kup_15 Provided by University Library of Tübingen Transcribed with Tesseract

This work is protected by copyright or related property rights but available in Open Access. The right of use and especially the right to reproduction is only granted within the legal limits of copyright law or due to the consent of the copyright holder.

 Improving criminal justice statistics, 15 Improving Criminal Justice Statistics National and International Perspectives Edited by Jorg-Martin Jehle and Chris Lewis Wiesbaden 1995 Die Deutsche Bibliothek - CIP-Einheitsaufnahme Improving Criminal Justice Statistics : National and International Perspectives / [Kriminologische Zentralstelle e.V.]. Ed. by Jorg-Martin Jehle and Chris Lewis. - Wiesbaden: KrimZ, 1995 (Kriminologie und Praxis ; Vol. 15) ISBN 3-926371-24-2 NE: Jehle, Jorg-Martin [Hrsg.]; Kriminologische Zentralstelle < Wiesbaden>; GT “"74 & it w (L noee® & 7oy © Eigenverlag Kriminologische Zentralstelle €.V KRIMINOLOGISCHE KrimZ ZENTRALSTELLE E.V. Adolfsallee 32, 65185 Wiesbaden Alle Rechte vorbehalten Printed in Germany ISBN 3-926371-24-2 Preface Since the beginnings of criminal statistics in the early 19th century their use as instruments for measuring the true amount of crime has been controversial. Nevertheless, they are still important up to the present day. These data are used to justify policies, to decide on measures of crime prevention and crime control and to make statements about the working of the criminal justice system. In spite of all methodological doubts, these figures are still referred to in criminology, as there are no real alternatives to them. Even victim (or crime) surveys, originally intended to replace recorded crime statistics, can only complement them; this is because - accor- ding to latest criminological thinking - such surveys measure different concepts from these statistics. Since criminal statistics have maintained or retained their relevance, it is all the more important to face and attempt to solve all those methodological problems brought up since the days of Quetelet and other pioneers in the field of criminal statistics. If these observations ar¢’already true for the respective national situa- tion, to which criminal statistics are normally related, they are all the more valid when one tries to compare the criminal statistics of different coun- tries. With the world closing up ("global village") and Europe growing together, not only economic or social comparisons are needed, but there also exists a demand to compare the developments of criminality as well as the criminal policies and the criminal justice systems of different countries. This was the starting point for a workshop organized by the editors and titled "Improving Crime and Criminal Justice Statistics - National and International Perspectives", which took place on the 11th International Congress on Criminology in Budapest in August 1993. In order to prepare the workshop the editors asked the contributors to answer the following questions in regard to their respective national situation: Are there any statistics covering the whole area of police and criminal justice in the respective countries? Are there any partial statistics, and if so, how are they connected? Are they compatible? What can be said about reliability and value of data of statistics? What will be the future developments regarding survey modes and computerization of data collection and evaluation? How can the availa- bility of the data be improved? How far can official statistics in future be replaced or complemented by victim surveys and microcensus? Is it possible to establish international comparability/compatability of national statistics? This volume contains most of the papers presented at the conference and revised for this publication as well as some complementary contribu- tions by other authors. After an introductory chapter, which describes the initial position and the present difficulties of the national criminal statistics and of their inter- national comparability, the situation in several West and East European countries is shown. This is supplemented by reports from Africa as an example of developing countries and from Canada as an example of a non- European developed country. A concluding chapter discusses problems of international comparability and of concrete initiatives to carry out such comparisons. The editors would like to thank those responsible for the execution of the 11th International Congress for Criminology, especially the head of the organization committee, Professor Katalin Gonczol, and the head of the scientific commission, Professor Hans-Jiirgen Kerner. But they would like to express their gratitude primarily to the speakers of the mentioned work- shop and all the other contributors to this volume. Wiesbaden, London Jorg-Martin Jehle, Chris Lewis May 1995 Contents Preface (1 OLONO QN0N0 UG 0RO 0N NG OR0 IGR O GRORORGE QR0 DL 0S0N0 00 O GrOR0Y 0RO OROEON QU U S0 0 D Introduction Improving Criminal Justice Statistics - 11 Needs and Prospects I T I R A I Jorg-Martin Jehle I1 Crime and Criminal Justice Statistics in Different Countries Danish Criminal Justice Statistics G40 OM0 O O CLUN 0RO 0 GG 0RO 100 O LT 31 Paul Henning Larsen Statistics of Crime and Criminal Justice in England and Wales G000 ORORO O IORGED O 0R0 GG EORON 0N G USR0S0 URQ UG G 35 Gordon Barclay CROY O G G0 O 0 47 Crime and Criminal Justice Statistics in Germany Richard Blath Greek Crime Statistics CRUY OO0 AORO O SOI0 SOr RN RO O OO 0P ORD Or GO U O 67 Calliope D. Spinellis and Maria Kranidioti Brief Review on the Situation of the ST el B1il6 6116 98] 6§ 4560 8) (6 ) o6 ekt ertoONIY) 89 Hungarian Crime Statistics Sdandor Erdosi Statistical Data Collections on Crime and Criminal Justice in Italy 109 Armando Caputo Crime and Criminal Justice Statistics in the Netherlands 44 a0 60 640588, S 0 8 &G TSHEE 117 Max M. Kommer Criminal Statistics in Poland 00 000 00000000000 000000 127 Jerzy Jasinski 00 000 d0 00000000 4d000000 135 Criminal Statistics in Portugal J. de Seabra Lopes 10 The Swedish System of Criminal Statistics The Scope of the Statistics © o 4 o s s s s s s s s e 6 s s s s e 0 o e s e e 141 Torbjormn Israelsson 11 An Example from the Developing Countries Criminal Justice Statistics in Africa 147 N. Masamba Sita 12 An Example from Overseas: Canada’s Justice Statistics Program 155 Sange de Silva 111 International Efforts and Prospects International Studies and Statistics on Crime and Criminal Justice 000000000000 00000000 167 Chris Lewis International Comparison of Crime and Criminal Justice Statistics 177 Max M. Kommer International Prospects for D T T AR 193 Statistics of Criminality Jean Gratia The European Criminal Justice Sourcebook Project GO0 0 0000000000000 00000000 203 Martin Killias List of Contributors OLOTO 0 00 0000 OO 000 000000000 Q0000 GG 213 I. Introduction e pa ST % §e 94"(35‘ 3 s T ‘i’&?x n ot o e I i o 4 i Y a0 o i 2 Ty e W i i 3 B i . o i i i s B o % 3 o a5 .‘.%;.’ iy /0 S £ & ! i ik e o i ipg) o s Ry v i o & oy Al b i i ek £ 2] e z B A % T &y e ! ¥ - R & e A5 i‘f{; o i o it i £ e i e 5 & i5e o e i e o i A g . il 5 Bk e ar . i R 3 pOs e D gty i b 23 i 1o g B e B e i B R 5 R ey I ] sy w G - MR i .4;;'1& s & 0 = Slaing s s ¥ . i - 7 4 e i i I X Y 5 e 2 i 3 o il x IR o s ey b a5 {,’j, i i e ot o £ ik s 'e e b i i 0 E0 A 6 v el gk iy e o b ‘%fl& R by e . B o 2 o e o 4 g Loy e i e L 5 o ot :*’ E i 7% i i e e o i gl ) @*" i 3 il G 3y, ::‘ji b % b o o A% s & {a e o 4 iy n’l o o f i & Y o g o il 2 o . ¥ i s 15 o & it I A i P o b o i s o 5 o o P g e & o> ) I e & o T i Aok Ty o e 5 & rh; e b e ;- . a i s % H s i 7 ;‘Q‘ o e T 3 e e A3 e o 5 B 25 o = L % e o o ad 2 ?s . ¥ 5 N o 5 k) {d o bt & i v o, b s e ¢ e &t i o i o o L 2} o ¥ o Es 1 g g 53 1 e s % Ey :a!" L) it gy vt g1 : e # e 1 i < 2 o e . e kY 5l kgt i i 5, b2 e :‘ B 1 e e g ’5’&‘ % o S N g o = % %‘ 2t fl oy T 5 (s ) : 7}5.'% Sk ol o ot 6 = el 3 it % ; i i G o e ey g & i o . & £ ) o 3 i b S o ot i s o ? (e i i ) " i fa i o cy’r A | e o - b 5 o (N 4 b 7 b e S 5 Gy & by i i3 o o ke i o o P ac % i ey A i . 4 o i o 7 i e bl i el b b0 i &5‘ it .i/‘» i i o i3l ey R o s o o 7 Improving Criminal Justice Statistics - Needs and Prospects Jorg-Martin Jehle I. Do we need Official Data Collections of the Criminal Justice System for Purposes of Research and Policy? As in all fields of state affairs, the prosecution of criminal offences and the execution of sentences have to be planned, personnel and material resources have to be provided and organisation measures have to be taken. Thus it is clear that the police and the criminal justice agencies need systematically to collect data for purposes of organization and planning. But ever since the creation of official criminal statistics in the early 19th century other objectives than these purely internal ones have been aimed at. The statistics enable people to make statements about the reality of crime and once there also existed the idea that the figures should indicate the "general moral health of a nation". Today expectations are not that high-pitched any more. But still now official statistics are used for the interests of criminal law policy and criminology. However, it is by now undisputed that criminal statistics offer no true picture of the reality of crime and its prosecution, but can only provide a framework of selective data. Insofar there are also no doubts that the figures cannot speak for themselves. They rather need careful interpretation, in which it is considered how, by whom and for what purpose the data are collected, how they are categorized and in which form they are represented. 1 Zacharii 1832, quoted by Heinz, W.: Die Deutsche Kriminalstatistik. Uberblick iiber ihre Entwicklung und ihren gegenwirtigen Stand; in: Kriminalstatistik, vol. 5 of the BKA- Bibliographienreihe, Wiesbaden 1990, p. 7. 12 Jorg-Martin Jehle Yet restrictions of the validity of the statistics are not only the result of these inherent limits, but in addition are due to the fact that already the collecting of the data is not always sufficient or reliable. This is true for the crime statistics of the police as well as of the personal and other statistics of the courts and public prosecutors. Should we therefore do better without collecting data, because officially collected data are completely unusable, as some authors® seem to suggest? The answer to this question is clearly no, as there are no alternative sources of data which can offer comparable information. Modern criminology has created methods like population surveys, the so-called victim or crime surveys, qualitative interviews and the analysis of criminal records and files. They all supply information about the actual extent of criminality and victimization and about the treatment of criminal offences and offenders by the police and criminal justice. These pieces of information are different and partly more exact and detailed than those in the criminal statistics®. Nevertheless, with regard to an overall picture of the official control measures of criminality these special instruments of criminology can only supplement, but never replace official criminal statistics. In spite of all their obvious shortcomings and one’s reasonable doubts one still can say that the official statistics in the fields of the police and criminal justice are indispensable for the criminal law policy, administration of criminal justice and for criminology. However, they have to be improved and moreover to be supplemented by other scientific survey methods such as crime surveys and the analysis of criminal records and files. These methodological observations refer to the national situation where improvements of the criminal statistics have to be undertaken (s. below IL.). 2 E.g. Pfeiffer/Strobl; on the present German debate see further Kriminologie und Praxis, vol. 8; Jehle, J.-M.: Pladoyer fiir bessere Kriminalstatistiken; Neue Kriminalpolitik 1994, pp- 22 ff. On an international level see the outstanding study of van Dijk/Mayhew/Killias: Experien- ces of Crime across the World. Key Findings of the 1989 International Crime Survey; Deventer, Boston 1990; in Germany surveys about victimization and fear of crime have been carried out in recent years; please see the respective studies published in Kaiser, G.; Jehle, J.-M. (ed.): Kriminologische Opferforschung. Neue Perspektiven und Erkenntnisse. vol. 102 of the Neue Kriminologische Schriftenreihe; Heidelberg; Teilband 1 1994; Teil- band II 1995. 13 Improving Criminal Justice Statistics - Needs and Prospects But these observations are also valid when one tries to compare the data of the criminal statistics of different countries. Therefore international comparisons are very complicated enterprises which need expert knowledge of the national peculiarities and international difficulties (s. below III). II. Reorganizing Statistics in the Field of Criminal Justice It is already stated above (I) that the official criminal statistics need to be improved. In the following it will be discussed in which direction the official criminal statistics should be developed further with the emphasis lying on the personal statistics of criminal justice. A model (see page 28 below) about the statistical recording in the course of public prosecution, of sentencing and of execution of the sentence gives an idea at what stage and by which agency data are collected and used for statistics and other purposes. The model refers to the German situation, but could be found in other countries in a similar way. The same is true for the weaknesses and shortcomings of the different statistics. 1. Purpose of the Statistics in the Field of Criminal Justice The purpose of the whole of the statistical data collections of the police and the criminal justice system is to record the essential elements of criminal prosecution, sentencing and execution of sentence in such a way that a comprehensive picture of the crime control by the state can be provided. The available statistics do not give this information: they either offer too many or not enough data. The aim, to receive a comprehensive picture of the crime control by the state, remains an ideal concept. It is clear that for theoretical and practical reasons cuts have to be made and therefore only extracts of the reality can be represented. However, the essential directions which set the course should be recorded, and especially the flow of proceedings on different levels should be registered - from the penal charge to the investigation by the public prosecution and from the decision of the court to the execution 14 Jorg-Martin Jehle of sentence and the discharge. Ultimately homogeneous statistics about the proceedings are needed or at least a comprehensive system of integrated partial statistics, which are harmonized in their recording procedures, is required.’ The present statistics in most of the countries are far from this. It is true that now statistical recording takes place during all stages of the proceedings (s. model) and altogether represents plenty of figures, however, in some parts there are important gaps, and the compatibility of the data is missing. In the following the crucial points will be listed: The survey of the criminal police statistics ends with handing over the proceedings to the public prosecution; any attempt to pursue and identify the cases on the next higher level is impossible in most of the countries.’ The statistics of the public prosecution and the statistics of the trials often refer to proceedings, which are not person-related and do not permit the connection of offence and type of disposal. The statistics of convictions start at court level, and do mostly not record the earlier disposals of cases and cases of pre-trial detention, which do not reach a sentencing; at the same time the actual sanctions, the conditions and instructions regarding court discharge and probation are not sufficiently differentiated. Data referring to the field of the execution of sentence are only partially recorded: for example by means of the probation service statistics and the prison statistics. The latter have the great disadvantage to be usually organized in reference to a fixed date. 4 Homogeneous statistics for police and criminal justice can be found only in a few coun- tries, such as Canada and Denmark (s. de Silva: An Example from Overseas: Canada’s Criminal Justice Statistics, Justice Statistics Program, pp. 155 ff., and Larsen: Danis pp. 31 ff.). Even compatible statistics, like in England/Wales (s. Barclay. Statistics of Crime and Criminal Justice. England and Wales, 1994, pp. 35 ff.) and Portugal (s. Lopes: Criminal Statistics in Portugal, pp. 135 ff.), are an exception. Only in England/Wales (s. Barclay. Statistics of Crime and Criminal Justice. England and Wales, 1994, . 35 ff.) and Denmark (s. Larsen: Danish Criminal Justice Statistics, pp- 31 ff.) the flow of individual cases through the whole course of criminal justice can be followed; there specific index numbers help to trace the cases. Improving Criminal Justice Statistics - Needs and Prospects 15 In some of the countries there exists a Central Register of criminal records which represents a further additional data collection and theoretically could also be used for statistical purposes. This instrument records mainly sentences and partly the person-related execution of sentence or the discharge. As the entries are kept for a while at least, the Central Register can especially be used for studies on reconvictions. In the field of criminal justice of most countries there exists either an excess or a lack of data. An excess, as partly the same data are recorded several times for different statistical purposes; a lack, as the statistics have gaps and are not quite compatible with each other. 2. Customer-oriented Presentation The stages and the data to be recorded depend on the needs of the different users which must be considered. The same is true for access to and publication of data from criminal statistics. In realizing the general objectives (see above 1) it is necessary to have a closer look at the needs of the potential data users. Research has surely the most comprehensive interest in recording statistical data. However, the researchers’ needs cannot determine the object of statistical recording, because even a very good statistical recording will never fully satisfy the scientific needs; on the contrary there will always remain an additional need for independent empirical research. Further- more, it is hardly predictable, what will become a future research object; rescarch profits most, if the data recorded are comprised and aggregated as little as possible in order to avoid restricting their message. The latter is principally also valid for criminal law policy. Everything in the process of criminal prosecution, sentencing and execution of sentence, which is subject to legal provisions, should be of interest to criminal law policy. It is evident that the total amount of possibly interesting data cannot always be present in the statistics to be published; here also exists a need 16 Jorg-Martin Jehle to provide access to the raw data, i.e. the data sets on the basis of individual cases, in order to touch relevant policy questions. In view of the necessity to obtain up-to-date information there is the special problem that statistics are usually updated only once a year and that publication is often considerably delayed.’ The principal interests of the administration agencies of justice are probably directed towards (questions of) organization and planning of personnel. Of special interest are here the number of proceedings and cases and data concerning the structure, personnel situation and costs of criminal justice’. The judges and public prosecutors, on the contrary, are more interested in person-related data concerning sentencing. Finally the interest of the public must be considered. For these purposes the current statistics of criminal justice offer too many detailed data; at the same time there is a lack of easily understandable presentations of data in form of tables, charts, comparisons and brief explanations®. The interests and needs of the different data users vary, they have to be taken into consideration and dealt with accordingly. Therefore the statistics in the field of the criminal justice system must on the one hand be published in a short version for a general public with additional comments and tables. On the other hand detailed material must be made available to an expert public. This can be done either conventionally on paper or - better - on 6 A possible solution could be publication at shorter intervals, for instance quarterly or half-yearly, as it is done in Denmark (see Larsen: Danish Criminal Justice Statistics, 31 ff.) or the Netherlands (see Kommer: Crime and Criminal Justice Statistics in the Ret herlands, pp. 117 ff.). Another alternative would be a possibility for certain authorities or agencies to have quick, short-term access to the data as it is practised in Hungary (see Erdosi, Brief Review on the Situation of the Hungarian Crime Statistics, pp. 89 ff.). In most of the countries data on costs and personnel are internally collected by the administration; sometimes they are published, for example in England/Wales (see Barclay. Statistics of Crime and Criminal Justice. England and Wales, 1994, pp. 35 ff.). Attempts to meet this need in a modern way have been made in several countries. In an exemplary way the Home Office in London presents its Digest "Information on the Criminal Justice System in England and Wales" (see Barclay: Statistics of Crime and Criminal Justice. ngland and Wales, 1994, pp. 35 ff.). 17 Improving Criminal Justice Statistics - Needs and Prospects electronical data media.’ To enable special evaluations for criminal law policy, administration of justice and science, one must provide access to electronical data sets on the basis of individual cases. 3. Organization of Data Collection and Recording Up to now the data collection for the different statistical purposes has not been coordinated and controlled enough and represented too little of the interests of the recording bodies. Therefore these data are not of sufficient reliability. All statistics are only as good as they are reliable in their recording of survey criteria. This is generally valid; so error rates can be stated in all areas of state affairs. Especially the data in the field of criminal justice are not always reliable, as criminological research has often shown. The sources of errors and error rates are different; only a few examples can be mentioned here, which criminological research in Germany - either based on a files analysis or by comparison of data from the Federal Central Register of Criminal Records with other statistics - has revealed." - A very general source of errors is the lacking accuracy of the officials when filling in the registration card, which might be due to their lacking motivation and lacking control. This problem will not change if the recording is done electronically. The criteria to be recorded are not easily ascertained and are therefore not or not accurately filled in, as for example information about prior convictions, victims, additional sentencing directives. In the Netherlands for example besides the "statistical yearbook" special statistics can be drawn up for scientific or political purposes. Additionally, "production statistics" and the data from the police statistics are available on flo mer. Crime and Criminal Justice Statistics in the y disks for certain institutions (s. Kom- etherlands, pp. 117 ff.). In Sweden, data are already available on disk and in computer tables, while the use of so-called "public use files" with de-identified, reduced data is discussed (s. Israclsson: The Swedish System of Criminal Statistics. The Scope of the Statistics, pp. 141 ff.) 10 See Jehle, J.-M.: Plidoyer fiir bessere Kriminalstatistiken; Neue Kriminalpolitik 1994, pp. 22 ff. 18 Jorg-Martin Jehle The instructions how to fill in the cards and the presentation of the various registration cards are too different and partly too complicated, or they allow different ways of handling (example: "most serious’ offence when offences have equally serious abstract ranges of punishment). These deficiencies can be largely overcome if the unconnected data recording is replaced by a coordinated, if possible homogeneous form, if recording instructions are as easy as possible, if formal data can be easily recorded and if the recording authorities themselves are interested in the recording." The collecting and recording of data must be reorganized; to avoid multiple recording and to increase the reliability, a homogeneous data concept should be established for all stages of the criminal justice process where data have to be recorded. This concept would establish compatibility of data of the different phases in the proceedings and at the same time facilitate data collection and recording. This requires that in the near future the recording bodies will exchange the conventional registration on paper for a computer system which enables a computerized data processing from the very beginning of the data inpu t'IZ The ideal concept would consist of a homogeneous data matrix, in which from stage to stage the new information is added and from which all the respective data to be recorded can be retrieved. 11 In Portugal, for example, "training sessions" are held to motivate those responsible for the data collection and to emphasize the necessity of reliable data (see Lopes: Criminal Statistics in Portugal, pp. 135 ff.). 12 In most of the countries electronic data & a rocessing (EDP) is available for the central office of statistics which aggregates the ta and produces the statistics. But in some countries efforts have been already made to use EDP during the whole process of bringing about the statistics, starting with the data input and ending with the publication (for the Netherlands s. Kommer: Crime and Criminal Justice Statistics in the Netherlands, p. 117 ff., for England/Wales s. Barclay. Statistics of Crime and Criminal Justice. Eng- land and Wales, 1994, pp. 35 ff.; for Hungary s. Erddsi: Brief Review on the Situation of the Hungarian Crime Statistics, pp. 89 ff.; for Sweden s. Israelsson: The Swedish System of Criminal Statistics. The Scope of the Statistics, pp. 141 ff.). Improving Criminal Justice Statistics - Needs and Prospects 19 This would very much simplify the transmission of data for statistics and criminal records purposes. From the user’s point of view it is especially important that the data sets on the basis of individual cases will be maintained in a pool for some time, so that, independent of the limitations resulting from categorization and aggregation in the respective statistics, special evaluations with whatever data connections will be possible. 4. Levels of Recording, Partial Statistics, Categorization Even with a homogeneous data concept it remains necessary to produce different, though integrated partial statistics, referring to the respective stages of the proceedings. Person-related process statistics, which have been demanded by criminologists for a long time already, can only partially be realized. Reasons for this are the change of the subject in the course of the proceedings and the participation of various recording bodies that need different information. Assuming that one survey unit represents the proceedings against one special person, it is not guaranteed, that the cases to be recorded are identical at each stage of the proceedings. The legal appraisal, detachment and (partial) discharge of special offences can change the earlier suspicion. And proceedings can be connected with others or detached from others. The sanctions imposed in earlier proceedings can be taken into consideration in a later decision. Finally the execution of several sanctions can be combined. It is clear that from stage to stage the survey unit and therefore also the number of cases to be recorded may change. To represent all these possible changes, detachments and combinations in uniform person-related or proceedings-related statistics is probably very difficult to realize. Therefore all the person-related data sets on the basis of individual cases, which were collected at all stages of the proceedings, should be gathered in one pool to provide for special process statistical analyses. 20 Jorg-Martin Jehle Besides that, (integrated and compatible) partial statistics have to be produced for each of the different stages of criminal proceedings, this being the police, public prosecutor, courts and the corrections system. They should basically record the flow of cases and refer to the proceedings concerning individual persons as survey unit. This concept comprises the fields, which today are already largely covered by the current statistics of police and criminal justice. However, in most of the countries the current statistics are not harmonized and lack compatibility (see above II. 1.). Especially here exist obstacles to harmonizing data recorded by the police on the one hand and by criminal justice agencies on the other hand. But these obstacles can be overcome. As far as uniform statistics or compatible partial statistics for police and criminal justice” do not exist, at least efforts should be made in order to establish common categories for both fields'. That means, the characteristics of the persons, the offences and judicial decisions should be recorded in the same way. Basically, only formal, easily recordable characteristics shall be registered. In every phase of the proceedings (and therefore for all partial statistics) the formal personal criteria, the offences committed and the decision, which closed the respective stages of the proceedings, are to be recorded. The committed offences should be completely recorded and not be restricted to the most severe offence as it has mostly been practised up to now". This results in a systematic underestimation of petty offences (for 13 This is the case e.g. in Canada, Denmark and England/Wales (see fn. 4). 14 In Portugal (see Lopes: Criminal Statistics in Portugal, pp. 135 ff.) criminological catego- ries have been established which can be transformed into penal law terms and vice versa; in Germany (see Blath: Crime and Criminal Justice Statistics in Germany, pp. 47 ff.) only harmonizing the different partial statistics in the field of criminal justice - without police - is aimed at in the near future. 15 For the situation in Germany see Blath: Crime and Criminal Justice Statistics in Germa- ny, pp. 47 ff. Improving Criminal Justice Statistics - Needs and Prospects 21 example driving without a licence). This suggestion makes it easier for the officials who do the recording, because they will no longer have to choose between different offences; however, for the organization of the statistics and the presentation of the tables or charts new concepts have to be developed. 5. Complementary research The statistics of the criminal justice system can only provide a rough framework. For detailed analysis regarding characteristics of victims and offenders, future convictions and new convictions, sentencing by the courts, and the intricate modalities of the criminal prosecution and the execution of sentence further sources of information, especially from scientific studies, are required. Already for reasons of work load and reliability of the data the recording for statistical purposes cannot be based on thorough and time- consuming activities of gathering information; moreover the data which are at the official’s disposal are limited. Therefore, for detailed information additional sources of information have to be used: Regarding reconvictions and prior convictions, for example criminal records, files analyses, regarding sentencing and the partly complicated execution modalities and specific victim- and offender-criteria regarding primary surveys. In these fields criminological research has to become active, though it still depends on the access to the official documents. The most important alternative of measuring criminality has become the so-called victim or crime surveys. In the United States of America they have been established for more than 20 years but only a few other countries 22 Jorg-Martin Jehle carry out crime surveys on a regular basis.'” Despite some specific methodological problems crime surveys can fill some of the gaps of the official statistics and thus lead to a more adequate evaluation of the statistical figures. They have proved an indispensible, though expensive instrument which complements the official crime and criminal justice statistics. Furthermore they can be used for international comparisons more easily than official statistics because international standardization especially regarding the offence definitions can be reached.” III. International Comparisons Attempts to compare data of national criminal justice statistics are very complicated enterprises. Therefore only few international comparisons have been carried out, mainly by international organisations, i.e. Interpol, the United Nations and the Council of Europe."® Whereas Interpol collects information restricted on crime recorded by the police, the United Nations Crime Surveys and the activities of the Council of Europe comprehend crime and criminal justice data. The latest UN Crime Survey is the fourth survey of crime trends and operation of criminal justice system, which deals with the years 1986 to 1990." The 16 E.g. in Canada (s. de Silva: An Example from Overseas: Canada’s Justice Statistics Program, pp. 155 ff.) crime surveys are carried out every 5 years, in the Netherlands (s. Kommer. Crime and Criminal Justice Statistics in the Netherlands, pp. 117 ff.) and England/Wales (s. Barclay. Statistics of Crime and Criminal Justice. England and Wales, 1994, pp. 35 ff.) even every 2 years. In the framework of an international project in some other uropean and Non-European countries crime surveys were carried out (van Dijk/Mayhew/Killias, fn. 3). In Germany after the unification different crime surveys and victimization studies took place (s. Kaiser/Jehle, fn. 3). 17 See van Dijk/Mayhew/Killias, fn. 3, especially The International Victimization Survey Questionnaire, pp. 153 ff. 18 See Lewis: International Studies and Statistics on Crime and Criminal Justice, pp. 167 ff. 19 A first description of the design and the methodological difficulties is given by Joutsen, M.: The fourth United Nations Survey of Crime Trends and O erations of Criminal Justice System (1986 - 1990); in: European Journal on Criminal Po| R cy and Research; 2-4; . 117 {f.; see also the comments by Roger Hood: Crime Trends and Criminal Justice in Europe, and by J6rg-Martin Jehle: Comparing Data of National Criminal Justice Statis- tics, papers held on the Fourth European Criminological Colloquium in Romain Motier Lausanne; 14-16 September 1994. Improving Criminal Justice Statistics - Needs and Prospects 23 Council of Europe has concentrated in former years on the prison statis- tics®; recently it has started an ambitious project, called the European Criminal Justice Sourcebook?. If one analyzes such ambitious internationally comparing projects like the UN Survey or the European Sourcebook, one has to differentiate between two different kinds of problems: (M Which problems do exist on a national level? @) Which problems come up with an international comparison? 1. Problems on the National Level It is clear that the state of the national statistics also impedes the international comparison, for in all cases the data collected have to be derived from given national sources. The first problems are the deficiencies concerning completeness and reliability of the statistical data. One factor is the bureaucratic inertia, which Matti Joutsen has also discoverd in the UN survey”. The motivation to record the data is not very high among the officials of the police and criminal justice, if their office has no own interest in the data or if the recording is not controlled. On the other hand also too much can be recorded, for instance when the statistics are connected with a proof of performance of the recording persons (this is the case for the German police). The second important problem, that also concerns the international comparability, is the lacking compatibility of the data on the different stages from the police treatment of the cases to the judicial proceedings. Different persons from different authorities record data for different statistics at 20 See the analysis of these data by Tournier, P.: The Custodial Crisis in Europe: inflated Prison Populations and Possible Alternatives; in: European Journal on Criminal Policy and Research; 2-4; 1994, pp. 89 ff. 21 The project is described by the chairman of an experts group which has prepared a draft model, Martin Killias: The European Criminal Justice Sourcebook Project, pp. 203 ff. 22 S. Joutsen (fn. 19) pp. 120 ff. 24 Jorg-Martin Jehle different times and different stages of the proceedings (s. above II. 1.). The only solution here is to urge on the national level that the collection and categorization of the data for the different statistics will be harmonized. In the long term a homogeneous (electronic) data entry form, in which the new data are filled in from stage to stage, should be aimed at. This would also establish the basis for process statistics, which has been demanded by criminologists for a long time already. International surveys which at the same time show the gaps and faults of the respective national data may on the other hand make an impact on the national level and give impulses for improvements there. 2. Specific Problems with an International Comparison Differences between the countries occur on all levels: of legal definitions, of the course of criminal prosecutions, and of the institutions involved, of the penal sanctions, and of the way all these data are statistically collected™. To give an example on the level of police recording: Even if the offence definitions would appear to be more or less comparable between the countries, comparisons can be seriously impaired by deviating statistical rules applicated in the different countries. Questions arise, such as: At what point of time will an offence be registered? What is the unit used in the statistics: the criminal act, the object of the criminal act, the victim, etc.? What happens in case an act violates different sections of law? How are serial offences counted? It is important not to attempt a comparison of the whole of the criminality in the different countries, which is always widely different in its definition, but to refer to some quite differentiated selected offences, as it is done for the European Sourcebook and the UN Survey. As favourable this might be, it also has disadvantages, because the selected offences cannot always be differentiated on all levels of the criminal justice system. Though, the basic problems of comparability still exist even for the selected offences. To give an example which also Matti Joutsen mentioned: 23 S. Joutsen (fn. 19); Kommer: International Comparison of Crime and Criminal Justice Statistics, pp. 177 ff.; Gratia: International Prospects of Criminality, pp. 193 ff. Improving Criminal Justice Statistics - Needs and Prospects 25 assault. There is no consistent legal concept even for this offence. While assault is understood in common law countries as physical attack, the element of physical injury is decisive in other countries e.g. in Germany. (By the way, this difference is not sufficiently considered in the victim surveys; even in German studies "physical attack" is asked for although there exists no analogous actus reus.) Further differences in the treatment of this offence can be seen in prosecution and sentencing. Partly assault is prosecuted only upon the formal complaint of the victim and in Germany the victim can be referred to private prosecution. This process of selection naturally has consequences for the sentencing, too. At the same time, this example might illustrate what enormous differences do exist between the countries especially on the level of criminal prosecution. This is particularly true for the possibility of the public prosecutor to dismiss cases or to impose sanctions by himself. All these difficulties may give the impression: The differences of the criminal justice systems and the statistical recordings are so big that one can not know for sure if either the differences or the similarities in the statistical figures can be interpreted as real ones. Though, one should not give up in view of the consequences of the hardly controllable differences. Instead of this, these differences have to be thoroughly worked out and described in the publication of the statistical figures - as it is attempted in the European Sourcebook Project. 3. A Special Problem: the Changes in East Europe The last point is concerned with the consequences of the radical political changes in East Europe®. Even when the states still exist in their former geographical shape, hardly solvable statistical problems occur, especially where longitudinal comparisons are concerned”. This is valid for comparisons with the time before 1989/90, because at this time the 24 Matti Joutsen treated this problem in his paper, held on the Fourth European Criminolo- gical Colloquium in Romain Motier/Lausanne, 14-16 September 1994. 25 S. Jasinski, Criminal Statistics in Poland, pp- 127 ff.; Erd6si, Brief review on the situation of the Hungarian Crime Statistics, pp. 89 ff. 26 Jérg-Martin Jehle structure of criminal opportunities as well as the criminal prosecution and the collection of statistical data were fundamentally different. But also the first years of the nineties were extra-ordinary; it is hard to differentiate how much of the measured enormous increase of criminality is real and how much of it is the result of the fact that the police and criminal justice had to be built up again and therefore could only partially deal with the criminality and record it. Germany is especially affected by these difficulties: Because of the totally different situation in the former GDR there cannot be any comparisons of the united Germany with the situation before 1990%. And even if the comparison is restricted to West Germany before and after the unification, it is difficult to be done, as the fall of the wall and the opening of the borders to East Europe have resulted in a so far unknown influx of people of whom some might come into conflict with the law. This can be best illustrated by the example of West Berlin: After the fall of the wall, the rate of reported cases of shoplifting increased 300 %, and the rate of pretrial detentions of young offenders rose 300 % as well. This specific development in East Europe and in Germany should be gradually consolidated in the middle or at the end of the nineties. 4. Final Remark The efforts for cross-sectional comparisons especially between the different European countries are important and deserving. However, no matter how carefully the differences between the criminal justice systems and their statistical recordings are considered, a totally satisfying 27 or "a comparability cannot be achieved, as far as a "standardization uniform way of data collection"” on national and international level will not be established. And this seems as desirable as impossible, at least in the near future. 26 An overview of the consequences of the German unification gives Kaiser/Jehle, fn. 3. 27 S. Kommer. International Compatison of Crime and Criminal Justice Statistics; pp. 182 ff. 28 S. Spinellis/Kranidioti: Greek Crime Statistics; pp. 67 ff., 88. 27 Improving Criminal Justice Statistics - Needs and Prospects Therefore longitudinal comparisons are here more fruitfu 12 Independently from the individual starting levels in the different countries one can find out, for example, in which countries the criminality recorded by the police increases or decreases, how far strategies of decriminalization have consequences and how far the measure of punitiveness is increased or weakened. Then one can observe and internationally compare parallel or contrary trends and connect them with accompanying circumstances. However, cross-sectional comparisons, like they are presented by the UN Survey and the Modell European Sourcebook of Criminal Justice, have to create the preconditions for such longitudinal comparisons. 