nalla’ is generally considered to be a primitive word, and a bond fide adjective : but if ‘ketta,” bad, is admitted to be a relative participle, from nd the ‘ked-u,’ to become bad, it is reasonable to suppose that ‘nalla,’ good, has also some such origin. Accordingly we find a root, ‘nal,” goodness, which is capable A good of being used adjéctivally, and then signifies good; and connected with the same ry 1‘00{3 root we find also the verb malg-u,’ to be bountiful, to be good. The preterite relative participle of this verb is ‘nalgiya,’ that was, or s, bountiful ; and from this, I believe, that the much-used adjective ‘nalla,” good, has been derived. elaf"® S ; 9 () THE NOUN. that in most of these examples, the signification of the genitive is as (ua was natural as that of the adjective, shows how intimately the genitive and the adjective are allied. Nevertheless, as used in these examples, I gre regard ‘a’ as an adjectival termination, rather than as a sign of the genitive, and as acquiring this force from its being the sign of the Tar relative participle. Indeed I would define these qualitative words pa (‘malei-(y)-a,” &c.) to be the relative participles of ¢appellative T6g1 verbs.’ See that class of words investigated in the section on ¢The 181 Verb. This usage, perhaps, explains the origin of the Tamil adjectives by 7 ) | ‘pala, many, and ‘sila,” few, viz., from the roots ¢pal’ and ‘sil bee (which are used in their crude state in the poets), and ‘a,’ the sign of Jen the relative participle. It is true that these words are also regarded uy as neuters plural,—and that in some instances they are correctly so higf regarded appears from the phrase ¢ palavin’ (¢ pala-v-in’) ¢pél, the oon Tamil designation of the neuter plural, literally the gender of the many ang (things). But when we look also at such phrases as ¢pala arasar, and many kings—phrases of constant occurrence, not only in the collo- quial dialect, but in the classics—it is evident that the ‘a’ of this ha\'f latter ¢ pala’ is used, not as a suffix of the neuter plural, or as a sign die of plurality of any sort, but as a sign of the relative participle, by low the use of which ¢ pal-a’ becomes an adjective. fron (3.) Many adjectives of this class are formed by the addition to hay nouns of quality of the sign of the relative participle of the future or 400 aorist, which is ‘um’ in Tamil, e.g., ¢ perum,” great. Beschi supposes ap this adjective to be derived from the abstract noun °perumei,” great- 3 g ness, by the rejection of the final ‘ei; and to all other adjectives of be this class he attributes a similar origin. ‘mei,” however, not ‘ei,” is ffe the suffix by which abstract nouns are formed (vide the section on ¢ The It is much better to befy Verb’), and as such it is one and indivisible. itig derive ‘perum’ from ¢per’,} the un-euphonised form of the root verk ‘peru,’ greatness, great, and ‘um,’ the ordinary relative participle of Tan the aorist; in the same manner as ‘ periya’ has been seen to be derived 2 and from ¢ per’ and ‘iya,’ the relative participle of the preterite. ‘um is ordinarily called the relative participle of the future, but this future will be shown, in the section on ‘The Verb,” to be properly an aorist, g and as such to be used very indeterminately, with respect to time. tivg ¢ Vipnil minang-um &udar,” Tam., means, not tkhe stars that will shine for in the sky, but the stars that shine in the sky, this tense being especially ag fitted to denote continued existence : and in consequence of this loose- Ing ness of reference to time, ‘um,’ the sign of the relative participle of W this tense is better fitted even than ‘iya’ to be suffixed to nouns of ad) FORMATION OF ADJECTIVES J 9 ) 18 48 quality, as an adjectival formative. Hence ‘perum,’ literally that s veand was or will be great, is a more expressive and more classical word for ples, 1 great than ¢ periya.’ of the It has already been shown, in the section on ¢ Sounds,” that ‘peim,’ of the Tam., green, is not a distinet form of adjective, but is softened from words ‘pasum’ (‘payum’) by a dialectic rule, whilst ¢ pagum’ is derived llative regularly from ‘pas-u,’ greenness, green, and ‘um,” the particle which 1 “The is now under consideration. 7. Dravidian nouns of every description may be used adjectivally octives by appending to them the relative participles of the verb signifying to d ‘sl become, which are in Tamil ‘4na’ and ‘agum’ (also ‘ulla,” an equiva- sign of lent word), in Telugn ‘agu’ and ‘ayana,” in Canarese ¢ ada I’ e.g., ggarded “uyarvana’ (‘uyarv’-ana’), Tam., lofty, literally that was or has become atly s0 ligh or a height. This mode of forming adjectives is especially used in 1 the connexion with Sanscrit derivatives, on account of their greater length 0 many and foreign origin. Such adjectives, however, are phrases not words; arasar,’ and they are incorrectly classed amongst adjectives by Europeans who - collo- have treated of Dravidian grammar. of this I may here also again remark, that certain words have been styled a sign adjectives by some European writers, which in reality are appellative ple, by nouns, not adjectives, and which ascquire the force of adjectives merely from the addition of the relative participles of the verb ¢o become, which fion to have been referred to above. Thus, the Tamil words ‘nallavan,’ « ture O good (man), ‘nallaval,” a good (weman), ‘nalladu,” @ good (thing), are pposes appellative nouns formed by the suffix to a noun of quality of the for- matives of the three genders; and the addition of ‘ana,” that has gre’r.‘tt- ves of become, to any of these words, though it constitutes them adjectives in ‘ei,’ s effect, leaves them in grammatical form precisely what they were before. Bonws may either qualify another noun, e.g., bonus vir, when ] ¢ The fer 10 it is an adjective, or it may stand alone and act as nominative to a ¢ 1006 verb, when it is a qualitative noun, e.q., bonus virtutem amat. The Tamil ‘nallavan,’ @ good (man), can only be used in the latter sense, lple Of rived and therefore is not an adjective at all. !’ Comparison of adjectives,—In all the Dravidian dialects comparison future aotisty is effected, not as in the Indo-European family by means of compara- fime. tive or superlative particles suffixed to, and combined with, the positive form of the adjective, but by a method closely resembling that in which | shine ially adjectives are compared in the Semitic languages, or by the simpler Joose: means which are generally used in the languages of the Scythian group. When the first of these methods is adopted, the noun of quality or plef ¢ adjective to be compared is placed in the nominative, and the noun or uns of s 2 260 THE NOUN. nouns with which it is to be compared are put in the locative and pre- compa fixed. It is generally stated in Tamil grammars that it is the ablative phrast of motion which is thus used, but I am persuaded that even when the follow case-sign is that of the ablative of motion, the signification is purely fals) that of the locative; and that in Tamil ‘il and ‘in,’ have in this Intens connexion the meaning of in (i.e., are locatives), rather than that of is pre Jrom : e.g., ‘avattr’-il idu nalladu,’ Tam., this s Obetter than those, in (i literally in those things this s good. It The conjunctive particle ‘um,” and, even, is often added, especially Super in the colloquial dialect, as an intensitive, e g., ‘avattr’-il-um idu differ nalladu,” Tam., t&is is better than those, literally even-in-those this is good. Buro Sometimes the noun with which comparison is to be made is put in the comp: dative instead of the locative. Sometimes, again, comparison is effected yenie by means of an auxiliary verb. The noun with which comparison is doned to be made is puat in the accusative ; it is followed and governed by point the subjunctive or infinitive of a verb signifying to see, to show, or to accor leave ; and the phrase is concluded by the subject of the preposition, Patis with the adjective to be compared. Thus, in Tamil we may say R ‘adei-(p)parkkilum idu nalladu,” literally even though looking at that m T this (is) good, or ‘adei vida idu nalladu,’ quetéing that this (is) good, 1.e., adher this s so good as to induce one to abandon that. Such modes of comparison, however, are stiff, cambrous, and little used except by Europeans; and in the Dravidian dialects, as in those I that of the Scythian group, direct comparison of one thing with another is wff ordinarily left to be understood, not expressed. The effect which is the aimed at is secured in a very simple manner by prefixing to the posi- fory tive form of the adjective some word signifying muck or wery, or by Stoo 3 appending to the subject of the proposition a sign of emphasis, or a a Jp word signifying indeed, e.g., ‘id-¢’ (or ‘idu tan’) nalladu,” Tam., THIS 8 th wndeed is good. to be In Telugu and Canarese the conjunctive particles ‘nu’ and ¢4’ are not necessarily required to help forward the former method of com- parison, like the Tamil ¢ am ;’ nor is this particle generally used in the higher dialect of the Tamil itself. The Canarese makes use also of the particles ¢ anta’ and ‘inta,” ‘antalu’ and ‘intalu’ (which, in their origin, are compounds of locatives and demonstratives), to assist in effecting comparison. In all these dialects the superlative is generally expressed by means A of prefixed adverbs signifying muck or wvery, or by the primitive Scythian plan of doubling of the adjective itself, e.g., ‘periya-periya,’ very great, literally great-great. 1f greater explicitness is required, the method by which it is effected is that of putting the objects with which COMPARISON OF ADJECTIVES. 261 L pre comparison is made in the plural and in the locative-case. Thus the lati \( phrase, the tiger s the fiercest animal, would be expressed in Tamil as on the follows,—¢ vilangugalil puli kodidu,” amongst animals (literally in amni- pute ly mals) the tiger is cruel. Sometimes, for the purpose of increasing the n thig intensity of the superlative signification the adjectival noun ¢ella, al/, hat of is prefixed to the plural noun which denotes the objects compared, ¢.g., th 03¢ in (t.e., amongst) all animals the tiger vs cruel. It is evident that the modes of forming the comparative and 01 lly superlative degrees of adjectives which have now been described, m idu characterize the Indo- differ greatly and essentially from those which (004 European family of tongues. If Drividian adjectives had ever been i the compared like those of the Sanscrit, it is inconceivable that so con- focted venient and expressive a plan should so completely have been aban- SO 1 doned. The Dravidian modes of comparison agree, up to a certain 1ed by point, with those of the Semitic tongues ; but they are in most perfect , 00 10 accordance with the Turkish method, and with the modes of com- gition, parison which are employed in the languages of Tartary generally. Robert de Nobilis and the Jesuit writers endeavoured to naturalise y sy ¢ that in Tamil the Sanscrit superlative particle ‘tama; but the Tamil 1, 16 adhered resolutely to its own idiom, and the attempt failed. ittle PrepositioNs or Post-PosiTions. — It has already been stated ‘When those that all the Dravidian post-positions are, or have been, nouns. et 18 suffixed to other nouns as post-positions, they are supposed to be in ich 18 the locative case ; but they are generally suffixed in their un-inflected form, or in the nominative ; and the locative case-sign, though under- ) ptM or by stood, is rarely expressed. It seems quite unnecessary to enter into o @ an investigation of the post-positions in a work of this kind, inasmuch THIS as they are sufficiently explained in the ordinary grammars, and are ) to be regarded simply as nouns of relation. K f com- 1 in the of the ; fheir i st 10 means miti\’e SI’i}'ﬁz’ 3(1, tlle Whi(’ll Th found numer of nur adject ments. langu: SECTION 1V, the n T the 1 hund, THE NUMERALS. of the which In the Dravidian languages, each of the cardinal numbers presents briefes itself to us in a two-fold shape. The first and more primitive form is It that of numeral adjectives : the second and more largely used is that of neuter nouns of number. appel The numeral adverbs (‘twice,” ‘thrice, two p &ec.) and also the distributive numerals (‘ by twos,” ‘by threes,” &c.) numb are formed from the nameral adjectives; whilst the ordinal numbers the ug (“second,” ‘third,” &ec.) are formed from the abstract numeral nouns. colloq In the colloquial dialects the neuter nouns of number are often | forme used, without change, as numeral adjectives: e.g., in Tamil, we may ‘oruy say ‘irandu pér,’ two persons; though ¢ iru pér,” or the still more “ling classical appellative noun, ¢iruvar, might have been expected to be but used. H This use of the noun of number instead of the numeral fives adjective is not ungrammatical ; but is in accordance with the charac- ] teristic Dravidian rule that every noun of quality or relation, though e in itself neuter and abstract, becomes an adjective by being prefixed of th to a substantive noun in direct apposition. The numeral noun “ondru,” Tam., ¢okati,” Tel., one, is the only numeral which is never ( used in this manner even in the colloquial dialects ; the adjectival Deiy numerals, ‘oru,” ‘ oka,” &c., being invariably prefixed to substantive ; allied nouns as numeral adjectives: the same forms are employed also as indefinite articles. fhe [ The abstract or neuter nouns of number are sometimes elegantly g post-fixed, instead of being pre-fixed to the substantive nouns which Cong they are intended to qualify. Z.g., instead of ‘nal’ erudu,” Tam., sg Jour oxen, we may say, not only ‘néng’ erudu’ (using the noun of Do number ‘nangu,” instead of the numeral adjective ‘nalu’), but also (o “erudu nangu;’ a phrase which literally means a quarternion of oxen. tiy, This phrase affords an illustration of the statement that the Dravidian iy nouns of number are properly abstract neuters. 1) ONE. 263 The primitive radical forms of the Dravidian numerals, will be In investigating the found to be those of the numeral adjectives. numerals one by one, it will be seen that the neuter or abstract nouns of number have been formed from the shorter and simpler numeral adjectives by the addition of neuter formatives and euphonic incre- ments. It is, therefore, the numeral adjectives of the Dravidian languages, not their numeral nouns, which are to be compared with the numerals of other families of languages. The compound numbers between ‘ten’ and ¢ twenty,” and especially the higher compounds (‘twenty,” ¢thirty, ‘two hundred,’ ‘three hundred,” &e.), afford much help towards ascertaining the oldest forms of the Dravidian numeral roots; seeing that the numeral adjectives which are employed in those compounds exhibit the numerals in their sents briefest, purest, and most ancient shape. I 1§ It is the adjectival form of the numerals which is used in forming that appellative nouns of number, such as ‘iruvar’ (‘iru-(v)-ar’), Tam., irice; two persons. The basis of this word is not ‘irandu,” the noun of number wo, but the numeral adjective ‘iru,’ with the addition of ‘ar, ‘&C.} ibers the usual suffix of the epicene or masculine-feminine plural. In the NS colloquial dialects, adjectival or appellative nouns of number are formed in this manner from the first three numeral adjectivesalone; e.g., often ¢oruvan, Tam., one person, (mase.), ‘unus,;” ‘orutti,’ one person (fem.), | may \e ‘una; Ciruvar, two persons; ‘mivar, three persons (both epicene ) more but in the higher or poetical dialects, almost all the numeral adjec- to be tives are converted in this manner into appellative nouns. neral From these circumstances it is evident that the Dravidian numeral arac- ough adjectives are to be regarded as the only essential portion of the roots of the cardinal numbers, and probably as the very roots themselves. fived nous ? One.—Two forms of the cardinal numeral ‘one’ are found in the ey er Dravidian languages, which will appear, I think, to be remotely fival allied. The first, < oru, is that which is used in all the dialects except ntive the Telugu ; the latter, ‘ oka,’ is used in the Telugu alone. 50 a5 (1.) The basis of the first and most commonly used form of this numeral is ‘or, to which ‘u’ is added for euphonisation; and this antly constitutes the numeral adjective ¢ one,” in all the dialects which make thich use of this base. ¢or-u, in colloquial Tamil, becomes ¢06r’ in the [am,y poetical dialect ; the essential vowel ‘o’ being lengthened to €6’ to i Of compensate for the rejection of the eaphonic addition ‘u.” The adjec- also tival form used in Tulu is * ori, in Ku ¢ra; with which the Behistun 00N The Canarese numeral adjective ¢irra’ or ‘ra’ may be compared. dian numeral adjective is identical with the Tamil, though its true cha- | 264 THE NUMERALS. racter 1s somewhat concealed. antlo Instead of ‘oruvan,’ Tam., ¢ unus,’ the Jias ] Canarese has ‘obban-u’ (or-b-an’), and instead of “oruval,” ‘una,’ ‘obbal-u’ (¢ or-b-al’). The ancient Canarese, however, uses ¢ 6rvyam’ ‘il for the former, and ‘6rval’ for the latter; the base of which, ¢ 6r, is Can,, the numeral root, and is identieal with the Tamil or-u’ or ¢ 6r. “kan The abstract neuter noun ‘one,” meaning literally one thing, or “an Tam unity, is in Canarese ‘ondu; in grammatical Tamil ‘onru’ (pro- nounced ‘ondru’ or ‘ondu,” and in vulgar Tamil ¢ onnu) ; in Malaya- and lam ‘onna; in Tulu ‘onji; in Génd undi ; in Tuda ¢vodda, ¢od oL softe ¢ood,’” or ‘vood; in Urfon “tinta.’ perc ‘or” being the adjectival form of this numeral, it claims by rule L - T R to be the representative of the erude root, as well as the basis of the the abstract or neuter nouns of number signifying ‘one’ or *unity,” which anee e i are used in the various dialects. It remains to be seen whether the I - derivation of each of those nouns of number from ¢ or’ can be clearly aboy made out. of ¢ At first sight the Canarese ‘ondu,’ and especially the Malayala WI‘:,’ ‘onna,’ appear to resemble the most common form of the Indo- our ] i European numeral ‘one, which is in Latin ‘un-us’ (in an older form hard “oin-0s’), in Greek ¢év,” in Gothic ‘ain’s.’ In the Koibal, a Samoiede empl dialect, there is a similar word for ‘one,’ viz., ‘unem: and we find in e % the Tungusian ¢ um,’” in the Marchu ‘emu.” Even in Sanscrit, though T ‘€ka’ is invariably used for one, a form has been mnoticed which Proc ( appears to be allied to the first numeral of the Western languages, viz., OR ‘lina-s,” less, which is prefixed to some of the higher numerals to Was express diminution by one (e.g., ¢ tnavinshati, nineteen), like the cor- nt responding prefix ‘un’ in the Latin ‘undeviginti’ It would be an interesting circumstance if the Malayéla ¢ onn-a’ and the Latin un-us’ “on were found to be allied; but the resemblance is altogether illusory, i% % and vanishes on the derivation of ‘onna’ from ‘or’ being proved. Iy It is reasonable to suppose that the numeral adjective of the Tamil, fory “oru,’ and its numeral noun ‘onru,’ must somehow be related. Now, Ty - whilst it is impossible on Dravidian principles to derive ‘oru’ from ot} “onru,’ it will be shown that the derivation of ‘onru’ from ‘oru’ is E the in perfect accordance with Dravidian rules: and if the Malayala {he ‘onna’ be simply an euphonised form of the Tamil ‘onru,’ as it | certainly is, every idea of the existence of a connection between any i n of these forms and the Latin ‘un-us’ must be abandoned. - of It was shown in the section on ‘Sounds’ that the Dravidian of languages delight to euphonise certain consonants by prefixing nasals 1ig to them. If the ‘r’ of ‘oru’ is found to have been converted in this manner into ‘nr,’ the point under discussion will be settled, What 1§ B E— ONE. 26 (9] P - i) the analogy, then, is there for this conversion ? “muru,” Canarese, three, “una, has through this very process become in Tamil, ‘minru’ (pronounced vyam' ‘mlndru, ‘mindu, or ‘minu’); in Malayilam, ¢minna. ¢karu,’ o, Is Can., a calf, becomes in Tamil, ‘kanru’ (pronounced ‘kandru’ or ‘kandu,’ and vulgarised in colloquial Tamil into ¢ kannu’); in Malayala ¢ kanna.’ 1, of Again, ‘kiru,’ the verbal suffix denoting present time in ' (pro- Tamil, has become in the poetical dialect ‘kinru,” pronounced ‘kindru ; alayd- and this, in the Malayalam present tense, is found to be still further In all these instances we ‘a {Od; softened into ‘kunnu,” and even ¢unnu. perceive that very euphonic alteration by which ‘oru’ has become y Tule progressively ‘onru,’ ¢ ondru,” ‘ondu,’ ‘onnu,” and ‘onna;’ and thus of the the derivation of ‘onna’ from ‘oru’ is found to be strictly in accord- which ance with analogy. er the It may be objected that the illustrations which have been given above exhibit a change of the hard ‘R’ into ‘ndr, whereas the ‘r’ Jearly of ‘oru’ is the soft medial ; and that, therefore, the analogy, though very remarkable, is not complete. I answer that, though the ‘r’ of .13)“&1& Indo- our present Tamil ‘oru’ is certainly the medial semi-vowel, not the » form hard ‘w,’ yet originally the hard ‘R’ must have been the very ‘r’ oiede employed. This appears from the Tamil adjective, odd, single. That nd In adjective is ‘orrei’ (pronounced ¢ottrei’); and it is derived from the numeral adjective, one. It has been derived, however, by the usual hougl which process of doubling the final consonant, not from ¢or-u, but from ‘¢ or,—evidently a more ancient form of the word, in which the ‘r’ | Vil ils to was the hard, rough ¢R,’—that very ‘r’ which is usually enphonised 0 001 into ¢ndr. be an It appears, therefore, that the origin which I have ascribed to iln-US’ é “onru’ is in complete accordance with analogy. Moreover, if the ‘n’ of ‘ondru,’ ‘ondu,” or ‘onna,” were part of the root of this 18015, od numeral, the ‘du’ which is suffixed to it could only be a neuter amil, formative ; and in that event ‘on’ should be found to be used as the ¢ Now, numeral adjective. on,” however, is nowhere so used ; and therefore from both the use of ‘or-u,” instead of ‘on,” as the numeral adjective, and 117 I8 the existence of the derivative ‘ or(Rr)ei’ (‘ ottr-ei’), single, prove that the root of this numeral must have been ‘or,” not ¢on.’ ayé}a a8 it Though ¢or,” in its primitive, unnasalised shape, is not now found in the cultivated Dravidian dialects as the first abstract neuter noun o ) any of number for one, or unity ; yet it appears in one of the ruder dialects of the family, viz, in the Rajmahal or Male; in which the numeral idian msals noun one is ‘art,’ or ¢ort, which is evidenly formed directly from ¢ Y] ar or ‘or. If it is true, as has been asserted that the Male art’ ) thiﬁ is appropriated to human beings, it must be identical with the Tamil Vhat 266 THE NUMERALS. ‘orutt-an,’ one man, ‘orutt-i,” one woman; the ‘tt’ of which is a for- 10t Mf mative, and is derived from the pronoun of the third person. See ¢ The ! Cl\d Noun.’ MM Compare also this form with the Brahui ¢asit,” one, of which Dravic “as,” the crude root, seems to bear as close an analogy to ‘or-u’ as Th ‘mus,’ the crude root of ¢ musit, the Brahui for #hree, undoubtedly does to the Canarese ¢ miir-u.’ If in the latter case the ‘s’ and ‘r’ i 01, are mutually convertible, it cannot be considered improbable that ally (3 ‘asit’ and ‘art,} and consequently ‘as’ and ‘or, bear a similar rami relation one to the other. This (2.) The basis of the Telugu numeral signifying one seems to be Lol essentially different from that which is used in the other Dravidian ot dialects. There is nothing extraordinary in the discovery in any 0ne-¢) language or family of languages of two roots for one. This would follow naturally arise from the very concrete character of this numeral, and abbre the variety of uses to which it is put. Even in Sanserit we find both (a ‘éka’ and ‘prathama.’ Two also is represented in Latin by ¢duo, doub “ambo,” and ¢secundus.’ some The Telugu neuter noun of number for one is © okati, literally one Telug thing, of which the adjectival form is oka.’ “okati’ is formed from poun ‘oka’ by the addition of the neuter and inflexional formative, “ti:’ otie Conng and by annexing the usual masculine and feminine suffixes, the Telugus form ¢okandu’ or ¢okadu,” one man, and ¢ okate,” one woman. inth ‘oka’ being found to be the crude root of this numeral, we have now um to inquire into its affinities. add Is the Telugu ‘oka’ derived, as has sometimes been supposed, e from the Sanscrit ¢éka, one ? It seems not improbable that the Leen Telugu word has some ulterior connexion with the Sanscrit one, to (ar which it bears so great a resemblance: but it is impossible to suppose the it to have been directly derived from the Sanscrit, like the Bengali may ok, or even the Persian ‘yak;’ for the Telugu has borrowed and tion ) occasionally uses the Sanserit numeral ¢éka,” in addition to its own the ‘oka;’ and it never confounds ‘oka’ with ‘éka, which Telugu off | It will grammarians regard as altogether independent one of another. | ‘k be seen also that words closely analogous to ‘oka’ are used in the ef whole of the Finnish languages, by which they cannot be supposed to have been borrowed from the Sanscrit. Thus the numeral one is in F, iy Wotiak ‘og, ‘odyg; in Samoiede ¢okur,” ‘ockur,” ‘ookur;’ in 2 A Vogoul ‘ak,” ‘aku; in Magyar ‘egy; in Lappish ¢akt;’ in Fin- 1¢ nish ‘yxi’ (‘yk-si);’ in Tcheremiss ik, ¢iktd ; in the Scythian of ih ok the Behistun tablets ¢ In the Sub-Himalayan languages, we find i “ako’ in Miri, ¢ akhet” in Naga, and ‘katka’ in Kuki. o These remarkable analogies to the Telugu < oka’ prove that it has an ONE. 267 s a for not necessarily, or even probably, been derived from the Sanscrit 0 ‘The ‘gka: 7 and if the two roots are allied, as they appear to be, it must which be in consequence of the relation of both the Sanscrit an‘cul the Lappo- M as Dravidian families to an earlier form of speech. ibtedly The Tamil infinitive ¢ okka,” which is used adverbially to mean md ‘r’ in one, all together, and which forms the ordinary Malayala word for lo that all, (with which compare Mordwin ¢ wok,” all), is supposed by the similar grammarians to be derived from the obsolete verbal root ¢o,’ to be one. This root ‘o’ is sometimes used adjectivally in Canarese instead of s to be ¢or-u, in which case it doubles the succeeding consonant; e.g., avidian ‘ottaleyu’ (‘o-(t)tale’), one head; ¢okkannanu’ (‘o-(k)kannan’), in any one-eyed man. It is evident from this, that o was originally would followed by a consonant; and that it must have been derived by al, and abbreviation either from the Tamil € or,” or the Telugu ok’ Can ‘oka’ and ‘or d both be in any way allied? It appears very doubtful whether there is any relation between them ; and yet duo, some few traces of affinity may be discovered. On examining the lly one Telugu word for eleven, ¢ padakondu,” the latter part of this com- | from pound numeral presents some peculiarities which are deserving of notice. } tl We should have expected to find ¢okati’ used for one in this connexion ; instead of which we find ¢ kondu,” a form which is found 5, the ) 0N in this compound alone. ‘kondu’ is here used as a neuter noun of 6 NOW number, and like all such nouns is formed from a crude base by the addition of some formative. If the ‘k’ is euphonic and intended to >posed, prevent hiatus, like the ‘h’ of ‘padihédu’ (¢ padi-(h)-édu,”) seven- teen, ‘kondu’ is identical with ¢k-ondu,” and ‘ondn’ is allied to the Al the Canarese ‘ondu,’” from the root ‘or:’ but if, as appears more likely, ne, 10 ,Ippose the ¢k’ is radical, the erude adjectival form from which it was derived may have been ‘kor: and if we are at liberty to adopt this supposi- engali | and tion, we may at once conclude that ‘kor’ was the original form of the Tamil-Canarese ‘or;’ for the initial ‘k’ might easily be softened oWl off (and there are several instances of the disappearance of an initial elugy t will i “k’), whilst it could not have been prefixed to or,” if it had not stood before it originally. n the Supposing ‘kor’ to be an older form of ¢or, it is not diffienlt to sed to s dn suppose ‘kor’ and ‘oka’ to be allied, by the corruption of both from a common root. If the old Scythian word for one was ‘okor, cor- ! }Fin- responding to the Samoiede °okur,’ both ‘kor’ and ¢or,” and also ‘oka,” would naturally be derived from it. A change analogous to this il Of appears in the Behistun tablets, in which we find that the numeral for 0 ﬁntl one which is used in the oldest extant specimen of the language of the ancient Scythians was ‘kir’ (corresponding to our supposititious t b 268 THE NUMERALS. of ‘in, Telugu ‘kor’), and that the numeral adjective derived from it was “irra’ or ‘ra.’ Here we have an ancient and authentic illustration of its especi both of the existence of a word for one containing both ‘k’ and ‘r,’ and of a derived persons numeral adjective from which the ‘k’ had been softened off: and and th it deserves special notice that ‘ra, the Behistun Tundre numeral adjective, is identical with ‘ra,” the numeral adjective of the ‘INUR Ku, a Dravidian dialect. In the Turkish, one is represented by of ‘i ¢ bir,” which seems to be allied rather to the Persian ‘bar’ in ¢béri,’ “irabg once (and ulteriorly to the Sanscrit ¢ var,’ zime), than to the Tamil ‘or.’ In The Caucasian numerals for one exhibit a closer resemblance to the instea Dravidian, viz., Lazian ¢ar, Mingrelian ¢arti,” Georgian °erthi ; stands and it may be noticed that as in the Dravidian ¢or, one, and ¢ir,” two, the ¢ so in those Caucasian dialects, ¢ v’ forms an essential part of both those numerals. adject Tamil Dravidian indefinite article—The Dravidian numeral adjectives, CDI]]pO ‘oru’ and ‘oka,” are used like similar numerals in most languages, as facilit a sort of indefinite article. The Turkish uses ¢bir,” one, in a similar b, ¢ furthe manner; and a corresponding usage prevails in the modern European charg languages, as well as in the colloquial dialects of Northern India. tirenn The only thing which may be considered as distinctive or peculiar Tela in the use of the Dravidian numeral adjective one, as an indefinite g ever, article, is the circumstance that it is not used in the loose general way diale in which in English we speak of a man, or a ¢ree, but only in those and cases in which the singularity of the object requires to be emphasized, Can when it takes the meaning of a certain man, a particular kind of tree, of { or a single tree. Europeans in speaking the native languages make in general too large and indiscriminate a use of this prefixed numeral, begn forgetting that the Dravidian neuter noun, without prefix or addition, Pats becomes singular or plural, definite or indefinite, according as the | connexion requires. i s L haye T Two.—The abstract or neuter noun of number signifying two, or Orig] duality, is in Canarese “eradu,” in Tamil ¢irandu,” in Telugu ¢rendu,’ Tan in Tulu ¢erad-u,” ‘raddha,” or ‘randu; in Malayilam ‘renda; in The Gond ‘rend’ or [‘ranu;’ in Seoni Génd ‘rund; in Tuda ‘aed’ or Tay ‘yeda.’ The Singhalese word for double is ‘iruntata.’ In all the earl Dravidian® dialects ‘the corresponding numeral adjective is ‘ir,” with heg such minor modifications® only as euphony dictates. This numeral 1§ | U3y adjective is in;Tamil ‘iru;’ in the higher dialect ‘ir, the increase in the quantity of the radical ‘i compensating for the rejection of the eup final euphonic ‘u.’ The ‘r’ which constitutes the radical qonsonant (du — TWO. 269 1t was of ‘ir, is the soft medial semi-vowel ; and it evinces in consequence ustration of its softness a tendency to coalesce with the succeeding consonant, ) especially in Canarese and Telugu. Thus, for ‘iruvar,” Tam., two and y, persons, the modern Canarese uses ‘ibbar-u’ (ancient dialect ¢irvar’), ad been Behistun and the Telugu ‘iddar-u.’ Instead, also, of the correct ¢iruntru,’ two hundred, of the Tamil, both the Telugu and the Canarese have ve of the ‘inntru; and the Canarese word for twenty is ¢ippattu,” instead ted by of ‘irupattu,” which would be in correspondence with the Tamil n ‘b, ¢jrubadu’ and the Telugu ¢iruvei.’ mil “or. < b4 In the Canarese neuter noun of number ¢eradu,’ two, € is used e to the instead of i’ as the initial vowel; but in this point the Canarese “orthi stands alone, and in all the compound numerals, even in the Canareses “Ir, o, the ‘i’ re-appears. Were it not for the existence of the numeral oth those 4 adjective ‘ir-u’ or ir, we might naturally suppose the ‘i’ of the Tamil ‘irandu’ and of the obsolete Canarese ‘iradu’ to be, not a component element of the root, but an euphonic prefix, intended to Jectives, ¢ facilitate pronunciation. i’ is very commonly so prefixed in Tamil; lages, 3 ? Qigis. S 14, night (from the Sanscrit €ratri’), becomes ¢ira,’ and by a similar further change ¢iravu.” This supposition with respect to the euphonic 9 h'u'opean character of the ‘1 of ¢irandu,” might appear to be confirmed by the lia, circumstance that it disappears altogether from the numeral nouns of the peculiar Telugu, the Malayalam, and several other dialects. The existence, how- definite ever, of the numeral adjective ¢iru’ or ‘ir,” in every one of the Dravidian mal way dialects, and its use in all the compound numbers (such as ¢ twenty’ 1 those and ‘two hundred’), suffice to prove that the ‘i’ of the Tamil- hasized, ; Canarese numeral noun ‘iradu’ is not merely euphonie, but is a part ) of e % of the root itself, and that ‘iradu,” the neuter noun of number, has nake 10 A com- been formed from ‘ir’ by the addition of a formative suffix. umeral, parison of the various forms shows clearly that ¢ir, euphonised into ddition, ‘iru, was the primitive form of the numeral adjective fwo: and we 8 the have now only to inquire into the characteristics of the numeral noun. The Canarese ‘eradu’ (or rather ‘iradu,” as it must have been originally) is the earliest extant form of the noun of number. The funo, 0F Tamil is ¢irandu, ¢d’ having been euphonically changed to ‘nd. ] ref"(.iﬂ,y Though there is a nasal in the Tamil word which is now in use, the 0a; i Tamil noun-adjective double bears witness to the existence of an ned’ o earlier form, which was destitute of the nasal, and which must have {1” ﬂ]e been identical with the Canarese. The Tam. word ¢iratt-u, double, D, with is formed directly from ¢irad-u, by the doubling of the ‘d as is et usually done when a noun is converted into an adjective; and the rease 1 euphonic change of ‘dd’ into ‘tt’ is according to rule. ‘du’ or of fHe ‘du’ is a very common termination of neuter nouns, especially of I 5000 2 Ll ( 0 THE NUMERALS. fsit, appellative neuters, in all the Dravidian languages. Thus, from the root ‘kira,” Tam., old, is formed ‘kiradu, that which s old. The howev ‘n’ which is inserted before ¢d’ in the Tamil ¢irandu’ is evidently initial euphonic, and is in perfect accordance with the ordinary phonetie appeat usages of the Dravidian languages. In Telugu every word ending in Magy: shown ‘du,’ receives in pronunciation an obscure nasal, whether it has a place Soythi in the written language or not; and there are many instances in Tamil also of the insertion of this nasal before a final ‘du’ for the (f(fr ‘] {) sake of euphonisation, when it is quite certain that there was no such mi ) nasal originally in the word in which it is found : e.g., ‘4ndu,’ there, vidias ‘indu,’ here, and ¢ yandu,” where ?’ are euphonised forms of ¢Aadu, by ‘k can b ¢idu,” and °yadu’ Compare also ¢ karandi,” @ spoon, Tamil, with the more primitive Telugu ¢ garite.’ The Tamil noun of number, signify- 1 ing two, must, therefore, have been ‘iradu’ originally. In the Gond T | ‘ranu,’ the ‘d’ of ‘irandu’ has disappeared altogether ; a change i Ca which is in accordance with the Malayala corruption of ¢ ondu,’ one, by Pl ‘miln into ‘onna.’ The Uréon word for two, ‘enotan,’ is probably Dravi- dian. Tn Urdon ‘otan’ (from the Hindi ¢ gotan’) is a suffix of each ﬁlmp of the first three numerals ; consequently ‘en’ is to beregarded as the il] [r‘ 2 Uréon root, and this seems to be analogous to the Dravidian ‘er 1 There are no analogies to ‘ir,” fwo, in any of the Indo-European fhirty The languages, and I am doubtful whether any real analogies to it are aoun discoverable even in the Scythian group, except perhaps in the Caucasian. e ‘my The Brahui vindicates its claim to be regarded as in part Drivi- Whi dian, or at least as the inheritor of an ancient Drividian element, by §6e and the close affinity of its second and third numerals to those of the Dra- In Brahui ¢wo is ‘irat ; and when this word is I vidian tongues. compared with the Brahui ¢asit,” one, and ‘musit,’ three, it is evident I 9 that in each of these instances the final ‘it or ‘at, is a formative ey suffix which has been appended to the root. Consequently ©ir,” the Sime ‘ root of ¢ir-at,’ is absolutely identical with the Dravidian ©ir Even 1o the Brahui formative evinces Dravidian affinities ; e.g., compare ¢irat’ with the Canarese noun of number ‘eradu,’” and especially with the oltey Tamil derivative ¢iratt-u,” double. ey The nearest analogies to the Dravidian ¢ir’ which I have noticed y lllfn in other families of tongues, are in the Cauncasian dialects; e.g., in the e\‘im Georgian “ori; in the Suanian (a dialect of the Georgian) “eru’ or 9 hag ‘ieru; in the Lazian ‘zur; thy and in the Mingrelian ¢ shiri: compare also the Armenian ergov. n In the Samoiede family of tongues, several words are found which bear at first sight some resemblance to the Dravidian ¢ir.’ These are fo THREE. 271 4 om the sit ‘side, and especially ‘sire’ or ¢siri.’ It seems improbable, ;T however, that the Dravidian ¢ ir arose from the softening off of the - 9 ridently initial ‘s of these words; for in the Finnish family this same ‘s honetic appears as ‘ k; whence two is in some dialects of that family ¢kit;’ in ding in Magyar ‘ket,” ‘ketto ; and in Lappish ¢quekt.’ It has also been a place shown that an initial ‘k’ is a radical element in the majority of the nces 1n Scythian words for fwo; and hence, though the Mongolian ¢kur-in’ for the (for ‘kayar-in’), twenty, becomes in Manchu ¢or-in,” in Turkish ¢igir- no such mi ;’ we cannot venture to compare this Manchu ¢or’ with the Dra- J thers, vidian “ir’ or ‘er ; for it is certain that the latter was never preceded iy by ‘k,” or any other consonant, so far back as the Dravidian languages ith the can be traced. signify- o Grond Three.,—The neuter noun of number signifying three or a triad, is change in Canarese ‘muru ; in Telugu ‘mtdu ;’ in ancient Telugu, as quoted I, one, by Pliny, and testified to by native grammarians ‘modoga ;’ in Tamil Drévi- ‘minru’ {pronounced ‘mindru,” ‘mindu,’ and ‘minu’) ; in Malay- b f each ﬁla,l}l ‘mlnna ;’ in Tulu ‘miji ; in Génd ‘mind ; in Tuda ‘mad’ in Urdon ‘man-otan.’ as the I The numeral adjective tkree, which is employed in three persons, Z‘OPGHII thirty, three hundred, and similar compounds, is either ‘mi ’ or ‘mi.’ it are The long ‘mt’ is found in the Tamil and modern Canarese epicene n the nouns, ‘muvar,’ ‘muvar-u,” three persons, and in the Canarese ‘muapattu,” thirty. The shorter form ‘mu,’” is used in three hundred ; Drivie which in every one of the Dravidian dialects is ‘miinniru ; and we see it also in the Tamil ‘muppattu,” and the Telugu ‘ mupphei,’ thirty, of, by i and in the Telugu ‘muggar-u,’ three persons. o Dri- The primitive and most characteristic form of the neuter noun of ord 18 number is evidently that of the Canarese ‘miir-u,” from which it is yident £ ® clear that the Tamil ‘ minr-u’ (‘mundr-u ’) has been derived, by the native same nasalizing process as that by which ‘oru,” one, was converted 1‘” the into ¢ onru.’ fiven It was shown in the section on “Sounds, that the Tamil ‘r’ is Gt often changed into ‘d’ in Telugu : hence ‘mir-u’ and ‘m{d-u,’ are h the identical ; and it is more probable that ‘m{d-u’ has been altered from ‘mfr-u, than that ‘mir-u’ was altered from ‘mid-u. ‘s’ and ‘r; oticed evince in many languages a tendency to interchange, generally by the i the hardening of ‘s’ into ‘r ;' consequently the Brahui ‘ mus’ (‘ mus-it’), 'll’ or three, is closely allied to the Canarese ‘mir’ (possibly‘it was the ori- ppare ginal form of the word), and still more closely to the Tulu ¢ miji. It is doubtful whether the ‘r’ of ‘mfr-u,” should be considered as a Whi(‘}l formative, or as a part of the ancient root. On the whole, it seems o 0 el s 272 THE NUMERALS. f probable that the ‘r’ is radical, for I have not met with any reliable instance of the use of a final formative in ‘r-u. bl The final consonants Tan ‘nal of ‘aru,” Tam,, siz, and of éru,’ seven, belong unquestionably to the nsf roots of those numerals; and the existence in the Brahui word for bl three of an ‘s, chan corresponding to the Tamil-Canarese ‘r,” would seem Test to decide the question, especially seeing that this ‘s’ is followed by the particle ¢ it,” which is itself a formative. Moreover, when we compare pare Alan ‘ mun-niiru,’ three hundred (the same in all the dialects), with “ nil ‘in-ntiru,” two hkundred, in Telugu and Canarese ; and when it is Tel remembered that the latter has certainly been softened from ¢ ir-nliru’ (in Tamil ‘iru-nGru’), it seems to be probable that ‘ mun-niiru’ has been formed in a similar manner from ‘mur-niiru,” and consequently ‘R that ‘mur,” not ‘mu,” was the original root of this numeral. The same a( conclusion is indicated by a comparison of the Telugu ‘iddaru,’” ¢wo 0N persons, and ¢ muggaru,’ three persons. 41‘ It seems probable, therefore, that ‘mu’ originally was followed by a consonant ; and the softening off of this consonant would naturally account for the occasional lengthening of ‘mu’ into ‘mi.’ il I have not been able to discover any analogy to this numeral, tie, either in the Scythian orinthe Indo-European tongues. The only Extra- ‘h Indian resemblance to it is that which is found in the Brahui ; and fua this circumstance is a striking proof of the existence in the Brahui of Tan a distinctively Dravidian element. The total absence of analogy to hut the Dravidian ‘mur,” in other families of languages, leads me to sup- i pose that it must have been derived directly from some Dravidian a verbal root. The Latin ‘secundus, is undoubtedly derived from Ta ‘sequor ;’ and Bopp connects the Indo-European ¢tri,” ¢three, with the Do Sanserit root ¢ tri,” o pass over, to go beyond, signifying that whick * goes beyond’ two. If this derivation of ‘tri’ be not regarded as too fanciful, i lﬂ‘ bl 10ste a similar derivation of ‘mur,” from a Dravidian verbal root, may easily be discovered. In those languages there are two verbal roots which e present some points of resemblance, viz., ‘mir-u,” Tam. and Can., to go 10 be beyond, to pass, to exceed, to transgress; and ‘mur,’ to turn, an obsolete root, which is contained in ‘mur-ei, Tam., order, succession, a turn chau; Whie (e.9., ‘idu un murei,” this is your turn). ‘ mar-u,” Tam., Tel., and Can. th0u< to change, and the Tel. noun ¢ méar-u,’ @ time, & course, seem to be cor- of il relative roots. lay that Four.—The Drividian noun of number signifying four, or a qua- oy ternion, is in Canarese ‘nélku ;’ in Telugu ‘nélugu;’ in TAamil ‘nan-gu;’ 9 Iy in Tuda ‘nonku’ or ‘nonk; in Gond ¢ nald ; in Uraon ¢ nikh- otan.’ i b FOUR. 273 eliable The adjectival or crude form of this numeral is “ nal’ or ¢nal. In onants Tamil it is ‘nal-u,’ in Malayalam ‘nal-a,’ in some Telugu compounds to the ‘nal ; and this adjectival form is often used as a noun of number ord for instead of ‘mnalku,’ &e. In composition ‘nél’ undergoes some 4 seem changes. The quantity of the included vowel, which is long in all the Dy the rest of the dialects, is short in Telugu compound numbers : e.g., com- ompare pare the Tamil tnénpadu,’ the Canarese ‘nélvattu,” and the Malay- - with alam ‘nélpadu,’ forty, with the Telugu ‘nilubhei; and the Tamil n it 1§ ‘nan-ntiru,” and the Canarese ‘nal-niru, four hundred, with the -nlRe’ Telugu ‘ndn-niiru.’ 1’ has The final “1’ also is subject to change. In Tamil it is changed into juently ‘R’ before ‘p,” as in ‘narpadu, forty ; and before ‘n’ it is assimilated 6 Same and becomes ‘n,’ in both Tamil and Telugu ; e.g., ‘nnniiru’ (in the u, fuo one), and ‘nanniiru’ (in the other), four hundred. These changes of ‘l,” however, are purely euphonic. ved by It is evident from a comparison of the above forms, that ‘nal’ (o, turally as the Telugu seems to prefer it, ‘nill’) was the primitive shape of this ey numeral ; to which ‘ku’ or ‘gu’ was subsequently added as a forma- meral, tive, in order to constitute it a neuter noun of number. This formative Fxtra- ‘ku’ (pronounced ‘gu’) isa very common one in the Dravidian lan- i and guages ; e.g., ‘kadu-gu, Tam., mustard, from ‘kadu, pungent. In hui of Tamil the only numeral to which ‘ku’ or ‘gu’ is appended is ‘nal gy b but in Telugu we find it used not only by ‘nalu-gu,” four, but also by fo sup- Jive, sux, seven, eight, and nine, in forming rational plurals ; e.g., from 2 jvidian “aru,’ siz, is formed ¢Aarugur-u, SUX PErsons. In such connexions the | from Tamil uses [4 v’ euphonic instead of ‘g’ (eg., ‘aru-(v)-ar), which ith the proves that ‘gu’ does not add to the grammatical expression, but is ‘g0 a mere euphonic formative. Even in Telugu ‘Aruvur-u’ may be used eiful, | instead of ¢arugur-u.’ / easily The change of ‘1’ in Tamil, into ‘n, before the ‘k’ of this which appended formative, ‘ku,” is an euphonic peculiarity which requires to be noticed. , 1000 In modern Tamil ‘1’ in this conjunction, would be bsolete changed into ‘R; but the change of ‘1’ into ‘n,’ before ‘k’ or g o furt which we find in the Tamil noun of number, ‘nan-gu,” is one which, d Can. though now obsolete, appears to have been usual at an earlier period he cor of the history of the language; e.g., compare ¢ Pang-guni, the Tamil name of the month March — April, with the Sanscrit name of that month, ¢Phalguna, from which it is known to have been o qur derived. This change of ‘1’ into ‘n, in ‘nan-gu,” must have been |én'gﬂi' made at a very early period, seeing that we find it also in the Tuda ik ‘nonk.’ In the entire family of the Indo-European languages there is not T —— ) 74 THE NUMERALS. Tan one language which contains a numeral signifying four, which in ‘the smallest degree resembles the Dravidian ¢nal’ Here the Brahui also fun fails us ; for it is only in the first three Brahui numerals that we find ‘e traces of Dréividian influences, and the rest of the numerals of that five, language from four to ten inclusive, are derived from the Sanscrit. it Though other analogies fail us, in this instance Ugrian affinities (‘e The resemblance between the Finnish int are more than usually distinet. tongues and the Dravidian, with respect to the numeral four, amounts not to identity, and cannot have been accidental. Compare with the Dra- Tel vidian ‘nal,” the Tcheremiss ¢nil ;7 the Mordwin ¢nile,” ‘nilen; the adj Vogul ‘nila ;* the Ostiak ‘niil,’ ‘nel,” ‘njedla,” ¢ nieda,” ‘ njeda ; the ad Finnish proper ‘mnelji; the Lappish ¢nielj,” ‘nelje,” ‘nella; the Si The root of all these nume- Magyar ‘négy’ (pronounced °neidj’). W( 0! rals is evidently ‘nil’ or ‘mel,’” the analogy of which to the Dréavidian ‘nal’ or ‘mal, is very remarkable. The Magyar ‘negy,” has lost ¢ the original ¢1,” through the tendency, inherent in the Finnish idioms, S The Ostiak ‘njedla’ or the to regard ‘1’ and ‘d’ as interchangeable. ‘nedla,” in which ‘d’ and ‘1’ form but one letter, a cerebral, consti- tutes the middle point of agreement. A similar softening down of al the ‘1’ of ‘nal,” appears in the Tulu, in which fourtcen (ten-four) is flex ¢ pad’naji.’ Five.—The Dravidian numeral noun five, is in Canarese and den Telugu ‘eid-u ; in Tamil ‘eind-u; in the poetical and colloquial dialects of the Tamil ‘anj-u; in Malayalam ‘afijcha ; in Tulu ‘ein-u’ or Sar ‘ein-i; in Tuda “Gtsh’ or “0j.” The Gond has ‘seighan’ or ¢seiyan, al —a word which is derived like ° sarun,” siz, from the use of ‘s’ as an % it euphonic prefix : ‘eiyan’ is to be regarded as the correct form of the fron Gond numeral. The ﬁréon, and other rude dialects of the North- ? 4 (0 Dravidian family, exhibit no analogy to any of the Dravidian nume- fom rals above four. In Telugu compounds, the word for five is not ‘eid-u,’ anl In this case the initial ‘h’ is but hén-u ; eg., ‘padibén-u,’ fifteen. Teng purely euphonic, and used for the prevention of /kiatus, as in the .'Onn parallel instances of ‘pada(h)aru,’ sizteen, and ‘padi(h)édu,’ seventeen. The Telugu possesses, therefore, two forms of five, ‘eid-u’ and ‘én-u; o and the Tamil ‘eindu’ shows how ‘eidu’ was converted into ‘énu,” viz., ity by the insertion of an euphonic nasal and the subsequent assimilation to it of the dental. five Al The numeral adjective five, is in most of the Dravidian dialects por ‘el in Telugu “é In Tamil, and also occasionally in Canarese, ‘ei’ In; is in combination converted into ‘ein’ or ‘eim,” by the addition of an - ey bej euphonic nasal. Thus fifty (five tens) is in Canarese eivatt-u;’ in (d] S T T e FIVE. 275 i the Tamil ‘eimbad-u’ (‘ei-m-pad-u) ; in Telugu ¢ ébhei’ (‘ é&-bhei).” Five i also hundred is in Canarese ein-ntir-u, in Tamil ¢eifi-fijira, in Telugu e find ¢é-nlir-u.” We see the numeral adjective five, and the noun of number ‘el {)f that Jive, in juxta-position in the Tamil ¢ ei-(y)-eind-u,’ five times five. I remains also in its pure, unnasalised form in the Tamil ¢eivar’ finities (“ei-(v)-ar’), five persons. The nasal ‘n’ or ‘m, which follows € ei,’ linnish in the compounds ¢ eimbad-u,” fifty, and °eififijir-u,” five hundred, is mounts not to be confounded with the ‘n’ of the Tamil eind-u,” or the 16 Drd- Telugu ¢ én-u,” but proceeds from a different source. It is an euphonic 1 the adjectival increment ; and is added by rule, not only to this numeral 1 the adjective “ei, five, but to many similar words which consist of a ) the single syllable, of which the final is a long open vowel, when such nume- words are used adjectivally. Thus, we find in Tamil not only such aridian compounds as ¢ eintinei’ ( ei-n-tinei’), the five conditions, and ‘eimpulan’ ag lost (‘ei-m-pulan’), the five senses; but also ‘keinnodi’ (‘ kei-n-nodi’), @ idioms, snap of the finger, and “ mangkai’ (“ mi-ng-kai’), @ mangoe, literally dla’ or the fruit of the < mi. consti- This adjectival, euphonic addition is an abbreviation of ‘am’ (or ywn of “an’ before a dental), and is probably from the same origin as the in- our) 18 flexional increment ‘in’ or €an. See the section on ¢ Nouns.’ ) It may be doubted whether the Tamil-Canarese ¢ ei,” or the Telugu ‘e Y is the better representative of the original numeral ; but the evi- 0 ad dence of the various dialects preponderates in favour of ei. dialects A remarkable resemblance must have been noticed between the 4 O Sanscrit panchan,” five (in Tamil ¢panja’), and the true Tamil ‘anju,” and the Malayéla ‘afijcha.’ The resemblance is so great that SEiyanJ " asan it has been supposed by some that the Dravidian word was derived i}j he from the Sanserit ; but instead of this supposition being confirmed by l?f f a comparison of the various Dravidian idioms, and of the various North nume forms under which this numeral appears, as would be the case if the ‘eld 1, analogy were real, it is utterly dissipated by comparison, like the appa- 1‘]1’15 rent analogy which has already been observed between the Malayéala “onna, one, and the Latin ¢ un-us.’ n the yenteens The primitive, radical form of the Dravidian numeral five is ‘ei’ | Gpelly or ‘& as appears from its use as a numeral adjective. The abstract L 0, Vi or neuter noun of number is generally formed from the numeral adjec- tive by the addition of some formative. The formative suffix which is milﬁli(}ﬂ added to ir-u, two, is “du ; and by the addition of <d-u,” & corres- ponding formative, ¢ ei’ becomes ¢ ei-du, five, or five things; which is dialecf;q in itself a neuter noun, though, like all such nouns, it is capable of 1086 o fan being used without change as an adjective. This formative suffix on 0 tH ! i “d-u” is an exceedingly common formative of neuter appellative T 2 2 oy { 6 THE NUMERALS. nouns in the Dravidian languages, particularly in the Tamil ; and is ‘ot doubtless borrowed from, or allied to, the termination of ¢ad-u, ¢, the 801 ¢ eid-u,” the numeral neuter singular of the demonstrative pronoun. bla noun of both the Canarese and the Telugu, is evidently the original for and most regular form of this word. ¢eid-u’ could not have been 1101 corrupted from ©anj-u,” or even from ¢eind-u,’ but the corrnption of “u ‘eind-u’ and ‘anj-u’ from an original ¢eid-u’ will be shown to be in el perfect accordance with usage. The first change was from ‘eid-u’ to €eind-u,’ by the insertion of ha an enphonic nasal, as in the former instances of ‘irad-u,’ two, changed of into irand-u.’ This euphonic insertion of ‘n,’ after certain vowels, is e 80 common in Tamil, that it may almost be regarded as a rule of the th language ; and hence preterite participles which end in Canarese in {h ‘ed-u,” always end in Tamil in ‘n-du; e.g., compare ‘aled-u,’ Can., fro having wandered, with ¢aleind-u,” Tam. When ¢eid-u’ had been aw. changed into ‘eind-u,” Tamil usages of pronunciation necessitated a ‘ke further change into ‘einj-u’ or ‘anj-u.’ It is a rule of the colloquial i Tamil that when ‘nd’ is preceded by ‘ei’ or ‘i’ it is changed in pro- ik nunciation into ‘nj.’ This change is systematically and uniformly and practised in the colloquial dialect ; and it has found its way into the hag classical and poetical dialect also. 5: e Moreover, in changing ‘eind” into ©einj,” there is a further change of the vowel from ‘ei’ to ‘a, in consequence of which ‘einj’ becomes 4 e anj This change is almost always apparent in the Malayalam, and E also in the pronunciation of the mass of the people in Tamil. Thus, [ ¢ pareindu,” Tam., having spoken, becomes in Malayélam °parafifiju ; and in this instance we see illustrated the change both of ‘ei’ into ‘a,’ W and of ‘nd’ into ‘fij ° consequently the perfect regularity of the fiv change of ‘eind-u, five, into ‘anj-u’ and ‘afijch-a’ is established. ~ o~ b4 Where the Malayalam does not change ‘nd’ into ‘ nfij, it changes it into ‘nn ; e.g., ‘nadandu,’ Tam. having walked, is in Malayalam ‘ natannu.’ to This illustrates the process by which ¢ eind-u’ became ein-u’ in Tulu, E al and ‘én-u’ in the Telugu compound, ¢ padi(h)én-u,’ fifteen. T; It is thus evident that the apparent resemblance of the Dravidian 1 “anju’ to the Sanscrit ¢panchan’ is illusory. It entirely disappears 8 on examination, and the slight resemblance which does exist is found to arise from the operation of DraAvidian principles of sound. Consequently ‘ei’ or ‘&’ must be regarded as the sole representative I¢ of the Dravidian numeral, and with thisit is evident that neither ‘pan- 6 chan,” nor any other Indo-European form has any analogy whatever. In some of the Finnish tongues the word for fve has some slight resemblance to the neuter Dravidian numeral ¢eid-u.’ U The Vogoul is “a 277 SIX. nd s at 7 the Ostiak ‘uut’ or ‘wet; the Magyar ‘ot’ (pronounced { the somewhat like ¢ét.’ I am doubtful, however, whether this resem- mera: blance is not merely accidental ; for the final “t’ of the Ugrian words 1212 for five appears to be radical ; whereas the final ‘d’ of the Dravidian been The Chinese noun of number, ‘eid-u,” is simply a neuter formative. 0n 0 ‘u be In [4 .’ may, perhaps, be compared with the Dravidian numergl adjective 2) el. In some languages the word used to signify five properly means @ jon. of hand, or is derived from a word which has that meaning,—the number need of fingers in each hand being five. In Lepsius’s opinion, the word for 618, 18 ten which is used in all the Indo-European dialects, had its origin in of the the Maeso-Gothic ‘tai-hun,’ two hands. Applying this principle to ese 10 the Dravidian languages, ¢ ei,’ five, might be presumed to be derived (an,, from ‘kei, Tam., @ hand, by the very common process of the softening been away of the initial consonant. On the other hand, I do not consider ited & I have ‘kei,’ @ hand, to be itself a primitive, underived word. mquial little doubt but that it is derived from ‘ki,’ ‘gé,’ ‘géy,” or ‘chey,” to do, i} pl’O' like the corresponding Sanscrit word ¢ kara,” a kand, from ‘kri’ to do ; ormly and in accordance with this opinion, I find that ‘kei’ Tam., a hand, to the has in Telugn become ‘ché,’ just as the Canarese ‘géy, to do, has become ¢ chey,”in both Tamil and Telugu. hange Hence the derivation of ¢ ei,” five, from “ kei, a hand, becomes im- ooomes peded by the previous question, is not “ kei” itself a derivative? 1, aﬂd Possibly ‘ei’ may be more nearly allied to the High Tamil abstract Thus, noun ‘ei-mei,’ closeness, nearness, a crowd, from an obsolete verbal rvot In this case the use of “ei’ as a numeral, fiju ) “ei, to be close together. nto 4, would take its origin from the close juxta-position and relation of the of the five fingers of the hand. Jised. Six.—In all the Dravidian dialects, the difference which is found it Into to exist between the neuter noun of number stz and the numeral mmm." The numeral noun is ‘4ru’ mn ) Tulty adjective is extremely small. Tamil, Telugu, and Canarese; “4ra’ in Malayalam; and €ar’ or <6r’ in Tuda; in Gond ‘s-artn.’ In Tulu it is “4ji,” a form which bears the yidian :i: same relation to ¢ 4aru’ that ‘mdji,” Tulu, three, does to the Canarese [ppeﬁys ¢ muru.’ {ish 18 The numeral adjective differs from the noun of number with souﬂd' respect to the quantity of the initial vowel alone; and in some cases mtative even that difference does not exist. In all Tamil compounds in which P tevel: o b D A b4 ‘4r-u’ is used adjectivally, it is shortened to “#r-u; e.g., <arubadu, stzty. The vowel is short in the Can. ‘aravattu,” and the Telugu o8 gt caruvel,” sizty; whilst it is long in the higher compound ¢aruniiru, ag()ﬂl is 278 THE NUMERALS. hos | Can., and “ Arn(ru,” Tel., stz hundred. In Tamil 1t i1s short in sz Wwas hundred, but long, as in the other dialects, in siz thousand. Probably ‘4r-u’ was the primitive form of this numeral, the initial vowel of which, like the included vowel of the personal pronouns, was euphoni- supp cally shortened in composition. (ans No analogy whatever can be traced between this Dravidian numeral and any word for siz that is contained in the Indo-European I languages; and I am very doubtful whether any Scythian analogies as th are discoverable. thos In Magyar siz is ‘hat;” in the Turkish languages “alty,’ <alte, &e. It may be supposed to be possible that the first they syllable of the latter word, ¢4l is allied to the Dravidian ‘AR’ in 16ga virtue of that interchange of 1’ and ‘r,” which is so common in the and Scythian tongues. It may be couceived also, that the Turkish alt, and the Magyar ‘hat’ are allied. I have no faith, however, in these the indistinet resemblances of sound; for the Magyar “at’ seems originally whic to have had an initial consonant. “kot’ is the corresponding numeral in Lappish, and ‘kisi’ in Finnisk; in Tcheremiss ‘kut; whereas, [ fe? e there is no reason to suppose that the Dravidian ‘4R’ ever commeneed i t with a consonant ; nor do I suppose it very likely that in the rude “het Scythian tongnes, in which even the numerals of cognate dialects differ from one another so widely, any real analogy with the Dravidian Nag numerals above four or five would be discoverable. hotk par Seven.—The Dravidian noun of number sewen is ¢ ér-u’ in Tamil, or “ér-a’ in Malayalam, < él-u’ in Canarese and Tulu, ‘éd-u’ in Telugu. These differences are in accordance with the rule that the Tamil deep, liquid, semi-vowel ‘r’ becomes ‘d,’ in Telugu, and ‘1’ in Canarese. 1S In the Tuda this numeral is “ér’ or ‘iid;’ in Mahadeo Gond ‘¢ y-ént fam or ‘ y-étli; in Seoni Gond ¢ éro. “att bt The numeral adjective seven, which is used in the compound L numbers seventy, seven hundred, &e., exhibits a few trivial differences ) U from the noun of number. In Tamil ‘ér-u is shortened to ‘er-u’ e —— ‘et when used adjectivally, like ¢4Rr-u,’ siz, which is similarly shortened dEl‘i\ to ‘ar-u.’ A thoy In Canarese seventy is ‘eppattu,’ in which not only is ¢ e shortened ang to ‘e, but the radical conmsomant 1’ answering to the Tamil Tege ‘;',’ has been assimilated to the initial consonant of the succeeding word. i In ‘élntiru, Can., seven hundred, this assimilation has not taken i 1 place. In Telugu, the ‘d” of -éd-u’ does not appear to be very 0f persistent. In ‘élnru,’ seven hundred, ‘d’ becomes ‘1’ as in the Canarese ; and in ‘ debhei,’ seventy (for ¢ edubhei’), the initial vowel ‘e’ the w— EIGHT. 279 L 8 has been displaced, according to a peculiar usage of the Telugu, which by was explained in the section on ¢ Sounds.’ wel of This displacement of the initial vowel shows that the ‘e’ of the phont- suppositious ¢ edubhei’ was short, as in the corresponding Tamil and Canarese compounds. ridian It cannot be determined with perfect certainty which of the three Opeal consonants, ‘r,” “d,’ or ‘1, was the primitive one in this numeral; but logies as the Tamil ‘r’ changes more easily into ‘1” or €d,’ than either of TIges those consonants into ‘r,” and could also be changed more easily than o finst they into the ‘n’ of the Gond, probably ¢ér-u,” as in Tamil, is to be R, i regarded as the primitive form of this numeral; from which ¢éd-u’ in the and ‘él-u’ were derived. alt, No resemblance to this Dravidian numeral is to be found in any of these the Indo-European languages; and the slight apparent resemblances ;inally which may perhaps be traced in some of the Scythian tongues, are not meral I fear, trustworthy. Compare with the Telugn ‘éd-u,’ the Turkish 1eI6AS; ‘yedi; the Turkish of Yarkand ¢yettah’ (the root of which appears enced in the Ottoman Turkish ¢ yet-mish,” seventy); and the Magyar rude ¢ hot:? alects In Armenian, seven is € yotn,” in Tahitian ¢hetu.’ The ¢h’ of the vidian Magyar numeral and the ¢y’ of the Turkish seem to be identical; but both have been derived from a harder sound, as will appear on com- paring the Magyar ‘het’ with the Lappish ‘kietya, and with the Tamil, corresponding Finnish ¢ geit,” in ¢ seitzeman.’ elugt, deep; Eight,——The Tamil numeral noun ‘ettu,” eight, bears a remarkable resemblance to the corresponding numeral of the Indo-European 1&1‘6?& ’ entl family, which is in Latin ¢octo,” in Gothic ‘ahtau; and especially to ‘atta,” the manner in which ¢ashtan, Sans. eight, is written and pro- nounced in classical Tamil, in which it is occasionally used in com- ound g008 pounds; hence it has naturally been supposed by some, that the Tamil ‘ettu’ has been derived from, or is identical with, this Sanscrit § er»ll’ derivative ¢atta.’ It will be found, however, that this resemblance, rtened though so close as to amount almost to identity of sound, is accidental ; and that it disappears on investigation and comparison, like the fened resemblance between € onna’ and ‘unus,” ‘anju’ and ‘pancha.’ Tamil eding The Dravidian noun of number eight is in Tamil ¢ ettu,” in Malay- alam ©ett-a,” in Canarese ‘ent-u,’ in Telugu enimidi’ or ¢ enmidi,” in Tulu ‘enuma,’ in Génd ¢ anumar’ or ‘armir, in Tuda etthu, ‘vet, taken ¢ oet’ or ‘yeta. vely The corresponding numeral adjective, which should by rule exhibit ] fhf’ ;g oK the primitive form of the word, is generally ‘en’ In Tamil ‘en’ is 280 THE NUMERALS. used adjectivally for eight in all compound numerals; e.g., ‘en-badu, consi esghty, < en-ntru,’ eight hundred, as also in miscellaneous compounds, eight, such as ‘en-kanan,’ ke who has eight eyes. The same form is used ten? adverbially in en-éru,” eight times seven. In Canarese, in which the It numeral noun is ‘entu, ‘en’ is used as the numeral adjective in nne ‘envar-u,’ eight persons; whilst in ¢ embattu,’ etghty, ‘n’ is changed wouid into ‘m,” through the influence of the labial initial of the second ‘enin member of the compound. In ‘entu-niiru,’ etght hundred, the numeral to ‘e noun is used adjectivally instead of the numeral adjective. The I Telugu noun of number ¢enimidi, eight, differs considerably from the that Tamil ‘ettu,” and the Canarese ‘entu; but the difference diminishes | Tely when the numeral adjectives are compared. The Telugu numeral above adjectivé used in ‘enabhei,” eighty, is ‘ena,” which is obviously identical simils with the Tamil-Canarese ‘en.’ In ¢ enamandru,’ or ‘enamandugur-u ’) woul eight persons, and € yenamann{iru,’ eight hundred, the ‘m’ of ¢ enimidi,’ }Er ‘e ewght, evinces a tendency to assume the place of an essential part of | the root. It will be shown, however, that ‘ midi’ is not a part of the elght root of this numeral, but a suffix; and consequently ‘en’ or ‘en,’ 0ne. without the addition of ‘ m,’ may be concluded to be the true numeral adjective and also the root itself. ¢ight iy 1] Thus, the apparent resemblance of the Tamil ¢ ettu’ to the Sanscrit heg T derivative ‘atta’ (euphonised from ¢ ashta ’), disappears as soon as the ( various forms under which it is found are compared. B The primitive form of the neuter noun of number derived from ‘en,’ e, Wit is evidently that which the Canarese has retained, viz, ‘entu,” which 10f} is directly formed from ‘en’ by the addition of ¢tu,” the phonetic 61, equivalent of ‘du’—a common formative of neuter nouns, and one of which we have already seen a specimen in ¢ eradu,” fwo The Tamil B 1gng ‘ettu’ has been derived from ‘entn’ by a process which is in accordance with many precedents. regm It is true that in general, the Tamil refrains from assimilating the nasal of such words as ‘entu,” and oftentimes it I { nfly inserts a nasal where there is none in Canarese; e.g., irandu,” Tam., fory :; two, compared with the Can. “eradu:’ still this rule, though general, is not universal, and is sometimes reversed., Thus, ¢ pente,” Can., a hen, Sﬂp Cay (in modern Canarese “henteyu,’) has in Tamil become ‘ pettei’—a Dy change exactly parallel to that of ¢ entu’ into ‘ettu.’ Wi Some difﬁculty is involved in the explanation of ‘enimidi,’ the Telugu noun of number which corresponds to ‘entu. ‘eni’ or ‘ena’ (as in ‘enabhei,’ eighty), is evidently identical with the Tamil- lay | Canarese ‘en :’ but what is the origin of the suffix ‘midi?’ This be ‘midi’ becomes ‘ma’ in some instances ; ei:; Dy ‘enama-ndru,’ eight persons; and the Tulu noun of number, elght, is ‘ enuma.’ Shall we fo dia EIGHT. 281 ‘o0 bady, consider ‘midi’ to be synondmous with ‘¢ padi,’ fen; and ¢enimidi,’ ompounds, eight, to be a compound word, which was meant to signify two from 11 used ten ? which the It will be shown under the next head that in the Telugu ‘tommidi,’ jective in nine, ‘midi’ is without doubt identical with padi,” ten If so, there 5 changed would seem to be a valid reason for supposing that the ‘midi’ of he second “enimidi,’ eight, is also derived from the same source, and appended evnumeml to ‘en’ with the same intent. ive, The It will be shown in our examination of the Dravidian numeral Zen from the that ‘padi’ has become greatly corrupted in compounds, especially in [iminishes Telugu ; in which the second syllable has disappeared in compounds | nameral above twenty. If ‘midi, identical with ‘padi, were liable to a ridentical similar corruption, as is probable enough, we may see how ¢enimidi’ ndugur-y, would be softened into ‘enama’ (in ‘enamandru), and also into ¢ enimid, “enuma’ in Tulu. 1 part of It is a characteristic of the Scythian languages that they use for art of the eight and nine compounds which signify ten minus two and ten minus o ‘en) one In some instances an original uncompounded word is used for numeral eight: but nine is always a compound. The Dravidian word for nine is, I have no doubt, formed in this very manner; and this seems to » Sanserit be a rational explanation of the origin of the Telugu word for eight. on as the On the other hand, in the Tamil-Canarese idioms, ‘en’ by itself is used to signify eight, without any trace of the use in conjunction rom ‘GH; with it of the word ¢pattu’ or ¢ padi,’ ten It is also deserving of 1 which notice that in the Telugu ¢enabhei, eighty, the second member of “enimidi’ has disappeared. phonetic d one of It is difficult to determine whether the disuse of fen as a compo- e Tnmil nent element in the numeral eight of the Tamil and Canarese is to be (;ordilllce regarded as a corruption, or whether the use of ¢en by the Telugu pofrains in the construction of eight is itself a corruption, arising from the ptimes it influence and attraction of the principle which was adopted in the g, Tamy formation of the next numeral nine. On the whole, I consider the latter eneral 8 supposition as the more probable, and therefore regard the Tamil- Canarese ‘en’ (in Telugu ‘ena,’) as the primitive shape of this ), 0 ety Dréavidian numeral. Max Miller’s supposition that ‘en’ is identical el b 3 with €er, properly ‘ir, fwo, is quite inadmissible. idi, the ‘en’ has no resemblance to any numeral belonging to any other language, whether Indo-European or Scythian; and it cannot, I think, i fens’ be doubted, that it was first adopted into the list of numerals by the - Tanil Dravidian people themselves. We have not to go far out of our way 7 This Lo to seek for a derivation. [‘ll: M/” ‘en’ is a primitive and very common Dravi- dian root, signifying either ¢o reckon or a number, according as it is Shal 78 282 THE NUMERALS. used as a verb or as a noun. searel In Tamil it is ‘en, in Telugu fenn-u,’ in Canarese ‘en-usu.’ We have an instance of its use as a noun in ‘fomp? thirty, ‘en-suvadi, Tam., @ book of Arithmetic, literally a number book. After the cor the Dravidians of the first age had learned to count seven, they found the ins they required a higher numeral, which they placed immediately above certain seven and called ‘en,’ @ number—an appropriate enough term for e 1 perhaps the highest number which they were then accustomed to reckon. into tk In A similar mode of seizing upon a word which denotes properly a ‘pade number or any number, and using it restrictively to denote some one (Cana number in particular —generally a newly invented, high number—is found in other languages besides the DrAvidian. concl Thus, in Lappish, “onds ¢lokke,” ten, means literally, a number, from ¢lokket,” Zo count. witl Nine.—In all the Dravidian idioms the numeral nine is a com- prefix pound word, which is used indifferently, and without change, as a to ‘o noun of number and as a numeral adjective. that | The second member of the compound numeral nine is identical other with, or evidently derived from, the numeral ten, the differences 0 prig between it and that numeral being such as can be accounted for by Telug the phonetic tendencies of the various Dravidian dialects. T The principal forms which this numeral assumes are the following : expre in Tamil it is ‘onbad-u, in Canarese ‘ombhatt-u,’ in Telugu 1§ n ‘ tommidi,” in Tulu ‘ormbo,” ‘orambu,’ or ‘worambu,” in Tuda yen- finst J bot,” in Kota ¢ worpatthu: in each of which instances the second give member of the compound plainly represents zen. | b In Tamil ten is ¢ patt-u; nine is ‘ onbad-u’ (“ on-pad-u,” euphoni- varig cally, ‘on-badu’) : and not only is it evident that ¢patt-u’ and idion ‘pad-u’ are allied, but the resemblance becomes identity when Conlp( ‘pad-u,’ the second member of ¢ onbad-u,’ is compared with the repre- onbg sentative of ten in ‘irubad-u,’ twenty’—literally twice ten and — ‘T similar compound numerals. Moreover ¢ onbad-u’ itself becomes in th ‘onbatt-u’ when used adverbially, e.g., ‘onbatt'—eér-u, nine times ‘ol seven.” In ancient Canarese ten was ¢ patt-u,” as in Tamil. In modern hut I Canarese it changes by rule into ‘hatt-u;’ nevertheless the original ﬂlﬁ‘ug labial retains its place in the compounds ¢ombhatt-u,’ nine, and Plag ¢ embatt-u,” eighty ; from which it is evident that in Canarese nine is formed from fen, by means of an auxiliary prefix, as in Tamil. TIn iy Telugu alone there is some difference between the word which separately signifies ten and the second member of ¢tommidi, the compound the numeral nine. Ten is in Telugu padi,” whilst nine is not ¢ tompadi,” or [0 “tombadi,” but ‘tommidi;’ and nine personsis ‘tommandugur-u.’ It can Schu NINE. 283 e, scarcely be doubted, however, that ‘tommidi’ has been euphonised from 1000 in ‘ tompadi.’ In the other compound numerals of the Telugu (‘fwenty, o After thirty, &c.,) in which ‘padi’ forms of necessity the second member, ey found the corruption of ¢ padi’ into ‘bhei’ or ¢ vei’ is still greater than in the instances now before us. ly abore It may be regarded, consequently, as term for certain that the second member of the various Dravidian words for omed to nine 1s identical with Zen. We have, therefore, now to inquire only into the origin and signification of the first member of the compound. opetly 1 In the Tamil ‘onbad-u,’” ‘on is the auxiliary prefix by which ome one ‘padu’ is specialised ; and this ‘on’ is evidently identical with the mber—is Canarese [4 om, ) in ¢ ombhatt-u.’ Max Miller, naturally enough, Lappish, coucluded it to be derived from ¢on,” the first portion of the Tamil il “ondru,” one:’ bnt this derivation, though plausible, is inconsistent with many facts which will be adduced. In Telugu the auxiliary 4 C0- prefix of ‘tommidi,” nine, is ¢ tom,” which is undoubtedly equivalent to ‘om ge, a8 8 in Canarese, and ‘on’ in Tamil; and as it is more likely that the ‘t’ of ‘tom’ should have been softened away in the identical other idioms than arbitrarily added by the Telugu, it seems probable, a priort that the original form of this numeral prefix was ‘tom,” as in ferences for by Telugu, rather than ‘om,” or ‘on. The Tuln appears to have preferred a different prefix ; nine being expressed in that language by ¢ wormbn’ or ¢orambu’—a word which llowing: Telugu is probably identical with the Kota ¢ worpatthu,” and of which the first member ¢ wor’ or ‘or’ seems to be allied to ‘or,” one, so as to la ‘yen- give the compound the meaning of one from ten. , seeond When the various compounds into which nine enters, and the various shapes in which it presents itself in the different Dravidian euphoﬂi- 1 ﬂHd idioms are compared, it is evident that the first member of the Telugu / when compound ‘tommidi,’ of the Canarese ¢ ombhatt-u,” and of the Tamil ‘ onbad-u,” must originally have commenced with ‘t. Though this 6 IEPTE‘ n—and “t’ is not found in the Tamil ‘ombadu,’ yet it has retained its place in the higher and less used members, viz., ¢ tonnir-u,” niney, and becomes ‘ tol-ayiram,” nine hundred. In Telugu it is found not only in nine, 0 Tnes modern but in all compound numerals into which nine enters: and in Canarese, i: though it has disappeared from nine and nine hundred, it retains its Original place in ‘tombatt-u,’ ninety. e, 1 aine 18 Additional light is thrown upon this prefix by the Tamil ¢tol- e ayiram,” nine hundred, in which ¢ tol’ is used as the equivalent of ¢ton’ i | In or ‘tom;’ and this is evidently the most primitive shape in which arafel) the prefix appears. Even in the Tamil ¢tonniir-u,’ ninety, the prefix mp[)lmd to ‘niRu is not ‘tom’ or ‘ton, but is really ‘tol,” as every Tamil _uadi; or scholar knows. The ‘1’ is assilllilgted to the ‘n’ of ‘nir-u, and e ot 2 84 THE NUMERALS. used both nasals are changed into the nasals of the cerebral row, by a is onl recognized Tamil law of sound. The operation of the same law is apparent in the trite word ‘enney,” Tam., oi/; which is properly ‘el pOllll( (ana ney, the oil or ghee of the ‘el’ or sesame, and in which the concur- pkm { rent consonants ‘1’ and ‘n’ are converted by rule into ‘nn.’ Seeing that the ‘ton’ of the Tamil compound ‘tonniru,’ ninety, comp: on th thus resolves itself into ‘tol, in accordance with the higher number of I “ tol-dyiram, mine hundred, it is evident that ¢on,” ‘om,” ‘tom,” and thoss “ton,” are but different forms of the same word ; and that ¢tol,” being frue, the most distinctive form, must be the parent of the rest, and the « b truest representative of the root. In composition ‘tol’ will become n ‘ton,” or ‘tom,” without difficulty; but none of those forms can be converted into ¢ tol’ under any circumstances whatever. A final “1° reg: but is constantly and regularly euphonised in Tamil, through the influence : i of the initial consonant of the succeeding syllable or word ; and this the rule especially applies to the final ‘1’ of the first syllable of com- nea pounds. When ¢1° is followed by ¢d, it is changed by rule into ‘n,’ e.g., ‘kol-du,’ having taken, becomes ‘kondu in Tamil; and this 1neal e euphonised form of ¢1d’ occupies in the Telugu ‘konu’ the position V of the root of the verb. When ‘1’ is followed by ‘p’ or ‘b, it is ordinarily changed into the cerebral ‘t;” and this is the consonant the | of { which we should expect to find in the compound numeral nine, viz., “otpad-u’ or ‘totpad-u,’ instead of ‘onbad-u,’ ‘ombhatt-u,” and ‘ tom- Can Wi midi.’ The true explanation of the change of ‘tolpad-u’ into (a ¢ tonbad-u’ is furnished by the poetical dialect of the Tamil, in which there are traces of the existence of a system of euphonic changes, dial the somewhat different from those which are now in use. Thus, from nal; a verbal root, the ordinary Tamil even of the poets forms ‘natpu, poin | {l friendship ; but we also occasionally meet in the poets with ‘nanbu,’ a rarer and older form ; and this shows the possibility of tolpad-u’ I fo ? becoming, by the earlier euphonic process, ¢ tonbad-u. The possibility Tetp San becomes a certainty when we find in the Tamil classiecs a word for nine of which the basis is actually not ‘on,’ but “ton,’ viz., ‘tondu, nine, a word which is unquestionably and directly devived from ¢tol, Pref 01y and which shows, not only that ‘onbadu’ must originally have been ¢ tonbadu,” but also that tol’ is the basis of the first member of every 1N o Dravidian form of this compound numeral. When the Telugu and Canarese compounds ninety, nine hundred, the frg &c., are compared with their equivalents in Tamil, we cannot fail to be struck with the great simplicity and regularity of the Tamil com- A pounds. In Telugu and Canarese, ninety is ¢ tombei’ and ¢ tombattu’ iy (literally nine tens), in each of which compound numerals tom’ is ga} o NINE. 285 ow, by 4 used to signify nine, though properly it does not represent nine, but e law is is only the first member of the numeral nine, which is itself a com- pound. In like manner nine hundred is in Telugu ¢ tommanntru,’ in petly ‘el ¢ concur- Canarese ‘ombayniru; compounds which are formed on the same ] i plan as nenety, though with a fuller representation of both parts of the | minel, compound nine, which they adopt as their first member. In Tamil, 1 number on the other hand, the composite numeral nine is altogether lost sight of in the construction of the compounds ninety and nine hundred; and fom," and those compounds are formed (by means of the same expedient, it is ol, being 1) true, as nine, but independently of it) by prefixing ¢tol’ to the word and the || become @ hundred in order to form minety, and the same ‘tol’ to a thousand in order to form mnine hundred. s can be In these instances ‘tol’ must be final ‘1 regarded as an adjective, not signifying any one numeral in particular, but having some such general signification as defect, diminution, or nfluence manus, and thus we arrive at the conclusion that it must have had and this the same meaning in nine also. As ‘tonnir-u’ (‘tol-nir-u’), ninety, of com- ] iﬂtﬂ ‘Hj, means the ‘ tol’ (or defective) hundred, and as ‘tol-Ayiram,” nine hundred, and this means the ‘tol’ (or defective) thousand; so ‘onbad-u’ (¢ tonbadu’), nine must mean the ‘tol’ (or defective) ten. ; positiou We have here, doubtless, the primitive Dravidian mode of forming P, it 38 the higher compounds of which nine is the first member. The absence consonant of this idiom in the higher compound numerals of the Telugu and pint, Vit Canarase is an illustration of the greater antiquity of the Tamil; nd ‘ tom- whilst the formation of nine on this peculiar plan in both Telugu and . into Canarese shows that originally 1t was the common property of all the n which dialects. Its mode of forming the higher compounds corresponds to changes, the Scythian mode of forming eight and nine, as has already been o ‘aly pointed out. The higher as well as the lower compounds are formed : ‘ﬂﬂt])ﬂy’ in this manner in Ostiak ; e.g., from nit,” eight, and st a hundred, ‘Hﬂl}bu; is formed ‘nit sot, not only eight hundred, but also eighty. It cor- ;jol'pil(l'u / responds also to the use of ‘ln’ to denote diminution by one in the bty Sanscrit ¢ Gnavinshati,” nineteen, and the parallel Latin ¢ undeviginti.’ gord for It only remains to inquire into the origin and meaning of the , ‘fOI}(,I“):' prefix ¢ tol.” It is not to be confounded with ‘tol,” ancient, the <1’ o ‘toly of which belongs to a different ¢ varga:’ and yet that ‘1’ also supplies qve been us with several good derivations. Though ancient is the meaning of of every 5 “tol” in Tamil, it is used to signify jfirst in Telugu; e.g., ‘toli-viram,’ the first day of the week, and the meaning, first, might naturally flow ]z//Zfl"ﬂ”l’ ‘ from an earlier meaning, one, of which however no trace remains. of fﬁd fo b Another possible derivation is from ‘tol-a,” Tel. and Can., a b3, or mil com division (as in an orange or jack-fruit) ; and a better one still is ¢ tola- e ity ga,” Tel., an infinitive, which is used adverbially to signify of; e.g., :tollll I 286 THE NUMERALS. The objection to all these derivations is that from ¢ ‘tolaga tiyu,” to take of. the Tamil numeral prefix is not ‘tol, but ‘tol; and that those con- pound fenty sonants are most carefully discriminated in all Tamil dictionaries. There are two similar roots containing the surd ‘1’ in Tamil, viz., ‘bbhe “tol,” loose, lax, and ‘tul-ei’ or ‘tol-ei’ (Can., ¢ toll-u’), ¢o ollow, to by th would perforate ; as an adjective, perforated; e.g., ‘tollei kadu,” a long, pen- In dent, perforated ear. These meanings do not harmonize very well 1§ usu with the use of ‘tol’ as a diminuent prefix to the higher numerals, and yet it appears probable that the word is the same. We must, I8 whicl therefore, take refuge in the supposition that originally ‘tol” conveyed “hatt the meaning of deficient or diminished. It seems scarcely necessary now to add that there is no affinity [n tl of n whatever, as some have surmised, between the initial portion of the Tamil ‘onbadu’ and the Greek ¢<évvéa. The Manchu ‘onyan, nine, and s has not only some resemblance to the Dravidian word, but seems to be Iemaj Nevertheless the ulti- T a compound formed on similar principles. mate component elements of the Manchu word—¢emu,’ one, and ‘juan,’ Tume; ten—have no resemblance whatever to the Dravidian. Tt derivs Ten.—In all the Dravidian languages the words used for ten are uncon (R)Q virtually the same; in Tamil ‘patt-u,’ in Canarese ¢hatt-u,’ in Telugu “padi,” in Tuda ‘pota,” in Gond ¢ pudth.’ mo\og In those Tamil compound numerals in which ten is the second in, 1 member; e.g., ¢ irubadu,” twenty, ¢ pattu’ becomes ¢ padu’ (euphonically ot “badu’), which is apparently the oldest extant form of this numeral, Tecg] deciy and in close agreement with the Telugu ¢ padi.’ In the Tamil compound numerals under fwenty, in which ten con- ‘pady stitutes the first number, e.g., ¢ padin-ARru,’ sizteen, literally fen 4 siz ; 0 pag “pad-u’ becomes converted into ¢ pad-in,” the “in”’ of which I consider Tami] to be either an adjectival formative used as an euphonic augment, or have In ‘patton- an euphonic augment used as an adjectival formative. di‘d‘ec badu,” nineteen, pattu, which I regard as an adjectival form of Tesey ¢padu,” is used instead of ¢ padin’—the prefix of all the rest of the 1 teens in Tamil, except ¢ pannirandu,” fwelve. That is simply a cor- any ] ruption of ¢padin-irandu.’ from The Telugu ¢ padi,” tem, is evidently identical with the Tamil that ‘padu; as ‘adi, Tel., 4, is identical with the Tamil ‘adu.” The Telugu ¢ padi’ undergoes more changes in composition than its equi- A valents in the other idioms. In the compounds under twenty, in Tequj which ten is represented by ‘pad,” ‘padi, ‘pada,’ or ‘padd,’ the of a changes are trifling ; the principal being in ¢ pandhommidi” (‘ pan,” for Word “padin,’ ten, and ‘tommidi, nine), nineteen; but in the compounds I N TEN. 287 18 that from twenty upwards, in which ten is the second member of the com- 56 C0D- pound and is a numeral noun, ‘padi’ is materially changed. In maries, twenty and sixzty it is altered to vei; in thiréy to ‘phei; in seventy to 1, viz,, ‘bbhei; and in the other numbers to ¢ bhei.’ This change is effected llon, to by the softening of the ‘d’ of ¢ padi, after which ‘pa-i’ or ¢ba-i’ g por would naturally become “bei,” and then ¢ vei.’ 1y well In Canarese, ten is ‘hatt-u,” by the change of ‘p’ into ‘h,” which merals, is usual in the modern dialect: in the ancient dialect, as in Tamil, it ¢ s, is ¢ patt-u’ In the compounds from eleven to mineteen inclusive, in which ten is nsed adjectivally, and is the first portion of the word, uveyed “batt-u’ is represented by hadin,’ as ‘patt-u’in Tamil by ¢padin.’ affinity In the compounds in which ¢en is placed last, and is used as a noun of the of number (from twenty and upwards), ‘patt-u’ is found in twenty ) i and seventy, ‘batt-u’ in eighty and ninety, and ¢ vatt-u’ in all the remainder. s to be e ulti- The Tulu uses patt’’ for the noun of number, and ‘pad’’ as the ¢ jnan,' numeral adjective. The vocabulary of the Dravidian languages throws no light on the derivation of ‘pad’) the normal form of this numeral. It is quite fen, ate unconnected with ¢ parru,” Tam. (pronounced ‘pattru’), o receive, the ‘w’ of which is radical, and connected with ¢ pari,’ to catch. Ety- Telugu mologically, the nearest Tamil root to ‘pad-u’ is ‘padi,’ to be fized second wm, to be imprinted. The noun ‘ padi’ hence may mean anything that vs noted down, imprinted, or recorded; and the numeral Zen might have mically meral, received this name from the use to which it has always been put in decimal calculation. Another possible derivation is the Telugu ‘padu-vu’ or ‘padu-pu,’ @ collection, a crowd; e.g., ¢ kukka padu-pu, ), 001 48105 a pack of hounds. This word, however, is not recognised by the Tamil. The only analogy or resemblance to this numeral which I )nsidel‘ have observed in any other language, is in the ¢ Koibal,’ a Samoiede ent, O dialect, in which ten is ¢ bet.’ It seems improbable, however, that the patton— resemblance is other than accidental, seeing that none of the other om of numerals of that language, with the exception of ‘okur, one, bears of the q €O any resemblance to the Dravidian. It is only in the lower numerals, from ome to four, that any real affinity is to be found or looked for in that rude and distant Scythian dialect. Tamil ) The A Hundred. — The only cardinal number above ten which g eqlli- requires to be noticed in a Drividian Comparative Grammar is that rltya i of a hundred. In all the Dravidian dialects, without exception, this e word is ‘mur-u.’ aﬂ;’ fOT I have not been able to discover any resemblance to this numeral Pouﬂds 288 THE NUMERALS. in any other family of tongues. In no two Seythian stems do we find signif the same word used to express this high number ; nor indeed amongst of fir such rude tribes could we expect to find it otherwise. of ‘A One and the same word for Aundred, slightly modified, is used in itself When every language of the Indo-European family, a remarkable proof of the unity and ancient intellectual culture of the race; and the Finnish ticipial word for a hundred, ¢ sata, has evidently, like some other Finnish ‘modal words, been borrowed from that family of tongues. the po The Tamil has a verbal root ‘nfiru,’ Zo destroy, which is identical Texior and in sound with ‘nlru,’ @ hundred; and there is also a related root ‘nir-u’ (in Tel., ‘nlir-u’), to reduce to powder; but it cannot be sup- Eump posed that the numeral ‘niir-u’ proceeded from either of these roots. “prath A word which may be supposed to be the origin of it is ‘nfir,’ to keap allied up, the suppositious verbal root of ‘ nliril,’ Tamil, a Aeap. ultimat Tam,, A Thousand.—The Dravidian words for thousand are ‘ayiram,’ ‘mudy] Tam. and Mala.; ¢sivira,” and also ‘savara,” Can.; ¢ vélu,’ softened numer into ¢ veyyi,” Tel.; ¢sara,” Tulu. ‘sAvira’ or ‘¢ savara, and ‘sara,’ are 1001, evidently identical; and we may safely derive both from the Sans. the oo ¢ sahasra.’ Probably also the Tam ¢Aayira-m’ was originally ¢ asira-m’ these or ¢ Asra-m,” and therefore an old corruption of the Sans. A priory we (muJ i might have expected to find the Dravidian languages borrowing from { pra; ? the Sanserit a word for expressing this very high numeral. The Proper] Telugu word for thousand, ¢ vél-u,’ is a purely Dravidian word, and is of py The origin of ‘veyi’ the plural of ‘veyi’ or ‘veyyi’ (‘veyu-lu’). In does not appear; but I am inclined to connect it ultimately with the Upwer Tamil root ¢ ve,” to be excessive, to be hot, harsh, &e. tion of in Sueg OrDINAL NumBERS. — It is unpecessary in this work to devote suffy it much attention to the Ordinal numbers of the Dravidian languages, The seeing that they are formed directly, and in the simplest possible - mﬁq;{\- manner, from the cardinal numbers, by means of suffixed verbal N participles or participial forms. ﬂdding The only exception is that of the first ordinal, viz., the word signifying first, which in most of the Dréavidian languages, as in the certaiy Vulgay Indo-European, is formed, not from the cardinal number one, but iHHStra from a prepositional root. In the Canarese and Malayalam, the numeral one is the basis of the word used for first. Which, The base of the first ordinal in Tamil and Telugu is ‘mudal,” a verbal noun, signifying priority—in time or place, or a beginning. The Uggapg | This, like all other Dravidian nouns, may be used adjectivally with- 10 begg, out any addition or change; and therefore ‘mudal’ alone, though \'évadu" S ORDINALS. 289 e find signifying @ beginning, is often used as an ordinal number in the sense ongst of first. More frequently, however, it receives the addition iu Tamil of ‘4m, which is the usual suffix of the ordinal numbers, and is in el In itself an aoristic relative participle of the verb ‘4g-u,’ to become. of of When ‘mudal’ is used in Telugu without the usual ordinal or par- nnish ticipial suffix, it requires to be put in the inflected form; e.g., not nnish The verbal noun ‘mudal’ is connected with “modal,” but ‘modati.’ the post-position ‘mun,” Tam., before; so that there is the same con- ntical nexion between the ordinal number first in the Dravidian languages, | 100t and the post-position before, which is observed to exist in the Indo- 6 SUp- European languages between the preposition ¢pra, Sans., before, and roots, prathama,” ¢ mp7os, &ec., first. Though the Tamil ‘mun, before, is ) hetp allied to ‘mudal,” first, yet neither of those words exhibits the ultimate root. The ‘n’ of ‘mun’ appears in the verb ‘mundu,’ Tam., to get before; but it does not appear to have had any place in iram, ‘mudal;” of which ¢dal’ is a formative termination belonging to a frened numerous class of verbal nouns, and ‘mu’ alone is the (suppositious) 2, a8 root. Mudal,” though itself, T believe, a verbal noun, is also used as I have no doubt that all Sans. the root of a new verb, signifying to begin. oy these words and forms spring from ‘mu,’ as their ultimate base. 1 We ‘mu’ is evidently a word of relation, signifying like the Sanscrit , from ¢ pra,” priority; and with it I would connect ‘mi,’ Tam., ¢o be old, The properly ‘mu, as found in ‘mudu,’ antiquity, which is a species :md I of priority, viz., priority in tume. In all the Dravidian idioms, the other ordinal numbers from zwo ‘ \'eyi ’ th the upwards, are formed directly from the cardinal numbers by the addi- tion of formative suffixes. The same suffix is added to every numeral in succession, without change either in the cardinal number or in the suffix itself. Jevote The ordinal suffix of the grammatical Telugu is ‘va’ or ‘ava,’ eg., 19265 ;gsible ‘midava, ¢hird: the Canarese adds ‘ni’ or ‘ani’ to the cardinal numbers, e.g., ¢ miranf,” third : the ordinal of the Tamil is formed by \'el‘b i adding ‘4m’ to the cardinal; e.g., ‘mindram, third. The clear and yord certain origin of the Tamil suffix ‘Am’ from ‘4gum,” poetically and vulgarly ‘4m,” the aoristic relative participle of ‘agu,’ to become, i the g, bt illustrates the origin of the suffixes of the Telugu and Canarese, which, though considerably changed, are undoubtedly identical with ], the the Tamil in origin. (1'11)’ s The adverbial forms of the Drividian numerals are formed by means of another class of suffixes from the same auxiliary verb ‘agu,’ il r “I[b to become. In this instance the suffixes which are used by the Tamil, ‘avadu,’ &c., are neuter participial nouns used adverbially. Often- ibou'gh U 290 THE NUMERALS. . times, however, adverbial numerals are formed by the addition of nouns {ha il signifying succession, &e., to the cardinal or ordinal numbers; e.g., ‘iru- murei,” Tam. twice, literally two times. belo The multiplicative numbers, as has already been stated, are the same as the numeral adjectives. lste It only remains to inquire what evidence respecting the affiliation fofc of the Drividian family of tongues is furnished by the preceding inves- Blan tigation of the numerals of that family. nls s The evidence is not only decidedly opposed to the supposition that the Drividian languages are derived from the Sanserit, but is equally un inconsistent with the supposition of the connexion of those languages thos with the family to which the Sanscrit belongs, either as a member of that family, or even as a remote offshoot. Not the smallest trace of resemblance has been discovered between the Dravidian numerals and those of any Indo-European language, with the exception of the resemblance of the Telugu oka,’ one, to the Sans- crit ‘éka,’ as well as to the Ugrian ‘og, ‘ak,’ and ‘okur; and in that instance I have no doubt that the Sanserit itself has inherited a Scythian numeral, the numeral for one of the Greek, Gothic, Celtic, &c., being derived from a different base. All the other numerals of the Indo-European languages spring from one and the same root, and are virtually identical ; with the soli- tary exception of the Gaelic word for jfive ; and hence, when we find in the Dravidian numerals no resemblance to those of the Indo-Euro- pean tongues, with the exception of the abnormal Sanscrit ¢ éka, we are compelled to conclude that the Dravidian languages belong to a different stock from the Indo-European. On the other hand, a comparison of the Drividian numerals with those of the Scythian tongues appears to establish the fact of the exis- tence of Scythian, and especially of Ugrian, or Finnish, analogies in the Dravidian family. The resemblance between the Dravidian one and four, and the corresponding numerals in the Ugrian languages is so complete, that we may justly regard, and cannot but regard, those numerals as identical. The same statement applies to the word for one, which is found in the Scythian version of Darius’'s Cuneiform inscriptions at Behistun. The numeral four, and the other numerals above one, are not contained in that unique relic of the ancient Scythian speech of Central Asia ; and in this case the negative argument concludes nothing. It may perhaps be thought that the resemblance of only two nume- rals (one and four) out of ten, cannot be considered to prove much ; but it is to be borne in mind that this resemblance is all, or nearly all, s AFFILIATION. 291 IIO.KIHS that is observed in the Scythian languages themselves between the i numerals of one language of a family and those of other languages belonging to the same family. re the Thus, it cannot be doubted that the Magyar and the Finnish are sister tongues, essentially and very closely allied ; and yet with respect iation to four numerals, seven, eight, nine, and ten, no distinet trace of resem- 1nyes- blance between them survives ; and it is only in the case of the nume- rals one, two, and four, that it can be said, without hesitation, that the n that same root appears to be used in both languages. The Dravidian qually numerals are therefore almost as closely allied to the Finnish as are ;uages those of the Magyar itself. ber of tween 3 with Sans- nd 1o ited 2 eltic, spring 6 SOH- ¢ find Juro- ;3; W gtoii g Wlth e ity gies n i one ﬂges iS " those [Iﬂd n jstun. fained Agia) pué mﬂfh ) r[y {Lll s v 2 Inth ‘D, quial ) appes singu these SECTION V. its In done this poud of th THE PRONOUN. leads In MucH light is thrown by the pronouns on the relationship of languages and families of languages ; for the personal pronouns, and especially extan those of the first and second person singalar, evince more of the and ¢ of the quality of permanence than any other parts of speech, and are gene- the in rally found to change but little in the lapse of ages. They are more in the permanent even than the numerals, the signs of case, and the verbal inflexions : and though, like every thing else, they are liable to change, Sppo High yet their connexions and ramifications may be traced amongst nearly all the languages of mankind, how widely soever sundered by time or of th hiau place. iming In some instances the personal pronouns constitute the only appre- { ciable point of contact or feature of relationship between languages nad of m which belonged originally to one and the same family, but which, in the lapse of time, and through the progress of corruption, have become word natur: generically different. This remark especially applies to the pronoun of the first person, Clmst which of all parts of speech is the most persistent. Verte( Ch&ng( I.—Personal Pronouns. i deg 1. ProNoUN oF THE FIrsT PERSON SINGULAR. the l chang' Comparison of dialects.—Our first inquiry must be ‘What was the primitive form of this pronoun in the Dravidian languages?’ | In Tamil the form which is used in the colloquial dialect is ‘nan,’ ﬂexi(m Dy the inflexion of which is not ‘nan,” as might have been expected, but g j ‘en ; and this inflexion ‘en’ indicates the original existence of a Casey 9 Caseg nominative in ‘én. Though “én’ is no longer found in a separate 0 shape, it survives in the inflexions of verbs; in which the sign of the 7 Oﬂg first person singular is ‘én, sometimes poetically shortened into ‘en.’ Ity dis(‘w FIRST PERSON SINGULAR. 293 In the higher dialect of the Tamil ‘y4n’ is more commonly used than ‘nan,” the inflexion of which is not ‘yan,” but ‘en,’ as in the collo- quial dialect.* From the examples which have been adduced above, it would appear that there are three forms of the pronoun of the first person singular recognized in Tamil, viz., ‘nan,” ¢ yan,” and “én.’ The first of these forms, though the most common, was probably the primitive one: its initial ‘n’ was first, I think, softened into ¢y, and finally aban- doned. It is not so easy to determine whether the included vowel of this pronoun was originally ‘4’ or ‘é A comparison of the corres- ponding plurals ‘ném,” ¢ ydm,” and ‘em’ (the inflexion of ‘ém’), and of the plural terminations of the verb, ¢ 6m, 25 6GA anm ‘am’ and “em, leads to the conclusion that ‘4’ was most probably the original vowel. In the singular, ‘en’ is the only inflexion of this pronoun which is uages extant in Tamil; but in the plural we find not only ¢ em,” but also ‘nam’ ecially 9 and ¢ yam. Though ‘nam’ is most frequently used as the inflected form of the of the isolated pronoun (e.g., ‘namar,” they who are ours, ‘nostrates’), 6 goIe the initial ‘n’ has altogether disappeared from the corresponding form ) 0016 in the pronominal terminations of the verb. At first sight we might verbal suppose ‘nam’ and ‘nem’ to be the pronominal terminations of the h‘(ﬂlge,' High Tamil ‘nadandanam,” or ‘nadandanem, we walked ; but the ‘n’ peatly of these terminations is merely euphonic, and is used to prevent {img 0f hiatus. When it is omitted, the vowels which it had kept separate immediately coalesce; e.g., ‘nadanda-am’ becomes ‘nadandam’ and ap ‘ nadanda-em’ becomes ‘nadandém ; a more common form than either uages The final ‘6m’ of this of which, but not so correct, is ¢ nadandém.’ ﬂiCha iﬂ word could not well have been corrupted from ¢ém,” but would spring hecome b naturally enough from ¢&m ; and of this we have a proof in the cir- cumstance that ‘4m’ (from ¢Agum,” 4t s, yes) is also sometimes con- _gersonr verted into ¢om.’ Moreover, whilst there are many instances of the change of ‘a’ into ‘e’ or ‘ei,” there is not any of the converse. It is deserving of notice also, that in this change from the heavier ‘a’ to the lighter ‘e, the Dravidian dialects exhibit the counterpart of the change of the corresponding Sanscrit pronominal root ‘ma’ into ‘éué,’ 748 the * In explanation of the abbreviated form of the pronoun called ‘the in- flexion,” which has been referred to above, it may here be repeated that in the ; {Il‘:m)’ personal and reflexive pronouns of the Tamil, Canarese, Malayilam, and Tulu, and in the reflexive pronoun of the Telugu, the ¢inflexion,” or basis of the oblique ol Dt cases (which by itself denotes the genitive, and to which the signs of all other 30“ cases are suffixed), is formed by simply shortening the long included vowel of the nominative. The included vowel of each of the personal pronouns is naturally epgf& f long ; and if in any instance the nominative has disappeared whilst the inflexion f 8 remains, we have only to lengthen the short vowel of the ‘inflexion,” in order to o) discover the nominative from which it was derived. 294 THE PRONOUN. In me, &c., in other dialects of the same family. A I conclude, therefore, that ¢ a > was originally the included vowel of the Tamil pronoun of ‘o) W In the first person, and that ¢nén,” the ordinary colloquial form of the higher pronoun, is the most faithful representative of the primitive Tamil 7. nomin As we proceed in our comparison of the various dialects, it will be (anare found that the evidence is cumulativeand gathers strength as we proceed. in the It might appear, indeed, at first sight that ¢ yin’ was an older form ‘06’ 1 than ‘nan ; but before our investigation is concluded, we shall be con- vinced, I think, that the ‘n’ is radical. ‘n ’ is known to change into and 1 alone, “y; but ‘y’ evinces no tendency to be changed into ‘n.’ Th In Malayilam, the nominative is ‘fijan’ (‘ny,” ‘jii, or ‘iij, the nasal of the palatal ¢varga,’ is to be pronounced as one letter, like the the no '’ ( ‘ni’ of onion); but the oblique form, or inflexion, is ‘en,’ as in Tamil, import except in the dative ‘inikka,’ in which ‘en’ is altered to ‘in.” The and th ordinary Malayala verb is destitute of personal endings: but in the ab mog poetry an inflected form of the verb is occasionally used, in which the pronominal termination of the first person singular is ‘én,” precisely as ‘né shows in Tamil. as well The compound sound of ‘iij’ or ‘ny,” in the Malayala ‘fijan’ or ‘nyan, is a middle point between the ‘n’ of ‘nin,’ and the ‘y’ of accusy (comps ‘yan, It is a softened and nasalized form of ‘n, from which the change to ‘y’ is easily made. In like manner, ‘nin,’ the original eviden Th form of the pronoun of the second person singular in all the Drévidian dialects, has become in Tamil, first ‘niy,” then ‘ni; and in the verbal Ancie terminations ‘aiy,’ ‘i, and ‘ei.’ ol din T In Canarese, the nominative of this pronoun which is used in the colloquial dialect is ‘nén-u,” as in Tamil, the inflexion of which (as tense, seen in ‘mnanna,’ my) is ‘nan. The ancient dialect uses ‘4n,’ the In inflexion of whieh is ¢ en’—identical with that of the Tamil. ‘4n’ is Teluo‘u evidently softened from the Tamil ¢ yén,’ as ¢yin’ from ‘nyin,’ or Torg | ‘fijan,’ and that from ¢pan’; and the same softening is apparent in the Proper Canarese plural ‘4m’ (instead of ‘yam’ or ‘nam’), we. The crude In form of this pronoun (‘na’) is sometimes used in Canarese as a Brahy; nominative, instead of ‘nanu; e.g., ‘na bandenu,’ 7 came; and in the kane same manner in Tamil, ‘ni,’” the crude form of ‘nin,’ thow, has altogether Mtaﬂy superseded ‘nin.’ The pronominal terminations of the first person areq singular of the Canarese verb are ‘en’ in the ancient dialect, and ‘éne, (1\11,‘ ¢énu, and ‘enu’ in the modern. I}t The Tulu nominative is ¢ yan,’” the inflexion ¢yén,’ the pronominal fﬂ\mm ending of the verb ‘e,” which is probably softened from ‘en.’ EXf) The Tuda nominative is ‘6n’ (plural ‘6m’), of which ‘en’ is the ang ¢ inflexion ; the singular terminations of the verb are ‘ én’ and ini.’ —— - .y S B, FIRST PERSON SINGULAR. 295 In the dialect of the Kotas the nominative is ‘4ne,’ the inflexion efore, un of ‘en,’ and the pronominal ending of the verb ¢e,’ as in Tulu. of the In Telugu the nominative of this pronoun is ‘nén-u;’ in the higher dialect ‘én-u’ (answering to ‘én,” the Tamil-Canarese pro- mil /, il be nominal ending of the verb, and ¢en, the Tamil and Ancient oceed, Canarese inflexion); and this preference of ‘&€’ to ‘4’ appears also in the plural, which is ‘mém-u,’ and in the higher dialect ‘ém-u.’ 1 form )6 C0l- ¢ né’ may be used at pleasure instead of ‘nén-u; like ‘n4’ in Canarese; ge info and in the higher dialect ‘én-u’ is sometimes represented by ‘é alone. The verbal inflexions of the Telugu, use only the final syllable of j, the ke the the nominative of each of the pronouns, viz., ‘nu’ (from ‘nénu,’ 7), ‘yu’ (from ‘nivu, thow), and ‘du’ (from ¢véadu,’ he). The most Tomil, important and essential part of each pronoun has thus been omitted ; " The 1 the and the fragments which have been retained are merely formatives, or ich the at most signs of gender and number. ‘nénu,’ 7, takes ‘na’ for its inflexion or oblique form; and this sely a9 C A2 e was shows that ¢4’ not originally the included vowel in Telugu, as well as in Tamil and Canarese. This view is corroborated by the i’ or accusative of this pronoun in Telugu, which is ‘nanu’ or ‘nannu,’ me, ‘y’of (compare the Canarese accusative ‘nanna’ or ¢ nannu’), and which has ich he evidently been derived from a nominative, ¢ nan’ or ‘na. )riginal The Ku nominative is ¢ An-u, which is identical with that of the yidian Ancient Canarese. In the inflexion, which is ‘na’ as in Telugu, the verbal g old initial ‘n’ retains its place. The verbal terminations of the Ku are “in”’ or ¢ in, in the present in the tense, and ‘ e’ in the past; e.g., ‘méin,’ I am, ‘ masse,” I was. ich (58 In Gond the nominative is ‘and,” and the inflexion is ‘nd,’ as in n,’ the Telugu and Ku. In the verbal inflexions ¢&n’ is commonly found, ‘{m’is more rarely ‘na.’ The Seoni Génd nominative is ‘nék,” which is ﬁn; or properly a dative. 1n the In the Rajmahali the nominative is ‘en; in Uréon ‘enan.” The ) c]fllde Brahui nominative is ¢ i; but in the oblique cases (e.g., ‘ kand,” of me; g “kane, me, to me), the personal base is ‘ka’ or ¢ kan,” a root which is in the totally unconnected with the Dravidian ‘na, and which is to be com- )gethef pared rather with the Cuneiform, Scythian, Babylonian, and Gujarathi pel‘SOﬂ S Chn,’ &c. ‘éney From this comparison the weight of evidence appears to be in favour of our regarding ‘nin, the Tamil nominative, as the best oitd existing representative of the old Dravidian nominative of this pronoun, and ‘na,” the crude form of the Canarese, as the primitive, unmodified is the ol D —— — o 296 THE PRONOUN. merel L root. This conclusion will be found to gain strength from the inves- tigation of the pronoun of the second person, the root of which will gxtrac 1 A nitia appear to be not ¢ or ‘ yi,” but ‘ni Each consonant of nén’ evinces a tendency to be softened away. gethe The initial ‘n,” though the more essential of the two, has been the fi g softened first into ¢dfij’ or ‘ny, then into ‘y,’ and finally has disap- peared ; and in none of the dialects has it, or any relic of it, been prono utter retained in the personal terminations of the verb. The final “n,” though not a part of the root, has shown itself more refer persistent, especially in the verbal terminations; but in the Telugu othe g ¢ and Ku inflexion ‘na,” in the Canarese crude nominative ‘n,” and in the corresponding Telugu ‘né,’ it has disappeared altogether. full £ The origin of the final ‘n’ of ‘nan’ is doubtful; but whatever term were its origin, it does not appear to belong to the root. In the Inent plural, it is uniformly rejected, and “ m,” the sign of plurality which is stil used in connexion with the personal pronouns, is not added to it, but Telu substituted for it. In Tamil the singular is ‘ nd-n,” the plural ‘na-m ;’ Is > 2 and a similar change from ‘n in the singular to ‘m in the plural, addi takes place in the other dialects also. This appears to prove that ‘na’ pea alone forms the pronominal base of both numbers of the pronoun of term the first person; that it denotes either 7 ‘or we according to the h&\ { singularity or plurality of the suffixed particle (‘na-n,’ 7 alone, ‘na-m,’ that Is); and that the final ‘n” of ‘nan,’ no less than the final ‘m’ of ‘nm,’ e 1s a sign, not of personality, but merely of number. i Is the final ‘n’ of ‘nén’ a sign of gender as well as of number 7— Is it a sign of the masculine singular, and connected with “an’ or ‘n,’ of ¢ the ordinary masculine singular suffix of the Tamil? The pronouns of of ¢ ; the first and second personsare naturally epicene; but it is not unusual for to find them assuming the grammatical forms of the masculine. Thus, the in Sanscrit, the terminations of the oblique cases of the pronouns of tha i the first and second persons are those which are characteristic of the sing masculine gender. The analogy of the Scythian pronouns, however, the inclines me to the supposition that the final ‘n’ of the Dravidian ar ‘nén’ was not in its origin a sign of gender or a means of grammatical hay expression, but was merely euphonic, like the final nasal of the Tartar Pl ‘man,’ /. Whatever were its origin, it must have had a place in the personal and reflexive pronouns from a very early period, for we an find it in the Brahui ‘ten,’ seff (Compare Tam. ‘tin’), and in the ng Ostiak ‘nyn,’ ¢hou (Compare Old Dravidian ‘nin).’ for If, as we have seen, ‘na’ is to be regarded as the primitive form 10 of the Dravidian pronoun of the first person, and the final ‘n’ as In¢ th FIRST PERSON SINGULAR. 297 ) nves- merely a sign of number, or as an euphonic formative, it might appear ich will extraordinary, that in the pronominal terminations of the verb, the initial ‘n,” the primitive sign of personality, has invariably and alto- | anay. gether disappeared ; whilst the first person singular is represented by 15 been the final ‘n’ alone. We might almost be led to suppose the initial s disap- ‘n’ to be a formative prefix, and the succeeding vowel to be the real f, been pronominal base. Formative and definitive pre-fixes, however, are utterly unknown to the Dravidian languages; and the anomaly If more referred to accords with similar anomalies which are discovered in Telogu other languages. In Hebrew, ¢ anachnu,” we, from ¢ anach’ (in actual and in use ‘andki’), 7, with the addition of ‘nu,” a sign of plurality, is the full form of the plural of the first personal pronoun; yet in the verbal atever terminations, ‘anachnu’ is represented solely by ‘nu,’ the final frag- In the ment, which originally was only a suffix of number. Another and hich is still more reliable illustration of this anomaly is furnished by the it, but Telugu itself. The pronoun of the second person singular in Telugu 3-m ;) is ‘nivu,” thou, from ‘ni,’ the radical base, and ‘vu’ an euphonic : lural, addition. This ‘vu’ is of so little importance that it totally disap- ‘i’ pears in all the oblique cases. Nevertheless, it forms the regular yun of termination of the second person singular of the Telugu verb; and it o the has acquired this use simply through the accident of position, seeing n4-m, that it is not even a sign of number, much less of personality, but is pam, merely an euphonisation. op {— Eaxtra-Dravidian relationship.—W e now enter upon a comparison of ‘na,’ the Dravidian pronoun of the first person, with the pronouns r'n, ns of of the same person which are contained in other families of tongues, pstal for the purpose of ascertaining its relationship. As ‘na’ constitutes the personal element in ¢ nm,” we, as well as in ¢ nan,’ 7, it is evident [hus, s of that our comparison should not be exclusively restricted to the singular, but that we are at liberty to includ¢ in the comparison, [ the the plurals of this pronoun in the various languages which are com- ovel, dian pared; for it is not improbable & priori that some analogies may have disappeared from the singular, which have been retained in the tical riar plural. All pronouns of the first person singular that have been used at the we any time in Asia, Europe, or Northern Africa, whether it be in con- the nexion with the Indo-European, the Semitic, or the Scythian family of tongues, are traceable, I believe, to two roots only. Kach of those il roots has been preserved in the Sanscrit, and in the more primitive members of the Indo-Kuropean family—one (‘ah’) in the nominative, 5“ the other and by far the more widely prevalent one (‘ma’) in the 298 THE PRONOUN. O 1l oblique cases. In order, therefore, to investigate thoroughly the affiliation of the Drividian promoun of the first person, it will be Finnis necessary to extend our inquiries over a wider area than usual. [ and If (1.) Semitic analogies. — The Semitic pronoun presents some be un remarkable analogies to the Dravidian. This will appear on com- (c paring the Dravidian ‘ni’ (Gond. ‘and,’) with the corresponding a 801 Hebrew ¢ani;’ with the prefix ‘an’ of the Hebrew ¢&ndki,’ of the ‘am, Egyptian ‘anuk, and of the Babylonian ‘anaku,’ ¢anaka,’ or ‘y-a ‘anku;’ and especially with the Jewish-Syriac ‘ané,’ the Christian- he s Syriac ‘eno,” and the Althiopic and Arabic ‘ani The plural of the and s Aramaic ‘and’ is formed by suffixing ‘n’ (the final consonant of nasal ‘in’ or ‘4n’): we may therefore compare the Tamil °nam,” we, prefer with the Aramaic plural ‘anin,” and also with the Egyptian plural termt ‘anen.’ nstes Notwithstanding this remarkable resemblance between the Semitic perso pronoun and the Drividian, it is doubtful whether the resemblance is espec not merely accidental. The Semitic initial syllable ‘an,” in which of the the resemblance resides, is not confined to the pronouns of the first T person. We find it not only in ‘ana’ (from ‘anah,’ and that again Tegan from ‘anah,’) 7; but also in the Arabic and Old Hebrew ‘anta’ and allied the Aramaic ‘ant,” thou, (Egyptian ‘en-tek, ‘en-ta’). The prefix petso being precisely the same in both cases, the pronoun of the second ke person seems to have as good a claim to it as that of the first. It does We not seem, moreover, to be an essential part of either pronoun; for we ultiy find a similar prefix in the third person in some of the Semitic dialects, deng e.g., in the Egyptian ‘entuf,’ ke, ‘entus,” she, and the Chald. and Vowe Heb. suffix ‘enhu,’ Ze. Moreover, the alliance of the Semitic pronouns Secon of the first and second persons with the Indo-European comes out When the the ] into more distinct reliefl when this prefix is laid aside. been initial ¢an’ is removed from the pronoun of the first person, we cannot Dety doubt the connexion of the remaining syllable (‘oki,’ ¢ah, ‘ah,’ ‘uk,’ the | ‘aku,” or ‘ak,”) with the Sanscrit ‘ah,’ the Gothic ‘ik,’ and the Of P( Greek-Latin ‘eg:’ and itis equally evident that when ‘an’ or ‘en’ is betw‘ rejected from the pronouns of the second person (‘antd,” ¢anti, ‘ant,’ eIy <entek,” ‘ enta,’) the ‘ta, ‘ti,” ‘te, or ‘t,” which remains is allied to the Sanscrit ‘tva’ and the Latin ‘tu. | It has sometimes been supposed that this Semitic prefix ‘an’ is that simply euphonic—a sort of initial numnation like that which is the admitted to exist in the Talmudie ¢inhu,’ Ae, when compared with 0ng the ordinary and undoubtedly more ancient Hebrew ¢hi.’ On this supposition, it is allied, in nature and origin, to the euphonic suflixes lang Pron FIRST PERSON SINGULAR. 299 ly the or nunnations which may be observed in the Greek ¢éyw-vy,’ in the vill be Finnish ‘mi-na, 7, and in the final nasal of the North—Indian ‘main,’ 1, and ‘tain’ or ‘tun, thou. §0me If this be the origin of the Semitic prefix ‘an,’ it must certainly be unconnected with the Dravidian ‘ni’ or ‘ani. I Com- Colonel Rawlinson supposes ‘an’ to be a particle of specifieation, onding a sort of definite article; and he also considers it 1o be identical with of the ‘am, the termination of the Sanscrit personal pronouns ‘ah-am,” 7 (g, OF ‘tv-am, thou, ‘ va-y-am, we, ‘yu-y-am, you. The only difference, ristian- he says, is that the particle is prefixed in the one family of languages, | of the b4 and suffixed in the other, with a change of ‘m into its equivalent ant of nasal ‘n. I am unable, however, to adopt this supposition, and A prefer to regard the Sanscrit termination ‘am’ as the ordinary plural termination of the nominative of the neuter singular, and as used instead of the masculine and feminine, simply because of the intense Somitie personality which is inherent in the first and second personal pronouns, ance 18 especially in their nominatives, and which renders the terminations which of those genders unnecessary. o first The only remaining argument which ean be adduced in favour of again regarding the Dravidian ‘n&’ and the Semitic ‘an’ as ultimately ' and allied, is the following. In the Semitic lunguages the first and second preﬁx personal pronouns have one element in common, the prefix ‘an.’ In seoond like manner, when we compare the Dravidian ¢ nd,’ 7, and ‘ni,” thou, t does we find that they also have one element in common, the initial and 2 for we ultimate base ‘n If it can be supposed that this initial consonant alects, denotes personality in general, and that it is the office of the succeeding , and vowel to inform us whether the person referred to is the first or the nounS second, then an ultimate connexion may be conceived to exist between 5 00t the Drévidian ‘n’ and the Semitic ‘an:’ for whatever may have 1l the been the origin of the latter particle, it appears to be wsed like the qnnob Dravidian ‘n,” as a sign of personality in general, and to constitute 1k, the basis to which ‘ah,” the sign of person No. 1, and ta,” the sign ] the of person No. 2, are suffixed. Probably, however, the resemblance en’ 8 between the Semitic and the Dravidian languages in this point, though anty remarkable, is altogether accidental. llied (2.) Indo-European analogies.—It has already been remarked n’iS that there are but two pronouns of the first person singular known to the Indo-European family of tongues, as to the Semitic and Scythian, h s yith one of which appears in the nominative of the older Indo-European languages, the other in the oblique cases. The nominative of this ths s pronoun is ‘ah-am’ in Sanscrit, ‘ad-am’ in Old Persian, ‘az-em’ in 300 THE PRONOUN. Zend, ‘eg-o’ in Latin and Greek, ‘ik’ in Gothic, ‘ih’ in the Old and ‘b German, ‘az’ in the Old Sclavonie, ‘asz’ in Lithuanian, and ¢ g’ (erma in Bohemian. On We find substantially the same root in the Semitic €ah, ‘ah, ‘uk, ‘aku,’ ¢ 06ki’ &ec., and in several languages of the cited, Malayo-Polynesian group; e.g., Malay °4&kd,” Tagala ¢aco, Tahitian ? favour 4 au It is evident that there is not the smallest resemblance between yiation any of these and the Dravidian ‘na.’ But though the Indo-European lertve: nominative has no connexion with the Dravidian pronoun, we shall langu: probably be able to establish the existence of some connexion between vowel ) the Dravidian pronoun and the base of the obligue cases in the Indo- site pr European languages. change The oblique cases of this pronoun in the Indo-European family are 4 Fin formed from a totally different base from that of the nominative; and forms, of this oblique base the best representative is the Sanserit ‘ma.” ‘m> L s forms the most prominent and essential portion of ‘ma;’ and this Wi “m is followed either by ‘a’ or by some vowel which appears to the D) have been derived from it. In the oblique cases of the Sanscrit, this I: pronoun has the form of ‘ma,” whenever the nature of the succeeding 7 18 der syllable allows ‘a to remain unchanged; eg., ‘ma-yi, n me, nasal ¢ ma-ma,’ of me. In the secondary forms of the dative and the genitive Eump it becomes ‘ma.’ In Zend and Old Persian, ‘ma’ preponderates; Ance whilst compounded and abbreviated vowels appear in the Zend dative- dialeg { ) genitives ‘meé,” ‘moéi;’ and a pronominal base in ‘ama’ is found in < i) I\ some of the Old Persian prepositional compounds. In the Greek ME natio “éué, poty ¢ mod, &ec., the vowel which is employed librates between t9. ‘e’ and ‘o, each of which is naturally derived from ‘a;’ whilst the 13y Indo initial ‘e’ of ¢ €ue is in accordance with the tendency of the Greek to ’I)Zg/’ f] prefix a vowel to certain words beginning with a consonant, e.g., been ‘ovopa’ for ¢ vwpa’ The Latin has ‘me,” except in the dative, which ( is ‘mihi.’ The Gothic has ‘mi’ and ‘meil.’ The Lithuanian uses my’ of h ‘man’ as the basis of its oblique cases; though possibly the final ‘n’ and of this form belongs properly to the sign of the genitive. Tafed In the pronominal terminations of the verb in the Indo-European aq | languages, the first person singular almost invariably makes use of “ada¢ this oblique pronominal base, in preference to the base of the nomina- () tive, with such modifications as euphony may require. The termina- by Ry tion of the first person singular is ‘mi’ or ‘m in Sanscrit and Zend, Gy in all primary and secondary verbs. We have the same ending in B, Greek verbs in ‘u’ and in the ‘uac’ of the middle voice; in the | k] ‘m’ of the Latin ‘sum’ and ‘inquam; 9 in the Lithuanian ‘mi;’ in lay the Polish ‘am;’ in the Armenian ‘em; in the New Persian ‘am.’ | IHd( It becomes ‘m’ in the Old High German ‘gim,” 7 go; ‘tuom,” Zdo; ( m J ST ————— - ? FIRST PERSON SINGULAR. 301 o 0ld and ‘bim’ or ‘pim’ (Sansc. ‘bhavami’), I am; converted in Modern d AN ga German to ‘bin. Nemitic On comparing the pronominal terminations which have now been of the cited, it is evident that the preponderance of use and authority is in ahitian favour of ‘mi;’ and that ‘m’ has been derived from ‘mi’ by abbre- otween viation. It is equally clear, however, that ‘mi’ itself has been ropean derived from ‘ma, the normal base of the oblique cases; for in all e shall languages ‘a’ evinces a tendency to be converted into some weaker tween vowel, ‘i, ‘e, or ‘o;’ whereas no instance is adducible of the oppo- Indo- site process. Perhaps the best illustration of the regularity of this change from ‘ma’ to ‘mi’ is that which is furnished by the Esthonian, ly are a Finnish dialect, in which each of the personal pronouns has two ; and forms, the one primitive, the other euphonised ; e.g., ‘ma’ or ‘minna,’ {m) I; ‘sa’ or ‘sinna, thou. d ﬂlla We have now to inquire whether anyanalogyis discoverable between ars 10 the Dravidian ‘na’ and the ultimate Indo-European base ‘ma.’ , this I am inclined to believe that these forms are allied, and that ‘na’ oding is derived from ‘ma.- A change of ‘m’ into ‘n’—of the stronger L ey nasal into the weaker—is in accordance both with Dravidian and Indo- nitive Thus ‘am, the accusative case-sign of the European precedents. 1ates; Ancient Canarese is weakened into ‘an’ (‘annu’), in the modern ative- dialect ; um, the conjunctive or copulative particle of the Tamil, is nd ‘nu’ in Telugu; and even in Tamil itself, ‘am,” the formative termi- (el ey nation of a large class of nouns, is optionally converted into ‘an ; In the teen e.g., ‘uran, strength, is used by the poets:instead of ‘uram.’ it the Indo-European family, in like manner, the change of ¢ mama,” Sans., ok 10 my, into ‘mana’ in Zend, and ‘mene’ in Old Slavonian, has already been noticed ; but proofs exist also of the special change of the ‘m’ of ] f,g ) ihich ‘ma’ itself—the most radical part of ‘ma’—into ‘n. The final ‘m’ 11865. of the first person of Sanscrit and Latin verbs (the abbreviation | ‘n and representative of ‘mi’ or ‘ma’) has in some instances degene- rated into ‘n t in Greek ; e.g., compare the Sanscrit asam,” I was, and the corresponding Latin ‘eram,” with the Greek ‘’jv; and )peaﬂ o of ¢ adada-m’ with €éél-w-v. We see a similar change of ‘m’ into ina ‘n,’ on comparing the Irish ¢chanaim, 7 sing, with the Breton 1nd- ‘kanann; the modern German ‘bin,” 7 am, with the Old High k) Zend, German ‘bimM’ or ‘pim ; and the Persian ‘hastam,’ 7 am, with the Beluchi ¢hastjan.’ Compare also the Laghmani ¢ pakan,’ 7 go. g in the The ‘n’ which alternates with ¢v,” as the initial and radical conso- nant of the plural of the pronoun of the first person in many of the 1 Indo-European languages, has been derived, I conceive, from the same all B Rl O R B “m’ (] It was shown in the section on ¢Sounds,’” that the Dravidian S o 302 THE PRONOUN. {hink ‘m’ is sometimes euphonically degraded either into ‘n’ or into ‘v and that whenever ‘n’ and ‘v’ are found to alternate, we have ‘i’ 2 former reason to consider both to be derived from an older ‘m. In like manner the ‘va’ of the Sanserit ‘va-(y)-am,” we, and the ‘na’ of ‘nas,’ the secondary, oblique form of the same pronoun, appear to be (3 of the mutually connected ; and both have probably been derived from ‘ma.’ The oldest form of the plural of this pronoun is that which is employed above in the verbal inflexions, and which in Sanserit is ‘mas,” in Latin ¢ mus,’ pende in Greek ¢ uev’ (for the more ancient and more correct Aolic ¢ pes’): broug 1 the most natural explanation of which pronominal ending is to consider (2965 it as derived from ‘ma,’ the old first person singular, by the addition of ‘s, the sign of plurality. The ‘m’ of this primeval ‘mas’ some- the ¢ 10min times becomes ‘v,” eg. in the Sanscrit ‘vayam, the Zend °vaém, and the Gothic ¢ veis ;’ and sometimes also it becomes ‘n,’ e.g. in the stanti in al Latin ‘nos,” the Welsh ni,’ the Greek ‘vac; and also in the Sans- crit secondary forms, ‘nas’ and nau,” the Zend ‘né,’ and the Old oven | Slavonic ‘nas’ This ‘n’ is evidently a weakening of ‘m,” and repre- into t sents the personality of the prononn of the first person, irrespective of the N the idea of number ; which is expressed by the subsequent portion of in ex the word. This being the case, we seem to be warranted in consider- opera ing it as identical in origin with the ‘n’ of the Dravidian ¢na’ (as ap- Turki parent in the singular ‘né-n,’ and the plural ‘ni-m’). by It has been suggested by Col. Rawlinson that the Sanscrit ¢nas,’ 868 the Latin ‘ nos,’ and the Greek ‘v&.” (like the ‘nu’ of the Hebrew has ‘anachnu’), were originally signs of plurality, which have made gran themselves independent of the bases to which they were attached. I and am unable, however, to adopt this view : for the ‘n’ of these forms their naturally interchanges with ‘m,” and evidently conveys the idea of hand ’ personality; and the ‘s of the Latin ‘nos’ (as of the correspond- tong ing ‘vos’) is more likely to be a sign of plurality than an abbrevia- b’ tion (as Bopp conjectures it to be) of the syllable ¢sma.’ Suffiy It may here be mentioned, as some confirmation of the supposition 1 that the Drévidian ‘ na’ is derived from an older ‘ma,’ that in Telugu Pron ‘m’ is used as the equivalent of ‘n,” and as the representative of the fom personality of the pronoun of the first person in the plural ‘mém-u,’ Quenf we. | The second ‘m’ of this word is undoubtedly a sign of plurality ; c\xan; and though the first ‘m’ may possibly be derived from ‘n,’ through 1 the attraction of ‘m,” the sign of the plural, yet this change would byt more naturally take place, if an initial ‘m’ had originally been used. San On this supposition ‘mém-u’ corresponds to the Sanserit ¢ mas ; and Ill&) has been weakened into ‘ném-u’ or ‘nim,” in the same manner as thig ‘mas’ has been weakened into ¢nas. On the whole, therefore, I FIRST PERSON SINGULAR. 303 It ‘y 2 think we are warranted in coming to the conclusion that the Dravidian 76 have ‘na’ and the old Indo-European ‘ma’ are allied, and, if so, that the Tnlike former has been derived from the latter. ‘i’ of ar to be (8.) Scythian analogies.—When we examine the personal pronouns mm, of the Scythian group of tongues, the views which have been expressed above are found to be corroborated : in addition to which, some inde- mployed ‘mng, pendent and very interesting analogies to the Dravidian pronoun are brought to light. J yfs"’): consider The pronominal root which constitutes the basis of the oblique lition of cases in the Indo-European languages is adopted in the languages of ' some- the Scythian family, not only in the oblique cases, but also in the nominative itself. ‘vatm, Whilst in both families the oblique cases are sub- , inthe stantially the same, the Indo-European uses as its nominative the base e Sans- in ‘ah ; the Scythian, the base in ‘ma.’ There are a few langnages even in the Indo-European family in which ‘ma’ has found its way the Old [ repre- into the nominative ; e.g., the Celtic has ‘mi,” the New Persian ‘man,’ the North-Indian vernaculars ¢ main.’ It is observeable, however, that tive of tion of in each of these exceptional cases Scythian influences have been in The New Persian has been influenced by the Oriental and msider- operation. Turkish, the Celtic by the Finnish, and the idioms of Northern-India (s ap In some by the Secythian vernacular which preceded the Sanserit. cases also, especially in the later dialects of this family, the accusative ‘nas, Jebrew has come to be used instead of the nominative, in violation of existing grammatical rules. Thus, the Singhalese ‘mama,” the Kawi ‘mami,’ ) made ed, and the later Cuneiform Persian ‘mém,” are probably accusatives in their origin, like the Italian ‘mi’ and the French ‘moi. On the other , forms hand, we are met by one, and only one, exceptional case in the Scythian idea of tongues. The Scythian of the Behistun Inscriptions makes use of lspond- ‘hu’ as its nominative ; but in ‘mi,’ the corresponding possessive hrevia: suffix, the ordinary Scythian base re-appears. The nominative (as well as the oblique cases) of the first personal osition pronoun in all existing languages of the Scythian group, is derived Telugu from a base in ‘ma ; and it will be shown that this ‘ma’ not unfre- of the quently comes into perfect accordance with the Dravidian pronoun, by ém-ﬂ,’ changing into ‘na ality ) In those languages ‘ma’ is very generally euphonised or nasalised ;rough by the addition of a final ‘n,” or of an obscure nasal resembling the ould Sanserit ; ‘anusvira :* in consequence of which, not ‘ma,” but ‘man, used. may be stated to be the normal form of the Scythian pronoun, and ) aﬂd this bears a closer resemblance than ‘ma’ to the Drividian ‘néan. er 8 The addition of this euphonic nasal is not unknown even to the o, 304 THE PRONOUN. Indo-European languages. It may be seen in the Persian ‘man,’ the O to th Sindhian méAn,” and the Bellich? ¢ menik; and a similar inorganic addi- i and ¢ tion is apparent in the Old Greek € yov’ and ¢ éydvy,’ as also in aly, “rovy’ and ‘7Tovv. The first nasal is much more common, however, On examining the ‘m' | and more characteristic in the Scythian tongues. ji inflex Turkish family of tongues, we find ‘men’ in Oriental Turkish ; ‘mAn’ Sessly in Turkoman ; ‘m&m’ in Khivan; ‘ben’ (‘m’ degraded to ‘b’) in defini Ottoman Turkish. Tn the Finnish family, the Finnish proper has o). ‘miné ; the Lappish ‘mon ; the Esthonian ‘ma’ or ‘minna; the the | Mordwin and Votiak ‘mon; the Ostiak ‘ma’ (dual ‘min, plural the f “men’). The Samoiede dialects have ¢man, ‘mani.’ In both from Mongolian and Manchu the nominative of this pronoun is ‘bi; but It this is evidently corrupted from ‘mi’ (like the Ottoman ‘ben’ from prong the Uigur ‘men ’) ; and it is “mi,” with a final nasal, which forms the 3 Sl basis of the oblique cases. [n both languages the genitive is ° mini :’ and the dative is ‘men-dou’ in Mongolian, ‘min-de’ in Manchu. The guage distin Tibetan pronoun is ‘gnya, ‘gna,’ or ‘nga’ (identical doubtless with In ‘ma’) ; the Chinese ‘ngo; the sub-Himalayan °gna; the Avan ‘nga ; the Mikir ‘ne; the Corean ‘nai. “Ining sente It is evident from the above comparison (1), that the true and from essential representative of the personal of this pronoun in the Scythian T tongues is ‘ma ; and (2) that as ‘ma’ has been euphonised in the from Western families of that group into ‘man,’” so it also evinces a tendency in the eastern stems to change into ‘nga’ or ‘mna.’ In many of those euphy have idioms ‘ma’ still retains its place unchanged, or may optionally be the there used instead of the later ‘man. The Mingrelian has ‘ma The Finnish has first i Suanian ‘mi,’ the Lasian ‘ma,” the Georgian ‘me. ] both ‘me’ or ‘ma’ and ‘mina,” and also ‘mia: the Ostiak both I I ‘min’ and ‘ma.’ Dl‘i‘l\'i( ‘n, { It is found also in those langnages in which ‘man’ constitutes the i gl] isolated pronoun, that ‘m’ is used as its equivalent in the personal ter- nomyy minations of the verbs, and generally in all inflexional compounds. We see this usage illustrated in the colloquial languages of Northern India but a for and in the Persian. For example, whilst ‘man’ is the nominative of Biny the Persian pronoun, the basis of the oblique cases is not ‘man’ but Ratiy ‘ma’ (e.g., ‘ma-14,” me, of me); and the pronominal ending of the verb in the first person singular is ‘m In a similar manner, in the the 1 Turkish family of languages, ‘m’ is used in composition as the equi- | disti; valent of ‘man’ or ‘men.’ Thus, in Oriental Turkish, whilst ‘men’ 1 is retained in the present tense ; e.g., ¢ bold-men,” 7 am ; the preterite is contented with ‘m’ alone ; e.g., ‘boldi-m,” 7 was. and ¢ The same suffix is used to denote the first person singular in the 01)“(11 o FIRST PERSON SINGULAR. ) 05 1, the possessive compounds of the Turkish, a class of words which is peculiar caddi- to the Scythian family ; e.g., ¢ baba-m,” my fuather, from ‘baba,’ father, also and ‘m,” the representative of the first person singular. In the Magyar wever, also, though the isolated pronoun of the first person singularis ‘én,’ yet ‘m’ is used instead of ‘n’ in the possessive compounds and determinate g the iy inflexional terminations : e.g., from ‘atya,” father, is formed the pos- sessive compound ‘atya-m, my father; and the first person singular of b)in et has definite or determinate verbs ends in ‘mjy e.g., ‘szeretem, I love (some one). It is also to be noticed, that whilst the Magyar has ‘én’ as the 1 the the singular of the isolated pronoun, its plural is ‘mi’ or ‘mink ; pIural the former of which is evidently pluralised from ‘ma’ or ¢ me,’ the latter 1 both from ¢min. ;i ;’ bﬂt ) It was shown that the initial and radical ‘m’ of the Indo-European '’ from K pronoun was occasionally converted in ‘n we have now to show that s the 9 it a similar change from ‘m’ to ‘n’ is apparent in the Scythian lan- . The guages also, and that in some of those languages ‘n’ has become as s with distinctive of the first person as in the Dravidian family itself. - Avan In Finnish, though the isolated form of this pronoun is ‘ma’ or ‘mina,” yet in all inflexional additions and compounds ‘m’ is repre- <3 1 sented by ‘n; e.g., from s8, father, is formed ‘isd-ni, my father, and e and from “0l, to be, is formed ‘0l-en,” 1 am. eythian This final ‘n’ is not derived from the euphonie ‘n’ of ¢mina ;’ but in the from a direct conversion of ‘m’ into ‘n ;’ for though we see the same ndency euphonic addition of ‘n’ in “sind’ (from ‘se’ or ‘sia’), thou, yet we Qf those have ‘t’ alone (the equivalent of ‘s’) in ¢dl-et,’ thou ast. ‘i has, lly be therefore, become in Finnish, as in Dravidian, the ordinary sign of the n, the first person singular of the verb, by conversion from an older ‘m.’ ish has The Magyar “én,’ [, appears to be still more nearly allied to the lk both ¢ Dravidian pronoun ; and in this case n is certainly derived from ) ‘m,” for whilst ‘n is found in the nominative, ‘m is used instead utes the in all possessive compounds and verbal inflexions. With the Magyar b nal e nominative ¢én,’ compare not only the Tamil-Canarese ‘én’ or ‘en J Is. We but also ‘an’ or ‘awn,” 7, in the Lar, a Sindhian dialect. A similar 1 Tndia form of this pronoun is found in the Mordwin, another idiom of the at ¥innish or Ugrian family, in which, whilst ‘mon’ is the isolated nomi- a1 b b native, ‘an’ is used instead in verbal inflexions ;’ e.g., " paz-an,’ I (am) r Of the the Lord. In the Olet or Calmuck dialect of the Mongolian tengue, there are 11 the distinct traces of the same change of ‘ma’ into ‘na.’ men The nominative of this pronoun in Calmuck is “bi’ (from ‘mi’), t { ?reterite and the same base appears in the genitive ‘mini :” but the rest of the oblique cases are formed not from ‘bi’ or ‘mi,’ but from ‘na ; eg., X r D the e—— —— . 306 THE PRONOUN. ‘na-da,’ to me, ‘ na-da-edze,’ from me, and also ‘na-mai,” me. We here 1688 of the discover the existence of a pronominal base in ‘na’ (probably derived In from ‘ma’), which is in perfect agreement with the Dravidian. used In a few of the Scythian languages the isolated pronoun, including its nominative, seems to be absolutely identical with that of the Dra- woule vidian family ; e.g., ‘na’ in the Quasi Qumuk, a Caucasian dialect, singul b and ‘ne in Motor, a dialect of the Samoiede. Compare also the Tamil East-Asian forms ; e.g., Avan, ‘na’ or ‘nga; Tetenge, an Assam agree dialect, ‘ne ; Corean, ¢ nal ; I whick Tibetan, ‘nga; Chinese, ‘ngo.’ doubt not that the ‘nga,’ 7, and ‘nge,” we, of the Tibetan, are identical siona with the ‘ma’ and ‘ me’ of the other languages of High Asia ; and to as ‘nga’ is plainly identical with ¢ pa,” as well as with ‘ma,’ the sup- ‘n'o position that the Dravidian ‘na’ is derived from the Indo-European ably is confirmed. begon and Scythian ‘ma We may here also compare the Australian pronouns of the first person, viz., ‘nga,” ‘nganya,’ / P4 . entire its dnal ‘ngalee,” we two; and the plurals ‘ngadlu’ and °nadju, initil we. nitial On the whole we appear to have reason to conclude that the howey various forms of the pronoun of the first person singular which have the in now been compared, ‘ma,” ‘nga,” and ‘na,’” are identical ; and that this it is I word was the common property of mankind prior to the separation of of the the Indo-European tribes from the rest of the Japhetic family. this I each ¢ 2. PRONOUN OF THE StECOND PERSON SINGULAR. peare Comparison of dialects. — In Canarese the nominative of this firs pronoun is ‘nin’ or ‘nin-u; and in the oblique cases the included danei‘ vowel of ‘nin’ is shortened by rule; e.g., ‘ninna,’ ¢tAy. The plural Yeuq differs from the singular only in the use of ‘m’ as a final, instead of ‘uir ‘n: it may therefore be concluded that those finals are marks or for- Separy matives of number, not of personality ; a conclusion which is converted the g into a certainty by the circumstance which has already been pointed Pogtic out that in the pronoun of the first person also the final ‘n of the Coales singular is converted in the plural into ‘m which In Canarese not only are ‘nad’ and ‘ni regarded as the crude It bases of the pronouns of the first and second persons, but they are Persoy occasionally used also as nominatives of verbs instead of ‘nan’ and SV ‘nin.’ the i} In the personal terminations of the verb, this pronoun is much iy changed in all the Dravidian dialects. It not only loses its initial ‘n,’ del‘iv like the pronoun of the first person; but its final ‘n’ also disappears. (asey Generally nothing remains in the verbal inflexions but the included ordiy vowel, and that also is more or less modified by use. In the Cana- often SECOND PERSON SiN(} ULAR. 307 lierg rese verb it appears as ‘i SIS AN 1 and ‘e: in the ancient dialect iye rived of the Canarese it is ‘4¥,” as in Tamil. In Tamil “ni,” which is properly the crude base, is invariably uding used as the isolated nominative, instead of ‘nin’—the form which Dig- would correspond by rule to ‘nan,’ the nominative of the first person alect, singular. That ‘nin’ originally constituted the nominative even in o the Tamil, appears from this that the oblique cases in the higher dialect \ssam agree in using ‘nin’ (shertened by rule from ‘unin’) as the base to o i which the case-suffixes are attached. Another form which is occa- ntical sionally used by the poets is ¢nij,” in which the final ‘n’ is softened ) and to ‘y’—in the same manner as the initial ‘n of ‘nan. The final S0P ‘n’ of this pronoun, though totally lost in the nominative, is invari- opeal ably retained in the oblique cases; in which it is the initial ‘n’ that mpare becomes liable to alteration. In the colloquial obliques the initial ‘n’ ), [ ,‘ entirely disappears, and does not leave even a ‘y’ behind it, as the initial ‘n ) When the adjn, of the first personal pronoun generally does. 4 bd K initial is discarded, the included vowel changes from 1 to ‘u ¢ u” t the however, constitutes the included vowel of this pronoun, not only when have the initial ‘n’ is lost, but sometimes, in the higher dialect, even when it 1s retained. b this ‘nin, nun,” and ‘un’ are severally used as the basis jon of of the oblique cases. In the persoual terminations of the Tamil verb, this pronoun is represented by the suffixes ‘4y, ‘ei,” or ‘i:’ from each of which suffixes the final ‘n,” as well as the initial, has disap- peared. In the poetical dialect of the language the initial ‘n’ at this tirst sight appears to have retained its place in such forms as ‘nadan- Juded danei,’ thow didst walk, and in the corresponding plural, ‘nadandanir,’ ye walked :> but the ‘n’ of these pronominal terminations (‘nei’ and p]ural ad of ‘nir’) is merely euphonic, and is inserted for the purpose of keeping p for- separate the contiguous vowels of ¢ nadanda-ei’ and ¢nadanda-ir. In erted the same manner, in the first person, ‘nadandanen, I walked, is used |iﬂted poetically for ‘nadanda-en;’ and when its vowels are allowed to coalesce, instead of being kept separate, they become ‘nadandén,’ f the which is the more ordinary form. crude The root of the verb is regularly used in Tamil as the second person singular of the imperative, without any pronominal suffix, and y a1re -l even without any euphonic addition: but the second person plural of the imperative in the colloquial dialect is formed by the addition of ‘um’ (the ordinary plural base of the oblique cases); which ‘um’ is ch 1 derived from a singular form in ‘un,” one of the bases of the oblique ,ea!‘S. cases already referred to. In the higher dialect ‘4j’ and ¢ir, the ,uded ordinary representatives of these pronouns in the verbal inflexions, are often added to the root to form the singular and plural imperative ; and % X2 ;i 308 THE PRONOUN. B e T e e.g., ‘kelay,’ hear thou, ‘ kélir, hear ye. [These forms are apparently identical with ‘kéldy, thou hearest mot, and ¢ kélir) ye hear not: but abli they are not really identical, as Beschi supposed, for it will be shown in the Section on the ‘Negative Verb’ that ‘a,’ a relic of ‘al’ no, N TR addi is'an element in all negative forms; though in these and in some other in f instances it has been absorbed in the succeeding long vowel.] the s With respect to the consonantal elements of the pronoun of the Tejec second person, there is little room to doubt that they consisted in an base initial and final “p, the foriner essential, the latter formative : but there €xar seems to be some doubt with respect to the included vowel. Authority ﬁppc 2 2 preponderates in favour of ¢i;’ ‘u ranks next, and ‘4 next to and that; but ‘ei’ and ‘e’ are also found. Sometimes in Tamil, ‘1’ is diale tical converted in pronunciation into a sound resembling ¢f;’ whilst the converse never takes place. It may therefore perhaps be concluded that ‘i’ constituted the included vowel of the original base of this ‘mi and pronoun. Beschi, in his grammar of the High Tamil, represents ‘di’ as being form used occasionally by the Tamil poets as a suffix of the second person relat singular of the imperative; and if this representation were correct, it of tf would be necessary to regard ‘di as a pronoun, or as the fragment cham of a pronoun, of the second person singular. It is founded, however, the | on a mistake: for the word which Beschi cites in proof (¢adi, become thou, from ‘Agu,’ abbreviated into ‘4, to become’) is not really an por imperative, but is the second person singular of the preterite; and “di’ tion‘ is compounded of ¢d, the sign of the preterite tense, and ‘i, the usual fragment of ‘ni,” ‘thow. ¢ Adi’ means properly thow hast become, Case: and it is used as an imperative by the poets alone to convey an emphatic prediction of a result which is regarded as already certain. iS 9 We find the same suffix in such poetical preterites as ¢ varu-di’ (for ‘yandiy ’) thow camest, and ‘kedu-di’ (for kettdy’) thou art ruined. S ——————— In Malayilam the nominative of this pronounis ‘ni;” but ‘nin’ laugl is used, as in classical Tamil, in the oblique cases. The dative has Use( ‘nan,” instead of nin; e.g., ‘nanikka,’ to thee —as if from a nomi- native in ‘nan,” with ‘a’ as the included vowel. This use of ‘a’ is in accordance with the colloquial Tamil personal termination of ("on@J 13 the verb, which is ‘4y, instead of ‘iy. &WQ In Tulu the nominative is ‘1;” but the oblique cases are formed lang In the personal terminations of the upon the base of ‘ni’ or ¢ nin. Ine verb the second person singular is represented by ‘a.’ The Tulu A €y nominative ¢ 1 illustrates the fact already stated that each of the for nasals of ‘nin’ (both the radical initial and the formative final) has sometimes been worn off, of SECOND PERSON SINGULAR. 3( ; 1€ g ently The Tuda has ‘ni’ as its nominative, ‘nin’ as the base of its but ? oblique cases, and ‘i as the personal termination of the verb. hown The Telugu nominative is ‘nivu,’ expanded from ‘ni’ by the " fiot, addition of the euphonic particle ‘vu:’ ‘nivu,’ Tel , thou, is identical ofher in form, though not in meaning, with the modern Canarese plural of the same pronoun, viz., ‘niva,” you. In the oblique cases the Telugu f the rejects the euphonic addition of ‘vu,” and uses ‘pi’ as its inflexional I an base, and also as its possessive. The objective alone follows the there example of the other dialects in abbreviating the included vowel, and hority appending a final nasal. That case is ‘ni-nu,” ‘nin-u,” or ‘nin-nu,’ xt 10 and is evidently formed from a nominative ¢nin-u.’ In the higher ) 18 dialect of the Telugu, ‘ivu, from an old nominative I, which is iden- st the tical with the Tulu, is occasionally used instead of ‘nivu. Juded The Telugu plural of this pronoun has ‘miru’ as the nominative, f this Both ‘miru’ ‘mi’ as the inflexion, and ‘mimu’ as the accusative. and ‘mimu’ indicate a base in ‘mi’ from which they have been being formed by the addition of signs of plurality ; and ‘mi’ bears the same peIson relation to the ‘ni’ of the other dialects that ‘m4,” the Telugu plural Sct, it of the first person, does to the ordinary Dravidian ¢na. How this change from ‘n’ to ‘m’ has taken place will be inquired into under gmem ever, the head of ¢ The Plurals.’ hecome In the personal terminations of the verb, the Telugu rejects every H) an portion of the pronominal root, and employs only the euphonic ad ’ 5 tion ‘vu or-Sivi.’ ‘. d ’,di i the In Gond the nominative is ‘ima;’ but ‘ni’ is used in the obliqube K 00y cases, ‘n or ‘i’ in the personal terminations of the verbs. Tts nominative The Ku corresponds on the whole to the Telugu. 'ey an ' ertalll is ‘inu;’ its inflexion ‘ni;’ the personal termination of its verb ‘i. The Rajmahal nominative is ‘nin;’ the Urdon, ‘nien.’ (for [4 The Brahui nominative is 11 ) as in most of the Dravidian e, languages; and its nominative plural is ‘num’—a form which is much i used in the higher dialect of the Tamil. Ve hﬂS pomi- On a comparison of the various Dravidian dialects we arrive at the of 8 conclusion that the primitive form of this pronoun was ‘ni’ ‘nd, or oo of ‘na ;'—niost probably the first. The only essential part of the pronoun appears to be the initial consonant ‘n;’ just as in the Indo-European languages ‘t’ is the only essential part of the corresponding pronoun. ormed In each family the vowel by the help of which the initial consonant is f the enunciated varies considerably, but evinces, on the whole, a preference Tulu ) for ¢i’ in the Dravidian languages, for ‘u in the Indo-European. of the Supposing “ni’ to be the primitive form of the Dravidian pronoun l) has of the second person, and comparing it with ¢na,” which we have seen j Jl 310 THE PRONOUN. cat to be the primitive form of the Dravidian pronoun of the first person, of it is deserving of notice that the only difference between the two is the difference between the two included vowels, ‘a’ and “i. The conso- fin nant ‘n’ seems to be the common property and the common sign of both pronouns, and the means by which their personality is expressed; 10( whilst the annexed ‘a’ restricts the signification to the first person, or that of the speaker; i,” to the second person, or that of the person de ;r addressed. The only resemblance to this arrangement with which I ou am acquainted is that which is found in the personal pronouns of the Hebrew, in which 7 is ¢an-6ki; thou, ‘an-td’ (corrupted into ‘at-ta’) % 1 fl The method adopted by the Dravidian languages of expressing the difference between the first person and the second by means of the W E vowels ‘a’ and ‘i, used as auxiliaries, does not appear to be the th result of accident. It is probably founded on some ultimate principle; though it may be difficult or impossible now to discover what that 1€ ‘1 principle is. If “a’ and ‘1’ be considered as identical with the demonstratives, 0 an idea which would suit the signification, and which is corroborated by the circumstance that ‘u’ is also a demonstrative, we are met by the apparently insurmountable difficulty that in all the Dravidian tongues, f‘i and (as far as the use of these demonstrative vowels extends) in all the tongues of the Indo-European family, ‘a’ is not the proximate, but the remote, demonstrative; and ‘i’ is not the remote, but the proximate ; whilst ‘u’ is used in Tamil as an intermediate between those two. bd Is any weight to be attributed to the circumstance that ‘a has naturally the first place in all lists of vowels, and ‘i’ the second ? Extra-Dravidian relationship.—1t has been shown that the Dravi- dian pronoun of the first person has affinities with each of the great Japhetic groups, with some special Scythian affinities. It will be found that the relationship of the pronoun of the second person is less extensive, but more distinetive: it is specifically Scythian. Throughout the Scythian as well as the Indo-European group, the most prevalent form of the pronoun of the second person singular is that which is formed from the consonant “t’ (e.g., ‘tu’), or its euphonised equivalent ‘s’ (e.g., ‘ov’); and the only other form which is found to be used in any family of either of those groups is that which is formed from the consonant ¢n,” and of which the Cuneiform-Scythian and the bl Dravidian ¢ ni is the best representative. Thege roots appear to have been always perfectly independent. I — — P SECOND PERSON SINGULAR. 311 1So1, cannot discover any reliable trace of a connexion between them, or 1 the of a gradual change in any instance of the one form into the other. Conso- In order to place this point in a clear light, it is desirable, in the ign of first place, to trace out the connexions and alliances of the pronominal root ‘tu.’ essed; eTson, It has been conjectured that this pronoun had its origin in the DEIson demonstrative base ¢t;’ but the investigation of this point is beyond hich [ our purpose, which is merely that of tracing its relationship. f the In Sanscrit the pronoun of the second person singular is ‘tva-m;’ ) in Zend ¢tl-m, and also ‘thw’’ as included in the accusative ¢thwa,’ thee. g the Connected with the Sanserit ¢ tva,” there is a simpler form, ‘ ta, f the which is appareut in ‘tava,” thy,; and we have analogies to this in The e the the Kawi ‘ta’ and the Semitic ta’ (included in ‘anta,” thow’). ciple; Semitic ‘tA’ is changed in the inflexions to ‘k&,” a change which that resembles that of the Kawi, which has ‘ta’ as its nominative and ‘ko’ as its possessive. Bopp supposes that ‘yu,” the base of the most commion form of the plural of this pronoun, is derived from ¢tu,” and fives, that ¢va,’ the base of the Sanscrit secondary plural ¢ vas’ and of the d by / the Latin ¢ vos,” is derived from ¢ tva.’ ‘v, however, is more frequently derived from £mi than from any other letter. ‘tva-m’ becomes oues, nall ‘tuva-m’ in the Old Persian; and from ‘tu’ (itself derived from nate, ‘tv ’) proceeds the Sanscrit dative ¢ tu-bhayam:’ the base of which is allied to, or identical with, the Latin, Armenian, and Pehlvi ¢tu;’ the - the yeen Holic and Doric ‘7?;’ the Persian, Afghan, and Singhalese ‘to;’ and the Gothic ¢thu.’ The ‘th’ of the Gothic and Zend points out the path by which the Old Greek ‘¢’ was converted into ¢ 2.’ 43)’ In the personal terminations of the verbs, in Sanscrit and most the other languages of the same family, the earlier ‘t’ of this pronoun has very generally been weakened into ‘s’ in the singular, whilst in most of the plural terminations, ‘t,” with some trivial modifications, avl reat and with a sign of plarality annexed, has succeeded in retaining its place. In our investigation of the pronoun of the first person it was be found that ‘ma’ was converted in the personal terminations of the Jess verb into ‘mi, and ‘mi’ still further weakened into ‘m:’ so also ‘su’ (for ‘tu’) generally becomes °si in the verbal terminations ; the and °¢si in like manner afterwards becomes ‘s hat sed In the Scythian group of tongues, the pronoun of the second person which is in general use is substantially the same as in the Indo- kil European—another evidence of the primeval identity of both groups: ed the but in the Scythian tongues the weaker ‘s’ has obtained wider preva- lence than the older ‘t; and the vowel by which ‘s’ is enunciated is more frequently i’ or ‘e,” than ‘u’ or ‘a.’ = ; 312 THE PRONOUN. The Magyar has ‘te’ in the singular, ‘ti’ or ‘tik’ in the plural, with which we may compare the Armenian ‘tu,” thou, and ¢ tuk, you. The Mongolian ¢ tchi’ or “dzi,’ thou, exhibits the progress of “ti’ 2 towards softening into ¢si In Finnish proper, the isolated pronoun K of the second person singular, is ¢se or ‘sina; but ‘t’ retains its place in the plural; and the personal termination of the verb even in the singular is ¢ t. The chief peculiarity apparent in the Scythian form of this pro- noun is, that it has generally been euphonised by the addition of a final nasal—the consonant ‘n,’ precisely in the same manner as the pronoun of the first person singular. (:() In the older Greek ¢ 7wy’ and ¢7odw,” correspond to ¢ éyivy’ and ‘éywr;’ and in like manner, in the languages which belong to the Scythian family, or which have been subject to Scythian influences, R T where the pronoun of the first person is found to be nasalised, the pronoun of the second person generally exhibits the same feature. a In the vernaculars of Northern India, we see this euphonic addition to the pronoun of the second person in the Hindi, Panjabi, and Sindhi, ‘tun,’ and in the Marathi and Gujarathi ¢ttm. In some of those § idioms, especially in the Gujarathi and Panjabi, the euphonic nasal appears in the oblique cases as well as in the nominative, but more commonly it is found in the nominative alone. | In the Turkish family of tongues, ‘sin’ or ‘sen’ is the usual form of the pronoun of the second person singular. The ¢n’ retains its place in the oblique cases, but is lost in “siz’ the plural. Compare also the Georgian ‘shen; the Samoiede tan, ‘tani; the Lappish § “don;’ the Votiak and Mordwin ‘ton’ (plural ‘tin '}; and the Finnish 1 ‘sind,” which alternates with se,” ¢sia,” and sie.’ The euphonie origin of this ‘n” is most evident in the Esthonian dialect of the Finnish, which uses indifferently ‘sa’ or ¢sinna’ for the second person, and ‘ma’ or ‘ minna’ for the first. ] In the Mongolian and Manchu, ‘n’ appears in the oblique cases 1 only. In Mongol the nominative is ¢tchi,’ in Manchu ‘si; but the genitive in the former is ‘tchini, in the latter ‘sini,” and the corres- ponding datives are techim-dou’ and ¢sin-de.’ In Calmuck the nominative is ¢dzi’ or ‘dzima,’ genitive ¢dzint,’ ctj dative ‘dzimadou,” accusative ¢dzimai.’ In the pronouns of this language we may observe several instances of ‘m’ being used as an euphonie, instead of ‘n.’ It is evident that there is no resemblance whatever between any of the pronouns compared above and the Dravidian ¢nj The final ‘na’ of the Finnish sina,” and its equivalent, the final vy’ of the Greek SECOND PERSON SINGULAR. 313 | “1dwy,’ are separable, euphonic, inorganic additions, and can have no real connexion with ‘ni,” which is an ultimate root. We have seen that the Indo-European and Scythian ¢m’—the initial of the pronoun of the first person—was probably the origin of the ‘n’ of the Dravidian ‘na. Is it possible that the radical ¢t’ of the pronoun of the second person in both those families of tongues was changed in like manner into ‘n,’—so as that ‘tu’ or ‘ti” was the origin of the Dravidian ‘ni? I think not. This is supposed, indeed, by Castrén, a very high authority, to be the history of the ‘n’ by which 1 the second person singular is often represented in the perscnal affixes of the Finnish and Turkish families. It may also be mentioned here, that a change of ‘t’ into ‘n’ is not quite unknown even in the Indo-European e languages. It is somewhat frequently found to take place in Pali; e.g., ‘té they, masculine, becomes optionally ‘né; ¢t4’ t¢hey, feminine In Sanserit becomes ‘n&; and ‘tani,” they, neuter, becomes ‘nani.’ also ¢étam,” hum, is sometimes changed into ¢énam. There is no evidence, however, that the ‘n’ now under consideration—the initial L of the Dréividian ‘ni’—arose from any such process of change; the supposition would be wholly a gratuitous one; and the discovery of ‘ni” in languages of such high antiquity as the Cuneiform-Scythian of the Behistun Inscriptions and the Chinese, shows that ‘ni’ claims to be regarded as independent of ‘tva,” or ‘tu, and as an ultimate pro- nominal root. It thus appears that there are probably two Japhetic bases of the second personal pronoun, as well as two of the first. There are traces, more or less distinet, in various languages of the h Scythian group, of the existence of a pronoun of the second person identical with, or evidently allied to, the Dravidian ¢ni.’ I begin with the most remarkable and decisive analogy, because the most ancient which is capable of direct proof, viz., the second per- sonal pronoun in the Scythian tablets at Behistun. T His S e 2 precisely as in the Dravidian idioms; and the possessive which is used in compounds is ‘ni,’ which is identical with the similarly abbreviated basis of the Dravidian oblique cases of this pronoun. The plural of this pronoun 1is, unfortunately, unknown. The personal termination of the verb is not “ni,” but ‘nti; which I suspect to be a compound of ‘ni’ and ‘ti,” like the ‘anta,” ‘anti,’ of the Semitic languages. The antiquity and distinctively Seythian character of the DirAvidian pronoun of the second person is thus clearly proved; and this proof of its antiquity entitles us to regard as allied to the Dravidian ‘ni’ certain resemblances to it which otherwise might be thought to be accidental. In the Ostiak, the most Dravidian of the Finnish dialects, in that compound of nouns with possessive suffixes which is so characteristic 314 THE PRONOUN. of the Scythian group, the first personal pronoun is represented by ‘m; the second, as in the Dravidian languages, by ‘n; e.g., ‘ime-m,” my wife, ‘ime-n, thy wife. In the Syrianian, another Finnish idiom, the 11 second person of the verb, both singular and plural, is formed by annex- ing a pronoun of which ‘n’ is the initial and radical; e.g., ‘kery-n,” thou f hast done (from ¢ kery,” ¢o do), ‘kery(ninyd,’” you have done. In ‘nyd, you, we see indication of a singular ‘ny,” thow, which has been plu- fl ralised, as is usual in these languages, by suffixing to it ¢d’ or “t.’ 1 In addition to the allied forms discoverable in these compounds, we find in the Ugrian tongues several instances in which the isolated pronoun of the second person which is used as a nominative is plainly 0 allied to the Dravidian. In the Ugro-Ostiak, or that dialect of the Ostiak which is treated of in Castrén’s grammar, thow is ‘nen; youw fi Here ‘ne’ or ¢ni con- two, ‘nin; you (indefinitely plural) ‘nen. P stitutes the pronominal base; and the final ‘n’ of the singular ¢ nen,’ (U] is a formative or euphonic addition like that which has converted the o Dravidian ‘ni A3 into ‘nin.’ The strong pronunciation of this Ostiak fl In other Ostiak final ‘n’ reappears, as we shall see, in the Turkish. of dialects we find ‘num’ and ‘ma,” and also (which is more deserving of 1 notice) ‘nyn,’” with a plural ‘nynt.’ In the Vogoul we find analogies fi which are no less remarkable than the above; e.g., ‘nei, ‘ny,’ ‘nan, by ‘nyngi,” and ‘nank. Compare also the Vogoul plurals ‘nen’ and €0 ‘non.’ J\ In the Finnish proper, the only trace of this pronoun which we 1§ observe is one which, but for the existence of such express analogies in 1§ other members of the family, we should probably have overlooked. In the plural of the second person of the Finnish verb (e.g., ‘olette,” ye § are, pluralised from ‘olet,” thou art), the suffixed pronoun corresponds § to that of which “t’ or ‘s’ is the initial; but in the possessive com- ‘i pounds, in which we should expect to find precisely the same form, we find instead of it a plural possessive of which the initial and radical is a 9 Je ‘n Thus, the expression thy hand, being ‘kates,” we should expect to ‘k find your hand, ‘katesse,’ or, more primitively, ‘Lkatette,” like the or corresponding Magyar ¢ kezetek,” (from ‘tek,” yow, another form of ‘te’); whereas the form actually used in Finnish is ¢kéatenne. It L U thus appears that two pronouns of the second person retain their place 1d in the Finnish; one, the singular of which is ‘si,” or more properly ‘ti,’ th the plural ‘te; and another, hidden in the ancient compounds, the tl plural of which is ‘ne,” and of which, by dialectic rules, the singular must have been ¢ni.’ 1n Even in the Turkish, we shall find traces of the existence of a fal similar pronoun. In the possessive compounds, the second person I SECOND PERSON SINGULAR. 315 by singular is not represented, as we should have expected it to be, by lﬂy ‘sen,” as the first person singular is by ‘m; but ‘n’ or ‘ng’ is used he instead (a nasal which corresponds to that of the Ostiak ‘nen’); X e.g., ‘baba-n, thy father,; and as the final ‘m’ of ‘baba-m’ is derived 0l from ‘mi’ or ‘me, 7, we seem to be obliged to deduce also the final ’d, ‘n’ of baba-n,” from an obsolete ‘ni’ or ‘me,” thou, which is allied to lu the corresponding forms that have been pointed out in other Scythian tongues. We find this possessive ‘»’ or ‘ng’ not only in the Osmanli We Turkish, but even in the Yakute, the Turkish of Siberia. ol The same ‘n I’ makes its appearance in the personal terminations ly of the Turkish verb. ‘sen’ is more commonly used than ‘% ;" but the ‘n’ is found as the representative of the second person in those verbal o forms which must be considered as of greatest antiquity ; e.g., in the o preterite of the auxiliary substantive verbs, ‘idum,” 7 was, ‘idun,’ thow )l\ wast, ‘1di,” he was. In the Oriental Turkish the forms corresponding the to these are ¢béldim,” ‘béldizn,” ‘boldi; and the same termination of ak the second person singular—the nasal ‘»’—appears in all the preterites jak of that language. We may compare also the plural forms of this pro- of nominal suffix. The Turkish pronouns are pluralised by changing the 65 final formative ‘n 2 into ¢z, K or rather by adding ¢z’ to the crude AL base. Thus, we is ‘biz’ (for ‘miz’), and you is ‘siz In possessive 2 dld compounds ‘i changes into ‘u; and hence owr father is ¢ baba-muz.’ In the same manner, your father is ‘baba-nuz, indicating a suppositious, e isolated pronoun, ¢niz,’ you, corresponding to ‘miz,’ we. Whilst ¢ u’ 510 is used instead of ‘i’ in Osmanli Turkish, the older and more regular ed ‘1’ retains its place in the Oriental Turkish ; e.g., ¢ uzt-ziz,’ you your- e selves ; in which you is ‘niz’ or ‘ngiz,’ and from which, when ¢z, the i sign of plurality, is rejected, we deduce the singular ¢zi’ or ‘ngi’ Il The same mode of forming the plural termination of the second person We appears in all regular Turkish verbs ; e.g., compare ‘kérkdu-nuz,’ ye 11 Jeared, with ‘korkdu-n, thou fearedst. We see it also in the imperative {10 ¢korku-nuz,” fear ye. In all these instances, I consider the Turkish ¢z’ the or ‘ng’ to be dialectically equivalent to the Finnish ‘n; and the 10f pronominal root which is thus found to underlie so many Turkish and It Ugrian compounds of the second person may, I think, be regarded as Jace identical with the Dravidian and Behistun-Seythian pronoun. Even tly the libration between ‘i’ and ‘u,’ which we noticed in considering the the Dréavidian forms of this pronoun, meets us again in the Turkish. ilar In the Himalayan dialects, though Tibetan or Indo- Chinese influences generally seem to preponderate over Drividian, we cannot fail to see Dravidian analogies in the Dhimal ‘n4, in the Miri ‘no, f 4 S0 in the Garo ‘nd4 ; and in the ‘n’ which forms the first and most 316 THE PRONOUN,. essential radical of the pronoun of the second person in all the rest of [ the Lohitic dialects. Still more remarkable is the Chinese ‘ni,” which is absolutely identical both with the DrAvidian and with the Behistun-Scythian pronoun : so also is the ‘ni of the Horpa, a dialect of Tibetan { nomades. Compare also the pronouns of the second person in various Australian dialects: e.g., ‘ninna, ‘nginnee,’ ‘nginte; the duals § ‘niwa, ‘nura ; and the plural ¢ nimedoo. al On a comparison of the various forms of this pronoun which have gW been adduced above, it must be evident that the affinities of the Dravi- 81 dian ‘ni’ are wholly Scythian ; and this important circumstance, taken in conjunction with the predominance of Scythian influences over i Indo-European in the formation of the first personal pronoun, con- il tributes largely to the establishment of the Scythian relationship of the Dravidian family. il 3. Tre RerrLeExive ProNouUN ‘SELF. al The Dravidian pronouns of the third person are, properly speaking, demonstratives, not personal pronouns; and they will, therefore, be I investigated under a subsequent and separate head. The pronoun, th which is now under consideration is entitled to a place amongst per- a9 sonal pronouns, because it possesses all their characteristics, and is €l declined precisely in the same manner. It corresponds in meaning to I the Sanscrit ‘svayam,’ and to the defective Greek ¢’ and the Latin I ¢ sul ‘sibi,” “se; with a range of application which is more extensive {(I than theirs. In Tamil the nominative singular of this pronoun is ‘tén; the d plural of which (by the usual pronominal change of ‘n’ into ‘m’) is “tam 2’ and the inflexion, or basis of the oblique cases (which, taken by itself, has the force of a possessive), is formed, as in the case of the LI other personal pronouns, by simply shortening the included vowel ; sl eg., ‘tin, of self, ‘swr, or (adjectivally) °suws,’ sua,” ‘suum. In_ qQu all its cases and connexions ‘tan’ is found to be more regular and q U persistent than any other pronoun. )‘19 The Canarese nominative is ‘tan’ in the ancient, ‘tan-u’ in the modern dialect : the inflexion is formed, as usual, by the shortening of the included vowel ; and the crude root ‘ta’ (without the formative an ‘n’) is sometimes used instead of ‘tAn-u,” just as ‘na,’ of the first ph person, and ‘ni’ of the second, are occasionally used instead of sh ‘pan-u’ and ¢nin-u. Po In Telugu the reflexive pronoun is more regularly declined, and is (‘n more in accordance with the Tamil-Canarese, than any other pronoun i THE REFLEXIVE fRONOUN. 317 0 of the personal class. The nominative is ¢tin-u; the inflexion and possessive ¢ tin-a ;’ the plural nominative ‘tim-u’ or ‘tar-u :’ ¢ t2’ may ely be used at pleasure, as in Canarese, for ¢tin-u.’ lan A similar regularity of formation and of declension is apparent in all w 1l the Dravidian dialects, so that further comparison of the forms of this OUS pronoun seems to be unnecessary. The root or base is evidently “ta,’ als seff. The final ‘n’ of the singular, though probably only a formative addition (like the final ‘n of ‘né-n, 7, and ‘ni-n, thow), is one of Ave great antiquity, for we find it even in the Brahui: e g., the nominative V1= singular is tenat’ (compare with this the inorganie ¢t,” which is suf- e, fixed to the personal pronouns in Gond) ; gen, ‘tena; dat. ‘tene.’ Ver ‘tan,’ self (like ‘nén,” 7, and ‘nin,’ thow), is of no gender, and is on- used in connexion with each personal gender indiscriminately; so that of this final ‘n’ has evidently a different origin from the ‘n’ or ‘am, which constitutes the sign of the masculine of rationals. Bhie cnveof the singular of the personal pronouns has nothing to do with gender, and is a sign of the singnlar number alone. ing, The wuse of this pronoun agrees with the use of the corresponding he Indo-European reflexive. It always agrees with the principal nomina- Ui, tive of the sentence, and with the governing verb, or that which is in agreement with the principal uominative. It i1s also used as an pEI 11 g emphatic addition to each of the personal and demonstrative pronouns, o t0 like the Latin ‘ipse,” or like the English self, in the compounds myself, afil yourself, &e.: e.g., we say in Tamil ¢ ndn-tan,” 7 myself ; ‘ ni-thn,’ thou 51v6 thyself ; ‘avan-tan,) he humself; ‘aval-tin, she herself; ¢adu-tan,’ atself or that itself : and ‘tam,” the plural of ‘tan’ (or, in the colloquial dialect, its double plural ‘tdng-gal’), is in like manner appended to the the plurals of each of those pronouns and demonstratives. )la ‘tan’ acquires also an adverbial signification by the addition of the ] by usual adverbial formatives; eg., ¢ tandy’ (for tan-agi’), Tam., of my- the self, of yourself, or spontancously : and when appended to nouns of el § quality or relation its use corresponds to that of our adverbs really, In quite, &c.; e.g., ‘mey tan,” Tam., <t 4s really true, ‘$ari tan,” quite ﬁﬂd right. the One use to which the reflexive is put is peculiar to these languages, [ of —vViz., as an honorific substitute for the pronoun of the second person; Hve and in this connexion either the singular, the plural, or the double first plural may be used, according to the amount of respect intended to be | of shown. When used in this manner, it is not annexed to, or com- pounded with, the pronoun of the second person, but is used alone : and though, when it stands alone, it generally and naturally denotes d 18 the third person, yet when thus used honorifically for the second person, 11()\111 318 THE PRONOUN. the verb with which it is connected receives the pronominal termina- ‘fa tions not of the third person but of the second. This use of ‘tan,’ as ot an honorific pronoun of the second person, illustrates the possibility, if not the probability, of the origin of the Indo-European pronoun ¢ tu,’ Wit thou, from a demonstrative base. tho e A very interesting class of Tamil words, the nature of which has generally been overlooked, has originated from the honorific use of the fror reflexive pronoun. Its inflexion, or possessive, has been prefixed hono- the of ¢ rifically to most of the pure Dravidian words which denote parentsand other near relations, in a manner which somewhat resembles our modern of { periphrasis Her Majesty, your worship, &e. In general the plural stra ‘tam’ has been used in this connexion instead of the singular ¢tan,’ that as a prefix of greater honour, but in some instances these compound The words have become so corrupted that their constituent elements can fron scarcely be recognized. The Tamil ‘tagappan,’ fatker, is formed from 0l d ¢ tam-appan,’ therr (honorifically Ais) father, meaning, as it were, Ais i paternity. ‘tammei,” mother, is from ‘tam-Ayi,’ her maternity ; and find “tAy,” mother, the more common word, is in like manner from °ta-ayi,’ the in which wegfind the crude ‘ta’ for ¢tam.’ ¢ tamayan,’ elder brother from (Tel., ‘tammudu’), is from ‘tam’ and ‘ayan’ or ‘eiyan,’ an n- den structor, meaning kis tutorship. ‘ tang-gei, younger sister, is from suff; ‘tam, and ‘kei, a handmaid (literally a hand), meaning her hand- and mardenship.™ ‘tambi’ (Tel., ‘tammu’), younger brother, and ‘tandei’ Tefle (‘Tel., ‘tandri’), father, evidently include the same honorific prefix Seco ‘tam’ or ‘tan; but the nouns which form the basis of these words WOP( are so changed that they cannot now be recognized. ‘tambirén,” a W title of God, which is commonly applied to a particular class of ety Hindu abbots is formed from the same ‘tam’ and ‘pirdn’ (pro- or ¢ bably from ¢pra,” Sans., before, first), Lord or God, meaning Ais lord- 1 ship,literally his godship : and this is perhaps the only word of this the class the derivation of which has commonly been admitted by lexico- tongl graphers. dem Another remarkable use of the reflexive pronoun is the adoption of ]le, ¢ its possessive, or inflexional base, ‘tan,’ of self or self’s, as the base of “toy the abstract noun ‘tan-mei’ or ‘tanam, quality or nature, literally 1 selfness ; ‘mei’ is the regular formative of Tamil abstracts, like our and This word is identical in meaning English ness, or the Latin ‘tas.’ Dr. with the Sanserit tatvam, nature, property, which is derived from 1_“6&[ i * Compare with this meaning of ‘a younger rister,” the name of ‘spinster’ ‘tuny which is applied by ourselves to unmarried females; and also the derivation which is attributed to ‘dubtri’ (‘duhitar’), Sans., daughter, viz., a mitk-maid Origi (or as Bopp regards it, a suckling), from duh,’ to malk. T S— N — — P —— — THE REFLEXIVE PRONOUN. 319 12 ‘tad’ or état,’ that, and is possibly allied to it (though indirectly) in a8 origin. A ‘ta,’ the base of the Dravidian reflexive pronoun, has no connexion i, with, or resemblance to, any other pronoun of this family of languages; though it is unquestionably a pure Dravidian root. If we look at its has meaning and range of application it must, I think, have originated the from some emphatic demonstrative base; and it will be found that 10 there is no lack either in the Indo-European or in the Seythian family nd of demonstratives closely resembling ‘ta’ or ‘t4-n.’ We see examples eIn of this resemblance in the Sanserit ¢ta-d,” that (from ta, the demon- ol strative, and ‘d,’” the sign of the neuter singular) ; in ‘tada,” then, at A, that time ; and also (with the ‘t’ weakened into “97)-in aaly Be ind The reflexive pronouns of this family, ¢ sva,” &ec., are probably derived cail from the same base, though considerably altered. Compare also the 0l Old Greek article, which is properly a demonstrative pronoun, ¢rds,’ H s ‘7')7,’ S and the corresponding German ¢ der,” ¢ die, ‘das.’ We and find the same or a similar demonstrative (with an annexed nasal, as in iy the Dravidian tan’) in the Doric, ¢ 7fjv-os,” ke, that, which is the form her from which the Aolian ¢ «fjv-os,” and the later Greek ¢ é-xev-os, was m derived (by a change similar to that by which the Hebrew pronominal o1 suffix ‘k&’ was derived from €ti’). The resemblance between ¢ v’ nd and ‘tan’ is certainly remarkable : and may not this Drividian ;dei’ reflexive pronoun, which is used honorifically as a pronoun of the ofix second person, throw some light on that curious, indeclinable Greek onds word which is sometimes used as a form of polite address, viz., ‘rav’ or ), & “@ zav, Sir, My good friend, &c., and which has been derived by some sOf etymologists from ‘fjv-os,” by others from an obsolete vocative of ‘¢’ O if 7151/77 ¥ pIO il The same demonstrative, with a similar final ¢n,’ appears also in this the Old Prussian ‘tan’s’ (for ‘tana-s’), he; and in the Scythian 160~ tongues, we find it, either nasalised or pure, in the Finnish remote demonstrative ¢tuo,” and the proximate ‘tama ; in the Lappish ‘tat’ ke, ‘tan,” of him (root ‘ta’); and in the Ostiak remote demonstrative I Uf ‘ toma,” and proximate ‘tema. o of The reflexive pronoun is used by the Seoni Géond both as a reflexive ally our and as a demonstrative. Thus, in the Song of Sandswmjee, in ning Dr. Manger’s paper (‘Journal of the Bengal Asiatic Society’), ¢ten,’ 7ol means kim (not ¢se, but “illum’); ‘tunna,’ his; and ¢ tane, her and 2t The reflexive signification also appears in the same song in ‘tunwa’ (Tam. ‘tan’), ¢ suus-a-um.” This seems to prove that ‘ta’ was el aciqﬂ originally a demonstrative. v The strongest argument, perhaps, for considering the Dravidian ¢t4’ 2 THE FPRONOUN. d 0 or “tan,” self, to be allied to the Scytho-Sanserit demonstrative ta’ is al the circumstance that ‘tan,’ the inflexional base of ‘tan,’ is used in 1§ ] i the formation of the word ‘tanmei’ or ‘tanam, quality, selfness, in precisely the same manner as the Sanscrit ‘tad,’ that; which forms of the basis of the corresponding Sanserit word ‘tatvam, gquality, that- 0 ness. The Dravidian word may have been, and probably was, framed in imitation of the Sanscrit (for so abstracta term is necessarily of late $01 origin), but it cannot have been directly derived from the Sanscrit SIN; word. It seems very probable that both bases are remotely allied, and nc if they are so allied, th&r alliance carries us back to a very remote of period: for whilst the Dravidian reflexive pronoun retains the original tla demonstrative t,’ the corresponding reflexive in every one of the Indo- $ec European tongues ‘sva, ‘se,; &c.) had already allowed ‘t” to be ref weakened into ‘s, before those tongues separated from the parent stem. e 4. PLURALS oF THE PERSONAL AND RerLEXIVE PRONOUNS. s I class the plurals of these pronouns together because they are the nal formed from the same pronominal bases as their singulars (which have already been investigated), and because they are all formed on one and of the same plan, viz., either by the addition of a pluralising particle it wh (generally ‘m’) to the pronominal base, or by the substitution of that particle for the singular formative. Exceptions exist, but they are few Tl and trivial. It of Comparison of dialects—In the classical dialect of the Tamil, the Ty plurals of the personal and reflexive pronouns (‘néan,’ 7, ‘ni,’ thou, ‘tn,’ ob self) are ‘ndm’ or ‘yam,’ we; ‘nir ‘niyir, or ‘nivir’ (instead of the more regular ‘nim’), you ; and °tam, selves. In the colloquial IEp €1l dialect a double plural has got into extensive use, which is formed by the addition to the classical plurals of ‘gal,’ the sign of plurality ial which especially belongs to the class of wrationals. In consequence of P Uiy the existence of these two sets of plurals, a difference in their use and application has gradually established itself. il we The classical or pure and simple plurals are now used in the collo- int quial dialect as honorific singulars ; whilst the double plurals—¢nang- gal’ (‘ndm-gal’), we; ‘ninggal’ (‘nim-gal’), you,; and ‘tanggal’ (tam- of { or | gal’), selves—are used as the ordinary plurals. A double plural has Ta crept into the Telugu also; eg., ‘mirulu’ (for ‘miru’), yow, and ¢yarulu’ (for ‘varu’), they. Another point of difference between ‘nam’ als and ‘nanggal,” the two Tamil plurals of the first personal pronoun, will be inquired into under a subsequent head. The formation of these 10y secondary, double plurals of the Tamil and Telugu is in harmony with g the oo PLURALISATION OF PERSONAL PRONOUNS. 321 18 a usage which is observed in the Turkish. In that language ‘ben,” 7, n is regularly pluralised into ¢biz, we; and ‘sen,’ thow, into ‘siz,’ you 1 bat those plurals are sometimes pluralised over again by the addition 1§ of ¢ler, the ordinary suffix of plurality ; e.g., ¢ biz-ler, we, ¢siz-ler,’ (- you led In the verbal inflexions of the Tamil, the initial or radical con- afe sonant of each of the pronominal plurals (as of the corresponding OL1G singulars) disappears; and the pronoun is represented solely by the nd ineluded vowel and the sign of plurality. The personal termination ote of the first person plural in the colloquial dialect is ‘6m; in the nal classical dialect ‘4m, ‘am The termination of the ‘ém, ‘em. do second person plural is ‘ir’ or ir,” the representative of ‘nir. The he reflexive pronoun ¢ tdm,” selves, has no place in the verbal inflexions. ent Of the three High Tamil or classical plurals which have been mentioned—¢ ndm, ‘nir, and ¢tdm’—two form their plurals by sub- stituting ‘m’ for the final ‘n’ of the singular, or by adding ‘m’ to are the crude root. This I consider to be the regular method of plu- Ave ralizing the personal pronouns; and the use of ‘mnir,) you, instead nd of ‘nim, is an abnormal exception. This appears on comparing it icle with ‘ning-gal,’ the corresponding plural in the colloquial dialect, which is formed from ¢nim ’— the plural that is required by hat o rule, and which is found in the ancient dialect of the Canarese. 9 It also appears from the circumstance that ‘nir is not the base of the oblique cases of the plural of this pronoun in any dialect of the the Tamil. ‘m’ constitutes the sign of plurality instead of ‘r ? in the all oblique cases of nir,” precisely as in those of ‘ném,” we. ‘nim’ is of represented in the oblique cases in the classical dialect by ‘nam’ and il <em; and by ‘nam’ and ‘enggal’ (‘em-gal’) in the colloquial by dialect. In like manner, the oblique cases of the plural of the second ity personal pronoun are ‘um’ and ‘num’ in the higher dialect; and oof ¢unggal’ (‘um-gal’) in the colloquial. ‘nin,’ the abbreviation of ud ¢nin, being used in the classics as the inflexion of the old singular, we should have expected to find the corresponding ‘nim’ (from ‘nim’) lo in the plural: but both in the oblique cases and in the termination 1g of the plural of the imperative, ‘i’ has given plaee to ‘u,” and ‘ num’ i or ‘um’ has supplanted ¢nim. ‘num,’” the plural inflexion of the has Tamil, is identical with the nominative plural of the Brahui, which is and also ‘num.’ all In Telugu the second personal pronoun is pluralised in the Ut nominative by ‘r’ instead of ‘m,’ e.g., mir-u,” you,; and in Telugu, as 66 in all the other Dravidian dialects, ‘r’ invariably forms the plural of fﬂh the persona,l terminations of the indicative mood of the verb. It will Y 322 THE PRONOUN. be seen, however, in the sequel that there are indications in Telugu that this use of ‘r’ is abnormal. i In Canarese the plurals of all the personal pronouns are formed { in the ancient dialect with perfect and beautiful regularity ; e.g., ¢ an, In the I, ‘am, we; ‘nin, thou, ‘nim,” you,; °tan, self, ‘tam, selves. 