diminish all punishments, and to abolish a certain number, by a magnanimous reaction of humanity against the arbitrary harshness of medi<ae>val times. It had also, and still has, a method of its own, <p xviii>namely, to study crime from its first principles, as an abstract entity dependent upon law.

Here and there since the time of Beccaria another stream of theory has made itself manifest. Thus there is the correctional school, which Roeder brought into special prominence not many years ago. But though it flourished in Germany, less in Italy and France, and somewhat more in Spain, it had no long existence as an independent school, for it was only too easily confuted by the close sequence of inexorable facts. Moreover, it could do no more than oppose a few humanitarian arguments on the reformation of offenders to the traditional arguments of the theories of jurisprudence, of absolute and relative justice, of intimidation, utility, and the like.

No doubt the principle that punishment ought to have a reforming effect upon the criminal survives as a rudimentary organ in nearly all the schools which concern themselves with crime. But this is only a secondary principle, and as it were the indirect object of punishment; and besides, the observations of anthropology, psychology, and criminal statistics have finally disposed of it, having established the fact that, under any system of punishment, with the most severe or the most indulgent methods, there are always certain types of criminals, representing a large number of individuals, in regard to whom amendment is simply impossible, or very transitory, on account of their organic and moral degeneration. Nor must we forget that, since the natural roots of crime spring not only from the individual organism, but also, in large measure, from its physical and social environment, <p xix>correction of the individual is not sufficient to prevent relapse if we do not also, to the best of our ability, reform the social environment. The utility and the duty of reformation none the less survive, even for the positive school, whenever it is possible, and for certain classes of criminals; but, as a fundamental principle of a scientific theory, it has passed away.

Hitherto, then, the classical school stands alone, with varying shades of opinion, but one and distinct as a method, and as a body of principles and consequences. And whilst it has achieved its aim in the most recent penal codes, with a great, and too frequently an excessive diminution of punishments, so in respect of theory, in Italy, Germany, and France it has crowned its work with a series of masterpieces amongst which I will only mention Carrara's ``Programme of Criminal Law.'' As the author tells us in one of his later editions, from the <a!> priori principle that ``crime is a fact dependent upon law, an infraction rather than an action,'' he deduced—and that by the sheer force of an admirable logic—a complete symmetrical scheme of legal and abstract consequences, wherein judges are compelled, whether they like it or not, to determine the position of every criminal who comes before them.

But now the classical school, which sprang from the marvellous little work of Beccaria, has completed its historic cycle. It has yielded all it could, and writers of the present day who still cling to it can only recast the old material. The youngest of them, indeed, are condemned to a sort of Byzantine discussion of <p xx>scholastic formulas, and to a sterile process of scientific rumination.

And meantime, outside our universities and academies, criminality continues to grow, and the punishments hitherto inflicted, though they can neither protect nor indemnify the honest, succeed in corrupting and degrading evil-doers. And whilst our treatises and codes (which are too often mere treatises cut up into segments) lose themselves in the fog of their legal abstractions, we feel more strongly every day, in police courts and at assizes, the necessity for those biological and sociological studies of crime and criminals which, when logically directed, can throw light as nothing else can upon the administration of the penal law.

CHAPTER I.

THE DATA OF CRIMINAL ANTHROPOLOGY.

THE experimental school of criminal sociology took its original title from its studies of anthropology; it is still commonly regarded as little more than a ``criminal anthropology school.'' And though this title no longer corresponds with the development of the school, which also takes into account and investigates the data of psychology, statistics, and sociology, it is none the less true that the most characteristic impetus of the new scientific movement was due to anthropological studies. This was conspicuously the case when Lombroso, giving a scientific form to sundry scattered and fragmentary observations upon criminals, added fresh life to them by a collection of inquiries which were not only original but also governed by a distinct idea, and established the new science of criminal anthropology.

It is possible, of course, to discover a very early origin for criminal anthropology, as for general anthropology; for, as Pascal said, man has always been the most wonderful object of study to himself. For observations on physiognomy in particular we may go as far backwards as to Plato, and his comparisons of the human face and character with those<p 1> of the brutes, or even to Aristotle, who still earlier observed the physical and psychological correspondence between the passions of men and their facial expression. And after the medi<ae>val gropings in chiromancy, metoscopy, podomancy and so forth, one comes to the seventeenth century studies in physiognomy by the Jesuit Niquetius, by Cortes, Cardanus, De la Chambre, Della Porta, &c., who were precursors of Gall, Spurzheim, and Lavater on one side, and, on the other, of the modern scientific study of the emotions, with their expression in face and gesture, conducted by Camper, Bell, Engel, Burgess, Duchenne, Gratiolet, Piderit, Mantegazza, Schaffhausen, Schack, Heiment, and above all by Darwin.

With regard to the special observation of criminals, over and above the limited statements of the old physiognomists and phrenologists, Lauvergne (1841) in France and Attomyr (1842) in Germany had accurately applied the theories of Gall to the examination of convicts; and their works, in spite of certain exaggerations of phrenology, are still a valuable treasury of observations in anthropology. In Italy, De Rolandis (1835) had published his observations on a deceased criminal; in America, Sampson (1846) had traced the connection between criminality and cerebral organisation; in Germany, Camper (1854) published a study on the physiognomy of murderers; and Ave Lallemant (1858-62) produced a long work on criminals, from the psychological point of view.

But the science of criminal anthropology, more <p 3>strictly speaking, only begins with the observations of English gaol surgeons and other learned men, such as Forbes Winslow (1854), Mayhew (1860), Thomson (1870), Wilson (1870), Nicolson (1872), Maudsley (1873), and with the very notable work of Despine (1868), which indeed gave rise to the inquiries of Thomson, and which, in spite of its lack of synthetic treatment and systematic unity, is still, taken in conjunction with the work of Ave Lallemant, the most important inquiry in the psychological domain anterior to the work of Lombroso.

Nevertheless, it was only with the first edition of ``The Criminal'' (1876) that criminal anthropology asserted itself as an independent science, distinct from the main trunk of general anthropology, itself quite recent in its origin, having come into existence with the works of Daubenton, Blumenbach, Soemmering, Camper, White, and Pritchard.

The work of