The question is, whether or no the circuit court erred in rejecting the evidence of William» King’s widow, as to declarations of her husband in his lifetime? And I have seldom examined a question, about which my mind felt at first more doubt, not about the general doctrine, but as to its application to this particular case.

It is well settled, “that husband and wife cannot be witnesses for each other, because their interests are identical, nor against each other, on grounds of public policy, for fear of creating distrust, and sowing dissension between them, and occasioning perjury.’’ It would also violate that confidence, which from the nature of the relation ought to be regarded as sacred, and W'ould be arming each of the parties with the means of offence, which might be used for very dangerous purposes. So important is this rule, that the law will not suffer it to be broken even by agreement; for in Barker v. Dixie, Rep. Temp. Hardw. 264, lord Hardwicke would not suffer a wife to be examined against her husband, though he consented that she might. And this rule is adhered to, though the marriage tie be dissolved by the death of one of the parties, or by a divorce. Aveson v. Kinnaird, 6 East. 192; Monroe v. Twisleton, Peake’s ev. app. lxxxvii. It is very clear from these and other cases, that where either is a party, the other is altogether incompetent.

The difficulty with me, arises from the fact, that in the case before us, neither the husband, nor his representative is a party; nor would the judgment in this case, as I apprehend, be evidence in a suit by the vendee of the slave against the representative of the husband, further than to shew the fact of the recovery. Still this case comes within the class treated of by Starkie in his law of evidence (part IV, p. 708,) under the second head, viz. “Where one of them (husband and wife) not being a party, is interested in a proceeding ^between others.’’ For here, though a judgment for the plaintiffs would not charge the representative, it would lead inevitably to an action against him, while a verdict for the defendant would render him secure. The husband then if alive would feel an interest in this case. But I find several cases in the books, which say that it is not such an interest, as of itself will render the wife incompetent. Thus, in Williams v. Johnson, 1 Stra. 504, the action was against the daughter’s husband for her wedding clothes: defence, that the goods were furnished on the credit of her father: and to prove this, the mother who was present at the choosing of the goods, was called to charge her husband, and allowed. And this case has not only never been overruled, but other cases are founded on it, and I see it referred to as good law in the latest books. It is this class of cases which has puzzled me. But upon reflection, I think I see a distinction between them and the case at bar, which becomes material, when we look at the reason of the rule, which disqualifies husband and wife from giving evidence against each other. It is not merely the interest, nor indeed, chiefly the interest of the parties, that thus disqualifies them; but the near relation in which they stand, and the unbounded confidence so essential to that relation, that the law will not suffer it to be violated. Tnus, in Aveson v. Kinnaird, where the declarations of a wife (dead at the time of trial) were sought to be used against the husband, lord Ellen-borough, among the reasons for admitting them, says, “No confidence has been violated, nothing extracted from the bosom of the wife, which was confided there by tne husband.” So, in Monroe v. Twisleton, when the wife, then divorced, and again married, was called to prove a contract against her former husband, lord Alvanley said, “to prove any fact arising after the divorce, this lady is a competent witness, but not to prove a contract or any thing else, which happened during the coverture. She was at that time bound to secrecy; what she did might be in consequence of the trust and confidence reposed in her by her husband; and *miserable indeed would the condition of a husband be, if when a woman is divorced from him, perhaps for her own misconduct, all the occurrences of his life, intrusted to her while the most perfect and unbounded confidence existed between them, should be divulged in a court of justice.” In the case of Le Texier v. Margravine of Anspach, 15 Ves. 165, the wife being made a party with her husband, and charged as his agent, and having documents in her possession, and made a party for discovery only as to her, and relief against the husband, lord Eldon said, “Before a married woman can be made a party, upon the same principle that an agent of a corporation is made a party, other principles must be broken through, long acknowledged by the law; arising out of the relation of husband and wife: a case in which it is almost impossible to represent the degree of inconvenience that will result, from the adoption of this principle; and the particular mischief must be met with the observation, that it must be endured, rather than that a general mischief should be introduced, *284•which is against all policy, and would break in intirely upon the happiness of private life.” These are cases in which the husband was a party, but the principle applies also where he is no party: for, in the one case or the other, it is equally a violation of the confidence reposed, to divulge in a court of justice, what was imparted in the sacred privacy of domestic intercourse. And of this opinion Starkie seems to be (part IV, p. 709,) when he says, “Where neither of them is either a party to the suit, or interested in the general result, the husband or wife is, it seems, competent to prove any fact, provided the evidence does not directly criminate the other, or, as it seems, involve the disclosure of some communication made by the other.” Now, the case from Strange did not violate this rule: the wife disclosed no communication : but being present when the goods were bought, she was called to prove on whose credit the sale was made. But is not our case very different? I think so. The husband was at the time holding in slavery, the mother and her children: if she was *an indian woman, they were all entitled to their freedom. Can we possibly suppose, that he meant to make such a declaration public? It is stated, that the party offered to prove that the declarations were made repeatedly in the presence of the family, and they were not requested to keep them secret. This could only (I presume) be proved by the wife; and I question the propriety of permitting her thus to qualify herself to disclose such communications. But suppose it proved, that the declarations were so made, and no secrecy injoined ; would it follow, that the husband wished or expected they should be divulged? Are we to say, that every word spoken, in the thoughtless, careless confidence of the domestic circle, is free for public disclosure, unless secrecy be expressly injoined? Is not the converse of the proposition true? And would it not have a most mischievous effect, would it not seriously break in upon that confidence which is the charm of domestic life, if men should, from our decisions, have cause to fear, that after they were in their graves, their reputation might be injured, and their children ruined, by the declarations they had made in the bosoms of their families? This freedom from restraint or apprehension, in the intercourse of one’s own fireside, seems to me so necessary to the quiet and repose of society, that I am fearful of trenching upon it in the slightest degree.

The case of Baring v. Reeder, 1 Hen. & Munf. 154, mentioned at the bar, was not like this. Baring had lent Mrs. Claiborne some personal property; her husband made Reeder an absolute bill of sale for it, but the possession remained; Baring hearing of the bill of sale, took back into his own possession the property he had lent; Reeder brought trover against him for it; and Mrs. C. was offered as a witness to prove the loan, and that she had told Reeder of it: there was no confidence violated there, nor was her husband’s interest involved; and I think with judge Roane, that he, as well as his wife, might have been examined.