OPINION

CANTIL-SAKAUYE, C. J. —

Irving Duke prepared a holographic will providing that, upon his death, his wife would inherit his estate and that if he and his wife died at the same time, specific charities would inherit his estate. The handwritten will, however, contained no provision addressing the disposition of his estate if, as occurred here, he lived longer than his wife. The specified charities contend that at the time the testator wrote his will, he specifically intended to provide in his will that the charities would inherit his estate in the event his wife was not alive when he died. The courts below excluded extrinsic evidence of the testator's intent, finding that the will was unambiguous and failed to provide for the circumstance in which his wife predeceased him. Therefore, finding that Duke died intestate, the court entered judgment in favor of the heirs at law, Seymour and Robert Radin.

(1) We granted review to reconsider the historical rule that extrinsic evidence is inadmissible to reform an unambiguous will. We conclude that the categorical bar on reformation of wills is not justified, and we hold that an unambiguous will may be reformed if clear and convincing evidence establishes that the will contains a mistake in the expression of the testator's intent at the time the will was drafted and also establishes the testator's actual specific intent at the time the will was drafted.

(2) We further conclude that the charities' theory that the testator actually intended at the time he drafted his will to provide that his estate would pass to the charities in the event his wife was not alive to inherit the estate is sufficiently particularized, with respect to the existence of such a mistake and the testator's intent, that the remedy of reformation is available so long as clear and convincing evidence on both points is demonstrated. Therefore, we will direct this matter to be remanded to the probate court for consideration of whether clear and convincing evidence establishes that such a mistake occurred at the time the will was written and that the testator at that time intended his estate to pass to the charities in the event his wife was not alive to inherit the estate when he died.

I. FACTS

In 1984, when Irving Duke was 72 years of age, he prepared a holographic will in which he left all of his property to "my beloved wife, Mrs. Beatrice Schecter Duke," who was then 58 years of age. He left to his brother, Harry Duke, "the sum of One dollar." He provided that "[s]hould my wife ... and I die at the same moment, my estate is to be equally divided — [¶] One-half is to be donated to the City of Hope in the name and loving memory of my sister, Mrs. Rose Duke Radin. [¶] One-half is to be donated to the Jewish National Fund to plant trees in Israel in the names and loving memory of my mother and father — [¶] Bessie and Isaac Duke."

Irving1 further provided in his will that "I have intentionally omitted all other persons, whether heirs or otherwise, who are not specifically mentioned herein, and I hereby specifically disinherit all persons whomsoever claiming to be, or who may lawfully be determined to be my heirs at law, except as otherwise mentioned in this will. If any heir, devisee or legatee, or any other person or persons, shall either directly or indirectly, seek to invalidate this will, or any part thereof, then I hereby give and bequeath to such person or persons the sum of one dollar ($1.00) and no more, in lieu of any other share or interest in my estate."

The will appointed Beatrice the executrix of the estate. The only change Irving ever made to his will was the addition, in 1997, of the statement that "[w]e hereby agree that all of our assets are community property." Beatrice died in July 2002, but the will was not changed to select a new executor.

Irving died in November 2007, leaving no spouse or children. In February 2008, a deputy public administrator for the County of Los Angeles obtained the will from Irving's safe deposit box. In March 2008, two charities, the City of Hope (COH) and the Jewish National Fund (JNF), petitioned for probate and for letters of administration. In October 2008, Robert and Seymour Radin (the Radins) filed a petition for determination of entitlement to estate distribution. The Radins are the sons of Irving's sister, Rose Duke Radin, who predeceased Irving. Their petition alleged that they are entitled to the distribution of Irving's estate as Irving's sole intestate heirs.

The Radins moved for summary judgment. They did not challenge the validity of the will. Instead, they asserted that the estate must pass to Irving's closest surviving intestate heirs, the Radins, because Irving did not predecease Beatrice, nor did Irving and Beatrice "die at the same moment," and there is no provision in the will for disposition of the estate in the event Irving survived Beatrice. In opposition to the motion, COH and JNF offered extrinsic evidence to prove that Irving intended the will to provide that in the event Beatrice was not alive to inherit Irving's estate when Irving died, the estate would be distributed to COH and JNF. The probate court concluded that the will was not ambiguous, and on that ground, it declined to consider extrinsic evidence of Irving's intent, and granted summary judgment for the Radins.

The Court of Appeal affirmed, based on our opinion in Estate of Barnes (1965) 63 Cal.2d 580 [47 Cal.Rptr. 480, 407 P.2d 656] (Barnes). In Barnes, the testatrix's will provided that all of her property was to go to her husband, and if she and her husband died simultaneously or within two weeks of each other, her entire estate was to go to her nephew, Robert Henderson. Her will included a disinheritance clause, stating that "`I hereby declare that I have thought of and considered each and every person who would inherit from me had I died intestate and who is not mentioned in this Will, and I hereby declare that I do not desire to devise or bequeath to such person or persons any sum whatsoever and I hereby disinherit such person or persons.'" (Id. at p. 581, fn. 5.) The testatrix's husband predeceased her, but she did not alter her will after his death.

When the testatrix died, 13 years after executing the will, she had various heirs at law, but Robert Henderson was not an heir at law because his mother was still alive. In the heirship proceeding, Henderson's mother testified that at the time the will was executed, Henderson frequently visited the testatrix at her home and spent many holidays with her, the two had a close relationship, and the testatrix was fond of him and often introduced him as her son. She also testified that the other relatives did not visit. The trial court found the will ambiguous, admitted the extrinsic evidence, and construed it in favor of Henderson. (Barnes, supra, 63 Cal.2d at p. 582.)

We reversed the judgment. We stated that the extrinsic evidence concerning Henderson's relationship with the testatrix did not assist in interpreting the will. Although that evidence might have explained why the testatrix named Henderson as an alternate beneficiary in the event she died within two weeks of her husband, it shed no light on her intention in the event her husband died years before she did. We further observed that the will made no disposition of her property in the event she outlived her husband by several years, and noted that although a disinheritance clause could prevent a claimant from inheriting under the will, it could not prevent heirs from inheriting pursuant to the statutory rules of intestacy. (Barnes, supra, 63 Cal.2d at pp. 582-583.)

Turning to the will, we acknowledged that "a will is to be construed according to the intention of the testator, and so as to avoid intestacy." (Barnes, supra, 63 Cal.2d at p. 583.) We added, "However, a court may not write a will which the testator did not write." (Ibid.) The terms of the testatrix's will reflected that she wanted all of her property to go to her husband, and "also demonstrate[d] an awareness that [if she died within two weeks of her husband] she might well have no further opportunity to designate an alternate, and therefore she named [Henderson]. However, ... the will is devoid of a provision or suggestion as to testatrix' intent if, as occurred, she was afforded sufficient time to review the will following the death of her husband." (Ibid.) We noted that if the absence of a disposition of her property had come to her attention after her husband died, she might have provided that her estate would go to Henderson, or she might have made other provisions. "Under such circumstances any selection by the courts now would be to indulge in forbidden conjecture." (Id. at p. 584.) Finally, we found no "`dominant dispositive plan'" that might warrant the finding of a gift by implication. (Ibid.) Therefore, finding the extrinsic evidence offered no assistance, we reversed the order distributing the estate to Henderson.

