Opinion
The dispositive issue in this appeal is whether, pursuant to General Statutes (Rev. to 1995) § 45a-257 (a), the trial court should have admitted extrinsic evidence regarding the decedent’s intent that *361his will would not be revoked automatically by his subsequent marriage.2 The named plaintiff, Alicia Erickson,3 who is the daughter of the decedent, Ronald K. Erickson, appeals4 from the judgment of the trial court *362in favor of the defendant, Dorothy Erickson,5 the executrix of the estate of the decedent, dismissing the plaintiffs appeal from the decree of the Probate Court for the district of Madison. The Probate Court had admitted the will of the decedent to probate. The trial court ruled that the decedent’s will, which had been executed shortly before his marriage to the defendant, provided for the contingency of marriage.
The plaintiff claims on her appeal that the trial court improperly concluded that the decedent’s will provided for the contingency of marriage.6 The defendant claims on her cross appeal that the trial court improperly excluded certain extrinsic evidence regarding the decedent’s intent. We conclude that the trial court should have permitted the defendant to introduce extrinsic evidence of the decedent’s intent. Accordingly, we reverse the judgment of the trial court and order a new trial.
Certain facts in this appeal are undisputed. On September 1, 1988, the decedent executed a will. At that time, he had three daughters and was unmarried. Two days later, on September 3, 1988, he married the defendant. He died on February 22, 1996.
The six articles of the will provide as follows.7 The first article provides for the payment of funeral *363expenses and debts by the estate. The second article states that the residue of the estate will pass to the *364defendant. The third article provides that if the defendant predeceases the decedent, one half of the residuary estate will pass in equal parts to the decedent’s three daughters, Laura Erickson Kusy, Ellen Erickson Cates *365and Alicia Erickson, and one half of the residuary estate will pass in equal parts to Thomas Mehring, Christopher Mehring, Maureen Mehring and Kathleen Mehring, the children of the defendant. The fourth article appoints the defendant as the executrix of the will, with Attorney Robert O’Brien as the contingent executor in the event that the defendant is unable to or refuses to serve as executrix. The fifth article gives the executrix or executor the power to dispose of property of the estate as necessary. The sixth article appoints the defendant as the guardian of any of the decedent’s children who have not reached the age of eighteen at the time of his death.
The Probate Court admitted the decedent’s will to probate. The plaintiff appealed from the Probate Court’s judgment. Prior to trial, the plaintiff filed a motion in limine to exclude extrinsic evidence of the decedent’s intent. The plaintiff argued that “[§] 45a-257 makes the Court’s inquiry very simple: to determine whether the will was revoked, the Court need examine only [the decedent’s] will, his marriage certificate to [the defendant], and his death certificate. Extrinsic evidence regarding [the decedent’s] intentions is inadmissible because the language of [the decedent’s] will is unambiguous, and therefore under ...[§] 45a-257 the operation of the marriage to revoke the will is automatic and mandatory.” The defendant, in opposition to the plaintiffs motion, made a detailed offer of proof to show the contrary intent of the decedent.8
*368The admission of certain evidence was undisputed, namely, the will, the marriage certificate of the decedent and the defendant, and the decedent’s death certificate. The trial court denied the plaintiffs motion in limine with respect to the evidence that Thomas Mehring, Christopher Mehring, Maureen Mehring and Kathleen Mehring, who were named beneficiaries in the will, are the children of the defendant. The court granted the motion in limine, however, with respect to any other evidence regarding the decedent’s intent.
*369With respect to the other issue at trial, namely, whether the decedent’s will provided for the contingency of his marriage to the defendant, the trial court, in a de novo proceeding, concluded that the Probate Court properly had admitted the will to probate because the will provided for the contingency of marriage. The trial court reasoned that “[the decedent’s] will bequeathed all of his estate to the woman he was licensed to marry and did marry two days later. In his will, he named her executrix and designated her the guardian of his daughters, whose mother had previously died. The nature of these provisions, coupled with the extreme closeness in time of the marriage constitutes clear and convincing evidence of provision for the contingency of marriage. It would be preposterous to assume that [the decedent] was instead executing a will to make provisions that were to be revoked two days later.” Accordingly, the trial court rendered judgment affirming the Probate Court’s judgment admitting the will, and denied the plaintiffs appeal. This appeal followed.
