ar V 
 
 13429 
 
 r-**^5 
 
(^mmll Wimmxtf ^xhxm^ 
 
 ^.3iWl1i«l5 
 
 2236 
 
CORNELL UNIVERSITY LIBRARY 
 
 56 796 
 
M^ Cornell University 
 WB Library 
 
 The original of tliis bool< is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924101156796 
 
HISTORY AND ORGANIZATION 
 
 OF CRIMINAL STATISTICS IN 
 
 THE UNITED STATES 
 
 A THESIS 
 
 PRESiaWED TO THE FACXTLTT OF THE GKADUATE 
 
 SCHOOL OF COBNELIi UNIVBHSITT 
 
 FOB THE DEGKEE OF 
 
 DOCTOR OF PHILOSOPHY 
 
 BT 
 
 LOUIS NEWTON ROBINSON 
 
COPYRIGHT, 19X1, BY HART, SCHAFFNER Sc MARX 
 ALL RIGHTS RESERVED 
 
TO 
 ANDREW D. WHITE 
 
NOTE 
 
 i HOUGH a flood of articles on special phases 
 of the general problem of criminology have 
 recently appeared in the United States in re- 
 sponse to the demand on the part of the public 
 for greater knowledge concerning this subject, 
 the statistical side of the question has been, 
 for one reason or another, very inadequately 
 treated. In the hope of partly filling this gap 
 in our knowledge, I began not long ago a 
 statistical study of the increase in criminality 
 in the United States. As the work progressed, 
 I soon found that I could never hope to use 
 the statistics intelligently without a thorough 
 understanding of their history and organiza- 
 tion. Since no complete study along this line 
 had ever been made, I set to work on this 
 new problem, relinquishing for the time being 
 my original plan. The results of my labor are 
 embodied in this book, which I hope will prove 
 useful to all those who may feel inclined to 
 subject to a statistical test some of the theo- 
 ries of the criminologists. 
 
 I wish to make acknowledgment here of 
 certain changes suggested by Professor Walter 
 
F. Willcox of Cornell which led to a recast- 
 ing of the first chapter. I owe him, more- 
 over, a debt of gratitude for teaching me to 
 appreciate statistics and to understand some 
 of their mysteries. My sincere thanks are 
 also due to those other kindly and inspir- 
 ing teachers under whom I prepared for the 
 doctorate. 
 
PREFACE 
 
 This series of books owes its existence to the 
 generosity of Messrs. Hart, Schaffner, and 
 Marx of Chicago, who have shown a special 
 interest in directing the attention of American 
 youth to the study of economic and commer- 
 cial subjects, and in encouraging the system- 
 atic investigation of the problems which 
 vitally affect the business worid of to-day. 
 For this purpose they have delegated to the 
 imdersigned Committee the task of selecting 
 topics, making all annoimcements, and award- 
 ing prizes annually for those who wish to 
 compete. 
 
 In the year ending June 1, 1910, the fol- 
 lowing topics were assigned : 
 
 1. The effect of labor unions on international 
 
 trade. 
 
 2. The best means of raising the wages of the 
 
 unskilled. 
 
 3. A comparison between the theory and the 
 
 actual practice of protectionism in the 
 United States. 
 
viii PREFACE 
 
 4. A scheme for an ideal monetary system for 
 
 the United States. 
 
 5. The true relation of the central govern- 
 
 ment to trusts. 
 
 6. How much of J. S. Mill's economic system 
 
 survives ? 
 
 7. A central bank as a factor in a financial 
 
 crisis. 
 Any other topic which has received the 
 
 approval of the Committee. 
 ;The present volume was awarded Hon- 
 . arable Mention. 
 
 Pbofessor J. Latteence Laughlin, 
 
 University of Chicago, Chairman. 
 Professor J. B. Clark, 
 
 Columbia University. 
 Professor Henry C. Adams, 
 
 University of Michigan. 
 Horace White, Esq., 
 
 New York City. 
 Professor Edwin F. Gat, 
 
 Harvard University. 
 
CONTENTS 
 
 I. Inthoduction 1 
 
 n. FeDEHAL CEIMINAIi STATISTICS — ObIGIN AND 
 
 Geowth 12 
 
 in. PEmEEUL CRiMiKAii STATISTICS (continued) — 
 
 ESSENTIAI. NaTOBE AND MeANINQ .... 27 
 
 IV. State Cbiminal Statistics — Judicial ... 38 
 
 V. State CniMiNAii Statistk!S (continued) — 
 
 Pkison y. P9 
 
 VI. Plan fob the Repwjanization op Cbiminal 
 
 Statistics in the United States .... 91 
 
 Appendix 101 
 
HISTORY AND ORGANIZATION 
 
 OF CRIMINAL STATISTICS 
 
 IN THE UNITED STATES 
 
 CHAPTER I 
 
 INTRODUCTION 
 
 Ihe subject of this work is the history and 
 organization of criminal statistics in the 
 United States. The aim is not merely to pre- 
 sent the main lines of development, but to 
 interpret and criticise the facts. It is hoped 
 that such a study will contribute something 
 to the further development of criminal sta- 
 tistics in the United States. 
 
 In this country the statistical side of crim- 
 inology is very imperfectly developed, and 
 while the same cannot be said with equal 
 force of other English-speaking countries, it 
 yet remains true that the statistical termin- 
 ology of this social science is characterized, so 
 far as the English language is concerned, by 
 great vagueness and uncertainty. Whenever 
 such a state of affairs exists, a primary requi- 
 site of sound discussion is to define carefully 
 the sense in which the terms are to be under- 
 stood. So at the start I must define and dis- 
 
2 CRIMINAL STATISTICS 
 
 tinguish crime, criminal, judicial statistics, 
 prison statistics, statistics of crimes, statistics 
 of criminals, and criminal statistics. 
 
 The definitions of a crime given in the text- 
 books on criminal law dififer in some particu- 
 lars, but the thought in them all seems to be 
 that a crime is an act or an omission which 
 permits of the following characterization:^ 
 it harms the general public (it may at the 
 same time harm an individual) and the state 
 as the representative of the general public, 
 taking cognizance of it, would, if the suspected 
 perpetrator could be produced, appear at the 
 bar to demand justice in the form of a pen- 
 alty, and would also have the sole power to 
 stay proceedings at any point.'^ 
 
 1 A good discussion of what constitutes a crime is given by 
 Courtney Stanhope Kenny, Outlines of Criminal Law, 1902, 
 pp. 1-25. For definitions of crime, consult: Clark and Mar- 
 shall's Treatise on the Law of Crimes, 2d edition, 1905, ch. 
 i, sec. 1, p. 1; Bishop's New Criminal Law, vol. i, ch. iii, 
 sec. 32. p. 14; McClain's Treatise on the Criminal Law, 
 1897, vol. i, part i, ch. i, sec. 4, pp. 15-16; May's Law of 
 Crimes, 3d edition, 1905, sec. 1. p. 1. 
 
 2 The word crime is also frequently used by the public in a 
 non-legal sense to designate the number of crimes committed 
 within certain geographical limits, or perhaps the number of 
 criminals suspected or detected, or even the general state of 
 public morals. When used thus, the word seldom conveys 
 an exact meaning and generally needs further elucidation and 
 explanation. It seems to me that the word crime should be 
 used only to characterize the act, and that the word criminality 
 ought to be employed in all instances where the state of being 
 criminal is meant. The common phrase "increase of crime" 
 
INTRODUCTION 3 
 
 In a broad sense a criminal is one who has 
 committed such an act or is guilty of such an 
 omission. But since many crimes are never 
 known nor the perpetrators of known crimes 
 so deeply branded by the mark of Cain that 
 all may recognize them, the statistician counts 
 as criminals only those whom the courts of law 
 have so adjudged. From the statistical point 
 of view, then, a man simply held for trial is 
 not a criminal, and cannot be so listed imtil 
 the law has spoken its final word. 
 
 Judicial Statistics are statistics wherein the 
 imit is some fact regarding the judicial sys- 
 tem or the administration of justice. It may 
 be a court, a judge, a case, a crime, a criminal, 
 an arrest or any other of the numerous facts 
 pertaining to the judicial system or the ad- 
 ministration of justice which can be measured 
 quantitatively. 
 
 Prison Statistics diflfer from Judicial Statis- 
 tics in that the unit is some fact regarding the 
 penal system or the administration of this 
 system. The unit may be a keeper, a jail, a 
 penitentiary, a cell, a crime, a prisoner, or, as 
 in the case of judicial statistics, any fact of 
 the penal system or its administration which 
 can be measured quantitatively. 
 
 should be replaced by the phrase "increase of criminality." 
 There is a clear distinction in the meaning of these two words, 
 and vagueness and confusion can be best avoided by using 
 them with their rightful significance. 
 
4 CRIMINAL STATISTICS 
 
 Statistics of Crimes are statistics wherein 
 the unit is a crime, no reference being made 
 to the agent. If these statistics are found 
 among Judicial Statistics or compiled from 
 the court records, they are Judicial Statistics 
 of Crimes; if they are found among Prison 
 Statistics or compiled from the prison records 
 they are Prison Statistics of Crimes. 
 
 Statistics of Criminals are statistics wherein 
 the unit is a criminal, no reference being made 
 to his crime or crimes. As in the case of Sta- 
 tistics of Crimes and depending on the same 
 reason, they also may be either Judicial Sta- 
 tistics of Criminals or Prison Statistics of 
 Criminals. 
 
 Criminal Statistics are statistics wherein 
 two units are employed — crime and criminal 
 — and wherein there is a correlation between 
 the two units. They, like the Statistics of 
 Crimes and the Statistics of Criminals, may 
 be either Judicial or Prison,^ 
 
 As has been explained in the preceding 
 paragraphs, there are two great groups of 
 statistics which furnish important data re- 
 
 * The best definition of Criminal Statistics that I have seen 
 is that given by H. von Scheel in his article, "Kriminal Sta- 
 tistik," in Conrad's Handwirrterbuch der Staatsivissenschaften, 
 Dritte Auflage, Band vi, pp. 246-247. The Rev. Douglass 
 Morrison in the Journal of the Royal Statistical Society for 
 March, 1907, pp. 1-24, gives many good hints on a definition 
 in his article, "The Interpretation of Criminal Statistics." 
 
INTRODUCTION 5 
 
 garding criminality. These are Judicial Sta- 
 tistics and Prison Statistics. In each group 
 are three main sub-groups. Statistics of 
 Crimes, Statistics of Criminals, and Criminal 
 Statistics. It is now necessary to analyze 
 these terms more fully in order that the na- 
 ture and meaning of these various kinds of 
 statistics may be clearly apparent. 
 
 Statistics of Crimes possess a marked defect 
 — they fail to indicate the "penchant au 
 crime," as Quetelet calls it, in the community. 
 One man might have committed several 
 crimes, or several men might have combined 
 to commit one crime; for this reason the use- 
 fulness of such statistics is limited, and since 
 through those who commit crimes must be 
 traced and sought the causes which lead to 
 crimes and the measures needed for their re- 
 pression, these statistics will never satisfy the 
 student of social conditions. 
 
 Statistics of Criminals serve to give some 
 idea of the criminal instinct and are useful for 
 administrative purposes. On the other hand, 
 the need of knowing the number and the na- 
 ture of the crimes of which they have been 
 adjudged guilty is very great, for without this 
 knowledge no idea of the true character of the 
 criminals can be had. 
 
 Criminal Statistics, which combine the 
 essential features of Statistics of Crimes and 
 
6 CRIMINAL STATISTICS 
 
 Statistics of Criminals, that is, statistics 
 wherein the two units — crime and criminal 
 — are correlated, are the only statistics which 
 are at all adequate as a guide in the study of 
 criminality. One must know not merely the 
 crime or the criminal, but both. A chapter 
 could be written on the subject alone of cor- 
 relation of the two units employed in criminal 
 statistics, but as it is not my purpose to treat 
 of the theory of criminal statistics any more 
 than is necessary to an understanding of the 
 work in the United States, this question will 
 be passed over with the brief remark that all 
 criminals should at least be listed with their 
 principal crime, if they are guilty of more 
 than one, and that the surplus of crimes 
 should in addition be recorded.^ 
 
 It will be remembered that these three 
 kinds of statistics may be either Judicial or 
 Prison, depending on whether they are found 
 among Judicial Statistics or Prison Statistics 
 — the two great groups of statistics which 
 include these three sub-groups as well as 
 others not related except very indirectly to 
 
 ' For a good exposition of this feature, see "Influence du 
 choix de I'unite sur les resultats de la statistique criminelle," 
 by M. Ch. de Lannoy in the Bulletin de I'lnstitut Interna- 
 tional de Statistique (Budapest, 1903), tome xiii, pp. 325-333; 
 also "Rapport sur la Statistique de la Criminalite," by E. 
 Cheysson (de I'lnstitut), in the Remie PSnitentiaire, vol. 
 xxvii (1903), pp. 1116-1125. 
 
INTRODUCTION 7 
 
 the study of criminality. I shall now attempt 
 to point out that the real difference between 
 a sub-group under Judicial Statistics and a 
 corresponding sub-group under Prison Sta- 
 tistics is one of inclusiveness. 
 
 It is perfectly clear, I think, that no state, 
 however well equipped it may be for gather- 
 ing such knowledge, can ever know or even 
 suspect all the crimes that are committed 
 against her ward — the general public. Some 
 crimes are known only to the men who com- 
 mit them; some are known, it may be, to 
 other men who for one reason or another do 
 not care to make their knowledge public. The 
 crimes that are most apt to be called to the 
 attention of the state are those that not only 
 injure the general public, but at the same time 
 also injure some individual. 
 
 As it is with crimes, so it is with criminals 
 — the state does not pretend to know of all. 
 Certainly a state can have no knowledge of 
 those criminals whose crimes it does not even 
 know, and often it cannot identify in the 
 least those who have committed the crimes of 
 which it is aware. Hence none of the statistics 
 can lay claim to include all the facts. Of the 
 two, the sub-groups of Judicial Statistics are 
 far more complete than are those of Prison 
 Statistics. The explanation is simple. The 
 prison records relate only to individuals who 
 
8 CRIMINAL STATISTICS 
 
 have been placed in confinement within the 
 institutions. Consequently they take no ac- 
 count of individuals who have paid fines as 
 punishment, and of course the crimes of which 
 no individuals are proven guilty find no place 
 in such records. All statistics compiled from 
 these records, whether they be Statistics of 
 Crimes, Statistics of Criminals, or Criminal 
 Statistics, will thus of necessity fail to include 
 data obtainable from the records of the courts. 
 
 So far as crimes and criminals are con- 
 cerned, this matter of inclusiveness is, in my 
 opinion, the sole distinction between the two 
 main groups of Judicial and Prison Statistics. 
 
 Up to this point I have assumed that the 
 crimes to be enumerated are acts or omissions 
 stamped as crimes by final official action, that 
 is, by the trial of the accused, and that the 
 criminals likewise are only those deemed so by 
 the final verdict of the courts. It remains, 
 therefore, to speak of certain other statistics 
 concerning which there is little obscurity as 
 to what they actually are. Statistics of Arrest 
 are, of course, a sub-group of Judicial Sta- 
 tistics, since the unit or units is some fact or 
 facts regarding the administration of justice. 
 These statistics may include a statement of 
 the number of individuals arrested or this 
 fact plus the crime or crimes for which ar- 
 rested. It is to be noted here that one is deal- 
 
INTRODUCTION 9 
 
 ing not with criminals but merely with sus- 
 pected criminals, and that the crime is not 
 well defined. Statistics of Commitments, if 
 Judicial, refer to those who have been com- 
 mitted to bail or to jail; if Prison, they refer 
 only to the latter. Statistics of Indictments 
 are necessarily Judicial Statistics and show 
 the results of another step in the judicial 
 procedure. Both the individual and the crime 
 or crimes have by this time been somewhat 
 more clearly indicated, but the indefiniteness 
 which stUl clings to both renders such statis- 
 tics of doubtful value. ^ 
 
 It is needless to say that these statistics all 
 relate to a period of time and not to a point of 
 time. The confusion which has existed on this 
 point in regard to the criminal statistics col- 
 lected by the federal government will be 
 cleared up in a later chapter. 
 
 The influence of the state upon the nature 
 of the criminal statistics is also a factor al- 
 ways to be considered. From what has already 
 been said, it is evident that no matter at what 
 
 * For a somewhat similar discussion in German, see "Zur 
 EinfUhrung in die Kriminalstatistik, insbesondere diejenige 
 des deutschen Reichs," by H. von Scheel in AUgemeinea 
 Statistisches Archiv for 1890, pp. 184-211. 
 
 An interesting study of the probability of conviction and 
 its relation to criminal statistics is the article of Bela Foldes 
 in the Bulletin de VInstitvt International de Statistique (Co- 
 penhagen, 1908), tome xvii, pp. 378-394, entitled "Les bases 
 mathematique et statistique de la condamnabUite." 
 
10 CRIMINAL STATISTICS 
 
 time the statistics are gathered the size and 
 number of the fish caught by the net thus 
 spread will vary greatly with the kind of laws 
 prevailing in the country and with their en- 
 forcement at the hands of the officers. An 
 interesting example of this is furnished by the 
 records of the courts and prisons of Massachu- 
 setts. During the last thirty years, several 
 changes have been made in the liquor laws; 
 and the enforcement of the laws, from the 
 evidence at hand, seems not to have been 
 uniformly strict. Now as a result of these two 
 circumstances, the records of the courts and 
 prisons referring to offenders against the 
 liquor laws are such that it would appear to 
 the casual observer that during certain years 
 drunkenness was a disorder common to all 
 and that during other years strong drink pro- 
 duced but little effect on the people of this 
 state. Comparisons, then, of the criminal 
 statistics of two different periods in the his- 
 tory of a state are a matter requiring the ut- 
 most care and insight. Likewise, comparisons 
 of the criminal statistics of states having dif- 
 ferent laws or enforcing them differently are 
 extremely difficult, if not impossible. 
 