The Court of Appeal noted that the will in this case is similar to the will in Barnes, supra, 63 Cal.2d 580. "Just as the court concluded in Barnes, Irving's will is not ambiguous.... It simply made no disposition whatsoever of the property in the event Irving outlived his wife by several years, as eventually occurred." The Court of Appeal also found the will sufficiently similar to the will in Barnes to compel the conclusion that it does not reflect a dominant dispositive plan to leave the estate to JNF and COH. Finally, it rejected the admission of extrinsic evidence because the evidence did not address any ambiguity in the will.

The Court of Appeal added that it was "mindful of the fact that the ultimate disposition of Irving's property ... does not appear to comport with his testamentary intent. It is clear that [Irving] meant to dispose of his estate through his bequests, first to his wife and, should she predecease him, then to the charities. It is difficult to imagine that after leaving specific gifts to the charities in the names and memories of beloved family members, Irving intended them to take effect only in the event that he and his wife died `at the same moment.'" It concluded, however, that because the will is unambiguous, Barnes, supra, 63 Cal.2d 580, precluded consideration of the extrinsic evidence.

We granted review to consider whether the rule applied in Barnes, supra, 63 Cal.2d 580, should be reconsidered. For the reasons set forth below, we hold that the categorical bar on reformation of unambiguous wills is not justified and that reformation is permissible if clear and convincing evidence establishes an error in the expression of the testator's intent and establishes the testator's actual specific intent at the time the will was drafted.2

II. DISCUSSION

(3) California law allows the admission of extrinsic evidence to establish that a will is ambiguous and to clarify ambiguities in a will. (Prob. Code, § 6111.5; Estate of Russell (1968) 69 Cal.2d 200, 206-213 [70 Cal.Rptr. 561, 444 P.2d 353].) As COH and JNF acknowledge, however, California law does not currently authorize the admission of extrinsic evidence to correct a mistake in a will when the will is unambiguous. (Estate of Dominici (1907) 151 Cal. 181, 186 [90 P. 448]; Estate of Page (1967) 254 Cal.App.2d 702, 719 [62 Cal.Rptr. 740]; 14 Witkin, Summary of Cal. Law (10th ed. 2005) Wills and Probate, § 236, pp. 315-316.) To evaluate whether there are circumstances in which this court should authorize the admission of extrinsic evidence to correct a mistake in an unambiguous will, we first consider whether the Legislature's actions in this field preclude this court from altering the rule. As explained below, a review of the development of the law in California reflects that the Legislature has codified judicial expansions of the admissibility of evidence with respect to a testator's intent, but has not acted in a manner that restricts the authority of courts to develop the common law in this area. Second, we consider whether the common law rule categorically barring the reformation of wills is warranted, in light of the evolution of the law of probate and modern theories of interpretation of writings, and we conclude that the categorical bar on reformation is not justified. Third, we consider principles of stare decisis, and conclude that a change in the law is warranted to allow the reformation of an unambiguous will when clear and convincing evidence establishes that the will contains a mistake in the expression of the testator's intent at the time the will was executed and also establishes the testator's actual and specific intent at the time the will was executed. Finally, we conclude that the remedy of reformation is potentially available with respect to the theory of mistake articulated by COH and JNF in this case. Therefore, we will direct the case to be remanded for the probate court's consideration of whether clear and convincing evidence establishes that the testator intended, at the time he drafted his will, to provide in his will that his estate was to pass to COH and JNF in the event his wife was not alive at the time the testator died.

A. Statutory and judicial development of the law concerning the admission of extrinsic evidence regarding wills

Beginning with the original Statute of Wills in 1540, statutory law has required that wills be in writing.3 (32 Hen. VIII, ch. 1, July 20, 1540; see Prob. Code, §§ 6110, 6111; Ann. Civ. Code, former §§ 1276, 1277 (1st ed. 1872, Haymond & Burch, commrs.-annotators) (further references to former provisions of the Civil Code are to that code as enacted in 1872); Stats. 1850, ch. 72, § 3, p. 177.) The principles governing the interpretation and enforcement of wills, however, have been developed by the courts. In California, common law principles governing the interpretation of wills were generally reflected in the Civil Code as enacted in 1872, which was based on the draft Field Code's summary of New York common law. (Rosenberg v. Frank (1881) 58 Cal. 387, 404 [the rules set forth in the Civ. Code concerning the interpretation of wills "seem to be in accord with the rules heretofore laid down by [the] Courts for such interpretation"]; see Standard Livestock Co. v. Pentz (1928) 204 Cal. 618, 640 [269 P. 645] [the Civ. Code as adopted in 1872 reflected the Field Code, which summarized N.Y. common law]; Siminoff v. Jas. H. Goodman & Co. Bank (1912) 18 Cal.App. 5, 11 [121 P. 939] ["`the Civil Code ... was not designed to make any general alterations in the established doctrines and rules of the common law' ..."].) "[E]xcept in those instances where [the Civil Code's] language clearly and unequivocally disclose[d] an intention to depart from, alter, or abrogate the common-law rule concerning a particular subject matter, a section of the code purporting to embody such doctrine or rule [would] be construed in the light of common-law decisions on the same subject." (In re Estate of Elizalde (1920) 182 Cal. 427, 433 [188 P. 560]; see Li v. Yellow Cab Co. (1975) 13 Cal.3d 804, 814 [119 Cal.Rptr. 858, 532 P.2d 1226] [sections of the Civ. Code that were declarative of the common law were not intended "to insulate the matters therein expressed from further judicial development"].) Where the Civil Code reflected a clear change from the common law, we construed its language narrowly. (Estate of Dominici, supra, 151 Cal. at p. 185 [because the code's bar on the receipt of declarations of the testator regarding the testator's intentions was contrary to the common law rule, it was construed narrowly to apply only to "mere incidental fugitive utterances or declarations of intent, as distinguished from specific instructions as to testamentary disposition which it may be proved were given"].)