29 S. Kommer. International Comparison of Crime and Criminal Justice Statistics, 177 ff.; Gratia: International Prospects for Statistics of Criminality, pp. 193 ff.; Killias: The European Criminal Justice Sourcebook Project, pp. 203 ff. Jorg-Martin Jehle Modell of the German statistical data recording in the process of criminal prosecution, sentencing and execution of sentence (simplified) Stage of proceeding Recording body Data collection/statistics Preliminary Investigation Suspicion of a criminal offense Police Transmission to public criminal police statistics prosecution (Polizeiliche Kriminalstatistik) Pending proceedings Public prosecution statistics of the criminal proceedings (in planning) decisions statistics of the public Public prosecution (charge, suspension, prosecution (Geschiftsstatistik etc. der Staatsanwaltschaft) Interlocutory Court statistics of the triais proceedin (Geschiftsstatistik der (Zwischenverfahren) Gerichte) Main proceedings Court Conviction Public prosecution statistics of convictions (Strafverfolgungsstatistik) Sentencing Public prosecution Federal Central Register (for criminal records) Exccution of sentence probation Public prosecution Federal Central Register Court probation service statistics (Bewihrungshilfestatistik) prison sentence Public prosccution Federal Central Register Prison administration prison statistics (Strafvollzugsstatistik) remission of penalty Public prosccution Federal Central Register and discharge Reconvictions Public prosecution Federal Central Register (analysis of recidivism - in planning) II. Crime and Criminal Justice Statistics in Different Countries "’ e o o it " i A = . 0 5 S 8 ! b K A 3 3 vy ) . 7 e o ) b 5 s i b g{ Y & v i ’ . SH a5 & ;{k i o gi.x e d 5 " o e L o i y o Nt e v i i 4 ot ey £ = s . . i AR o i i 'w,!m ar il A e ] {0 Fally "} - o o ¢ o . o i o > * 4 ok 9 - e & ¥ o1 e oy Q o i . - e g o {0 L i 4 3 i b 2% i 4 o s b e - B L i 9 e " % ?' 4 i . I A e s i i . By b By Ir i s o S8 ! Ao st o % )?'c e e = ol o b i a7 b 1 i) v e i E . e b o W A 4 % EYe g b e 8 ot . s i . i i . #, & i i3 i ;%‘ 5 i i = i i W P Wi % ot e o i IS & e SR {52 g ;h . b 3 5 iy b & i :w gt £ " et G e i i i i s Ve i . £ ) i i e i " P it 3;; % :; g e & 0 (L ikt i pis 7 i 5 e e k3 i a s el [ & ol ) o ¢ ; e % 2 i i 'y e o ‘\‘i‘“;g% e 5 . o Ex # . i ek B o 1 it e i ; o Gt R i el 0 " 4 xé s § fobe i #i ! i e i R 3 . i i L y i i el k) é:% . 00 o N e ! Ean a4 W 2 & o i ‘, S e ! b & . o e 2 et A S T i o o iy Aot 2 . I B i i piy o g (0 St e % i 0 o i o . & ¥ ¥ o v . s & E: ‘1 = i 3 4%‘; it b i L e : o il b g il W B . il o i ) e i3 : i el L i e 45 i B 7 b Es i s o o o g i B . i " Ha Wi i #)ng ot & I g 3 {& % g i th oW I g i s 2 i s i iy in o i i A G v iy i L4 iy k¢ i i i 5 i o 9 e 4 it b o i % W 2 P Danish Criminal Justice Statistics Paul Henning Larsen 1. Denmark has been publishing annual criminal statistics since 1832. Their main publication is called Kriminalstatistik, which is published annually. The volume is about 120 pages and comprises mainly tables with little textual or graphical comment. It is however very clear and well-produced, with an English translation of main table headings and clauses at the back. In addition to Kriminalstatistik, they also publish an annual brief on clear- up rates, quarterly briefings on recorded crime (publishing data 20 days after the end of the quarter to which they refer) and assorted press releases and special studies. 2. The statistical section dealing with criminal justice statistics comprises two academic staff and two clerical staff. Clearly, even given that most of their data comes to them through AD], it is a very small section. There are also around ten criminologists at the University who carry out various pieces of research using the section’s data. 3. Due to the nature of the Danish statistical system being able to link individuals to different registers via their unique Central Population Regis- ter (CPR) number, it would be possible to carry out analyses based on the education of criminals, their nationality, income, residence and so forth. However, their limited staff mean that they have been unable to make the most of this source. The CPR number also means that they can link records to produce criminal histories, like the Offenders Index. 4. The usual statistics that they provide are summarised in the diagram below: 32 Paul Henning Larsen The Danish Criminal Justice statistical system Crime committed Crime recorded People charged People sentenced (no. of crimes) (no. of persons) People given custodial sentences Clear-up rates, Place of type of crime, per residence, Penal Main and lesser Type of prison, 100,000 of Code, Traffic Art, charges, guilt, length of population other Special _— disposal = | sentence, over 15 etc Laws, offence, on remand age and sex Reported but NFA Other sanctions Not guilty Not reported and not known Data on reported crime published 20 days after end of quater (by ADI) The numbers of crimes record by the police and the proportion that are cleared up for the various categories of crimes and also as a rate of crimes committed per hundred thousand population (aged over 15). Kriminal- statistik gives these for the years 1933 to 1991. Examples of subsequent information to be found there are given below. Danish Criminal Justice Statistics 33 5. Recorded crime has been decreasing in Denmark since 1989 but still stands at over half a million recorded crimes per annum (all categories of offence). Over half of all recorded crime comprises burglaries and thefts. Violent and sexual offences represent only 2% per cent of all reported crime. Over 96 per cent are property crimes. 6. The clear-up rate has fallen from over 40 per cent in the 1950’s to around 20 per cent throughout the 1980s. This coincides with a sharp rise in the number of recorded crimes per 100,000 of the population over age 15 - which has risen from under 4,000 to over 12,000 over these four dec- ades. The clear-up rates for burglary and theft are around 11 per cent and 23 per cent respectively. The clear-up rate for robberies is over 40 per cent. Of reported homicides and attempted homicides, 83 per cent were cleared up in 1991. 7. The total number of people convicted for crimes in Denmark in 1991 was 141,020 (18,792 of whom were women). This represents a rate of 3,295 persons per 100,000 of the population aged over 15. Just under a half of the total were motoring offences of some sort (- for women, this proportion is nearer a third). There were 62 cases of homicide or attempted homicide, and in total about 5,600 offenders convicted of acts of violence. In addition there were over 700 convictions for sexual offences. The peak age of offen- ding in Denmark in 1991 (as a proportion of the population) was 20 for males and 24 for females. However, for property crimes only, the peak for males is 18-20, and for violence against the person offences committed by males it is 18-22 years. 8. Of the 55,276 offenders under the Penal Code, which exclude motoring offences and those offences relating to ’special laws’, 16,637 received custodial sentences (7,984 of which were unsuspended and 8,653 sus- pended). 18,870 offenders received fines, 16,365 were recorded as ’no charges’, and 1,661 had the charge withdrawn. The ’charge withdrawn’ figure has fallen from 2,600 in 1981. The number of convictions resulting in fines has increased in almost every year over the same period (1981-91). Motoring offences resulted in 5,607 custodial sentences in 1991, the vast 34 Paul Henning Larsen majority of which were unsuspended. The ’special laws’ group contributed a further 1,072 custodial sentences. The average sentence length in Denmark appears to have fallen over the past decade, particularly for the more serious offences (falling from 7.9 months to 6.2 months for cases resulting in sentences of ‘non-lenient’ custody). Statistics of Crime and Criminal Justice in England and Wales Gordon Barclay Responsible Authority Statistics of crime and criminal justice in England and Wales are the re- sponsibility of a single authority, the Research and Statistics Department (RSD) of the British Home Office [Contact, Gordon C. Barclay, Dissemina- tion and Intemational section, 18th Floor, Lunar House, Wellesley Road, Croydon, Swrrey, UK: 0181-760-2326]. The RSD sets central standards for the criminal justice agencies re- garding counting rules, accuracy, timeliness, format, storeage and transmission of data and collates the data received from these agencies into extensive data bases. It also conducts household, victim and other surveys. The RSD not only serves the British Government and Parliament with policy and statistical advice but is the main agency for analysis and publication of crime and criminal justice statistics. Although Scotland and Northern Ireland are integral parts of the United Kingdom, they are really separate states as far as criminal justice is con- cerned, with their own laws and administrative structures for law and order. Scotland, in particular, has a very different system of law. The statistics of these two parts of the UK are very different from those of England and Wales and are not covered in this chapter. Basis for Data Collection in England and Wales Although the Police Act 1964 gives the Home Office the authority to demand statistics in any detail of police authorities, this power is not used. 36 Gordon Barclay Rather a series of consultation processes exists with all agencies regarding data collection which aim to balance the cost of collection against both local and national needs for data, taking into account likely changes in both technology and administrative structures. Such procedures have existed for nearly two centuries. The most important present developments are: On the demandside, the pressure from government for regular and early monitoring of their justice policies and for improved value for money. And, from all users the pressure for statistics to be made available in user- friendly fashion, including more and more on electronic media. On the supply side, the developments in information technology which enable large amounts of data to be collected relatively cheaply as a by- product of administrative systems, the potential of modern databases to produce speedy and detailed analyses and to link different data collections. As in most countries, statistics are always in a process of development. In the middle of the 1990s they still maintain a good deal of the character of the original 19th century systems. The migration to make full use of present IT will certainly take until the end of the century. The Criminal Justice System (CJS) The CJS in England and Wales differs from nearly all European coun- tries in that it developed from Anglo-Saxon concepts of Common law. More details are given in Barclay (1993). Although similar agencies exist as in other countries, their powers and procedures vary. Broadly, police record about a quarter of all crime that occurs, mainly because it is not reported to them. They clear up about a quarter of the crime they do record and caution about 40% of offenders known to them and pass the rest to the Crown Prosecution Service (CPS). Both the police and some other agencies have the power to give fixed penalties for (mainly traffic) offences and the number of such penalties is Statistics of Crime and Criminal Justice in England and Wales 37 about four times the number of court cases. The CPS pass most cases to the lower (magistrates’) courts, where the vast majority of court cases are dealt with. Other authorities, e.g. licencing authorities also pass cases to the lower courts when their own procedures have failed. A proportion of the more serious cases are passed to the higher (Crown) court. Penalties awarded by the courts range from discharge, through fines, community penalties, up to life imprisonment. Two agencies are concerned with the offender after sentence: the Probation Service supervise communi- ty penalties, including supervised release. The Prison Service supervises those in custody. Each of the agencies: Police, CPS, Magistrates’ Courts, Crown Courts, Probation Service and Prison Service pass Statistics to the Research and Statistics Department. However, for organisational and historical reasons, these statistics often have slightly different characteristics: e.g. detail of offences, timeliness, definition of units. Cases are not linked statistically across agencies. The sections below give a general discussion of the availa- ble statistics and regular publications. There is a long-term strategy to coordinate the computer systems of all these agencies (CCCIJS), which is likely to come about by the end of the century and the compatibility of statistics which will then be produced will be very much greater. This CCCJS work is also the responsibility of the Research and Statistics Department [contact Martin Narey, 0171-273-3527]. Statistics of Crime This topic is covered in more detail in Lewis (1992). Most crime does not get reported to the police so that the starting point must be victim surveys to estimate the total extent of crime. The British Crime Survey is now conducted every two years. In it questions are asked of around 10,000 households, including their experience of crime. A large 38 Gordon Barclay number of publications are associated with the surveys and the latest results are given in Mayhew, Aye Maung and Mirrlees-Black (1993). The survey estimated a total of 15 million crimes in 1991. The vast majority of offences were against property. Motor vehicles (cars, vans and motorcycles) were a very common target: over a third of all BCS incidents involved theft of, theft from or damage to vehicles. In contrast, burglaries formed a smaller proportion of survey crimes, around 9%. Wounding and robbery accounted for only 5% of BCS crime although less serious assaults made up another 12%. Over the ten years from 1981 to 1991 the proportion of BCS offences reported to the police has increased. For the comparable sub-set, 36% of offences were reported in 1981 but 50% in 1991. There have also been changes in the proportion of reported crime police record. The overall result has been a decrease in the ratio of BCS crime to the number of offences recorded by the police. In 1981 it was estimated that, for the comparable subset- there were 4.5 BCS incidents for every one recorded by the police. This had dropped to 3.4 by 1992. The better known set of crime statistics are those recorded by the police. The counting rules for this series are set by RSD. Statistics are published on a half-yearly basis, in considerable detail: e.g. by 70 offence types, for each police forces area, and by whether the incident was cleared up by the police and how it was cleared up, e.g. by a charge, caution, taken into consideration or otherwise. In 1993 the total number of offences recorded by the police was some 5.5 million, the same as in 1992. However, for over 30 years this series has increased by an average of about 5%: in some years, typically those of economic recession, the increase is greater, in a few years, typically those of economic recovery, the increase is less, or there has even been a decrea- s€ The recent BCS results confirm that much of the increase in police recorded offences during the last decade was due to recording changes and not to a real increase in crime. However, this has not diminished the politi- 39 Statistics of Crime and Criminal Justice in England and Wales cal interest in recorded crime figures, at both national and local levels. They are regarded as an indicator of malaise in society, and played up by the media. Certain commentators regard the statistics themselves as causing a fear of crime and the Home Office has been accused of statistical ter- rorism by continuing to publish them. Whatever the media interest, it is not possible to judge what is hap- pening to crime without considering both victim survey results and police recorded crime. Since the usual statistic used to compare countries is the police figure, this must throw doubts of the sense of trying to do so. Even when it is known that police definitions in two countries are compatible, the changes in recording over a particular period may well have been different. However, some crimes are known to be well recorded: e.g. serious vehicle theft and house burglaries and comparisons across countries of trends in such offences would be more valid. This dual approach will continue to be developed. If funding could be obtained an annual victim survey, as in the USA, would be a possibility. There is also the need to conduct surveys of business premises, hospitals, schools, which are not included in the BCS. On the police side there is a continuing need to bring the counting rules up to date, to ensure that new offences are included and that the rules for offences such as cheque-book fraud reflect what is happening in society. However, there is a tension here between the need for more accurate measurement, the growing cost of such measurement and the political embarrassment in explaining in public that a once and for all step up in the series is an artefact of the recording system. Other Police Statistics The most important series is the number of cautions which are issued by the police. This is a procedure whereby the police can stop a case before it goes to the prosecutor if the suspect admits his guilt. He is cau- 40 Gordon Barclay tioned and set free, with no further action taken on that offence, although for a future offence his previous caution will be taken into account. Individual records of each caution are sent by the police to RSD who maintain a database of cautions and publish detailed analyses. A considerable amount of data is also collected on police activity on stops and search, prevention of terrorism, road checks, intimate searches, motoring crime, fixed penalties awarded, checks for jury service, drug addicts and offenders and other police activities. This information is pub- lished in the series of Home Office Statistical Bulletins. Statistics of Prosecutions Before 1986 most prosecutions were brought by the police and few statistics were available. The CPS was set up in 1986, but the introduction of a full case-handling computer system for the CPS will not take place until around 1996. At present few statistics are published the main ones being in the annual CPS report. However, the amount of detail available is very limited, being based on the need to measure and forecast workload, rather than to pass management information to other agencies of the CJS or to measure a social phenomenon for wider interest. Statistics for the Magistrates’ Courts These statistics have a long history, with results of individual cases having been sent to the Home Office since the start of the 19th century. From the middle of the 19th century, these statistics have, most illogically, been collected by the police. At present most results still come from the police on paper but a move to automatic data interchange direct from courts computers is in progress and electronic data interchange is expected by 2001. The Home Office maintains a data base of information on all disposals from magistrates’ courts, with information on offence type, disposal (in- Statistics of Crime and Criminal Justice in England and Wales 41 cluding amount of fine, length of order/sentence), age, sex, name, court, remand. This enables a very large amount of detailed analysis, for policy and public use. Statistics for the Crown Courts The history of these statistics is similar to that for the lower courts, al- though the move to automatic data collection is scheduled for 1995. The type of information collected is the same as in the lower courts and a similar database is maintained. Although courts data is, in theory, 100% complete, such a large scale data processing task does not always get carried out correctly, and some deficiencies creep in. However, the main missing element is that the infor- mation simple records the situation at the end of the case, ie when it leaves that particular court. There is no detail on what has taken place during the court hearing: for example, a person charged with wounding who was prepared to plead guilty to common assault (a lesser offence) would simply have his disposal recorded for the lesser offence and the original charge would be lost statistically. Also, if a person appealed against his sentence and thus had his sentence quashed, this would also not be recorded. This inability to ’follow through’ the offender from charge to final disposal is probably the most important deficiency in British criminal statistics. The most important analyses from courts results are: counts of the total number of known offenders (defined as those cau- tioned or found guilty at court) by age, sex, offence, court, police force area, etc. Use made of particular disposals, both in absolute numbers and in relation to other disposals. This is where ministers wish to monitor the effect of a new law. 42 Gordon Barclay Use made of particular disposals at a court in a particular town, to see whether sentencing is being carried out uniformaly across the country. There are few official rules on sentencing, other than maximum sen- tences although guidelines are from time to time set by senior members of the judiciary. Analyses showing the workload of courts, delays, manpower, etc. These court results are published annually in Criminal Statistics, with workload figures in Judicial Statistics. Court Statistics present difficulties in international comparisons. The disposals available to courts differ from country to country, so comparisons of proportions given imprisonment will reflect as much differences in the law as in practice. The existance of penalties outside the courts system will influence the sense of any comparisons as will the resouces which are available to the CJS: e.g. judges will be unlikely to make a specific commu- nity order if the facilities do not exist locally for that order to be carried out. Like all other comparisons, differences should not be taken at their face value but used as the starting point for further investigations. Statistics of the Probation System The Probation Service is organised locally: it has a role in preparing reports before sentence is given in courts as well as supervising the various community penalties which can be given and in supervising prisoners after their release up to the time their sentence expires. Statistics are collected on each case, by automatic data interchange with local Probation Services. The main analyses are in connection with the workload on the services. Statistics are published annually in Probation Statistics. Because the roles of the Probation Service are so very different from the roles of similar agencies in other countries, it is doubtful if valid compari- sons could be made across countries. More should be made of the differing roles of the service in different countries and their different funding: for Statistics of Crime and Criminal Justice in England and Wales 43 example, the importance a government gives to treating offenders in the community. Whether the local community feels it has a role in rehabilita- tion or whether this is left to the State. Statistics on the Prison Service The Prison Service is organised centrally and its local and central ad- ministrative computer systems are all linked via electronic network. Thus, statistics are passed to RSD from the local prisons giving information on an individual basis on offence, age, sex, name, length of sentence, date of release, ethnic characteristics etc. Analyses are made of the numbers admitted to prison, by offence, age, sex, etc, and of the numbers in prison. These analyse are particularly useful, both to ministers who wish to know the effect of their policies on the prison population and on those running the service, in order plan most efficiently their resources. The last 20 years has been characterised by many changes of policy with regard to the use of prison for offenders, in a general reaction to the increase in recorded crime and a generally growing prison population. One of the main uses of statistics is in predicting the demand for prison places both short and long term. In general British sentencers have not taken into account the availability of prison places when sentencing, so that short term in- creases in incarceration rates have to be met by using marginal prison population (e.g. police cells, ships, army camps) until there is time to build a new prison. The statistics are published in the annual Prison Statistics. On the face of it, comparison of the numbers of people in prison is something which should be quite valid across different countries. However, account should be taken of the policy on remanding in prison before trial or before sen- tence and of the basic level of crime in the country. Both can affect the prison numbers: e.g. one would expect a country where the crime rate was double another’s to have a higher prison population, even if the justice system were identical. 44 Gordon Barclay Ad Hoc Statistical Collections A large proportion of the statistics collected are on an ad hoc basis, to supplement the regular figures noted above. For example, a National Prisons Survey has looked in much more detail at the characteristics of offenders in prison: self-reporting studies have given more information on the number of offences committed by offenders: surveys of violent crime look in more detail into the circumstances of crimes: court surveys look much more quickly at the short-term effects of new laws. The results of such surveys are published in two RSD series: either Home Office Research Studies or Home Office Statistical Bulletins. Linked Statistical Analyses of Offenders The existence of large databases enables both large and small scale studies of particular types of offender to be carried out. The Offenders Index is a database created from the courts results database to give an alphabetic list of all those who have been convicted over the last 30 years, with details of their convictions. This enables the following types of studies: following through cohorts born in particular periods. Following through those given particular disposals to show how soon, and how often they are reconvicted. Following through the criminal careers of particular types of people to see if they are different from others: e.g. drug addicts, those given a particular type of treatment during a community or prison sentence. Following through those given cautions to see if this treatment is cost- effective. Comparative studies of different types of disposals. Results are published in the two RSD series. Statistics of Crime and Criminal Justice in England and Wales 45 Digests of Statistics It must be acknowledged that when there is such a large amount of statistics available, these can be quite indigestible even to the more serious reader. In the last 3 years an attempt has been to produce a much more user-friendly summary of the more important points. This Digest of the CJS is available free and updated every two years. It presents information in a series of Bull Points and graphs and is confined to under 90 pages long. Publications Series - Research Studies: Home Office Research Studies (HORS), starting with No. 1 in 1969 now reached No. 132 in 1993. Research Findings: Short Summaries of Research, started 1992. Series - Statistical Studies Home Office Statistical Bulletins, some 30-40 published each year, including annual series on cautions/sentences, motoring, breathtests, drug addicts, drug offenders, police powers, prevention of terrorism, delays in court, recorded crime (half-yearly). Statistics Findings: Short Summaries of Statistical results, started 1993. Annual Criminal Statistics - England and Wales Judicial Statistics - England and Wales Probation Statistics - England and Wales Prison Statistics - England and Wales Gordon Barclay Biannual Digest of Statistics on the Criminal Justice System The British Crime Survey Other publications referred to: Barclay (1993) The Criminal Justice System in England and Wales [updated regularly] Crown Prosecution Service - Annual report Lewis (1992) The measurement of Crime, Social Trends 1992 Mayhew, Aye Maung and Mirrlees-Black, The 1992 British Crime Survey, HORS No. 132 Crime and Criminal Justice Statistics in Germany Richard Blath' The authors of this booklet were asked to give answers to six questions on crime and criminal justice statistics in their respective countries. For the Federal Republic of Germany, the following holds true: 1. Are there any statistics covering the whole area of police and criminal justice in these respective countries? In the Federal Republic of Germany, there are no statistics covering the whole range of police and criminal justice. We do not even have statistics which collect data on all crimes committed in our country. There are various reasons for this situation. One is that some offences, for example customs and tax offences, are dealt with not by the police but by special agencies who are entitled to prosecute these offences. Another reason is that the different statistics can and do actually have different aims: criminal policy matters on the one hand, and organizational planning and control on the other. So at least it seems difficult to establish or combine statistics which serve both ends. These different purposes of statistics make it diffi- cult to create a uniform concept of statistics which covers all aspects and levels of crime and the criminal justice system. In Germany, there are not only practical and historical reasons which make the introduction of such statistics more difficult. In addition, aspects of data protection would make it very difficult to introduce this kind of "Verlaufsstatistik", as we call it in Germany. * I would like to thank Mrs. Beyer and members of the staff of the translation office of the Federal Ministry of Justice for their assistance in preparing this paper. 48 Richard Blath 2. Are there any partial statistics, and if so how are they connected? Are they compatible? In the Federal Republic of Germany, there are a lot of crime and crimi- nal justice statistics. In order to comprehend them, it seems necessary to have at least a good idea of the law enforcement and criminal justice system. We should also know how these statistics work, which data are collected by these statistics, and where the gaps are. I should therefore like to outline German crime and criminal justice statistics against the background of the different levels of the criminal law enforcement system. In No. 1 of the supplement, you will find a short overview of the sections of the proceedings and of the relevant statistics. a) Crime statistics of the police Only those crimes which are reported to the law enforcement agencies can be recorded by crime and criminal justice statistics. In Germany, we do not have an official crime survey telling us something about the unreported figures of crime in recent years. However, some local and nationwide crime surveys has been carried out by criminal research institutes.' The great majority of crime, but by no means all of it, is handled first of all by the police. The results of the investigations carried out by the police are recorded in crime statistics of the police. 1 Schwind, Ahlborn, Eger, Jany, Pudel, WeiB, Dunkelfeldforschung in Gottingen 1973/74, BKA-Forschungsreihe, Band 2, Wiesbaden 1975; Stephan, Stuttgarter Opferbefragung, BKA Forschungsreihe, Band 3, Wiesbaden, 1976; Schwind, Ahlborn, WeiB, Empirische Kriminalgeographie, BKA-Forschungsreihe, Band 8, Wiesbaden 1978; Schwind, Ahlborn, WeiB, Dunkelfeldforschung in Bochum, BKA-Forschungsreihe, Band 21, Wiesbaden 1989; Kury, Dérmann, Richter, Wiirger, Opfererfahrungen und Meinungen zur inneren Si- cherheit in Deutschland, BKA-Forschungsreihe, Band 25, Wiesbaden 1992; Bilsky, Wetzels, Mecklenburg, Pfeiffer, Subjektive Wahrnehmung von Kriminalitdt und Opfer- erfahrung, in: Kaiser, Jehle (Hrsg.), Kriminologische Opferforschung. Neue Perspektiven und Erkenntnisse, Bd. 102 der euen Kriminologischen Schriftenreihe, Heidelberg 1994, Gutsche, Sozialer Umbruch und Kriminalitdtsentwicklung in den neuen Bundesliandern, in: Kaiser, Jehle (Hrsg.), Kriminologische Opferforschung. Neue Perspektiven und Erkenntnisse, Bd. 102 der Neuen Kriminologischen Schriftenreihe, Heidelberg 1994; Kury, Zur Bedeutung von Kriminalitdtsentwicklung und Viktimisierung fiir die erbrechens- furcht, in: Kaiser, Jehle (Hrsg.), Kriminologische Opferforschung. cue Perspektiven und Erkenntnisse, Bd. 102 der Neuen Kriminologischen Schriftenreihe, Heidelberg 1994. Crime and Criminal Justice Statistics in Germany 49 These statistics include information on the type, place and time of commission of crime, use of weapons and amount of damage (see No. 2 of the supplement). They include further information on the clear-up rate, as well as on the suspect, for instance age, sex and nationality. Finally, they include - in the case of criminal violence - some information on the victims (age, sex and relationship to the offender). According to German law, in criminal proceedings the police are not allowed to discontinue criminal proceedings. Only the public prosecutor and the court have the right to do this. Therefore, the police have to submit the files to the public prosecutor who takes the further necessary decisions. The data for crime statistics of the police are collected at the time when the police close the file of a case prior to handing it over to the public prosecutor. It has already been mentioned that not all crimes are included in crime statistics of the police. For example, road traffic offences are not included, even though they are handled by the police. Another example are customs and tax offences. The figures for crime statistics of the police are published annually by the Federal Criminal Police Office, Wiesbaden.? b) Statistics of the prosecution service When the public prosecutor has received a file, he may - if necessary - make further investigations, or he may take his decisions on the basis of the file. He is obliged to discontinue the case if in his opinion no crime has been committed, or if the suspect has not been proved to have committed a crime. In less severe criminal cases, the public prosecutor is allowed to discontinue the proceedings, in some cases with, and in other cases without, the permission of the court. In most serious cases, he will bring charges against the defendant or - in less severe cases - apply for a court order imposing a criminal punishment (suspended sentence up to one year or fine). 2 Last volume: Bundeskriminalamt, (Hrsg.), Polizeiliche Kriminalstatistik 1992, Wiesbaden 1993. 50 Richard Blath At the level of the public prosecutor, we do not have the same type (and quality) of statistics as for the police. At the level of the prosecution service, we only have statistics concerning the proceedings, but not on the defendant (although some information on the number of defendants is included). These statistics of the prosecution service include information on initiation of proceedings, on the duration of proceedings so far, the type of proceedings, and the type of the dismissal at this stage (see No. 3 of the supplement). These statistics have a primarily administrative function (for instance planning the distribution of the workload and the allocation of personnel). Nevertheless, these statistics have some relevance for crime policy matters. They show for instance that in the last two decades, the number of proceedings dismissed by the public prosecution service has increased markedly. The annual figures for the statistics of the prosecution service and the statistics described below are published by the Federal Statistical Office, Wiesbaden.? c¢) Statistics of the criminal court If the public prosecutor has terminated the proceedings at this level by indictment or by an application of a court order imposing a criminal punishment, he sends the file to the criminal court. In case of indictment, the court examines the so-called interim proceedings as to whether there is a real chance of reaching a conviction. If this has been established, the court then opens the so-called main proceedings, which means: preparation of the trial and the trial itself. When the court has terminated the proceedings by a conviction or by another decision, the court registry provides data for another type of statistics, i.e. the statistics of the criminal court. These statistics compare the statistics of the prosecution service, mentioned above. They include, for example, information on the kind of proceedings, the duration of the proceedings and the kind of termination 3 Last volume: Statistisches Bundesamt, (Hrsg.) Staatsanwaltschaften 1989, Arbeitsunter- lage, Wiesbaden 1991. Crime and Criminal Justice Statistics in Germany 51 of these proceedings (see No. 4 of the supplement). Information about the crimes is not included. Concerning the defendent these statistics provide only information on their number and on the type of termination of the proceedings." These statistics are also of a primarily administrative character. d) Statistics of Convictions At court level, there is another type of statistics which is important mainly for criminal policy matters: the so called "Strafverfolgungsstatistik", the criminal prosecution statistics. But this term is misleading. For the purpose of translation we should rather call them statistics of convictions. In contrast to the above mentioned statistics, these relate to persons whose criminal proceedings at the court level have been concluded. That means that a conviction or another decision (for instance discontinuation, acquittal, order of transfer to a psychiatric hospital) taken in respect of the accused person has come into force. These statistics include information on the most severe crime of which the defendant has been accused or for which he has been convicted. They include further information about the person (with respect to age, sex and nationality) and information on the type of court decision, in particular sanctions (see No. 5 of the supplement).’ e) Prison statistics Once the decision of the court has entered into force, the next phase of the criminal proceedings begins: the execution of the sentence. Whilst we do not have comprehensive statistics on this phase, there are however statistics relating to certain types of sanctions. Several statistics concern prison inmates. In this field, there are statistics which are collected on a monthly basis and which concern prison accommodation, the number of admissions and the number of releases. In addition, we have statistics which, on 31 March of every year, count the number of inmates serving a prison sentence. These statistics include some information on the prisoner 4 Last volume: Statistisches Bundesamt, (Hrsg.), Strafgerichte 1990, Wiesbaden 1993. 5 Last volume: Statistisches Bundesamt, (Hrsg.), Fachserie 10 "Rechtspflege", Reihe 3 "Strafverfolgung 1991", Stuttgart: Metzler-Poeschel 1993. 52 Richard Blath (age, sex and family status), the most severe crime committed, the type of custody and the duration of the prison sentence. There is also some statistical information on convicted persons in mental hospitals and in hospitals for alcoholics and drug addicts (see No. 6 of the supplement).® f) Probation statistics In accordance with the Penal Code, a prison sentence of up to two years may be suspended. The remainder of a prison sentence can also be suspended when 2/3 (or more) of the sentence have been served. In these cases of a conviction, a sentenced person is placed on probation if the conviction was subject to juvenile criminal law. In cases of such a conviction according to the general - i.e. adult - criminal law, the convicted person may, but need not, be placed under the supervision of the probation service. We collect some statistical data on the procedure and the outcome of probation. These probation statistics include information on the person (age, sex, nationality, most severe crime committed, type of sanctions, the type of outcome, duration of the probation periods; see No. 7 of the supplement).’ From this it can b;: ascertained that these crime and criminal justice statistics are not connected with each other. Only some statistics are compatible. For instance, the statistics of the criminal prosecution service and those of the criminal courts are designed so as to be comparable. Both series of statistics count proceedings, and they both include information on the type and the duration of the proceedings and the decisions finally taken. Last volume: Statistisches Bundesamt, (Hrsg.), Fachserie 10 "Rechtspflege", Reihe 4.1 und 4.2, "Strafvollzug am 31.3.1991", Stuttgart, etzler-Poeschel 1993. Last volume: Statistisches Bundesamt, (Hrsg.), Fachserie 10 "Rechtspflege”, Reihe 5, "Bewihrungshilfe 1990", Stuttgart: Metzler-Poeschel 1992. Crime and Criminal Justice Statistics in Germany 53 Comparisons are often made between the number of suspects recorded in the crime statistics of the police and the number of convicted persons, recorded in the criminal prosecution statistics, who are suspected and convicted of a certain offence or crime. But differences in the type of data collection, time lags and the change of the judicial evaluation of the crime committed in the course of the criminal proceedings make such a comparison really difficult. In addition, there is a gap in the information available on persons whose proceedings are terminated by the public pro- secutor. Also, this gap creates some difficulties in interpreting the differences between the number of suspects and the number of persons convicted of a particular offence. There are also gaps in the information available with respect to the execution of sentences, besides the prison statistics, for instance in cases where a conviction is suspended and no transfer takes place to the probation service. There is also a lack of information in the area of prison statistics. Therefore, we do not have sufficient information on prisoners admitted to and released from prison. Finally, we do not have nationwide statistics on the execution of fines and on recidivism. A summary of the existing crime and criminal justice statistics is to be found under No. 8 of the supplement. No. 9 of the supplement gives a rough impression of the numerical volume of crimes, defendants and proceedings at the different levels of the criminal justice system. 3. What can be said about the reliability and the value of data of statistics? Generally, it can be said that the existing crime and criminal justice statistics in Germany are extremely reliable. Certainly each statistic contains mistakes, but all the statistics mentioned are collected by specialised regional offices. Crime statistics of the police are gathered by the criminal police offices of the Linder, and by the Federal Criminal Police Office. The statistics of the criminal justice systems are collected by the statistical offices of the Lander and by the Federal Statistical Office. Every statistic is examined with respect to accuracy and completeness. So, in principle, we can rely on the results of these statistics. 54 Richard Blath Also, statistical data are of high value in respect of the administration of criminal justice and crime policy matters. They are widely used for these purposes. There are, however, certainly some shortcomings in our statistical data collection which we are trying to surmount at present. Working groups consisting of representatives of the Linder and Federal Ministries of Justice and statistical offices are about to fill some of the gaps in our gathering of statistical data. 4 What will be the future developments regarding modes of survey and computerisation of data collection and evaluation? How can the availability of data be improved? Computerisation of data collection in the criminal justice and prison administration systems will improve the quality of statistics and make the data collecting processes easier. In addition, it will open the way to filling at least some gaps in our system of criminal justice statistics. On the other hand, availability, which means the use of data stored in data banks of the administration of justice, is now, and will remain, restricted by legal provisions. So, we need to be a little patient as regards the opening of these data bases for statistical purposes. Insofar as the existing legal provision does not allow the use‘of these data for statistical purposes, new legal provisions allowing the statistical use of such data are being planned or are in a state of preparation. 