¶1The authorities all agree that the testimony of a deceased witness may be given in evidence on a subsequent, trial of the same cause; but they differ upon the question raised in this case, it being held in some, that the person -called to prove what the deceased witness said, must undertake to repeat it all with literal exactness; in others, that it is sufficient if he can state the whole substance.' We are not aware that the point has been distinctly raised and settled in this State.
¶2*373The cases in England are not numerous, and none of them very early, in which questions relating to the admissibility of such evidence, have been discussed ; which may be owing in part to the fact that the practice of granting new trials on rules and cases reserved, is of comparatively modern origin ; and in the ancient practice, repeated trials of the same cause were extremely rare, and would still seem to be much less frequent in that country than with us. Rex v. Woodbray, 8 Term. Rep. 622; Bright v. Eynon, 1 Burr. 398.
¶3Breckworth’s case, Sir T. Raymond, 170, in which the general admissibility of such evidence appears to have been conceded, though Kelyng, C. J., dissented on a special ground, is of no importance to the present inquiry.
¶4In Pike v. Crouch, 1 Lord Raymond, 730, it is said “If a man was sworn as a witness at a former trial, and gave evidence and died, the matter that he deposed at a former trial, may be given in evidence at another, by any person who heard him swear it at a former trial.” There is perhaps nothing in the language here used that can be regarded as decisive of the present question. But when the court say that any person, who heard the witness testify, may prove the matter that he deposed, we can hardly suppose they undérstood that they were laying down a rule requiring an exact verbal recital of the former evidence, a thing so difficult as to be practically impossible in a vast majority of cases.
¶5In Coker v. Farewell, 2 P. Williams, 563, the question under discussion was, whether an issue directed out of Chancery, should •be sent to another trial. A witness who had once testified before the jury, had died, and this was one of the facts urged on the application for a new trial. The court say, “ The testimony which the witness had given on the former trial might be given in evidence again between the same parties.” Here it would seem to have been taken for granted, that the party who had called the witness before, would be able to give competent evidence of his former testimony, which the court would hardly have assumed as matter of course, if the rule had been supposed to require a degree and *374kind of proof, which it would seldom be in the power of the party to produce.
¶6The King v. Joliffe, 4 Term. Rep. 285, is the authority which has been principally relied on to support the strict rule, requiring the literal statement of the former evidence. That was a rule to show cause, why an information should not be filed against the defendant for improper conduct in regard to a trial at Nisi Prius. The question before the court was, whether affidavits used at Nisi Prius on an application to put off the trial, were admissible against the defendant in the hearing on this rule to show cause against filing the information. Lord Kenyon, in delivering, his opinion on that point, is reported to have said, “ The evidence which the witness gave on a former trial, may be used in a subsequent one, if he die in the interim, as I remember was agreed on all hands on a trial at bar, in the instance of Lord Palmerston; but as the person, who wished to give Lord Palmerston’s evidence, could not undertake to give his words, but could merely swear to the effect of them, he was rejected.” There is nothing else in the case, that bears on the point we are considering.
¶7In The King v. Joliffe, there was no question as to the admissibility of the testimony which a deceased witness had given on a former trial. The subject was merely alluded to by one of the judges in the way of illustration, and his brief and very unsatisfactory statement of the ruling on trial in an anonymous case, where the former testimony of Lord Palmerston was attempted to be proved, was the only authority given for the rule stated. We are not informed how the question arose in that case, nor what was the nature of the fact to which Lord Palmerston had testified.
¶8On the trial of Ennis v. Donnisthorne, Corn. Sum. Ass. 1789, Lord Kenyon is said to have ruled that the witness ought to recollect the very words, “ for the jury alone can judge of the effect of words ; ” and to have cited from his own words the case of The King v. Deborah, 1 Phillips, Ev. 231, and note (3}.
¶9These three rulings of the point on trial, of which we have seen no reports, except the loose statements that have been referred to, are the only English cases we are aware of, that give any *375countenance to the rule, requiring a literal recital of the testimony given by the deceased witness. We have met with nothing to show that the rule has in England been acquiesced in and applied in practice. The subsequent cases, in which the evidence of deceased witnesses has been proved, would rather lead us to infer the contrary.