The 1872 Civil Code recognized the principle that "[a] will is to be construed according to the intention of the testator" (Civ. Code, former § 1317), but it also reflected the restrictive rules of its day concerning the admission of extrinsic evidence to assist in determining the testator's intention. In particular, the code recognized the admissibility of extrinsic evidence in only two situations. First, in the event of an ambiguity appearing on the face of a will, the court was authorized to consider the circumstances in which the will was made. (Civ. Code, former § 1318.)4 Second, in the event of an ambiguity in the description of a person or property, which could "appear[] from the context of the will or from extrinsic evidence," the court was authorized to correct the description. (Civ. Code, former § 1340.)5

In 1931, when California's first Probate Code was adopted (Stats. 1931, ch. 281, p. 587), the Civil Code's provisions concerning the admissibility of extrinsic evidence were moved, without substantive change, to section 105 of the new Probate Code6 (see In re Estate of Armstrong (1937) 8 Cal.2d 204, 208 [64 P.2d 1093] [Cal. Code Commission's policy was to compile, consolidate and clarify existing law, and not to make substantive changes]; 14 Witkin, Summary of Cal. Law, supra, Wills and Probate, § 21, pp. 79-80), but the codification did not end the judicial development of principles concerning the admission of extrinsic evidence. Our decision in Estate of Russell, supra, 69 Cal.2d 200, constitutes a prime example of this judicial common law development. There, we acknowledged the two circumstances in which extrinsic evidence historically was admissible (id. at pp. 206-208), but we observed that "[i]n order to determine initially whether the terms of any written instrument are clear, definite and free from ambiguity the court must examine the instrument in the light of the circumstances surrounding[] its execution so as to ascertain what the parties meant by the words used" (id. at pp. 208-209). We further observed that the shift in focus from the ordinary meaning of the words of a document to the intended meaning "reflects the modern development of rules governing interpretation, for in the words of Wigmore `The history of the law of Interpretation is the history of a progress from a stiff and superstitious formalism to a flexible rationalism.'" (Id. at p. 209.) We concluded that "[t]here is nothing in these rules of interpretation which confines their application to contracts. Indeed, quite the contrary. The rules are a response to `problems which run through all the varieties of jural acts,' are therefore not necessarily solvable separately for deeds, contracts and wills, are not peculiar to any one kind of jural act, but involve a general principle applicable to all." (Id. at p. 210.) Finally, we noted that in the context of wills, only the meaning the testator intended by his or her words is relevant. Therefore, the principle that courts should consider extrinsic evidence of the peculiar meaning the author intended by the written words is unrestricted. (Ibid.)

With respect to Probate Code former section 105, we stated that although that provision "delineates the manner of ascertaining the testator's intention `when an uncertainty arises upon the face of a will,' it cannot always be determined whether the will is ambiguous or not until the surrounding circumstances are first considered." (Estate of Russell, supra, 69 Cal.2d at pp. 212-213.) Thus, although the Legislature had codified the historical grounds on which courts had authorized the admission of extrinsic evidence, we did not perceive its provisions to be a limitation on the continued development of the common law.

The California Law Revision Commission and the Legislature followed this court's lead. In 1982, in response to a direction from the Legislature to study whether the Probate Code should be revised, the Commission recommended changes related to wills and intestate succession. (Tentative Recommendation Relating to Wills and Intestate Succession (Nov. 1982) 16 Cal. Law Revision Com. Rep. (1982) pp. 2301, 2305, 2317-2510 (Tentative Recommendation).) Among the goals of the revisions was "to carry out more effectively the intent of the decedent who dies leaving a will." (Id. at p. 2305.) Among its recommendations was to repeal Probate Code former section 105, which reflected the historical limitations on the admission of extrinsic evidence concerning the meaning of a will. In its comment to this recommendation, the Commission stated that former section 105 "purported to codify the much-criticized distinction between patent and latent ambiguities in a will." (Tentative Recommendation, supra, 16 Cal. Law Revision Com. Rep. at p. 2503; see Jevne v. Superior Court (2005) 35 Cal.4th 935, 947 [28 Cal.Rptr.3d 685, 111 P.3d 954] [acknowledging Commission's reports as evidence of legislative intent].) As recommended by the Commission, the Legislature repealed section 105 (Stats. 1983, ch. 842, § 18, p. 3024), and enacted former section 6140, which provided that "[a] will is to be construed according to the intention of the testator." (Stats. 1983, ch. 842, § 55, pp. 3049, 3052; see Tentative Recommendation, supra, 16 Cal. Law Revision Com. Rep. at p. 2503; Cal. Law Revision Com., Annual Rep. (Dec. 1983) 17 Cal. Law Revision Com. Rep. (1984) pp. 822, 873 & appen. IX.)

Subsequent statutory revisions reflect no more than an intention to codify the case law concerning the admission of extrinsic evidence. In 1984, section 6140 of the former Probate Code was amended to substitute language from the Uniform Probate Code, and it then provided: "(a) The intention of the testator as expressed in the will controls the legal effect of the dispositions made in the will. [¶] (b) The rules of construction expressed in this article apply where the intention of the testator is not indicated by the will." (Stats. 1984, ch. 892, § 24, p. 2995.) The California Law Revision Commission commented that "[t]his change is nonsubstantive. Nothing in Section 6140 limits the extent to which extrinsic evidence admissible under former law may be used to determine the testator's intent as expressed in the will." (Cal. Law Revision Com., Annual Rep. (Mar. 1985) Communication of Law Revision Commission Concerning Assembly Bill 2290, 18 Cal. Law Revision Com. Rep. (1986) p. 86 [appen. XII]; see id. at p. 77 [appen. XII].)

In 1989, the California Law Revision Commission recommended that the Legislature enact a new Probate Code that would "omit obsolete material and make numerous technical, clarifying, conforming, and minor substantive revisions in the existing provisions." (Recommendation Proposing New Probate Code (Dec. 1989) 20 Cal. Law Revision Com. (1990) p. 1035.) In 1990, the Legislature repealed the former Probate Code and enacted a new Probate Code. (Stats. 1990, ch. 79, §§ 13, 14, p. 463.) Among the provisions reenacted was former section 6140 of the Probate Code. (Stats. 1990, ch. 79, § 14, pp. 463, 687, enacting new Prob. Code, including § 6140.) Later in 1990, before the new Probate Code became effective, the Legislature added section 6111.5 to both the former Probate Code and the new Probate Code. (Stats. 1990, ch. 263, § 2, p. 1534, adding former § 6111.5 to the Prob. Code; Stats. 1990, ch. 710, § 47, p. 3294, enacting new § 6111.5 without change in language; Stats. 1990, ch. 79, § 14, p. 463.) Section 6111.5 provides: "Extrinsic evidence is admissible to determine whether a document constitutes a will ..., or to determine the meaning of a will or a portion of a will if the meaning is unclear." As was noted in a summary prepared for the Assembly Committee on the Judiciary regarding the Senate Bill that added former section 6111.5 to the Probate Code, the legislation "essentially codifie[d] existing law regarding extrinsic evidence." (Assem. Com. on Judiciary, Rep. on Sen. Bill No. 1984 (1989-1990 Reg. Sess.) The inclusion of sections 6111.5 and 6140 in the new Probate Code continued the predecessor provisions without change. (Revised and Supplemental Comments to the New Probate Code (Sept. 1990) 20 Cal. Law Revision Com. Rep. (1990) p. 2090.)