5. How far can official statistics be replaced in future or complemented by victim surveys and microcensus? According to the existing legislation on microcensus, it will not be possible to include in these statistics questions concerning undetected crimes or offences. Crime statistics should be complemented by victim surveys. A number of victim surveys have been carried out in Germany in recent years. The main problems involved in carrying out such victim surveys over fixed periods are financial ones. In the light of the present fiscal situation, a number of doubts have arisen as to whether it will be possible in future to carry out such surveys repeatedly. Crime and Criminal Justice Statistics in Germany 55 On the other hand, existing crime statistics should not be replaced by victim surveys. Victim surveys cannot cover offences with low incidence rates. A consequence of such a replacement would be that we would not get information on the whole range of criminal offences or crimes, as is almost the case with our police crime statistics. 6. Is it possible to establish international comparability/compatibility of national statistics? Countries differ in respect of their law enforcement and criminal justice systems. In addition, the penal codes of countries differ. Consequently, the countries have different crime and criminal justice statistics, and the results of these statistics may be of varying significance. This holds true at least with regard to the details of the provisions of the Penal Code and of the law enforcement and criminal justice system. On the other hand, the different penal laws and structures of the criminal law enforcement and criminal justice systems may share some common features. As far as comparability exists and the relevant statistics are compatible, it must be possible to establish international comparability or compatibility of national statistics. A group of specialists set up by the European Committee of Crime Problems of the Council of Europe, is about to examine the feasability of a "European Sourcebook of Criminal Justice Statistics". I do hope that the results of this group of experts will open up the possibility of establishing internationally comparable crime and criminal justice statistics, at least on the European level. 7. Concluding Remarks The new developments in Europe, above all the opening of the borders, the breakdown of the old regimes and the building of new political structures in Eastern Europe, as well as the process of growing integration between the countries of Eastern and Western Europe - in the economic field - have an enormous impact on criminality and on the suppression and prevention of crime by police forces and by the criminal justice system. 56 Richard Blath To cope with these new developments, there seems to me to be a need to intensify mutual cooperation between countries while combating and preventing crime. A first step in this respect can be, and should be, the exchange of information, not only in relation to the solving of concrete criminal cases, but also in respect to the general situation of crime and the criminal justice system. Crime and criminal justice statistics provide important and broad information of this kind. Supplement Table of the most important crime and criminal justice statistics 58 Richard Blath 1. section of the criminal statistics name proceedings criminal investigation: police reporting to the police crime statistics of the police starting proceedings end of the investigations by the police prosecution service - starting proceedings statistics of the prosecution service - end of the proceedings [criminal prosecution on the prosecution level statistics] Interim and main proceedings. - receipt of the case statistics of the criminal court proceedings of penal statistics of order convictions interim proceedings main proceedings procedure concerning means of redress final decision Crime and Criminal Justice Statistics in Germany 59 statistics name sections of the proceedings proceedings concerming execution correctional systems/probation supervision of conduct fine none juvenile punishment with suspension probation statistics without suspension suspension of the remaining prison statistics prison sentence probation statistics imprisonment with suspension: without probation none with probation probation statistics without suspension: prison statistics suspension of the remaining prison sentence without probation none with probation probation statistics new proceedings none - police none - prosecution service - court envisaged recidivism statistics probation statistics 60 Richard Blath 2. crime statistics of the police statistical units statistical characteristics 1. crimes crime place; time of crime use of weapons volume of damage clearance rate 2. suspects age gender nationality domicile group offence alcohol previous record drug addiction 3. victims (violent acts) age gender relationship to the suspect Crime and Criminal Justice Statistics in Germany 61 3. statistics of the prosecution service statistical units statistical characteristics proceedings (prosecution level): number of proceedings person unknown known defendants agency the duration of proceedings type of proceedings final decision accusation other decisions termination with conditions termination with- out conditions number of defendants appeal road traffic offence economic crime 62 Richard Blath 4 statistics of the criminal court statistical units statistical characteristics proceedings (court level) duration of proceedings type of starting of proceedings number of defendants number of trials number of trial days members of the last trial type of termination appeals road traffic offence 5. statistics of convictions statistical units statistical characteristics defendants month/year of the final decision gender age nationality most severe crime victimized child type of decision/sanctions remand custody duration/reasons previous convictions Crime and Criminal Justice Statistics in Germany 63 6. prison statistics statistical units statistical characteristics prisoners prisons capacity type of custody remand custody imprisonment juvenile imprisonment preventive custody penal arrest custody pending deportation prisoners (at the 31 march) gender age nationality familiy status domicile previous convictions type of custody duration of custody most severe crime persons committed to a mental type of institution hospital or to an institution temporary commitment for curing drug addicts and alcoholics type of addiction revocation age family status 64 Richard Blath 7. probation statistics statistical units statistical characteristics clients gender age nationality most severe crime type of probation reason for terminating probation duration of the remaining prison sentence duration of the probationary period previous probation probation officer number Crime and Criminal Justice Statistics in Germany 65 8. crime and criminal justice statistics name period/date statistical unit type of data crime year case single suspect victim public year proceedings single prosecution 1.1 -"— aggregate " 31‘.12. n month criminal court year proceedings n single 1.1 aggregate n 31.12. n month criminal year defendant single prosecution prison 31.3 prisoner single aggregate 31.12 " year " aggregate month probation year client " single 31.12. - aggregate 31.12. probation officer 66 Richard Blath 9. Crime and Criminal Justice Statistics Federal Republic of Germany Some Figures 1990 Crime Statistics of the Police number of crimes 4 455 333 number of cleared up cases 2093 130 number of suspects 1437 923 Statistics of the Prosecution Service (country court) 2 575 996 number of proceedings, unknown defendant 2 876 090 number of finalized proceedings number of charges 413 924 number motions to fine order 473 766 number of proceedings discontinued with conditions 177 983 without conditions 428 427 774 712 no sufficient reason for charge Statistics of Criminal Courts number of finalized proceedings 642 902 Criminal Prosecution Statistics number of defendants 878 305 692 363 number of sentenced persons adult penal law 615 089 prison sentence without suspension 32 749 prison sentence with suspension 69 705 ine 512 343 77 274 juvenile criminal law 4319 prison sentence without suspension prison sentence with suspension 7 784 youth detention 12 785 other means of correction 37 408 corrective disciplinary action 14 978 Prison Statistics number of prisoners 31.12. 44 335 number of untried prisoners 14 070 number of convicted prisoners adult penal law 25 581 juvenile penal law 3421 189 preventive custody Probation Statistics number of probations 31.12. 131 381 terminated probations 40 795 adult penal law 27 686 with success 19 304 revocation 8 382 juvenile criminal law 13 109 with success 9 793 revocation 3 316 Greek Crime Statistics Calliope D. Spinellis and Maria Kranidioti Introductory remarks Historical context Statistical categories and some legal explications On advantages and disadvantages of particular sources of data 4.1. Police statistics or information on suspected crimes 4.2. Criminal court statistics or information on convicted individuals 4.3. Prison statistics or information on detainees Discussion and assessment 5.1. Police and criminal justice "system" from "above" and "below" 5.2. Compatibility 5.3. Reliability and value of data 5.4. Future developments 5.5. Victim surveys and self-report delinquency studies 5.6. International comparability and compatibility Speculations on the potential of Greek statistics 68 Calliope D. Spinellis and Maria Kranidioti 1. Introductory Remarks Crime statistics in Greece are accused for unreliability, for being pub- lished with great delay, for uncomparability and for being amenable to mis- interpretation or misuse. Some of those accusations are true for some types of statistics on the national level only, and some are true for statistics of other countries as well. In the following discussion an attempt is made to describe the sources, structure and content of published statistics in relation " to the criminal justice "system and to assess them taking into considera- tion the international perspective. 2. Historical Context Greek crime statistics have a history of more than 100 years. The first official data on crime in Greece were published in 1883 in the "Govern- ment’s Gazette". These old data referred to: a) prosecutions, b) convictions, and c) acquittals. Some years later, the Ministry of Justice published its own statistics. These included statistical information on prison population of the years 1889, 1901 and 1908. For the years 1911-1913 data on court agents’ activi- ties, namely, "statistics on civil justice" and "crime statistics" were collected by the same Ministry. These publications were interrupted by World War I and some years later "crime statistics" were published again - for the years 1926 to 1937 - this time by the Ministry of National Economy. This Ministry published "prison statistics", which in fact continued appearing until World War 112 1 For a discussion of this "system" as non-system or as a process perspective, see A. Ruther- ford, in Council of Europe, Interactions within the Criminal Justice System, Strasburg, 1987, pp. 9-10). C. D. Spinellis, Eglimatologia. Synhrones kai palaioteres katefthynsis, Athina, 1985, pp. 59 ff. Greek Crime Statistics 69 After the dramatic events of World War II and the Greek Civil War, a more systematic collection of "court statistics" and of "prison statistics" started in 1957. In 1970, the first "police statistics" were published including data on "committed felonies and misdemeanours and their alleged perpetra- n3 tors . Since then, the Greek crime statistics include data stemming from three basic sources: the police, the criminal courts and the prisons and other custodial institutions. These statistics are published (annually but with a considerable delay) by the National Statistical Service of Greece in a volume entitled "Statistics of Justice. Civil Justice, Criminal Statistics and Prison Statistics". The volume "Statistics of Justice" consists of three parts: i) Civil Justice ii) Police Crime Statistics and Criminal Justice Statistics and iii) Prison Statistics. This publication of the National Statistics of Greece should be differentiated from another, entitled "Statistical Yearbook of Police" published by the Ministry of Public Order since 1983. From the above information, one may conclude that Greek statistics appear to be scattered in many publications. Many years are missing and their structure often changes. Thus, before 1957, one cannot find compara- ble data on convictions or on prison statistics. As far as police statistics are concerned, before 1970, a lot of data are missing. After 1984, however, with the integration of police and gendarmerie forces and with the computeriza- tion of these statistics the "Statistical Yearbook of Police" is the best source of relevant information®. Therefore, generally speaking, crime statistics in Greece are not very useful in providing reliable information for a long period of time. Nevertheless, they do not differ from crime statistics of other countries with respect to discontinuities or alterations due to wars, or to socio-political changes’. It is then safe to suggest that at least for the past two decades, Greek statistics provide some systematic information on crime and the activities within the criminal justice. This statement, however, needs some elaboration and elucidation. 3 Ibid., p. 60. 4 This assessment is made after classical criminological literature. Quetelet was the first to rely on crimes reported to the police, followed by Sellin and thus, creating a tradition for using these statistics as a basis for most theorizing on crime rates (see A. K. Bottomley, Criminology in Focus. Past trends and future prospects, 1979, p. 22). See, i.e., the brief historical account on crime statistics in the United States, in Suther- land/Cressey, Criminology, New York, 1978, pp. 35 ff. 70 Calliope D. Spinellis and Maria Kranidioti 3. Statistical Categories and Some Legal Explications As already stated Greek crime statistics are of two kinds: Statistics published by the Ministry of Justice (which also include some information on police statistics and may be found in "Statistics of Justice" volume) and those published by the Ministry of Public Order (appearing in "Statistical Yearbook of Police"). Both publications follow the classifications of the Greek Penal Code but they also contain data on violations of other Special Laws (Nebengesetze). Parenthetically, it should be mentioned that these violations account for approximately 75 % of the total criminality of the country. The data contained in the above publications are fragmentary on many issues and it is not so clear according to which criteria certain offences are selected to be presented individually, others in categories and still others omitted (e.g. violations of Customs’ Laws). The lack though of a particular system of selection leaves space for assessment to the researcher and thus, bias with regard to this matter® is avoided. More specifically, Greek statistics include information on: i) 20 broad categories of offences defined according to "legal goods threa- tened" (Rechtsgiiter) - e.g. crimes: "against life nn ) nn non against public order", n on "against sexual freedom b against property b against honour ) against religious peace" etc.’. 6 Compare Uniform Crime Reports of the FBIL. The selection of information and the classification of offences leads to biased description of the crime problem in the United States, where the white-collar crime and drug offences are underrepresented (S. E. Brown/F-A. Erbsen/G. Geis, Criminology. Explaining crime and its context, Cincinnnati, 1991, pp. 130 ff.). Nevertheless, it seems to be practically difficult to accumulate statistical information on the so-called "modern criminality" (Nordic Criminal Statistics, 1950-1989. Summary of a Report, 1991, p. 10). See i.e. table B: 7 in National Statistical Service of Greece, "Statistiki tis Dikeosinis. Politikis Dikeosinis, Eglimatologikis ke Sofronistikis", year 1988, 1992 (edited), pp. 76 ff. Greek Crime Statistics 4! 1) approximately 80 offences considered basic, serious or frequent, are presented separately (e.g. "intentional homicide", negligent homicide", "theft of all kinds", "thefts and embezzlements of low value" etc.?, n"nn i) more than 15 Special Laws (e.g. Laws concerning "illegal arms ) public health violation Jon o) protection of Labour", "Market Law", "Motor Ve- hicle violations", "Drugs", "Gambling", "Tax Law violations" etc. With respect to age-group categories the Greek Penal Code’s relevant provisions are followed (art. 121-133 GPC), i.e. data refer to the following groups: 7-12 years (provision for educational or therapeutic measures only), 13-17 years (provision for the same measures as a rule and for deprivation of liberty in a reformatory in exceptional cases), 18-20 years (provision for the same treatment as adult offenders, but the judge may apply more lenient sanctions because of mitigating circumstances); the adult age-groups also represent unequal age intervals: 21-24 years, 25-29 years, 30-34 years, 35-44 years, 45-59 years, 60 and more. Finally, attention should be drawn to classifications according to seri- ousness of the offences and their penalties. The Greek Penal Code distin- guishes offences into felonies, misdemeanours and petty offences. Life incarceration or incarceration of 5 up to 20 years is provided for felonies, imprisonment of 10 days up to 5 years and/or a pecuniary penalty is provided for misdemeanours and detention of 10 days up to 1 month or a fine is provided for petty offences. There are other penalties as well imposed on special cases: i.e., supplementary penalties and security measures (art. 18, 50-58 GPC)’. Last but not least, it should be pointed out b i) that crime statistics do not include petty offences (e.g. "change of "nn nn name", causing disturbance by noise ’ unlawful preparation and fur- nishing of poison" etc.), 8 Ibid., pp. 43 ff. 9 See Dionysios D. Spinellis, "Criminal law and Procedure" in Kerameus/Kozyris eds., Introduction to the Greek Law, Athens, 1993, pp. 348-351. 72 Calliope D. Spinellis and Maria Kranidioti ii) that criminal court statistics include irrevocable only decisions on con- victed individuals. 4. On Advantages and Disadvantages of Particular Sources of Data In order to assess official crime statistics a more specific description of the three different sources of official data, namely police, criminal courts and prisons, is needed. This description should refer to: i) the method of data collection, ii) the selection processes, and i) the advantages and disadvantages of each entity of data for the user. It should be mentioned that, together with reliabilty and validity, a related significant matter for the user is to gather detailed information from the already planned and organized collection of data. 4.1 Police Statistics or Information on Suspected Crimes The police statistics which are included in the "Statistics of Justice" volume give information on felonies and misdemeanours of alleged of- fenders. This information refers to the offender’s sex, age and the geographic region where the crime was allegedly committed. Data included in this publication are selective. The selection is made in a way that facilitates comparability with court statistics in which similar categories are included. These police statistics are meaningful provided that they are used in a study focusing on offenders and offences more and less on the criminal justice "system" per se (Aetiologisches Paradigma vs. Kontrollparadigma)'®. The comparability between police statistics and court statistics as presented in the present volume is not the only advantage of these police statistics. Police statistics: 10 See S. Lamnek, Theorien abweichenden Verhaltens, Basel/Stuttgart, 1979, pp. 238 ff. Greek Crime Statistics 73 - Distinguish between attempted and accomplished crimes. - Include crime rates per 100,000 inhabitants aged 7 years and above. Yet, they have a lot of weak points as well. For instance: Data are presented in an overgeneralized manner as the National Statis- tical Service is a secondary source of selection - the first being the relevant section of the Ministry of Public Order. Data on occupation, education or nationality of the alleged offender do not exist. No data are given on crimes committed by more than one person (gangs, bands, conspiracies, organized crime etc.)". However, a presentation of Greek police crime statistics cannot be complete unless reference is made to the already mentioned "Statistical Yearbook of Police" - a separate publication which encompass information on police activities. In this publication, information about offences is more detailed, and among others, it includes data on crimes known to the police, detected and undetected crimes, months of the year that crimes are com- milted, some categories of victims of crime, offences committed in football and basketball games etc.'’. These statistics appear in a less systematic form than the police statistics in the "Statistics of Justice" volume but they contain more information than the latter. 4.2 Criminal Court Statistics or Information on Convicted Individuals Court statistics, namely "statistics of convicted individuals" are more detailed than police statistics. They provide information on the offender’s sex, age, education, occupation, marital status, and area of residence, as well as on geographical region where the crime was committed, the type or 11 C. D. Spinellis, op. cit., p. 61. 12 Ministry of Public Order, Statistiki Epetirida tis Astinomias", 1992, pp. 9 ff. 74 Calliope D. Spinellis and Maria Kranidioti category of crime, recidivism, penalties or other therapeutic or custodial measures. Moreover, court statistics include information on juveniles’ treatment and sex, geographical region where the offence was committed, as well as data on the criminal activity of non-Greek nationals. The method of collection of court statistics may be briefly described as follows: Special cards for each convicted person are filled in by the courts’ secretaries and they are sent to the Statistical Service. Each card is called "statistical bulletin of the convicted person" and is divided into three parts: The first part includes the personal characteristics of the convicted indivi- dual (area of birth and residence, age, occupation and marital status at the time the offence was committed). The second part includes information on the type of crime, place and time referring to the "commission" of the act, type of conviction (sentence or custodial measure and suspension of sen- tence). The third part contains information on the offender’s criminal record before the commission of the act. The basis of calculation is the "individual conviction". In other words, if somebody is convicted for three different offences during the same year, he/she appears three times in the court statistics. Needless to add that, his/her personal characteristics will be registered three times, as if he/she was three persons. However, his/her sentence will be registered only once on a cumulative basis". Court statistics referring to convicted individuals chronologically are far away from the time of the commission of the offence; therefore, they describe crime in a less accurate way than police statistics. Nevertheless, Greek court statistics and especially the way of their presentation, reveal some positive aspects: They encompass a "crime rate" calculated on the basis of 100,000 inhabi- tants aged 7 years and above (sece under 3.). Moreover, there exists a recidivism rate constructed in the same way. (Recidivists per 100,000 inhabitants). 13 C. D. Spinellis, op. cit., p. 62. Greek Crime Statistics 75 They include detailed information about the way the data are tabulated with references to specific provisions of the Penal Code and the Special Penal Laws. They contain comparative tables referring to criminality, sentences and custodial or therapeutic measures for a period of five years. They encompass particular tables referring to the handling of juvenile offenders aged 7 to 17 years and to young adult offenders aged 18 to 20 years. They include data on convicted aliens, the category of their offence and the country of their permanent residence. Court statistics though, as police statistics, are not free from disadvan- tages. Namely: - They have omissions in the registration of offenders’ demographic or other data. Therefore, categories like "sentence not declared", "age unknown" etc. are included in the statistical tables. - They lack information on cases where imprisonment is converted to pecuniary sentence - a kind of "day fine". Such a conversion is provided by the Penal Code, art. 82, and is often used by the judge in Greece. For instance, there is no information on the numbers of persons who remain detained because they do not have the means to pay their converted sentence or on the number of recidivists from those whose sentence was converted etc. - They include age - categories with arbitrary intervals, i.e. 18-20, 21-24, 25-29, 30-34, 35-44, 45-59 etc. Data on the suspension of sentences are presented according to the type of offence and sex of the offender only, and not according to other characteristics of the offender: e.g. age, occupation, marital status etc. 76 Calliope D. Spinellis and Maria Kranidioti - There is no information on the population of each geographical region, where the offence is committed, or a crime rate for each region. As far as aliens’ criminality is concerned there is no information on the number of temporary residents from each country and therefore ab- solute numbers are almost meaningless. For instance, if 21 Australians, 37 Italians and 2 Bulgarians are convicted in 1988 in Greece, this infor- mation is not amenable to any meaningful comparative conclusions, since the total number of these countries’ temporary residents in Greece is not given. The categories of occupation are vague. For example, in a category like "persons occupied in agriculture, farming, logging etc.", various profes- sions are included which are far from being arranged in a scale of so- cio-economic status. Therefore, these categories are almost useless for socioeconomic comparisons. - No data exist on prosecutions or on acquittals'. 4.3 Prison Statistics or Information on Detainees The directors of custodial institutions provide every three months: i) statistical cards for each adult or juvenile detainee who is in their institution, i1) information on persons on pre-trial detention, and i) data on persons detained because of debts to the state. These cards are also filled in for each individual who is released from the institution and they include the released person’s data: sex, level of education, type of sentence,offence for which he/she was sentenced and date and reason of release. A similar card is filled in for released juveniles. In 14 1Ibid., pp. 63-64. Greek Crime Statistics 77 the juvenile’s card data on the following are included: his/her family’s environment, occupation, level of education, training and time spent in the institution, reason of incarceration and release and previous record. Individual cards are also completed for every person who is detained in any custodial institution or other penitentiary at the 31st of Decem- ber of each year. Therefore, in prison statistics, one may find information in particular on: The total number of persons who are detained in a year, in any institu- tion for adults or minors (psychiatric institutions for convicted indivi- duals are also included). The total number of persons who are detained in one custodial institu- tion or in all of them in a particular day. For example, during 1982, 6,921 persons were detained, whereas in the 31st of December 1982 the number of detained persons was 2,991", The total number of first offenders and of recidivists. S. Discussion and Assessment So far statistics have been mainly presented in a static fashion, in other words, as particular sources of data which supply information on various crime issues. In the following discussion these sources will be considered as inputs and outputs of sub-systems of the criminal justice "system". Within this frame, substantial information on the real input to the system - i.e. an input which would include most of the offences found in the dark area of crime - may be acquired through victimization and self-report studies which will be considered as well. In this context six main issues will be discussed. The first refers to the question whether Greek statistics cover the whole area of the police and 15 National Statistical Service of Greece, op. cit., p. 121. 78 Calliope D. Spinellis and Maria Kranidioti the criminal justice "system". The second is related to the compatibility of statistics stemming from the various sub-systems. The third concerns the issue of reliability and value of data of crime statistics'®. Future develop- ments regarding computerization of data and the replacement of official statistics by victim surveys or self-report studies are the next issues to be considered. Finally, the possibility to establish comparability and compatibi- lity of Greek statistics with crime statistics of other countries is discussed. 5.1 Police and Criminal Justice "system" from "above" and "below" The answer to the question whether "Greek statistics cover the whole area of the criminal justice "system" from police down to custodial and non-custodial institutions, passing through the prosecutor’s office and the criminal courts" depends on the researcher’s approach. If the selective nature of crime statistics is not questioned and is taken for granted, a look "from above" may lead to the conclusion that both the police and the administration of criminal justice are basically covered by these statistics. In other words, the main sub-systems, namely police, courts and prisons are adequately represented in the officially published volumes on crime statis- tics (see above 4.1, 4.2 and 4.3), since they encompass - at least formally - data from all police stations, all criminal courts and all custodial institutions of the country. The integration of police and gendarmerie has also con- tributed to the uniformity of data presentation, i.e., there are now fewer differences in the collection and tabulation of data in the big cities and provincial areas than before. On the other hand, if the researcher approaches the system "from below" certain aspects of the police organization are crucial to the answer. Such an aspect is that crime data are initially written in books which are kept by 16 Reliability in this context refers to the consistency of repeated data from a single observer as well as agreement among observers (See, for instance, A. Binder/G. Geis, Methods of Research in Crimininology and Criminal Justice, New York, 1983, p. 128). Value of data relates to the notion of validity, namely the degree the researcher has measured what he set out to measure (see, for instance, H. W. Smith, Strategies of Social Research. The Methodological Imagination, Prentice/Hall Intern., London, 1975, p. 61). See, also C. M. Glinnie, Improvements to crime statistics, in Council of Europe, Means of Improving Information in Crime, Strasburg, 1976, pp. 39-40. Greek Crime Statistics 79 officers in local police stations. There are approximately 52 police stations only in the area of Athens, and the information sent to the General Securi- ty of the Police, then to the Ministry of Public Order and finally to the National Statistical Service could be diversified, unwillingly or even willing- ly, in the process of transfering the information from one agency to an- other. Consequently, many offences - especially those considered petty offences - do not reach the central bureaus of the police. Because of this and other organisational parameters, even the datum "crimes known to the police" in the "Statistical Yearbook of Police" could be the result of either an erroncous classification of a crime, or a selective registration process that could take place within the police. Hence, one may argue that crime rates in certain areas of the city of Athens and of the country as well, are over- or underrepresented in police crime statistics. A more formal, detailed and accurate registration of vic- tims’ reports at local stations together with a system that would allow direct flows of information through computers to the Statistical Service is needed. Such changes though should be accompanied by a general shift towards professional policing'” and appropriate personnel training, 5.2 Compatibility It is a truism to state that every stage of the criminal justice "system” is a different organization and that the justice agents’ interests and purposes are also different. It can be argued that the information they provide to the researcher is a "side product” of various aims which are irrelevant to his/her work. In this respect, compatibility among statistics produced by the sub- 17 "Professional policing" is characterized by commitment of the officers to "community relations" and use of technology (S. Holdaway, "Changes in Urban Policing", British Journal of Sociology, v. 28, 1977, p. 124). The term refers to the British experience of the Unit Beat Policing system which was introduced in all forces of the country following the Home Office Report in 1967. Furthermore, two notions of "professional policing" can be discerned. Managerial professionalism, which is represented by the function of community liaison, training and administrative officers, theory of police management and the olicy statements of police officers. This notion is compatible to the public image of po f cing. Practical professionalism remains a series of hedonistic, protective and highly practical activities which are largely opposed to managerial professionalism (ibid., p. 134). It can be argued that practical professionalism is crucial in many aspects of police work and determinative for the information on crime rates. 80 Calliope D. Spinellis and Maria Kranidioti systems of the system of justice cannot be attained. The Greek statistics give easily support to this argument. For example, there are differences in the way data are organized in police and court statistics. Police statistics refer to allegedly committed offences primarily and then to alleged offen- ders, while court statistics refer to convicted individuals. The argument that compatibility of crime statistics cannot be reached, could be employed for any source of crime data, like victimization or self- report studies. Nevertheless, a number of prominent criminologists in the United States have found, in relation to FBI, victimization (NCR) and self-report data, that the various sources of information about crime are somewhat more compatible than first believed possible, since the crime patterns they record are often quite similar'®. Despite these similarities though, comparing the sources can result in confusing and contradictory findings because of the use of different methods of acquiring data. These differences must be carefully considered in any interpretation of crime trends'. And in any case, such observations of crime trends in the Greek statistics as well as comparisons with other data would require multiple sources of information, which do not exist. An approach of the Greek criminal justice "system" as a sequential process reveals an additional problem. There is no statistical information regarding the intermediate stages of this process. No data are published on cases at the public prosecutors’ offices or on the Judicial Councils®. Besides that, court delay in Greece, appears to be significant, as compared to other European countries, where courts are organized in a more efficient way”'. Therefore, the flow of cases cannot be observed in the statistics, neither from police to courts, nor from courts to prisons. (A rough des- 18 This opinion is supported by Alfred Blumstein, Jacqueline Cohen and Richard Rosenfeld (L. J. Siegel, Criminology. Theories, patterns and typologies, New York, 1993, p. 76). 19 Ibid., p. 77. 20 For the pre-trial stage and the stage before the Judicial Councils, see Dion. Spinellis, op. cit., pp. 358360. 21 Council of Europe, Delays in the Criminal Justice System, Strasbourg 1992. A systematic study on this matter is missing. One may draw this conclusion by comparing the date of incident and date of decision in Greek jurisprudence (see legal journals Poinika Chronica, Armenopoulos etc.). Greek Crime Statistics 81 cription of flows would be the following: In 1988 the input of police was 311,179 or 3,397 per 100,000 inhabitants and the output of the criminal courts was 132,925 convictions or 1,451 per 100,000 inhabitants and the input in custodial institutions was 6,921 individuals, while the output (re- leased) 3,905 individuals)®. Yet, the only instance, where the flow may be accurately estimated is, as already mentioned, at the end of the process, in the prison statistics (the number of the released persons from those de- tained in a year, see above, under 4.3.). In custodial institutions, though, as one would expect, ends up only (approx.) 1/45 of the alleged offences. 5.3 Reliability and Value of Data The question of reliability and value of data in the statistics may have a different answer according to the three main sources of this data. The most credible statistics with the criterion of what is set out to be measured® are those of prisons, provided that the goal is simply to measure detainees and not crime rates or severity of sanctions. With the same criterion, and if the goal is to measure criminality, police statistics deem most reliable for they meet the well-known criterion of closeness of the source to the point of generation of the information being provided™. Moreover, in Greece the principle of legality applies and police, with the exception of petty offences, does not have much discretion in selecting the cases to be forwarded to the Public Prosecutor’s Office. However, inaccura- cies may be found in police statistics because of erroncous classifications of the crime (e.g. a crime may be classified as misdemeanour by police and as felony by courts or as intentional homicide by police and negligent homicide by courts). Furthermore, as already mentioned, police statistics refer to suspects and some of them my be proved innocent at the later stages of the administration of justice. Nevertheless, the existence of dark figure neutralizes this potential overrepresentation. 22 Calculations based on data of the National Statistical Service of Greece, op. cit. 23 The criterion of value of data (see note 16). 24 C. M. Glinnie, op.cit., p. 40. 82 Calliope D. Spinellis and Maria Kranidioti However, the reliability of police statistics may be questioned on the basis of the criteria of i) the degree of self interest involved on the part of the person responsi- ble for keeping the records and it) the extent to which the person making the return has been trained to know what is required of him®. It is exactly on the basis of these two last criteria that court statistics as compared with police statistics are more reliable, provided that one is interested to acquire through infor- mation more on the courts’ activities and less on the quantity and quality of criminality or on the profile of the offender. Yet, the fact that Greek statistics do not include data on acquittals blurs somewhat the picture decribing the "production” of criminal courts within a cer- tain period of time. Moreover, despite the absence of the principle of expedience or oppotunity, some selectivity process occurs, especially through the public prosecutors’ powers and the intervention of well- known lawyers. Overall, due to the lack of statistical information at intermediate stages on the one hand, and to the court delay on the other, it could be suggested that Greek statistics do not describe the criminal justice process accurately. The information they provide is, in any case, fragmentary. Nevertheless, this information appears for twenty years uniformly in the "Statistics of Justice". That is, consistent patterns of crime can be observed for this period in the official statistics. 5.4 Future Developments Official crime data collection has traditionally relied upon complicated police and courts’ burcaucracy. One of the most serious handicaps of the Greek statistics, to which attention has been initially drawn (see above, under 1.) is their publication with great delay. This delay leads the resear- cher to crude estimations regarding recent changes in crime rates. Given 25 Ibid. Greek Crime Statistics 83 the fact though that computerization in the public sector is expanding, it could be expected that in the very near future, information provided by the National Statistical Service of Greece will be collected, processed, analysed and presented faster than before. The use of information technology by the criminal justice should have a positive impact on the quantity and quality of data as well. However, the fulfilment of such expectations depends always on organizational parameters and economic priorities. With respect to the availability of data one could assume that modern technology which is in process of being applied in Greece, will improve interconnections among various organizations which produce and elaborate crime information. Such organizations are: i) the criminal justice sub- systems, ii) Ministries and iii) the National Statistical Service on the one hand, and the autonomous research centres and the University research institutes, on the other. One could hope that parts of (raw) data produced by the National Statistical Service could be in the future quickly and easily available to research centres before being published. 5.5 Victim Surveys and Self-report Delinquency Studies In the present context, the interest lies not only on official crime and recidivism rates stemming from the criminal justice "system" but also on additional or alternative sources of data like the victimization and self- report crime and delinquency studies. If, however, such empirical data are deemed necessary and useful to supplement crime statistics, their validity and reliability has to be established on both the national and the inter- national level. Undoubtedly, a periodic (every two years) repetition of these surveys is indispensable as well as the broadening of their scope (e.g. victimization of business, of museums, of EEC fraud etc.). In this arca the Research and Documentation Centre of the Dutch Ministry of Justice has pioneered®. 26 See, for instance, J. J. M. van Dijk/P. Mayhew/M. Killias, Experiences of crime across the eventer, Netherlands, 1990, World: Key findings of the 1989 International Crime survey, also J. J. M. van Dijk/P. Mayhew, Criminal victimization in the Industrialized World. 