The plaintiff claims that because the will did not expressly provide for the contingency of marriage, the trial court improperly concluded that the decedent’s will provided for the contingency of marriage and, therefore, that his subsequent marriage did not automatically revoke his will under § 45a-257 (a). The plaintiff argues that in determining whether a will provides for contingency of a subsequent marriage, the court may consider only the language of the will, and that the will in question in this case does not include any language referring to the contingency of marriage, such as “the words ‘if I marry,’ ‘when I marry,’ ‘my future wife’ or even ‘my fiancee.’ In fact the words ‘wife,’ ‘spouse’ and ‘marry’ do not appear within the four corners of the will.”
*370In contrast, the defendant claims that the trial court improperly excluded extrinsic evidence of the decedent’s intent establishing that his subsequent marriage to her would not result in the revocation of his will.9 In the alternative, the defendant argues that there is sufficient language within the will itself, without resort to extrinsic evidence, to indicate that the decedent did not intend for his will to be revoked upon his marriage to her.
We conclude that the will, in and of itself, did not provide for the contingency of the subsequent marriage of the decedent and, therefore, under existing case law, properly would have been revoked by that marriage pursuant to § 45a-257 (a). We also conclude, however, that under the circumstances of this case, the trial court improperly excluded evidence of a mistake by the scrivener that, if believed, would permit a finding that the will provided for the contingency of marriage. We therefore reverse the judgment of the trial court and order a new trial in which such evidence may be considered by the trial court.
On the basis of existing case law, the question of whether a will provides for the contingency of a subsequent marriage must be determined: (1) from the language of the will itself; and (2) without resort to extrinsic evidence of the testator’s intent.10 Fulton Trust *371 Co. v. Trowbridge, 126 Conn. 369, 372, 11 A.2d 393 (1940) (execution of will followed by adoption of child; no language in will providing for such contingency); Strong v. Strong, 106 Conn. 76, 79-80, 137 A. 17 (1927) (execution of will followed by birth of child; no language in will providing for such contingency); cf. Czepiel v. Czepiel, 146 Conn. 439, 442, 151 A.2d 878 (1959) (language in will provided for possible subsequent marriage); Blake v. Union & New Haven Trust Co., 95 Conn. 194, 196-97, 110 A. 833 (1920) (language in will provided for subsequent birth of child). Applying this standard, we conclude that the trial court should not have admitted the will because, notwithstanding the inferences that the trial court drew from the dates of the marriage license and the will, and from the identity of certain of the named beneficiaries in the will, there was no language in the will providing for the contingency of the subsequent marriage of the decedent. Thus, this will fell on that side of the dividing line encompassing those wills that do not, in and of themselves, provide for such a contingency; see Fulton Trust Co. v. Trowbridge, supra, 372; Strong v. Strong, supra, 80; and beyond that side of the line encompassing those wills that do provide for such a contingency. See Czepiel v. Czepiel, supra, 440; Blake v. Union & New Haven Trust Co., supra, 196.
This conclusion does not, however, end our inquiry in this case. In Connecticut Junior Republic v. Sharon Hospital, 188 Conn. 1, 2, 448 A.2d 190 (1982), this court considered the issue of “whether extrinsic evidence of a mistake by a scrivener of a testamentary instrument is admissible in a proceeding to determine the validity of the testamentary instrument.” In a three to two decision, this court held that such evidence is not admissible. Id., 9. Upon further consideration, we now conclude that the reasons given by the dissent in that case are *372persuasive and apply to the facts of the present case. We, therefore, overrule Connecticut Junior Republic, and hold that if a scrivener’s error has misled the testator into executing a will on the belief that it will be valid notwithstanding the testator’s subsequent marriage, extrinsic evidence of that error is admissible to establish the intent of the testator that his or her will be valid notwithstanding the subsequent marriage. Furthermore, if those two facts, namely, the scrivener’s error and its effect on the testator’s intent, are established by clear and convincing evidence, they will be sufficient to establish that “provision has been made in such will for such contingency,” within the meaning of § 45a-257 (a).
