The Queen Vs Elahi Bax

Sir Barnes Peacock, Kt., C.J.@mdashI am of opinion that a conviction upon the uncorroborated testimony of an accomplice is legal. This is not new law, nor founded upon a new principle. The point was decided in England as far back as the 10th December 1662, after conference with all the Judges, Several cases to that effect are cited by Sir Mathew Hale in his "Pleas of the Crown," Volume I, pages 303-304. He, however, remarks,-- "Yet though such a party be admissible, as a witness in law, yet the credibility of his testimony is to be left to the jury, and truly. It would be hard to take away the life of any person upon such a witness, that swears to save his own, and yet confessed himself guilty of so great a crime, unless there be very considerable circumstances, which may give the greater credit to what he swears." In The King v. Attwood 2 Lea, 521, which is a leading case upon the subject, two prisoners were convicted of highway robbery upon the uncorroborated evidence of an accomplice as to their identity. The question was referred for the opinion of the twelve Judges, who were unanimously of opinion that an accomplice alone is a competent witness; and if the jury, weighing the probability of his testimony, think him worthy of belief, a conviction supported by such testimony alone is perfectly legal. In sentencing the prisoners, Buller, J., made the following remarks: "Prisoners, you were convicted of a highway robbery at the last summer Assizes at Bridgewater, The material circumstances of the trial were these: The prosecutor gave in evidence that he was robbed by three men on the day laid in the indictment, mentioning the conversation that passed during the robbery, and proving all the facts that are necessary in law to constitute that offence; but as it was dark, he could not swear to the person by whom it was committed. The accomplice was then called, who swore, that he and you had, in company of each other, committed this robbery; and he mentioned all the circumstances that passed, which exactly corresponded with those which the prosecutor had before related. On the testimony of these two witnesses the jury found you guilty; but on a doubt arising in my mind respecting the propriety of this conviction, I thought it proper to refer your case to the consideration of the twelve Judges. My doubt was, whether the evidence of an accomplice, unconfirmed by any other evidence that could materially affect the case, was sufficient to warrant a conviction?-- And the Judges are unanimously of opinion that an accomplice alone is a competent witness; and that, if the jury, weighing the probability of his testimony, think him worthy of belief, a conviction supported by such testimony alone is perfectly legal. The distinction between the competency and the credit of a witness has been long settled. If a question be made respecting his competency, the decision of that question is the exclusive province of the Judge; but if the ground of the objection go to his credit only, his testimony must be received and left with the jury under such directions and observations from the Court as the circumstances of the case may require, to say whether they think it sufficiently credible to guide their decision on the case. An accomplice, therefore, being a competent witness, and the jury in the present case having thought him worthy of credit, the verdict of guilty, which has been found, is strictly legal, though found on the testimony of the accomplice only." His Lordship then passed sentence of death upon the prisoners, but intimated that it was his intention to recommend them to mercy. In the case of Rex v. Jones 2 Camp., 131. Lord Ellenborough says:-- "No one can seriously doubt that a conviction is legal, though it proceed upon the evidence of an accomplice only. Judges, in their discretion, will advise a jury not to believe an accomplice, unless he is confirmed, or only in as far as he is confirmed, but if he is believed, his testimony is unquestionably sufficient to establish the facts which he deposes it is allowed that he is a competent witness, and the consequence is inevitable that if credit is given to his evidence, it requires no confirmation from another witness. Within a few years a case was referred to the twelve Judges, where four men were convicted of a burglary upon the evidence of an accomplice, "who received no confirmation concerning any of the facts which proved the criminality of one of the prisoners; but the Judges were unammously of opinion that the conviction as to all the four was legal, and upon that opinion they all suffered the penalty of the law. Strange notions upon this subject have lately got abroad, and I have thought it necessary to say so much for the purpose of correcting them;" see The King v. Durham 2 Lea., 538. At the Old Baily Sessions, 1784, Smith and Davies were tried for robbing Hunter. During the night the prosecutor was attacked by four ruffians, whose persons he was unable to identify; but during the scuffle he had torn a piece of the coat which one of them had on, who, on being discovered by these means, turned King''s evidence and implicated the two prisoners. But the Court, although it was admitted as an established rule of law that the uncorroborated testimony of an accomplice is legal evidence, thought it too dangerous to suffer a conviction to take place under such unsupported testimony; and the prisoners were acquitted. The law, as above laid down, that a conviction is legal though supported by the uncorroborated evidence of an accomplice, has been admitted by Lord Denman in Rex v. Hastings 7 C. & P., 152, by Alderson, B., in Rex v. Wilkes 7 C, & P., 272, and by many other learned and eminent Judges; and it was so ruled by the Court of Criminal Appeal in Rex v. Stubbs Dears, C.C., 555; S.C., 25 L.J.,. The law of England, therefore, upon this subject is beyond doubt. The law of America is the same, and in that country, where in most of the States new trials are granted in criminal cases, new trials have been refused even when the verdicts were obtained upon the uncorroborated evidence of an accomplice. The cases upon the subject are collected in Whartou''s Criminal Law of the United States of America, page 366. It does not appear that in the cases in which new trials were refused, the Judge who tried the case had omitted to make such observations to the jury with reference to the evidence of the accomplices as the circumstances required. But in civil cases it is clear that, both in that country and in England, a new trial will be granted where, from the absence of proper instructions from the Judge, the jury fall into an error. Formerly, the rule was that the mere commission of a crime did not render a witness incompetent, but persons convicted of treason, felony, or certain other crimes, were rendered incompetent by conviction. The incompetency created by conviction was removed in England by Act of Parliament, and was subsequently removed here by Act XIX, 1837, by which it was enacted that no person shall, by reason of a conviction for any offence whatever, be incompetent to be a witness in any stage of a cause, civil or criminal, or before any Court in the territories of the East India Company.