84 Calliope D. Spinellis and Maria Kranidioti In Greece it appears that no much change may be expected in the near future in the use of alternative sources and methods of data collection. Researchers, however, have already started collecting information on crime from additional sources. Perhaps, the time has come to consider police crime statistics as one of the sources of crime-reporting. Customs Authori- ties and Naval Authorities could be compelled to submit their statistics to the National Statistical Service in order to include this information in the volume "Statistics of Justice". These statistics would involve custom’s viola- tions, currency violations or drugs and arms trafficking via the sea (cross- boarder crime). Moreover, the Ministry of Environment and City Planning should also systematize its own statistical information on environmental crime and submit it to the National Statistical Service for the same purpose. Furthermore, data on individuals on bail or on "conditional release without bail" should be collected in order to give a more accurate picture of pre- trial detention. Data on recidivism should be enriched by statistics on parole and probation violations. For certain types of crimes, e.g. domestic violence, hospital records or private doctors report may be useful. The same is true for statistics provided by Insurance companies with respect to car thefts. As far as victim surveys are concerned in Greece only one preliminary victim survey has been conducted” and government funds have to be allocated in this area for further similar studies. Moreover, at least four self-report delinquency studies have been relatively recently conducted. Their scope however differs: Two of them focus on small samples and on delinquent behaviour®, one of them on a large national sample but it is centred on drugs and other substances and the last one deals with delin- quency in various schools of the country. Obviously, if victim surveys and self report delinquency are to be used as supplementary source of crime statistics in Greece, they have to be 27 C.D. Spinellis/A. Chaidou/T. Serassis, "Victim Theory and Research in Greece", in Kai- ser/Kury/Albrecht (eds.), Victims and Criminal Justice, Freiburg i. Br. 1991, pp. 125-128. 28 See C. D. Spinellis/E. Apospori/M. Kranidioti/P. Symiyianni/K. Angelopoulou, Key findin of a preliminary self-report Delinquency study in Greece: A Contribution to the 1992 fht ernational Study on Self Report Delinquency, 1993 [to be published], where relevant bibliography exists. Greek Crime Statistics 85 systematized and to be conducted on a national basis periodically. Needless to add, that primary source of crime statistics is the criminal justice "system". However, the other sources mentioned above are also useful but they cannot substitute official data at present. 5.6 International Comparability and Compatibility International comparability and compatibility of crime statistics are challenging enterprises but they strike against numerous problems. Crime definitions, methods and ways of collection, processing, analysis and presen- tation of data vary from country to country. The United Nations Crime Surveys are examples of meeting certain of these problems®”. Moreover, the Council of Europe, on the theoretical n30 level, held a discussion on "Means of Improving Information on Crime H and on the practical, an effort is made i) to publish a report on criminal justice statistics™, and 11) to ameliorate the European prison statistics through the project S.P.A.C.E.”. More specifically, the problem of definition of crimes is approached through broad classification of offences (e.g. offences causing death or injury, sexual offences, theft, robbery, damage to pro- perty, offences of fraud and cheating etc.)”. On the other hand, the Nordic countries suggest two methods of compi- lation of international data. The first method starts with an enquiry into comparability of legal definitions of crime in various countries, which is followed by an enquiry into comparability of the equivalent statistical 29 P. Tornudd in Heuni, Course on United Nations Criminal Justice Policy, Helsinki 1985, p. 132. 30 Council of Europe, Means of Improving Information, op. cit. 31 Personal communication with M. Killias on a "European Source-Book of Criminal Justice data", 32 Counseil de I' Europe, Conseil de cooperation penologique, "Statistiques Penales Annu- elles du Conseil de I' Europe" (Project S.P.A.C.E.), 1992. 33 C. M. Glinnie, op. cit., p. 44. 86 Calliope D. Spinellis and Maria Kranidioti definitions. The second, starts with the search through statistical publica- tions for those types of offences which appear to be comparable. The legal rules and statistical procedures applied in each country are documented and the final assessment of comparability is left to the user of statistics. The latter method in practice has been proved to be the only effective one™. This means that, if the ideal of absolute compatibility is left aside, there is hope for compiling and comparing international data. In any case, such a hope may find support to contemporary critical contributions in the field of Criminology®. Crime categories and some individual offences appear in a consistent form in the Greek statistics and problems of continuity” do not arise with respect to the last twenty years (see 1.) Hence, Greek data comparison with international data, according to the second method, should not be difficult. It should be though impossible to compile all data deriving from the various systems of justice. A prerequisite would be a choice of offences, similar to the choices of the Scandinavian project”’. Additionally, it should be mentioned that the Greek penal legislation is not that different from other European countries’ relevant legislation (e.g. provisions for theft, robbery, drug trafficking etc.). Therefore, it is expected that problems of congruence® with respect to international comparisons will not be ensurmountable. Nordic Statistics, op. cit. p. 5. 35 See Leavitt G.C., "Relativism and Cross Cultural Criminology: A Critical Approach", Journal of Research in Crime and Delinquency, v. 27, 1990, pp. 5 ff. and agg, 1" Context and Linkage. Reflections on Comparative Research and Internationalism in Criminology" British Journal of Criminology, Vol. 33, 1993, pp. 541 ff. 36 Continuity concerns questions of whether an individual time series (for example, registe- red robberies in country A) reflects the same legal and statistical content at all points of measurement and how possible changes will be perceived (Nordic Statistics, op. cit., p. 13). 37 The choice has been made after Interpol’s International Crime Statistics. Basic liberties were taken with the intended offence definitions. The overall goal was to choose as well defined categories as possible (op. cit., p. 8). 38 Congruence concerns the question of whether the data being analysed from each country are comparable (op. cit., p. 13). Greek Crime Statistics 87 6. Speculations on the Potential of Greek Statistics Criminologists in Europe as well as in other parts of the world share some common fate. Official statistics have gone through a severe critique and research has often relied on fragmentary information. Therefore, problems arising with respect to Greek crime statistics are not unique and the reader may find similarities regarding difficulties he/she faces in his/her realm of work. Besides the already discussed problems it should be men- tioned that there is lack of information in the statistics on very important issues like the economic and social cost of crime. One may assume that such information is almost non-existent in other countries’ statistics, as well. However, the Greek user of statistics, being a researcher, a politician or a journalist, could speculate on some improvement of official data, particu- larly in the area of police statistics. Police forces®, have already enriched their data with information on victims of certain offences. Additionally, in the light of technological developments, new techniques which would ensure the accuracy and completeness of evidence on crime could be employed by the police” in order to increase clearance rates. Yet, improvement in collecting and presenting data should be accompanied by changes in the criminal justice "system" and in financing policies. Besides that, formulating concrete suggestions on improving information on crime and on applying new techniques requires assiduous study of the systems’ operation and relevant research which is presently lacking in Greece due to limited economic resources. On the other hand, a more optimistic view would lead to speculations that set aside concrete problems of crime statistics. Greece is a relatively 39 The FBI is preparing to implement some important changes in the Uniform Crime Reports. These changes involve an attempt on gathering more detailed information on individual criminal incidents. Some departments will provide detailed reports on 23 crime patterns, including incident, victim and offender information. This will allow a national data base on the nature of crime, victims and criminals to be developed. Moreover, the definitions of many crimes will be revised. For instance, rape will be defined in neutral terms (Siegel, op. cit., pp. 72-73.). Such a new definition of rape is already provided by the law in Greece since 1984 (law 1419/84). 40 Stringent auditing techniques, such as those to be implemented for the Uniform Crime Reports (Siegel, op. cit., p. 73.). 88 Calliope D. Spinellis and Maria Kranidioti small and homogeneous European country. In principle, it could be sur- veyed easily compared to other countries which are larger and have a federalist structure. Contemporary developments on information technology and especially international cooperation and potential assistance by technical advisors from the United Nations leave some hope that in the future, positive changes might materialize in the area of crime statistics. Such changes, among other, should give the possibility of collecting, processing, analysing and presenting data on crime in a uniform way from the following sources of the criminal justice "system": The police, the procecutors, Judicial Councils, Courts and Custodial Institutions. The uniform way of collection of data from all these sources should allow observations of most of the flows from the criminal justice "system". Brief Review on the Situation of the Hungarian Crime Statistics Sandor Erdosi With this study my intention is to give a short review on the situation of the Hungarian criminal statistics especially on the statistical system I myself are working in. During the preparation of the International Congress in Budapest the organizers of the workshop asked the lecturers to concentrate in their contributions on some main questions. First of all I would like to give brief answers to these questions. After that, I will complement the short answers with comments with which my ambition is to refine the picture on the situation and give some motives relating to development ideas. The Questions and Short Answers a) Q Are there any statistics covering the whole area of police and criminal justice in the respective countries? At present, no integrated crime statistics system exists in Hungary which reflects the process of jurisdiction in a wider sense by the help of a common system of terminologies and data-processing procedures. b) Q Are there any partial statistics, and if so how are they connected? Are they compatible? Hungarian law enforcement agencies, courts and penal authorities operate their own statistics systems which are independent and incompatible. An additional statistics system relating to minor offences attaches to these three sub-systems of the official 90 Sdndor Erdosi statistics. In addition to and in completion of the official statistics (or statistical sub-systems) data collections of scientific purpose are performed aiming at the disclosure of the regularities of latent crime and victimization. c) Q What can be said about reliability and value of data of statistics? The inner validity and reliability of the statistics systems mentioned can be evaluated as good in general. d) Q What will be the future developments regarding survey modes and computerization of data collection and evaluation? How can the availability of the data be improved? From the point of view of information technology the sub-systems of the Hungarian Crime Statistics are of different level. e) Q How far can official statistics in future be replaced or complemented by victim surveys and microcensus? Under the circumstances available in Hungary victimology surveys can have only a supplementary role beside the official crime statistics. f) Q Is it possible to establish international comparability/compatibility of national statistics? We, from Hungarian side, also hold the development of a crime statistics system as desirable which would allow international - and mainly European - comparisons. Brief Review on the situation of the Hungarian Crime Statistics 91 Comments 1. To the question of integrated Crime Statistics (ad a, b) Relating to the question of integration it seems to be worth while differentiating between two trends of the development of crime statistics: one is the completeness of the measuring system and the other is the ability to reflect the complex law enforcement/jurisdiction process. The two directions of the development are motivated by diverging interests. The institutions responsible for the different phases of law enforcement and jurisdiction are interested first of all in particularizing data-circles connected to their own functions; a higher (whole society/nation-wide) interest, however, would need a picture about the characteristics of the whole process and its mode of operation be it harmonic or disharmonic. The rapid increase in crime has made this diverging trend obvious. Parallel with the increase in crime the demand for related information has also increased and information connections between institutions separated on the basis of the principle of the division of power have become more valuable. Recently one of the eminent home experts of the topic has written the following: "Until 1952 one of the organizational units with the biggest personnel was the one called then jurisdictional statistics. This responsibility was delegated at the time of the decentralisation of the judicial statistics partly to the Ministry of the Interior, partly to the Ministry of Justice and partly to the Prosecution. Since then, there is no adequate information on the jurisdictional process as a whole. (...) The judicial statistics was split between organizations." Part of the truth is, at the same time, that the most remarkable progress in the history of the home crime statistics until now was reached just about 1 Kovacsics, J.: The role of statistics in the state administration and jurisdiction after the change of the political system | A statisztika szerepe a rendszervéltozds utdn az dllami- gazgatdsban és az igazsdgszolgdltatdsban) - Magyar Kozigazgatds, 1993. No. 2 .p. 78. 92 Sandor Erdosi a decade after the date indicated. It was at that time namely, that the authorities accepted the existence of criminology and the so called Hungarian Unified Police and Prosecution Crime Statistics (HUPPCS) was introduced. This system obtained its present form after long professional debates and is standing in the cross-fire of debates also today. An important characteristic of the system is that it takes note of and classifies the reported crimes according to the bearing of a case contained in the prevailing Criminal Code. It has four basic observational units: (1) reported crimes, (2) suspected persons, (3) criminal proceedings, and (4) injured organizations or persons. Beside the other observational units mentioned the HUPPCS contains - among the firsts also on a world scale - the natural persons among the victims of crimes as well.? But we are not satisfied with this yet. We attach great importance to the victimological aspect of crime, therefore we intend to expand the data-circle relating to the victims during the next development process. The establishment of the HUPPCS - beside the enforcement of the criminological approach - was an important step forward to the integration of crime statistics. With the (re)unification of both independent systems of the police and the prosecution, much unnecessary parallelism could be abolished.? The integration - which started - became short after the introduction of the HUPPCS an item on the agenda and several working committees of different level dealt with it. The process, however, remained unfinished and this statistical system does, as indicated, include neither crimes judged by courts and the convicted persons nor the execution of sentences imposed. 2 T6th, T:: Contribution to a scientific seminar under the title: "Current questions of the development of home affairs informatics" (Hozzdszdlds "A beliigyi informatika fejlesztésé- nek aktudlis kérdései" cim tudomdnyuos szemindriumon) - BM. Adatfeldolgoz6 Szolgdlat 1990. (manuscript) Kovacsicsné Nagy, K.: The perspectives of the unified criminal information system (Az egységes blniigyi informdcids rendszer perspektivdi) - Beliigyi Szemle, 1987. No. 10. p. 41. 93 Brief Review on the situation of the Hungarian Crime Statistics The rocketing increase in crime - which coincided with the process of the change of the political system - took up again the question of streamlining crime statistics. In the present debates two standpoints are outlined. According to the first, the police is under a constant pressure of performance and therefore it needs up-to-date information on the development of crime on the basis of which it can "measure" and improve the efficiency of law enforcement activity. Apparently, this requirecment could be met by making police statistics anew independent. Actually, at this point the question arises if it is admissible in criminal statistics to prefer the requirement of up-to-dateness to the principle of legality. Present information technology gives almost unlimited possibilities to ensure for crime statistics to meet the requirement of up-to-dateness. Crime statistics, however, can only fulfil its original function if it meets also the criteria of legality. Legality is, of course, not an inherent requirement of crime statistics but also one of the cardinal principles of criminal jurisdiction. Following from its object, however, outer validity of crime statistics depends exactly on the fact in what degree judicial practice itself meets the requirements of legality in the given country.* One of the fundamental characteristics of criminal statistics systems is to what degree they are able to establish a balance between the requirements of up-to-dateness and legality by the help of practical compromises. The antagonism between these two requirements cannot be dissolved theoretically. In theoretical sense, the difficulties of establishing integrated criminal statistics systems arise from this. To introduce anew an independent police crime statistics would mean just the giving up of this balance and the preferring of the principle of up-to-dateness. 4 This statement holds truc only if one supposes a legitimate legal system. The examination of the legitimacy of the (criminal) law belongs, however, to the field of law sociology, or in a narrower sense, of the criminal sociology. 94 Sandor Erdosi This solution would, in our opinion, mean a retreat because - although it is true that the supervisional and legal guarantor role of the prosecution would prevail in the actual proceedings but the data from the independent police statistics would not go through this filter. Consequently, three types of crime statistics would circulate simultaneously: police, prosecution and court crime statistics, so different criminal statistics would overlap without determined contour as a picture on tv-screen tuned in badly. According to the other standpoint, the integration of the crime statistics systems yet separated should proceed. This solution would have several advantages at which only a hint can be made here. Not only would it give a more comprehensive view about crime and jurisdiction but it would abolish further overlapping and make for example measuring the organizations possible. The efforts for integration must be brought, however, into harmony with the requirements of a constitutional state. Earlier the experts conceived the establishment of an integrated crime statistics - or more accurately - crime informatics system by connecting the data files of the registers of the police/prosecution, the court, the penal authorities and the criminal registry. But this is limited by the principle of the separation of the branches of power and the protection of personal data. The Statistical Bill passed recently by the Hungarian Parliament prohibits the transfer/acceptance of individual data and permits the link of data files only within the same organization. The question arises whether the organizations of law enforcement and jurisdiction can be considered as organs with the same functionality. It is obvious that the integration conceptions should be thought over systematically in the light of the requirements of a constitutional state. We think that a coordinated development based on a carefully weighed plan would make an integration to a greater degree possible than today without violating legal guarantees. The real question is which interest will prove to be stronger: the separated organizational interests (the adherence to the usual, or traditional) or the common interest of restraining and repression of crime. Brief Review on the situation of the Hungarian Crime Statistics 95 2. To reliability and validity of data (ad c) It is very difficult to give short answer to the question relating to validity and reliability. The answer requires namely different approaches depending on the fact whether we refer the question to the whole system, one of the sub-systems or its variables. It can be here advert only to some general questions. ca.) The Hungarian criminal statistics is also criticised mostly because it cannot give a picture about latent crime. The proportion (of the totality or different parts) of the crime become known and latent crime is the archetopic of criminology which the Hungarian criminal statistics could not solve until now, either. The last developments of criminal movement make this question especially topical today in Hungary. A new phenomenon of Hungarian crime is that the increase in the number of known delicts practically stopped in 1992, and began to decrease in the last half-year. In the first half-year of 1993 the number of crimes become known decreased by 7%, the number of offenders become known by 9%, the number of injured become known by 14% and last the number of crime cases decreased by about 16% where police concluded investigation as compared to similar period of last year. Two further circumstances are worth special attention in connection with this. First is, that the decrease in the numbers of delicts has begun not in the last half-year. (See to this Figure 1.) It emerges clearly from the quarter time-series that crime become known was in a descending line already in 1992. But it was then concealed by the aggregation of data on year level and its comparison with 1991 data which represented the end of an ascending line. The second is, that the HUPPCS record violations of norms at the closing of investigations, that is, these crimes were committed in periods before their becoming known. From this, the consequence follows that the Hungarian crime wave broke already in 1991. All this should be taken into account as one of the fundamental facts of Hungarian crime and law enforcement. 96 Sandor Erdosi It is unfavourable, however, that it is rather difficult for professional public opinion to take notice of statistical data relating to the decrease of known crime. What could be the reason for this? I believe the reason for this is, on the one hand, an uncertainty about probable development of crime trend and on the other hand doubts concerning the validity of the data of crime statistics. Prognosis of crime would mean practically a forecast of social, economic and political conditions which determine crime. Today, however, branches of science and political programs dealing with this undertake, at best, to outline heuristic scenarios. It is a fact that the moderately decreasing trend of crime become known for law enforcement agencies lasts for at least one and a half year. In spite of this no grounded answer can be given today to the question if in the near future further increase, stagnation or decrease of crime should be reckoned with. If, remaining strictly within the circle of linear statistical extrapolation, we wish to give a forecast on the probable number of crimes becoming known, then (1) we should decide whether we choose the streching out of a more lasting but more unfavourable or a shorter but more favourable trend and (2) we should take into consideration seasonal fluctuations in both cases. Regarding the above mentioned, we have made two versions on the extrapolation of the number of crimes becoming known every quarter: with stretching out of the trend of fact-data M in an optimistic version for the period between I. quarter of 1991 and IV. quarter of 1992 and ) in a pessimistic version for the period between 1. quarter of 1988 and IV. quarter of 1992. Brief Review on the situation of the Hungarian Crime Statistics 97 Figure 1.: Number of crimes become known in some quarters between 1988 and 1992, and its linear extrapolation from the first quarter of 1993 (on the basis of additiv seasonal model) Figure 1 shows the probable values of the number of crimes becoming known every quarter on the basis of both types of linear extrapolation. In both cases trends forecasted were corrected with seasonal deviations (calculated on the basis of an additive model). The number of crimes become actually known in the first quarter of ’93 was lower by 8% than it was expectable on the basis of the pessimistic extrapolation (122.512) and exceeded by 3% the value extrapolated optimistically (109.203). Crime become known in the second quarter, however, remained below by 25% of the volume indicated for this period on the basis of optimistic extrapolation (130.851). So from quantitative point of view, crime become known took shape in the first quarter of 1993 on a level between optimistic and pessimistic estimation and in the second quarter on a level well under the optimistic estimation. These data can be explained from different aspects in different ways. From statistical point of view, however, it is not only the increasing pace of crime which has broken, it is not the stagnation of crime’s volume but it is a definite decline. The question relating to the development of crime in our days is also not whether increase has continued in the last one and a half year but the one: how long can decrease in the volume of crime last. It would be naturally carly to speak of the formation of a new trend line or "decreasing path". It seems to be possible rather that decrease experienced since one and a half year will stop in the near future and the volume of crime will become stable on a lower level as compared to that at the turn of the decade. I think that both research workers and leading experts of law enforcement should pay greater attention to these data. If data are valid, therefore, if not, then therefore. These data refer namely to the part of crime become known for the authorities and they are not conform with the perception of the public about crime and the press coverage of crime. 98 Sdndor Erdési It can be declared of the official crime statistics system of a given country that it reflects the state of real crime if law enforcement agencies operating in that country are informed about the activities violating penal norms. In the contrary case, namely, statistics give a picture not about the activities which violate prevailing legal norms and its offenders but only about the information of the organs of law enforcement and criminal jurisdiction relating to this. Criminal statistics, as one of the indirect means of social knowledge is, in itself, not destined for and not able, either, to give compensation for deficiencies in this field. It is partly the reason why some of our experts suppose that the (perhaps temporary) stagnation or decrease of known crime does not reflect the culmination of actual crime. It is possible namely, that merely the rate of latency has increased. This supposition contradicts criminology experiences relating to the relative constancy of latency-rates, still it cannot be excluded in advance. There are namely two possibilities. One is that latency-rates did not change significantly in the period under review. In this case criminal statistics give a reliable indication about the decrease of the volume of both crime become known and actual crime. The question remains, however, where is the threshold value of difference at which this decrease can be perceived also by public opinion. The other possibility is that latency-rates have grown significantly in the past few years. In this case, present criminal statistics indicate only the decrease in crime become known reliably but do not reflect stagnation or increase in the actual volume of crime. The reason behind this should not, however, looked for in the reliability of statistical methods but in the fact that an increasing part of the actually committed crimes does not come to the notice of law enforcement agencies. One of the conditions of the choosing of a successful law enforcement strategy would be, therefore, the methodological tracing of the changes in the latency-rates. This strategy should be directed, namely, in the first case towards the stabilization of the state reached, in the second, however, 99 Brief Review on the situation of the Hungarian Crime Statistics towards the reduction of the increasing latency; especially that the increase of latency has in itself a stimulating impact on crime. Possibility of increase in latency would deserve additional greater attention because the reasons behind this can be - among others - (1) the adjustment of the population to a higher risk level and (2) the so called selection on the part of the authorities. I think it is not necessary to stress that in both cases greater attention should be paid to the connections between law enforcement agencies - first of all the police - and the citizenry. ¢b.) As second, a deforming impact deserves attention which is, in my opinion, in connection with the reliability of the data of the HUPPCS. The direction of this impact is just in contradiction with what changing rate of latency can exercise on the validity of crime statistics data. One of the functions of the Hungarian criminal statistics is, namely, to provide data to the measurement of the workload and success of the police authorities. As a consequence, some local authorities tend to give a more favourable picture about their practices. I have to note that efforts have been made in Hungary for a longer time to measure efficiency of police work but no special measuring system exists, such as the PPPM (Police Program Performance Measures), for instance, proposed by O’Neill W., Needle J. A. and Galvin R. T. in 1980.° Therefore police leadership uses - among others - statistical data of the HUPPCS for this purpose. Naturally we themselves do not hold this as absolutely adequate and know professional doubts arisen in connection with this elsewhere.’ The mentioned effect to the opposite direction is in connection with the measurement of efficiency of police authorities - and within this - the S O'Neill, M. W. - Needle, J. A. - Galvin R. T:: Appraising the performance of police agencies: the PPPM system - Journal of Police Science & Admin., 1980. No. 3. 6 See for instance reference of Walker to the inflation of different efficiency indexes. Walker, M. A.: Do we need a clear-up rate? - Policing & Society, 1992. No. 2/4. pp. 293-306. 100 Sandor Erdosi application of the so called omissions of investigation. As to the omissions of investigation, under our national regulations, investigative authorities may apply omission of investigation if in the same case investigation extends to other more important crime(s) as well as compared to which crimes with omissions are of no significance from the point of view of impeachment. Since 1988 omissions of investigation were calculated as successful investigations. After the transformation of the evaluation system then, their number continuously increased from year to year. Figure 2: Number of proceedings ended with omission of investigation between first half years 1988 and 1993 As a consequence of mass occurrence which was to be observed from January until March this year, they constituted a share in relevant investigations which deformed figures on the efficiency of police proceedings. In order to eliminate this deformation, from the first half-year of this year on, when calculating efficiency indexes - both at numerator and denominator - proceedings closed with omissions of investigation are disregarded. Naturally there are debates concerning all this within local (county level) police authorities but correction was made with the support of the Chief of the National Police. As a consequence of modification, efficiency indexes of the police became naturally lower by some per cent (efficiency of investigation by 6%, clear-up index by 9%) and a more moderate increase is to be seen in the efficiency as compared to previous periods. We think, however, that corrected indexes are nearer to reality than the previous ones. Figure 3: Investigation efficiency index (left) and clear-up index (right) together with omissions of investigation and calculated in a corrected way between January and June 1993 Brief Review on the situation of the Hungarian Crime Statistics 101 The example of the omissions of investigations shows, in my view, what dangers it contains if official statistics are used not only for getting to know the crime situation and criminal movement in a country but at the same time for the measurement of the successfullness of police organs. In our national relations we cannot meet this demand by the help of other means yet. According to our experiences, however, effects of this increase rather than decrease the volume of known crime. Purely from statistical point of view this is possible because (at least the Hungarian) crime statistics handle crimes of different weight, that is serious crimes and bagatelle offences, together. This makes it possible for investigators with more ambition than the average to carry on paper or magnetic data carrier investigations or omit them in petty offences. The essence of the problem is here, however, also not of statistical character. The question is actually, how should law enforcement and criminal jurisdiction handle the more serious and less serious forms of crime during a crime boom which has been characteristic of Hungary in the last years and is also today. The competency of statistics is limited to the prevention of the deforming impacts similar to the above mentioned by the help of control of data-supply. I can only mention in this study questions, like: whether weighting of crimes would be necessary?; if this weighting could be completed in an overall crime statistics?; if the weights themselves should be established on an empirical or axiological basis? and how a weighting like this would react upon criminal political practice? cc.) At the examination of the time changes of crime the frequent modifications of criminal laws and sentencing practice - compared to the international practice - cause great difficulties. As a result, temporary fluctuations and lasting changes occur in the volume of (some sectors) of known crime. In factual analyses temporary increases or decreases can be 102 Sandor Erdosi smoothed by the separation of formal and material factors which evoke fluctuations.’ The filtering out of longer - sometimes decennial - cyclical effects is a harder task. The alternations of the sentencing practice between restricting and expanding interpretation of cumulativity are the examples for this. According to some presumptions the number of the registered violations of law is about 10% higher than the number of actual (material) violations of law as a consequence of the sentencing practice.® Apart from the fact that this presumption awakens some natural law reminiscences it is hardly controllable. The parallel classification of the law violating acts according to a set of norms supposed unchanged would, namely, require the doubling of the activity statistics at the least which is hard to execute. An international criminal code would give, however, a good base of comparison for the filtering out of similar turbulencies. cd.) Finally it should be mentioned relating to the question of accuracy that no important deformations of political purpose have been made in crime statistics in the last 2 - 3 decades. The characteristic of the past state socialist system manifested itself rather in the application of the criminal law as indirect means of politics. 3. To information technology (ad d) Data-processing in all the three sub-systems mentioned is executed by computers: at the investigative authorities and the courts on mainframe and at the penal authorities in a PC environment. 7 Kovacsicsné Nagy, K.: Measurement of the effects of the modifications of statutory rovisions in crime time-series (A jogszabdlymédositdsok hatdsdnak mérése a binozési idOsorokban) - Beliigyi Szemle 1982. No. 9. pp. 32-41. Rozsa, J.: Development of crime from the standpoint of the statistician (A bnozés alakuldsa a statisztikus nézOpontjdbdl) - In: Studies on social adjustment problems, Kossuth 1988, pp. 94-118. Brief Review on the situation of the Hungarian Crime Statistics 103 The sub-system operated by the Ministry of the Interior and the Police can be considered the relatively most sophisticated one. In this system data processing has been performed centrally by electronic computers from the very beginning. At present, adequate hardware and software conditions are available for the county authorities for the processing of their own data. For the sake of supervision and information of the state management we wish to maintain central data-processing. For this, a software is available which was prepared 10 - 15 years ago. This program cannot, however, cope with increased demand. Therefore we plan to lease a fourth-generation statistical or information management software system by which applications can be developed for use both in mainframe environment and at the workstations of the local area networks connecting to the mainframe. Data are entered decentrally by the informatics departments of local police authorities. The development of a nation-wide area network is near completition. It make yet possible for data entered to be transferred on line. Moreover this network make possible for the workstations installed at the central and local law enforcement agencies to access to the data-bases containing the results of the nation-wide data-processing. At present, about 100 workstations or rather local arca networks are operating from where the users can get the results of the monthly processings within 10 days. 4. To alternative measures (ad e) The Hungarian criminal statistics - which can be now called traditional - measures the known crime totally. Consequently the objective of victimology survey can be - in addition to the examination of the phenomenon of vic- timization - indirectly the assessment of the degree of latent crime. This would be very important from several points of view. Systematical surveys would be necessary by applying adequate methods suitable for learning possible changes in latency-rates. I note at this point that valuable victimology surveys have been and are carried out in Hungary but not usually and often enough. Therefore we do not have victimology 104 Sandor Erdési time-series which could verificate or falsificate the assumption on the increase in latency. For lack of registration and verification, victimological surveys can undertake only assessments. The alternative of a survey of a few thousand and a microcensus is, from this point of view, a technical or financial question. A microcensus carried out as part of the official statistics would guarantee a smaller rate of those refusing answer. 5. To international developments (ad f) We deem the establishment of a compatible international crime statistics system possible only on the middle term, mainly because of the differences in the legal systems (nomenclatures) and the lack of institutional/organi- zational conditions. Hungary is interested, at the same time, in the success of the initiative aiming at the harmonization of the national statistics systems and the establishment of a joint data-base. Therefore we are prepared to take part in the exchange of experiences and ideas and to contribute to the activity of a preparatory expert committee. Brief Review on the situation of the Hungarian Crime Statistics 105 Thousand 200 A A 150 — 100~ -------- o 50 — — Become known 4 Optim. estimate A Pessim. eslimate * Fact - I-1l. 9. 1993 T T T T T T T T T T T T I T T T T T T T T T T T T — 1929 39 49 19 29 3q 49 19 29 39 4q 19 29 3q 49 19 2q 3q 49 1q 29 39 4q 1q 2q 3q 4q 1988 1989 | | 1990 | 1991 | 1992 | 1993 l 1994 Sdndor Erdési 106 30000 q ” 6 25000 A Y/ 20000 —: . S 2 15000 = i 10000 3| g g 3 5000 o oP \Q@ > \q% \‘5@ 2 & Ky ’\q‘b \q& @& K o> O > X o Brief Review on the situation of the Hungarian Crime Statistics 107 % 100.0 it 90.0 80.0 70.0 60.0 50.0 40.0 1 30.0 20.0 10.0 0.0 Jan Feb Mar Apr May Jun ;Without omission | 43.2 44.7 52.7 48.3 51.2 62.0 Together w. omis. []| 43.8 52.6 68.9 49.4 52.5 63.3 108 Sdndor Erdési % 100.0 M without omissions K it Together w. omis 90.0 80.0 1 : 70.0 60.0 50.0 3 . A 40.0 1 ......... A e 5 P e o 30.0 20.0 10.0 0.0 Jan Feb Mar Apr May Jun Without omissions 31.7 30.8 43.7 38.0 38.7 48.1 Together w. omis. 32.7 43.2 66.0 39.7 40.7 50.5 Statistical Data Collections on Crime and Criminal Justice in ltaly Armando Caputo General remarks Statistical data collections on crime and criminal justice in Italy are carried out by the National Institute of Statistics (ISTAT). These data collections record those events violating criminal law which have been reported to the authorities responsible for social control, as well as in- cluding some personal characteristics of the persons responsible for these violations. The main data collections are on crime; reported offences; sentenced persons; population in custody. The concepts of real, apparent and legal criminality are present in the quantitative study of crime. 1. Real criminality is crime which is effectively present in society and results from the total number of offences committed, whether these have been reported to the authorities or not (hidden crime); Apparent criminality includes offences made known to the judicial bodies regardless of the subsequent steps take, if any; Legal criminality concerns offences which, having been brought to court, have resulted in a sentence being pronounced by the judge. 