 The form of the state complicates the ques- 
 tion still further. If the state is unitary it is 
 apt to have uniform codes of laws applying 
 equally to all parts of the country, as in France 
 
INTRODUCTION 11 
 
 and in England. If, however, the state is 
 federal, it may have uniform codes or again 
 it may not, depending somewhat upon the 
 extent to which centralization has developed. 
 Germany, for example, has uniform civil and 
 criminal codes, but the states of the United 
 States have each its own set of laws. The 
 compilation, therefore, of criminal statistics 
 for the United States as a whole is a more dif- 
 ficult task than it is for France or England or 
 Germany. 
 
 Having defined the important terms and 
 having pointed out the elementary principles 
 relating to their collection, it remains to speak 
 of the plan of this book. Fortunately, the na- 
 ture of the subject makes a simple plan possi- 
 ble. I shall take up first the criminal statistics 
 collected by the federal government, then 
 those collected by the states, and lastly I shall 
 attempt to outline a scheme for the organiz- 
 ation of criminal statistics in the United States 
 which my study has led me to believe would 
 be not only better than the existing organiz- 
 ation, but also feasible and more in line with 
 the present trend of statistical organization in 
 the United States. 
 
CHAPTER II 
 
 FEDERAL CRIMINAL STATISTICS — ORIGIN AND 
 GROWTH 
 
 JD EDERAL Criminal statistics begin with the 
 census of 1880. But statistics, called in the 
 census volumes "statistics of crime," were 
 collected at the censuses of 1850, 1860, and 
 1870, which, though they were statistics of 
 criminals only and not true criminal statistics, ^ 
 were yet, as it proved, the embryonic form of 
 the statistics of 1880. Thus with the statis- 
 tics of criminals collected at the census of 1850 
 the history of federal criminal statistics may 
 properly begin. 
 
 One of the sections^ of the act of May 23, 
 1850, was that, in the event of no further legis- 
 lation, the provisions there laid down were to 
 govern the taking of any subsequent cen- 
 suses. This law sufficed for the censuses of 
 1850, 1860, and 1870. Consequently, the 
 organization of these three censuses was the 
 
 1 I am judging from the actual published reports. The in- 
 structions in 1860 to the assistant marshals were such as to 
 make possible criminal statistics, but the facts if gathered 
 were not utilized. 
 
 ' Section 23, "General Act providing for the Census of 
 1850 and for every subsequent Census." Given in Statistics 
 o/ United States, 1850, pp. xix-xxi. 
 
FEDERAL CRIMINAL STATISTICS 13 
 
 same.^ The task of carrying out the provisions 
 of the act was given to the Secretary of the 
 Interior, who was authorized to appoint a 
 superintending clerk, known as the Superin- 
 tendent of the Census. The actual collection 
 of the statistics was performed by the United 
 States marshals. They divided their districts 
 and appointed assistant marshals to do the 
 collecting in each division thus created. These 
 made their reports to the marshal, who in 
 turn reported to the Secretary of the Interior. 
 The Superintendent of the Census, with the 
 aid of a clerical force at his disposal, compiled, 
 analyzed, and published the statistics and 
 when all was finished, or as much as Congress 
 saw fit to have finished, resigned his ofiice, and 
 the census oflBce was then disbanded. Such 
 in brief was the organization under which 
 statistics of criminals, along with other sta- 
 tistics, were gathered and presented to the 
 public under the censuses of 1850, 1860, and 
 1870. 
 
 The inclusion of inquiries regarding crimin- 
 als in the schedules of the seventh census was 
 due, it seems, to Mr. Shattuck, one of the ex- 
 perts whom the census board in preparing its 
 
 1 See Carroll D. Wright and Waiiam C. Hunt, History 
 and Growth of the United States Census, pp. 39-58, also 
 law mentioned above, and Annals of the American Acad- 
 emy of Political and Social Science, vol. xii (1898), pp. 358- 
 386. 
 
14 CRIMINAL STATISTICS 
 
 plans called into consultation.^ Two sched- 
 ules of this census bore on criminals. One of 
 the questions asked on the population schedule 
 for free inhabitants was "whether deaf and 
 dumb, blind, insane, idiotic, pauper, or con- 
 vict."^ On the schedule for social statistics, 
 four questions were asked : first, as to the num- 
 ber convicted of crime during the year ending 
 June 1, 1850; second, as to the number of 
 those in prison on June 1, 1850; third, as to 
 nativity; and fourth, as to color of the native 
 born.^ That the questions regarding criminals 
 on these two schedules were ever designed to 
 supplement each other I do not believe. More 
 likely the word convict was added to the ques- 
 tion on the first schedule simply through force 
 of association of ideas. 
 
 To understand what people would be la- 
 beled as convicts by the assistant marshals 
 in filling out the first schedule, a knowledge 
 of the instructions sent out to them is neces- 
 sary.^ According to the definition, the in- 
 mates of a jail, a penitentiary, or a similar 
 
 ^ "He pVIr. Shattuck] drew up the mortality schedule by 
 request, though against his own advice, and also furnished the 
 schedule of social statistics and that in part of slaves." In- 
 troductory remarks of J. D. B. DeBow to the Compendium of 
 the United States Census of 1860, p. 13. 
 
 ^ See the census volume for 1850, p. xii, where schedules 
 are given. 
 
 ' To be found ibid. pp. xxi-xxv; also found in pamphlet 
 form. 
 
FEDERAL CRIMINAL STATISTICS 15 
 
 institution would constitute a family, and, 
 since all families were to be visited by the 
 assistant marshals, it may be supposed that 
 those found in these institutions would be 
 thus characterized. But this is not all. In- 
 structions were given that "when persons 
 who had been convicted of crime within the 
 year resided in families on the 1st of June, the 
 fact should be stated, as in the other cases of 
 criminals." Since it was felt that this inter- 
 rogatory might give offence, the assistant 
 marshals were advised to obtain this inform- 
 ation from the county record. This question, 
 it will be remembered, was on the population 
 schedule for free inhabitants, and thus the 
 general characteristics of sex, age, color, etc., 
 applying to all inhabitants would be recorded 
 of all who were termed convicts. Since the 
 question was not asked on the population 
 schedule for slaves, only those found in the 
 penal or correctional institutions would be 
 enumerated as convicts, and not those found 
 in private families as would be the case of free 
 inhabitants.^ 
 
 In the collection of the information called 
 for by the schedule of social statistics, public 
 records and reports and town, county, and 
 
 ' It is a curious fact that the instructions called for a return 
 of the crime for slaves found in prison, but not for the crime 
 in the case of free inhabitants. Pamphlet of Instruetiona for 
 1850, p. 18. 
 
16 CRIMINAL STATISTICS 
 
 state offices were to be looked upon as the 
 chief X sources of information. No definite 
 method of procedure was laid down, and it is 
 to be feared that very untrustworthy sources 
 were quite generally consulted.^ 
 
 Though, as already stated, the censuses of 
 1860 and 1870 were taken in accordance with 
 the provisions of the act of 1850, yet a few 
 changes were made in the schedules of inquiry 
 and in the instructions sent out to the mar- 
 shals and assistant marshals. The number of 
 questions on the population schedule for 1860 
 was increased, a circumstance which ipso facto 
 would increase the number of facts to be re- 
 corded concerning criminals. From the popu- 
 lation schedule of 1870 the question as to 
 whether the individual was a convict was 
 omitted, but this, of course, did not mean that 
 the population schedule no longer related to 
 criminals.^ All institutions were to be visited 
 
 » In theForum, vol. xxix (1900), p. 399, Roland P. Falk- 
 ner relates the following, which has to do with the collection 
 of these statistics in 1870: "A census enumerator rushed into 
 the office of one of our State Boards of Charities and Correc- 
 tions to inquire how many prisoners there were in that state. 
 On being informed that the office did not have the exact in- 
 formation he replied that there was nothing left for him to do 
 but to guess the number; and the figure which he evolved 
 from his inner consciousness became the record of the pris- 
 oners of the state." 
 
 ' In the History and Growth of the United States Census it is 
 stated (p. 55) that "the inquiries under 'pauperism and 
 
FEDERAL CRIMINAL STATISTICS 17 
 
 and the name on the schedule would indicate 
 whether penal or otherwise. The instructions 
 issued in 1860 were much more detailed than 
 those of 1850, and called for a statement of 
 the crime for which confined. In 1870, the 
 instructions relating to that part of the work 
 which concerns us were very brief. In the col- 
 lection of the social statistics a slight change 
 was made. The marshal could appoint a 
 deputy marshal to take charge of this particu- 
 lar work for the entire district or state. 
 
 No great amount of time was spent on the 
 compilation of the statistics of criminals for 
 these three censuses, and the published facts 
 were so brief that even the whereabouts of 
 those for 1850 and 1860 seem not to be gener- 
 ally known. The figures for 1850 are to be 
 found in the Compendium of the United States 
 Census for 1850, pp. 165-168, where they first 
 appeared. Both the population schedules* 
 and the schedules for social statistics were 
 used in making up the tables as well as a 
 
 crime' as to the number of native paupers and criminals on 
 June 1, respectively, were extended to distinguish native 
 whites and native blacks." This must be a mistake, as this 
 inquiry was on the schedule of 1850. 
 
 * Mr. Falkner is partly mistaken when he says (Pm5. of 
 American Economic Association " Federal Census Essays," 
 p. 171) that the facts regarding criminals on the population 
 schedules were not tabulated. Those for 1850 were. 
 
18 CRIMINAL STATISTICS 
 
 prison society report and a report made by 
 the Secretary of State in 1845. The totals 
 used for purposes of comparison in subsequent 
 reports are those obtainable from the table 
 made up from the schedules of social statis- 
 tics. In 1860, even less attention seems to 
 have been paid to them. So far as I am aware, 
 the only reference to them in the volumes of 
 the census for 1860 is that found on page 512 
 of the volume called Mortality and Miscella- 
 neous Statistics. Though positive evidence is 
 lacking, I am inclined to think that the table 
 there given was made from the schedules of 
 social statistics. The statistics of criminals 
 collected at the census of 1870 appear in the 
 volume. Population and Social Statistics, page 
 568, and again in the Compendium of this 
 census on page 531. Accompanying these 
 there is a brief discussion of their meaning 
 and limitations. 
 
 Prior to the taking of the tenth census, a 
 complete change in organization took place. 
 The appointment of the superintendent was 
 placed in the hands of the President; the work 
 of supervising the enumeration was taken 
 from the marshals and given to special officers 
 under the direct management of the superin- 
 tendent; and the actual enumerating was en- 
 trusted to a set of men who were subject to 
 the special oflScers mentioned above. The 
 
FEDERAL CRIMINAL STATISTICS 19 
 
 great change, however, which influenced di- 
 rectly the criminal statistics of this census 
 was the appointment of experts to take charge 
 of the various subjects included in the Social 
 Statistics.^ Mr. Frederick H. Wines was se- 
 cured for that division of the Social Statistics 
 which related to "pauperism and crime," and 
 to him must be given the credit for enlarging 
 the scope of these particular statistics. 
 
 The plan determined upon by Mr. Wines 
 was to enter all prisoners on the regular popu- 
 lation schedule and then to supplement the 
 information thus gained by the use of a second 
 schedule especially designed for the require- 
 ments of this class. Not content with these 
 inquiries pertaining to the prison population, 
 he sought for the additional knowledge con- 
 cerning criminality to be gained from judicial 
 records. Abstracts of the court dockets and 
 of the justices' returns, together with detailed 
 reports from police departments, were re- 
 quired. The supplemental schedule contained 
 something like nineteen inquiries and the 
 special blanks for judicial records about one 
 hundred and thirty-nine items. ^ 
 
 * In the instructions sent out to the enumerators, it is 
 stated that Schedule 4 — Social Statistics — was withdrawn 
 entirely from their hands. This is only partly true, since the 
 regular enumerators obtained the information called for by 
 the supplemental schedules, which were really a part of the 
 Social Statistics. 
 
 ' See History and Growth qf the United States Census, p. 97. 
 
20 CRIMINAL STATISTICS 
 
 The information called for by the popula- 
 tion and supplemental schedules was collected 
 by the regular enumerators who received ad- 
 ditional pay for filling out the supplemental 
 schedule.^ The police departments seem to 
 have filled out their own schedules, but 
 whether special enumerators were obtained to 
 secure the information on the two remaining 
 judicial schedules I do not know. The omis- 
 sion of the word convict on the population 
 schedule of this census, as in 1870, and the 
 lack of instructions on this point make it 
 clear that the statistics of prisoners would 
 not include those who had been in prison but 
 were then residing in private families. 
 
 As I stated in the first part of the chapter, 
 federal criminal statistics began with the cen- 
 sus of 1880. The schedules just referred to con- 
 tain inquiries regarding the nature of the 
 crime, and the volume of the census which 
 relates to these statistics, including, as it 
 does, the results of these inquiries as well as a 
 record of the criminals, makes applicable to 
 these figures the designation criminal statis- 
 tics. 
 
 These statistics are to be found in the Com- 
 pendium of the census and in Volume XXI 
 which pertains to the defective, the depend- 
 
 * See Compendium for 1880, p. 1661, and Census for 1880, 
 vol. xxi, p. X. 
 
FEDERAL CRIMINAL STATISTICS 21 
 
 ent, and the delinquent classes. The tables in 
 the Compendium relate only to prisoners in 
 confinement on June 1, while those in Volume 
 XXI include, in addition to these, tables re- 
 ferring to those reported to have been sen- 
 tenced to imprisonment dm-ing the preceding 
 year and to police statistics. Little attention 
 was paid to the compilation and analysis of 
 the judicial criminal statistics. Owing to the 
 closing of the census office, Mr. Wines was not 
 able to accomplish as much as he had planned,^ 
 and this fact probably accounts for the meagre 
 tables derived from judicial sources. 
 
 The act of March 1, 1889, provided for the 
 same general organization of the work as did 
 the act under which the census of 1880 was 
 taken. Minor changes were, of course, intro- 
 duced, but these did not alter the general 
 scheme of organization. Mr. Frederick H. 
 Wines was again placed in charge of the crim- 
 inal statistics, and undertook their collection, 
 compilation, and analysis. 
 
 As in 1880, each individual belonging to the 
 criminal class was to be entered first upon the 
 regular population schedule and then upon a 
 supplemental schedule. The only additional 
 source of information utilized at this census 
 was the reports from police stations and lock- 
 ups, other judicial reports being omitted. The 
 
 » See Cemrn of 1880, vol. xxi, pp. v and xlvi. 
 
22 CRIMINAL STATISTICS 
 
 number of distinct inquiries on the supple- 
 mental schedule of 1890 was thirty-two, and 
 the number on the blank for police returns 
 thirty-seven.^ 
 
 On the population schedule for 1890, the 
 question was asked whether the individual 
 enumerated was a prisoner or a convict. This 
 seems to be a return to the policy of 1850 and 
 1860, but the instructions to the enumerators 
 lead me to believe that this question pertained 
 to those found in institutions only. 
 
 The published results of the investigation 
 of Mr. Wines are contained in two volumes. 
 Pages 123-263 of the volume called Crime, 
 Pauperism, and Benevolence and pages 161- 
 207 of the Compendium of the Eleventh Census 
 relate to the criminal statistics. A discussion 
 of all the classes of which Mr. Wines had 
 charge is found on pages 7-120 of the volume 
 first mentioned. 
 
 No attempt seems to have been made to 
 obtain any results from the police returns; 
 and, as the schedules — both population and 
 supplemental — referred to the de facto popu- 
 lation of the penal and correctional institu- 
 tions on June 1, 1890, no records of convic- 
 tions or commitments for the preceding year 
 were secured. 
 
 * Taken from History and. Growth of the United States Cen- 
 sus, pp. 97-98. 
 
FEDERAL CRIMINAL STATISTICS 23 
 
 By the provisions of the act of March 3, 
 1899, which provided for the census of 1900, 
 certain changes were brought about in the or- 
 ganization of the work. The office staff was 
 increased and the director of the census, for- 
 merly known as the superintendent, was given 
 greater power in the direction and control of 
 the work. But the most important change, 
 and the one which directly affected the sub- 
 ject imder consideration, was the limiting of 
 the work of the decennial enumeration to the 
 four inquiries relating to population, mor- 
 tality, agriculture, and manufactures. Pro- 
 vision was made, however, for the collection 
 of statistics on various special subjects after 
 the completion of the decennial work, and 
 among these special subjects was that of crim- 
 inality. Consequently the investigations in 
 this field did not start until three years after 
 the general enumeration.^ 
 
 During the intervening period the census 
 office was made permanent and was trans- 
 ferred from the Department of the Interior 
 to that of Commerce and Labor. Mr. Roland 
 P. Falkner was employed as an expert special 
 agent to take charge of the work, and the 
 plan and the scope of the inquiry were de- 
 
 1 For a description of the work of this investigation see the 
 introduction to the special report, Prisoners and Juvenile 
 Delinquents, 1904, published in 1907. 
 
24 CRIMINAL STATISTICS 
 
 termlned by him. Mr, Falkner soon resigned 
 and the work was then turned over to Mr. 
 John Koren, who is responsible for the treat- 
 ment and analysis of the statistics secured. 
 