2. It was contended in the course of argument in the present case, that, in India, the rule of evidence in the mofussil is different from the law of England with respect to the legality of convicting upon the uncorroborated evidence of an accomplice. If there had been a long uniform course of decisions in the late Sudder Court, that the uncorroborated evidence of an accomplice was insufficient in law for the conviction of a prisoner, we should have been disposed to bow to those decisions, and to act upon the rule "stare decisis." One case only was cited from 7 Nizamut Reports, page 57 ⁽²⁾, in which a Judge of the Sudder Court stated that he did not think it legal to convict upon such evidence. There may be other cases to the same effect, but there is no uniform current of decisions which would justify us in holding that the law in this respect in the mofussil was different from the established law of England, and from that which was administered in the late Supreme Court, and is now administered by this Court, in the exercise of Original Criminal Jurisdiction. It would require a uniform train of decisions to justify us in holding that the law of evidence to be administered by the Court upon such a point as this is different in the exercise of the Appellate Criminal Jurisdiction from that which is acted upon in the exercise of Original Jurisdiction. When called upon to give effect to particular expressions which have been made use of by the Judges of the late Sudder Court with regard to the rules of evidence, we must bear in mind that, up to a very recent period, when trial by jury was established in certain districts, it was the province of the Sessions Judges, and of the Judges of the late Sudder Court, to determine questions of fact as well as questions of law in criminal cases; and that, in dealing with such cases, it was not very frequently necessary to determine whether the evidence of a particular witness was insufficient in law to justify a conviction, or merely insufficient to induce them, as Judges of fact, to declare that a prisoner was guilty. There is a wide distinction, however, between disbelieving evidence and determining that it is not legally sufficient if believed; but this distinction is not always sufficiently adverted to by Courts, which are Judges of fact as well as of law. Act II of 1855, s. 28 ⁽³⁾, was referred to by the appellant''s pleader, by whom the case was very well argued, and it was contended that that Act rendered corroboration necessary. Upon that point it is sufficient to say that it was not the intention of the Act to render inadmissible any evidence, which, but for the Act, would have been admissible--see s. 58; nor was it intended to lessen the legal effect of any such evidence.

3. We have, therefore, no hesitation in answering the first question in the affirmative, and declaring that a conviction may be legally had on the uncorroborated evidence of one or more accomplices.

4. It is unnecessary, as regards this part of the case, to answer the second question.

5. Holding that the Judge was not bound to direct the jury that the evidence was not legally sufficient for a conviction, we shall probably mislead if we do not go on to consider whether there was any error or defect in the summing up which constitutes a valid ground of appeal. The question of misdirection is raised by the Judges who referred the case, and is, I think, substantially before us, and ought to be considered, although there is no specific question as to whether there was a misdirection or not. I proceed, therefore, to consider whether there is any ground for setting aside the conviction upon the ground of error in the summing up.

6. In the case of Regina v. Farler 8 C. & P. 106 a very learned and eminent Judge, than whom no one was better able to deal with evidence, and to determine the degree of credibility to which particular witnesses were entitled,--I mean the late Lord Abinger,--in summing up the case to the jury, made the following remarks:-- "I am strongly inclined to think that you will not consider the corroboration in this case sufficient. No one can hear the case without entertaining a suspicion of the prisoner''s guilt, but the rules of law must be applied to all men alike. It is a practice which deserves all the reverence of law, that Judges have uniformly told juries that they ought not to pay any respect to the testimony of an accomplice, unless he is corroborated in some material circumstance. Now, in my opinion, that corroboration ought to consist of some circumstance that affects the identity of the party accused. A man who has been guilty of a crime himself will always be able to relate the facts of the case, and if the confirmation be only on the truth of that history, without identifying the persona, that is really no corroboration at all. If a man were to break open a house and put a knife to your throat, and steal your property, it would be no corroboration that he had stated all the facts correctly, that he had described how the person did put a knife to the throat, and did steal the property. It would not at all tend to show that the party accused participated in it. Here you find that the prisoner and the accomplice are seen together at the house of the landlord--now look at his evidence. If they were seen together under circumstances that were extraordinary, and where the prisoner was not likely to be unless there were concert, it might be something. But he lives within one hundred and fifty yards, and there is nothing extraordinary in his being there; and he left when they were shutting up the house. Therefore it is perfectly natural that he should have been there, and left when he did. The single circumstance is, that the prisoner was seen in a house which be frequents, where he may be seen once or twice a week, and there the case ends against him. All the rest depends upon the evidence of the accomplice. The danger is, that when a man is fixed, and knows that his own guilt is detected, he may purchase impunity by falsely accusing others. I would suggest to you that the circumstances are too slight to justify you in acting on this evidence." The prisoner was accordingly acquitted. In Rex v. Wilkes and Edwards 7 C. & P., 272, a similar rule was laid down by Alderson, B., in a case of sheep stealing. He said:-- "There is a great difference between confirmations as to the circumstances of the felony, and those which apply to the individuals charged; the former only prove that the accomplice was present at the commission of the offence; the latter show that the prisoner was connected with it. This distinction ought always to be attended to. "In summing up, the learned Judge said:-- "The confirmation of the accomplice as to the commission of the felony is really no confirmation at all; because it would be a confirmation as much if the accusation were against you and me, as it would be as to those prisoners who are now upon their trial. The confirmation which I always advise juries to require, is a confirmation of the accomplice in some fact which goes to fix the guilt on the particular person charged. You may legally convict on the evidence of an accomplice only, if you can safely rely on his testimony; but I advise juries never to act on the evidence of an accomplice, unless he is confirmed as to the particular person who is charged with the offence. With respect to the prisoner Edwards, it is proved that, meat of a similar kind was found in his house. The meat cannot be identified, but it is similar; that is, therefore, some confirmation of the accomplice as to Edwards more than any one else. It is also proved that the skin was found in a whirley hole; that is no confirmation, because it does not affect the prisoners more than it affects any other persons. With respect to the prisoner Wilkes, it is proved by the witness Meek that the prisoner Wilkes told him nearly the same story as the accomplice has told you to-day. If you believe that witness, there is confirmation of the accomplice as to the prisoner Wilkes: you will say whether, with these confirmations, you believe the accomplice or not. If you think that his evidence is not sufficiently confirmed as to one of the prisoners, you will acquit that one; if you think he is confirmed as to neither, you will acquit both; if you think he is confirmed as to both, you will find both guilty." The jury found both prisoners guilty. In the case of Rex v. Stubbs Dears, C.C., 555; S.C., 25 L.J., Mag. Ca., 16, in the Court of Criminal Appeal above referred to, Parke, B., said:-- "My practice has always been to tell the jury not to convict the prisoner unless the evidence of the accomplice be confirmed, not only as to the circumstances of the crime, but also as to the person of the prisoner." And Creswell, J., added:-- "You may take it for granted that the accomplice was at the committal of the offence, and may be corroborated as to the facts; but that has no tendency to show that the parties accused were there." Jervis, C.J., in the same case remarked:-- "There is another point to be noticed. When an accomplice speaks as to the guilt of three prisoners, and his testimony is confirmed as to two of them only, it is proper, I think, for the Judge to advise the jury that it is not safe to act on his testimony as to the third person, in respect of whom he is not confirmed, for the accomplice may speak truly as to all the facts of the case, and at the same time, in his evidence, substitute the third person for himself in his narrative of the case." In Rex v. Moores ⁽⁴⁾ indictment against A as principal, and B as receiver, where the evidence of an accomplice was corroborated as to A, but not as to B. Baron Alderson thought it was not sufficient as to B.