110 Armando Caputo In general, these data collections focus upon aspects of apparent and legal criminality given the difficulty in quantifying hidden crime, especially where no victim is involved. However, ad hoc surveys such as the Multi- purpose Household Survey carried out by ISTAT are used and the wide range of social-demographic data in this sample survey enables us to analyse some aspects of victimisation caused by events which took place in the twelve-month period prior to the survey and experienced by some members of the family interviewed. Despite the fact that it provided a sample estimate only and this only for specific offences against the person or property, this survey indicated the extent of "hidden" criminality, that is, unreported offences. During the last phase of this survey (December °89 - May ’90) it was estimated that 47.0% of criminal events were not reported to the authorities. In order to provide a more in-depth analysis of victimisation there are plans to widen the scope of the Multipurpose Household Survey as well as that of surveys from administrative sources (concerning reported crime). In general, although this is not absolutely true, during normal periods crime surveys can be considered to be at least representative of the reality being investigated, given the similarity of the trend of offences committed and offences reported. When analysing the data, one must bear in mind that the number of crimes reported depends on the efficiency of the police force and the courts in crime prevention and repression activities as well as the various criminal laws in force. Also, as the crime figures are based on the information supplied by the respondents, they depend on the promptness or otherwise of this information and its completeness. There are data collections right across the field of criminal justice. However, the different data source are not directly comparable; different judicial bodies, police forces, etc. have different counting rules; there are also varying survey methods and varying data reference times. In the course of the last ten years, the processing phase for criminal justice statistics in Italy has been entirely computerised. Changes and updates are systematically introduced following the frequent changes in the 111 Statistical Data Collections on Crime and Criminal Justice in Italy law with the aim of highlighting certain topical aspects of crime (foreigners, minors, etc). The use of magnetic media has been widely developed for the storage of basic data and the dissemination of the information produced. The latter is carried out using both the most up-to-date means of communication (Videotel) and the means of communication reaching the largest audience (press release). It is very hard to compare crime surveys internationally, even among countries with economic and social similarities, due to the considerable differences in classification and collection criteria, in the criminal laws in force and in the elficiency of the survey systems. Reference is frequently made in international comparisons to such offences as murder, robbery and violence, but other aspects of crime which are even more dangerous and damaging to democratic society are often neglected. Despite these difficulties, joint work between the researchers and operators in this sector in various countries could make it possible for new fields of investigation to be developed and could be useful in helping to define a basic list of major crimes, specifying nature and legal content as far as possible, and at the same time activating various procedures for close collaboration in order to reduce the risks of mistakes or misinterpretations. Measurement of Crime Time and territorial series regarding crime are generally based on the monthly ISTAT survey on reported offences for which the judicial authorities have commenced penal action. The data concerns violations of the penal code, excluding fines, and the persons responsible for these violations. The main unit of study is the reported offence which is analysed according to type of offence, the date on which and place in which it was 112 Armando Caputo committed, the source of information and the number of people reported plus their sex and citizenship. On the basis of the current criminal procedure code, which came into force on 24th October 1989, in the case of offences by known offenders, penal action begins as soon as the Public Prosecutor formulates the charge, according to Art. 405 of the criminal procedure code, against the person being investigated. There have been no changes as far as offences by unknown offenders are concerned, but penal action is considered to have begun upon registration of the offence in the "Unknown Offenders Register". As far as the disaggregation of the offences is concerned, the rule used for the survey method is that concerning the single items corresponding to the same number of crime categories foreseen by the Penal Code, as well as some specific categories foreseen by the main special laws concerning crime (drugs, arms, smuggling, etc). Although useful for special study pur- poses, this classification, called "analytical classification" by ISTAT, is too large for practical purposes. For this reason a "reduced classification" of offences is adopted in ISTAT publications, grouping the single items in groups and sub-groups. Offences Reduced Classification Against the Person Against life Massacre Murder Attempted murder Infanticide Manslaughter (voluntary) Manslaughter (involuntary) Against personal safety and freedom Assault and battery Deliberate personal injuries Involuntary personal injuries Brawls, etc Violence, threats, etc. Abuse and slander 113 Statistical Data Collections on Crime and Criminal Justice in Italy Against the Family, Public Morality and Decency Against the family Failure to pay maintenance Ill-treatment within the family Bigamy, incest, etc. Against public morality and decency Rape Indecent assault Indecent behaviour Corruption of minors, etc. Instigation and exploitation of prostitution Illegal abortion Against Property Larcenies Theft Burglary Robbery, extortion and kidnapping Robbery Extortion Kidnapping Damage to objects, animals, land, etc. Swindling and other fraud Swindling Embezzlement Receiving of stolen goods or property Fraudulent insolvency, etc. Against the Economy and Public Trust Against public economy, industry and trade Trade fraud, etc. Sale of non-genuine food Arbitrary entry into companies, etc. Bankruptcy Issue of dud cheques Against public safety Illegal drug manufacture and dealing Against public trust Counterfeiting of money Counterfeiting of seals False impersonation 114 Armando Caputo Against the State, Other Social Institutions and Public Order Against the image of the state Against the public administration Assault, resisting and insulting public official Peculation, embezzlement, etc. Failure to fulfil official duties, etc. Against the administration of justice Against religious feeling, etc. Against public order Other Offences Other ISTAT Data Collections In addition to the above-described survey of reported offences for which penal action has commenced, a further source of statistical information in this field is the ISTAT data collection on reported offences carried out with the help of the police forces and together with the Ministry of the Interior. This data collection concerns offences reported to the judicial authorities by the State Police, the Carabinieri, the Guardia di Finanza, and, unlike the data collection on crime, does not include offences which have directly reached the judicial authorities by other means (private, public officials or civil servants not forming part of the police force, judiciary, etc.) while including those offences, which, although reported, do not lead to penal action. The two data collections not only differ in their source but also with respect to the moment when the offence is registered, in as far as penal action may begin some time after the offence is reported. It becomes clear that despite the essential differences between these data collections, a suitable comparison of the data contained in them could provide interesting conclusions which would not otherwise be available if data from a single source were considered. A further data collection carried out by ISTAT in the field of crime is that on sentenced persons which concerns persons convicted or acquitted in a given period following a final judgement resulting in a sentence. 115 Statistical Data Collections on Crime and Criminal Justice in Italy This collection provides a number of interesting variables concerning the person being tried (age, marital status, education, professional status, etc.) not taken into consideration in other data collection, and which are widely used for the carrying out of more in-depth studies of crime from a social point of view. The phenomenon of crime can also be studied by means of collection for preventive and penal institutions which study the entry and exit flows into and out of preventive and penal institutions and the number of detainees present at the end of every month. The preventive and penal institutions are divided into the following categories: a) Preventive custody institutions b) Institutions for serving sentences ¢) Institutions for the application of security measures d) Penal institutions for juveniles. With some exceptions, accused persons awaiting trial, convicted persons, detainees, young detainees are assigned to the above-mentioned institutions, respectively. Data on the reasons for entry into and exit from these institutions and the number of prisoners present at the end of each month is supplied to ISTAT by the preventive and penal institutions, while data relative to characteristics of the prisoners (e.g. age, sex, citizenship, marital status, profession) are transmitted to ISTAT by means of magnetic media by the Penitentiary Administration. Finally, the Court Surveillance communicates data to ISTAT concerning alternative measures to imprisonment. fig ot T 7 { RS o 0y 2 Bt 9 5 W o B 8 o i e i 4 S s Nl i T o 740 i Ty e o g i e 5 ol e e 5 i a o & % =4 i i At ey A Li > ko) GE Wik ) ; 3 % o b o & kéd 8 e X i (58 7% b i i > 5 4 o, . P i A Ry A0 o b = K, s 4 Ry e B a a; . %30 ‘v, {44 o e i e e 3&! i S b gie i i s i k {7 (.3 ) i L I feh 3 MEY 5 ‘:‘ i) 2l e i i = ‘éfl K {0 ; bed B ) o & i i 5 Al he 2 gy i 3 Ve =08 4 2 .’ b o s e &% & bt B e & Wi i L s e e e - k! £ e e L iE. S8 v % o 4?.‘ ey B % i s b b s o b ; e e e 4 i P 2y e i vl % (¥ A 4ehd 5 % i i Garisl 7 ity i 4 2% 2;- A ] & & i oy 5 e s e i o g £ Y (3 2 g p i v gk b i i B e el e B A i i P A g i ol G - .: ) iy 45 o & ok b o e i b <& Ex s 0 s 5 s b, o or o b N o e e el e i £ =) A st S, o ] i ¥ SR e e i * " 0 b . Y e 2 & ekl n £ £ o 0 o L b L i g e e 2 e B s ¥ ) B I e e E e b P b B . £y 1 o ol 8 " i e iy i i b [ %{ e e i & Fa i e . b ;i 555 3 ) i A K o o s ) i chun N2 AR o b I v i (¢ i et o o R i * s t. A T e 3 e hs g t % i & - i e EX ko Yk 2 g f;“‘: i RS e e i o 4 4 (s ok i & g b 3 @ b h i et b % % s T o A ge o o0 S e o i & 4 i s i it 7 it gy i % i o i 4 i 5 % % o i X 3 2 i i Ty P 4 B s ) Y o} o o ) i B it 50 e % :2.‘. [ bt \: b PNiss i s G4 o o o i ¥ i i 5 3 v % $hs e ) o ooy 3 5 N i 53 2 i £ % i . o i Ry oy e I s Loy é % i o t« S i o i 5 ¥ ! e X X i o il i i 5‘ i 4% o i s e i . 2 5 5. z e g b el " bo i - o ) . }fig : i i g Aol o 5 Jona Je 4l A3 i o} ey e ; . 21 ol Aio 5 « o 0 i gy di AR (& b i e & = i3 S ¥ v £ o S i i ey A o ! i i) LT 8 hE 4 7% 5 i B i 2 i 4l t‘v, it s i .v: k! R » ¥ i o £ e (i e i g i ;:L’im il i i\ 0 it =0 5 A o I ot 4 - b & i e a i ; s S 5% Hh i o b o ey ey hi i el A il j? 2 i o s ;; Ak v 7 AL X i o 5 e 3 s Py o % = & L i 5 iz Q; 4 s : 5% 4 00 i A e e o i % % g b 2 o 5o 6 (] % 3L 2 4 T A i a4 ok ok i e i o & is & - P s g0 §i iks ks o ? ey ¥ S o i Hy iy Sk R o o A ? 5 [ e e o o b S 5 o L gon b b :‘, 5 e i s el : : b ',:: o) i ot 5 g i i i r % s Fon o 4 i Wfiv‘ o L 4 " 9 3 . o st ar W 2 L S ot e i i S i A u% ":r;‘ iy B o 21 o i s P s oo, i i . 4 ] i 2 b ¢ £ b3t s e o : e : i o S Sk i e B i i ke é}?':"‘ . 4 e e o et = L R49 & G f A% & & i B i & e, i i R e 1 Ay a0 s A 0 . % b i . i s Ju i . : b k% %) e Wt . e i = i b i i - i i e 4 i % 5 rfl‘ g : % T &, 5 i 5 Nak @ . b 54 -?‘ A e Ca ¢ e w 5§ & e & -{ &) i o v £ A ey & 5 e A €5 A v e 0 L s ! i Sy ) ¥ k% g i i 2 i o gt e {3 s 3 P ) A i g Lot be i e R g i b o e i R 1% v 7 % it i e i 3 i c £ En w i o e " i id i o k2 % s 4 » b o ¥ B ks ) (1 i 5 g B i o 5 % é § v e v i e g = 54 : & s i s i g Y " i s Y & e N e e e E i 2 o 4 & L : . aie % Ty i ) W £ A & A L i et i s i i EAs ) N g A ..v (e i H L% e . i 5 E‘;'. B b ey i e i § . o i i e vy s e 4 i s o ¢ il s B & i) i 5 i i b g 5k B & o b L e i B s M, 20 % 2, 14 i Ea g z e o b i e o b &5 SR i % e il 0 i 0 0y r {, % Sat?) b i B i £ A 3 Fm 4 t5h e, e < s kL % o e i iy ke e 15 g at 5 o o s b s ! : 7@ A . e i & i e b o T o . et & i 7 e ¥4 < ol ; 7] n i - i " e 0 i 1 o it o e = i aaa) o E . g . 1 o o e ik e Wi 0 ) & ey i i e 5 G pr i L) A% 5 b st o (LR 4 B S 5 o &% i % e i it i i iy ) o ’} it it e i Crime and Criminal Justice Statistics in the Netherlands Max M. Kommer' The Netherlands have a long history in compilation and publication of crime and criminal justice statistics. In fact, the first publications date from 1836 (prison statistics) and 1847 (criminal justice statistics). From the year 1900 on, these statistics have been the responsibility of the Central Bureau of Statistics (CBS); before that year they were compiled and published by the Ministry of Justice. Scope, definitions and methods of data-collection have, of course, changed over the years but long running time series of basic statistics are available. In this paper I will describe the present-day situation: coverage, reliability and validity, availability and foreseeable developments. Coverage At present, Dutch crime and criminal justice statistics cover the whole spectrum from victimization to imprisonment. There are, however, no statistics that provide an insight in the flow of cases or people through the criminal justice system: victimization data from the regularly held victim surveys are not compatible to the reported crime rates from the police statistics (cf. Geerlof, 1987, p. 32), the number of cleared cases/offenders from the police statistics is inevitably different from the number of cases processed by the public prosecutors office and there is no connection between the court statistics giving the number of unconditional prison sentences and the prison statistics giving the number of prisoners received. There is a number of obvious reasons why such an over-all coverage is 1 The author is acting head of the research team on legislation and administration of law in the Research and Documentation Centre of the inistry of Justice. 118 Max M. Kommer absent: victimization, police, prosecution, court and prison statistics are all produced with different goals, by different sub-departments of the CBS, and they have different counting units. The Victim Survey was held bi-annually by the CBS between 1985 and 19892 It refers to sixteen types of crimes (see appendix 1); the sample is apr. 4.500 persons aged 15 years or older (on a total population of over 14 million in 1993). The 1983 and 1987 Victim Surveys also included items on civil legal problems encountered by respondents and their recourse (cf. Huls e.a., 1991) and on offenses committed by the respondents themselves (cf. Geerlof, 1988). Currently, the Victim Survey has been replaced by the Justice and Security Survey, which also includes items on civil and administrative legal problems. Apart from the national surveys there are two other sources of information on victimization: local victim surveys and the international victim survey in which The Netherlands participated from the beginning (Van Dijk e.a., 1990). Police statistics are published quarterly, semi-annually and annually. They cover crimes and misdemeanours that have been reported to or detected by the municipal and state police or the Royal Marechaussee’. All police corps® send a monthly report to the CBS, in which they state among others the number of crimes they registered (broken down into a large number of categories, see appendix 2), the number of known suspects and the number of suspects taken into custody (both broken down by sex and minor/adult). Statistics regarding the Public Prosecutors and the Courts are published yearly. They cover all cases dealt with by the Public Prosecutors and/or the Courts, including those that were reported by non-police investigation Between 1974 and 1984 the Victim Survey was held yearly by the Research and Documen- tation Centre (RDC) of the Ministry of Justice. The Royal Marechaussee is the military police force; it assists the municipal or state police where and when necessary and it is responsible for border control. formerly about 165. Presently 26, of which 25 regional and 1 national (the Royal Marechaussee excluded); Crime and Criminal Justice Statistics in The Netherlands 119 departments (for instance the economical offenses investigation department of the Ministry of Economic affairs and the fiscal fraud unit of the Ministry of Finance). Data on criminal cases is collected on an individual basis: as soon as a case is received from the police, the Public Prosecutor’s office sends an information sheet to the CBS.” These information sheets, which contain particulars of the offence® and the suspect, are followed by reports on the disposal of the case: by the Public Prosecutor or the District Court.’ In the event of an appeal, a report on the subsequent disposal(s) is sent to the CBS by the appropriate Public Prosecutor’s office. Public Prosecutor and Court statistics cover a large number of items. Not all possible (cross-)tables are published, but they may be made available on request. In the publications some tables show a breakdown by type of offense (see appendix 3); this breakdown differs from the one used in the police statistics. Data on misdemeanours and economic offenses are aggregated and reported quarterly. Prison statistics were, until recently, based on yearly reports by the apr. 60 penal institutions. These reports covered capacity, staffing and prisoners. The yearly published Prison Statistics provide information about the type of detention (remand or prison sentence), population on the first resp. last day of the year, average population, flow of population as well as particulars of the prisoners (sex, age-category, provenance). Nowadays prison statistics are based on individual reports of prisoners admitted / released. S A caseis something different from a crime. A crime that is completed by two or more suspects will result in two or more cases; multiple crimes completed by one suspect may result in one or more cases. When a case consists of multiple offenses, only the three most serious ones are mentioned in the information sheet and subsequent reports. One case may result in multiple reports, e.g. when a case is transferred from one district’s Public Prosecutor’s office to another. Multiple reports like these may or may not be linked or excluded in the official statistics, depending on the nature of the table. On the other hand, cases may be combined when brought before the judge; in that event, there will be n-1 reports of the Public Prosecutor’s dealing with the case and 1 report of the Court’s dealing with the combined cases. 120 Max M. Kommer Reliability and Validity Reliability and validity of victim surveys have been extensively discussed over the years (see e.g. Dijk, 1990); there is no indication that the Dutch victim survey as conducted by the CBS differs in any respect. Reliability and validity of the Police, Public Prosecutor, Court and Prison Statistics are probably not as high as might be understood from the above description of data collection procedures. Although the CBS takes a pride in making the data as complete and accurate as possible, using several audit and cleaning procedures, data quality, at the end of the day, is in the hand of people within the criminal justice system who may not always be very motivated in completing the CBS-forms accurately. Over the past two decades, various statistics could not be published or are known to be unreliable due to data-quality problems or incompleteness of data: there are no Prison Statistics over the year 1974, in the early nineties one state police district refused to send in their reports and Public Prosecutor and Court Statistics over the year 1992 are probably corrupted as a consequence of the introduction of an electronic data-processing system in the Public Prosecutors’ offices, which led to the loss of data on an unknown but substantial number of cases. In general, however, reliability and validity of the various statistics may be regarded as being high. A special type of validity is the usefulness of the statistics / the data they are based upon for criminological research and policy making. Here, validity might be considered less than desirable. One important factor is the usage of a crime typology based on the Penal Code (and other relevant Acts) rather than a more criminological or criminalistic one; an other is the impossibility to follow cohorts of cases / defendants through the different stages of their processing by the criminal justice system (even the Public Prosecutor and Court statistics, though based on reports that can be linked, are input and output statistics rather than flow statistics). Availability All statistics mentioned here are published by and available from the Central Bureau of Statistics. Most statistics are only available in Dutch, but Crime and Criminal Justice Statistics in The Netherlands 121 selected tables are included in the yearly "Statistical Yearbook of the Netherlands". Notes and short articles on Dutch crime and criminal justice statistics appear frequently in the quarterly journal of the CBS "Netherlands Official Statistics". Currently, police statistics are also available on floppy disk (cf. Kalleveen, 1992, p. 29). Special selections and/or tables can be produced for scientific or policy purposes; it is even possible to obtain the (anonymized) micro-data of the Public Prosecutor and Court statistics. The department of Statistical Information and Policy Analysis of the Ministry of Justice makes selected and annotated national and regional statistics available to policy makers, scientists and the general public. National and regional statistics regarding crime prevention, victimization and policing are (also) available from the Directorate for Crime prevention of the Ministry of Justice; statistics regarding the prison service and imprisonment are available from the Prison Department of the Ministry of Justice. The Research and Documentation Centre can play an intermediary or referring role. Addresses can be found in appendix 4. Developments in the Near Future Over the last few years, efforts have been made to develop modules for the police and public prosecutors’ electronic dataprocessing systems that will automate the reporting to the CBS. It may be expected that these modules will become fully available in the near future. As a result, reporting will probably be more complete and exact. It may, however, also lead to the use of definitions that are more cut to the needs of the respective institutions than those of criminology or policy making. A development running parallel to the former one is the production of "production statistics" by the various institutions. Here also the introduction of EDP plays an important role, as well as accountability and planning and control needs. This development may lead to a less important role of official statistics for policy makers, and hence for a (further) reduction of 122 Max M. Kommer means for the Justice and Security Department of the Central Bureau of Statistics. Victim surveys and self-report studies will probably play a role of growing importance for criminological research as well as policy making.* They can, however, never replace police, prosecution and court statistics in so far as these give an insight in the production of the criminal justice system. References Dijk, J. J. M. van, P. Mayhew and M. Kilias: Experiences of Crime across the World: Key findings of the 1989 international crime survey. Deventer/Boston, 1990, Kluwer. Geerlof, J.: Trends in criminality. In: Netherlands Official Statistics, vol. 2, no. 4 (1987), pp. 32-36. Geerlof, J.: Self reported delinquency. In: Netherlands Official Statistics, vol. 3, no. 2 (1988), pp. 51-53. Huls, F. W. M., J. G. C. Kester and J. Naafs: In: Netherlands Official Statistics, vol. 6, no. 4 (1991), pp. 44-47. Kalleveen, P. W. van: Police recorded criminality. In: Netherlands Official Statistics, vol. 7, no. 3 (1992), pp. 27-29. 8 For example, the RDC plans to run a self-report study aimed at children under 12 years of age. As they can not be prosecuted under criminal law, their criminal activity is not visible in the official statistics. Crime and Criminal Justice Statistics in The Netherlands 123 Appendix 1 Types of crimes covered in the Dutch national Victim Survey Bicycle theft Moped theft Car theft Theft from car (exterior) Theft from car (interior) Car vandalism Other vandalism Theft of purse / wallet Burglary Other theft Sexual harassment outdoors Sexual harassment indoors Assault / threat outdoors Assault / threat indoors Indecent exposure Hit-and-run accident 124 Max M. Kommer Appendix 2 Types of crimes covered in the Dutch Police Statistics OFFENSES AGAINST THE PENAL CODE MILITARY PENAL CODE MALICIOUS DAMAGE and PUBLIC ORDER ROAD TRAFFIC ACT offenses against public order driving under influence, bike (excl. discrimination) driving under influence, car discrimination driving under influence, other offenses against public authorities hit-and-run accident crimes against bodily security driving while disqualified (arson,culpable explosion and flood) refusal to cooperate indecency culpable homicide or grievous harm malicious damage joy-riding damage of car other Road Traffic Act damage of public transport damage of public building DRUGS ACT other damage hard drugs soft drugs VIOLENT CRIME murder/manslaughter (completed) (FIRE) ARMS & attempted murder/manslaughter AMMUNITION ACT help suicide or abortion ECONOMIC OFFENSES ACT assault Nuisance Act threat Open Waters Pollution Act Manure Act culpable homicide or grievous bodily harm Waste Products Act rape Chemical Waste Products Act sexual assault other environmental other sexual offenses protection acts Working Conditions Act theft with violence Other economic offenses act blackmail OTIIER ACTS, DECREES insult ETIC. PROPERTY CRIMES simple theft theft with breaking other aggravated theft embezzlement counterfeiting currency other forgery fraud receiving stolen goods other crimes Penal Code Crime and Criminal Justice Statistics in The Netherlands 125 Appendix 3 Types of crimes covered in the Dutch Public Prosecutor and Court statistics: PENAL CODE DRUGS ACT Offenses against public order hard drugs and authority soft drugs Offenses against life / the person of which TELECOMMUNICATIONS ACT simple assault Violence against property WEAPONS AND AMMUNITION ACT of which malicious damage OTHER LAWS Property crimes theft ECONOMIC OFFENSES ACT of which simple theft TAX LAWS aggravated theft theft with violence receiving stolen goods Sexual offenses Other offenses of the Penal Code ROAD TRAFFIC ACT driving under influence hit-and-run accident driving while disqualified refusal to cooperate culpable homicide or grievous bodily harm joy-riding other offenses of the Road Traffic Act 126 Max M. Kommer Appendix 4 Addresses of relevant institutes Central Bureau of Statistics Crime Prevention Directorate dept. of Justice and Security Statistics Ministry of Justice Prinses Beatrixlaan 428 Schedeldoekshaven 100 PO BOX 959 PO BOX 20301 NL-2270 AZ Voorburg NL-2500 EH Den Haag tel. +31 (70) 3 37 38 00 tel. +31 (70) 370 67 10 fax +31 (70) 3 87 74 29 fax +31 (70) 370 79 05 Research and Documentation Centre Prison Department (Policy Ministry of Justice Information) Schedeldoekshaven 100 Ministry of Justice PO BOX 20301 Schedeldoekshaven 131 PO BOX 20301 NL-2500 EH Den Haag tel. +31 (70) 370 76 17 NL-2500 EH Den Haag fax +31 (70) 3 70 79 48 tel. +31 (70) 3 70 26 23 fax +31 (70) 3 70 29 21 Statistical Information and Policy Analysis Department Ministry of Justice Schedeldoekshaven 100 PO BOX 20301 NL-2500 EH Den Haag tel. +31 (70) 3 70 65 68 fax +31 (70) 3 70 79 02 Criminal Statistics in Poland Jerzy Jasinski 1. The beginning Criminal statistics in Poland have a fairly long tradition. It goes back to the rebirth of the Republic of Poland that occurred after World War 1. Already the first issue of the Statistical Yearbook of Poland' that appeared in 1923 contained a chapter on the administration of justice with some data on the business of the court and on prison population. In the next issue for the year 1924, the scope of information expanded. It covered basic data on offences known to the police in the year 1922, and for some most serious crimes even for the years 1919-1922, on persons convicted by criminal courts, on prisons and prison population, on cases dealt with by the courts, and finally on policemen, judges, barristers, and notaries. During the twenties and thirties some changes were introduced in the amount of information provided and in the way the data were arranged. These changes were due mostly to the unification of the Polish criminal law?. The last of these Yearbooks appeared in 1938 and covered the years 1936-1937. Rocznik Statystyczny Rzeczpospolitej Polskiej (Annuaire statistique de la Republic Polonaise), volume I rt 11 1920/22, Glowny Urzad Statystyczny, Warszawa 1923. Volume VIII of the Yearboo {:a appeared in 1930 and then was suspended due to financial con- straints. In the same year the first volume of Maty Rocznik Statystyczny was issued. The scope of the chapter on crime and administration of justice was not substantially reduced. The volume XII and last of Maly Rocznik Statystyczny appeared in 1939. 2 After the World War I on the Polish territory there were three different legal systems in force. They were inherited from the powers who took art in the partition of Poland at the end of XVIII century (viz. Russia, Germany, and A ustria). 128 Jerzy Jasinski 2. The development Soon after World War 11, the publication of the Statistical Yearbook was resumed, but it was suspended after a few years as a result of the prevailing Stalinist secrecy mania’. Since 1955 the Yearbook has been published without any break and it always contains a chapter on the administration of justice. There have been some alterations in the scope of data published but, with only one temporary exception, the following data have always been included: of the number of offences committed, of juvenile delinquents dealt with by the courts, of adults convicted, of sentences imposed, of prison population, of prisons, courts, prosecutors, and judges. In the seventies, no data on prison population were published for several years. This unusual omission was caused by an article in a legal journal which, among other things, demonstrated that the number of prisoners in Poland, per 100 000 population, exceeded 3 to 5 times that noted in the Western European countries. This difference was interpreted as a proof of the excessive punitiveness of the Polish criminal justice system. In order to stop such comparisons the authorities banned the publication of the data on prison population which at that time was on the rise. The information on what is published in the Yearbooks should be supplemented by the information on what is withheld. The most notable examples in this respect are: the number of policemen and prison staff, and the budgetary appropriations for each segment of the criminal justice system. In the Communist state, public expenses for securing "law and order" were one of the treasured secrets. This secret was guarded so eagerly that even when a few years ego the basic budgetary figures became accessible they turned out to have been arranged in such a way that it was impossible to establish the cost of the administration of justice in penal matters. 3 The first post-war issue of the Statistical Yearbook (Rocznik Statystyczny 1947, Glowny Urzad Statystyczny, Warszawa 1947), contained some data on persons convicted by the courts. In the next issue the scope of these data - for the years 1946 and 1947 - was expanded. In the issue for the year 1949 criminal statistic nearly disappeared: the only information provided concerned the financial criminal cases instituted in 1948. In the issue for 1950 there were no data on crime or criminal justice, and there were no statistical yearbooks published in Poland for the years 1951-1954. Criminal Statistics in Poland 129 Apart from the Statistical Yearbook there are several series of internal publications related to various aspects of the work of the police, prosecution, courts, and prisons. Nearly all of them are for internal use only, therefore, there is no point to mention them here. The only exception is the Court Statistics published by the Ministry of Justice since 1962. It is not for sale, but it is widely distributed and is available in all university libraries. It consists of three parts: the first one contains information on the business of the courts and prison statistics, the second one includes information on the activitiecs of the family courts and on juvenile delinquency, and the third one provides data on persons convicted and sentences imposed. 3. The idea of uniform criminal statistics The main feature of the Polish criminal statistics is that they are not so much the statistics of crime or administration of justice but they consist of separate sets of data related to the activities of different segments of the criminal justice system. At least twice, in the mid-fifties and in the mid-sixties, there were some altempts made to create uniform criminal statistics. These statistics were to have had two parts. One of them was to follow offences from the moment they had been reported to the police till the moment - at the farthest end - of sentencing their perpetrators. The other part was to follow persons from the moment of becoming suspects till the moment - at the farthest end - of their conviction or even till the termination of the execution of the penalty imposed. After long deliberations, both attempts were abandoned offically due to the inherent difficulties in their implementation. The main reason for the discontinuance of the work seems to have been, however, different. Some of the parties involved lost their interest in the affair. Uniform criminal statistics would inevitably become an instrument of outside insight into the work of each segment of the criminal justice system, and each of these segments had some secrets rather hidden than revealed by its own statistics. At present, the idea of uniform criminal statistics is not likely to take root in the foreseeable future. 130 Jerzy Jasinski The above attempts were not entirely futile. In the sixties, the common statistics for the police and public prosecutors were established. They cover the known offences, suspects, and the results of the preliminary investigation. However, the public prosecutor’s office as well as the police retained some kind of statistics that show their dealings in criminal and in some other matters. Also the courts and the prison system have statistics of their own. 4. Police statistics The police statistics have been reorganized several times. At first, it was based on the returns prepared by each police territorial unit. The returns had a form of tables and the contents of these tables were added, first on regional and finally on the country level. At the beginning of the sixties, the system of police data collection for statistical purposes was changed, and it has remained basically unchanged. For each offence two separate statistical sheets are filled up, a short one at the moment the offence is reported to the police, and a longer one at the end of the preliminary investigation. The longer sheet contains some information on the suspect, if available, and on the way the preliminary investigation is terminated. If the preliminary investigation is conducted by the public prosecutor himself, it is his office, not the police, that fills up the longer sheet. This makes the statistics of offences a common endeavour of the police and the prosecutor®. In the sixties and seventies the sheets were sent to the Police Headquarters where the statistics were compiled. At present, the statistics are prepared at the regional level according to uniform scheme, and only the tables are transmitted to the Police Headquarters for the compilation of the statistics for the whole country. The problem with the police statistics is that the definition of the main unit in the statistics used, is not clear: it should always be one offence, but quite often it is not obvious whether one offence or more offences were committed. This is particularly true when there is more than one 4 In all criminal cases the decision on how the preliminary investigation is terminated rests with the public prosecutor. Criminal Statistics in Poland 131 perpetrator, more than one victim, more than one object stolen or damaged or when the violations of law occurred over a longer period of time or at separate but similar occasions. These difficulties are well known, so there is no need to go into the details here. However, these difficulties have to be mentioned because they have not been solved satisfactorily in the Polish criminal statistics, which creates some problems in looking at the numbers of offences known to the police over a long time span. The police statistics do not cover all known offenses. There are two groups of such offences. The first group consists of offences brought to court on a private accusation. By the very nature of these offences there is very little police involvement in their investigation. The other group consists of a variety of offences related to health, food, borders, customs, taxes, financial regulations, in which the investigation is conducted not by the police but by different administrative authorities which also have accusational powers. S. Court statistics The court statistics consist of two distinctly different parts. One of them is compiled from the returns sent to the Ministry of Justice by the common courts of Ist and IInd instance from all over the country. In this part of the judicial statistics several different units of counting are used. The most notable of them are: the criminal case, the decision taken by the court, and the person in relation to whom the decision is taken. For the purpose of this statistics the courts used to present to the Ministry of Justice many returns, most of them on a monthly and quarterly basis. The form of these returns was often complicated ,which made it difficult for the courts to complete them. They were frequently changed and there was a tendency, well known in every bureaucratic system, to €xpand the scope of the data requested. Slowly, the number of returns and the frequency at which they were requested decreased as a result of the pressure from the Central Board of Statistics (Glowny Urzad Statystyczny). In the seventies, the above Board had been bound by law to control the activities of all Ministries and other units of central administration in the domain of statistical data collection for internal needs. This step has proved 132 Jerzy Jasinski quite effective: first, in arresting the steady growing number of returns required; second, in improving the quality of the returns; and third, in reducing the frequency the returns had to be supplied. At present, the Ministry of Justice requires from the courts some statistical returns every half a year, and most of them once a year. This part of the judicial statistics has been used mostly for the day-to-day management of the court business. In the past, it provided also a basis for "guiding" the criminal policy executed by the courts. At the beginning of the nineties this latter task was abandoned since it became obsolete when the independence of the judiciary had been re-established. The other part of the judicial statistics was for many years based on special sheets filled up by the courts in every case a sentence passed became valid. The sheets were sent to the Ministry of Justice, and after their contents had been used for statistical purposes, were transferred to the Criminal Register. In the late seventies the entries to the Criminal Register and the supply of information for compiling the court statistics were separated. In this part of the judicial statistics the unit used is the convicted person. This statistics are utilized mostly for detailed analyses of the penalties imposed. This is reflected in the Court Statistics published by the Ministry of Justice. The bulk of the tables provided there relates to the number of persons convicted for separate offences and to the penalties these persons received. In this respect, the Polish judicial statistics do not seem to differ much from those in other countries. What is missing from the polish judicial statistics is first of all the data on victims of offences committed by the sentenced offenders. There is also only very little information on the damage caused by the sentenced offenders. This information is limited to that done by offenders sentenced for larceny and some related offences, and to the damage caused to the state property only. The other missing type of data is related to recidivism. What can be learned from the judicial statistics is how many previously convicted persons are sentenced each year, for what offences, and what penaltics are imposed on them. It is, however impossible to learn how many persons who had been convicted in a given year have been reconvicted during the first, Criminal Statistics in Poland 133 second, third, etc., year after being sentenced or after being released from prison. The judicial statistics cover the sentences of common courts (courts of general jurisdiction) only, that is they do not cover the sentences of courts martial. In two periods this limitation became very important. In the years 1945-1955, and in the years 1982-1983, the courts martial had, apart from the exclusive jurisdiction over servicemen, a wide jurisdiction over civilians as well. 6. Prison statistics The prison statistics are based only on the returns from prisons . They are centrally compiled at the Ministry of Justice. The main unit used is the person deprived of liberty. After some years of confusion, the double counting which stemed from transferring prisoners from one prison to another, and from the change of their status from remand to convicted prisoners, was eliminated. All categories of persons deprived of liberty as a result of committing (or being suspected of committing) an offence are included. That is: (1) sentenced prisoners’, (2) persons remanded in custody, (3) persons doing their time because of default of paying a fine or because of failing to comply with the requirements of the penalty of limitation of liberty, and (4) persons ordered to be arrested because of committing a transgression. Till the early nineties some of the above persons, in spite of being deprived of liberty, were missing from the prison statistics. According to a section of the Code of Execution of Penalties, in certain circumstances some remand or even sentenced prisoners could be held in police cells. This section was abolished in 1990. Since these persons were not in custody executed by the Prison Service they were not counted in prison statistics. Occasionally (e.g., in times of political crises) their numbers might have been substantial. S Persons sentenced by the court of first instance are, according to the Polish law, included for the purpose of the execution of the penalty of imprisonment into the category of sentenced prisoners in spite of the fact that their sentences are not valid yet. 134 Jerzy Jasinski 7. Final Remark In Poland, criminal statistics are composed of several independent and only loosely related parts. Most of the information gathered reflects, first of all, the dealings of the separate segments of the criminal justice system of the country, and not so much crime or offenders. It seems very likely that this situation will remain unchanged for many years. Criminal Statistics in Portugal J. de Seabra Lopes I. Development of criminal statistics In 1983 the production of statistics on the administration of justice was committed to an agency in tyhe Ministry of Justice, the Gabinete de Estudos e Planeamento do Ministerio da Justica (GEPM]J),which acts as a delegation of the national statistical office. For a long time, in harmony with the Continental European tradition, criminal statistics mainly consisted of judicial statistics, which tended to be interpreted as fairly relaible sources of information on the criminal situation of the country. Recent developments led to the emergence of a new programme of recording crimes derived from law enforcement agencies, most notably ’police statistics’. In January 1993, the preliminary stage of the programme was set up with the three main police forces in Portugal. Gradually, other specialised agencies concerned with specific crimes, such as tax evasion and crimes against public health and the economy, will also participate in the system. GEPMJ developed a summary reporting system which uses a uniform crime classification consisting of 81 numbers divided into five classes. Both criminal code and non-criminal code offences are included, but not law violations which, despite their inclusion within the penal field, are not conceptualised as crimes. Data recorded on suspects is limited to sex, age and detention. For practical reasons, no information is gathered on victims at this stage. 