Commonwealth v. Holmes

127 Mass. 424, 1879 Mass. LEXIS 108

Opinion

Gray, C. J.

It has always been held that a jury might, if they saw fit, convict on the uncorroborated testimony of an accomplice. Lord Hale, Lord Holt and Lord Mansfield treated the question of his credibility as one wholly for the determination of the jury, without any precise rule as to the weight to be given to his testimony. 1 Hale P. C. 304, 305. Charnock’s case, 12 Howell’s State Trials, 1377, 1454. Rex v. Rudd, Cowp. 331, 337; S. C. 1 Leach (4th ed.) 115, 120. The earliest case reported, we believe, in which there is any indication of such a rule, is one in which, on a trial at the Old Bailey in 1784 for robbery, the prosecutor was unable to identify the robbers, except one who turned king’s evidence, and implicated the two prisoners. “ But the court, though 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.” Smith & Davis’s case, 1 Leach, 479 note.

In 1787,

Footnotes

•

* at the trial of two men for highway robbery, the prosecutor testified that he was robbed by three men, stated the conversation between himself and them, and proved all the facts necessary in law to constitute the offence, but, as it was dark, could not swear to the persons of the robbers. An accomplice then testified that he and the defendants, in the company of each other, committed the robbery, mentioning all the circumstances that passed, which exactly corresponded with those that the prosecutor had before related. Mr. Justice Buller, before whom the trial was had, being in doubt (to use his own words) “whether the evidence of an accomplice, unconfirmed by any other evidence that could materially affect the case, was sufficient to warrant a conviction,” referred the case to the consideration of the twelve judges; and afterwards, in passing sentence, announced that the judges were 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 in 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 an accomplice only.” The King v. Atwood, 1 Leach, 464. According to another statement of that case on page 479 of the same book, “the ten judges who were present were unanimously of opinion that the cir cumstance of his being an accomplice went to his credit only, and that his evidence might be left with the jury, although it was entirely uncorroborated by any other testimony; and that the practice of rejecting an unsupported accomplice, is rather a matter of discretion with the court than a rule of law.” And see S. C. cited in Jordaine v. Lashbrooke, 7 T. R. 601, 609. Shortly afterwards a case of burglary was submitted to the jury on the uncorroborated testimony of an accomplice, and the conviction affirmed by the twelve judges. The King v. Durham, 1 Leach, 478.

Lord Ellenborough, sitting at nisi prius, said: “Ho 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 so 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.” Rex v. Jones, 2 Camp. 131. See also S. C. reported at length in 31 Howell’s State Trials, 251, 315, 325, and Despard’s case, 28 Howell’s State Trials, 345, 488, 489.

In Rex v. Birkett, Russ. & Ry. 251, in 1813, the twelve judges are said to have “ thought that an accomplice did not require confirmation as to the person he charged, if he was confirmed as to the particulars of his story.” The case came before them informally, and is very briefly and imperfectly reported, and perhaps presented the same point only as Atwood’s case, above stated. Yet in some trials about the same time, or soon after, the jury were instructed that corroboration was not necessary upon every material fact, nor as to all the prisoners, and that it was sufficient if the accomplice was confirmed upon such and so many material facts as to satisfy the jury that his statement was true. Thompson, B., in Swallow’s case, 31 Howell’s State Trials, 971, 980, 981. Le Blanc, J., in Mellor’s case, 31 Howell’s State Trials, 997, 1012. Bayley, J., in Rex v. Dawber, 3 Stark. 34. Hullock, B., in Rex v. Barnard, 1 Car. & P. 87.