 Three schedules were used in the investiga- 
 tion, one relating to the commitments to pe- 
 nal institutions during the calendar year 1904, 
 one to persons found in prison June 30, 1904, 
 and one designed to show the movements 
 of the prison population during the year. By 
 far the greatest emphasis was laid upon the 
 schedule of commitments,^ the desire being to 
 do away with the mistaken policy of basing all 
 knowledge of criminality in the United States 
 upon the record of those found in prison on a 
 certain day of the year. No success was had 
 with the third schedule and the returns were 
 not published. The results of the other sched- 
 ules appeared in 1907 as a special report of the 
 Bureau of the Census which bears the title 
 Prisoners and Juvenile Delinquents. 
 
 Increasing interest in criminal statistics is 
 responsible for another investigation on the 
 part of the federal government, the results of 
 which have not yet appeared. Most foreign 
 governments have for many years collected 
 
 ^ Mr. Koren in his introduction to the volume. Prisoners 
 and Juvenile Delinquents, makes a mistake in asserting that 
 statistics of commitments were never before gathered. Com- 
 pare his opening sentence with what is said in this chapter 
 concerning the censuses of 1850 and 1860. 
 
FEDERAL CRIMINAL STATISTICS 25 
 
 judicial criminal statistics, and the neglect of 
 the United States government to provide for 
 such collection has long been a source of re- 
 gret to many persons interested in social ques- 
 tions. In 1906, the Bureau of the Census was 
 authorized to collect such statistics and took 
 up the work in 1907. According to the plan 
 adopted the investigation is limited to certain 
 states and counties, and the collection is made 
 by special agents and clerks detailed from the 
 regular office force. These are sent into the 
 counties selected, and copies of the criminal 
 dockets of the courts are made by them. The 
 reports are to cover all cases for the calendar 
 year 1906. While the previous attempts of 
 the federal government to collect judicial 
 criminal statistics have all met with failure, 
 the chances of success of the present one seem 
 to be extremely good. The present efficient 
 organization of the census bureau argues well 
 for satisfactory results. 
 
 This brief history of the federal criminal 
 statistics may be summarized as follows: They 
 date from the census of 1880, though their 
 origin may be traced back to the census of 
 1850 when statistics of criminals were first 
 collected. With few exceptions they are prison 
 criminal statistics, and relate either to those 
 found in prison on a certain day of the year or 
 to those committed during the year preced- 
 
26 CRIMINAL STATISTICS 
 
 ing the census inquiries. In all cases, they 
 were collected and published through the 
 agency of the census office, or Bureau of the 
 Census, as it is now called; and, omitting the 
 investigations of 1904 and 1906, the collec- 
 tion of statistics was made at the time of the 
 general enumeration of the population and by 
 the regular enumerators.'^, 
 
 • While the statistics which have just been considered are, 
 so far as I am aware, the only federal statistics that compre- 
 hend the coimtry as a whole, some mention should be made 
 of the judicial and prison criminal statistics which are to be 
 found in the reports of the Attorney General of the United 
 States. As is to be supposed, these relate to crimes tried in 
 the federal courts and to United States prisoners. 
 
 By the Act of June 20, 1872 (R. S. U. S. sec. 384), the At- 
 torney General was required to made a report to Congress of 
 the business of the Department of Justice for the' preceding 
 fiscal year. This report was to include "statistics of crime" 
 and other information of a statistical nature. Considering 
 the fact that they are collected chiefly for administrative 
 purposes, they are well arranged. In the case of the prison 
 criminal statistics, the error of focusing the attention on the 
 number confined in the prisons on a certain day of the year 
 was not made. 
 
CHAPTER III 
 
 FEDEBAL CRIMINAL STATISTICS (continued) 
 
 ESSENTIAL NATURE AND MEANING 
 
 In the previous chapter it was shown that, 
 despite the attempts from time to time to ob- 
 tain judicial criminal statistics, the reports of 
 the census office which have up to this time 
 been issued are devoted almost exclusively to 
 prison criminal statistics.^ Emphasis was also 
 laid upon the fact that these statistics, not 
 counting those in the second part of the recent 
 report for the year 1904, relate only to the pris- 
 oners found in prison on a certain day of the 
 year. Now this last feature needs explanation. 
 It is one that affects the whole nature and 
 meaning of these statistics and one that has 
 caused them to be grossly misunderstood. 
 
 The purpose of criminal statistics is two- 
 fold : (1) that one may judge of the nature and 
 extent of criminality in a given geographical 
 area, and (2) that one may determine the 
 transformation, if any, which is occurring in 
 
 * The imimportant exceptions are as follows: The reports 
 for 1850, 1860, and 1870 contain tables showing the number 
 of people convicted during the year preceding the day of 
 enumeration. Crimes are not given. The report for 1880 gives 
 the number of those sentenced to imprisonment and some 
 police statistics. 
 
28 CRIMINAL STATISTICS 
 
 these two phases. The results, when known, 
 may give direction to many movements of one 
 kind or another, but the purpose of the sta- 
 tistics is to furnish these two sets of data. 
 Their application is another question. 
 
 Opinion may differ in regard to the kind 
 and number of facts which criminal statistics 
 must be able to yield in order thoroughly to 
 fulfil this twofold purpose.^ Much might be 
 said on this question, but such a discussion, 
 though interesting in itself, would be apart 
 from the aim of this chapter, which is to ex- 
 plain the nature and meaning of the federal 
 criminal statistics, by a critical analysis of 
 their distinctive and peculiar features. For 
 this particular task, the marrow of a theoreti- 
 cal discussion of criminal statistics is what, for 
 want of a better term, I have called the time 
 element. 
 
 Some statistics speak for a point of time, 
 others for a period of time. A census of popu- 
 lation is a good example of the first. Here the 
 question is, how many people are there in the 
 country, and the answer is, on such a day of 
 
 • * To constitute criminal statistics, they must, at least, 
 contain information regarding criminals and crimes. For a 
 discussion of the questions which it would be best to ask in 
 collecting criminal statistics, see" ZurEinfUhrung in die Krim- 
 inaJstatistik, insbesondere diejenige des deutschen Reichs,'' 
 by H. von Scheel in Allgemeines Statisches Arckiv for 1890, 
 pp. 200-203. 
 
FEDERAL CRIMINAL STATISTICS 39 
 
 the year there were so many counted.^ Ridic- 
 ulous as it must seem to those having any 
 knowledge of the nature of criminal statistics, 
 it is nevertheless true that to many the federal 
 criminal statistics seem analogous to popula- 
 tion statistics and appear to speak for a point 
 of time. It is, therefore, well to understand 
 what would be their nature if they spoke for 
 a point of time. Either they would be an 
 enumeration of all the crimes committed on 
 the day in question and of the criminals who 
 were responsible for them or they would be an 
 enumeration of criminals plus the crimes, not 
 of which they were guilty, but of those which 
 were committed on the day mentioned, for 
 as soon as one began to consider the crimes 
 of which criminals were guilty he would 
 most certainly be dealing with a period of 
 time and not with a point of time.^ Unlike 
 the concept of population or of mortality, the 
 concept of criminaHty has a two-fold essence 
 — criminals and crimes — not one or the 
 other. 
 
 Criminal statistics, however, belong to the 
 second category, and are classed along with 
 
 * The United States is content to let the day stand for the 
 point of time. Other countries, as Germany, are more exact, 
 their point of time being twelve o'clock midnight. 
 
 * In reality a day is a period of time and not a point of 
 time. But for the purpose of the discussion I have thought it 
 best to treat it as a point of time. 
 
30 CRIMINAL STATISTICS 
 
 mortality statistics, statistics of births, di- 
 vorce statistics, etc. They all speak for a 
 period of time. The length of this period is 
 governed by many circumstances. It may be 
 that the machinery for collecting the statistics 
 is adjusted to some particular length of time. 
 Perhaps they are collected chiefly for adminis- 
 trative purposes and, if so, the length of the 
 span will vary with the organization of the 
 work to be done. On the other hand, if their 
 collection is governed by those interested 
 mainly in the sociological results the span of 
 time will be such as will make possible a cor- 
 relation of the facts gathered with other 
 known facts of the social world.^ The well- 
 nigh universal custom of foreign nations is to 
 make the year the unit of time. In a continu- 
 ous series the span of time must always be the 
 same or, at least, reducible to the same. Com- 
 parisons, otherwise, would be impossible, for 
 statistical comparisons are always compari- 
 sons of number, and it is but natural to sup- 
 pose that the longer the span of time the 
 greater would be the number of events to be 
 recorded. 
 
 * As an example of such correlation, see the article by 
 Ferdinand Fonnies in Jahrbuch fiir Gesetzgehung, Verwaltung 
 und VolJesmrtschaft im Deutschen Reich, Zweites Heft, pp. 
 271-292, 1909, entitled "Eine neue Methode der Verglei- 
 chung statistischer Reihen (im Anschluss an Mitteilungen 
 liber kriminalstatistischen Forschungen "). 
 
FEDERAL CRIMINAL STATISTICS 31 
 
 If statistics do not belong to the class first 
 mentioned, that is, if they do not speak for a 
 point of time, but for a period of time, it is ab- 
 solutely necessary to know for what period of 
 time they do speak. If confusion exists here, 
 the statistics are worthless. The chief peculi- 
 arity of the federal criminal statistics lies in 
 this very fact that they do not speak for a 
 point of time but for a period of time, and that 
 the period for which they speak as a whole 
 is bounded by no clear limits. This is due to 
 the fact that there is a horizontal grouping 
 of the statistics themselves, each group hav- 
 ing its own individual period of time differ- 
 ent as to length from all the rest. The fail- 
 ure to recognize this circumstance in the 
 compilation and in the discussion offered 
 to the public has resulted in spreading 
 broadcast much information of a doubtful 
 character. 
 
 That part of the report for 1904 which is 
 similar to the previous ones furnishes material 
 with which to uphold the two contentions: 
 (1) that the statistics do not speak for a point 
 of time, and (2) that the period for which they 
 do speak is indefinite. In this report the fol- 
 lowing figures relating to those found in 
 prison on June 1, 1904, under sentence for the 
 crime of burglary are given: ^ 
 
 » See pp. 108-109. 
 
CRIMINAL STATISTICS 
 
 Life 
 
 92 
 
 IJ^ years 
 
 
 297 
 
 Indeterminate 
 
 
 2 years 
 
 
 1492 
 
 period 
 
 3146 
 
 2^ years 
 
 
 278 
 
 Under 1 month 
 
 19 
 
 3 years 
 
 
 1402 
 
 1 month 
 
 20 
 
 4 years 
 
 
 901 
 
 2 months 
 
 21 
 
 5 years 
 
 
 1983 
 
 3 months 
 
 45 
 
 6 years 
 
 
 466 
 
 4 months 
 
 17 
 
 7 years 
 
 
 532 
 
 5 months 
 
 7 
 
 8 years 
 
 
 312 
 
 6 months 
 
 165 
 
 9 years 
 
 
 112 
 
 7 months 
 
 1 
 
 10-14 years 
 
 
 1188 
 
 8 months 
 
 25 
 
 15-19 years 
 
 
 344 
 
 9 months 
 
 45 
 
 20 years 
 
 
 128 
 
 10 months 
 
 35 
 
 21 years and 
 
 over 
 
 110 
 
 11 months 
 
 37 
 
 Period not stated 
 
 34 
 
 1 year 
 
 794 
 
 
 
 
 
 
 
 
 Total, 
 
 14,048 
 
 The federal criminal statistics, of which an 
 example is given above, relate to criminals 
 found in confinement on June 1, 1904, and to 
 the crimes of which they were guilty, but these 
 crimes were not committed on that day but at 
 some time in the past. They therefore do not 
 speak for a point of time and must not be so 
 understood. 
 
 Taking up the second contention that the 
 total period is indefinite, a glance at the 
 figures here given reveals that there were, 
 omitting as too indefinite for discussion the 
 figures of those committed for life, an inde- 
 terminate period, and for a term not stated, 
 actually twenty-seven different time periods 
 
FEDERAL CRIMINAL STATISTICS 33 
 
 used, and of these four — those for varying 
 lengths of time — are not clearly defined. 
 The table, it is seen, states for example that 
 there were found in confinement 112 under a 
 sentence of nine years, and, therefore, since all 
 would be found in prison who had during the 
 last nine years been committed with a sen- 
 tence of nine years, the time unit for this 
 particular grade of burglary would be exactly 
 nine years. Likewise, the time unit for that 
 grade of burglary for which eight -year sen- 
 tences were imposed, and for which 312 were 
 found confined, would be eight years. The 
 time unit for the various grades of the crime 
 biu"glary would thus increase or diminish in 
 length with the increase or diminution in the 
 length of the sentence. Stated in other words, 
 in collecting the statistics the general crime 
 of burglary has been subdivided, as it were, 
 according to grievousness of character into a 
 number of classes, and each class examined 
 separately by the state, the examinations dif- 
 fering in length of time for each class, and 
 the longest being for the most grievous class. 
 All this discussion applies, of course, equally 
 well to the statistics of all the crimes in the 
 catalogue. 
 
 Now the consequences growing out of this 
 state of aflFairs are manifold and will be ap- 
 parent when the usefulness of the statistics 
 
84 CRIMINAL STATISTICS 
 
 is questioned. It will be remembered that 
 one of the purposes of criminal statistics is to 
 enable one to judge of the nature and extent 
 of criminality. The only set of facts which 
 can possibly form the basis for the answer to 
 the above question are data which tell of the 
 number of criminals and the kind and number 
 of crimes committed in a given area within 
 a definite time limit. Now the figures taken 
 from the report for 1904 state that there were 
 14,048 persons found in prison on June 1, 
 1904, who had been sentenced for the crime 
 of burglary. The period within which these 
 crimes took place, is, however, not indicated. 
 Now it is not enough to have given a state- 
 ment of a number of various crimes and crim- 
 inals. The finite mind must know the length 
 of time within which these phenomena took 
 place, otherwise it cannot come to any con- 
 clusion, since one of its basic principles of 
 thought is missing. The fact is that, as has 
 been shown, there is not one span of time 
 within which these crimes took place but over 
 twenty-seven, and by no means that I know 
 of is it possible to reduce them to one unit of 
 time, without which change one cannot ob- 
 tain an idea of the nature and extent of crim- 
 inality. If one should add 112, the number of 
 those who were found in prison with a nine- 
 year sentence, to 312, the number of those 
 
FEDERAL CRIMINAL STATISTICS 35 
 
 who were found in prison with an eight-year 
 sentence, he would have the sum of 424, that 
 is, 424 persons were convicted of the crime of 
 burglary and sentenced to serve a term of 
 either eight or nine years within — is it nine 
 years or eight years? The fact is that, though 
 the process of addition has united the two 
 groups of prisoners into one, it has not, nor 
 can it, imite the two separate time units which 
 are there involved. The two spans of time 
 remain as distinct and separate as before, and 
 "all the king's horses" cannot bring them 
 together. Four hundred and twenty-four does 
 not represent all who during the last nine years 
 were committed on an eight or nine year sen- 
 tence, for those who were committed on an 
 eight-year sentence during the first of the nine- 
 year series would by June 1, 1904, have served 
 their time and been dismissed. Conversely, 
 424 represents more than the number of those 
 who during the last eight years were com- 
 mitted to prison on an eight or nine year sen- 
 tence, since it would also include a ninth-year 
 contingent of those committed on a nine-year 
 sentence. Similarly one can add together the 
 various numbers of those who were committed 
 on sentences ranging from an indeterminate 
 period to a life term and obtain the sum of 
 14,048, but this process of addition has not 
 united the numerous time units which are in- 
 
S6 CRIMINAL STATISTICS 
 
 volved. Fourteen thousand and forty-eight 
 does not represent the number of those who 
 were committed to prison during the last four 
 score and ten years for the crime of burglary, 
 nor does it represent the number of those who 
 were committed during the last month. Since 
 then no one time unit is given for the general 
 classes of crimes known as burglary, murder, 
 arson, etc., the nature and extent of criminal- 
 ity can only indirectly be known by studying 
 each grade of the classes of crimes separately. 
 Thus one could take those grades of all crimes 
 for which a twenty-year sentence had been 
 given and, since he would have a definite time 
 unit of twenty years, make some sort of an 
 analysis. In the same way, he would be forced 
 to group together those grades of crime for 
 which a nineteen-year sentence had been im- 
 posed, and analyze them for a nineteen-year 
 period, the process having to be repeated as 
 many times as there were time units, or, in 
 other words, as there were sentences of differ- 
 ent length. His results would be most pecu- 
 liar on account of this horizontal grouping, 
 but I believe that this method is the only one 
 that borders on the principles of statistical 
 analysis as applied to criminal statistics. This 
 method offers, too, the only means of making 
 them subserve the second of the chief pur- 
 poses of criminal statistics. For example, a 
 
FEDERAL CRIMINAL STATISTICS 37 
 
 comparison of certain rates of two periods is 
 possible. For all crimes for which five-year 
 sentences were given, rates could be obtained 
 for the two five-year periods 1885-90 and 
 1899-1904, the rates beiag obtained by divid- 
 ing the number of crimes of each period by the 
 population of the middle year of that period 
 and multiplying the quotient by one thousand 
 or ten thousand as the circumstances seemed 
 to render advisable. 
 
 These statistics have passed among the 
 general public for much more than their true 
 worth. In reality, they tell little or nothing of 
 criminality in the United States. 
 