0 oblique cases the included vowel is shortened as usual; and the only other change which takes place is in the weakening (as in Tamil) of the i radical ‘a’ of the nominative of the first person into ‘e, e.g., ‘ emma,’ our. al In this particular, ‘ namma,” the form which has survived in the The colloquial dialect is more regular, and evidently more ancient. al modern dialect substantially agrees with the ancient, the chief dif- ference consisting in the softening, in the nominatives alone, of the final ‘m’ into ‘vu; e.g., ‘nivu, ‘nive,’ and ‘tavu,’ instead of ‘nam,’ 11 ‘nim,” and ¢ tAm.’ 8 In the personal terminations of the verb, the modern dialect uses fi ‘éve, ‘evu, and ¢évu,’ as representatives of ‘nivu,” we; the ‘e’ of 6§ which forms corresponds to ¢én,’ the termination of the Tamil sin- th gular. This final ‘vu’ of the modern Canarese is not euphonie, like an the ‘vu’ of the Telugu singular, ‘ni-vu,” thou; but is softened from, th 2 is the true and is the representative of, an older ¢m. Though ‘m of sign of the plural of the second person, as of the other personal pro- nouns, ‘r’ is used instead in all the Canarese verbal terminations, as in those of all the other dialects. The ancient Canarese uses ‘ir,’ the ta modern ‘iri’ and ¢iri.’ of Y In Telugu the ‘m which constitutes the pronominal sign of plu- of K rality is not softened into ¢ vu in the termination of the first person Ie plural of the verb, as in Canarese. That termination is ¢ amu,” ¢ 4&mu,’ “emu, ‘ému; and in the preterite it takes the shape of ¢imi,’ through i the influence of €ti, the preterite formative. The plural of the dia AR 1 second person is represented by ‘aru, ri, ‘eru ¢éru, ‘uru, and | farn of which ‘r, the pluralising suffix of ‘miru,’ you, is the only essential element. e.\p € The Telugu differs from the Tamil-Canarese in occasionally using 80f ‘ tar-u’ instead of tam-u,’ as the nominative plural of the reflexive aly PI‘OHOUH. This irregularity, however, like that of the pluralisation 0rg of the second personal pronoun by means of ‘r’ instead of ‘m,’ dis- it appears in the oblique cases; the plural inflexion or possessive of this pronoun being ‘ tam-a,” in Telugu, as in the other dialects. ‘tamar-u, Pl sometimes used instead of tdm-u,’ is properly a possessive noun. an The Telugu plurals ¢ mém-u,” we, and ‘mir-u’ (or ¢ miru-lu’), yow, 13 present some peculiarities which require to be separately inquired S a into. of | 3 PLURALISATION OF PERSONAL PRONOUNS. i In common with their singulars, the inflexions of these pronouns reject altogether the final consonant—the sign of number —and e retain the long included vowel of the nominative unaltered. Thus, 1, the inflexion or possessive of ‘mému’ is ‘mA,’ and that of ‘miran’ he ‘mi’—corresponding to the singular inflexions ‘nd’ and ‘ni.’ The ly objective case, however, follows the rule of the Tamil and Canarese ; the It may, e.g., ‘mamu’ or ‘mammu, us, ‘mimu’ or ‘mimmu, You. 1, therefore, be concluded that the mode in which the inflexions ‘mi’ the and ‘mé’ are formed is irregular and of late origin; and that in he Telugu, as in the other dialeets, ‘m’ is to be regarded as the received lif and regular sign of the plural of the personal pronouns. the The chief peculiarity of these pronouns (‘mém-u’ and ‘mir-u’) H n, in Telugu, is the change of the initial ‘n into ‘m It has been seen that ‘na’ is the root of the primitive Drividian pronoun of the 1568 first person singular, and ‘ni’ that of the second; that the most of essential portion of those pronouns is the initial consonant ‘n; and SI that the normal method of forming plurals from those singulars is by ke annexing to them a final ‘m.’ How then is it to be accounted for )i that the Telugu plurals have ‘m’ as their initial and radical, instead 106 of ‘n P—mém-u’ and ‘mir-u,” instead of ‘ném-u’ and ‘nim-u’ or 10 “nir-u?’ ,( a5 I believe that this ‘m’ is not to be considered as the represen- the tative of an older pronominal root; but that it is merely the result of the euphonic attraction of the final ‘m,” which constitutes the sign of plurality. I have been led to this conclusion by the following l 500 reasons :— 10, (i.) In the higher and more ancient dialect of the Telugn, oh ‘memu,’ we, is replaced by ‘ému; precisely as ‘énu’ is used in that dialect instead of ‘ nénu,” /. These older forms, ¢ én-u’ and ‘ém-u,’ are the in perfect accordance with the Ancient Canarese ‘4n’ and ‘Am,” and ;md especially with the personal terminations of the Tamil verb, ¢én’ and nly < em.” It 1s demonstrable that the Canarese ‘4n’ and ‘am’ have been sing softened from ¢ yan’ and ‘yam, of which another form is the Malay- ala ‘njan’ and ‘fijAm; and I believe that these are derived by the 176 {100 ordinary change of ‘n’ into ‘fij’ and ¢y, from the Tamil ‘nin’ and d 15 ‘nam.’ ff 118 We thus arrive at the conclusion that the ‘m’ of the Telugu plural is abnormal, and that ‘mém-u’ must have been formed from J-lly an older ‘ném-u; and if, as I have supposed, the normal Dravidian gl ‘na’ itself is allied to, and weakened from, a still older Scytho- p 4 Sanscrit ‘ma,” the remembrance of this, or the surviving influence > of the fact, would tend to facilitate a return of ‘n to ‘m 1m Y 2 324 THE PRONOUN. Telugu; though T doubt not that the euphonic attraction of the *m’ which is used as a sign of plurality, is to be regarded as the immediate the cause of that return. Sel (ii.) If the plural of the Telugu first person alone had ‘m’ for tw its basis, we might possibly suppose that ‘m’ to be radical and pri- €01 mitive, on account of ‘m’ being the basis of the corresponding Scytho- “al Sanscrit pronoun; but we find the same initial ‘m” in the plural of the Telugu second person also. Now, as it cannot be doubted that ter ‘ni,’ the singular of that pronoun—agreeing as it does with the thi Behistun-Scythian and the Chinese, as well as with many of the Th Finnish forms——faithfully represents the primitive Dravidian pronoun of the second person, it seems certain that ¢ mim * (the supposititious ten nominative from which the objective ‘mim-mu’ has been derived) the must have been altered from ¢nim.’ We may, therefore, conclude that Vel the same process took place in the pronoun of the first person also. (iii.) The Telugu is more addicted to harmonic changes than any i other Dravidian dialect. It alters both vowels and consonants for Th harmonic reasons so frequently, that the change from ném-u’ to ‘mém-u,’ and from ‘nim-u’ to ‘mim-u,’ would be thought by Telugu plu per people a very natural and trivial one. Possibly this change throws light on a termination of the impera- Pro tive in Tamil which has not been accounted for. the The ordinary representative of the plural of the pronoun of the Wit second person in Tamil imperatives is ‘um, a weakened form of Wi ‘num; but in the poets we find also ‘min,’ e.g., ‘kén-min’ (theme ‘kel, to hear,) hear ye, instead of the colloquial ¢kél-um.’ Possibly n this ‘min’ is an euphouic displacement of ¢ nim,” that very abbrevia- nf} tion of ‘nim’ which we should expect to find used (instead of ‘um’) in the older dialect. pexs 1y The Ku agrees with the Telugu, in the main, as to the mode in Dr which it pluralizes the personal pronouns. Its nominatives are ¢am-u, we, and ‘ir-u,” you ; but the inflexions of the same are ‘mi’ and ‘mi. fon In the personal terminations of the verb the plural of the first person annexes ‘Amu’; that of the second ‘éru’ or ¢éaru.’ tiop tha The Malayéala plurals are nearly identical with those of the Tamil. Wit The only difference deserving of notice is that the included vowel ver is abbreviated in the nominative plural, as well as in the oblique cases; fOlu e.g., ‘fjingal,’ we, instead of ‘ninggal, and ‘ningal,’ you, instead of ‘ninggal ;’ and that in the oblique cases the initial ‘n’ is not lost, and EXe 2 2 that “a’ changed to ‘i in the first person and ‘1’ to ‘u in the second, any, as in Tamil, but the nominatives themselves are used unchanged as the bases of the oblique cases. the diun PLURALISATION OF PERSONAL PRONOUNS. 3 - 9 5 i In Tulu the plural of the first person is ‘namma,’ instead of ‘nim,’ late the inflexion of which is ‘nama,’ as in High Tamil. The only repre- sentative of this plural in the verbal terminations is ¢va.’ There are for v two forms of the plural of the second person, as in Tamil; viz., ¢ir b I tho corresponding to the Tamil ‘nir,” and the double plural ‘ninggal’ “ar’ represents this plural in the personal terminations of the verb, ural tha The Tuda plural of the first person is ¢6m, as in the personal termination of the verb in colloquial Tamil. The representative of the the this plural in the verbal inflexions is ¢imi,” as in the Telugu preterite. The plural of the second person is ‘nima.’ 011 The Kota plural of the first person is ‘nme,” which in the personal 1008 terminations of the verb becomes ¢ emme.’ The nominative plural of ved) the second person is ‘niye,” of which the inflexion is ¢ nima,” and the that verbal ending ¢ irri.’ In Gond the plural of the first person is ‘amat,” of the second any ‘imat;’ the final t’ of which forms is inorganic and abnormal. for fo The inflexion, like that of the Telugu, is ‘mA’ for the first person plural, and ‘mi’ for the second. The personal termination of the first ugu person plAuraJl of the verb is ‘4m’ or ‘6m ; of the second, ‘rit’ or ¢it.’ eLd- The Uron and Ré4jmahal dialects form the plurals of their personal pronouns regularly by changing the final ‘n’ into ‘m.’ Compare the Urdon ‘em,” we, with enan,’ 7: the Réjmahal ‘nam, ¢om, we the with “en,’” 7, and ‘nim,’ the base of the possessive “nim-ki,” your, | Of with the singular nominative ¢nin,’ #how. eme In Brahui the plural of the first person is ‘nan;’ that of the second ibly ‘num,” which is identical with one of the Tamil plurals. In the verbal VI inflexions the final ‘n’ of ‘nan,” we, represents the plural of the first n’) person;’ e.g., ‘aren,’ we are: in the second person the final ‘m’ of ‘num’ disappears, and ‘ri’ is used instead, precisely as in the true o 10 Dravidian dialects; e.g., areri, you are. -l The result of the foregoing comparison is, that the first person and forms its plaral in all the Drividian idioms (with the solitary excep- first tion of the Brahui) by changing the final formative ‘n’ into ‘m:’ that the second person originally formed its plural in the same manner il without exception, viz., by substituting ‘m’ for ‘mn,” though the owel verbal endings and the nominative of the isolated pronoun are now 1968) found to prefer ‘r;’ and that there is but one solitary and trivial l of exception (viz., that of an optional Telugu nominative) to the rule and that the reflexive pronoun also forms its plural by discarding ‘n’ and tOIld; annexing ‘m Consequently we are now entitled to regard ‘m’ as d 88 the regular and ancient sign of plurality which is used by the Dravi- dian personal pronouns. 326 THE PRONOUN. ¢ ) ¢ ar, 1T or ‘r,’ is the plural of all ‘rationals’ in the Dravidian languages, with the exception of the three personal pronouns; and the existence of this exception constitutes ‘r’ a sign of the plural of the 1 third person. How then has a termination which is peculiar to the third person found its way into the second ? In this manner, I appre- hend:—¢nir,” or more fully ¢ niyir’ (ni-(y)-ir,) means literally thow+ [ they; and this compound would necessarily bring out the signification d you. The Sanscrit ¢ yushmé’ (¢ yu+smé’), you, is supposed to have a I similar ori gin. Bxtra-Dravidian relationship.—W e now proceed to inquire whether ] final ‘m,” the distinctive Dravidian plural of the personal pronouns, forms the plural of this elass of words in any other family of langunages. al Ul ‘m’ having a tendency to be weakened into ‘n’ (of which there are many examples in the terminations of Tamil nouns), and ‘m’ and fl ‘n’ being generally equivalent nasals, the use of a final ‘n’ as a sign P of the plural of pronouns, may possibly be equivalent to that of ‘m.” If so, we may adduce as examples of plurals allied to the Dravidian €0 the Brahui ‘nan, the Chaldee ¢anfn,” and the Ostiak ‘men,” we; as also the Persian ¢tan,” yow.’ A slight trace of the use of ‘m’ as a sign of the plural may be noticed in the Beluchi ‘ mimiken,” we, (2 th when compared with ‘menik,” 7. In the Ostiak,a Finnish dialeet, the first person plural of the verb terminates in ‘m, whilst the plaral of 9 § the corresponding pronoun terminates in ‘n On comparing the Finnish proper olen,” I am, with ‘olemme,” we are, we are struck with their resemblance to the Dravidian rule. The resemblance, how- i Cl ever, is illusory; for the ‘m’ of the Finnish ‘me” is a sign of per- ? sonality, not of plurality. ‘me, we, is the plural of ‘ma,” the old Finnish 7; of which ‘na’ (from which the ‘n’ of ‘olen’ arises) is, a8 as I have shown, an euphonic modification. We can scarcely indeed be expect to find in the pronouns of the Scythian languages any sign of 81 n plurality perfectly corresponding to that of the Drividian ‘m;’ for in those languages the personal pronouns are generally pluralized by P a change of the final vowel, not by any change or addition of conso- W, nants: e.g., Manchu “bi,’ 7, ‘be, we; Magyar ‘te, thou, ‘ti, you; ) Ostiak and Finnish ‘ma,” 7, ‘me’ (or ‘men’), we. I have reserved till now the consideration of a series of close and remarkable analogies which run through the whole of the Indo-Euro- J pean family of languages, and which are found also in the North- 4 Indian vernaculars. In those languages we find very frequent use of H “m’ in the plurals of the personal promouns, in which it either con- ] stitutes the final consonant, or occupies a place of evident importance ; ol T ————— — —— " PLURALISATION OF PERSONAL PRONOUNS. 327 9 dian and this ‘m’ in some instances appears to replace a final ‘n’ or ‘n | the which is used by the corresponding singulars. the In the vernaculars of Northern India we find the following ) ) the instances of the use of ‘n’ or ‘n in the singular and ‘m’ in the plural. Hindi ‘main, 7, ‘ham,” we; ‘td, ¢tin,” or ‘tain,’ thou, pe- 0+ ‘tum,” yow: Gujarathi ‘hun, 7, ‘hame,’ we; ‘tin, thou, ‘tame,’ tion youw: Marathi tun, thow, ¢tumhi,’ you. In Bengali and Uriya ‘n’ V6 & disappears from the terminations of the singulars, but in the plural ‘m retains its place as in the other dialects: e.g., Bengali ‘toma’ or “tumi,’ the inflexional base of the plural of the second person; and other ¢ Uriya ‘tumbha,’” the base of ¢ tumbhaméani.’ The same distinctive ¢ m ous, appears in the Pali-Prakrit ¢ tumhe, you, ‘amhe,” we. Compare also the New Persian ‘shumd,’ you, and the final ‘m’ of ‘hastém,’ ages there we are. " and Similar and very striking analogies meet us in Greek. Compare the singulars ¢éywv and ¢ 7odv, ‘éywrvy,’ and ¢7ovwy’ with the g it plurals ¢ 7},uefs- > and ¢ opels)’ This resemblance too is strengthened dian when the vowels of the Greek plurals are compared with some of the corresponding Drévidian ones: e.g., compare ‘yu-ets’ with the Telugu e a8 ‘ém-u,” we;’ and ¢ vu-cts’ with ‘um,” which is the base of the oblique " 10ty cases of the Tamil plural of the second person, and is used to represent that pronoun in the plural of the imperative. , the al of It also deserves to be noticed, that in the Greek, Persian, Gaurian, &e:, ‘m 9 is not used indiscriminately by all nouns, or even by all the pronouns, as a sign of plurality in general, but is invariably restricted Eruck how to the pronouns of the first and second person—a usage which pre- cisely accords with that of the Dravidian languages. per A strong case for regarding the ‘m’ of the Indo-European idioms e old as allied to the plural ‘m’ of the Dravidian family has doubtless now ) 5 been established ; and yet this resemblance, though so exact and con- deed sistent, will be found on investigation to be entirely illusory! On a ;ﬂ Of ' for more extended comparison it diminishes, and at last it disappears. Perhaps, indeed, no better illustration can be found than that which ’dby will now be adduced, of the danger of confiding in apparent resem- 10" blances, and of the value of comparison in philology. ) g/()lé g The resemblance of the final ‘n of the North-Indian and of some Greek singulars to the final ‘n’ of the singulars of the Dravidian and pronouns, though probably accidental, is to be classed in a rather ur0- different category from that of the plural ¢m. The final ‘%’ of the ortl- Hindi ‘main,” “tun,’ &c, is an euphonic and purely inorganic nasal, 156 o which adds nothing to the grammatical expression: this is also the (on chéracter of the ‘v’ of the Greek ‘éywr’ and ¢7odv;’ and the origin 168 § 328 THE PRONOUN il of these nasals is to be attributed to the euphonic influences from which the final ‘n’ of the Tartar men’ and sen’ proceeded ; not tum I think, to the neuter termination of the Sanscrit pronouns ‘ah-am I and ¢tva-m On the other hand, the final ‘mn of the Dravidian comp promouns is not a mere inorganic or euphonic addition, but is used bu ence distinctively as a sign of the singular, and in most of the dialects evinces greater persistency than the initial and radical ‘n’ itself. This Though, however, in actual use ‘n is a sign of the singular, it B may possibly have proceeded originally from an euphonic origin ; and expl this view is confirmed by the circumstance that in Canarese it is (for regarded as a formative, and accordingly is optionally dispensed with | and the crude, unformed roet, without this addition, is occasionally foune This ¢ used as the nominative singular. n may, therefore, after all, Prar have some ulterior connexion with the final ‘n’ of the Grazco- Gaurlan _(( as well as the Seythian singulars. Bop This disturbing element being eliminated, we come now to the re- 01121} semblance which is found to subsist between the Graaco-Gaurian plural 10 i m’ and the final ‘m’ of the Dravidian plurals On extending our thirg comparison a few stages, so as to include those dialects which exhibit Singy he original character of the Indo-European pronouns, no trace of a i an ) VR connexion between the ome ‘m’ and the other, will be found to Semi survive and evide 7//146(,8' vpers’ are not the oldest forms of the Greek plurals flan — For ¢yueis, the Doric and ARolic dialects have apes tL,LL/LLC“ and L3 (Llulll,LG for ¢vueis’ they hav v,uex UpLES, and ‘u/u./ue of which forms, the oldest and most reliable appear to be ¢duues,’ or its he o uninflected type ¢ dupe, and ‘ bupes’ or ‘ upe When “duue,” we, is Zag now compared with the corresponding Prakrit amh with the hey Gujarathi ‘hame,” with the Zend (supposititious) ‘ahme,” from which ot proceeds the possessive ahmékem’ (corresponding to the Prakrit have ‘amhakam’), our ; and finally with the Vedie-Sanscrit ¢asmeé,” we, it On is evident that the last-mentioned form, ¢asm is the normal type 13- from which all the rest are derived. The progression is very clear— he< ¢ asme ahmé amhé u/LL/.LE Y B g ) aulue-eg s — 7/ILLCLS’ ‘hame ‘ham ¢ he In like manner on comparing 'U/,l,LLES‘ or Yppe, you, with the New Persian ¢shum €Va with the Zend ¢ylishem’ (in the oblique cases aly yusma’), and with the Vedic-Sanscrit ¢ yushmé’ (for yusme ), it is Ve equally obvious that ¢yusmé’ is the root of the whole. ¢ yusmé,’ you, add1 the plural of tu,’ tkou, has probably been softened from ‘ tusme,’ that = ‘tu-smé’ (as ‘asmé’ from ‘masmé = ‘ma-smé’): and this suppo- bYG sititious ‘ tusmé’ (weakened into ¢ tuhmé,” like ‘asmé’ into ¢ahmé’) by becomes a reality, when we turn to the Prakrit ¢ tumheé,” you,—from 4 g PLURALISATION OF PERSONAL PRONOUNS. 329 s from which comes directly the Gaurian ¢ tumhi,” ¢ tumbha,’ ¢ tame,’ d; nat, ‘tum, &e. ab-am’ 5 ) It has now been ascertained that the Dravidian final ‘m is to be avidian compared, not with the apparently equivalent ‘m’ of ¢ ham’ and ‘tum,’ is used but with the Vedic particle ‘sme ;" and the improbability of the exist- dialects ence of any connexion between these two is evidently very great. olf, This improbability increases when the origin of ¢sme’ is investigated. ular, it Bopp believes ‘sma’ to be a pronoun of the third person, and n; and explains ‘a-smé’ (for ‘ma-smé’), we, to signify 7+ they, and ‘yu-smé’ % 1t B (for ¢ tu-smé’), you, to signify thou+ they. A bl d with, Moreover, though the ‘m’ which is derived from this ¢sm (& 1S tonally found only in the plural in Sanscrit and Greek, yet in Zend, Pali, and fror all Pracrit it is found also in the singular ; e.g., Pracrit ¢ mamammi i, = (‘mama-sm’-i’), m me, ‘tumammi’ (= ‘tuma-sm’-i’), in thee. Bopp supposes this use of ‘sm’’ in singular pronouns to be of late the re- origin, and to have arisen from imitation of the plurals: but there is 1 plural no proof whatever either that ‘sm’’ was originally a pronoun of the g our third person, or that the plural has a better right to it than the xhibit singular. Possibly it may have been a particle of specification, like the g of & ‘an’ prefixed to the pronouns of the first and second persons in the Semitic languages. But whatever may have been its origin, it is now evident that it is entirely unconnected with the ‘m’ that forms the plurals. sign of plurality which is used by the Dravidian personal pronouns, } and which Origin of this sign of plurality.—We have now to inquire whether o 1t the origin of this plural ‘m’ can be discovered in the Drividian lan- 10ty 18 guages themselves, seeing that no trace of it is discoverable in any b the other family. It appears to me to have been derived from ‘um, the which conjunctive or copulative particle of the Tamil, and which appears to have been the primitive form of this particle in the other dialects also. pakrit ulg} ]t On this supposition ‘ nam,” we, and “nim,” you, resolve themselves into ] type ‘na-um,’ ‘egoque, and ‘ni-um,’ ‘fuque.”’ This view is corroborated by the extensive use which is avowedly made of this very' “um' >y Jpar— am. the formation of Tamil distributive and universal pronouns. Thus, h the ‘evanum,’ every ome, ‘engum,’ every where, ‘ ubique,” and ¢ epporudum,’ always, every time, are unquestionably and avowedly derived from ) 3568 “evan, who? ‘engu, where? and ‘epporudu, what time? with the | it18 addition in each instance of the conjunctive particle ‘um, and ; so ) you, that the compound pronoun every one is regularly expressed in Tamil usﬂlé;’ }uppO, by who ? and— ; every where, like ‘ubique,” by where ? and — ; always, by what time? and—. In the same manner ‘um’ is annexed mt) from as an auxiliary to some affirmative universals for the purpose of 330 THE PRONOUN. widening their application ; e.g., ¢ ellA-(v)-um, Malayalam, all, literally h Qcy all and——, from “ella, all, and ‘um, and. This form is abbreviated Sanserl 1t 18 fon in Tamil into ¢ ellam ;’ which is regarded and treated by grammarians as a plural ; and if the addition of ‘um, abbreviated to ‘m,” un- spoken doubtedly constitutes pronominal distributives and universals, may not we—0f the sign of plurality which is employed by the personal promouns be lian ha an abbreviation of the same ‘um ? A parallel case appears in Ostiak, guages, in which the sign of the dual number (‘ ga, ‘ka, ¢gai, ‘gan,’ &e.) is Austra derived by Castrén from ‘ka’ or ‘ki, also. Al 1ouns Twofold plural of the Dravidian pronoun of the first person.—The which ordinary plural of the Dravidian first personal pronoun is constantly ‘nam, 1 used, not only as a plural, but also as an honorific singular, precisely prope] as the Royal and Editorial we is used in English ; and the plural of ‘ningg every other Drividian pronoun may optionally be used as an honorific ‘engng singular in the same manner. It is not, however, this twofold signifi- Th cation or use of the same pronoun to which I now refer ; but the exist- the Ta ence of two pronouns of the first person plural, which differ from one this 1 another in signification almost as much as the plural and the dual of restrict other languages. from 1] In all the Dravidian dialects, with the exception of the Canarese its hon and the higher dialect of the Tamil, there are two plurals of the pro- a8 an noun of the first person, of which one denotes, not only the party of than the speaker, but also the party addressed, and may be called the plural for enclusive ; the other excludes the party addressed, and denotes only the for | party of the speaker, and may be called the plural exclusive. Thus, if addres a person said we are mortal, he would naturally use the we which in- himge] cludes those who are spoken to, as well as the speaker and his party, Th or the plural vnclusive: whilst he would use the plural exclusive, or that Origin which excludes the party addressed, if he wanted to say, ‘we’ are which Hindus ; ‘you’ are Europeans. ()Miqu There is a similar distinction between the two plurals of the first person which are used in the MarAthi and the Gujarathi: e.g., hame’ Lespor “man; in Gujarithi, means we—the party speaking ; whilst ¢ 4pane’ means we Proby —the party speaking, and yow also who are addressed. There is no con- Cupho nexion between the particular pronominal themes which are used for this purpose in Northern India and in the languages of the South; but the existence of so remarkable an idiom in the North-Indian family, as well as in the Southern, demonstrates the existence in the Northern family of an ancient under-current of Dravidian, or at least of Scythian Ungg; influences. The idiom in question is a distinctively Scythian one, and of is one of those points which seem to connect the Dravidian family with tkm‘, ' DEMONSTRATIVE PRONOUNS. 331 , literall y the Scythian group. There is no trace of this twofold plural in the breviated Sanscrit, or in any of the languages of the Indo-European family, but mmarans it is found everywhere in Central Asia, in the languages which are ‘m’ u- spoken by the primitive, nomadic tribes. Thus the Manchu has ‘mi,’ , Inay not we—of the one party, and ¢be,” we—the whole company. The Mongo- nouns be lian has a similar idiom ; and it is found also in the Polynesian lan- n Ostiak, guages, in many of the languages of America, and also in those of the ) de) is Australian tribes. All the Dravidian languages do not use precisely the same plural pro- nouns as nclusive and exclusive plurals. The colloquial Tamil (with m,~1he which the Malayilam and Tulu agree) forms the plural exelusive from nstantly ‘nAm,” the ordinary and regular plural, by the addition of ‘gal,’ which is precisely properly a neuter sign of plurality ; by which addition ‘nim’ becomes plural of ‘nAnggal’ in Tamil; ‘djangal’ or ‘fijangngal’ in Malayélam ; and honenfie ‘engngal’ in Tulu. d siguifi- The Telugu, on the other hand, uses ‘mém-u’ (answering not to he exist- the Tamil ‘ninggal,” but to ‘ném’) as its plural exclusive; and as Tom 0ne this is the simplest form of the pronoun, it seems better suited to fthis dual of restricted use than the reduplicated form. The Telugu, though differing from the Tamil in this point agrees with the Tamil in using ‘ mému’ as lanarese its honorific singular ; and this use of the plural exclusive in Telugu as an honorific is more in accordance with philosophical propriety he pro- patty of than the Tamilian use of the plural inclusive for this purpose : g plural for when a superior addresses inferiors, it is evidently more natural for him to make use of a plural which excludes those whom he oulythe Thus, if addresses, than one in which they would be included together with himself. fich in- The Ku agrees with the Telugu, and uses ‘4m-u’ (identical in s partfs orthat origin with the Tamil ‘nim’) to express the restricted signification we' oIt which the Tamil gives to ‘nénggal’ Its plural inclusive is “&ju,’ the oblique form of which is ‘amma; and the Telugu plural which cor- responds to ‘4ju’ (but which in meaning corresponds to ‘nam’) is the first hame’ ¢manam-u,’ the base and inflexion of which is ‘mana.’ ‘manam-u’ is eans 1 probably derived from ‘mé4,’ the inflexional base of ‘mému ; with an euphonic addition, or possibly with a weakened reduplication. 110 €01 forthis hut the IT.—Demonstrative Pronouns. lﬁ}U a8 [)rt]lel'ﬂ The Dravidian languages, like most, if not all, other primifive, uncompounded tongues, are destitute of pronouns (properly so called) cythiml e, 4 of the third person, and use instead demonstratives signifying this or that, with the addition of suffixes of gender and number. In these it 332 THE PRONOUN. stiafive languages ‘he,’ means literally that man ; “she,’ that woman ; and “ they,” those persons or things. suffix 0 the mas The words which signify man and woman have gradually lost the ‘4, wi definiteness of their original signification, and shrunk into the position neuter of masculine and feminine terminations, They are no longer substan- in Telu tives, but mere suffixes or signs of gender ; and are so closely incorpo- rated with the demonstrative bases that it requires some knowledge of s {d’ the principles of the language to enable us to separate them. In lar whi comparison, therefore, with the Turkish and Ugrian languages, in which there is but one pronoun of the third person, the Dravidian languages, hiatus euphon which possess three, appear to considerable advantage. Nevertheless, the speech of the Dravidians was originally no richer than the other singula bases, Scythian idioms ; and it has at length surpassed them only by the uses ¢ Aryanistic device of fusing that-man, that-woman, that-thing, into single, euphonious words. ANDEXY The signification of man and woman still shines through in the narily masculine and feminine terminations; but no trace remains of the ‘aR-am words by which a thing and things were originally expressed, and it was and fe which are now represented only by *d,’ the sign of the neuter sin- gular, and ‘a,’ that of the neuter plural. the fin Four demonstrative bases are recognised by one or another of the ialeet Dravidian dialects, each of which is a pure vowel sl Pttt person remote, ‘1’ the proximate, and ¢u,’ the medial demonstrative ; e‘g'a ; together with ¢ A2 S which is the suffix of emphasis in most of the “ejg dialects, but is a demonstrative in Ku. The first two, viz., ‘a,” the T remote, and ‘i, the proximate demonstrative, are the most widely and ‘I, fo frequently used. and ¢ The medial ‘u’ is occasionally used by the Tamil poets, in Ancient suffixe Canarese, and in Tulu, to denote a person or object which is inter- for the mediate between the remote and the proximate ; and it will be found isthe that it has ulterior affinities of its own. ¢, the ordinary Dravidian follow suffix of emphasis, is used as a demonstrative in Ku alone,—in addition ‘avan however to ‘a’ and ‘i eg., ‘évaru, they. It appears also in the ‘i\’ar, that. Uraon “edah,’ this, the correlative of ¢hiidah,’ The use of <&’ Tamy] being chiefly emphatic, I refer the reader, for an account of it, to a gende subsequent head. b 1 The ordinary remote and proximate demonstratives of the Dravi- gende dian dialects are the simple, short vowels ‘a’ and “i; and it will be The, found that every other form which they assume is derived from this by only some euphonic process. T ﬂle e L. Demonstrative pronouns.—The original character of the demon- (ﬂlphc Spar; DEMONSTRATIVE PRONOUNS. 333 o, and strative bases is best exhibited by the neuter singular ; the formative suffix of which does not commence with a vowel, like ‘an’ and ‘al; ot the the masculine and feminine suffixes, but consists in a single consonant position ¢d, with an euphonic vowel following it. The remote and proximate substan- neuter singulars are in Tawmil ‘adu,” that (¢thing), ‘idu, this (thing) 5 in Telugu ‘adi,” ‘idi; in Canarese ‘adu,’ ‘idu; in Malayélam ¢ ata, incorpo- ledge of ‘ita ; in Gond and Tuda ‘ad, ¢id. em, In ‘d’ having already been shown to be the sign of the neuter singu- in which lar which is used by pronominals and appellatives, and there being no IguAges, hiatus between ‘a’ or ‘i’ and ‘d,’ and therefore no necessity for rtheless, euphonic insertions, it is evident that the ‘a’ and ¢i’ of the neuter singulars cited above constitute the purest form of the demonstrative he other bases. In addition to ‘adu’ and ‘idu,’ the High Tamil sometimes by the uses ‘adan’ and ¢idan.’ 1, 1ufo These forms are probably derived from the annexation to ‘ad’ and ‘id’ of ‘am,” which is dialectically and ordi- v h in the narily convertible into ‘an (E£.g., ‘ar-an,” wvirtue, is identical with of the ‘ar-am.’) ‘am’ is a formative of neuter nouns ; and I conceive that it was not added to ‘ad-u’ and ¢id-u,’ till it had ceased to be known od, and and felt that ‘d’ was itself a sign of the neuter singular. ‘dan,’ for §ill- the final portion of ‘adan’ and ‘idan’ is sometimes used in the high of the dialect, instead of ¢du,” as the pronominal termination of the third person neuter singular of the participial noun, especially in the dative; N the e.g., ‘Seygiradan-ku’ (euphonically ¢g$eygiradar-ku’), instead of rative ; ‘$eygiradu-kku,’ for or to the doing. of the The suffixes which are annexed to the demonstrative bases ¢a’ and %’the ¢i, for the purpose of forming the masculine and feminine singulars Iy and Those and the epicene and neuter plurals, commence with a vowel. suffixes are in Tamil ‘an,” for the masculine ; ‘al,’ for the feminine ; ‘ar,’ pclent for the epicene plural; and ‘ei’ or “a,’ for the neuter plural ; and ‘v’ Inter- The fonnd is the consonant which is most commonly used to prevent hiatus. following, therefore, are the demonstrative pronouns of the Tamil, viz., :vidiaﬂ ‘avan,’ ‘ille;’ ‘ivan, ‘hic;’ ‘aval’ ‘illa;’ ‘ival) ¢ heec;’ ‘avar, ‘dlli;’ dition ‘ivar, 9 P < avel Sillay Cived) fihee) I quote examples from the n the Tamil alone, because, though different formatives of number and of ‘6 gender are sometimes anuexed in the other dialects, those differences , t0d do not affect the demonstrative bases. All the above suffixes of gender have already been investigated in the section on ‘The Noun.’ rivi- The mode in which they are annexed to the demonstrative bases is the ill be only point which requires to be examined here. is by The demonstrative bases being vocalic, and all the suffixes, with the exception of the neuter singular, commencing with a vowel, the euphonic consonant ‘v’ had to be used, to keep the concurrent vowels o separate and pure. 334 THE PRONOUN. ) ‘v, though most frequently used to prevent ¢hiatus,’ is not the only consonant that is employed for this purpose. The Ku being but and ¢ a@tion little attentive to euphony, it sometimes dispenses altogether with the they euphonic ‘v, and leaves the contiguous vowels uncombined ; e.g., A -} ¢ A a afiju,’ ke, ¢ aalu, she. In Tulu ‘y is sometimes substituted for either ‘v, eg., ‘4ye, he: and even this ¢y’ disappears in the corresponding tfy 18 ot feminine, ‘al’ (for ‘aval’), ske; in which the two contiguous vowels are combined. Even the Tamil sometimes combines those vowels invar instead of euphonically separating them : e.g., ¢ yavar, who ? is com- Malay monly abbreviated into ¢ yar ; and this is still further softened to ‘ar,’ TOwe dialec in the colloquial dialect. nitial In the higher dialect of the Tamil, ‘n’ is often used euphonically instead of ‘v, especially in the personal terminations of the verbs. it bec T Thus, instead of ‘irundan’ (for ‘irundavan’), ke was, the poets some- times say irundanan; and for ‘irundava, they (neuter) were, the whick form which we should expect to find used, ¢irundana’ is universally emph used instead. demoy This euphonic ‘v’ has in some instances come to be regarded as Tacter an integral part of the demonstrative itself. In the nominative plural energ of the Gond neuter demonstrative, the final and characteristic vowel ‘a> folloy has disappeared altogether, without leaving any representative ; e.g., Secon ‘av, those (things), iv, these (things). In the oblique cases ‘a 18 In represented by ‘e.’ In Telugu, though the nominatives of the neuter wouds Tam plural demonstratives, ¢ avi’ and ©ivi,” use ‘v’ merely as an euphonie, yet in the oblique cases, the bases of which are ‘va’ and ¢vi,” the de- least monstrative vowels have got displaced ; and ¢ v’ stands at the beginning o of the word, as if it were a demonstrative, and had a right per se to be T represented. In the masculine singulars ‘vadu,’ ‘ille, ‘vidu, ‘hic,’ used and in the epicene plurals ¢ varu,” ¢4, ¢ viru,” ‘A, ‘v’ euphonic has (With« advanced a step further, and assumed the position of a demonstrative meth in the nominative, as well as in the inflexion. That this ‘v,” how- in ever, is not a demonstrative, and that the use to which it is put in donh) Telugu is abnormal, is shown by the fact that in ‘d4’ and °di, the ane |, inflexions of ‘adi’ and “idi, ¢¢llud’ and ‘’%oc, the neuter singular S0nay demonstratives of the Telugu, d,’ though certainly not a demon- arg g strative, nor even euphonic, but simply a sign or suffix of neuter di‘alee singularity, has been advanced to as prominent a position (by a similar haye euphonic displacement) as if it belonged to the root. Pr eﬁx In Tulu “avu,” which is properly the plural neuter, is used for the iﬂugt singular ; whilst ‘atu,” (corresponding to ‘adu’), is used to signify fur yes. A similar use of a plural form for the singular appears in the Old el Persian ‘ava,” ¢, which appears to be derived from ¢ava, those. \WE DEMONSTRATIVE PRONOUNS. 33 5 of the 2. Demonstrative adjectives.—When the demonstrative bases ‘a 10 hut and ‘i’ are simply prefixed to substantives, they convey the signifi- h the cation of the demonstrative adjectives that and this. When prefixed, ) they are indeclinable ; but on thus prefixing them to substantives, £ 101 either the initial consonant of the substantive is eaphonically doubled, nding e.g., ‘anndl’ (‘a-(n)nal’), Tam., that day ; or if this euphonic doubling YOWe is not resorted to, the demonstrative vowels are lengthened. The Tamil VOWels | invariably adopts the former plan: the latter is more common in the § COM Malayalam and Canarese. Where the substantive commences with a 0 “al, vowel, and ‘v’ is inserted as usual to prevent hiatus, the Tamil, by a dialectic rule of sound, doubles this ‘v,” as if it were regarded as an nically initial consonant : e.g., when “Gr,” Tam., ¢ willage, receives this prefix, verbs, it becomes not ‘avir’ (‘a~(v)-tr’), but ‘avvir. S0me- The origin of this doubling of the initial consonant of the word to which the demonstrative vowel is prefixed, is to be ascribed to the v, the ersally emphasis which is necessarily included in the signification of the demonstrative. Through this emphasis ‘a’ and ‘i’ assume the cha- Jed a8 racter, not of ordinary formatives, but of qualifying words; and the plural energy which they acquire influences the initial consonant of the following substantive, which is no longer an isolated word, but the el ‘ot ,€y, second member of a compound. In the same manner and from a similar cause, when Sanscrit ‘3’is neuter words which commence with ‘a’ privative are borrowed by the _)honic, Tamil, the consonant to which ‘a’ is prefixed is often doubled, at least in the colloquial dialect; e.g., ‘affijinam’ (‘a-(fij)-fijAnam,’ g de- nping tgnorance. The occasional lengthening of the demonstrative vowels, when ;1o be e used adjectivally, in Malayalam, Canarese, and the other dialects ic has (without the doubling of the succeeding consonant), is merely another method of effecting the same result. The emphasis which is imparted rafive how in this manner to the demonstrative, is equivalent to that which the doubled consonant gives; and hence when the demonstrative vowels put in ' are lengthened, from a and ‘1’ to ‘4’ and ‘i, the succeeding con- i the The fact that the demonstrative vowels sonant always remains single. nglﬂ{ll‘ are short in the pronouns of the third person in each of the Dravidian ool dialects without exception, shows that those vowels could not originally ﬂelltel have been long, and that the use of long ‘4’ and ‘i’ as adjectival milat prefixes, instead of ‘a’ and ‘i’ is owing to emphasis. Some curious o i 16 illustrations of the lengthening of a vowel through emphasis alone, are furnished by the common speech of the Tamil people; e.g., ‘adigam,’ 0] Iy Il much, large—a word which is borrowed by the Tamil from the Sanscrit eV —when it is intended to signify very much, is colloquially pronounced ne— 336 THE PRONOUN. den ¢ adigam.’ Similar instances might be adduced from each of the ‘an colloquial dialects. In addition to the use of the simple vowels ‘a’ and i, and their Tel (A2 Can equivalents a and ‘1, as demonstrative prefixes, the Tamil makes Al much use also of a pair of derived demonstrative adjectives, viz., ‘anda,’ that, and ‘inda,’ this; e.g., ‘anda maram,’ that tree, ‘inda nilam,’ “ind adye this piece of land. These demonstrative adjectives are unknown to the Sam other dialects of the family; and in the higher dialect of the Tamil itself they are unused. Their bases ‘and’) €ind’) are evidently iden- the anot tical with those of the Telugu adverbial nouns ‘and-u,” ¢ind-u,” that or I have n 0 this (place or thing) ; e.g., ‘indu-16,” in this, ‘andu-ku,’ to that. no doubt that these Telugu words, ‘andu’ and ‘indu,’ are either form I derived by euphonization from the Tamil demonstrative pronouns ¢adu’ and ¢idu,” or, which is more probable, from the addition to the whic demonstrative bases of the formative ‘du,” nasalized to ‘ndu; in the form: the g same manner as in Tamil ‘gu’ and ¢du,” nasalized to ‘ngu’ and ‘ndu,’ will be found to have been annexed to the same demonstrative bases, g for the purpose of forming similar adverbial nouns. If to these lere, that adverbial nouns, ‘and-u,” ‘ind-u,” we now add ‘a,’ the formative of the > relative participle, by annexing which so many adjectives are formed, W £ we arrive at ‘anda’ and ‘inda,’ the Tamil demonstrative adjectives. tigitkg 1) 3. Demonstrative nouns: their use as adwerbs. — The Dravidian ing § Giong languages form nouns of place, time, and relation into demonstrative bage adverbs (which are adverbs in use only, but nouns in form) by pre- fixing to them the demonstrative bases. Vidiay with The most frequently used words of this class are those which D Ky} signify then and now; and they are formed by simply prefixing the demonstrative vowels-—‘a’ (remote) and ‘i’ (proximate)—to any noun Stratiy With which signifies #ime: e.g., from ¢porudu,’ ¢ime, the Tamil forms ¢apporudu’ (‘a-(p)porudu’), then, that time, and ¢ ipporudu,’ now, this Tt time ; from the corresponding noun ‘pudu,’ the Telugu forms ‘appudu,’ e\'iden has 1 and ‘ippudu; and from ‘pol’ the Malayalam forms €appol’ and “ippol.’ Each of these adverbs has the signification of a noun in the Stangj locative case; so that the expressions, that time, this time, signify, in Canill’\ Tt that time, in this time, &c; but the case-signification, though understood, is not expressed. In the Dravidian languages indeed, not nouns of and ¢ time and place only, but all nouns may be used as adverbs, with or Otlipg withovt the addition of the suffixes of cases; and all adverbs are 0n R, either nouns or verbs. {d’ 0) A class of words which more nearly resemble our ‘adverbs’ than {]lm’ i fOl'm;u the compounds referred to above, are formed by annexing to the 4 DEMONSTRATIVES. 33 ( f the demonstrative bases certain formative suffixes: e.g., compare the Tamil “angu,’ there, ‘ingu,’ here; ‘andru,’ that day, ¢indru,’ to-day,’ the their Telugu ‘andu,” in that place, there; ‘indu,” in this place, here; the makes Canarese ‘andu,’ that day, ‘indu,’ to day; ‘alli,) in that place, there, A ‘illi,y) @n this place, here; also the High Tamil ¢ an du, there, then; Vi, ilam, ‘indu,’ here, now; and the Tulu ‘anchi, there, ¢inehi,” here. Similar to the adverbs are formed from the interrogative vowel ‘e’ by annexing the Tamil same formative terminations. In one of the examples given above, riden- the ideas of place and time are evidently regarded as identical ; and fhat ot another, ‘andu,’ is an adverb of place in Telugu, and an adverb of time [ have in Canarese. The Tamil ¢ andru,’ that day, is evidently an euphonised ither form of the same word. pnouns In the Tamil adverbs ‘angu,’ there, ‘inga,” kere, the formative fo the which we find to be employed is ¢ gu’—one which is much used in the in the formation of ordinary nouns, and which in those nouns as well as in the adverbs before us, is often euphonized into ‘ngu.’ The primitive, | ‘1}('1”1, unnasalised formative is seen in the corresponding Génd adverbs ‘iga, Dases, here, ‘haga,’ or ‘aga, there; whilst in the Gond ‘inga,” now, we see ) these that in that dialect also ‘g’ has occasionally been nasalised into ‘ng.’ of the We may compare also the Goénd adverbs ¢hike, Zither, ¢ hoke, ormed, thither. 1S, The resemblance between the Gond ‘iga,” Zere, and the correspond- ing Sanscrit ‘iha’ is remarkable; yet it cannot be supposed that the Avidian Gond word has been borrowed from the Sanserit. The demonstrative strafive base ‘i’ is the common property of the Indo-European and the Dra- by pre- vidian families; but, though ‘iga’ seems to be as closely connected with ¢iha,’ as the Latin ‘ego’ is with the Sanscrit ‘ah-am, yet the ylich Dravidian formative ‘gu,” ‘ ngu,” ‘ku’ (by suffixing which the demon- slﬂg the strative vowel becomes an adverbial noun) has no connection whatever with the merely euphonic ‘h’ of ¢ iha.’ | forms The Canarese adverbs of place ‘alli,) #here, and “illi, kere, are A s evidently derived from the High Tamil “il,) a house, a place, which apP“‘.iu; has been converted, by euphonic displacement, into 1i. Notwith- o Al v standing this, ‘il,’ @ house, is not contained in an isolated shape in Canarese. i in The formative ¢ du,’ euphonised to ‘ndu, which is found in ‘andu,’ &erstoody and ‘indu,” is one which is occasionally used in the formation of ouns of ordinary nouns, though not so frequently as ‘gu. (See the section gith O on ‘ Roots.”) It may have an apparent resemblance to the formative erhg are ‘d’ or “th’ of several Sanscrit adverbs, e.g., ‘athd,’ so, thus; ‘tada,’ then, at that time; but the connections and relationship of the Sanserit ps' 02 formative are widely different from those of the Dravidian one. Z ) 338 THE PRONOTUN. el In the High Tamil adverbs, ‘4 ndu’ and ‘indu’ (‘47 and ‘1, for ‘a and ‘i’), the cerebral ‘d’ is sufﬁxed as a formative instead of the i has dental ; and it is nasalised accordingly by ¢mn,” instead of ‘n; but I have no doubt that this ¢ du is from the same source, and is used for precisely the same purpose, as the dental suffix. In Telugu in a similar lan attu fha manner, that, an adjective and adverb, which is properly a demonstrative noun, exhibits the cerebral ¢t instead of the more “h usual dental d’ or nd. D andu,’ in Tamil, means not only thern and there, but also a year This word has some apparent resemblance to words which signify a sty year in various other languages; e.g., Ossete € ans Latin ¢an-nus Tur but the resemblance blar Mongol ‘on,” Manchu ¢ ania,” Tungusian ‘anyan disappears on investigation, for the Tamil <andu’ (Tel éndu’) is a modern corruption from ©yAndu,” the word which is used for year in app The the poetry and in all ancient inseriptions; and €yandu’ is derived from the interrocative base ya’ and the formative ¢du,” nasalised to thor ‘ndu’ (like ‘iradu,” fwo, nasalised to irandu’), and means properly dise when, secondarily o time, and lastly « year vidi 3! Afiliation of the demonstrative bases xtra-Dravidian affinities.— of ff There is only a partial and indistinet resemblance between the remote ¢ > > a, proximate ‘I, and medial €u, which constitute the bases of the gene Drividian demonstratives, and the demonstratives which are used by alve the languages of Northern India S0ng 8 A In Bengali and Singhalese, is used as a demonstrative; in me Marathi € ha,” ¢ hi ‘hen in the Hindustani we find ¢ vuh,” that yih,’ 0 this; but in the oblique cases the resemblance increases; e.g., s-k Mang to ths. ‘1 is used as the proximate demonstra,tlve in the North- Indian langua es more systematically than “a’ or any corresponding Tiveg vowel is used as the remote; eg., Marathi “ikade,” here; Hindi fion, idhar hather Mar itake,” so much 1ITepy A general resemblance to the Dravidian demonstrative bases is Inaf apparent in several of the Hlma]ayan languages; e.g., Bodo ‘imbe,’ this, < hobe, that ; Dhimal ¢ (u, Uraon édah,’ “hiidah.’ The h and 1 Dot Rajmahal are porfectly identical with the DraAvidian demonstratives, and form another evidence of the Dravidian character tself The\( of a portion of that idiom of my The connexion which appears to subsist between the Dravidian A medial demonstrative ‘u’ and the ¢ u’ of the Uraon cmd Dhimal is Pl Tbe | deselvmg of notice. Perhaps the Drividian medial ‘u’ (Dhimal ‘&, Whig Urdon ¢hidah ’) may be compared with the Old Hebrew masculine- feminine pronoun of the third person, ¢hti;’ and thus with the Old dy Pare DEMONSTRATIVES. 339 ) Persian remote demonstrative ‘hauva,” of which the first portion he appears to be ‘hu,” and the second ‘ava,—which ‘ava’ forms the Al base of the oblique cases. It may also be compared with the ‘u’ or for ‘0’ which forms the remote demonstrative in some of the Scythian lar languages: e.g., Finnish ¢tuo,’ tha?, ‘tama,’ this; Ostiak ¢toma,’ that, ‘tema, this. Compare also the Hind. ‘vuh,” #hat; Bodo 018 ‘ hobe.’ The Magyar demonstratives are mere in accordance with the (i Dravidian ‘o and £is eun, % thaw, ez b thas. The demon- ny stratives of the other languages of the Scythian family (e.g., the U, Tuarkish ‘bou,” that, ol, this’) are altogether destitute of resem- nee blance. When we turn to the languages of the Indo-European family, they pu is'a, appear in this particular to be closely allied to the Dravidian. ved Throughout that family both ‘a’ and ¢i’ are used as demonstratives ; fo though not to so large an extent, nor with so perfect and constant a 1ly discrimination between the remote and the proximate, as in the Dra- vidian family. In Sanserit ‘a’ is used iunstead of the more regular H £y in most of the oblique cases of ¢idam, this; and the correlative ) [4 of this word, ‘adas,’ means not only tkat, but also this. 9 Nevertheless, ot a is more generally a remote than a proximate, and ‘i’ more the generally a proximate than a remote demonstrative. In derived \ by adverbial words ‘i’ has always a proximate force; but ¢ta,” the con- sonantal demonstrative, is more generally used than “a.” The follow- n ing are examples of each vowel:—¢i-ha,” %ere; ¢i-dinim,” now; ”h)’ ¢ ta-danim,’ then: also ‘i-ti,” so, this much, ‘a-tha,” so, thus, in that manner. ,k('),' ith We may also compare the Old Persian ‘avada,’ thither, in that ling direction ; and the corresponding proximate ¢i-da, hither, in this direc- ndi fion. The resemblance between these forms, notwithstanding the irregularity of their application, and the Dravidian remote and prox- sis imate demonstrative bases, amounts to identity. All irregularity disappears in the New Persian, which in this lbe; The point accords as perfectly with the Dravidian languages as if it were itself a Dravidian idiom. fian Its demonstratives are ‘4n,’ that, in, this. These demonstratives are adjectival prefixes, and naturally destitute oter of number; but when plural terminations are suffixed, they acquire a [ian plural signification ; e.g., © &nan,’” those (persons), “inan,’” these (persons). The same demonstratives are largely used in the modern Turkish, by ll 18 ( ﬁ” which they have been borrowed from the Persian. ‘an’ and ‘in’ are 106 undoubtedly Aryan demonstratives. This is apparent when we com- pare ‘an’ with the Zend ‘aém, that; and that again with the old 7D 340 THE PRONOUN. Sanscrit ‘ayam; but ‘in’ is still more clearly identical with the Zend ¢1im, this. The same ‘im’ constitutes the accusative in Vedic | Sanscrit (and is also identical with ‘iyam,’ the masculine-feminine singular of the Old Persian, and the feminine of the Sanscrit); but in Zend <im’ is the nominative, not the accusative, and it is to this form that the New Persian is most closely allied. The demonstrative base ‘i’ (without being restricted, however, to a proximate signification) appears in the Latin “is’ and ‘id, and in the Gothic ©is; and the Dravidian and New Persian distinction between the signification of “a’ and that of ¢1i,” has been re-developed in our English that and this. Whilst the New Persian ‘4n’ and ‘in’ are closely connected with Sanscrit and Zend demonstratives, it does not follow that they are directly derived from either the one tongue or the other. On the contrary, the exactness with which the Persian discriminates between i the remote and the proximate, leads me to conclude that it has retained more faithfully than either of those languages the primitive charac- teristics of the Pre-Sanscritic speech. If so, instead of supposing the Dravidian dialects to have borrowed their demonstratives, which are % still purer than the Persian, from the Sanscrit (which are irregular and greatly corrupted), it is more reasonable to suppose that the Dravidian demonstrative vowels retain and exhibit the primeval bases from which the primary demonstratives of the Sanscrit and of all other Indo-European tongues have been derived. Emphatic ‘6’—1t has been seen that in Ku ‘¢’ is used as a demon- strative ; e.g., < évaru’ (‘é-(v)-ar’), they,; and this may be compared with the demonstrative €&’ of the Sanscrit ¢état, this (neuter), and the corresponding Zend ¢aétat.’ In the other Drividian dialeets, however, &’ is not used as a demonstrative, but is post-fixed to words for the purpose of rendering them emphatic. The manner in which ‘&’ is annexed, and the different shades of emphasis which it commu- nicates, are precisely the same in the various dialects, and will be sufficiently illustrated by the following examples from Tamil. When ‘e is post-fixed to the subject of a proposition, it sets it forth as the sole depositary of the quality predicated; e.g., ‘kalvi-(y)-& &elvam,’ learning (alone s) wealth; when post-fixed to the predicate, it intensifies its signification; e.g., ‘kalvi Selvam-&, learning is wealth (indeed): when post-fixed to a verb or verbal, it is equivalent to the addition of the adverb truly, certainly; eg., ‘alla-(v)-&’ (certainly) not In the colloquial dialect, it has often been annexed to the case-terminations of nouns without necessity, so that it has sometimes DEMONSTRATIVES. 341 he become in that connexion, a mere expletive; in consequence of dic Ine which, in such instances, when emphasis is really required by a sign of case, the &’ has to be doubled; e.g., ¢ ennéléyé’ (‘ennil-é-(y)-¢’), Juis through me (alone). s The same sign of emphasis forms the most common vocative case- sign in the various Dravidian dialects, the vocative being nothing {0 more than an emphatic enunciation of the nominative. Compare with n this the use of the nominative with the addition of the definite article 100 as the vocative in Hebrew and in Attic Greek. The Persian ¢é of ped supplication " may also be compared with it. Some resemblance to the use of ‘&’ as a particle of emphasis may yith be discovered in the Hebrew ¢/%e paragogic ’ (pronounced ‘eh’ or “ah,’ are the according to the connexion), which is supposed to intensify the signi- fication of the words to which it is annexed. The ¢he directive’ of een the same language is also, and not without reason, supposed to be a ned mark of emphasis; and its origin is possibly connected with that of 18 the definite article. A still closer resemblance to the emphatic ‘€’ of the the Dravidian languages is apparent in Chaldee, in which ¢4’ suffixed are to nouns constitutes their ‘emphatic state, and is equivalent to the llar <A definite article of many other languages. The Persian € of particu- the larity,” the & of ascription of greatness,” &e., in addition to the ¢& of 2568 supplication,” which has already been referred to, probably spring from ther a Chaldaic and Cuthite origin; though each of them bears a remark- able resemblance to the Dravidian emphatic <&.’ non Honorific demonstrative pronouns.—1 have deferred till now the qred consideration of a peculiar class of honorific demonstratives, which are and flound only in Telugu and Canarese, and in which, I think, direct oS, Aryan influences may be detected. onds (1.) In all the Dravidian dialects, the plural is used as an honorific hich singular when the highest degree of respect is meant to be expressed ; mi- but when a somewhat inferior degree of respect is intended, the pro- | be nouns which are used by the Telugu are ‘ayana,” ke, €ulle, and ¢ dme, Then she, “illa,;” with their corresponding proximates ‘iyana,’” ‘%ic,’ and h a8 ¢tme,” ¢ hec. These pronouns are destitute of plurals. a1, It can scarcely be doubted that an Aryan origin is to be attributed b to these words; and this supposition would account for the circum- ylth stance that they are found in the Telugu only, and not in any other ) the dialect of the family (except the Tulu ‘&ye,” ke, is to be regarded ully) as a connected form): it would also harmonise with their use as the honorifies. nies Compare ‘ayana’ with the Sanserit masculine ‘ayam, ‘ulle,” and 342 THE PRONOUN. “iyana’ with the Sanserit feminine, and the Old Persian masculine- feminine ‘iyam,” ‘hic, ¢ heac. §f B ¢ame,” “illa,” and ‘Ime,’ ‘hee,’ the corresponding feminine pronouns of the Telugu, may be compared not only with the plurals of the Sanserit Jil I pronoun of the third person (‘ime,” mas., ‘imé&h,’ fem., ¢ imani,” neut.), A but also with ‘amum’ and ‘imam,’ hum, which are accusative singulars, and from which it is evident that the ‘m’ of the plural forms is not a : sign of plurality, but is either a part of the pronominal base, or an euphonic addition. Bopp considers it to be the former, but Dravidian fo 0 analogies incline me to adopt the latter view; for it can scarcely be doubted that ‘Ame’ and ‘ime’ are identical with ‘Ave’ and ‘ive, (( another pair of Telugu pronouns (used when a little less respect is T intended); and the ‘v’ of these forms seems to be the ordinary euphonic ‘v’ of the Dravidian languages (which sometimes alternates to with ‘m’), and not part of the base. i (2.) When a little less respect is meant to be shown than is fo implied in the use of ‘4yana’ and ‘iyana,” and of ‘ame’ and ‘ime,’ the Telugu makes use of ¢ atadu,” “elle,” © ave,” ‘¢lla,” with their corresponding 9 proximates ‘itadu’ and ¢ ive Here the Canarese substantially agrees with the Telugu; e.g., ¢ atanu,’” “elle, ‘itanu,” ¢Aic’(Ancient Can. ‘atam,’ ‘itam ). The final ‘nu’ of the Canarese is the formative of the a8 masculine singular, corresponding to the Telugu ‘du;’ and the demon- ] th stratives ‘a’ and ‘i’ are often lengthened (as has already been shown) in Canarese. The Canarese feminines ¢ dke,” ‘illa,” ‘ike, ‘hec,’ do not ¢ A 1 appear so perfectly to accord with the Telugu ‘ave, ve; the only D real difference, however, is that the Telugu ‘v is purely euphonic, st whilst the Canarese ‘k’ is a formative, the use of which constitutes ¢ake’ and ¢ ike’ abstract pronominal nouns. ‘ti\' n Both the above sets of Telugu pronouns are destitute of plurals, [4) but both are pluralised in Canarese; e.g., ¢ tagalu,” ¢itagalu,’ those and these (men); ¢ akeyar, ikeyar,’ those and these (women). I do not think that the Tuda ‘adam,” ke, ske, i, is allied to any of the pronouns now referred to. I consider it to be a neuter singular r which is synonymous with ‘adu,’ the neuter singular of the Tamil- Canarese, and which is used corruptly for the masculine and feminine, Wl as well as for the neuter. The use of ‘adam’ as a neuter singular, an instead of ‘adu,’ exactly corresponds to the"use of ‘adan’ instead of il ¢adu’ in High Tamil. When the Telugu masculine of respect ‘ata-du,’” ‘ita-du,” and the corresponding Canarese honorifics ¢ Ata-nu,” “ita-nu,’ are scrutinised, it 4p he is evident that in addition to the vocalic demonstrative bases, ‘a’ and fht ‘i)’ which are found in Dravidian demonstratives of every kind, the ‘ta’ the INTERROGATIVES. 343 which is subjoined to ‘a’ and ‘i’ possesses also somewhat of a demon- strative or pronominal signification. It cannot be regarded like ‘v’ 1§ as merely euphonic; and its restriction to masculines shows that it is it not merely an abstract formative, like the ¢k’ of the feminine ¢ ake. ) It can scarcely be doubted, I think, that the origin of this ‘ta’ is Aryan; for we find in all the Aryan languages much use made of a similar ‘ta,’ both as an independent demonstrative, and as an auxiliary il to the vocalic demonstrative. ‘ta-d,” Sans. that, is an instance of the Al former; whilst the secondary or auxiliary place which ‘ta’ or ‘da’ be occupies in the Sanscrit ¢ étad’ (‘ é-ta-d’), this, and ‘adam, ‘adas,’ 6, (‘a-da-m,” ‘a-da-s’), this, or that, is in perfect agreement with the 18 Telugu and Canarese © 4-ta-nu,” ¢ a-ta-du.’ TS The final ‘e’ of “éave, 262 ive, ame, ‘ime,” ‘ake,’ ‘ike, is equivalent "y £8 to the Tamil ¢ el ‘e’ or ‘ei’ is an ordinary termination of abstracts in these languages, and a suitable one, according to Dravidian notions, 1§ for feminine honorific pronouns. he ne ITI.—Interrogatives. 03 N There are two classes of interrogatives in the Dravidian languages, he as in all others, viz., interrogative pronouns or adjectives such as, i who? which? what? and syntactic interrogatives, such as, is @¢? s there? m) 106 Interrogative pronmouns and adjectives resolve themselves in the ly Dravidian tongues into interrogative prefixes, resembling the demon- strative prefixes already considered, by suffixing to which the forma- 6, The tes tives of number and gender we form interrogative pronouns. interrogative particle itself, when simply prefixed to a substantive, constitutes the interrogative adjective what ? Js, nd ‘e ,(1.) The first and simplest interrogative prefix is the vowel ; In all the DrAvidian dialects this prefix is used in the formation of oy Jar pronominals, in precisely the same manner as the demonstrative bases 59 ¢a’ and ‘i It forms one of a set of three vocalic prefixes (‘a,” ¢ 1 (e’)’ ik which occupy one and the same position, obey one and the same law, Ny ay and differ only in the particular signification which is expressed by each. | of The unity of principle pervading these prefixes will be clearly apparent from the subjoined comparative view. The forms which are the here exhibited are those of the Tamil alone; but in this particular all B d the dialects agree so perfectly with the Tamil, and with one another, that it is unnecessary to multiply examples. 1} 344 THE PRONOUN. for PROXIMATE REemorE ha DEMONSTRATIVE DEMONSTRATIVE INTERROGATIVE ¢ . ‘1 ’ “n b Pr 1ts Masculine sing avan, ille evan, quis ? 10 ivan, hg‘c st Feminine do ival, hec aval, illa eval, que? of Neuter do idu, hoc adu, lud edu, quid ? Epicene plural ivar, hi, hee 10 avar, i, ille evar, qui? que? t0 Néuter do. ivei, hec avei, #lla evei, que ? i 1¢ The interrogative vowel ‘e’ forms the basis also of various inter- tqz rogative adverbs or adverbial nouns, in the same manner as the cor- m responding demonstrative adverbs or adverbial mouns are formed to from ‘a’ and ‘i7" eg., compare ‘engu,” where? with ‘angu, theres 1 and ‘ingu,’ here; ‘endru, which day? with “andru,” that day, and th ‘indra,” to-day. A few adverbs of this kind are peculiar to the T Canarese ; e.g., ‘entha,” of which kind? ‘ettaln,” on which side ? elli, T wn which place? where? The Tamil has also a regularly deeclined 1 interrogative noun ‘ennam,” what ? which is derived from ‘en’ (com- (0 monly lengthened to “én’), why ? 1l ‘e’ is also prefixed adjectivally to substantives, in the same man- P ner as ‘a’ and ‘i; and (as was observed in the case of those pre- fixes) it doubles the initial consonant of the substantive to which it is prefixed : c.g., compare ‘evvari’ (from ‘vari,” way), which way ? ) with €avvari, that way, and ‘ivvari, this way. In addition to this adjectival prefix, a regular interrogative adjective (‘enda,” what ? or whick ?) is formed by the Tamil, to correspond with the demon- ;: strative adjectives, ‘anda,” that, ‘inda,’ this, which are peculiar to that dialect. It is chiefly, however, in the colloquial dialect that | these forms are used. The higher and more ancient dialect prefers € the simple vowels ‘a,’” ¢ e I, e, to express ¢that, this, and which ? I need not call attention to the beautiful and philosophical regularity of this triple set of remote and proximate demonstratives ¢ and interrogatives. In no other language or family of languages in the world shall we find its equal or even its second. In addition to which, the eircumstance that the demonstrative vowels are not only used in these languages with an invariable and exact discrimination of meaning which is not found in the Indo—European tongues (with the solitary exception of the New Persian), but are also associated 1 with a corresponding interrogative vowel of which the Indo-European INTERROGATIVES. 345 tongues are totally ignorant, tends to confirm the supposition which I have already expressed, that the Dravidian family has retained some Pre-Sanscrit elements of immense antiquity; and, in particular, that its demonstratives, instead of being borrowed from the Sanscrit, represent those old Japhetic bases from which the primary demon- stratives of the Sanscrit itself, as well as of various other members of the Indo-European family were derived. The only peculiarity which requires notice in the use of this inter- rogative prefix, is the circumstance that it is occasionally lengthened ¢ A ) to ‘&, precisely as ‘a’ and ‘i’ are lengthened to a and ““1 In Tamil this euphonic lengthening is very rare. It is found only in the neuter-singular interrogative pronoun ‘edu,” what or which (thing)? te: ‘quid 2 which sometimes, especially in composition, becomes ‘édu ;’ and COr in the interrogative adverb ¢en,” why ? which is ordinarily lengthened ned In Telugu this increase of quantity is more common. It to ¢én.’ o1y appears not only in ‘émi’ and éla,” why ? but is ordinarily used as and the interrogative prefix, where the Tamil invariably has short ‘e.’ the Thus, whilst the Tamil has ¢evvidam, whkat manner? how? the Il Telugu says ‘évidham’ instead. So also, whilst the Tamil occa- ned sionally only uses ‘édu,” ‘quid 2’ instead of the more classical ‘edu,’ the corresponding interrogative of the Telugu is invariably ¢édi,” and its Ol 7] plural ¢évi On the other hand, the Telugu masculine interrogative Il pronoun ‘evvadu,’ ‘quis?’ preserves the same quantity as the Tamil ‘evan; and even when the prefix is used adjectivally, it is sometimes {7 t1s ‘e’ (not &’) as in Tamil, e.g., “eppudu’ (not ¢épudu’), what time ? when ? ! We have already scen that the base vowels of the Telugu demon- this G A gf? strative pronouns ‘vadu,” ‘ille, Vi du,” ke, have shifted their position, it the forms required by rule instead of ‘vadu’ and ‘vidu’ being ‘avadu’ and ‘ivadu:’ but as the interrogative pronoun is less used to hat than the demonstrative, and therefore less liable to corruption, it has 618 escaped this euphonic displacement of the base vowel; and in con- sequence we have not ‘védu,’” ‘quis?’ in accordance with ‘vidu’ and ‘vidu,” as we might have expected, but ‘evvadu’ (for ‘evadu’), in al close accordance with the Tamil ¢evan.’ yes 9 In each of the Dravidian languages, ‘e is commonly pronounced 1 [4 as ‘ye, and ‘€’ as ‘yé’ and in Telugu this X is often written as t0 well as heard; but as this is a characteristic of all words in each ply of the dialects which begin with this vowel, and not of the inter- 100 rogatives alone, it is evident that it is merely a peculiarity of pro- ith fol nunciation, and that the ¢y’ in question is no part of the interrogative base il 346 THE PRONOUN. thi Compare with this Dravidian interrogative the interrogative ‘é’ of the Hebrew; e.g., ‘épd, where? (compounded of ‘&’ and ‘po, il “u here) ; and also ‘éka,’” where ? or how ? Under the head of the Syntactic interrogative [ a, a resemblance bas (30 ‘ki which subsists between this interrogative ‘e’ or ‘&’ and the e of emphasis will be inquired into: and in that particular also it will be found that the Hebrew usage agrees with the Dravidian. a8 ¢ San (2.) The other interrogative of this class is ‘ya.’ ‘yA’ is not used at all in Telugu; but it is largely used in ] Sar Canarese and somewhat more rarely in Tamil. In High Tamil ‘ya’ Fin is not only prefixed adjectivally to substantives (like ‘a,’ ‘e, and ‘e’), e.g., ‘ya-(k)kalam, what time; but it is even used by itself as a ¥ pronoun; e.g., < ya-(8)Seyday, what hast thou done? It forms the basis of only one adverbial noun, viz., ‘yandu,’ Tam., when ? a year, a Tels correlative of ‘andu,’ then, and ‘indu,’ now. The only use to which the ‘ya’ is put in the colloquial dialect of the Tamil, is that of forming all the basis of interrogative pronouns; a complete set of which, in Je wh Tamil as well as in Canarese, are formed from ‘ya; eg., ‘yavan, ‘quis 2 ‘yaval, ‘que? ‘yadu, ‘quid ? ‘yavar, ‘qui? ‘quee ? ¢ yavei, (0 ‘quee ?’ The Canarese interrogative pronouns accord with these, with Tels a single unimportant exception. The neuters, singular and plural, 10 of the Canarese are formed from ¢yéva, instead of ‘ya; eg., but al ¢ yavadu,’ ‘quid 2’ (for yadu’), and ‘yavavu,” ‘que ?’ (for ‘yava’). This additional ‘va is evidently *derived by imitation from the euphonic ‘v’ of ‘yavanu, /e, and its related forms; but it is ount of place in connexion with the neuter, and is to be regarded as a corruption. fly In Tamil a peculiar usage with respect to the application of the the epicene plural ¢ yavar, ‘qui?’ ‘que ?” has obtained ground. It 1s a0 largely used in the colloquial dialect with the signification of the the singular, as well as that of the plural, though itself a plural only and bl without distinction of gender; and when thus used, ¢ yavar’ is abbre- the viated into ‘yar; eg., ‘avan yar, who is he? (literally ke who ?); 10g <aval y?ir,’ who is she? “yar’ has also been still further corrupted Wi into ar,” especially in compounds. thy The Gond interrogatives ‘ba’ and ‘bo appear to have been K/ hardened from ‘ya. tay C0) Extra- Dravidian relationship. — There is no analogy between la either ‘e’ or ‘y&’ and any of the interrogative bases of the Indo- 01 European family. Both in that fa-mily and in the Scythian group the U} ordinary base of the interrogative is the guttural ‘k; e.g., Sanserit, INTERROGATIVES. 347 “kim,” what ? The same base appears in the Sanscrit interrogative POy initial syllables ¢ka-) ‘ki-’,’ ‘ku-,” which correspond to the Latin ‘qu-,’ the Gothic ‘hva-,” and the English ¢ wh-. We find the same 1ce base again in the Turkish ‘kim ’ or ¢ kim,” who ? what? in the Magyar of ‘ki, who ? plural ‘kik; and in the Finnish ‘kuka’ (root ‘ku’). e In the absence of a real relative pronoun, the interrogative is used The base of the as a relative in many of the Scythian languages. Sanserit relative pronoun ‘ya’ (‘yas,’ ‘ya ‘yat’), bears a close n apparent resemblance to the Dravidian interrogative ¢ya. The A ya Sanscrit ‘ya,” however, like the derived North-Indian ¢j6, and the ;nd Finnish ¢yo’ is exclusively used as a relative, whereas the Dravidian S 4 ‘ya’ is exclusively and distinctively an interrogative. IS It has been conjectured that the Sanscrit ¢ya,” thongh now a & relative, was a demonstrative originally; and if (as we shall see that I there is some reason for supposing) the Dravidian interrogatives ‘e’ and ‘a’ were originally demonstratives, it may be supposed that il?g ‘ya’ was also a demonstrative, though of this no direct evidence 1n i, whatever now remains. If ‘ya’ were originally a demonstrative, the o, connexion which would then appear to exist between it and the Sansecrit relative would require to be removed a step further back; for it is ith not in Sanscrit that the relative ¢ ya’ has the force of a demonstrative, 1, s but in other and more distant tongues, viz., in the Lithuanian ¢ yis,” %e; and in the Slavonian ¢ yam, and the Zend ¢yim," Aim. ) he it 9 Syntactic tnterrogatives, ‘4’ and ¢6.’—The interrogative prefixes ‘e and ‘ya, are equivalent to the interrogative pronouns and adjec- 5 & tives, who ? which ? what ? &e. Another interrogative is required for he the purpose of putting such inquiries as are expressed in English by a change of construction ; e.g., us there ? is it ¢ by transposition from 1§ he there s, it 1s. This species of interrogation is effected in all the Dra- N a to nd vidian languages in one and the same manner, viz., by suffixing the noun, verb, or sentence which forms the principal subject of inter- (% rogation ; and in these languages it is by the suffix of ‘4’ alone, ?)’ ol without any syntactic change, or change in the collocation of words, that an interrogative verb or sentence differs from an affirmative one ; e.g., compare the affirmative ‘avan tandan,” Tam., Ze gave, with ‘avan ol tandan-a ¥° did he give ? and ‘avan-atandan ¥’ was it ke that gave ? compare also adu Gr,” that is ¢ village, with ¢ adu Ur-a ¥ s that a vil- lage ? This interrogative is never prefixed to nouns or pronominals, e or used adjectivally ; but is invariably post-fixed, ]ikg an enunciated or audible note of interrogation. o €6’ is not unfrequently used like ‘a’ as a simple interrogative ; ”) 348 THE PRONOUN. but its special and distinctive use is as a particle expressiv.e of doubt. Thus, whilst “ avan-A’ means 4s it he ? ‘avan-6’ means can i be he? or poi CA? ton 1 am doubtful whether it is he or not. (0] is post-fixed to words in pre- I cisely the same manner as ‘4, and is probably only a weakened form (1“ of it, in which, by usage, the interrogation has become merged in the 00! expression of doubt. It has acquired, however, as a suffix of doubt a the position and force of its own, quite independent of ‘4; in consequence the of which it is often annexed even to interrogative pronouns; e.g., ‘evan-6,” Tam., I wonder who he can be ; ¢ ennam-6,” what it may be I L know not—compound forms which are not double interrogatives, but which consist of a question evan,” who 2 or ‘ennam, what ? and an answer ‘0, I am doubtful.—I know not.—There is room for further pr thy mquiry. of ‘ko,” which seems to correspond in meaning to <6, is used as a prefix for of uncertainty by some Telugu pronominals: e.g., compare ¢anta,’ fro thus much, ¢ inta,’ this much, ¢enta,’ how much ? with ¢ konta,” some, or, il as if one should say, 7 know not how much. thy This prefix is rare in Telugu, unknown in Tamil, and is possibly derived from the Sanscrit interrogative. als to The use of 4’ as an interrogative suffix has not been derived from il any language either of the Scythian or of the Indo-European family. (4 It is altogether unknown to the Sanscrit; and the Cashmirian is the only Non-Drividian tongue in which it is found. 9 he I am inclined to consider ‘4, the Dravidian interrogative, as derived from, orat least as allied to, ‘a’ or ‘4,” the remote demonstrative (¢ of the same family. The quantity of that demonstrative ‘a’ is long 86 or short as euphonic considerations may determine; and though the ALY a il interrogative is always long, yet in consequence of its being used 6X as a post-fix, it is pronounced long by necessity of position, whatever it may have been originally. Hence the question of quantity may, in this inquiry, be left altogether out of account. The only real difference between them is the difference in location; ¢a’ demonstrative being invariably placed at the beginning of a word, a’ interrogative at the end of it. If the interrogative < a’ were really connected with ‘a,” the demonstrative, we should expect to find a similar connection subsisting between ‘e’ or ‘&’ the adjectival interrogative, and some demonstra- tive particle, with a similar interchange of places; accordingly this is found to be the case, for ‘&’ is not only the ordinary sign of emphasis in all the Dravidian tongues, but it is used in Ku as an adjectival demonstrative ; and it is curious that in this instance also, there is a change of location, ¢ &’ emphatic being placed at the end of a word, ‘e’ interrogative at the beginning. DISTRIBUTES. 349 oubt, A similar change in the position of particles, to denote or corres- el or pond with some change in signification, is not unknown in other Thus in Danish, the article en’ has a definite sense in one pre tongues. position and an indefinite in another; e.g., ‘en konge,’ a king, form | the ‘kongen,’ the king. But it is still more remarkable, and more corro- bt & porative of the supposition now advanced, that in Hebrew, one and 1ence the same particle, ‘he’ (for it is one and the same, and any difference that exists is merely euphonic), imparts emphasis to a word when -y post-fixed to it, and constitutes an interrogative when prefixed. el , but Distributive pronouns.—In all the Drévidiban tongues distributive d an pronouns are formed by simply annexing the conjunctive particle to e the interrogative pronoun. Thus, from ‘evan,’ who ? by the addition refix of ‘um, and, the conjunctive or copulative particle of the Tamil is formed, viz., ‘evanum,’ everyone, whosoever (literally who ?-and),; and it from ‘epporudu,” when ? is formed in the same manner ‘epporudum,’ ¢, O, always, literally when ? and-. In Canarese similar forms are found, though not so largely used as in Tamil ; e.g., ‘yavagall’ (‘ya-agal-ii’), sibly always ; and in Telugu ‘nu’ (the copulative particle which answers to the Tamil ‘um’ and the Canarese ‘i’) is used in the same manner from in the formation of distributives; e.g., ‘evvadunu’ (‘evvadu-nu’), uily. everyone, ¢ eppudunnu’ (‘ eppudu-(n)-nu’), always. s the The Dréavidian languages have no pronouns properly so called besides those which have now been examined. as ) Instead of relative promouns, they use verbal forms which are ative called by English grammarians relative participles; which see in the Jong section on ¢ The Verb.’ All other words which correspond either in ) the meaning or in use to the pronouuns of other languages will be found on psed examination to be nouns, regularly formed and declined. Eev'er v, 11 .:'ence eing : the ’ the ting st 18 18 9818 vl 158 [ i i 10 Tl il frs all SECTION VL qu i S he THE VERB. of 18 Toe object in view in this section is to investigate the nature, affections, and relations of the DrAvidian verb. I commence with D of some general preliminary remarks upon its structure. (1.) A large proportion of Dravidian roots are used indis- C0 an criminately, either as verbs or as nouns. Jef When case-signs are attached to a root, or when, without the addition of case-signs, it is used as the nominative of a verb, it is 8l an regarded as a noun: the same root becomes a verb without any internal change or formative addition, when the signs of tense (or th 0 time) and the pronouns or their terminal fragments are suffixed to 1t of Though, abstractedly speaking, every Dravidian root is capable of this two-fold use, it depends upon circumstances whether any particular root is actually thus used; and it often happens, as in VI other languages, that of three given roots one shall be used solely or So generally as a verbal theme, another solely or generally as the theme ol of a noun, and the third alone shall be nsed indiscriminately either on as a noun or as a verb. l Herein also the usus loquend: of the various dialeéts is found to th differ ; and not unfrequently a root which is used solely as a verbal i) theme in one dialect, is used solely as a noun in another. (o (2.) The inflexional theme of a Drividian verb or noun is not 1 always identical with the crude root or ultimate base. In many instances formative or euphonic particles (such as ‘vu,” ‘ku,” ‘gu’ or th ‘ngu, ‘du’ or ‘ndu,” ‘bu’ or ‘mbu’) are annexed to the root,—not al added on like isolated post-positions, but so annexed as to be incor- L porated with it. (See the section on ‘Roots.’) But the addition of €3 one of those formative suffixes does not necessarily constitute the ty root to which it is suffixed a verb: it is still capable of being used Pa as a noun, though it may be admitted that roots to which those of — | STRUCTURE. 351 suffixes have been annexed are more frequently used as verbs than as nouns. (8.) The structure of the Dravidian verb is strictly agglutinative. The particles which express the ideas of mood and tense, transition, intransition, causation, and negation, together with the pronominal fragments by which person, number, and gender are denoted, are annexed or agglutinated to the root in so regular a series aund by so quiet a process, that generally no change whatever, or at most only a slight euphonic change, is effected either in the root or in any of the suffixed particles. [See this illustrated in the section on ¢Roots.’] (4.) The second person singular of the imperative may perhaps be considered as an exception to the foregoing rule. The crude theme of the verb, or the shortest form which the root assumes, and which is capable of being used also as the theme of a noun, is used in the e, yith Drévidian languages, as in most others, as the second person singular of the imperative ; and the ideas of number and person and of the s conveyance of a command, which are included in that part of speech, are not expressed by the addition of any particles, but are generally left to be inferred from the context alone. Thus, in the Tamil, the {18 sentences ‘adi virundadu,’ the stroke fell; ‘ennei adi-ttan,’ ke struck me; any and ‘idei adi, strike thow this; the theme, ‘adi,” strike, or a stroke, is the same in each instance, and in the third illustration it is used with- (or ] to out any addition, and in its crude state, as the second person singular of the imperative. able Aty (5.) As the Drividian noun has but one declension, so the Dra- § 10 vidian verb has only one conjugation and but very few irregular forms. Some European grammarians have arranged the Dravidian verbs in yOl eme classes, and have styled those classes conjugations; but the differences on which this classification is founded, are generally of a trivial and ther superficial character. The structure of the verb, its signs of tense, and | to the mode in which the pronouns are suffixed, remain invariably the thal same, with such changes only as euphony appears to have dictated. Consequently, though class-differences exist, they are not of sufficient importance to constitute different conjugations. nof any Such is the simplicity of the structure of the Dravidian verb, that ’Or the only moods it has are the indicative, the infinitive, the imperative, not and the negative, and that it has only three tenses, the past, the present, and the aorist or indefinite future. The ideas which are 0T expressed in other families of languages by the subjunctive and opta- nOf the tive moods, are expressed in the Dravidian family by means of suffixed particles; and the imperfect, perfect, pluperfect, future-perfect, and uced other compound tenses, are expressed by means of auxiliary verbs. h0se 3 9 ~ 2 THE VERB. In these respects the Drividian verb imitates, though it does not f The modern PR equal, the simplicity of the ancient Seythian verb. pl Turkish has, it is true, an extraordinary number of moods—con- ditionals, potentials, reciprocals, inceptives, negatives, impossibles, &e., ¢ together with their passive, and also a large array of compound tenses; but this complexity of structure appears to be a refinement of a com- b paratively modern age, and is not in accordance with the genius of the e s SN T 0) Oriental Turkish, or Tartar properly so called. Remusat conjectures il that intercourse with nations of the Indo-European race, some time 8] after the Christian era, was the occasion of introducing into the Turkish language the use of auxiliary verbs and of compound tenses. ‘From CL the extremity of Asia, he says, ‘the art of conjugating verbs is un- known.’ The Oriental Turks first offer some traces of this; but the very sparing use which they make of it seems to attest the pre- [ existence of a more simple method.’ I All the Dravidian idioms conjugate their verbs, with the partial I exception of the modern Malayilam, which has retained the use of the 1 signs of tense, but has rejected the pronominal terminations. Never- ( theless, the system of conjugation on which the Dravidian idioms proceed, is one of primitive and remarkable simplicity. The Gond is the only Dravidian dialect which has adopted a complicated system ; and it has probably done so through the influence of its Kole or Himalayan neighbours. (6.) The Dravidian verb is more rarely compounded than the Indo-European one; and the compound of a verb with a preposition is especially rare. An inexhaustible variety of shades of meaning is § secured in Sanscrit and Greek by the facility with which, in those ¢ languages, verbs are compounded with prepositions; and the beauty of many of those compounds is as remarkable as the facility with which a they are made. In the Scythian tongues, properly so called, there is no trace of compounds of this kind; and though they are not unknown 1l in the Dravidian family, yet their use is not in harmony with the purer idiom; and when the component elements of such compounds are carefully scrutinised, it is found that the principle on which they { are compounded differs widely from that of Indo-Eurcpean compounds. 4 The Dravidian prepositions which are most frequently compounded with verbs are those which signify over and under, the use of which is 3] { illustrated by the common Tamil verbs ‘mér-kol,” to overcome, and kir-(p)padi, to obey. Drividian prepositions, however (or rather, post- bl positions), are properly nouns; e.g., ‘mél,” over, literally means over-ness, superiority; and ‘mél-kol’ (euphonically ‘mér-kol’), to overcome, literally signifies to take the superiority. These and similar verbal TRANSITIVES AND INTRANSITIVES. i;5t ) 108 themes, therefore, though compounds, are not compounds of a pre- lem position and a verb, but compounds of a noun and a verb; and the CON: Greek verbs with which they are to be compared, are not those which ke, commence with mepl, warda, ¢dvd, &ec., but such compounds as ses; [4 “ moh\wpréw,” to besiege a city, literally to city-besiege ; vavmwéw,’ to 01l In such cases, whether in Greek build a ship, literally to ship-build. the or in Tamil, the first member of the compound (that is, the noun) does ures not modify the signification of the second (that is, the verb), but ime simply denotes the object to which the action of the verb applies. It kish is merely a crude noun, which is used objectively without any signs of rom case, and is intimately combined with a governing verb. un Dravidian verbs acquire new shades of meaning, and an increase or the diminution in the intensity of their signification, not by prefixing or ple combining prepositions, but by means of auxiliary gerunds, or verbal participles, and infinitives—parts of speech which in this family of rtial languages have an adverbial force; e.g., ‘mundi (p)ponan,” Tam., ke the went before, literally having-got-before he went ; ¢SurRi’ (¢ Suttri) Vel (p)poméan,” he went round, literally rounding he went; ‘tara (k)kudit- Oms tan,’ ke leaped down, literally so-as-to-get-down he leaped. o Classification of Dravidian Verbs. ) O I. TRANSITIVES AND INTRANSITIVES. the Dréavidian grammarians divide all verbs into two classes which are 118 called in Tamil pira vinei’ and ‘tan vinei, ¢ramsitives and intran- These o 18 sitives, literally outward action-words and self action-words. 105 classes correspond rather to the ¢ parasmai-padam’ and ¢atmane- yot padam,’ or transitive and reflective voices, of the Sanscrit, than to the hich active and passive voices of the other Indo-European languages. g 18 The Dravidian ¢pira vinei’ and ‘tan vinei, or transitive and Wil intransitive verbs, differ from the ¢parasmai-padam’ and atmané- the padam ’ of the Sanscrit in this, that instead of each being conjugated nds differently, they are both conjugated in precisely the same mode. hey They differ, not in their mode of conjugation, but in the formative nds additions made to their themes. Moreover, all ¢ pira vinei,” or tran- 1d€d sitive verbs, are really, as well as formally, transitives, inasmuch as I ) 15 they necessarily govern the accusative, through the transition of their aﬂd action to some object; whilst the ¢ tan vinei,” or intransitive verbs, are 0st all necessarily, as well as formally, intransitives. The Dravidian 1685 transitives and intransitives exactly resemble in force and use the )il )iy determinate and indeterminate verbs of the Hungarian. The Hungarian lbll determinate verbs, like the Dravidian transitives, imply an object—an P4 o) A — 354 THE VERB. accusative expressed or implied, e.g., ‘szeretem,” I love (some person or thing); whilst the Hungarian indeterminate verbs, like the Dra- vidian intransitives, neither express nor imply an object, e.g., ‘szeretek,’ 1 love, t.e., T am in love. 8 al In a large number of instances in each of the Dravidian dialects, including entire classes of verbs, there is no difference between tran- b sitives and intransitives, either in formative additions to the theme, or fo i in any structural peculiarity, the only difference is that which consists in the signification. Thus in Tamil, all verbs of the class which take ‘1’ or ‘in’ as the sign of the past participle are conjugated alike, fi whether they are transitives or intransitives; e.g., from ¢ pann-u, trans., fo make, are formed the three tenses (first person singular) ({0 ¢ pannu-gir-éu,” I make, ¢pann-i-(n)-én,” I made, and ‘pannu-v-en,’ I will make: and in like manner from ¢ pés-u,’ intrans., to talk, are t formed, precisely in the same manner, the corresponding tenses ¢ pésu-gir-én,” I talk, pés-i-(n)-én,’ / talked, and ¢ pésu-v-én, I will 1 talk. th In a still larger number of cases, however, transitive verbs differ ‘t from intransitives, not only in signification and force, but also in grammatical form, notwithstanding that they are conjugated alike. ml The nature of the difference that exists and its rationale, are more clearly apparent in Tamil than in any other Dravidian dialect; my i illustrations will, therefore, chiefly be drawn from the Tamil. There are three modes in which intransitive Tamil verbs are con- verted into transitives. el (1.) Intransitive themes become transitive, by the hardening and doubling of the consonant of the appended formative; e.g., ¢ peru-gu, g to abound, by this process becomes ¢ peru-kku,’ #o increase (actively), to cause to abound. Transitives of this kind, which are formed from 1y intransitives in actual use, are often called causals, and they are as U well entitled to be called by that name as many causal verbs in the hay Indo-European tongues; but as there is a class of Dravidian verbs lay which are distinetively causal (and which are formed by the annexing Ty to the transitive theme of ‘vi, a causal particle; e.g., ¢ pannu-vi,’ ! to cause to make, from ¢ pannu,’ to make), it will contribute to perspi- of cuity to regard the whole of the verbs of which we are now treating, nf simply as transitives, and to reserve the name of causal verbs for the or double transitives in ¢ vi.’ 1y When transitives are formed from intransitives by doubling the consonant of the formative, it is in the theme or inflexional base itself t0j that the change takes place: there is no change in any of the signs of tense, or in the mode in which those signs are added; and the %13.1‘1 1ty TRANSITIVES AND INTRANSITIVES. 