But in 1829, where the testimony of an accomplice was confirmed as to an accessory, but not as to the principal, Mr. Justice Littledale directed an acquittal of both. Rex v. Wells, Mood. & Malk. 326. And for the past fifty years it has been the usual practice of English judges at nisi prius to advise the jury that the corroboration of the testimony of an accomplice ought to be of facts going to prove the guilt of the defendant, and that corroboration as to the guilt of one defendant only would not justify the conviction of another. Vaughan, B., in Rex v. Field, Dickinson Qu. Sess. (5th ed.) 520. Patteson, J., in Rex v. Ad-dis, 6 Car. & P. 388; and in Kelsey’s case, 2 Lewin, 45. Williams, J., in Rex v. Webb, 6 Car. & P. 595. Alderson, B., in Rex v. Moores, 7 Car & P. 270; in Rex v. Wilkes, 7 Car. & P. 272; in Rex v. Fletcher, 2 Lewin, 45 note; and in Regina v. Jenkins, 1 Cox C. C. 177. Lord Abinger, C. B., in Regina v. Farler, 8 Car. & P. 106. Gurney, B., in Regina v. Dyke, 8 Car. & P. 261. Jervis, C. J., Parke, B., and Cresswell, J., in Regina v. Stubbs, 7 Cox C. C. 48; S. C. Dearsly, 555. This practice, however, was never considered as establishing an- absolute rule of law. See Patteson, J., in Rex v. Hargrave, 5 Car. & P. 170; Lord Denman, Park, J., and Alderson, B., in Rex v. Hastings, 7 Car. & P. 152; Gurney, B., in Rex v. Jarvis, 2 Mood. & Rob. 40; Lord Abinger, in Regina v. Farler, ubi supra; Maule, J., in Regina v. Mullins, 3 Cox C. C. 526, 531.

In 1855 the question was directly brought before the Court of Criminal Appeal, held by Chief Justice Jervis, Baron Parke, and Justices Wightman, Cresswell and Willes. By the case stated by the chairman of the quarter sessions, it appeared that, on the trial of Stubbs and two others for larceny, three accomplices were examined, whose testimony was not corroborated as to Stubbs, but only as to the other prisoners; and that it was contended in behalf of Stubbs, that the jury should be directed that the evidence of the accomplices ought to have been corroborated as to him. But the chairman directed the jury that it was not necessary that the accomplices should be corroborated as to each individual prisoner being connected with the crime charged; that their being corroborated as to material facts tending to show that the other prisoners were connected with the larceny was sufficient as to the whole case; that the jury should look with more suspicion as to the evidence in the case of Stubbs, where there was no corroboration, than in the cases of the other prisoners, where there was corroboration, but that it was a question for the jury. Chief Justice Jervis and Baron Parke said that the chairman had departed from the usual practice. The Chief Justice said: “ Where an accomplice speaks as to the guilt of three prisoners, and is confirmed as to two of them only, the jury may, no doubt, if they please, act on the evidence of the accomplice alone as to the third prisoner; but it is proper for the judge in such a case to advise the jury that it is safer to require confirmation of the testimony of the accomplice as to the third prisoner, and not to act upon his evidence alone; for nothing is so easy as for the accomplice, speaking truly as to all the other facts of the case, to put the third man in his own place.” Baron Parke said: “My practice has always been to direct the jury not to convict unless the evidence of the accomplice be confirmed, not only as to the circumstances of the crime, but also as to the identity of the prisoner. An accomplice necessarily knows all the facts of the case, and his story, when the question of identity is raised, does not receive any support from its consistency with those facts.” And Mri Justice Cresswell said: “ I agree in.the view of the question taken by my brother Parke, and have always acted upon it. You may take it for granted that the accomplice was present when the offence was committed, and there may therefore be no difficulty in corroborating him as to the facts; but that has no tendency to show that any particular person who may be accused was there.” But all the judges agreed that the rule that a jury should be advised not to convict on the unsupported testimony of an accomplice was merely a rule of practice, and not a rule of law, and that the Court of Criminal Appeal, having power to decide questions of law only, could not interfere. Regina v. Stubbs, Dearsly, 555; S. C. 7 Cox C. C. 48.

In Taylor oh Evidence (3d ed.) 796, the general rule is thus stated: “ The degree of credit which ought to be given to the testimony of an accomplice is a matter exclusively within the province of the jury. It has sometimes been said they ought not to believe him unless his evidence is corroborated by other evidence; and, without doubt, great caution in weighing such testimony is dictated by prudence and reason. But no positive rule of law exists upon the subject; and the jury may, if they please, act upon the evidence of an accomplice, even in a capital case, without any confirmation of his statement. It is true that judges in their discretion will advise a jury not to convict a prisoner upon the testimony of an accomplice alone, and without corroboration, and the practice of giving such advice is now so general, that its omission would be deemed a neglect of duty on the part of the judge. Considering too the respect which is always paid by the jury to this advice from the bench, it may be regarded as the settled course of practice not to convict a prisoner, except under very special circumstances, upon the sole and uncorroborated testimony of an accomplice. The judges do not, in such cases, withdraw the case from the jury by positive directions to acquit, but only advise them not to give credit to the testimony." It may be observed that this statement is borrowed, almost word for word, from Professor Green leaf’s treatise published many years before. 1 Greenl. Ev. § 380.