CHAPTER IV 
 
 STATE CRIMINAL STATISTICS — JUDICIAL 
 
 America," says James Bryce, "is a Com- 
 monwealth of commonwealths, a Republic of 
 republics, a State which, while one, is never- 
 theless composed of other states even more 
 essential to its existence than it is to theirs."^ 
 This is the secret of our government, and 
 whoever wishes to understand the history of 
 statistical work in this country must keep it 
 continually before his mind. The real power 
 to collect criminal statistics lies not in the 
 federal government but in the states. Hence 
 it is not surprising that several of the states 
 began this work at a very early period. Both 
 judicial and prison criminal statistics are and 
 have been collected by some of the states, 
 though others have paid no attention to the 
 collection of either. 
 
 On account of certain fundamental differ- 
 ences in organization, the history of state 
 criminal statistics has been separated into 
 two parts — judicial and prison criminal 
 statistics. This chapter will be devoted to a 
 consideration of the judicial criminal statis- 
 
 1 American Commonwealth (edition of 1910), vol. i, ch. ii, 
 p. 15. 
 
STATE CRIMINAL STATISTICS 39 
 
 tics while the following chapter will treat of 
 prison criminal statistics. 
 
 The existence of state judicial criminal sta- 
 tistics seems to be a fact little known. In my 
 search for material I found almost no reference 
 to them, the prevailing opinion seeming to be 
 that Massachusetts was the only state that 
 collected such statistics.^ Yet the history of 
 these statistics in one state covers a period of 
 eighty years, and at the present time twenty- 
 five of the fifty states and territories that go to 
 make up the Union are collecting either judi- 
 cial criminal statistics or statistics of crimes.^ 
 
 The first states to undertake such work 
 were New York, Massachusetts, and Maine. 
 The impulse leading to the collection may have 
 come partly from abroad, for the studies in 
 criminal statistics of Guerry and Quetelet 
 appeared about the time that these states 
 first turned their attention to the collection of 
 such statistics.^ The extension to other states 
 
 1 See the Report of the Director of the Census to the Sec- 
 retary of Commerce and Labor for 1905 recommending the 
 compilation of judicial statistics. 
 
 2 Since this chapter was written a report of a committee 
 of the American Institute of Criminal Law and Criminology 
 has appeared in the journal of this institute for September, 
 1910. Li this report Mr. John Koren gives a brief history of 
 the work of the individual states in the collection of judicial 
 criminal statistics. 
 
 • Quetelet's Recherches statistiques sur h Royaume de Pays- 
 Bos was published in 1829; his Recherches sur U penchant au 
 
40 CRIMINAL STATISTICS 
 
 of the practice was no doubt due to two things. 
 Many of the western states modeled their 
 constitutions after that of New York, and this 
 fact probably brought about a similarity in 
 the statutes defining the duties of the offi- 
 cers. The second cause is, of course, to be 
 found in the likeness of needs which would 
 naturally beget like laws, especially since the 
 states seem to have used each other in this 
 particular as models. 
 
 In general the statistics are those derived 
 from reports sent by the states' attorneys or 
 clerks of criminal courts to some one of the 
 state officers as attorney general, secretary of 
 state or governor, and they usually appear in 
 the reports made to the legislature by the 
 officer in question. They are always for a 
 definite time interval, a year or two years, 
 depending somewhat on the time of meeting 
 of the state legislature. In some cases there 
 is a minute characterization of the court pro- 
 ceedings, while in others a brief summary is 
 
 crime aux diffirens dges was read in the session of July 9, 1831, 
 of the Royal Academy of Brussels, and the first edition of his 
 Physique soeiale, ou essai sur le dheloppement des facuttSa de 
 Vhomme came out in 1835. Note below the reference that is 
 made to this last work in the history of judicial criminal stat- 
 istics of New York on p. 47. Guerry's Essai sur la statistique 
 morale de la France appeared in 1833. This was preceded by a 
 small pamphlet on criminal statistics which was published in 
 1829. Von Mayr regards Guerry as the founder of Moral Stat- 
 istics. See_Staiistikund Gesellschaftslehre (1895), vol. i, p. 185. 
 
STATE CRIMINAL STATISTICS 41 
 
 all that is given. The tabulation is quite gen- 
 erally very defective, many of the reports giv- 
 ing no summary of the facts for the state as a 
 whole. But by far the gravest defect is the in- 
 completeness of the returns, a 'circumstance 
 which renders the statistics of many states of 
 little value. 
 
 In the paragraphs which follow, I have at- 
 tempted to present a brief history of the sta- 
 tistics for each state that collects them. I 
 have not, however, differentiated sharply 
 between criminal statistics and statistics of 
 crimes. Indeed in some cases it would have 
 been impossible to do so from the material at 
 hand. But such a distinction ought certainly 
 to be made in handling the statistics them- 
 selves. In the table summarizing the work of 
 the states which follows at the end of this 
 chapter and again in Chapter VI, the two 
 kinds are grouped together under the one 
 heading of Judicial Criminal Statistics. The 
 sketches are in every instance founded on a 
 study of the laws governing the collection and 
 in nearly every case on the reports in which 
 they appear. 
 
 In 1839, the legislature of this state passed 
 a law requiring the county at- . 
 
 torneys to report each year to the 
 attorney gfeneral the amount and kind of 
 
42 CRIMINAL STATISTICS 
 
 official business done by them, the number of 
 persons prosecuted, the offenses for which 
 such prosecutions were had, the results of the 
 prosecutions and the punishments meted out.^ 
 By the same law, the attorney general was 
 required to make out a similar report of his 
 work and to present it, along with an abstract 
 of the reports made to him by the county 
 attorneys, to the governor and his council. 
 The substance of this law was incorporated in 
 all the revisions which followed.^ Its pro- 
 visions, however, have not always been com- 
 plied with. For example, the attorney general 
 seems not to have made a report during the 
 period 1841-58, and at first the county at- 
 torneys failed to send to the attorney gen- 
 eral the reports that were required of them. 
 The arrangement in the reports of the attor- 
 ney general of the statistics thus collected is 
 such that a comprehensive view of the situa- 
 tion is not easily obtainable, the emphasis 
 being laid on the conditions prevailing in the 
 different counties. 
 
 The data which are available for a history 
 
 1 Law of 1839, ch. 408. 
 
 2 Revision of 1840, ch. 101, sec. 6; Revision of 18^0, ch. 102, 
 sec. 7; Revision of 1867, ch. 77, sec. 34; Revision of 1857, ch. 
 79, sec. 17; Revision of 1871, ch. 77, sec. 26; Revision of 1871, 
 ch. 79, sec. 15; Revision of 1883, ch. 77, sec. 57; Revision of 
 1883, ch. 79, sec. 19; Revision of 1903, ch. 79, sec. 64. 
 
STATE CRIMINAL STATISTICS 43 
 
 of the judicial criminal statistics for this state 
 are very incomplete. In the 
 Revised Statutes of 1901, there ^^^ Hampshire 
 is a provision requiring certain county officers, 
 among whom are mentioned solicitors, to 
 make up reports and to send them to the sec- 
 retary of state to be bound. ^ The reports of 
 the solicitors were to include full statistics of 
 crimes and misdemeanors. I have not been 
 able to secure the bound copies, and am there- 
 fore unable to speak ot their contents. 
 
 In 1832,^ it was made the duty of the at- 
 torney general to present to the legislature 
 
 a report concerning the suits ,, 
 
 J 4.- u- T, X, 4. Massachusetts 
 
 and prosecutions wnicn ne at- 
 tended, the report to include an abstract of 
 the reports made to him by the district at- 
 torneys.^ These reports were made to the 
 legislature each year until 1843, when the 
 office of attorney general was abolished.* The 
 attorneys were then required to make their 
 report to the secretary of the commonwealth, 
 who was to present to the legislature an 
 abstract of the same. In 1849, the office of 
 
 * Revised Statutes of 1901, ch. 17, sees. 5 and 8. 
 2 Law of 18S2, ch. 130, sees. 8 and 9. 
 
 ' Some time previous to this the legislature ordered returns 
 to be collected for the period 1826-30. See BejioTt of 1833, 
 p. 10. 
 
 * Law of 181^, ch. 99. 
 
44 CRIMINAL STATISTICS 
 
 attorney general was reestablished, and his 
 former duty of making reports was taken 
 from the secretary of the commonwealth and 
 given back to him.^ He continued to make 
 these reports showing the results of his own 
 work and that of the district attorneys until 
 1862. A law was passed in 1859 that the 
 clerks of courts, trial justices, and the clerks 
 or justices of police courts should make re- 
 turns to the secretary of the commonwealth." 
 He was required to make an abstract of these 
 and append it to the report of the attorney 
 general. From this time on, the report of the 
 attorney general contained only the results 
 of his own work, the abstract appended to it 
 by the secretary of the commonwealth evi- 
 dently taking the place of the abstract of the 
 reports of the district attorneys. An act of 
 1870 struck out the words, "and append the 
 same to the annual report of the Attorney 
 General. "* There is no evidence to show that 
 any abstract of these returns was thereafter 
 published or made by the secretary of the 
 commonwealth.* An act of 1881, however, 
 
 f • Law of 18Ji9, ch. 186, sees. 4 and 5. 
 
 ' Law of 1859, ch. 14, sees. 11 and 12. 
 
 ' Laws of 1870, ch. 274. 
 
 * See what Carroll D. Wright says in the Eleventh Ati- 
 nual Report of the Massachusetts Bureau of Statistics of Labor, 
 part iii, pp. 126-127, in regard to the origin of the criminal 
 statistics which he there uses. 
 
STATE CRIMINAL STATISTICS 45 
 
 provided that these returns should henceforth 
 be made to the commissioners of prisons in- 
 stead of to the secretary of the common- 
 wealth.^ These commissioners were required 
 to publish such abstracts and tabular state- 
 ments as should show the results of criminal 
 prosecutions in the courts of the common- 
 wealth. In 1901 a new board was created, 
 but the returns were made to this new board 
 in the same manner as they had been made to 
 the old board.* 
 
 One other set of reports maybe looked upon 
 as pertinent to this history. In 1852, a law 
 was passed requiring justices of the peace to 
 make returns to the secretary of the common- 
 wealth of criminal cases coming before them.^ 
 Abstracts of these reports were made by the 
 secretary of the commonwealth for the years 
 1853 to 1857, inclusive, and published in the 
 Documents of the General Court (excepting for 
 the year 1853, when the report appeared in 
 the legislative Documents of the House). The 
 reports of these justices for the period 1862-63 
 to 1868-69 were published in the abstract of 
 the secretary of the commonwealth by the 
 law of 1859. 
 
 Few of the reports published prior to 1881 
 are complete for all counties, and it is to be 
 
 1 Laws of 1881, ch. 66. ^ Laws of 1901, ch. 364. 
 
 » Laws cf 1852, ch, 289, sec. 1. 
 
46 CRIMINAL STATISTICS 
 
 greatly regretted that the commissioners, in 
 compiling these statistics since that date, did 
 not see fit to tabulate the statistics for all 
 courts by crimes and convictions. As it is, 
 they are of little use except to show the na- 
 ture of the work of the courts, since the 
 unanalyzed totals cannot be used as evidence 
 in a study of criminality.^ 
 
 The Revised Statutes of 1829 made it the 
 duty of the clerks of courts of records to enter 
 judgment of any conviction in their minutes 
 and to send to the secretary of 
 JN ew Yor state a transcript of these entries 
 
 ten days after the adjournment of such court. ^ 
 To aid the clerks in this work, the district at- 
 torneys could be called upon to prepare a 
 statement of the offense upon which the per- 
 son was convicted. At the time, the sole pur- 
 pose of filing these transcripts in the oflSce of 
 the secretary of state was that they might 
 furnish evidence of previous convictions when 
 an old offender was committed for trial on a 
 new charge. On April 13, 1838, the Senate 
 passed a resolution requiring the secretary of 
 state to present an abstract of the returns of 
 
 ^ Something is said of the history of criminal statistics in 
 Massachusetts in the article, "Statistics of Crime in Massa- 
 chusetts," by Frederick G. Pettigrove in Puhlicationa o/ the 
 American Statistical Association, 1892-93, pp. 3, 5. 
 
 ' Revised Statutes of 1829, part iv, title 6, sees. 5, 6, 7, 8., 
 
STATE CRIMINAL STATISTICS 47 
 
 convictions for criminal offenses. This ab- 
 stract appeared in 1838 and covers the period 
 1830 to 1837 inclusive. It includes, in addi- 
 tion to the returns from courts of record, 
 statistics from the mayors' courts in the cities 
 of Albany and Rochester and from the courts 
 of special sessions in the counties of Putnam 
 and Yates. It is interesting to note that the 
 secretary of state carefully analyzed these 
 statistics and compared the results with for- 
 eign statistics, making frequent references to 
 Quetelet's book. Physique soeiale, ou essai sur 
 le developpement des faeultes de Vhomme. It 
 may indeed be that the interest awakened by 
 Quetelet's book was responsible for the be- 
 ginning of judicial criminal statistics in New 
 York. In 1839, a similar abstract appeared, 
 and in this year a law was passed by the 
 legislature which served to make perma- 
 nent the Senate resolution of 1838. In sub- 
 stance, it was to the effect that the clerk 
 of every court of record should transmit to 
 the secretary of state a statement in regard 
 to the indictments, convictions, acquittals, 
 etc. It was furthermore provided that the 
 sheriff of the county in which the court of 
 record was located should transmit to the 
 secretary of state personal information re- 
 garding each individual thus convicted. The 
 clerks were at the same time to send to the 
 
48 CRIMINAL STATISTICS 
 
 secretary of state copies also of all certificates 
 of convictions made by any court of special 
 sessions and required to be filed with them. 
 The clerk of the court of general sessions in 
 the city and county of New York was required 
 to transmit to the secretary of state a tran- 
 script of every conviction had in a special ses- 
 sion of this city and county. It was also 
 made the duty of sheriffs to give the same 
 kind of information concerning persons con- 
 victed in courts of special sessions in the cities 
 of New York, Albany, Hudson, Schenectady, 
 Troy, Utica, Rochester, Brooklyn, and Buf- 
 falo as they did regarding those convicted in 
 courts of record. The secretary of state was 
 required to publish annually an abstract of 
 such returns for the use of the legislature.^ 
 Thus, it is seen, provision was made at this 
 early date for the collection and publication 
 of thoroughgoing judicial criminal statistics. 
 
 In 1861, a law was passed consolidating and 
 amending the previous acts relating to these 
 statistics. No important changes, however, 
 were made in the existing laws. The act was 
 more definite than previous ones regarding 
 the information to be forwarded to the secre- 
 tary of state. It made also such amendments 
 as were necessary to secure returns from the 
 courts which had in the meantime been cre- 
 1 See Laws of 1839, ch. 259, sees. 1-9. 
 
STATE CRIMINAL STATISTICS 49 
 
 ated.^ The law of 1866, made several changes 
 in the law of 1861, but they were all in the 
 nature of supplementing or revising the main 
 features of the previous law.^ For example, 
 provision was made that as soon as a city 
 should be incorporated reports from the court 
 of special sessions in that city should be made 
 by the sheriff. The law of 1867 did little more 
 than regulate the details of collection, ex- 
 tending and remodeling the existing laws 
 where the facts seemed to warrant it.* The 
 final statement of the law governing the col- 
 lection and compilation of these statistics is 
 to be found in the Criminal Code of the state.* 
 As already indicated these are true judicial 
 criminal statistics. In the reports of the sec- 
 retary of state, they are summarized and to 
 some extent analyzed. They would be of the 
 greatest importance were it not for the fact 
 that in no year have the reports been com- 
 plete for all counties.^ Penalties for failure 
 to comply with the law calling for the collec- 
 tion of the statistics there were, but the 
 proper pressure was not brought to bear to 
 ensure the enforcement of the law. 
 
 1 See Laws of 1861, ch. 97. ' See Laws of 1867, ch. 604. 
 
 2 See Laws of 1866, ch. 723. * See Code, title x. 
 
 ' The following references taken at random from my notes 
 will serve as partial evidence of this: 1841, v. 3, S. Doc. 67, 
 p. 8; 1860, V. 6, A. Doc. 190, p. 4; 1866, v. 5, A. Doc. 114, 
 p. 3; 1873, V. 2, A. Doc. %, p. 10; 1888, v. 6, .4. Doc. 43, p. 9. 
 
so CRIMINAL STATISTICS 
 
 The history of judicial criminal statistics 
 in this state goes back to the year 1847, when 
 an act was passed requiring 
 Pennsylvania prothonotaries or clerks of all 
 criminal courts to transmit to the secretary of 
 state statements regarding the criminal busi- 
 ness of such courts. By the same act it was 
 made the duty of the secretary of state to 
 make an abstract of these reports and to lay 
 it before each branch of the legislature.^ What 
 evidence I have found seems to point to the 
 conclusion that these provisions of the law 
 were never complied with until changed by 
 the act of 1872.^ This latter act was to the 
 effect that the prothonotaries or clerks should 
 send their reports to the Board of Public 
 Charities which had been created in 1869.' 
 Statistics for the year ending September 30, 
 1873, were secured by this board and prepared 
 for publication under the direction of Dr. A. 
 J. Ourt. Since that time, the board has con- 
 tinued to collect these statistics, and the re- 
 sults appear in the annual report which they 
 
 I * See Laws of 18^.7, Act no. 131, sees. 2, 4. 
 
 ^ The act of 1847 made provision also for the collection of 
 prison criminal statistics, and the secretary of state states in 
 1868 that up to that time this part of the law had remained a 
 dead letter. He says, however, nothing in regard to the court 
 statistics, but it is to be presumed that no attention was 
 paid to their collection. See the Pennsylvania Legislative 
 Documents for 1868, pp. 1324-5, for his statement of lie case. 
 