136 J. de Seabra Lopes I1. Data collection GEPMJ runs a national centralised system of data collection. Procedures vary from summary reports and individual reports submitted by contributors as raw data, to pre-organised tables sent in by some criminal justice agencies. With a few exceptions, the system is manual at the contributors’ level and automated at GEPMJ level. GEPMJ’s data base stores a vast amount of information, of which only a part is included in the annual report Estatisticas da Justica. This publication contains basic data on the administration of criminal and non-criminal justice. It includes a time series on topical subjects and a description of the volume of activity during the year. The main areas covered are input statistics (caseflows and caseloads in courts and law enforcement agencies); case characteristics (namely, persons dealt with by criminal and juvenile courts); prison population; and population of children and young people placed by juvenile court orders in Ministry of Justice institutions for minors. Most statistical information systems on the criminal justice system is based on legal concepts and definitions. However, the crime reporting system is an exception: behavioural or police terminology (like ’snatch theft’) are used when considered more informative. This has posed a problem of compatability; data on crimes of which persons are charged, convicted and sentenced rest entirely on legal concepts. As a rule, behavioural types can be aggregated and converted into legal ones. On the whole, criminal justice statistics are compatible with demographic and other social statistics. Such links have been made easier by the specialized supervision that the national statistical office exercisesover the content of statistical forms produced by GEPMJ. 137 Criminal Statistics in Portugal II1. Reliability of data In statistical production, reliability is an aspiration which has to be continuously sought after. It must be admitted that operational staff regard statistics as an unwelcome burden. Consequently, the tension between administrative and operational tasks on the one hand, and statistical tasks on the other, is probably inevitable and even more so in manually operated contributors systems. In training sessions conducted at GEPMJ for court clerks, motivation is an ever-present concern and the importance of reliable statistical information is stressed. Elaboration of rules of procedure and elucidation of individual doubts are routine ways to enhance the quality of the information collected at the suppliers level. At the central unit (GEPMJ) quality is easier to control. In a relativerly small centralized and computerized system, staff training and definition of rules of procedure do not pose great problems. Criminal justice statistics have evolved partly from tradition; partly from consideration given by GEPMIJ to the subject matter of coverage and contents; and partly in answer to requests by users within the criminal justice system. However, quantitative studies, as basis for evaluating and planning in this particular area, are not common. Therefore, users have not been prominent partners in the improvement of the criminal justice statistical system. Nevertheless, regular meetings provide the opportunity to exchange views, assess needs and define priorities. IV. Victimization surveys The regular production of criminal statistics at the national level has been supplemented recently, in 1989 and 1992, by the implementation of two victimization surveys. These surveys (the first limited to the Greater Lisbon arca and the second extended to the mainland) were stimulated by the crime reporting programme, as a means to obtain a view of crime - or of some crimes - independant from that conveyed by police statistics. Both 138 J. de Seabra Lopes surveys revealed that the frequency of crime reporting was extremely low, not even reaching a quarter of the incidenets mentioned by victims. This information is of undoubted relevance for the administration of justice. Besides further victimization surveys, the most important development is likely to be the integration of the operational information system for decision-making with the statistical one. This, however, will only be possible when the automation of the criminal justice system, at present at its early stage, is fully completed. It is then likely that the description of cases entering the system may receive the same attention now given to cases terminated. Another possibility might be the elaboration of longitudinal statistics, making it possible to link cases in each step of the criminal justice process. V. To international comparisons In spite of all future improvements, it is doubtful that criminal statistics from different countries can ever be comparable, if the aim is measuring levels of crime. Comparisons of global data are particularly dangerous, as the range of behaviour which is criminalised varies considerably from country to country. In Portugal, for instance, the legal system was traditionally based on the distinction between crimes and other violations of the law situated in the penal field (corresponding more or less to the French contraventions). A few years ago, a new category was introduced - contra-ordenacoes - similar to the German ordnungswidrigkeiten, meaning violations of the regulatory norms which lic outside the penal field. Eventually, anti-social behaviour will only be considered under the categories of crimes and contra-ordenacoes. It is therefore evident that different concepts of crime will have a direct bearing on crime figures. Even in the restricted field of traditional crime, comparisons may be risky due to differences in legal definitions and procedures, in statistical rules of reporting and, not in the least, in the frequency of crime reporting. The latter circumstance is evident in the case of Portugal, which has a considerably low rate of reported crimes: approximately 50% of the results Criminal Statistics in Portugal 139 from a group of countries which recently participated in an international crime survey. Statistics of persons before the courts mean different things in different countries: for instance in Portugal the prosecution has limited power and the police no power to dismiss cases but in some countries, such as Great Britain, the prosecution is discretionary and the police frequently resort to informal warnings (cautions) instead of prosecution. Even in later stages of the proceedings, legal differences may have a substantial influence on figures. Under Portuguese law, for instance, victims are entitled to terminate cases before the court decisions are given in a wide range of cases (such as common assault, slander and sexual crimes). This facility is frequently used, thus affecting figures at the adjudication stage. On the other hand, international comparisons, which highlight variations in law or practice between countries, can have a dynamic effect and suggest new ways of dealing with basically similar questions. In short, international comparisons are not valid as indicators of crime levels but can open the way to innovations in the criminal justice systems of different countries. Bt E Bt £e e I 13 i (e o i S kg L b, % o g o3 14 Fid : o L 15 " :;‘, §% H e ! [ 58 o i Lu ol i o A e #a () i ot R 4 ! g Bips B f%fl % o yid i A i o ) o X & o 25 w5 Sl Ey i ¥ G i o i 2 g 1 o i § ¥ @ i G b e % i . e : " % 5 i £ i e ol R St e 3 7 g e 44 h e i a 5 i b e 1%5 oy s o s b o !) i Al Gl S iy g Gl el 545 no ko 3 o & di oo rk é’ e [y e !fi % Ay % $ ~”'fiv!f'k" o0t [ i e %‘1 L o g Y i M he ol 5 e 4 1% Ehes g i at ¥ 9 g i o o i i e = Eiie Vi vh e &4, oA e i) [ S i el %4 o 1. B 1oe! f’ . e St o i oy ,»,/’** et 3 ‘:J\' !‘; e ¢ o i W % ey G al 2 7 & iz o e 5 b o8 ik 1 oo e e E . . % A “ DI o, b e o = “ b o o i b i1 £ H & i s el ba 55 b ) v i o i i1 3 Y 3 i Y ) i = 5 : 43 By R i oar o o o W i i o~ oo & gt \"C e o e e i1 i ol i o foat $E8 o I8 £ vy o ‘%‘«& i i e i s g o W o o ol . o RS g i i > : £t Yo et i 8 = as s W 5 i o J o3 it e f& i 5 s 0 il i Al 4] ot o g . S £ Rt n P ] 3 e i G o i o e i i o e - i ; T et ey = £l gl i i kL G . kg o e i i) i b 0 i e i e i g P s i i b o i i i " e - e 1% S %y s e g V. P o ed i [ B e i & k) v % = i p b 17 A i e T £ ot = we 2 e o Ar e e & i e Ea £ ) e L o e -~ The Swedish System of Criminal Statistics The Scope of the Statistics Torbjorn Israelsson General remarks Swedish criminal statistics provide information on crimes registered by the police, persons suspected and those convicted of crimes, persons in custodial as well as non-custodial correctional care, persons who recidivate into crimes, and, finally, special actions such as arrest and pre-trial deten- tion. The collection of data for criminal statistics is accomplished for all practical purposes entirely by means of data extracts from the judicial informational system, that is, information gathered for administrative pur- poses within the judicial system. It is the police, prosecutors, courts, and correctional system that submit information for the statistics. The statistics describe the handling within the criminal justice system of a large quantity of crimes that become known and registered by the police - the proportion of crimes that are cleared up, the conclusion by the police/ prosecutor of the preliminary crime investigation, which sentences/sanctions are selected, and how these are executed. By compiling several years of conviction statistics, it becomes possible to study recidivism, to describe criminal careers, and to illustrate the relatively efficiency of the sanctioning system. Special statistics on the criminality of successive birth cohorts have been developed. Statistics on special measures - primarily, taking into custody, formal arrest, and pre-trial detention - are quite incomplete and can be said to represent the area most poorly illustrated in the statistics. 142 Torbjorn Israelsson Are the Figures from the Different Parts of the System Compatible? The statistical information from the various points in the criminal justice system are not entirely comparable. For instance, statistics on crimes repor- ted to the police (hereafter, crime statistics) and conviction statistics (in- cluding some prosecutorial decisions implying findings of guilt) reflect activities in different parts of the criminal justice system. The conviction statistics cover all criminalized actions that result in findings of guilt, where- as crime statistics do not cover all crimes registered by the police (for example, many crimes under the special legislation with only fines in their penalty scale are excluded). Further, crimes are classified differently in the crime statistics than in the conviction statistics. The classification of crimes for the crime statistics is performed according to a particular standard for coding and counting developed by the statistical burecau (SCB). The coding is based on the relevant legislation but is often further broken down into subcategories to provide information on the object and location of the crime, the age and sex of the victim, the method used by the offender, and so on. The figures for convicted persons, on the other hand, are given only by the section of the law under which they are found guilty, without any further information about the crime. Comparable crime groups can be found in both areas of these statistical areas only on a very crude level - by chapter of the Crimi- nal Code. The various units of measure in the statistics - number of registered crimes, cleared crimes, suspected/convicted persons, prisoners, and so on - are measures of criminality but they measure criminality at different points of the criminal justice system s disposition of criminal cases. For example, it is not possible to follow individual cases through the judicial system, due in part to the incomplete links between the various statistical branches. The information that is given in the statistics thus is a type of cross-section for the year in question. The Swedish System of Criminal Statistics 143 The Reliability and Value of the Statistics The criminal statistics have been delegated an increasingly significant role in society. From having primarily been used for historical comparisons of crime trends, the statistics have started being used to a greater degree as arguments in the public debate on criminal policy, and in part as the basic material for planning and decision-making about actions/measures to be taken. These use-areas place greater demands on the reliability of the statistics. The reliability of the statistics can be improved by use of complementary data sources for illustrating the dark figure of crimes. Here, SCB s victim surveys constitute a valuable complement. In general, it can be said that reliability is high for the statistics on convicted persons and statistics on correctional care. Crime statistics reflect only that part of reality that consists of crimes that become known to and registered by the police. To answer questions about the level and structure of or changes in the actual criminality, further knowledge is often necessary. For many types of crimes, such knowledge is obtained from victim surveys, surveys of self-declared criminality, or analyses of convictions and police investigations. Criminal statistics are sulficiently reliable for those crimes that are regularly reported to the police (for example, break-ins, car thefts), whereas reliability is worse for crimes that are seldom reported (minor crimes of violence between acquaintances, crimes against public order or the state or such crimes that become known as a result of the police and other authority “s surveillance activities (drunken driving, tax crimes, and so on) Changes in Data Collection, Development of Surveys, Use of Computers, Improvements of Dissemination Electronic data processing is developing rapidly within the various sectors of the criminal justice system. Over a short period of time, in effect all routines within the police, prosecutors, courts, and correctional care will 144 Torbjoérn Israelsson be computerized. This may influence the manner in which information is gathered for the statistics, in that the data are gathered directly from the local authorities instead of as is the present practice, obtaining statistical basic material from the central judicial computerized system. So far, surveys have been primarily used for victim studies. Other feasi- ble areas for surveys include studies of public confidence in the justice system, how the "general sense of justice" has changed, and so on. The criminal statistics are now distributed by means of printed matter, computer tables, diskettes for further processing by the user, as well as through information being made available via central data bases. The distribution through access to data bases will increase. Within SCB, a conversion is being discussed from central computer-based production to client-server solutions. Further, the possibilities are being discussed for creating so-called public use files with de-identified, reduced information that various users, researchers, and the like, gain access to for further processing and analyses. Victim Studies or Microcensuses as Substitutes/Complements for Administrative Statistics Criminal statistics that are primarily based on material collected for administrative purposes can hardly be replaced by victim surveys or micro- censuses. On the other hand, victim surveys constitute an important com- plement to the traditionally gathered statistics. Victim surveys as well as microcensus surveys are expensive to conduct and in light of the current poor economy within the Swedish public administration, the possibilities of further developing this complement appear to be slim. More likely is that we will be forced to reduce sample sizes in the victim surveys which will limit their usefulness - in particular, if we want to break down the statistics on a regional basis. The Swedish System of Criminal Statistics 145 The Possibilities of Establishing International Comparability The question of whether criminal statistics can be used for international comparisons is impossible to answer generally. Each particular use must be examined with regard to whether or not the official statistics are appropri- ate for comparisons between countries. The international demands on the Swedish criminal statistics have until now consisted of requests from international organizations for reports to be sent in to their publications and data bases. It is primarily the U.N. and Interpol who request such information. Within the framework for the work that is conducted between the Nordic statistical bureaus, a Nordic criminal statistics publication has been produced. The method chosen for this endeavour implies proceeding from each countries” published statistics, compiling them, and describing the differences in their legislation and statistical routines. Then, it is left to the users to determine how possible differences in the statistical outcomes should be interpreted. Even if criminal statistics have not yet been the focus of any substantial interest within EUROSTAT, we expect to begin receiving requests for criminal statistics for European comparisons. What is usually asked for in comparative studies of criminality among countries includes information on the level of crime, crime trends, sentences and sanctions, and the size of prison populations. It should be possible here to compile statistics at least for well-defined groups of crimes as well as groups of sanctions. Inter- national statistics on the prison population are also compiled by the Coun- cil of Europe in their Prison Information Bulletin. i a» <y (e o) b £ 5 t e s o i S . i L Ak Il oy i . ot ik i % e I & P 0 il dy " o ¢! e i Y 0 ASEEL L ki A o % i 5 b ikt A d o4 i 5{ bt % B et St By 4 s o i ’ i il O " 4 £ B i d e i . 5 i i L gt i L b o e K s % i ey et B e u;;‘ o - . & i (5 i 0 ok T e e S by L : ke e 3 e v vy hy %v o 5 A s (5, i) pis o 21 i 08 e i kR Ax Fd 0 i kP il it 1 i % 33 il (21 - Sy e o g Lek &1 yata o P B e L 1 & = % A g T e Sidy i 4 g 8l ek Ed . 2 s e [ ¥ L 5 T i . ey i i o " i ;é%w kY o Pl B hy " o el bl 5, . i S b & i uA & L A ¢ i % " 2 Wi i LNl ) il £ & e a5 P H el s (E5VA] b e i o [ 5 r’ L %A i L i B o e A ity = ¥ 3 i ke e ) o b i 1% b # T v i ‘{D i s i v e N e o' i g kA g pen o B & 9 o b B SR L 5 s ik L i A i £ b Doy Ty i o Wl o4 = G % o o i 0 MR " ’: il i s o i & day ’:‘ gl L A% K th i 2 ‘: i o5 o e fl; 1 g S ™ " % it i i i : ; 4 2 i i [ & e [ d it I 5 ik o Fo % o5 e o s i i fail B 45 590 sl e 3 e " s A i h o g 452 £ s B 7 e ek i% e 2 o by i o ) o v & i 5 i s o '5 4 o e i . 4 o ) e e A . ,,E i Fe i An Example from the Developing Countries: Criminal Justice Statistics in Africa N. Masamba Sita 1. Introduction Many authors have written about the lack of good qualitative and quan- titative statistics needed to carry out research into the arcas of crime prevention, the working of the criminal justice system and the treatment of offenders. Such reasons explain the low participation rate of African coun- tries in United Nations surveys. The material brought together in seminars organised by the United Nations African Institute for the Prevention of Crime and the Treatment of Offenders [UNAFRI] has encouraged us to ask the following question on African crime statistics: which is, to know whether African countries do have adequate statistics on crime for their own purposes? This question can perhaps be better rephrased in the following way: Is there a lack of statistics or is it more a problem of the quality of statistics and their availability? This chapter will consider all these questions, making use of the material brought to UNAFRI seminars by delegates from dif- ferent African countries. 2. The Available Literature The authors noted below examine some of the aspects of the following questions which we consider important: to know 148 N. Masamba Sita i) Do the criminal statistics cover all the different agencies of the crimi- nal justice system and the penal system? Are statistics from different agencies compatible. i) What is the accuracy and usefulness of the statistics? if) What changes are necessary in the collection of the statistics? v) To what extent should measures of recorded crime be replaced or complemented by victim surveys or sample censuses. v) Are comparisons possible at the international level? 2.1 J. Thabard The first response to these questions has been made by J. Thabard, who writes about juvenile delinquence, saying that the available statistics are often of little use, the trends they show are now accurate and do not allow a simple comparison of one country with another'. 2.2 Y. Brillon Another author who has given a lot of attention to the problem of crime in Africa is Y. Brillon. He state that in general all authors deplore the lack of effective statistics, their accuracy and their poor usefulness.” 2.3 Femi Odenkule This author® introduces a further dimension to African criminal statis- tics. He reports: During my first working visit as Director of UNAFRI to a collection of African states of all sub-regions of the continent, in order to discover problems and priorities, the local responsible officials whom I met were unable to provide me with the latest published official statistics on offending and penal justice. 1 J. Thabard: Quelles ap roches de la delinquance juvenile en Afrique: in Environnement Africain, No. 79-39, 19 5 9, p. 9. 2 Y. Brillon; Ethnocriminologie de I'’Afrique Noir, VRIN/P.U.M. 1980, p. 12. 3 Femi Odenkule: Bilan des connaissances en Afrique: in Ph. Robert (sous la directioin): les Politiques de prevention de la delinquance a I'aune de la recherche; ed. I'Harmattan, 1991, V. note 5, p. 177. Criminal Justice Statistics in Africa 149 He went on to say: In one of these countries, 1 was told that this type of publication related to national security and that in order to have acces to such material I would need to .. make an official request!!! We note two points when considering the comments of these authors: i) The inability of countries to produce the latest officially published statistics on crime and the penal system. ii) The difficulty of access to these statistics What is the cause of this inability? Is it because the statistics do not exist in these countries or is it due to poor methods of collection? Is the difficulty of access in certain countries simply an administrative feature or is it due to the non-existance of the statistics? Whatever one can say about the problem of poor data collection, there is a further difficulty in that it is impossible to reconcile statistically the flow of police arrests and the figures of prison population because of the independence of various parts of the criminal justice system (S.A.J.P.).* To the extent that in many African countries administrative constraints restrict access to criminal statistics, victim surveys can supplement our knowledge. Such enquiries have the undoubted advantage that they are not limited by what the administrative system of justice can produce. 3. Country Reports 3.1 The Point of View of African Experts E. Kibuka observes that: One common factor identified in most countries is the inability of governments to collect statistics which are accurate, complete and useful: and 4 Cfr. G. Houchon; Proprietes et strategies pour une criminolo ¢ du tiers-monde; in UNICRI, Essays on Crime and Development; Publication No. 36 R oime, July 1990, p. 58. 150 N. Masamba Sita to prepare national statistics which are measure the volume of crime, its characteristics and trends.’ Having considered several commentaries on country reports, it is inter- esting to note an enquiry lead by UNAFRI. Despite its limited remit, the enquiry throws a certain light on the problem we are examining. We will limit ourselves to a sample of items on the questionnaire. Of the 19 countries which took part in this seminar’, only 9 had responded to the questionnaire. - Item 2: Only 6 countries had a uniform classification for statistics. = Item 5: 6 countries said that they organised the collection of statistics - Item 11: 6 countries had a centralised statistical service =) Item 15: Only one country had a structure for collecting, processing and analysing statistics The statistics from this enquiry indicated a very different situation amoung different african countries. If we postulate a hierarchy of measures, the position varies from one country to another. The collection of statistics in a pilot survey carried out by UNAFRI on crime, administration of juctice and victimization, was planned in two towns, Bujumburta and Kampala, although access was allowed and counting rules agreed, the coverage of the survey was quite low’. We thus return to the point we made before; the lack of structures for collection, processing and analysis of statistics. In most countries there is a data collection, because this is part of the law. But the rest of the work, the centralisation of the statistics with the aim of analysing trends (either E. Kibuka; Information et donnees sur la crime, la victimization et 'administration de la justice penale dans le context du developpement en Africa, 1987, p. 1. idem; p. 7-9. Seminaire de 'UNAFRI sur las planification pour la prevention du crime et la justice penale dans le contexte du developpement, Addis Abeba, juin 1987. Compare the report presented to UNAFRI on the co-ordination and carrying out of the African crime survey; Kampala (Uganda) 23-27 November 1992. Criminal Justice Statistics in Africa 151 increasing or falling) and the formualation of crime policies, is just not done. 3.2 Countries Reports The brevity of this paper only allows us to make a few references to these reports for illustrative purposes. 3.2.1 Burundi In this country, article 2 of the code of penal procedure (CPP) says: Officers of the police judiciare record the crimes which they investigate. They receive complaints about crimes, the times and the places where they occur ... On the basis of this aspect of the penal procedure and others, different registers are drawn up: a) transactions occuring b) cases opened ©) prison receptions d) documents sent to the Parquet These different registers are based on different counting rules, and the consistency of these rules is often ignored in Burundi. The different statis- tics of the criminal justice process, such as the police, courts and prisons, are not of equal validity. A further source of statistics is the Burundi Report® - this is the security council for each administrative sector - i.e. the national and regional security councils. This oversees the two corps - the Police and the Gendarmerie on a weekly basis, where the local provincial governor considers events where the two corps interact. 9 Idem. 152 N. Masamba Sita It is clear that there is no national statistical service. As far as crime is concerned, unfortunately, the annual statistics only reflect the work of the police. 3.2.2 The Congo The report from the delegate from the Congo showed that there was not a national statistical service which coordinated the reports from the local police commissariats to make a detailed report on the national situa- tion. 3.2.3 Madagascar The report of the Direction de la Police Judiciare is a very useful source of statistics. This report includes monthly, quarterly, half-yearly and annual statistics of all the police activites in the island. A report is pub- lished each year. 3.2.4 Ruanda Two important sources of the situation in Ruanda are the Police Judi- ciare and the Inspection Administrative. The presidential decree No 132/06 of 10 March 1989" instructed these two organisations, or behalf of the central administration, to collect, analyse and exploit statistics on crime and criminal justice. Courts, tribunals and the Parquets send regular statistics on their activities to the central administration where they are analysed and published in an annual report of crime in the country. 4. Summary The various reports noted above indicate what statistics on crime and criminal justice are available in African countries. Unfortunately, they are not very far advanced. 10 Idem, p. 15. Criminal Justice Statistics in Africa 153 In comparing one country with another, the level of detail varies. Nota- ble, social, economic and administrative conditions are different, but also the method of collecting the statistics varies. For example, police statistics take more interest in national security than courts or prison statistics. We have said before that it is very difficult to account statistically for the flow of people through from arrest by the police to prison population in the absence of a common database where each agency is able to record exactly what has happened while the particular person was in their charge. This characteristic compromises all attempts to make comparisons, either at the national or international level. National statistical services in these countries, where they exist, ought to be concerned with the compatibility of the figures produced by the criminal justice system. Whatever the qualites of criminal justice statistics in a country the potential for crime surveys remains high. These would supplement the statistics of the criminal justice system. Such a diverification of statistical data collection would serve a very important social purpose. i s heiZ o o o b pa . i e hrd S w5 e Y @'& L i i i s v 4 5 o i o ) it o M e T o et L Ve E § L/r, £ o ot % pos adion . o o 5 i 3 g I PR i . (& i i Sies vyl p¥ i o 0 o i i 4 ‘ i bfl' % ¥ 3 - 5 o & o s i e e e o P o s ik Bl - i b P i o i - 73 i K & e g i 5 £ i & st f}! e 5 # i 1 . i T a £ o b i & ¥ " e % ¥ ur 2 . ‘34 o s (RE) 434 R bR A i 1 i e e »“g i drig e Y o 0 biks o o € : ket i % be 4 s X & e s = I o o g i 3 S0t i a i i o e e d e i W L b o e i 5! 5 X gk i B 21 2k el 34 m W o Be) i [ £ P e b 0 ee & 4 i i e £ B 5 i o - i . b i i s 5 Sy g ip ki e n s i b e <:: i g 25 e ik, fik o & vf w‘s o e St i By ot i e s i : N . b il i G ey ks & 4 i " i i i 8 ) fifi‘ ‘A{I B i i! btir‘ . o A iR E 4 k) s i il 2 N i i it ¢ e e £ et b :% s 7 ¢ ¥ bl i e ifi‘ “ . R i 4 s .t i & L 9 bl 4 ] £ ) G i e ar e e B e e vt o 4 . i o o y : . e e () Y ¥ i [ g b X . Jiss b g & g4 R S P s o il e & gt i 9 - ) o 9 e e o i i 1 i I o B % - i - it o i o jie; s 4 X g Vi viny) byt Lo o B o s e & . v - i " s .l p i‘ e i & o 4 . o e Vi i e | T - 2 5;{ o e =W Et b i £ L . i i & L A o i b i& . i i o i 5 e Libs . i o i b o i R {3 o i v 5 e e Red o A i b = - ;r,f " 1 " i ! i g £ i e iir & . e K. o8 wiy n An Example from Overseas: Canada’s Justice Statistics Program Sange de Silva Introduction Established in 1981, the Canadian Centre for Justice Statistics has a mandate to produce information on the extent and nature of crime and the administration of criminal, civil and administrative justice in Canada. This information is intended to serve governments in the development, operation and evaluation of justice policies and programs as well as to broaden public awareness of the trends in crime, how the justice system operates and of its costs. The Canadian Centre for Justice Statistics also provides financial assistance and expert advice to federal, provincial and municipal agencies with justice responsibility in developing information systems that can serve local operations as well as national information requirements. Reflecting these responsibilitics, the Centre has two main operational arcas. The first, Statistics and Information Directorate, is responsible for the development on operation of statistical surveys, their analysis and special studies. The second, Technical Assistance Directorate, provides technical and financial assistance to individual, jurisdictions. It is also responsible for responding to information requests, marketing and client relations. 156 Sange de Silva Statistics and Information At the present time information is collected on the justice sectors of policing services, corrections, courts, youth justice and legal aid. For each of these sectors, four types of information are collected: caseload and case characteristics data quantify the volume and describe the nature of the cases dealt with; qualitative information describes the structure, legislative authority and programs of each sector; and, revenues, expenditures and personnel data (REP) quantifies the costs of administering the system. Data collection is on-going resulting in the availability of time series information as well as the latest data possible. The following is a brief description of each Canadian survey and the type of information collected. Policing Services The Uniform Crime Reporting (UCR)Survey has collected information on three categories of caseload and caseload characteristics data since 1962 (i.e., offence, clearances and persons), including the number of known offences or those reported to the police, unfounded and actual offences cleared by charge or otherwise, the number of adults charged (by gender) and young offenders charged or dealt with informally (by gender). Extensive development work was initiated in 1984 to collect incident- based data which will expand the information available from the survey. For example, the new survey captures age and sex of victims and offenders, victim-accused relationships, level of injury, weapons involved, location of incident, type of property and dollar values for property and drug crimes. With these data, issues such as family violence (i.e., elder abuse, child abuse and wife battering), the cost of crime, and the economic magnitude of the drug trade can be addressed. This new incident-based Uniform Crime Report Survey is now being implemented across Canada. 157 Canada’s Justice Statistics Program Information collected by the Homicide Survey can address specific questions such as methods used to commit homicide, characteristics of homicide incidents, victim-accused relationships and characteristics of victims and suspects. The survey was amended during 1990-91 to collect information on the incidence of drug-related and gang-related homicides as well as improved information in the area of suspect-victim relationship to help address areas such as family violence. Detailed information is collected on police officers, special constables and civilians (by rank, gender and function) and on police salaries and expenditures via the Police Administration Survey. Adult Courts Case characteristics and caseload information collected by the Adult Criminal Court Survey (ACCS) provides details relating to the disposition of criminal cases in Provincial-Territorial Courts involving federal statute offences, and caseload indicators for federal, provincial and municipal statute cases. The Revenues, Expenditures and Personnel Survey (REP) provides information on the operation and costs of the court system in Canada. The survey collects data on all levels of court, including the federal level, with the exception of the Municipal Courts in Quebec. Qualitative descriptions of the organization and jurisdictions of the courts sector are provided as part of this survey. The Legal Aid Survey collects aggregate provincial/territorial resource, caseload and caseload characteristics information associated with the delivery of legal aid services. Descriptive information (structure, administration, eligibility criteria, tariff rates and process) is also collected on the delivery of legal aid in all provinces/territories. 158 Sange de Silva The Prosecutions Survey provides information on the delivery of prosecutorial services in each jurisdiction. In addition to providing quantitative information regarding personnel and expenditures it also describes the unique service delivery mechanisms in jurisdictions. Corrections The Adult Corrections Survey collects aggregate data concerning custodial and non-custodial services provided to adults in the provincial, territorial and federal jurisdictions across Canada. Caseload, case characteristics and REP data are collected in addition to basic qualitative information which describes the organization and delivery of correctional services for each province or territory, Correctional Services Canada, and the National Parole Board. This survey is currently under review to identify information needs and corresponding data requirements. In addition, the Adultand Youth Key Indicator Report Projects collect and report aggregate key indicator data which provide a monthly time series overview and analysis of adult and youth custodial caseloads. Information collected under both surveys is limited to a single data element: "monthly average daily actual-in count” of persons in custodial facilities. Semi-annual reports are prepared, providing corrections managers with timely information on short-term trends. Youth Justice The Youth Court Survey collects statistical information on Criminal Code and other federal statute charges heard in youth courts. Caseload and case characteristics data are collected from participating jurisdictions through the largely form-based manual survey. Based on current system development plans, it is expected that most jurisdictions will achieve the capacity to provide national information from automated systems. Canada’s Justice Statistics Program 159 It is expected that system development work for the Youth Custody and Community Services Survey will be completed in most jurisdictions within five years. In the interim, data will be generated by an annual, aggregate survey. The Mentally Disordered Offenders (MDO)Survey is a micro data survey. It gathers statistical information about persons under the jurisdiction of the provincial Boards of Review. Virtually all of these patients have been found unfit to stand trial or not criminally responsible. These patients, who now number well over a thousand, are of particular interest partly because they raise complex legal, psychiatric and administrative issues, and partly because the cost of containment and treatment over the years is very high. The database contains information about the age, sex and marital status of mentally disordered offenders. It collects court information including alleged offence and the type of court order imposed. The name and address of the treating institution, medical information including diagnosis, and the mental history of the patient are also captured. The database also collects information on the Review Board process including which persons were present at the meeting, and the recommendation of the Board. The survey also collects information on discharge including diagnosis at discharge, destination of the patient, legal status and other pertinent hospital and psychiatric data. As mentioned earlier, these surveys are intended to describe the activities of the administration of justice. As such, statistics generated by these surveys may not necessarily reflect accurate trends of the justice phenomena. For Example, less than 40 % of the crimes are reported to the police nationally. This percentage may change over time and across jurisdictions making jurisdictional comparisons and trend analysis questionable. Changing practices and policy orientations are also reflected in the official statistics. Despite these obvious anomalies, official statistics remain a very useful source of information to monitor justice phenomena in Canada. Experience shows that public reporting of crime changes so slowly that its impact on official statistics becomes insignificant. The same could be said about justice policies and procedures. 160 Sange de Silva Other factors contributing to a high level of reliability of official statistics in Canada include a common criminal code for all provinces and territories, a response rate approaching 100 % for the police surveys, application of common units of counts, definitions, data elements and counting procedures by all respondents, and policies to ensure that each respondent has an opportunity to verify the accuracy and completeness of its own information prior to public release. Victimization Surveys The major goal of the Canadian justice statistics program has historically been the development of performance indicators in the administration of criminal justice. In the past few years however, criminal victimization surveys have been developed and these surveys produce data which are complimentary to the Uniform Crime Reporting data. Canada’s victimization survey is conducted every five years using a national sample of 10,000 adults. In addition to estimating criminal incidents, these surveys provide data on the financial as well as physical costs of victimization, on the fear of being victimized, and on the reasons for not reporting some incidents to the police. They also provide data on the opinions respondents hold of the administration of justice and on the kinds of activities people engage in to avoid victimizations. Special Studies The Special Studies program is given a high priority within the Centre’s activities. Special Studies are devoted to subjects which are of great importance to the policy and management agenda of the governments, and would require special planning, data collection and analysis. At present, several one-time surveys including a national survey of violence against women are being conducted to address the issue of family violence. Sentencing, Youth Recidivism and Aboriginal Justice are also being studied under this program. Canada’s Justice Statistics Program 161 Technical Assistance Technical Assistance is the second major organizational area within the Centre. Its major role is to support the development of statistical programs and the transfer of technology between jurisdictions, helping them to develop operational information systems through technical expertise and resource support. Technical Assistance also provides valuable systems and technical support internally. Technical Assistance comprises four program areas: Jurisdictional Assistance, Centre Assistance, Technology Co-ordination and Information and Client Services. A brief description of each follows. Jurisdictional Assistance The Jurisdictional Assistance Program provides technical and financial assistance to justice jurisdictions to develop information systems capable of contributing to national statistics and information; provides technical leadership, support and assistance in the development of strategies and systems which are supportive of the national statistical effort; keeps the justice community informed about technological innovations; and assists in the delivery and effective utilization of technical services to the Centre. The Jurisdictional Assistance Program system specialists work in close co-operation with analysts, program chiefs and jurisdictional representatives to identify opportunities, develop proposals, secure funding and proceed with system development. Information and Client Services The Information and Client Services unit was established in 1990-91 to promote the use and accessibility of information to users by responding to requests for statistics and information, by providing clearinghouse and brokerage services and by undertaking general communication activities. 162 Sange de Silva Centre Assistance This program provides data processing support to the Centre’s statistical programs for publications and responses to special requests. Services fall into three main categories: (i) information support for Centre products and services; (ii) technology support for the Centre’s hardware and software; and (iit) systems support to build tools which facilitate the analytic process or other activities of a repetitive nature. Technology Co-ordination Group This program is responsible for: (i) the development of in-house systems which facilitate the analytic process or other activities of a repetitive nature; (ii) the management, acquisition and maintenance of technological resources at the Centre; and (iii) the development of a long-term technology strategy. Organization The Canadian Centre for Justice Statistics is directed by a board of directors composed of federal, provincial and territorial deputy ministers responsible for justice and the Chief Statistician of Canada. It is also advised, in a formal organizational setting, by the Canadian Court Administrators. This ensures that the work of the Centre reflects the priorities for those who are charged with the responsibility to deliver justice services in Canada. To ensure autonomy, clarity of purpose and impartiality, the Centre is located outside justice policy departments, and within the central statistical agency - Statistics Canada. Unlike the experience in many other countries, Canada’s justice statistics program is highly centralized with the Centre having the sole responsibility to produce all national information pertaining to crime and the administration of justice. This helps the program to pursue a holistic organizational approach, allowing for more effective 163 Canada’s Justice Statistics Program program planning and program integration. By virtue of its greater size, it offers a critical mass of professional expertise, visible career opportunities and greater continuity of staff. It also offers a clear focal point for the user community. Summary In 1981, the Canadian authorities made a solid commitment to the establishment of a national justice program. Despite very considerable progress made during the last decade, several significant information gaps remain to be filled. Canada has virtually no information on civil or administrative justice. Official records which form the basis for most criminal justice statistics on prisons, on courts and on police need improvement. Improvement can occur in coverage, completeness and consistency and more importantly, in automation. In Canada, the demand is rapidly increasing for information which goes beyond the administration of justice. Reliable measures of wife assault, child abuse and elder abuse are being sought. Canada faces the challenge of refocussing its statistical program to support a growing policy emphasis on crime prevention through social development. 