In Regina v. Boyes, 9 Cox C. C. 32; S. C. 1 B. & S. 311, 7 Jur. (N. S.) 1158; which came before the Court of Queen’s Bench on a motion for a new trial, Chief Justice Cockbum quoted this passage from Taylor on Evidence with approval, and said of the judge’s instructions to the jury: “ If he told them the practice was generally not to act on the evidence of an accomplice without being confirmed, but if the evidence made out to their, minds that he was speaking the truth they ought to believe him, I think his direction was right. I protest against its being the duty of the judge to direct the jury to acquit because the evidence of an accomplice is uncorroborated.” 9 Cox C. C. 35, 36. And Justices Wightman, Crompton, Hill and Blackburn were of opinion that the application of the rule was a matter of discretion with the judge before whom the trial was had, depending upon the circumstances of the case, the nature of the crime, and the extent of the complicity of the witnesses; 1 B. & S. 320-322; 7 Jur. (N. S.) 1161, 1162; in this respect substantially concurring with the opinions of the twelve judges of England in the ease of Atwood, already cited, and of all the Irish judges in Rex v. Sheehan, Jebb, 54.

The leading case in this court upon the subject is Commonwealth v. Bosworth, 22 Pick. 397, decided in 1839, in which, on the trial in the Court of Common Pleas of an indictment for larceny, an accomplice in the commission of the crime testified in behalf of the Commonwealth, and the defendant, being convicted, brought the case to this court upon two exceptions to the admission of evidence, which were as follows:

First. The accomplice, being asked by the defendant’s counsel, upon cross-examination, whether he had not been promised indemnity from prosecution and a reward in money if he would become a witness for the Commonwealth, gave an account of several interviews between himself and a deputy-sheriff and the magistrate before whom the preliminary examination was had, and testified that the officer and the magistrate had given him assurances that he would not be prosecuted if he would tell all he knew in regard to the transaction, and should be paid the sum of $100 if he would tell where certain stolen cloth was, provided that was all he knew about the matter. The district attorney, as the bill of exceptions stated, “ claiming a right to corroborate the testimony of this witness as to all matters about which he had been properly examined, in order to support his genera] credit,” called the officer and the magistrate as witnesses, who “testified to the same assurances and to the circumstances and conversations which took place at the several interviews between them respectively and the accomplice. To this testimony the' defendant objected; but the court overruled the objection, and admitted the evidence.”

Secondly. “ The defendant, for the purpose of impeaching the testimony of the accomplice, introduced in evidence a letter from him to the defendant, in which he admitted that his testimony in relation to this case, on a former occasion, was false. To explain this evidence, and to show that the letter had been obtained unfairly, the district attorney asked the witness a variety of questions, in answer to which he stated, among other things, that this letter was a part of a correspondence which had been carried on in prison, after he and the defendant had been confined there. He also stated the means by which the correspondence had been carried on, the situation and the relative position of the several rooms, and the arrangement of the prisoners therein at different times while the correspondence was carried on. The district attorney, in order to corroborate the testimony of the accomplice, for the purpose of supporting his general credit, then called the sheriff and jailer to prove that the situation of the rooms and arrangement of the prisoners therein corresponded with the account given by the accomplice. To the admission of this evidence the defendant objected; but the court overruled the objection, and the evidence was admitted.”

Mr. Justice Morton, who delivered the opinion of the court, began by affirming, and supporting with references to authorities, these two propositions:

“ 1. It is competent for a jury to convict on the testimony oi an accomplice alone. The principle which allows the evidence to go to the jury necessarily involves in it a power to believe it. The defendant has a right to have the jury decide upon the evidence which may be offered against him; and their duty will require of them to return a verdict of guilty or not guilty, according to the conviction which that evidence shall produce in their minds.”

“ 2. But the source of this evidence is so corrupt, that it is always looked upon with suspicion and jealousy, and is deemed unsafe to rely upon without confirmation. Hence the court ever consider it their duty to advise a jury to acquit, where there is no evidence other than the uncorroborated testimony of an accomplice.”