 = See Laws of 1872, no, 34. 
 
STATE CRIMINAL STATISTICS 51 
 
 issue. Since 1888 the handling of these statis- 
 tics has been done by those unacquainted 
 with the principles of statistical investigation, 
 and the effect has been to render the tables 
 thus compiled of very little use. The written 
 analysis is entirely misleading.^ 
 
 A law of 1889 required the clerks of crimi- 
 nal com-ts of record, or of superior courts hav- 
 ing criminal jurisdiction, to 
 transmit to the attorney gen- ° arolina 
 eral reports regarding the indictments disposed 
 of at such courts. These reports were to give 
 information about the number convicted, the 
 kind of crime, the sex, the race, etc. It pro- 
 vided also that the secretary of state should 
 make a tabulated report regarding these. ^ 
 This, however, he did not do, as the intent of 
 the law seemed to be that the attorney gen- 
 eral should make the report regarding these 
 returns,^ and in the Revisal of the Code for 
 1905 the section referring to the work of the 
 secretary of state was omitted.* Of the reports 
 presented by the attorney general only one, 
 that dated 1892, admits of use. The peculiar 
 
 ' ^ For a number of years the analysis was on the basis of 
 the number of those convicted, no attention being paid to 
 those who plead guilty. 
 
 2 See Laws of 1889, ch. 341. 
 
 ' See Report c^ Attorney General dated Dec. 8, 1892. 
 
 * See Code for 1905, vol. i, sec. 917. 
 
52 CRIMINAL STATISTICS 
 
 tabulation designed to show the business of 
 the courts leaves the facts of criminality very 
 much in the dark. 
 
 There is no statute that I can find which 
 expressly stipulates that the solicitors shall 
 make reports to the attorney 
 South Carolina g^^^^^i j^ ^is report of No- 
 vember 10, 1886, the attorney general states 
 that it is submitted in compliance with the 
 requirements of section 503 of the General 
 Statutes.^ This section requires an annual 
 report from him of the cases which he has 
 argued, tried or conducted during the preced- 
 ing year "with such other information in re- 
 lation to the criminal laws, and statements, 
 as, in his opinion, the criminal jurisdiction 
 and the proper and economic administration 
 of the criminal laws warrant and require." 
 Though reports from the solicitors are not 
 called for by this section, the attorney general 
 might have so interpreted it. At any rate his 
 reports since 1887 have included statistics of 
 their work. These are not summarized as a 
 whole, nor are they complete for the state, 
 some solicitors failing nearly every year to 
 report. 
 
 Another set of returns of interest in this 
 connection are those which an act of 1900 re- 
 
 > Found in Civil Code of 1902, ch. xix, art. iv, sec. 640. 
 
STATE CRIMINAL STATISTICS 63 
 
 quired the clerks of courts of general sessions 
 to make to the attorney general and to the 
 solicitors of their respective circuits.^ In a 
 way, they seem to supplement the reports 
 of the solicitors since they give the name, the 
 race, the sex, the age, and the alleged crime of 
 every person brought to trial in their courts. 
 They appear also in the annual report of the 
 attorney general and bear the heading " Crim- 
 inal Statistics." They likewise are not sum- 
 marized, and it is only through patient and 
 painstaking effort that they can be made to 
 yield results. 
 
 In his report of March, 1903, the attorney 
 general gives a copy of the letter which he had 
 sent to the state and prosecut- 
 ing attorneys of the criminal °" 
 courts of record.^ He states in this letter that 
 he finds it imperative to require the enforce- 
 ment of section 91 of the Revised Statutes 
 providing for reports from them. His instruc- 
 tions are that they send him quarterly reports, 
 the first one to cover the period October 1, 
 1904, to January 1, 1905. Statistical abstracts 
 of the reports of the attorneys appear in the 
 attorney general's report for 1905-1906. 
 Several counties seem not to have responded, 
 
 1 Laws of 1900, no. xxiii, p. 442. 
 * See p. 4 of this report. 
 
54 CRIMINAL STATISTICS 
 
 and the tables given are not well sum- 
 marized. 
 
 The office of attorney general was created 
 by an act passed in 1846, By section 14 of 
 that act, the prosecuting attorneys 
 ° were required to -send to the at- 
 
 torney general a particular statement of 
 crimes in their respective counties, specify- 
 ing the number prosecuted, the crimes, the 
 punishments, etc.^ The first report of the 
 attorney general was made in 1846, and con- 
 tained the information forwarded by the 
 prosecuting attorneys. Several of the counties 
 made no report for this year, and continued 
 negligence along this line finally compelled 
 the attorney general to recommend in his 
 report for 1866 that this method of collecting 
 statistics be done away with, and that the 
 clerks of courts be required to furnish the 
 statistics to the commissioner of statistics. 
 This suggestion was embodied in a law in 
 1867,^ and that year these statistics appeared 
 in the commissioner's report. This office was 
 abolished in 1868, and a Bureau of Statistics 
 in the office of the secretary of state was 
 created.^ Since the creation of this office the 
 clerks have sent their reports to the secretary 
 
 > See law of Feb. 16, 1846, sees. 14 and 16. 
 
 2 See Laws of 1867, pp. 17-18. = Laws of 1868, pp. 92-93. 
 
STATE CRIMINAL STATISTICS 55 
 
 of state, and an abstract of these appears 
 yearly in his report. 
 
 In revising the laws in 1879, it was again 
 made the duty of the prosecuting attorneys 
 to send reports to the attorney general, the 
 revisers evidently forgetting that a similar 
 law had been purposely repealed in 1867.^ 
 The attorney general suggested in his report 
 of 1881 that this statute be repealed, but the 
 step was not taken. He experienced no better 
 success in obtaining these reports than his 
 predecessors had had under the act of 1846. 
 Not being able to secure the repeal of the 
 statute, he ceased finally to pay any attention 
 to it, and his report for the year 1892 is the 
 last one that contains such returns. 
 
 "With certain exceptions, the statistics found 
 in the manual of the secretary of state seem 
 complete enough for use. The tabulation is 
 peculiar, but with care some information re- 
 garding criminality can be obtained from 
 them. The reports of the attorney general are 
 all too defective to constitute any evidence in 
 the study of such a question. 
 
 In 1899 it was made the duty of the chief 
 of the Indiana Bureau of Statis 
 tics "to collect, compile, system- 
 atize, publish, print and report biennially" 
 
 » See Revised Laws of 1879. sec. 1282. ' 
 
56 CRIMINAL STATISTICS 
 
 criminal statistics for the state. The clerks of 
 all courts were required to furnish the infor- 
 mation demanded by the chief. ^ This law has 
 been carried out. The statistics are fairly well 
 arranged, but owing to changes introduced in 
 1907 comparisons with former years are next 
 to impossible. From a perusal of the published 
 figures it is difiicult to say whether the statis- 
 tics are complete for the whole state or not. 
 
 The reports of the attorney general cover 
 the long period of time from 1839 to the 
 present.'^ They have always con- 
 Michigan tained in some form or other 
 the reports of the prosecuting attorneys 
 which were required to be made to him. The 
 early reports are, however, very unsatisfactory 
 both on account of the poor tabulation and 
 the incompleteness of the returns. Since 
 1890, they have been very much better, and, 
 though not summarized, present some inter- 
 esting facts. 
 
 Though clerks of all courts having criminal 
 jurisdiction were, by an act of 1873, required 
 
 > Act of 1899, Laws of 1899, p. 515; Revision of 1908, vol. 
 3, sees. 9342-9344. 
 
 ' The first reference to any law on the subject that I found 
 is taken from the Revised Statutes of 1838, part i, title 3, ch. 
 1, sees. 21 and 22. These sections require a report from him 
 and from prosecuting attorneys. 
 
STATE CRIMINAL STATISTICS 67 
 
 to make out and transmit to the governor 
 of the state annual reports of the number 
 
 of convictions and other inf orma- 
 . J. . J . . Wisconsin 
 
 tion regarding crime and crimi- 
 nals,^ Wisconsin is sadly deficient in the mat- 
 ter of judicial criminal statistics. I find that 
 the secretary of state occasionally made ab- 
 stracts of these reports, as in 1879 and 1883. 
 They are, however, so meagre and so few in 
 number as scarcely to deserve mention. 
 
 In his report for 1861, the attorney general 
 
 called the attention of the legislature to the 
 
 fact that other states required 
 
 .1 4.- 4.4. 4. J Minnesota 
 
 the prosecuting attorneys to send 
 
 criminal statistics to the attorney general.^ 
 This fact having been brought to their no- 
 tice, the legislature passed a law providing 
 for such returns from their own prosecuting 
 attorneys. Beginning with the year 1862, 
 individual abstracts of the attorneys' reports 
 have appeared in the report of the attorney 
 general. They are very incomplete, the pro- 
 secuting attorneys failing from time to time 
 to send in returns. 
 
 The Code of 1851 made very full provisions 
 
 1 SieeLaws of 1873, ck.lQ9, "An Act relating to the Fur- 
 nishing of Certain Statistics of Crime." 
 * See p. 327 of his report for this year. 
 
58 CRIMINAL STATISTICS 
 
 regarding the collection and the publication 
 of judicial criminal statistics.^ According to 
 it, the clerk of each district court 
 ^^^ was required to report annually 
 
 to the secretary of state the number of con- 
 victions for all crimes and misdemeanors 
 which were had in that court. It also stated 
 that the report should show the character of 
 the offense, the sentence and the occupation 
 and habits of the convict. An abstract con- 
 taining all the facts reported was to be made 
 by the secretary and presented to the General 
 Assembly. This law has stood until the present 
 day, and is incorporated in the Code of 1897. '^ 
 Up to the year 1874, the counties were very 
 lax in making reports. In some years, as many 
 as thirty were deficient in this respect. Since 
 that time, much improvement has been made 
 in the work, and the abstracts of the secretary 
 of state constitute a valuable source of know- 
 ledge. 
 
 The Revised Codes of 1905 required the 
 
 state attorneys to make monthly reports to 
 
 „ , ^ , . the attorney general of all pro- 
 North Dakota ,. . "'^f . ^ ^ 
 ceedings m the various courts 
 
 of the counties (justices' courts excepted) 
 
 wherein the state was a party or was inter- 
 
 » Code of 1851, sec. 148. 
 
 » Code of 1897, ch. 8. title 3, sec. 293. 
 
STATE CRIMINAL STATISTICS 59 
 
 ested. Various facts were requested to be in- 
 cluded in the reports, such as the purpose of 
 the action and the disposition of the same."^ 
 Concerning the pubUcation of an abstract of 
 these reports by the attorney general I have 
 found no reference in the laws, and have been 
 unable to secure access to the Public Docu- 
 ments of the state published since 1904. 
 
 Since 1883, it has been the duty of all cir- 
 cuit and county solicitors of city and criminal 
 courts to send to the attorney 
 general a certified statement ^°^ 
 
 of the number of cases disposed of at each 
 term, the number of convictions, the number 
 of acquittals, the number sentenced to death, 
 to the penitentiary or to jail, etc. The attorney 
 general is required to make a report to the 
 governor of the business of his office and of the 
 criminal cases disposed of in the entire state 
 as shown by the reports of the solicitors,^ The 
 general statements appended to the reports of 
 the attorney general and showing the charac- 
 ter and results of the prosecutions are very 
 well gotten up. There is a good summary of 
 the state as a whole, and the information 
 seems to be complete and carefully prepared. 
 
 * See Revised Codes of 1905, sec. 2494. 
 = See Laws of 1883, p. 39. Also Codes of 1907, vol. iii, sec. 
 7785; vol. i, sees. 634-35. 
 
60 CRIMINAL STATISTICS 
 
 An act of 1855 (incorporated in subsequent 
 revisions) provided for a report from the 
 ^ . . attorney general to the legisla- 
 
 ture, the report to include the 
 annual reports of the several district attor- 
 neys,^ The items called for in the reports are 
 the usual ones of convictions, crimes, acquit- 
 tals, etc. As found in the reports of the attor- 
 ney general, the statements of the district 
 attorneys are separate and distinct, and in no 
 year, I believe, are all counties there repre- 
 sented, though, as in most cases, a penalty is 
 attached to the non-fulfillment of the law. 
 
 By an act passed in 1846, the attorney gen- 
 eral was required to present to the governor a 
 Texas report of the criminal business 
 
 of the state. To this end he 
 could demand that district and county attor- 
 neys and clerks of district and county courts 
 should furnish him with information.'^ The 
 Revision of 1879 made the duty of district and 
 county clerks more specific, and that of 1897 
 did the same for district and county attorneys. ^ 
 
 » Laws of 1855, p. 6. This does not specify that the dis- 
 trict attorneys shall send reports, though it speaks of includ- 
 ing them. This part seems to have been added in revising the 
 laws in 1870. See Revised Laws of 1870, sees. 131 and 1147. 
 Also Revised Laws of 1904, vol. i, pp. 53-54. 
 
 » Act of May 11, 1846, p. 206. ^ 
 
 ' Code of Criminal Procedure, art. 39. 
 
STATE CRIMINAL STATISTICS 61 
 
 The reports of the attorney general are, how- 
 ever, very disappointing. He says in his report 
 dated December 31, 1869 : "There are one hun- 
 dred and twenty-seven counties in the State 
 that report organization. I have full reports 
 only of the criminal proceedings for 1868 from 
 twenty-seven counties. I have reports from 
 thirty other counties for 1868,| covering one 
 half of the year, or one term of the District 
 Court therein. From'the balance of the coun- 
 ties, no report at all for that year. It will be 
 seen, therefore, that it is impossible to make 
 such a report as is by the law contemplated." ^ 
 The continuation of this state of affairs has 
 destroyed whatever value the reports might 
 have had. 
 
 The Political Code of 1895 called for a report 
 from the attorney general on the condition of 
 affairs in his department. It 
 also stated that his report was 
 to be accompanied by copies of the dockets 
 and reports sent to him by the county attor- 
 neys.^ The biennial report of December, 
 1900, 1 believe to be the first one made under 
 this act. Neither it nor the succeeding ones 
 contain reports from all counties, and hence 
 their usefulness is much impaired. 
 
 » See pp. 49-50 of the report of 1869. 
 ' Political Code. 1895, sec. 460, 12. I believe this to be the 
 first law on the subject, as I find no other references. 
 
62 CRIMINAL STATISTICS 
 
 In 1896, an act was passed which required 
 the attorney general to report to the governor 
 biennially and to include in his 
 report copies of the returns from 
 county attorneys.^ In no one report have all 
 the counties of the state been represented, 
 though it is possible, as in some of the other 
 states, to study the statistics of a certain 
 section of the state represented by the coun- 
 ties which reported each year. 
 
 The legislature of 1889 passed an act en- 
 titled, "An Act to require District Attorneys 
 to make certain reports to the 
 Nevada attorney general." ^ The re- 
 
 ports were to specify the number of persons 
 convicted, the average punishment, the num- 
 ber acquitted, and certain information regard- 
 ing the circumstances of crimes for which per- 
 sons were sentenced to state's prison. In his 
 biennial report for 1893-94, the attorney gen- 
 eral states that he had submitted in his previ- 
 ous reports the returns made to him by the 
 district attorneys, but that he had not felt it 
 worth while to do so that year on account of 
 their meagre and unsatisfactory character.* 
 Though the returns were afterwards included 
 
 1 See Laws of 1896, pp. 122-124. 
 
 ' Laws of 1889, ch. liv, sees. 1 and 2. 
 
 ' Report of the Attorney General for 1893-94, p. 6. 
 
STATE CRIMINAL STATISTICS 63 
 
 in some of his reports, his description, just 
 referred to, continued to be appHcable to them. 
 
 Though the Revised Laws of 1887 required 
 a report from the attorney general and ac- 
 companying reports from the 
 district attorneys, I am not ° 
 
 aware that the reports of the attorney general 
 ever contained such returns as were required 
 to be made to him by the attorneys.^ 
 
 An act of 1886 required all prosecuting at- 
 torneys to report each year to the governor 
 the nature and amount of 
 business transacted by them.* ^^ ngton 
 
 Now though the law of 1888, requiring a re- 
 port from the attorney general, said nothing 
 about including abstracts of the returns of the 
 prosecuting attorneys,^ yet the attorney gen- 
 eral's biennial report for the period 1891-92 
 contains such abstracts. This report, how- 
 ever, is the only one that does include them, 
 and, so far as I am aware, no other abstract 
 of these returns has yet been published. 
 
 "^The situation at the present time in this 
 state resembles that in the early period of the 
 
 1 See Revised Laws of 1887, sec. 250, 12. 
 ' Laws of 1886, p. 62, see. 13. 
 1 • Laws qf 1888, p. 8, sec. 7. ^ 
 
64 CRIMINAL STATISTICS 
 
 history of judicial criminal statistics in the 
 state of New York. Since 1862 there has been 
 a law requiring the district at- 
 Oregon torneys .to file in the office of 
 
 the secretary of state a written report of the 
 amount and kind of official business done by 
 them, specifying the prosecutions, their re- 
 sults, the punishments, etc.^ No abstract of 
 these, I believe, has ever been published. 
 
 On page 13 of the attorney general's report 
 for the year 1900, the following interesting 
 . sentence was found: "Section 
 
 470 of the Political Code has 
 long required reports from the several district 
 attorneys to accompany the biennial report 
 of the Attorney General, but, so far as I am 
 advised, that requirement is complied with 
 this year for the first time." The law to 
 which he here refers was passed in 1850,^ and 
 as it provided no penalty for refusing to com- 
 ply with its provisions the attorney general 
 was unable to obtain the returns from the 
 district attorneys.' The report of 1900 lacks 
 
 » Laws of 1863. 
 