In 1994, without input from the Law Revision Commission, the Legislature enacted provisions governing the construction of all donative documents. (Prob. Code, § 21101 et seq.; see id., § 45; Stats. 1994, ch. 806, § 41, pp. 4022-4028, enacting Prob. Code, §§ 21101-21140].) According to the Commission, "[a]ll of the rules of construction [were] based on previously existing Probate Code provisions applicable to wills. The basic idea of the 1994 extension to trusts and other instruments was to achieve uniformity among the common estate planning instruments." (Recommendation: Rules of Construction for Trusts and Other Instruments (Nov. 2001) 31 Cal. Law Revision Com. Rep. (2001) pp. 171-172 (Recommendation on Rules of Construction).)

As enacted in 1994, Probate Code section 21102 provided: "(a) The intention of the transferor as expressed in the instrument controls the legal effect of the dispositions made in the instrument. [¶] (b) The rules of construction expressed in this part apply where the intention of the transferor is not indicated by the instrument." (Added by Stats. 1994, ch. 806, § 41, p. 4022.) Thereafter, the California Law Revision Commission recommended that language be added to make clear that the Legislature did not intend section 21102 to restrict the circumstances in which extrinsic evidence was admissible. In its recommendation, the Commission stated that "[l]anguage in Section 21102 suggests that the rules of construction may only be overridden by an expression of contrary intention in the instrument itself. However, existing law allows extrinsic evidence of a testator's intent to rebut the presumptive effect of the rules of construction."7 (Recommendation on Rules of Construction, supra, 31 Cal. Law Revision Com. Rep. at p. 174.) The Commission further stated that, "although the intention of a donor `as expressed in the instrument' controls the legal effect of dispositions made in the instrument, expressions in the instrument are not the exclusive means by which a donor's intention may be ascertained. Under the parol evidence rule, for example, extrinsic evidence is admissible on the issue of a mistake or imperfection of the writing."8 (Recommendation on Rules of Construction, at p. 175, fns. omitted.) The Commission expressed the view that "[t]he role of extrinsic evidence in the determination of the donor's intention should be recognized in the statute." (Ibid.) Therefore, it recommended that the following language be added to Probate Code section 21102: "`Nothing in this section limits the use of extrinsic evidence, to the extent otherwise authorized by law, to determine the intention of the transferor.'" (Recommendation on Rules of Construction, at p. 175.) The Legislature added the recommended language as subdivision (c) of section 21102. (Stats. 2002, ch. 138, § 11, p. 721, amending § 21102.)

(4) This history of statutory provisions concerning the admissibility of evidence of a testator's intent reflects that the Legislature has codified legal principles developed by the courts, and has taken steps to ensure that its enactments do not restrict the admissibility of extrinsic evidence beyond the principles established by the courts. Nothing in this history suggests that the Legislature intended to foreclose further judicial developments of the law concerning the admissibility of evidence to discern the testator's intent, and "we see no reason to interpret the legislation as establishing a bar to judicial innovation." (American Motorcycle Assn. v. Superior Court (1978) 20 Cal.3d 578, 601 [146 Cal.Rptr. 182, 578 P.2d 899] [California's contribution statutes did not preclude judicial adoption of comparative partial indemnity]; see Leung v. Verdugo Hills Hospital (2012) 55 Cal.4th 291, 301 [145 Cal.Rptr.3d 553, 282 P.3d 1250] ["nothing in the statute's legislative history suggests an intent to foreclose the courts from rendering future decisions that would further the statute's main purpose of ameliorating the harshness and the inequity of the common law rule at issue"].) Moreover, it does not appear that the Legislature has addressed the issue of reformation of wills.9 Therefore, as we did in Estate of Russell, supra, 69 Cal.2d 200, we may continue to develop the law concerning the admissibility of evidence to assist in the determination of the testator's intent when the language of the document is clear on its face.

B. No sound basis exists to forbid the reformation of unambiguous wills in appropriate circumstances

As discussed below, extrinsic evidence is admissible to correct errors in other types of donative documents, even when the donor is deceased. Extrinsic evidence is also admissible to aid in the construction of a will, and in some cases, the resulting "construction" has essentially reformed the will. Extrinsic evidence is also admissible to determine whether a document was intended to be a will, and to prove the contents of a will that has been lost or destroyed. Because extrinsic evidence is not inherently more reliable when admitted for these various purposes than when admitted to correct an error in a will, concerns about the reliability of evidence do not justify a categorical bar on reformation of wills. To the extent categorical resistance to reformation is based instead on a concern that reformed language would not comply with the formalities required by the statute of wills,10 principles developed in the context of the statute of frauds, which similarly requires a signed writing to evidence specified documents, illustrate that the purposes of the statute of wills are satisfied by the testator's execution of a writing that complies with the statutory requirements. With the statutory purposes satisfied, only the concerns regarding the reliability of evidence might justify a categorical bar on reformation of wills, and those concerns are addressed by imposing a burden of clear and convincing evidence.

(5) In California, extrinsic evidence is generally admissible to correct errors in documents, including donative documents other than wills. (Civ. Code, §§ 1640, 3399 [contracts];11 Hess v. Ford Motor Co. (2002) 27 Cal.4th 516, 524-526 [117 Cal.Rptr.2d 220, 41 P.3d 46] [release agreement]; Giammarrusco v. Simon (2009) 171 Cal.App.4th 1586, 1603-1604 [91 Cal.Rptr.3d 50 [irrevocable trust]; Bilafer v. Bilafer (2008) 161 Cal.App.4th 363, 368-369 [73 Cal.Rptr.3d 880] [irrevocable trust]; Ike v. Dolittle (1998) 61 Cal.App.4th 51 [70 Cal.Rptr.2d 887] [after trustors' deaths, reformation allowed to correct a drafting error]; American Surety Co. v. Heise (1955) 136 Cal.App.2d 689, 695 [289 P.2d 103] [mutual mistake in insurance policy]; Reina v. Erassarret (1951) 103 Cal.App.2d 258, 266 [229 P.2d 92] [gift deed]; Genuser v. Ocean Acc. etc. Corp. (1941) 42 Cal.App.2d 673, 674 [109 P.2d 753] [unilateral mistake in insurance policy]; Merkle v. Merkle (1927) 85 Cal.App. 87, 104 [258 P. 969] [deed]; Robertson v. Melville (1923) 60 Cal.App. 354, 356 [212 P. 723] [deed]; Parker v. Hardisty (1921) 54 Cal.App. 628 [202 P. 479] [purchaser of land from recipient of gift deed could obtain reformation to cure flaw in gift deed]; 2 Witkin, Summary of Cal. Law, supra, Insurance, §§ 216, 217, pp. 322-324; 1 Witkin, Summary of Cal Law, supra, Contracts, § 276, pp. 306-307.)