3% L o = £ " b o . PRt L e i le o s b A oF 4 & !'I 15 8 4 i S{s,}‘fi#f}?{ o i i %{4 5 auniy e n #0 ¥ o , b £ P §d & i A RN i i WL g i, Py Jo i % s s & B " i o %< A i e e W & v % . i & o % 4 2 R . i i i % s = e e e 2 & ¥a i pe 0 i i W 4 e e [y 4 s ¥l : e il iy s Ay B e Nl 4 il i ;,%n i A i Ao 2 e ¥ s .%D i o i & i el S b e 3‘» s b 'y g ol ; S ot e ¢ X L e it 1% i s 5 i AL 25 &t 44 B e b i 5 i i i Gags fga e & 1 o ¥ g ) i 54 b L f e ks sy ¥y FA. o e g Y b Y o - 0 WAl o7 o e ‘e b o g ot sl i 4 L Ged poe 0 g - pe e o v on o 4 o bl i ot b it i ad & W 5 o A gk o i E+ o 4 B oA o P P i ok i > i e it ! . e bl 3% i i i€ e % o G o i % i A 3 i 4 " 5 e 2 Jou ok o b o . . £ ek by o iy £ie o 5 i o ih! e i e g ks 7 g e i b ik i o) : et e o f i i O o i 5 5 - 2 i e P h 1 i £ ! e N o . I e o o e L " o T fl. N ! i . 5 = i & e o Al - III. International Efforts and Prospects R o e o i - ol e = &, o i piex 4 . 7 oy % it A ¥ % ¥ i i i o, V] o e i =y o vy h i Fi g e el 3 o 7 gis T % 454 2 b fi s L ; i 9 i R o v '3:‘,‘ o 153 o 2% b Ry i W g ), 0 A e L / y o iy a o e e Ry S48 g e e s i o P g : Ly & o, S by e 7 M . i b i e & fi? 4,: 3 A o S0 i . S o i1 i i i pit - P 5 7 b i e i £k NS "%n v o £ & ! i A sl ' s 4 . . gy N BN Al e A WA AR 5 e i Nt jol ) v S é& i 0 & B b 5 T & Sated i ) : ¢ 3 i i A % e G & o o e L o N i e o o iy . £ o - £ i L o J 0 s i o o % L bt A . it et i L i o . el B s iy :: o i i o & &% ge .‘:} k b G e N 1 e P By 535 i % e (s e . aha i g o4 o . o o i - i o) 4 & & Lo . gL £ e a ALY £54 f i i LA o g . 4o 3 o o 5 SRy 4 . )4 il . 4% e r Lo i Sl ; . % . By oY 5% D SE £ ; e & it s o8 i ol phe L - i e TN 3 ) i ¢ 7z T o a4 o B e b L o o5 s sia) i o N & ol :‘4 % 4 e o i i i, S i e v e % ¥ 7 . o i 3 %p\ft’* N e ol ik - & S I .«f{fi it 2 i i o et e o g ot . k=) i " d:;;' b e, ik, e & 5t R i i o i - oy e B & L : st 43 58 ki b Ee" Ay - 4 e o s 20 o X % o i [ i i A he o g 158 o s o) Etons ¥ 2 LE¢ B Pl iR - Py o o e P g% g e i1 i {3 i i E il oy e 72 b ud’ S 5 i ik S ot 24 o 5 T & £ ¥ o e B i e 5 5 i 3 iy 5 i Vo, i ¥ e RN b i e e all % i o s e Seae ke e s e i S e 7, i b i b IR e ?: b | 5 ki 2 ¥ (o i o o ke 5 i sl ? vl i iz G R i o S 54 e noe 3 F i bt e ?.41;{7 Ay be 25 o i . 5 g . ¢: &5 \ff‘( s oL L e B i - i S t e . i o ey 5 2 . - i e pS s il '. 5: ol % L8 s ; g i 6k e ke i e i frbds % o International Studies and Statistics on Crime and Criminal Justice Chris Lewis This chapter considers work by International organisations to produce statistics on crime and criminal justice for different countries. The main agencies covered are Interpol, the United Nations and the Council of Europe. INTERPOL 1. The longest running set of statistics are those on crime recorded by the police which are produced by Interpol. 2. Interpol, the International Criminal Police Organisation, collects and publishes information on crimes recorded by the police. The first set of statistics was published in 1950 and the most recent set contained information from almost 100 countries on crimes recorded in 1989 and 1990. 3. Offences are divided into thirteen major categories: murder; sex offences; rape; serious assault; aggravated thelt; robbery and violent theft; breaking and entering; theft of motor cars; other thefts; fraud; counterfeit currency offences and drug offences. For each category of offence the number of cases known to the police, the percentage of attempts, the percentage of cases solved and the number of crimes per 100,000 inhabitants is recorded. The total number of offenders and the percentage of women, juveniles and foreigners is also recorded. 168 Chris Lewis 4. Detailed instructions are given on what should be included in each category of offence but even so the different legal definitions in each country mean that the figures have limited use as a basis for comparison. Different police procedural practices also affect the comparability, as does the fact that the volume of unrecorded crime varies from country to country and from year to year.' UNITED NATIONS 5. The first attempt by an international organisation to produce a international set of statistics for the criminal justice as a whole seems to have been the Statistical Report on the state of crime 1937-1946 produced by the UN Secretariat in 1950. This covered 31 countries over a half of the membership of the UN at that time. However, it was many years before this attempt was repeated and it was not until 1977 that the UN Crime Prevention and Criminal Justice Branch in Vienna started to carry out a periodical world survey of crime trends and trends in criminal justice systems. UNITED NATIONS CRIME SURVEYS 6. The first survey covered the years 1970 to 1975 [see report A/32/199]: the second survey the years 1976 to 1980 [sec report A/CONF.121/18 and Corr.1]: and the third survey the years 1981 to 1985 [see report A/CONF.144/6]. Information on these surveys is now available to the International Community, including governments and academic institutions, not only in these reports but also on computer diskettes, via the UNCJIN computer network and, in the form of country criminal justice profiles in the UN Bibliographical IS of the Dag Hammarskjold Library in New York.? 1 The longest running set of statistics are those on crime recorded by the police which are produced by Interpol. 2 For further details on how these sources can be accessed, please contact the coordinator of UNCIIN, School of Criminal Justice, State University of New York, 135 Western Avenue, Albany, New York State 12222, USA. International Studies and Statistics on Crime and Criminal Justice 169 7. In addition the UN regional agencies have produced regional reports dealing with specific topics using and interpreting data from these surveys. For example, the Helsinki Institute for Crime Prevention and Control (HEUNI) has produced Criminal Justice Systems in Europe (Helsinki 1985) and Criminal Justice Systems in Europe and North America (Helsinki 1990): the Australian Institute of Criminology has produced Crime and Justice in Asia and the Pacific (Tokyoand Canberra): the Latin American Institute for the Prevention of Crime and the Treatment of Offenders produced Crime prevention in the context of development {for the UN Havana Congress 1990: and the Asia and Far East Institute for the Prevention of Crime and the Treatment of Offenders produced Crucial issues of criminal justice in Asia (A/CONF.121/UNAFEI) (Funchu, Tokyo, 1985.) 8. The United Nations has just completed the collection of information from their 4th Survey, which deals with the years 1986 to 1990, thus making a total of some 21 years of detailed information. Each survey has seen an increasing number of countries participating, although there are still many countries which do not return any statistics. 9. The major goal of the 4th survey is to improve knowledge about the incidence of reported crime and the structure of criminal justice systems, as a basis for improving the international exchange of information and cooperation. The specific objectives are first, to determine which data are generally available in national databases and sccondly, to serve as an instrument for strengthening cooperation among member states, by setting the review and analysis of national crime-related data in a broader context. 10. The questionnaire, a copy of which can be obtained from the UN Crime Branch in Vienna, is divided into 4 sections: Police, Prosecution, Courts and Prisons. Each section asks questions, not only about the numbers of people involved, but of the manpower and financial resources devoted to the particular part of the justice system. 170 Chris Lewis 11. Each section includes a table requesting data for various different offences. A common classification system is used, which took into account 1S .the likely availability of statistics within member states. This classification Intentional and non-intentional homicide (grand total) Intentional homicide, total, committed and attempts. Non-intentional homicide Assaults, total and major Rape Robbery Theft, total and major Burglary Fraud Embezzlement Drug-related crimes, total, trafficking, simple possession Bribery and/or corruption Other (please specify) 12 A summary contents list of the questionnaire is: Police: A. Manpower resources B. Financial resources C. Crimes recorded nationally by offence D. Crimes recorded in largest city Prosecution: A Manpower resources Financial resources Number prosecuted by offence E Number prosecuted by age/sex International Studies and Statistics on Crime and Criminal Justice 171 Courts: Types of courts Total convicted and acquitted Total convicted by type of sentence Manpower resources Financial resources Numbers convicted by offence QmmoOw» Numbers convicted by age/sex Prisons: Category of prisoner Average time spent awaiting trial by offence Average length actually served by offence Numbers on probation Numbers on parole Manpower resources Prison admissions by offence Prison populations S“STQTMmUOwW> Financial resources 13. After a considerable amount of tidying up work on the database cassettes of the data are now available. Regional reports on the data were produced for the UN Crime Congress in Cairo in 1995. Plans are now being made to continue the data collection into the future so that the 1970- 1990 series can be extended. INTERNATIONAL VICTIM SURVEYS 14. The United Nations has also taken part in the second victimization survey in 1992. This was supervised by the Ministry of Justice in the Netherlands and UNICRI, the UN Research Institute in Rome. A full account is given in Criminal Victimization : a global perspective in the first issue of TRENDS, the UNCIJIN Crime and Justice Letter in 1993. 172 Chris Lewis 15. This survey takes as its starting point that crimes recorded by the official criminal justice agencies such as the police are only a part of the actual crimes committed. In victim surveys an attempt is made to discover the full extent of crime: this in turn enables governments to know whether the extent of reporting to the police is changing. The experience of victim surveys has been that the propensity to report crime varies from country to country and over time, as well as by offence, residential status, age, sex, lifestyle, and, to some extent the development profile of a country. 16. The first survey covered 15 countries (see van Dijk, Mayhew and Killias, Experiences of crime across the world. Deventer, Kluwer, 1990) The second survey comprised 11 developed and 12 developing countries as well as 6 from Eastern Europe. The countries involved were: 1989 Survey Austria Belgium Canada Finland Indonesia France Federal Republic Japan Netherlands of Germany Norway Poland Spain Switzerland UK USA 1992 Survey Austria Canada Finland Belgium Italy Japan Netherlands New Zealand Sweden UK USA Czechoslovakia Georgia Hungary Poland Russian Federation Slovenia Brazil Costa Rica Argentina Egypt India Indonesia Papua New Guinea Philippines South Africa Tunisia Uganda United Republic of Tanzania International Studies and Statistics on Crime and Criminal Justice 173 The results of the second survey are published in Understanding Crime : Experiences of Crime and Crime Control, UNICRI, Rome 1993. 17. The 1992 International Victim Survey used a questionnaire which obtained responses about crimes against individuals aged 16 years and over. The questionnaire also identified opinions and attitudes about reporting to the police, fear of crime, crime prevention and attitudes to punishment. many difficulties were encountered in attempting to ensure comparability between diverse national contexts, which required certain methodological adjustments. 18. In some countrics, the survey was carried out on a national basis, in others only in major cities. In developed countries, data were collected using computer-assisted telephone interviewing; in developing countries and countries of Central and Eastern Europe, by face-to-face interviewing. The survey was based on a population sample of, on average, between 1,999 and 2,000 respondents in each area surveyed. 19. The type of information obtained by the 1992 survey covered crimes against clearly identifiable individuals and households, disregarding, therefore, victimless crimes, such as drug abuse. Perceptions of the latter, which present important victimization risks to individuals and populations, may be included in future surveys. Future of the International Victim Surveys 20. These surveys represent part of a continuous effort to develop the international and national crime and criminal justice data-base. The results of the surveys will be put at the disposal of the international community through publications and database access. The wealth of material collected merits much further analysis, including data validation, secondary analysis, testing of criminological theory, and the development of relevant policy considerations. Chris Lewis 174 21. Preliminary negotiations are under way regarding a third international victim survey to be carried out later in the 1990s, which will cover, it is hoped, some countries not previously involved. An international survey of crime against retailers has also been carried out. COUNCIL OF EUROPE PRISON STATISTICS 22. The Council of Europe covers most countries of Western Europe, and, in the last few years, has extended its membership to include some Eastern European countries such as Hungary. The Council has, since 1983, exercised an important role in bringing together statistics for certain aspects of the Criminal Justice System. The most important of these is their Prison Information Bulletin. 23. The Prison Information Bulletin and related publications gives information, inter alia on the prison population per 100,000 general population, the population of remand and sentenced prisoners per 100,000 general population, the percentage of female prisoners, the percentage of young prisoners, committals to prison per 100,000 general population by type of prisoner and an indicator of average time spent in custody. 24. In June 1990, a new project was introduced called Statistique Penale Annuelle du Conseil de ’Europe. The results of the 1992 enquiry were made available in September 1993, in Statistique Penale Annuelle du Conseil de PEurope Enquete 1992, by Pierre Tournier, of CESDIP, Paris, France, who has been the main worker on this project. 25. This annual enquiry goes very much wider than a collection of tables, including also commentary on the criminal justice systems of different countries and noting trends in the figures. The quality of the data received is examined and useful remarks are made on definitions. Recent changes in legislation, and administrative changes such as amnesties, etc. are also noted. Some very useful analyses are made and presented as charts: for International Studies and Statistics on Crime and Criminal Justice 175 example, the suicide rate in prison is shown to correlate positively with the occupancy rate in prison. CRIME AND CRIMINAL JUSTICE STATISTICS 26. The success of the project to make available prison statistics has encouraged the Council of Europe to go ahead with a pilot study of the possible of producing a Sourcebook of European Statistics on more wider aspects of the Crime and the Criminal Justice System. An expert group under the chairmanship of Professor Martin Killias of Lausanne has reported favourably on the feasibility of collecting relevant and comparable data on crime, prosecutions, courts, prisons, non-custodial penalties, resources, and recidivism. Professor Killias describes this project in more detail in the final chapter of this book. 27. The initial strategy of this group has been to define a set of tables, etc. which, prima facie, secem appropriate for a Euro Sourcebook and to attempt to compile these tables for a small group of countries, represented on the expert group. This pilot will then be used to define what common statistics exist for a larger group of European countries, which could serve for the same sort of comparative purposes as the prison statistics. T i it B SPUCH k2 3 i - i W ke P % ) PN L % A SR i 5L ol i &) o - ooy iy i % i ' it it ¢4 i die f‘ r} iy e i B i B 1 i bed i it o i P b i i B & La iy L b ¥4 B e Y E g i ¢ %’” i i i A 255 % b % 5 o i) o 2 & “} B e o e, i Yoty i i 1 b i & o y e e o i 0 B e b fat} i i 7 B 5. : iy 5" i i ¥ & i i g & ie v & 3 N ot i i L ra 4 o A " ; E. s i = i i L k t; o o i 8 % 5 05,3 Gt o . #i i bl 1 3 pri) k2 3 g 4 i i Ey e = h& 4 A 3 i i 7%, ko i o A Haiilt L vl 3 & 4 % e o G L s ik ke il A & 54 ey o % I Ve o Ty 3 oy s [ £ i i ! e e 8 A4 . e K b b i3 e ol Mr 5 Iy st £ ) gt el L e o 4 o i i iy ) il e e 4 i 4 E k't o B s b s i e Sl e 5 1 ki (o Al % i3 s o or s U N 4 i S e i & 5 o i 7% ‘s st i i o s e v 2 FVu i P » Sy iy 53 s 3 e 1;;3, r - it (; 2 i o 4 s i L B o o e g & HE 0 24 o e o b P £ A 24 qos 3 i i o i Y b #f o ) bt i g I > T ¥ 4 i i . N ki i - i 4 i y; 5 i s "~ 5 ,};2 ) :& i o o i . . i = 2 i b . e i i o 5 i bt ) e 58 5 ke e i i i (e e 0 h ] ey é?e 3 b b bl i v e o o - it P T s [ ¥4 e " % el Ei o i (4 e b g 5 fid - e i 3" R o International Comparison of Crime and Criminal Justice Statistics Max M. Kommer' Introduction In 1937 the Dutch criminologist prof. W.A. Bonger wrote an extensive state-of-the-art article on crime statistics, in which he expressly discussed the subject of international criminal statistics (Bonger, 1937). At that time, he could look back at over 50 years of attempts in this area, and he had to admit that these fifty years had not resulted in much progress. Since then, another 50 years have amply passed, but it is questionable if he would come to other conclusions, could he look back again. In this paper, I want to shortly tackle three major problems of inter- national comparison of crime statistics, and then suggest some practical ways to deal with them. I will restrict myself to what one could call the "classic’ statistics (police and criminal justice statistics), and leave out international crime surveys and self-report studies. The latter offer some advantages over the classic ones, but have their specific disadvantages as well. Discussing them in this paper too would not really clarify things. 1. Problems Anybody who has ever tried to compare the crime level or reactions to crime in one country with that or those in one or more other countries by using the official crime or criminal justice statistics must at least have 1 The author is acting head of the Research Team on Legislation and Administration of Law, Research and Documentation Centre, Minist of Justice. Address: RDC, PO-box 20301, NL-2500 EH Den Haag, The Netherlands 178 Max M. Kommer encountered three major problems (apart from securing the statistics at all). These problems are: - incompatibility of national crime statistics - different criminal law/criminal justice procedure - different criminal justice systems 1.1 Incompatibility of National Statistics Suppose that we want to perform a relatively uncomplicated task, name- ly ranking a number of countries by their murder-rate, i.e. the number of intentional homicides per 100.000 capita. Suppose that we also may call ourselves the proud owners of the official crime statistics of these countries (One could, for instance, go through the sources listed in appendix 1 for an impression of what that would mean). What could, then, be easier? About everything — apart from ranking the same countries by their burglary rate, perhaps. First we will have to make sure that the statistics cover the same period, and preferably the whole country we think they refer to.? We then have to find out if the statistics are broken down to the level we need. In the case of murder it is very probable that we will find a separate category — but it may turn out not to be separately enough.’ Once we have taken this hurdle, we come to the tricky part. Are like cases treated likewise in the different countries? How, for instance, are euthana- sia and illegal abortions registered? Are they registered as murder in one 2 This is not always the case. From my Dutch background I would be inclined to see England, Wales and Scotland as a unity, as well as the United States. We all know, however, that there are no British crime statistics, and I am still puzzled by the fact that ‘Waco (Texas) announced plans to reduce its crime rates by neglecting to reports some of its violent crimes to the FBI’ (Texas Monthly, january 1980, cited in Carley, 1981, p. 1) - I thought that this was only done by the Dutch police in some cities with respect to bicycle-thefts. In the Dutch official statistics the category ‘murder’ has for many years included com- pleted as well as attempted homicides and manslaughters. This implicated a 10 time higher number than would have been the case had attempts been excluded. As a matter of fact, in the 1990 HEUNI Expert Group report on the third UN Crime Survey, The Netherlands were ranked number one on homicide (table 1, p. 54), precisely for this reason (apparently the authors overlooked a remark to this effect made on the question- naire). International Comparison of Crime and Criminal Justice Statistics 179 country, and not at all in another? Are cases registered by the label given by the police, the prosecutor or the judge?’ I do not think there is any need to elaborate on this point any further. We all know the pitfalls, and we all know there is no one way to avoid them apart from being very, very careful. 1.2 Differences in Criminal Law / Criminal Justice Procedure Though we may assume that in our quest for truth about murder rates we will not be hindered by trivialitics as murder not being a crime in one or more countries, we might come across this problem when we shift our attention to less serious crimes. For instance, in some countries shop lifting or damaging of goods may (under circumstances) be defined in criminal law as a misdemeanour instead of a crime. In the case of murder, we may (at least seen from the Dutch perspective) find the opposite: acts that are not considered criminal (like abortion), or that may - under conditions - not be prosecuted (like euthanasia), may be defined as murder in other countries. Now it would seem quit easy to locate differences like these, and to control for them, but in practice it is not. You would need at least three things: a thorough knowledge of the (judicial) language of all countries you want to compare’, figures that are broken down to the level you want and lots of time. As long as it comes to crime statistics it would scem that differences in criminal justice procedure can not be of too much influence on the figures. But what if the police have extended discretionary power in one country (opportunity principle), and very little in another (legality principle)? Then again figures on less serious crime may be incomparable. The real problem, however, arises when we wish to know more about how cases are dealt with in different countries. Criminal law and criminal justice procedure do differ 4 It is common practice for the Dutch police to label a failure to stop your car when summoned to do so by a police officer as an attempted murder if the police officer has had to jump aside in order not to be hit. Most of these cases are eventually dealt with as the small traffic offence they are in essence. I would, for instance, not know what type of offence is meant by the category "child destruction" in the England and Wales criminal statistics, especially where "infanticide" is included in homicide, and "procuring illegal abortion" a separate category. 180 Max M. Kommer markedly, and that makes it very hard to compare outcomes at higher levels of the criminal justice system. Take for instance the "dismissal" or the "transaction” by which a Dutch prosecutor can decide to deal with a case. Do they compare to something more or less the same in other countries? 1.3 Different Criminal Justice Systems Not only may the possible outcomes on different levels in the criminal justice procedure be incomparable, the structure of the criminal justice system itself may differ in such a way as to make comparisons very difficult to make. This can already be the case at the police level. Not a few coun- tries have more than one police force (for instance organized at federal, state and local level, or one reporting to the civil authorities, one reporting to the judicial ones and one reporting to the military), and it is not necessa- rily so that the results of their efforts are combined in one statistical source. At the prosecution level there is of course the difference between sys- tems that include a separate prosecution service for all or specific offenses, and the ones that do not. And then there are the different ways the courts are organized, with various competencies. In the best case this can make it quite difficult to find the figures you are interested in, or to gather them from diverse sources. In the worst it makes it quite impossible to compare criminal procedures in a straightforward statistical way.’ 6 Just as a matter of interest: problems like these can also be encountered when comparing national statistics over different years. In the Netherlands, between 1990 and 1991, a number of small traffic offenses were transferred from criminal law to administrative law. This was done gradually, in the sense that the new procedures were first implemented in only one court district and, subsequently, over a period of two years in the other ones. The transfer brought with it some new jurisdictions for the police, a new organization for the administration of the sanctions, completely different roles for the prosecutor and the judge and a new type of sanction. Statistics on the new system are not yet available, but in appendix 2 the results are presented of an evaluation study, comparing the outcome of the criminal law procedure with that of the administrative procedure. I find them very illustrative of the problems I mentioned here. International Comparison of Crime and Criminal Justice Statistics 181 1.4 Conclusion Efforts to compare crime and/or criminal justice statistics may be hazar- dous to your health, or at least to your good temper. Even if you can draw upon reliable national material’, there are a lot of problems to overcome. You need to have a lot of background information on the statistics them- selves, on the criminal laws, the criminal justice procedures and the criminal justice systems you want to compare. And you will have to find a way to present your findings in a way that they are accessible to a public that will be very interested, but not too willing to spend much time on subtleties: policy makers and politicians. 2. Solutions Taking into account the character of the problems discussed, it looks at least pretentious to propose solutions. That I still dare to do so has a lot to do with the fact that I am not the first and only one who did. In fact, all the ideas I had (and some more) when I first set out to survey the field, proved to be at least 60 years late. The suggestions I have to offer can be divided into three categories: restriction, standardization and explication. I will discuss them in the fol- lowing paragraphs. 2.1 Restriction If national crime figures in general can not be compared, one might be wise to restrict oneself to those figures or types of comparisons that can, or that at least would yield a not too distorted picture. An example of such a restriction was introduced in paragraph 1.1, where I suggested that we restrict ourselves - for the sake of argument - to mur- 7 Of course, you should know about reporting procedures - and preferably not the official but rather the actual ones. Max M. Kommer 182 der. This suggestion was also made by Bodio in 188 and Zahn in 1925% the idea behind it is that there is a limited number of crimes of which the definition will not differ too much between countries, and that by their seriousness or number can be regarded as valid indicators of the crime situation in the respective countries. Of course, as I indicated by taking this example, there still are a lot of problems to cope with, but restriction has the advantage that it makes it easier to implement some of the other solu- tions that may help to overcome those. A second type of restriction is in the type of comparison. If differences in criminal law and criminal justice systems make it impossible to compare crime or intervention levels, it may still be possible to compare changes over the years: does crime rise (or fall) at a higher or lower rate in one country than in another. Of course it would be necessary to also have an indication of the possible sources of changes in the rates compared; it would be easier to provide these if the number of crimes compared would be restricted too. Just for the record: using dynamic in stead of static comparisons was already suggested by Starke in 1888, in his report to the Institut Internatio- nal de Statistique. 2.2 Standardization One of the best ways to secure comparability is to standardize the things you want to compare and the instruments you use for comparison. Here again we can go back at least a century to find suggestions in this sense. Bodio (1891) proposed to draw up a comparative terminology, and Yvernes (1911) put forward the idea to devise a uniform outline for national crime statistics. Both suggestions are, in my opinion, quite valuable. A comparative terminology would make official statistics a lot easier to read, even for polyglots. What is even more, it would free the national statistical institutes from the ’obligation’ to translate part of their publications in English (or whatever language would seem most suitable), and the readers from the 8 All references to publications from the period 1888 - 1937 are based on Bonger, 1937. International Comparison of Crime and Criminal Justice Statistics 183 guesswork that is sometimes needed to understand what exactly is meant by the result. It is the more interesting where formulating it seems quite feasible, and at the same time would force those who set out to do so to go quite deep into a lot of the (definitional) problems I mentioned above. Designing a uniform outline is, in itself, probably the most tempting proposal of all. It seems, however, an illusion that it will ever be adopted by more than a few countries - that, from then on, would have to cope with a considerable problem when comparing their national statistics over the years before and after adoption. Still, if such an outline could be devised for a limited number of offenses and stages in the criminal justice process, national statistical institutes might be prepared to add a ’uniform crime report’ to their range of publications - especially if they knew that a consi- derable number of other countries would do the same. A very drastic level of standardization was suggested by Gini and Spal- lanzi in 1931. They proposed to fix a general judicial content of a limited number of specific crimes, so as to make comparison easier. It is not difficult to think of a number of objections that could be brought forward against this idea. It would, in the most extreme case, mean that national criminal law should be adapted to fit statisticians needs, which is not very likely to happen. In less extreme cases it would mean that in the whole process of registration and reporting on the national level these specific crimes should be isolated from crimes that under prevailing law would be considered the same. This would be very impractical, and at least produce a lot of extra work. 2.3 Explication A combination of elements of the two former categories, together with a new approach, was suggested by De Roos in 1927. He proposed to draw up a manual for ideal national crime statistics and to give guidelines for the best ways to interpret these for international comparison. A manual, of course, would be part of an effort to standardize (and probably to restrict), but giving guidelines for interpretation is an interesting extension. It would be very helpful to all who want to take comparing national statistics as a 184 Max M. Kommer serious matter, but I doubt if it would do much against the inexpert (ab)use of statistics by newspapers, policy makers and politicians. Manuals and guidelines tend to be left unread by practically all people who would bene- fit most from them, and there is no reason to assume that in this case it would be otherwise. Still, it would probably be one of the most practical things to do, especially where preparing the guidelines would bring with it a very intense confrontation with all the problems one can meet, and might thereby stimulate to think of new solutions. In fact, in 1929 De Roos proposed yet another solution, that I would be inclined to see as a first step towards general guidelines: let national experts write annexes to national statistics, giving an interpretation of the figures. If this would be done in close con- tact with national experts from other countries, it would help enormously in providing the material on which guidelines could be based. 2.4 Conclusion All these solutions have been proposed before, some more than a cen- tury ago. Does that mean that they are impractical? I do not think so, at least not of all of them. In fact restriction, standardization and explication are at the basis of the Uniform Crime Reports used in the USA to com- pare crime figures over more than 50 states, each having their own national statistics, laws and criminal justice systems. This shows that each of these ways is at least partly possible, though it may be a hard going.’ 3. The Way Ahead Apart from the mere existence of the Uniform Crime Reports, there are two reasons why I think we should not let the failures of the past cast a shadow over the future. Firstly, a proposal that included most of the solu- 9 When devising international comparability of crime and criminal justice statistics, it would be quite unwise not to draw on the experiences with the UCR. That will probably guard us from both over-optimism and needless pessimism. International Comparison of Crime and Criminal Justice Statistics 185 tions mentioned above was part of a report that was accepted by the con- gress of the International Statistical Institute in 1937, and then sent to most European governments. It is most probable that at that time their interests were not really in international comparisons of any statistics, apart from arms-production. Maybe, therefore, we should not conclude that all these suggestions led to failures, but rather that they were never implemented. Of course I know that since 1937 a lot of work has been done, or at least has been the subject of ongoing discussion. The Council of Europe at least seems to have a long going tradition of organizing workshops and estab- lishing expert committees on the subject. Inevitably, this must lead to some- thing. Secondly, over the years technology has reached a level that, in my opinion, is more suitable for an endeavour as complicated as this one. Tele- fax and E-mail make direct communication a lot easier, or at least can do so if organizations involved decide to give their members/employees full access to them. Equally important, especially when it comes to the presen- tation of results, is the availability of sophisticated database systems and of hypertext. Using these media rather than paper would make it a lot easier to have comments, annotations and explications available at every given point and moment. They would also be very helpful in dealing with dif- ferences in breakdown level and in preparing customized' versions of a global data set. If we want to go on developing comparable crime and criminal justice system statistics, there seem to be two things we can learn from the past. The first is that we can never hope to succeed when the governments involved do not see the importance of having these statistics readily availa- ble. Not only will the development be quite expensive in terms of both financial and human resources, it is also probable that changes will have to be made to national statistics and procedures. I am not too sure that national governments will be prepared to support any real efforts in this field. What, I wonder, would they really benefit from its results? Inter- national comparisons are currently used by governments only for political 10 Especially: with descriptions and comments in any desired language, and with special references to the user’s familiar criminal law / criminal justice system. f 186 Max M. Kommer goals: showing that their policies are fruitful, but that extra money is nee- ded in order to prevent the country to become even more criminal than the other ones. To make these comparisons, they do not need a continuous production of comparisons, and not even reliable tools for comparing. What counts is the result, and in the present situation it is not too difficult to go on looking for an interesting result between a lot of non-interesting (i.e. politically not useful) ones - who would be able to fight them? Our best hope, therefore, would be the supra-national bodies, as well as continuous pressure on national governments. The second thing that will be needed is an enormous effort from a lot of people: statisticians, experts on crime and criminal justice, and specialists in related fields. They must be prepared to set out on a long, uncertain journey—with little hope of any other reward then their own satisfaction in having reached yet another problem. After all, what is so pleasant in making your own national statistics accessible to others, if all you get in return is the possibility to compare them with those of other countries? And if you are interested in the comparison itself, why bother to go on making your own statistics accessible? 187 International Comparison of Crime and Criminal Justice Statistics References Bodio, L.: L’organisation de la statistique pénale en Italie et sur les difficultés qui sopposent aux comparaisons internationales dans cette branche de la statistique. In: Bulletin de DInstitut International de Statistique 1V, 2e Livraison, 1888, p. 165 ff. Bodio, L.: Rapport sur la statistique judiciaire penale. In: Bulletin de I'Institut International de Statistique VI, 2¢ Livraison, 1891, p. 122 ff. Bonger, W. A.: Over criminele statistiek. Een bijdrage tot haar geschiedenis en haar theorie. In: Tijdschrift voor strafrecht, vol. 48, 1937. Reprinted in: W. A. Bonger, Verspreide geschriften. Amsterdam, 1950, De Arbei- derspers. Carley, M.: Social Measurement and Social Indicators; Issues of Policy and Theory, London, 1981, George Allen & Unwin. Gina & Spalmanzi: Sulla comparazione dei dati di Statistica Criminale dei diversi Stati. In: Bulletin de I'Institut International de Statistique XX VI, 2e Livraison, 1931, p. 567 ff. Roos, De: Consonnes et voyelles, Communication sur la Statistique crimi- nelle. In: Bulletin de I'Institut International de Statistique XXIII, 2e Livraison, 1928. Roos, De: Rapport de la Commission pour 'étude comparative des statis- tiques criminelles dans divers pays. In: Bulletin de I'Institut International de Statistique XXIV, 2¢e Livraison, 1929, p. 567 ff. Starke, W.: Des ¢lements essenticls qui doivent figurer dans la statistique criminelle et des moyens de les rendre comparables. In: Bulletin de IInstitut International de Statistique 1V, 1e Livraison, 1888, p. 69 ff. 188 Max M. Kommer Yvernés, M.: De la recherche des causes du crime par les procédés statis- tiques. In: Bulletin de I'Institut International de Statistique XIX, le Livraison, 1911, p. 270 ff. Zahn, F.: Internationale Kulturstatistik. In: Bulletin de I'Institut Internatio- nal de Statistique XXII, 3e Livraison, 1925. Appendix 1: Some sources To get an impression of the problems of comparing national statistics, it would suffice to go through the information on murder as it can be found in the following sources. Federal Republic of Germany Polizeiliche Kriminalstatistik 1989, pp. 105-108. Wiesbaden, 1990, Bun- deskriminalamt. USA Crime in the United States; Uniform Crime Reports 1989, pp. 7-13. Washington, DC, 1990, U.S. Department of Justice/FBI. The Netherlands Kwartaalbericht rechtsbescherming en veiligheid 90/2, pp. 19-20. Voorburg, 1990, Centraal Bureau voor de Statistiek. England and Wales Criminal Statistics England and Wales, 1988, p. 37. London, 1989, Home Office. 189 International Comparison of Crime and Criminal Justice Statistics Finland Oikenstilastollinen vuosikirja (Yearbook of Justice Statistics) 1992, pp. 30- 31. Helsinki, 1993, Tilastokeskus. HEUNI Criminal Justice Systems in Europe and North America, p. 54 and p. 189. Helsinki, 1990, Helsinki Institute for Crime Prevention and Control, affiliated with the United Nations Appendix 2: comparison of dealing with small traffic offenses under crimi- nal vs. administrative law: The Netherlands, 1990/91. 190 Max M. Kommer ADMINISTRATIVE LAW traffic ticket drop notice payment to police police level / central collection office | impos reminder Appeal payment to P.P. sible et ] I granted not Open Payment Bailiff granted ] s notice payment Open payment (open) Open No succes RESULT PER 1.000 CASES AFTER 1 YEAR: PAID: 940 DROPPED: 20 OPEN: 40 International Comparison of Crime and Criminal Justice Statistics 191 CRIMINAL LAW traffic ticket g 1 drop to prose- payment cutor police level drop to court payment 80 é prosecutor’s level SESSION ] dismissal Fine order to court level arrest QO ='N x g R payment / Open police level impriss. RESULT PER 1.000 CASES AFTER > 2,5 YEARS: PAID / SUBSTITUTE IMPRISONMENT: 938 DROPPED: 33 OPEN: 29 a it g e b i o i o i 5 o e ) R b & n it N, e Wi e i ¥ o Lo §< B N7 o ) P 43 g %1 9 e o (e Vs 7 b5 3 pRA. R g E ' B e i i At i % ¥ % = 0 > i 5 = i g & . "5'4 L 78 47 fla A e 0 RS i P ki % B i it S, QA e o . e & (i . . 3N e i s Y S ,%« o ke £ by’ il ‘s‘*f o 5, ‘/ o £ (i ) . . e A 5 S 3 i i % e i " i 5" 5 hid o 0 7 9 R &) PR A . i} e ’ = D 4 [ L i #) # b Ay b X S, ;? oA ) ey s A e o e e Ry i i 3 @ A e ik 4 Y i PR b A ) s P Yo o e i i e 4 i b & i I 5 g i % i P By it o B ¢ aeny & o L Ty 5 b Lo 2y i e ¥y ; i W 0 i e L e % % Ll . £ o . % g E b1 S S e ,}!’ B 5 = e o e £ i hY o it digte i ! . . ;‘ :' . @ i g bLi b hots ok 7 Y o8 ¥ e b i A% i :: ooy e oA e ¥ P 5 Lt o o L e iy . ‘:; < oy o & s P I, 8 o i el 4 Nl i i - 7, 3 Pl g ot £ . e £ i & P, 0 s s e o e N : 1o 5 &) Lt 3 o SR 4y 4 R i i a4 e i o Tt i il g - 63 i s o S ql 3 ?: i il 0 Al b o E. {S"fi, s i N i ] % 47 i & s b ) i ol o N bk s e 6 &G et E by Ja g s A S He g s '?t i E ) o PRV = g ok & e 9 i g3 i) &t i i i i g & % ISy <A it b = o e o S i 5 e hi it ® X e 4 iy, & e bl s 4 o F i T i ) 5 g (4 ;;‘,nr :g 15 34 g ok e i ke U i % 3 0 ! hot 5, 0 b < s 3 j}? ot & M e i v g i o (s 13 1 Y g i e it I o i e St 3 et b e 5 il S v o ko o it b o 5 i “ TP, ;l e bl i e e PERN & i e oY o % Kt i S Ve, ,}A T e . & 4 b it A e 7 h<] 5 g b4 p h e s % o i ¥ 0 3 ' i iz 3 £ v A bor Sy s ‘3 : ‘f g s 0 i st iic 4 4 $ e 1 A = o B SR b ms o iy el 88 i £y o e i i . / % Gig By & g s thad i 2 ant ¥ a0 i g {f{{f’w e ke 2 b e5s ’; o % ; o & 5! b b e ik i [ i % . i =y o o) < 9 %, 2 iE: i g2 % R i e & - b B 15 o A o i 2 ,_‘; ¥ Ay K] § Ay X B st . 