To Be Argued By: SETH P. WAXMAN Time Requested: 20 Minutes APL-2015-00052 New York County Surrogate’s Court File No. 2604/08 Court of Appeals STATE OF NEW YORK Proceeding of KEVIN AOKI, KANA AOKI NOOTENBOOM, KYLE AOKI and KENNETH PODZIBA, as Trustees of the Benihana Protective Trust, for Relief with Respect to the Benihana Protective Trust dated June 8, 1998 by and between Rocky H. Aoki, as grantor, and Kevin Aoki and Darwin C. Dornbush, as trustees, for the benefit of Rocky H. Aoki and others. KEVIN AOKI, KANA AOKI NOOTENBOOM, KYLE AOKI and KENNETH PODZIBA, Respondents, -against- ECHO AOKI, OLIVIA YUMI NOOTENBOOM, NATALIE EMI NOOTENBOOM, SKY AI AOKI and NOA AOKI, DEVON AOKI and STEVEN AOKI, Respondents, KEIKO ONO AOKI, Appellant. BRIEF FOR APPELLANT KEIKO ONO AOKI d ALAN E. SCHOENFELD WILMER CUTLER PICKERING HALE AND DORR LLP 7 World Trade Center 250 Greenwich Street New York, New York 10007 Telephone: (212) 230-8800 Facsimile: (212) 230-8888 Attorneys for Appellant Keiko Ono Aoki Of Counsel: SETH P. WAXMAN (pro hac vice) DANIEL S. VOLCHOK (pro hac vice) ADAM I. KLEIN THOMAS G. SPRANKLING (pro hac vice) WILMER CUTLER PICKERING HALE AND DORR LLP 1875 Pennsylvania Avenue NW Washington, D.C. 20006 Telephone: (202) 663-6000 Facsimile: (202) 663-6363 May 11, 2015 TABLE OF CONTENTS Page TABLE OF AUTHORITIES ................................................................................... iii INTRODUCTION ..................................................................................................... 1 QUESTIONS PRESENTED ...................................................................................... 6 JURISDICTION ......................................................................................................... 7 STATEMENT OF FACTS ........................................................................................ 7 A. Background ........................................................................................... 7 B. The Irrevocable Release ........................................................................ 9 C. Rocky Signs A New Codicil-Forcing Dornbush And Shaw To Belatedly Inform Him Of The True Meaning Of The Release ................................................................................................. 13 D. After The Children Retain Other Counsel, Shaw Continues Drafting Releases That Would Assist Them ....................................... 15 E. Proceedings Before The Surrogate’s Court ......................................... 16 F. The Appellate Division’s Decision ..................................................... 23 SUMMARY OF ARGUMENT ............................................................................... 25 ARGUMENT ........................................................................................................... 27 I. THE APPELLATE DIVISION ERRED IN LIMITING APPLICATION OF THE CONSTRUCTIVE-FRAUD DOCTRINE ................................................................. 27 A. The Hallmark Of Equity Jurisprudence, Including The Constructive-Fraud Doctrine, Is Flexibility ........................................ 28 B. The Appellate Division’s Rigid Rule Finds No Support In New York Precedent ........................................................................... 30 - i - C. The Appellate Division’s Rule Is Not Supported By Decisions From Outside This State ..................................................... 35 D. The Appellate Division’s Rule Would Have Far-Reaching Deleterious Ramifications ................................................................... 39 E. Properly Applied, The Constructive-Fraud Doctrine Requires That The Release Be Invalidated ......................................... 42 II. THE FACTS HERE SATISFY THE APPELLATE DIVISION’S RULE ....................... 48 A. The Attorney-Client Relationship Between The Lawyers And The Children Made The Lawyers Effective Parties To The Release ......................................................................................... 49 B. Dornbush And Shaw Had Their Own “Interest In The Transaction” ........................................................................................ 53 CONCLUSION ........................................................................................................ 57 - ii - TABLE OF AUTHORITIES CASES Page(s) Adams v. Cowen, 177 U.S. 471 (1900) .............................................................. 35, 36 Addis v. Grange, 192 N.E. 774 (Ill. 1934) ......................................................... 37, 38 Albany City Savings Institution v. Burdick, 87 N.Y. 40 (1881) .............................. 46 Amalfitano v. Rosenberg, 12 N.Y.3d 8 (2009) ........................................................ 41 Amend v. Hurley, 293 N.Y. 587 (1944) ............................................................. 24, 45 Baker v. Humphrey, 101 U.S. 494 (1879) ........................................................... 4, 41 Barnard v. Gantz, 140 N.Y. 249 (1893) .................................................................. 42 Beatty v. Guggenheim Exploration Co., 225 N.Y. 380 (1919) ................................ 29 Benihana of Tokyo, Inc. v. Benihana, Inc., 891 A.2d 150 (Del. Ch. 2005) ....... 55, 56 Callahan v. Callahan, 514 N.Y.S.2d 819 (3d Dep’t 1987) ..................................... 33 Cowee v. Cornell, 75 N.Y. 91 (1878) ..............................................1, 2, 6, 27, 30, 35 Cowen v. Adams, 78 F. 536 (6th Cir. 1897) ............................................................. 36 Fisher v. Bishop, 108 N.Y. 25 (1888) ................................................................ 32, 33 Fisher v. Bishop, 36 Hun 112 (4th Dep’t 1885) ...................................................... 33 Gardine v. Cottey, 230 S.W.2d 731 (Mo. 1950) .......................................... 38, 39, 44 Hecht Co. v. Bowles, 321 U.S. 321 (1944) ........................................................ 28, 29 In re Thelen LLP, 24 N.Y.3d 16 (2014) ................................................................... 42 Kirschner v. KPMG LLP, 15 N.Y.3d 446 (2010) .................................................... 42 Marx v. McGlynn, 88 N.Y. 357 (1882) .................................................................... 40 - iii - Matter of Aoki v. Aoki, 985 N.Y.S.2d 523 (1st Dep’t 2014) ............................passim Matter of Aoki, 948 N.Y.S.2d 597 (1st Dep’t 2012) .............................................. 7, 9 Matter of Co-operative Law Co., 198 N.Y. 479 (1910) .......................................... 53 Matter of Feinberg, 543 N.Y.S.2d 300 (2d Dep’t 1989) ......................................... 52 Matter of Gordon v. Bialystoker Center & Bikur Cholim, Inc., 45 N.Y.2d 692 (1978) .................................................................................................. 2, 42 Matter of Greiff, 92 N.Y.2d 341 (1998) .............................................. 2, 4, 28, 31, 32 Matter of O’Hara, 445 N.Y.S.2d 201 (2d Dep’t 1981) ..................................... 24, 34 Matter of Paul, 482 N.Y.S.2d 121 (3d Dep’t 1984) ................................................ 47 Matter of Poggemeyer, 449 N.Y.S.2d 12 (2d Dep’t 1982)...................................... 52 Matter of Smith, 276 N.Y.S. 646 (4th Dep’t 1935) ................................................. 47 Matter of Stortecky v. Mazzone, 85 N.Y.2d 518 (1995) .......................................... 17 Matter of Wilson, 298 N.Y. 398 (1949) ................................................................... 17 McLenithan v. McLenithan, 710 N.Y.S.2d 674 (3d Dep’t 2000) ............................ 49 Meinhard v. Salmon, 249 N.Y. 458 (1928) ............................................................. 41 Messner Vetere Berger McNamee Schmetterer Euro RSCG Inc. v. Aegis Group PLC, 93 N.Y.2d 229 (1999) ........................................................... 