7. Conflicting opinions have been expressed as to whether, in a case in which an accomplice accuses two persons, and is corroborated as to one, but not as to the other, a jury ought to be advised to acquit the one as to whom there is no corroboration. The opinion expressed by Jervis, C.J., in Rex v. Stubbs ⁽⁵⁾, as above-mentioned, appears to be the correct one; for nothing is more easy than for an accomplice to accuse an innocent person, in order to get off his real companion in guilt, and to attribute to the person falsely accused acts which were really committed by the guilty companion.

8. In the present case, two accomplices gave evidence against Elahi Bax; but that does not seem to carry the case much further, In Rex v. Noakes 5 C. & P., 326, in which two accomplices spoke distinctly as to the prisoner, Littledale, J., told the jury that if their statements were the only evidence, he could not advise them to convict the prisoner; that it was not usual to convict on the evidence of one accomplice, without confirmation; and that, in his opinion, it made no difference whether the evidence was that of one accomplice only, or of more than one. This, as a general rule, is correct, for otherwise two companions in guilt might get off by confessing and falsely accusing two innocent persons. But if two or three persons should be apprehended at different places, at long distances from each other, and should each confess and give a similar account as to the persons associated with them in a particular dacoity, the statement of each, if made under such circumstances as not to raise a presumption of collusion, might be proved in corroboration of his evidence; such statement being admissible as corroborative evidence, under Act II of 1855, s. 31. The evidence of several accomplices, so corroborated, might be sufficient to satisfy a jury, although the evidence of one of them alone could not have been safely acted upon. These are matters to which the attention of a jury ought, under all circumstances, to be specially directed, with proper remarks from the presiding Judge, according to the rule laid down by Buller, J., in the case already cited.

9. The danger of acting upon the evidence of an accomplice, who is admitted to give evidence for the Crown, arises not merely from the fact of his having committed a crime, for that would go to the credit of every witness who had recently committed or been convicted of a crime, but from the fact of his giving his evidence under the hope or expectation of pardon, and of his obtaining immunity from punishment if his evidence be believed. Suppose, two gentlemen of previously undoubted honor and good character should in a moment of irritation, not amounting in law to provocation, get out of a dak-carriage, and thrash the coachman or since for not giving them a good horse; suppose the man should die, and that both should be convicted of culpable homicide, would any one say that either of them would be so wholly unworthy of credit as witnesses in any other case, that a jury ought to be advised not to act upon his testimony, except so far as it was corroborated? If he would not after conviction be unworthy of credit, if called upon to give evidence against a stranger for another offence, why should he be unworthy of credit before conviction against his own companion and friend, if compelled to give evidence against his will? Suppose that, immediately after the commission of the offence, one should be apprehended, and the other should escape without being identified with sufficient certainty for conviction; suppose that the one who escaped should be apprehended and brought to trial, and that the one who had been apprehended in the first instance should be called as a witness against his will, and being compelled to give evidence (as he might be, under Act II of 1855, s. 32), should identify his companion, would any Judge, in the exercise of a sound judicial discretion, feel himself bound to tell a jury that, because the witness was an accomplice, it would be dangerous to act upon his evidence alone uncorroborated?