 » Laws of 1850, ch. 11. 
 ^ ' See p. S of the report of the attorney general for 1860, 
 also p. 3 of his report for 1861, where he speaks as follows: 
 "It would afford me pleasure to give some reliable data in 
 reference to the amount of crime committed in this state 
 during the past year. . . . But such is out of my power owing 
 to the defect of the law relating to district attorneys." 
 
STATE CRIMINAL STATISTICS 65 
 
 returns from seventeen counties, and as the 
 later ones are about equally deficient, their 
 use is extremely limited. 
 
 Even Homer had to descend to tediousness 
 when telling of the forces that set out for 
 Troy, and these histories, which in this re- 
 spect resemble the famous catalogue of ships, 
 claim no other merit than that they honestly 
 attempt to represent the facts of the situation. 
 Presented in this fashion, a rapid survey of 
 the main features of the state judicial crimi- 
 nal statistics is somewhat diflBcult, and it has 
 therefore seemed best to summarize the main 
 facts in a tabular form. 
 
CRIMINAL STATISTICS 
 
 TABLE SHOWING WORK OF STATES IN THE 
 
 
 Date oJ 
 
 i 
 
 
 
 first law 
 
 
 STATE 
 
 authoriz- Collected by 
 ing col'ec- 
 tioR 
 
 Sent to 
 
 Maine 
 
 1839 
 
 County Attorneys 
 
 Attorney General 
 
 New Hampshire 
 
 1891 
 
 County Solicitors 
 
 Secretary of State 
 
 Ma8Sacliu3ett3 
 
 1832 
 
 Clerks of Courts, 
 
 Commissioners of 
 
 
 
 Trial Justices 
 
 Prisons 
 
 New York 
 
 1829 
 
 Clerks of Courts and 
 
 Sheriffs 
 
 Secretary of State 
 
 Pennsylvania 
 
 1847 
 
 Prothonotaries, 
 
 Board of Public Chari- 
 
 
 
 Clerks 
 
 ties 
 
 North Carolina 
 
 1889 
 
 Clerks of Courts 
 
 Attorney General 
 
 South Carolina 
 
 1886 
 
 Solicitors, Clerks 
 
 Attorney General 
 
 Florida 
 
 1904 
 
 State and Prosecut- 
 ing Attorneys 
 
 Attorney General 
 
 Ohio 
 
 1846 
 
 Clerks 
 
 Bureau of Statistics 
 
 Indiana 
 
 1899 
 
 Clerks 
 
 Bureau of Statistics 
 
 Michigan 
 
 1838 
 
 Prosecuting Attorneys 
 
 Attorney General 
 
 "Wisconsin 
 
 1873 
 
 Clerks 
 
 Governor 
 
 Minnesota 
 
 1862 
 
 Prosecuting Attorneys 
 
 Attorney General 
 
 Iowa 
 
 1851 
 
 Clerks 
 
 Secretary of State 
 
 North Dakota 
 
 1905 
 
 State's Attorneys 
 
 Attorney General 
 
 Alabama 
 
 1883 
 
 Solicitors 
 
 Attorney General 
 
 Louisiana 
 
 1855 
 
 District Attorneys 
 
 Attorney General 
 
 Texas 
 
 1846 
 
 District Clerks and 
 County Attorneys 
 
 Attorney General 
 
 Montana 
 
 1895 
 
 County Attorneys 
 
 Attorney General 
 
 Utah 
 
 1896 
 
 County Attorneys 
 
 Attorney General 
 
 Nevada 
 
 1889 
 
 District Attorneys 
 
 Attorney General 
 
 Idaho 
 
 1887 
 
 District Attorneys 
 
 Attorney General 
 
 "Washington 
 
 1886 
 
 Prosecuting Attorneys 
 
 Governor 
 
 Oregon 
 
 1862 
 
 District Attorneys 
 
 Secretary of State 
 
 California 
 
 1850 
 
 District Attorneys 
 
 Attorney General 
 
STATE CRIMINAL STATISTICS 
 
 67 
 
 COLLECTION OF JUDICIAL CRIMINAL STATISTICS 
 
 Where tonnd 
 
 In reports of Attorney General « Not summarized but fairly good 
 
 In reports of Commissioners 
 
 In reports of Secretary of State 
 
 In reports of Board 
 
 In reports of Attorney General 
 
 In reports of Attorney General 
 In reports of Attorney General 
 
 In reports of Bureau 
 
 In reports of Bureau 
 
 In reports of Attorney General 
 
 In reports of Secretary of State 
 
 In reports of Attorney General 
 
 In reports of Secretary of State 
 
 In reports of Attorney General 
 
 In reports of Attorney General 
 
 In reports of Attorney General 
 
 In reports of Attorney General 
 In reports of Attorney General 
 In reports of Attorney General > 
 In reports of Attorney General ' 
 In reports of Attorney General' 
 Not published 
 In reports of Attorney General 
 
 1 Not now publiehed. 
 
 The tabulation not suited to crimi- 
 nal statistics 
 
 Well summarized but returns are in- 
 complete 
 
 Poorly tabulated 
 
 Not tabulated by crime and convic- 
 tion 
 
 Returns incomplete 
 
 Incomplete returns and not sum- 
 marized 
 
 Good 
 
 Fairly good 
 
 Fairly good 
 
 Poor and meagre 
 
 Incomplete returns 
 
 Fairly good 
 
 Not published 
 
 Well summarized and seemingly 
 complete 
 
 Incomplete returns and not sum- 
 marized 
 
 Incomplete returns mostly 
 
 Not summarized 
 Incomplete returns 
 Very unsatisfactory 
 
 Incomplete and not tabulated 
 3 Not found. 
 
68 CRIMINAL STATISTICS 
 
 From this table, the extent and organiz- 
 ation of state judicial criminal statistics is at 
 once seen.^ Their development has already 
 been portrayed, and the following chapter 
 will take up the history and organization of 
 state criminal statistics which in a manner 
 supplement the judicial criminal statistics. 
 
 1 Some might feel that I should have included Connecti- 
 cut in this group of states. Yet so far as I am able to 
 perceive, the statistics published in the report of the 
 comptroller of public accounts are merely judicial statis- 
 tics and do not include criminal statistics. 
 
CHAPTER V 
 
 STATE CRIMINAL STATISTICS — {Continued) 
 
 PRISON 
 
 As was indicated in Chapter I, prison criminal 
 statistics differ from judicial criminal statis- 
 tics mainly in that they are less comprehen- 
 sive. To be sure, they differ, also, in the kind 
 of subsidiary information that they may yield ; 
 for example, the manner of carrying out the 
 sentence is a fact which might be obtained 
 from prison criminal statistics but not from 
 those compiled from judicial records. The de- 
 tails of collection, such as time, place, and 
 authorities are also all usually dissimilar, and 
 yet, after all, the main difference is, as stated 
 in the opening sentence, one of comprehen- 
 siveness. This fact is due, of course, to the 
 circumstance that prison criminal statistics 
 take no account of any but those committed 
 to the various penal or correctional institu- 
 tions. 
 
 In the United States, the real power to col- 
 lect prison criminal statistics lies, as it does 
 with the collection of judicial criminal statis- 
 tics, with the separate states. A number of 
 these have collected prison statistics of crimi- 
 
70 CRIMINAL STATISTICS 
 
 nals and a very few prison criminal statistics. 
 This chapter includes the history of those that 
 have collected prison statistics of criminals, 
 for I am confident that the evidence points to 
 a development of these into prison criminal 
 statistics, and the omission of the work of the 
 states along this line would, I believe, consti- 
 tute a distinct loss to the value of this study. 
 In the collection of prison criminal statis- 
 tics or prison statistics of criminals, the states 
 have uniformly followed two policies. One 
 plan has been to have the sheriffs and keep- 
 ers of state correctional and penal institutions 
 send in returns to the secretary of state, who 
 has been required to publish an abstract of 
 these returns for the benefit of the legislature. 
 The other plan is for the same local officers 
 to send their returns to a state board of chari- 
 ties and corrections, or board of similar name, 
 which makes such use of them- as it may 
 desire, usually publishing a summary of the 
 results. These boards are all of about the 
 same character. They are composed of from 
 three to nine members, serving with or with- 
 out pay for a term of years, and have either 
 the control and management of state institu- 
 tions and an investigating power over county 
 institutions or are devoid of all control but 
 empowered to investigate and report. Some- 
 times the charity group of institutions is 
 
STATE CRIMINAL STATISTICS 71 
 
 separated from the penal and correctional 
 group and the supervision and control of the 
 two vested in two distinct boards. In this 
 case, the criminal statistics or statistics of 
 criminals would, of course, be collected by the 
 board dealing with that class of institutions.^ 
 The histories which follow have been made 
 as brief as possible, and only the reports of 
 importance are touched upon. 
 
 The State Board of Charities and Correc- 
 tion was created in 1895.^ As organized under 
 
 the law, the board consisted of „ _ . . 
 o • i 1 !• ^ New Hampshire 
 
 nve persons, appointed tor nve 
 
 years and receiving no compensation. It con- 
 stituted a committee to which the overseers 
 of the poor and the county commissioners re- 
 ported concerning minors cared for by them. 
 In 1897, its powers and duties were greatly 
 enlarged.' It was given supervision over the 
 whole field of charitable and correctional 
 institutions with the exception of the state 
 prison and the asylum for the insane at Con- 
 cord. The board has made biennial reports 
 since 1896, and all but the first one contain 
 
 ' Prof. Frank A. Fetter discusses the work of these 
 boards in a report read at the 36th National Conference 
 of Charities and Corrections, held at Buffalo, N. Y., 
 June, 1909. 
 
 ' Laws of 1895, ch. 116. 
 
 ' Laws of 1897, ch. 91. 
 
72 CRIMINAL STATISTICS 
 
 statistics of the inmates of the various institu- 
 tions, relating chiefly to the number found 
 confined on a certain day of the year. The 
 offenses are not given except for those of per- 
 sons committed to the house of correction. 
 The board has made several recommendations 
 in its reports, and if these are adopted the sta- 
 tistics will be more valuable. 
 
 In addition to the statistics collected by the 
 State Board of Charities, there are the returns 
 of the sherififs which are sent to the secretary 
 of state. These, as well as the reports of the 
 solicitors which were mentioned in the preced- 
 ing chapter, have not been obtainable and 
 consequently I am unable to speak of their 
 nature. 
 
 In 1896, the legislature repealed a law 
 which had stood on the statute books since 
 1860. This law had required the 
 sheriffs to transmit to the sec- 
 retary of state various facts regarding every 
 person committed to jail. The offence was to 
 be specified, and therefore these may be con- 
 sidered as prison criminal statistics. The 
 reports were not published but filed in the 
 secretary's office. 
 
 Though there are two boards in charge of 
 state institutions, no statistics for the state 
 as a whole are given. 
 
STATE CRIMINAL STATISTICS 73 
 
 In 1834, a law passed the legislature which 
 required the keeper of each jail and house of 
 
 correction to make semi-annual ,_ . ^, 
 , ,, 111 Massachusetts 
 
 returns to the mayor and aider- 
 men of the city of Boston, or to the county 
 commissioners of the county, in which the 
 jail or house of correction happened to be lo- 
 cated.^ The commissioners sent the reports 
 to the secretary of the commonwealth, who 
 made an abstract of these and presented it to 
 the House or the Senate as the case might be. 
 These abstracts cover the years 1833 to 1839. 
 Their nature can be best understood by quot- 
 ing from the abstract of 1839, prepared by 
 Mr. J. P. Bigelow, then secretary of the com- 
 monwealth. He says: "The abstract, as in 
 several preceding years, has been made princi- 
 pally from the returns of the keepers, there 
 being no returns (as required in the Revised 
 Statutes, chapter 143, section 33) from the 
 Inspectors, in most of the counties. From 
 Franklin and Hampshire, no returns have 
 been received either from Keepers or Inspec- 
 tors. It is understood that in the county of 
 Hampshire, during most of the year, there did 
 not exist a legal board of County Commis- 
 sioners, which will account for the omission 
 of returns from that county. 
 " There being no special form of returns con- 
 * Laws of 1834, ch. cli, sec. 15. 
 
74 CRIMINAL STATISTICS 
 
 cerning prisons prescribed by law, it will be 
 observed by the Abstract, that the informa- 
 tion derived from the returns is, in some cases, 
 much more full than in others. It is also im- 
 portant to notice that the returns as at present 
 made exhibit a much larger aggregate of 
 crime than the facts, if properly set forth, 
 would warrant, for the number stated as com- 
 mitted to the houses of correction of course 
 include a larger portion of those who were in 
 the first instance committed to the common 
 jails." 1 
 
 In 1840, the secretary of the commonwealth 
 was required to furnish the sheriffs and over- 
 seers blanks for their returns.^ Though this 
 was a change in the right direction, the gen- 
 eral nature of the statistics, owing to incom- 
 pleteness and reduplications, was not thereby 
 very much improved. Moreover, returns 
 from the state prison and from the state re- 
 form school were not Included. 
 
 In 1863, the Board of State Charities was 
 established with the power to investigate and 
 supervise the whole system of public chari- 
 table and correctional institutions of the 
 state.' A year later, the secretary of this 
 board was by law required to furnish sched- 
 
 • See House Documents, 1840. Document no. 39, p. 3.' 
 
 * Laws of 1840, ch. 15, sees. 1-2. 
 ' Laws of 1863, ch. 240. 
 
STATE CRIMINAL STATISTICS 75 
 
 ules to the keepers of prisons and workhouses, 
 including the state prison and the house of 
 industry, reformatory and correction in Bos- 
 ton, and to the sheriffs.^ The work of the sec- 
 retary of the commonwealth in collecting 
 prison criminal statistics was thus given to 
 the Board of State Charities. The statistics 
 collected by the board were the most compre- 
 hensive yet obtained. A great many ques- 
 tions were, however, asked which by their 
 very nature precluded an answer. The tables 
 were not well summarized and it is diflSeult to 
 make use of the data found in them. 
 
 In 1870, three commissioners of prisons were 
 appointed to supervise the jails and houses of 
 correction.^ The jailers and keepers now sent 
 their returns to the commissioners instead of 
 to the Board of State Charities. Two years 
 later, these officials were again directed to 
 send their returns to the Board of State Char- 
 ities, provision being made to give the com- 
 missioners access to them.' A rettu"n to the 
 previous policy was made in 1874 when the 
 keepers of all penal institutions were required 
 to keep records and to make returns to the 
 commissioners.* It was not, however, until 
 1877 that the act requiring returns to be made 
 
 » Laws of 186Jf, ch. 307, sec. 1. » Laws o/ 1870, ch. 370. 
 > Laws o/ 1872, ch. 24, sec. 1. 
 « Laws of 1874, ch. 264, sec. 1. 
 
76 CRIMINAL STATISTICS 
 
 to the Board of State Charities was finally 
 repealed.^ Although by the act of 1874 it 
 would seem that the prison commissioners had 
 the power to require returns from all institu- 
 tions, yet they never included in their report 
 returns of the state prison. The tables are not 
 well constructed and for a comparative study 
 of criminality are of little use. 
 
 In 1879, the Board of Prison Commissioners 
 was created, and all reports and returns of the 
 warden of the state prison, of the reformatory 
 for women, of jailers, keepers of houses of 
 correction, county commissioners or directors 
 were henceforth made to it.* This board was 
 supplanted in 1901 by a new one bearing 
 the same name and having the same powers 
 and duties. It consists of five persons, the 
 chairman being the only paid member. These 
 two boards have collected and tabulated 
 prison criminal statistics, and it is not too 
 much to say that they are the best of all such 
 state statistics. 
 
 A board of State Charities and Correction 
 consisting of six persons and having charge 
 _.,_., of certain specified state insti- 
 tutions was created in 1869.^ 
 It was reconstituted in 1877, at which time 
 
 » Laws of 1877, ch. 120, sec. 1. ' Laws of 1879, ch. 294. 
 ' Laws of 1869, May session, ch. 814, pp. 1-6. 
 
STATE CRIMINAL STATISTICS 77 
 
 the number of members was increased to nine 
 and its powers further extended,^ Its reports, 
 made annually to the General Assembly, have 
 included since about 1877 criminal statistics 
 of institutions and of jails, those for jails re- 
 lating to all those found in confinement at a 
 certain time. Needless to say they are worth- 
 less for statistical studies. 
 
 From 1843 to 1861, the county commis- 
 sioners reported regularly to the legislature 
 on county jails. Under the act 
 of 1861, the secretary of state 
 was required to compile an abstract of these 
 returns. The number of those found in prison 
 on a certain day of the year and the number 
 of those committed during the year are both 
 given in the secretary's abstracts. In addi- 
 tion, various facts are recorded of those com- 
 mitted during the year, such as sex, age, color, 
 and offense, although no distinction seems to 
 have been made between those convicted in 
 the courts and those awaiting trial. 
 
 This state also has a State Board of Chari- 
 ties, created in 1873.^ It consists of five per- 
 sons appointed for four years, and its chief 
 
 » Public Laws of 1877, ch. 603. 
 
 ' Laws of 1873. It was called Board of Charities in this law, 
 and the name was changed to the present one by the law of 
 1884 (ch. 77, sees. 1-6). 
 
'78 CRIMINAL STATISTICS 
 
 duty is the inspection of penal and charitable 
 institutions. The biennial report of this board 
 gives a few statistics of those found in state 
 institutions on day of visit and some of the 
 later reports give a summarized statement of 
 commitments to the jail, though not by crimes. 
 