2, 40 Olitkowski v. St. Casimir Savings & Loan Ass’n, 4 N.W.2d 664 (Mich. 1942) .............................................................................................................. 39 Poucher v. Blanchard, 86 N.Y. 256 (1881) ............................................................. 53 Richter v. Richter, 60 So. 880 (Ala. 1913) .............................................................. 39 Rogers v. Rogers, 63 N.Y.2d 582 (1984) ................................................................. 28 Russell v. Briggs, 165 N.Y. 500 (1901) ................................................................... 35 - iv - SEC v. Capital Gains Research Bureau, Inc., 375 U.S. 180 (1963) ....................... 29 Seymour v. Freer, 75 U.S. 202 (1869) ..................................................................... 29 Smith v. Smith, 134 N.Y. 62 (1892) ................................................................... 46, 47 State v. Barone, 74 N.Y.2d 332 (1989) ................................................................... 17 Sucese v. Kirsch, 606 N.Y.S.2d 60 (3d Dep’t 1993) ......................................... 50, 52 Ten Eyck v. Whitbeck, 156 N.Y. 341 (1898) .............................................................. 2 Tropp v. Lumer, 806 N.Y.S.2d 599 (2d Dep’t 2005) ............................................... 49 Weinberger v. Romero-Barcelo, 456 U.S. 305 (1982) ............................................ 28 Wood v. Rabe, 96 N.Y. 414 (1884) .......................................................................... 34 STATUTES N.Y. C.P.L.R. § 5602(a) ............................................................................................ 7 N.Y. Sur. Ct. Proc. Act § 201[3] .............................................................................. 17 OTHER AUTHORITIES 27A Am. Jur. 2d Equity (2015) ............................................................................... 28 The Federalist No. 80 (Alexander Hamilton) .......................................................... 30 Long, Alex B., Attorney Deceit Statutes, 44 U.C. Davis L. Rev. 413 (2010) ............................................................................................................. 42 Story, Joseph, Commentaries on Equity Jurisprudence (13th ed. 1886) .................. 2 Symons, Spencer, Pomeroy’s Equity Jurisprudence (5th ed. 1941) ........... 17, 29, 30 - v - INTRODUCTION Two of Rocky Aoki’s children colluded with his longtime lawyers to mis- lead Rocky, founder of the Benihana restaurant chain, into signing a document that they expected would result in their inheriting his fortune, rather than Rocky’s wife. The legal question here is whether the Appellate Division erred by adopting a rule that severely limits when such deception by fiduciaries can be remedied through the constructive-fraud doctrine. That rule would have serious ramifications, encouraging unscrupulous conduct by fiduciaries toward their beneficiaries. More specifically, the rule would harm attorney-client relationships, reducing lawyers’ incentives to avoid conflicts of interest and other conduct detrimental to their clients. The Appellate Division’s rule is also inconsistent with this Court’s precedent: This Court has long held that constructive fraud, a fixture of New York’s equity jurisprudence for over a century, is implicated “[w]henever … the relations between the contracting parties appear to be of such a character as to render it certain that they do not deal on terms of equality.” Cowee v. Cornell, 75 N.Y. 91, 99 (1878). Where one party “justifiably reposed” trust regarding a transaction in a fiduciary, id. at 100, the arrangement is “scrutinized with extreme vigilance, and clear evidence is required that the transaction was understood, and that there was no fraud, mistake or undue influence,” Matter of Gordon v. Bialystoker Ctr. & Bikur Cholim, Inc., 45 N.Y.2d 692, 698 (1978) (internal quotation marks omitted). And to uphold a transaction in such situations, “there must be clear proof of [its] integrity and fairness.” Id.; accord Ten Eyck v. Whitbeck, 156 N.Y. 341, 353 (1898) (citing Cowee, 75 N.Y. at 99). The purpose of the constructive-fraud doctrine is simple: to “grant relief” from unjust bargains struck in situations where “confidence is reposed” in a fiduciary “and that confidence is abused.” 1 Story, Commentaries on Equity Jurisprudence § 308, at 310 (13th ed. 1886). To serve that end effectively, constructive fraud must, like other equitable doctrines, “remain[] malleable to address a myriad of facts and circumstances.” Messner Vetere Berger McNamee Schmetterer Euro RSCG Inc. v. Aegis Grp. PLC, 93 N.Y.2d 229, 235 (1999) (discussing the part-performance doctrine). As this Court explained in its most recent exposition of constructive fraud, that doctrine cannot be confined within “absolutist rubrics.” Matter of Greiff, 92 N.Y.2d 341, 346 (1998). The Appellate Division contravened these principles here. Rather than preserve equity’s hallmark flexibility, the court imposed a novel straitjacket on the constructive-fraud doctrine, limiting it to circumstances in which the fiduciary charged with fraud is “a party to or ha[s] an interest in the subject transaction.” Matter of Aoki v. Aoki, 985 N.Y.S.2d 523, 527 (1st Dep’t 2014). That rule has no - 2 - grounding in the jurisprudence of this State and would deprive the doctrine of much of its equitable utility, with far-reaching consequences. The infirmity of the Appellate Division’s rule is illustrated by the facts here. Rocky Aoki’s longtime lawyers worked with his children-and without Rocky’s knowledge-to freeze out Rocky’s wife, appellant Keiko Ono Aoki, and secure Rocky’s business empire for themselves. Rocky, trusting his counsel, signed an apparently innocuous document buried in a stack of papers during a meeting scheduled for other purposes. As his lawyers knew, but Rocky did not, that document would irrevocably disinherit Keiko against his wishes. Indeed, that was the reason the lawyers had drafted the document. Those lawyers had a duty to advise Rocky of these consequences and to ensure that they reflected his desires. Instead, through what the trial judge here called either “deliberate[] obfuscat[ion] or accident[],” they said nothing about the document’s true purpose and effect. RA 36 n.43. The lawyers’ motive for participating in this scheme was a desire to find favor with the children and thereby secure for themselves millions of dollars in future legal and consulting fees. After conducting a bench trial and hearing directly from the key witnesses (including Rocky’s lawyers), the Surrogate (Kristin Booth Glen, S.) found that the lawyers had committed “misrepresentations, omissions or concealment” in the course of their fiduciary relationship with Rocky, RA 42 (brackets omitted), and - 3 - therefore concluded that the constructive-fraud doctrine required the release’s invalidation. The Appellate Division did not dismiss the Surrogate’s factual findings-they are in fact amply supported by the record-but instead held that constructive fraud did not apply as a matter of law because Rocky’s lawyers acted for the immediate benefit of a third party (Rocky’s children) rather than themselves. That decision did not rest on the “particularized … scrutiny” that this Court required in Greiff, 92 N.Y.2d at 347, but rather on the Appellate Division’s novel rule categorically limiting the circumstances under which constructive fraud applies and hence the burden of proving the fairness of a transaction shifts. This rule, which arbitrarily excludes a broad swath of deceitful conduct from the doctrine’s burden-shifting framework, will harm trusting clients, undermine attorney-client relationships, and prevent courts of equity from protecting victims of fiduciary malfeasance. Under the