vestigations to which I allude are exclusively inductive; they are based on collections of almost innumerable facts, extending over many countries, thrown into the clearest of all forms, the form of arithmetical tables; and finally, they have been put together by men who, being for the most part mere government officials, had no particular theory to maintain, and no interest in distorting the truth of the reports they were directed to make.

The most comprehensive inferences respecting the actions of men, which are admitted by all parties as incontestable truths, are derived from this or from analogous sources; they rest on statistical evidence, and are expressed in mathematical language. And whoever is aware of how much has been discovered by this single method, must not only recognise the uniformity with which mental phenomena succeed each other, but must, I think, feel sanguine that still more important discoveries will be made, so soon as there are brought into play those other powerful resources which even the present state of knowledge will abundantly supply. Without, however, anticipating future inquiries, we are, for the moment, only concerned with those proofs of the existence of a uniformity in human affairs which statisticians have been the first to bring forward.

The actions of men are by an easy and obvious division separated into two classes, the virtuous and the vicious; and as these classes are correlative, and when put together compose the total of our moral conduct, it follows that whatever increases the one, will in a relative point of view diminish the other; so that if we can in any period detect a uniformity and a method in the vices of a people, there must be a corresponding regularity in their virtues; or if we could prove a regularity in their virtues, we should necessarily infer an equal regularity in their vices; the two sets of actions being, according to the terms of the division, merely supplementary to each other.21 Or, to express this proposition in another way, it is evident that if it can be demonstrated that the bad actions of men vary in obedience to the

20

Dufau, Traité de Statistique, pp. 75, 148.

21 Some moralists have also established a third class of actions, which they call indifferent, as belonging neither to virtue nor to vice; and hence there arose the famous doctrine of probability, set up by several eminent Romish casuists, and hotly attacked by Pascal.. But this, if we put aside its worst feature, namely its practical bearings, is merely a question of definition; inasmuch as every indifferent act must lean on the side either of evil or of good, and may therefore be referred to the category to which it inclines; and certainly every increase of vice diminishes virtue relatively, though not always absolutely. Among the Greek philosophers there was a schism on this point: 'Αρέσκει δὲ αὐτοῖς (i. e. Stoics) μηδὲν μέσον εἶναι ἀρετῆς καὶ κακίας· τῶν περιπατητικών μεταξὺ ἀρετῆς καὶ κακίας εἶναι λεγόντων τὴν προκοπήν. Diog. Laert. de Vitis Philosophorum, lib. vii. segm. 127, vol. i. p. 445.

changes in the surrounding society, we shall be obliged to infer that their good actions, which are, as it were, the residue of their bad ones, vary in the same manner; and we shall be forced to the further conclusion, that such variations are the result of large and general causes, which, working upon the aggregate of society, must produce certain consequences, without regard to the volition of those particular men of whom the society is composed.

Such is the regularity we expect to find, if the actions of men are governed by the state of the society in which they occur; while, on the other hand, if we can find no such regularity, we may believe that their actions depend on some capricious and personal principle peculiar to each man, as free will or the like. It becomes, therefore, in the highest degree important to ascertain whether or not there exists a regularity in the entire moral conduct of a given society; and this is precisely one of those questions for the decision of which statistics supply us with materials of immense value.

For the main object of legislation being to protect the innocent against the guilty, it naturally followed that European governments, so soon as they became aware of the importance of statistics, should begin to collect evidence respecting the crimes they were expected to punish. This evidence has gone on accumulating, until it now forms of itself a large body of literature, containing, with the commentaries connected with it, an immense array of facts, so carefully compiled, and so well and clearly digested, that more may be learned from it respecting the moral nature of Man than can be gathered from all the accumulated experience of preceding ages. But as it will be impossible in this Introduction to give any thing like a complete statement of those inferences which, in the actual state of statistics, we are authorized to draw, I shall content myself with examining two or three of the most important, and pointing out the connexion between them.