10. When the Judges speak of the danger of acting upon the uncorroborated evidence of accomplices, they refer to the evidence of accomplices who are admitted as evidence for the Grown, in the hope or expectation of a pardon. If, in such a case, the accomplices admitted to give evidence act fairly and openly, and discover the whole truth, though, according to the law of England, they are not, except in certain cases, for which special provision is made by statute, entitled as of right to pardon, yet, the usage, the lenity, and the practice of the Court is to stop the prosecution against them, and they have an equitable title to a recommendation to the mercy of the Crown (Cowper''s Reports, 334). The origin of the practice of admitting accomplices to give evidence for the Crown without approvement is explained by Lord Mansfield in Rex v. Rudd Cowp., 331, at p, 335. He there says:-- A person desiring to be an approver, must be one indicted for the offence, and in custody on that indictment. He must confess himself guilty of the offence, and desire to accuse his accomplices. He must likewise upon oath discover, not only the particular offence for which he is indicted, but all treasons and felonies which he knows of; and after all this, it is in the discretion of the Court, whether they will assign him a Coroner, and admit him to be an approver or not; for if, on his confession it appears, that he is a principal, and tempted the others, the Court may refuse and reject him as an approver. When he is admitted as such, it must appear that what he has discovered is true; and that he has discovered the whole truth. For this purpose, the Coroner puts his appeal into form; and when the prisoner returns into Court, he must repeat his appeal, without any help from the Court, or from any "bystander. And the law is so nice, that if he vary in a single circumstance, the whole falls to the ground, and he is condemned to he hanged. If he fail in the color of a horse, or in the circumstances of time, so rigorous is the law, that he is condemned to be hanged; much more if he fail in essentials. The same consequences follow if he does not discover the whole truth; and in all these cases the approver is convicted on his own confession. See this doctrine more at large in Hale''s Pleas of the Crown, Vol. II, pages 226 to 236; Stan: Pleas of the Crown, Lib. 2 C, 52 to C, 58; 3 Inst., 129. A further rigorous circumstance is, that it is necessary to the approver''s own safety, that the jury should believe him; for if the partners in his crime are not convicted, the approver himself is executed. "Great inconvenience arose out of this practice of approvement. No doubt, if it was not absolutely necessary for the execution of the law against notorious offenders, that accomplices would be received as witnesses, the practice is liable to many objections. And though, under this practice, they are clearly competent witnesses, their single testimony alone is seldom of sufficient weight with a jury to convict the offenders; it being so strong a temptation to a man to commit perjury, if by accusing another he can escape himself. Let us see what has come in the room of this practice of approvement. A kind of hope that accomplices who behave fairly, and disclose the whole truth, and bring others to justice, should themselves escape punishment, and be pardoned. This is in the nature of a recommendation to mercy. "But no authority is given to a Justice of the Peace to pardon an offender, and to tell him he shall be a witness against others. The accomplice is not assured of his pardon, but gives his evidence in vinculis, in custody; and it depends on the title he has from his behavior, whether he shall be pardoned or executed." Sir Mathew Hale, speaking of approvement, says:-- "This course of admitting approvers has long been disused, and the truth is, that more mischief had come to good men by this kind of approvements, by false accusations of desperate villains, than benefit to the public by the discovery and convicting of real offenders; jailors for their own profits often constraining prisoners to appeal honest men, and, therefore, provision made against it by I.E. 3, C. 7." (See 2 Hale''s Pleas of the Crown, 226.) The modern practice of admitting accomplices to give evidence under a hope of pardon, though not so dangerous as the old practice of approvement., would still be attended with the greatest danger, but for the safeguard which has to some extent been provided by the practice of the Judges in recommending juries not to act upon such evidence without requiring corroboration as to the identity of the person accused.

11. The danger of acting upon the uncorroborated evidence of accomplices is at least as great here as it would be in England, for here, as in England, the accomplices are not actually pardoned before they give evidence. In England, by confessing and giving evidence, they acquire an equitable right to a recommendation for the mercy of the Crown. Here the Magistrate is merely authorized to tender a pardon ⁽⁶⁾; and if it appear to the Court of Session at the time of trial, or to the High Court as a Court of Reference, that the person who has accepted the offer of pardon has not conformed to the conditions under which the pardon was tendered, either by concealing anything essential or by giving false evidence or information, it is competent to the Court to direct the commitment of such person for trial for the offence in respect of which the pardon was tendered ⁽⁷⁾. The witness, therefore, does not give his evidence under an absolute certainty of immunity. In Scotland the law is different. There, as remarked by Mr. Alison ⁽⁸⁾, "it has been long an established principle of our law that by the very act of calling a socius and putting him in the box, the prosecutor debars himself from all title to molest him for the future with relation to the matter libeled." This is always explained by the presiding Judge to the witness as soon as he appears, and consequently he gives his testimony under a feeling of absolute security, as to the effect which it may have upon himself. This privilege is absolutely, and altogether independent of the prevarication or unwillingness with which the witness may give his testimony. Justice may, indeed, be often defeated by a witness retracting his previous disclosures, or refusing to make any confession after he is put into the witness-box; but it would be much more put in hazard if the witness were sensible that his future safety depended on the extent to which he spoke against his associate at the bar. It is quite as necessary here as it is in England, if not more so, that the evidence of accomplices should not be left to a jury without such directions and observations from the Judge as the circumstances of the case may require. The question is, whether the omission of the presiding Judge, on a trial in the mofussil to make such observations, is not such an error in his summing up as to justify the Court, on appeal or revision, in setting aside a verdict of guilty.

12. It has been said by the learned author, Mr. Starkie ⁽⁹⁾, speaking of the administration of civil justice in England, that "it is the practice for the Judge at Nisi Prius not only to state to the jury all the evidence that has been given, but to comment on its bearing and weight, and to state the legal rules upon the subject and their application to the particular case, and even to advise them as to the verdict they should give, so that it may accord with his view of the law and with justice." He proceeds:-- "Indeed, without this assistance from the Judge, few juries would, in a contested case, be able to come to an unanimous opinion, being frequently left in a state of great perplexity by the influence of the speeches of the contending pleaders. The accuracy of the summing up by the Judge is, therefore, of the very utmost importance, because if the jury, after hearing the evidence and the powerful arguments which probably have been urged in favor of quite opposite views of the question, were entirely left to decide for themselves, without an impartial direction as to what just and legal weight ought to be attached to this or to that view of the case, it would be difficult, if not impracticable, for them to come to a just conclusion; and hence, in the administration of civil justice, it is incumbent on the Judge correctly to state the law upon, the case, as well as the evidence and the bearings of the latter." If the above remarks as to the impracticability of juries coming to a just conclusion are correct as regards the administration of civil justice in England, they are still more "so as regards the administration of criminal justice in the mofussil, where trial by jury is in its infancy, and where the persons of whom juries are generally composed are necessarily more dependent upon the Judge than they are in England for sound and proper advice and assistance as regards the degree of weight which may be fairly and safely attached to the testimony of particular witnesses. The jury, it is true, are not legally bound to act upon the advice or recommendation of the Judge, as there is no appeal from a verdict of acquittal or from a verdict of guilty upon a mere matter of fact. By s. 379 of the Code of Criminal Procedure it is enacted that, in a trial by jury, the Judge will sum up the evidence on both sides, and the jury shall then deliver their finding upon the charge, and "a statement of the Judge''s direction to the jury shall form part of the record." There can be no doubt that that section requires the Judge to sum up properly, and there would be very great danger in holding that there is no remedy by appeal against a verdict of guilty, if it appears clearly to the High Court that failure of justice has been caused by improper advice upon a question of fact, or by an omission to give that advice which a Judge, in the exercise of a sound judicial discretion, ought to give upon questions of fact, or as to the degree of credit to be given to particular witnesses. It appears to me that it amounts to an error in law in the summing up, which on appeal is a ground for setting aside the verdict, subject, however, to the limitation provided by the Code of Criminal Procedure in ss. 439 and 426, viz., that the Appellate Court is satisfied that the accused person has been prejudiced by the error or defect, and that a failure of justice has been occasioned thereby. It was said by Tindal, C.J., in Davidson v. Stanley 2 M. & Gr. 721, 728, that "it is no objection that a Judge lets the jury know the impression which the evidence has made upon his own mind," and that "at all events the party objecting to such a course should show that the impression entertained by the Judge was not justified by the evidence." And it has been already shown that it is the practice of Judges in England to advise juries not to convict merely upon the uncorroborated evidence of an accomplice. If a Judge, in a criminal trial in the mofussil, were to tell the jury that, in his opinion, the evidence was sufficient to justify them in finding the prisoner guilty in a case in which, if the Judge had been trying the case with the aid of assessors, the High Court would on appeal have reversed his judgment if upon the same evidence he had convicted the prisoner, I have no doubt that the Court ought, on appeal, to set aside a verdict of guilty found by the jury, notwithstanding the advice was merely as to the weight of evidence. So, if a Judge, instead of advising a jury not to convict upon the mere uncorroborated evidence of an accomplice, were to advise them to convict upon such evidence, or were to tell them that the uncorroborated evidence of an accomplice, given under a tender of pardon, was admissible, and that it was for them alone to form their opinion upon it, that a conviction founded upon such evidence would be legal, and that such evidence, without corroboration, might be acted upon with as much safety as that of any other witness, I think the error in the direction would form a good ground of appeal.