rule, constructive fraud does not apply even where attorneys work in secret against their own clients in a transaction-a classic example of fiduciary misconduct. That contravenes the black-letter principle that courts of equity do not enforce agreements produced by attorneys’ deception of their own clients. See, e.g., Baker v. Humphrey, 101 U.S. 494, 502 (1879) (attorney’s “breach of duty” to client “is ‘constructive fraud,’ and is sufficient” to void the transaction). Indeed, the constructive-fraud doctrine exists precisely to protect innocent parties from the harm such fiduciary deceit causes. - 4 - Even if the Appellate Division were correct that constructive fraud can apply only if the fiduciary charged with fraud is a party to or has an interest in the subject transaction, the facts here satisfy that rule. As explained, Rocky’s longtime attorneys worked in secret at the behest of two of his children to draft and execute a release that irrevocably disinherited his wife from the Benihana empire, even though Rocky had never expressed any such desire. While the children’s interest in that result is self-evident, the lawyers also stood to benefit directly, by cementing their relationships with children whose control over a large business could generate lucrative legal work. Regardless of whether the lawyers are treated as agents of the children-direct parties to the challenged transaction-or simply individuals with their own interest in the release, their faithless conduct satisfies the Appellate Division’s rule for invoking the constructive-fraud doctrine’s burden-shifting framework. The court erred in concluding otherwise. This Court should reject the Appellate Division’s novel rule as an improper departure both from established equitable principles and from this Court’s precedent. And because the evidence and the Surrogate’s findings fully support her conclusion that the children did not prove that Rocky signed the release voluntarily rather than because of his lawyers’ constructive fraud, the Appellate Division’s judgment should be reversed and the case remanded with instructions to reinstate the Surrogate’s invalidation of the release. - 5 - QUESTIONS PRESENTED In Cowee v. Cornell, 75 N.Y. 91 (1878), this Court held that “[w]henever … the relations between the contracting parties appear to be of such a character as to render it certain that they do not deal on terms of equality,” the burden of proving fraud “is shifted, the transaction is presumed void, and it is incumbent upon the stronger party to show affirmatively that no deception was practiced, no undue influence was used, and that all was fair, open, voluntary and well understood,” id. at 99-100. The questions presented are: 1. Does Cowee’s burden shifting apply only if the disloyal fiduciary is “a party to or ha[s] an interest in the subject transaction?” Matter of Aoki v. Aoki, 985 N.Y.S.2d 523, 527 (1st Dep’t 2014). The Appellate Division answered yes. Appellant contends that this ruling was incorrect. This issue was preserved in Respondent’s Appellate Division Brief (at 31-32), Respondent’s Appellate Division Motion for Leave To Appeal (at 18- 25), and Respondent’s Court of Appeals Motion for Leave To Appeal (at 20-31). 2. If the constructive-fraud doctrine is limited in the way the Appellate Division held, is that court’s test satisfied where a client’s attorneys, secretly working at the behest of the client’s counterparties, misled the client into signing a release that directly benefited those counterparties and indirectly benefited the attorneys themselves? - 6 - The Appellate Division answered no. Appellant contends that this ruling was incorrect. This issue was preserved in Respondent’s Appellate Division Brief (at 2-3, 35), Respondent’s Appellate Division Motion for Leave To Appeal (at 25- 33), and Respondent’s Court of Appeals Motion for Leave To Appeal (at 31-41). JURISDICTION The Appellate Division’s decision and order reversing the judgment of the Surrogate’s Court and granting summary judgment to respondents Devon and Steven Aoki is a final determination of all causes of action. This Court therefore has jurisdiction under N.Y. C.P.L.R. § 5602(a)(1)(i). Appellant moved this Court for leave to appeal on November 19, 2014. This Court granted leave on February 24, 2015. STATEMENT OF FACTS A. Background Rocky Aoki founded the Benihana restaurant chain in 1963. Matter of Aoki, 948 N.Y.S.2d 597, 598-599 (1st Dep’t 2012). For decades thereafter, he relied on Darwin Dornbush-“his trusted friend and attorney,” RA 38-for legal advice concerning both his personal affairs and Benihana. RA 645-646, 723. In 1998, Dornbush and his future law partner, Norman Shaw, created the Benihana Protective Trust (“BPT” or “the Trust”) to hold Rocky’s Benihana stock. RA 723. The Trust was designed to give Rocky the same power of appointment - 7 - over his Benihana assets that he would have if he owned them outright, and thus the trust instrument empowered Rocky to appoint BPT trustees and distribute BPT assets in his will. RA 645-646, 723. This appointment power was virtually unlimited; Rocky could designate anyone except himself, his estate, or any creditor of either. RA 723 & n.4. Dornbush was appointed as one of three trustees; two of Rocky’s children (Kevin and Kana) were the others. Matter of Aoki v. Aoki, 985 N.Y.S.2d 523, 524 (1st Dep’t 2014). In 2002, Rocky married Keiko Ono Aoki. RA 723. The couple did not enter into a pre-nuptial agreement, which prompted concern among Rocky’s children about their hoped-for inheritance of the Benihana empire. RA 305, 628, 723. Motivated by a worry that “Keiko is gonna take Dad’s wealth,” RA 18, 305, Rocky’s children enlisted Dornbush-Rocky’s attorney-to assist them in effectuating their own agenda for the disposition of Rocky’s assets, including his interests in Benihana, upon his death, see RA 18-19, 305, 308-311, 628. Though having no reason to believe that Rocky shared his children’s concerns in this regard, Dornbush “advised [Kevin and Kana] that a postnuptial agreement” between their father and Keiko “would resolve their concerns.” Aoki, 985 N.Y.S.2d at 525. Acting on this advice, Kevin and Kana went to dinner with Keiko and Rocky in 2002 (just two months after their wedding) and asked the couple to sign a post- - 8 - nuptial agreement. RA 723. Neither agreed to do so. Id. Subsequently, “Dornbush pressured Rocky to have Keiko sign a postnuptial agreement, but he too was unsuccessful.” RA 18 (citation omitted). B. The Irrevocable Release Dornbush and Shaw, who were law partners by this point, continued to pursue options that would effectuate the children’s agenda, irrespective of Rocky’s (i.e., their client’s) testamentary wishes. Having concluded that Keiko “would not sign anything,” RA 18, 388, the two lawyers set out to find an alternative that would replicate the effect of the rejected post-nuptial agreement, RA 18. The animating concern, as Shaw wrote in an August 2003 memorandum, was that “Rocky might be influenced to exercise his testamentary power of appointment under the BPT in favor of Keiko, thereby diverting [Rocky’s] stock to her at his death, and leaving Kevin and the other children out in the cold.” RA 628. This concern about Rocky being “influenced” could not have involved his competence, however. As the Appellate Division found in a prior case, Rocky was at all relevant times competent to make testamentary decisions. Matter of Aoki, 948 N.Y.S.2d at 598. Instead, the concern was that Rocky would voluntarily make decisions-which he had every right to make-that would benefit Keiko at the expense of his children (for whom Shaw and Dornbush were advocating at this point despite the obvious conflict of interest). - 9 - Motivated by this concern, “Shaw proceeded to explore ways to bar Keiko from access to BPT assets,” RA 724, although, again, Rocky had expressed no wish to do so, RA 18, 319-320. Shaw drafted two memoranda-neither of which was requested by or shared with Rocky-outlining options for ensuring that the children and not Keiko would inherit Rocky’s interest in Benihana. RA 724 & n.7; see RA 573-575, 661-668 (the two memos). Among the options Shaw considered was a partial release of Rocky’s interest in Benihana, which would “leave open the possibility of appointing some of the BPT assets ([Benihana] stock) to his wife.” RA 574. Shaw ultimately pursued a more extreme option, however (again without consulting his client), drafting an irrevocable release of Rocky’s power to appoint BPT assets to non-descendants. RA 398-401; see also RA 661.1 On September 24, 2002, Rocky arrived at Dornbush’s office for a meeting with the BPT Trustees (Dornbush, Kana, and Kevin). RA 724. Rocky was scheduled to sign several legal documents, including a codicil to his will. Id.; see RA 522-523, 591-592 (the documents). These documents constituted, in Rocky’s words, “a lot of papers.” RA 23; see also RA 35 (a “group of documents”). 