355 n hardened formative appears in the imperative, as well as in the other parts of the verb. i The nature of these formatives has already been investigated in the section on ‘Roots; and it has been shown that they are euphonic s, aceretions, which, though permanently annexed to the base, are not to I be confounded with it. I subjoin a few illustrations of this mode of or forming transitives by the doubling and hardening of the consonant of st the formative. \Le (i.) ‘gu,’ or its nasalised equivalent, ‘ngu, becomes ‘kkuj; e.g., ke, from ¢ po-gu,” to go (in the imperative softened into pd’), comes ) ‘po-kku,’ to drive away; from ‘ni-ngu,’ to quit, comes ‘ni-kku,” fo put a1) away. )n’ (ii.) “Su’ becomes ‘&éu’ (pronounced ‘chu’); e.g., from ‘adei-su,’ e to take refuge, comes °adei-chu,” to wnclose. €5 (iii.) ‘du,” euphonised into ‘ndu,” becomes ‘ttu; e.g., from °tiru- jill ndu,” fo become correct, comes tiru-ttu,” fo correct: in like manner the cerebral ‘ ndu’ becomes ‘ttu;’ e.g., from ‘ti-ndu,’ to fouch, comes for “ti-ttu,” to whet. 1 (iv.) “bu, euphonised into ‘mbu,” becomes ¢ ppu;’ e.g., from ¢ nira- {8 mbu,’ o be full, comes ‘nira-ppu,’ to fill. 016 When intransitives are converted into transitives in this manner ny in Telugu, ‘gu’ or ‘ngu’ becomes, not ‘kku’as in Tamil, but ¢ chu,’ —a difference which is in accordance with dialectic rules of sound. )} Thus from ¢ ti-gu,’ or ‘tl-ngu,” fo kang, to sleep, comes ‘ti-chu, or euphonically ti-nchu,” to weigh, to cause to hang. and The Telugu also occasionally changes the intransitive formative o ol ‘gu,” not into ¢ chu,” the equivalent of ‘kku,’” but into ‘pu;’ e.g., from o ‘méy, to graze, comes ‘mé-pu,’ fo feed: and as ‘ppu’ in Tamil is 0m invariably hardened from ‘bu’ or ‘mbu,’ the corresponding Telugu ) 85 ‘pu’ indicates that ‘bu’ originally alternated with ¢gu; for the the hardening of €gu’ into ‘pu’ is not in accordance with Drividian orbs laws of sound. This view is confirmed by the circumstances that in g Telugu the use of ‘pu’ instead of ‘chu’ (and of ‘mpu’ instead of vl ‘nchu’) is in most instances optional, and that in the higher dialect of the Tamil the formative ‘pp’ sometimes supersedes ‘kk; e.g., the "\Pl fiﬂgy infinitive of the verb fo walk, may in that dialect be either ‘nada-kka’ the or ‘nada-ppa.’ It is obvious, therefore, that all these formative ter- minations are mutual equivalents. the If the transitive or causal ‘p’ of such verbs as ‘nira-ppu, Tam., t ¢l to fill, < mé-pu,’ Tel., to feed, were not known to be derived from the hardening of an intransitive formative, we might be inclined to affiliate 50f the it with the ‘p,” which is characteristic of a certain class of causal verbs 9 A2 356 THE VERB. in Sanserit ; e.g., ¢jivh-p-ayami, I cause to live, ¢ jié-p-ayami, [ make al to know. Tt is evident, however, that the resemblance is merely acci- th dental, for etymologically there is nothing of a causal nature in the Dravidian formatives; it is not the formative itself, but the hardening tlli of the formative which conveys the force of tramsition; and on the ]ec other hand, the real sign of the causal in Sanscrit is ‘aya, and the “p’ which precedes it is considered to be only an euphonic fulerum. 1] It has already been shown (in the section on ‘ Roots’) that the th various verbal formatives now referred to are used also as formatives sh of nouns, and that when such nouns are used adjectivally the consonant Iy of the formative is doubled and hardened precisely as in the transitives €0 of verbs ; e.g., ©marattu,’ medicinal, from ¢marundu,’ medicine : Iy ‘ pAppu,’ serpentine, from ¢ pambu,’ a snake. When nouns are used to bes qualify other sounds, as well as in the use of transitive verbs, there [0l is a transition in the meaning of the theme to some other object ; and iy the idea of transition is expressed by the doubling and hardening of Th the consonant of the formative, or rather by the forcible and emphatie 10 enunciation of the verb which that hardening of the formative neces- sitates. § ? (2.) The second class of intransitive verbs become transitives by doubling and hardening the initial consonant of the signs of tense. ad Verbs of this class are generally destitute of formatives, properly so ‘i called ; or, if they have any, they are such as are incapable of change. an The sign of the present tense is in Tamil ¢gir’ ;’ that of the preterite 2 W oricav “d,” ordinarily euphonised into ‘nd ;" and that of the future ‘b ZOZ& These are the signs of tense which are used by intransitive verbs of low this class ; and it will be shown hereafter that they are the normal fbel tense-signs of the Dravidian verb. When verbs of this class become 1t] transitives, gir’’ is changed into ‘kkir; ¢d’ or ‘nd’ into ‘¢t 2R Pa and ‘b’ or ‘v’ into ‘pp.’ Thus, the root ¢ ser, to join, is capable i both of an intransitive sense, e.g., o join (@ society), and of a transitive wh sense, e g., to join (things that were separate). The tense-signs of the the intransitive remain in their natural condition; e.g., ‘sér-gir’-en, 7 aly Jjoin, ¢ §ér-nd-én,’ I joined, ‘&ér-v-én,” I will join: but when the signifi- 0y -eation is active or transitive, e.g., to join (planks), the corresponding Nm parts of the verb are ‘&ér-kkir-én,” I join, ¢Sér-tt-én,” 7 joined, ©&er- 1§ . pp-én, 1 wnll join. ace The rationale of this doubling of the case-sign is evident. It is an beg emphasized, hardened enunciation of the intransitive or natural form why of the verb ; and the forcible enunciation thus produced is symbolical begg of the force of transition by which the meaning of the transitive theme ety overflows and passes on to the object indicated by the accusative. 1§ § TRANSITIVES AND INTRANSITIVES. 3567 e In verbs of this class the imperative remains always unchanged ; Yol and it is the connexion alone that determines it to a transitive rather the than an intransitive signification. ing It should here be mentioned that a few intransitive verbs double the the initial consonant of the tense-sign, and that a few transitive verbs the leave the tense-sign in its original, unemphasized condition. Thus, “iru,’ fo sit, to be, is necessarily an intransitive verb ; nevertheless, in the the present tense ¢iru-kkir-én,” / am, and in the future ‘iru-pp-én,’ 1 [ves shall be, it has made use of the ordinary characteristics of the transi- Jang tive : so also ¢padu,’ to lie, though an intransitive, doubles the initial 1Ve8 consonant of all the tenses ; e.g., ¢ padu-kkir-én,’ 7 lie, ‘padu-tt-én,’ 7 ne: lay, ‘padu-pp-én, I shall lie. On the other hand, ‘i) fo give, to d to bestow, though necessarily transitive, uses the simple, unhardened, un- here emphatic case-signs which are ordinarily characteristic of the intransi- and tive ; e.g., ‘i-gir-én, I give, ‘i-nd-én,’ I gave, ‘i-v-én, I will gve. g of These instances are the result of dialectic rules of sound, and they are hatie not in reality exceptions to the method described above of distinguish- oS- ing transitive and intransitive verbs by means of the hardening or softening of the initial consonant of the case-signs, s by (3.) A third mode of converting intransitives into transitives is by This particle is adding a particle of transition to the theme or root. ¢du’ in Canarese, and ‘ttu’ (in composition ‘tu’ or ¢du’) in Tamil; lyso ange and may be regarded as a real transitive suffix, or sign of activity. terl te We have an instance of the use of this particle in the Can, ‘tal-du,’ o Py lower, from thl-u,’ to be low, and the corresponding Tam. ‘tar-ttu,” to lower, from ¢ tar’ or ‘tar-u,’ to be low. When the intransitive Tamil ! of theme ends in a vowel which is radical and cannot be elided, the tran- ormal :c()me sitive particle is invariably ‘ttu,’ e.g., ¢ padu-ttu,’ to lay down, from “tty ¢ padu,” o le. It might, therefore, be supposed that ¢ttu’ is the pri- pa.ble mitive shape of this particle ; but on examining those instances in which it is compounded with the final consonant of the intransitive sitve theme it appears to resolve itself, as in Canarese, into ‘du.” It is (,f the always thus compounded when the final cousonant of the theme is % o1l [ ]f’ﬂm or €17 ¢d’’ or ‘®’; and in such cases the ‘d’ of ‘du’ is not merely placed in juxta-position with the consonant to which it is attached, but udin is assimilated to it, or both consonants are euphonically changed, gel Thusély andi“dn’ according to the phonetic rules of the language. I\ ﬂ;n become ‘Rr-u’ (pronounced ‘ttr-u’), e.g., from ¢sural,’ intrans., fo be whirled, comes ¢ surarr-u’ (‘surattr-u’), trans., to whirl. ‘]’ and ‘du’ orm become ‘ttu,” e.g., from ‘mil’ ¢o return, comes ‘mitt-u,’ to cause to )0 Jical return, to redeem. From these instances it is clear that ‘du,” not ¢ ttu,’ i heme is to be regarded as the primitive form of this transitive suffix. 358 THE VERB. dist What is the origin of this transitive particle, or sign of activity “ttu’ or ‘du? I believe it to be identical with the inflexion, or ad- by na jectival formative, ‘attu’ or ‘ttu,” which was fully investigated in the ity section on ‘The Noun,” and of which the Canarese form is ‘ad’,’ the ek ¢ 1 ot 541 There is a transition of meaning when a noun is dong used adjectivally (7.e., to qualify another noun), as well as when a from verb is used transitively (7.e., to govern an object expressed by some noun in the accusative) ; and in both cases the Dravidian languages use (with respect to this class of verbs) one and the same means of expressing thin 0nw transition, viz., a particle which was originally a neuter demonstrative. Nor is this the only case in whicll the Tamil transitive verb tion L exhibits the characteristics of the noun used adjectivally, for it was nath shown also that the doubling and hardening of the consonant of the formative of the first class of transitive verbs is in exact accordance RIIUI this with the manner in which nouns terminating in those formatives double and harden the initial consonant when they are used to qualify app other nouns. Another illustration of this principle follows. lang‘ (4.) The fourth (a distinctively Tamil) mode of converting intran- they sitive verbs into transitives consists in doubling and hardening the final 16ce b consonant, if ‘d’ or ‘wr This rule applies generally, though not oy invariably, to verbs which terminate in those consonants; and it {Hly applies to a final ‘nd-u’ (euphonised from ¢d-u’), as well as to ‘d-u’ 1)) itself. The operation of this rule will appear on comparing ¢ vad-u, to aph wither, with ¢ vatt-u,’ to cause to wither ; ¢ 6d-u,’ to run, with ¢6tt-u,’ ¢o s drive ; ‘tind-u, to touch, with ‘titt-u, fo whet ; ‘mar-u,’ to become afy changed, with ¢ marRr-u’ (pronounced ‘ mattr-u’), to change. The cor- tg, responding transitives in Telugu are formed in the more usual way by s¢id, adding ‘chu’ to the intransitive theme, e.g., ‘miru-chu,’ to cause to 1 change, ¢ vidu-chu,’ to cause to wither. Cangg Tamil nouns which end in ‘d-u,” ‘nd-u,” or “r-u,” double and harden Iy the final consonant when they are used adjectivally, or placed in an atcl adjectival relation to a succeeding noun ; e.g., compare ‘kid-u,’ a Jungle, with “ katt-u vari, a jungle-path ; ‘irand-u,’ two, with ‘iratt-u casy nil, double thread ; ‘aRr-u, @ river, with ‘arru’ (pronounced ‘attru’) gove manal, 7tver sand. Thus we are furnished by words of this class with Lay another and remarkable illustration of the analogy which subsists in gove the Dravidian languages between transitive verbs and nouns used (eg adjectivally. e g 1I. CAausAL VERBS. Persg There is a class of verbs in the Dravidian languages which, though f avay l“;l( generally included under the head of #ramsitives, claim to be regarded CAUSALS. 359 ity distinctively as causals. They have been classed with transitives both rad- Beschi alone places them by native grammarians and by Europeans. 1 the in a class by themselves, and calls them ¢éval vinei,’ verbs of command, ! the t.¢., verbs which imply that a thing is commanded by one person to be un 15 done by another. 160 & Causals differ from transitives of the ordinary character, as well as Some from intransitives, both in signification and in form. RN The signification of intransitive verbs is confined to the person or 5500 thing which constitutes the nominative, and does not pass outward or The significa- itive, onward to any extrinsic object ; e.g., ¢ po-gir-én,” I go. veth tion of tramsitive or active verbs, or, as they are called in Tamil, out- t was ward action-words, passes outwards to some object exterior to the nomi- f the native, and which is generally put in the accusative ; e.g., ‘ unnei dance anuppu-gir-én,” I send thee : and as to send is to cause to go, verbs of atives this class, when formed from intransitives, are in some languages, appropriately enough, termed causals. Hitherto the Indo-European nalify languages proceed ¢ pari passu’ with the Dravidian ; but at this point 1tran- they fail and fall behind : for if we take a verb which is transitive of necessity, like this one, to send, and endeavour to express the idea of final | nob causing to send, t.e., causing one person to send another, we cannot by nd 1t any modification of structure get any single Indo-European verb to express the full force of this idea : we must be content to make use of id_u" a phrase instead of a single verb ; whereas in the Dravidian languages, L, 10 as in the Turkish and other languages of the Scythian stock, there is 1, 1 a form of the verb which will express the entire idea, viz., the causal : pcoine 6 (oI~ e.g., < anuppu-vi,’ to cause to send, which is formed from €anuppu,’ fo send, by the addition of the particle ‘vi’ to the theme. Ay by Transitives are in a similar manner converted in Turkish into st 10 causals by sutfixing a particle to the theme ; e.g., ‘sev-dur,’ to cause to love, from ¢sev, to love; and € étch-our, o causz to work, from qrden “atch,’ to work. 1 an -y, 0 There is a peculiarity in the signification and use of Dravidian causal verbs which should here be noticed. Indo-European causals raftl govern two accusatives, that of the person and that of the object ; e.g, i) I caused him (ace.) to build the house (acc.): whereas Dravidian causals g Wlth govern the object alone, and either leave the person to be understood s 0 (e.g., ‘ vittei (k)kattuvittén,” Tam., 7 caused to build the house, or as ,' USCd we should prefer to say, I caused the house to be built) ; or else the person is put in the instrumental; e.g., 1 caused to build the house “avanalé’ or ‘avanei (k)kondu,” through him, or employing him; that is, I caused the house to be built by him. hough Though the Dravidian languages are in possession of a true causal mf([@d - 360 THE VERB. the —formed by the addition of a causal particle,—yet they sometimes Ta1e8 resort to the less convenient Indo-Eurcpean method of annexing an auxiliary verb which signifies 0 make or t0 do, such as ‘sey’ and heen ‘pann-u’ in Tamil, ‘mad-u,” in Can., and ‘chéy-u,” in Tel. These - auxiliaries, however, are chiefly used in connexion with Sans. deriva- vl tives, it being contrary to the Dravidian idiom to combine indigenous natu particles with foreign themes. The auxiliary is annexed to the in- Teval I finitive of the principal verb. el Tamil idiom and the analogy of the other dialects require that causzls should be formed, not from neuter or intransitive verbs, but shut ) be th from transitives alone; but sometimes this rule isfound to be neglected. Even in Tamil, ¢ vi,’ the sign of the causal, is in some instances found In 16 to be annexed to intransitive verbs. Thisusage is not only at variance Telu with theory, but it is unclassical and unidiomatical. In each of those ordin cases a true transitive, derived from the intransitive in the ordinary Telu manner, is in existence, and ought to be used instead. Thus, ¢varu- origi vi,” Tam., to cause to come, is less proper, as well as less elegant, than anne ¢ varu-ttu;’ and ‘nada-ppi,” to cause to walk, to guide, than nada-ttu.’ the 1 The use of the causal, instead of the active, where both forms and { exist, is not so much opposed to the idiom of the other dialects, as to a8 801 that of the Tamil. The use of one form rather than another is Ly optional in Telugu and Canarese ; and in some instances the active inchy bas disappeared, and the causal alone is used. Thus ‘rappinchu,’ or pincl ‘ravinchu,’ fo cause to come, the equivalent of the Tamil ¢varu-vi, pin is preferred by the Telugu to a form which would correspond to mug “varu-ttu: and instead of ‘akk-u,” Tam., to cause to become, to make, in {] which is the active of ‘4g-u’ and is formed by the process of bUCOF doubling and hardening which has already been described, the Telugu “may uses the causal ‘ka-vinchu,” and the Canarese the corresponding diffey causal ¢ag-isu.’ of th The causal particle which is most commonly used in Tamil is ¢vi; verh e.g., ¢ pannu-vi,’ to causc to make, from ¢pannu,’ to make; and ‘kattu- 01 ) vi, to cause to build, from ‘kattu,” to build. Instead of ‘vi’ we thep sometimes find ¢bi’ or ‘ppi,” according to the euphonic requirements ])y il of the preceding syllable. When the theme ends in a nasal, which it I does but rarely, ‘bi’ is added to form the causal; e.g., ‘kin-bi,’ #o 13gy) cause to see, to show. [A more idiomatic word, however, is the proper {l 10 | transitive of ‘kéan,” see; viz., ‘katt-u, i.e., ‘kén-ttu, fo show.] When the theme ends in a vowel which is of such a character that if a eVﬂj i be sonant follows it it will necessarily be hardened and doubled, ¢ vi’ or deryy }i ‘bi’ changes dialectically into ‘ppi; e.g., from ‘edu,’ to take up, is i L formed the causal ¢ edu-ppl,’ to cause to take up. is undoubtedly 18 th i ;i _— CAUSALS. 361 fimes the ‘most common form of this particle in Tamil, and ¢bi’ is the 1g an ] rarest; yet ‘bi’ seems to be the original from which the others have ana been derived : for ‘ppi’ caunot have been derived from ¢vi,’ whereas Thiese ¢bi,” after certain vowels, dialectically and necessarily becomes ¢ppi ;’ eriva- ‘vi’ also will not harden into ¢bi, whereas ‘bi’ will readily and enols naturally soften into ¢vi.’ I conclude, therefore, that ‘bi’ may be he in- regarded as the normal form of the causal particle. In Telugu, causal verbs ordinarily end in ‘inchu; eg., ¢chéy- e that inchu,’” to cause to do, from °chéy-u,’ ¢o do; <miy-inchu,” fo cause to b, but shut, from ¢miy-u,” to shut, and this ‘inchu’ might be supposed to Jected: be the Telugu shape of the causal particle: but it will be seen that found in reality ¢i’ alone is to be regarded as the causal particle of the ) the rance Telugu ; and that ‘i is a softened, degraded form of ¢vi, those The final ‘nchu’ of the ordinary causal particle of the Tamil. Telugu causal is merely a nasalised formative, and is identical in linary “yari- origin with the Tamil formative fklku.? This formative ‘kku’ is , than annexed to ‘vi' in certain parts of the Tamil causal verb, viz., in bt the infinitive, and in the third person neuter of the aorist or future; forms and the identity of the Tam. ‘kku’ with the Tel. ‘nechu’ is apparent asto as soon as the Tamil infinitive is compared with that of the Telugu. E.g., compare ¢§ey-vikka,” Tam., to cause to do, with the Tel. ¢chéy- het 18 active incha,” and ‘ tira-ppikka,’ Tam., to cause to open, with the Tel. ¢ tera- n, of pincha.’ Here ‘vikka’ and ‘ppikka’ alternate with ¢incha’ and -V, ‘pincha; and hence it appears that the ‘nch’ of the one dialect must as certainly be a formative as the ‘kk’ of the other is. Even nd to ke in the Tamil of the southern Pandiya country, ‘kk’ systematically Tl 3 becomes ‘ch Thus the correct Tamil ‘mara-kka,” Zo forget, is 088 of ‘ mara-cha’ in the southern patois, precisely as in Telugu. The chief [elugu nding difference between the Tamil and the Telugu, with respect to the use of this formative ‘kk,’ is that it is used by two parts of the Tamil verb alone, viz.,, the infinitive and the indefinite neuter, future, or 115 aorist ; whereas in Telugu it is added to, and compounded with, the katto- theme itself, and is used accordingly by every part of the verb, even 1 we by the imperative. Ordinarily this formative ‘chu’ is used unchanged ments in Telugu; but when it follows i, it is invariably euphonised or hich nasalised into ‘nchu: hence the caunsal verb of the Telugu terminates bi,' not in ¢i-chu, but in ‘i-nuchu,” proncunced ‘intsu.’ In like manner proper every verb the base of which ends in ‘i, terminates in ‘inchu,” though TVhen it be not a causal; e.g., ¢ jayi-nchu, to conquer, from ‘jayi,’ a Sans. i1 if derivative. g’ or We thus come back to the conclusion that ¢i,” softened from vi, Ilp; iS is the regular causal particle of the Telugu. In a few instances the ptedly 362 THE VERB. softening process by which ‘vi’ was changed into ‘i has been dialects resisted ; and in those instances ‘vi’ is the sign of the casual in Telugn, Telugu, as in Tamil; eg., compare ¢vida-vi-nchu’ or ¢vidi-vi-nchu,’ {JayiI to rescue, to cause to leave, with the Tamil ¢vidu-vi. The Tel. < ké-vi- (anares nchu,” fo cause to become, has retained this particle; whilst it has been the Tel lost by the corresponding Can. ‘Ag-i-su” We occasionally find the which 1 causal formed by ‘pi,” and even ‘ppi’ in Telugu as in Tamil; and The formati though the use of these hardened forms is rare, yet their existence in Telugu serves still further to identify ¢i’ with the Tamil ¢vi,” ¢bi,’ identiq and ‘ppi’ ‘ tera-pi-nchu,’ Tel., o cause to open, is an example of the and of use of ‘pi; and ‘teppinchu’ (‘ te-ppi-nchu’), o cause to bring, from verbs e “te-chu,’ to bring, illustrates the use of ‘ppi.’ The Telugu verbs i, b in ‘chu,’ ‘nchu,’ “pu, ‘mpu, &ec., which are destitute of this causal them in particle under any form (e.g., vidu-chu’ and ¢ vidu-pu,’ ¢o cause to foundin quit, to rescue; ‘vanchu,’ to bend; ©lépu,’ to raise) are to be regarded older as transitives, not as causals. Tami They are formed, not by annexing ¢ vi’ /: or ‘i) but by the doubling and hardening of the final consonant by the of the formative (e.g., compare ‘1épu,’ to raise, with the corresponding particle Tamil ‘eruppu,’ the transitive of ¢erumbu’); and the verbs from ‘Vi’ha which they are so formed are not actives, but neuters. Instead, there- Kk’ i fore, of saying that ¢ tir-u,’ to end, forms its causal either in ¢ tir-chu’ The or ¢ tir-pinchu,’ it would be more accurate and more in accordance betweer with Tamil analogies, to represent ©tir-u’ as the neuter, ¢tir-chu’ as The try the transitive, and ¢ tir-pi-nchu’ as the causal. It 1s of the essence the (y of the true causal that its theme is a transitive verb; e.g., ¢ katt-inchu,’ like t to cause to build, from ‘katt-u,” to build. be ful) In Canarese, causal verbs are formed by suffixing ¢is-u,’ or rather itlf), ‘i-su,’ to the tramsitive theme; e.g., from ‘mad-u,” to do, is formed €. ‘i ¢ mad-i-su,’ to cause to do. This causal particle ¢i-su’ (in the ancient Ellid ‘n{; dialect “i-chn ’) is annexed to the theme itself before the addition valk, of the signs of tense, so that it is found in every part of the causal The verb, like the corresponding Telugu particle ¢i-nchu,’” with which it Particle is evidently identical. It has been shown that the Telugu ¢i-nchu’ & Tqf must have been nasalised from ‘i-chu’ (the phonetic equivalent of Ulang | the Tamil ¢i-kku,” for < vi-kku’); and now we find this very ‘i-chu’ the dey in Canarese. The change in modern Canarese from ‘i-chu’ to ¢i-fu’ ( a‘ya?? I is easy and natural, ‘§’ being phonetically equivalent to ¢ch,” and The ‘chu’ being pronounced like ‘tsu’ in Telugu. Othey D Au additional proof, if proof were wanting, of the identity a ghe of the Can. ‘i-su’ with the Tel. ‘i-nchu,” is furnished by the fomg class of derivative verbs, or verbs borrowed from the Sansecrit. indyg Sans. derivative verbs are made to end in ‘i’ in the Dravidian ) .’/0,‘ an( CAUSALS. 363 ' been dialects (e.g., ¢ jay-i, comquer); and those verbs invariably take in sual in Telugu, as has been said, the formative termination ‘nchu,’ eg., nchu, ¢ jayi-nchu.’ The same verbs invariably take ‘i-su, or ‘y-isu,’ in kd-vi- Canarese. Thus from the Sans. derivative theme, ¢ dhari,’ to assume, s been the Telugu forms the verb ¢dhari-nchu,’ the Canarese equivalent of nd the which 18 ¢ dbhari-§u.’ 1; and These verbs are not causals; bnt the use which they make of the ence In formative ¢ nchu’ or ‘su, preceded by i, illustrates the original 4 identity of the Canarese causal particle ‘i-su’ with the Tel. ¢ i-nchu,’ > of the and of both with the Tamil ¢vikku.’ In Tamil, Sans. derivative g, from verbs end in ‘1 as in Canarese and Telugu, but the causal particle | verbs ‘vi, ‘bi, or ‘ppi, is never added, except it is desired to convert them into causals. causal Hence in Tamil no one is in any danger of con- e {0 founding the true causal with the Sans. derivative. Generally the 3garded older and harsher sounds of the Canarese have been softened by the (2] g 1 Tamil ; and in particular, the Canarese k’ has often been softened sonant by the Tamil into ¢z S or ‘ch: but in the instance of the causative onding particle, exactly the reverse of this has happened; the Tamil ¢bi’ or g from ‘vi’ having been softened by the Canarese into ¢1,” and the formative ¢kk'iinto ‘&’ there- r-chn’ The Canarese, like the Telugu, does not so carefully discriminate between transitive and causal verbs as the Tamil has been found to do. rdance The true causal of the Tamil is restricted to transitive themes; but ho' a8 pssence the Canarese, notwithstanding its possession of transitive particles, incha, like those of the Tamil (e.g., compare *nera-hu,’ to fill with ‘neri,’ fo be full, and ¢ tiru-pu,’ fo turn (actively), with ‘tiru-gu,’ to turn (of wself), yet it often annexes the causal particle to intransitive themes; ¢ ather formed e.g., ‘0d-i-su,’ to cause to run (Tam. ‘ott-u’), from ‘6d-u,’ to run; jicient and ‘nad-i-su,” fo cause to walk (Tam. ‘nada-ttu’), from ‘nadi,’ ¢o walk. Jdition ca,uSal The oldest and purest form of the Indo-European causative i 1t particle is supposed to be the Sanserit ‘aya,” with ‘p’ prefixed after A a root In ‘a ., ‘aya’ becomes ‘i’ in Old Slavonic; and the resem- acht’ 2> blance between this and the Telugu €1 is very close ; nevertheless, pJent of the derivation of the latter from < w1 or ‘bi’ and of the former from Gl ‘aya,” proves that the resemblance is purely accidental. L The Tulu forms its causal verbs in a different manner from the by and Py to the verbal theme, other Dravidian dialects, viz., by suffixing a and then adding the signs of tense: e.g., from ‘marp-u,” fo make, is Jentity formed ¢ marp-d-vu, to cause to make. This greatly resembles the hy the Hindustani causals; e.g., ‘chal-wa-na, fo cause to go, from ¢ chal-na,” to et go; and as the Hind. causative particle ‘wa’ has probably been ﬁvidiaﬂ 364 THE VERB. a A D derived from the Sanscrit ‘aya’ or ¢p-aya,” the Tulu ¢ may be agent, ¢ supposed to proceed from the same or a similar source. ool In Gond “ha’ or ‘h’ is the causal particle, and is added to the (2) present participle of transitive verbs, not to the theme. the pre by the III. Tue Passive Voick. t b, Each of the primitive Indo-European languages has a regular ‘mugin passive voice, regularly conjugated. The Sanscrit passive is formed finished by annexing the particle ‘ya’ (derived from ¢y4,’ to go), to the verbal of pass theme, and adding the personal terminations peculiar to the middle comple volce. Tra Most of the languages of the Scythian family also form their forms, passives by means of annexed particles. To the verbal theme Thy the Turkish suffixes in order to form the passive, *il ot il and is the Finnish ‘et ; the Hungarian ¢at ‘et, ‘tet; and to these atfi 4 particles the pronominal terminations are appended in the usual become, manner. the agg The Dravidian verb is entirely destitute of a passive voice, pro- may g perly so called, nor is there any reason to suppose that it ever had a become, passive. None of the Dravidian dialects possesses any passive particle Vo or suffix, or any means of expressing passivity by direct inflexional used in changes: the signification of the passive voice is, nevertheless, capable tion of of being expressed in a variety of ways. We have now to inquire done, } into the means which are adopted by the Drividian languages for which conveying a passive signification; and it will be found that they cor- the ([01' respond in a considerable degree to the means used for this purpose Iy by the vernaculars of Northern India—which also are destitute of a Substy regular passive voice. Same fg In the particulars that follow all the Dravidian dialects agree: 2 pay what is said of one holds true of all. The (1.) The place of a passive voice is to a large extent supplied by the use of the neuter or intransitive form of the verb. This is in third p \‘Oice W every dialect of the family the most idiomatic and characteristic mode or by ﬂ of expressing the passive; and wherever it can be used, it is always the gt preferred by classical writers. Thus, it was broken, is ordinarily Sitate 9 expressed in Tamil by ¢udeindadu,” the preterite (third person sin- toa ty gular neuter) of ‘udei,” intransitive, ¢o break or become broken; and mplie; though this is a neuter, rather than a passive properly so called, and of the, might literally be rendered ¢ has come into a broken condition, yet it for ¢, is evident that for all practical purposes nothing more than this is As required to express the force of the passive. The passivity of the 00t iy expression may be increased by prefixing the instrumental case of the lery : THE PASSIVE VOICE. 365 13y be agent, e.g., “ ennal udeindadu,’ ¢ was broken by me, ov literally ¢ came wnto @ broken condition through me. | to the (2.) A very common mode of forming the passive is by means of the preterite verbal participle of any neuter or active verb, followed by the preterite (third person singular neuter) of the verbs o become, to be, to go, or (occasionally) to end. Thus, we may say either regular ‘mugindadu,’ i is finished, or ¢mugind’ Ayittru, literally Aawving formed Jinashed it s become. This form adds the idea of completion to that verhal of passivity: not only is the thing done, but the doing of it is middle completed. Transitive or active verbs which are destitute of intransitive 1 their forms, may in this manner acquire a passive signification. theme Thus ‘katt-u,’ to bind or build, is necessarily a transitive verb, ‘Al and is without a corresponding intransitive ; but in the phrase ‘kévil ) these katti ayittru,’ the temple is built, literally the temple having built has y ustal become, a passive signification is acquired by the active voice, without the assistance of any passive-forming particle. ¢ poyittru, it has gone, 8, - may generally be used in such phrases instead of ‘ayittru, 4t s had a become. article Verbal nouns, especially the verbal in “dal’ or ¢al’ are often exional used in Tamil instead of the preterite verbal participle, in the forma- mpab]@' tion of this constructive passive; e.g., instead of ‘Seyd’ Ayittru,’ 4 4s inquire done, literally having done it has become, we may say ‘Sejdal Ayittrn,’ g8 for which though it is used to express the same meaning, literally signifies 1ey coI- the doing has become, i.c., it has become a fact, the doing of it is completed. In these instances the use of the active as a passive (with the purpose i ofa substantive verb) corresponds to the New Persian rule of using the same form of the verb as an active when it stands unsupported, and 3 gree . as a passive when followed by the substantive verb. The Dravidian constructive passives now referred to require the ied by third person mneuter of the auxiliary verb. The force of the passive s is 10 voice will not be brought out by the use of the masculine or feminine, or by the epicene plural. If those persons of the verb were employed, ¢ mode always the activity which is inherent in the idea of personality would neces- sitate an active signification ; it would tie down the transitive theme narly to a transitive meaning ; whereas the intransitive relation is naturally o Sl i, 4l implied in the use of the neuter gender, and therefore the expression of the signification of the passive (viz, by the intransitive officiating 3d7 aﬂd for the passive) is facilitated by the use of the third person neuter. this A somewhat similar mode of forming the passive has been pointed out in the Hindustani and Bengali; e.g., < jana yay,” Beng., it is known, Of thf,' literally it goes to be known. of the ‘jana’ is represented by some to be a 366 THE VERB verbal noun, by others to be a passive participle but, whatever it be, able ] found there is some difference between this idiom and the Dravidian one; beaten for in the corresponding Tamil phrase °terind’ Ayittru,” o s known, 2 ¢ terind- is unquestionably the preterite verbal participle of an (4 n eac! intransitive verb, and the phrase literally means having known it is become literally Aaving known it is gone, con- 1means ¢ terindu poyittru veys the same signification It is remarkable that a verb signifying ANnex to go should be used in the Drividian languages as a passive-making 4, ) auxiliary, as well as in the languages of Northern India or ¢ B Occasionally Dravidian active or transitive verbs themselves are by Tience used with a passive signification, without the addition of any intran- sitive auxiliary whatever Relative participles and relative participial of the nouns are the parts of the verb which are most frequently used in this which manner; e.g., erudina suvadi unduj ach’ aditta pustagam véndum (p)pad Tam., 7 have got a written book ; 1 want a printed one In this phrase Siifere both erudina written, and ‘ach’-aditta,” printed, are the preterite nore It relative participles of ¢ranmsitive themes The former means literally that wrote; yet it is used passively to 51gn1fy written, and the latter finit means literally that printed or struck off, but is used passively as equl— It valent to that is printed. her The relative participial noun, especially the preterite neuter, is to act oftentimes used in the same manner e.g., in ‘Sonnadu pédum, Tam., tive ¢ what was said 1s suficient, < Sonnadu,’ literally means that which said Necess but the connexion and the usage of the language determine it to Slomiff signify passively that which was said; and so distinetively in this case Sibivg is the passive sense expressed by the connexion alone, that the use of m T¢ the more formal modern passive $olla-(p)pattadu, would sound formg Tam., ¢anéde, Tel that s called awkward and foreign endra, action literally that spoke, is another very common instance of the same rule Tany ‘Iyésu enbavar,” Tam., signifies literally, Jesus—he who speaks; but Tamj] usage determines it to mean ke who s called Jesus thergh (3.) The verb ‘umn,” ¢o eat, is occagionally used in the Dravidian wel] It is inva- languages as an auxiliary in the formation of passives Tl ably appended to nouns {substantives or verbal nouns), and is never 80 ¢a] compounded with any part of the verb; e.g., ‘adi undan,’ ke was beaten, Seneys or got a beating, literally he ate a beating, padeipp’ undén,” I was created, literally 7 ate a creating h ey 0 The same singular idiom prevails also in the North-Indian ver- 0 the naculars The particular verb signifying fo eat which is used in those auxljy languages diflers, indeed, from the Dravidian u but the idiom is thOrou identical, and the existence of so singular an idiom in both the In It is remark- northern and the southern family is deserving of notice %0 gq) — THE PASSIVE VOICE. 367 It be, able that the same peculiar contrivance for expressing the passive is i 00e found in the Chinese, in which also 0 eat a beating, means to be kngun, beaten. of an (4.) The mode of forming the passive which is most largely used m ot s in each of the modern colloguial dialects of the Dravidian family, is by 6, CON- means of the auxiliary verb ‘pad-u,” to suffer, to experience, which is nifying annexed to the infinitive of the verb signifying the action suffered; making e.q., ‘kolla~(p)pattan,” Tam., ke was killed, literally, he suffered a killing or @ to kill. It is also annexed to nouns denoting quality or condition; jes are e.g., ‘vetka-(p)pattin,’ he was ashamed, literally he suffered or expe- mntran- rienced shame. The ultimate base of a verb is sometimes used instead tiipial of the infinitive or verbal noun in construction with this auxiliary, in in this which case the base is regarded as a noun; e.g., instead of ‘adikka- ndum, (p)pattan,” we may say ‘adi pattin,’ ke was beaten, or literally Ze phrase suffered a beating; and where this form can be used, it is considered reterite more idiomatic than the use of the infinitive. terally It is evident that this compound of ‘pad-u,” fo suffer, with an infinitive or noun of quality, is rather a phrase than a passive voice. ) Jafter 5 equ- It is rarely found in the classics; and idiomatic speakers prefer the other modes of forming the passive. ‘pad-u’ is often added, not ounly iter, 18 to active, but also to neuter or intransitive verbs; but as the intransi- ' Tam, tive expresses by itself as much of a passive signification as is ordinarily necessary, the addition of the passive auxiliary does not alter the h suid; e lt tO signification ; e.g., there is no difference in Tamil between the intran- 1is ¢ase sitive ‘ teriyum,’ it appears, or will appear, and teriya (p)padum; or in Telugu between ‘ teliyunu” and ¢ teliya badunu,” the corresponding y 156 Of sound forms. In ordinary use ‘pad-u’ conveys the meaning of continuous action or being, rather than that of passivity; e.g., ‘irukka-(p)patta, il e rule. Tam., is vulgarly used for ‘irukkira, thatis; and I have heard a S5 hut Tamilian say, ‘nin nandrij $ppida-(p)pattavan, Tam., meaning thereby, not J have been well eaten, but I have been accustomed to eat well. svidian The Dravidian languages, indeed, are destitute of passives properly s 10va: § pever so called ; and, therefore, they resist every effort to bring ¢pad-u’ into Dty general use. Such efforts are constantly being made by foreigners, who are accustomed to passives in their own tongues, and fancy that r [ s they cannot get on without them; but nothing sounds more barbarous to the Dravidian ear than the unnecessary use of ‘padu’ as a passive ﬂ ver- ) those auxiliary. It is only when combined with nouns that its use is thoroughly allowable. jom i In none of the Dravidian dialects is there a middle voice, properly h the so called. The force of the middle or reflective voice is expressed 368 THE VERB. then constructively by the use of an auxiliary verb, viz, by ‘kol,” Tam., to natic take (Tel. ‘kon-u’); eg., “panni-(k)kondén,’ I made it jfor myself, 116Ce! literally, I made and took it. This auxiliary sometimes conveys a 18 60 reciprocal force rather than that of the middle voice; eg, pési-(k) kondargal, Tam., they talked together; ¢adittu-(k)kondargal,” they beat 118 one another. The same usage appears in the other dialects also. hut ¢ pIon IV. Tue NEGATIVE VOICE. whi Properly speaking, the Dravidian negative is rather a mood or affin voice than a conjugation. All verbal themes are naturally affirmative, ar and the negative signification is expressed by means of inflexional Lo additions or changes. Nevertheless, it will conduce to perspicuity to of ar inquire now into the negative mood or voice, before entering upon the consideration of the pronominal terminations and tenses. Tl The regular combination of a negative with a verbal theme is a 8100 peculiarity of the Scythian family of tongues. Negation ig generally the s expressed in the Indo-European family by means of a separate particle tens (IR used adverbially; and instances of combination like the Sanserit ‘nasti,’ Teali it is mot, the negative of ¢ asti,” ¢t is, are very rare, and are found only of g in connexion with substantive or auxiliary verbs: whereas, in the ab\ st Scythian languages, every verb has a negative voice or mood as well as an affirmative. The Scythian negative voice is generally formed 100 of of by the insertion of a particle of negation between the theme and the aece pronominal suffixes; and this is as distinctive of the Dravidian as of the Turkish and Finnish languages. Different particles are, it is true, Satls used in the different languages to express vegation; but the mode in Vallg which such particles are used is substantially the same in all. In general, the Dravidian negative verb has but one tense, which trace (o) is an aorist, or is indeterminate in point of time ; e.g., pogén,” Tam. hs re]an (‘ povanu, Tel., ¢ pogenu,’ Can.), 7 go not, means either 7 did not, I do deriy not, or I will not go. The time is generally determined by the con- text. tnide The only exception is in the Ku, in which there is a negative In most of the dialects there 1s preterite, as well as a negative aorist. only one mood of the negative in ordinary use, viz., the indicative. If 15 thy an infinitive and imperative exist, it is only in classical compositions 13t1o that they appear; and they are ordinarily formed by the help of the hemg adu : infinitive and imperative of the substantive verb, which are suffixed as Mo auxiliaries to the negative verbal participle; e.g., ¢ éeyyé‘d’—iru,’ Tam., do not thou, literally be thow doing not. In the Telugu alone, a negative infinitive, and a prohibitive or The negative imperative, are in ordinary use even in the colloquial dialect. 0cea; In the Dravidian negative voice, as in the affirmative, the verbal (Ingt, THE NEGATIVE VOICE. 369 Tam,, to theme remains unchanged; and in both voices the pronominal termi- P myself, nations are precisely the same. The only point, therefore, which it is necessary to investigate here is the means whereby the idea of negation ODVEYS & is expressed. “pisi () they bent The Dravidian negative is altogether destitute of signs of tense : it is destitute, not only of the signs of present, past, and future time, U but even of the sign of the aorist; and in Tamil and Canarese the Thus, pronominal suffixes are annexed directly to the verbal theme. mood o whilst the present, past, and future tenses (first person singular) of the irmative, affirmative voice of the Tamil verb ¢ var, to flourish, are ‘var-gir-én,’ ¢ var-nd-én,” ¢ var-v-én ; the corresponding negative is simply ¢var-én,’ flexional I flourish not—Tliterally, as appears, flourish-1,—without the insertion icuity 10 upon the of any sign of time between the theme and the pronoun. What is the rationale of the Dravidian negative ? The absence of signs of tense evidently contributes to the expres- eme 1§ & sion of the idea of negation : it may at least be said that it precludes generally the signification of the aflirmative. In consequence of the absence of 3 p:u‘ticle tense-signs the idea expressed by the verb is abstracted from the it ‘ﬂéSti:T ind only realities of the past, the present, and the future: it leaves the region of actual events, and passes into that of abstractions. Hence, this ) n the abstract form of the verb may be supposed to have become a negative mood, not by a positive, but by a negative process,—by the absence y formed of affirmation, not by the aid of a negative particle. Is this to be accepted as the rationale ? dian 88 of If we examined only the Tamil and the Canarese, we might be it 19 true, satisfied with this explanation of the origin of the negative ; for in the o mode in various persons of the negative voice in both languages there is no trace of the insertion of any negative particle; and though the vowel 6, which ‘a’ has acquired a predominant and permanent place in the verbal and ) Tan. relative participles, we should not feel ourselves warranted in consi- not, I[do dering that vowel as a particle of negation, without distinet, reliable - the cob- evidence from some other source. ) negﬂ,ﬁ\'e The only peculiarity in the personal forms of the Tamil negative i there 3 is the invariable length of the initial vowel of the pronominal termi- 'f[tive' H nations. Thus, the initial ‘a’ of the neuter singular demonstrative 111})05iti ons being short, we should expect the Tamil of & Aourishes not to be ‘var- EIP []f the adu ; whereas it is ‘vaAr-Adu’ or ¢ vAr-4’ This increase of quantity Sm‘ﬁxed o might arise from the incorporation and assimilation of some inserted m: Tﬂl‘n'7 vowel ; but we might also naturally suppose it to be merely euphonic. The corresponding vowel is short in Telugu ; but even in Telugu it is pive occasionally lengthened for the sake of emphasis ; e.g., ¢ palukdka’ g} dilth (instead of ¢ palukaka’), without speaking. In the Canarese negative 2B t — 370 THE VERB. we miss even this lengthening of the initial vowel of the pronominal The T terminations ; e.g., we find invariably ¢bal-adu,’ instead of the Tamil any trace ¢ var-ddu.’ In the verbal and relative participles in both languages the thow goes vowel ‘a’ is inserted between the theme and the formative, and this is cerfain ¢ a 9 is invariably short in Canarese and long in Tamil ; e.g., ‘bal-a-de,’ The le Can., not having lived, or without living ; Tam., ¢var-ddu’ or °var- acoordane The verbal noun in Tamil is ¢ var-a--mei, the instead of a-mal,” without lwing. not lLiving. The relative participle that lLived or lives not, is in Can., We ‘bal-a-da,” in Tam., ¢ var-4-da.’” In these instances, if euphony alone negation had been considered, ‘u,’ the ordinary enunciative vowel, would have the forma appeared where we find “a: it may, therefore, be concluded that ‘a’ appeared (euphonically ¢4’ in Tamil) has intentionally been inserted, and that the analo it contributes in some manner to grammatical expression. Tamil an It will be found that much light is thrown upon this subject by the een the Telugu. The pronominal terminations of the negative voice of the suceeeded Telugu are identical with those of the present tense of the affirmative. sultx, apy In Tamil and Canarese the pronominal terminations of the verb com- dence of mence with a vowel; but in Telugu verbs the pronoun is represented Pronomin; by the final syllable alone, and that syllable invariably commences Itis with a consonant. Hence, if no particle of negation were used in the formatiye conjugation of the Telugu negative voice, the pronominal suffix would The il be appended directly to the verbal theme, and as every Telugu theme ‘W’ o terminates in the enunciative ‘u,’ that ‘u would not be elided, but Without g would invariably remain. ‘What then is the faet ? formatiy On examining the Telugu negative, it is found that the vowel ‘a’ 0 slipp invariably intervenes between the theme and the pronominal suffix ; and as the final enunciative u’ of the theme has been elided to make DOUHS, al of verbs way for this “a,’ it is evident that ‘a’ is mot an euphonic insertion, o the by but is a particle of negation. Compare ¢chéy-a-nu,” Tel., 1 do not, ideutical ' with Tam., ‘ey(y)-én; ¢chéy-a-vu, thou dost not, with Tam., ﬁ\'e, iS Y “Sey(y)-ay ;. ¢ chéy-a-mu, we do mot, with Tam., sey(y)-om ;" ¢ chéy- From this comparison it it iS More a-ru, you do mot, with Tam., ‘sey(y)-ir. cannot be doubted that ‘a’ is regularly used in Telugu as a particle added by (mia’ Wwhi of negation. We find the same ‘a’ used in Telugu, as in Canarese and Tamil, in the negative verbal participle; e.g, chéy-a-ka,” without Natioy of nellter 10 dotng ; in the relative participle, e.g., ¢ chéy-a-ni,’ that does not ; and in In each of these par- ey gy the verbal noun, e.g., ‘ chéy-a-mi,’ the not doing. ticipials ‘a’ is used in the same manner by the Canarese, and ‘4’ o f}wu). by the Tamil: and that those vowels are not euphonics or conjunctives, i Plopey but signs of negation, even in Tamil-Canarese, is now clearly proved Veth iy by the evidence of the Telugu, in which a similar ¢ a’ is used, not only 4 a}) ﬂle Si by the participles, but by all the personal forms of the verb. 8 ugyy) Q; b= { 1 THE NEGATIVE VOICE. ninal amil The Tel. verb to go, forms its ordinary negative, it is true, without 9 any trace of this vowel of negation : eg., “pénu,” I go not, ‘podvu, sthe thow goest not. This, however, is only an apparent irregularity, for it this is certain that the correct forms are ‘pév-a-nu’ and °pov-a-vu.’ e, The lengthening of the included a’ of ‘kéanu,’ 7 become not, is in yar- accordance with the Telugu law of displacement, ¢kanu bl being N i instead of ‘ak-a-nu’ or ‘ag-a-nu,’ the equivalent of the Tamil ‘agén.’ Can,, We have thus arrived at the conclusion that ‘a 2 is the sign of alone negation which is systematically used by the Dravidian language in have the formation of the negative voice of the verb. It has, it is true, dis- b appeared from the conjugated forms of the Tamil and Canarese; but {hat the analogy not only of the Telugu personal forms, but also of the Tamil and Canarese participles, proves that it must originally have ythe been the common property of all the dialects. The negative ‘a,” being f the succeeded in Tamil and Canarese by the initial vowel of the pronominal ative, suffix, appears gradually to have got incorporated with it: and an evi- com dence of this incorporation survives in the euphonic lengthening of the ented pronominal vowel in Tamil and Tulu. 60068 It is desirable now to inquire into the participial and imperative n the formatives of the negative verb. yould The negative verbal participle of the Tamil is formed by suffixing heme ¢4-du’ or ‘a-mal ; e.g., < fey(y)-a-du’ or ‘Sey(y)-a-mal,’ not doing, or , but without doing. In the highest and lowest Tamil “mei’ i1s used as the formative of this participle instead of mal,’ e.g., ¢ varuv-a-mei,” with- b ‘e’ out slipping. ‘me1 constitutes the ordinary termination of abstract offs 5 nouns, and is added both to crude roots and to the relative participles make of verbs ; e.g., ¢ tAr-mei,’ lowness, humility ; ¢ iru-kkindr-a-mei,” a being }I‘tiOn) '| or the being. The formative termination of negative verbal nouns is ) 0l identical with this abstract ¢mei ;’ and ‘mal,’ the participial forma- I‘fbm') tive, is evidently equivalent to it, and probably the original form: for chey- it is more likely that a final ‘1’ should have been softened away than qon it added by use. The verbal noun of the Telugu negative verb ends in The other Tamil termi- uticle “mi,” which is virtually the same as * mel.’ 1686 nation of negative verbal participles, ‘du,’ is an ordinary formative of rlﬁo thout neuter nouns of quality. The corresponding Canarese termination is Gn 2 ¢de ;” and in Tamil ¢ du,” with a subsequent emphatic = is commonly ¢ pM used as a negative imperative or prohibitive ; e.g., Sey(y)-4-d-&,” do % not thou,—a proof that the negative verbal participle in ‘du’ or ¢de’ d‘a 1175 is properly a verbal noun. The relative participle of the negative )ro\’ed verb in each of the dialects, except the Telugu, is formed by suffixing “a, the sign of the relative, to the verbal participle in ¢d-u,” eliding { onlf as usual the enunciative ‘u; eg., ‘fey(y)-4-da, Tam., ‘géy-a-da,’ 282 372 THE VERB. Can., that does or did not. Many additional forms are constructed by (ol the addition of the various tenses and participles of the substantive bl verb, and it is by the help of that verb that the negative imperative ¢ipl and negative infinitive in both Canarese and Tamil are ordinarily ‘tel formed. The negative relative participle of the Telugu is formed by adding ¢ni,” instead of the usual relative ‘a,’ to the negative particle; by whi e.g., ‘chéy-a-ni,’ that does or did not. This ¢ni1 is one of the Telugu inflexional increments, and is also used as a particle of conjunction, as Case abs will be seen under the head of the relative participles. bee The negative verbal participle and negative imperative of the Telugu require to be separately investigated. caus with Mr. A. D. Campbell, in his Telugu Grammar, states that the nega- tive verbal particle is formed by suffixing ‘ka’ to the infinitive of the affirmative voice ; and that the prohibitive is formed in like manner the Tely by suffixing ‘ku’ or ‘ka’ to the infinitive, with the ordinary addition of ‘mu’ or ‘mo.’ o i In consequence of this representation, Dr. Stevenson has been led to consider ‘kun’ as a Telugu sign of negation, and to belie search for allied or equivalent particles in other Indian languages. the The comparison of the negative verbs in the various Dravidian dialects which has just been made, proves that this representation is inaccurate, of ¢ and that the ‘a’ to which the ‘ka’ and ‘ku’ aforesaid are suffixed is not the ‘a’ which forms the sign of the infinitive, but the negative prou particle ‘a The suffixes of the forms in question, therefore, are not tho, ‘ku’ or ‘ka, but ‘a-kun’ and ‘a-ka,” or ‘4-ka; and thus ‘chéy-a- Phat ka,” without doing, or not having done, and ‘chéy-a-ku’ or ‘chéy-a-ka,’ 181 do not, come into harmony with the other Telugu forms, viz., ‘chéy-a- a"l\‘a'/ | ni, that does not, ‘chéy-a-mi,” the not doing ; and also with the negative participles and verbals of the other dialects. Pond; The “a’ of the Telugu imperative and negative verbal participle o, being undoubtedly the sign of negation, it only remains to inquire into and g the origin of the ‘ka’ or ‘ku’ which is suffixed to it. Teso The participial suffix ‘ka’ is evidently used in Telugu for the for same purposes as the Tamil suffixes ‘du,” ‘mal, and ‘mei,’ and the akg. ) Can., ‘de’ Those suffixes, though used by verbal participles, are un- of th; I consider doubtedly to be regarded as formatives of verbal nouns. 110un@ ‘ka’ also as proceeding from a similar origin ; for in Telugu many I8 1de verbal nouns are formed in this very manner by adding ‘ka’ to the tiVe, ( root: e.g., ‘nammi-ka,’ confidence, from ‘ nammu,’ o confide ; and ‘kori- thoy, ka, hope, from ‘koru,’ to hope. This ‘ka’ is ‘kkei,” in Tamil (eg., (l\a, ‘nambi-kkei, confidence), and ‘ge’ or ‘ke, in Canarese : it is a very im pe].2 common formative of verbal nouns, and is equivalent in use to the for- they, ( matives of which d’ or ¢t ‘b’ or ¢p, is the initial. When we tiﬂpm f[)l'mat THE NEGATIVE VOICE. 373 ed by tive compare Telugu derivative nouns ending in ‘ka’ (eg., ‘teliyi-ka,’ ative semblance, from °teliyu,’ o appear) with the negative verbal parti- uarily ciples of the same language, which invariably end in ‘ka’ (eg., el by ‘ teliy-a-ka,” not seeming), it is evident that the particle ‘ka’ is not that The vowel ‘a’ ticle; by which the difference in meaning is expressed. which precedes ‘ka’ is evidently the seat of the difference. In those dlugu on, 8 cases in which the derivative noun and the negative participle are absolutely identical in sound and appearance, the negative ‘a’ has f the been absorbed by the preceding long ¢4’ of the root. This is the cause of the similarity between rika,” a coming, and ‘raka, not or without coming, the latter of which is for ‘ra-a-ka.’ nege of the In the dialect of the Kotas of the Nilgherry hills, ¢ p is usea as janner the formative suffix of the negative verbal participle instead of the dition Telugu ‘k’ and the Tamil-Canarese ‘d; e.g., ‘hogi-pe,” without enson going, corresponding to the Can. ‘hdgade, and the Tel. ¢ povaka.’ The Telugu prohibitive suffix ‘ku,” or more commonly ‘ka,” is, I nd to believe, identical with ‘ka,” the suffix of the verbal participle, just as 1ages, ialects the ¢d’ of “dé,’ the vulgar Tamil prohibitive, is identical with the ¢ d’ of “du,’ the negative verbal participle in the same dialect. urate, fixed Dravidian imperatives are in general nothing but verbal nouns pronounced emphatically. Hence, the Tamil ¢&ey(y)-a-dé, do not 5g&ti\'e 16 106 thou, is simply ¢sey(y)-A-du, doing not, with the addition of the em- phatic CA (&) :héy_a. and the Telugu ¢ chéy-a-ku,” or ‘chéy-a-ka,” do not thow, is in like manner, I conceive, identical with the verbal participle ‘chéy- ’-ﬂ'ka;’ hey-4- a-ka,” doing not, or without doing, with an emphasis understood. There is in classical Tamil a prokibitive particle which corres- ggative ponds to this Telugu prohibitive, viz., ‘arka; e.g., ‘Sey(y)-arka,” do not The Tamil prohibitive is used in connexion with both numbers ticiple and every gender; and I believe that it is by usage only that the cor- ¢ 10t0 responding Telugu form is restricted to the second person singular ; for when we compare the Tam. ‘Sey(y)-arka’ and the Tel. ¢chéy- for the aka,” we cannot doubt that they are identical. What is the origin nd the e 1l of this Tamil prohibitive suffix ‘arka? it is derived from “al’ (pro- nounced ‘ar’ before ‘k’), the particle of negation, and <ka, which Onsidef 1 maky is identical with ‘ka’ or ‘ga,’ a sign of the Tamil infinitive, opta- tive, or polite imperative, apparent in such words as ‘var-ga,” may (ke, t0 the thou, you, they, &c.) flourish. ~All verbal nouns in Malayalam end in kit | (€.g~9 ‘ka’ or ‘ga,” and each of those verbals is used also as a polite imperative ; e.g., ‘ wari-ka’ or ¢ wari-ga,’ is either a coming or mayest 3 very thou come, according to the context: so that the infinitival, par- he for ticipial, or imperative formative appears to have been originally the en e formative of a verbal noun. 374 THE VERB. We should here notice the prohibitive particle of the Gond, viz., of th gl ‘mani’ or ‘minni,’ This is not suffixed to the verb, but prefixed, like the Latin ‘noli.’ used ‘minni’ closely resembles the Tamil suffix ‘min,’ in such words as ‘fey(y)an-min,” do not ye: but the resem- Tami blance is purely accidental ; for the prohibitive particle of ‘Sey(y)- ol an-min’ is ‘an’ (euphonised from ¢al’), and ‘min’ is not, as often Beschi supposes, a prohibitive particle, but is a sign of the second and Tam person plural of the imperative, and as such is systematically used in the higher dialect by the imperative of the affirmative voice, as well the 4 as by the prohibitive; e.g., ‘poru-min,’ bear ye. Possibly the Gond (L prohibitive, ¢ mani,” is connected rather with the Hindustani ¢ mat’ and very the Sanscrit < ma,” and remotely with the Turkish particle of negation, comp “me’ or ‘ma; which is used like the Dravidian ‘a’ in the formation elem of the negative voice of the verb. ‘minni’ also closely resembles “inni,’ the probibitive particle of the Scythian tablets of Behistun. Tam congl Orwgin of ‘a’ the Dravidian megative particle.— We bave seen from that ‘a’ is the Dravidian sign of negation, and that it is inserted final between the theme and the signs of personality and other suffixes to Tami form the negative voice of the verb. Tam Is this ‘a’ connected with the ¢alpha privative’ of the Indo- Inur European tongues ? I think not, though this would be a more natural of 4} use of the ‘alpha privative’ than that of forming the temporal alreg augment in Sanscrit and Greek, according to Bopp’s theory. There appe is no trace of ‘alpha privative’ or any equivalent privative prefix in verh the Dravidian languages ; and its place is supplied by some post-fixed | relative participle or verbal noun formed from ¢il’ or ‘al; e.g., from from ‘nér, Tam., straight or straighiness, is formed ‘nér-inmei’ (“il-mei’ Prohi euphonised), crookedness, want of straightness. The The negative ‘a’ of the Dravidian negative verb is, I have no Fiup; doubt, softened from ‘al’ or ‘il the ordinary isolated particle of Simi negation. This very sign of negation is sometimes used by the Tamil classics instead of ‘a’ in“verbal combinations; e.g., ‘arig-il-ir, yow know not, takes the place of the more common ¢ari-(y)-ir:’ compare i also ¢ ninei-(y)-ala,” not considering ; Seyg-al-adar,” they wio will not thoge do or they will not do. In all these examples the ‘al’ is, I conceive, dil‘ec the negative particle ‘al,” not the ‘al’ of the verbal noun. There or a] cannot be any doubt whatever of the negative force of < al’ in the nega-~ lup| tive appellatives, which are formed from ‘al-an’ or ¢il-an,” %e is not, fory combined with verbal roots; c.g., ¢ pés-al-ém, we speak not, ‘und-il-ei,’ lord’ thow eatest not or hast not eaten. The Gond regularly forms its nega- Teyg) tive voice by suffixing ‘halle’ or <hille,” a barbarous euphonisation ppli — APPELLATIVES. 375 ind, viz, of the more correct ‘al’ or ¢il; and the dialect of the Kétas makes prefixed, a similar use of the particle <illa,’ This particle is also systematically nl suffix used in forming the prohibitive, or negative imperative, of the High 16 16500~ Tamil, in which connexion ‘al’ is ordinarily lengthened to ‘4l’ or eyl “ely eg., ‘Sel-€l,) go mot, < muni-(y)-€l,’ be not angry. But it is also 1ot as often retained unchanged ; e.g., “ sey(y)-ar-ka’ (‘ar’ for ‘al’), do not, o second and ‘sey(y)-an-min’ (‘an’ for ‘al’), do mot ye. In modern vulgar y used in Tamil ‘illei” (for “illa ’) is commonly sabjoined to the infinitive of o, a8 vell the affirmative verb to form an aoristic negative; e.g., ¢ vara~(v)-illei,’ he Gind (4, thou, ke, &ec.) did not, do not, or will not come. This form, though mat’ and very common, is not classical, and has arisen from the tendency which negation, compounds evince to break up in process of time into their component ormation elements. esembles “al’ or “il’ being the isolated particle of negation in the oldest lstun, Tamil dialect, and being still used in various verbal combinations, I conclude that ‘a’ the verbal sign of the negative, has been softened ave Seen from ¢al.’ Several parallel examples of the softening away of a inserted final ‘1’ can be adduced. ¢dal,’ the formative of many verbal nouns in iffixes t0 Tamil, has become ¢ ta’ in Canarese and Telugu ; e.g., “§ey-dal,” doing, Tam., is in Tel. ¢ché-ta; ¢muri-dal, Tam., breaking, is in Canarese e Tndo- ‘muri-ta.’ It will also be shown to be probable that ¢a,” the suffix o patural of the infinitive, has been weakened from another “al; and we have temporal already seen that ¢ mei,’ the Tamil suffix of the negative verbal noun - There appears to have been softened from ¢mal,’ the suffix of the negative preix in verbal participle. Whatever opinion we entertain respecting the derivation of ¢a’ st Axed 28 from from ‘al’ the widely extended affinities of €al’ <&l or €&l the [ il-mei’ prohibitive or negative imperative particle, are deserving of notice. The prohibitive particle of the Séantal, a Kol dialect, is €4l4; the have 10 Finnish prohibitive also is “4l4; the Ostiak ‘ila; and we find a similar prohibitive particle even in the Hebrew, viz., <al;’ Chaldee ¢14a.’ lrticle of o Tamil V. ArPPELLATIVE VERBS. b g In some languages of the Scythian group, verbal terminations, or : wlﬂ ot those pronominal fragments in which verbs terminate, are suffixed directly to nouns; which nouns become by that addition denominative concei"e} or appellative verbs, and are regularly conjugated through every | There number and person ; e.g., from the noun ¢ paz,’ the Lord, the Mordwin Jip neg” forms ‘paz-an,” I am the Lord; and from the possessive ¢ paz-an,’ o i3 10 Lord’s, it forms ‘paz-an-an,” I am the Lord’s. Adjectives being H']d.,‘i l'eif, merely nouns of quality in the Scythian languages, every rule which its 0§ g applies to nouns applies to adjectives also. In the New Persian, ml]’.?ﬂ — —— i 376 THE VERB. probably through the influence of the conterminous Scythian lan- fion w (not ¢ gnages, there is a similar compound of a noun or an adjective, with the verbal terminations; e.g., ‘merd-em, /-am a man, from ¢ merd,” @ (y)-4n, v man, and ‘em, the contracted form of the substantive verb / am. appeat The agreement between the Dravidian languages and those of the the Scythian family with respect to the formation of appellative verbs ance w of this character is complete. Any Dravidian noun and any adjective Th may be converted into a verb in the more ancient dialects of each oblige of the Dravidian languages, and in some connexions even in the persor colloquial dialects, by simply suffixing to it the usual promominal unders fragments: and not only may nouns in the nominative case be thus willage conjugated as verbs, but even the oblique case-basis, or old genitive, its a may in High Tamil, as in Mordwin, be adopted as a verbal theme. gender Tamil grammarians call these verbs ¢ vinei-(k)kurippu,” literally e'g ) fI' verbal signs,; and they have, not inappropriately, been styled conjugated f nal(l) nouns by an English writer on Tamil Grammar: but I think the best good; name is that which was given them by Beschi, viz., ‘appellative they (1 verbs.’ Th Appellative verbs are conjugated through every number and 0D £xa person, but they are restricted to the present tense ; or rather they are the ) of no tense, for the idea of time is excluded from them. nion Thus, from ¢kén,” Tam., ¢ king, may be formed ¢kén-én,’ 7 am a euphor king; < kon-ei,) thow art a king; ‘kon-ém,” we are kings; ‘kon-ir, ye Th are kings. So also we may annex to the crude base the oblique or third ] genitival formative ¢in, and then from the new constructive base verh, ‘kon-in,” of the king, or the king’s, we may not only form the appel- pial 1 lative nouns, ¢ kén-in-an,’ ke who is the king’s; © kon-in-ar,’ they who and ifs are the king’s (each of which may be used also as an appellative verb, two g when it signifies ke is the king’s, or they are the king’s); but we may either | also form the more distinctively verbal appellatives, ¢ kon-in-¢n,” 7 It am the king’s, ¢ kon-in-ém,” we are the king’s, &c. This use of the the for oblique or ‘inflexion’ as the basis of appellative verbs is a pecu- Person liarity of High Tamil; but the formation of appellative verbs from o dif the nominative or crude base of nouns is common to the whole Dra- Verbg vidian family. Thus, in Telugu (in which the vowel of the pro- WhiCh _: nominal termination varies by rule in accordance with the preceding vowel), from ‘sévakudu, @ servant, or ‘kavi, a poet, we form the Qi g appellative verbs sévakunda-nu,’ 1 am a servant; ‘kavi-ni, [ am a Torg § poet : “ sévakunda-vu,’ thow art a servant; ‘kavi-vi) thow art a poet. Uhigl, In the plural the Telugu has allowed the base of the noun (to which verh the pronominal terminations are affixed) to be pluralised, apparently It from having forgotten that the plural sign of the pronominal termina- Or oo APPELLATIVES 307 ) lan tion was sufficient of itself ; e.g., it says ‘sévakula-mu,’ we are servants with (not ¢sevakunda-mu’) ; whereas in Tamil the difference between ‘adi- o, (y)-én,’ I am (your) servant, and ¢ adi-(y)-ém,” we are (your) servants, i appears in the pronominal terminations alone; and the plan of denoting f the the plural which the Tamil has adopted is evidently more in accord- verbs ance with the true theory of the appellative verb Ithis lective The Telugu appellative verb is destitute of a third person f each obliged to be content with placing the isolated pronoun of the third in the person and the substantive noun in apposition, ith a substantive verb ominal understood vadu kavi e.4., he (is) a poet adi Gru,’ that (is) @ 0 thus village The Tamil is in this particular more highly developed ; for nifive, its appellative verbs are freely conjugated in the third person in each me. gender and number, by suffixing the final fragment of the pronoun terally e.g., from ‘nal, goodness or good, is formed ¢mnal(l)-an,” ke is good nal(l)-adu’ or nan-dru (for “nal-du’), it ¢s gMd ‘nal(l)-al, ske is good he best good ; nal(l)-ar, they (epicene) are good nal(l)-ana’ or ‘mnal(l)-a, llative they (neut.) are good. The neuter singular may appear to take a variety of forms ; but r and on examination those various forms will be found to be identical ; and 1oy 418 the apparent differences which exist are owing either to the euphonic union of the final ‘du with some previous consonant, or to its ] it @ euphonic reduplication The third person neuter, singular and plural (and occasionally the third person masculine and feminine also), of every species of Dravidian 1@w0F verb, is often used not only as a verb, but also as a verbal or partici- e b'l‘e pial noun Its primary use may have been that of a participial noun L but at all events the ]‘6/ (I]m and its use as a verb may be a secondary one two uses are found to be interchangeable ; e.g., irukkiradu means o veld e Ay either it is, or that which s, or the being, according to the context It is especially with relation to appellatives that this twofold use of 16y [ f the the forms of the third person must be borne in mind for in the third ) pecu person (singular and plural, masculine, feminine, and neuter) there is no difference whatever in spelling or pronunciation between appellative b from o Dra verbs and appellative noun and it is the contextalone that determines which meaning is the correct one Generally the appellative verb is he pio more commonly used in the Hich dialect, and the noun in the collo- el but to this there are exception i the quial dialect and (e.g.) ‘nalladu more frequently signifies in the colloquial dialect it is well than that ]ﬂmﬂ which is good ;—that is, it is used more frequently as an appellative poe verb than as an appellative noun )thh It is certain, however, that the appellative verb, whatever person areﬂ“y or gender it take is used more largely in the higher dialect of the e1ml 378 THE VERB. Tamil than in the lower ; and its brevity and compression render it peculiarly adapted for metaphorical use. dire Adjectives are formed into appellative verbs as well as nouns: but iy as the Drividian adjective is merely a noun of quality used adjecti- liter vally, the difference is more in terms than in reality : e.g., <oli-(y)-ei, (7 Tam., thow art bright, is literally thow art brightness ; and ¢ ini-(y)-ei,’ who thow art sweet, is thow art sweetness. Appellative verbs are formed appe from adjectives, or nouns of quality, not only in the cultivated Dravi- the: dian dialects, but even in the Ku, which is spoken by a barbarous 18 1 race ; e.g., ‘negg-anu,’ Ku, 7 am good, ‘negg-Amu,” we are good. adje When nouns of quality are used as the bases of appellative verbs or nouns they are generally adopted in their crude shape, as in the instances which have just been cited ; but in many cases we find the particle ‘iya’ intervening between the crude base and the pronominal termination or sign of gender ; e.g., ‘kod-iya-n’ (as a verb), ke ds cruel; Drdy (as a noun) one who is cruel, or a cruel man ; * val-iya-n,’ a strong man form or ke is strong, &e. This is the same particle which we have already fragl seen to be used as an adjectival formative; e.g., ‘val-iya,’ strong, A ‘ per-iya, great, ¢ sir-iya, little, &c., and I have stated that I conceive to ve words like these to be relative participles. 6l 1s identical with the 4 seti i’ of the past verbal participle, which is often used in Telugu as an of th adjectival formative without any addition ; and the final ‘a’ is the also sign of the relative, which is kept separate from ‘i’ by an euphonic ¢ \ Tely ‘lya’ is therefore the formative of the relative preterite participle, and bot} “val-i-(y)-a,’ strong, means properly that which was strong. But though tern the form of the preterite tense is employed, the signification (as add; often happens, especially in the case of relative participles) is aoristic 1 I | or without reference to time. This being the origin, as I conceive, formg of such forms as val-iya,” an appellative noun like ‘val-iya-n, a strong Inent man, 18 in reality a participial noun, signifying ke who is strong, and so ]&ngu of the other genders; and this explanation brings such forms into per- g fect harmony with other parts of the Drividian conjugational system, ing f] for participial nouns are regularly used in these languages as verbs. In some instances, ‘a,’ the sign of the relative participle, is dis- tong[ latjg pensed with, and the pronominal signs or signs of gender are elegantly suffixed to ‘1, the sign of the verbal participle, e.g., ¢ peri-du,” Tam., gy Vary it is great, or that whick is great, instead of peri-(y)-a-du.’ antj, On the other hand, in another class of instances, i’ disappears, of ¢} and ‘a’ alone remains. Words of this class, when deprived of their 1 signs of gender, are commonly called adjectives, and undoubtedly it is as adjectives that they are used ; but looking at their construction and bEil]g iﬂ [O% force I would term them relative participles of appellative verbs. iy PRONOMINAL SIGNS. 379 fer it In the words referred to, ‘a,’ the sign of the relative participle is directly annexed to crude substantive roots; e.g., ¢ udei-(y)-a, belong- : but ing to, more literally which is the property of. ‘malei~(y)-a, hlly, Jecti literally which is a Iill; “ti-y-a,” evil, literally which is evil. As ¢ udei- :y :"ei; (y)-an, considered as a noun, is certainly an appellative, signifying /e y )'ei; who owns, a proprietor, and as the same word is used poetically as an ormed appellative verb, when it signifies /e is the owner ; it seems evident that Drév- the proper light in which to regard ‘udei-(y)-a’ (and every similar word) is to consider it as the relative participle of an appellative verb used ?&IOUS adjectivally. verbs n the Conjugational System. d the minal MopE oF ANNEXING PronomiNAL Sieys.—The persons of the Dréavidian verb, including the related ideas of gender and number, are el [ fman formed by suffixing the personal or demonstrative pronouns or their ready fragmentary terminations, to the signs of tense. fromfy The change which the pronouns undergo when they are appended celve to verbs as signs of personality have already been exhibited in the section on ¢ The Pronoun.” They consist chiefly in the softening away h the a8 an of the initial consonant ; but in a few instances the final consonant has is the also been softened away, and nothing left but the included vowel. In Telugu, ‘ni-vu, the pronoun of the second person singular, has lost iy, ¢, and both its radical initial and its formative final ; and in the personal terminations of the verb it is represented only by ¢vu,” an euphonic hough addition. aristic In the Indo-European languages the personal signs of the verb are formed by suffixing pronominal fragments to the root ; and those frag- ceIve, hrong ments are disguised in a still greater degree than in the Dravidian languages, not only by frequency of use and rapidity of enunciation, 0d 80 but also by the love of fusing words and particles together, and form- 0 PEI- ystem: ing them into euphonious componnds, which distinguishes that family of tongues. Sometimes one dialect alone furnishes the key to the expla- b is die- nation of the inflexional forms which are apparent in all. Thus, the origin of ‘unt’ or ‘ant,” the sign of the third person plural in the gantly various Indo-European languages (e.g., fer-unt, ¢ep-ovr,’ ¢ bhar- Tam, anti, &e.) is found in the Welsh alone, in which ¢ hwynt’ is a pronoun of the third person plural. .peal's! their The various changes which the Dravidian pronouns undergo on being used as the pronominal signs of verbs have already been stated yit I in order. In Telugu, and partly also in Canarese, the pronominal ter- I il minations vary according to the tense ; but this arises from the opera- 380 THE VERB. 1t 18 tion of the law of harmonic sequences (see the section on ‘Sounds’), aalle by which a vowel is affected by a preceding vowel, and changed so as sign to harmonise with it. What requires here to be investigated is simply fime the mode in which the pronominal signs are attached to the Dravidian othe verb. retal (1.) The pronominal signs of the Dravidian verb, like those of the but primitive Indo-European and Scythian languages, are suffixed, not pre- Thu fixed. In the modern Indo-European vernaculars, most of the verbs dids have lost their old pronominal terminations, and the pronouns which simi ave used as nominatives to verbs are usually isolated and placed first. beate Thus, instead of love-Z, in accordance with the ancient ‘am-o,” we have L beg learnt to say 7 love,—an alteration of position which produces no 1 change in meaning. they In the Semitic languages a change in the position of the pronoun barb from the termination of the verb to its commencement produces an important change in grammatical signification : the position of the liar]) ( pronouns or pronominal fragments determines the tense. When the i w prouominal fragments are prefixed, the tense of the verb is regarded as 0ons] future or aoristic: it is regarded as past when they are suffixed. Pre- fixing the pronominal fragments denotes that the action of the verb Euro] the r has, as yet, only a subjective existence in the mind of the speaker or lows agent,— ¢.e., it is future ; suffixing them denotes that the action of the verb has already acquired an objective existence, apart from the will 18 00 by | or wish of the speaker or agent, .c., it is past. 100t No peculiarity of this kind characterises the Dravidian languages: the tenses are formed, not by means of the position of the pronouns, and | it 1 but by particles or signs of present, past, and future time suffixed to the theme ; and the personal signs, as in the Turkish and Finnish § families, are suffixed to the signs of tense. The only exception to this chang Whic} rule is that which forms the most characteristic feature of the Malay- alam—a language which appears to have been directly derived from inry, with the Tamil,—but which, in so far as its conjugational system is con- cerned, has relapsed into a condition nearly resembling that of the Sing Mongolian, the Manchu, and other rude primitive tongues of High Place tenge Asia. In ancient times, as may be gathered from Malayéala poetry, and especially from inscriptions preserved by the Syrian Christiansand Iy ey the Jews, the pronouns were suffixed to the Malayala verb, precisely as they still are in Tamil. At present, the verb is entirely divested, bingy at least in the colloquial dialect, of signs of personality ; and with the the T pronouns, the signs of number and gender also have necessarily dis- appeared ; so that the pronoun or nominative must in every instance Drgy be separately prefixed to the verb to complete the signification, and f&mﬂ} PRONOMINAL SIGNS. 381 it is chiefly by means of this prefixed pronoun that a verb, properly so s, called, is distinguished from a verbal participle. Though the personal 80 48 mp: signs have been abandoned by the Malayéla verb, the signs of tense or ly time have been retained, and are annexed directly to the root as in the idian other dialects. Even in modern English some persons of the verb retain archaic fragments of the pronominal signs (e.g., lovest, loveth) ; f the but in the MalayAlam every trace of those signs has disappeared. b pre verbs Thus, whilst we would say in Tamil ‘adittén,” 7 beat ; ‘adittay,” thou didst beat; ‘adittan,” ke beat ; the Malayalam uses in these and all which similar cases the verbal participle ¢atichu’ (for ¢adittu’), kaving finst beaten, with the prefixed pronouns I, thou, ke, &c. : e.g., ‘ fjan atichu,’ have I beat ; “ni atichu,” thou didst beat ; ‘avan atichu, ke beat. 68 10 Though the pronominal signs have been lost by the Malayéla verb, they have been retained even by the Tuda ; and notwithstanding the Onoun barbarity of the Gonds and Kus, their conjugational system is pecu- 8 anl liarly elaborate and complete. f the (2.) Another and distinctively Scythian peculiarity in the manner n tne in which the personal signs are suffixed in the Dravidian languages ded a8 consists in their annexation, not directly to the root, as in the Indo- Pre The first suffix to European family, but to the temporal participles. verb the root in the affirmative voice is that of the sign of tense, then fol- er or lows the suffix of personality. Every pure Dravidian affirmative verb of the is compounded of three elements, which are thus arranged and named g will by Tamil grammarians, viz., (i.) the ‘pagudi’ (¢ pracriti,’ Sans.) or root ; (ii.) the ‘idei nilei, or medial particle, i.e., the sign of tense ; 1ages and (iii.) the ‘vigudi’ (¢ vieriti, Sans.), the variation or diferentia, HOUNSy t.e., the pronominal termination, \ed 10 When the signs of tense are attached to the theme, some euphonie inpish changes take place (not in the theme, but in the signs themselves), 0 this which serve, as has been shown, to distinguish transitive verbs from [al ay intransitives. Other euphonic changes also take place in accordance f Tom with Drividian laws of sound, which will be inquired into when those § C00 signs of tense are one by one examined. The changes which take ()f the place in the pronominal signs when they are annexed to the signs of High tense have already been stated in the section on the ‘Pronoun.’ 0ellYs In the Indo-European languages we meet with no instance of the ncﬂﬂd annexation of the pronominal signs to the participles, .., to the com- (1564 bination of the root with the signs of tense. We have no instance of asted the use of any form like ¢amant-o,” instead of ‘am-o,’” to signify 7 love. h the This, however, is the method which is invariably employed in the g dis Dréavidian languages, and which constitutes an essential element in the It is also distinetive of tance family likeness by which they are pervaded. } qnll THE VERB. 382 the Turkish. Thus, the Turkish ‘ollirsen,’” thou art, is formed from sing the ¢6llir,” being, the present participle of the verb €6l ¢o be, with the So also the Oriental Turkish y addition of the pronoun ‘sen,’ thou. ‘bdlamen,” 7 am, is formed from ¢bola,” being (theme ¢bol,” to be), and whi T the pronominal suffix ‘men,” 7 pret; An important difference which is generally found to exist between verb the Dravidian langnages and the North Indian vernaculars should here Heb be stated. In the languages of Northern India the present tense of a verb is ordinarily formed by annexing the substantive verb to its nout the present participle, e.g., ‘karitechi,” Beng. (‘karite-achi’), 7 am doing, instead of 7 do. ¢4, In Telugu, probably through the influence of the North-Indian subst vernaculars, a similar usage prevails; but it is found in the present Com tense only, it may readily be dispensed with, and the simpler usage, “kan Iom which accords with that of all the other Dravidian dialects, is un- In Tamil and Canarese this use of the { { doubtedly the more ancient. substantive verb, as an auxiliary in the formation of the present tense, 108 ‘( whet is unknown : it is used as an auxiliary only in the formation of the compound preterite and future tenses. ing d The Malayalam occasionally uses the substantive verb in a similar Thus, tions manner to the Telugu, but with a somewhat different signification. In Telugu ‘naduchutunnénu,’ I walk (from ‘naduchu-tu,’ walking, and Yary and y ‘unnénu,’ 7 am), has simply the meaning of the present tense, and is equivalent to the simpler form ‘naduchutinu,’” answering to the Tamil and ‘nadakkirén,” and the Canarese ‘nadeyutténe; but in Malayalam, F whilst ‘fijan natakkunnu,” means 7 walk, ¢fijain natakkunnunta’ has generally a progressive sense, e.g., I am walking or continuing to walk. fofm& (3.) Tt is a peculiarity of the Telugu that the third person of the tiCipk preterite is sometimes left altogether destitute of the signs of time, person, tenses number, and gender; and this peculiarity applies also to the third (calle Thus, whilst ‘unditini, 7 was, and ¢unditivi,’ the Te person of the aorist. thou wast, are supplied with the usual signs of tense and person, the this 8 third person of the same tense is simply ‘uude-nu,’ ke, she, or it was, or and they were, without distinction of number or gender, and without even fengeg the particle <ti, which constitutes the usual sign of the preterite. Datj The aorist third person, with a similar absence of distinction, is ‘undu- from ¢ nu ; and in both cases the final ‘nu’ is merely a conjunctive suffix, Myge] like the corresponding Tamil ‘um.” Sometimes even the aorist for- of this mative ¢ nu’ is discarded, and the root alone is used as the third person singular. Thus (e, she, or wt) falls or will fall, may either be ‘padu- Ve nu,’ or simply ‘padu.’ The usage of poetical Tamil occasionally agrees ang in with that of the Telugu with respect to the neuter gender, both tense; FORMATION OF TENSES. 383 from singular and plural, especially in connexion with the negative voice of ) the the verb : e.g., Sey(y)-4," it will not do, is often used for ‘sey(y)-adu.’ rkish A usage similar to that of the Telugu prevails in many languages ) and which are widely different one from the other. Thus, the New Persian uses for the third person singular of the ween preterite the contracted infinitive, as grammarians style it—an abstract 1 here verbal noun, which may be regarded as the theme of the verb. The of 9 Hebrew third person masculine of the preterite tense is also a verbal to Ifs noun, without pronominal addition. We see a similar peculiarity in Jotng, the third person of the present tense of the verb in some languages ; e.g., compare the three persons of the present tense of the Turkish ndian substantive verb, ¢o6lirum,” 7 am ; ¢oltrsen, thow art; ‘6lir,” he is. resent Compare also the Armorican ‘kanann,” 7 sing; ¢kanez,” thou singest ; “kan, he sings. Compare with these examples the Hungarian 1sage, Iy - ‘ismerek,’ I know ; ‘ismersz, thow knowest ; and ©ismer,” he knows. (4.) The Dréavidian verb, like that of many other languages, does of the fense, not distinguish the genders of either the first person or the second, of the whether singular or plural ; but in the third person it marks all exist- ing distinctions of gender with peculiar explicitness and minuteness. milar Thus, without the use of isolated pronouns, and employing the inflec- tions of the verb alone, we can say in Tamil ¢ varugiran,” he comes : D, In ‘ varugiral,’ she comes ; ‘varugiradu,’ it comes; ¢ varugirér,’ they (men j, and and and women) come, or honorifically ke comes ; ¢ varugirirgal,’ they (men Tamil and women) come ; ¢ varugindrana,’ they (things) come. yélam; ForMATION OF THE TENSES.—Most of the Drividian tenses are a’ has ) w!]lk. formed from participial forms of the verb; an inquiry into the par- ticiples is therefore a necessary preliminary to an inquiry into the of the tenses. Dravidian verbs have two species of participles, one of which yersoly third (called relative participles, because they include the signification of the relative pronoun), will be inquired into in a subsequent part of @iﬁvi; this section ; the other, commonly called werbal participles or gerunds, oD, the and which are now to be considered, constitute the bases on which the s, OF tenses are formed. The forms which are assumed by the verbal j evel etel'lte participles will be inquired into in connexion with the signs of tense, from the consideration of which they cannot be severed. I content ‘andr myself here with some general remarks on the signification and force ‘ Sllfﬁx', of this class of words. g for- pel‘SOﬂ Verbal participles: their signification and force.—In ordinary Tamil ( pgl('ill‘ and in Malayalam, there is but one verbal participle, that of the past 3%‘1’963 tense; in all the other dialects there is a verbal participle of the , both < © 384 THE VERB. present tense as well as of the past. In this particular, therefore, the g Tamil (with its daughter, the Malayalam) may be considered as the of e poorest of the Dravidian dialects. It partly makes compensation for I the deficiency by the use in the classical idiom of a verbal participle a i of the future, which none of its sister dialects possesses. Even the 8 us classical idiom, however. is destitute of a present verbal participle. that Properly speaking, the, words which are called wverbal participles are prete not participles at all, seeing that they do not participate in the nature of ti of adjectives, as all the Indo-Kuropean participles do. They have used somewhat of the signification of gerunds, inasmuch as in addition to main the idea of time, they include more or less of the idea of cause. apre Nevertheless, as each of the Indo-European participles is commonly tense used also as a gerund, without losing the name of a participle, and as every ‘the gerund in do’ (to which alone, amongst Latin gerunds, the 0 Dravidian participles have any resemblance) has a very restricted T application, it appears advisable after all, to style these words par- ot ¢ ticiples instead of gerunds,—or more fully wverbal participles, to dis- done, tinguish them from what are called relative participles. One ¢ The following sentences will illustrate the force of the Dravidian that verbal participles. use ¢ (1.) Present verbal participle.—This verbal participle is unknown ad p lgyer in Tamil and Malayalam; but is commonly used both in Canarese and in Telugu. I quote the illustration which follows from the Canarese. which and g ‘Vikramdrka, punishing the wicked and protecting the good, reigned over the kingdom.’ Here the English words punishing and protecting, are Whet] <ontip participles of the present tense, used gerundially; and the Dravidian words which they represent, (in Canarese, sikshisutta’ and ¢ rakshi- leed suttd) have precisely the same force. In this respect only there is a (and ; Qecisg difference between them, viz., that the English participles are capable of being used also as adjectives, whereas the Dravidian words, though Proper tlags { called participles, cannot be used adjectivally, or in any other way than that here exemplified. defet lete (2.) Preterite verbal participle.— Saliwahana, having killed ‘Vikra- marka, assumed supreme power.” Though the English participle Zaving l killed which is here used, is a compound one (being formed from the tense ( present participle having, and the passive participle killed), its signi- Sign ¢ fication is that of a simple, uncompounded participle of the past tense, & eupt and the Dravidian word which it represents (‘ kondru,” Tam, ¢ kondu,’ The Can.) is also a preterite active verbal participle. In this instance, 1, neither the English participle nor the Dravidian one is capable of ing fhe being used as an adjective. In reality, they are both preterite gerunds of the { THE PRESENT TENSE. 385 e, the the or gerundials, though they retain the name of participles as a matter of convenience. on for In those Drividian dialects in which there is a present, as well as aiple 1 the a preterite, verbal participle (as in Canarese and Telugu), the present is used to express subordinate actions which are contemporaneous with ciple. that which is denoted by the principal and finite verb; whilst the % a8 preterite expresses subordinate actions which are antecedent in point 1atire of time to the principal action. In Tamil, the preterite participle is have used to express all subordinate actions, whether simultaneous with the ion to main action or antecedent to it; but though that participle is always Gause: a preterite in form, it possesses the force of a partieiple of the present monly v tense, when the connexion requires it. In each of the dialects and in nd a8 every connexion, the nominative of the final governing verb is the 5, the nominative of all the subordinate verbal participles. icted The Dravidian verbal participles may be eompared with the Sans- § pat- crit ‘indeterminate past participle’ in ‘tva; e.g., ‘krutva, hawving o dise done. Like that participle they are indeclinable and indeterminate, One of the chief peculiarities, however, of these verbal participles is, yidiant that they have a continuative force, dispensing altogether with the use of eonjunctions. In the Dravidian languages, though nouns and pronouns are united by means of conjunctions, finite verbs are known never so united. In ewery sentence there is but one finite verb, se and which is the last word in the sentence, and the seat of government; 1a16se. and all the verbs which express subordinate actions or circumstanees, o 00’ whether antecedent or comtemporaneous, assume an indeterminate, i, are continuative character, as verbal participles or gerundials, without the yidian need of eonjunctions or copulatives of any kind; so that the sense akshi- (and in Tamil the time also) waits in suspense for the anthoritative e 54 decision of the final governing verb. Hence those participles might gpable properly be called continuative gerundials. Tamilian grammarians hough class them, with infinitives and subjunctives, as ‘vinei echam,” verd- 7 Wiy defects, or verbal complements, i.c., words which require a verb to com- plete the sense. Jikra 1. The present tense.