13. Now, there are errors of omission as well as errors of commission, and I have no doubt that it would form a good ground of appeal against a verdict of guilty if a Judge were to call the attention of a jury to all the evidence against the prisoner, and to omit altogether to allude or call attention to the evidence in his favor. By such a summing up the Judge would not comply with the requirement of the Code of Procedure, and a verdict found upon such a summing up ought, I think, to be set aside, if the Court should be of opinion that the evidence was not sufficient to justify a conviction. I put the case merely to try the principle. It appears to me that such an omission, or an omission to follow a practice which is universally adopted by the Judges in England, and is described by Lord Abinger to be "a practice which deserves all the reverence of law" In Reg. v. Farler, 8 C. & P., 106, would be a ground of appeal against a conviction upon a verdict of guilty based upon such evidence alone, and found by a jury upon such a summing up. So, also, I think it would be error in a summing up, if a Judge, after pointing out the danger of acting upon the uncorroborated evidence of an accomplice, were to tell the jury that the evidence of the accomplice was corroborated by evidence of a fact which did not amount to any corroboration at all. When Lord Ellenborough said,-- "Judges in their discretion will advise a jury not to believe an accomplice unless he is confirmed, or only in as far as he is confirmed" In Rex, v. Jones, 2 Camp., 131, he must have intended that it was their duty to do so. "Discretion," says Lord Mansfield, "when applied to a Court of Justice, means sound discretion guided by law. It must be governed by rule, not by humour. It must not be arbitrary, vague, and fanciful, but legal and regular."

14. But although I am of opinion that the Legislature intended that the Sudder Court should have the power of setting aside a verdict of guilty, pronounced by jury upon an erroneous or defective summing up of the evidence by the presiding Judge, yet I think that it was not their intention that a verdict of guilty should be set aside in every case, in which there is a defective or erroneous summing up. It was their intention to provide protection for the innocent, but not chances of escape for the guilty. The power, therefore, of reversing a finding or of setting aside a trial was carefully guarded by ss. 426 and 439 of the Code of Criminal Procedure, by which it was enacted that "no finding or sentence passed by a Court of competent jurisdiction shall be reversed or altered, on appeal or revision, on account of any error or defect either in the charge or in the proceedings on the trial, * * * unless in the judgment of the Appellate Court the accused person shall have been prejudiced by such error or defect," and that "no trial, held in any Criminal Court, shall be set aside, and no judgment, passed by any Criminal Court, shall be reversed, either on appeal or otherwise, for any irregularity in the proceedings of the trial, unless such irregularity have occasioned a failure of justice."

