responsibility of sifting the facts to those whose natural and proper duty it is to sift them. As for cross-examination by counsel, many false impressions prevail. People who take their view on the subject from actual experience are well aware that counsel of any experience never try to prove their case by cross-examination. In respect to prisoners, counsel, in my experience, usually regard their duty as done when they have committed the prisoner to contradicting witnesses not likely either to commit perjury or to be mistaken. I have indeed been greatly struck with the moderation and brevity with which prisoners have usually been cross-examined before me. I think indeed, as I have already said, they have been cross-examined rather too little than too much.

A French criminal trial-and it is from the reports of French trials that English people get the notions unfavourable to the examination of prisoners which commonly prevail-is quite a different process from an English one, and proceeds from entirely different principles. It is in its essence an inquiry into the truth of a charge brought forward and supported by public authority, and the duty of the judge is rather to inquire than to direct and moderate. His examination of the prisoner is directed to this object, and the result, no doubt, is to produce scenes much at variance with what our notions, founded as they are upon principles and on practice of an entirely different kind, approve. It is no part of my present purpose to compare the two systems, or to criticise either of them. It is enough to say that there is no danger that a change in the procedure of the English system, made in exact conformity not only with its principles, but with the practice already established and in use in a large and important class of cases, should introduce amongst us what strike us as the defects of a system founded upon and administered according to totally different principles.

One point which appears to me of great practical importance in the matter of the evidence of prisoners is that provision should be made for their being examined as witnesses before they are committed, as well as at their trial. There cannot be a greater pledge of truthfulness and good faith. It is a common form for solicitors to advise their clients, when asked before their committal whether they wish. to say anything, to answer,' I reserve my defence.' How far this may be a convenient course in the case of a guilty person I do not say, but in the case of an innocent person who has a true and substantial defence to rely upon it is a great advantage to be able to say, 'This defence of mine is not an after-thought, it is what I have said all along. It is what I gave my accusers notice of as soon as I had an opportunity.' An alibi in particular is greatly strengthened if it is set up at once, and that for many reasons. In the first place, such a course gives the prosecution an opportunity of making inquiries and testing the evidence of witnesses. In the second place, the evidence of the

witnesses is less open to attack, either on the ground of a failure of memory or on the ground of subsequent contrivance.

It is more difficult to say how this desirable result is to be obtained. One way of doing it would be to make the accused person not merely a competent but a compellable witness at every stage of the inquiry; to authorise the magistrates or the prosecutor before the magistrates to call him as a witness; and to provide that unless he gave evidence at the trial his deposition might be given in evidence. This course would no doubt be effectual, and I do not myself see why it should not be taken. I can understand, however, that there might be a feeling against it. It might be regarded as oppressive, and it might not improbably invest a certain number of police officers with a discretion which they are not fit to exercise. It is not uncommon for officers of the police to act as prosecuting solicitors in some parts of England and Ireland, and it may well be that such an addition to their powers would be objectionable. In matters of this sort the popularity of the law is more important than an increase of its efficiency, unless the increase of its efficiency is very great indeed. It is, however, important to obtain as general as possible a recognition of the fact that to keep back a defence is a suspicious thing, and that to bring it forward on the first opportunity is the strongest pledge of sincerity and truthfulness that can be given.

One point closely connected with this subject is the propriety of adding to the permanent and general law a provision to the same effect as that one which lately proved so useful in Ireland for the detection and suppression of systematic crime-power, namely, to the police authorities to hold an inquiry upon oath with a view to discover the authors of a crime, although no one may have been charged with it. It was one of the proposals of the Criminal Code Commission of 1878 that such a power should be given, and a clause to that effect was introduced into the Criminal Code which that Commission prepared. Upon general grounds I cannot understand the objection to such a measure. The practice exists in most parts of the world, and in England the principle is recognised by one of the oldest of our judicial institutions-the coroner's inquest. Of its utility for the discovery of crime it is necessary only to refer to the case of the murder of Lord Frederick Cavendish and Mr. Burke. It is, of course, possible to lament that discovery, but there can be no question at all as to the means by which it was brought about. With regard to all questions of the reform of the criminal law, whether in regard to the rules of evidence or otherwise, it must never be forgotten that those who fear that the criminal law may be applied to themselves or their friends for political offences of which they do not morally disapprove do not wish to see the efficiency of its administration increased.