—It may be stated generally that the present hawin] tense of the Dréavidian verb is formed by suffixing the pronominal § ngﬂi’ signs to the present verbal participle, with such trivial changes only as euphony requires. fpIEEy The exceptions to this general rule are as follows :— koﬂdn,’ (1.) In poetical Tamil the tenses are sometimes formed by suffix- stances ing the pronominal terminations to the relative participles, instead ble of of the gerunds or verbal participles; e.g., ‘nadanda(n)an’ (equivalent :eruﬂds 2 ¢ BRSNS 386 THE VERB. to the colloquial ‘nadanda(v)an’), he walked, literally a man who walked. TIn such instances a verbal or participial noun is used with ing Vi the force of a verb. This is not an uncommon usage in other lan- guages also; and in colloquial Tamil the third person neuter of the dia verb, both singular and plural, is certainly a verbal noun in its origin, for though used with the force of a verb; eg., ¢nadandadu,’ i walked, literally means a thing which walked ; and the plural ¢nadanda(n)a,’ i Bl means literally things which walked. The peculiarity of the poetical dialect is the extension of this usage to each person of the verb; e.g., I8 nal ¢ nadanda(n)en,” I walked, literally I who walked; * nadanda(n)am’ or der ‘ nadanda(n)em, we walked, literally we who walked. This mode of forming the tenses has been developed from the the hay Dravidian custom of using participial and verbal nouns as the conjuga- tional bases of verbs, and, so far, is in accordance with the genius of the iz the language; but it has a constructive, artificial look, and it is an por exception to the mode which prevails throughout all the other dialects (1 of the family, whether colloquial or classical. ad (2.) The Tamil has, properly speaking, no present verbal par- ticiple, but only a particle denoting present time, which is suffixed to 101 the theme of the verb, and to which the pronominal signs are then Int suftixed for the purpose of forming the present tense. The combina- isp tion, however, of the root and the particle of present time, forms 0011 I think it may, therefore, be assumed ch virtually a present participle. that the Tamil had a verbal participle of the present tense at a former “ta period, which bas now become obsolete, except in combination with orj the personal terminations, when it constitutes the present tense of the Ingt % verb. the (3.) The sign of the present tense in the Ancient Canarese verb, sam is altogether unconnected with the formative of the present verbal participle. The present verbal participle of the ancient dialect is of | identical with that of the modern one, the temporal sign of which is derg <tta’ or ‘utta, whilst ‘dap’ or ¢p’ is the sign of the present tense this of the verb in the ancient dialect; e.g., ¢ baldapen’ (‘bal-dap-en ), I To be live. (4.) The Telugu usage of employing the substantive verb in a (ta, modified form (viz., ‘unnanu,’ 7 am, ¢unnavu,’ thow art, &c.) as an ig] auxiliary in the formation of the present tense, can scarcely be called Sang an exception to the general rule specified above; for this auxiliary is softe annexed to the present verbal participle, which is closely allied to diffe that of the Canarese; and its use in this connexion is only a refine- 10t ¢ ment of the grammarians, not a necessary element in the formation of the present tense. 0 ¢ THE PRESENT TENSE. 387 who These real or apparent exceptions being disposed of, it remains to with inquire into the formation of the present verbal participles in the lan various dialects. f the ForMmATION OF THE PRESENT.— In both the ancient and the modern rigin, dialect of the Canarese the verbal participle of the present tense is lked, formed by suffixing to the verbal theme, ‘ uta,” ‘ute,” ‘utta,” ¢ utte,” or (n)a, “uttd; e.g., < bal-uta,’ living; 6d-ute,” reading; °ond-utta,’ joining; atical “ili-(y)-utte,” descending ; ¢ mad-utta,” doing. Of these particles ‘utta’ ) 6y is most commonly used. Probably this particle had but one ‘t’ origi- ' or nally; and it seems also probable that the initial ‘u’ is euphonic, and derived from the final euphonic ‘u of the majority of the verbal n the themes. The primitive form of this particle would therefore appear to have been “ta’ or ‘te. 1juge- The final vowel, “a’ or ¢e,” is elided before mius of the initial vowel of the pronominal signs, or rather perhaps, incor- i3 an porated with it; e.g., ‘balutténe’ (‘bal-utt’-éne’), I live; ‘balutti’ falects (¢ bal’-utt’-1’), trow livest. The present verbal participle of the Telugu is ordinarily formed by | par- adding ‘chu’ (pronounced ‘tsu’) to the theme of the verb. Occa- xed 10 sionally ‘ka’ is used to form the present participle instead of ¢chu.’ o then In the colloquial dialect ‘tu’ is used instead of ¢chu; and though it mbind- is possible that ‘chu’ may be the original, and “tu’ (from ‘tsu’) the foms corruption, yet it would be more in accordance with analogy to derive “chu’ from ‘tu; and this ‘tu’ so nearly resembles the Canarese ssumed former “ta’ or ‘te, that we may safely conclude both forms to have been originally identical. Probably also ¢du,” the particle which in most instances is inserted as a sign of tense between the verbal theme and the pronominal terminations of the Telugu aorist, springs from the same origin as ‘tu. eV erb, : verbal ¢chunnu’ or ‘tunnu,’ the ordinary termination of the participle alect iS of the present temse in grammatical Telugu, is a compound form derived from ¢chu’ or ‘tu,’ the real and only sign of present time in shich i8 ot fense this language, and unnu,” a participle of the substantive verb ‘undu,’ to be, used as an auxiliary. 9-6[1 )y I I cannot offer any opinion respecting the origin of the use of ¢tu,’ ‘ta, or ‘te,” as a sign of present time in Telugu-Canarese. We h i 8 might propose to compare it with at’ or ‘t, the formative of the b )as an ‘at,” however, is e callefi Sanscrit present participle, e.g., jayat,’ conquering. softened from ‘ant’ or ‘nt,’ the affinities of which lie in a widely iliary 8 aﬂled 0 different direction ; besides which, this form is used only as a participle, not also as a gerund. i refine We might also compare the Telugu-Canarese formative with “te’ eratioIl or ¢ite, the formative of the Bengali present participle, e.g., ‘karite,’ 2%c 2 THE VERB. 388 doing; but this form is identical with that of the infinitive in the n same language, and ‘te’ or ‘ite’ has been supposed, with reason, to ‘} be the dative or locative of a verbal noun. ‘ka,” the secondary and less common formative of the Telugu present, may perhaps have been th derived from ‘gir’ or ‘kir, the sign of the present tense in Tamil. fix The Tulu sign of the participle of the present tense is ¢v,” which The is identical with the Tamil-Canarese sign of the future or aorist. an present participle is formed in Ku by suffixing ‘i’ or ‘pi; and in M Gond by suffixing ¢i,” “si’ (properly signs of the preterite), or ‘kun,’ JilR which is identical with the Malayéla ¢ kunnu.’ o1 The sign of present time used by the Tamil and Malayalam, differs One considerably from that of the Telugu-Canarese. alay The present tense in Tamil is formed by suffixing ¢gir-u,’ Sl ‘ gindr-u,” or ¢ anindr-u,” to the verbal theme, to one or other of which o ¢ Anindr-u’ is a com- particles the pronominal signs are annexed. pound form, which is rarely used even by the poets, and is derived, 1 L m conceive, from ¢4, the ultimate base of ‘A-gu,” fo be or become (and soft which is not unfrequently used in this shape in the poets), and Cum ‘nindr-u,’ standing, abiding. The other particles of present time, 80 § ‘gir-u’ and °‘gindr-u,” are in common use, especially the former ; e e.g., ‘varu-gir-an’ or ¢varu-gindr-an,’ %e comes. The only difference and between them is that ¢ gindr-u’ is considered more euphonious and elegant than ¢gir-u,” and more suitable in consequence for poetry and GO elevated prose. I have no doubt that they are identical in origin, and naty that the one is merely an euphonised form of the other. In some is be connexions ‘gir-u’ and ‘gindr-u’ are changed by dialectic rules of Sents euphony to ‘kkir-u’ and ¢kkindr-u,” viz., when they are attached lu31 to roots consisting of two short syllables (like ¢ padw’, to lie, ©iru,’ to Soun( be, ‘nada, to walk) the final vowel of which is regarded as a part of the root, and is incapable of being elided. It 1s a rule of the T of th language that if in such cases the sonants ‘g’ ‘d,” ‘b, immediately follow, they shall be hardened, that is, converted into the correspond- Prong in ) ing surds ‘k, ‘t) and ‘p; and in Tamil the only method of harden- &g, ing sonants is by doubling them,—for it has already been shown that in this language the same consonant is a sonant when single and a ﬂle ( klup, surd when doubled. Hence we say in Tamil not ¢iru-gir-én,’ I am, but ‘iru-kkir-én.’ A similar result follows in another and more 4 It has been shown numerous class of instances from a different cause. or ‘k Plesey in a former part of this section that transitive or active verbs are in many instances made to differ from intransitives by the hardening In such or “k and doubling of the initial consonant of the sign of tense. cases ‘gir-u’ and ‘gindr-n’ become (not for the sake of euphony Whigh THE PRESENT TENSE. 389 16 1 merely, but as a means of grammatical expression) ‘kkir-u’ and ‘kkindr-u.’ i The Malayalam uses the same sign of tense somewhat modified : 1l the sign of present time in Malayilam is ‘unnu’ or ‘kkunnu,’ suf- fixed to the verbal theme. oh he Where the Tamil would use ¢gindru,” the Malayalam has ¢ unnu; and where the Tamil has ‘kkindru,’ there ¢ kkunnu’ is used by the 1 Malayalam. The Malayala particle is clearly a softened and eupho- 1, nised form of the Tamil one. The Tamil compound sound ° ndr’ 18 constantly converted into ‘nn’ in Malayalam ; e.g., < ondru, Tam., ol one, is in Malayalam ¢onna,” and ‘m{ndru,” Tam., three, is in Malay- alam ‘munna.’ Even in vulgar colloquial Tamil the same or a 1l ch similar tendency appears; ‘ondru, one, being commonly pronounced ‘onnu,” and ‘mundru,’ three, ¢ munu.’ The Tam. ‘gindru’ and IS ‘kkindru’ would, therefore, naturally and dialectically be converted i in Malayalam to °ginnu’ and ‘kkinnu.’ The next point is the nd softening away of the ‘g’ of ¢ginnu.’ This has arisen from the cir- id ? cumstance that in Tamil ‘g is pronounced in the middle of a word ne, so softly as to be little more than an indistinet, guttural breathing: )1’ in consequence of which it is used to represent the ‘h’ of the Sansecrit, Nce and in the colloquial dialect it is often discarded altogether; e.g., nd ‘ pogirén,’” I go, is commonly pronounced ¢ pé-rRén; and ¢ varugiran,’ e nd comes, ¢ vari-ran’ or ¢ va-Ran.’ Hence ‘ ginnu’ (from ¢ gindru’) would nd naturally become in Malayalam ¢innu.’ The only remaining difference me is between the ‘i’ of ‘innu’ and the ‘u’ of ‘unnu; and this pre- g of sents no difficulty, for even in Tamil ¢i’ is very often pronounced as hed bl ‘u’ by the vulgar, and the ‘u of the Malayala ‘unnu’ is a middle "t sound between €1 and ‘u.’ qart The identity of the Malayéla sign of the present tense with that the of the Tamil, cannot be doubted. Sometimes in Malayéla poetry the ely pronominal signs are suffixed to the signs of tense, as in Tamil; and nd: in that connexion the identity of the signs of tense is clearly apparent ; Jen e.g., compare ‘adikkindran’ (‘adi-kkindr-an’), Tam., ke beats, with that the corresponding form in poetic Malayalam, ¢atikkunnan’ (“ati- id 8 kkunn’-4n”). (lilly A priort it might have been supposed that the Malayala ¢unnu’ jore or ‘kkunnu’ was related to ¢chunnu’ or ‘tunnu,’ the sign of the Wi present participle in Telugu. The resemblance, however, is altogether eill illusory ; for the Malayala particle is derived from the Tamil ¢ gindru’ ing or ‘kkindru,” whilst the Telugu ¢chunnu’ is compounded of ¢chu, uch the real sign of present time, and ‘unnu,” a participle of ‘undu,’ Zo be; LOH)’ which participle is in Malayalam ¢unta.’ 390 THE VERB. Rl I have said that T believe the Tamil ¢gir-u’ and ‘gindru’ are identical in origin, and that the one is merely an euphonised form of the other. I have no doubt that ¢ gindr-u’ is the secondary form, and that it has been derived from ¢ gir-u.’ There are many instances of words ending in ‘rRu’ converted euphonically into ¢ ndru,” of which one will suffice as an example, viz., ‘miru,” Can., three, which has been converted into ¢ mindru,” Tam., but per contra there is no instance extant of ‘ndr’ being simplified into ‘r’ or ‘rR; and the fondness for nasal sounds which is inherent in the Dravidian languages, forbids If this the supposition of any such change ever having taken place. view of the case is correct, it detracts somewhat from the claim of the Malayalam to high antiquity, for it proves that it was subsequent to the change of ‘kkir-u’ to ¢kkindr-u,’ te., subsequent to the com- meneement of the phonetic refinement of the Tamil language, that the Malayalam acquired a separate existence and a distinet place of its own amongst the Drividian dialects. The origin and ulterior relationship of ‘gir,” as a sign of present time, is as completely enveloped in mystery as that of the correspond- ing Telugu-Canarese ‘chu,” ‘tu,” ‘ta,” ‘te. I notice (but it is scarcely deserving of notice) the slight resem- blance in sound between °gindr-u’ and ‘ant,’ ‘ent ‘and,” ‘ende, &c., the formative of the Indo-European present participle, of which sometimes the nasal is discarded, as in the Sanscrit ¢ jayat,” conquer- ing, and sometimes the dental, as in the English singing and the Seotch sengin. No greater importance is to be attributed to the slight resemblance of ‘gir’ to ‘our’ or ‘ur, the formative of the present participle in Turkish ; for I have no doubt that this ‘ur’ is derived from © dur,’ est, the impersonal substantive verb. 2. The preterite tense.—The mode in which a language forms its preterite, constitutes one of the most distinctive features in its gram- matical character, and one which materially contributes to the deter- mination of the question of its relationship. In the Semitic languages past time, or the objective reality of past events, is denoted by placing the verbal theme first, and suffix- ing to it the sign of the personal agent. In the primitive Iudo- European languages the preterite appears to have been most commonly formed by means of the reduplication of the root or verbal theme; but this reduplication has in many instances been so softened and euphonised, that it has dwindled into the mere use of a different vowel in the preterite from that which forms part of the root. The Indo- d — ~ THE PRETERITE TENSE. 391 16 Europeé,n preterite was also frequently formed by means of a prefixed of temporal augment ; a prefix which Bopp considers to be identical with nd “alpha privative, but which is supposed, with greater probability, by of Meyer, to be identical with ‘a,’ a relic of the auxiliary verb fo have, ¢h which is still prefixed to verbs in the Celtic languages as a temporal 12$ augment, 7.c., as a sign of past time. 106 In a large proportion of the verbs in the Germanic tongues, in the eSS Modern Persian, in the Turkish and Finnish families of languages, ids in the vernacular languages of Northern India, and, with a few excep- his tions, in the Drividian languages, the preterite is formed by suflixing he to the verbal theme a particle, gemerally a single consonant only, which is significant of past time. to 1 The Dravidian preterite tense is ordinarily formed, like the pre- he sent, by annexing the pronominal signs to the preterite verbal par- 1ts ticiple. It is in that participle that the idea of past time resides: by it alone that idea is expressed: the changes that are made when the pronominal signs are added, will be shown to be euphonic merely, not o1 1 structural ; and in the Malayélam (in which the pronominal signs are not annexed), that part of speech which corresponds to the Tamil m preterite verbal participle, expresses by itself the past tense of the le verb. Consequently an inquiry into the Dravidian preterite tense, resolves itself into an inquiry into the formation of the preterite ch o verbal participle. The preterite verbal participle is used in Tamil with a wider the range of signification than in any other dialect, though its proper and inherent meaning is that of the preterite alone. The Tamil being 106 n destitute of a present verbal participle, uses the preterite verbal ésl, participle instead ; in consequence of which, in a Tamil sentence, the question of time is in abeyance till it is determined by the tense of the final governing verb. This statement applies to the verbal participle alone, not also to the preterite tense of the finite verb; 1t8 )i which is restricted in Tamil to the expression of past time, precisely as in the other dialects. el We have now to inquire particularly into the Dravidian methods of of forming the preterite. They divide themselves into two—(i.) by reduplication of the final consonant; and (ii.) by suffixing a sign X do of past time, nly (1.) THE FORMATION OF THE PRETERITE BY REDUPLICATION OF 16, e Finan CoxsoNanT.—This mode of forming the preterite is adopted ad by a very small number of verbs in each of the Dravidian dialects; but el its existence cannot be doubted, and it is a mode which is as interesting o as it is remarkable. In the Indo-European languages, when the pre- 392 THE VERB. Al terite is formed by means of reduplication, it is the root which is doubled, or at least the first syllable of the root; but in the Dravidian of The dialects the reduplication is that of the final consonant alone L4 verbal themes which form their preterites in this manner are those Ta which end in ¢d-u g-u, or ‘Rr-u,” preceded by a single short vowel the e.g., in Tamil i pad-u,” fo suffer pug-u,’ fo enter; and ‘per-u, fo obtain, the preterites of which are ‘patt-én, 7 suffered; ¢pukk-én,’ 7 T entered; and ‘ pettr-én,” 7 obtained In each of the above examples plo the final consonants—‘d,” ‘g’ and ‘®r’—are doubled, and being thus JUR doubled, are converted by rule into the corresponding surds ‘tt, ¢ kk,’ and ‘RR’ (pronounced °ttr’) Whilst the above and similar verbs Vel form their preterites in this manner in the classical dialect of the Be the Tamil, in the modern colloguial dialect some of those very verbs have adopted the more ordinary method of denoting past time by means of ‘b Thus a suflixed particle or consonant pukk-én,’ 7 entered, has been ten superseded in the modern dialect by ¢ pugu-nd-én,” and ‘nakk-én,’ 7 U laughed, by ‘nagei-tt-én The Canarese forms the preterites of this bei class of verbs in exact agreement with the Old Tamil ; e.g., nakk-anu, Pl he laughed, {from nag-u, to laugh and the Telugu, though less wh systematic in this point, exhibits the operation of the same rule, a8 | especially in the relative participles of the preterite This Dravidian reduplication differs materially in form from that e of the Indo-European languages; but it appears to proceed from a the similar principle, and it constitutes, so far as it goes, an mterestmg dia point of resemblance between the two families. pre dia (2.) THE FORMATION OF THE PRETERITE BY SUFFIXING SOME an PaArTICLE OR sIGN OF PAst Time.—This, with the exception of the the very few verbs included in the previous class, is the method of form- clu ing the preterite which is invariably adopted by the Dravidian IEDI languages, and which may be regarded as their characteristic mode. suff For the purpose of thoroughly investigating this important subject, it lo will be desirable to inquire into the practice of each dialect seriatim of | 1 The Camarese preterite. — The most characteristic Canarese a0 preterite is formed by annexing ‘d’ (euphonically ¢d-u’) to the Ay verbal theme This addition constitutes the preterite verbal par- Clig ticiple; e.g., ili-d-u,” Zaving descended, nudi-d-u,’ having spoken : to lia which the pronominal terminations are suffixed to form the preterite di tense; ¢g., ¢ ili-d-enu,’ 7 descended, ¢ nudi-d- thou sarvdst All verbal tho themes (both in the ancient and in the modern dialect, and whether transitive or intransitive) which end in ‘i or ‘e, form their pre- Prg da terite in this manner, together with many themes ending in ‘u THE PRETERITE 393 TENSE. 18 All the apparent irregularities that exist are merely modifications an of the ¢d’ in question. Thus, sometimes ‘t’ is substituted for d;’ e e.g., ‘aritanu,’ ke knew, instead of aridanu’ (corresponding to the 08 Tamil ‘arindan’): sometimes the “d’ of the preterite combines with el, the final consonant of the root, and converts it into ¢dd’ or ‘ tt; e.g., fo “iddanu,’ %Ze was, instead of ¢irudanu’ (Tam, ‘irundin’); ‘eddu,’ having risen, instead of ‘eludu’ (Tam. ¢erundu’); ¢uttu,’ Aaving les ploughed, instead of ‘uludu’ (Tam. ‘urudu’); ‘nintu, Aaving stood, s instead of ‘niludu’ (Tam. ‘nindru p k, Another Canarese preterite is formed by suffixing i’ to the erude ths verbal theme; e.g., ‘mad-i, Aaving done, from ‘mad-u, o do. the Between this ‘i’ and the pronominal terminations, ‘d’ is inserted in 1ve the formation of the preterite tense; e.g., ‘mad-i-(d)-enu, 7 did; yof ¢ bal-i-(d)-anu,” ke lived. This mode of forming the preterite charac- el terises most verbs ending in ‘u’ in the modern dialect. The final 1 2.9 ‘u’ of such verbs is merely euphonic, not radical, and is elided on ¢ his being annexed; and the ‘d’ which is inserted between ‘i’ and the 1, pronominal signs, though possibly identical in origin with the ¢d’ oss which constitutes a sign of the preterite, is merely euphonic, in so far e, as the use to which it is now put is concerned. In a considerable number of instances the formation of the pre- b hat terite in ¢1 appears to be a modern corruption. Intransitive verbal 0 a themes ending in ‘u’ form their preterite in ‘d’ in the ancient dialect; and it is in the modern dialect alone that i’ forms their ing preterite: e.g., instead of ¢bal-i’ (modern), kaving lived, the ancient dialect has ‘bal-d-u; and as the ancient dialect is undoubtedly more authoritative than the modern, ‘d’ or ‘d-u’ may be considered as ME the the legitimate form of the preterite of this class of verbs. This con- {il clusion is confirmed by the analogy of the Tamil, in which the cor- jal responding verbal theme forms its preterite verbal participle by e, suffixing ‘nd’—an euphonised form of ‘d; eg., ¢ var-nd-u,” having 1t Sowrished, which is the equivalent, not of the modern Can. ¢bAl-i,’ but /] of the ancient ¢bal-d-u.’ How is this diversity in the formation of the preterite to be accounted for ? Jan ‘1’ have been derived in any manner from ‘d 2’ 1650 the An argument in favour of this supposition may be deduced from the Al circumstance that the ancient ‘bal-d-en,’ 7 lived, which is in perfect to dialectic agreement with the Tamil ¢var-nd-én, has in the modern 1t dialect become ¢ bal-i-d-enu. Even in the ancient dialect itself, though this ‘i’ is generally unknown, it makes its appearance in the hal el preterite relative participle; which is ¢ bél-i-d-a,” that lived, not ¢bal- 10 d-a,” though the corresponding Tamil is ¢ vér-nd-a. If we may U 394 THE VERB. judge, therefore, from these instances, ‘i’ seems to have come into existence as a vocalic bond of connexion between the root and the 001 sign of the preterite. ha In a similar manner, the future, both in Canarese and in Tamil, often makes use of ‘u as a bond of union between the verbal root 1 and ‘v, the sign of tense; eg., ‘bal-u-v-enu,” modern Can., aund r Ta ¢ var-u-v-én,” colloquial Tam., 7 skall live, instead of the ancient and more correct ‘bal-v-en,” Can., and ¢ var-v-én,” Tam. In this case the oh S Te is certainly euphonic; though it has not come to be used as ‘i’ has, to express grammatical relation, or in lieu of the sign of tense its which it is employed to euphonize. g If we had to account for the insertion of ¢i % before (d’ n the such instances only as have been mentioned, we inight be content €xl k] as a ha with the supposition of its euphonic origin ; but the use of ¢i sign of the preterite, has a much wider range. All transitive verbs an ending in ‘u,’ both in the ancient dialect of the Canarese and in the to fro modern, form their preterite verbal participles by suffixing ‘i; and there is nothing to show that those verbs ever formed their preterites fro in any other manner. A very large number of verbs of this class form their preterites in Tamil also by suffixing ¢i;’ and in Telugu the che preterite is formed by suffixing €i’ to the root, not of one class 0D of verbs only, but of all, with the exception of the small class of car reduplicative verbs. JU th This statement applies, it is true, to the preterite verbal participle of the Telugu, not to the preterite tense of the verb, which generally tey suffixes or inserts, as a tense-sign, some additional consonant or par- the ticle: but in Malayalam the preterite verbal participle constitutes by the itself the preterite tense, without the addition of any pronominal signs; and in that dialect €i’ is the only sign of past time which is Ca used by a large number of verbs. Thus ¢péad-i,” which means kaving (3! sung, in the other dialects, signifies in Malayalam (/e, ske, or 1t) sang : (w ¢i’ is, therefore, in that dialect a distinctive sign of the preterite in b the class of verbs referred to; and it is to be remembered that the th (] addition of the pronominal terminations, though the means of express- by ing personality, effects no change in the means whereby time is an expressed. k] It The extent and prevalence, therefore, of the use of ‘i as a sign of the preterite, may seem to forbid our supposing it to have been in by all cases derived from an euphonization of ‘d; and as ‘d,’ on the Iey 2 other hand, cannot have been derived from ¢i, it might appear pro- 11 bable that ‘d’ and ‘i are distinct and independent signs of past JUL time. (a THE PRETERITE TENSE. 395 nto I have no doubt that of these two signs of past time ¢d’ is to be the considered as the older and more characteristie. We have seen that in many instances in which the modern Canarese il, has ‘i, the ancient dialect and the Tamil have ¢d. Not in those ? oot instances only, but universally, the Telugu uses i as the sign of the nd preterite ; but the greater antiquity of the grammatical forms of the nd Tamil and the Old Canarese, precludes the supposition that their most the characteristic sign of past time has been corrupted from that of the ‘I’ Telugu. In addition to which, it will be shown that in the Telugu 156 itself there are traces of the existence of an old sign of the preterite agreeing with that of the Tamil and the Ancient Canarese. It would, 11 therefore, appear that two modes of forming the preterite being in ont existence, an older in ‘d’ and a more recent in ‘i, the modern form & has in many instances, particularly in Telugu, superseded the more in Telugu and Gond, would seem lba ancient: and the prevalence of ‘i he to prove that this form, whether an indigenous corruption or derived nd from foreign influences, entered the South-Indian family of languages tes from the Telugu quarter. 85 In the Indo-European family of languages we find similar inter- changes amongst the signs of past time; and though in some instances he 8§ one form or mode may have been derived from another, yet this of cannot have been the case uniformly; e.g., the weak Germanic con- jugations cannot have been corrupted from the strong, or vice versd; le though it seems certain that the strong method of forming the pre- terite was more ancient than the weak, and though it is also certain ly Al that the former mode has in very many instances been superseded by the latter. by al It remains to inquire into the origin of the ¢d’ which is inserted in Canarese between ‘i’ and the pronominal terminations, and also between 15 H ‘3 and the sign of the relative participle. It appears to be used 1 (whatever be its origin) merely for the purpose of preventing /Aiatus between concurrent vowels; e.g., ‘madi-(d)-enu,” /7 did, ‘mad-i-(d)-a,’ n he that did. Hiatus is generally prevented in the Dravidian languages by the insertion of a nasal, or of one of the semi-vowels, ¢y’ and ‘v and it seems extraordinary that ‘d’ should be used for this purpose. 18 It is true that in some of the inflexions of Canarese nouns, eg., ‘mara-d-a,” of a free, ¢d’ might seem to be used euphonically; but it il has been shown in the section on ¢The Noun, that that ¢d’ is the i1} remnant of a neuter demonstrative, and is used as an inflexional 10 )= increment: it is not, therefore, a precedent for the use of ‘d’ for the prevention of Aiatus merely. Possibly the use of this d’ by the st Canarese verb may thus be accounted for :—a consonant for preventing Ll € 396 THE VERB. f hiatus between the sign of the preterite and the subsequent signs of i personality and relation being required, the Canarese preferred using 1l for this purpose an old sign of the preterite which still survived. h Thus, ¢d’ was not a new invention, but an old and partially obsolete particle used for a new purpose, and placed in a position in which it fi would not have appeared, but for the use to which it had already been C put 11 The Tamil preterite. — The preterite is ordinarily formed in b Tamil, as in Canarese, in two ways; viz., by suffixing either ‘d’ or ‘1’ to the verbal theme. In the former case, ‘d’ itself is more rarely used than some euphonization of it or related consonant ; but 1§ such secondary forms invariably resolve themselves into ¢d. Thus, 0 when a theme with 1’ as its final letter is followed by “d’ as the sign 18 of the preterite, the compound becomes ‘ndr; e.g., the preterite Some- verbal participle of ¢ pol,’ like, is not ¢ pol-d-u’ but ¢ pon-dr-u. times, however, when ‘d’ follows ‘1, the compound becomes ‘RR, a pronounced ¢ttr; e.g., from ‘kal,’ to learn, comes not ¢kal-d-u,’ but 11 ‘karr-u’ (“kattr-u’), having learned. ‘1’ followed by “d’ be- a comes ‘nd; eg., from ‘mal’ to die, comes ‘méand-u,’ having died. 18 Sometimes, however, when ¢d’ follows ‘1’ the compound becomes ‘tt; eg., from ‘kél} ¢o hear, comes ‘kétt-v,) having heard. These ) i and similar combinations are merely instances of euphonization, in accordance with the fixed phonetic rules of the language; and in each case it is in reality ‘d’ alone which constitutes the sign of past time. 1§ In some verbs the primitive ‘d’ still remains unchanged and P pure; e.g., ‘uru-d-u, having ploughed, from ‘uru, to plough; or 2 a with a conversion of the dental ‘d’ into the cerebral ‘d ; e.g., ‘kan- il d-u,” having seen, from ‘kén,” to see. The euphonization of ‘d’ which occurs most frequently, and is P 1§ most characteristic of the Tamil, is its conversion into ‘nd. This W conversion takes place without phonetic necessity, and solely through that fondness for nasalisation which is so deeply inherent in the Tamil (L ] and Telugu, and by means of which the formatives ‘gu,” “du,’ and u ‘bu’ have so generally been changed to ‘ngu,” ‘ndu,” and ‘mbu.’ In the majority of cases in Tamil in which ‘d’ (preceded by a vowel or semi vowel) once formed the sign of the preterite, it has been nasalised into ‘nd; whilst the Canarese wherever it has preserved th n the primitive ‘d,’ has preserved it unnasalised and pure. Thus whilst the Tamil preterite of ‘iru,’ fo be, is “iru-nd-én,” 7 was, the corresponding Canarese is ‘iddenu’ (for ¢iru-d-enu’); and whilst the Su preterite of the Tamil verb ¢var, to Afowrish, is °var-nd-an,” /fe J L THE PRETERITE TENSE. 397 Mourished, the equivalent in Ancient Canarese is ‘bal-d-am.’ The 1§ ()f higher dialect of the Tamil retains some traces of the primitive, sing ved, un-nasalised purity of this sign of the preterite; e.g., ¢viru-nd-u,’ having fallen, from ‘viru,” to fall, is occasionally written by the poets lete ch it ¢ vir-d-u.’ (vir’ is phonetically equivalent to ‘viru.’) It is curious to notice the progress of nasalisation which is apparent in this verb on been comparing the Canarese ‘biddu’ (for ¢bil-du’), the High Tamil ¢ virdu,” the modern Tamil ¢ virundu,” and the Malayalam ¢ vinu.’ Another change which ‘d’ undergoes in Tamil consists in its d " o being hardened and doubled in certain cases, so as to become ¢ tt.’ This happens to ‘nd’ as well as to ‘d, a clear proof of the 11016 hut development of the former from the latter; and when the ¢d’ of ‘nd’ is doubled, the nasal entirely disappears. Just as the doubled form hus, of ‘ng’ is ‘kk, and that of ‘mb’ ‘pp; so the doubled form of ‘nd’ sign I In some instances, this change is merely euphonic; e.g., enite ‘padu,’ to lie, an intransitive verb, takes for its preterite, not ‘padu- ome- d-én’ or ‘padu-nd-én,’ but ¢ padu-tt-én,” 7 lay. Such cases, however, RR, are rare, and in general the use of ‘tt’ as a sign of the preterite but instead of ‘d’ or ‘nd’ is a means of distinguishing transitives or be active verbs from intransitive: e.g., the “tt’ of ¢tar-tt-én,” 7 lowered, lied. is formed by the doubling and hardening of the ‘nd’ (the equivalent jmes of ‘d’) of the corresponding intransitive ¢tar-nd-én, 7 got low. See hese the further explanation of this subject under the head of ¢ The Classi- I, 1 fication of Verbs.’ ench The second mode of forming the preterite in Tamil, as in Canarese, e The themes which form their is by suflixing i’ to the verbal theme. and preterite in this manner are those which terminate in ‘u’ euphonie, 0r and of which the radical portion consists either in one long syllable or k- in two syllables, whether short or long. In this connexion, as in prosody, a vowel which is long by position is equivalent to one which Bt is naturally long. The following are examples of the classes of verbs This which take ¢i’ for their preterite:—(long syllable) ‘padu, to sing; ngh (long by position) ¢ pann-u,’ to make,; (two short syllables) ¢ erud-u, o ﬂmﬂ write; (one syllable short, and one long by position) ¢tirupp-u,’ ¢o and turn. b All verbs of which the final consonant is a liquid semi-vowel el ¢l <L’ ‘r) ‘r’ mot ‘v’ or ‘r’), whatever number of syllables el they may contain, form their preterite by means of ‘d’ or some of its yed modifications: such verbs are therefore exceptions to the above rule. Even in the class of Tamil verbs which take ‘i Jii as their preterite the suffix, there are traces of the prevalence of ‘d’ at a more ancient the period. Thus, whilst thow didst go is in the ordinary dialect ¢ p6-(n)-aj’ ) ]t’ 398 THE VERB. (properly ‘pog-i-(n)-4y,” from ‘pd,” or ‘pé-gu,’ to go), in the poets ‘ po-d-i’ is sometimes used instead: so instead of ¢4-(n)-aj’ (for Ag-i- Jli Ji (n)-ay,” from ¢ &-gu,’ to become), thou becamest, the poets sometimes use ‘a-d-1.” In these instances the Canarese also, even in the colloquial pa T dialect, says < pédi’ and ‘adi’ Even ‘nd’ is sometimes ‘d’ only in Tamil poetry; e.g., ¢ varu-d-i, te thow camest, is found instead of the more modern ‘va-nd-aj’ (for el varu-nd-ay’) ; and it is evident that this form, ¢ varu-d-i,” exactly (n corresponds to the forms quoted above, ¢ pd-d-i’ and ¢4-d-i.’ m 1 Notwithstanding, therefore, the prevalence of ¢i’ as a sign of the th preterite in Tamil, as in Canarese (though in a less degree than in Canarese), there seems to be some reason for regarding it as an inno- vl th vation, or at least as a less ancient, less characteristic, and less widely used sign than ¢d. ‘n J of is inserted in Tamil (as ‘d’ in Canarese) between the ‘i’ which constitutes the sign of the preterite of certain I of classes of verbs and the pronominal terminations, and also between the sign of the preterite and the sign of the relative participle; e.g., p from ¢pad-i,’ having sung (the preterite verbal participle of ¢pad-u,’ v to sing), is formed ¢pad-i-(n)-4n,’ 7 sang; ‘pad-i-(n)-ay, thow didst Pa sing; ¢ pad-i-(n)-an,” he sang: so also ¢pad.i-(n)-4, the relative par- I 2 ticiple that sang. Whatever be the origin of this ‘n, it cannot be (¢ doubted that its use in Tamil is at present wholly euphonic; and this a8 statement applies also to the use of the same ‘n’ in the preterite relative participle of the Telugu. It in no respect contributes to the expression of grammatical relation; and when used by the relative it participle in Tamil, it may optionally and elegantly be changed into di ¢ of ¥y, which is one of the semi-vowels that are systematically used for the prevention of Aiatus; e.g., instead of ‘padi(n)a, that sang, we I may write with perfect propriety °padi(y)a.’ We see a parallel use st of “‘n’ in the Turkish verb, in the frequent insertion of an euphonic Ir ‘n’ between the theme and the infinitival particle, and also between U the theme and the sign of the passive. The most weighty argument g in confirmation of the euphonic origin of the Tamilian ‘n’ in question, is derived from the use of ‘n’ as an euphonic fulerum, or means of fo preventing Aiatus in the Drividian languages generally, and even in an connexion with another part of the Tamil verb. Thus,in the classical lat plural neuter of the present tense, ¢ varugindrana’ (‘ varu-gindr-ana’), il they (things) come, the ‘n’ of the pronominal termination ¢ana’ is he undoubtedly equivalent to the ‘v’ of the isolated plural neuter ¢avei’ fo (for ‘ava’); and is used merely for the euphonic prevention of Aiatus g between the first ‘a,” or the demonstrative vowel, and the final ‘a,’ or the sign of the neuter plural. (‘a(n)a’or ‘a(v)a’ isequivalent to ¢ a-a.’) — THE PRETERITE TENSE. 399 poets It the Tamil and the Telugu alone were concerned, we should agi- perhaps be justified in considering the purely euphonic origin of the 8 Use ‘n’ in question to be a settled point; but a difficulty arises on com- quial paring those languages with the Canarese. Wherever the Tamil and 2 Telugu use ‘n in the formation of the preterite tense and the pre- -, terite relative participle, there the Cunarese uses ‘d: e.g., ‘ madi-(d)- (for enu, / did, not ‘méadi-(n)-enu; and ‘madi-(d)-a,” that did, not ‘ madi- :actly (n)-a.’ Now, though this ¢d’ of the Canarese is certainly euphonic in its present use, it has been shown that there is reason for suspect- f t]le ing it to be derived from ‘d,” the old sign of the preterite; and if an I this supposition be correct, it would follow that the Tamilian ‘n, nno- which corresponds so perfectly to the Canarese ‘d,” is derived from idely the same source as ‘d,” and euphonically altered from it. The ‘n’ of the Tamil preterite, therefore, as well as the ‘d’ of the Canarese, 1ese) rtain may testify to the primitive universality of the use of ‘d’ as a sign Weel of past time. Whether ¢d’ (= ‘n’) was originally a sign of the preterite or not, the conversion of ¢d’ into ‘n’ in this connexion, by éd'u;’ viz., in the preterite tense, and especially in the preterite relative participle, is analogous to the change of “ta’ or ‘da’ to ‘na’ in the st past participle of the Indo-European tongues; especially in the p‘dl‘ of be German, from which the final ‘n’ of cur own past participles (such as ‘fallen’) has been derived. ] this terite iii. The Malaydla preterite.—The Malayala preterite is substan- o the tially the same as the Tamil: the only real difference consists in the ative | mto disuse in Malayalam of the pronominal terminations. The sign of past time is invariably the same in each language; with only such q for modifications of sound as are dialectic and regular. That which con- l we | use stitutes the preterite verbal participle in Tamil, is in Malayalam the preterite tense of the verb; e.g., ‘nadandu,” in Tamil signifies having honie walked ; the corresponding Malayala word ‘natannu,” means (ke, ske, ween ment it or they) walked. The only thing which it is necessary to notice here, is the dif- sSﬁOﬂ;‘ ference which exists in Malayalam between the past tense of the verb 10s Of and the past verbal participle or gerund, and the agreement of the el ill latter in appearance with the past relative participle. By analogy sical A8 ‘natanna, having walked, which is the past participle, should have it ’>’ been used as the past tense of the verb, whereas ‘natannu’ is the ]a,ﬂ iS 3 form used instead: ‘natanna’ is also the relative participle that avel’ wa‘lked. s How are we to account for these things? a Uf I conceive that ¢ natannu,’ the finite verb, is an abnormal form: it iy ? — 400 THE VERB. should have been ‘natanna,” and thus identical with the past verbal P Rl R participle. But after the pronominal terminations were laid aside, it th appears to have been felt that something was necessary to distinguish PL the past participle, which is a continuative, from the past tense of the i verb, which is a final; and from this feeling the merely enunciative half sound of the ‘a’ of ‘natanna’ was emphasized, and thus gra- 18 dually transformed into ¢u,” which, though merely an enunciative in he ‘9 Tamil, has a more distinctive position in Malayalam. Whilst this 9 change was going on, the enunciative ‘a of the past participle 18 remained unchanged, inasmuch as it was a continuative word, and not Pa a seat of emphasis. e The explanation of the resemblance between ‘natanna, hawving th walked, the past verbal participle in Malayalam and ‘natanna,’ that s walked, the past relative participle, is very easy. A reference to the th: Tamil shows that the resemblance is only apparent. ‘natanna,’” the th past verbal participle, corresponds to the Tamil ¢nadand-u,” the final U ‘u’ of which is merely enuanciative and euphonie, and is invariably of elided when followed by another vowel ; and in like manner the final 1D ‘a’ of the Malayala past participle is merely enunciative. It is that 1ne euphonic, constantly elided ‘a’ which dialectically answers to the the Tamil ‘u.’ (e.g., compare ‘ad-u,” Tam., i, with the Malayéla at-a ; GUI ‘ar-u, Tam., @ rwer, with ‘4r-a,” Mala.; ‘ondr-u, Tam., one, with e ‘ onn-a, Mala.) Hence arises the rule that this final ¢a 1s not (! to be dwelt upon in pronunciation, but enunciated with rapidity. all Whereas the final ‘a’ of ‘natanna,’ that walked, is identical with ¢ a,’ fes the sign of the relative participle in all the Dravidian languages, and Wa This which was in its origin, as I conceive, a sign of the genitive. dia latter ‘a’ contributes largely to grammatical expression, and cannot Wit be elided without destroying the sense, whilst the former ‘a’ is ide inorganic and merely euphonie. I (& 1V. The Telugu preterite. —In Telugu all preterite verbal par- (1\0 ticinles, without exception, are formed by adding ‘i’ to the theme, ng Even those verbs which form their preterites by suffixing ‘d’ or the some modification of it in Tamil, Canarese, and Malayalam, form eVl their preterites in Telugu by suflixing ‘i; e.g., ‘kon-du, Tam. and hee Can., having bought, is in Telugu ‘kon-i’ and ‘kan-du, Tam. and thoy Can., having seen, is ‘kan-i.’ g Notwithstanding the universality of this rule, there are traces even in Telugu of the use of a particle corresponding to the ‘d’ of Ver the other dialects as a sign of past time. Though the preterite verbal i]]y participle never takes any suffix but that of ‘i, some parts of the di&]i THE PRETERITE TENSE. 401 thal preterite teinse of the verb in the higher idiom of the langunage, viz., o, it the first and second persons both singular and plural, insert the 2 ish particle ‘ti’ between the ‘i of the verbal participle and the pro- " the nominal terminations. It cannot be doubted, I think, that this ‘ti,’ tive which is found nowhere but in the preterite, is allied to the ‘d’ which gra- is inserted in the same place in the Canarese preterite. Thus, whilst el both in Canarese and in Telugu the preterite verbal participle of this ‘ad-u,” to play, is ‘4d-i, having played; in both dialects “ti” or ¢d’ ciple is suffixed to ‘i’ before adding the personal terminations; e.g., com- not pare Can. ‘ad-i-d-enu,” 7 played, Tel. ¢&ad-i-ti-ni. It has already been shown to be probable that the ¢d’ thus inserted by the Canarese, ping though now used to so large an extent euphonically, was originally a that sign of the preterite, identical with the ‘d’ which is still used [for This view derives confirmation from o the that purpose by many verbs. ' the the Telugu, in which the corresponding ¢ti’ does not appear to be final used euphonically at all, and certainly is not used for the prevention iably of hiatus; for there is no Aiatus and no necessity for an euphonic final insertion between the aforesaid ¢4di’ and ‘ni,’ the pronominal frag- that ment, or in the second person between ‘Adi’ and ‘vi.’ Moreover, ) the there is no instance of such a particle as ‘ti’ being used merely for =i euphony in any of the Drividian dialects. It therefore follows that with we must regard ‘ti’ as a sign of past time—subordinate indeed to g 106 ‘1, and unused in the third person of the preterite, but immediately ity allied to ¢d,” the past tense-sign of the Tamil and Canarese, and testifying to the existence of a time when ¢d,” or its equivalent ¢ ti,’ h ¢ 3; was the ordinary sign of the preterite in Telugu, as in the other , and dialects. This In some Telugu verbs, “ti’ is combined in such a manner with the final consonant of the theme, as to prove beyond doubt its annot 2| I8 identity in origin and force with the Tamil ‘d:" e.g., € chés-ti-ni,” Tel., I did (for °chési-ti-ni’), is evidently equivalent to the Tam. &ey- d-én; and ‘kon-ti-ni, 7 bought (for <koni-ti-ni’), is equivalent to | pat ‘kon-d-én.” So also when ¢ &) the Telugu conditional particle, answer- ing to the Tamil ‘&l is suffixed to the preterite tense of a verb for eme, 4’ or the purpose of giving to it the meaning of the subjunctive, it appears form evident that the ancient sign of the preterite of the Telugu must have been, not i, but “ti’ or ‘it; eg., compare the Telugu ¢ chést-&,” +f (7, ., alld v a,n(] thou, he, &c.) did or do (abbreviated from ‘chés-it-&’), with the Tamil ‘Seyd-al’ We have seen that the Tamil inserts <n’ between the preterite traces i of verbal participle and the pronominal terminations in many instances in which ‘d’ is used for this purpose in Canarese. The colloquial ,'erbfll dialect of the Telugu makes much use of ‘n4’ or ‘na’ in the same f the 2'n 402 THE VERB. connexion ; e.g., ‘4d-i-(nd)-nu,” 7 played, (answering to the Tamil in ¢ 4d-i-(n)-én "), instead of the more ancient and elegant ¢&d-i-ti-ni.’ P ilh The Tamil ‘n’ used in this connexion separates two vowels, and might therefore be considered as purely euphonic; but the Telugu ch ‘na’ or ‘na comes between a vowel and a consonant, and must, “H therefore, have a more important use than that of preventing hiatus. ‘ti’ and ‘na’ alternate in the formation of the Telugu preterite T tense, ‘ti’ being preferred in the more classical dialect, ‘na’ in the more vulgar: and this seems to confirm the supposition that the ‘n’ of We of ‘na,’ like the Tamil ‘n,’ is derived from the old preterite ¢d.’ can hardly fail to regard these particles as identical, when we examine Vel an instance in which they are used as equivalents in all three dialects; suf In e.g., compare ‘ay-i-(n&)-nu,’ Tel., 7 became ; < 4-(n)-én,” Tam. for (‘ag- ¢ A ° i-(n)-én’); and ¢4-(d)-enu,” Can. (for ag—l -(d)-enu’). tg On the whole, therefore, it may be concluded that the Telugu suf agrees with the other dialects in exhibiting distinct and deep-seated suf traces of the ancient use of d’ or ‘t’ as a sign of the preterite, not- An withstanding the universal prevalence in Telugu at present of the use 10§ of ‘i as the sign of the preterite verbal participle. the T may here take occasion to guard against an illusory resemblance to which my attention was once called, viz., the resemblance which 1m0 subsists between the Telugu preterite verbal participle ¢veichi,” having D placed, and the corresponding Tamil participle ¢veittu,” which is vulgarly pronounced °veichi.’ The ‘tt’ of the Tamil ¢vei-tt-u,’ being simply the hardened and doubled form of ¢d.’ is the ordinary sign of the preterite; and if there were any real alliance between 0] “tt-u,” through its provineial pronunciation, and the Telugu ‘ch-i,” we tha should undoubtedly have here an instance of the use of ‘tt,” i.e, tior of ¢d,” in modern Telugu as well as in Tamil, as a sign of the pre- the terite verbal participle, and consequently of past time. The resem- blance, however, is altogether illusory. The “ch’ of the Telugu I eql “veichi’ corresponds, not to the ‘tt’ of the Tamil ¢veittu,” but to ‘b the ¢kk’ which constitutes the formative of so many verbs and nouns in Tamil. atty ‘kk’ makes its appearance in the infinitive of this very soft verb, viz., ‘vei-kk-a,” to place, the Telugn of which is ¢ vei-ch-a.’ See ¢kk’ is vulgarly pronounced ‘ch’ in the southern part of the Tamil tug country; and the same pronunciation universally obtains in Telugu. Stea The imperative or theme of this verb in Telugu is not ‘vei,’ as in 2 of the formative Prgf Tamil, but ¢veich-u’ (with the addition to ¢ vel beiy ‘ch-u,” which is equivalent to the Tamil ‘kk-u’); and from this te) ¢ veich-u,’ the preterite verbal participle veich-i, is regularly formed, 9 tion; in thig, as in all other cases, by the addition of ‘i If the correspond- Wep THE PRETERITE TENSE. 403 mil ing Tamil verb formed its preterite in the same manner, its verbal 11 participle would be ¢vei-kk-i, not ¢vei-tt-u.’ A case in point in and Hlustration of this, is the Tam. ‘ti-kk-u,’ to lift, to weigh (Tel. ¢ ti- ugu ch-u’), the preterite verbal participle of which is ¢ti-kk-i’ (Tel. 10, ‘ tit-ch-i’). s, erife v. Preterites of minor dialects.—In Tulu, ‘t’ constitutes the cha- the racteristic mark of the preterite tense, and “ti,” the preterite insertion of the Telugu verb is the sign of the Tulu preterite verbal participle. We In Gond ‘si’ or ji,) apparently softened from “ti, forms the nine verbal participle of the preterite; but the perfect tense is formed by oS suffixing “tt, e.g., ‘kei-tt-an,” 7 have called; °kei-si,) having called. (% In Seoni Gond also the preterite or conjunctive participle suffixes si; eg., ‘ wunk-si, having spoken: but the past participle is formed by lugn suffixing ‘tlr; e.g., ¢ wank-tir,” spoken; and the past tense simply ated suffixes ‘t; eg., ‘wunk-t-an,” / spoke, ‘wunk-t-i, thow didst speak. not An imperfect or progressive tense is formed in both those dialects by RILH inserting ‘und’ or ‘nd,” the substantive verb, between the root and the pronominal terminations. a1ce These instances tend to eonfirm the supposition that ¢ d, or some rhich modification of it, is the oldest and most characteristic sign of the 2 iy Dravidian preterite, and that the use of ‘i is of secondary origin. h i ‘tt'u;’ Origin of the Dravidian signs of past time. ety (1.) The most probable conjecture that I can offer respecting the weel origin of ‘i, is one which confirms the supposition of its secondary character. , Ve I coneeive it to have been originally a vowel of conjunc- | " tion, employed for the purpose of euphonically connecting the verbal PIE theme, and the true sign of past time, ‘d’ or ‘d-u’ ssenl- Where the theme terminated in a hard consonant, euphony would elugt require some such vocalic bond of connexion ; e.g., the Old Canarese, ¢ béal-d-en, [ lived, is undoubtedly somewhat harsh to an ear that is b fo n0uDs attuned to Dravidian phonetics; and it was natural that it should be We yery softened, as it has been in modern Canarese, into ‘bAal i-d-enu. ‘ch-&ul see a precisely similar euphonic insertion of ‘i’ in the Latin ¢ dom-i- Tamil tus’ (instead of ‘dom-tus’), famed, and the Sanscrit ¢ pid-i-tah’ (in- stead of ‘pid-tah’), pressed. Subsequently we may suppose the true elugll. 2 in preterite ‘d” to have gradually dropped off; whilst ‘i’ remained, as ative being the easier sound, with the adwventitious signification of the pre- - fhis terite. There are many instances in all languages of euphonic addi- ;meth tions coming to be used instead of the parts of speech to which they were attached ; e.g., in the Telugu verb ¢ vu’ is used to represent the pond‘ 2D 2 404 THE VERB. second person singular of the pronoun instead of ‘bm,’ thou, though ‘vu’ was originally only an euphonic addition to ‘ni,’ by which it b was converted into ¢ nivu.’ It deserves motice that wherever ‘i’ is used in Canarese or in b Tamil, instead of ‘d,” as a sign of the preterite, the use of ‘d’ would t in that instance be harsh and uncouth; and that on comparing the . d > with those that suffix i Tamil verbs which form their preterite in ‘1 “d, no reason but euphony can be alleged why the one suffix should 0o be employed rather than the other: conmsequently euphonic causes I must have contributed to the development of ‘i.’ 11 This supposition of the origin of ¢i’ from the vocalic conjunction of ¢4’ with the verbal theme, would also account for the circumstance fr that wherever ‘i’ is followed by a vowel (whether the initial vowel of the pronominal terminations, or the ‘a’ which constitutes the sign € of the relative participle), it picks up again the ‘d’ which it had th 9 gradually lost, and uses it as an euphonic bond of conjunction, either in its original shape of ¢d,’ as in Canarese, or in its nasalised shape di of “n, as in Tamil and Telugu. The manner in which ‘ti’ is sepa- rated from the theme in some Telugu preterites, e.g., ‘kon-i-ti-ni’ di (‘kon-ti-ni’), I bought, confirms this supposition of the euphonic de origin of ¢i.’ 10 (2.) ¢d, the older and more characteristic sign of the Dravidian ¢ preterite, has many interesting affinities with corresponding signs fa of past time in various Indo-European and Scythian languages. th I have no doubt that it has an ulterior, though remote, connexion fr with “t> or ‘ta’ (alternating with ‘na’), the ordinary suffix of the te Indo-European passive particle; eg., ¢jiid-ta-h,” Sans., known ; Greek ng ¢ yvw-7o-s; Latin “(g)nd-tu-s;” ¢ bhug-na-s,” Sans., bent; Gothic bug- a-n(a)-s.’ In Gothic this suffix is “d’ or ‘t; in New Persian inva- = & riably ¢d. § il In Sanscrit the participle which is formed from ta’ is in general 1§ distinctively passive ; but a few traces exist of a preterite signification, 81 only however in connexion with neuter verbs; e.g., ‘ga-ta-s,’ one who al went ; ¢ bhil-ta-s,” one who has been. A preterite signification pre- 3 dominates also in the active participles formed by suffixing ¢tavat’ pe (derived from the passive ‘ta’); e.g., ‘kra-tavat,” was making; and Per in the indeterminate past participle, or gerund, which is formed by suffixing ‘tva; eg., ‘kru-tvd,” having made or through making. pa Though there is probably an ultimate connexion between the for preterite ‘d’ of the Dravidian languages and the passive (and for secondary preterite) ‘t’ of the Sanscrit, the use of this suffix is too 8y essential a characteristic of the Dravidian languages, and too rare and aly THE PRETERITE TENSE. 405 igh exceptional in Sanscrit, to admit of the supposition that the former B borrowed it from the latter. The <1’ which constitutes the sign of the preterite in Bengali, has n been supposed by Professors Max Miiller and Bopp to be derived from uld the past participial ‘t’ of the Sanscrit; e.g., ‘karilam,’ 7 did, is the derived by them from *karita,” Sans., done, followed by the personal liibs termination ¢4m This supposition is confirmed by the conformity yuld of ‘karilam’ to the New Persian ¢kardem,” 7 did, and by the use in uses Marathi of a similar preterite in ‘1, which is supposed to be derived in like manner from the Sans. passive participial ‘t; e.g., mi kelo-m,’ tion I did, ‘min gélo-n, I went. The interchance of ¢d” and ‘1’ is of ance frequent occurrence ; and possibly the Sanscrit ‘t’ may have become owvel “d’ before it was corrupted into ‘1. There is no proof of this, how- sign ever, and the ‘1’ which is used as the equivalent of “t’ or ‘d’ in had the formation of the Slavonian preterite, ‘byl’ (Pers. ¢btd,” Sans. ther ¢ bhiita-s’), ke was, shows that ‘t’ may have passed into ‘1’ imme- hape diately, without the middle point of the cerebral ¢d. QPLL Whether the preterite 1’ of the Bengali and Marathi is derived -0’ directly from the Sanscrit passive participial ¢t,” or whether it has oni6 descended from the Old Scythian vernacular of Northern India, it is interesting to notice the fact of the conformity in this important par- dian ticular, between the Dravidian languages and those of the Gauda family. We should notice, however, this important difference between gns the two, that whilst the Gauda preterite ‘1 in so far as it is derived }Xiull from the Sanscrit, appears to be only a secondary constructive pre- f the terite, the Dravidian ¢d’ exhibits no trace whatever, either of con- reek nexion with any passive participle or of a constructive origin. bug- In the New Persian, ‘d’ invariably forms the sign of the preterite; nva- e.g., ‘bl-d-em,” I was,; ‘bur-d-em,” I bore. The participle which con- stitutes the verbal theme in Persian, and which has a formative that peral is passive in Sanscrit, has an active as well as a passive-preterite signification; e.g., ‘burdeb, means either borne or having borne, JﬁOYb o according to the context. The preterite tense has in Persian been I)I'e' developed out of a passive participle; and this appears to have hap- pened through the influence of the past time which is inherent in the a'\"at ! and perfect passive. In the Germanic tongues, ‘t’ or ‘d’ not only forms the perfect iy passive participle, as in Sanscrit, Latin, and Zend ; but is used also to ] the form a regular preterite tense. The Sanscrit ‘t’ of ¢ gata-s,” who went, forms the preterite of neuters only; but the Gothic ‘t’ appears (31](1 systematically in the preterites of a numerous class of active verbs; ig too 6 {lﬂd and is found not only in the participle, but in the regular preterite 406 THE VERB Vi tense; e.g., ¢ bauh-ta,” 7 bought; ¢ thah-ta, 7 thought; and ‘¢ vaurh-ta, I made. St It is not my object to endeavour to trace the origin of the suffix in JLi question. Whether the ‘t’ of the passive participle is identical, as I conceive, with that of the indicative preterite, or whether it springs fu from a different origin,—whether ‘t’ or ¢d,” the sign of the preterite, by is derived from ¢ dha, ¢o set, to make, from °thun,’ to do, or from the th demonstrative ¢ ta,—I am not about to inquire. My belief is, that the real origin of this suffix cannot now be ascertained; it has simply been my object to point out in various languages of the Indo-European ( family, and in the western branches more than in the Sanscrit, the di existence of a mode of forming the preterite (viz., by suffixing ‘d’ or a0 “t’ to the root) which closely resembles that which forms a charac- teristic of the Dravidian languages. fu The formation of the preterite by suffixing ‘d,’ is not confined to the Indo-European family, but prevails also in the Turkish and Ugrian st tongues. U8 “d’ is the sign of past time which is used by the Turkish; e.g., be compare ‘sewer-im,’ I love, with ¢ sewer-d-im,” 7 loved ; and this ‘d’ is A inserted, as in Tamil and Canarese, between the root and the pro- Y0 nominal signs. Compare the present ‘im,” 7 am, with the preterite 0 “i-d-um,” £ was. Notice also ¢6l-d-um,” 7 was, and the equivalent form in Oriental Turkish, ¢bol-d-im.’ In Finnish, the preterite is regularly formed by suffixing ¢t.” The Ul) al preterite participle from which the perfect tense is formed terminates al in ‘ut, ‘yt, ‘et &e.; eg., ‘oll-ut, having been, from the theme ‘ol,’ to be. The Hungarian forms its preterite in a similar manner; e.g., the ¢ 18 preterite participle of ‘le-uni,” to become, is le-tt, having become; and EXi from this is regularly formed the perfect ‘ le-tt-em,’ 7 have become. n It especially deserves notice, that these Turkish, Finnish, and i Hungarian signs of the preterite are totally unconnected with the passive participle. They are distinetive signs of past time, and of that alone ; and as such they are suffixed to all indicatives, whether active ©) or neuter, and are appended, in addition to the sign of passivity, to passive forms, only when those passives are also preterites. In this Wl particular, therefore, the analogy between the Dravidian preterite and sig the Turko-Ugrian is closer and more distinctive than the Indo- I European analogies which have been pointed out: it may be said 0n indeed to amount to identity. the 3. The future tense.—The present and preterite tenses of the Dra- g B — —— Re— e e — THE FUTURE TENSE. 407 & vidian verb are formed from present and preterite participles, by suffixing the pronominal terminations. The future is without a verbal n participle or gerund, except in High Tamil, in which there is a rarely s used verbal participle of the future, ending in ¢véin; nevertheless, the 125 future tense is formed virtually in the same manner as the other tenses, te, by suffixing a sign of future time to the verbal theme, and adding to the that sign the pronominal terminations. the In the Dravidian languages there are two future formations. gell One, which is more distinctly a future than the other, is found in can Canarese and Telugu alone; the other, which is contained in all the the dialects, inclusive of the Canarese and Telugu, is an indeterminate, 0r aoristic future, and is called by Telugu grammarians the aorist.’ [0 It should here be observed also, that the use of the present for the future is exceedingly common in all the Dravidian dialeets. | to il (1.) The more distinctive future—In modern Canarese this con- stitutes the second form of the future, in consequence of being less used than the other. It is formed by inserting ¢iy, or ‘i’ or ‘d; (il between the theme and the pronominal signs, and lengthening the 15 vowel which immediately follows this future particle, viz., the initial L0 vowel of the pronoun; e.g., ‘méd-iy-énu,’ 7 will do, or, ‘nudi-d-énu,’ 7 11t will say. Jenb In Telugu also, this future assumes a two-fold form, from the €L Iy The optional use of two inserted particles, corresponding to the ‘iy,” or and ‘d’ of the Canarese. One form inserts ‘¢’ between the theme ates and the pronominal terminations; e.g., ‘chés-é-nu,” 7 will do,; which (5335 ol ¢’ is optionally changed to 1 in the third person neuter plural; e.g., ¢ chés-i-ni,” they (neut.) will do. The other form of the future, which the is still more rarely used, inserts ‘eda; e.g., ¢chés-eda-nu,’ 7 will do; and except in the third person singular, and the third person neuter plural, in which ‘edi’ is used instead of ‘eda; e.g., ‘chés-edi-ni,’ they (neut.) and will do. the that (2.) The aoristic future, or aorist.—Of this future also there are several forms. (Hive 1\ In Tamil, the most commonly used form of the future is that Jy to fhis which inserts ‘v, ¢b,” or ¢pp, between the theme and the pronominal signs; e.g., ‘ fey-v-én, I will do; ‘kan-b-én,” I will see; ‘nada-pp-én,’ x ﬂﬂd 1 will walk. ‘b’ appears to be the original form of this particle; for ndo on this supposition we can easily account both for ‘v’ and ¢pp.’ aid In certain cases the initial consonant of the temporal particle, or the particle itself, if composed of a single consonant, would require to be doubled, e.g., after the vowels a’ and ‘i) and when the doubling Dra 408 THE VERB. 186 of the tense sign is the method adopted for converting an intransitive verb into a transitive. In those cases we find ‘ pp ’ used as the sign of the the future instead of ‘v; e.g., ¢ valar-pp-én,” 7 will rear, is the future the transitive of ¢valar-v-én,” 7 will grow. And we may hence conclude il that ‘b,” not ¢ v, was originally the sign of the future of the intran- 81 of sitive, for ‘b’ when doubled becomes ‘pp’ by rule; and whilst it is certain that ‘b’ will readily change into ‘v,” no instance of the change fen of ‘v’ into ‘b’ in Tamil can be adduced. Notwithstanding this, ‘b’ doe is now used only after a final nasal; eg., tin-b-én, 7 will eat; ‘ en-b-én,” 7 will say. thi In consequence of ‘b’ invariably becoming ‘pp’ after the vowel (18] ‘1 ‘pp’ is the sign of the future of all Tamil causals; e.g., ‘kattu-vi- | One pp-én,’ 7 will buald. In classical Tamil there is a future verbal participle or gerund, in wh ‘van,” which when hardened becomes ppan; eg., ‘Sey-van,’ being ver about to do,; padi-ppan,’ being about to learn. The principal element of In this is the future tense-sign ‘ v;’ and this participle in ¢van’ or the ‘ppan,’ constitutes the sign of the infinitive in Malayilam; eg., ‘iri-ppan,’ to be, ‘ urukku-van,’ ¢o dissolve. 488 The Tamil future formed from ‘v’ or ‘b, is destitute of a relative 10 § i participle, and uses instead the aorist future in ‘um.’ Generally also, that aorist is used instead of the more distinctive future in the third the person singular neuter. Thus, whilst e will be is ¢iru-pp-an,’ ¢ will 16 be is ordinarily ¢iru(kk)-um,” not ‘iru-pp-adu; and forms like ass) ‘iru-pp-adu’ are in general used only as participial nouns. In this kS i S LR e i respect the Tamil is less regular than the Canarese, in which the po ordinary third person neuter singular of the future tense is ‘iru-v-adu.’ e, The future is sometimes formed by the Tamil poets by adding ‘g Iy (or ‘kk’) to the root, instead of ‘b’ (or ‘pp’); e.g., “Seygén’ (for the I ‘seyvén’), L will do; ‘adeikkén’ (for ‘adeippén’), 7 will obtain. hax am inclined to think these forms identical in origin, for € g’ often futy changes into ‘v.’ Whi Another future formation of the Tamil may be called the defective (ch aoristic future, inasmuch as its reference to future time is still less the distinct and determinate than the future in °v,” and as it is restricted Dag to two forms, the third person singular neuter, and the relative par- Wit ticiple. This defective future is formed by suffixing ‘um to the ) The future in ‘um’ is not formed theme ; ‘erud-um, @t will write. considered by Tamil grammarians as distinct from, and independent of, the future in ¢ v; but is strangely enough considered as a part of it. fory Its claim, however, to be regarded as a distinet future formation is 18y confirmed by the Malayalam, in which it is the only future in ordinary apl THE FUTURE TENSE. 409 lvé use; e.g., ‘fijan erut-um, 7 will write, ‘ni erut-um, thow wilt write; 1of the other form corresponding to the Tamil future in ¢ v, being used in 1re the poetry alone. In the Tamil of prose and conversation the future ude in ‘um’ is used in connexion with the neuter of the third person an: singular alone; but in the poetry it occasionally takes a wider range t1s of application, and is sometimes construed even with the masculine- nge feminine plural, as in Malayalam. (b’ The future in < um, when used in Tamil as a relative participle, fat; does not differ from the form of the same future which is used as the third person singular neuter. The forms are identical ; e.g., ‘pog-um,’ )\Vt?l it will go, € pég-um, which will go; they may therefore be regarded as V1 one. ‘um’ is added, not to the crude root of the verb, or that form | in which is used as the imperative, but to the formed theme, or that el verbal noun which forms the basis of the infinitive, and the equivalent nent of which constitutes in Telugu the inflexional basis of every part of or the verb. The base to which the future ‘um’ is suffixed, may safely be e:gl’ assumed to be a verbal noun, even in Tamil, though it rarely appears Lti\'e in a separate shape. also, The following instances will show the relation subsisting between hird the Tamil infinitive and the aoristic, impersonal future, in virtue of the formation of both on the basis of the formed verbal theme, or will like assumed verbal noun, in question :—compare pog-a,’ fo go, ¢ pég-um,’ it this will go, inflexional theme ‘pé-gu:’ “pokk-a,” to cause to go, to get rid of, ¢ pokk-um,’ @t will get rid of; inflexional theme ¢ po-kku:’ irukk-a, ¢o the adu be, < irukk-um,’ vt will be, inflexional theme ¢iru-kku.’ In those cases in which intransitive verbs are converted into transitives by doubling the initial consonant of the tense-sign (e.g., ¢ valar-gir-én, 7 grow, (for i hardened into ¢ valar-kkir-én,’ 7 rear), the infinitive and the aoristic often future of the transitive verb are formed upon the basis of a theme which terminates in the formative ‘kk-u’ (the equivalent of which is “ch-u’ in Telugu), whilst the unformed theme, or ultimate root, is epli\’e Jess the basis of the corresponding forms of the intransitive; e.g., com- pare ¢ valar-a,’ to grow, ¢valar-um, ¢ will grow, theme ¢ valar, icted with ¢ valar-kk-a,’ ¢o rear, °valar-kk-um, < will 7rear; theme pﬁl ) t[le ‘valar-kku.’ It is evident from a comparison of these illustrations that the above g ]lot ¢g’ or ‘k’ is no part of the sign of future time; it belongs to the nt of 1t formative, not to the future ; the infinitive as well as the aoristic future is built upon it; and the Telugu formative which corresponds to it has on iS a place in every part of the verb. nay & # 410 TIHHE VERB. fhat The fature in ‘um’ is altogether impersonal, no pronominal termi- nations are ever added to it, and in consequence it is well adapted to (ans be used as a relative participle, the relative participles being used alike Q071 by all persons, numbers, and genders. The particle ‘um’ which constitutes the sign of future time, is with identical in form, and is also, I believe, identical in origin and force, gran with “um,’ the conjunctive or copulative particle of the Tamil. It is a0t also identical with ‘nu,’ the impersonal suffix of the third person pror singular and plural of each gender of the Telugu aorist, a tense which sing perfectly corresponds with the one now under consideration. ‘nu’is 3 J i a conjunctive particle in Telugu also; and it is probable that this Deaon will particle has been chosen, both in Tamil and in Telugu, to be the characteristic sign of the aorist, because of its suitableness to express the T the idea of continuity. This tense, it is true, frequently denotes the witl; future; but it is much more frequently used to express continuous .‘iﬁ action, or what is habitually dcne. Thus, “mad-u pul tin(n)-um, e, of tl Tam., is to be translated, not the ow will eat grass, but the ox eats (i.e. habitually eats) grass, or grass s the ox’s food. When the relative participle of this aoristic future, coupled to a L noun signifying time, is followed by a finite preterite verb, the future 100§ in Tamil takes the sense of the imperfect; e.g., ‘nin var-um porudu, Is porei (k)kandén,” when I was coming (literally when I shall come), I saw 18 1¢ the battle. In respect of this capacity of the aoristic future for becom: Louy ing a historical preterite, it resembles the future tense of the Semitic witl Lat languages. The High Tamil (and also the Telugu) often uses the formed B R e e e verh theme or verbal noun referred to, without the addition of um,” as an wid C0ng aorist; e.g., ¢ parapp-u,’ instead of ¢ parapp-um,’ it will spread, or which will spread. This form is rarely used except by the poets, and is even Som more distinctively an aorist than the aorist future in <um.’ The ) (‘-Iose final “u’ does not belong to the particle ‘um, but is the ordinary euphonic, enunciative < u,” and accordingly is often elided. or The Canarese, with which the Tulu agrees, forms its ordinary aly] future by inserting ¢ v’ between the theme and the pronominal termi- and nations, in accordance with the first Tamil future, viz., that in ‘v.’ be g Like it, this Canarese future has often an indeterminate, aoristic sense; and but it is more regular than the Tamil, inasmuch as it never changes Dart N 9 into ‘b’ or ‘pp, in the modern dialect, but uses ‘v’ as the 665715 invariable sign of future time. It is not obliged also, like the Tamil, [ to borrow its third person singular neuter from another formation, but tion. forms it, like the other persons, by means of ‘v; eg., ¢irn-v-adu, they will be; and it has also a relative participle of its own; e.g., ‘balu-v-a, Varg THE FUTURE TENSE. 411 oI- that will live; compare ‘4gipa, Ancient Canarese, that will become, ted to (answering to the modern ¢ Ag-iruva’). alike The Telugu tense which corresponds to the Tamil and Canarese aoristic futures is still more distinctively an aorist than they, though e, 18 with an inclination in general to the idea of futurity. By English force, grammarians this tense is commonly called, not ¢ the future,” but ¢the aorist.’ It is formed by inserting ‘du’ between the theme and the It1s pronominal terminations; with the exception of the third person JErS0n which singular and plural, in which ‘nu’ alone, the equivalent of the Tamil o' s ‘um, is added to the theme. Compare the Tam. ‘ag-um,” it will b this become, it will be, with the Telugu aorist ‘avu-nu, (e, she, it, they, &c.) will become. e the Probably the Tel. aoristic formative “du’ is allied to ‘tu,’ the particle of present time. xpress s the The Gond makes use of ‘ k’ as the sign of the future, in connexion with the first and second persons of the verb; e.g., wunki-k-a,” 7 will nuous speak. Compare the ‘g’ or ‘kk’ which is sometimes used as the sign J-um, )(lt of the future by the High Tamil. toa A ffinities of the sign of the future.—The most characteristic and most extensively used sign of the future in the Dravidian tongues, uture is evidently the ‘v’ or ‘b’ of the Tamil, Canarese, and Tulu. It rudu, ¢ [ saw is remarkable that in Bengali also, the sign of future time is v, pro- nounced ‘b ; and this Bengali ‘b’ has been connected by Max Miller eeon with the ‘b’ or ‘bo’ which forms the most characteristic sign of the pNIIC Latin future, and which is considered to be a relic of an old substantive verb. The ¢d’ of the Dravidian preterite has been proved to have so med a8 all wide a range of affinities both in Europe and Asia, that it cannot be considered improbable that the Drévidian futuric ‘b’ also possesses uv]“r]l some ulterior Indo-Euaropean affinities. gven The As in the case of the sign of the preterite, it will be found that the closest analogies are those of the Ugrian languages. In Finnish, ‘wa’ inary or ‘va’ is the sign of the future participle which is used as an auxiliary in the formation of the future tense; e.g., ‘ole-wa,” about to be: inary and the sign of the future infinitive is ¢ wan; e.g., ¢ ole-wan,” to be, to prml be about to be; with which we may compare the Tamil future gerund, ‘v and Malayéla infinitive in ¢ van.’ In the Hungarian, the future Genses participle is formed by suffixing ‘v6;" e.g., ‘1é-v6’ (Finnish ¢ole-wa’), anges being or about to be. g the 4. Compound tenses.—It is unnecessary to enter into an investiga- Fﬂm]]; tion of the Dravidian compound tenses, inasmuch as in all the dialects )y hot they are formed in the simplest possible manner, by suffixing the o, 1t various tenses of the substantive verb to the verbal participles of active -4 412 THE VERB. verbs. Thus, doing I was will represent the imperfect (also doing 7 came); doing-keeping (i.c., keeping a doing) I was, a more continnative e it imperfect ; having done I am, the perfect ; having done I was, the plu- perfect; having done 1 shall be, the future perfect. the A vast number of auxiliary verbs are used in all the Dravidian ced bel« dialects, in conjunction with infinitives and verbal participles, for the purpose of expressing compound ideas; but as the use of those auxiliaries bio pertains rather to the idiom or syntax of the language than to the the W grammatical structure, and is sufficiently explained in the ordinary grammars, it would be out of place to inquire into them here. i &, it | RELATIVE PARTIOIPLES.—It is a remarkable peeuliarity of the Dravidian languages, that they bhave no relative pronouns whatever, ver and that the place of the relative pronoun is supplied by a part of the wh Ul verb which is called ‘the relative participle,” a participle which is invariably followed by a noun, and preceded by the words or phrases w that depend upon the relative. 18 The vernaculars of Northern India have relative pronouns derived 4, from the Sanscrit relatives ¢ yah,” ‘ya,’ ¢ yad,” who, which; but of those plo pronouns they make little use, probably through an under-current of 1 Dravidian, or at least of Pre-Sanscrit influences. In those languages a sentence which contains a relative is ordinarily divided into two 0on members; and the demonstrative pronoun which forms the nominative fon ticr of the second member of the sentence, is used instead of a relative. Thus instead of saying, the man who came yesterday has come again to- el day, they would prefer to say, a man came yesterday, he is come again I to-day. The Dravidian languages sometimes make use of a similar 1ey idiom, but only in the hurry of conversation. They are not obliged to have recourse to any such arrangement, the signification of the relative, Tar together with that of the definite article, being contained in, and Tely distinctly expressed by, the relative participle of the verb. Thus they dy would say in Tamil, ‘vanda-al,’ the person who came, literally the-who- Cay came person. In like manner instead of ¢vanda,” the preterite, they 0f { might use the present relative participle; e.g., ¢ varugira al,” the-who- i €] is coming person, or the future ¢ varum 4al,” the-who-will-come person. fom The name given to the relative participle by Tamil grammarians, F is, { is ‘peyr echam,” noun-defect, or noun-complement, i.e., a word which (%7) requires the complement of a noun to complete its signification. This uh name is given to it because it participates so largely in the nature of an adjective, that it is invariably followed by a noun, to which it Pre stands in the relation of a relative, and which it connects with the e, antecedent clauses. in\’a THE RELATIVE PARTICIPLE. 413 ing | Like other Dravidian adjectives, it undergoes no alteration on lative account of the number or gender of the related noun; but inasmuch as e plu it is a verb as well as an adjective (¢.c., a participle participating in the nature of both parts of speech), it is capable of governing a pre- ndian ceding noun, equally with any other part of the verb to which it or the belongs; e.g., ‘nilleil erudina pulavan,” Tam., the poet who wrote the liaries book, literally the-who-the-book-wrote poet; ¢kéttil tirigira yauei,” Tam., 0 the the elephant that wanders in the jungle, literally the-ﬁz‘hat-in-tl&e-j@mgle- wanders elephant. linary The relative suffix most largely used in the Dravidian languages is “a,” which is appended to the verbal participle or gerund, to convert of the it into a relative participle. Thus, in Tamil, the (assnmed) present tever, verbal participle of ‘uru,’ to plough, i “uru-gir,” ploughing; from of the which, by suffixing ¢a, is formed the present relative participle ioh 18 ‘urugir-a, that ploughs. The preterite verbal participle of the 9 hrases same verb is ‘uru-d-u, having ploughed, (of which the final ‘u is merely enunciative,) from which, by the addition of the same 14 erived a, is formed the preterite relative participle °uru-d-a,’ ¢that ‘those ploughed. 3 When the preterite verbal participle ends, not in ¢d-u,” but in ¢ i b nt of ‘n’ (alternating with ‘y’), is euphonically inserted between the 1ges & concurrent vowels ‘i’ and ‘aj eg., from ‘erud-i, having written, is ) tWo formed ¢ erud-i-(n)-a,’ or ‘erud-i-(y)-a, that wrote. In all these par- native The future afive. ticulars, the Malayalam perfectly agrees with the Tamil relative participle of the Tamil is not formed from *a,” but terminates fin fo- in ‘um,’ being identical with the aoristic future third person singular g neuter. imilar The Canarese has in this point the advantage not only of the ged t0 ative, Tamil, but of all the other dialects; inasmuch as it forms its future relative participle also by suflixing ‘a; eg., < midu-v-a,’ which will , and do On the other hand, the relative participle of the present tense in ¢ they g-il,']lf)- Canarese is defective, being formed by means of the relative participle of the future, used as an auxiliary: e.g., ¢ bal-utt-iruva,” which lives, ) they literally which will be living. The preterite relative participle is g-w/t()‘ formed, like that of the Tamil, by suffixing ‘a;’ the only difference Il b is, that between the final ‘i of the verbal participle and the relative rians; Whlch ‘a, d’ is inserted euphonically instead of ‘n; eg., ‘mad-i-(d)-a,’ which did, from ‘mad-i, kaving done. This The Telugu agrees with the Tamil in forming its present and e of ioh it preterite relative participles by suffixing ‘a,’ and in inserting ‘n’ between the 1’ in which the preterite verbal participle of that dialect h the invariably ends, and the relative ‘a; e.g., from ‘avu-tunnu,’ becoming, 414 THE VERB. is formed ‘avu-tunn-a,” ¢hat becomes; and from ‘ay-i, having become, £XI18 is formed ¢ay-i-(n)-a,” that became. of t and The suffix of the relative participle of the negative voice of the verb is ‘a’ in Tamil, Malayilam, and Canarese, in Telugu it is lang 2 verl £ny Ma It is now evident that ‘a’ may be regarded as the characteristic relative suffix of the Drévidian languages. The only exceptions are “ni S5 ‘ni,’ the negative relative suffix of the Telugu; the suffix of the aoristic future relative in several of the dialects, viz., ‘ni’ in Ku, the 2 ‘um’ in Tamil, and ‘edu, ‘edi,’ “é or ‘éti’ in Telugu; and “ti’ prec Wil 1 ‘ the sign of the preterite relative participle in Tulu and Ku. Not only are the greater number of relative participles formed by el suffixing ¢a,’ but, as was observed in the section on ¢ The Noun, hawi Ultimate most Dravidian adjectives also receive the same suffix. nouns of quality or relation are capable of being used as adjectives, rela without any change or addition; e.g., ‘sir-u, small, ‘per-u, great; appe but more commonly these nouns are converted into quasi relative of 1 participles, and rendered thereby more convenient for use as adjec- ] tives; e.g., ‘sir-i-(y)-a,” small, ‘per-i-(y)-a,’” great. The preterite relative participles of regular verbs are also frequently used as adjec- ‘gen,h a tives; e.g., ‘uyvar-nd-a,” high, literally that was high ; ¢ tar-nd-a,’ low, tic;al Case, literally that was low. Tamil adjectives like per-i-(y)-a,” agree so exactly with preterite relative participles like °pann-i-(y)-a’ (for will ¢ pann-i-(n)-a’), which made, that they may safely be regarded as preterite relative participles in form, though unconnected with the whi preterite or any other tense in signification. teat Another class of Tamil adjectives receive the suffix of the future 8lve or aorist relative participle, ¢.e., ‘um,” which is suffixed like ¢i-(y)-a’ part; to the crude noun of quality; e.g., ¢ per-um, great, ‘pas-um,” green. of ¢ There is no difference in meaning between these two classes of adjec- parti tival formatives, the use of the one rather than the other being deter- of ) mined solely by euphony or usage; but on the whole ‘um’ is con- 01}y sidered more elegant than ‘i-(y)-a.’ Pond langy Origin of the relative suffixes—The Tamil aorist or future suffix 1 ‘um, has already been shown to be identical with the conjunctive ony or copulative particle. I regard all the other relative suffixes as ang originally signs of the inflexion, or possessive case signs, express- U8g ¢ ing the signification of endowed with, possessed of, having, which has, &e ]&ngl gllag; In the older Scythian languages, a relative participle is used, as in lag } the Dravidian languages, instead of a relative pronoun; and the thoug THE RELATIVE PARTICIPLE. 415 e, existence of a family likeness in so remarkable a particular is a proof of the existence of a family relationship between the Scythian group f the and the Dravidian. The particle which is suffixed in the Scythian 1t 18 languages for the purpose of forming a relative participle out.of a verbal participle, is identical with the sign of the possessive case. In Tistic Manchu this particle is ‘ngge’ or ‘ninge’ (corresponding to the Turkish 1§ are ‘ning’); in Mongolian ¢ don’ or ‘ton: and the addition of this pos- f the sessive case-sign converts the verbal participle (i.e., the theme with 1 Ku, the tense-sign attached) into a verbal adjective or relative participle, | ‘i precisely as in Tamil or Canarese. Thus in Manchu, from ¢aracha,’ written, which is the verbal participle of ¢ara,’ to write, is formed the d by relative participle ‘aracha-ngge,’ whick wrote, literally the-written- Voun, having. imate The language of the Scythian tablets of Behistun has also a tives, relative suffix, ‘pi,” answering to the Mongolian ‘ki,’ which is g/reat f appended, as in the Dravidian languages, to the theme in the formation lative of relative participles. Looking at the analogy of the Scythian languages, and at the adjec- torite genius of the Drévidian languages themselves, I have no doubt that 9 (a7 which forms the most common Dravidian relative suffix, is iden- adjec tical with ‘a,” the oldest and most characteristic sign of the possessive L; Zﬁ‘ll’v e 80 case. The other particles also which are used as suffixes of the relative will be found to have a similar nature. ’ (for od 28 Though the sign of the relative participle in Ku differs from that which prevails in the other dialects, yet ‘ni,’ the sign of the aorist b the relative participle, is identical with the sign of the inflexion or posses- sive case, which is also ‘ni.’ future ‘ni,’ the sign of the negative relative L] participle in Telugu, appears to bear the same relation to ‘ni,’ a sign (v} of the Telugn inflexion. ‘ti, the sign of the preterite relative gl'ééﬂ'. participle, both in Tulu and in Ku, is the most commonly used sign adjec- of the inflexion in Telugu; and the various suffixes of the Telugu deter- g (00 aorist relative participle are apparently adjectival formatives, corres- ponding in origin to ti,” the sign of the neuter inflexion in the same language. quffis Though the use of a relative participle, instead of a relative pro- noun, is characteristic of the Scythian tongues; yet both the Turkish petive The and the Finnish languages possess a relative pronoun as well. 7{65 a8 use of such a pronoun is foreign to the grammatical structure of those {pl'eSS‘ languages, and has evidently been borrowed from the usage of lan- It is certain that the Turkish guages of the Indo-European stock. has been much influenced by the Persian; and the Oriental Turkish, 7&5 in though it has borrowed from the Persian a relative pronoun, rarely d the 416 THE VERB. uses it, and ordinarily substitutes for it a suffixed particle of its own, A in a genuine Scythian manner. of n Formation of Moods. Tel whi The investigation of the structure of the Dravidian verb may now Jess be considered as completed ; for in each dialect of the family the verb par has, properly speaking, only oue mood, the indicative ; and the forms i which correspond to the conditional, the imperative, and the infinitive forr moods of other languages, are verbal nouns or compounds, rather than wil moods. Nevertheless it is desirable, at this point, to inquire into the alx manner in which those moods are formed. but &Hy (1.) The conditional or subjunctive.—In most of the Indo-Kuropean i languages, and even in the Turkish and Finnish, the subjunctive is a Can regularly conjugated mood, distinct from the indicative, with pro- verl nominal terminations of i1ts own. In the Dravidian languages the 01T subjunctive is formed by simply postfixing to different parts of the and verb, either a particle corresponding in meaning to ‘sz’ or ¢f, or the (it conditional forms of the substantive verb, which includes the same tqu particle, and which signifies if it be. Different particles are used for this purpose in the different dialects, and they are not in each dialect if | suffixed to the same part of the verb; but the principle on which oom they are suffixed, and the use to which they are put, are the same in plea all as ! In Canarese the conditional particle is ‘re,” which is derived pro- bably from ¢ir-u,” the theme of the verb fo be: it is appended to the tenit b0 relative participle of the preterite, and that participle being inper- sonal, the condition applies, without change of form, to all persons, [ ey Peth numbers, genders, and times; e.g., ‘méadida,’ tkat did, on receiving the this suffix becomes ‘madida-re,” if (I, thou, he, she, they, &ec.) do, did, or shall do. Person, number, and gender are expressed by the pre- 8y The use ! fixed pronoun, and time by the subsequent finite verb. of the relative participle—a form which always requires a noun to dinle ticle complete its signification—shows that ‘re is regarded as a noun, and fixeg that a closer rendering of the construction would be in the event of (my, your, &c ) doing, more literally in the event that (I, you, &e.) COII]F { ﬂyi t have done (so and so). The most essential and ancient form of the Telugu conditional an g ¢ mn tel‘it{ consists in annexing or ‘ina’ to the ultimate conjugational base; e.g., ¢ chlch-in,” of (I, thou, he, &c.) should see. This “in’ is Irope < evidently identical with the mn which is used for the same purpose of‘a I =5 THE CONDITIONAL. 417 0w, and in the same manner in Tamil; and as the Tamil ‘in’ is a sign < of the ablative or locative, signifying in the event, so must the Telugu i’ or ‘ni’ be identical in origin with the ‘na’ or ‘ni’ which the Telugu uses as a locative. In Telugu the various conditional particles which are in ordinary use are parts of the substantive verb, more or now less regular in form, each of which is used to signify if it be. The verh forms particle which is commonly used for this purpose in the higher dialeet is ‘é-ni,” the conditional form of the verb ‘avu,’ to be or become, a itive form which corresponds to the High Tamil ¢Aj-in,” and means, as than will be seen, in being, i.e., in the event of being. This particle or o the auxiliary, ¢é-ni,” is appended not to the verbal or relative participle, but to the personal terminations of the verb. It may be appended to any tense, as to any person; but whatever tense it is attached to, the opean time of that tense is rendered aoristic, and is determined, as in eI5 4 Canarese, by the context, especially by the tense of the succeeding L 1)1’0- verb. The manner in which ¢éni’ is post-fixed in Telugu exactly s the f the corresponds to the use that is made of ¢4gil’ or ‘4nal’ in Tamil, and of “4yin’ in High Tamil; e.g., ¢ chésitin’-éni, o 7 did or do 1 the (lLiterally df it be (that) I did), and ¢ chésitim’éni, f we did or do, are same equivalent to the Tamil ‘&ejdén-ayin,” if I did, and *§ejdom-4jin,’ od for of we did. alect In the colloquial dialect of the Telugu the conditional particle which commonly used is simply &, which is suffixed, not to any tense at me i pleasure like ¢é-ni,’ but only to the preterite; and is not appended, as ‘é-ni’ is, to the personal termination, but to the root of the pre- | pro- terite, or as I conceive it to be, the old preterite verbal participle; to the e.g., ‘chésit-&,” or rather ¢chést-&, of (7, thou, he, &c.) did or do. impet- I consider this ‘€’ to be either the particle of emphasis, or rather 18018, perhaps an abbreviation of ¢ é-ni,’ the ¢ (S A ¢ A? * of which alone represents oiving the substantive verb. e is equivalent to “avu’ or ‘ayi,’ and ‘ni, )y fhd, 9 as will be seen, is a locative case-sign, equivalent to the Tamil in. [} pl‘e‘ Another mode of expressing the conditional mood in the colloquial o USe dialect of the Telugu agrees with the Canarese in this, that the par- Dﬂn tO ticles are suffixed to the relative participle. The particles thus suf- l, aﬂd fixed are ‘att-ayité’ and att-Ayend; the first part of both which at of compounds, ‘att-u,” is a particle of relation meaning so as, as if. ! ) “ayité’ (‘ayit-’) is the orvdinary conditional of ‘avu,’ o be, being an emphasised form of ©ayi-ti, the impersonal preterite, or old pre- ition&l terite verbal participle of ‘avu.’ ‘ayena, is emphasised from ‘ayenu,’ fional properly ‘ayenu,” it was, the third person of the preterite tense in, iS of ‘avu. urpos? In Tamil, the most ancient and characteristic mode of forming 2B 418 THE VERDB jl&sor pa the conditional mood “is by suffixing the locative case signs in,” to the formed verbal theme, s.¢., that assumed verbal noun which 3l 118 forms the basis of the infinitive and the aoristic defective future mn Thus, from the formed theme ¢ pég-u,’ going, is formed the infinitive pog-a,’ to go, and ‘pég-um,’ wt will go and from the same base by ohe il the addition of the locative, ¢il’ or in,” is formed the conditional From ¢var- coming, 18 Cor pog-il Pog-1m of (I, thou, &c.) go ;% formed var-a, Infinitive, to come ¢ yar-um it will come; and also o w var-il’ or ¢var-in, ¢f (I, &c.) come In like manner, from ag-u, ag-um, it will don being, is formed the infinitive ¢ag-a,” to become or be be,; and also ‘A4g-il,” if (£, &ec.) be 4g-in’ (the equivalent of ‘4g-il’), 1l has been softened into ‘aj-in; and this is identical in origin and he SltOl meaning with the Telugu ¢é-ni’ (‘for ‘avu-m ) referred to above, and is subjoined to the pelson'ﬂ terminations of verbs in the same sha manner as ¢ e-n1 This conditional 1l on ‘In is undoubtedly g identical with €il’ or ¢in,” the Tamil sign of the ablative of motion, which is properly a sign of the locative, signifying in, at, or on; and (A of this ‘in,’ the Telugu equivalent, in accordance with dialectic laws (Z \") mg is “ni,” which is also occasionally used as a locativ This being the case, the signification of ¢ag-il or ‘ay-in, 18 pa evidently in being, i.e., in the event oj being; and this is equivalent to Defy Hence ¢ 4g-il, the phrase of ¢ be. ¢4y-in, and ‘é&-ni, are well suited perf to be used as conditional auxﬂlaues, and &ppended to the various we sl e personal terminations of verbs. m The second mode of forming the conditional in Tamil, consists 1n find the use of the above-mentioned conditional forms of the substantiv ang verb, viz., ‘ag-il’ and ¢Ay-in’ (and also a commoner form an-al’) 9., as auxiliaries to other verbs; and when thus used, they are post-fixed den [4 S to any person of any tense: e.g., like the corresponding Telugu verh, Seydén-agil, if it be that [ did, or if I did, literally in the (event but of its) being (that) 1 did Seyven-agil, of I shall do, literally in the g a’ (event of 1ts) being (that) I shall do This mode of forming the Tamil conditional, though not confined high 1t 18 to the classics, is but rarely used in the colloquial dialect )EI‘\O chiefly used in elegant prose compositions Juye A third form of expressing the sense of a conditional mood in 4 [ Tamil is by appending the particle or noun ‘kal’ to the past relative 8¢ seyda-(k)kal, if (I, &ec.) do or did ‘uvari olitta- participle; e.g., the I (k)kal, if the sea should .roar The conditional form which is form most commonly used by the vulgar, is a corruption of this, viz or even ¢eydakki; and the Ku conditional also is seydakka, I\al’ formed by appending ‘kka ‘kal’ being appended to a relativ fon i THE CONDI TIONAL. 419 or participle, it is evidently to be considered as a noun; and it may hich either be the crude Sanscrit derivative ‘kil’ (for ‘kal-am’), time, ture used adverbially to signify when, a use to which it is sometimes put 1tive in Tamil; or, less probably, the pure old Dravidian word ‘kal,’ a eby channel, a means. The literal meaning, therefore, of ¢&ejda-(k)kal’ 1002 will be, when (I) do or did, a form which will readily take from the Iy 1§ context a conditional force ; e.g., in the following Tamil stanza,— When 180 you have done (“seyda-(k)kal’) @ good action to any one, say not, < when ag: 11, will that good action be returned ?’—it is evident that when you have tuill W done is equivalent to if you have done. The signification of when is ril'), still more clearly brought out by the use of ‘kil’ in connexion with and the future relative participle; e.g., ‘Sej(y)ung-kal, if (ke, they, &c.) hove should do, literally when (they) shall do, or in the time when (they) same shall do. This mode of expressing the conditional mood is exceed- tedly ingly common in the Tamil poets. tion, The fourth Tamil mode of forming the conditional is by suffixing and “al’ to the abbreviated preterite relative participle, e.g., ‘§eyd-al) f [aw ) (Z, &ec.) do. If we looked only at examples like ¢g&eyd-al, we might naturally suppose ¢4l” to be suffixed to the preterite werbal 2 n, 1 participle (¢ &eyd-u’), the final ‘u of which is regularly elided before a vowel ; and this form of the conditional would then nb 10 qited perfectly agree with the second Telugu mode, e.g., ¢chést-é If arions we look, however, at the class of verbs which form their preterite in “i,” and their preterite relative participle in ‘n-a,’ we shall sfs 1 find that ¢Al” is added to the relative, not to the verbal participle, and that the two vowels (‘a and 4’) are incorporated into one; antive ) e.g., the conditional of ¢4g-u,” to be, is not ‘Aag-i-al,” but ‘4an-al, evi- Besides, the ‘/.ﬁXed) dently from ¢&n-a’ (‘4g-i-(n)-a’), that was, and €al’ verbal participle must be followed by a verb or some verbal form ; A (eveni but ‘41’ is a noun, and therefore the participle to which it is suffixed must be a relative participle, not a verbal one. In colloquial Tamil, in e “al’ is suffixed to impersonal forms of the verb alone; but in the higher dialect €&l or its equivalent ‘é&l,” may be suffixed to any )Ilﬁﬂed person of any tense; e.g., ¢ Seydanei-(y)-el, if thow hast done; °Sey- it 8 guvén-él,’ if I shall do. It is also suffixed to the relative participle, as I conceive ‘4l’ is in the ordinary dialect; e.g., ¢$eygindra-(v)-al, 100d iﬂ ¢Seyda-(v)-al, of (I, thou, &ec.) should do. This ‘&eyda-(v)-al’ of the High Tamil illustrates the origin of the more common colloquial olifta- form ¢g&eyd-al. ich i8 This conditional particle ‘4l’ has been corrupted, I conceive, from (Y ) ¢kal’ the particle already mentioned, and is not, I think, to be con- 350 8 cal’ is founded with €al, the sign of the instrumental case in Tamil. gltif? 2.0 .2 420 THE VERDB. rarely used as a sign of the conditional in the higher dialect, in which I ‘kal’ is generally preferred. fl to One form of the conditional mood is expressed by if (e.g., if I do); another is expressed by though, or although (e.g., though I do, or th though I have done). This second form of the conditional is generally g expressed in the Dravidian languages by suffixing the conjunctive al particle to one of the conditional particles already referred to. Thus, 40 in Tamil, ‘éeyd-al’ signifies ¢f (I, &ec.) do; whilst ‘Sejd-al-um’ il signifies though (I, &e.) do. ‘um,” the conjunctive or copulative particle, having the sense of even, as well as that of and, the literal ul meaning of this phrase is even ¢of (1) do. The same particle ‘um’ is II suffixed to the preterite verbal participle to bring out a preterite for signification ; e.g., ‘Seyd’-um,’ though (I, &ec.) did, literally even having tly done. Ta In the superior dialect of the Telugu the conjunctive particle ¢ yu’ of (answering to the Tamil ‘um’) is appended to the conditional par- the ticle, when the reference is to the present time, and to the preterite (L( relative participle (in that case ‘nu’ is substituted for ¢ yu’) when the past time is referred to. (Ul The Canarese adds ‘ri’ and ‘Agyu’ to the relative participle, Pt when the conditional sense is that of although. ‘ri’ 18 ‘re, with the copulative ‘1’ annexed; and ‘Agyu’ is ‘Agi, having been, with and the addition of the same ‘1,’ like the Tam. ¢ AnAl-um.’ Con tak 2. The imperative.—In the Dravidian languages the second person imp singular of the imperative is generally identical with the root or theme of the verb. verl This is so frequently the case, that it may be pers regarded as a characteristic rule of the language. Dort In a few instances in Tamil there is a slight difference between but the imperative and the verbal theme; but those instances scarcely I8 constitute even an apparent exception to the general rule, for the difference is caused not by the addition of any particle to the root, for $poi] un 3 the purpose of forming the imperative, but merely by the softening away of the formative suffix or the final consonant of the theme, for Sign) of i the sake of euphony; e.g., ‘var-u,” to come, takes for its imperative '] ¢va,” Tel,, ‘ra; the plural (or honorific singular) of which is in High the \ Tamil ¢ vammin,” in Telugu ‘rammu.’ It has been seen that there is a class of Tamil verbs which form Samg their transitives by doubling the initial consonant of the sign of tense. time; ~0] Such verbs also, however, use the simple unformed theme as their imperative, and, in so far as that mood is concerned, make no distine- 81y Wi[h( tion, except in their connexion and force, between transitives and Iy THE IMPERATIVE. 421 ch intransitives. Thus, ‘ked-u,” is either spoil or be spoiled, according to the connexion, whilst every other part of the verb takes a form suited 0); to its signification; e.g., the infinitive of the intransitive is ‘ked-a,’ or that of the transitive ‘kedukk-a.” The Telugu, on the other hand, lly generally makes a distinction between the imperative of the transitive Ve and that of the intransitive; e.g., whilst the intransitive be spoiled, is 115, ¢ chedu,’ the transitive is not also ¢ chedu,” but ‘cheruchu’ (for ¢che- Ji} duchu ’), a form which would be ‘kedukku’ in Tamil. 1ve A large number of Telugu verbs use as their verbal theme, not the il ultimate root, but a species of verbal noun ending in ‘chu,” ‘pu,’ or 18 ‘mpu This accounts for the presence of ¢chu,” which is in itself a 1ite formative, in the imperative ‘cheruchu,’ and not only in the impera- i tive, but through all the moods and tenses of the Telugu verb. The Tamil uses the equivalent verbal noun (ending in ¢ kku’) as the base yll of its transitive infinitive, and of the third person singular neuter of par the future or aorist of its transitive; e.g., ‘kedukk-a,’ to spoil, and nite ‘kedukk-um,’ ¢t will spoil; but in every other part of the verb it uses hen the root alone (including only the inseparable formative, if there be Hence it is easler to ascertain the one) as its inflexional theme. ple, primitive, true root of a verb in Tamil than in Telugu. with The ascertained use of a species of verbal noun as the imperative yith and inflexional basis of certain classes of Telugu verbs, leads to the conclusion that every Dravidian imperative, whatever form it may take, is to be considered as a verbal noun. The crude root is the 1501 imperative in Tamil; and yet that even the Tamil imperative is a f Or verbal noun appears from this, that the pronoun which the second person plural of that imperative employs, is not the nominative, or any y e portion of the nominative, of the Tamil pronoun of the second person, ween but the oblique case or genitive, viz., that form of the pronoun which ‘1'ce1y is used in construction with nouns; e.g., ‘ked-um,” spoil ye, or be ye ¢ the spoiled, is formed by suffixing to the verbal theme, not nir,” you, but t, for ‘um,” your,—a proof that the imperative in Tamil has the grammatical ening significance of a verbal noun, and that it is the context and the energy of its enunciation that constitute it an imperative. [ fOl’ afive The particle ‘mu’ or ‘mi,’ is often added to the inflexional base of the verb, or verbal theme, to form the imperative in Telugu. The I]O‘h same practice obtains in the Ku; and even in Tamil ‘mé’ is some- 1 il times suffixed to the singular of the imperative and ‘min’ to the plural, —only, however, in the classical dialect. In Telugu, nevertheless, tense et as in Tamil, the verbal theme is more commonly used as the imperative without the addition of any such particle ; and it seems probable that jstinc g ﬂﬂd ‘mu’ or ‘mi,” the only remaining relic of some lost root, is added as 422 THE VERB. b, an intensitive or precative, like the Tamil ‘én; eg., ‘virum-én,’ Ok do come. kke “andi,” which is added to the root in Telugu, to form the oy second person plural of the imperative, is the vocative of an obsolete U noun, Sirs (sometimes used honorifically to mean Sir); it is probably lar identical with ¢andar,” an honorific plural suffix of the Canarese (e.g., Riu ol of ‘av-andar-u,’ they), and the other signs of the same part of the verb [ 10 in Telugu (“di,” ‘udi,’ and ‘udu’ or ‘du’), are evidently abbreviations 1se of ¢andi.’ a1 ‘min’ (also ‘minir’), the particle occasionally used to form the the plural of the imperative in classical Tamil, is perhaps only a metamor- Db phosed pronoun, and equivalent to ‘nim,” the oblique case of the old ‘nim,” yow,; and probably €in,” the suffix of the imperative plural in tive Malayalam, has been softened from ‘min.’ wit 3. The infinitive.—It has been customary in Dravidian grammars, tha tru especially in the Telugu, to call various verbal nouns infinitives ; as ‘tke mfinibive wn ‘uta,’ 2?9 and ¢the infinitive vn 16| “the infinitive in ‘adam-u, €éd1.’” This use of terms is not sufficiently discriminative ; for though 18 i Infi each of those forms may be used with the force of a guast infinitive in cha certain connexions, yet the two first are properly verbal nouns, and the third is a participial noun: each is capable of being regularly Lay Dih declined, and each possesses a plural. The Tel. ¢padu-ta,’” is identical Com with the Tamil ‘padu-dal, suffering,; whilst the infinitive proper, to suffer, is in both languages ‘pad-a. I have no doubt that the true tha infinitive was originally a verbal noun also, and this origin of the my Dréavidian infinitive will, I think, be proved in the sequel; but the usus loquendi of grammatical nomenclature requires that the term may infinitive should be restricted to those verbal nouns which have ceased thel 0y to be declined, which are destitute of a plural, and which are capable of being used absolutely. ﬂ]io Both in Tamil and in Telugu an infinitive in ‘u’ is occasionally Wl whi used : it is identical with the root in Telugu and with the conjuga- tional theme in Tamil; and therefore seems to be rather a verbal fivy noun used absolutely than a formed infinitive. fror The Malayala infinitive ‘vAn’ or ‘ppén,” is properly a future Ty gerund, and is used as such in High Tamil. W he o The true Drayidian infinitive is generally formed by suffixing ‘a’ the to the verbal theme. This is invariably the mode in which the infi- Suel nitive is formed in Telugu; e.g., € chéy-a,” to do. Ordinarily in Tamil 1ng] and Canarese the infinitive is formed in the same manner; but a verbal noun is also much used in Canarese as an infinitive, with the ang dative case-sign understood or expressed; e.g., instead of ‘mad-a,’ to po A_ E—— 2 dad € THE INFiN ITIVE. do, they prefer saying ¢ méd-al-ke’ (in the modern dialect ¢ mad-ali- Ok kke’), for doing, or (without the case-sign) ‘mad-al’ or ‘méad-alu,’ the doing or to do. Similar constructive infinitives are occasionally used Jete in classical Tamil also, instead of the true infinitive in ‘a; eg., ¢ $ol- bly larku’ (‘éollal-ku’), for saying, and follal, saying, with the sign &5 of the dative understood, instead of ‘8oll-a,’ to say. There is also erh another infinitive, or honorific imperative in € ga,” which is occasionally T used in classical Tamil; e.g., ¢ari-ga,” to krnow, or mayest (thow) know, a form which will be inquired into presently. Notwithstanding the these apparent exceptions, ‘a’ is to be considered as the regular 101 Drévidian sign of the infinitive. old Professor Max Miiller, noticing that the majority of Tamil infini- | tives terminate in ¢ ka, supposes this ‘ka’ to be identical in origin with ‘ku,’ the dative-accusative case-sign of the Hindi, and concludes It is that the Dravidian infinitive is the accusative of a verbal noun. Jars, true that the Sanscrit infinitive and Latin supine in ‘tum’ is cor- ‘the rectly regarded as an accusative, and that our English infinitive to do, e 1 is the dative of a verbal noun; it is also true that the Dravidian ough infinitive is a verbal noun in origin, and never altogether loses that e 1 character ; nevertheless, the supposition that the final ‘ka’ of most and T'amil infinitives is in any manner connected with ‘ku;’ the sign of the larly 2 Dravidian dative and of the Hindi dative-accusative, is erroneous. tical comparison of various classes of verbs and of the varions dialects shows o1, 10 that the ‘ka’ in question proceeds from a totally different origin. 108 The Tamil infinitive terminates in ¢ ga’ (‘ g-a’) only in those cases f the in which the verbal theme ends in a formative ‘gu’ (g-u’) ; and in t the many instances in which g’ appears in the infinitive (as in the verbal tert theme) in the ordinary dialect, ¢ v’ replaces it in the poets : e.g. ‘noga,’ eased to be pained, is not so much used by the classics as ‘néva.’ ‘ppa’ is Pa])le also used in the higher dialect instead of ‘kkaj e.g., ‘madappa,’ fo These interchanges of the formative consonant, walk, for ‘ nadakka.’ nally which is the termination of the verbal theme, and to which the infini- jug tival “a’ is added, are in perfect agreement with the Telugu ; and ;el‘bill from both it is apparent that ‘a’ alone is the sign of the infinitive. 9 Tamil verbs ending in the formative ‘g-u are intransitives; and ufure when they are converted into transitives, the formative is doubled for the purpose of denoting the increased intensity of signitication. In i3 ‘J) such cases the formative ¢ g-u’ is converted into ‘kk-u;’ and accord- o g infl, ingly the infinitive of all such verbs ends in ‘ kk-a.’ Tﬂmﬂ Thus, the verb ¢ pé,” to go, takes ¢ gu’ for its intransitive formative, bu[’, b and hence its verbal theme is ¢ po-gu; from which is formed the acrist th the ‘pog-um,’ @ will go, the verbal noun ° pog-al,’ going, and the infinitive PR i 424 THE VERB. ‘ pog-a, to go. The corresponding transitive verb is ¢ pé-kku,” to drive 1t away (‘gu’ being converted into ‘ kku’) ; and from this is formed in i a like manner ¢pokk-um, ¢ will drive away, and also the infinitive ‘pokk-a,’ to drive away. In some instances the intransitive shape of be the verb has no formative ; and when it is converted into a transitive, 101 the initial consonant of the tense-sign is hardened and doubled : <.e., th ‘gir’ becomes ‘kkir, ‘d’ or ‘nd’ becomes ‘tt, and ‘v’ or‘b’ be- que In such instances the verbal theme on which the infinitive s eomes ‘ pp.’ is constructed takes the doubled formative, ‘kk-u: e.g., compare (‘76 ‘valar-a,’ to grow, with ¢valar-kk-a,” to rear. This formative, ‘kk’ ) oM however, appears not only in the infinitive but also in the aorist ‘valar-kk-um,’ 4t will rear. A very large number of Tamil verbs, we including many transitives, have no formative termination whatever ; Th Wi and the infinitive of such verbs is formed by simply suffixing ‘a’ to the root ; e.g., ‘ var-a,” to flourish, and ‘kan-a,’ to see. In the event of be the root of a verb of this class ending in ‘i’ or ‘ei,” ‘y’ is inserted and between the root and the sign of the infinitive ; e.g., ‘ari-(y)-a, fto Ma know ; ‘adei-(y)-a, to obtain. This ‘y,” however, is clearly euphonic. Mal When an intransitive root is converted into a transitive by annexing ]Zﬁ ‘tt-u’ to the root, e.g., ‘tAr-tt-u,” fo lower, the infinitive simply elides 15 ] the euphonic ‘u’ and suffixes ‘a ; e.g., ¢ tar-tt-a.’ From a comparison of these instances, it appears indubitably certain nf that ‘a’ alone is the normal suffix of the Tamil infinitive, and that the fory ‘g’ or ‘kk’ which so often appears, belongs to the formative of the ver verbal theme, not to any supposititious case-sign. of ¢ What then is the origin of the infinitival suffix ¢ga,” which is conr occasionally used in classical Tamil; eg., ‘ari-ga,’ to know, instead of the the ordinary ¢amri-(y)-a,’ and ‘8ey-ga, to do, instead of ey (y)-a?’ the This form is chiefly used as an optative, or as conveying a wish or Clat] It does not polite command; e.g., ‘ni ari-ga, mayest thouw know! Suff follow, however, from this, that it would be correct to regard it as a Tap form of the imperative originally ; for the ordinary infinitive in ‘a’ is Rppe often used by the poets in the same manner, and not unfrequently even fomy by the peasants. lecth I am persuaded that the ‘g’ of ‘ga’ is simply the usual formative s ‘g’ or ‘g-u’ of verbal nouns, and that its use is primarily euphonie. infly The same formative ‘g’ is found to be used in connexion with other tige parts also of the very verbs which are given as examples of this rule. nfg Thus, not only is ‘ari-ga,” o know, used instead of ‘ari-(y)-a, but Cony ‘ari-g-il-ir, you know not, instead of ¢ ari-(y)-il-ir,” or ‘ari-(y)-ir; and Moy just as ‘Sey-ga, to do, is used instead of ‘Sej(¥)-a,” so we find I ¢ éey-gu-v-én,” I will do, instead of ‘Sey-v-én.’ The ‘g’ which makes i“Ste; & THE INFINITIVE. 425 e its appearance in these instances, is in its origin the formative ‘g-u,” as I appears by the second example; but it is used rather for euphony than five any other cause. It is also to be noticed, that the formative ‘gu’ may 6 of be appended to any verbal root whatever, as a fulerum to the inflex- Ive, ional forms, provided only that the euphony is improved by it, or that Uty the prosody requires it. This view of the origin of the ‘ga’ in be question, is conformed by the evidence of the Malayalam, for in that itive dialect ¢ ga’ is the formative of verbal nouns, answering to the Tamil ‘gei; e.g., ‘wariga, a coming,; and yet the very same form is used as pal'e Kk’ a polite imperative; e.g., ‘ni wariga’ (Tam. ¢ varuga’), mayest thou orist come! Here we see not only a verbal noun used as an imperative, but erbs, we see the infinitive of one dialect treated as a verbal noun in another. ver The Tamil verbal noun which directly answers to the Malayala 10 ‘ wariga, @ coming, is ‘ varugei; and ¢ varuga’ in Tamil has ceased to f 0t [ 01 be used as a verbal noun, and been restricted to the use of an infinitive orted and imperative; but it is evident from the identity of both with the 3,; fo Malayala ¢wariga,’ that both are verbal nouns in origin. The onlc, Malayala € wariga’ is regularly declined ; e.g, ¢ wariga-(y)-al, through xing the coming. We thus come back to the conclusion, that ¢a, not ¢ ga,,’ is the true infinitival suffix of the Tamil. ides On examining the Telugu we shall find that the only sign of the 3.1‘tiiiﬂ infinitive which is recognised by that language is ‘a.’ The various formatives which, as we have seen, are inserted between the Tamil i the f the verbal root and the suffixes of the infinitive, form in Telugu part of the verbal theme itself, and are found not only in one or two connexions, but in every mood and tense of the verb, including ich 18 ad of the imperative. In Telugu, therefore, the only difference between the imperative and the infinitive is, that the latter elides the enun- ! ish or ciative ‘u’ of the former, and substitutes for it its own distinctive sufix ‘a.’ 5 nob Thus, whilst the imperative of the verb ¢o open, is in Tamil ‘tira,’ and the infinitive ‘tira-kk-a; the formative ‘kk ’ which j s ‘a,’iS appears in the Tamil infinitive, and which might be supposed to form part of the infinitival suffix, appears in Telugu (in its dia- / even lectically softened form of ‘ch’), not only in the infinitive, but ﬁative also in the imperative and throughout the verb ; e.g., ‘tera-ch-a, infinitive, fo open,; ‘tera-ch-u,’ imperative, open thouw. At the same Jonic ] othe I time, the Telugu sign of the dative case ‘ku’ or ‘ki’ is never softened into ‘ch’ in any connexion ; consequently, there is no possibility of s vule: connecting the Telugu sign of the infinitive with that of the dative. .a” bub ') and Moreover, the formative ‘ch’ is often replaced, especially in the imperative and infinitive, by ‘p; eg., ‘nadu-p-a,’ infinitive, to walk, 0 ﬁud mukC‘S instead of ‘nadu-ch-a,” corresponding to the Tamil ‘nada-kk-a,’ of 1 426 THE VERB. Hence, it cannot be fre which the imperative and theme is ‘nada.’ doubted, that the Tamil ¢ g’ and ‘kk,” and the corresponding Telugu be “cl’ and ¢p,” alternating (after ‘i’) with ‘nch’ and ‘mp,” are merely formatives, without any special connexion with the suffix of the th infinitive, which is ‘a’ alone. il In most instances in Canarese the formatives referred to above are fo discarded altogether, and the ¢a’ which constitutes the sign of the infinitive is suffixed to the crude verbal root. Thus, whilst the verb th pr “ir-u,” to be, takes ‘iru-kk-a’ for its infinitive in Tamil, the simple? Canarese infinitive is ‘ir-a.’ C0] inf Origin of the infinitival sufix ‘a’— 1 conceive that ‘al’ was originally the sign of the infinitive in all the Dravidian dialects, and ) that ‘a’ is a weakened form of ‘al’ Several analogies may be to adduced which render the softening of al’ into ‘a,” not only possible, ‘b but probable. It has been shown that a,’ the verbal sign of negation, “ti The is probably derived from ‘al,’ the ordinary negative particle. following analogy is more decisive. Much use is made in Tamil of a verbal or participial noun ending in ‘dal;’ e.g., ‘alei-dal,’ ¢ wandering, vid from ‘alei,’ to wander; ¢ muri-dal, @ breaking, from ‘muri,” to break. a8 In Canarese the final ‘1’ of those and similar verbal nouns has lan systematically disappeared; eg., ‘ale-ta,; a wandering, ‘muri-ta,’ « to breaking. Dy The Telugu also has softened away the final <1’ of the same class oy of words; e.g., compare the Tamil ‘ méy-(t)tal, pasturage, with the m corresponding Telugu ‘mé-ta; ‘ché-ta,” Tel., an act, with ¢gey-dal, dll Tam. ; and ‘nada-ta,” Tel., walk, conduct, with ¢nada-(t)tal, Tam. g Even in Tamil also, ¢nada-(t)tei’ alternates with ¢ nada-(t)tal.’ ady It bas already been stated that the verbal noun in ‘al, with or fof without the dative case-sign, is used instead of the infinitive in ‘a’ in | nif both dialects of the Canarese and in classical Tamil. In Gond also, Opt the sign of the infinitive is € alle;’ ampliﬁed from ‘al; e.g., ‘aille,’ ¢o il be, which is evidently identical with the Tamil verbal noun, ¢agal, f being—a form occasionally used in the higher dialect as an infinitive. ll'llp Now, as the Dravidian infinitive undoubtedly partakes of the character se) of a participial or verbal noun, and is considered by native gram- beey marians as a verbal participle or gerund of the present or aorist tense; ki as it is certain that it is intimately associated with a verbal noun in Tay ¢al,” one of the most characteristic in the language, and which denotes Vid not the abstract idea of the verb, but the act; and as ‘al’ in other hug connexions has been softened into ‘a,” we seem to be justified in coming 0rg to the conclusion that ¢a, the infinitival suflix, has been weakened gy THE INFINITIVE. 427 e from ‘al and, consequently, that ‘4g-a,” to be, is identical with ‘4g-al,; ugu being. rely A parallel instance of the softening away of the final consonant of the the infinitive appears in the Indo-European tongues. The sign of the infinitive is in Persian and Gothic ‘an,” but in Frisian ‘a; e.g., ‘mak-a,’ ) are to make. the The present infinitive of the Finnish is apparently identical with verb the Dravidian, being also in ‘a; eg., ‘oll-a’ to be; but it appears IIET probable that this ‘a was originally preceded by a nasal, for the correspouding Esthonian infinitive is ‘olle-ma,” and the sign of the infinitive in Hungarian is ¢ ni.’ Was On the supposition that the Dravidian infinitive terminated origi- and nally in ‘I, there is a remarkable, but probably accidental, resemblance | be to it in the Armenian, in which ‘1’ is the infinitival suflix; e.g.s ible, ¢ ber-e-l, to carry (compare Tam. ¢por-al,’ bearing or to bear); fi o, “ta-l,” ¢o give (compare Tam. ‘ta(r)-al,” giving or to give). Tl 16 | ofa Use of the infinitive.—1If we look at the force and use of the Dra- i, vidian infinitive, we shall discover conclusive reasons for regarding it peal as a verbal or participial noun. It is not only used as in other b 128 languages to denote a purpose or end, e.g., ¢ var-a (§)Sollu,” tell (him) , 4 to come, but also in the following connexions. (i.) The majority of Dravidian adverbs are infinitives of neuter verbs; e.g., he knocked class down, would be in Telugu ¢ pada gottenu,” in Tamil ¢vira (t)tallinan ; | the in which phrases down means to fall, i.e., so as to fall. Through the dal, same idiom ¢ Ag-a,” the infinitive of the verb ¢o become (in Tel. < k4, or Tam. ¢gh’), is ordinarily added to nouns of quality to convert them into adverbs ; e.g., ‘nandr’-aga,” Tam. well, from ¢ nandr-u,” good, and ‘ ag-a,’ th or to become. (ii.) The infinitive is elegantly used with an imperative sig- a’in nification (in accordance with the Hebrew idiom), or rather as an optative, seeing that it conveys a wish rather than a command; e.g., {1150» ‘ni var-a (more frequently ¢vér-ga’), mayest thouw flourish! The le, fo ﬁgnj; infinitive of the verb to be, also regularly forms an optative, or polite tive: imperative, by being annexed to the future tense of any verb; e.g., ¢ seyvay-aga, mayest thow do, from ‘Seyvay, thou wilt do, and ¢ Aga,’ to racter become, literally, may it be (that) thow wilt do. (iii.) It is used as a gram- onse kind of ablative absolute; e.g., ‘porudu vidind’ irukk-a, én tingugiray,’ un ill Tam., the sun having arisen, why sleepest thou? In this instance, ‘vidind’ irukk-a,’ (literally ¢o be—having arisen,) is in the perfect tense, Jn{)tes but ¢irukk-a’ is not a preterite infinitive, but is the ordinary present other or aorist infinitive of the verb ¢ir-u,” fo be. The infinitive used in this ming kcHC‘.1 manner is styled a verbal noun in Malayalam, and is capable of being 428 THE VERB. 101 regularly declined, which proves that it is a noun. The Malayalam prefers to use as an infinitive ¢ van,” which is properly a gerund of the il (a future; but some of the uses to which it puts its verbal noun in ¢ ga,’ ‘ka,” or ‘kka, show that that verbal is identical with the Tamil tho infinitive ; e.g., ‘ellawarum kélkka,’ Mal., (Tam. ‘ellarum kétka,’) in cip the audience of all is literally so as that all should hear, or, whilst all Jur were hearing. (iv.) A series of infinitives is often elegantly used, 1101 somewhat as in Latin, to express minor actions that take place pal contemporaneously with the principal action; e.g., they would say in to Tamil ‘mugil erumba’ (whilst the clouds were rising), ¢ vanam irul-a’ (whilst the sky was gathering blackness), < marei porindu pey(j)-a’ bei (whilst the rain was falling abundantly), ‘Grar tiru-vird nadattinirgal’ ukl (the villagers celebrated their sacred festival). (v.) The reduplication of any infinitive expresses exactly the force of the Latin gerund in ‘do e.g., - pog-a pog-a, balan kollum, ‘wires acquirit eundo,” more closely, tha as it goes—as it goes (literally to go—to go) it gathers strength. These illustrations prove that the Dravidian infinitive has the force par 151 of a gerund or verbal participle, or of a verbal noun, as well as that of the infinitive properly so called. The examples adduced are all from of ¢ the Tamil, but parallel examples could easily be adduced from each of par the other dialects. itde I by Formation of Verbal Nouns. dist the Dravidian verbal nouns divide themselves into two classes, viz.— participial nouns, which are formed from the relative participle, and exc am verbal nouns of each tense and retaining the time of the tense to which Ong they belong, which are always formed directly from the theme, and are determinate in point of time. tive. 86y 1. Participial mouns.— The greater number of nouns of this class g are formed by suffixing the demonstrative pronouns, or their termi- g nations, to the present and preterite relative participles; e.g., from how ‘feygira,’ that does (the present relative participle of ‘&ey, to do), is regu formed ‘Seygira-(v)-an,” ke that does; ‘seygira-(v)-al, she that does, &ec. ﬁpm In like manner from the past relative participle ‘&ejda, that did, is time formed ‘$eyda-(v)-an,’ ke that did ; ‘seyda-(v)-al,’ she that did, &e.; and the { by simply adding the appropriate terminations, participial nouns of the } any number or gender (but always of the third person only) may be made fory at pleasure. A similar series of future participial nouns exists, or may the be constructed if required ; e.g., ¢ 6duvan,’ ke who will read, or is accus- loyy tomed to read. The Tamil future in v’ or ‘b’ is destitute of a rela- Not tive participle ; but its existence is implied in that of future participial g ( VERBAL NOUNS. 429 alam nouns, like ¢pdva-du,’ that which will go, and ‘kanba-(v)-an,” ke who f the will see, and must have ended like the future relative participle of the & Canarese, in ‘ va,” ‘ba,” or ‘ppa.’ The Tamil aoristic future in ‘um,’ [amil though a relative participle as well as a future tense, forms no parti- ) 1 cipial nouns, probably in consequence of ‘um’ being in reality a con- st all junctive particle, not a true suffix of relation. Negative participial used nouns of each number and gender are formed exactly like the affirmative place participial nouvns, by suffixing the various demonstrative terminations Ay 10 to the negative, instead of the affirmative, relative participle. -4 These participial nouns are declined like other nouns; nevertheless, j)a being parts of the verbs, they have the same power of governing nouns freal as the verbs to which they belong; e.g., ‘vittei (k) kattinavan- jon of ukku,’ ¢o him who built the house. In these respects all the Dravidian dialects are so perfectly agreed “do; o3¢ that it is needless to multiply quotations. There is a peculiarity about the words which are used as neuter 01¢6 participial nouns in Tamil which requires to be noticed. Each of them ; hat of is used in three different significations, viz.—as the third person neuter of the verb, as a neuter relative-participial noun, and as a verbal- | from ach of participial noun. Thus, ‘&eygiradu’ in the first connexion means it does ; 1 the second, that which does; in the third, the doing or to do. I have termed it in the third connexion ¢a verbal-participial noun,” to distinguish it from the ordinary verbal nouns, which are formed from the theme, not from participles, and from which the idea of time is vin— excluded. I It is a verbal noun in use, though participial in origin. p, and am persuaded that of these three senses the original and most correct ) Whi(:h one is the last, viz., that of the verbal-participial noun ; for the rela- 16, and tive-participial noun ought by analogy to be ‘seygira-(v)-adu,” not ‘Seygir-adu ;’ and whilst it is certain that a participial or verbal noun s class might easily be used as the third person neuter of the verb, in accord- ance with the analogy of many other languages, it is difficult to see feri how the third person neuter of the verb could come to be used so ; from regularly as it is as a verbal-participial noun, This species of parti- | do), i cipial noun, though neuter or without personality, includes the idea of Oes‘ &C. did, time : it has three forms, in accordance with the present, the past, and the future tenses of the verb ; eg., ‘§eygiradu,” the doing; ‘Sejdadu,’ 0 ad Each of these the having done ; and ‘Seyvadu,’ the being about to do. ouns of forms may be pluralised, as far as usage permits, when it is used as p made the third person neuter of the verb, or as a relative-participial or may noun ; but when used abstractly as a verbal-participial noun it is g qoetls” not pluralised. The participial noun formed from the future is 10 1'913‘ one of the most commonly used forms of the verbal noun in Canarese, riiﬁiPi"ll VR 430 THE VERB. 1 e ‘i\liyu-v—adu,’ or ‘ilivu-du,” the act of descending, from €ili, to descend. a1 Words of this kind have sometimes been called infinitives ; and it is true that they may generally be rendered in the infinitive on trans- fl fn lating them into English, e.g., ‘appadi seyjgiradu sari (y) alla,” Tam., it s not right to do so. But this is simply because the English infini- U tive itself is sometimes used as a verbal noun, and ¢o do is equivalent Ir to the participial noun, the doing. The phrase might be more closely Verbal nouns of this class rendered, the doing thus (is) not right. i ka become more allied to infinitives when they are put in the dative ; 6.1, ‘seygiradu-kku,” for the doing, i.c., to do. As the pronoun ‘adu’ Wi becomes optionally ‘adan, so the participial noun ‘Sejgiradu’ may an become ‘seygiradan.’ This change, however, is exceedingly rare except in the dative ; and in that connexion ‘§ejgiradan-ku,” euphoni- an cally ‘&eygiradar-ku,” is more common in written compositions than Wo ‘seygiradu-kku.’ i The Tamil alone possesses an abstract relative-participial noun, ex- i pressing in the form of a declinable participle, the abstract idea denoted Vo by the verb. It is formed by appending ‘mei,” the suffix of abstracts, S to the present or preterite relative participle of any verb: e.g., from In] ‘irukkindr-a,” that 7s (the present relative participle of ¢iru,’ o be), by the addition of ¢mei’ Tamilians form “irukkindra-mei,” being. i The use of this form is confined to classical compositions ; but the ver abstract derivative nouns which are formed by annexing ‘mei’ to the lite crude verbal theme (e.g., ‘poru-mei, patience, from ‘poru,’ to bear) lge are much used even in the colloquial dialect. The relative-participial thog noun in ‘mei, whilst it is declined like a uoun, has the governing it power of a verb ; but the corresponding verbal derivative in ‘ mei’ has al the force of a substantive only. for The Tamil suffix ‘mei’ is ‘me’ in Canavese, ‘mi’ in Telugu. In tive several of the Scythian tongues we find a suffix used which bears a fory considerable resemblance to this. The suffix of the participial noun ! na@ in Finnish is ‘ma’ or ‘mA : in Esthonian ‘ma’ is the suffix of the stan infinitive : supines are formed in Finnish by suffixing ‘man :’ the gt Turkish infinitival suffix is ‘mak’ or ‘mek.’ Possibly we may also also compare with this Drividian ‘me’ or ‘mei,” the old Greek infinitive g in ¢ pev,” and such nouns as ¢ woly-pa, ¢ dec-po-s,” and ¢ oxio-un,’ each tion of which exhibits an old participial suffix. byu ang 2. Verbal nouns.—Dréavidian verbal nouns are indeterminate with Clrey respect to time, being formed, not from participles, but from the verbal Verh root or the formed theme ; but they express the act, not the abstract tong; VERBAL NOUNS. 431 fo idea, of the verb to which they belong, and hence are called by Tamil grammarians ¢ toril pejr,’ nouns of operation. dit Verbal nouns are carefully to be distinguished from verbal deriva- WS tives, or substantives derived from verbs. The latter, though derived I, from verbs, are used merely as nouns; whereas the verbal noun, fint properly so called (like the participial noun), is construed as a verb. lent In several Dravidian grammars written by Europeans this distinction sely has not been attended to; and Tamil words like ‘nadei’ or ¢ nadappu, class walk, have been classed with verbal nouns like ‘nadakkei,” ¢nadak- 0.0y kudal, and ‘nadakkal, walking. Though, however, each of these o words may be translated walking, the first two are simply substantives; may and adjectives, not adverbs, must be used to qualify them; whereas 118 ‘nadakkudal,’ the corresponding noun of operation,is a true verbal noun, o0l and is qualified by adverbs, precisely as the verb itself, ¢ nada,’ to walk, than would be. Thus, we can say ‘nidi(y)4dy nadakkudal, acting or walk- wng Justly ; but we could not use the adverb ¢nidi(y)ay’ to qualify ex: either ‘nadappu’ or ‘nadei.’ It would be necessary to qualify those oted words by the adjectival form °¢nidi(y)ina,” there being nearly the acls, same difference between ‘nadappu’ and ° na@akkudal’ that there is in English between bekaviour and behaving. from A verbal noun in ‘gei’ or ‘kkei’ is often used in Tamil, e.g., ) le), et “irukkei,” the being, Seygei, the doing ; but though this is used as a verbal noun, e.g., “appadi irukkei-(y)-al,’ seeing that 4t is so, more the o the literally ¢hrough its being so, yet the forms which are most commonly used as verbals, and which have the best claim to that character, are bear) those which terminate in ‘al i’ e.g., ‘Sej(y)-al,” or ‘Sey-dal,’ doing ; cipial g ‘nadakk-al’ or ‘ nadakkudal,” walking. Whether the suffix appended be “al’ or “dal,’ it is generally suffixed, not to the crude root, but to the - has formed verbal theme, i.e., to that which forms the basis of the infini- Tn tive and of the defective future or aorist: e.g., the verbal noun that is formed from ‘ir-u,’ o be, is not ¢ir-al,’ but ¢ iru-kk-al,’ being ; and from 18 & ‘nad-a,’ to walk, is formed not ¢na-d-al,” but ‘nada-kk-al’ Notwith- nout f the standing this, ‘al’ or ¢dal’ is sometimes added directly to the ulti- mate base: e.g., not only have we ‘pédg-al’ or ‘ pogu-dal,’ going, but * the also ‘po-dal ;7 and not only ‘4g-al’ or ‘Agu-dal, becoming, but also / alSO ¢a-dal. Probably, however, in these instances the right explana- pitive tion is, that the formative ‘g’ of ¢ po-gu’ and ‘4A-gu’ has been softened ' ach by use. The ¢d’ of “dal’ is clearly a formative of the same character and force as the ‘g’ of “gei’ or ‘kkei;’ and this is proved by the circumstance that the ¢d’ is doubled and converted into ‘tt > when the ; With verb becomes a transitive instead of an intransitive, or when euphonic vel‘bal considerations require: e.g., comp. ‘kurei-dal,” intransitive, a being >sh‘ﬂ0t o 432 THE VYERB. fro cw*tailed, with ¢ kurei-ttal,’ transititve, a curtailing. It is evident COL that this ‘d’ is unconnected with the ¢d’ which constitutes the sign 16 of the preterite tense of many verbs; for the verbal noun in ‘dal’ is fon as indeterminate with respect to time as thatin ‘al’ or that in ‘kkei;’ V0 and the corresponding Telugu forms are ‘ta’ and ‘dam-u;’ eg., son ‘chéyu-ta’ or ¢ chésu-ta,’ or more commonly ‘chéya-dam-u,” doing. oD The distinction which has been shown to exist between verbal i nouns, properly so called, and verbal derivatives, furnishes, I con- ceive, some confirmation of the hypothesis that the infinitive in ‘a’ the has been softened from al,” the Tamil suffix of verbal nouns. it fon 8. Verbal derivatives.—It is not my intention to enter fully into e the investigation of the formatives of werbal derivatives, or substantives derived from verbs, most of those formatives being merely euphonie, qua it and their number in the various dialects being very great. It may be desirable, however, to direct the reader’s attention to a few of the al more characteristic and interesting modes in which the Dravidian of languages form nouns of this class. (1.) The first class of derivative nouns (if indeed it is correct “nd “nd to consider them as derivatives) consists of those that are identical with verbal themes ; e.g., compare ‘katt-u, a tie, and katt-u,’ ¢o ¢ie. fror bees (2.) Some verbal themes become nouns by the doubling and (o hardening of the final consonant; e.g., ‘erutt-u,” @ letter, from ¢ erud-u,’ to write; patt-u,” a song, from ¢pad-u,’ to sing. This is especially a V1 Tamil method of forming derivative nouns, for some of the corres- ponding Telugu nouns are formed differently; and where they do the resemble the Tamil, the resemblance consists only in the hardening, of { (or and not also in the doubling of the final consonant; e.g., ¢ pata,” Tel., a song, from ¢ pad-u,’” to stng. The Telugu differs also from the Tamil and e in changing the final or enunciative ‘u’ of the verbal root into ‘a.’ Compare ‘ét-a, play (Tam. ‘att-u’), from ¢ad-u, to play. The uvi Tamil mode of doubling, as well as hardening, the final consonant, seems most in accordance with Dravidian analogy; for it is when a ohj sonant is doubled that it is naturally converted into a surd, and when i 1t iy it is not doubled, it should be pronounced as a sonant. It is remarkable how many purposes are served by the doubling Byl of Dravidian final consonants. (i.) Tt places substantives in an El'll‘l adjectival relation to succeeding substantives; (ii.) it converts intran- iS U sitive verbs into tramsitives; (iii.) it forms a sign of the preterite di&h tense; and (iv.) it forms derivative nouns from verbal themes. (3.) A remarkable mode of forming derivatives is that of lengthen- and ing the included vowel of monosyllabic verbal roots: e.g., in Tamil, fiat Y I — sdEEEEE—— VERBAL DERIVATIVES 433 dent from ¢pad-u, to suffer, comes pad-u,’ sufering; from ‘min,’ to glitter, g comes ‘min, a star. Nor is this method found only in the classies : 5 it appears in words of the most familiar class; e.g., ¢ nikk-u,’ the kel ;' tongue, from ‘nakk-u,’ to lick. The Tamil simply lengthens the root vowel in forming derivatives of this class, and leaves the final con- 60, sonant unchanged; but the Telugu and Canarese harden the final ethal congonant, in addition to lengthening the root vowel; e.g., from ~ < ee ¢ pad-u,’ to suffer, they form not ‘pad-u,’ but ¢ pét-u,” suffering. 000 the section on ¢ Roots.’ e (4.) Abstract nouns are formed from verbal themes by suffixing The same suffix ‘mei; e.g., < tAr-mei,” humility, from ¢ tar,” to be low. forms abstracts also from nouns of quality or relation and pronominals; Info e.g., peru-mei, greatness, from fper-u, great, and ¢tan-mei,’ nature, tives quality, from ©tan,’ wself, literally self-ness This suffix is in Telugu onic, ‘mi; e.g., ‘kali-mi,” wealth, from ‘kalu-gu,’ fo accrue. Y he (5.) Many nouns are formed from verbs in Tamil by suffixing f the ¢am,” and at the same time doubling and hardening the final consonant idian of the verbal theme. ‘ng’ being the equivalent of ‘g’ ‘nd’ of ‘d; rrect ‘nd’ of °d, and ‘mb’ of ‘b, ‘ng’ on being doubled becomes ‘kk,’ ‘nd’ becomes ‘tt; ‘nd’ becomes ‘it and ‘mb’ becomes ‘pp;’ e.g., tical from ¢ tlng-u, fo sleep, is formed ¢ tikk-am,’ sleep ; from ¢ tirund-u,’ fo fo tie. become correct, comes ©tirutt-am, @ correction; from ¢tond-u, to dig, and comes ‘tott-am, a- garden; and from ¢virumb-u,’ fo desire, comes -1, ¢ yirupp-am,’ a desire. lly & In most instances the Telugu (and the Canarese always) rejects ITES~ the final ‘m’ of nouns of this class; e.g., ¢ ting-a,” Tel., sleep, instead Y do of the Tamil ¢tikk-am.’ Though the final consonant, if ‘g’ “d, ‘b’ ing, (or their equivalents), is always doubled before this ‘am’ in Tamil Tel., and Malayalam, verbal themes which end in other consonants often Jamil become nouns by simply annexing ‘am; eg., ‘uyar-am,’ keight, from )‘a. <uyar,’ to be high; ‘Ar-am,’ depth, from ‘ar, fo be decp. The Possibly this ‘am’ is derived from the Old Canarese sign of the )Il?lﬂt’ objective case. ‘am’ being in that dialect a sign of objectivity, it en & might naturally be used as a formative of abstract neuter nouns; and when it is certain that this would be in accordance with analogy, for this is evidently the reason why the nominative singular of so many Indo- ibling European neuter nouns is formed by suffixing ‘am’ or ‘um, which 4 is properly the sign of the accusative. ntral- (6.) A vast number of verbal derivatives in all the Dréavidian terite dialects, are formed by suffixing to the verbal themes those favourite and multifariously used formatives, ‘g, ¢d,” ‘b,” under various modi- fications, and with various vowel terminations. ;theﬂ‘ 2 F — — ——— s 434 THE VERB. {ire i. The ‘g’ formative becomes in Tamil ‘gei, e.g., $ey-gei, an action, from < &ey, to do, it is nasalised to ‘ngei, e.g., ¢ ka-(ng)gei,’ “u( SR heat, from ‘kay, to burn; or is doubled and hardened into ¢kkei,’ (Tar e.g., ¢ padu-kkei, @ bed, from ¢ pad-u,’ to lie. The corresponding g Canarese formatives are ‘ke’ or ge, with not unfrequently the chan [3 1. prefix of an euphonic The Telugu nouns which take this for- ‘s mative terminate in ‘ka’ or ‘ki; eg., ‘éli-ka, government, from of fe ¢ él-u, to govern, and ¢ uni-ki,’ residence, from ¢ undu,’ to be, to dwell. 18 ¢ ii. The ¢d’ formative is in Tamil ‘di; e.g., ‘kedu-di,” ruin, from ufl ¢ ked-u, Zo spoil : being doubled and hardened it becomes tti;’ refer e.g., ‘unar-tti, sensibility, from ‘unar, to feel, to be sensible. This also ‘tt’ is generally softened into ‘chi; eg., pugar-chi’ (instead of used ‘ pugar-tti *), praise, from ‘pugar, to praise. This formative 1s ‘t’ “kay instead of ‘d’ in Canarese and Telugu. It appears in Canarese under Sang the forms of ‘ta’ and ‘te; eg., ‘hogal-te,” praise, from ‘hogal ’ (Tam. o} ‘pugar’), to praise; ‘ kay-ta, producing fruit, from ‘kay, to fruit. I In Telugu we find ‘ta’ ‘or ‘ta’ and ‘ti; eg., ‘alafa-ta, fatigue, spruy from ‘alay-u’ (‘ala-u’) to be tired; °tin-ta,’ ecating, from ‘tin,’ to S0me eat ‘mi-ta,” a lid, from ‘mi-yu,’ fo shut; and ‘nadi-ti,) conduct, diree from ° nadu-chu,’ to walk.’ have iil. The ‘b’ formative is in Tamil generally softened into ‘v, d.e., Ity 3 (4 Vi or ‘vu; eg., ‘kél-vi, hearing, from ‘kél,” to hear, and ‘marei- from In some instances, how- vu, concealment, from ‘marei,’ to conceal. ( ever, ‘b’ is euphonised into ‘mb’ (‘mbu’); e.g., ¢ vé-mbu,’ the Mar- arg | gosa tree, from ‘veé-y,” to be umbrageous; ‘pa-mbu, a snake, from confe ¢ pA-y, to spring. ‘b’ cannot retain its proper sound before a vowel, @ Py and when single either becomes ‘v’ or ‘mb; and that the ¢vu’ hecay which is so common a formative in each Dravidian dialect was Tiatiy softened from ¢bu,” appears from the circumstance that when it is cu) | doubled it becomes ‘ppu;’ e.g., ‘nada-ppu,” @ walking,; ‘iru-ppu,’ @ thegs betng ; ¢ mu-ppu,” old age. In Telugu this formative is ‘vu,” “vi,’ or pomnt ‘pu; e.g., ‘cha-vu, death, from °cha-chchu,’ to die (corresponding appl Tam. and Can. ‘§i-vu,” from “84°); °digu-vu, ¢the bottom, from “dig-u, to descend; ‘teli-vi,” understanding, from ¢teli-yu,’ to know (2 dey ¢ chéru-pu,” nearness, from ¢ chér-u,” to draw near; ¢ édu-pu,’ a weeping, cly from ¢édu-chu,” ¢o e¢ry (corresponding Tam. ¢ ara-ppu,’ from ‘ara’). The Canarese uses in this connexion ‘va’ alone; eg., ‘ira-vu,’ a kil (3 being, corresponding to the Tamil ¢iru-ppu.’ oftey that 4. Nouns of agency or operation.—The participial nouns of the (]efect Dravidian languages are largely used as nouns of agency; but such nouns are also formed in each of the Dravidian dialects in a more (4 DF&\] NOUNS OF AGENCY. 435 g > direct and primitive manner by suffixing ‘1’ to the verbal root; e.g., lgel, ‘an(n)-i’ (Tam. and Can.), an eater, from ‘un, fo eat; ‘kol(l)-i’ kkel, (Tam. and Can.), a killer, from ‘kol, ¢o ill. The Dravidian lan- ding guages in borrowing feminine derivative pouns from the Sanserit, ¢ A 9 y the change the final 1 of the Sansecrit feminine into short ¢i; eg., g for- ¢sunda-ri,” Sans., ¢ fair woman, becomes ¢ sundari.’ But this final ‘i’ from of feminine derivatives, which is directly borrowed from the Sanserit, I is not to be confounded with the more distinctively Dravidian ¢i,’ by from suffixing which nouns of agency or operation are formed, without ‘i reference to gender, whether masculine, feminine, or neuter. Tt is This 9 also to be distinguished from the ‘i which in Sanserit is sometimes ad of used as a suffix of noums of agency, gemerally masculines, e.g., is ‘4’ ‘kav-i-s,” a poet, literally a speaker, in borrowing which from the under Sanserit, the Dravidian languages invariably reject the sign of the (Tam. nominative, and use the crude theme (e.g., ¢ kavi’) instead: it Possibly i, the Drividian suffix of nouns of agency, may have e, sprung from the same origin as the ‘i’ by which similar nouns are in, 10 sometimes formed in Sanserit; but it is certain that it has not been ndicty directly borrowed from the Sanscrit, and it does not appear even to have been introduced into the Drividian languages in imitation of it. v; i,e., Its independence of a direct Sanserit origin will sufficiently appear naRel- from the following statement of the manner in which it is used. )y how- (1.) Dravidian nouns of agency which are formed by suffixing ‘i, o Mor- are destitute of gender: their gender depends entirely upon the , from context; e.g., ¢ panei-(y)-ér-i,” Tam., a Palmyra climber (from ¢ panei,’ VOW@L a Palmyra, and ¢ ér-u,’ fto climb), may be considered as masculine, 0 (‘ru, because men only are climbers of the palmyra; ‘man-vett-i,” Tam., a i Was native spade, a hoe (from ‘man,’ the ground, and ‘vett-u,” to dig or 3 it 18 cut), is in like manner neuter by the necessity of the case: but both ¢ these nouns, and all similar nouns, when regarded from a grammatical i, OF point of view, are destitute of gender in themselves, and may be onding applied at discretion to objects of any gender. from ) (2.) Nouns of agency may be formed in this manner from primitive, ] l[ TLO”-’I )' underived nouns, as well as from verbal roots; eg., ¢ nir-kal-i, Tam., !lf/fp iﬁf/ ) : a chair, literally that which has four feet, from ‘nal-u, four, and .'Ma ’). ‘kal, a foot. {"wl;" 0 (8.) When nouns of agency are formed from verbs, the suffix is often added, not to the crude root, but to the conjugational theme, or that form of the root which appears in the infinitive and in the defective aorist; e.g., ‘ung-i, Tam. (as well as ‘ un(n)-i’), an eater. ot suc (4.) My chief reason for regarding this suffix as a true and ancient 9 mOI’e Dravidian form, and as independent of the Sanserit, whatever may LR 2 436 THE VERB. have been its ulterior relation to it, consists in the very extensive use which is made of nouns of agency formed by means of this suffix, not only in the Tamil classics, but also in the language of the peasantry. It appears in the names of plants and animals, in the names of many of the objects of nature, in old compounds, in proverbs, in nicknames, in the very highest and in the very lowest connexions, and to a much larger extent in all these varieties of use, than in Sanserit. The following Tamil examples cannot be supposed to have been derived from Sanscrit precedents :—¢kal(l)-1,” cactus, from kal,’ toddy, sweet sap; < vel(l)-i, silver, from ‘vel to be white; ¢ puld, the cheetah, or 9 the sea, from ¢uvar,’ saltness; leopard, from ‘pul, small; ¢uvar-i “ar-i,) the sea, from ‘ér-u,’ to be deep. Compare also the following compounds, ¢ vari-katt-i,) a guide, literally a way-shower ; ¢ vanam- pad-i,’ the lark literally the heaven-singer ; ‘ tottal-vad-i, the sensitive Tt plant, literally if (one) touch, the witherer, or as we should prefer to msol say, touch-me-and-I-wither. out I Of gra AvverBs.—It is unnecessary in a work of this kind to enter into fallen the investigation of the Drividian adverbs, for, properly speaking, & com the Dravidian languages have no adverbs at all: every word that is used as an adverb in the Dravidian languages is either a verbal theme, less ) and Ve or the infinitive or gerund of a verb; and illustrations of the manner in which those words acquire an adverbial force will be found in the ad()pt { overly ordinary grammars of each of the Dravidian dialects. the b and (] Tegar] ducted; and gt Whig) Drivig With p Burgp, gtoup, and g Ve use ¥, not antry. - many names, a uch . The derived SECTION VII. !y, suect il oF ——— altnes; llowing GLOSSARIAL AFFINITIES. yanan- sensifive TuE comparison of the vocables of languages was often conducted refer t0 in so loose and unscientific a manner, without definite principles, with- out regard to dialectic changes, and to the neglect of the comparison of grammatical forms and structure, that this branch of philology has ter 1nto fallen into indiscriminate and not undeserved disgrace. I admit that peakinr' a comparative vocabulary, however carefully prepared, is of much | that 1§ Isolated nouns less philological value than a comparative grammar. ] theme, and verbs are very apt to get corrupted in the lapse of time, and to manner adopt one pbase of meaning after another, till the original meaning is lm the overlaid or forgotten; whilst declensional and conjugational forms— the bones and sinews of a language—retain for ages both their shape Nevertheless, I and their signification with marvellous persistency. regard the comparison of vocables, when carefully and cautiously con- ducted, as an important Aelp to the determination of lingual affinities ; and it will be found, I think, that the following vocabularies bear independent testimony, in their own degree, to the very same yesult at which we arrived by grammatical comparison, viz., that whilst the Dravidian idioms exhibit traces of an ancient, deep-seated connexion with Pre-Sanscrit—the assumed archaie mother-tongue of the Indo- European family—their relationship to the languages of the Seythian group, especially to the Ugrian tongues, is closer, more distinctive, and more essential than any other. 438 G’L()SSARIAL AFFINITIES. | T I vidi Indo-Europea;n Affinities. YO0 Section 1. fain SANSCRIT AFFINITIES. foll | Before entering upon the comparison of Dravidian with Sanscrit and vocables, it is desirable to disentangle the subject from extraneous witl matter by a preliminary examination of words which appear to have Poss been borrowed by the Sanscrit from the Dravidian languages. tong I have long felt persuaded that some words of pure Dravidian foun origin have found their way into Sanscrit vocabularies; and I have no 15 fo doubt that a still larger number of words have been introduced into diale Sanserit from the North-Indian vernaculars. I have also already Jexic stated my opinion (in the section on ‘Sounds’) that it was from the whil Dravidian languages that the Sanserit borrowed its ¢cerebral’ con- verb: sonants. varig There is probably alinost as large a proportion of Drévidian words tion in Sanserit, as of British words in English: but this fact has generally and | remained unnoticed or unknown; and wherever any word was found only to be the common property of the Sanserit and any of the Dravidian withs tongues, it was at once assumed to be a Sanscrit derivative. Doubt- t]le g less, the number of Sanscrit derivatives, properly so called, which a g have been introduced into the Dravidian languages is very great; but EXist, those words are always recognised and admitted to be derivatives by Celye Tamil and Telugu lexicographers, and carefully distinguished from 10t g national or native Dravidian words. In a few cases, as might be expected, but in a few cases only, some doubt exists whether a par- Ji ticular word was borrowed by the Sanscrit from the Tamil, or by the Tamil from the Sanserit. Sanscrit lexicographers and grammarians akky were not so discriminative as their Dravidian brethren; and if any writer had happened to make use of a local or provincial word, that is, a word belonging to the Dravidian vernacular of the district in att which he resided (and it was natural that such words should occasion- b ally be used, for variety of metre or some other cause, especially after Sanscrit had ceased to be a spoken tongue), every such word, provided =lt:wi) only it were found written in Sanscrit characters, was forthwith set down in the vocabularies as Sansecrit. Some words of Greek or Roman origin, such as ¢ denarius,” ¢ dpa, Nerrov’ (in the sense of @ minute of a degree), and even the Greek names of the signs of the Zodiac, have found their way into Sanscrit. If so, it may be concluded that a e S —— S —— T \ SANSCRIT. Q 9 much more considerable number of words belonging to the old Dra- vidian vernaculars must have obtained a footing in the Sanscrit vocabularies. The grounds or conditions on which I conclude any word con- tained in the Sanscrit lexicons to be of Dravidian origin, are as follows : (i.) When the word is an isolated one in Sanscrit, without a root mserit and without derivatives, but is surrounded in the Dravidian languages aneous with collateral, related, or derivative words; (ii.) when the Sanscrit 0 have possesses other words expressing the same idea, whilst the Dravidian tongues have the one in question alone; (iil.) when the word is not yidian found in any of the Indo-European tongues allied to the Sanserit, but ave 10 is found in some of the Scythian idioms, or at least in every Dravidian o info dialect however rude; (iv.) when the derivation which the Sanserit lready lexicographers have attributed to the word is evidently a fanciful one, om the whilst Dravidian lexicographers deduce it from some native Dravidian 1’ con- verbal theme of the same or a similar signification, from which a variety of words are found to be derived; (v.) when the significa- | words tion of the word in the Dravidian languages is evidently radical and physiological, whilst the Sanscrit signification is metaphorical, or perally s found only collateral ; (vi.) when native Tamil and Telugu scholars, not- aidian withstanding their high estimation of the Sanscrit, as the language of Doubt- the gods, and the mother of all literature, classify the word in question as a pure Dravidian one;—when any of these reasons are found to , which ‘lt,' but exist, and more especially when several or all of them coincide, I con- tives by ceive we may safely conclude the word in question to be a Dravidian not a Sanscrit derivative. I here subjoin a selection of such words. ed from jght be 8 O Words borrowed by the Sanscrit from the Dravidian tongues. I by the akka, a mother. For the exclusive Scythian relationship of this word, nn}ﬁl‘iaﬂs and proof of its derivation by the Sanscrit from the Scytho- | if any ord, that Indian vernaculars, see the list of ¢Scythian Affinities.’ strict in attd, a mother, an elder sister, a mother’s elder sister.—See ¢ Scythian Affinities. lly alter 70 Vide atavi, a jungle, a forest. The root of this word is represented in Sansecrit dictionaries to be ¢ ata,’ to go, because a forest is a place ith b where birds, &e., go; which is evidently a fanciful derivation. i Boma? All the Dravidian languages contain a primary root ¢ad,” the qafe of 0 1, b AL radical signification of which is nmearness, closeness; and this monosyllabic root is modified and expanded so as to signify ] that & ey e _— 440 GLOSSARIAL AFFINITIES. every variety of closeness. Amongst other derived words we have in Tamil ¢ adar, to be crowded, to grow thick together (like the trees of a forest); and T have no doubt that it was from this verbal root, not from any native Sanscrit one, that ‘atavi’ (in Tamil and Telugu ‘adavi,’) was derived. Even the forma- tive ¢ vi’ is one which is distinctively Dréavidian; e.g., ¢kélvi,’ Tam., hearing, from ¢ kél, ¢o hear. ali ani, ani, }t]ze pin of the axle of a cart; derived, it is said, from ‘an-a,’ o sound. On comparing this word with the Tamil ¢éni, a nail, katy a pin or peg of any kind, it is evident that they are not different words, but one and the same; and the only question is which vs the original ? The Tamil word is connected with a family of roots, each of which has a real affinity in signification to that of @ nail, considered as a fastening,; e.g., ‘an-ei, to embrace, to tie; ‘an-i, to put on,; ¢ an-avu,’ to cleave to; ‘an-u,’ to touch. The derivation of the Sanscrit word from this Dra- vidian root is, therefore, beyond comparison more natural than that whieh Sanscrit lexicographers have devised. amba, amma, } mother, vocative ‘ammi.’ This word is found also in some of the Western Indo-European dialects; e.g., Old High German and Oscan ‘ amma; Icelandic ‘amma’ (grandmother); German ‘amme’ (nurse). Notwithstanding this, I am inclined to believe that it was from the Drévidian languages that this word found its way into the Sanserit.—See proofs of its Seytho-Indian character in the ¢ Scythian Affinities. aré, interjection of calling to an inferior: no derivation given. Com- pare with this the Telugu ¢aré,” and the Tamil € adé”’ or ¢ ada which are used in exactly the same manner as the Sanscrit The dialectic interchange of d” and ‘r, has so interjection. kaly often been illustrated, that ¢aré’ may safely be considered as equivalent to °adé.’ The supposition of their identity 1is 9. strengthened by comparing the Sanscrit reduplication €aréré, with the corresponding Tamil reduplication ¢adada.’ Whilst the Sanserit interjection is underived and perfectly isolated, the equivalent Tamil intel‘jection claims to be derived from a Tamil root; viz., from ‘adi,’ a slave, the primary signi- fication of which word is the lowest part of anything, e.g., the sole of the foot. The corresponding interjection addressed to women R R Rt CTRT DI e SANSCRIT. 441 } We in Tamil is ‘adi,” and this becomes ‘adé’ by the addition of ‘&’ like emphatic, which is the ordinary sign of the vocative case. So om closely is the meaning of @ slave still connected with these inter- an jections in Tamil, that I have heard persons when thus TR addressed turn round angrily and say, did you buy me, that elyi you call me “adi ¥’ ali, @ woman’s female friend. Probably from ‘alu,” a wife, @ woman, in poetical and vulgar Telugu, and also in Géond. 3, 10 natl katu, sharp, vehement, pungent: assumed Sans. derivation ‘kat-a,” to-go. o 100 In The corresponding Dravidian word is in Tamil ¢kad-u.’ stion Telugu ¢ katu’ is represented as Sanserit, and "kac}-u,’ with the rith 8 very same meaning, as native Telugu. The most important ation derivative of this root in Sanscrit is ‘katuk-ah,” masculine, or e, 7] ¢katuk-i, feminine, mustard. The word ‘katu’ is deeply an-l, rooted in Sanscrit, and is @ prior? unlikely to have been bor- y Drﬁ- rowed from the Dravidian tongues; and yet it cannot be than doubted, I think, that its origin is Dravidian. Not only are the direct derivatives of this word much more numerous in Tamil than in Sanserit, but collateral themes and me of meanings are also very abundant, whereas in Sanserit no cor- relative root exists. ‘kad-u,” Tam., o be sharp, is one of a erinan cluster of roots which are united together by a family resem- emal blance. Some of those are ¢kad-u-gu,” to make haste,; *kad-i, to cut, to reprove; ‘kad-i’ (with another formative), o bite; it was <kari,” identical with ¢kadi,’ curry; ‘kadu-kadu’ (a mimetic y mto word), to appear angry. Moreover, the Sanscrit ¢ katuk-(ah in the or 1),” mustard, appears to have been derived from the Tam. ‘kadugu, mustard ; for nouns formed from verbal themes in this manner, by suffixing the formative ‘ku,” pronounced ‘gu,’ Com- ol ) are exceedingly abundant in Teymil. ﬂﬂSCﬂt kala, any practical art, mechanical or ﬁne: assumed derivation ¢ kala,’ ias §0 to sound. The Tamil makes uze of the:same word (“kalei’ for ored 85 ‘kala’), but includes in the signification every science, as well fity 18 as every art. We cannot, I think, doubt the derivation of ‘ al‘él'évl ¢kalei’ or ‘kal4, from the primitive Tamil root ‘kal,’ to learn, (another derivative of which is < kalvi,” learning.) afectly The other meanings of the Sanserit word ‘kald,” are so entirely unconnected with this, that it is evident that two ! siglli' different words spelled in the same manner (one of them Dra- b \4’"’ th e vidian), have erroneously been supposed to be one and the same, b 442 GLOSSARIAL AFFINITIES. kavér-i, saffron, also the river Cavery (from its muddy colour): assumed root ¢ kava,’ to colour. Greek name of the same river, yapypes. Possibly this word may be of true Sanscrit origin. I may suggest, however, the possibility of the origin of the name of the river Kévéri, from the Dravidian ¢kavi,) red ochre, and “ér-u,” Tel.,, a river, or ¢ ér-i,” Tam., a sheet of water. kiicha-h, a female breast, especially that of a young unmarried woman : k ko derivation ¢ku,’ to sound. There is a much more natural deri- vation in the Tamil word ‘kiicha-m,” bashfulness, a verbal noun from ¢ kas-u,’ to be shy, to be ticklish. kij-a, to utter a cry, as a bird. Probably this word is mimetic; we may compare it, however, with corresponding Dravidian words, which also appear to be mimetic; viz., Tam. ‘k{i-vu; Tel. ‘ki-su; Can. ‘ki-gu; each of which is derived from a Dra- vidian root, ‘kl,’ an <narticulate cry. Compare also our English ¢ coo0.’ kuti, a house; related words ¢ kutiram,” also ¢kutérah,’ a cottage, a hut, and ‘kutumba,’ @ family: assumed derivation kuta, crooked. There cannot be any doubt of the derivation of ¢kuta-m,” « water pot, from ¢ kuta,” crooked; but the other words are pro- bably of Dravidian origin. In Tamil ‘kudi’ means a house, 1\ h habitation, also an vnhabitant, a farmer; related Tamil words are ‘kudil * and ‘ kudifei,” a Zu¢; a provincial form of the latter of which is ‘kuchu.’ In Telugn and Canaresc ‘gudi’ means a temple, and ¢ gudishi’ or ¢ gudise,” a hut. In Hindustani ¢ guti’ means a house. Say By all native grammarians, these words are considered to be of pure DrAvidian origin; and the existence of sy the same root in all the Finnish tongues favours the supposition that it was not borrowed by the Drividian langnages from the Sanscrit, but by the Sanscrit from those languages. Compare the Finnish ‘kota,” Tscheremiss ‘kuda,” Mordwin ‘kudo,” Ostiak ‘ chot,’—each signifying a house. Was the Saxon ‘cot’ also derived from this same Scythian or Finnish source ? kuni, kini, }having a crooked or withered arm. Compare this with ¢kiin,’ Tam., crook-back; and especially with ‘kun-i’ Tam. to stoop, an undoubted Dravidian root, from which it is probable that both the Tam. ‘kiin,” and the Sanscrit ¢ kuni’ or ¢ kiini’ have Y been derived. kula, a pond or pool : assumed derivation "kﬁi-a,’ to cover, Com- SANSCRIT. 443 med £ pare the Tamil ‘kul-am’ and the Tel. ‘kol-anu,” a tank, « s pool. nay We can scarcely doubt that the Sanserit word is identical 6 of with the Dravidian one ; and if so, it must have been derived and from it, for the Tam. ‘kul-am, a tank, is unquestionably a verbal noun from ‘kul-i,’ o bathe, a pure Dravidian root. W kotta lert kota, }a Jort, a stronghold : assumed derivation ‘kuta,’ to be crooked, Jount or ‘kutta,” to cut. The Dravidian dialects make use of the same or a similar word for a fort, viz., ‘kéta’ in Tel., ‘kéte’ in Can., and ¢ kottel ’ in Tam. we The Tamil having another and very ancient word for (I I Tel. stronghold, viz., aran,” which is certainly a Dravidian root, it may be concluded that ¢koéttei’ has been borrowed from the Dra Sanserit. But where did the Sanserit itself obtain this word ? our Probably from a Tamil root after all ; for the Sansecrit deri- vations of ‘koéta’ are very fanciful, whilst we could not desire T, a better or more natural derivation than the Tamil ‘kdd-u, line, a Uline of circumvallation, which is sometimes used to 4;];411 0, denote also a walled village, a fortification. ‘kod-u,” when used pIO adjectivally, becomes ‘kott-u.’ 186, (0 khatva, a couch, a cot : assumed derivation ‘khatt-a,” #o screen. Com- ords pare the Tam. ‘katt-il, a cot, from ‘katt-u,” ¢o tie or bind, to Jatter at build. The word ‘ katt-u’ is thoroughly and essentially Dravi- ans dian, and one which abounds with derivatives and related words, qutl sava-m, @ corpse. are 106 of sava-m, adj., relating to a dead body. 1100 These words are said to be derived from * sava, to go, but m he this derivation is much less probable than the Dravidian verbal ZA S np‘lle root ¢ sa, Tam. and Can., to die. The vowel of ‘64’ is short )stiak (18 in the Telugu cha’ (for ‘%a’), in the corresponding verbal ‘1130 theme ‘chachu ;’ and both in Tamil and Canarese it is short ¢ LA in the preterite tense. Sa is undoubtedly a pure Dravidian root ; and it re-appears in the Samofede ¢ chawe,” dead. kun Probably also the Sanscrit ‘shei’ (‘sayati ), to waste away, and ‘shé,’ 0 be destroyed, have some ulterior connexion with it. §t00py g [hﬂt siya, the cvening : assumed derivation ¢sho,’ 7o destroy, to put an eﬁd " Jave to. The Tamil ‘84§, to lean, to incline, a pure Dravidian word, seems to be a much more natural derivation, the evening being the period when the sun inclines to the west. (om- 444 GLOSSARIAL AFFINITIES. nana, several, various, multiform. Bopp derives ‘nana’ from certain assumed obsolete demonstratives signifying this and that. It is more likely to have been derived from the Tamil ‘nal-u’ or ‘nan-gu,’ fowr, this numeral being constantly used in the Dravidian languages to signify several, various, or an iﬁd@ﬁnite number of moderate extent. By a corresponding usage the numeral ¢en is taken to represent any large indefinite num- ber. Thus a Tamilian will say, 7 was told so and so by four persons,—i.e., by several persons; or, we must do as ten people do,—t.c., as the world does. The numeral adjective ‘nala’ (from ‘nil-u,’ four) is regularly used in Tamil to signify | various, though literally meaning four-fold ; and the euphonic pa change of 1’ into ‘n’ in the High Tamil ¢ nin-gu,” four, shows a how ‘ nand’ may have originated from ‘nAlA. The Tamil Die- a tionary gives us, amongst other instances of the use of ¢ nala, one which is identical with the instance of the use of ‘nani’ ;;? given in the Sanscrit Dictionary, viz., ‘nAla vidam, in various ways, literally in a four-fold way; which we may compare with the corresponding Sans. ‘nana vidha,” in various ways. With respect to the Dravidian origin of this word, the testi- mony of the Tamil standsalone ; for in the Canarese and Telugu Dictionaries ‘nana’ is regarded as Sanscrit. Kl nira-m, water : assumed derivation °ni 7) to obtain. This derivation shows that the word was not familiar to the Sanserit lexico- graphers. ji ‘niram’ is rarely used, in comparison with ‘ap (connected with ‘aqua’), and ‘uda’ (connected with ¢unda’ i and ¢ 178wp 7). ‘jala,” another Sanscrit word for water, is sup- posed to have been borrowed from the northern vernaculars ; whilst I have little or no doubt that to ‘nira’ a Dravidian origin should be ascribed. The corresponding Dravidian vord is ‘nir’ or ‘nir-u; and this is the only word properly signify- ing water which the Dréavidian dialects possess. The Telugu ordinarily uses ‘nillu’ for ‘niru,’ 4.e., the plural (‘nirulun,” cor- 2 rupted to ‘nillu’) for the singular ; but ‘niru, the singular, is also occasionally used. Pan ‘nir is in Go6nd softened to ‘ir and in Brahui it has il s s Sl become * dir.’ The Malayalam alone commonly uses for water another word, viz., ‘vellam,’ which properly means a food. This word is used in Tamil to denote the water with which rice- pall Jields are flooded ; and it has probably thence come to signify water in Malayalam. Even in that dialect, however, ‘nir’ is :: Bt i i s SANSCRIT. 445 L(‘ 1“ also used. In Tamil the adjective ¢tan,’ cool, is o frequently prefixed to ‘nir,” that in the colloquial dialect the compound mil “tannir,” water, literally cold water, has superseded the original sed and simple noun. (n Whilst I have no doubt that ‘nir’ is a true Dravidian word, g6 it may have descended to the Drévidian family from some (it Japhetic source older than the Sanscrit; and hence, it may foui have some ulterior connection with the Greek “vypos’ and /)p/e, ‘vapos, wet, (and throngh them with the modern Greek ¢ vipo,’ ila water), though these words are supposed (and perhaps cor- nify rectly) to be derived from ¢ vdw, to Aow. 0016 patta-m J0WS pattana-m ( @ city, town, or village: assumed derivation ¢ pata,’ to sur- pattana-m, f Dic ald, round. The Tamil has borrowed the word ‘ pattanam ’ from ind’ the Sanscrit ; and yet, as in the case of ¢ kota,’ a fort, it will be 1101 found, I think, that the Sanscrit word itself was derived origi- 1pm’e nally from the old Dravidian vernaculars. Professor Wilson ) conjectures that, ¢ pattam’ is probably identical with the st ‘pettah,” of Southern India; but the word from which I con- lugt ceive it to have been derived is ‘patti,” Tam. a fold for cattle, a pound, a small village,—a word which constitutes the final portion, or termination, of the names of so many towns and villages in the south; eg., ¢ Kovil-patti, Zemple-town. In gtion xi00- Canarese the same word is hatti; e.9., Dim-hutty. The Old ) ap Sanscrit seems to have adopted this word ¢ patti,” in addition to its own ‘pura’ (which is a true Indo-European word), and inds’ 5 sup- formed from it first ¢ patta-m,” and then ¢pattana-m.’ The word ¢pettah,” @ suburb (Tam. ‘péttei’), which is re- lars ; »’idmn ferred to by Wilson, belongs to a different root from ‘patti,’ and o cannot have been the origin of the Sans. ¢pattam.” * péttei’ is derived from ‘pédu, Tam. « sufix to the names of willages; gnify, ‘elugt which, again, is identical with ‘padu’ and ¢ padi,’ a place, each of which is suffixed to names of villages like ¢ pédu.’ 0o panno, llm’; 8 Prakrit, gold. This word is supposed by Ellis to be derived from the Sans. ¢ suvarna.’ I think it much more probable that 1t ha it was adopted into the Prakrit from the Tam. ‘pon, or the qaler Tel. ¢ ponn-u,” gold. vt palli, @ city, a town, a village. This is without doubt identical with j i the Drividian word ¢ palli,” which is added to various names of :} g’ ll’lft\' places in the South ; e.g., Zrichinopoly, properly ¢ Trisirapalli.’ e ! 446 GLOSSARTIAL AFFINITIES. The DréAvidian origin of this word is sufficiently proved by the circumstance that it is chiefly, if not exclusively, used to denote places which are within the limits of the Dravidian tongues. bhaj-a, to share. Va bhéga, a portion. I am doubtful whether to regard these words as SEa derived from the Tamil ‘pag-u,’ to divide, to share, or to suppose = LS both the Sanscrit and the Tamil to be derived from a common = and earlier source. Probably the former supposition is in this case the more correct. At all events the Tamil ‘pag-u 9 1S a pure, underived Dravidian root. The noun formed from it, signifying @ share, is ‘pang-u’ (‘ng’ for ‘g’ as is often the case); and a collateral root is pag-ir, meaning also to share. al The Sanscrit word ‘pangu’ means lame, and is altogether un- connected with the Tamil one. vl ~mina-m, ¢ Jish; assumed derivation ‘mi’ (‘minéti’), o hurt. 5 The Dréividian word for fisk is ‘min,” a word which is found in every dialect of the family, and is the only word signifying fish which these languages possess. ‘min’ is found even in the small list of decidedly Dravidian words contained R R e L SRR S The Goénd has ¢ mind.’ It seems in the Rijmahal dialect. much more probable that the Sanscrit-speaking people bor- rowed this word from the Indian aborigines, and then incor- porated it in their vocabulary with other words signifying the same object, than that the Dravidian inhabitants of the Malabar and Coromandel sea-boards were indebted for the word which denoted so important an article of their food and commerce, to a race of inland people coming from the North-West. Moreover the derivation of ¢min,” which is supplied by the Dravidian languages, is as beautiful as the Sanscrit derivation is uncouth. The root of ‘min, a fish, is ‘min, Tam., to glitter, to be phosphorescent. Hence the glow-worm is min-mini’ by redu- plication; and ‘min,’ a verbal noun which is formed from ‘min’ by the lengthening of the included vowel (like ©péru, Tam., a birth, from ‘peru,’ to bear, and “kol,’ reception, from ¢kol, to receive), signifies in poetical Tamil a star, as well as a fish; eg., ‘van-min, a star (literally a sky-sparkler); and ¢aru-min, the Pleiades, literally the siz stars. Who that has seen the phosphorescence flashing from every movement of the fish in tropical seas or lagoons at night, can doubt the appro- e A R —— o d SANSCRIT. 44 ( the priafeness of denoting the fish that dart and sparkle through ote the waters, as well as the stars that sparkle in the midnight sky, by one and the same word, viz, a word signifying that which glows or sparkles? valacsha-m, white; assumed derivation ‘vala,” to go. Much more a8 probable is the derivation of this word from the Dravidian )0S6 “ vel, white. Compare also the related Dravidian words ‘veli, o1 The space, the open air; ‘velli) silver; ¢velicham,” light. this Hungarian ¢vilaga, light, appears to be an allied word. Has 1§ a the Slavonian ¢veli, white, been borrowed from a Seythian 1t source ! or is it one of those ultimate analogies which bind the both families together ? (11, val-a, fo surround. Uil valaya-m, a circlet, a bracelet. The Dravidian languages have bor- rowed the Sanscrit noun, with or without modification ; but the verb from which the noun has been formed was itself, I b s doubt not, borrowed by the Sanscrit from the Dravidian ord languages. The corresponding Dravidian root is ‘val-ei,” to yund bend, to crook, metaphorically to surround. This word has a ined larger store of secondary meanings and wider ramifications oIS than the Sansecrit verb. It is also used as a noun, without any hor formative addition, when it signifies a /hole, SInuosity ; e.g., “eli-valei,” Tam., a rat-hole. Whilst the Tamil makes occa- 1COT- sional use of the Sanserit ¢ valayam, « bracelet, an armlet; it g the also uses valeiyal,” a verbal noun formed from ‘valei,’ its [abar lel(‘h own verbal root, to signify the same thing. Taking these 181Cey -arious ecircumstances into consideration, I conclude that the Dravidian verb has certainly not been borrowed from the Sanscrit, and that the Sanscrit verb has probably been derived y the from the Dravidian. ation In the foregoing list of Dravidian words which have found a place in the vocabularies of the Sanserit, I have not included {0 e the names of various places and tribes in Southern India which redu- are mentioned in the Sauscrit historical poems, and which from have in consequence found a place in the dictionaries. In PéRlli. general the vernacular origin of those words is admitted by ) from Sanserit lexicographers. In one case, however, a Sanscrit ell 88 origin has erroneously been attributed to a Dravidian word of ), 3,11"1 this class. ‘Malaya,” a mountain or mountainous range in bas Southern India, is represented as being derived from ‘mala, if the Sans., to kold or contain (sandal-wood). The real origin is 3PP1‘O 7 448 GLOSSARIAL AFFINITIES. { unquestionably the Drévidian ‘mal-a’ or ¢mal-ei,” a hll or mountain, and also a hilly or mountainous country; and the 1t th range of mountains referred to under the name of ¢ Malaya’ is doubtless that of the Southern Ghauts or the ‘Malayéla’ country, 10 which was called ¢ Male’ by the Arabian geographers. he [ I now proceed to point out the existence of some real Sanscrit & The words affinities in the vocabularies of the Dravidian languages. which are contained in the following list are true, underived Dravidian ad roots, yet they are so closely allied to certain Sanscrit words, that ud they must be concluded to be the common property of both families of tongues. ad Possibly one or two words may have been borrowed at an early period by the one language from the other; but in most cases, if not in every case, there is a preponderance of evidence in favour of the an €l mutually independent origin of both the Sanscrit word and the Dra- vidian one, from a source which appears to have been common to - both. The various words appear to be too deeply-seated in each family of languages, to have too many ramifications, and (whilst they Sl retain a family likeness) to differ too widely, either in sound or in kad signification, to allow of the supposition of a direct derivation of the one from the other. Moreover, notwithstanding the general resem- kan blance of the Dravidian words contained in the following list to the Sanserit ones with which they are compared, and notwithstanding the prejudice of mative grammarians in favour of everything Sanscrit, j? these words are invariably regarded by native scholars as independent Con- of the Sanscrit, and as underived, ‘national’ Dravidian words. sequently, if a connexion can be traced, as I think it can, between these words and the corresponding Sanscrit ones, it must be the kiny connexion of a common origin. I place in another and subsequent list those Dravidian words which are more directly allied to the Greek sji or Latin, the Persian, or some other Extra-Indian member of the In this list 1 place those Indo-European family, than to the Sanserit. Dravidan words which appear to be allied to the Sanscrit alone, or more directly to the Sanserit than to any other Indo-European lan- guage ; and it is remarkable how few such words there are, compared with those of the other class. A comparison of the two following lists kugj will, T think, lead to the conclusion that the Indo-European elements which are contained in the Dravidian languages were introduced into those languages before the Sanscrit separated from its sisters, or at least before the Sanserit, as a separate tongue, came in contact with the Dravidian family. ca SANSCRIT. 449 (U} The Dravidian words which follow are quoted from the Tamil, if the Where it is certain 18 1t 1s not expressly mentioned that it is otherwise that the final vowel or syllable of a Dravidian word is no part of the 17, root, but is a separable formative accretion, or a particle which has been added merely for euphony, or for the purpose of facilitating enunciation, I have separated such vowel or syllable from the genuine orlt portion of the word by a hyphen. ou Jian d to strike, to beat, to kill that ud-ei, to kick, to stamp Compare ‘ut-a,” Sans to strike, to knock il1es down ad-ei, to get in, to obtain, to possess. Compare ‘ad-a,” Vedic Sans., fo atly 106 occupy or possess he an-u, Tel Dra en, Tam } to speak, to suy. Compare ‘an-a,” Sans., to sound n 10 Qr-u, to creep; in the higher dialect of the Tamil, ¢o ride (as m a palan— each quin). hey Compare Sans. ¢ ur-a,’ ¢o go. or 1 kad-a, to pass, to pass by or over Compare ¢kat-a,” Sans. to go f the psent: karudei, an ass Tel. ‘gadide; Can. ‘katte Compare Sans. ‘khara, ar ass o the The Sanscrit word is borrowed and used by the Tamil ¢ the poets; but it is never confounded with ¢karudei, which is qSC]lt considered to be a purely Dravidian word. Nevertheless, dent “karudei’ is evidently allied to ¢khara’ in origin, and also to (on the Persian ‘char,’ and the Kurdish ‘kerr.’ Compare especially fyveel the Laghmani ‘karatik,” a female ass kinna, Can. small; Tuda ‘kin; Tel. ¢ chinna; Tam. smna Com- aquent pare ‘kana,” Sans. small a minute partacle also ‘kamga the (treek smallest, the youngest. There is no doubt of the Tamil ¢ éinna’ f the having been softened from ¢kinna but I have some donbt ) those whether the ‘n has not been corrupted from ‘r, for the one, OF ultimate root to which ‘§inna’ is referred by Tamil scholars is l hﬂ SIR-U mp’tlel ng ]1‘t kudirei, a %orse; Can gudure Compare Sans. ‘ghoda,” a horse, el ghorati,” to go as a horse. The Dravidian lanwuages have o 010 borrowed ‘ghdda’ from the Sans. (in Tamil ‘goram,’ ‘gédagam’); o o but ‘kudlrel is regarded as an underived, mdxgenous Dravidian word. It is evident, however, that the two WOIdS are ulti- d’, Wlth mately related, 2¢a — 450 GLOSSARTAL AFFINITIES. kir-u, fo cut, to scratch, to rend. kir-i, 2o tear. ‘Compare ¢ khur-a,” Sans. to cut, to scratch. ked-u, fo spoil or destroy, or (intransitively) to be spoiled or destroyed : verbal noun ‘ kéd-u,’ ruin, relative participle ¢ ketta’ (‘tt’ for 1l « dd’), bad. Compare the Sanscrit ¢ khid-a,” to suffer pain or misery, and its verbal noun ‘khéda,” sorrow, distress. Compare also ‘khit-a,’ to terrify, and its derivative ‘khéta,’ bad, low. If these words are allied to the Dravidian one, as they appear to be, it must be in virtue of a common origin; for there is not a more distinctively Dravidian word in existence than ‘ ked-u.’ fir-el, to shave: base ‘sirR’. Compare Sans. ¢ kshur-a,” fo scrape, and ¢‘kshaura,” shaving. Compare also the corresponding Greek verbs ¢ fvp-dw,” ¢ Evp-éw, to shave; our English shear, from the ’ pa Germ. ¢ scheer; and even the Greek ¢ «éip-w,” to share. gil-ir, to tremble, to have the hair standing on end. Compare °chél-a,’ Sans. to shake, to tremble.—See also subsequent list under ¢ kulir,” cold. §e, to be red. This root forms the basis of many Tamil adjectives and nouns (e.g., ‘Sen,” red); but is not used anywhere in its pri- mitive unformed shape. Compare ‘sona,’ Sans. to be red. ﬁd-j, a stick, a club; verbal theme ‘tadi,’ ¢o be thick or heavy. Com- pare“ tad-a,” also ¢ tad-i,” Sans. to strike, to beat. fﬁv'-u, to sprinkle gently (as dust). ti-ru, to drizzle, to scatter, to spread abroad (as a report). The transitive of ‘tu-ru’ is ‘ti-rRu’ (pronounced ‘tuttru’), pal, to winnow; and a derivative from ¢ tivu’ is ‘tasi,” dust. The ultimate root of all these words evidently is ‘ti.’ Compare ¢ dht, Sans. to shake, to agitate; a derivative from which is iy ¢ dhuli, dust. Compare also ‘tista-m,” dust (assumed deriva- tion tlis-a,” fo sound), with which our own word ‘dust,’ is evidently identical. From the Sanscrit ‘dhli; the Tamil has borrowed °tali’ and also ‘tll,’ dust: °tis-i,” on the other hand, is a pure Dra- vidian word, allied to the Turkish ‘tus’ or ¢tusan,’ powder, and possibly to the Mongolian ‘toghoz; and there cannot be any doubt of the Tamil verbs ‘tiv-u’ and ‘tir-u’ being un- derived Dravidian themes. “dhli’ or ‘ti’ appears, therefore, SANSCRIT. 451 to be the common property of both families of languages; whilst it is in the Dréavidian family that the original meaning of this ol root appears to have been most faithfully preserved. for nad-a, o walk. Compare the Sanscrit theme ¢ nat-a,” fo ddﬁlﬁ?e, to act, to shake; derivatives from which are ¢ nata-m,’” dancing, ‘nata- and ka-m,’ @ drama, a play. It seems improbable that the Sanscrit 4, word has been borrowed from the Dravidian tongues ; and yet it is certain that the Dravidian word has not been borrowed ey here from the Sanscrit, for the Telngu and Canarese make a broad distinction between the Sans. derivative ‘natinchu’ or ‘natisu, han to dance, and their own theme ‘naduchu’ or ‘nadi, to walk, and whilst the Sanscrit has many words signifying walking, the and Dravidian languages have ‘nad’,) alone. Probably, therefore, both words have been derived from a common sonurce. reck the pad-u, Tam. Tel,, and Can. to sing. Compare Sans, ¢ path-a,” to read, to recite. The Sans. “path-a’ is, I have no doubt, the theme 3, from which the corresponding Tel. ‘path-i, and the Tamil nder ‘ pad-1,’ o 7ead, have been borrowed; and the Tamil ¢ pAda-m,’ @ lesson, is clearly derived from the Sans. ¢ patha,” reading. ‘pad-uy’ to sing, however, and ¢ patt-u,’ @ song, (Tel. ¢ pata,’ | anfl Can. ‘pat-u;’ Gond ‘ pita,’) are certainly not derivatives from PI'I the Sanserit; but I suspect them to be ultimately related to ‘path-a’ and ‘patha,” as descended from some ancient source (Jom common to both, The ideas expressed are nearly related ; for the reading of all Hindus (and of all Orientals) is a sort of intoned ‘cantilena;’ and even the Sanserit derivative ¢ padi,’ 7o read, often receives in colloquial Tamil the meaning ¢o sing. pal, Tam. and Can. « portion, a part, a class. Compare Sans. ‘phal-a,’ ru') / The to divide; also Lat. ¢ par-s) a portion.