15. The Code of Criminal Procedure provides that, if a person be convicted on a trial by jury, the appeal shall be admissible only upon a matter of law. But it certainly is not against the principle, or even the letter of the Code, that the Court should have power to set aside a verdict of guilty, for an insufficient or defective summing up of the evidence, in a case in which, in their judgment, the verdict is not warranted by the evidence. If a verdict and conviction could not, under such circumstances, be set aside, trial by a jury, in the Courts of Session in this country, would be fraught with the most dangerous consequences. On the other hand, if every convict, against whom a verdict of guilty is pronounced by a jury, has a right to have that verdict set aside upon appeal, and to obtain his discharge, whenever it can be shown that the presiding Judge has not properly directed the jury as to the degree of weight which ought to be given to particular evidence, a wide door would be thrown open for the escape of guilty men, and the due administration of the criminal law of this country would be placed in the greatest jeopardy, in those districts to which trial by jury has been extended. A verdict of acquittal by a jury cannot be reversed, and ample protection is afforded to prisoners by allowing the High Court to reverse a verdict of guilty for any error or defect in the summing up, whenever the Court is of opinion that a failure of justice has been thereby occasioned. It has been suggested that the word "reverse" means to change to the contrary, and that to reverse a verdict of guilty is to change it into a verdict of not guilty, and that, although the Court, as a Court of Revision, may grant a new trial, as a Court of Appeal it has not power to do so. But I am of opinion that the word "reverse" is not used in so restricted a sense. The word "reverse," in ss. 419 and 426, is applicable not merely to findings or verdicts, but also to sentences; and in s. 439 the same word is used with reference to judgments only. But if the word "reverse," when applied to a verdict, means, "to change or turn into the contrary," it must also mean the same when applied to judgments or sentences. Thus, a judgment of conviction must be turned into a judgment of acquittal. S. 420 shows that such was not the meaning of the word when applied to sentences, even if without that section it would have been possible to put such a construction upon it. The Court, upon revision, may grant a new trial ⁽¹⁰⁾. But the person convicted cannot obtain a revision as a matter of right. I think that the Court has as great a power in this respect on appeal;, as it has on revision, and that it may set; aside a verdict of guilty, and a conviction founded upon it, for any error in law, such as a misdirection of the Judge in point of law, or an error or defect in the summing up of the evidence, or the improper rejection or admission of evidence provided the Court is of opinion that the person convicted has been prejudiced by the error or defect, and that a failure of justice has been occasioned thereby-I am of opinion that the word "reverse" is used in its legal sense, and means "to make void," "to set aside," or "annul." The Legislature, when giving a power to a Court of Revision to order a now trial, may have thought it necessary to do so by express words, as a Court of Revision may act of its own motion and without any application or consent of the person convicted, but an appeal must be preferred by the person convicted; and it seems to follow that, if he asks to have a finding and conviction set aside for error in law, he cannot set up that conviction in bar of a second trial, That was the principle acted upon by the Judges in America, who held that a new trial might be granted in cases of felony notwithstanding the words in the Constitution, "nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb," which words were interpreted by Mr. Justice Story to mean that "no person shall be tried a second time for the same offence, where a verdict has been given by jury." "I am aware," said Mr. Justice Kane, "that one of the most eminent of our jurists--Mr. Justice Story--has found an inhibition in the Constitution against the grant of new trials in cases involving jeopardy of life. But I cannot realize the correctness of the interpretation which, anxious to secure a citizen against the injustice of a second conviction, requires him to suffer under the injustice of the first. Certainly, I would not subject the prisoner to a second trial without his consent. If being capitally convicted, he elects to undergo the sentence, it may be his right. When, however, he asks a second trial, it is to release himself from the jeopardy in which he is already, and it is no new jeopardy that he encounters when his prayer is granted, but the same divested of the imminent certainty of its fatal issue." The same distinction was noticed by other Judges between jeopardy incurred with the consent of a prisoner, and jeopardy incurred without that consent. If a new trial may be granted for error in law by a Court of Revision, even without the prisoner''s consent, can it be doubted that the same Court, as a Court of Appeal, may grant a new trial when an appeal is preferred by a prisoner against a verdict and conviction? It appears to me that, in all cases in which finding of guilty is set aside upon appeal, the Court, if it considers it necessary, may order a new trial. In some cases it may be necessary, for example, where evidence is improperly rejected, or where, for other reasons, the Appellate Court is unable to form a correct opinion as to the guilt or innocence of the appellants. But when the finding and conviction are objected to upon the ground that the Judge did not properly direct the jury as to the degree of weight which ought to be given to the evidence, it appears to me that this Court, sitting as an Appellate Court, is not necessarily bound to send the case back for a new trial. If the Court are of opinion that the evidence could not, in any proper view of the case, support a conviction, it would be worse than useless to send the case back for a new trial, in order that a jury might have the opportunity of convicting upon such evidence under a proper summing up. S. 419 of the Code of Criminal Procedure allows the Appellate Court to alter or reverse the finding. It does not compel them to send the case back for a new trial, in cases in which they see that it is useless, and may be injurious to do so. Regard for the prosecutor and witnesses forbids it; the prisoner is amply protected by the section which prohibits an appeal from a judgment of acquittal, and a failure of justice is sufficiently guarded against by allowing the Court to order a new trial whenever, upon appeal, they are satisfied that there has been a failure of justice. It would tend to defeat, and not to promote justice, if a verdict of guilty were set aside, and a new trial granted, for a defective summing up with reference to the weight of evidence in a case in which the High Court would, upon the evidence given on the trial, have affirmed a conviction if, instead of a trial by jury, the trial had been before a Judge and assessors. In determining whether the verdict ought to be set aside, and a new trial granted, for a defective summing up of the evidence, it appears to me that the question to be considered is not whether, upon a proper summing up of the whole evidence, a jury might possibly give a different verdict, but whether the legitimate effect of the evidence would require a different verdict. If the evidence is such that the High Court would have affirmed the conviction if the trial had been before a Judge and assessors, I think that they ought not to set aside a verdict of guilty found by a jury, merely because the Judge has not, in summing up, given proper caution or advice to the jury as to the weight which they might properly give to the evidence. If a verdict is set aside for such a cause, upon the ground that the error of the Judge has caused a failure of justice, and that the prisoner has been prejudiced thereby, it may be necessary in some cases to grant a new trial. But if the Court is satisfied that a failure of justice has been caused, and that the evidence is wholly insufficient to support any conviction against the prisoner, and would upon the same evidence have reversed a conviction if the case had been tried without the intervention of a jury, there is no necessity, and I think it would be improper, to grant a new trial. In such a case, the Court having set aside the verdict, may order the prisoner to he discharged.

(His Lordship commented on the evidence in the case, and proceeded):--

The case should be returned to the Divisional Bench with an expression of the opinion of the Full Bench:

1st,--That a conviction found upon the uncorroborated evidence of one or more accomplice or accomplices alone is valid in law.

2nd,--That, for the reasons above stated, there was error in law in the summing up of the evidence, which would warrant the Court in setting aside the verdict of guilty, if the Court is satisfied that the prisoner was prejudiced by the error, and that there has been a failure of justice.

3rd,--That the verdict and conviction ought not to be set aside, if the Court he of opinion that the verdict was warranted by the evidence, and that, upon that evidence, they would have upheld the conviction on appeal, if the trial had been by the Judge with the aid of assessors, instead of by jury.

Kemp and Phear, JJ.

concurred.

Seton-Karr, J.

16. I have previously read the exposition of the law by the learned Chief Justice with that attention which its full and exhaustive nature merits.