He then proceeded to discuss and define the nature and extent of such corroboration, thus: “ 3. The mode of corroboration seems to be less certain. It is perfectly clear that it need not extend to the whole testimony; but it being shown that the accomplice has testified truly in some particulars, the jury may infer that he has in others. But what amounts to corroboration ? We think the rule is, that the corroborative evidence must relate to some portion of the testimony which is material to the issue.” The meaning of this is made clear by what immediately follows: “ To prove that an accomplice had told the truth in relation to irrelevant and immaterial matters, which were known to everybody, would have no tendency to confirm his testimony involving the guilt of the party on trial. If this were the case, every witness, not incompetent for want of understanding, could always furnish materials for the corroboration of his own testimony. If he could state where he was born, where he had resided, in whose custody he had been, or in what jail or what room in the jail he had been confined, he might easily get confirmation of all these particulars. But these circumstances having no necessary connection with the guilt of the defendant, the proof of the correctness of the statement in relation to them would not conduce to prove that a statement of the guilt of the defendant was true.”

Taking the whole paragraph together, it is manifest that the phrase “material to the issue ” is used as equivalent to “ involv ing the guilt of the party on trial,” or “ having necessary connection with the guilt of the defendant.” This interpretation receives support from the reference, at the end of the paragraph, to the case of Rex v. Addis, 6 Car. & P. 388, in which Mr. Justice Patteson said, “ The corroboration of an accomplice ought to be to some fact or facts, the truth or falsehood of which go to prove or disprove the offence charged against the prisoner.”

Having thus stated his general propositions, Mr. Justice Morton added: “4. But these principles, though plain, are not always easy of application. Questions of competency are so numerous and various, are distinguishable from each other by such nice shades of difference, and many of them come so near the line, that it oftentimes is extremely difficult to determine whether they fall on the one side or the other.”

He then announced the decision of the court upon the particular questions before it, in the following terms: “ The inquiries of the accomplice by the defendant’s counsel, whether he had been offered a reward or promised an indemnity, were relevant questions, and the answers to them became material evidence. We are therefore inclined to think that the testimony in confirmation of these answers was admissible. But this can scarcely be brought within the line ; and we are of opinion that the testimony of the sheriff and jailer, as to the location of the rooms in the jail and the situation of the prisoners, &c., falls on the other side of the line. We cannot perceive how the circumstance that the witness told the truth about these public and common objects, concerning which he knew that proof was at hand, has any tendency to confirm the material parts of his testimony, involving the guilt of the defendant. We think the Court of Common Pleas erred in the admission of this evidence. And although there is very little reason to suppose that it had any influence upon the minds of the jury, yet, as it cannot be known that it had none, and as this is a criminal case, we feel bound to order a new trial.”

The evidence which the court was “ inclined to think admissible,” though it could “ scarcely be brought within the line,” was the testimony in confirmation of the answers of the accomplice to the inquiries made of him on cross-examination, whether he had been offered a reward or been promised an indemnity. This evidence was doubtless “ admissible ” to affirm or prop up the credit of the accomplice, as of any other witness whose credit had been attempted to be impeached. Commonwealth v. Ingraham, 7 Gray, 46. But as to its being “ scarcely within the line” of which the judge was speaking, namely, the line that divides evidence which is sufficient from that which is insufficient to establish such corroboration of an accomplice as will make it safe for a jury to convict, it is difficult to see how it could be brought within that line at all; for how could the evidence of the circumstances as to the promise of reward and of indemnity tend to render the accomplice more worthy of belief, than if no such inquiry had been put to him on cross-examination ? But, if it did come within the line in question, it was not because it tended to confirm him as to a fact which did not tend to connect the defendant with the crime, but because it related to the degree of bias under which the accomplice testified, which would affect the credit of his whole testimony, and no less that part which directly bore upon the defendant’s guilt than the other parts.

To construe the hesitating expression of opinion in favor of the admissibility of the evidence concerning the offer of reward and promise of indemnity, as warranting the admission, for the purpose of establishing the corroboration of an accomplice, of testimony to facts not connecting the defendant with the crime, would not only be wholly inconsistent with the test previously stated under the third head of the opinion, but would be in direct conflict with the judgment sustaining the other exception and granting a new trial for the very reason that the court could not perceive that the evidence admitted as to the location of the rooms at the jail and the situation of the prisoners had “ any tendency to confirm the material parts of his testimony, involving the guilt of the defendant.”