(6) In addition, California courts have admitted extrinsic evidence to apply to the construction of a will to accomplish what is arguably, or has the effect of, reforming a will. For example, admission of extrinsic evidence that the testator referred to her siblings as the "`Broude Trust'" allowed the court to correct the testator's error in leaving everything to the "`Broude Trust Fund'" instead of to her siblings. (Estate of Glow (1962) 208 Cal.App.2d 613, 616-617 [25 Cal.Rptr. 416] [evidence of testator's "lay" usage of a technical term is admissible]; see Estate of Kime (1983) 144 Cal.App.3d 246, 264 [193 Cal.Rptr. 718] [testator's statements were admissible regarding whether she understood language merely appointing an executor to be effective to designate a beneficiary]; Estate of Fries (1963) 221 Cal.App.2d 725, 727-730 [34 Cal.Rptr. 749] [will identified husband as executor but did not identify to whom testatrix bequeathed property; court relied in part on husband's testimony that testatrix said she would give him everything]; Estate of Karkeet (1961) 56 Cal.2d 277, 281-283 [14 Cal.Rptr. 664, 363 P.2d 896] [extrinsic evidence was admissible to establish that a will that named the testatrix's good friend as executrix, but made no testamentary disposition, was intended to leave the estate to the good friend]; Estate of Akeley (1950) 35 Cal.2d 26, 30 [215 P.2d 921] [evidence that testatrix was unmarried, had no relatives, and drafted her will herself was cited in support of ruling that three charities that were each given one-quarter of the residue of the estate would each receive one-third].)

(7) Principles allowing the admission of extrinsic evidence to identify and resolve ambiguities in wills have also been invoked to correct attorneys' drafting errors and thereby to reform wills. For example, in Estate of Taff (1976) 63 Cal.App.3d 319 [133 Cal.Rptr. 737], the testatrix directed her attorney to provide that if the testatrix's sister did not survive her, the residue of the estate would go to her sister's children. Her will, however, stated that if her sister predeceased her, the residue would go to "`my heirs in accordance with the laws of intestate succession.'" (Id. at p. 322.) The court found that the extrinsic evidence of the testatrix's instructions to her attorney and statements to her sister "exposed a latent ambiguity, i.e., that when the testatrix used the term `my heirs' in her will, she intended to exclude the relatives of her predeceased husband, Harry." (Id. at p. 325.) Similarly, in Estate of Anderson, supra, 56 Cal.App.4th 235, a later will expressly revoked all prior wills, but due to attorney error, the later will failed to include a provision exercising a testamentary power of appointment over a portion of a trust. As a result, the later will failed to carry out the testatrix's intent that the half of the trust over which she had a power of appointment go to her daughter rather than to the issue of her late husband. The court held that extrinsic evidence was admissible to determine whether the testatrix intended to revoke the earlier will's provision exercising her power of appointment. (Id. at pp. 242-248.)

(8) Extrinsic evidence is admissible not only to aid in the construction of a will, but also to determine whether a document was intended to be a will. (Halldin v. Usher (1958) 49 Cal.2d 749, 752 [321 P.2d 746] [parol evidence is admissible to prove a document was intended as a will rather than a contract]; Estate of Sargavak (1950) 35 Cal.2d 93, 96 [216 P.2d 850] [evidence is admissible to prove a will was executed in jest, as a threat to induce action, under the mistaken belief it was a mortgage, to induce illicit relations, or in response to annoyance from one who seeks to inherit].) In addition, courts have long recognized that extrinsic evidence is admissible to prove that a will has been lost or destroyed, and to prove its contents. (Prob. Code, § 8223; Swift v. Superior Court (1952) 39 Cal.2d 358 [247 P.2d 6]; Estate of Flood (1941) 47 Cal.App.2d 809 [119 P.2d 168]; 14 Witkin, Summary of Cal. Law, supra, Wills and Probate, §§ 550-551, pp. 630-633.)

Thus, extrinsic evidence is admitted to correct donative documents other than wills after the donor's death. Moreover, myriad circumstances exist in which California courts appropriately admit evidence to establish a testator's intentions. Because extrinsic evidence is not inherently more reliable when admitted for these various purposes than when admitted to correct an error in a will, Professors John Langbein and Lawrence Waggoner, leading advocates of an extension of the doctrine of reformation to unambiguous wills, conclude that evidentiary concerns do not explain or justify the bar on reformation of wills. In their view, a greater obstacle to reformation has been concern with the formalities required in the execution of a will by the statute of wills. (Langbein & Waggoner, Reformation of Wills on the Ground of Mistake: Change of Direction in American Law? (1982) 130 U.Pa. L.Rev. 521, 524-529 (hereafter Langbein and Waggoner).) To overcome the objection that reformed language is unattested, they look to principles related to the statute of frauds.

(9) Like the statute of wills, the statute of frauds requires certain documents to be evidenced by a writing subscribed by the party. If not evidenced by such a writing, a contract subject to the statute of frauds is invalid. (Civ. Code, § 1624.) "`The primary purpose of the Statute [of Frauds] is evidentiary, to require reliable evidence of the existence and terms of the contract and to prevent enforcement through fraud or perjury of contracts never in fact made.'" (Sterling v. Taylor (2007) 40 Cal.4th 757, 766 [55 Cal.Rptr.3d 116, 152 P.3d 420].) Once sufficient written evidence of an agreement is presented, the evidentiary purpose is served, and extrinsic evidence is admissible to clarify ambiguous terms and to reform the writing to correct a mistake, even when the writing is intended to be a complete and exclusive statement of the parties' agreement. (Id. at p. 767; Civ. Code, §§ 1640, 3399; Code Civ. Proc., § 1856, subds. (b), (e); Hess v. Ford Motor Co., supra, 27 Cal.4th at pp. 524-525.)

(10) In correcting a contract subject to the statute of frauds, a court is not enforcing an oral contract, but is instead enforcing a written contract in accordance with the parties' actual agreement. To overcome the presumption that the writing is accurate, we have required clear and convincing evidence of a mistake before allowing reformation of a contract. (Nat. Auto. & Cas. Co. v. Ind. Acc. Com. (1949) 34 Cal.2d 20, 25 [206 P.2d 841]; Burt v. Los Angeles Olive Growers Assn. (1917) 175 Cal. 668, 675 [166 P. 993]; R & B Auto Center, Inc. v. Farmers Group, Inc. (2006) 140 Cal.App.4th 327, 382 [44 Cal.Rptr.3d 426].)