17. There appears from this and from other cases to have existed some slight doubts amongst us as to how the evidence of an accomplice or of two accomplices should be treated, and as to whether a conviction is legal, and ought to be affirmed, if founded on such evidence alone and uncorroborated. I have now, after full consideration, arrived, substantially, at most of the same conclusions as the Chief Justice. I may observe, however, that on such questions we do quite right to search for information, guidance, and aid in the decisions of the highest judicial authorities in English as well as in American law books. But, as was remarked by Campbell, J., in a late case, I may take the liberty of doubting whether dicta of English law, or even the most elaborate English decisions, are imperatively to rule us on all points in the discharge of the Appellate Criminal Jurisdiction of this High Court. It is almost superfluous to observe that we deal hero with a state of society very different from any European society, and we must apply the law either of particular statutes, or that which is best suited to the people. We are not necessarily to be guided by English law on all points. The substantive criminal law of the Penal Code is unquestionably different from English criminal law. On the other hand, I would observe that the utmost that the old Sudder decisions established is, to my thinking, that it was a rule of practice, rather than an established rule of law with Sudder Judges, not to convict on the uncorroborated testimony of accomplices. In this country such a rule may practically, in many cases, be a sound rule, though it is easy to conceive some cases in which there could be no reason why a conviction should not ensue on the uncorroborated evidence of an accomplice.

18. I think it unnecessary, after a citation of so many high authorities by the Chief Justice, after his full statement of the particular case before us, and after his general remarks, with many of which I entirely agree, to do more than state my own conclusions. I trust that the law on this important subject may henceforth he, in a great measure, settled. Some of the cases quoted, especially that in which Lord Abinger delivered judgment, were referred to by me on a very recent occasion.-- The consideration, then, which I have given to this subject, has enabled me to arrive at the following conclusions:--

1st,--A conviction upon the uncorroborated evidence of an accomplice is legal, and failure in corroboration of the same is not a ground for refusing to convict, or for reversal of conviction.

2nd,--Judges, for all that, ought to be most careful in this country to direct juries that the evidence of accomplices should be received with caution, and that, if possible, corroboration should be required. The extent of such corroboration should be matter for the jury. In cases tried by assessors and open to appeal on the facts, the Judge should himself, I think, act on the same principle.

3rd,--A new trial can be granted, if necessary, by a Bench of the High Court, sitting as well in appeal as on revision. In a late case Kemp, J., and myself, in this view, ordered a new trial on an appeal from a conviction by a jury.

4th,--In the case now referred to us, the failure of the Judge to direct the particular attention of the jury to the nature of the evidence of the accomplices, did amount to an error in law; and the Divisional Bench may deal with the case accordingly.

5th,--Convictions ought not to be reversed, nor should new trials be granted, unless the accused has been really prejudiced within the meaning and scope of s. 426 of the Code of Criminal Procedure.

Jackson, J.

19. Upon the pure legal point before us, I agree in the conclusions at which the Chief Justice has arrived, and generally in the reasons which he has stated.

20. I think it must be admitted that a conviction by a jury upon the uncorroborated testimony of an accomplice is good in law.

1st,--Because the accomplice is a competent witness, even though he may have been previously convicted of an offence ⁽¹¹, and because a single witness, if entitled to full credit, is sufficient, except in cases of treason, to prove any fact, unless there be a rule or practice in our Courts that requires corroborative evidence in support of his testimony.

2nd,--Because cases are conceivable in which the accomplice would be thoroughly credible.

3rd,--Because there is no such established rule or practice as is referred to in the latter part of the section just cited.

21. There can be no doubt that the Chief Justice has indicated how it is that we have no such rule. This came about, first, because for many years exclusively, and until quite recent times, in a large proportion of eases, the procedure in criminal trials was governed by Mahomedan law, and the rules taken from that law and applied to particular cases were never accurately defined and laid down for general adoption; second, because the finality of the verdict of a jury has only arisen under the Code of Criminal Procedure, and the Judges of the late Sudder Court, or Nizamut Adawlut, being supreme judges of fact as well as of law in criminal trials, were not under the necessity of discriminating between what was legally insufficient and what their judgment refused to accept. This being so, in the vacancy, as it were, of any rule upon the subject, we ought probably to adopt on the Appellate Side of this Court the same principles of evidence which are recognized in the exercise of Original Jurisdiction. At any rate, we are not at liberty to adopt any principle of exclusion which is not admitted there, and has not the sanction of ancient practice in the late Nizamut Adawlut. And although I should fully adhere, as a Judge of fact, to the principle which I stated in the case of Dwarka 5 W.R., Cr., 18, who was tried by a Judge with assessors, and which in that case had the concurrence of Kemp and Seton-Karr, JJ., yet, as matter of law, I am bound to say that a conviction by a jury founded upon the evidence of an accomplice without corroboration is not invalid. But before the jury can deliver any finding upon a charge, they must have the evidence on both sides summed up to them, by the Judge, and this function, also called his "direction to the jury" (s. 379, Code of Criminal Procedure) must be fully, discreetly, and conscientiously performed. A statement of the direction is to form part of the record, and a report of it forms part of the matter for this Court''s consideration when it acts as a Court of Revision. There can be no doubt but that an erroneous direction to the jury, where such direction has caused a failure of justice, is a ground for setting aside the verdict, and for either discharging the prisoner or ordering a new trial, as the circumstances of the case may require. And I am not acquainted with any kind of error which is more important in criminal trials than a direction which misleads or omits to guide the jury as to the nature or the weight of evidence.

22. I think it immaterial for the purposes of the present decision, whether the word "reverse," in the 419th and 420th section of the Criminal Procedure Code, means simply to "turn to the contrary," as in the natural sense, or "to make void," which is, no doubt, a legal interpretation, as convictions before a jury can only come up in appeal to this Court, as this Court under its powers of revision can order a new trial and can exert those powers upon hearing an appeal as well as on other occasions.