It thus appears that the decision in Commonwealth v. Bosworth establishes two points: 1st. That if any evidence is admitted as competent by way of corroborating an accomplice so as to make it safe for the jury to convict, which is not legally entitled to that effect, it is a subject of exception and ground for a new trial; 2d. That no evidence can be legally admitted as competent and s ifficient for such corroboration, which does not tend to confirm the testimony of the accomplice upon a point material to the issue, in the sense that it tends to prove the guilt of the defendant.

That case did not present any question of the form of instructions to the jury upon the effect of the whole evidence in the case, but only as to the competency and sufficiency of particular portions of evidence for the purpose under consideration. Subsequent cases in this court have presented both classes of questions, and it will be convenient to state the decisions on each class separately.

In the instructions which Mr. Justice Metcalf, in Commonwealth v. Brooks, 9 Gray, 299, said “ conformed to the settled law,” and in those which Mr. Justice Wells, in Commonwealth v. Snow, 111 Mass. 411, declared to be unexceptionable, the jury were told that the evidence, to be corroborative, must be of facts which connected the defendant with the commission of the crime charged. But, as in each of those cases the defendant was convicted, and no exception could be taken by the Commonwealth, it is evident that the only point really adjudged was that the instructions were sufficiently favorable to the defendant.

By the instructions in Commonwealth v. Price, 10 Gray, 472, to which the defendant was held to have no legal ground of exception, the jury were told that, the evidence of the accomplices being unsupported by any corroboratory evidence, it was unsafe, on account of its corrupt and suspicious source, to convict upon it without confirmation; and the jury were advised to acquit: but that, nevertheless, it was competent for them to convict upon the uncorroborated testimony of 5"he accomplices alone; and if, upon the whole evidence, they were convinced beyond a reasonable doubt of the guilt of the defendant, their verdict should be guilty, otherwise not guilty. There being no corroborative evidence whatever, no question was there presented as to how far such evidence, if offered, must go.

But in Commonwealth v. O’Brien, 12 Allen, 183, the instruction, to which this court held the defendant to have no ground of exception, was merely that, unless the testimony of the accomplice was corroborated upon a material point, the defendant was entitled to an acquittal. And in Commonwealth v. Larrabee, 99 , the judge instructed the jury that, although he could not say to them that they could not convict upon the uncorroborated testimony of an accomplice, yet it was not safe for them to do so; and that, before considering the testimony of an accomplice, they ought to see whether he was corroborated; that it was not necessary for his whole story to be corroborated, but that corroboration in some material particular was sufficient; and exceptions to these instructions were overruled. It is to be observed, however, that in neither of these two cases had the judge been specifically requested to instruct the jury that the accomplice must be corroborated in some point tending to prove the defendant’s guilt.

In Commonwealth v. Scott, 123 Mass. 222, there was evidence tending to corroborate the testimony of the accomplice which went to connect the defendants with the crime, as well as the other parts of his testimony; and the judge instructed the jury that the testimony of the accomplice should be scrutinized with extreme caution, and that it was not safe or prudent to convict upon the evidence of the accomplice alone, unless he was corroborated in important and material respects in matters vital to the issue in the case. The point adjudged was, that there was no established rule of law which required the judge to advise the jury to acquit, unless there was corroboration of the statements of the accomplice connecting the defendants with the crime; and that the defendants therefore had no ground of exception to the instruction given, or to a refusal to instruct the jury that the corroboration required of the accomplice was not the corroboration of that part of his story which related to his own acts and declarations, but corroboration of that part of his story which connected the defendants with the crime.

We have no doubt of the correctness of that adjudication. The opinions of the twelve judges of England in Atwood’s case and of the Court of Criminal Appeal in Stubbs’s case, and the general current of authority in England, as we have already seen, sustain the position that the refusal of a judge to advise the jury that it is not safe to convict on the testimony of an accomplice, unless corroborated as to his statements connecting the defendant with the commission of the crime, though a departure. from the usual practice is not such an error in law as to authorize an appellate court, having, in the form in which the case is brought before it, jurisdiction of questions of law only, to set aside the verdict. And the decisions of the greatest weight in this country are to the same effect. State v. Haney, 2 Dev. & Bat. 390. People v. Costello, 1 Denio, 83. State v. Potter, 42 Vt. 495. State v. Litchfield, 58 Maine, 267. Carroll v. Commonwealth, 84 Penn. St. 107.