¶10In Doe v. Passingham, 2 Car. & Payne, 440, a witness was allowed to prove that a deceased witness, on a trial which took place thirty-two years before, had deposed to a declaration made by Gwin Lloyd, relative to the legitimacy of the plaintiff’s mother. No question is reported to have been made as to the manner of proving the former evidence; but it is extremely improbable that after the lapse of thirty years a witness could undertake to repeat the words of the former testimony.
¶11In Todd v. The Earl of Winchelsea, 3 Car. & Payne, 387, the testimony of the deceased witness was read from the notes of a short-hand writer. The evidence of the deceased witness was to prove the situation of the rooms in a house where a will was executed, and from its nature is not likely have been brief and simple in its character, such as would allow the short-hand writer to take down the whole, literally.
¶12In The King v. Whitehead, 1 Car. & Payne, 67, the evidence of a witness for the prosecution, examined on a former trial, who had since died, was read from the Lord Chief Justice’s notes, by order of the Court above. It is quite difficult to understand how such an order could have been made, if it were understood that the evidence was incompetent, unless the words of the deceased witness were exactly stated. It is possible that the new trial may have been granted upon condition that this evidence should be admitted; but no such ground for receiving the evidence is reported in the case.
¶13The rule has not been adopted and approved by respectable English writers on the law of evidence. On the contrary, it is sharply criticised by Phillips, and its application limited to cases where the words of the deceased witness were material, and of course belonged to the substance of his testimony. 1 Phillips, Ev. 231.
¶14*376And Starkie, instead of laying down this as the rule of law, merely states the ruling in Lord Palmerston’s case, and, in a note, questions its correctness. 2 Starkie on Ev. 280, and note 00-
¶15On this state of the authorities in England, it cannot he said that any rule has been established there, by deliberate decision or settled practice, which requires the exact words of the deceased witness to be proved. In this country the decisions have been conflicting. In Massachusetts, the strict rule was adopted in Commonwealth v. Richards, 18 Pick. 434, and affirmed in Warren v. Nichols, 6 Met. 261, and must be considered as established in that State. The authority of these cases is, however, much weakened by the strong dissenting opinion of Hubbard, J., in Warren v. Nichols.
¶16So it has been held in Indiana, that the words of the former testimony must be proved. Ephraims v. Murdock, 7 Blackf. 10.
¶17In New York, the substance of the former evidence would seem to have been considered as admissible in some of the earlier cases. Jackson v. Bailey, 2 Johns. Rep. 17; White v. Kibling, 11 Johns. Rep. 128.
¶18In Wilber v. Selden, 6 Cowen, 162, 165, the evidence was rejected because the witness could not state the exact words of the former evidence, and this rule appears to have been recognized, though the point was not raised, in Crary v. Sprague, 12 Wendell, 41. But the later case of Clark v. Vorce, 15 Wendell, 195, shows that the question was then to be regarded as unsettled in New York. In that case the Chief Justice says, “ If nothing will answer but an exact transcript of the testimony of the witness, in his very words, and all his words, it will exclude all such1 testimony. There are few or no cases, where a cautious and prudent man will swear that his notes of testimony of a witness, taken down at the time, contain his very words, and all his words.”
¶19In Pennsylvania, it was held, in the Circuit Court of the United States, on the trial of United States v. Wood, 3 Wash. C. C. R. 440, that the exact words must be proved; and so in Lightner v. Wike, 4 Serg. & Rawle, 203. But a series of subsequent de*377cisions have established the contrary rale in that State. Cornell v. Green, 10 Serg. & Rawle, 16 ; Wolf v. Wyeth, 11 Ib. 149 ; Watson v. Gilday, 11 Ib. 337; Chess v. Chess, 17 Ib. 409; Moore v. Pearson, 6 Watts & Berg. 51.
¶20In Ohio, it appears that the practice on this point had not been uniform in the Circuit Court of that State; but in Wagers v. Dickey, 17 Ohio, 439, the more liberal rale was established.