(11) In contrast to cases involving the statute of frauds, which may or may not involve a party who is deceased, cases arising under the statute of wills always involve a testator who is deceased and therefore cannot explain his or her intentions. We have already recognized, however, in the context of inheritance rights, that imposing a burden of proof by clear and convincing evidence is a means to address evidentiary concerns related to the circumstances that the principal witness is deceased and statutory formalities were not followed. (Estate of Ford (2004) 32 Cal.4th 160, 172 [8 Cal.Rptr.3d 541, 82 P.3d 747] [one seeking to inherit based on the doctrine of equitable adoption must meet the clear and convincing evidence standard, in part because the adoptive parent is deceased and can no longer testify, and the relief sought is outside the ordinary course of intestate succession and without the formalities required by the adoption statutes].) In addition, discerning intent in the context of a will is eased by the fact that the court must ascertain only the subjective intent of a single individual. (Langbein & Waggoner, supra, 130 U.Pa. L.Rev. at p. 569.) Therefore, the fact that the testator will always be unavailable to testify does not warrant a categorical bar on the admission of extrinsic evidence to reform a will.

Applying the analysis developed with respect to the statute of frauds, Langbein and Waggoner observe, "Whereas an oral will instances total noncompliance with the Wills Act formalities, a duly executed will with a mistakenly rendered term involves high levels of compliance with both the letter and the purpose of the Wills Act formalities. To the extent that a mistake case risks impairing any policy of the Wills Act, it is the evidentiary policy that is in question."12 (Langbein & Waggoner, supra, 130 U.Pa. L.Rev. at p. 569, italics added.) With respect to evidentiary concerns, the authors advocate that reformation be allowed only in cases of clear and convincing evidence of the alleged mistake and the testator's intent. (Ibid.) As noted, we have previously imposed a clear and convincing evidence standard to support a claim of inheritance based on equitable adoption. (Estate of Ford, supra, 32 Cal.4th 160, 172.)

(12) In cases in which clear and convincing evidence establishes both a mistake in the drafting of the will and the testator's actual and specific intent at the time the will was drafted, it is plain that denying reformation would defeat the testator's intent and result in unjust enrichment of unintended beneficiaries. Given that the paramount concern in construing a will is to determine the subjective intent of the testator (Estate of Russell, supra, 69 Cal.2d at p. 205; 4 Page on Wills (Bowe-Parker rev. 2004) § 30.1, p. 2), only significant countervailing considerations can justify a rule categorically denying reformation.

The Radins cite various factors in support of their contention that reformation of wills should never be allowed, some of which we have addressed above. First, they distinguish wills from other written instruments, noting that probate of a will always occurs after the testator's death, whereas contract litigation typically occurs when the parties to the contract are alive, and trust administration "frequently" begins before the testator's death. In addition, anyone may claim to be an intended beneficiary of a will, but the parties to a contract typically are few. We are not persuaded by these arguments in favor of a categorical bar on reformation. As we have noted, the death of a principal witness has not been viewed as a reason to deny reformation in other contexts. Also, although anyone may claim to be an intended beneficiary of a will, an appropriately tailored reformation remedy will alleviate concerns regarding unintended beneficiaries; it is unlikely that there will be many persons who have a connection to a testator and can produce clear and convincing evidence both of a mistake in the drafting of the will at the time the will was written and of the testator's specific intentions concerning the disposition of property. Categorically denying reformation may result in unjust enrichment if there is a mistake of expression in the will, and imposing a burden of clear and convincing evidence of both the existence of the mistake and of the testator's actual and specific intent at the time the will was drafted helps safeguard against baseless allegations. (See Estate of Ford, supra, 32 Cal.4th at p. 172 [equitable adoption must meet the clear and convincing evidence standard]; see also Dillon v. Legg (1968) 68 Cal.2d 728, 736 [69 Cal.Rptr. 72, 441 P.2d 912] ["the possibility that fraudulent assertions may prompt recovery in isolated cases does not justify a wholesale rejection of the entire class of claims in which that potentiality arises"]; Notten v. Mensing (1935) 3 Cal.2d 469, 477 [45 P.2d 198] [acknowledging the strong temptation to fabricate evidence of an oral agreement to leave property to the plaintiff, but concluding that "if such agreement is proved by full, clear and convincing evidence, such agreement should be enforced according to its terms"].)

Second, the Radins express concern that reformation overrides the formalities required to execute a will. The fact that reformation is an available remedy does not relieve a testator of the requirements imposed by the Statute of Wills. (See Prob. Code, § 6111.) Therefore, the formalities continue to serve various functions associated with the rituals of will execution, such as warning the testator of the seriousness of the act and clearly identifying the document as a will. (See fn. 12, ante.) To the extent reformation is inconsistent with the formalities' evidentiary purpose of establishing the testator's intent in a writing, the inconsistency is no different from the tension between reformation and the statute of frauds. As explained above, the evidentiary concern is addressed by requiring clear and convincing evidence of a mistake in expression and the testator's actual and specific intent. We should not allow stringent adherence to formalities to obscure the ultimate purpose of the statute of wills, which is to transfer an estate in accordance with the testator's intent.

Third, the Radins assert that allowing reformation in circumstances in which the estate would otherwise pass pursuant to the laws of intestacy constitutes an attack on the laws of intestacy. We disagree. The purpose of reformation is to carry out the wishes of the testator, and the remedy reflects no judgment other than a preference for disposition pursuant to the wishes of the testator. This preference is consistent with the statutory scheme. (See Prob. Code, §§ 6110 [a will that is not properly executed is enforceable if clear and convincing evidence establishes it was intended to constitute the testator's will], 21120 ["Preference is to be given to an interpretation of an instrument that will prevent intestacy or failure of a transfer, rather than one that will result in an intestacy or failure of a transfer."].)

Fourth, the Radins assert that allowing reformation will result in a significant increase in probate litigation and expenses. Claimants have long been entitled, however, to present extrinsic evidence to establish that a will is ambiguous despite the fact that it appears to be unambiguous. (Estate of Russell, supra, 69 Cal.2d at pp. 206-213.) Therefore, probate courts already receive extrinsic evidence of testator intent from claimants attempting to reform a will through the doctrine of ambiguity. (Cf. Buss v. Superior Court (1997) 16 Cal.4th 35, 57 [65 Cal.Rptr.2d 366, 939 P.2d 766] [in rejecting the contention that requiring only a preponderance of the evidence to establish an insurer's right to reimbursement will open the floodgates of litigation, the court noted that "the `floodgates' have been open for quite some time ..."].) The task of deciding whether the evidence establishes by clear and convincing evidence that a mistake was made in the drafting of the will is a relatively small additional burden, because the court is already evaluating the evidence's probative value to determine the existence of an ambiguity.13 To the extent additional claims are made that are based on a theory of mistake rather than a theory of ambiguity, the heightened evidentiary standard will help the probate court to filter out weak claims. Finally, fear of additional judicial burdens is not an adequate reason to deny relief that would serve the paramount purpose of distributing property in accordance with the testator's intent. (See Buss, at p. 58 [acknowledging that the future might bring more claims for reimbursement, "[b]ut the possible invocation of this right — or any other — is not a sufficient basis for its abrogation or disapproval"]; Ochoa v. Superior Court (1985) 39 Cal.3d 159, 171 [216 Cal.Rptr. 661, 703 P.2d 1] [rejecting a proposed limit on the circumstances in which negligent infliction of emotional distress may be established, despite claim of "`infinite liability'"].)

