Finally it must be noted that the fact that there are a number of offenses which do not appear in the criminal statistics, does not mean that there are many criminals of whom the same can be said. It is the Italian school in particular that has maintained the proposition that in the struggle between the criminal and society it is the criminal that has the upper hand. This is a mistake; the criminal generally loses, and in the great majority of cases very quickly. Certainly the criminals are not punished for each crime, but the cases in which they remain unpunished are very rare. In the world of criminals itself there is no other opinion on this point. [84] Although the value of statistics in studying the etiology of crime is certain, it must not be thought that they tell us everything. [85] I pass over here numerous things which most statistics still lack, for example, a classification by motive rather than according to the technical distinctions of the penal laws, [86]—and will simply call attention to the difficulty of making international comparisons. The difference merely in the penal laws makes the comparison very difficult, and in many cases even impossible. Further, the dissimilarity of procedure, the differences in the organization of the police, etc., increase this difficulty. [87] In general, we can say that international comparisons give results where the nature of the crime (as homicide, for example) minimizes the difference between the codes, and where the figures show considerable differences. We can truly say that for the etiology of crime statics are much less important than dynamics. When we apply the dynamic method to a fairly long period, we have also to allow for the changes that have taken place in the penal codes and in the police organization, etc. The statistical method certainly contains many sources of error. It goes without saying that this is not a reason for not using it at all, but simply for being careful. We must guard against conclusions too hastily drawn. If statistics show less criminality among the Jews—as is generally the case—it is not safe to say, therefore, that the innate morality of the Jew is greater than that of other men. If crime increases during a certain period together with irreligion, we have no right to conclude that there is a causal connection between the two. The connection between the movement of the price of grain and that of crimes against property has been proved many times, as we have seen; if this parallelism of the two curves no longer occurs, we cannot say on that account that economic conditions no longer play a part in the etiology of these crimes; we have seen that in manufacturing countries the industrial situation in general dominates the course of economic criminality. Similar examples could easily be multiplied.