But the question how far a defendant has a right of exception to the refusal of a judge, in submitting the whole case to the jury, to advise them in a particular form as to the amount of corroboration which will make it safe for them to convict, is wholly different from the question of the right of the defendant to except to the admission of evidence, against his objection, for such a purpose as to attribute to it an effect to which it is not by law entitled.

The decision in Commonwealth v. Bosworth has for forty years been treated as settling that, if evidence is admitted for the pur pose of so far corroborating the testimony of an accomplice as to make it safe for a jury to convict, which is not legally to be considered as corroborative in that sense, the error may be revised by bill of exceptions. Commonwealth v. Desmond, 5 Gray, 80. Commonwealth v. Savory, 10 Cush. 535, 538. Commonwealth v. Larrabee, 99 Mass. 413, 416. And we are not aware of any case in which evidence which fell short of proving such acts or admissions of the defendant, or such participation with the accomplice at some stage of the transaction, as tended to prove the defendant’s guilt, has been held by this court to be legally sufficient to constitute such corroboration.

In this Commonwealth, indeed, as in England, evidence which tends to prove the guilt of the defendant is sufficient by way of corroboration, although it does not directly confirm any particular fact stated by the accomplice; as, for instance, evidence of the possession of stolen goods by one indicted for stealing ór receiving them. Commonwealth v. Savory, 10 Cush. 535. Rex v. Wilkes, 7 Car. & P. 272. Regina v. Birkett, 8 Car. & P. 732. Regina v. Mullins, 3 Cox C. C. 531. So where the defendant attempted to prove an alibi, and there was evidence tending to show his presence at the time and place of the commission of the crime, and it appeared that his brother, who, the accom plice had testified, was present when the defendant confessed facts showing his participation in the crime, was not called by the defendant as a witness. Commonwealth v. Brooks, 9 Gray, 299. See also People v. Dyle, 21 N. Y. 578; State v. Wolcott, 21 Conn. 272.

So it has been held that evidence that the defendant said, to the officer who arrested him, that the accomplice had nothing to do with the robbery, warranted the inference that the defendant knew the circumstances attending it and the persons who were engaged in its commission, and that this knowledge was derived from his own participation in the crime ; and therefore tended in some degree to corroborate in a material particular the testimony of the accomplice as to the complicity of the defendant. Commonwealth v. O’Brien, 12 Allen, 183.

In Commonwealth v. Larrabee, 99 Mass. 413, the testimony which was ruled to be sufficient, if believed, to constitute corroboration in a material particular, was of two kinds': 1st. Evidence tending to show that the defendants and the accomplice were seen driving together on a certain day in apparent intimacy towards the place where the horses were stolen on the evening of that day, and were also seen together at another place with the horses on the next day, was held to tend materially to- corroborate the accomplice, because it proved joint action at some stages of the transaction, and tended to show the participation of the defendants in the larceny; 2d. The defendants having introduced their own and other testimony tending to disprove the whole story of the accomplice, evidence to contradict their testimony as to the place where they met the accomplice was held admissible, because it went to show the falsehood of the defendants’ own testimony, and indeed of their whole defence, which was a most material particular directly tending to prove their guilt.

In Commonwealth v. Elliot, 110 Mass. 104, the testimony which was ruled at the trial of an indictment for breaking and entering a building, and held by this court, to be to a material fact tending to corroborate the accomplice, was admitted in connection with testimony that the offence was committed about midnight, and was that the defendant and the accomplice were seen together going towards the place just before midnight, and re turning an hour or two afterwards. And in Commonwealth v. Snow, 111 Mass. 411, there was independent evidence of various acts and statements of the defendant himself, tending to prove that he had committed the crime.

In Commonwealth v. Scott, 123 Mass. 222, no exception was taken, at the time of the introduction of any portion of the evidence, to its admissibility or effect for the specific purpose of corroborating the accomplice; and the dicta in the opinion, which might, taken by themselves, seem to support the theory that evidence may properly be held competent as corroboration of an accomplice, in the sense of rendering it safe and prudent for a jury to convict, which does not tend to connect the defendant with the crime, were not requisite to the decision, and appear to the court, upon further consideration, to be based upon too limited a view of the judgment in Commonwealth v. Bosworth, for the reasons that we • have already stated at large in commenting upon that case. >

In Commonwealth v. Drake, 124 Mass. 21, in which a woman was indicted for procuring an abortion, and denied (as the opinion assumes, by her own testimony) that the woman upon whom the abortion was committed, or her companion, Wyman, was ever in the defendant’s house, as Wyman had testified; and it was held that, even if Wyman was an accomplice, evidence from other sources that the two did go there, and that Wyman accurately described the interior of the house, corroborated Wyman in a material point; it was a material fact, bearing on the defendant’s connection with the crime, that the woman on whom the illegal operation was performed, and whom the defendant had testified to have never been in her house, had been there.