1 The Appellate Division stated that Shaw “recommended” such a release. Aoki, 985 N.Y.S.2d at 525. Although the court appears to have been suggesting that such a recommendation was made to Rocky (as opposed to the children), nothing in the record supports that suggestion. RA 18-20 & n.12, 724. Indeed, Shaw testified that “the only time he met with Rocky about the Release was on the day of signing.” RA 20. - 10 - Buried somewhere in this “thick [stack of] paper” was the two-page irrevocable release that Shaw had drafted. RA 23; see RA 541-542 (the release). Innocuously titled “Partial Release of Power of Appointment Under New York Estates, Powers & Trusts Law §10-9.2,” the release uses the word “irrevocably” only once-outside the operative clause (that is, the clause that describes how the release will operate and that is set apart on the page, in block text, from the prefatory language). RA 541. Further obfuscating its true meaning, that clause is couched as an affirmative power rather than a limitation, providing: “I shall have a testamentary power to appoint any of the principal and accumulated net income remaining at my death to or for the benefit of any one or more of my descendants.” Id. Rocky, seeing (and indeed hearing of) this document for the first time, e.g., RA 34-35 & n.41, had it explained to him on the spot. (The Surrogate expressly did not credit Dornbush’s testimony that he had met with Rocky “at least three times to explain the release and read it” to him prior to this meeting. RA 35 n.41.) As Dornbush later testified, his explanation consisted of telling Rocky only that “when you sign this piece of paper, that means you will no longer be entitled in your will to leave the [Benihana] stock to anyone you choose. Right now you can give it to the ASPCA. After you sign this piece of paper, you can only leave it to your children.” RA 365. - 11 - Dornbush therefore did not mention the critical fact that this change in Rocky’s testamentary arrangements-unlike many others they had made previously-was irrevocable. See RA 366 (Surrogate: “Did you say anything to [Rocky] about irrevocability or revocability?” Dornbush: “I never used those words with Rocky.”). Neither did Shaw, who did not recall precisely what he told Rocky, see RA 403, but who testified that his abbreviated explanation of the release was made “in the context of” Rocky having met with Dornbush “for approximately an hour beforehand … the purpose of which was to explain what he was hopefully going to sign,” RA 411. Because his attorneys omitted the key point about irrevocability, the Surrogate found, “Rocky could reasonably have believed that, as with previous documents, he was always free to change his mind.” RA 35 n.42; see also RA 725 n.9 (concluding in denying summary judgment that “a trier of fact could find[ that] it was not unreasonable for Rocky to make the … assumption” that, like “the other documents presented for his signature,” the release was “similar to [documents] he had previously signed and was now changing”). Dornbush also failed to caution Rocky that, although the release did not mention Keiko or use the word “wife” or “spouse,” it would permanently disinherit her. RA 365. Nor-contrary to his ordinary punctiliousness about the tax consequences of Rocky’s testamentary decisions, see RA 43 n.53-did Dornbush - 12 - mention that by eliminating Keiko as a potential beneficiary, the release “would necessarily deprive Rocky’s estate of a marital deduction of what was then worth approximately $13 million,” RA 22. Dornbush and Shaw never sent Rocky a copy of the executed release. RA 37 n.44.2 C. Rocky Signs A New Codicil-Forcing Dornbush And Shaw To Belatedly Inform Him Of The True Meaning Of The Release In 2003, Rocky retained a new attorney, Joseph Manson, who drafted a codicil to Rocky’s will. RA 26, 728. That codicil (RA 648-652) appointed Keiko to receive twenty-five percent of the BPT assets outright upon Rocky’s death. See RA 729. It also provided that during her lifetime, Keiko would receive the income from the remaining seventy-five percent (which, upon her death, she could appoint in her will to one or more of Rocky’s descendants). RA 26, 648. The codicil made no mention of the irrevocable release. See RA 648-652. On August 28, 2003, Manson met with Dornbush to discuss Rocky’s will and related issues. RA 27. During the meeting, Manson asked for Dornbush’s 2 Rocky signed an amendment to the release in December 2002 (RA 543- 544), to conform his will to an IRS ruling that Shaw learned about after Rocky signed the release. See RA 24. This amendment, also drafted by Shaw, was “entirely technical.” Id.; see also RA 727-728. (Both the Surrogate and the Appellate Division treated this amendment as a second release, and thus frequently referred to the plural “releases.” E.g., RA 31; Aoki, 985 N.Y.S.2d at 524.) - 13 - legal opinion on the validity of the codicil. Id. Dornbush said nothing about the release-even though, if valid, it would nullify the codicil. Id. After meeting with Manson, Dornbush revealingly recorded his thoughts in a memorandum to file. RA 633-634. Dornbush speculated “as to why Rocky was asking our firm to express an opinion as to the validity of the Codicil” prepared by Manson. RA 633. “My guess,” Dornbush continued, “is that Kevin, in discussions with his dad, made mention of the fact that Rocky had signed ‘some paper’ when in [Dornbush’s] office, giving up his right to leave the assets of the Benihana Protective Trust to anyone other than his children and their descendants.” Id. Manson’s request for an opinion, in Dornbush’s view, “clearly resolve[d] the issue as to whether or not we had a duty to disclose the … Partial Release at th[at] time.” Id. (emphasis added). Dornbush was trepidatious about that disclosure, writing: “Undoubtedly, the fur will fly when … Keiko and Rocky[] discover the existence of the executed Partial Release.” Id. (emphasis added).3 Shaw sent Manson the requested opinion letter regarding the validity of the codicil on September 8. See RA 911-913. The letter opined the codicil “was invalid because Keiko was not a permissible appointee” under the release. RA 30; 3 Indeed, anticipating Rocky’s dissatisfaction with the state of affairs that the lawyers had concocted, Dornbush’s partner Shaw was at the time exploring “whether the partial release was binding” or “whether it c[ould] be revoked or otherwise disregarded or bypassed by Rocky.” RA 626 (emphasis added). - 14 - see also RA 911-912. Manson later testified that Rocky was “visibly upset” by the opinion letter. RA 251-252, quoted in RA 31. D. After The Children Retain Other Counsel, Shaw Continues Drafting Releases That Would Assist Them Four days after sending Manson the opinion letter, Shaw sent Kevin and Kana a letter (RA 636-637) that “tracked the language” of the opinion letter. RA 31. This follow-up letter also went to the two children’s (new) counsel, who had been retained in the wake of Dornbush’s much-belated recognition that his firm could not simultaneously represent Rocky and his children. RA 31, 657. Shaw had already sent the new attorney both his internal August 2003 memo (see supra p.9) and the irrevocable release. “There is no indication that Shaw sought or obtained Rocky’s consent to sharing these confidential documents with an attorney likely to represent interests adverse to [the] firm’s client, Rocky.” RA 29 n.32. Although the retention of new counsel made it clear that Shaw was not representing the children, he “continu[ed] to work on the issue of the releases, though for whom is unclear.” RA 31. Specifically, based on the new lawyer’s “discussions … with the children” (Kevin and Kana), RA 638, 641, Shaw drafted two further proposed releases (RA 639-640, RA 642-643) that Rocky could sign. Shaw shared these not with Rocky (his client and the proposed signatory) but with the children’s lawyer, RA 638, 641-who, again, the Surrogate found “likely … - 15 - represent[ed] … interests adverse to … Rocky,” RA 29 n.32. In neither case had Shaw “discussed this with Rocky.” RA 638, 641. Rocky, meanwhile, worked to effectuate his actual testamentary intent: In 2007, he signed a new will, his last. RA 545-570. Like the 2003 codicil, this will allocated to Keiko 25 percent of the BPT assets outright, income from the other 75 percent, and the right to appoint in her will one or more of Rocky’s descendants to receive that 75 percent. RA 547; see also RA 39. Rocky’s new will also provided a contingency plan in case the release he had signed was-“contrary to my desires” (RA 547)-upheld in court. In that event, two of Rocky’s children, Devon and Steven, were each to receive half the BPT assets. RA 547-548; see also RA 39. E. Proceedings Before The Surrogate’s Court 1. Rocky died in 2008. Aoki, 985 N.Y.S.2d at 526. The following year, the BPT trustees-consisting now of Kevin and Kana, their brother Kyle Aoki, and Kenneth Podziba-petitioned the Surrogate’s Court for a determination of the validity of the irrevocable release. RA 46-63. Those named as respondents included Keiko and three of Rocky’s other children: Devon, Steven, and Echo. RA 47-48. As the Surrogate explained, however, “[d]espite the way in which the case is titled, it basically pitted Devon and Steven against Keiko.” RA 14 n.1. Devon and Steven moved for summary judgment on the validity of the irrevocable release. RA 736-737. The Surrogate granted the motion with regard to - 16 - several defenses not relevant here, but otherwise denied it, holding that Keiko had advanced triable issues of fact as to whether Rocky’s signing of the release had been procured by constructive fraud. RA 733-734. The Surrogate noted there was “no question that Kana and Kevin had sought Dornbush’s legal advice and assistance for the purpose of limiting or denying Rocky’s ability to provide for his new wife on his demise.” RA 726. She also deemed it “clear that Dornbush had an impermissible conflict of interest,” adding, “there is also no question that, when the [irrevocable] Release was signed, neither he nor anyone else present informed Rocky of that conflict.” RA 726-727. In view of these (and other) “undisputed facts,” the Surrogate concluded that the release “clearly” could be challenged on grounds of constructive fraud. RA 730; see also 3 Symons, Pomeroy’s Equity Jurisprudence (“Pomeroy’s”) § 960(c), at 836-837 (5th ed. 1941) (under the constructive-fraud doctrine, an attorney “cannot, without the free and intelligent consent of his client … act both for his client and for one whose interest is adverse to or conflicting with that of his client in the same general matter”).4 4 This Court has regularly cited Pomeroy’s in expounding New York’s equity jurisprudence in general, and the equitable powers of the Surrogate in particular. See, e.g., State v. Barone, 74 N.Y.2d 332, 336 (1989); Matter of Wilson, 298 N.Y. 398, 404 (1949). There is no dispute that the Surrogate sits in equity. See Matter of Stortecky v. Mazzone, 85 N.Y.2d 518, 524 (1995) (“[T]he Surrogate’s Court Procedure Act … provides that the court shall ‘exercise full and complete general jurisdiction in law and in equity to administer justice in all matters relating to estates and the affairs of decedents.’ ” (emphasis added) (quoting N.Y. Sur. Ct. Proc. Act § 201[3])). - 17 - 2. After a three-day bench trial-at which the Surrogate heard live testimony and thus had the opportunity to weigh the credibility of witnesses, see RA 33-she issued a detailed opinion addressing two overarching factual questions. The first was whether Rocky was aware when he signed the release that it was irrevocable. RA 33. The second was whether, assuming Rocky was unaware, his signature was nonetheless “voluntary and not the result of misrepresentation or omission by his counsel and fiduciaries, Dornbush and Shaw.” Id. The Surrogate found that Keiko had carried her burden of persuasion on the first issue. RA 34. She then found that the children-who had the burden on the second issue under the constructive-fraud doctrine, because “the evidence … establishes … a fiduciary relationship between Rocky and Dornbush,” RA 41- had failed to carry that burden, RA 40. She therefore invalidated the release. More specifically, the Surrogate first found that Rocky was not told before signing the release that it was irrevocable. See RA 35 (“the language [Dornbush and Shaw] used” to explain the release, “could easily have been understood by Rocky as something less than ‘irrevocable’ ”). Rather, the Surrogate found, Rocky likely assumed that the release-like previous testamentary documents he had signed-was revocable, and hence that “he was always free to change his mind.” RA 35 n.42. That finding was supported, the Surrogate elaborated, by Rocky’s execution of the 2003 codicil and the subsequent “request that Dornbush and Shaw - 18 - confirm the validity of the codicil”-steps Rocky would not have taken had he known the releases were irrevocable. RA 17. Crediting testimony that Rocky “had no idea about the Releases,” the Surrogate thus found that “Rocky had no idea that he had forever surrendered his power to appoint some or all of BPT to Keiko.” RA 38. The Surrogate’s finding was also “strongly bolster[ed]” by Rocky’s deposition testimony in an earlier case. RA 37. There, Rocky had been unaware that a similar document-which created the BPT-was not revocable. See RA 37- 38 (“ ‘I have to say that I did not understand at the time the meaning of the word “irrevocable.” … I had asked my lawyer [Dornbush] [to] make this contract in order to protect me, so I had-have no idea about that, not being able to take it back.’ ” (quoting the deposition transcript)). The Surrogate explained that while: this testimony relate[d] to an earlier event, and a different document, it reinforces the view of a man who: believed that he always had the power to change his mind; was cursory, if not careless, about legal documents when prepared by his trusted friend and attorney, Dornbush; and may not have understood the implications of revocability and irrevocability…. RA 38; see also RA 729 (summary-judgment ruling noting that Rocky stated in an affidavit that September 2003-roughly a year after the release was signed-“was the first time he realized he had executed [an] irrevocable release[]-and that he had never had any intention of doing so”); RA 729 n.18 (quoting the affidavit). - 19 - The finding that Rocky did not know the release was irrevocable when he signed it was further supported, the Surrogate found (RA 44), by Dornbush’s August 2003 memo expressing concern about the need to “disclose” the (already- signed) release to Rocky-and predicting that the “fur will fly” when Rocky “discover[s] the existence of the Partial Release.” RA 633. This memo “clearly demonstrate[d] what had been concealed” from Rocky. RA 44. (The Surrogate “discredit[ed]” Dornbush’s claim that his comments in the memo were “no more than ‘sloppy’ drafting.” Id.) Devon’s and Steven’s contrary argument at trial (and on appeal) was that if Rocky truly had not known the release was irrevocable when he signed it, he would have sued to have it declared unenforceable as soon as he learned that fact. His failure to do so, they contended, was “compelling, if not dispositive, evidence of Rocky’s knowledge of the full content and effect of the Releases.” RA 38-39. The Surrogate considered this point but-in her role as fact-finder, and in view of all the testimony and evidence taken at trial-rejected it. Rocky, she explained, “may not have wanted to engage in such litigation, which would literally play out the conflict between Keiko’s interests and those of the children during his lifetime,” even though he was aware of “the possibility, necessity or likelihood of such litigation.” RA 40. Tellingly, “in his final will,” he anticipated such litigation by stating his “desires” that the release be invalidated, and providing a - 20 - contingency plan in the event those desires were frustrated. Id.; RA 547. The Surrogate thus found that Rocky’s choice not to litigate “the validity of the releases during his lifetime does not … necessarily lead to an inference that he knew that the Releases were valid, as Devon and Steven argue.” RA 40. 3. In the course of finding that Rocky did not know the release was irrevocable when he signed it, the Surrogate made two related findings about the conduct of Rocky’s lawyers. First, she found Dornbush’s explanation to Rocky of the release-that Rocky could no longer leave his interest in Benihana to “the ASPCA”-to be either “deliberately obfuscatory or accidental,” but in no event “relevant to what the Releases were actually intended by Dornbush and Shaw to do: keep Keiko … from ultimately taking the BPT assets.” RA 36 n.43; see also id. (“Although the loss of the ability to leave the stock to Keiko was what Rocky would undoubtedly have cared most about, Dornbush never told him that this would be a result of signing the release[.]”). Second, the Surrogate found that Dornbush’s and Shaw’s failure to apprise Rocky that signing the release could trigger roughly $13 million in additional tax liability “[c]urious[]” (RA22) and indeed “almost inexplicable” (RA 43 n.53) in light of the care otherwise taken to “consider-and provide for-tax consequences” in the firm’s work on Rocky’s estate planning. RA 43 n.53. “[T]his omission-of a financial consequence that might well have caused Rocky to change his mind, or not sign-… strongly - 21 - suggests a similar reason for not explaining to Rocky, who thought he could always change his mind,” that this decision was forever. Id. The Surrogate also discussed the conflicts of interests that beset Dornbush and Shaw, and the breaches of their fiduciary duties that resulted in Rocky’s signing the release. The Surrogate exhaustively cataloged the various ways in which Dornbush and Shaw had acted at the children’s behest or adverse to their client’s best interests (or both). She wrote, for example, that “Rocky’s children … had consulted with [Dornbush, who] enlisted his partner Shaw to craft a document that would be the ‘equivalent’ of [a] post-nuptial agreement.” RA 43. She also wrote that “Dornbush believed he was acting in Rocky’s-and the children’s-best interest, but [did not] clearly convey[] the meaning of [the release] to Rocky, and may have even misled him.” Id. Noting “the not insubstantial charges of conflict of interest urged against [Dornbush and Shaw] and accepted, at least in part, by the summary judgment decision,” which were “reinforce[d]” by “the exhibits admitted at trial,” RA 36 (citation omitted), the Surrogate found that: [T]he evidence … establishes … a fiduciary relationship between Rocky and Dornbush, … that Rocky reasonably reposed trust and confidence in his longtime attorney and friend, Dornbush.… [B]ecause of that trust and confidence, Rocky did not read the Release, but simply signed it; and further, that such reliance was reasonable. RA 41 (footnotes omitted). In short, the Surrogate wrote, “[b]ased on all the evidence presented here, and my assessment of the credibility of the witnesses, … - 22 - ‘misrepresentations, omissions or concealment’ [by Dornbush and Shaw] … is precisely what happened here.” RA 42 (other brackets omitted).5 4. Having found that Rocky did not know the release was irrevocable, and having further found the existence of a fiduciary relationship between Rocky and his lawyers in connection with the signing of the release, the Surrogate held that the constructive-fraud doctrine shifted the burden to the children to show that Rocky’s signing of the release was truly voluntary. RA 40-41. On the strength of the same evidence and findings addressed above-and the additional finding that “Rocky would not have signed the Releases had he understood their true import,” RA 43-the Surrogate ruled the children had not shown that Rocky’s “signing [of] the Releases was voluntary.” RA 33-34, 40; see also supra pp.1-2 (quoting this Court’s cases explaining what must be shown to uphold a challenged transaction under these circumstances). She therefore held the release invalid. RA 41, 44-45. F. The Appellate Division’s Decision The Appellate Division reversed “on the law,” announcing a novel categorical rule: “[F]or constructive fraud to apply, the fiduciary must be a party to or have an interest in the subject transaction.” Aoki, 985 N.Y.S.2d at 524, 527. 5 Shaw at one point drafted a memorandum (RA 644-647) addressing a possible legal challenge to the irrevocable release based on “the not insubstantial charges of conflict of interest.” RA 36. The memo self-servingly disclaimed “divided loyalt[ies],” RA 644, yet simultaneously observed that Rocky had “astonishingly married [Keiko] without the protection of a prenuptial agreement,” RA 646 (emphasis added). - 23 - Because “neither Dornbush nor Shaw w[as a] part[y] to the releases and thus could not benefit from them,” the court held, the doctrine did not apply and therefore the burden rested with Keiko “to prove that the releases were … procured by fraud.” Id. at 527. The only authority the Appellate Division cited to support its adoption of this rule was a decades-old Second Department decision. See id. (citing Matter of O’Hara, 445 N.Y.S.2d 201, 204 (2d Dep’t 1981)). In addition to placing the burden on Keiko, the Appellate Division largely disregarded the Surrogate’s findings, notwithstanding that the credibility-based findings of a trial judge “who saw and heard the witnesses … should be given the greatest weight.” Amend v. Hurley, 293 N.Y. 587, 594 (1944). For example, the Appellate Division asserted that there was a dearth of evidence that Dornbush or Shaw “represented to [Rocky] that the waivers were anything but irrevocable, or misled him regarding their effect.” Aoki, 985 N.Y.S.2d at 527. That assertion was contrary to the Surrogate’s detailed findings, which the Appellate Division did not explicitly address. (The assertion just quoted may itself have been intended as such a rejection; whether it was or not, the evidence recounted above-including that Dornbush, when asked to explain the release’s effect, never mentioned Keiko or irrevocability but simply stated that it restricted Rocky’s ability to give his assets to “the ASPCA,” RA 365-demonstrates that the assertion is incorrect.) And whereas the Surrogate in her role as fact-finder had considered but ultimately - 24 - rejected the inference that the children contended arose from Rocky’s failure to sue over the release during his lifetime, see supra pp.20-21, the Appellate