General Considerations. 341 



accused has arisen from defective proof of the actual commission of the 

 crime. On the other hand, were the state of crime estimated only from 

 the number of persons convicted, the errors would probably be serious.* 



It would be a task of great interest to compare together the criminal 

 Statistics of different states, but as yet there are insuperable difficulties in 

 the way. As a preliminary requisite, we must establish a common stan- 

 dard of legislative morality, and an uniform activity in the executive police. 

 From differences in these respects, the statistical records of one country 

 may shew a much smaller amount of crime than those of another, where 

 perhaps no real superiority exists : and without bearing these consider- 

 ations in mind, we should fall into great errors in comparing different 

 epochs in the history of a nation, or the morality of different countries at 

 the same period. 



It appears that on the average, 72 out of 100 accused are connected 

 either by birth or residence with the department in which they are brought 

 to trial. This proportion increases to 84 in 100, in respect to crimes 

 against the person, and falls to 69 in 100 as to offences against property. 

 Not above 3 in 100, of the accused throughout France, are found to be 

 foreigners. 



To give value and certainty to Criminal Statistics, it is necessary to 

 arrange and combine facts, so as to divest them of whatever is merely ac- 

 cidental. Adopting this principle, M. G. assures us that general results 

 can be established, so uniform and constant as to preclude the supposition 

 of tlieir mere fortuitous recurrence. t In proof of this position, M. G. in 

 the first place institutes a comparison between the number of crimes com- 

 mitted within a given period, in different parts of the kingdom. For sta- 



• The criminal records of our own country confirm the opinion, that the number 

 of convictions only allords very inadequate data from which to estimate the amount 

 of crime. It is constantly found that the proportion of convictions to committals is 

 greatest in the case of oflences, of which the punishment is least severe. Thus in 

 1832, out of 107 persons committed for arson, only 35 were convicted ; and for 

 murder, out of G6 committed, only 20 were convicted ; while for larceny, 10,130 

 convictions were obtained out of 13,469 committals. In Scotland the whole number 

 of acquittals and liberations bears a considerably higher ratio to the committals than 

 in England ; and this arises from the very large number who in Scotland iire libe- 

 rated without ever being brought to trial, being very nearly 1-3 of the whole com- 

 mittals, instead of about 1-7, as in England. It is understood, however, that by far 

 the greater number of these persons are so liberated, not because they are believed 

 to be innocent, or even because there would be any difficulty in proving their guilt, 

 but under the feeling that they are sufficiently punished by their incarcei-ation previ- 

 ous to the time of trial. (Vide British Almanack and Companion for 1834, p. 66.) 



t " Cliaquc annee voit se reproduire le mSme nombre de crimes dans le mfime 

 ordre, dans les mfmes regions ; chaque classe de crimes a sa distribution particuli^re 

 ct invariable, par sexe, par Age, par eaison : tous sont accompagnes, dans dcs pro- 

 portions pareilleg, de faits accessoires, indifferens en apparence, et dont rien encore 

 n'exjiliquc le retour." — p. !). 



No. 5.— Vol. I. 2 Y 