(13) Fifth, the Radins discount justifications for allowing reformation in appropriate circumstances. They assert that Probate Code section 6110, subdivision (c)(2), which allows the probate of a will that was not executed in compliance with statutory attestation requirements if clear and convincing evidence establishes that the testator intended the writing to be a will, was not intended to lessen required formalities. Although section 6110 does not reduce the formalities of attestation, it reflects a judgment that the formalities should not be allowed to defeat the testator's intent when clear and convincing evidence satisfies the evidentiary concerns underlying the formalities of the statute of wills. The Radins also reject as a factor in support of a reformation remedy the avoidance of unjust enrichment. They state that no one has a right to inherit, and they recite various facts that they believe reflect that it is more just for Irving's relatives to inherit his estate than for the charities to receive it. If, however, a testator did not intend to devise property to a particular party, that party's receipt of the property as a result of a mistake constitutes unjust enrichment.

In sum, the Radins identify no countervailing considerations that would justify denying reformation if clear and convincing evidence establishes a mistake in the testator's expression of intent and the testator's actual and specific intent at the time the will was drafted.

C. Principles of stare decisis do not compel adherence to precedent in this context

(14) The Radins assert that because the existing rule has withstood the test of time, any change should be left to the Legislature. "`... It is, of course, a fundamental jurisprudential policy that prior applicable precedent usually must be followed even though the case, if considered anew, might be decided differently by the current justices. This policy, known as the doctrine of stare decisis, "is based on the assumption that certainty, predictability and stability in the law are the major objectives of the legal system; i.e., that parties should be able to regulate their conduct and enter into relationships with reasonable assurance of the governing rules of law." [Citation.]'" (Freeman & Mills, Inc. v. Belcher Oil Co. (1995) 11 Cal.4th 85, 92-93 [44 Cal.Rptr.2d 420, 900 P.2d 669].) The rule is flexible, however, and it "`"should not shield court-created error from correction."'" (Id. at p. 93.)

Although allowing reformation of an unambiguous will in appropriate instances will overturn many decades of precedent, we conclude that principles of stare decisis do not compel continued adherence to the rule at issue. The interest in ensuring certainty, predictability, and stability has been undermined by the inconsistent application of the principles applicable to the construction of wills. As noted above, in the course of applying the doctrine that an ambiguous will may be clarified through the admission of extrinsic evidence, courts have essentially reformed wills. (See, e.g., Estate of Akeley, supra, 35 Cal.2d at p. 30; Estate of Kime, supra, 144 Cal.App.3d 246, 264; Estate of Fries, supra, 221 Cal.App.2d 725, 727-730; Estate of Glow, supra, 208 Cal.App.2d 613, 616-617.) The results have been inconsistent, however, depending on whether a court perceives a gap in a will to reflect an ambiguity. (Compare Estate of Karkeet, supra, 56 Cal.2d 277, 281-283 [testatrix named an executrix but did not name any beneficiaries; it was unreasonable to think the decedent intended to make a will and also allow entire estate to escheat to the state; extrinsic evidence was admissible to resolve ambiguity] with Estate of De Moulin (1950) 101 Cal.App.2d 221, 225 [225 P.2d 303] [will left money to children, described residuary, and appointed his spouse executrix, but did not name beneficiary of residuary; extrinsic evidence that scrivener mistakenly omitted provision leaving residuary to spouse was not admissible]; cf. Estate of White (1970) 9 Cal.App.3d 194, 201 [87 Cal.Rptr. 881] ["scholarly analysis of the cases interpreting [former] section 105 of the Probate Code has resulted in the justifiable conclusion that it is impossible to determine when the testator's oral declarations would be deemed admissible in any given case ..."].)

Rather than introducing uncertainty into estate planning, allowing reformation of a will upon a clear and convincing showing of a mistake in expression and the testator's actual and specific intent helps ensure that the testator's affairs are settled as intended. And because the doctrine is relevant only in the context of litigation, and it affects the distribution of an estate only upon a determination by clear and convincing evidence of a mistake in the will and of the testator's actual intent at the time the will was drafted, adoption of the doctrine will not diminish the principles of law that encourage the preparation of well-drafted, properly executed wills. "Precisely because the reformation doctrine is a rule of litigation, no draftsman would plan to rely on it when proper drafting can spare the expense and hazard of litigation." (Langbein & Waggoner, supra, 130 U.Pa. L.Rev. at p. 587.)

In addition, the principles we are reconsidering are entirely court created, and the Legislature's inaction does not weigh against allowing reformation. As explained above, the Legislature has followed the courts' lead in adopting more flexible rules concerning the interpretation of wills, and has been attentive to codifying principles established by our cases without barring the continued evolution of the law. Furthermore, the technical requirements applicable to wills have become more flexible, the evidence admissible to interpret wills has been expanded, and existing doctrines concerning the resolution of ambiguities in wills have been stretched to allow the correction of mistakes.

Finally, allowing reformation in these circumstances is consistent with the Legislature's efforts to apply the same rules of construction to all donative documents (see Prob. Code, § 21101 et seq.), and will promote fairness in the treatment of estates, regardless of the tools used for estate planning. According to the legislative history of Probate Code section 21101, the State Bar's estate planning, trust, and probate law section sought uniformity of treatment of various instruments because trusts were increasingly being used in lieu of wills. (Assem. Com. on Judiciary, Analysis of Assem. Bill No. 3686 (1993-1994 Reg. Sess.) Apr. 20, 1994, pp. 2-3.) Moreover, allowing reformation of trusts and other instruments, but never of wills, appears to favor those with the means to establish estate plans that avoid probate proceedings, and to deny a remedy with respect to the estates of individuals who effect their plans through traditional testamentary documents. Denying reformation in these circumstances seems particularly harsh with respect to individuals who write wills without the assistance of counsel, and are more likely to overlook flaws in the expression of their intent.