22

Of all offences, it might well be supposed that the crime of murder is one of the most arbitrary and irregular. For when we

I say this advisedly: and whoever has examined these subjects must be aware of the way in which writers on morals repeat the commonplace and hackneyed notion of their predecessors; so that a man, after reading every thing that has been written on moral conduct and moral philosophy, will find himself nearly as much in the dark as when his studies first began. The most accurate investigators of the human mind have hitherto been the poets, particularly Homer and Shakspeare; but these extraordinary observers mainly occupied themselves with the concrete phenomena of life; and if they analyzed, as they probably did, they have concealed the steps of the process, so that now we can only verify their conclusions empirically. The great advance made by the statisticians consists in applying to these inquiries the doctrine of averages, which no one thought of doing before the eighteenth century.

VOL. I.-2

consider that this, though generally the crowning act of a long career of vice, is often the immediate result of what seems a sudden impulse; that when premeditated, its committal, even with the least chance of impunity, requires a rare combination of favorable circumstances for which the criminal will frequently wait; that he has thus to bide his time, and look for opportunities he cannot control; that when the time has come, his heart may fail him; that the question whether or not he shall commit the crime may depend on a balance of conflicting motives, such as fear of the law, a dread of the penalties held out by religion, the prickings of his own conscience, the apprehensions of future remorse, the love of gain, jealousy, revenge, desperation ;—when we put all these things together, there arises such a complication of causes, that we might reasonably despair of detecting any order or method in the result of those subtle and shifting agencies by which murder is either caused or prevented. But now, how stands the fact? The fact is, that murder is committed with as much regularity, and bears as uniform a relation to certain known circumstances, as do the movements of the tides, and the rotations of the seasons. M. Quetelet, who has spent his life in collecting and methodizing the statistics of different countries, states, as the result of his laborious researches, that "in every thing which concerns crime, the same numbers re-occur with a constancy which cannot be mistaken; and that this is the case even with those crimes which seem quite independent of human foresight, such, for instance, as murders, which are generally committed after quarrels arising from circumstances apparently casual. Nevertheless, we know from experience that every year there not only take place nearly the same number of murders, but that even the instruments by which they are committed are employed in the same proportion." This was the language used in 1835 by confessedly the first statistician in Europe, and every subsequent investigation has confirmed its accuracy. For later inquiries have ascertained the extraordinary fact, that the uniform reproduction of crime is more clearly marked, and more capable of being predicted, than are the physical laws connected with the disease and destruction of our

23 "Dans tout ce qui se rapporte aux crimes, les mêmes nombres se reproduisent avec une constance telle, qu'il serait impossible de la méconnaître, même pour ceux des crimes qui sembleraient devoir échapper le plus à toute prévision humaine, tels que les meurtres, puisqu'ils se commettent, en général, à la suite de rixes qui naissent sans motifs, et dans les circonstances, en apparence, les plus fortuites. Cependant l'expérience prouve que non-seulement les meurtres sont annuellement à peu près en même nombre, mais encore que les instrumens qui servent à les commettre sont employés dans les mêmes proportions." Quetelet sur l'Homme, Paris, 1835, vol. i. p. 7; see also vol. ii. pp. 164, 247.

bodies. Thus, for instance, the number of persons accused of crime in France between 1826 and 1844 was, by a singular coincidence, about equal to the male deaths which took place in Paris during the same period, the difference being that the fluctuations in the amount of crime were actually smaller than the fluctuations in the mortality; while a similar regularity was observed in each separate offence, all of which obeyed the same law of uniform and periodical repetition.24

This, indeed, will appear strange to those who believe that human actions depend more on the peculiarities of each individual than on the general state of society. But another circumstance remains behind still more striking. Among public and registered crimes, there is none which seems so completely dependent on the individual as suicide. Attempts to murder or to rob may be, and constantly are, successfully resisted; baffled sometimes by the party attacked, sometimes by the officers of justice. But an attempt to commit suicide is much less liable to interruption. The man who is determined to kill himself, is not prevented at the last moment by the struggles of an enemy; and as he can easily guard against the interference of the civil power, 25 his act becomes as it were isolated; it is cut off from foreign disturbances, and seems more clearly the product of his own volition than any other offence could possibly be. We may

24 Thus, in twenty years' observations, the number of persons accused of various crimes in France, and registered under their respective ages, scarcely varies at any age from year to year, comparing the proportion per cent. under each age with the totals. The number of persons accused in all France, in the years 1826 to 1844, was about equal to the deaths of males registered in Paris; but singularly enough, the former results are more regular than the latter, notwithstanding the accidental causes which might affect them;-notwithstanding even a revolution in Paris, which convulsed society and brought in a new dynasty." Brown on the Uniform Action of the Human Will, in The Assurance Magazine, no. viii. July, 1852, pp. 849, 350. That the variations in crime are less than those of mortality, is also noticed in Statistique Morale, pp. 18, 34, in Mémoires de l'Académie de Belgique, vol. xxi. Bruxelles, 1848, 4to.