¶21In Maryland, it is sufficient to prove that the deceased witness swore to certain facts, without proving the precise words. Garrott v. Johnson, 11 Gill. & Johns. 173.
¶22It is held sufficient to prove the substance of the former testimony in Vermont, New Jersey, Virginia, North Carolina, Alabama, and Mississippi. State v. Hooker, 17 Vermont, 658; Sloan v. Somers, 1 Spencer, 66; Caton v. Lenox, 5 Randolf, 31; Ballenger v. Bames, 3 Dev. 460 ; Gildersleve v. Carraway, 10 Ala. 260; Smith v. Steamboat, 1 Howard, Miss. 479.
¶23The rule which admits the substance of the deceased witness’ testimony is approved in 1 Greenleaf’s Ev. § 165, and in Cowen & Hill’s note, No. 441, to 1 Phil. Ev. 231, and on the whole the weight of American authority appears to be decidedly in favor of that rule.
¶24The rule which requires an exact recital of all the words used by the deceased witness, has been defended on the ground that there is danger in allowing a witness to state what he understands to be the substance of the former evidence; that the jury ought to hear the precise language, and draw their own conclusions ; that the witness is inclined to give his own impressions and inferences, instead of the real substance of the testimony.
¶25An exact recital, if it can be had, is doubtless preferable to any statement of the substance. So the testimony of the witness himself on the stand is far more satisfactory than the most literal repetition of his former testimony; and if the witness is living and can be produced, neither the substance nor the exact words of what he testified before can be received. The evidence is not admitted, whether in the words or in the substance, because in its general character it is the best; but because it is the best, that, in the particular instance, the party is able to produce. It is secondary evi*378dence; and allowed because the best is lost. It is conceded that the former testimony may be proved, though not in its nature the best evidence ; and the question is, whether a strictness of proof shall be required, which in most cases it is quite impossible to produce ; for it is manifest that to insist on a literal repetition of all that the deceased witness testified, would exclude the evidence in a great majority of instances. Where the former testimony has any complication or any considerable extent, no cautious and conscientious witness would take it on himself to repeat it in the exact words from memory, or from any notes that could possibly be taken. To hold a rule so stringent would be likely to encourage rash and unscrupulous witnesses to undertake an exact recital of the evidence,.and exclude the cautious and guarded statement of others, who were conscious of the extreme difficulty of such a task, and would venture to give no more than the substance. It is at least doubtful, whether a witness who would undertake for a thing so difficult, not to say impossible, as a literal recital of the former testimony would deserve more credit than another, who would attempt no more than a statement of the substance. A witness,'who testifies to a conversation or declaration, which is not very brief and simple, is usually understood to throw something of discredit on his whole story, by professing to repeat exactly all that was said, because the substance is 'all that the best memory can ordinarily retain; and to attempt more leads to a just suspicion that the witness is willing to go beyond his knowledge and recollection.
¶26In other analogous cases, where evidence is given of a conversation, an admission, a verbal contract, or the contents of a writing, the substance is generally held to be sufficient.
¶27In an indictment for perjury, it is not necessary to prove with verbal exactness the testimony of the defendant. Rex v. Meenton, 3 Car. & Payne, 498.
¶28'The substance of dying declarations in a prosecution for murder is sufficient. Montgomery v. The State, 11 Ohio, 424.
¶29So where a deed or other writing is lost, and the exact contents cannot be proved, the substance is sufficient. 1 Greenl. Ev. § 84, note. Renner v. Bank of Columbia, 9 Wheaton, 581.
¶30*379A verbal bargain may be proved by showing the substance of what was said. In such case, the witness, if he do not recollect the language used, may even say what he understood to be the effect and result of the conversation. Eaton v. Rice, 8 N. H. Rep. 378 ; Maxwell v. Warner, 11 N. H. Rep. 568.
¶31The weight of authority appears to be in favor of the more liberal rule ; and on general principles and grounds of convenience, we are lead to the same conclusion.
¶32There must be
¶33Judgment on the vei'diot.