Division deemed that inference “most significant[]” in concluding that Rocky understood the release was irrevocable, Aoki, 985 N.Y.S.2d at 528. Keiko moved the Appellate Division for leave to appeal to this Court. The Appellate Division denied leave on October 14, 2014. Keiko then moved this Court for leave to appeal, which this Court granted on February 24, 2015. SUMMARY OF ARGUMENT I. This case involves paradigmatic constructive fraud: Dornbush and Shaw took advantage of their longtime fiduciary relationship with Rocky, getting him to sign the release by concealing its irrevocability (which Rocky’s trust in them enabled them to do), and doing so for the benefit of Rocky’s children, on whose behalf the lawyers were working without Rocky’s knowledge. This Court has long held that bargains resulting from such deception will not be enforced by courts of equity. The Appellate Division reached a contrary result by embracing a novel rule that limits the constructive-fraud doctrine to situations in which the disloyal fiduciary is a party to, or has an interest in, the transaction. While even that rule, properly applied, covers the facts here, this Court’s precedent makes clear that constructive fraud should not be artificially limited in that way. Rather, it- - 25 - like all equitable doctrines-should be applied flexibly so courts can remedy injustice as the facts of each case require. Neither the Appellate Division nor Devon and Steven ever identified any case, from any jurisdiction, concluding that constructive fraud is inapplicable on facts analogous to those here. To the contrary, other states (to which this Court has often looked in situations like this) have held that constructive fraud does apply to transactions based on fiduciary deceit even though the fiduciary was not a party or direct beneficiary. The Appellate Division’s rigid limitation on constructive fraud would have grave and far-reaching ramifications, making it easier for fiduciaries to act to the detriment rather than the benefit of their trustees. The court’s new rule would also erode public trust in the legal profession and undermine a powerful tool for remedying attorney misconduct. This Court should reject that rule and order the Surrogate’s judgment reinstated. II. Even if the Appellate Division had articulated the correct legal test, it erred by concluding that, as a matter of law, the facts here do not satisfy it. First, Dornbush and Shaw were effectively parties to the transaction here. Because they had formed an attorney-client relationship with children who were direct beneficiaries of the release, and thus were acting as their agents (secretly), the attorneys were on the opposite side of the transaction from Rocky. Upholding the Appellate Division’s decision would therefore preclude finding constructive - 26 - fraud in situations where an attorney acts simultaneously in the service of their separate clients, concealing their conflicted loyalty from one side while surreptitiously aiding the other. Second, Dornbush and Shaw had a direct interest in inducing Rocky to sign the release. By putting Benihana under the children’s control, the attorneys ensured that lucrative legal fees from Benihana would continue for years after Rocky’s death. Upholding the Appellate Division’s decision would thus preclude a finding of constructive fraud in situations where it is assuredly warranted because a party has an indirect financial interest in the transaction. ARGUMENT I. THE APPELLATE DIVISION ERRED IN LIMITING APPLICATION OF THE CONSTRUCTIVE-FRAUD DOCTRINE This case is an archetypal example of constructive fraud: “[S]uperior knowledge of the matter derived from a fiduciary relation” was deployed by the fiduciary-lawyers against their client at a counterparty’s behest. Cowee v. Cornell, 75 N.Y. 91, 99 (1878). As a result, “unfair advantage” was not merely “probable,” id. at 100, but-the Surrogate found-actually present, RA 41. The Appellate Division could justify refusing to apply constructive fraud’s burden-shifting framework on these stark facts only by adopting a novel, categorical rule: “[F]or constructive fraud to apply,” it announced, “the [disloyal] fiduciary must be a party - 27 - to or have an interest in the subject transaction.” Matter of Aoki v. Aoki, 985 N.Y.S.2d 523, 527 (1st Dep’t 2014). That rule is an unprecedented and unwise development in the law of constructive fraud. No prior New York case has limited the doctrine to instances in which the faithless fiduciary either was a party to or had a direct interest in the challenged transaction. To the contrary, this Court has explained that its constructive-fraud precedent “eschew[s] absolutist rubrics,” in favor of an “enduring, nuanced balance of fair assessment.” Matter of Greiff, 92 N.Y.2d 341, 345, 346 (1998). The Appellate Division’s decision is inconsistent with those principles and would have profound effects well beyond this case. Its judgment should be reversed. A. The Hallmark Of Equity Jurisprudence, Including The Constructive-Fraud Doctrine, Is Flexibility 1. “The essence of equity jurisdiction” is “the power … to mould each decree to the necessities of the particular case.” Hecht Co. v. Bowles, 321 U.S. 321, 329 (1944), quoted in Weinberger v. Romero-Barcelo, 456 U.S. 305, 312 (1982). “In other words, equity’s purpose is to promote and achieve justice and to do so with some degree of flexibility.” 27A Am. Jur. 2d Equity § 2 (2015). Or as this Court has put it, equity’s purpose is to “soften[] where appropriate the harsh consequences of legal formalisms.” Rogers v. Rogers, 63 N.Y.2d 582, 586 (1984). - 28 - To accomplish this objective, a court of equity has “unquestionable authority to apply its flexible and comprehensive jurisdiction … as might be necessary to the right administration of justice between the parties.” Seymour v. Freer, 75 U.S. 202, 218 (1869); accord, e.g., SEC v. Capital Gains Research Bureau, Inc., 375 U.S. 180, 194 (1963). As Pomeroy’s explains, “the Chancellor always has had, and always must have, a certain power and freedom of action, not possessed by the courts of law, of adapting the doctrines which he administers.” 1 Pomeroy’s § 60, at 78; see also Beatty v. Guggenheim Exploration Co., 225 N.Y. 380, 389 (1919) (Cardozo, J.) (“A court of equity in decreeing a constructive trust is bound by no unyielding formula. The equity of the transaction must shape the measure of relief.”). So important are these attributes of equity that they endure even in the face of apparently mandatory statutory language: In Hecht Company v. Bowles, the U.S. Supreme Court held that, notwithstanding a statutory directive that an injunction “shall be granted” in particular circumstances, courts retained discretion to deny one. 321 U.S. at 328 (emphasis added). Cabining courts’ injunctive power, the Court explained, would constitute “such a major departure” from equity’s “long tradition” of “[f]lexibility,” “mercy,” and “practicality” that it could not “be lightly implied.” Id. at 329-330. - 29 - 2. The foregoing equitable principles apply fully to the constructive- fraud doctrine. “Constructive fraud is simply a term applied to a great variety of transactions, … which equity regards as wrongful, … and for which it gives the same or similar relief as that granted in cases of real fraud.” 3 Pomeroy’s § 922, at 626; see also The Federalist No. 80 (Hamilton) (“It is the peculiar province … of a court of equity to relieve against what are called hard bargains: these are contracts in which, though there may have been no direct fraud or deceit, sufficient to invalidate them in a court of law, yet there may have been some undue and unconscionable advantage taken of the necessities or misfortunes of one of the parties, which a court of equity would not tolerate.”). Accordingly, courts have been careful not to restrict the doctrine’s application to particular transactions. Rather, the doctrine applies wherever there is an “element of untruth,” “and even perhaps an intention to deceive, towards a third person, not a party to the transaction.” 