 The State Commission of Prisons was cre- 
 ated in 1895. It consisted of three persons, 
 , serving for four years, and 
 
 having as its object the inspec- 
 tion of all institutions used for the detention 
 of sane adults charged with or convicted of 
 crime or detained as witnesses or debtors.-^ 
 All such institutions were required to report 
 certain facts regarding the inmates received 
 and confined therein, but such data as were 
 called for are not sufficient to form the basis 
 for prison criminal statistics. A very slight 
 change in the law would remedy this, and such 
 development may be looked for in the future. 
 
 An attempt to obtain prison criminal sta- 
 tistics seems to have been made in 1883 when 
 
 New Jersey ^ ^^^ "^^^ passed creating the 
 Council of State Charities and 
 Correction.^ Section 9 of this act required the 
 warden, physician, steward or other officer in 
 charge of any institution of charities or cor- 
 1 Laws of 1895, ch. 1026. ' Law of March 23, 1883. 
 
STATE CRIMINAL STATISTICS 79 
 
 rection to keep an exact register (the form of 
 which was to be prescribed by the council) of 
 each person in his charge and to send to the 
 clerk of the council an abstract of the same. 
 This board attempted to secure criminal sta- 
 tistics from each county, but owing to a defect 
 in the law it was unable to obtain any but the 
 most meagre statistics. Its third report is the 
 last one that I have found, and I have reason 
 to believe that the law requiring these statis- 
 tics was thereafter not enforced.* 
 
 The movement to collect prison criminal 
 statistics began in 1847. In this year an act 
 was passed requiring sheriffs, _ . 
 inspectors or persons in charge 
 of any jail or penitentiary to make detailed 
 reports to the secretary of state concerning 
 the institutions over which they had charge.^ 
 This law remaiued on the statute books until 
 1872, but to the best of my knowledge it was 
 never enforced.' Three years after the crea- 
 
 1 See Laws of 1897, ch. 124, p. 179, sec. 2, which says, 
 "Nothing in this act contained shall be held to revive an act 
 entitled : An act providing for the record and report of Jail 
 statistics." 
 
 ' Laws qf 18i.7, no. 131. 
 
 ' Diligent search has failed to find the abstracts of these 
 reports which the secretary of state was required to make. 
 The early reports of the Board of Public Charities make much 
 of the difficulty of securing statistical information and por- 
 tray the beginning of such work under their regime. 
 
80 CRIMINAL STATISTICS 
 
 tion of the Board of Public Charities, the law 
 directed that the reports be sent to it, and in 
 the report of this board for 1873 certain prison 
 statistics of criminals appear.^ From that 
 time on statistics of a similar nature have 
 been published in the annual report of this 
 board. The report aims to give the number 
 committed to all penal and correctional insti- 
 tutions and the number found on a certain 
 day of the year. The statistics of the peniten- 
 tiaries and of the Huntington reformatory 
 give the kind of crime for which the individuals 
 were committed, but the sole distinction 
 made along this line in the statistics of the 
 county prisons and of the workhouse is that 
 between felonies and misdemeanors — a dis- 
 tinction of little use to the student. 
 
 A Board of Charities and Correction for 
 Virginia was established in 1908.'^ It con- 
 -^. . . sists of five members appointed 
 
 for a term of five years. It is 
 strictly advisory in its nature but has the 
 power to secure statistics. In fact the law 
 says: "that said board shall collect and pub- 
 lish statistics regarding the dependent, defec- 
 tive and delinquent classes both in and out of 
 
 ' Laws of 1872, no. 84. 
 
 ' Law approved March 13, 1908, being eh. 276 of Acts of 
 1908. 
 
STATE CRIMINAL STATISTICS 81 
 
 institutions, within the state, and such other 
 data as may be of value." I have not yet had 
 the opportunity to examine the report which 
 the board has recently made and thus am un- 
 able to say whether this provision of the law 
 has been carried out. 
 
 A Board of Public Charities was created by 
 
 the legislature which held its session in 1868- 
 
 69. It consisted of five mem- „ ,, ^ ,. 
 , 1 X J i! . • X North Carolina 
 
 bers elected tor a term of two 
 
 years by the General Assembly. Like so many 
 of the state boards, its duty was to investigate 
 and supervise the whole system of charitable 
 and penal institutions,^ Detailed information 
 regarding the inmates of prisons, poorhouses, 
 etc., was required to be sent to the Board of 
 Public Charities by the board of commission- 
 ers of each county. The statistics in the re- 
 ports of the Board of Public Charities are 
 merely prison statistics of criminals found in 
 confinement and are not complete. 
 
 The Board of State Charities was organized 
 under the act of April 17, 1867. Five persons 
 constituted the board, and it pos- . . 
 
 sessed the general power of in- 
 vestigation and supervision of state chari- 
 table and correctional institutions. This 
 
 1 Laws of 1868-69. ch. 170. 
 
82 CRIMINAL STATISTICS 
 
 board lasted until 1872. During this time it 
 made five annual reports whicli contain sta- 
 tistics of jail prisoners for the entire year and 
 for the number in prison on a certain day of 
 the year. The board was revived by the act 
 of April 3, 1876, and as at present organized 
 consists of six members serving without pay 
 for a term of three years. The reports which 
 this board has annually made to the legisla- 
 ture have universally contained statistics of 
 jail prisoners, number received during year 
 and found on certain dates in confinement. 
 Returns from certain institutions are also 
 found from time to time, but none of the sta- 
 tistics can be considered as true prison crimi- 
 nal statistics. 
 
 A prison census is taken each year by the 
 
 Board of State Charities which was created 
 
 , ^. in 1889.^ This board is com- 
 
 Indiana j » • i 
 
 posed of six members serving 
 
 for three years and has had from its beginning 
 the power to obtain statistical information 
 from sheriffs and officers in charge of other 
 penal or correctional institutions. It has had 
 great difficulty in obtaining complete returns, 
 and such as have been secured are acknow- 
 ledged to be inexact. A few of the offenses, as 
 intoxication and vagrancy, are enumerated in 
 
 1 Laws of 1889, ch. 37. 
 
STATE CRIMINAL STATISTICS 83 
 
 some of the later reports of the board, but 
 in general the number found constitutes the 
 extent of the information. 
 
 In 1872 a law was passed entitled "An 
 act to secure uniform and reliable statistics 
 
 concerning the dependent and _,. . 
 
 • . , , J ii • Illinois 
 
 crinunal classes, and then- 
 treatment in state and county institutions." ^ 
 This law required, among other things, that 
 the keepers of state institutions and sheriflFs 
 should send reports to the Board of State 
 Commissioners of Public Charities which had 
 been organized in 1869. '^ The board, it seems, 
 had at the beginning been impressed with the 
 need of statistical information, and the law of 
 1872 was, no doubt, the result of its eflforts.* 
 The law, however, was of little avail, and, in- 
 deed, the criminal statistics of prisoners in 
 jail published in their first report before the 
 passage of the law of 1872 were better than 
 any offered to the public afterwards, the 
 later reports containing only a few scattered 
 statistics of jail prisoners. 
 
 In 1909 a law was passed which brought 
 about a great change in the management of 
 the penal and charitable institutions of the 
 state.^ There are now two boards known re- 
 
 1 Laws of 1872, p. 751. = Law of 1869, approved April 9. 
 ' See First Biennial Report, p. 173. 
 * Law approved June 15, 1909, Laws of Illinois, Forty- 
 Sixth General Assembly, 1909, pp. 103-123. 
 
84 CRIMINAL STATISTICS 
 
 spectively as the Charities Commission and 
 the Board of Administration. Both were em- 
 powered by the law to collect statistics. An 
 interesting fact and one that bodes much good 
 for the statistics of that state for the future is 
 the appointment of Dr. Frederick H. Wines as 
 statistician for the Board of Administration. 
 
 This state has attempted since 1873 to 
 secure prison criminal statistics for the entire 
 . state. In that year, a law 
 
 passed the legislature making 
 it incumbent upon the sheriffs or jailors to 
 send to the secretary of state detailed infor- 
 mation regarding those committed to their 
 care.^ The statistics relate to the number 
 found confined on a certain day of the year, 
 the number committed during the year, and 
 the average number. The only reference to 
 crimes is the distinction made between high 
 crimes and minor offenses. The abstracts 
 made by the secretary of state appear in the 
 Joint Documents, and the Board of Correction 
 and Charities has from time to time included 
 in its reports excerpts from these, giving 
 at the same time commitments to the state 
 institutions. A few changes in the forms sent 
 out by the secretary of state would make 
 these statistics yield valuable results. 
 
 » Law8 of 1873, no. 167. 
 
STATE CRIMINAL STATISTICS 85 
 
 This state has had four state boards, all of 
 which relate in some way to the history here 
 attempted. In 1871, the State ^. 
 
 Board of Charities and Reform 
 was created.^ It had full power of supervision 
 and investigation over state institutions and 
 investigating power over county or municipal 
 institutions of a penal, correctional, or chari- 
 table character. This board was reorganized 
 in 1876,^ and in 1881 it was relieved of all 
 duty in regard to six state institutions by the 
 creation of the State Board of Supervision of 
 Wisconsin Charitable, Reform and Penal In- 
 stitutions.^ Both boards were abolished in 
 1891, and their places taken by a new board 
 called the State Board of Control of Reforma- 
 tory, Charitable and Penal Institutions, which 
 at the present time possesses all the powers 
 and duties of the two former.* It is difficult 
 to describe the statistics which these various 
 boards have published. A shifting, faltering 
 plan has been pursued, but in general the 
 attempt seems to have been made to se- 
 cure statistics of those received during the 
 year and of those found in prison on a 
 certain day. Seldom are the crimes of the 
 prisoners enumerated, and the illogical and 
 irregular tabulation deprives the statistics of 
 
 » Laws of 1871, ch. 136. » Laws of 1881, ch. 298. 
 
 » Laws of 1876. ch. 414. * Laws of 1891, ch. 221. 
 
86 , CRIMINAL STATISTICS 
 
 any usefulness they might otherwise have 
 possessed. 
 
 Statistics designed to show the movements 
 of population in jails and other penal institu- 
 _^. tions have been collected since 
 
 Minnesota ^gg^ The State Board of Cor- 
 
 rections and Charities began this work, and 
 after its abolition in 1901 the new Board of 
 Control of State Institutions continued the 
 work.^ A census of the prison population is 
 taken at various times during the year with 
 the idea that this method will show the shift- 
 ing character of this class. The offenses for 
 which committal was ordered are not specified. 
 
 A few jail and penitentiary statistics of 
 prisoners have been collected by the Board of 
 j^. . Charities and Correction which 
 
 was organzied in 1897.^ In 
 some cases the sentences for murder and lar- 
 ceny are given. In the main, however, the at- 
 tention of the board is focused on the sex, color 
 and number of those found in confinement.' 
 
 The Board of State Charities, organized 
 in 1895,* has collected a few prison statistics 
 
 1 Laws of 1883, ch. 127; 1901, ch. 122. ^ Laws of 1897, p. 45. 
 ' The report of the board for the biennial period 1899- 
 1901 includes an article entitled "Is Crime Increasing?" 
 * Laws of 1896, ch. 193. 
 
STATE CRIMINAL STATISTICS 87 
 
 of criminals. The number on hand, the num- 
 ber received during the year, the sex and 
 race of the prisoners are some _ 
 of the facts set forth. I have 
 no report since 1903, but it is doubtful if the 
 character of the statistics has since xmdergone 
 any change for the better. 
 
 It is now the duty of sheri£Fs to make out a 
 jail report and send it to the secretary of state, 
 who in turn must communicate 
 these reports to the Legislative 
 Assembly at its annual session.^ As I have 
 not been able to obtain access to the report of 
 the secretary, I am not in a position to speak 
 of the nature of these statistics. 
 
 This state has a State Board of Charities 
 
 and Reform* and its reports have at times 
 
 contained a very few statistics _ _ . 
 
 ... , . • i-i .• Montana 
 
 of state and county institutions 
 
 giving the admissions, discharges, etc., during 
 
 the year. 
 
 The State Board of Charities and Reform 
 was created by that session of the legislature 
 which was held in 1890-91.' The ^yoniing 
 
 act which organized this board 
 
 » See Revised and Annotated StcUutea of OUakoma, 1903, 
 ch. 68, art. 24, sec. 620. 
 
 « Created in 1893. » Laws, 1890-91, ch. 37. 
 
88 CRIMINAL STATISTICS 
 
 also required sheriffs to report to it regarding 
 persons confined in the county jails. As the 
 board had definite charge of the various state 
 penal and reformatory institutions it was in 
 a position to obtain complete prison criminal 
 statistics. The first reports of the board gave 
 criminal statistics of institutions and a mere 
 enumeration of prisoners in jail on a certain 
 day of the year. The later reports have given 
 up the jail statistics entirely. 
 
 The reports of the State Board of Charities 
 and Corrections (created in 1891)^ have at 
 times contained tables designed 
 to show the movements of 
 population both in state and county institu- 
 tions. The jail censuses, taken at various 
 times during the year, were omitted from some 
 of the reports and only those for state insti- 
 tutions given. 
 
 Since 1903, California has had a State 
 Board of Charities and Correction.^ This 
 _ ... . - board presents in its biennial 
 
 report tables designed to show 
 the movements of population in the state and 
 county penal and correctional institutions. 
 
 » Laws, 1891, p. 327. 
 
 * Statutes and Amendments to the Codes of California, 1903, 
 chaptpr ccclxiii. 
 
STATE CRIMINAL STATISTICS 89 
 
 Although several facts are recorded regarding 
 those in jails the important question regarding 
 crime for which committed has been omitted 
 from the category. 
 
 As in the preceding chapter, the facts of the 
 situation are shown in tabular form on the 
 next page. 
 
 It will be noticed that Massachusetts was 
 the pioneer in this work, and from long ex- 
 perience has been able to evolve an efficient 
 scheme of organization which has produced 
 good results. The trouble with most of the 
 states is that the emphasis has been put on 
 the accommodations for the prisoners, the 
 aim of the boards evidently being to gain 
 some idea of the population with the idea of 
 providing suitable structures. To say, how- 
 ever, that they have never been concerned 
 with the study of criminality is to waive all 
 the evidence offered by the essays on crimi- 
 nality which are to be found in the reports of 
 the various boards. They are all groping for 
 light on the subject of statistics and the need 
 for some centralized leadership is evident. 
 
90 
 
 CRIMINAL STATISTICS 
 
 is 
 
 fPj H M h M 
 O O O O O 
 
 Iu u u t-* 
 o o o o 
 
 I 
 
 oJOO 
 
 o o o 
 
 ii 
 
 Mg 
 •23 
 
 Q/d C t4 »j 
 
 S b a S << 
 
 M C8 2 o o 
 ogMM 
 
 g O.S an 
 
 h h h »4 l4 N 
 
 O O O O O O 
 
 H H H ^ h H 
 
 CO n Q ri <^ a 
 
 O O O o O O 
 
 pqpq P3PqfflfQ 
 
 O O O O O O 
 
 o o o o o o 
 
 Q) 0} 0) <U V V 
 
 a 
 
 si 
 
 ^cB oj 5 00 » <s a) «o£SS o 
 
 00 IW -O » - " 
 
 ■g* p.fl gp.'g p, 
 
 ro f#l (11 K wi n- ™ — 
 
 s 
 
 ^ 
 
 S ■2„ .S 
 
 S ^0 b. b. O 
 
 5 o a > ^ a 
 
 S ja g £« S 
 
 _ cgwoot- bi 
 O CO OOQOCO oo 
 
 I 
 if g.s I 
 
 0„ 
 
 •Sgo 
 oHfl S 
 
 ""Hi 
 
 Sfl t8 fe rt 
 
 »«•- o o 
 
 .2o 
 
 ■Sja <^ ^ 
 
 .o o o o 
 
 DO (D so 00 
 
 |£a |5 |u 'Si 
 'h 'C 'C 'fl 
 
 i O n ffi n ri 
 
 JOUUQ 
 
 I^OO O O 
 
 . t- fcM t, »H 
 
 >>rt d c8 d 
 
 »g o o o o 
 
 V 0) 01 « o 
 
 s u rt rt rt rt 
 
 
 _co oi oi o O) J, oi eS 
 
 cow 0000OS&®000» 
 
 ^^ -nlhM 
 
 - cu ^ c 
 
 PiSlI I 
 
 §2 § ^SS ago 
 
CHAPTER VI 
 
 PLAN FOR THE REORGANIZATION OF CRIMINAL 
 STATISTICS IN THE UNITED STATES 
 
 JjEFORE submitting a plan for the reorgani- 
 zation of criminal statistics in the United 
 States, I wish to present briefly the facts of 
 the present situation. 
 
 The federal census bureau has recently 
 changed its plan of collecting criminal statis- 
 tics. It will be remembered that prior to the 
 last census such criminal statistics as it ob- 
 tained were mostly gathered at the time of 
 the regular population enumeration and re- 
 lated mainly to those found confined in penal 
 and correctional institutions on a certain day 
 of the year. The limiting in 1900 of the decen- 
 nial work to four main lines of inquiry made 
 necessary a change in the manner of collecting 
 criminal statistics. In 1903, schedules were 
 sent out to the keepers of jails, penitentiaries 
 and reformatories, and this time the emphasis 
 was laid upon commitments during the calen- 
 dar year. The intention is, I believe, to pur- 
 sue the same policy, in the future. The next 
 report of this nature will then appear some 
 time after the main work of the 1910 census 
 
92 CRIMINAL STATISTICS 
 
 is completed. The census bureau has also 
 started to collect judicial criminal statistics. 
 Certain counties in various states have been 
 selected and clerks from the census office sent 
 there to fill out prepared schedules by refer- 
 ence to the criminal dockets. The first report 
 will be for 1906. What plans for the future of 
 this work are being contemplated I am unable 
 to say. 
 