—See especially the npare Semitic affinities of this qud. oh 18 piRa, other; eg., ‘piRa-n,’ another man; <hera’ (for ‘pera’), Can. el,iv 9 Compare ‘para,” Sans. in the sense of different, a sense which'it g, 18 often bears. It is with this preposition, and not with *pra,’ before, forward, that I think the Tamil ¢ pira,’ other, should be ) tuh compared. ) Dfé‘ The use of the Tam. ‘ pira,’ and that of the Sans. ‘para’ (in ey the signification adduced above) are identical ; and we might of be naturally suppose the Tamil word to have been derived from the Sanscrit. The Tamil, however, whilst it admits that g m op1016) ’ paraf was borrowed from the Sanscrit, regards ‘ pira’ as an 262 s A P i 45 [} GLOSSARIAL AFFINITIES. indigenous theme. The ‘®’ of ¢pira, is unknown to the Sanscrit, and is considered to be a distinctive mark of Dra- vidian words. Moreover, the Tamil ascribes to ‘pira’ the meaning of other alone, whilst the Sanscrit ¢ para’ has a much wider range of signification. It is remarkable that the Tamil has another root, ‘pir-a-gu,’ D after, (ultimate base ¢ pir,’) which is considered to be perfectly independent of, and unconnected with, ¢ pira,’ other; and yet that this very meaning, after, is one of the many significations which are attributed to ¢ para’ in Sanscrit. It may be concluded, I think, that ‘para’ and ‘ pira,” are radically allied; and yet the supposition that the one is derived Sa from the other, is quite inadmissible. Each is too deeply e seated in its own family of tongues to allow of this supposition, Ar and we are therefore driven to conclude, that both have been Sar derived from a common source. th Wit pél, milk. The Dravidian languages do not contain the verbal theme oon from which this word is derived. We may compare it with the Sanscrit ¢ pAyasa, milk, and also with °phya,” water, Zend blis ¢ péo,” Affghan ¢ poi; all of which words are derived from pa Sans. to drink, a root which runs through almost all the Indo- ear to ¢ European languages. Possibly the Dravidian ¢pal,’ milk, may be a verbal noun tha formed from this very theme; for a large number of verbal ang nouns are formed in Tamil by simply adding ¢al’ or €1’ to the root. Notwithstanding this, the purely Dravidian character the | Sey and connexions of this word ¢pal, preclude the supposition of its direct derivation from the Sanscrit ¢ pa Toay the pés-u, to speak. Compare ‘ bhash-a,’ Sans. to speak. el pl, a flower, or to blossom, Tam., Tel., and Can. Compare ‘phull-a,’ fam Sans. to blossom, and ¢ pushpa,” a flower. 1t g i IS al ol Vg WEST INDO-EUROPEAN FAMILY. 453 the Ia Indo-European Affinities. the uch Sectron I1. EXTrA-SANScRITIC OR WEsT INDO-EUROPEAN AFFINITIES, Viz.: U atly Dravidian words which appear to be specially allied to words that are yet contained in the languages of the Western or non-Sanscritic branches 1008 of the Indo-European family. are Some of the words which are contained in the following list have ived Sanscrit as well as Classical or West-Aryan analogies; but they have eply been placed in this, rather than in the preceding, list, because the West- jon, Aryam affinities are clearer, more direct, and more certain than the been Sanscrit ones. The greater number, however, of the words that follow, though indubitably connected with the Western tongues, and especially eme with the Greek and Latin, exhibit no analogy whatever to any words with contained in the Sanscrit. Jend If the existence of this class of analogies can be clearly esta- blished, it must be concluded either that the Dravidians were at an ( pa, [ndo early period near neighbours of the ‘Vest-f&ryan tribes, subsequently to the separation of those tribes from the Sanscrit-speaking people, or noun that both races were descended from a common source., atbal The majority of the Dravidian words which exhibit West-Aryan analogies, do not belong to that primary, rudimental class to which the a6t q0lel the words that the Dravidian languages have in common with the Scythian are to be referred. Nevertheless, they are so numerous, on Of many of them are so remarkable, and, when all are viewed together, the analogy which they bring to light is so distinet, that an ultimate relation of some kind between the Dravidian and the Indo-European families, may be regarded as conclusively established. ull'aa’ As before, the Dravidian words are to be regarded as Tamil, except it is stated that they are taken from some other dialect. K as-el, to shake. Compare ‘ sel-w, to shake, to move to and fro. arn-vi, @ waterfall; from ar-u,’ to ebb, to trickle down. ’ Compare ‘riv-us,” Latin, @ brook, English ‘river, also the verbal theme of those words, ¢ gé-w’ or ‘pv-w’ (asin piy’), to Sow. al-ei, to wander, to be unsteady: ‘alei, as a noun, means « wave. Compare a\d-opar,” to wander; Germ. ° welle, Armen. ‘alik, a wave. av-a, desire; also ‘av-al’ a verbal noun, derived from an obsolete 454 GLOSSARIAL AFFINITIES. root ‘av-u,’ to desire. Compare Sans. ‘av-a, of which one of [} the rarer meanings is fo desire. The affinity between “ava’ and the Latin ‘ave-o,’ fo desire, is still more complete, inas- much as this is the only meaning of the word in Latin, as in Tamil.—See also ¢ Semitic Affinities.’ I ayv-a, Tel., a grandmother. In Tuda ‘avva, means a mother,; in Canarese ‘avva’ or ‘avve’ means either @ mother or grand- mother, or, generally, an old woman. The ordinary Tamil form of this word is ‘auv-ei,” an honorific term for a matron, an ul< elderly lady, but ‘avv-ei’ is also used. Compare the Latin The root of Ul ‘av-us, a grandfather; ©avi-a,” a grandmother. the Latin word appears to have been applicable to any elderly relation ; e.g., ‘av-unculus, a maternal uncle. 6y, avi, a spirit, literally wvapour, breath; then life, and also a spurit : verbal theme ‘avi,) to yawn, to breathe. Compare the Gothic 611 ¢ahma,’ spirit; the Sanserit ‘atma,” self, or soul; the Greek ‘ dvepos, wind,; the Latin ‘anima,” and the Tamil ‘anma,’ soul or spirit. Compare also the High Tamil ¢én-dal, the ) breath. The resemblance of the Tam, ‘avi’ to the Greek aw e to blow, is especially remarkable. ir-u, to draw, to pull. Compare ‘épv-w, to draw. Compare also “ele,” the Canarese equivalent of ¢iru, with ¢ é\k-w, fo drag, a word which is probably related to ¢épd-w,” through that alliance of ‘r’ to ‘1’ which is apparent in all languages. or F the ultimate root, and ‘mbu,’ iru-mbu, won,; from ‘iru ’ a formative, euphonised from ‘bu’ or ‘vu Tel. ¢inumu.’ Compare the Saxon ‘iren,’ Danish ‘iern,’ Armenian ‘ergad.’ The ‘r’ of these words has been hardened from ‘s, as appears from comparing them with the German ‘eisen’ and the Sans- flr-a[ crit ¢ ayas. None of these words, however, though possibly they may have some ulterior connexion with the Tamil, seems to be so nearly velated to ‘iru’ as the Motor (a Samoiede dialect), ‘ur.’ in-u, to bring forth young, said of cattle only. Compare Engl. ‘¢ yean, Sax. ‘eanian.’ kaq.j uyar, high; when used as a verb, fo raise. Compare ¢ deip-w’ to ¢ raise up ; also “dep’ in aep-Bels’ (Aor. pass. past), and in the adverb ¢ aép-dyv, lifted up. Compare also ‘dzyp, the awr; Armenian ‘wor, high, Ossete ‘arw, heaven. T WEST INDO-EUROPEAN FAMILY. 455 o of Ad ur-i, Can to burn Tam., ¢er-i Compare ‘ur-o, Lat., to burn, Ya Armenian There are also or, fire; Affghan ‘or,” ¢ wur 135 very remarkable Semitic analogies e.g., Hebrew ¢ur, fire, § 1 and ‘or, light ur-u, to plough Compare Lat. ‘ar-o; Greek apo-w , Lithuanian 11l ‘aru ‘uru-dal tillage, may also be compared with the equi- ind valent Greek ¢ apo-7os Orm an ul-ei, mare Compare €N-0s, marsh atin ul-el a howling Lat ululo,” to Aowl Greek oNo\v{w ;' English t of ‘howl.’ erly ¢ ey, to shoot (an arrow), to cast (o dart) Compare LOS‘ an arrow it to shoot, to cast; éw, to send ot hie er-u, to rse, to get up Compare Lat. ‘ori-or, to 7ise, to get up 1ok ‘eru jiiAyiru,” Tam., the rising sun, may be compared with the lma, Latin ¢ ori-ens sol " the (!’ ell-a, all The Canarese ‘ellar, all they, corresponding to the Tamil a, ellor’ (for ‘ellar’), together with the Tamil ‘ellir,) all ye ¢ A ] iv. (from ¢ ell,’ all, and for “nir,” you), prove that the ultimate also Dravidian root of this word is el. A vowel has been eupho- afterwards drag, nically added to el’ (at first a short vowel hat lengthened), in consequence of which addition the consonant is doubled by dialectic rules mbll; Compare Ossete ‘al,” “ali, “all; Saxon ‘eal; Danish al English “all Probably the Greek G \-058 and the Hebrew umll. “kol’ are allied rather to our own ‘whole,” Lat. ¢salv-us,’ Sans tod sarva, than to the Dravidian and Germanic el all )peara ans or-am border, brim, margin, coast Compare Lat ora, border ssibly MaArgin ¢6r-am’ has no connexion with any Dravidian word gpems signifying mouth ; and possibly the derivation of the Lat ¢ ora ofede from ‘os,” ¢oris may be open to question The correspond- ing word in Gujarathi, Marathi, and Hindi, is ‘k ], o kad-l, to eut, to rend, to reprove. “katti, a knife. Compare Sans ¢ krit-a,” to cut, but especially the Enghsh ‘cut;’ Norman ‘cotu, Welsh ‘cateia ;’ Latin ¢ caed-o. Compare also the Persian o 10 and Ossete ‘kard, a knife, and the Sanserit ¢ karttari, which n the g 01y words, however, are more nearly related to ‘krut-a, Sans., than to the English cut.’ EERER | + D » 6 GLOSSARIAL AFFINITIES. kar}, the eye; ‘kan’ (in the preterite ‘kan’), to see, also to mark, to consider, to think. In the latter sense it becomes euphonically ‘kanpu’ in Tamil, but the base remains unchanged.” In Telugu, the ordinary ‘n, the nasal of the dental row, is used instead of ‘n, the cerebral nasal. Compare the Welsh ceniaw, to see; English ‘ken,’ view, power or reach of wvision. In Webster’'s English Dictionary ‘kanna’ is said to be an eye in Sanscrit; whereas it is exclu- sively a Dravidian word. This mistake may be compared with that of Klaproth in representing ‘kuruta, bdlind, as a Sanscrit word, instead of referring it to the Dravidian lan- guages, to which alone it belongs. Possibly the Dravidian ‘kan,” to see, to consider, may have some ulterior connexion with the Gothic ‘kunn-an,’ to know,; Greek ¢qvid-vac; Sans. ‘jia; Latin ‘gna’ (‘gnarus’), Old High German ¢chann.’ The different shades of meaning which are attributed in Greek to ‘qvid-var’ and ‘éwdé-var,’ seem to corroborate this supposition ; for the latter is represented as meaning to know by reflection, to know absolutely, whereas the former means fo per- cewve, to mark, and may therefore have an ulterior counexion with the Dravidian root. karadi, ¢ bear,; from ‘karadu,’ rough, knotty, unecven, the ultimate base of which must be ‘kara’ or ¢ kar. The Tuda word for a bear is ¢ kar.’ Compare the Persian ¢ chars,” Kurd ‘harj,” and even the Latin ‘urs-us.’ Compare also the Samoiede °korgo,” and the Tungusian kuti.’ karug-u, an eagle. Compare Persian ¢ kergish,” Ossete ¢kartziga.’ ki S kal-a, Can., to steal;, Tam., ‘kalavu,’ a theft; Malayalam, ¢ kall.am,’ a lee. Compare Latin ¢clep-o, to steal; Greek ¢ r\am-els.’— ki See also ¢ Scythian Affinities.’ gav-i, Can., a cave, a cell; Tam., < keb-i,’ a cave. The equivalent Sanscrit words are © gubd,” a cavern, from ¢ guh-a,’ to conceal, and ‘gaha’ a cave, @ forest, from ‘gah-a, to be vmpervious. ki ‘guhd’ has become in Tamil < kugei; but the Tamil ¢kebi’ e A and the Can. ¢ gavi,’ are altogether independent words. Compare with them the Latin ‘cave-a, a cavity, a den, from ¢cav-us, hollow, theme ¢cav-o, to hollow out: and with this compare the Tel. ¢ kapp-u,’ fo cover over, the origin of the Tamil ‘kapp-al’ a ship, and also, probably, of ‘keb-i,’ and ‘gav-i, a cave.—See also ¢ Scythian Affinities.’ WEST INDO-EUROPEAN FAMILY. 457 , b0 kay, to be hot, to burn. Thé Telugn ‘kagu,” Can. ‘kay-u, to burn, lly and the Can. ‘kéage, heat, compared with the Tamil ‘ kanggei,’ In show that the ultimate root is “ka, to which ¢y’ or ‘gu’ is bEd added dialectically as a formative. The only Sanscrit word which secems to be at all related to this Dravidian one, is il “kam-a,’ fo desire; and we should not, perhaps, have suspected \&I‘y it to be related, were it not for its connexion with the Hebrew clu ‘ham-ad,” to desire, and the derivation of that word from red ‘ham-am’ (base ‘Lam’), to be warm. S 4 Compare with the Dravidian ‘ka’ or ‘kay,’ the Greek lan The words seem ‘karw’® (Attic € kd-w’), to burn, to be hot. Jian absolutely identical. Liddell and Scott represent ¢ xaw’ to be xion connected with the Sanserit ¢such-a,” ¢0 dry. How much more a0, nearly connected with the Dravidian ‘kay !’ Besides, the Dra- vidian languages have another word which seems to have a d.in real relation to ‘sush-a,’ viz., sud-u,” o burn. this ki, Gond, to do; Ku ‘gi;’ Kéta ‘ke; Can. ¢géy-u; Tel. ¢ chéy; Tam. by ‘§ey.” The harder form is always to be regarded as the primitive per one, and hence ‘ chéy’ and ¢ §ej ’ are to be referred to the Can. xion ‘géy,’ and that to an ultimate ‘ke,” allied to the Gond ¢ ki.’ Compare the Old Persian ‘ki,” to do; a root which is pro- mate bably related to the Sans. ¢kri,” but more nearly still to the Dravidian words now quoted. The Pracrit form of “kar’ (for rslan ‘kri’) is ‘ka; eg., ‘ka-da,” made, instead of ¢kar-da: and the corresponding Marathi is ‘ke; e.g., ‘ke-la, made. The ]pare Kotra has ¢ kek,” to do. kind-u, to stir, to search, to turn up the ground. Compare ¢ kevr-ew, to prick, to goad, to spur on. !,am,’ () kira, old (not by use, but with respect to length of life). Compare (1) Sans. ¢jaras, age, but especially the Greeck words signifying age, aged, Viz., ‘ yfjpa-s, ‘epa-os, € qepar-os, oép-wv.’—See ralellt also the Scythian affinities of this word. et kira-mei (base -kira’), a weck, literally property, possession, each iU, portion of a week being astrologically regarded as the property fehi' or inheritance of some planet. Compare Ossete ¢ kuri, ¢ kore,’ a week; Georgian ‘kuire.’ Possibly these words are derived ety qnd from the Greek ¢ kvpi-axy,” Sunday, the Lord’s day; but whence is the Greek word derived ? from ¢ «xdpe-os,” a Lord, a possessor, arigit the base of which seems to be allied to the Tamil ¢kira,’ keb—i;' POssession. 458 GLOSSARIAL AFFINITIES. kil-ei, @ young branch. Compare Ossete ‘kalius; Servian ¢galusa; Greek ‘w\ddos, o young shoot, @ branch. The theme of the Greek word is ¢ k\d-w,” to lop, to break,; and the Tamil ¢ kil-ei, considered as a verbal theme, means not only to sprout, but also to pluck off. ‘kill-u, to pinch, to pluck, is a collateral theme, kupp-ei, sweepings, refuse, dung, a dung-heap. Compare ¢ kompos,’ dung, dirt, a farm-yard. kur-u, skort, brief: derivative verb ¢ kuru-gu,’ to diminish: collateral root ‘kur-ei) a defect, to be or make defective. Compare Persian “chord,” short, German ‘kurz; Latin ¢curt-us,” short, small, defective. On comparing the Latin word ¢ curt-us, with such words as ¢sert-us, connected, from ser-o,” it may be concluded that °curt-us,” is derived from an obsolete verbal theme ¢cur-o,’ which would bé identical with the Tamil ‘kur-u. kuru-du, blindness, blind; ultimate base ‘kuru’ (like *kira,’ the ultimate base of ‘kiradu,’ old). Compare Persian ¢ kir ;’ Kurd ¢kor; Ossete ‘kurm, blind. kuri, Can., a sheep, Tuda °gurri.’ Compare Irish kaora; Georg. ¢ chhuri.’ kul-ir, cold, Tam. and Can.: ultimate base, by analogy, ‘kul; Telugu and Canarese ‘chali, cold : collateral root, <ilir, Tam., o tremble. ¢kiudal’ and ‘kudir, cold, are doubtless derivative or allied words. Compare German ‘kiihl; Saxon ‘eyl,” ¢ col, ‘cele;’ Russian ‘cholod;’ English ¢ cool,’ ‘cold; Latin ¢gelu;’ English ¢ chill.’—See also ‘Scythian Affinities.’ kél, to hear, Tam. and Can. Compare Latin ‘aus-cul-to, to hear, to listen; also the Greek kh\i-w,” to hear; Welsh ©clyw, hear- tng; Irish ‘cluas,’ the ear; Lithuanian klau,” to hear; Latin ‘clu-o,” to be called.—See also the Scythian affinities, which are still closer than these. kol, to kull. Compare Russian ‘kolyu,’ o stab; and especially the English ¢ kill” and ‘quell.’—See also ¢ Scythian Affinities.’ sakk-u, @ sack. Compare Greek °cdrk-os’ or ‘edr-os, @ sack.—See also ¢ Semitic’ and Scythian Aflinities. Greek lexicographers derive this word from a Greek etymon; but we can suppose it to have sprung from a Greek ta WEST INDO-EUROPEAN FAMILY. 459 base only on the supposition (which is an inadmissible one) 16 that the Greeks were the great carrying traders of antiquity. l The Tamil word ©sikk-u,” denotes a loose bag of coarse cloth, It but not also as in Hebrew and Greek, the coarse cloth itself of al which the bag or sack is made. Those languages, therefore, would appear to come nearer than the Tamil to the original source of the word. >\, satt-u, to close a door, to shut; Saxon ¢seytt-an,’ to shut tn; Dutch ¢ schutt-en,” o stop ; English ¢ to shut.’ 9l (7 ad-i (pr‘onounced ‘jadi’), a jar. Compare Spanish ‘jarra; English < 1ty jar th be sal, @ bucket. Compare oy\-ia,” any flat board or tray witk a raised rvm.—See also ¢ Semitic Affinities.’ al fivar-u, Can., a splinter. iil Compare English ¢a shiver. &ir-u (pronounced ¢siir-u’), to Aiss. Compare ©ovp-i{w, to ppe, also he to hiss; Latin €su-surr-us, a whispering, or whistling. Our English word ¢%iss’ is evidently mimetic; but ‘siru’ and its allied roots bear no trace of an imitative origin. ' $ud-u, Tam. and Can., to heat, to burn, to fire: related root ¢8ud-ar,’ to shine. Compare. Persian ‘sus-an; Kurd ¢sodj-an; Ossete ¢ suds-in, to burn. Compare also Sans. ‘sush-a,” and Latin ol ‘sicc-0, to dry. Probably Sans. ¢ kud-a,” to heat, to burn, 1] contains the same base. V6 ol éepp-u, to speak. Compare ‘ém-w’ (for “fém-w’), to speak. U el to go, to proceed. This is unquestionably a pure Dravidian root, and abounds in derivatives; e.g., ‘&el,” the white ant; *gel-avu,’ f0 expenditure; ¢Sel-vam, prosperity. It forms its preterite also (l in a manner which is peculiar to pure Dravidian verbs. It is obviously allied to the Sans. ¢shal-a,” fo go or move, o ich ¢shél-a,” to move, to tremble; chal-a’ and °char-a,’ Zo go, to shake, to totter; and also to the Hindustani derivative, ¢chal,’ to go. Close as these analogies are, ‘&el’ appears to bear an the equally close resemblance to ‘cel,’ the obsolete Latin root, sig- %: nifying to go, from which are formed celer, and also ex- cell-o’ and ¢ pree-cell-o.” The same root is in Greek ©«el ; S8 e.q., ‘kéx-ns, @ runmer, and ¢ kéX\w, to urge on. ek tag-u, fit, proper, worthy. Compare ¢ élc-awos, right, proper, just. ok B bl i . 460 GLOSSARIAL AFFINITIES tayir, curds. Compare ¢ rvp-os, cheese. tin, 0 eat,; tindi, food. Compare ©7évé-w, to gnaw, to eat daintily, “TeévOys, a gourmand. tir-a (pronounced nearly like ¢tora’), fo open; °tir-a-vu,’ an open- MY, & WAY, & MEAns. Compare Greek ¢ 6Ypa,” a door; German ‘thur; Old High German ‘tor; Gothic ¢daur; Sanscrit “dvara.’ These words are commonly derived from the Sansecrit theme € dvru,’ fo cover,; but as they all mean not the door-Zeaf, but the door-way, and metaphorically @ way, or means, this derivation is far inferior to that of the Dravidian ‘tira’ or ‘tora’ (Can. tera ), fo open. tind-u, to touch, to kindle. Compare Gothic ‘tandya, 7 kindle. Possibly there may be a remote connexion also with the Sans. ¢danh,’ ¢o burn, the intensitive of which is ¢ dandah.’ On the other hand, the ‘n’ of the Tam. tind-u’ is probably euphonic, for it disappears in the Can. °tid-u.’ tel, clear. Compare éfjA-os,” clear, manifest. tol-ei (base ‘tol’), distance, adverbially distant. As a verb, ‘tol-ei’ signifies to end, or come to an end. Compare ¢ Tij\-¢, far of, which Buttmann derives from ¢ 7é\-os, an end. tripp-u, Tel., to turn; also, by corruption, ¢tippu;* Can. ¢ tiru-pu ;’ Tamil ‘tiru-ppu.’ These are causal or active verbs, and the corresponding mneuter or intransitive verb signifying o turn, is in Tel. ‘tiru-gu, in Tamil ‘tiru-mbu.’ The Canarese has ‘tiru-hu,” ‘tiru-vo,” and °tiru-gun. There are also a few related themes; e.g., < tiru-gu, Tam., to twist or twrn; from which is derived ¢ tirugal,” a mill. From a comparison of all these words, it is manifest that their common base is ©tiru,” to which various formative additions have been made, for the purpose of expressing modifications of meaning. “tiru’ itself, also, has evidently arisen through the phonetic necessities of the language, from ‘tru’ or ‘tri, which is to be regarded as the ultimate base. Compare Greek ¢ wpém-w, to turn,; which bears a remarkable likeness to the Tel. ‘tripp-u,” and the initial portion of which (with that of our English ¢turn’), seems closely allied to the Dravidian base, ‘tru’ or ¢tri.’ Probably the Sanscrit ‘tarkun,” @ spindle, is not a collateral word, but one which has been directly borrowed by the Sanserit from the Dravidian tongues. WEST INDO-EUROPEAN FAMILY. 461 nas-u, fo crush, to squash. Compare ¢ vico-w,” to squeeze close, to stamp down. nar-a, Can., a tendon, a sinew, catgut; sometimes, but improperly, @ vein or artery: adjectivally wiry, stringy. Tel ‘narammu ; Tam. ¢narambu; Rajmahal ¢néru.’ ‘nara’ or ‘nar’ appears to be the ultimate base; with which compare the Latin ¢nerv- us’ and the Greek ¢ wvebp-ov, a tendon, a ligament. I nin-ei, ; to think, to remember; Can., ‘nen-i.” This word is undoubtedly a Tamil primitive, and is probably connected with “nefijju,’ § I Tam., the soul, literally the upper part of the chest, the dia- phragm. [With respect to this double signification of the word ‘ nefijj-u,’ compare the twofold meaning of ‘@pyv, in Greek, viz., the diaphragm or chest (supposed to be the seat of the mental faculties), and also the mental faculties themselves.] 6 If there is any analogy between the Dravidian “nin’ or Gy ‘nen,’ and the Sanscrit ‘man,’ fo think, it comes to light only by comparing it with the corresponding Greek word ¢ vé-cw,’ by reduplication ‘vevo-yuar. ‘ pva-opal, to think on, to re- member, and ¢ pév-os,” wish, are in perfect accordance with the Sanserit ‘ mana,” and are probably more ancient than ¢ vé-ew;’ I of which the initial ¢»* has possibly been changed from ¢ u.’ The Dravidian ‘nen’ or ‘nin’ has in like manner, I conceive, been changed from an older ‘men’ or ‘min,” allied to ‘man-a’ and ¢ ,ue'v-oe.’ 16 18 nind-u (also ‘mifij-u’), to swim; €nich-u’ and ‘nitt-al, swimming. 18 Tel. ¢id-u,; Can. ¢ich-u,’ to swim; Tel. derivative noun, ‘ita,’ W swimming. A comparison of these words shows that the final ‘ndu’ of the Tamil verb has been euphonised from ¢du.’ Il I have no doubt that the base of this verb is simply “nid’ or > {0 e of which ‘mi, Can., to bathe, is probably a collateral form. 16 Compare ¢ni with the Latin ‘no’ (‘navi’), fo swim; lf, of Greek vé-w, also € vyj-xw; Sanscrit ‘nau; Greek ¢vad-s, a boat. ih) Compare also ‘nid-u’ (the supposititious original of h both ‘nind-u’ and ‘id-a’) with the Latin secondary verb 16 ¢ nat-o. Bopp derives these Indo-European words from ¢sna,’ Sans., to bathe; but their root is not, I conceive, in Sanscrit. 1)’ It is only in the Classical and in the Dravidian tongues that it is found, al b ney, to weave. Probably ‘nil, @ thread, and also to spin, is a word 462 GLOSSARIAL AFFINITIES. of collateral origin. As pal, milk, may have been derived rom ‘pa,’ to drink, and ‘til, dust, from ‘td;’ fo scatter, so ‘0l @ thread, may be supposed to be derived from an obsolete ‘nd,” ¢o spin: and this root would naturally be concluded to be a correlative of ¢ney, to weave. Compare ©vé-w, to spin, ‘vi-pa,’ @ thread; and more especially the Latin ‘neo,” which not only means to spin, to entwine, but also, secondarily, to weave; e.g., tunicam quam molli neverat auro.’ Virg. A collateral root, and one which bears, perhaps, a still closer avalogy to the Dravidian ¢ ney,’ to weave, is that which we find in the German ‘nih-en,” ¢o sew; Latin ¢nec-to, to knit, to join; and Sanscrit ‘nah-a, fo bind, to tie. pad-u, to suffer, to recetve or feel an tmpression,; a word which is used as an auxiliary in all the Dravidian langnages in the formation of passive verbs: derivative noun, ¢ pat-u,” Tel. and Can., « suffering. Compare Latin ¢pat-ior, and Greek ¢maf-eiv,’ each of which has precisely the same meaning as the Dravidian verb. pa(}—u, Tel. to fall. This verb is identical in Telugu with the pre- ceding one; but the meaning, to fall, which it bears in Telugu, in addition to that of fo suffer, suggests a different set of affinities. Even in Tamil it means ¢o k¢, or to light, as well as to suffer, or receive an impression. Compare Slavonian ¢ pad,’ to- fall; Sanscrit ‘pat, to fall, to fly; Zend ‘pat, to fly; Latin ¢pet’ in im-pet-o,’ to fall upon; Greek ‘mér-ouar, to My, and alse ¢ wirr-w,” to fall. L pan, to make, to work, to produce: colloquial form ¢pannu; Tel. ¢ pannu.’ This word is evidently allied to the Sanscrit ¢ pan-a, pa to do business, to megotiate; the noun corresponding to which, ‘pana,” means business, hence property. This noun, ‘pana, has been borrowed by the Dravidian languages; but the only signification which it bears is money. Whilst ¢ pana-m,’ money, is always admitted by Dravidian grammarians to be a BSanscrit derivative, they regard ‘pan(n)-u,’ to make, to work, as a primitive Dravidian word; and this view is confirmed by the circumstance that it stands at the head of a large family of derivatives and collaterals; some of which are pann-ei, tillage, a wrice-field; ©pan-i,) serwice, humility; °panikku,’ « design, a clever performance; ‘pani,” Malayalam, dificulty, toil ; — A o — T DR— WEST INDO-EUROPEAN FAMILY. 463 { ‘pani,’ Tel., work. It is especially worthy of notice that 0 ‘pan-i, as a verbal root, signifying ¢o be subservient, to obey, to £ worshap, has become in its turn the parent of a host of deri- 0 vative words. I have no doubt that ¢ pan,’ to make, to work, has an ulterior { connexion with the Sanscrit ¢ pan-a,’ fo negotiate; but it appears fo to have a still closer connexion with the Greek ¢mov-dw, to toil, to work hard, ¢ mov-os, work, a task, and ¢wév-ouar, to work, to toil. Compare also the Babylonian ¢ban-as,’” to do, to il make. ch 1] pamp-u, Tel, to send: a softened form of the same word is ‘ampu,’ from which is derived the Tamil ¢anuppu,” to send, and ‘ambu,” an arrow; also the Tel. ‘ampa,” an arrow, and ed ‘ampakam,’ dismission. It is obvious from a comparison of these words, that the Telugu has best preserved the original on form. { Telugu grammarians suppose ‘pampu,’ to send, to be a causal from ¢ pd,” to go; and it is certain that some causals are formed in Telugu by adding ‘mp’ to the root. This supposi- of tion, however, would lead us to expect ‘pémpu’ instead of ‘pampu;’ and it is inconsistent with the existence of a causal 16 formed from ¢pampu’ itself, viz., ¢ pamp-i-inchu, to cause to 4 send, corresponding to the Tamil ‘anuppu-vi. I therefore, of think that ¢ pamp-u,’ ¢o send, should be regarded as a primitive word. 3 d, Compare the verb ¢ pamp-u’ with Greek ¢meum-w, to send, i and the noun ,_‘ pamp-u,’ @ dismission, with ¢ woum-1). {0 pal-e, Can., old, long in use, of ancient date; Tamil ¢ para.’ Compare ¢ rakat-os, lold, ancient, antiquated ; * mi)\azl,’ in olden time. el par-u, to fruit, to become ripe; *para-m, a ripe fruit; Can. ‘pala’ Hy (‘r’ changed into ‘1’); Tel. ‘pandu’ (‘r changed dialec- lCh, tically into ‘d’ and then nasalised). Compare Persian ¢ ber,’ nd Jruit; Armenian ¢ perk; Latin ¢fru-or, ¢fru-x, Compare »nl): also the Sanserit ‘phala, fruit; a word which has been Iy borrowed by the Tamil in the sense of effect or profit, but 0 & which is never confounded by it with its own ¢para-m.’ I 07"1'7 suspect this root to be identical in origin with the preceding by one In Tamil to be old or long vn wuse, is ‘ para; to be ripe, pily ‘paru; and both the words themselves and the ideas they £y express seem to be allied.—See also, however, the Semitic ! analogies of this word. 'O‘l'l} | 464 GLOSSARIAL AFFINITIES. N A oA pal, many, wvarious; ¢pal-ar, many people; ©pal-a, many things. The ordinary adjectival form of this word, which is used with- out discrimination of number or gender, is pala;’ but pal’ is more classical. There is also a verb formed from the same base, ¢ pal-gu,’ to become many, to be multiplied, to increase. If there is any connexion between this word and the Sans- crit “puru’ (for ¢paru’) much, it is a very distant one; whereas ‘pal’ appears to be closely allied to the Greek “molvs, ¢ mohv, many, much, and the Latin ¢ plus.’ Compare ¢palar,” many persons, with ‘6. wo\\o,’ the many, the majority. The Sanscrit ¢ puru’ is derived from ¢ pri,’ to fil{ (‘ pi-par-mi’); but the Tamil ‘pal’ is an ultimate root. pall-i, a town, a willage, a school, @ mosque; in Travancore a church, and generally a place of concourse. Compare molws, a city, i from ¢ 7oléw, to haunt, to frequent. palli’ is found in Sanscrit dictionaries; but I consider it to be a purely Dravidian word. jul piykk-u, to rend in pieces, to card, to comb cotton, to pick. Compare ¢ mék-w, to comb; English ¢to pick.’ pir-i, to divide; also ‘por,’ to cleave. Compare Sanscrit ¢ phal-a,’ o dwvide; but especially the Latin ¢ par-s,” @ portion, also ¢ por- tio,” from the supposititious root ¢por-o’ or ‘par-o,” fo appor- pe tion, to divide. The Greek ¢ mop-w,” in the sense of im-par-ting, is doubtless an allied word. The closest and most remarkable analogies, however, are those which we find in the Semitic * vocabulary,—which see. pill-ei, Tam., @ ckild; Tel. “pilla; Can. pille;’ Latin ¢ puell-us,’ If the Latin word is derived from ‘puell-a, a boy, a girl. ¢ puer-ulus,” it is probably unconnected with ‘pill-ei.” Perhaps a more reliable affinity is that of ¢fil-ius,” “fil-ia, @ som, a daughter, supposed to mean literally a suckling.—See also the PeR ¢ Scythian Vocabulary.’ pugar, to praise. Compare Old Prussian ¢ pagir-u’ 7 praise, and the corresponding noun ¢ pagir-sna, praise. pur-am, a side, especially the outside, the exterior; e.g., ‘appuRam,’ that side; ippuram, this side: adjectivally ¢ pur-attu,’ ex- ternal: adverbially ‘ purambéga’ ( puram-b’-aga’), externally: as a verbal theme ¢ purappadu’ (‘ pura-(p)-padu’), to set out : Can. ‘pora-ge,’ outside: ‘pora-du,’ to set forth. . There is, 465 WEST INDO-EUROPEAN. S doubtless an ulterior connexion between °pura-m,’ the outside, externally, and ¢ pira,” other yet they are not to be regarded 1§ as one and the same word; and ‘puram’ has affinities of its own, as well as meanings of its own 16 Compare Greek ¢ mapa, beside, in which one of the meanings of the Drividian word § appears, whilst the meaning of side is not conveyed by the correlative Sanscrit ) para Compare especially the Latin ek ¢ foris abroad ‘forum,” a public place; *fori the decks of a e ship, with the Canarese pora, outside This seems a more natural derivation of ¢ foris iy than the Greek ¢ 0ipa,’ a door word which I have connected with the Dravidian ¢tira,” fo 7)’ open In the Dravidian Janguages ‘f’ is unknown, and ‘p’is always used instead ch, pls-ei, @ cat, especially in the South-Tamil idiom In the Cashgar by dialect of the Affghan, pusha ’ signifies @ cat Compare Irish ‘ pus a cat English ¢puss 1t pill-1, Tel., a cat pul-i signifies @ tiger, or more commonly a cheetak, or hunting leopard, in all the Dravidian dialects, and e a cat also in Canarese Compare Persian ‘pelang,’ a tiger, but especially the Latin ‘feles’ or ‘felles,’ @ cat, a word which f0 is also used to denote various animals of a similar character or" per- great, also ‘per-ia another form of the same adjective in 07 Tamil, and probably a more ancient one, is ¢par-u Possibly /2‘/, pal-a,” many, is a related root, seeing that there appears to ble be the same relation between ‘per-u 0 par-u,” great, and i SIR-U1 7 E small, and sil-a, pal-a, momy, that there is between Jfew ‘per-u,” great, is also used as a verbal theme, and in 18 that connexion it signifies ¢0 increase. ‘varh’ and ¢vih,’ to grow o Compare Sanscrit ¢ puru,” much but especially the Zend ‘berez’ and ‘barez,’ great aps y verbal ‘noun PeRr-u, to bear, to bring forth, to obtain, to get or beget the Péer-u, a bringing forth or birth, a thing obtained or a benefit PIR-a to be born, to proceed from, is doubtless a the b related word; and there is probably a relationship between these words (especially the latter) and ‘pira,’ otker, foreign F pura-m, the exterior, and even ‘por-u, to bear or sustain. Ll g s Compare the Latin par-10 pe-per-i, to bring jorth, to 73 acquare Possibly the ultimate base of all these words is the Iy Indo-European preposition, ¢ pra,’ signifying progresswwe motion, i expansion, excess, &c.; and the Zend form of this preposition, = 15 2 H o A —— 466 GLOSSARTAL AFFINITIES, ‘fra, indicates the propriety of classing the Latin ¢ fru-x’ with the other derivatives. See also the Semitic Vocabulary. péy, a demon. Compare the English ‘fay,” an elf; French ‘fée. peiy-an, a boy, a servant; also ‘ pay-an,” ‘pay-al,” and ¢ peiy-al:’ Mala. fal is that of ¢ pei-tal ;7 Can. ¢hei-da.’ The termination the neuter verbal noun; and consequently ¢ payal’ might be applied to a youth of either sex, though restricted in Tamil to the masculine. The Malayalam has ‘an pei-tal,’ @ boy, ¢ pen pei-tal,” a girl. Compare Greek ‘wais, ¢ 7waié-os, a boy or girl, a servant; Laconian ‘mowp ; Latin ‘puer; Persian ¢ bach,” a boy, ¢ puser, o 1l son; Swedish ‘poike; English ‘boy.'—See Scythian Affinities. POR-1t, to sustain, to bear, to suffer patiently: ‘por-u-ppu,’ respon- sibilaty ; ¢ por-u-mei,” patience. Compare Gothic bair-an,” to bear,; Greek ‘@ép-w; Latin ‘fer-o. The Tamil distinguishes between this word and ¢pir-a,” to be born, though both are probably from the same base. The Latin in like manner dis- tinguishes between ¢par-io’ and fer-o,” whilst the Gothic tongues make no difference between ¢bear, fo sustain, and “bear,” to bring forth. They constitute one word, from which is formed the past participle to be born or borne, and also the 1l noun birth. por-u-du, fvme: theme por. Compare Sanscrit ¢ var-a,’ fume; Per- sian ¢ bar,” the theme of ¢ bari, once; Latin ‘ber, the suffix of time, which appears in the names of the months from Sep- tem-ber to Decem-ber. povrv-u, to 7ise, to be puffed up, like bread: a real Tamil word, though a local, vulgar one. Compare English ‘to puff, and Dutch ‘boff-en,’” also ‘pof,” a blast which swells the cheeks. po, fo go; also ‘pd-gu’ (with the usual formative addition of ‘gu’). The second person singular of the imperative of ¢pdgu’ is ‘ po.’ Laghmani (an Affghan dialect), ‘pak,’ to go; Greek ‘Ba-w,’ to go; Latin ‘va-do,” to march; Hebrew ¢bo,’ to come, occasionally Zo go. pod-u,’ to put. Compare Dutch ‘poot-en,’ to set or plant; Danish ‘pod-er, to graft; English to put. Bil—u, Can., fo fall; Tam. ‘vir-u” Compare English “to fall; Ger- man ‘ fallen.’ y WEST I'NDO-EUROPEA&. 46 i mag-an, ¢ son, @ male; Tulu ‘mag-e. Compare Gothic ‘mag-us,’ « boy, a son, from the verbal theme ‘mag,’ originally to grow, then Zo be able; Gaelic ‘mak,” « son,; Tibetan ‘maga,’ son-in- law. Compare also Latin ‘ mas,’ a male. b, of may-ir, hair. Compare Persian ‘mui; Armenian ‘mas,” Aair. e maRr-a, to forget. Compare Lithuanian ‘mirsz,” fo forget. to \e]‘] ma, a male, particularly the male of the lion, elephant, horse, and swine; e.g., ‘ari-ma,” a male lion. Compare Latin ‘mas,” a it male. i] marg-u, to die, to be bewildered, to mingle: velated theme ‘mar-u,” 7o 5, be confused, to be lazy. 01l Compare Latin < marc-eo, to wither, to be faint, to be 1] languid or lazy, and also the Greek ¢puapaivw, which in the hes passive voice signifies fo waste away, or die. Possibly all these are words have a remote connexion with ¢ mri,” Sanscrit, to die. li It is evident, however, that there is a closer and more hic special connexion between the Latin and Greek secondary nd themes here adduced and the Tamil. lich mig-u, much, great: as a verbal theme, fo be much. ‘ mifij-u,’ o the abound (from ‘mij,’ nasalised), is probably a collateral root. Related words, Tel., ‘migal-u,’ remainder, that which s too Per- much,; ‘migula’ and ¢migala,” adverb and adjective, much, ex- ifiix ceedingly, also ‘mikkili,’ the same. Can. ‘mig-u,’ to exrceed, also Sep ‘migil-u,’ both as a verb and as a noun; ancient dialect of Canarese ‘ migal,” much; °mogga,’ and also * mokkala,” @ mass, a heap, an assemblage. uch f The Sanserit ‘maha,’ great, from ‘mah,’ fo grow, is fre- ateh E quently used in the Dravidian dialects, but it is always considered to be a Sanscrit derivative, not the original base from which the Dravidian words have been derived. This o) view is confirmed by the circumstance that the Dravidian ook languages have no word signifying much, except ‘ mig-u,’ and . 1ts correlatives. (ot The Drévidian words quoted above, bear a much closer resemblance to the corresponding words in the Thus, Classieal and Germanic tongues than to the Sanserit. )a,nish the Latin ‘ mag-nus,’ ‘mag-is; the Persian ‘mib’ or ‘meah; the Greek ¢ uéya’ or ¢ peyalos; the Old High German ¢ mih- hil; Norse ¢mikil; Danish ‘megen; English ¢migh-t ; (rer- Scottish ¢ mickle,” are more closely connected with the Tam. 2. °H 2 468 GLOSSARIAL AFFINITIES. ‘migu,’ the Can. ‘migal’ and ¢ mokkala’ and the Tel. ‘migala’ and ‘mikkili, than with the Sans. ¢ mah-at.’ The final ¢1° of the Dravidian words seems to be ¢a particle of specialisation.’ —See the section on ¢ Roots.’ mirg-u, to plunge, to sink. ‘amir’ appears to be a softened form of the same word; and probably the ‘g’ of ‘mirgu’ is only a formative. Compare Latin “merg-o,” to plunge, to immerse. The preterite being ‘mersi,” not ¢ merxi,” possibly the ‘g’ of ‘mergo’ is a formative addition like the g’ of the Tam. ‘mir-g-u.’ mugil, Tam. and Ancient Can., @ cloud. Compare Sanscrit ‘mégha,’ a cloud, from ‘mih’ (‘méhati’), to sprinkle. The word ‘ mégha,’ has been borrowed from the Sanscrit By the Dravidian lan- The guages, and is now more commonly used than ¢mugil.’ latter, however, is found in the classics, is much used by the peasantry, and is undoubtedly a pure Dravidian word. Doubtless ‘mégha’ and ¢mugil’ are ultimately allied; but there is a direct and special connexion between the Dravidian word and the Greek ¢ ¢-u/x\-1,” a cloud, the Lithuanian ‘ migla,’ the Slavonian ‘mgla, and the Gothic ‘milh-ma; in each Qf which the ‘1’ of ‘mugil’ retains its place. muyal, fo labour, to endeavour. Compare Latin ¢ mol-ior,” to endea- ,% vour, to strive; Greek ¢u@d\-os, the toil of war,; English ‘to moil,” to labour or strive. i) muRumuru, to grumble, to murmur. A very similar word, ‘ moru- moRru,” to murmur, would naturally be regarded as identical with ¢ murumuru; but a different origin is ascribed to each. ‘morumoRru’ is said to be simply and solely a mimetic word, § one of a large class of imitative, reduplicated exclamations; e.g., ‘ he said ¢ moru-moru ; i.e., he spoke angrily: ‘his head said ¢ kiru-kiru i.e., 16 went round. ‘murumuru,’” on the other hand, is not purely imitative, but seems to be regularly ra) formed by reduplication from ‘muru,’ the base of ¢ murukku,’ to twist, to chafe; and the signification of grumbling, and being discontented, has arisen from that of ckafing. Yal Whatever be the derivation of the Tamil word, it may be compared with the Latin ¢ murmuro,” to mutter. The Latin R B T e R word is evidently an imitative one, the reduplication: of the syllable ¢ mur’ being used to signify the continuance of a low muttering sound. ¢mur’ has doubtless some connexion with s st ah e N T O R T e e WEST INDO-EUROPEAN. 469 the base of ‘musso,” ‘mussito,’ to mutter or grumble. Compare also the Greek expression fo say ©ud us,” to mutter, to grumble. The Old Prussian ¢ murra,” to murmur, is evidently related. —=See also the ‘Scythian Vocabulary.’ The Tamil word means not only ¢o utter a muttering sound, f but also to ewpress discontent, to be angry; and in this it goes i) beyond the meaning of the corresponding Latin ¢ murmuro. Muttering, is in Tamil expressed by ‘muna-muna,” a some- of what similar, yet independent, imitative word. | mukk-u,” the nose: theme ‘mig-u, Can., to smell; related Tamil verbs ‘mugar’ and ‘mokk-u,” to smell. Compare Greek ¢ uvemip, the nose. The Greek word is dy said to be derived from ¢ udCw, to moan, to mutter, to suck in, or from °uifa, the discharge from the nose (Latin € mucus’). he It 1s worth consideration, however, whether the Dravidian by derivation is not, after all, a more probable one. 1 b mett-a, Tel., a bed, a cotton bed, a cushion; Tamil ‘mett-ei; Canarese ‘ mott-e.’ an The word seems to be a derivative from ¢ mel, 3, Tam., soft, fine. of Compare Latin < matta,” @ mat, @ mattrass; Slavonian ‘mat ;’ Saxon ‘meatta:’ also the Hebrew “mittah,” a bed, a cushion. The Tamil seems to give the best and most natural deriva- el tion of these words. 4t0 mel, fine, thin, soft, tender; mell-a,’ softly, gently. Compare Latin ¢ moll-is,” soft, tender, pliant; Greek ¢ paa- RI- Ko<, Soft, gentle, tender. The derivation of the Latin ‘mollis, il from ‘movilis,’ is inconsistent with the connexion which cer- ach. tainly subsists between ‘mollis’ and ¢ patacos; and the re- '01'(]1 semblance of both to the Dravidian ¢ mel > is remarkable. IH Compare Sauscrit ‘mridu,” soft, which is in Tamil ¢ med-u.’ head the ray, Tel., a stone. Bearing in mind the mutual interchange of ‘r’ and larl)" ¢l we may perhaps compare this word with the Greek ? ‘Na-as or ‘\ai-a, a stone. Ko, jein] val-i, strength; ¢val-i-ya,’” strong; ‘van-mei’ (‘val-mei’), strength. The Drévidian languages bave borrowed, and frequently use, }y be the Sanscrit “bala’ (in Tamil ¢balan,” ¢balam,’ and even [atin f the ‘valam’) ; and it might at first be supposed that this is the origin of ‘vali, &e. I am persuaded, however, that the words ) Jow cited above have not been derived from the Sanscrit, but have with | 470 GLOSSARIAL AFFINITIES. been the property of the Dréavidian languages from the begin- ning. The Dravidian ‘val’ has given birth to a large family, not only of adjectives and nonns, but also of derivative verbs, which have no connexion whatever with anything Sanscrit ; e.g., ‘vali) a spasm; ‘vali, to row, &c.; and if this word is not to be regarded as Dravidian, this family of languages must be supposed to be destitute of a word to express so necessary and rudimental an idea as strong. “val, also, more closely resembles the Latin ¢ val-eo,” to be strong, and °val-idus,” than the Sanscrit ¢ bala-m.’ val, fertility, abundance; ¢ val-ar, and many related verbs, to rear, to cause to grow. Compare Latin ‘al-o,” to nowrish. Connexion doubtful. E The vind-u, the wind. Compare Latin ‘vent-us; English ¢ wind. 2 Tamil word is said to be derived from ¢ vin, the sky: its resemblance to ¢ vent-us’ is, the}refore, probably accidental. virel, fo shiver from cold, to grow stiff from cold. °vR1’ was pro- I bably the primitive form of this theme. Compare Greek ° ¢pios-w, to tremble, to shiver; puy-éw,’ to ¢ It shiver or shudder with cold,; © piy-os, frost, cold, a shwvering 0 from cold; also Latin ‘frig-eo,’ to be cold; °frig-us,’ cold; § ‘rig-eo, ‘rig-or,” to be styf, as from cold; English ‘to freeze.’ vin, wuseless, vain. Compare Latin ‘van-us, empty, unreal, frivolous, €8 vaIn. h th vend-u, fo wish, to want. Compare English ¢ want’ from Saxon If the ‘n’ of the Tamil word is euphonic, fu ¢ wanian,’ fo fal. as it appears to be from comparison with ‘bed-u,” the corre- hy sponding Canarese word, this resemblance is merely accidental. i W] VER-U, different, other. Compare Latin ¢ var-us,’ the secondary mean- Vi ing of which is different, dissimilar; ‘also ‘ var-ius,’ diversified, 1t various, different from something else. Elpw 001 vray, Tel., to write; Tam. ‘erud-u. If these words were originally be _identical, as it is probable they were, the Telugu °vray’ must have lost a final <d.’ or Compare English ‘to write’—See also ‘Scythian Affinities.’ tuf ap] —— R— R ——— — EE—— - S—— 471 IT [’v N Semitic Affinities, OR, 18 st Dravidian words which appear to be allied to the Hebrew and its stster tongues. i) )h The number of such words in the Dravidian languages is not all great; and it might be supposed that in attempting to prove the existence of this class of affinities, in addition to aflinities of the Indo- fo European and Scythian classes, I prove nothing by attempting to prove too much. [ submit, however, the following list of words to 01 the scrutiny and judgment of those who may ‘entertain this supposi- tion ; and they will find in it, if I mistake not, clear evidence of the fie existence of a few highly interesting points of resemblance between Its the Dravidian vocabulary and that of the Hebrew. In some of the instances which will be adduced, the Semitic vocables are allied to the Indo-European as well as to the Drividian 10 languages; but it will be found that the Dravidian analogies are closer and more direct than the Indo-European, and it is for that fo reason that the words are inserted in this list rather than in the pre- ing ceding one. I In some instances, again, the only analogies to th'e Semitic vocables are such as are Dravidian. If the existence of Semitic affinities in the Dravidian languages is DIl established, those aflinities cannot be explained by supposing them to have been introduced by the Jews who have settled on some parts of the Malabar eoast ; for the Jews, whether ‘black’ or ‘white,” have care- X0l fully preserved their traditional policy of isolation : they are but a small »niC, handful of people at most; they have never penetrated far into the 17e- interior, even on the Malabar coast, whilst on the Coromandel coast, tal. where the Tamil is spoken, they are entirely unknown ; and the Dra- vidian languages were fully formed, and the Tamil and Telugu were, oAl i it is probable, committed to writing long before the Jews made their appearance in India. Whatever words, therefore, appear to be the common property of the Hebrew and the Dravidian languages, must mlly be regarded either as indicating an ancient, pre-historic intermixture nust or association of the Dravidians with the Semitic race, or as consti- also tuting traces of the original oneness of the speech of the Noachidae. This word is found unaltered in app-a, father! vocative of ¢app-an.’ all the Dravidian dialects. The Mech also, a Bhutan dialect, has “appa’ for father; the Bhotiya ‘aba; the Singhalese p— 472 GLOSSARIAL AFFINITIES “appa.’ Analogies will also be found in the Scythian voca- bulary. In all the languages of the Indo-EKuropean and Semitic families, the ultimate base of the words which denote father, is ‘p’ or ‘b,” and that of the words which denote mother is ‘m.’ The difference between those two families consists in this, that the Indo-European words commence with the consonants ‘p or ‘m; e.g., ‘pater, ‘mater; whilst in the Semitic languages, those consonants are preceded by a vowel; e.g., Hebrew €ab,’ father, < ém,” mother. In this particular the Dréavidian lan- guages follow the Semitic rule; e.g., Tam., ‘app-an,’ father, ‘amm-al,” mother. The resemblance between ‘appan’ (vocative appa’), and the Chaldee €abba,” father (Syriac € b ’) is very remarkable. It is so close, that in the Tamil translation of I Gal. iv, 6, < ‘abba,’ father, there is no difference whatever, either in spelling or in sound between the Aramaic word ¢ abba’ (which by a phonetic law becomes ¢appa’ in Tamil), and its natural and proper Tamil rendering ‘appa; in consequence of which it has been found necessary to use the Sanserit derivative ¢ pita-(v)-6, instead of the Tamil ‘appa,’ as the translation of IR the second word. amm-4, mother! vocative of ‘amm-ei’ or ‘amm-al,’ mother. ur Compare Hebrew ém’ or ‘imm,’ motker; Syriac ‘4mo.” See also the Scythian and Indo-European affinities of this word, which are still closer than the Semitic. ur K Ar-u, @ riwer,; Tel. ‘ér-u; Malayalam ‘4Rr-a correlative root ¢éri,’ Tamil, @ natural reservoir of water. Compare Hebrew ¢or’ or R ‘yeor,” a river; Coptic ‘jaro.’—See also ‘Scythian Analogies.’ al, not. In Tamil, ‘al’ negatives the attributes of a thing; ¢il,” its il The existence: ‘él’ (and sometimes ‘al’), is prohibitive. vowel is transposed in Telugu, and ‘1&’ (the base of ‘lédu’), used instead of ¢il. Compare the negative and prohibitive particles of the Hebrew, al’ and €16 also the corresponding Arabic and Chaldaic ‘1. ¢16° in Hebrew negatives the pro- perties of a thing, like “al’ in Tamil, and another particle (‘ain’) is used to negative the existence of it. This idiom is ki one which remarkably accords with that of the Dravidian languages. ik Compare also the Chaldee ‘léth,’ it is mot, a compound of “]a,’ the negative partiqle, and ¢ith,’ the substantive verb i s SEMITIC. 473 0~ (a compound resembling the Sanscrit nasti’), with the cor- responding Telugu ‘lédu,” 7t is not, which is compounded of tie 1é,” the negative particle, and ‘du,’ the formative of the third I person neuter of the aorist.—See also ¢ Scythian Affinities.’ gl i av-a, desire: a related word is ‘Aval, also desire, which is a verbal D noun derived from the assumed root €Av-u,’ to desire (Marathi ‘avad,” love). Compare Hebrew ¢ avvah,” desire, a verbal noun g, ( ﬂb,, derived from ‘avah, to desire. The ultimate base of the Hebrew ‘4v’ or ‘av’is identical Jan- not only with the Tamil ¢av’ or “av,” but with the Latin ‘av- ther, tive eo,” to desire, and the Sanscrit ¢ av-a,” of which to desire is one of the rarer meanings. Compare also Hebrew ¢abah,” to will. very n of ir-u, to be; Brahui ‘ar’ Compare Babylonian ‘ar,’ fo be; also Coptic ther ‘er’ or ‘el, and the Egyptian auxiliary ‘ar. hich The Dravidian word appears to mean primarily fo si, fural secondarily to be, i.e., to be ¢ simpliciter,” without doing any- iich thing. tive n of iR—a, the ultimate base of ‘irRa-ngu,’ neuter, fo descend, and its tran- sitive € ira-kku, to cause to descend. Compare Hebrew ¢ yarad’ (biliteral base ¢ yar’), fo descend. 5 mo. ur-i, Can., to burn; Tamil ‘er-i. Compare Hebrew ‘tur, fire, ‘or, light.—See also Indo-European Affinities.” vord, [4 A 9 ar, a city, a town, a village. Compare Hebrew ar or *Mr a oy, Babylonian ¢er. ‘b, ' o eR-1, to cast, to shoot. Compare Hebrew ¢ yaradh’ (biliteral base gies.’ yar’), to cast, to shoot. erum-ei, a buffalo; Gond ‘armi, a ske buffalo; Telugu ¢yenumu ;’ ) its Canarese ‘emmeyu.’ These synonyms (in which ‘m’ is used The as a radical) seem to prove that the final ‘mei’ of the Tamil i) word ¢erumei, has no connexion with ‘mei,’ the formative jitive termination of Tamil abstract nouns, but represents an essen- yding tial part of the root. Compare Hebrew ‘rém, a bugfalo or I)I'O wild ox. ricle *)m iS kir, a sharp pownt. Compare Hebrew ¢Kkir,” to pierce, to bore. dian gakk-u, a loose bag of coarse cloth, a sack ; Ma]a,yélam ¢ chakk-a.’ Compare Hebrew ‘sak, a sack.—See also the Greek affinities od of of this word. 1 it 18 474 GLOSSARIAL AFFINITIES. We find the same word in many languages; e.g., Celtic ‘sac; Finnish ¢sakki; Magyar ‘saak.’ The use of this word in Genesis xlii., is unquestionably more ancient than its vse in it Greek, or in any other Indo-European tongue. The Baby- i lonians having been the great carrying traders of the earliest period of human history, it would seem probable that the word It is remark- “sak ’ was originally theirs, and if so, Semitic. able that though this word is in Tamil, it is not in Canarese 10 or Telugu; nor is it in Sanserit. Possibly it is a relic of the commerce which the old Phenicians and Hebrews carried on with the Malabar coast. They brought with them to Palestine A the Old Tamil and Malayalam name of the peacock, ¢ tok-a’ (‘thuka’ in the Book of Chronicles); and they left behind them in India their word for sack. 2 A sa l, a bucket. Compare Hebrew ‘sal, a basket; Greek ¢op\-la, a tray. §ay, to lean, to wncline. Compare Hebrew ‘sha’an’ (biliteral base, ‘sha’’ or ‘sha’), to lean. péll §ina-m, anger: verbal theme ‘&ina-kku,’ to be angry. Compare Hebrew ¢sané; Chaldee ‘sené, to hate; Hebrew °sinah, hatred. The corresponding Canarese word being ¢kini,’ fo be offended, ‘sina-m’ is probably softened from ‘kina-m, The analogy is therefore somewhat doubtful. §ir-u, to hiss. Compare Hebrew ‘sharak’ (biliteral base ‘shar’), o hiss; Greek ¢ ovplCw,” to pipe, to hiss. fum-ei, a burden: verbal theme ‘suma-kku,’ ¢o bear, to carry. Com- pare Hebrew ‘simak’ (biliteral base ‘sam’), to support, to uphold, to weigh heavily on. suv-ar, @ wall. Compare Hebrew ‘shir, « wall. gevv-el, equal, level, correct: base ‘Sev’ or ¢Se. A nasalised, adjec- JéR tival form of the same root is ‘sen; e.g., ‘en-Tamir, correct Tamil, the classical dialect of the Tamil language. From ¢se, “sev,” or ‘sen,’ is formed ¢ semm-ei’ (‘sen-mei’), an abstract of the same meaning as ‘sevvei. Compare Hebrew ¢shavah; Chaldee sheva’ (biliteral base ‘shav’ or “shev’), to be equal, to be level. If the Sanscrit € sama,” even, is at all connected with the Tamil ‘sev’ or ‘sen,’ the connexion is very remote ; whereas the Tamil ‘and the Hebrew word seem to be almost identical. SEMITIC. 475 tie natt-u, to fix, to set up, to establish: ulterior verbal theme ‘nad-u,’ to plant. Compare Hebrew ¢natd’’ (biliteral base ‘nat’), to »rld 10 plant, to set up, to establish. )y est nitt-u, to lengthen, to stretch out,; formed by causative reduplication of the final consonant from nid-u’ (also “nil”’), long. Compare i Hebrew ‘natadh’ (biliteral base ‘ nat’), to stretch out. k Bse nokk-u, to look direct at, to address. Compare Hebrew ‘nokah’ (base the ‘nok’), straight forward, over against. on 116 par-u, to become ripe, to fruit; para-m, a ripe fruit. Compare Hebrew ¢ pardh,” to be fruitful, to bear fruit; ¢parah,’ fo blossom, to break forth: biliteral base of both ¢par.” Especially \ind compare peri, fruit. Compare also Arnienian ‘perk, and Persian ¢ ber,” fruit. Doubt is thrown upon the aflinity of 1} these words with the Dravidian ¢ par-u,” by the apparent con- nexion of that word with accustomed. ‘ par-a,’ Tam., to become old, to be 156, pal, a part, a portion; « class; Canarese ‘pal-u:’ collateral Tamil are roots ‘pir-i,) to divide; °pil-a, also ¢ por, to cleave. The 7 and ¢1’° confirms ah, regularity of the dialectic interchange of ‘r the supposition that ¢ pal’ and ¢ pir-i’ are related roots. Com- o be The pare Hebrew ¢pélah, ¢pala, ¢palah, ¢palag, ‘palal; and also (by the interchange of ‘r’ and ¢1’), ¢parash,” ¢paras; and Chaldee ‘peras,’ to separate, to divide, to distinguish, &ec. ) { All these words, like the Tamil ¢pal’ and “pir-i’ (and also ¢ pagir,’ to divide), include the idea of separation nto parts.— See also the Indo-European analogies of these roots; e.g., lom- 10 Sanserit ¢ phal-a,” fo divide; Latin °par-s, and ¢por-tio,” a portion. It is evident, however, that the Semitic analogies are the closest. peRr-u, to obtain, to bear or bring forth, to get or beget: verbal noun Jjec- ‘per-u,’ @ bringing forth or birth, a thing oblained, a benefit: preet collateral root °pir-a,” to be born; Gond ‘pirra, fo spring ‘gp, Jorth ; also ‘pira, Tam., other, different. Compare Hebrew tract‘ ‘parah,’ to be fruitful; ‘ peri, fruit; ©parah, to blossom, to jah break forth. Whether the connexion between ¢ par-am, Tam., qllﬂl’ and ‘peri,’ Heb., fruit, be real or only apparent, I have no el doubt of the existence of an intimate relation between ¢ per-u,’ 0te ) to bear, ‘pir-a, fto be born, and the Semitic words which are mosh here adduced, as well as the Latin ¢ par-io,’ ¢ pe-per-i.’ 476 GLOSSARIAL AFFINITIES. ba, Can., to come, Gond ¢ wai; Tamil ‘va’ or ‘var; Telugu ‘ra; :; Tuda ‘wo.’ Compare Hebrew ¢ bo,” to come, to come in; Babylonian ba,” ¢o come | Com- may, to die, to put to death. Compare Hebrew ‘mith,” to dee. pare also ‘muwo, dead, in the Lar, a Sindhian dialect. P ey RSP s LSS | maRr-u, to change; Can. to sell. Compare Hebrew ¢ miur,’ to change % or exchange, of which the ‘niphal’ is ¢namar, as if from a T base in ‘marar’ or ‘mar. The corresponding Syriac ¢ mor,’ means to buy. misukka-n, a poor, worthless fellow; ‘misukk-ei, a worthless article. Compare Hebrew ‘miskeén,” poor, unfortunate. i The Hebrew word is commonly supposed to be derived from ¢sakan; but Gesenius considers the ‘m’ of ‘miskén’ to be a radical letter, not a servile. This word has found its way (probably by means of the Saracens) into several Euro- pean languages; e.g., French ‘mesquin.’ The Tamil does not contain the root of this word ; and it may therefore be supposed to have borrowed it an early period from some Semitic dialect, after the manner of ¢ sikk-u,” a sack. At all events, the coincidence between ¢ miskén’ and ¢ misukkan,” is one which deserves to be noticed. R R S o o R A S D mett-a, Tel. (Tam. ‘mettei;’ Can. ‘motte’), a bed, a cotton bed, a cushon. The Dravidian word appears to be derived from ‘mel,” soft. Compare, however, the Hebrew ‘mittah,” a bed, cushion, a litter, from ¢natah,’ to stretch out. | I Scythian Affinities, OR, Dravidian words which exhibit a near relationship to words contained in some of the languages of the Scythian group, particularly to the Finnish dealects. The majority of the affinities that follow are clearer, more diréct, and of a more essential character than the Indo-European or Semitic affinities which have been pointed out in the preceding lists. Many of the words which will be adduced as examples are words of a primary character, and of almost vital necessity—words which carry SCYTHIAN 4 (s authority and convey intuitive convietion, in comparisons of this kind I Some of the Dravidian words in the following list have Sansecrit or Indo-European affinities, as well as Scythian; a very few also have i} Semitic affinities ; but I have preferred placing them in this list, because the Scythian affinities are not only the most numerous, but the closest that appear to exist. Such words, though they are but few, are of e peculiar interest, as tending to prove the primitive oneness of tbe | & Scythian and Indo-European groups of tongues. o, For the important Scythian affinities which are apparent in the Dravidian pronouns and numerals, see the sections devoted to those dle parts of speech akk-a, Can and Tel elder sister, Tamil ¢akk-al Marathi aki ved In Sanscrit {0 akka, signifies a mother and an improbable Sanserit derivation has been attributed to it I believe this 1ts 110 word to be one of those which the Sanserit has borrowed from the indigenous Dravidian tongues; and the proof of this sup- not position is furnished by its extensive use in the Scythian \ed The Sansecrit signification of this word, a mother ot group differs, it is true, from the DrAvidian, an elder sister; but a and comparison of its significations in various languages, proves that it was originally used to denote any elderly female relation, and that the meaning of the ultimate base was probably that 1 of old rom The following are Seythian instances of the use of this root d ( with the meaning of elder sister, precisely as in the Dravidian languages —Tungusian ‘oki’ or akin ;7 Mongolian ¢ achan; Tibetan achche Mordwin a dialect of the Turkish ‘ege ‘aky ; other Ugrian idioms ‘iggen The Lappish akke,” signifies both wife and grandmother I'he Mongol ‘aka,” Tungusian ‘aki,” and the Uigur ‘acha,’ sig- nify an elder brother whilst the signification of old man is conveyed by the Ostiak ¢iki,’ the Finnish ukko,” and the mﬁ] Hungarian ‘agg Even in the Ku, a Dravidian dialect The ultimate base of all these o the akke,” means grandfather words is probably ‘ak, old. On the other hand ‘akka, Osmanli Turkish, means a younger sister; and the same mean- ing appears in several related idioms. It may, therefore, be con- rect sE sidered possible that ‘akka’ meant originally sister; and then it any elder sister, or younger sister, by secondary or restricted usage It is proper here to notice the remarkable circumstance of & i a1ty that the Dravidian languages, like those of the Secythian group :l 478 GLOSSARIAL AFFINITIES. in general, ave destitute of any common term for brother, sister, uncle, aunt, &c., and use instead a set of terms which combine the idea of relationship with that of age; e.g., elder brother, younger brother, elder sister, younger sister, and so on. a The derivation of ¢akka,’” from a root signifying old, would i a appear to be the more probable one. att-an, father. att-al, mother. We find in the Sanscrit lexicons ‘atta, a mother, an elder sister, a mother’'s elder sister ; also ‘atti, in theatrical langunage, an elder sister. I regard this word also as probably of Dravidian origin; and it will be found that in one or another of the related meanings of father or mother, it has a The wide range of usage throughout the Scythian tongues. difference in quantity between the Tamil €att-al’ or €att-ei, and the Sauscrit ‘attd,” does not appear to be of much con- 1 sequence ; and the change of “tt’ in some Dravidian dialects into ‘ch’ or ‘tch, is in perfect accordance with generally prevalent laws of sound. Hence the Malayala ‘achchh-an’ (pronounced ‘atchhan’), and the Canarese ajj-a,’ are iden- tical with the Tamil ‘att-an; and probably the Hindi and Marathi ¢ 4ja, @ grandfather, is a related word, if not identical. The related words, ¢ dtt-ei,” Tamil ; ‘att-e, Canarese; ‘att-a,’ Telugu, have also the meanings of mother-in-Zuw, sister-in-law, paternal aunt; and the coresponding Singhalese ‘att-a,” means a maternal grandmother ; meanings which are not found in Sanserit. For the Seythian analogies of these words, compare Finnish “aiti, mother, together with the following words for father, viz., Turkish ¢ ata; Hungarian ‘atya; Finnish ‘atta; Tchere- - Compare also miss ‘atyad; Mordwin ‘atai; Ostiak ‘ata’ i Lappish ‘aija,” grandfather, and also attje.’ It is remark- able that ‘atta’ is also a Gothic theme ; e.g., ‘attan,” father, al “aithein,” mother. il ‘ann-el ’ and ‘amm-ei’ are ann-ei, mother: honorifically elder sister. probably correlative forms of the same base, ‘m ’ being some- times softened into ‘n.’ vy Compare however Finnish ‘anya,’ mother ; Mordwin ‘anai ; Ostiak ‘ane; and also ‘anna’ and ‘ana’ in two dialects of the Turkish. The Hindi ‘anni,’ a nurse, is probably the same word. &Lei: SCYTHIAN. by 4 { 9 o, app-an, futher. Compare the following words for Jather-in-law, viz., e Ostiak “{p,” “6p;’ Finnish ‘appi; Hungarian ‘ip, ‘ipa,’ ¢y ‘apos.’—See also ¢ Semitic Analogies.’ amm-al, uld amm-el, mother : the word is also used honorifically in addressing amm-an, matrons. The following are correlative words, ‘amm-ay,’ maternal grandmother, aunt by the mother’s side, and amm-an,’ an mother’s brother. Compare Samoiede ‘amma,” mother; Jenesei cal ‘amma’ or ‘am ;" ¢ Estrian ‘emma ;" Finnish ¢ em4.’ Compare bly also Ostiak ‘in-a,” woman, wife; Hungarian ¢eme.’—See also Or Semitic analogies. The Sanscrit ‘amba’ or ‘amma,” mother, 8 & properly a name or title of Durgh, is doubtless derived from the Dravidian word. ' The bloody rites of Durga, or Kali, el were probably borrowed from the demonolatrous aborigines by (oM the Brahmans; and ‘amma,’ mother, the name by which she R was known and worshipped—her only Dravidian name—would ally naturally be borrowed at the same time: all From the same source is derived the Scindian ‘ama’ and ell- the Malay ¢ama,” mother. and It is remarkable that in one or two Dravidian dialécts the ical words which denote father and mother, have mutually changed it-a, places. In Tulu, ‘amm-e,’ is father; appe,” mother; and in Ju i, Tuda the former is “en,’ the latter <aph.’ Compare the Mon- gans golian ‘ama,’ father. d In Tibetan and its sister dialects, ‘pa’ or ‘po’ denotes a man; ‘ma’ or ‘mo, @ woman: and these words are post-fixed ish to nouns as signs of gender; e.g., ‘Bot-pa,” a Tibetan man, !l‘]{eh ‘ Bot-ma,” a T%betan woman. here ar-u, also ar-u, }preaious, dear, scarce. Compare Hungarian ‘aru,’ ° ar,” price; ‘lﬂll\ Finnish and Lappish ‘arwo.’ qiher al, &l } the prohibitive particle, ‘noli; e.g., ¢ kodél’ (from ‘kod-u, give), - giwe mot: Séntal prohibitive ¢4l Compare Lappish ¢ali’ or L) {ue ‘ele; Ostiak €ila,’ and Finnish €Ala.’—See also ‘Semitic ome- Analogies.’ avva, Tel., a grandmother; Tuda ¢avva,’ a mother; Tamil avv-a, « 13l }. matron, an elderly woman. Compare Mordwin ‘ava,” mother. fs of —See also ¢ Indo-European Analogies.’ sl al-ei, @ wave; Can. ‘ale:’ as a verbal theme ‘alei,” means o wander, GLOSSARIAL AFFINITIES. 480 to be unsteady. Compare Finnish ¢allok,” @ wave; Armenian ¢ alik.’ 4r-u, a river; Telugu ‘ér-u.’ Compare Lesghian ‘or; Avar CHoks Yakutan (Siberian Turkish), ¢ oryas; Lappish ¢ wiro;’ Ostiak ‘jeara.’ Compare also Armenian ‘aru ; Coptic ‘jaro; and Hebrew ¢or,” ¢yedr. Am, 1t s, yes. Compare Vogoul ‘am,’ yes. iru-mbu, wron. Compare Motor (a Samoiede dialect), ‘ur,’ eron.— See also ‘Indo-European Analogies.’ id-u, Tel., to swim; Tamil ‘nigj-u.’ Compare Ostiak ¢fidem ; Finnish “uin,” to swim. uyarka, high: infinitive (used adverbially) of ‘uyar, to be high, Compare Samoiede ©arka,’ Rajmahal idiom, ‘arka,’ htgh. hagh. ul, o be in, to be: as a noun, a being, an entity : as a post position un, within ; Ancient Canarese o6l As a verb ‘ul’ is very irre- gular; and the €1’ though radical, has generally been eupho- nised into ‘n. The primitive form and force of the root are apparent in the Tamil appellative verb ¢ulladu’ (‘ul(l)-adu’), it 1s, there is ; the Canarese ‘ullava’ (‘ ul(l)-a-vu’), there are; and such nouns as ‘kadavul’ (‘kada-(v)-ul’), Tamil, God, literally the surpassing or transcendent Being. ‘ulladu’ has in Tamil been euphonised into ¢ undu’ (like ‘kol-du, having taken, into < kondu’); and this euphonised appellative forms Compare with the inflexional base of the Telugu verb to be. “ul, to be, the Ugrian substantive verb ¢ ol,” to be; e.g., Tchere- miss ‘olam, I am; Syrianian ‘voli, I was; Finnish ¢olen, 7 am. Compare also the Turkish ‘0l," o be. The primitive meaning of the Dravidian ‘ul,’ seems to be ‘within, in which sense it is still used as a post position in Tamil. ka erud-u, to write. Compare Hungarian ‘ir, fo write; Manchu ‘ara; Finnish ¢ kir. elu~mbu, bone. Compare Finnish ¢lua; Samoiede ‘luy,” bone. Compare Mordwin ‘wok,” all. okk-a, Malayélam, all. katt-i, @ knife. Compare Tungusian ‘koto,” a knife.—See also ¢ Indo-‘ European Affinities.’ SCYTHIAN, 481 All kad-i, { to bite. kar-i, J Compare Lappish ¢ kask,” ¢ bite.—See also ¢ Indo- European Aflinities.’ ak katt-u, to bind, to te. Compare the following words, each of which nd has the same signification :—Hungarian ‘kot; Ostiak ¢katt- em’ (to fasten, to catch); Syrjanian ¢kuta; Finnish ¢keitt; Lappish ¢ karet." kannir, tears. Compare Finnish ¢kényv.” The Tamil word (¢ kén— nir’) literally signifies eye water; so that it is doubtful whether this resemblance is not accidental. l]\ll kapp-al, @ ship, a vessel, originally a verbal noun from ¢ kapp-u,” Tel., to cover over : derivative Telugu noun ‘kapp-u,” @ covering. The oh; verb is not found in Canarese or Tamil, but the Canarese noun ka, ‘kapp-u,” a subterraneous room, a pit-fall for catching elephants (covered over with branches of trees and grass), and the | Ul Tamil noun ‘kappal,’ @ ship, properly a decked wessel, in con- 16 tradistinction to ‘padugu,” an open wessel, are evidently iden- ho tical in origin with the Telugu verb and noun. are The Malay word for skip is ‘kapal;’ but this has probably ), been borrowed directly from the Tamil, and forms one of a e small class of Malay words which have sprung from a Dra- (od, vidian origin, and which were introduced into the Eastern Archipelago, either by means of the Klings (Kalingas) who 18 iIl g settled there in primitive times, or by means of the Arab traders, whose first settlements in the East were on the Malabar TS with coast, where the Malayalam, the oldest daughter of the Tamil, is spoken. The following Secythian words for skip appear to here- plen, be really analogous to the Tamil, and have certaiuly not been borrowed from it:—Vogoul ‘kap’ or ¢kaba ; Samoiede {0 be ‘kebe; Jenesei ‘kep; Yerkesian ¢ kaf; Ostiak ¢chap.’— See also the analogies adduced under the word ‘kebi,” a cave. i iﬂ kar-u, black,; euphonised from ¢kéar; Gujarathi ¢karo. Compare ' ar&}’ Tuarkish ‘kara ;7 Calmuck ¢ chara ;7 Mongolian ¢ chara,’ ‘ kara; Manchu ¢kara ;’ Japanese ¢ kuroi.’ These Scythian affinities are too distinet to admit of the smallest doubt. There is pos- sibly an ulterior, but doubtful Py , connexion between this Scytho- Dravidian root and the Sanscrit ‘kala, black, or Tamil ‘kal- am,” from which there is a derivative, ‘ kira-g-am,” that throws Iudo light on the relation of ¢ kala’ to < kar-u.’ Possibly also ¢kri,’ oI 482 GLOSSARIAL AFFINITIES. the radical portion of ¢ krishna,” Sanscrit, black, may be related to the same Scythian theme. kara-di, @ bear; probably from ¢kara-du,” rough. Compare Samoiede ‘korgo; Tungusian ¢kuti,” ¢kuuti.’— See also ¢Indo-Euro- pean Affinities.’ Compare Ostiak ¢ kuru karug-u, an eagle. k,’ an eagle.— See also ¢ Indo-European Affinities.’ karutt-u, the throat; also ¢ kur-al,’ the wind-pipe. Compare Vogoul ‘kuryd, the throat; Finnish ¢kurko,” ‘kero,” ‘kerri; Kurd ¢g’eru; Lappish ‘karas,” ¢kirs; Slavonian ‘gorlo.’ kal, @ stone. Compare Lappish ‘kalle,” also *kedke’ or ‘kerke; Lesghian ‘gul; Kamtschadale ¢kual,” ¢kualla.’ Probably these words have an ulterior connexion with the Finnish ‘kiwi; Hungarian ‘ké; Ostiak ‘key, ¢kaiick.’ Compare 1 also (through the interchange of ‘1’ and ‘r’) the Tamil ‘kar,’ gravel, a pebble; with the Greek ¢ xep-ds,” gravel, and xep-pas, The Dra- a stone,; and the Armenian ‘K’ar,” ‘kuar, a stone. vidian root cannot be traced farther than ¢kal’ a stone; but the corresponding Lappish ¢kalle,” appears to be derived from, or connected with, ‘kalw-at,” to become hard. Compare also ‘karra,” Lappish, kard, rough. k kall-am, Malayalam, e le; Tamil ¢kala-vu, a theft. Compare Lappish ‘keles,” a lie. karr-u (pronounced ‘kattr-u’), wind. Compare Kangazian (a Turkish dialect) ¢ kat,’ wind; Sojoten (a Samoiede dialect) ¢ kat; other Samoiede dialects ‘chat, ‘kada’ (also a storm, °charru’); Georg. ‘kari; Jurazen ‘chada.’ ke kay, to heat, or be hot, to burn, to boil. Compare Finnish ¢ keite,’ ¢ keitta,” to boil, to cook; Hungarian ‘keszil.’ Compare espe- cially the Indo-European Affinities of this word. kal,” oot dLuda . kol.;” .Tulu. © kir Compare Mongol ¢ kol; Ostiak ‘kur; Tungusian ‘chalgan,” ‘balgan; Permian ‘kok; Ossete ‘kach,” ‘koch; Vogoul ‘lal;’ Korean ¢ pal; Canton- Chinese ‘koh.’ kir-a, old, aged. Compare Oriental Turkis‘h “chari; other Turkish idioms, ‘kar, ‘kart; Wotiak ‘keres; Lesghian ‘heran.’— See also the Indo-European analogies of this word. SCYTHTIAN. 483 d kil, Can., below,; Tamil ¢kir. Compare Wolgian ‘kilgi, ‘kelga, deep. From the Tamil ¢kir’ is derived ¢kir-angu,” a bulbous le FEASERRT root ; with which we may perhaps compare the Slavonian [0 ‘koren; Jensei ‘koryl, a root. kudir-ei, @ horse; Canarese ¢ gudur-e. The Sanscrit ¢ ghoda,” a horse, has doubtless an ulterior connexion with the Dravidian word ; la() but I cannot suppose the Dravidian word to have been directly borrowed from the Sanscrit one, for the Tamil occasionally oul borrows and uses ‘ghéda’ (in Tamil ‘ ghéram,” also < gédagam ;’ Both ind Telugu ¢gurram-u’), in addition to its own ‘kudir-ei.’ words seem to be derived from a common origin. The Scythian analogies are Jenesei ¢ kut’ and Lesghian ‘kota.’ Compare ¢ also Malay ¢kuda.’ bly ish kud-i, a habitation; ‘kud-il, ¢kudis-ei,’ a hut, a cottage. In Telugu are and Canarese ‘ gud-i,” means a temple. A similar word, ¢ kuta’ i, or ‘kuti, is also contained in Sanserit ; but it appears to be one of those words which the Sanscrit has borrowed from the 5, ra: Dréavidian tongues. It has a place in each of the dialects of but the Finnish family ; e.g., Mordwin ¢ kudo,” @ kouse; Tscheremiss ‘kuda; Finnish ‘kota; Ostiak ¢chot; Lappish ‘kata.’ I 0ml, = als0 suspect the Saxon ‘cot’ had a Finnish origin. kul-ir, cold, to become cold: ultimate base ‘kul; related words ‘kiid-al” and ‘kid-ir, cold; also Telugu and Canarese chali,’ pare cold. ¢&il-ir, Tamil, to tremble, seems to be a collateral root. With ¢kul-ir,” compare Lappish ¢ kal-ot,” to freeze; Finnish kish “eyl-ma; and with ‘chali’ (Telugu and Canarese), compare fher Permian ¢ cheli,” cold.—See also ‘Indo-European Affinities.’ ll)’ kei, hand,; Canarese ‘kye, keiyyi; Telugu chéyi’ A comparison of these words seems to show that ‘kéy,’ was the primitive ite, form of this root. Possibly there is a remote ulterior con- 8Pt nexion between the Sanscrit ‘kara,” the hand, and the Dravi- dian word : possibly, also, as ‘kara’ is supposed to be derived 1\01 ) from ‘kri, to do (or ¢hri) to take), so ‘kéy,’ the hand, may b be derived from, or connected with, the Dravidian ¢ki, ‘gi, xOl\ ‘gé’ ‘chey’ or ‘gey, to do. There appears also to be a ytoD- special resemblance between the Tamil ‘kei’ and the Greek ‘xelp; yet when the Greek gemitive ¢ xep-os’ is compared with the Old Latin “hir’ and the Sanserit ¢ kara,” it is obvious r]\l»h tu.'" that it is with the latter that ‘e’ is to be connected, rather than with the Tamil ¢kei. The analogies of the Dravidian 2 12 484 GLOSSARIAL AFFINITIES. word seem to be exclusively Scythian. Compare Hungarian “kéz’ (pronounced ¢keis’); Finnish ¢kchési’ (root Ska s €. genitive ‘ké-an’); Estnian ‘kési; Ostiak ‘ket; Lappish ‘kat; Permian ‘ki; Lasian ¢ke; Mingrelian ‘che 7 Quasi- Qumuq (a Turkish dialect), ‘kiiya.’ Compare also Persian ¢‘ kef. The Hungarian has both ¢kar’ and ‘keéz 7 but the former is used to signify arm, the latter hand, a distinction which seems to prove that those roots, though perhaps ulti- mately related, have long been independent of one another. keb-i, @ cave; Canarese ‘gav-i; also ‘kapp-u,’ Canarese, a subter- raneous room, a pit-fall. Compare Mongol and Manchu ¢ kobi,’ a cavity, a cave ; Ostiak ‘kaba, “kebi, ¢ kavi, a chamber. Compare also ‘kapp-al,” Tamil, a ship, from ¢ kapp-u,’ Telugu, to cover over.—See ¢ Indo-European Affinities.’ kev-i, Can., the ear ; Telugu (euphonically softened) ¢chevi; Tuda ‘kavi; Brahui ‘khaff: related words ‘kad-u, the ear, and “kél,” to hear. Compare the following Scythian words signify- ing the ear :—Samoiede dialects ‘ko, €ku, ¢kus ;’ Korean ‘kui; Ossete ‘kus; Kurd ¢goh; Turkish dialects ¢kulak.’ kél-u, Ancient Telugu, the hand. Compare Kuralian ‘kell’ and Georgian ‘cheli, the hand. kél, to hear; ‘kél-vi) hearing. Compare Finnish ‘kuul-en,’ to hear; 9 Syrjanian ‘kyla; Tcheremiss *kol-am; Hungarian ¢ halla ; Lappish ‘kull-et’ (‘kullem, hearing); Ostiak ©kidl-em. 8 Notice the change of the final ‘1’ of the other Finnish dialects into “dl’ in Ostiak, a single cerebral consonant, precisely similar in sound to the final ‘1’ of the corresponding Tamil ‘kél.’—See also the Indo-European affinities of this word. kol, to kill. Compare Finnish ¢kuol’ to die; Tcher. ‘kol-em; Syrj. ‘kula ; Hung. ‘hal. See also ‘Indo-European Analogies.’ b kon, a king, a ruler; in honorific usage a shepherd, or man of the ¢h shepherd caste. Another form of the same word is ‘ko,” a king, a god. It is hard to determine whether k6’ or ‘kon’ is to be regarded as the primitive form of this word. Compare the tal Turkish and Mongolian ‘khén,’ also ‘khagan,’ a ruler; Ostiak ¢ khon ; a Scythian of the Behistun tablets, ¢ ko king. kor-i, the domestic fowl; Goénd korh ' Seoni Gond kor, a hen t 485 SCYTHIAN. This word is the common term which 1s (¢ gogori, a cock). used in the Dravidian languages for both the cock and the hen. h If it is required to express the gender °saval’ a cock, or ‘ pettel, a ken, is prefixed adjectivally to the common term Al “Ko.’ o The Sanscrit ¢ kukkuta, a cock, from ‘kuk-a,’ to scratch on does not seem to have any affinity with the Drividian €kor-i.’ fl The Scythian analogies, on the other hand, are close and direct. Compare Vogoul ‘kore; Ostiak ‘korek,’ ‘kurek; Permian 7] ‘korech,’ ‘kuryg,’ ‘kuraga.’ Probably the North-Asian tongues borrowed this word directly from the Dravidian ; for the domestic i, fowl had its origin in India, where the wild variety still exists ; and when it was introduced into Upper Asia, the name by Uy which it was known in India would naturally be introduced along with the fowl itself. That name being, not Sanscrit, 1da but Dravidian, it would appear that the domestic fowl must nd have been introduced from India into Central and Northern Asia, prior to the irruption of the Aryan race, and the con- fy pall sequent cessation of intercourse between the Dravidians and k? the other members of the Scythian family. 'The Dravidian word has found its way into two languages and of the western branch of the Indo-European family, viz., the Persian and the Russian. Compare Persian ‘khor-os,” a cock ; ‘kour-ek, @ poulet ; and the Russ ‘kilir,) a cock; ¢kir-itsa,’ o'y a fowl ; diminutive ‘kar-otchka,” a chicken. la , en garal, rain driven by the wind: in the usage of the Southern Tamilians, lef}ta the rain brought by the south-west monsoon. Compare Samoiede sely ¢ sarre ; Permian ‘ser ;) Wotiak ‘sor, rain. gl 4, or &ig-u, fo die; Telugu ‘chachu’ (base ‘cha’). Compare Samoiede ¢ chawe’ and ¢chabbi,” dead. em 7’ 7168 ched-u, Telugu bad. Compare Ostiak jat, bad. f fhe chér-u, mud. Compare ‘chedo,’ ‘zerta,” €choti,” and ¢ chat’,—Les- klltﬂ: ghian words for clay. i t0 tal-a, Telugu, the head ; Tamil ‘tal-ei.” Compare Mongol ‘tologoi ;’ 0 the Buriat ¢ tulgai; Samutan (a Tungusian i) ¢ Calmuck ¢tolgo ; dialect) € dcll, ¢dollokin ;' other Tungusian dialects ‘diill, sO;y e ¢del, € deli’ ti, fire. The more commonly used Tamil word for fire is ‘neruppu 7 g b 486 GLOSSARIAL AFFINITIES, ‘*i Telugu *nippu:’ but ‘ti is the more classical word ; and it SRR R is much used by all classes of people in the southern districts It is used also in Tulu. of the Tamil country. The Scythian affinities of this word ¢ti,) are peculiarly distinet ; e.g., Samoiede ‘tu, ‘tui,) ‘ti, ¢ty ;' Mantchu ¢ tua; Hungarian ‘tiiz ; Ostiak ‘tiit ;7 Tungus ‘togo; Lesghian “tze,” ¢ zi ‘zle Finnish ¢tuli; Lappish ¢tall’ Compare also Gaelic ‘teine; and Welsh ‘taan ; Persian ‘tigh; Sanscrit ‘téjas.’ tus-1, dust, powder. Compare Turkish ‘tus, ‘toosan; Mongol ‘toghoz,” dust. tol, skin. Compare Vogoul ‘toul, ‘towl, skin. nakk-u, fo lick: derivative noun ‘nakku’ or ‘na,’ the tongue. Com- pare Ostiak ‘nal’ to lick, and ‘nal,’ the tomgue: Samoiede e ‘nawa,’ the tongue. nag-el, to laugh, laughter. Compare Ostiak ‘nag-am,” to laugh ; ‘nach,’ laughter. nay, a dog ; Tuda ‘noi. Compare Mongol ‘nogai,” @ dog; Calmuck ‘nokoi,” ‘nochoi.’ nu, Telugu copulative particle and. Compare Ostiak ‘no,” and. nett-i, the jforchead ; Telugu ‘nud-ur. Compare Lesghian ‘nata,’ ‘nodo,” ‘nete-bek,’ the forehead. i ney, ghee, clarified butter, ovl. Compare Avar (Turkish family of idioms), ‘na, ‘nah,’ ‘nach,’ butter. pal nod-u, Canarese to see, to look. Compare Mongol ‘niidu,’ the eye. fijayir-u, the sun. Compare Hungarian ‘nar, tie sun ; ‘nyar’ (‘fjar’), summer ; ‘nap, a day : also Mongol ‘nar-an,’ the sun ; Ostiak pal ‘nai ;; Affghan ‘nmar.’ pas-u, green ; ‘pul; grass. Hungarian ‘pusit,’ grass; Vogoul ‘piza ;’ Ostiak ‘pady.’ i pay-an, peiy-an, 1 -a boy, a servant, Malayalam ¢ peidal ;’ Canarese ©heida,” a PIR- pay-al, peiy-al, boy or guirl. The words terminating in ‘an’ are masculines ; those in ‘al’ and ‘dal’ are verbal nouns. ‘dal’ 1s as commeon a Pl " S —— — - 54 SCYTHIAN. 48 ( i formative of verbal nouns even in Tamil as “al,” and the two ts forms are mutually convertible. Both ‘payal’ and ‘peidal’ are necessarily abstracts, and are therefore capable of denoting either sex. ly ¢ payan,” Tamil, is restricted to signify a boy. b The theme or base of these words is evidently ¢pay’ or il ‘peiy,’ which are equivalent sounds, and of which the ‘y’ seems ¢ to have been converted from, or is convertible into ‘8, if we may I judge from ‘pagan-gal,” which is often used as the colloquial plural, instead of ‘payan-gal.’ Compare the following Ugrian words for son : Vogoul ‘py,’ rol ‘pu; Mordwin and Syrj. ‘pi; Wotiak ‘pyes;’ Finnish ‘poika ;’ Hungarian ‘fiu; Estrian ‘poeg; Ostiak ‘pach,’ ‘poch,” ‘pagul, ‘pagam,’ ‘pyram ;’ Lappish ¢ patja.’ The Swedish ‘poike’ is evidently derived from the Finnish ‘poika ; and the Greek Il ‘mai-¢’ and the Latin ‘pu-er’ are evidently related roots. See e ‘Indo-European Affinities.’ The Dravidian languages appear to contain the ultimate h, theme of all these words, viz., “pei,” Tam., to be green or fresh, a word which has been softened from ‘pas-u’ (‘pay-u,’ convert- ible into ‘pei’), green, by a common Dravidian law. The deri- 1k vation of ‘pay-an,’” @ boy, from this root would account for its conversion in vulgar usage (in the plural) into ‘pas-an,’ and it would also explain why ¢al’ and ‘dal, which are formatives of verbal nouns, are often used as terminations. fa, par-a, old (by reason of wse); Canarese ‘pala-ya; ‘paras-u,” Tamil, old, what s old. Compare Mordwin ‘peres; Syrj. ¢ porys ; ns and Ostiak ¢ pirich, old. pal, tooth. Compare Lappish ‘pane, ‘padne; Wolgian ¢padne, “pai, ‘pin ;7 Ostiak ‘pank, ‘penk, ‘pek ; Techer, ‘py. al A‘) ) pal, a part, a division. Compare the following Ugrian words signify- tiak ing a half:—Samoiede ‘pealed; Tcher. ‘péle; Lappish ‘bedle ; Ostiak ‘pélek ; Hungarian ‘fél.” See also ¢ Semitic Affinities.’ 7 pid-t, to catch. Compare Finnish ¢ pidan,” to caich. pir-agu (base ‘pir’ or ‘pin’), behind, after. Compare Ostiak ¢pir,’ [a’ {0 ¢ pira,” behind, hindermost ; Finnish ¢pera.’ pill-ei, a child. Compare Yarkand Tartar ¢billa, @ child; Hindi 165 ‘pilla,” a cub, a pup. See also ¢ Indo-European Affinities.’ on & 488 GLOSSARIAL AFFINITIES. pug-ei, smoke. Compare the following words signifying vapour in the Turkish dialects, ‘bug,” ¢buch,’ ‘bugu’ Compare also the English “fog.’ pen, a female; Canarese ‘henn-u.’ Compare Lappish ‘hene,” a female. pokkul-i, Telugu tke navel. Compare Ostiak ¢ puklam,’ the navel. bayir, Canarese the belly ; Tamil ¢ vayir-u; Gond pir. Compare Kangazian (a Turkish dialect), “bar,’ tke belly ; Armenian ‘por;’ Albanian ‘bark ;? Ostiak ‘perga ;7 Mordwin ¢ pak.’ bal, Canarese to exist; Tamil var,” to flourish, to live prosperously. Compare Oriental Turkish ¢bél,” to exist. man-a, Canarese @ Aouse: classical Tamil ‘man-ei.” Compare Samoiede ‘men,” @ house, Vogoul ‘unneh’ mar-am, @ tree, wood,; Canarese ‘mar-a ; Telugu ‘méan-u’ (for ‘mran-u’). Compare Lappish ‘muor,” ¢ muorra,’ @ tree, wood ; Quasi-Qumuk Turkish ‘murm, ¢murch ;7 Mongol ‘modo ; Tomsk. ‘madji; Finnish ‘mezza ; Lettish ‘mes.’ maR-i, @ foal, the young of the horse, the ass, &e. Compare Mongol “mori, @ horse; Manchu ‘morin; Breton ‘mor; German ¢ mahre.’ This Dravidian mal-a, Canarese a &ill, a mouniain ; Tamil ‘mal-el.’ root has found its way into the Sanscrit lexicons as the base of ¢ Malaya, the Sanserit name of the Southern and Western Ghauts—* Malayalam,’ or as the Arabian geographers called it, Male. It has probably given their name also to the Mal-dives, the ¢dives’ (Sanscrit ‘dwipa’), or wslands, pertaining to Male or Malayalam.’ Compare Albanian ‘malli) @ %l ; Vogoul ‘molima ; Permian ‘mylk ;; Wolgian (by a change of ‘1’ into ‘r’), ‘mar ;’ Samoiede ‘mari; Avar ‘mehr ; Finnish ‘magi.’ muRumuRuy, to mutter, to grumble. Compare Finnish ‘muraj, and Hungarian ‘morog, to murmur. See also “Indo-European Affinities.’ menj, Gond, an egg : plural ‘mensk.’ Compare Hungarian ¢ mony,’ an egg : Finnish ‘muna; Samoiede ‘muna.” Canarese ‘motte’ 18 more remote. SCYTHIAN. 489 van, heaven. Compare Mordwin ¢ ménel,’ heaven ; Tungus. ‘iiyan ; 4 dialect of the Kookies and Bungoos in the Chittagong hills, ‘van.’ vay, the month. Compare Samoiede ¢ aiw-a,’ month ; Lappish ‘saiwe ;’ Hungarian ‘ayak. vir-i, to watch. Compare Finnish ¢wir-ot, fo watch; Hungarian ¢ vir-ad.’ ¢ velich-am, light. Compare Hungarian ¢ vilaga,” light. Y y de o )y ol Al g‘ an of | on i (7 ale % ,” )y nd oal Hy;: .tte ;: | (i lc b al | t i €l 00 1¢ ! g W i 0! ip the Tll dann Lt e DI or ey By Th % SEFENNIIEX I Are the Pariars of Southern India Dravidians ? It is commonly supposed by Anglo-Indians, that certain tribes and castes inhabiting Southern India, especially the Pariars and similar low-caste tribes, belong to a different race from the mass of the inha- bitants. The higher castes are styled ¢ Hindus, or else ¢Tamilians, ¢ Malayalis,” &c., according to their langua o o e and nation; but thos names are withheld from some of the ruder and more primitive tribes, and from the Pariars and other agricultural slaves, As this supposi- tion, and the use of words to which it has given rise, are frequently met with both in conversation and in books, it seems desirable ft« enquire whether, and to what extent, this opinion may be regarded as correct. The term ‘Hindu’ as used by some, is one which pertains to religious nomenclature. When they speak of certain classes as ‘Hindus,” they mean that they are followers of the Brahmanical religion, or the religion of the Purinas; and according to this use of words (which is open to serious objection, inasmuch as it is the use, not of a theological, but of a geographical term, to denote one out of several religions which prevail within the region to which the term applies), the tribes and classes whose religion differs from that of the Brahmans are not ¢ Hindus. In this sense it is true, that the Tudas and the Goénds are not Hindus, and that the majority of the predatory, wandering tribes, and of the lower castes are not Hindus, or at least are not ‘orthodox Hindus; though, geographically, it is certain, that they have as much right to the name of Hindus as the Brahmans themselves. Some, again, use the term ‘ Hindu’ as synonymous with Aryan.’ They call the Brahmans and the higher castes of Northern India 492 APPENDIX This seems ¢ Hindus,” but withhold the name from the aboriginal races a still more improper use of words, inasmuch as it denationalizes not only the low-caste inhabitants of the northern provinces, but also the whole of the Drividian inhabitants of the Dekhan and the Peninsula ; notwithstanding the proofs that exist that they crossed the Sind Hind, or Ind-us, and occupied the ¢Sapta Sindhu,” or ‘country of the seven rivers,—the Védic name of India—before the arrival of the Aryans, and that they have, therefore, a better claim to be called ¢ Hind-us’ than the Aryans themselves To deprive the Dravidians and other aboriginal races of the name of ¢ Hindu,” is as unjust as it would be to deprive all persons of Anglo-Saxon descent of the name of ¢ Englishman,” and to restrict that name to the descendants of Nor- man families There are some, again, who with the error now mentioned, conjoin an additional one. They suppose the higher castes of the Tamil, Telugu, and other Drévidian peoples, to be identical in origin with the Aryan races of Northern India, and the lower castes alone to have a Non-Aryan origin. Hence they call the high-caste Dravidians ¢ Hindus,” and withhold that name from the Pariars, &c., not on geo- graphlcal but on ethnological grounds. I apprehend, however, that the Non- AryaJn origin both of the higher and of the lower castes of Dravidians, has been proved when the Non- Aryan structure of the Dravidian languages has been established ; and, therefore, this use of words may be passed by without further remark, as arising simply from misapprehension The Pariars (called in Telugu Malars) are not the only caste or class of people in the Dréavidian parts of India, who are commonly regarded as outcasts, nor are they the lowest or most degraded of those classes ; but partly because they are the most numerous servile tribe (their numbers amounting on an average to at least a tenth of the scadsiel entire population), and partly because they are more frequently brought into contact with Europeans than any similar class, in consequence of the majority of the domestic servants of Kuropeans throughout the Madras Presidency being Pariars, they have come to be regarded by some persons as the low-caste race of Southern India Hence, besides the above-mentioned errors in the application of the name ‘Hindu, there are various popular errors afloat respecting the origin of the Pariars and their position in the caste scale, which require to be noticed before entering on the question now to be discussed, ‘are the Pariars Dravidians ? Europeans were generally led to suppose, on their arrival in India several generations ago, that the Pariars were either the illegitimate ARE Tm: PARIARS OF SOUTHERN INDIA DRAVIDIANS? 493 offspring of adulterous intercourse, or were persons who had been f excluded from caste for their crimes. This notion was invented and propagated by the Brahmans and the higher castes, and originated, in part, in their wish to justify their exclusive, unsocial behaviour : towards the Pariars, on principles which they supposed that Euro- peans would approve. In part, also, it originated in an error arising from the uncritical habit of the Hindu mind ; viz., the error of trans- ferring to Southern India and to the Dravidian tribes, the fictions § which were devised in Northern India to account for the origin of the i Those northern new castes or, so called, ‘mixed classes,” of the North. Y castes or classes came into being through the operation of two causes; first, from the sub-division of the original castes of Vaisyas and servile or Sudra Ary:ms, in accordance with the progressive sub-division of labour ; and secondly, from the introduction of one tribe after another within the pale of Aryan civilization, as the religion and civil polity of the Sanscrit-speaking race spread throughout the country, and the aborigines were transformed from Mléchchas into Stdras. In Manu b and the Shastras, no mention is made of either of these causes; but the new or mixed castes are attributed exclusively to fictitious mixtures of i the older castes. The more respectable of the new castes are attributed f to the legal intermarriage of persons belonging to different castes of recognized respectability. Another and inferior set of castes are f attributed to the adulterous intercourse of persons of equal respecta- bility, but of different caste, or of high-caste men with low-caste women ; whilst the lowest castes of all are represented to have sprung from the adulterous intercourse of high-caste women with low-caste I men, and are said also to constitute the receptacle of persons who had [ been socially excommunicated for offences against their caste. Whatever amount of truth may be contained in this representation [7 of the origin of the castes of Northern India (and I think it most t z probably a fiction throughout), it may confidently be affirmed that the f Dravidian castes had no such origin. The only ¢ mixed caste’ known in Southern India, is that which consists of the children of the dancing girls attached to the temples. Of this class the female children are ] brought up in the profession of their mothers, the males as temple florists and musicians. In all ordinary cases, when children are born in adultery, if there is no great disparity in rank or caste between the :(; parents, the rule is that the caste of the child of adulterous intercourse is that of the less honourable of the two castes to which its parents belong. Where considerable disparity exists, and where the derelic- tion of rank is on the woman’s side—as for example, where a high- ‘caste woman, or even a woman belonging to the middling castes, has 494 APPENDIX. formed an intimacy with a Pariar man (and in the course of a residence amongst the Hindu people for seventeen years, I have heard of several such cases), neither the caste of the father nor any other caste has any chance of being recruited or polluted by the addition of the woman’s illegitimate offspring. The child never sees the light; the l mother either procures an abortion or commits suicide. To suppose, therefore, as Europeans have sometimes been led to suppose, that the entire caste of Pariars (including its subdivisions, and the ‘left hand’ castes corresponding to it) bas come into existence { in the surreptitious manner described above, or that it is composed of il persons who have been excluded from caste for their crimes, is a base- less dream, which is too preposterous for serious refutation. Though a h it is probable that it was from the statements of natives that the { Anglo-Indian community originally derived this notion, yet I never met with any natives, learned or unlearned, by whom the notion t appeared to be entertained; and the Pariars themselves, who regard their lowly caste with feelings of pride and affection, which are very b different from what might be expected of them, would resent this 80 representation of their origin, if they had ever heard of it, with indignation. . P Anglo-Indians who are not acquainted with the vernacular lan- guages, often designate Pariars as outcasts’ as persons who are tl fo “ without caste,” or as persons who have ‘no caste to lose.” It is true that the Pariar servants of Europeans will sometimes vaunt that they th N belong to ‘master’s caste; and many masters know to their cost that f their Pariar servants practise no scrupulous, superstitious distinctions respecting meats and drinks. Notwithstanding this, to suppose that i The - th the Pariars have literally ‘no caste,’ is undoubtedly a mistake. Pariars constitute a well defined, distinct, ancient caste, independent ar to of every other; and the Pariar caste has subdivisions of its own, its own peculiar usages, its own traditions, and its own jealousy of the The encroachments of the castes which are above it and below it. il T Pariars, though, perhaps, the most numerous caste in the country, belong to the lowest division of castes, and are not fabled to have Te( sprung from even the least noble part of Brahma ; nevertheless, they are not the lowest of the castes which are comprised in this lowest b division. I am acquainted with ten castes in various parts of the D Tamil country, which are certainly lower than the Pariars in the social scale ; and in this enumeration I do not include the Pallars, a of caste between whom and the Pariars there is an unsettled dispute By respecting precedence. The treatment which the Pariars receive from 8n the castes above them, is doubtless unjust and indefensible; but it is Dy ARE THE PARTARS OF SOUTHERN INDIA DRAVIDIANS ? 