In Connecticut Junior Republic v. Sharon Hospital, supra, 188 Conn. 9, this court reasserted the familiar rule that, although extrinsic evidence is not admissible to prove an intention not expressed in the will itself or to prove a devise or bequest not contained in the will, such evidence is admissible to identify a named devisee or legatee, to identify property described in the will, to clarify ambiguous language in the will, and to prove fraud, incapacity or undue influence. In rejecting the claim that extrinsic evidence should also be admissible to prove a scrivener’s error, the majority relied principally on our existing case law and on the risk of subverting the policy of the statute of wills. Id., 23 (Peters, J., dissenting). The majority acknowledged, however, that, as with any rule of law, time and experience could persuade to the contrary. “[Pjrinciples of law which serve one generation well may, by reason of changing conditions, disserve a later one. . . . Experience can and often does demonstrate that a rule, once believed sound, needs modification to serve justice better.” (Citations omitted; internal quotation marks omitted.) Id., 17-18. We are now persuaded to the contrary.
*373The dissent in that case by Justice Peters and joined by Justice Shea, concluded that it “would permit extrinsic evidence of a scrivener’s error to be introduced in litigation concerned with the admissibility of a disputed will to probate.” Id., 22. The dissent gave three principal reasons for its conclusion, each of which we consider to be persuasive and each of which applies to this case. See id., 22-26.
First, given that extrinsic evidence is admissible to prove that a will was executed by the testator “in reliance on erroneous beliefs induced by fraud, duress, or undue influence”; id., 22; there is no discernible policy difference between that case and a case in which “a will is executed in reliance on erroneous beliefs induced by the innocent error, by the innocent misrepresentation, of the scrivener of a will.” Id., 23. In each instance, “the testamentary process is distorted by the interference of a third person who misleads the testator into making a testamentary disposition that would not otherwise have occurred.” Id., 22-23. “In each instance, extrinsic evidence is required to demonstrate that a will, despite its formally proper execution, substantially misrepresents the true intent of the testator.” Id., 23.
Similarly, in the present case, there is no discernible policy difference between extrinsic evidence offered to show fraud, duress or undue influence, and extrinsic evidence offered to show that a scrivener’s error induced the decedent to execute a will that he believed would survive his subsequent marriage. In both instances, the testamentary process was distorted by the interference of a third person who misled the testator into executing a will that would not otherwise have been executed — in the present case, a will that would be revoked upon his marriage because it did not contain language providing for the contingency of marriage. Thus, as in the case of fraud, duress or undue influence, extrinsic evidence is required to demonstrate that the *374will that the testator executed did not substantially state his true intention.11
Second, the dissent recognized that, based on the policy of the statute of wills, the “risk of subversion of the intent of a testator, who cannot personally defend his testamentary bequest, is without doubt a serious concern.” Id., 24. The dissent, however, persuasively underscored the counterbalancing “risk of blindly enforcing a testamentary disposition that substantially misstates the testator’s true intent.” Id. Again drawing on the analogy to the case of fraud, duress or undue influence, the dissent stated that “[h]ad the decedent’s lawyer deliberately and fraudulently altered the second codicil, the relevant extrinsic evidence would unquestionably have been admitted.” Id., 25. The dissent contended that “innocent misrepresentation is treated as generally equivalent to fraud in terms of its legal consequences.” Id. Therefore, the dissent asserted, the “[statute of [w]ills does not compel enforcement of testamentary dispositions that a testator never intended to make.” Id.
Similarly, in the present case, had the decedent’s attorney deliberately and fraudulently, rather than innocently but mistakenly, misrepresented to the decedent that his will would be valid despite his subsequent marriage, it is at least arguable that the beneficiaries of that fraudulent conduct, namely, the heirs-at-law of the decedent who would inherit in the event of his intestacy, would not be permitted to take advantage of that fraud, and that a court of equity could impress a constructive trust on their inheritance. We conclude that, analogously, in this case, the extrinsic evidence should be admissible to establish the decedent’s true intent.