 As has been shown, many of the states have 
 collected judicial criminal statistics, a very 
 few prison criminal statistics, and a number 
 prison statistics of criminals. The following 
 table will give some idea of the nature and 
 extent of this work of the states : 
 
 TABLE SHOWmO EXTENT AKD NATURE OP THE 
 WORK OP THE STATES IN THE COLLECTION OP 
 CRIMINAL STATISTICS AND PRISON STATISTICS 
 OP CRIMINALS 
 
 
 Collects 
 
 Collects 
 
 Collects 
 
 State 
 
 Judicial 
 
 Prison 
 
 Prison 
 
 Criminal 
 
 Criminal 
 
 Statistics of 
 
 
 Statistics 
 
 Statistics 
 
 Criminals 
 
 Maine 
 
 Yes 
 
 
 
 New Hampshire 
 
 1 
 
 1 
 
 Yes 
 
 Vermont 
 
 No 
 
 Yes 
 
 No 
 
 Massachusetts 
 
 Yes 
 
 Yes 
 
 No 
 
 Rhode Island 
 
 No 
 
 Yes 
 
 No 
 
 Connecticut 
 
 No 
 
 Yes 
 
 No 
 
 * Nature not known as data were not at hand. 
 
REORGANIZATION OF STATISTICS 93 
 
 
 Collects 
 
 Collects 
 
 Collects 
 
 State 
 
 Judicial 
 
 Prison 
 
 Prison 
 
 Criminal 
 
 Criminal 
 
 Statistics of 
 
 
 Statistics 
 
 Statistics 
 
 Criminals 
 
 New York 
 
 Yes 
 
 No 
 
 Yes 
 
 New Jersey 
 
 No 
 
 No 
 
 No 
 
 Pennsylvania 
 
 Yes 
 
 Yes 
 
 No 
 
 Virginia 
 
 No 
 
 1 
 
 1 
 
 North Carolina 
 
 Yes 
 
 No 
 
 Yes 
 
 South Carolina 
 
 Yes 
 
 No 
 
 No 
 
 Florida 
 
 Yes 
 
 No 
 
 No 
 
 Ohio 
 
 Yes 
 
 No 
 
 Yes 
 
 Indiana 
 
 Yes 
 
 No 
 
 Yes 
 
 Illinois 
 
 No 
 
 No 
 
 Yes 
 
 Michigan 
 
 Yes 
 
 Yes 
 
 No 
 
 Wisconsin 
 
 Yes 
 
 No 
 
 Yes 
 
 Minnesota 
 
 Yes 
 
 No 
 
 Yes 
 
 Iowa 
 
 Yes 
 
 No 
 
 No 
 
 Missouri 
 
 No 
 
 No 
 
 Yes 
 
 North Dakota 
 
 Yes 
 
 No 
 
 No 
 
 Tennessee 
 
 No 
 
 No 
 
 Yes 
 
 Alabama 
 
 Yes 
 
 1 
 
 No 
 
 Louisiana 
 
 Yes 
 
 No 
 
 No 
 
 Texas 
 
 Yes 
 
 No 
 
 No 
 
 Oklahoma 
 
 No 
 
 No 
 
 No 
 
 Montana 
 
 Yes 
 
 No 
 
 Yes 
 
 Wyoming 
 
 No 
 
 No 
 
 Yes 
 
 Colorado 
 
 No 
 
 No 
 
 Yes 
 
 Utah 
 
 Yes 
 
 No 
 
 No 
 
 Nevada 
 
 Yes 
 
 No 
 
 No 
 
 Idaho 
 
 Yes 
 
 No 
 
 No 
 
 Washington 
 
 Yes 
 
 No 
 
 No 
 
 Oregon 
 
 Yes 
 
 No 
 
 No 
 
 California 
 
 Yes 
 
 No 
 
 Yes 
 
 . * Nature not known as data were not at hand. 
 
94 CRIMINAL STATISTICS 
 
 The character of the statistics in each state 
 has been noted in the histories given above. 
 It is sufficient here to state that the statis- 
 tics are, from the point of view of a student 
 of criminology almost without exception bad. 
 The reasons for this state of affairs are, 
 broadly speaking, four. It must be remem- 
 bered that the states had a double purpose 
 in view in the collection of these statistics. 
 They were to furnish information for admin- 
 istrative purposes as well as to give an index 
 of the nature and extent of criminality in the 
 community. The administrative purpose was 
 perhaps less evident in the judicial criminal 
 statistics than in the others. But these judi- 
 cial criminal statistics enabled the state au- 
 thorities to gain an oversight of the work 
 of the courts and the attorneys, though the 
 compilation and analysis seems quite gener- 
 ally to have been made with the sociological 
 idea uppermost. Likewise the prison criminal 
 statistics or prison statistics of criminals were 
 to afford information to the state officers, or 
 to state boards, of the needs of prisons, jails 
 and reformatories for money, buildings, etc., 
 as well as to give information on the state of 
 criminality. As the first reason, then, for the 
 bad quality of the state criminal statistics, I 
 would give the presence of two purposes in 
 their collection and the non-recognition of 
 
REORGANIZATION OF STATISTICS 95 
 
 the fact that this circumstance rendered nec- 
 essary a plan of procedure somewhat dififer- 
 ent from one required where the success of 
 but one purpose is sought. A second cause 
 may be found in the ignorance on the part of 
 those charged with the collection and analysis 
 of the statistics of the principles and methods 
 of statistical science. This ignorance is excus- 
 able since until recently very little opportu- 
 nity for training along these lines existed in 
 this coimtry. A proper understanding of the 
 task before them would have made possible 
 the elimination of many grievous errors which 
 cannot at this late date be rectified. The atti- 
 tude of officials may be given as a third cause. 
 The work has been done, as so much official 
 work is regularly done, not so much for the 
 sake of obtaining trustworthy and valuable 
 results as to carry out the letter of the law. 
 This has resulted in a lax enforcement of the 
 law and incomplete returns. A fourth cause 
 is to be traced to the presence of the spoils 
 system in the appointment of the members, 
 or of the secretary, of the state boards of 
 charities. These positions often mean a com- 
 fortable berth to the hangers-on of the party 
 in power, the position of secretary, at least, 
 generally carrying with it a fair salary. 
 
 This, then, is the situation, and without 
 some change in the present organization there 
 
96 CRIMINAL STATISTICS 
 
 is little hope of improvement for some time to 
 come. What is needed, if this country is to 
 have criminal statistics comparable in value 
 to those of European nations, is a reorganiza- 
 tion which shall be, at the same time, a revivi- 
 fying force. The plan for reorganization which 
 I here submit is not radical, nor even new, 
 since it is one already in force in another field 
 of statistical work. 
 
 From 1850 to 1890 inclusive, much the 
 same situation existed as regards mortality 
 statistics. At each census, statistics of those 
 who had died during the year preceding were 
 collected by the enumerators at the time of 
 the regular enumeration. The unquestioned 
 inaccuracy of these statistics brought about a 
 change of method in 1900. 
 
 The power to collect mortality statistics 
 rests with the individual states and cities. 
 Many of these had from time to time enacted 
 provisions on this subject. Without reference 
 to any uniform plan these inevitably differed 
 widely in their requirements. Some few of the 
 states and cities had succeeded in obtaining 
 adequate results, but most of these attempts 
 were from the beginning doomed to failure 
 because of a disregard for the statistical prin- 
 ciples involved.^ All this the census oflSce 
 
 ' See Special Report of the Census on Mortality for 1900- 
 04, p. ix, column i. 
 
REORGANIZATION OF STATISTICS 97 
 
 knew, and it elaborated a plan the main fea- 
 tures of which I shall here try to state. Briefly 
 put it was to join forces with the states to 
 secure mortality statistics of such a nature 
 that they would both satisfy the requirements 
 of the state boards and organizations and at 
 the same time give to the census office the 
 requisite data for the compilation of mor- 
 tality statistics comparable with those of other 
 nations. Needless to say a good deal of pre- 
 liminary work and study was necessary before 
 the plan could be put into operation. Congress 
 passed a resolution asking for the cooperation 
 of the states in the work; and the census office, 
 after trying to determine the reasons for fail- 
 ure on the part of the states and cities, worked 
 out the details of a system which it hoped 
 would meet with success. This included 
 among other things a standard form for re- 
 porting details and the adoption of the inter- 
 national classification of the causes of death. 
 It issued pamphlets showing exactly what 
 must be done to procure mortality statistics 
 of value. It also sought to secure the coopera- 
 tion of all societies and organizations which it 
 thought might be interested in the work. The 
 state legislatures were called upon to act; and 
 the registration area, a district comprising 
 those states and cities which "possess records 
 affording satisfactory data in necessary de- 
 
98 CRIMINAL STATISTICS 
 
 tail," has during the period 1900 to 1909 
 grown from ten to eighteen states. It is the 
 hope and expectation of the census office that 
 this registration area will be gradually ex- 
 tended through the admittance from year to 
 year of other states which shall have in the 
 mean time attained the required standard of 
 excellence in the collection of mortality sta- 
 tistics, either through the passing of new and 
 better laws or the stricter enforcement of 
 existing ones. 
 
 From the brief sketch of mortality statistics 
 in this country which I have just given, the 
 close analogy between the organization of 
 mortality statistics as it existed in 1900 and 
 the present organization of criminal statistics 
 is, I believe, clearly apparent. The power to 
 collect criminal statistics as well as mortality 
 statistics resides in the individual states. 
 Many of these states are now attempting to 
 collect criminal statistics, and their efforts 
 are attended with the same results as those 
 which followed their endeavors to obtain mor- 
 tality statistics. A few are meeting with suc- 
 cess, but in many cases failure has been the 
 sole result of many years of wearisome trial. 
 I fail to see where the present plan of the cen- 
 sus office ofiFers any permanent solution of the 
 question. It can obtain judicial criminal sta- 
 tistics from a few states perhaps, but the 
 
REORGANIZATION OP STATISTICS 99 
 
 expense must necessarily be very large, — too 
 large to render the extension of this scheme 
 advisable. Furthermore, it means in many 
 states a duplication of work and a consequent 
 economic loss of time and money. Neither 
 does it aid in the solution of the difficulties 
 which confront the state boards. Their re- 
 ports will be based on worthless statistics, 
 and no uniformity in the matter of presen- 
 tation of facts will exist among the several 
 states. 
 
 I believe that the plan which has been so 
 successful in reorganizing the mortality statis- 
 tics of the country would work equally well 
 in the case of criminal statistics.^ I have tried 
 to show that the conditions are virtually the 
 same, and surely the success of the plan in the 
 one instance argues strongly for its adoption 
 in the other. I shall not attempt here to work 
 out the details of the plan. I have no doubt, 
 
 1 This chapter appeared in the Jowmal of the American 
 Institute of Criminal Law and Criminology for May, 1910, 
 and in the September issue of the same year (pp. 429-430) 
 will be found what seems to be a criticism by Mr. John Keren 
 of the plan of reorganization which I have outlined in this 
 chapter. I feel, somehow, that his own plan for a reorgan- 
 ization (pp. 430-431) does not diflFer materially from mine. 
 I leave that, however, for others to determine. But I am, I 
 must confess, slightly curious to know how Mr. Koren, with- 
 out the aid of a classification which he manifestly deems im- 
 possible, could weld the statistics from different states "into 
 a whole" and reduce "statements so far as need be to com- 
 mon terms." 
 
100 CRIMINAL STATISTICS 
 
 however, that the method of procedure which 
 was adopted in the case of mortaUty statistics 
 would prove equally suitable to the reorganiza- 
 tion of criminal statistics. It would certainly 
 involve a careful study of the needs of the 
 state boards of charities and of state officials 
 and of the causes of their failure to obtain 
 the statistics required. The classification of 
 crimes ought also to be undertaken, and here 
 cooperation with European authorities might 
 prove very beneficial. The help of all the so- 
 cieties and organizations for the study of the 
 dependent and criminal classes ought to be 
 enlisted, and specialists called in to draft a 
 program of work which would give the re- 
 quired results. This work would naturally 
 be done by the census office since it is the 
 only body in a position to procure the serv- 
 ice of experts. The registration area would, of 
 course, be small to start with, but I am of the 
 opinion that its growth would be more rapid 
 than that for mortality statistics has been. 
 Since the same statistics would serve both the 
 purpose of the states and that of the federal 
 government there would be no duplication 
 of work. Uniformity in the reports from the 
 states would be secured, and the influence of 
 the permanent census bureau ought to make 
 for an ever increasing standard of excellence. 
 
APPENDIX 
 
 ON THE "iNCBEASB OF CRIME " 
 
 fee much has been said lately regarding the "increase 
 of crime " that a few words concerning the premises of 
 a judgment on this question may be of some interest. 
 
 At first thought, it would seem that the method of 
 determining an increase or decrease in criminality 
 would be analogous to that used in studying the changes 
 in mortality. The points of resemblance between crim- 
 inal and mortality statistics are many, and it is but 
 natural to suppose that the two are susceptible of the 
 same treatment. The fact of resemblance in some par- 
 ticulars, it seems to me, has quite often been responsi- 
 ble for the slight recognition given to the dissimilar 
 features. It is these that I wish to touch upon. 
 
 The increase or decrease of mortality is determined 
 by the death rate. This rate is obtained by dividing 
 the number of deaths which occur within the year by 
 the population as it stood at the middle of the year 
 and multiplying the quotient by one thousand. Varia- 
 tions in this rate indicate an increase or decrease of 
 mortahty. This method has been quite generally used 
 to determine the increase or decrease in criminality. A 
 rate has been obtained by dividing the total number of 
 crimes of a year by the population and multiplying the 
 result by one thousand. This rate has then been com- 
 pared with the rates of previous periods and judgment 
 pronounced accordingly. ■ ' 
 
102 APPENDIX 
 
 The reason that changes in mortality can be indi- 
 cated by the method which has just been outUned is 
 that the figure which represents the number of deaths 
 is the sum of phenomena which are exactly ahke in 
 their effect on mortality, or in other words it is com- 
 posed of homogeneous units. The demise of an indi- 
 vidual may be caused by scarlet fever or by appendi- 
 citis, but from the point of view of mortality statistics 
 the character of the event is in each case the same. 
 Because of this fact, the ratio of deaths to population, 
 which is what the above mentioned rate is, can well 
 serve as an index of mortality and furnish a basis for 
 the calculation of changes in mortality. 
 
 In the case of criminal statistics, the situation is 
 very different. The figure which represents the total 
 number of crimes which have occurred within the year 
 is not made up of homogeneous units. True it is that 
 each unit is called a crime, but here all homogeneous- 
 ness ends. Each fatal disease produces a death, and a 
 death is a phenomenon whose essential characteristics 
 never vary. Each criminal act is called a crime, but the 
 term crime is used to designate a great variety of acts 
 which differ widely in all their essential characteristics. 
 No one would claim that the criminality of a country 
 where one hundred murders to every one thousand of 
 population occur annually is the same as that of a coun- 
 try where one hundred larcenies to every one thousand 
 of population constitute the yearly criminal record. 
 Nor is it logical to believe that the criminality of a 
 country where the criminal record is one hundred 
 crimes to every one thousand of population, the crimes 
 consisting of fifty murders and fifty robberies, is the 
 
APPENDIX 103 
 
 same as that of a country where the record is one hundred 
 crimes to every one thousand of population, the crimes 
 in this case including fifty assaults and fifty larcenies. 
 Just as one would not say that the criminahty of the 
 two countries was the same, even though the criminal- 
 ity rate was the same, so also he would not conclude 
 that the criminality of a country had remained sta- 
 tionary if the above rates were those of two periods in 
 the history of that country. In spite of this simple 
 truth, the arguments to prove an increase or decrease 
 of criminality are often based on the comparison of 
 general criminality rates. It is against just this sort of 
 thing that I protest. It assumes that each period's 
 contingent of crimes contains the same relative propor- 
 tions of arsons, murders, larcenies, assaults, etc. The 
 assumption is one which ought not to be made without 
 good evidence of its truth. Proof should first be oflFered 
 that this is true, and then no one would question the 
 fitness of allowing such a rate to serve as an index by 
 which the increase or decrease of criminality might be 
 deduced. In case it were shown that the relative propor- 
 tions of the various crimes were not the same for two 
 periods, such an index would serve no scientific purpose. 
 The increase or decrease of any one particular crime 
 is told by that particular crime rate. If all the rates 
 for the various crimes of two periods tend to increase 
 or to decrease, then the increase or decrease in criminal- 
 ity is at once manifest. If, however, some of the rates 
 increase and some decrease all that can be said is that 
 the character of criminahty is changing. Yet the mind 
 seems to need some sort of an index which will show 
 the increase or decrease in criminality. If it were not 
 
104 APPENDIX 
 
 for the increasing use of the indeterminate sentence, 
 such an index might be obtained by dividing the sum 
 in years of all the sentences imposed during the year 
 by the population as it stood at the middle of the year. 
 In sentencing the prisoner, the judge rates the crime 
 according to its effect on society, and the only figure 
 which would be composed of homogeneous units would 
 be this sum of the sentences. The problem of the in- 
 crease or decrease of crime is one of great difliculty, and 
 a confession of ignorance is far better than an answer 
 based on unscientific reckoning. 
 
CAMBRIDGE . MASSACHUSETTS 
 U . S . A