23. Upon the duty of a Judge in summing up I need not add anything to what has been said by the Chief Justice. But upon the necessity for that duty being carefully performed, and upon the special danger of relying on approvers'' testimony in this country, I think it right to say that which my official experience has suggested. It is not too much to say that native juries in the mofussil are generally quite incapable of appreciating evidence unaided, when the question before them is at all critical. On the one hand, they readily, and even greedily, listen to positive assertions regarding the guilt or innocence of the prisoner, frequently without discriminating between that which the witness declares of his own knowledge and that which is pure hearsay. On the other hand, they commonly disregard circumstantial evidence, though it be of the strongest and moat trustworthy character. And though it may come out in cross-examination that the statement is hearsay, though it may even be apparent at once, and the words might not be taken down, yet they have been heard by the jury, and the impression is made, and though the facts which constitute circumstantial evidence he well put together and their effect be obvious to the trained judicial mind, they will seem of little importance to the ignorant juror. With these proclivities the native juror plainly stands in need of intelligent guidance, and without that guidance, will, in difficult cases, often go entirely wrong. It is remarkable, too, as it is notorious, that the jurors are, as a rule, decidedly less intelligent, as well as less instructed, than the persons employed as assessors in criminal trials, and yet, by a strange anomaly of modern law, the verdict of the ignorant, inexperienced, unsworn jury, is final upon facts; while upon facts not only are the assessors overruled by the Judge, but the opinions of Judge and assessors together may be set aside by the Appellate Court.

24. And then as to testimony:-- I feel bound to say, after many years of conversance with Courts of Justice of India, that veracity is not regarded in this country as it is in the countries of Western Europe. Whether this he due to willful falsehood, to imperfect memory, to inexact conditions of mind, to fear, or to all these causes combined, I am not called upon at present to enquire. I need only say that the care with which witnesses must be watched, and the deductions which have to be made from their credit, are much greater than in England. It must in fairness be remembered that, as witnesses, we have to do almost universally with the meaner classes; that the respectable native avoids being made a witness, as we should shun the small-pox, and that witnesses, therefore, are scarcely a fair sample of the population. But the fact remains: and when the witness is, moreover, a person stained, by his own confession, with the commission of atrocious crimes, most of all, where to the desperate ruffianism of the dacoit he adds the depravity of the retained approver, can the unsupported word of such a person be a safe ground on which to convict any prisoner? I need not say that this, (and not the unlucky gentleman who, in a moment of irritation, has committed an act of violence), is the kind of approver, or accomplice, whom we have in view "when we speak of approvers'' testimony. The other case is of extremely rare occurrence--this of every day.

25. Now, when in the course of a long trial in which many persons are on their defence, there is against particular prisoners only the kind of evidence we are speaking of, and the Judge in his direction, instead of pointing out the defect and warning the jury against the danger, actually throws a veil over that nakedness, and disguises the danger by the use of general words to the effect that "the tendency of the evidence is to establish the prisoner''s guilt," in such a case can it be doubted that the Judge has greatly miscarried, that the jury have been wrongly directed, and that the prisoner has been seriously prejudiced? I think not, and I am sure that the nature of both witness and juror, the finality of, and absence of sanction to, the verdict, make it even more incumbent on the Judge in this country, than it is in England, to perform with care and fidelity the office of direction.

26. I have heard it said that if the jury go wrong, it does not very greatly matter; the prisoner can he pardoned. No doubt he can, and there may be persons so constituted as to find this a satisfactory assurance. It is not so to me. No doubt, after an improper conviction has taken place, when the matter can be properly represented through the proper channel, when the head of the Government can be communicated with at Darjeeling or at Simla, the convict may, after weeks or months of unmerited suffering, receive a free pardon for an offence of which he ought never to have been found guilty. For my part, I should prefer to be tried by a careful and regular administration of justice.

27. The Chief Justice has pointed out that the prisoner is not in every case of misdirection entitled to a new trial, and there has been some apprehension expressed that the admission of the principle we are laying down may open a door to the escape of criminals, merely by reason of some shortcoming of the Judge in point of form. But it seems to me that the simple test, "Has there been a failure of justice?" may be applied in most cases with perfect case and perfect safety.

28. In regard to the proposed rule, that we should not interfere in case of misdirection where the facts are such that, if the trial had been held before a Judge and assessors, we should have affirmed the sentences I have only one misgiving. It is not always safe, I might say it is rarely safe, for an Appellate Court with papers before it, to put itself in the place of the Court below which has heard the witnesses; and it might be that, in affirming the conviction on the faith of some unnoticed circumstance of corroboration, found in the evidence, we might be using that which the Judge and jury would not have relied upon. But this, perhaps, only suggests caution in the application of the rule rather than an objection to the rule itself. With these observations, therefore, I concur both in the judgment on the general point, and in the course which it is proposed to take with the particular case before us.

⁽¹⁾ Act X of 1872, s. 297, provides as follows:--

"If, in any case either called for by itself or reported for orders, or which comes to its knowledge, it appears to the High Court that there has been a material error in any judicial proceeding of any Court subordinate to it, it shall pass such judgment, sentence, or order thereon as it thinks fit.

"If it considers that an accused person has been improperly discharged, it may order him to be tried, or to be committed for trial;

"If it considers that the charge has been inconveniently framed, and that the facts of the case show that the prisoner ought to have been convicted of an offence other than that of which he was convicted, it shall pass sentence for the offence of which he ought to have been convicted:

"Provided that if the error in the charge appears materially to have misled and

prejudiced the accused person in his defence, the High Court shall annul the conviction, and remand the case to the Court below with an amended charge, and the Court below shall thereupon proceed as if it had itself amended such charge.

"If the High Court considers that any person convicted by a Magistrate has committed an offence not triable by such Magistrate, it may annul the trial and order a new trial before a competent Court.

If it considers that the sentence passed on the accused person is one which cannot legally be passed for the offence of which the accused person has been convicted, or might have been legally convicted upon the facts of the case, it shall annul such sentence and pass a sentence in accordance with law.

If it considers that the sentence passed is too severe, it may pass any lesser sentence warranted by law; if it considers that the sentence is inadequate, it may pass a proper sentence,."