*375Third, the dissent examined and rejected the two main objections to the admission of extrinsic evidence of a scrivener’s error. One objection was “that whatever error the scrivener may have made was validated and ratified by the testator’s act in signing his will.” Id., 26. The dissent responded, correctly in our view, that, although “signing [a] will creates a strong presumption that the will accurately represents the intentions of the testator, that presumption is a rebuttable one.” Id. Similarly, in the present case, although the fact that the decedent signed the will may create a rebuttable presumption that he did not intend it to survive his subsequent marriage, that presumption should be rebut-table by persuasive extrinsic evidence to the contrary.
The other objection was “that allowing extrinsic evidence of mistake will give rise to a proliferation of groundless will contests.” Id. The dissent presented a two part response, with which we also agree. First, it noted that, “[i]n the law of contracts, where the parol evidence rule has undergone considerable erosion, this risk has not been found to have been unmanageable. In the law of wills, the risk is limited by the narrowness of the exception that this case would warrant . . . [namely, to] permit the opponent of a will to introduce extrinsic evidence of the error of a scrivener, and [to] require proof of such an extrinsic error to be established by clear and convincing evidence.” Id., 26-27
Similarly, in the present case, the admissibility of such extrinsic evidence, in our view, will not prove to be any less manageable than in cases of parol evidence in contract disputes. Furthermore, we would impose the same elevated burden of proof on the proponent of the will in a case such as this. The proponent would have to establish the scrivener’s error by clear and convincing evidence.
*376We recognize that the dissent’s position in Connecticut Junior Republic would have resulted in the consideration of extrinsic evidence of a scrivener’s error offered for the purpose of preventing admission of a testamentary document to probate, rather than for the purpose oí procuring such admission, as in the present case. Regardless of that distinction, however, the mistake doctrine advocated by the dissent in Connecticut Junior Republic applies equally to the present case. The dissent in Connecticut Junior Republic phrased the issue in that case as follows: “Must the true intent of the testator be thwarted when, because of the mistake of a scrivener, he has formally subscribed to a written bequest that substantially misstates his testamentary intention?” Id., 22. That is precisely the issue in the present case. The dissent in Connecticut Junior Republic answered that question in the negative, recognizing that evidence of a scrivener’s mistake should be admissible where offered to establish that a written bequest should not be admitted to probate because its execution was the product of a mistake of the scrivener and, therefore, did not embody the disposition intended by the testator. Id., 22-23. Likewise, in the present case, evidence of a scrivener’s mistake should be admissible to establish that a written bequest should be admitted to probate because the disposition provided by the bequest would have obtained, in accordance with the decedent’s intent, but for the scrivener’s mistake.
Finally, we address one other consideration that was not present in Connecticut Junior Republic, which might be perceived to be present in this case. That is the potential for past reliance by testators on our prior case law. We do not believe that this is a persuasive consideration. It is very unlikely that, in reliance on such case law, any testators executed wills and deliberately omitted language providing for the contingency of marriage in order to be sure that their wills would be *377revoked by a subsequent marriage. Moreover, it is very unlikely that any testators, having married after executing such wills, deliberately did not make new testamentary dispositions in reliance on the proposition that their prior wills had been revoked by that subsequent marriage.
Applying these principles to the facts of the present case, we conclude that the extrinsic evidence offered, if believed, could prove clearly and convincingly that there was a scrivener’s error that induced the decedent to execute a will that he intended to be valid despite his subsequent marriage. The offer of proof indicates that the evidence would be susceptible to an inference by the fact finder that there had been an implied assertion by the scrivener that the will would be valid despite the decedent’s subsequent marriage. This inference could have been bolstered, moreover, by the evidence of the conversations between the decedent and the scrivener shortly before the decedent’s death.
The judgment is reversed and the case is remanded for a new trial.
In this opinion KATZ, PALMER and PETERS, Js., concurred.