As the Radins note, to date only a few states allow reformation of wills. However, both the Restatement Third of Property and the Uniform Probate Code now support the remedy.14 The Restatement's reformation provision appeared in the tentative draft of March 1995, and in the final draft issued in 2003. The Uniform Probate Code's provision authorizing the reformation of wills was added in 2008, and five states have adopted that provision.15 In addition, Washington, which has not adopted the Uniform Probate Code, has provided by statute that an unambiguous will "may be reformed ... to conform the terms to the intention of the testator" upon clear and convincing evidence of a mistake. (Wash. Rev. Code § 11.96A.125 (2011).) In Connecticut, extrinsic evidence is admissible to prove a scrivener's error in a will, and a correction will be made upon a clear and convincing showing of error. (Erickson v. Erickson (1998) 246 Conn. 359 [716 A.2d 92, 98-100].) Courts in New York and New Jersey have also applied a more liberal approach to correcting flaws in wills. (In re Herceg (2002) 193 Misc.2d 201 [747 N.Y.S.2d 901, 905] ["it seems logical to this court to choose the path of considering all available evidence as recommended by the Restatement in order to achieve the dominant purpose of carrying out the intention of the testator" (italics added)]; Engle v. Siegel (1977) 74 N.J. 287 [377 A.2d 892, 894] [the "doctrine of probable intent" allows consideration of "`competent extrinsic evidence and common human impulses ... to ascertain and carry out what the testator probably intended should be the disposition if the present situation developed'"].)16

(15) For the reasons discussed above, we are persuaded that authorizing the reformation of wills under the circumstances and with the protections discussed above serves the paramount purpose of the law governing wills without compromising the policies underlying the statutory scheme and the common law rules. If a mistake in expression and the testator's actual and specific intent at the time the will was drafted are established by clear and convincing evidence, no policy underlying the statute of wills supports a rule that would ignore the testator's intent and unjustly enrich those who would inherit as a result of a mistake. (Cf. Rowland v. Christian (1968) 69 Cal.2d 108, 118-119 [70 Cal.Rptr. 97, 443 P.2d 561] [in rejecting common law classifications based on the plaintiff's status as a trespasser, licensee, or invitee with regard to the liability of a possessor of land, the court looked to the "basic policy" that everyone is responsible for injuries caused by his or her negligence in managing property].)

D. The charities have articulated a valid theory that will support reformation if established by clear and convincing evidence

COH and JNF contend that Irving actually intended at the time he wrote his will to provide that his estate would pass to COH and JNF in the event Beatrice was not alive to inherit his estate when he died, but that his intent was inartfully expressed in his will and thus there is a mistake in the will that should be reformed to reflect his intent when the will was drafted. Their contention, if proved by clear and convincing evidence, would support reformation of the will to reflect Irving's actual intent.

First, the alleged mistake concerns Irving's actual intent at the time he wrote the will. As explained above, reformation of a document that is subject to the statute of frauds or the statute of wills entails the enforcement of the written document in a manner that reflects what was intended when the document was prepared. If Irving's only intent at the time he wrote his will was to address the disposition of his estate in the circumstances in which he died before Beatrice or they died simultaneously, his will accurately reflects his intent. In that circumstance, his mistake, if any, would be in failing subsequently to modify the will after Beatrice died, and that mistake would not be related to the will he wrote and that COH and JNF seek to have reformed. (See generally 1 Witkin, Summary of Cal. Law, supra, Contracts, § 278, p. 308.)

Second, the alleged mistake and intent are sufficiently specific. The allegations are precise with respect to the error and the remedy: the charities assert Irving specifically intended when he wrote his will to provide that his estate would pass to COH and JNF not only upon the simultaneous death of Irving and Beatrice, as the will expressly states, but also in the event Beatrice was not alive to inherit the estate at the time of his death. Although COH and JNF do not allege that the error was merely clerical, but instead assert that Irving's intent was inartfully expressed, their theory alleges "a mistake in the rendering of terms that the testator has authored or approved. The remedy in such a case has exactly the dimensions of the mistake. The term that the testator intended is restored." (Langbein & Waggoner, supra, 130 U.Pa. L.Rev. at pp. 583-584.)

The charities' theory, which sets forth a specific disposition of assets Irving allegedly intended when he wrote his will, distinguishes this case from circumstances in which it is alleged that the testator had a more general intent regarding the disposition of the estate which was not accomplished by the will as written. An example of an error involving general intent would be a case in which a testator intended in his or her will to provide adequate resources to one of the will's beneficiaries to support that beneficiary for a lifetime, but the specific gift set forth in the will proves to be inadequate for that purpose. Thus, that will accurately sets forth the testator's specific intent with respect to the distribution of assets, but due to a mistake with respect to the value of those assets or the needs of the beneficiary, the will fails to effect the testator's intent to provide adequate assets to support the beneficiary. In contrast to cases in which the alleged error is in the rendering of the specific terms intended by the testator, cases in which the alleged error is in failing to accomplish a general intent of the testator would require a court to determine the testator's putative intent: if the testator had known of the mistake, how would the testator have changed the will? The case before us presents only the issue of whether a will may be reformed when extrinsic evidence establishes that the will fails to set forth the actual specific intent of the testator at the time the will was executed, and we express no opinion on the availability of reformation in cases involving claims of general and putative intent.

Finally, for the reasons discussed above, evidence of the testator's intent must be clear and convincing. Among the evidence to be considered is the will itself, but when reformation rather than construction of a will is at issue, the rules of construction, which set forth principles for determining disposition of estate assets where the testator's intention is not reflected in the will (Prob. Code, § 21102), do not apply where extrinsic evidence supplies the missing terms. (Langbein and Waggoner, supra, 130 U.Pa. L.Rev. at pp. 579-580.) Other doctrines of interpretation are also supplanted by the remedy of reformation. For example, although the terms of a will may be inadequate alone to establish a dominant dispositive plan that would warrant a gift by implication (see Estate of Barnes, supra, 63 Cal.2d at p. 584; Brock v. Hall (1949) 33 Cal.2d 885, 890-892 [206 P.2d 360]), those aspects of the will that tend to reflect an intent to make a particular gift should be considered together with the extrinsic evidence of intent to determine whether there is clear and convincing evidence of an intent to make a gift. Similarly, although a disinheritance clause cannot prevent heirs from inheriting pursuant to the statutory rules of intestacy (Estate of Barnes, supra, at pp. 582-583), any intent reflected in such a clause may be relevant when reformation is sought.

III. CONCLUSION

We hold that an unambiguous will may be reformed to conform to the testator's intent if clear and convincing evidence establishes that the will contains a mistake in the testator's expression of intent at the time the will was drafted, and also establishes the testator's actual specific intent at the time the will was drafted. We reverse the judgment of the Court of Appeal and remand the matter to the Court of Appeal with directions to remand the case to the trial court for its consideration of extrinsic evidence as authorized by our opinion.

Werdegar, J., Chin, J., Corrigan, J., Liu, J., Cuéllar, J., and Kruger, J., concurred.