495 06 not generally known by those Europeans who sympathize in the l ;Z wrongs of the Pariars, that, whenever they have an opportunity, the 4§ Pariars deal out the very same treatment to the members of castes he which are inferior to their own, e.g., the caste of shoemakers, and the he low-caste washermen; that they are, equally with the higher castes, filled with that compound of pride of birth, exclusiveness, and Jjealousy to which is called caste feeling ) P & and that there is no contest for pre- I, cedence amongst the higher castes of longer standing, or of a more eager character, than that which is carried on between the Pariars and 106 of the Pallars. In the insane dispute about pre-eminence, which is se always being carried on in Southern India, between the ‘right hand’ oh and the ‘left hand’ castes, the Pariars range themselves on the right the hand, the Pallars on the left; and it is chiefly by these two castes that the fighting part of the controversy is transacted. yer ion Now that Europeans are better acquainted with Indian affairs, the ad ;: theory of the illegitimate origin of the Pariars is more rarely found to 24| be entertained ; and, as the study of the native languages extends, the his supposition that they are ‘ outcasts,” or that they have ‘no caste,” will soon disappear likewise. ith fz The question which is really before us having been cleared of popular errors and extraneous matter, we now come to the considera- an are tion of that question itself. ¢ Are the Pariars Dravidians 7 Are the rue forest tribes, the lower castes, and the so-called ‘out casts,” that speak the Dravidian languages, especially the Tamil Pariars and Telugu hey :; hat Malars (who may be taken as the representatives of the class), of the same origin and of the same race as the Dravidians of the highei‘ 008 hat castes} Whilst both classes have a right to be called ‘Hindus,” are 4 the higher castes alone Dravidians, Tamilians, Malayalis, &ec.? and The are the Pariars and people of similar castes to be regarded as belonging ent to a different race ? 1t On the whole I think it more probable that the Pariars are Dra- the The vidians; nevertheless, the supposition that they belong to a different try; race, that they are descended from the true aborigines of the country ave ! —a, race older than the Dravidians themselves—and that they were hey reduced by the first Dravidians to servitude, is not destitute of plausi- bility. mest the It may be conceived that as the Aryans were preceded by the Dravidians, so the Dravidians were preceded by an older, ruder race, the N 3‘ of whom the Déms and other ¢ Chandilas,” of Northern India, and the Py Pariars and other low tribes of the Peninsula, are the surviving repre- ;ute sentatives. If this primitive race existed prior to the arrival of the rom Dravidians, it would naturally happen that some of them would take f 38 496 APPENDIX refuge from the intruders in mountain fastnesses and pestilential jungles —like the Rajis or Déms of the Himalayas, the Weddas of Ceylon, and the Mala-(y)-arasers of the Southern Ghauts; whilst others, probably the majority of the race, would be reduced to perpetual servitude, like the Pariars, Puliars, and Pallars. The history of the subjection of the Pre- Aryan Stidras of Northern India, would thus form the counterpart and supplement of the history of the subjection of a much older race Though, however, all this may be conceived to be poss1ble, and though there may not be any @ priori improbability in it, it is more to the purpose to state such circumstances and considerations as appear to be adducible in its support (1.) The Pariars, the Pallars, the Puliars, and several other low caste tribes, are slaves to the higher castes, and appear always to have been in an enslaved condition ; and it is more natural to suppose that they were reduced to a servile condition by conquest, than to suppose that entire tribes were enslaved by the operation of ordinary social causes If then, the castes referred to were a subjugated people, they must have settled in the country at any earlier period than their conquerors, and probably belonged to a different race (2.) The low-caste inhabitants of Southern India (whether they be slaves like the Pariars; vagrants like the Korawas, or basket makers; or freemen and proprietors of land, like the Shanars, or palmyra cultivators), are distinguished from the entire circle of the higher castes by clear, unmistakable marks of social helotry The or title of ¢Stdra, which has been assumed by the higher caste which was conferred upon them by the Brahmans, is withheld from 1 the low-caste tribes; they are not allowed to enter within the precincts of the temples of the Dii majorum gentium; and wherever old Hindu usages survive unchecked, as in the native protected states of Travan- t core and Cochin, the women belonging to those castes are prohibited 1] from wearing their ‘cloth ver their shoulders, and obliged to leav the entire bust uncovered, in token of servitude It may be argued, that broadly marked class distinctions like the ! above-mentioned, which separate the people of at least twenty different 18 castes or tribes from the rest of the population, are incompatible with the supposition of an original identity of race g (8.) There are various traditions current amongst the Pariars to the effect, that the position which their caste occupied in native society it at some former period was very different from what it is now, and t much more honourable Wilks observes that there is a tradition that the Canarese Pariars were once an independent people, with kings of § = ARE THE PARIARS OF SOUTHERN INDIA DRAVIDIANS? 497 s their own. The Tamil Pariars sometimes boast that at an ancient nd period, theirs was the most distinguished caste in the country. They ly 2 say that they were reduced to their present position, as a punishment \1\9 for the haughty behaviour of their ancestors to some ancient king; on which occasion the Vellalars, or caste of cultivators, who are now gl called ¢ Tamirar,” or Tamilians, par excellence, were raised to the place 0Ly s e g which had previously been occupied by themselves. There is a similar s but more distinct tradition that the Korawas, or gipsy basket makers, iy were once ‘ kings’ of the hill country in the extreme south. uch T " (4.) In various parts of the country Pariars enjoy peculiar pri- 1t8 vileges, especially at religious festivals. Thus, at the annual festival i s of ¢ Egattal, the only mother—a form of Cali, and the tutelary goddess low of the ¢Blacktown’ of Madras,—when a ¢tali, or bridal necklace ave (answering to our wedding ring), was tied round the neck of the idol hat B in the name of the entire community, a Pariar was chosen to represent )05 the people as the goddess’s bridegroom. Similar privileges are claimed cial by Pariars in other parts of the country, especially at the worship of divinities of the inferior class, such as the village ‘ammas,’ or mothers, hey heir and the guardians of boundaries; and these peculiar rights which are conceded to them by the higher castes, may be supposed to amcunt to hey an acknowledgment of their Pre-Dravidian existence, or at least to sket an acknowledgment of their ancient importance; like the privileges claimed at the coronation of Rajput princes by the Bhills, a northern , O the race of aborigines. The It has always been the policy of Hindu rulers, to confer a few or empty privileges upon injured races as a cheap compensation for from injuries; and it has generally been found, where an inquiry has been 1nc ts made, that such privileges possess a historical signification. indu (5.) The strongest arcument which can be adduced in support of val the Pre-Dravidian origin of the Pariars and other low-castes, consists \ited in the circumstance that the national name of ¢ Tamilians, ¢ Malayalis,’ ¢ Kannadis,” &ec., is withheld from them by the wusus loquend: of the Jeave Dravidian languages, and conferred exclusively upon the higher castes. s the When a person is called a ‘ Tamiran,” or ‘Tamilian,” it is meant that he is neither a Brahman, nor a member of any of the inferior castes, but erellt a Dravidian Stdra. The name is understood to denote, not the lan- with guage which is spoken by the person referred to, but the nation to I8 {0 which he belongs; and as the lower castes are never denoted by this national name, it would seem to be implied, that they do not belong to sciety and the nation, though they speak its language, but belong, like the Tamil- ) that speaking Brahmans and Mahommedans, to a different race. T may here mention an argument which is occasionally urged in 08 Uf 2 x i — ——— 498 APPENDIX. support of the same view of the case, which is founded upon a mistake. It has been said that the name Pariar, or Pariah, is synonymous with that of the ¢ Paharias,’ a race of mountaineers near Calcutta; and hence it is argued that the Pariahs may be considered, like the Paharias, as a race of Un—Aryan, Un-Dravidian aborigines. It is a mistake, however, to suppose that there is any connection whatever between those two names. The word ¢ Pariar,’ properly ¢ Pareiyar,” is the Tamil plural of pareiyan’ (¢ parei-(y)-an’), which denotes not mountaineer, but a drummer, a word which is regularly derived from The name ‘parei,” a drum, especially the great dirwm used at funerals. ¢Pariar’ is, in fact, the name of a hereditary occupation, the Pariars being the class of people who are generally employed at festivals, and It is true that their numbers are especially at funerals, as drummers. now so great, that many of them are never so employed, and that the only employment of the great majority is that of agricultural slaves; but whenever and wherever the din of the ‘parei’ happens to be heard, we may be assured that a ‘Pareiyan’ is the person who is engaged in beating it. As the whole caste, though the most numerous in the circle of the low-castes, is denominated by this name, it appears é probable that originally drumimning was their only employment. If so, f they must have been much less numerous at a former period than they are now. The origin of the epithet ¢ Malar, which is applied to the [3) Telugu Pariars, is unknown, 8 Though the circumstances and arguments that have now been fi alleged in favour of the Un-Dravidian origin of the lower castes, he possess a considerable degree of strength, I proceed to show that they th are not perfectly conclusive, and that they are to some extent counter- al balanced by considerations which are adducible on the other side. he (1.) The argument which is drawn from the servile condition of the Pariars fails to establish the conclusion; because it is certain that Py there are many slaves in various parts of the world who do not differ The Russian 1 from their masters in race, though they do in status. po serfs are Slavonians, and the Magyar serfs Magyars, equally with ‘their masters. Illustrations of the inconclusiveness of the argument oy may be drawn also from Dravidian life. the The Shénars, the highest section of the lowest division of castes, tioy are generally proprietors of the land which they cultivate, and many i | fl‘ek of them are almost on a level with the Dravidian Stdras. The more beﬂ wealthy of the Shanérs have slaves in their employment, some of whom, called ¢ Kalla Shéanars,” belong to a subdivision of the Shéanpér ing) caste. These servile Shanars appear to have been slaves from a very S0} early period; and yet they are admitted even by their masters to ddm T ARE THE PARIARS OF SOUTHERN INDIA DRAVIDIANS? 499 belong to the same race as themselves. There are also servile sub- divisions of some of the higher or unquestionably Dravidian castes. Thus, a portion of the Maravas of the southern provinces, are slaves to the Poligars, or Marava chieftains; and even of the ¢ Velldlas, or Tamilian cultivators, there are not a few families who are slaves to the temples. Various circumstances might contribute to the reduction of the Pariars, &ec., to servitude, irrespective of difference or identity of race. In the wars of barbarous nations, it often happens that both conquerors 6 and conquered belong to the same race, and even to the same tribe. bl In a civilized age, the conquerors may be content with governing and taxing the conquered ; but in a ruder age, and especially in a tropical {7 16 climate, where labour is distasteful, the vanquished are ordinarily reduced to the condition of slaves. In such cases we shall meet with %) a phenomenon exactly parallel to that of the Pariars, viz., a servile be tribe speaking the language and exhibiting the physiological pecu- 18 liarities of their masters, and yet separated from them by an impassable L1} barrier. Other causes, however, in addition to that of war, may have I8 been in operation, such as poverty, or a state of society resembling the 80, feudal system, or even a trade in slaves like that which in Africa sets £y not only nation against nation, but village against village. At all he events, taking into account the probability that these and similar social evils may have existed, it does not seem more difficult to account pen for the enslaved condition of the Pariars, without supposing them to have been a different race from their masters, than it is to account for 68, ey the serfdlom of the Russian peasantry, or the existence of slavery for amongst nearly all the primitive Indo-European races, without the help of any such supposition. 1 of It is worthy of notice also, that whilst the Pariars, Pallars, and hat Puliars are slaves, a much larger number of the castes that are included \ffer in the lower division—including some of the very lowest—are com- ‘sian posed of freemen. wiﬂl (2.) The traditions that have been mentioned respecting the pent honourable position formerly occupied by the Pariars, do not establish the point in hand. Supposing them to rest upon a historical founda- tion, they prove, not an original difference of race, but only the ancient bitesf jany freedom of the Pariars, and the respectability of their social rank, more before their reduction to slavery. i6 of (8.) The circumstance that the entire circle of the lower castes, ndt including the Pariars, are separated from the higher by badges of social distinction, and denied the national name, is one which must be very r 0 admitted to possess great weight: Though the argument which may 2K 2 500 - APPENDIX. be deduced from this circumstance is a very strong one, it does not appear, however, to be absolutely conclusive, for it is in accordance with the genius of Hindu legislation, to punish poverty by civil and social disabilities; and high-caste pride might naturally take the shape of an exclusive appropriation even of the national name. We find a parallel use of words in the Sanscrit Shastras, in which nations that are admitted to be of Kshatriya origin (e.g., the Yavanas and Chinas), ave termed Mléchchas, not in consequence of difference of race, but solely in consequence of their disuse of Brahmanical rites.’ There is a still closer parallel in the law of Manu, that Brahmans who took up their abode in the Dravida country—in Manu’s time an un- cleared forest—should be regarded as MIléchchas. (4.) There is nothing in the physiology of the Pariars, in their features, or in the colour of their skin, which warrants us to suppose that they belong to a different race from their high-caste neighbours. The comparative blackness of their complexion has led some persons to suppose them to be descended from an imaginary race of Negro aborigines; but this hypothesis is unnecessary, as well as gratuitous. The swarthiness of the complexion not only of the Pariars, but also of the Puliars of the Malayala country—a still blacker caste,—is ade- ; quately accounted for by their continual employment for many ages in the open air, exposed to the full force of the vertical sun. If the Fellahs, or labourers, and Bedouins, or wandering shepherds, of Egypt, are admitted to be Arabs of pure blood, notwithstanding the deep brown of their complexion, it is unnecessary to suppose the Pariars, who labour in a hotter sun than that of Egypt, to be of a different race from the rest of the Dravidians, in order to account for their com- plexions being a shade darker. Such of the Pariars as bave had the good fortune to be placed in more favourable circumstances, are found to be as fair as the high- castes. When Pariars have risen to a position of competence and comfort, and Sfdras have become impoverished, and been obliged to work hard in the sun all day, their mutual difference of complexion is reversed, as well as their social position; and in the second, or at least in the third generation the Stidra becomes dark, the Pariar fuir. § v 1 admit that the features of the Pariars differ from those of the 1] high-caste ¢cultivators,” as the features of every caste in India differ f from those of every other caste; yet there is no difference between the ¢ cultivator’ and the Pariar in the shape of their heads. Not only from their peculiarities of feature and dress, but even from the shape V] of their heads, we are generally able to distinguish Tamilians or il Telugus from the Turco-Tartar Mahommedans of India. But looking *H I ARE THE PARIARS OF SOUTHERN INDIA DRAVIDIANS? 501 at the shape of their heads alone, and leaving complexion and features out of account, it is impossible to distinguish a Tamilian, or high-caste Dravidian, from a Pariar or any other member of the low-castes Difference in feature is of little or no account in this inquiry, for it is notorious that castes which proceed from the same origin differ from one another both in features and in mental characteristics, as widely as if they inhabited different and distant countries t The robber castes of Kallars and Maravars, differ as much from the higher castes in their features as the Pariars, and in habit of mind 0 they differ still more Nevertheless, they claim to be considered as Stdras The caste title of the Maravars, ‘Déva,’ is the same as that of the old kings of the Pandiya and Chola dynasties Chieftains of their i race still possess the principalities of Shevagunga and Ramnad, which are called ‘the tw Maravas;’ and the latter, the prince of Ramnad \\e has claimed from an ancient period to be considered as Sétupati, or [§ JIE hereditary ¢ guardian of Rama’s bridge.’ The other predatory tribe, the Kallars, have a king of their own, 10 s the Tondiman Rajah, or Rajah of Poodoocottah; they claim a relation- ship to the ancient kings of the Chéla country ; and they are regarded of by the Tamilian Vellalars, or cultivators as next in rank to them- de selves 10 It is possible, and even probable, that these robber castes settled the in the Tamil country subsequently to the settlement of the mass of the S Pt) eep population ; but it does not follow that they belonged to an Un-Dra- vidian race For the course which I have supposed the Kullars and lﬂrDY Maravas to have followed, is precisely that which was followed on the 106 decline of the power of the Pandiyas, by various Telugu and Canarese il castes that are unquestionably Dravidians J 4 1n (5.) The essential unity of all the Dravidian dialects argues the unity of the race, inclusive of the lower castes igh The mixed origin of the Hindus of the Gaura provinces may be and conjectured, not only from historical notices, but from an examination ed to of the component elements of the northern vernaculars In those on 18 vernaculars we can trace the existence of two lingual currents, the Jeast Sanscrit and the Scythian, the one running counter to the other; but in no dialect of the Dravidian languages are such traces discoverable f the of any extraneous idiom which appears to have differed in character differ from that of the mass of the language o the All the grammatical forms of primary importance in all the Dra- L Oﬂly vidian dialects cohere together and form one harmonious system If hape the Pariars and the other servile castes are supposed to be a diffe- s O rent race from the Dravidians, and the only surviving descendants of i 502 APPENDIX. the true aborigines, it will be necessary to regard the isolated moun- tain tribes, the Tudas, Go6nds, &ec., as rempants of the same aboriginal race; and if this theory were correct, the languages of those long isolated tribes, should be found to differ essentially from the Telugu and the Tamil, On the contrary, no essential difference in gramma- tical structure, or in the more important names of things, has been discovered in them; but the Gond and the Ku, the Tuda and the Koéta dialects, belong demonstrably to the same family as the more cultivated Drévidian tongues. It is also worthy of notice, that though the Pariars and the other servile classes in the plains live in hamlets by themselves, removed to a considerable distance from the villages in which their high-caste masters reside, there is no trace amongst them of any difference in idiom, of peculiar words, or of peculiar forms of speech. The only difference which is apparent, consists in their mispronunciation of Sanscrit derivatives, arising from their general want of education; and in many instances, even this difference is not found to exist. 1 On the whole, therefore, the supposition that the lower castes in the Dravidian provinces belong to a different race from the higher, LRSS R appears to me to be untenable. It seems safer to hold, that all the indigenous tribes who were found by the Aryans in Southern India, b belonged to one and the same race. It is probable enough that the Dravidians were broken up into hostile tribes before the Aryan immi- gration, and that the distinctions, not only of richer and poorer, but t also of master and slave, had already come into existence amongst them. Dl Those distinctions may have formed the foundation of the th caste system, which their Brahmanical civilizers introduced, and which e was moulded by degrees into an exact counterpart of the caste system al of Northern India. /i pe 1§ Py COl tig i Jag hut h (] [ | the; 11 Are the Nilgherry Tudas Dravidians ? Tue Tudas, or aboriginal inhabitants of the Nilgherry Hills, are commonly supposed to belong to a different race from the Dravidians of the plains, The reasons that have been adduced in support of this supposition appear to me inconclusive. Unfortunately, so much exaggeration and error are included in those reasons arising from the sentimental interest with which everything connected with the Tudas has been invested by tourists, that there is not much satisfaction in dealing with the question. (1.) The difference of the religion of the Tudas from what is called Hinduism, or the Brahmanical religion, is alleged to prove that they belong to a different race to their low country neighbours, It is quite true that the Tuda religion differs greatly from the Brahmanical ; but it will be shown in another portion of this Appendix that the original religion of the majority of the Dravidians of the plains differed from Brahmanism as widely, and that the religion of the Gonds and Kus, who are as certainly Dravidians as the Tamil ; people themselves, is very different from the religion of the Brahmaus, I and not unlike that of the Tudas, i (2.) The mannersand customs of the Tudas are said to be altogether sut generis, and such as to indicate an origin different from that of the people of the plains. Many of the customs of this tribe are certainly remarkable, but it i is a mistake to regard them as peculiar to the Tudas, and sui generis. Polyandria is practised by the Tudas, but it is practised also by the Coorgs, whose Drévidianism cannot be questioned ; and female infan- g ticide is not confined to the Tudas, but is unfortunately too well known in various parts of India. The Tudas are not the only Indian people who live a wandering, | pastoral life ; who subsist entirely upon milk and grain; who dwell in huts formed of twisted bamboos ; who wear no covering upon their s heads ; who let their hair grow to its full length ; or who never wash their clothes or bodies from their birth to their death. | % o 504 APPENDIX. Each of those customs is practised by various other Indian tribes, though not all of them, perhaps, by any tribe but the Tudas: and though the Tudas may observe some customs of minor importance which are quite peculiar to themselves—(e.g., the Tuda men do not, like other long-haired Dravidians, tie their long hairin a knot like women, but allow it to cluster round the head in natural curls. The bamboo huts, also, in which they dwell are built on the plan of a perfect equi- lateral arch),—yet the observance of a few peculiar customs by a caste which is so isolated as the Tudas, cannot be regarded as a proof of difference of race ; for every caste in India, whether :A\ryan or Dra- vidian, whether high or low, has some custom or another which is entirely peculiar to itself,—-generally some peculiarity in dress, in the ornaments worn by the women, or in the manner in which their houses are built. (8.) The Tudas are said to be a fine manly race, with European features, Roman noses, hazel eyes, and great physical strength ; and hence it is concluded that they differ from the Tamilians and other Dravidians in origin, as well as in appearance. It is certain that the Tudas are an athletic, hardy, fine-looking race, as might be expected from their simple mode of life and the bracing mountain air which they breathe ; but it isalso certain that many of the statements that are commonly made, both in conversation and in books, respecting their physical characteristics are mere romance. As regards size and strength of body they will not bear a comparison with the natives of the north-western provinces, or even with the Telugu palanquin bearers. The supporters of the Celtic or Indo-European origin of the Tudas are wont to rest the chief weight of their theory in the Roman noses of their protégés ; but aquiline noses are not un- frequently met with amongst the people of the plains, though they have not had the good fortune to attract so much of the notice of tourists : and after all, the nose which is most commonly seen on the Tuda face is not an aquiline nose, but simply a large nose. Even if it were universally aquiline, it would reveal nothing respecting the origin of the Tudas ; for physiology makes little account of noses, but much of heads, and the shape of the heads of the Tudas does not differ Even their SR e aaieid materially from that of the low country Dravidians. features do not differ from those of the people of the plains to a greater degree than their isolated situation for many ages would lead us to expect. It is true that the Tudas have hazel eyes and naturally curling hair ; and this alone would give them a different appearance from the black-eyed, straight-haired people of the plains. The colour A s of their eyes may be, and probably is, the result of their long residence ST ARE THE NILGHERRY TUDAS DRAVIDIANS? 505 in the temperate climate of the hills ; but this circumstance, when con- sidered as an argument for difference of race, is neutralized by the dark colour of their hair, approaching to black, and especially by the darkness of the colour of their skin. Tt has not becn noticed by writers on the Nilgherries, but it is nevertheless a fact, that, notwithstanding the long residence of the Tudas on a cold, cloudy, mountain region, the P colour of their skin is considerably darker than that of the more modern hill race, the Badagars, a race of people who immigrated from the Canarese country not many centuries ago, and is many shades darker than that of the majority of the natives of the Malabar coast. The darkness of the complexion of the Tudas appears to prove that they came originally from the eastern or sun-burnt side of the range of Ghauts ; and that long before they took up their abode on the hills they had formed a constituent portion of the low country population. It should be observed also, that this inference exactly accords with the results that were deduced from the examination of the Tuda language which is contained in the Introduction. It was there shown that the language of the Tudas was essentially Dravidian, and that it is, on the whole, more nearly allied to the Tamil, the language which is spoken in the plains on the eastern side, than to any other dialect. % After weighing the various considerations that have now been ) adduced, we may, I think, safely adopt the conclusion that the Tudas 5 belong to the same race and stock as the mass of the Dravidians, though long separated from the rest of the race, and isolated from its civiliza- tion. It may, at least, be confidently asserted that the evidence of the Dravidian origin of the Tudas greatly preponderates over that of every other supposition. ;:11, ] n h er ) (78 {0 Iy 06 ur 08 111 Dravidian i’hysical Type. My object in making a few remarks on the physical type of the Dravidian race is merely that of guarding the reader against certain commonly received errors. Lingual comparison is, I believe, the only guide to a knowledge of the pre-historic relationship of the Dravidian family on which any reliance can safely be placed ; and though I admit that, in some instances, physiology has contributed much to the dis- covery of the affiliation of races, it seems to me, in so far as the study hags hitherto been pursued, that it is at fault in this instance. I must here premise that my remarks relate exclusively to the Dra- vidian race properly so called, not to the aboriginal or Nishada races of India generally. Many of the physical characteristics which Mz Hodgson attributes to the ¢ Tamulians,” may undoubtedly be observed in the Bhutin and Sub-Himalayan tribes, and in a smaller degree in the Santals and other Koles ; but the inexpediency of using, as a general appellation so definite a term as ¢ Tamulian,” appears from the error into which some recent writers have fallen of attributing the same or similar physical characteristics to the Dravidians or Tamulians of Southern India, who differ as much from the Himalayan tribes as do the Brahmans themselves. Mr. Hodgson thus distinguishes the ¢ Arians’ from the ¢ Tamu- lians ¥ ¢ A practised eye will distinguish at a glance between the Arian and Tamulian style of features and form—a practised pen will readily make the distinction felt—but to perceive and to make others perceive, by pen or pencil, the physical traits that separate each group or people of Arian or of Tamulian extraction from each other group, would be a task indeed ! In the Arian form their is height, symmetry, light- ness and flexibility : in the Arian face an oval contour with ample forehead and moderate jaws and mouth ; a round chin, perpendicular with the forehead, a regular set of distinct and fine features; a well raised and unexpanded nose, with elliptic nares ; a well-sized and freely opened eye, running directly across the face ; no want of eye- e e e A s s il DRAVIDIAN PHYSICAL TYPE. 507 brows, eye-lash, or beard ; and lastl y a clear brunet complexion ; often not darker than that of the most southern Kuropeans. In the Tamulian form, on the contrary, there is less height, less symmetry, more dumpiness and flesh : in the Tamulian face, a some- what lozenge contour caused by the large cheek bones, less perpendi- cularity in the features to the front, occasioned not so much by defect of forehead or chin, as by excess of jaws and mouth ; a larger propor- tion of face to head, and less roundness in the latter; a broader, flatter face, with features less symmetrical, but perhaps more expression, at least of individuality ; a shorter, wider nose, often clubbed at the end and furnished with round nostrils ; eyes less, and less fully opened, and less evenly crossing the face by their line of aperture ; ears larger, lips thicker ; beard deficient ; colour brunet, as in the last, but darker on the whole, and, as in it, various. Such is the general description of the Indian Arians and Turanians.’ Mr. Hodgson states also in several places that a Mongolian stamp is impressed on all the aborigines of India. “Look steadfastly,” he says, ‘on any man of an aboriginal race, and say if a Mongol origin 1s not palbably inscribed on his face.’ R R R S A Probably there was little if any reference to the Tamulians, pro- perly so called, in this striking and accurate description of the Brah- ? mans of Northern India and of the forest tribes of the Himalayas and the Vindhyas ; but through the vague use of the appellation ¢ Tamu- | lian,” it is evident that Prof. Max Muller has been led to suppose the same description applicable to the Dravidians proper, or aboriginal in- habitants of the south. Founding his theory on this description, which he quotes and eulogizes (in his Zuranian Lesearches, included in Bun- sen’s Outlines of Universal History),—he says: ‘ From the most ancient times to the period of the Puranas, we meet everywhere with indications, more or less distinet, of two races brought into contact in the Indian peninsula and again, ¢ The traveller in India to the pre- sent day, though he would look in vain for the distinctive features of a Brahman (?), a Kshattriya, or a Vaisya, feels the conviction irre- sistibly growing upon him, as he passes along the streets of cities, or ¢l the roads of villages, whether north or south of the Vindhyas, that every- where he is brought in contact with at least two races of man, distinet [ in mind as well as in hody.’ It is evident also from a quotation from a paper of Dr. Stevenson’s, which he subjoins, that by those ¢ two races o i of man’ he understood ‘the higher and lower orders of natives’— I x]l “the Brahmans and other castes allied to them, and the lower or non- :;i Arya,n castes of the Hindu population.’ We thus arrive at the con- d clusion that Mr. Hodgson’s description of the physical peculiarities 4 sz % APPENDIX. 508 which he calls ¢ Tamulian,’—that is, as he understands the term Turanian or Mongolian—has come to be accepted as a faithful pour- traiture of the Un-Brahmanical Hindus generally, including the Un- Brahmanical classes south of the Vindhyas, ¢.c., the entire mass of the Drévidian people. The Professor quotes also those notices from the Puranas in which the type of the Nishada features is given.—He is ‘a being, they say, ‘of the complexion of a charred stake, with flattened features, and of dwarfish stature.’ ¢ The inhabitants of the Vindhya mountains are called his descendants. According to the Mat- syapurana, they were as black as collyrium. According to the Bhaga- rata-purdna, they had short arms and legs, were black as a crow, with projecting chin, broad and flat nose, red eyes and tawny hair. The Padma-purina adds a wide mouth, large ears, and a protuberant belly, and particularises their descendants as Kiratas, Bhillas, Bahanakas, In the next chapter the Professor states Bhramaras, and Pulindas.’ that he ‘accepts for his starting point this general distinction between Arya,s and Nishadas, which, whether suggested by physical features or proved by the evidence of grammar, may be considered as an undis- puted fact ; and he then proceeds to inquire ¢ whether they can be subdivided into distinct groups.’ Finally he distinguishes, yet on lingual evidence alone, between two classes of Nishadas, the Tamulic, in the narrower sense of the word, and the Bhotiya or Sub-Himalayan.’ Another recent writer, Dr. Logan, treating of the Dravidians ex- clusively, thinks that there is a strong Melanesian or Indo-African element in the Tamil physiology ; and accounts for it by the supposi- tion that a negro race overspread India and Ultra India, not only before the arrival of the Aryans, but even before the arrival of the Scythians. He sces an evidence of this in the colour of the Dravidians, and in the exceeding variety of physical type and features which he observes amongst them. Yet even in his opinion, and in this point at least I think he is quite correct, the Tamilians are ¢intellectually more Europeanised than any other Tartaro-Iranian race.’ The tide evidently runs so strongly against my Dravidian friends, that it seems almost fool-hardy to attempt to resist it; and yet I am persuaded that it has arisen in the main from misapprehension. I am persuaded that the physical type of the Dravidians is not Mongolian, is not tinged with negro peculiarities, is not essentially different from that type which is called Caucasian or Indo-Germanic ; and that %’ whilst the distinct, indubitable evidence of their language requires that the Dravidians should be affiliated with the Scythian race, physio- logical considerations throw, in this instance, so little light upon the o DRAVIDIAN PHYSICAL TYPE. 509 = subject that they would admit of our affiliating them, if it seemed necessary, with the Indo-Europeans. Leaving out of account, for the present, the question of colour, it does not appear to me that there is any essential difference between the heads or features of the Dravidians and those of the Brahmans. There is, it is true, a great variety of feature, as well as of colour, apparent amongst the Dravidians; but though the varieties of feature, or rather of physiognomy, which one observes are numerous, they are generally so minute and unimportant that in the absence of any class- difference in the shape of the head, they are consistent with the sup- position of oneness of blood, and may safely be referred to local, social, and individual causes of difference. The long continued operation of the caste-law of the Hindus appears to me to be quite sufficient to account for the differences of feature, colour, and expression that are observed M to exist. Like oil and water in the same vessel, or ingredients which may be I ; mixed mechanically, but will not combine chymically, the various castes into which the Dravidians were arranged by their Brahman preceptors have lived side by side for ages, probably in some instances éi for twenty-five centuries, without commingling. For ages there has 10 been no intermarriage, no social intercourse, no common bond of sym- pathy. Rank has become hereditary, as well as caste ; and not only rank, but even intellect, temperament, character, and physical charac- teristics. In consequence of the separation of caste from caste for so i many ages through the prohibition of intermarriage, unmistakeable 1 points of difference both in features and in mental temperament have e been developed. 1 It would be surprising indeed if under such circumstances ¢varieties he of man’ did not make their appearance, and if ethnologists, looking {2 at the question from a distance, did not sometimes doubt whether they 1 could all be referred to a single race of pure blood. ‘Some,’ says )16 Dr. Logan, speaking of the Tamilians in particular, ‘are exceedingly Iranian, more are Semitico-Iranian, some are Semitic, others Australian, ds, some remind us of Egyptians, while others again have Malayo-Poly- A nesian, and even Simang and Papuan features.’ am In no country in the world are features and complexion so variable ally as in India ; but caste, as it exists in India, and especially as it affects il the condition of the lower classes, is unknown in every other country in the world. hat 768 Separate for ever from the society of their fellow countrymen a si0- class of agricultural labonrers or slaves: prohibit all intermarriage the with families in more easy circumstances: require them to live by 510 APPENDIX themselves in wretched wigwams, removed to a cousiderable distance from the village which is inhabited by the respectable householders compel them to work hard the whole year round in the open air in an inter-tropical climate—in a country where the sun comes twice in the year right over head let all possibility of their rising to a higher condition of life, or obtaining a more sedentary, shady employment be pay them no wages feed for ever precluded : prohibit education them scantily and clothe them still more scantily : encourage drunken- ness and the eating of carrion prohibit the women from dressing themselves with ordinary regard for decency :—treat them, in short for twenty centuries as the Brahmans and high-caste Dravidians have treated the Pariars and other low-castes, and it will be quite unneces- sary to have recourse to Dr. Logan’s theory of their intermixture with a primitive race of Africans or Negritoes in order to account for the coarseness of their features, their dwarfishness, or the blackness of their skin Notwithstanding all this, though the Pariars, as a class are darker than any other class in the Carnatic, we find amongst them as great a variety of colour as amongst other classes of Hindus ; and occasionally we may notice complexions that are as clear as those of ] the higher castes, together with considerable regularity of features The question, however, which is before us now 1 not ‘are the Pariars of the same race as the high-caste Dravidians ¥ but ‘to what race do the DrAvidians themselves belong ¥ — ¢ Do those who are admitted to be Dravidians of pure blood resemble most the Aryans or the Mongohans ¥ this is the question really at issue. In comparing the physical type of the Dréavidians with that of Mongolians and Aryans, it is unfair to restrict the comparison to the lower classes of Dravidians; for the South-Indian Stdras, or high caste Dravidians, claim to be regarded as the purest replesentatwes of their race Their institutions and manners have been Aryamsed but it is pure Dravidian blood which flows in their veins There may possibly be some doubt whether the lower castes were not intermixed with an anterior race: but the higher castes call themselves Tamilians Telugus, Malayéalis, &c. par excellence ; and their special right to those national appellations is always admitted, in terms at least, by the lower tes themselves When we compare the physical type of cultivated, high caste Dra- vidians with that of the Brahmans, no essential difference whatever and very little difference of any kind, can be observed In many instances the features of the high-caste Dravidian women are more delicately formed and more regular than those of Brahman women themselves, whilst their complexions are at least equally fair e DRAVIDIAN PHYSICAL TYPE. 511 and if any difference appears, it consists not in Mongolian breadth of face, but in greater elongation and narrowness. The Dravidian type of head will even bear to be directly compared with the European. Compare, for instance, the heads of the Tamil or Telugu Munshis, translators, and Pandits in any Zillah court with that of the presiding English judge ; and it is evident that the Dravidian heads differ from the English only in being smaller and narrower,—with a preponderance in the former of the signs of subtilty and suppleness, in the latter of 4 straightforward moral and mental energy. & It is especially deserving of consideration, that the Nilgherry Tudas, who of all Dravidian tribes have been most thoroughly guarded by their secluded position from Bhramanical influences, instead of E being more Mongol-like or Negro-like than the AryaniSed Dravidians, are so distinctively Caucasian in the opinion of many persons, that { they have been regarded as Celts, Romans, Jews, &c., and the chief difficulty that exists is that of inducing people to be content with the 1l statement, that the Tudas are proved by their language to be identical in origin with the Dravidians of the plains. it Amongst the lower class of the Dravidians, I have occasionally observed a type of head which is somewhat inclined to be what is 16 called Mongolian, that is, it exhibits unusual breadth across the cheek it bones, a pyramidal forehead, a somewhat oblique position of the eyes, and a pyramidal nose with a broad base. {7 Or On the other hand, Mongolian smoothness of skin, seantiness of hair, flatness of face, and the peculiar monotonous olive hue of the Mongolian complexion are never met with ; and it should be observed of he that the other elements of the Mongolian type which one does occa- oh sionally notice, though it is chiefly, if not solely, amongst the lower of classes that they are seen, yet they do not constitute the class-type of t any caste whatever; nor are they ordinarily or frequently met with, 0y but are exceptional instances, which scarcely at all affect the general ed rale; and I have no doubt that similar exceptional instances could 0 easily be pointed out amongst the lower classes of our own race. The physical type of a race should be determined by the shape of )58 jer the head and the more permanent peculiarities of feature alone, irre- spective of the complexion or colour of the skin; for every one who has lived in India must have learned to regard colour as a most T () deceptive evidence of relationship and race. It is true that the Brahmans as a class are much fairer than the Pariars as a class ; but e the conviction is forced upon the mind of every observer, by the hundreds of instances which he meets with in daily life, that the a1 colour of the features of the Hindus is purely a result of the external it D S 512 APPENDIX. circumstances in which they are placed, with respect to climate, occu- pation, and mode of life. They are dark-complexioned in proportion as they are exposed to the sun in out-door labour, and fair in propor- tion as they live a sedentary life; and consequently colour, if an evidence of anything specific, is an evidence only of the social status of the individual and his family. It is vain, therefore, to expect from considerations of colour and complexion any real help towards deter- mining the race to which the Dravidians belong. The influence of climate alone, in darkening or blanching the colour of the skin is greater than is commonly supposed; and India One of the best of furnishes innumerable instances of this influeuce. Indian instances of the influence of climate in modifying colour with which I am acquainted, is furnished by the extreme fairness of the complexion of th e greater proportion of the natives of the Malabar coast, compared with the very dark hue of a like proportion of the natives of the coast of Coromandel, who belong to the same or similar castes, and who follow similar occupations. The natives of the Coro- mandel coast are exposed for ten months in the year to a very high degree of dry heat, in a level country, bare of wood.* The natives of the Malabar coast are exposed to a similar degree of heat for not more than two months out of the twelve, and a similar degree of drought is on that coast unknown: their sky is almost always laden with moisture ; the quantity of rain that falls is always double, gene- rally treble, the quantity that falls on the eastern coast in the same latitude. The country is everywhere well wooded, and the houses of the people are generally nestled in deep, cool groves; and, in conse- quence, in the same degree of latitude, and with a difference in longitude of only a degree or half a degree, the skin of the people on the western side of the Ghauts (or central mountain range of Southern India) is as much fairer than that of the people on the eastern side, as the complexion of the Brahmans of any province is fairer than that of the labouring classes in the same province. Notwithstanding this difference in complexion, there is no difference in race, for the Malay- alis are demonstrably descended from an early colony of Tamilians; and an equally remarkable difference in complexion is apparent amongst the members of those Tamil castes, of whom a portion have settled in Malayalam. 1 never knew the thermometer * In my own n eighbourhood in Tinnevelly, fall lower at any p eriod of the year, day or night, than 7 6°. For about eight months in the year it ave raged 84°, and for about two months rose above 90°. j The maximum, which it rarely reached in the shade, was 96°. The average fall of rain in the same neighbourhood, during the six years in which I registered it, was only 25 inches! | e — R—— DRAVIDIAN PHYSICAL TYPE. 513 Towards the southern extremity of the Peninsula, the breadth of the central mountain range is greatly diminished, and there is easy access from the Tamil country into Travancore by the Aramboly pass. By this pass, and by similar breaks and gaps still farther south, the Tamilians of the old Pandiya kingdom forced their way into Malay- alam, and possessed themselves of the province of South Travancore. The government of this province has again reverted to the Travan- corians, in whose hands it has been for several centuries; but the bulk 4 of the population continue to be Tamilians, as far as the vicinity of Trivandrum, the Travancore capital. Up to that limit the majority | of the people on the Travancore side of the mountain barrier belong to the same castes as in the East India Company’s district of Tinne- velly, on the eastern side ; they speak the same language, and follow the same occupations; they occasionally intermarry, and their features are perfectly similar; yet, notwithstanding all this, they differ so % materially with respect to colour, that a stranger would naturally suppose them to belong to different races. A remarkable instance of difference of colour under these circumstances is furnished by the 3 Shéanérs, or palmyra cultivators, who are found in considerable numbers it on each side of the Ghauts, up to the very foot of the mountains. In of the vicinity of Neyattangkarei, a Travancore village, the Shanars on the western side of the mountain range, are separated from their fellow 1 caste-men on the eastern side, by a space of only about fifteen miles 16 as the crow flies; and the only difference in their circumstances, is the of difference in the climate, which is caused by the precipitation of the moisture of the south-western monsoon on the western side of the Ghauts, and its interception from the eastern side. In consequence of 1n )} this difference in the climate alone, the Shénars who reside on the eastern side of the Ghauts, are amongst the blackest of the Tamilians, 1 while on the Travancore side, the same class of people, engaged in the e same occupations, are as fair as the Brahmans of the Carnatic, and in ab some Instances even fairer. This fact, which is patent to the observa- 11§ J tion of every one in the neighbourhood, is perhaps the most remarkable proof in existence, of the immense influence of climate in modifying 15 nb the colour of the skin. Another and better known evidence, is furnished by the circum- Vo stance that many of the descendants of the Portuguese who settled in India several centuries ago, are now blacker than the Hindus them- fer selves. The class of people referred to are a mixed race, descended ght )0 from European fathers and native mothers, yet instead of being the fairer for their admixture with European blood, many of them are of a B darker colour than the natives from whom, on the maternal side, they 2.1 o 514 APPENDIX. are descended. Even amongst the Brahmans, though a perfectly pure, unmixed race, differences of colour are frequently observed. It is supposed to be unlucky to meet ‘a black Brahman,” or ‘a fair Pariar, the first thing in the morning. The Brahmans of Northern India are generally fairer than those of the south, with the exception of the Nambiris, or high-caste Brahmans of the Malabar coast, who appear to be the fairest of their race. Professor Max Miiller (in his valuable paper on the Bengali in the Reports of the British Association), thinks he finds in the Gonds, and other Un-Aryanised Dravidians, evidences of the existence of a race ‘ closely resembling the Negro; and says that, ¢the existence of the On same dark race in the South of India, is authenticated by Strabo.’ the contrary, Strabo’s statement, when not merely alluded to, but translated, will be found to corroborate the view which I have taken. He says, ‘the Southern Indians resemble the Althiopians in colowr: but in features and in hair, they resemble the rest of the Indians (for on account of the moisture of the climate the hair does not become woolly) ; but the Northern Indians resemble the Egyptians.’ This statement of Strabo throws light on a passage in Herodotus, in which a black race, apparently Hindus, are said to have been brigaded with the fairer Indians in the army of Xerxes. He says, ¢ Athiopians from the eastward—from the sun-rising—from Asia— marched with Indians, but differed not from other Athiopians except in their language and their hair; for the Libyan Aithiopians have the woolliest hair of all men, but those people are straight-haired.’ Hero- dotus supplies us with a fact, Strabo with the right explanation of that fact. Herodotus is silent with respect to the features of the Eastern Athiopians; Strabo asserts that their features resembled those of the rest of the Indians. Though there is little or nothing of a distinctively Mongolian character in the features and heads of the Dravidians, considered generally ; and though consequently physiology does not furnish any reliable evidence in support of their Mongolian or Scythian origin, it is unsafe to draw any conclusion from this circumstance. The danger of arguing from negative evidence respecting matters which are so fleeting and changeable as features and complexion, is illustrated by the change which has passed over the features of the Mahommedans of India. The Mahommedans of India are partly descended from the Affghan, and partly from the Mogol invaders; but the great majority are descendants of the Tartar-Turkish soldiers and camp-followers, who accompanied both the Affghans and the Mogols. Probably most of the Affchan invaders of India were Seljukian Turks; the Mogols i R a a S i DRAVIDIAN PHYSICAL TYPE. 51 (%) 4 were, as their name imports, Mongolians; and the hordes that followed the fortunes of both classes of invaders, were a mixed race —a ‘collu- vies gentium ’—comprising various tribes and races of Mongolian and Tartar-Turkish origin, called by the Hindus Turushkas, in Tamil Turukkar, or more commonly Tulukkar, 7.e., Turks. The proportion of Persians and other races of Indo-European origin, who accompanied the Affghans and Mongols in their expeditions, was exceedingly small; and thongh the Mahommedans have occasionally made proselytes amongst the Hindus, by force or the prospect of secular advantages, and have -occasionally robbed Hindus of their wives and daughters, the disturbing influence of these accessions to their ranks has been so small, that it may be left altogether out of account, Hence, the Mahommedans of India may be regarded as a Tartar-Mongolian people; and we might naturally expect to observe in them those physiological peculiarities of the High Asian races which must have characterised the majority of their ancestors on their first arrival in India, and which are still apparent in all their distine- tiveness, not only in the Mongolians, but in the Siberian Turks. Notwithstanding this, we generally search in vain amongst the Indian Mahommedans for signs of their Tartar origin. With the exception of a somewhat greater breadth of face and head, and a more olive complexion, they do not now differ from the Hindus, properly so called, in any essential point. They exhibit, it is true, special pecu- liarities of physiognomy and expression; but every Hindu tribe or caste has, in like manner, a peculiar physiognomy of its own, by which it differs from every other tribe. A change appears to have st e passed over the physiology of the Mahommedans of India similar to that which the Magyars and the Osmanli Turks have experienced since they settled in Europe, and which has transformed them from Tartars into Europeans. As, therefore, there cannot be any doubt of the original Mongolianism of the majority of Indian Mahommedans, or of the absence from them now of almost every thing that is Mongolian, 8o, though little or nothing that is distinctively Mongolian is now apparent in the features or physiology of the mass of the Drividians, they may, notwithstanding this, be descended from as purely a Scythian or Mongolian ancestry as the Mahommedans are known to be; or at least, we may conclude that there is nothing in the physiological view of the question which is opposed to the argument derived from lingnal comparison. Perhaps, however, on the whole, the safest conclusion is, that the mass of the Dravidians, though as truly Seythians as the Mongolians themselves, were, even at the time of their entrance into India, free 272 J V-3 16 APPENDIX. from those peculiarities of feature that are called Mongolian We cannot safely conclude that the Mongolian type of features was from the beginning the inheritance of the whole of the Scythian tribes It appears more probable that that type was developed in the course of time in the steppes of High Asia; and it is certain that the tribes amongst whom it has acquired a peculiar degree of permanence, are the Tibetans and the Mongolian nomades, who still inhabit the original seats of their race It is remarkable that the only Indian tribes which are now dis- tinetly characterized by Mongolian peculiarities, are those which entered India by the north-east, and which are probably of Tibetan origin The Garos and other forest tribes on the Bhutan frontier, as described by Mr. Hodgson, seem to be decidedly Mongolian ; and the Koles and Santals are probably descended from the same or a similar stock. The existence at an early period in the vicinity of Orissa, of barbarous tribes differing in appearance from the rest of the Hindus, and exhibiting a Mongolian or foreign type, is attested by the follow ing passage in the Periplus Maris Erythrae: After referring to the region watered by the Godavery and Kistna, the author says: ¢ After this, keeping the sea on the right hand and sailing northwards, we come upon certain barbarous tribes, as the ¢ kippadac’ (Sans. ¢ Kiratus’), a race of people with flattened noses (evidently Mongolians), also the horse-faces and the long-faces, all of whom are said to be cannibals Then sailing eastwards, and having a certain sea on the right, we come to the Ganges.’ The distinct statement of Strabo wh]ch has already been quoted, joined to the negative evidence of this passage, proves that at the Christian era, the civilized, cultivated Dravidians (the Pandiyas Calingas, Andhras, &c.), did not materially differ in physiognomy or personal appearance from the northern Hindus; and that certain bar- barous inhabitants of the jungles, who are barbarians still, were the The Gondali only tribes that appeared to be distinctively Mongolian of Ptolemy, probably the Gonds, who are classed among ¢ the Bitti,” and distinguished from ¢ the Phyllitae’ (probably the Bhills), are not said to have differed in appearance from the more cultivated Dravidians Some writers, I think erroneously, speak of the ‘jet blackness’ of the Gonds; and the Rajmahal people are said to be black. Notwith- standing this, according to the account of that accurate observer, Dr. Buchanan Hamilton, the features of the Males or Rajmahal hill people, do not essentially differ from the Aryan type. ‘Their lips are full, but not at all like those of the Negro. Their faces are oval, not shaped like a lozenge as those of the Chinese are Their eyes, instead DRAVIDIAN PHYSICAL TYPE. 517 of being hid in fat and placed obliquely like those of the Chinese, are exactly like those of Europeans.’ We have seen that some of the Vindhya Nishadas are described in the Purinas to be ‘as black as crows; but without debating the accuracy of the amiable portrait of those primitive tribes, which the Puranas have drawn, and which waits to be tested by Mr. Hodgson’s pencil, it will suffice for the present to remind the reader that those very Purina writers entertained so different an impression respecting the true Dréavidians of the south, that they fell into the opposite error of Aryanising them, and supposed the Calingas, Pandiyas, Kéralas, and other Dravidians to be descended from colonies of Aryans from Oude. N 1V Ancient Religion of the Dravidians. ReL1crous usages are sometimes found to throw light on the origin or relationship of races. Similarity in the religiousideas and practises of any two primitive tribes strengthens the evidence of their relation- ship derived from similarity of language. Let us see whether any light can be thrown on the ¢uestion of the relationship of the Dravi- dians by an inquiry into their religious usages. A priori, this inquiry seems likely to lead to some definite result, inasmuch as the religions of the ancient Indo-European nations and the old Scythian religions of Upper Asia present many essential points of difference. In the earliest times we find amongst the nations of the Indo-Euro- pean family the universal prevalence of certain tenets and usages, which each of those nations appears to have inherited from the common pro- The docirine which was most characteristic of genitors of the race. the whole family was that of the Metempsychosis ; their objects of worship were either the elements of nature personified, or a Pantheon of heroes and heroines ; and the most characteristic of their religious usages was the maintenance of a distinet order of priests, generally hereditary, who were venerated as the depositaries of all ancient tradi- tions and spiritual power. In whatever race these religious peculiarities appear to have pre- vailed, we shall probably find on inquiry that there are weighty reasons for attributing to that race an Indo-European origin or rela- tionship : and in like manner a family likeness (exceedingly dissimilar from the particulars now mentioned) will be found to characterize the religious practises of the nations of the Scythian group. In endeavouring to ascertain the characteristics of the primitive Dra- vidian religion, we are met by a serious but not insurmountable difficulty. The Brahmans, by whom the Aryan civilization was grafted on the ruder Dravidian stock, laboured assiduously to extirpate the old Dra- | vidian religion, and to establish their own in its room ; and they are generally supposed to have succeeded in accomplishing this object. Notwithstanding their success however, it is still possible in some S— —— —— — S— I EE—— ANCIENT RELIGION OF THE DRAVIDIANS. 519 degree to discriminate bet\;reen the doctrines and practices which were in- troduced by the Brahmans and the older religion of the Dravidian people. If, for instance, any usages are found to prevail extensively in Southern India, and especially amongst the ruder and less Aryanised tribes, which are derived neither from the Védas nor from the Puranas, neither from Buddhism nor from Jainism, such usages may be concluded to be relics of the religious system of the Drividian aborigines. Many such usages do actually exist. Several religious systems widely differing from the Brahmanical are discoverable amongst the Dravidian nations, and are especially prevalent amongst the rude inhabitants of the jungles. Hence, we are not quite destitute of the means of comparing the characteristics of the ancient Dravidian religion, prior to the intro- duction of Brahmanism (or what is commonly called Hinduism}), with the religious usages that prevailed amongst the Scythian races. The system which prevails in the forests and mountain-fastnesses throughout the Dravidian territories, and also in the extreme south of the Peninsula amongst the low caste tribes, and which appears to have been still more widely prevalent at an early period, is a system of demonolatry, or the worship of evil spirits by means of bloody sacrifices st v e and frantic dances. This system was introduced within the historical period from the Tamil country into Ceylon, where it is now mixed up with Buddhism. On comparing this Dravidian system of demonolatry and sorcery with ¢Shamanism "*—the superstition which prevails amongst the Ugrian races of Siberia and the hill-tribes on the South- Western frontier of China, which is still mixed up with the Buddhism of the Mongols, and which was the old religion of the whole Tartar race before Buddhism and Mohammedanism were disseminated amongst them—we cannot avoid the conclusion that those two superstitions, though practised by races so widely separated, are not only similar but identical. I shall here point out the principal features of resemblance between the Shamanism of High Asia and the demonolatry of the Dravidians,+ as still practised in many districts in Southern India. * This word Shamanism is formed from Shaman, the name of the magician- priest of the North Asian demonolaters. ‘Shaman,” though a name appropriated by demonolaters, is of Buddhist origin, and was adopted from the Mongolians. It is identical with ‘Samana,” the Tamil name for a Buddhist, and is derived from the Sanserit word ‘¢ Srimana.’ The use of this word Shaman in Siberia, must be of comparatively modern origin; but the system of religion into which it has been adopted and incorporated is one of the oldest superstitions in the world. + I beg to refer the reader for a full account of the peculiarities of the Drévi- dian demonolatry, to a small work of mine called 7he Shdndrs of Tinnevelly, I think I have published by the Society for the Propagation of the Gospel. proved in that work that the demonolatry of the Shanars (the palmyra cultivators 5 92 - 0 APPENDIX. (1.) The Shamanites are destitute of a régular priesthood. Ordi- narily the father of the family is the priest and magician ; but the office may be undertaken by any one who pleases, and at any time laid aside. Precisely similar is the practice existing amongst the Shanars and other rude tribes of Southern India. Ordinarily it is the head of the family, or the head-man of the hamlet or community, who performs the priestly office ; but any worshipper, male or female, who feels so disposed, may volunteer to officiate, and becomes for the time being the representative and interpreter of the demon. (2.) The Shamanites acknowledge the existence of a supreme God; but they do not offer him any worship. The same acknowledgment of God’s existence and the same neglect of his worship characterize the religion of the Dravidian demonolaters. (8.) Neither amongst the Shamanites nor amongst the primitive, un-brahmanized demonolaters of India is there any trace of belief in the metempsychosis. (4.) The objects of Shamanite worship are not gods or heroes, but demons, which are supposed to be cruel, revengeful, and capricious, The offi- and are worshipped by bloody sacrifices and wild daneces. clating magician or priest excites himself to frenzy, and then pretends, or supposes himself, to be possessed by the demon to which worship is being offered ; and after the rites are over he communicates, to those who consult him, the information he has received. The demonolatry practised in India by the more primitive Dravi- dian tribes is not only similar to this, but the same. Every word used in the foregoing description of the Shamanite worship would apply equally well to the Dréividian demonolatry ; and in depicting the cere- monies of the one race we depict those of the other also. ; of Southern India) did not originate with the Brahmans, or in any local develop- ment of the religion of the Brahmans; but that on the contrary, the element of demonology which is contained in the later Puranic system, was borrowed from this old Drévidian superstition. ; It is admitted to be a fact that the Buddhists of Ceylon borrowed their demo- nolatry from the DrAvidians of the Old Péindiya Kingdom: if so, it cannot be unreasonable to suppose that it was from the same or a similar source that the Brahmans borrowed the demoniacal element which is contained in the later Puranas. I apprehend that we have a mythical record of the adoption of the aboriginal demonolatry into the later Brahmanical system, and of the object in view in this alliance, in the Puranic story of the sacrifice of Daksha. According to that story, Siva (7.e., Vedantic Brahmanism) found himself unable to subdue the old elementary divinities, and to secure to himself the exclusive homage at which he aimed, till he called in the aid of the demons (the demonolatry of the abo- rigines), and put himself at their head in the person of his (‘ pro-re-nata’) son, Vira-Bhadra; a demi-god, whose wife, emanation, or representative, Bhadra-Cali, is regarded by the Shanirs as their patroness and mother. AR e ANCIENT RELIGION OoF THE DRAVIDIANS, 521 Compare the following accounts of the demonolatrous rites of the Shamanites of Siberia and those of the demonolaters of India. The description of the Shamanite worship is formed from a series of arranged quotations from Prichard’s account of the deseriptions which various Russian travellers and ecclesiastics have given of the superstitions of the Ostiaks, the Samoiedes, the Siberian Turks, and The account of the Dra- other Pagans inhabitants of Northern Asia. vidian superstitions is taken from my paper on ‘the Tinnevelly Shanars,” a paper which was written before I was aware of the identity of the demonolatry of Siberia with that of Tinnevelly. SoHAMANITE DEMONOLATROUS RirteEs.—¢ When the Shaman, of magician, performs his superstitious rites he puts on a garment trimmed with bits of iron, rattles, and bells : he cries horribly, beats a sort of drum, agitates himself, and shakes the metallic appendages of his robe ; and at the same time the bystanders increase the din by striking with their fists upon iron kettles. When the Shaman, by his horrible con- tortions and yells, by cutting himself with knives, whirling and swooning, has succeeded in assuming the appearance of something pre- e o e e ternatural and portentous, the assembled multitude are impressed with the belief that the demon they are worshipping has taken possession of the priest, and regard him accordingly with wonder and dread. When he is quite exhausted with his exertions, and can no longer hold out, he makes a sign that the spirit has left him, and then imparts to the people fche intimations he has received.’ SuiniAr DemoNornaTrROUS RITEs.—¢ When it is determined to offer a sacrifice to a devil, a person is specially appointed to act the part of priest ; for devil-worship is not, like the worship of the deities, appro- priated to a particular order of men, but may be performed by any one who chooses. The officiating priest is styled a ‘devil-dancer.” Usually the ‘head man,” or one of the principal men of the village officiates ; but sometimes the duty is voluntarily undertaken by some devotee, male or female, who wishes to gain notoriety, or in whom the sight of the preparations has awakened a sudden zeal. ‘The officiating priest is dressed up for the occasion in the vestments and ornaments which are appropriated to the particular devil that is worshipped. The object in view in donning the demon’s insignia is to strike terror into the imagination of the beholders ; but the party- coloured dress and grotesque ornaments, the cap and trident and jingling bells, of the performer, bear so close a resemblance to the usual adjuncts of a pantomime that an European would find it difficult to look grave. r J 22 APPENDIX. ‘The musical instruments, or rather the instruments of noise, which are chiefly used in the devil-dance are the tom-tom, or ordinary Indian drum, and the horn; with occasionally the addition of a clarionet when the parties can aflordit. But the favourite instrument, because the noisiest, is that which is called ‘the bow.” A series of bells of various sizes is fastened to the frame of a gigantic bow ; the strings are tightened so as to emit a musical note when struck ; and the bow rests on a large empty brazen pot. The instrument is played on by a plectrum, and several musicians join in the performance. ‘One strikes the strings of the bow with the plectrum, another produces the bass by striking the brazen pot with his hand, and a third beats time and improves the harmony by a pair of cymbals. ‘When the preparations are completed, and the devil-dance is about to commence, the music is at first comparatively slow, and the dancer seems impassive and sullen ; and either he stands still or moves about in gloomy silence. Gradually, as the music becomes quicker and louder, his excitement begins to rise. Sometimes to help him to work himself up into a frenzy he uses medicated draughts; cuts and lace- rates his flesh till the blood flows ; lashes himself with a huge whip ; presses a burning torch to his breast ; drinks the blood which flows from his own wounds ; or drinks the blood of the sacrifice, putting the throat of the decapitated goat to his mouth. Then, as if he had acquired new life, he begins to brandish his staff of bells, and dance with a quick, but wild, unsteady step. Suddenly the aflatus descends. There is no mistaking that glare, or those frantic leaps. He snorts, he stares, he gyrates. The demon has now taken bodily possession of him ; and though he retains the power of utterance and of motion, both are under the demon’s control, and his separate consciousness is in abeyance. The bystanders signalize the event by raising a long shout attended with a peculiar vibratory noise. ‘The devil-dancer is now worshipped as a present deity ; and every bystander consults him respecting his disease, his wants, the welfare of his absent relations, and the offerings which are to be made for the accomplishment of his wishes. ¢ As the devil dancer acts to admiration the part of a maniac, it | requires some experience to enable a person to interpret his dubious or unmeaning replies, his muttered voices and uncouth gestures ; but the wishes of the parties who consult him help them greatly to interpret his meaning.’ It seems to me unnecessary to say anything more in proof of the substantial identity of the demonolatry of the Dravidians of India with the Shamanism of Northern Asia. It may be alleged that simi- ANCIENT RELIGION OF THE DRAVIDIANS. 523 larity in mental characteristics and social circumstances alone might give rise to this similarity in religious ideas and practises, but it geems far more probable that both the superstitions which bave now been described have sprung from a common origin: and I may add that the conformity which has been traced between the old religion of the Dravidians and that which was once the religion of all the Scy- thian nations corroborates the supposition of the Seythian relationship of the Dravidian race. Whilst the demonolatrous rites which I have now described appear to have constituted the prevailing superstition of the ancient Dravi- dians, we meet also with traces of the existence of systems that correspond in part to those which prevailed amongst the Indo-European races. The religion of the Kunds or Kus, though it contains a demonola- trous element, may be described as in the main a worship of gods of rivers and mountains, of gods of the earth and the sky, and of the gods of elements and genii loc. It is in part an elementary wor- ship, which may be allied in principle to that of the Aryans, but &hich differs widely from it in spirit and form, and appears to be quite inde- pendent of it in origin. This remark especially applies to that section of the Kus which practises human sacrifices, and delights in cruelty and gloom. A worship of gods of rivers and mountains similar to that of the Kus is found amongst the Koles, and also amongst the Sub-Himalayan and Bhutén tribes described by Mr. Hodgson ; and it seems not imprqbable that it was from those tribes that the Ku religion was derived. Amongst the Dravidians of the plains no trace of the worship of the elements has ever been discovered. Indeed there is reason to believe that the old Védic or elementary worship of the Brahmans had already merged into the mythological and mystical system of the Puranas, before the Brahmans effected a settlement in the South. So far as appears, every Drividian usage which is not of Brahma- nical origin is either identical with Shamanism or allied to it. The religion of the Tudas of the Nilgherry hills exhibits some peculiarities which are analogous to the earliest Brahmanical religion, or the religion of the Védas, together with some which are regarded as Seytho-Druidical. The peculiar veneration with which the Tudas worship the manes of ancestors ; their sacrifices to secure the peace of the dead ; their worship of genii loci by means of offerings of milk and clarified butter; their freedom from the worship of idols ; the religious veneration with which they appear to regard a sacred bell, which is hung up in- their 524 APPENDIX temples, or dairies; their abstinence from flesh, and living entirely on grain and milk ; their exclusion of women from all share in the rites of worship, and even from the precincts of their temples ; their prac- tice of polyandria and female infanticide ;—these and analogous pecu- liarities of the religious system and social life of the Tudas accord to a certain extent with usages which prevailed in the earliest ages amongst most of the tribes of the Indo-European race. There is no trace amongst the Tudas of hero-worship or of Aryan mythology, of the doctrine of the transmigration of souls, or of the existence of a priestly caste,—all of which are distinctive portions of the Indo-European system. Nevertheless the peculiarities of the reli- gion of the Tudas theh have been mentioned above may be suspected to have had an Aryan origin, or at least to have been shaped and Our ignorance of the history of the X T R e tinged by Aryan influences Tudas (an ignorance which has not been dispelled by the speculations of Captain Congreve), and of the circumstances which compelled them to take refuge in the Nilgherry hills, renders it exceedingly difficult if not impossible, to determine whether their religion sprang from the same Scythian origin as the Dravidian demonolatry, or whether it is to be placed to the account of their early association with We must look to further and more accu- some Indo-European race rate research for the solution of this problem The religion of the Tudas has sometimes been regarded as ¢ Druidical ¢ Celto-Druidical identical with the religion of the ancient Celts ;' but, with the exception of the performance of some of their rites in the deep gloom of sacred groves, a practice which was not ey peculiar to the Celts alone, but which prevailed amongst various ancient naticns, it does not appear that there is anything dlstmctlvely or certainly Druidical in the existing system of the Tudas. The supposition of the Druidical character of the Tuda religion has arisen from the error of attributing to the Tudas various remains and usages which were peculiar to an earlier and probably extinct race Those remains consist of cairns or burrows, eromlechs, kistvaens, and circles of upright, loose stones, which are nearly identical in form with those that are found in Europe in the ancient seats of the Celts and whatever mystery may hang over the origin of those remains and g8 over the race of which they are the only surviving relics, there seems no reason for hesitating to style them, in a general sense, Druidical In the cairns or barrows referred to, vases, cinerary urns, and other vessels of glazed pottery are often found, which sometimes contain SRR e I R ANCIENT RELIGION OF THE DRAVIDIANS. 525 human bones, more or less charred, and mixed with ashes, sometimes a little animal charcoal alone. Most of these vessels have a peculiar glaze of a rich red colour, with a zig-zag ornamentation : some have a black glaze. Brass and iron implements of agriculture and of war have often been discovered in them : in several instances a bell has been found, as in some of the Celtic barrows in England ; and occa- sionally gold ornaments have come to light. Though these remains seem to be undoubtedly Druidical, they can hardly lay claim to an antiquity equal to that of many Druidical remains found in Europe. The rich glaze of the pottery ; the elegance of the shape of some of the vessels (compared with the rude cinerary urns discovered in the British barrows); the presence of implements of iron ; the representa- tions of processions with musieal instruments and led horses, which are rudely sculptured on the sides of some of the cromlechs ; the presence of gold ornaments ; all these circumstances denote a superior civilization to that of the primitive Celts, and therefore a later origin of the relics. If it be true, as it is confidently asserted (though I have been unable to ascertain the truth of the statement), that a Roman aureus was dis- covered in one of the barrows, the race by which those Druidical rites e e were practised must have survived for several centuries after the Christian era. At first it was supposed that cairns and other Druidical remains were discoverable only on the Nilgherry hills; and hence it was natural that these remains should at first be attributed to the Tudas, the aborigines of the Nilgherries, and who are as peculiar in their cus- toms as in their language. On further research it was found that the people to whom those remains belonged had practised agriculture and made use of horses; whereas the Tudas were ignorant of agriculture, appeared to have always lived a pastoral, wandering life, and were ignorant even of the existence of the horse. It was subsequently dis- covered that the Tudas neither claimed the cairns and cromlechs as B i l©n lH\ih» it b i°™ belonging to themselves or their ancestors, nor regarded them with reverence ; that their rites of sepulture are altogether different from those of the ancient people who used those cairns; and that they ascribed them to a people still more ancient than themselves, by whom they assert that the plateau of the Nilgherries was inhabited prior to their arrival, Sometimes they designated the cairns as burial places of the ¢Curbs,’ i.e., of the Curubas or Curumbars, a race of nomade shepherds who once overspread a considerable part of the Tamil country (probably the ‘nomadic Soras’ of Ptolemy), and of whom a few scattered relics still inhabit the slopes of the Nilgherries. It appeared, however, on making inquiry of the Curubas, that they 526 APPENDIX. neither practised Druidical rites themselves, nor supposed the barrows to be the work of their ancestors; so that the problem still remained unsolved. It was at length ascertained that similar cairns or barrows, containing a great variety of similar remains, but of a more advanced order and in a better condition, existed in immense numbers on the Ana-mala hills,—a réJnge of hills on the south side of the great Coim- batoor gap, which form the commencement and the northern face of the Southern Ghauts ; and further investigation proved their existence, not only in mountain ranges, but in almost every part of the Dekhan and Peninsular India, from Nagpore to Madura, and also in various districts in the presidency of Bombay. Similar remains are found also in Circassia and Russia ; and circles of stones surrounding ancient graves are found both on the Southern Arabian coast and in the Somali country in Africa. This discovery has had the effect of disconnecting the cairns, and other Druidical remains of the Nilgherries from the Tudas, almost as completely as from any other Dravidian race or tribe that now exists ; and the question of the origin of the relics which have been discovered in such numbers not only in the Nilgherries, but in many other parts of India, and in the plains as well as on the mountains, and also the ulterior question of the relationship and history of the people of whom these relics are the only monuments that remain, have now become problems of a more general and of a deeply interesting character. Captain Meadows T aylor has discovered and examined a large number of these remains at Rajan Koloor, in Sorapoor, and also at Siwarji, near Ferozabad, on the Bhima; and has devoted much attention to the comparison of them with similar remains found in England. He calls them ¢Scytho-Celtic, or ‘Scytho-Druidical.’ It is probably correct to regard them as Druidical; but they are not on this account necessarily Celtic, for the practice of rites of a Druidical character and the use of cairns and barrows were not con- fined to the Celts, but appear to have prevailed also amongst the Finns, the Euraskians, and the other Scythians by whom Europe was inha- bited prior to the arrival of the Celtic race ; and traces of the same system of religion and sepulture have been discovered in various parts of Northern and Central Asia. The other term, ¢ Scytho-Druidical,’ seems an unobjectionable one. It is a remarkable circumstance that no class of Hindus know any- thing of the race to which these Druidical remains belonged, and that neither in Sanscrit literature nor in that of the Dravidian languages 3 is there any tradition on the subject. The Tamil people generally call the cairns by the name of ‘pandu-kuris.” ‘kuri’ meaus a pit or grave, % S i ANCIENT RELIGION OF THE DRAVIDIANS. 527 and ‘ pandu’* denotes anything connected with the Pandus,’ or Pandava brothers, to whom, all over India, ancient mysterious structures are generally attributed. To call anything ‘a work of the Pandus’ is equivalent to terming it ¢ Cyclopean’ in Greece, ‘a work of the Picts’ in Scotland, or ‘a work of Nimrod’ in Asiatic Turkey ; and it means only that the structure to which the name is applied was erected in some remote age, by a people of whom nothing is now known. When the Tamil people are asked ‘by whom were these pandu-kuris built and used ¥ they sometimes reply, by the people who lived here long ago ;' but they are unable to tell whether those people were their own ancestors or a foreign race, and also when and why those ¢kuris’ ceased to be used. The answer which is sometimes given is that the people who built the cairns were ‘a raceof dwarfs who lived long ago, and who were only a span or a cubit high, but were possessed of the strength of giants.’ The supposition that the builders of the cairns had settled in India earlier than the Dravidians, and were expelled by the Dravidians from the plains and forced to take refuge in the hills and jungles, where they gradually died out, would accord with some of the circumstances now mentioned ; but it is inconsistent with the proofs which we meet with of the civilization of the race, and in particular with their acquaint- ance with the art of glazing pottery, an art which is unknown to the modern Hindoos themselves. If we should suppose, on the other hand, that they were a race of nomadic ¢ Scytho-Druidical * shepherds, who wandered into India, affer it was peopled and settled (perhaps about the Christian era), and then wandered out again, the circum- stance that the Druidical remains are found most plentifully in remote mountainous regions renders this supposition an improbable one. The improbability of the supposition would, however, be diminished, if we were to suppose that this shepherd people, instead of retracing their steps and wandering out of India, formed alliances with the Dravidians, and gradually merged in the mass of the Dravidian race. Whether the people to whom these Scytho-Druidical remains belonged were or were not Dravidians (a point which cannot be settled till we know something more of them), it cannot be regarded as pro- * This word ‘Pandu’ is not to be confou nded (as Captain Congreve has confounded it) with ‘Pandi’ or ‘Pindiya,’ the name of the ancient dynasty of Madura. Possibly both words may be "derived from the same etymon; but his- Pamil scholars derive ¢ P4 torically they are unconnected and independent. It may be added also that some g ndi,” the title of the Madura dynasty, not from the Sanserit ¢ Pandu, but from ‘P¥ndu,’ a Tamil word, signifying ancient, from the ultimate root par-a,” old. e e }?(-y 528 APPENDIX. bable that their religious usages _and rights of sepulture had their origin in India. The resemblance of the barrows and t heir conten’gs (with the crom- lechs, &c.) to the Druidical remains which are discovered in the ancient seats of the Celtic race in Europe, is too exact and remarkable to be accounted for on any other supposition than that of their derivation from the same origin. Hence the people by whom Druidical rites were introduced into India must have brought them with them from Central Asia ; and this favours the conclusion that they must have entered India at a very early period—a period perhaps as early as the intro- duction of Druidical rites into Europe. On this supposition it is necessary to suppose that they kept themselves separate from the various races that entered India subsequently, and that they imitated the civilization of the newer immigrants without abandoning their own peculiarities. Tt remains, however, as great a mystery as ever that those people have everywhere disappeared, and that not even a tradition of their existence survives. On a review of the various particulars which have been mentioned above respecting the religious usages of the Un-f&rya»nised Dréavidians, including the Kus and the Tudas, and also the unknown race that practised Druidical rites, it seems unquestionable that the majority of the ancient Dravidian inhabitants of India were demonolaters or Sha- manites, like the majority of the ancient Scythian tribes of Upper Asia, whilst it also seems probable that there existed amongst some Dravidian tribes a strong under-current of Indo-European, and possibly of Druidical tendencies.’ This result exactly accords with the supposition which has already been deduced from lingual comparison respecting the relationship or affiliation of the Dravidian race, viz., that in basis and origin it is dis- tinctively Scythian, with a small but very ancient admixture of an Indo-European element. PRINTED BY HARRISO-N AND SONS, ST, MA#TIN’S LANE, PR N S m - R i s — g e ~ — e o ,,r = e S — aww | 1 4 '3 3 % . ..n . i & 5 Sy " 4 o s s - o VA i 24 o e i % gy i A L S a3 e ' - o 5 s e P i o ” /3] g % e N // e /ff o { % S A Al Wr LS e ..wuf . &:MW Mooy, oy e he s Pk i, o) S g{t Biiin e XN M i i v Z:vaff. i g P A e (T3 R = et R, i S el (Ul % Friow wffsr £ 55, b G bl faty i, ) e Z ~ S g e g s Gt o ) F‘g é}g gqm Sk m b ).}!rxr = i i a2 > Yo Ve i e Vo e gyl a.pas.,ﬂ..k?r Sy, e P S ey % S o g st v g{ g g, ? R i 5% g S e fimse i Do o1 f o Gt | S ot o g o i ot AN o S i}»ﬁ b v b oy o, o B iy Gl S ¥ St - e .‘ b .!@m”& o . o i o bias gy T8 S e e lgff ghe e e oy R ey = 3 it erinny i b Wi e o ?M S i o) i % o s o ¥ Sk S & & i e i ey I'10 TR i ] ] ‘AN N Focus (0] A n Balance 1 2 3 4 5 6 7 8 0 11 12 13 i Y y4 VierFarbSelector Standard * - Euroskala Offset ‘;‘ Copyright 4/1999 YxyMaster GmbH www.yxymaster.com‘ =3 [ =) Y = P D THE ; IELLY, SOUTHERN INDIA % LDWELL, BA, PALL M A I E GRAMMAR FATION OF THE GOSPEL IN FOREIGN PARTS, AT ILY OF LANGUAGES.