495 o o g 5 Bigich e X s . i % 5 o % & g i ¢ 4 & % s A1 oy [ ke 8 £ & i 9 it i 5 s 5 i 9 = ey y i A e : i 'e . i R e i ¥ i, - e e i b § i o 3 " o ¥ G - et i . g7 i vl b e s ’: P L fch g o AR e o e h i WYy (s 55 ) . 5 Bt s L Pe L o b oM i e e s ¥ 2 4 fa i 3. o 3t E i g T e P i i s s i i o i 3 % 6 ; ) G o i3 o s . B o i s Pl o - i 2 i " o o ] L R . S e i i B o e 5y A ne Loy i P iy fid = e 3 o o W ek Lo W & e it £ i ps e . s e 8 o S 1 A s ,? £ b £ b % i i ke i 5o o =3 B gt o oo k3 ey B 5 N o i = o, i o o s g g i e o e, SAT% peis . i L £ o 5 i N & & ¢ L 5 g : ", v g A bR i* s e L i i o s v 2 3 e b = R gt ¥ o A B e il o) o el i i 3 g e i . % i e i ks i : s A0 ik g g A 6 A i i i 1 & 3 i oF iy s s 2 Fa R, i 5 ey ? & & . i 4oL E . P 5 e :‘:’ Sl B £ . £ 5 1 & : e by e i e i A % e o i i o & = i = e e % s sty a4 Iri e 2 £ iy i il o . " o b i 5 o T fifg i = i Sae s i pa e g ey i L g £ i A e e Ry o & 2 5 & 5 7 e s : £ fil. a0 % s g it b g j i o -3 o i g e ? & 5 g Gt i ; % ,x R4 ) i '; s el 5 ol i o g o e e [ 3 ; %5 i b d e g s i i ok S s oh = 2 & 0 i P 3 i 2 i Lo (! o R X 5 0 e g E i & R o dr oot ik o A i b & ) ? ¥ 44 & s o i o o - & 5 & e £ E ,‘:: b, e S : i o 35 s s i e 5. !’ ) ‘;:4 e @ o % i (R, A o h Pt gk ey o e ) ‘« o ey o i S s fa o 4 Ref his : i v ;: AR e o i (s s & . b 3 i i : o s jisnd %t o § A i = i R o International Prospects for Statistics of Criminality Jean Gratia The first question we need to address is what we are trying to measure with criminal statistics. For the criminologist, the question is how far crime penetrates into the life of society. More pragmatically, it is how we can portrait, with as precise a scientific instrument as we can obtain, a picture of society in which crime is the main agent. International crime statistics are somewhat like photographs taken from a satellite. Such photographs provide neither the detail nor the beauty of an aerial snap but nevertheless give objects an extraordinary scale and are extremely important in such activities as military operations. In short the criminologist needs basic information as a painter needs colours. Such information is found by counting and we first must consider the question, what is it that we should count. Different Techniques of Crime Measurement There are three possible ways of measuring crime: using police data using data from victims using data on known criminals Each of these approaches the problem in a different way, uses different techniques of measurement, and different measuring instruments. Tradi- tionally, the first measures of crime were those obtained on the basis of the work of the Criminal Investigation Department; that is, statistics of known criminals, who were apprehended and dealt with in some way. 194 Jean Gratia Since the second world war, new techniques of appraisal have emerged: on the one hand, the victimization inquiries, which are practised today with some success: on the other hand the self-confession inquiries, which are based on interviews with the criminals and the offenders themselves in front of a criminologist. Self-confession is a recent introduction in criminology and is hardly widespread: it still has to prove its worth. Victimization Enquiries Victimization enquiries were devised in the USA by Reiss and Ennis in the mid-1960s under presssure from a commission of enquiry on crime, whose chairman was dissatisfied with the statistics placed at his disposal. Today they work very well as instruments relying on the country’s population rather than on police machinery, because whereas victims are numerous, they have no really representative organizations at their disposal: they cannot be questioned as such. It is therefore the mass of a given territory’s inhabitants that must be questioned. Fortunately for the community but not for the enquirer, the victims are like the gold nuggets in the sand of a river; isolating them requires time and effort. If one takes France as an example; official statistics refer to 3,831,000 crimes and offences in 1992 for a population of 56.7 million inhabitants, or a stated offence per 15 inhabitants. But in the course of the inquiries, victims must declare themselves and that is a weakness. Answers are very subjective; they are also prone to errors: they do not follow official definitions: many people will interiorise the fact they are a victim and deny to themselves unpleasant memories. The main aim of victimization studies is to indicate the true number of criminal acts without the compulsory filter of police forces. We know that police, for numerous reasons ignore some crimes: some of these crimes are minor offences such as acts of vandalism or petty crime: others are more serious such as rape or incest. This source of information is on the other hand entirely focused on the victim and, as a consequence, overlooks a whole range of offences with no specific victim: these include offences such 195 International Prospects for Statistics of Criminality as shoplifting from large store, vandalism of public buildings: defrauding the railways. Neither would we expect offences to be reported which are usually only discoverd by the police themselves: for example, drug-taking. Self-confession Investigations These use the criminals and the offenders themselves as the source of information. There are of course a great many criminals; they are at least as numerous as offences committed or attempted, but numerous authors can slip into a single creature of flesh and blood. This latter remark is important because it happens rather frequently that the same individual commits several, even several hundred offences. Therefore, it is not possible to count the total number of criminals. At the most, one may know those who are caught. These can be a pre- cise source of information when they are interrogated in prison or else- where, in the course of self-confession inquiries. One will immediately understand that this manner of obtaining information is somehow difficult to implement and it will be added, to stress the line of the criticism, that their partiality is such that one may wonder how real the information col- lected can be. Nevertheless, these inquiries give us information on a field of observation which is indeed limited, but extremely accurate; it is there- fore an instrument of measurement which is not be neglected, in particular in the field of behaviour sanctioned by penal law, but which does not lead to the production of an identifiable victim and can therefore not be assesses by the device of victimization inquiries. The Criminal Investigation Department Statistics In this way, victimization inquiries and self-confession inquiries open, in theory at least, on spaces which are boundless, or near boundless. But, in reality, although they allow the writing of some case studies limited both in space and in time, they have never permitted a general and regular meas- urement of criminality. This role, continous in time and space, is therefore 196 Jean Gratia carried out by the instrument for measuring criminality which is the Crimi- nal Investigation Department statistics in nearly every nation of the world. The basic function for criminology has always been fulfilled from the procedures intended for the judiciary power. In any case it was probably never set up with the aim of contributing to the criminological work, but with an operational purpose to direct the police action. International Criminal Statistics Today The UN has already studied the subject with an international vision. Its personnel and its computers have stored a multitude of data supplied by member states. But they soon revealed themselves ill-assorted according to the particular area of the globe being considered. And that is why some years ago, the UNO presented a first survey on the state of crime in the world, by limiting itself to general considerations through a lack of as standardised gauge. This was a wise decision in view of the lack of homogeneous data. Similarly, INTERPOL, the International Criminal Police Organization, which has at its disposal a strong network of exchange of information between policemen, set up, during the 1970s, a statistical collection device in member countries (c. 150) on the basis of common definitions. This system rests today on simple and not very constraining rules unanimously accepted in 1976 by the Organization’s General Meeting. But, paradoxically data gathered in that fashion are not in the least meant to establish comparisons between the countries by reason of the definitions of punishable actions, differing according to the various legislations and, it could be added, because of the diversity of counting methods. For all that, INTERPOL’s criminal statistics are the only ones to be kept internationally according to a common norm. They are regularly published and are accessible to criminologists. The latest publication, for 1991, showed that over two-thirds of member countries had not only agreed to these statistics but had actually produced them. International Prospects for Statistics of Criminality 197 These commendable actions on the part of the international organiza- tions arouse nevertheless the scientific dissatisfaction of the entire world’s criminologists whereas the the developed countries - and even some others - have rather satisfactory internal instruments at their disposal. Therefore, it does not seem impossible within a reasonable period to create one or several universal, or at least semi-universal instruments for measuring crime. Doctor Kube, of the German BKA, recently stated in the International Review of Criminology and Technical Police (April 1992) that: it is the task of the EEC’s national research and statistics services to devise common directives concerning police criminal statistics and to publish and analyse periodically data relating to criminality. A Field of Observation in Keeping with our Time Since the most ancient civilisations of Egypt and Mesopotamia, counting and taking a census have been tasks which are a matter for the scribes or, more prosaically, the civil servants who have succeeded them. So, apart from disputing the principle of sovereignty, international instruments ought not to question, in their very existence, the internal devices of the States. In other words, national and international instruments must not exclude one another. An international vision must simply be obtained by broadening the permitted sphere of observation inside the country. One can as well say that this field of observation is not the real crimi- nality nor even the known criminality, but rather the treated criminality. Indeed, a great number of statistical systems do not retain the least serious offences within the known criminality (for example, France does not classify fines) nor the offences which will not be the subject of a procedure aimed at the judiciary power because of the attitude of the victims helped by policemen who do not wish any traces of them kept. 198 Jean Gratia Moreover police recorded crime statistics, in general, reflect the characteristics of the administrative system for their collection and we should recognise this. For example, recent years have seen a growth in police initiatives which make it easier to report crimes, for example, through the growth of telephones and computers and better welcoming services. These factors certainly make police crime statistics an imperfect instrument for assessing criminality, but nevertheless these are the best we have, because they are the closest to reality. The Counting One can count the criminal or the offending facts which have been registered by the police; one can also count those which have been elucidated, solved, resolved ... It is the basis of the criminal statistics. Accounting the authors and even the victims could also be added. It is nevertheless difficult to obtain an accountancy of the victims from the police simply because police look for the offenders ... not for the victims. With a luxury of detail regarding the criminals, statistics change into an operational analysis of criminality which allows a more efficient fight against crime that makes the investment worthy. The description of the victims does itself only bring about criminological satisfaction and does hardly contribute to police action. = The offenders. It is here that the relevance of the device will appear because there can be an unlimited number of classifications. It will be possible to anticipate the age group, sex, socio-professional categories of the implicated persons or even indicate the judicial means used (measures of custody, control by a doctor, presence of a counsel ...) - The victims. It will finally be possible to count and even classify them also. This said, statistics have a price and these data being of less interest to the police, the financing of this part of the statistical instrument shall probably have to be sought outside Criminal Investiga- tion Department powers. International Prospects for Statistics of Criminality 199 The Index To classify crimes and offences, France uses 106 indices plus a "various non specified" category. This is both a little and a lot. Little because it is not possible to compile a register within a unit of offences provided for in the French Code of Criminal law ... and the other texts of the law. A lot, because the imputation at the initial stage of collection carried out through the 107 indices by the police, requires a certain skill. That is however a reasonable option if one wants to avoid a too rough picture of criminality. It must be stressed that in international applications, the number of indices must not be excessive so that it may be adjusted to local legislations. Of course, the indices must first be gathered under headings which group themselves in categories. The latter must be a number below ten. In France, the choice remains every classic; today one distinguishes: thefts economic and financial offences crimes and offences against persons offences in the matter of drugs offences against the alien police destruction, degradation and damage of property other crimes or offences These seven categories correspond to the present centres of interest. Besides a short time ago, the last four categories were regrouped under the hardly enlightening label, it must be acknowledged of "drugs, public peace and regulations". The Accounting Units In France, statistics of facts noted by the services of the CID represent the criminality denounced by both the victims in their own name and by the police on behalf of the whole community. It must, however, be specified that the contraventions remain outside the field of statistical observation because the offender is neither a criminal nor an offender: he perpetrates 200 Jean Gratia acts which are indeed antisocial, but which are not perceived as a source of insecurity. The statistics’ starting points are the facts of crimes or offences recounted in the procedure intended for the judiciary powers. By convention, the accounting unit is proper to each index: in the case of murder, it is the victims who are counted and not the crimes themselves (one crime may cause several victims). On the contrary, for receiving stolen goods, it is not the number of objects received which serves as an accounting unit, but in fact the established procedure for this offence. It is therefore always the socially sought information which must be used as an accounting unit. These French examples draw attention to the conventional aspect of statistics. At an international level, one must also integrate conventions relating to the accounting units. It is illusory to compare statistics which have not been prepared according to the same norm, even when the defini- tions of the facts are the same. The guide of statistical methodology used in France specifies for each index what the accounting unit is; such a document is indispensable to make our national statistics scientifically acceptable. The same would be the case for international statistics. The Collecting If the accounting units have to be the same, the technical methods of collection need not necessarily be uniform. This is indeed the case in France where two collecting techniques co-exist without prejudicing the final product; one is entirely centralised (that of the National Gendarmerie) and the other largely decentralised (that of the National Police). For all that, France is a country where the administrative fabric is of a rather centralised type, although it is less so nowadays than some thirty years ago: collecting in the shortest time is therefore unquestionably made easier. International Prospects for Statistics of Criminality 201 The Future If an international device was put into place, it should also quickly process information so that a true analysis of criminality could be carried out, not only by university criminologists who do not integrate time constraints, but also by the police who have to fight crime in real time. Therefore, two functioning hypotheses can be presented: Either a Euro- pean Bureau defines the rules, ensures the collecting and redistributes the information, or an assembly of experts defines the rules and agrees to exchange data mutually ... and quickly. In both cases, periodicity has to be defined. This second hypothesis is the easiest to set up immediately. It even seems that numerous countries view it favourably. g g o e o \‘ B 4&1{ i = . L t:2 St 3 - 5 & t & i fis 3 % s (X% o (% el i i 9 i o i & T i Iy 0 }," s b i B i Bl i’ e s 5 fis H A o 5 1y o v o Ll,’ 5 L 4R R i o b 4 s L) i i fi}%‘ iy . ! ¥ ¢ g o 2 Sder® r L [ 4 e e s H o o SN 3 gt i bl it i % s IE oy et Vi g ey e i 5 )yéj‘ > 5, ; gl '3 fame 3 i i b W i i A g - i . hl b o, I e il i e 5 Mé’? i e '5{:” it s it . e i M i 1’ o : b e i & i & Y s i ¥ gt ey i i N1k E'5 " i3 al oy o O v g i 5, e 580 g Esy o ’bé% b el i a0 i1 i s e b : s gy 74 i i i phd iy 7 i . i ol i LR o W S o ) i b A N (i} e il e Ao £ i b by . 2 o & g ey B) N i i E = 2 o '%f §. - £ Gl b 3 i ¥ £ i o s % s i b Ll - i 0% s i o ot o ® i i o E v fig«, P o fo A B P . Taes % Pt & B b 5 b . i A o {5 has, g T b A 15 i o i g R ol o o iy % i 7 e [na i 2 é)';‘ i 2 e K ¢ fiad a0 ¥, i e 2 B ok 4 Ra s K b e e g e Vs A N 4 b g g ; i £ it 5 iy B e o ok it & :‘ b By s i ” g Bt o i | i 10 5 ,5: . o o S . e o & e L 3 e - o e % £ i) ot i ot o e D P 2 ALy = &y i) i ory % ot A ¥ o i e . i }‘s o 7 ¥ o S S A b it et 42 ! 8 i) N b B % o 05 s o ) I 5 € ik o i P b i o e it (8 e ) e L7 P S 5 § A - i ) Uy st ot ay e pi (9 Sy A 2] i =y . ol Sy i % 5 ¥ ¥, et el 5 i i o . . . L %i’s s " i o i e o /i & ! Ay 3 5 . % [ R i [ o i B ) A % o e o i) 5 el Hod ol 0 s ) & b R :%? & e 5 i z i i i s fl: B i il & s s s Hid e o i " ! L o s i i J g b 8 B £ o o) Bt iy ¥ oY 28 5 £ ik i 8 o . ¢ 0 L 55 Vi P 2y g s % N g2 e it it e £ g ot W o X o e . o A i S g e o ) ik 4 o5 o Yt £ by Ty . A0 " o & e . o Lot o e o » i e, i3 b 2 s 'i : 12-‘%% ' ok 5 o i L 2] ¥ g A, b e ¥ W e % 5 S oy o P v . Fd [ e o e el e - . o s % ig\k s s g e 2 Bl g o ! & 0 oA s i i o ety s i LA i EoAsh b/ The European Criminal Justice Sourcebook Project Martin Killias 1. Background Interest in comparative criminology is as old as criminology itself. Tocqueville was sent to America by King Louis-Philippe in 1831/32 to learn what he could about the American prison system. At the same time, Quetelet started his interregional and international statistical comparisons. Indeed, one might suspect that interest in comparative rescarch was greater at that time than later in the history of criminology. Since data collection in the field of criminal justice started in Western Europe early in the 19th Century, particularly after the Restoration in France of 1815, Quetelet and his colleagues had few time series to analyse. They were left with cross- sectional data as the only relevant material available. Given the absence of surveys and studies based on police or court files at that time, these early criminologists were virtually obliged to turn to international and/or interregional statistical comparisons. With this focus of research, they had to struggle with all the difficulties and pitfalls of comparative research which many researchers have encountered and criticized ever since. It is interesting to note that criticisms that valid comparisons across jurisdictions are virtually impossible given the many differences in legal definitions, reporting and recording practices, are about as old as the earliest writings in comparative criminology. Already in 1832 - Quetelet’s main works were first published in 1830 and 1832 - a Geneva biologist, Alphonse de Candolle, came up with most of the criticisms we have been familiar with ever since. These difficulties and the resignation they have provoked may have contributed to the relative scarcity of international comparative data in criminal justice, compared to other fields such as public health, the economy, public finance etc. Moreover, judging by the scepticism of many published reviews towards the recent Martin Killias 204 international victimization surveys and self-report studies, it may be fair to say that critical reservation is the prevailing attitude today towards international research among most criminologists. This general reservation may do as much harm to criminology as uncritical use of questionable data. If our ultimate aim is theory testing, we have to rely upon data in order to reject hypotheses. Without data, no theory, no matter how invalid or even absurd it may be, can ever be disproved. The less data available, the better for the survival of theories! In this sense, an excessively critical attitude towards data ultimately leads to uncritical tolerance towards wrong theories. This pitfall is particularly sad in an area where, more than with national or local data, policy-related variables could be studied: Does a lower age of criminal responsibility reduce or foster juvenile delinquency? How do sentences affect crime rates? Does a higher certainty of conviction reduce crime rates? What impact do policies towards drug use have on drug-related crimes? All these and many other questions can reasonably be studied only across jurisdictions with different approaches. Europe’s legal fragmentation may offer many opportunities for evaluating the effects of different criminal justice policies. (This, by the way, may be one of the best reasons for not centralizing Europe’s criminal law and justice systems: too much chance for innovation and learning may be lost if we abandon diversity in this field.) But we will not be able to take advantage of this chance without data measuring the outcomes in countries with different policies. 2. The American Model Since 1973, the US Department of Justice has published every year the Sourcebook of Crimminal Justice Statistics, a manual full of tables and graphs of impressive proportions. This volume contains data on offences recorded 205 The European Criminal Justice Sourcebook Project by the police, offenders known to the police, victimization, public attitudes towards criminal justice related topics, corrections, and some varying topics of general concern. The American sourcebook is a great model to follow. As such, it stimulates interest and efforts at imitation, but may also discourage any initiative to try something similar elsewhere, in Europe in our case. In this connection one easily forgets that criminal law and procedure are, in the United States, matters regulated by the States. This means that many difficulties which seem to make impossible any European data collection in this field harass US statisticians just as much. The FBI Index, on which substantial parts of the US Sourcebook are based, is indeed a compilation of many items of data collected at local and state level without too much coordination. If Americans were able to overcome the measurement issues we are struggling with, we may conclude that efforts in the same direction in Europe should not be viewed as wasted energy from the beginning. The admiration for the American sourcebook has already produced a number of paradoxical consequences in European criminology except perhaps for colleagues from major research centres with large international libraries, criminal justice statistics even from neighbouring countries might be hard to locate, even harder to read and almost impossible to use, given the lack of familiarity with legal and statistical definitions and collection routines. In daily work to set a domestic figure against some external data it is often easier to give a comparative figure from the US rather than from a bordering country. This is a most unfortunate situation. It certainly weakens our European identity, and it does not help us to learn from experiences in other European countries. As far as reference is made to current innovations in criminal justice somewhere else in Europe, it is done typically in an anecdotal manner without any attempt at systematic comparison. To put it simply: European criminology is rarely comparative and hardly European. 206 Martin Killias 3. The Council of Europe Sourcebook Project 3.1 Organization and Coverage As a modest attempt to change things in this domain, the Council of Europe’s Comite Directeur des problemes Criminels (CDPC) approved, during its plenary session of June 1993, a proposal to charge a small experts’ group to work out a so-called feasibility study on a data handbook This experts’ committee started its work in September 1993 and held five meetings; it presented its final report to the CDPC plenary session in June 1995. Members of this committee are G. Barclay and C. G. Lewis (England and Wales), P. Tournier (France), J.-M. Jehle (Germany), I. Kertesz (Hungary), H. von Hofer (Sweden), M. Kommer (Netherlands), M. Killias (Switzerland, Chairman), with Wolfgang Rau acting as the secretary. During the first meeting HEUNI was represented as an observer. The members of the group agreed that the best way to carry out a feasibility study was to draft a pilot data handbook covering the seven countries represented on the group. For these countries members felt they could, with a reasonable investment in time and resources, gather the necessary data and indicate all the definitional problems on law and routines in the handling of statistical data. In order to extend the coverage an effort was made to gather data from 5 more countries on which some members felt confident they could locate data with sufficient background information on definitions and statistical routines, often with helpful colleagues in the respective countrics. These additional countries are Belgium, Ireland, Italy, Northern Ireland and Scotland. The experts agreed to use data for 1990. The European Criminal Justice Sourcebook Project 207 3.2 Areas covered by the project The sourcebook project covers the usual areas of data collections of this sort: a) Police data on known offences The group established a list of offences to be included. This list includes: Homicide (Total and substantive) Assault Rape Robbery (Total and with a weapon) Theft of which Motor vehicle thefts Bicycle thefts Burglary Burglary from a dwelling Drug offences (Total and trafficking) In order to make the data as comparable as possible, the group has opted for a standard definition of all these offences. In addition, tables on legal definitions give in summary form the essential information on deviations from the standard definitions for each offence, so that the reader may be able to relate differences in the figures to definitional variations. In connection with homicide, for example, this table indicates whether, in some countries, the concept of homicide (as used in national crime statistics) includes or not assault with fatal consequences, abortion, infanticide, euthanasia and help with suicide. In addition to these definitional tables, the document contains tables on statistical routines in the process of recording the several offences. These tables inform, among other things, on the existence or not in the respective 208 Martin Killias countries of written guidelines how offences are to be recorded and classified, on the time at which recording occurs (i.e. at the time police learn about it, or when they have concluded their investigation), on the unit of counting (i.e. the offence, the victim), on whether or not secondary (usually minor) offences are recorded, and how serial offences (of the same type) are counted. The group hopes that this way of treating the information will allow potentially important sources of variation to be identified, without bothering the user with endless comments on national legal definitions or recording practices. b) Police data on known suspects This chapter of tables aims to give, for the same types of offences, the available information on suspects known to the police. The tables indicate the proportion of females, minors, and foreign citizens among the suspects. Again, tables on the definitions (e.g. the age of criminal responsibility) and statistical routines follow the quantitative tables. c) Other police data Additional tables indicate the number of police officers and the expenditures related to police work. Tables on the definition of police officers and categories of expenditures are included in this section. d) Data on prosecution, courts and sentencing The Group has established tables on prosecution, courts and sentences. The table on prosecution includes the total number of cases treated by prosecutors, the proportion of cases dismissed, the percentage of cases where the prosecutor himself sentenced the (consenting) defendant (a 209 The European Criminal Justice Sourcebook Project common practice in some countries of Europe) and the proportion of cases brought before a court. In the section on persons convicted, information is given on convictions for all offences included in previous sections, with indication of the percentage under age 18, females, foreign nationals and the rate per 100,000 population. For all the offences included, detailed information is given, in additional tables, on the sentences imposed (total, fines, non-custodial sentences, suspended custodial sentences, unsuspended custodial sentences.) For the unsuspended custodial sentences and every offence included, details are given on the length and time served in prison (total, 0-6 months, 6-12 months, 1-2 years, 2-5 years, 5 years and over, life sentences, other determinate sentences, average sentence length, time served in prison). Finally, this section contains rates on convictions and unsuspended custodial sentences (for any of the offence included and for the total) per 100,000 population, per 100 recorded offence and per 100 suspects. These tables illustrate the relative ’certainty’ of convictions and imprisonments for a given offence in different countries. In order to complement the picture of sentencing styles in the several countries, a table includes the percentage of unsuspended custodial sentences shorter than 1 year. ¢) Corrections Taking advantage of previous work done under the umbrella of the Council of Europe, by Pierre Tournier and his colleagues on corrections, the pilot document includes data on the number and the capacity of prison facilities, and of prisoners on remand or serving sentences. Additional tables include the number of persons serving community sanctions and expenditures related to corrections including community sanctions. We also add a table summarizing recidivism rates from studies carried out in different countries. It is uncertain whether data on the life prevalence of convictions and imprisonment will be available for a sufficient number of countries. 210 Martin Killias f) Survey data We planned at first to include survey data from national and regional studies. But the methodology of the existing studies varied too much between countries. Thus we have decided to stick with of existing internationally designed studies. Thus, the main findings from the two waves of the International Crime Survey on victimization, crime reporting, crime prevention measures and attitudes are included. It has also been desirable to include some data from the International Survey on Crime against Retailers. This has been conducted in 1994 in the Netherlands, England and Wales, France, Germany, Italy, France, and Switzerland. It covers topics such as theflt against shops, restaurants, and businesses: it also includes, for some countries, extortion of protection money and corruption. Finally, we include some key data from the International Self-Reported Juvenile Delinquency Project. This survey has been conducted on a national basis in England, Netherlands, Portugal, Spain and Switzerland, and in a few cities in other countries. The inclusion of data from these surveys will promote their dissemination among users of the Criminal Justice Sourcebook. By finding survey data and statistical materials in the same volume as crime statistics, policy makers might become more aware of the usefulness of international survey research in this area. 3.3 Future Perspectives The experts’ committee has presented its final report, a so-called Model Sourcebook of Criminal Justice, to the CDPC plenary session in June 1995. It was decided that this Model Sourcebook should be published and that the work of the experts’ committee should continue to June 1996 in order to prepare the steps for establishing a European Sourcebook of Criminal Justice. The European Criminal Justice Sourcebook Project 211 The experts’ group has gained so far a few impressions which might be useful for the next decisions which have to be made. First, it seems to us that a data handbook is feasible, i.e. it is possible to gather statistical data from different countries and to present them in synthetic tables. It is also possible to collect the necessary background information on statistical procedures and legal definitions and to present these in an easily readable form. Second, the pilot study’s relative success has been possible thanks to a close cooperation between eight specialists from the countries included. In other words, an approach by questionnaire, no matter how well designed, might never have produced the same result. What is needed is personal involvement of competent specialists, individualized (and publicly acknowledged) responsibility as well as motivation and commitment to the common product. Should the Council of Europe ever consider to include a data handbook of this sort among its standing activities, it might, therefore, be crucial to have at least one responsible contributor per country, and to create among them a firm commitment to the common goal by appropriate measures (for example, a common meeting at Strasbourg to get the message through and to receive all sorts of helpful comments and reactions from their side). A steering committee might eventually be needed to manage and coordinate the initiative, with a role comparable to the one Pierce Tournier plays in the Council of Europe’s correctional data collection. Which countries should be included? Basically, the understanding is that all the Member States of the Council of Europe should be covered in a future Sourcebook. On the other hand, this might be a too ambitious standard to start with. Our feeling, therefore, is that the project might start on a somewhat narrower basis and then continually add new countries in future editions. 212 Martin Killias Which Subject areas should be covered? The same sort of solution might be attempted. If the experts’ group has decided on including only a few offences at the start, this list could be added to in the future. Which years should be included? For the time being, the experts’ group has chosen the year 1990, for obvious practical reasons. Although we made some additional effort to gather some data on trends, the bulk of the materials to be presented in the pilot project is cross-sectional. In the future, it will be much easier to cover trends. This will be the more appealing as many limitations to cross-sections comparisons do not necessarily affect data on trends. How often should a handbook of this sort be produced? The answer will clearly depend on resources available. Given the rather pessimistic outlooks for the next few years, a five year interval might be a realistic idea. How should a European data Handbook be coordinated with the UN Surveys on Crime Data and the Operations of the Criminal Justice System? Our fecling is that the two do complement and not exclude each other. The UN Surveys cover an admirably large array of countries, whereas the Council of Europe’s data collection will necessarily be limited in geographical respects. Of course, HEUNI might no longer need to proceed by the questionnaire method to gather data from the nations covered by the European Sourcebook, but could directly take them over from Strasbourg. This would allow the UN to invest more energy in the collection of data from other countries, and especially in their assessment of their reliability and validity. Contributors Barclay, Gordon Head of Dissemination and International Section, S1 Division, Home Office, Lunar House, 40 Wellesley Road, GB-Croydon, Surrey Cro 9YD Blath, Dr. Richard Regierungsdirektor, Bundesministerium der Justiz, Referat fiir Kriminologie, D-53170 Bonn Caputo, Armando ISTAT, Statistische Giudiziarie, Viale Liegi, 13, I-00100 Roma Erdosi, Sandor Data Processing Office, Statistics and Information Department, Ministry of Home Affairs, Budapest, Hungary Gratia, Jean Charge du service Centrale d’Etude de la Delinquence a la Direction Centrale de la Police Judiciare, Ministere de I'Interieur, 13 rue des Saussaires, F-75008 Paris Israclsson, Torbjorn Statistics Sweden, Karlavagen 100, S - 115 81 Stockholm, Sweden Jasinski, Prof. Jerzy, Institute of Legal Studies, Polish LL.D. Academy of Sciences, Nowy Swiat 72, 00-330 Warszawa, Polen Jehle, Dr. Jorg-Martin Direktor der Kriminologischen Zentral- stelle e.V., Adolfsallee 32, D-65185 Wiesbaden Killias, Prof. Dr. Martin Institut de Police Scientifique et de Criminologie, Université de Lausanne, CH-1015 Lausanne 214 List of Contributors Kommer, Dr. Max M Research Coordinator, Research and Documentation Centre, Ministerie van Justitie, Schedeldoekshaven 100, NL-2511 EX Den Haag Kranidioti, Dr. Maria Center of Penal and Criminological Research, University of Athens, Faculty of Law, Section of Criminal Law and Criminology, 57, SOLONOS Str. (Rm 307), 106 79 Athens, Greece Larsen, Paul Henning Danmarks Statistik, Sejrogade 11, 2100 Copnhavn 0, Danmark Lewis, Chris Chief Statistician, Home Office, Research and Statistics Department, 50 Queen Anne’s Gate, GB-London SW 1 H 9 AT Lopes, J. de Seabra Director-General, Gabinete de Estudos e Planeamento do Ministerio da Justica, Av Oscar Monteiro Torres, 39, 1016 Lisboa Codex, Portugal Silva, Sange de Executive Director, Canadian Centre for Justice Statistics, 19th Floor, R H Coates Building, Ottawa, Canada K1A OT6 Sita, N. Masamba UN African Institute for the Prevention of Crime and the Treatment of Offenders, PO Box 10590, Kampala, Uganda Director of the Center of Penal and Spinellis, Prof. Dr. Calliope D. Criminological Research, University of Athens, Faculty of Law, Section of Criminal Law and Criminology, 57, SOLONOS Str. (Rm 307), 106 79 Athens, Greece Publications of the KrimZ Kriminologie und Praxis (KUP) Series of the Kriminologische Zentralstelle Band 1 Anwendungsbezogene Kriminologie zwischen Grundlagenforschung und Praxis / hrsg. von Jorg-Martin Jehle u. Rudolf Egg. - Wiesba- den, 1986. - 158 S. ISBN 3-926371-00-5 Band 2 Datenzugang und Datenschutz in der kriminologischen Forschung / hrsg. von Jorg-Martin Jehle. - Wiesbaden, 1987. - 426 S. ISBN 3-926371-01-3 Band 3 Drogentherapie und Strafe / hrsg. von Rudolf Egg. - Wiesbaden, 1988. - 323 S. ISBN 3-926371-03-X Band 4 Datensammlungen und Akten in der Strafrechtspflege / hrsg. von Jorg-Martin Jehle. - Wiesbaden, 1989. - 346 S. ISBN 3-926371-06-4 Band 5 Criminological Research and Planning in State and Supranational Institutions /ed. by Jorg-Martin Jehle. - Wiesbaden, 1990. - 223 S. ISBN 3-926371-07-2 Band 6 Feuerhelm, Wolfgang: Gemeinniitzige Arbeit als Alternative in der Geldstrafenvollstreckung. - Wiesbaden, 1991. - 403 S. ISBN 3-926371-09-9 Band 7 Individualpréaventionund Strafzumessung :ein Gesprachzwischen Strafjustiz und Kriminologie / hrsg. von Jorg-Martin Jehle. - Wies- baden, 1992. - 388 S. ISBN 3-926371-11-0 Band 8 Die Zukunft der Personenstatistiken im Bereich der Strafrechts- pflege : Materialien und Diskussion einer Expertensitzung / hrsg. vom Bundesministerium der Justiz und der Kriminologischen Zen- tralstelle. - Wiesbaden, 1992. - 162 S. ISBN 3-926371-12-9 Band 9: Die Therapieregelungen des Betdubungsmittelrechts : deutsche und auslandische Erfahrungen / hrsg. von Rudolf Egg. - Wiesba- den, 1992. - 262 S. ISBN 3-926371-14-5 Band 10 Kriminologie als Lehrgebiet : Kriminologische Aus-, Fort- und Weiterbildung in verschiedenen Studienfachern und Berufsfeldern / hrsg. von Jorg-Martin Jehle. - Wiesbaden, 1992. - 365 S. ISBN 3-926371-15-3 Band 11 Block, Petra: Rechtliche Strukturen der Sozialen Dienste in der Justiz - eine vergleichende Analyse der Léndervorschriften. - Wiesbaden, 1993. - 282 S. ISBN 3-926371-16-1 Band 12 Kurze, Martin: Strafrechtspraxis und Drogentherapie. Eine Implementationsstudie zu den Therapieregelungen des Betau- bungsmittelrechts. - Wiesbaden, 1994. - 2., erg. Aulfl. - 320 S. ISBN 3-926371-23-4 Band 13 Die strafrechtliche Unterbringung in einem psychiatrischen Krankenhaus : Probleme und Perspektiven / hrsg. von Michael Gebauer u. Jorg-Martin Jehle. - Wiesbaden, 1993. - 232 S. ISBN 3-926371-20-X Band 14 Organisation und Kooperation der Sozialen Dienste in der Justiz / hrsg. von Jorg-Martin Jehle und Werner Sohn. - Wiesbaden, 1994. - 381 S. ISBN 3-926371-21-8 Band 15 Improving Criminal Justice Statistics : National and International Perspectives /ed. byJorg-MartinJehle and Chris Lewis. - Wiesba- den, 1995. - 214 S. ISBN 3-926371-24-2 Band 16 Die strafrechtliche Unterbringung in einer Entzichungsanstalt / hrsg. von Axel Dessecker und Rudolf Egg. - Wiesbaden, 1995. - ca. 250 S. ISBN 3-926371-25-0 Berichte - Materialien - Arbeitspapiere (BMA) Heft 1: Der Kriminologische Dienst in der Bundesrepublik Deutschland / hrsg. von Jorg-Martin Jehle. - Wiesbaden, 1988. - 117 S. ISBN 3-926371-02-1 Heft 2 Feuerhelm, Wolfgang: Geldstrafenvollstreckung und Gemeinniitzi- ge Arbeit : e. Zwischenbericht. - Wiesbaden, 1988. - 106 S. ISBN 3-926371-04-8 Heft 3 Egg, Rudolf; Kurze, Martin: Drogentherapie in staatlich anerkann- ten Einrichtungen : Ergebnisse e. Umfrage. - 3. Aufl. - Wiesbaden, 1991 (1. Aufl. 1989). - 134 S. ISBN 3-926371-05-6 Helt 4 Jehle, Jorg-Martin; Feuerhelm, Wolfgang; Block, Petra: Gemein- niitzige Arbeit statt Ersatzfreiheitsstrafe : Forschungskolloquium zu einer bundesweiten Untersuchung; mit einem Anhang zur Situation in der DDR. - Wiesbaden, 1990. - 180 S. ISBN 3-926371-08-0 Heft 5 Sohn, Werner: Referatedienst Kriminologie : ausgewihlte Zeit- schriftennachweise aus der Literaturdatenbank JURIS fiir das Jahr 1990 / zsgest. und bearb. von Werner Sohn. Wiesbaden, 1991. - 259 S. ISBN 3-926371-10-2 Heft 6 Sohn, Werner: Referatedienst Kriminologie : ausgewihlte Zeit- schriftennachweise aus der Literaturdatenbank JURIS fiir das Jahr 1991 / zsgest. und bearb. von Werner Sohn. Wiesbaden, 1992. - 330 S. ISBN 3-926371-13-7 Heft 7 Sozialtherapie in den 90er Jahren gegenwirtiger Stand und aktuelle Entwicklung im Justizvollzug / hrsg. von Rudolf Egg. - Wiesbaden, 1993. - 205 S. ISBN 3-926371-17-X Heft 8: Sohn, Werner: Referatedienst Kriminologie : ausgewiéhlte Zeit- schriftennachweise aus der Literaturdatenbank JURIS fiir das Jahr 1992 / zsgest. und bearb. von Werner Sohn. Wiesbaden, 1993. - 323 S. ISBN 3-926371-18-8 Heft 9: Sohn, Werner: Referatedienst Kriminologie : ausgewihlte Zeit- schriftennachweise aus der Literaturdatenbank JURIS / zsgest. und bearb. von Werner Sohn. - Folge 4. Ausg. 1994. - Wiesba- den, 1994. - 336 S. ISBN 3-926371-22-6 Other Monographs Kriminologie: Forschungsdokumentation ; 1980 - 1986 / Hrsg. Informa- tionszentrum Sozialwissenschaften und Kriminologische Zentralstelle. - Bonn; Wiesbaden, 1988. - 689 S. + Reg.anh. ISBN 3-8206-0039-6 Further Publications Furthermore numerous articles in journals and readers on various crimi- nological fields have been published.