In the case at bar, the indictment charges the defendant with burning a barn and shed in Great Barrington, in the night-time of September 11, 1878. At the trial, a youth, who was admitted to have been an accomplice, was called as a witness, and testified, in substance, that on the evening of Saturday, September 2, he was at certain places with other persons whom he named in Great Barrington, and on his way home alone about eleven o’clock discovered the defendant attempting to set fire to the buildings, but they were not burned that night; that the defendant then said to him that the buildings would be burned on the next Saturday or Sunday night, and that, if he would keep silence, and would be where he could have witnesses to prove where he was during those nights, and, after the burning, would give out that he burned the buildings himself, the defendant would make it all right with him; that they then separated, and the witness went directly home to his father’s house, and arrived there and was let in by his father a little after midnight, and went to bed; that about the middle of the following week the defendant met him, and gave him four ten-dollar bills; that on the morning of Monday, September 11, the witness heard that the buildings had been burned; that several weeks after the fire the defendant made statements to him, amounting to admissions of guilt; and that once, in the following spring, when the defendant and one Warner were working on some railroad ties, the witness spoke to the defendant, but could not remember what was said.

The evidence offered in corroboration of this witness was merely that he was at the other places and with the persons whom he named in Great Barrington on the evening of September 2, and returned to his father’s house a little after midnight, and his father got up and let him in, and he went to bed, that on Sunday, September 10, he showed his brother four ten-dollar bills, and on several occasions, some days after the fire, he was seen to have in his possession one or more ten-dollar bills; and that, in the following spring, at the time when the defendant and Warner were at work on the railroad ties, the defendant and the accomplice were seen standing a few feet off, and the witnesses thought they talked together, but did not know anything they said.

The whereabouts of the witness, when not in the defendant’s company, on the evening of September 2, was wholly immaterial. His possession, on September 10, the day preceding the night of the fire, and at other -times afterwards, of bank-bills, (corresponding in number and amount with those that he testified to having received from the defendant, but not otherwise shown to have ever been in the defendant’s possession,) even if it tended to show that the witness had been hired to commit the crime, had no tendency to prove that the defendant was the person who so hired him. Hone of the testimony offered in cor roboration of the accomplice had the slightest tendency to prove any facts connecting the defendant with the crime charged, or any participation of the defendant with the accomplice at any stage of the transaction, or would have been competent independent evidence against the defendant for any purpose. Except as to the part relating to the standing and talking together in the following spring, (which hardly proved anything,) it related wholly to facts not shown to have occurred in the presence of, or to have been known to, the defendant; and, therefore, gained no weight as corroborative evidence from the fact, stated in the bill of exceptions, that the defendant, testifying in his own behalf, denied his guilt, and denied all the statements of the accomplice as to the interviews before the fire and the payment of money to him by the defendant, and all his statements connecting the defendant in any manner-with the fire.

The bill of exceptions shows that all the evidence in question was offered in corroboration of the accomplice, and was, before its admission, objected to by the defendant as immaterial, incompetent, and not admissible to corroborate the accomplice, and as not tending to corroborate him in any material point; that its admission was accompanied by a ruling of the judge, in the hearing of the jury, which would naturally, if not necessarily, be understood by them to declare that all the evidence so admitted ought to be considered as so corroborating the testimony of the accomplice in its material parts as, if believed, to make it safe and proper for the jury to convict the defendant; and that to such admission of the evidence the defendant excepted. The objection to the effect allowed to the evidence by way of corroboration is much more distinctly and fully reserved' in this case than by the exception which was sustained by this court in Commonwealth v. Bosworth.

It is proper to add, that Mr. Justice Morton finds himself unable to concur with the other judges in the view above taken of the cases of Commonwealth v. Bosworth and Commonwealth v. Scott. But the court is unanimous in the opinion that the evidence objected to did not tend to corroborate the accomplice in any material point, and that the

Exceptions must be sustained.

1788 in the text of 1 Leach, 464, but clearly shown to be a mistake by referring to 1 Leach, 478, 479, and 7 T. B. 609