25 The folly of lawgivers thinking that by their enactments they can diminish suicide, is exposed by M. C. Comte in his Traité de Législation, vol. i. p. 486. See also some good remarks by Jefferson, in his observations on Criminal Law, in Appendix to Jefferson's Memoirs, by Randolph, vol. i. pp. 126, 127. Heber (Journey through India, vol. i. pp. 389, 390) found that the English Government had vainly attempted to check the suicides frequently committed at Benares by drowning; and in our country the interference of legislators is met by the perjury of jurors, since, as Bentham says, English juries do not hesitate to violate their oaths by declaring the suicide to be non compos. Principles of Penal Law, in Bentham's Works, edit. Bowring, 1843, vol. i. pp. 479, 480. In regard to the determination of the individual, and the impossibility of baffling his intention, there are cases recorded of persons who, being deprived of the ordinary means of destruction, put an end to life by holding their breath; while others effected their purpose by turning back their tongue so as to exclude air from the larynx. Elliotson's Human Physiology, pp. 491, 492.

also add that, unlike crimes in general, it is rarely caused by the instigation of confederates; so that men, not being goaded into it by their companions, are uninfluenced by one great class of external associations which might hamper what is termed the freedom of their will. It may, therefore, very naturally be thought impracticable to refer suicide to general principles, or to detect any thing like regularity in an offence which is so eccentric, so solitary, so impossible to control by legislation, and which the most vigilant police can do nothing to diminish. There is also another obstacle that impedes our view; this is, that even the best evidence respecting suicide must always be very imperfect. In cases of drowning, for example, deaths are liable to be returned as suicides which are accidental; while, on the other hand, some are called accidental which are voluntary.26 Thus it is, that selfmurder seems to be not only capricious and uncontrollable, but also very obscure in regard to proof; so that on all these grounds it might be reasonable to despair of ever tracing it to those general causes by which it is produced.

These being the peculiarities of this singular crime, it is surely an astonishing fact, that all the evidence we possess respecting it points to one great conclusion, and can leave no doubt on our minds that suicide is merely the product of the general condition of society, and that the individual felon only carries into effect what is a necessary consequence of preceding circumstances.27 In a given state of society, a certain number of persons must put an end to their own life. This is the general law; and the special question as to who shall commit the crime depends of course upon special laws; which, however, in their total action, must obey the large social law to which they are all subordinate. And the power of the larger law is so irresistible, that neither the love of life nor the fear of another world can avail any thing towards even checking its operation. The causes of this remarkable regularity I shall hereafter examine; but the

26 This also applies to other cases besides those of drowning. See Taylor's Medical Jurisprudence, 1846, pp. 587, 597; and on the difficulty of always distinguishing a real suicide from an apparent one, see Esquirol, Maladies Mentales, vol. i. p. 575. From a third to a half of all suicides are by drowning. Compare Dufau, Traité de Statistique, p. 304; Winslow's Anatomy of Suicide, 1840, p. 277; Quetelet, Statis tique Morale, p. 66. But among these, many are no doubt involuntary; and it is certain that popular opinion grossly exaggerates the length of time during which it is possible to remain under water. Brodie's Surgery, 1846, pp. 89-92.

27 Tout semble dépendre de causes déterminées. Ainsi, nous trouvons annuellement à peu près le même nombre de suicides, non-seulement en général, mais encore en faisant la distinction des sexes, celle des âges, ou même celle des instruments employés pour se détruire. Une année reproduit si fidèlement les chiffres de l'année qui a précédé, qu'on peut prévoir ce qui doit arriver dans l'année qui va suivre." Quetelct, Statistique Morale, 1848, p. 35; see also p. 40.