For these various reasons I think that the old rule as to the

exclusion of persons accused of crime from competency as witnesses ought to be entirely abolished, and that criminal and civil proceedings should so far be put upon the same footing. It would, however, be wrong, in advocating such a measure, not to point out one inevitable consequence. It is a consequence which has already been incurred in respect of all civil proceedings, and which I believe to be nearly inseparable from all improvements in the law. There are in all legal proceedings two interests which are diametrically opposed to each other, though their opposition is for the most part concealed, because its existence is one of those disagreeable truths which no one likes to admit. They are goodness and cheapness; either object may be attained, but not both. Up to a certain point it is no doubt possible to combine and promote the two objects at once. If you have a system at once inefficient and costly, a system in which fees are imposed at every step for the purpose of providing for useless officials, it is no doubt possible to increase efficiency and economy at the same time by a reduction of establishments and alterations in the law. This state of things did at one time exist to a considerable extent in regard to litigation in England, and it was possible to get the work better done at a less cost by proper alterations, but even at that time reforms usually were found to mean increased expenditure in the long run; and I think that, in regard to the administration of justice, the question in most cases is whether new elaborations are worth the price paid for them. I have a very decided opinion that in civil cases the procedure in the present day is too elaborate, though some recent efforts have been made for its simplification, I hope with success. I do not think this is so with regard to criminal justice. A certain number of criminal trials are still dealt with, not unfairly, not hastily, but without that degree of care to find out the truth which ought to be employed in every case in which liberty and character, and, indeed, a man's whole prospect of leading a respectable, prosperous life, may be at stake, but which an ignorant unadvised man cannot be expected to employ for himself. Many circumstances, some of which I cannot now remember, have produced a conviction in my mind that, if the whole truth were known, it would be found that many crimes are not so simple as they look, and that prisoners might often, if fully examined, bring to light facts which would set their conduct in an unsuspected light. This, I think, would certainly lengthen trials and might tend to complicate them considerably. Unless some means were taken to secure the taking of the prisoner's evidence fully before the magistrates, it would in all probability lead to the raising of false issues before juries, and make occasional adjournments for the purpose of summoning new witnesses necessary, and thus in various ways give a good deal of trouble to all the parties concerned; but I think it would contribute largely to the fairness of the ultimate result, and this is the main thing to consider.

J. F. STEPHEN.

COMTE'S FAMOUS FALLACY.

CIRCUMSTANCES, which I need not specify, have led me to consider of late, more carefully than I had ever considered before, the grounds upon which Comte's famous theory or dictum concerning the three progressive states of human knowledge rests, and the amount of truth which it contains. I have long doubted the accuracy of the law of progress as Comte has stated it; the very neatness and plausibility of the statement seem to suggest that it is not likely to be strictly exact; at the same time these qualities also suggest the probability of the existence in it of some strong element of truth. There may be in this case, as in so many others in which mathematical accuracy is impossible, a basis of reality of which it is important to ascertain the nature and limits, while the claim of absolute universality may be incapable of being substantiated, and may tend to throw doubt upon the claim to acceptance which the theory may really possess.

I propose in the following pages to offer to such persons as care for discussions of the kind some observations upon Comte's three states, and to suggest the limitations necessary for the acceptance of the same as an exposition of truth. Or perhaps it may be more correct to say, that I shall lay before the reader such modifications— and they are important modifications-of Comte's statement as seem to me to be necessary, in order to free it from exaggeration and from virtual error. First, however, let us have Comte's own enunciation of his theory, which shall be quoted from Miss Martineau's translation of the Philosophie Positive:

From the study of the development of human intelligence, in all directions, and through all times, the discovery arises of a great fundamental law, to which it is necessarily subject, and which has a solid foundation of proof, both in the facts of our organisation and in our historical experience. The law is this—that each of our leading conceptions-each branch of our knowledge-passes successively through three different theoretical conditions: the theological, or fictitious; the metaphysical, or abstract; and the scientific, or positive. In other words, the human mind, by its nature, employs in its progress three methods of philosophising, the character of which is essentially different, or even radically opposed: viz., the theological method, the metaphysical, and the positive. Hence arise three philosophies, or general systems of conceptions on the aggregate of phenomena, each of which excludes the others. The first is the necessary point of departure of the

human understanding, and the third is its fixed and definitive state. The second is merely a state of transition.

Now in this enunciation of the supposed necessary law of progress, the following are the material points :

1. Each branch of knowledge passes through three states: the Theological, the Metaphysical, and the Scientific.

2. The progress is in the order above indicated.

3. The three states are mutually opposed to each other, and cannot harmoniously co-exist.

I trust to be able to show that no one of these propositions is universally true, but by way of introduction let me give an illustration or two of the philosopher's meaning, in order that we may be in a better position to consider the limitations which should be imposed upon it. I will borrow the first from the writer of the article on Comte in the Encyclopædia Britannica, who in his turn borrows from an able English disciple of Comte' :

Take the phenomenon of the sleep produced by opium. The Arabs are content to attribute it to the will of God.' Molière's medical student accounts for it by a soporific principle contained in the opium. The modern physiologist knows that he cannot account for it at all. He can simply observe, analyse, and experiment upon the phenomena attending the action of the drug, and classify it with other agents analogous in character.

A still better, because wider, illustration is afforded by the general view of nature taken by thinking men in different epochs of the earth's history. Here we have undoubtedly something which corresponds very much to Comte's theory. In early days natural phenomena were attributed by those who at all thought about such things to direct divine action; the rising and setting of the sun, the phenomena of thunder and lightning, rain, famine, and pestilence, and all the multiform facts of the material universe connected themselves instinctively with the action of a Being, or of Beings, more powerful than man. The only escape from the thought was to be found in not thinking at all—an escape of which probably many availed themselves. This is Comte's theological stage in palpable manifestation. Then comes the metaphysical stage as exhibited by such speculations as those of the Greek philosophers, concerning which we may truly say that they were only transitional, scarcely caricatured by Molière's medical student with his soporific principle. Yet these speculations had a marvellous hold upon the human mind, and in no small degree probably affect it still; it was only after hard battles and long-continued struggles that nature's abhorrence of a vacuum and the notion of inherent tendencies, and such hypotheses as that of the fortuitous concurrence of atoms, and the like, yielded to the overwhelming claims of inductive science. To this last step, which has conducted the human mind to some real knowledge of