3 Pomeroy’s § 922 n.2, at 626 (emphasis added). B. The Appellate Division’s Rigid Rule Finds No Support In New York Precedent Adhering to the equitable tenets just discussed, this Court has for over a century embraced a flexible approach to the constructive-fraud doctrine, explaining that application of the doctrine is warranted “[w]henever … the relations between the contracting parties appear to be of such a character as to render it certain that they do not deal on terms of equality.” Cowee, 75 N.Y. at 99; see also supra pp.1- - 30 - 2. Neither Devon and Steven nor the Appellate Division cited a case adopting the contrary approach. 1.a. To begin with, the Appellate Division’s decision contradicts this Court’s decision in Greiff. There, this Court rejected two inflexible (though diametrically opposed) theories of how constructive fraud applied to prenuptial agreements. One theory held that “prenuptial agreements [a]re presumptively fraudulent due to the nature of the relationship between prospective spouses”; the other, that they “may never be subject to burden-shifting.” Greiff, 92 N.Y.2d at 346. The Court rejected both positions, explaining that the “dispositive tests” for whether constructive fraud applies should not “pivot on [a] legalism.” Id. at 347. “Instead, a particularized … scrutiny,” i.e., case-specific analysis, is required. Id. Notably, the Surrogate’s ruling in Greiff (as here) rested on the particular facts presented, but the Appellate Division had “reversed, on the law.” 92 N.Y.2d at 344. This Court reversed the Appellate Division, holding that “the burden shift” provided by the constructive-fraud doctrine should be “neither presumptively applicable nor precluded.” Id. at 346. The Appellate Division repeated here the error that it committed in Greiff. Rather than considering, on the particular facts presented, whether shifting the burden would “serve the interests of fair conflict resolution as between proponents - 31 - [and] opponents” of the release, the court created what Greiff labeled an impermissible “absolutist rubric[].” 92 N.Y.2d at 346. In opposing leave to appeal, Devon and Steven argued that in Greiff this Court refused “to supply a categorical rule [only] to the determination of whether a confidential relationship exists.” Opp. to Motion for Leave To Appeal at 33 (N.Y. Ct. App.) (hereafter NYCA Opp.). That is incorrect. Greiff declared the need to avoid rigid rules not just about whether a specific relationship exists, but more broadly about when the burden to prove constructive fraud shifts-precisely the type of rule imposed here. Greiff’s key passage makes this clear: “We emphasize … that the burden shift is neither presumptively applicable nor precluded. We eschew absolutist rubrics that might ill serve the interests of fair conflict resolution as between proponents or opponents of these kinds of ordinarily useful agreements.” 92 N.Y.2d at 346 (emphasis added). b. Other decisions from this Court (as well as from other Departments of the Appellate Division) also support reversal here. For example, in Fisher v. Bishop, 108 N.Y. 25 (1888), this Court invalidated a mortgage extracted from an elderly farmer by his creditors after three days of pressure and threats-including by Wattles, a justice of the peace whom the farmer had often employed “as a legal adviser and conveyancer,” id. at 27. Expounding on the doctrine of undue - 32 - influence (a species of constructive fraud), this Court explained the basic principle in terms equally applicable here: One who has by reason of his supposed ability and integrity been employed by another as a confidential adviser … occupies a confidential position towards his employer, which, in good faith and common honesty, should preclude him from taking advantage of his situation, and using the information thus acquired to the detriment or disadvantage of his employer. Id. at 29 (emphasis added). Contrary to the Appellate Division’s rule here, Fisher did not describe the availability of an equitable remedy as hinging on Wattles being a party to the transaction or having any interest in it. Indeed, although the Appellate Division in Fisher described Wattles as a party to the mortgage, 36 Hun 112, 113 (4th Dep’t 1885), this Court did not even mention that point, much less treat it as dispositive. Rather, in this Court’s view it sufficed that the trusted fiduciary had exploited his position of trust “to the detriment or disadvantage of his employer.” Fisher, 108 N.Y. at 29; see also Callahan v. Callahan, 514 N.Y.S.2d 819, 821-822 (3d Dep’t 1987) (wife stated claim for constructive fraud where her husband’s attorney-“a trusted friend” on whom she “relied” for legal advice- induced her to surrender valuable property to the husband for a pittance). Fisher supplies the appropriate rule here. Where a party’s trusted fiduciary secretly works to advance the interests of another party to a transaction, and ultimately induces his client to agree to that transaction, the principles of equity and the concerns underlying the constructive-fraud doctrine apply with full force. - 33 - See Wood v. Rabe, 96 N.Y. 414, 425 (1884) (“[W]hen a person through the influence of a confidential relation acquires title to property, or obtains an advantage which he cannot conscientiously retain, the court, to prevent the abuse of confidence, will grant relief.”). The Appellate Division’s categorical bar here on the application of that doctrine cannot be reconciled with this Court’s precedent. 2. The only decision the Appellate Division cited to justify its new rule here was Matter of O’Hara, 445 N.Y.S.2d 201 (2d Dep’t 1981). See Aoki, 985 N.Y.S.2d at 527. Even that case does not support the rule. O’Hara involved deceit not by a current fiduciary but by former fiduciaries: Mrs. O’Hara’s former lawyers, who had “discontinued their representation” of her. 445 N.Y.S.2d at 203. (She signed the release of her attorneys in the presence of her new attorney, who represented her loyally. Id.) O’Hara’s rejection of constructive fraud therefore turned not on any categorical rule but on the court’s fact-specific conclusion that Mrs. O’Hara’s former lawyers were not her fiduciaries at the time the release was executed to their benefit. As the court explained, “no relation of attorney and client subsisted between her and appellants at the time she signed the release.… Appellants were thus not representing her by that date and she had the burden of coming forward with evidence sufficient to allege a prima facie case of fraud or other misconduct.” Id. at 204. O’Hara simply did not consider the scenario here, where a conceded fiduciary’s superior knowledge is secretly deployed against his - 34 - trusting client in a transaction. Like the Appellate Division, Devon and Steven have cited no prior case holding the constructive-fraud doctrine categorically inapplicable in those circumstances. To be sure, Cowee referred to “the relations between the contracting parties,” 75 N.Y. at 99, and several subsequent cases dutifully recite that formulation. That is an understandable shorthand, as in most cases the disloyal fiduciary is the deceived client’s contractual counterparty. But it is not an ironclad rule-as is clear from the fact that no prior decision holds the burden-shifting framework inapplicable where, as here, a fiduciary’s superior knowledge and position of trust are deployed on behalf of, and indeed in concert with, his client’s counterparty. Conversely, many cases have applied constructive fraud where the disloyal fiduciary was not a formal party to the transaction. See infra pp.35-39. C. The Appellate Division’s Rule Is Not Supported By Decisions From Outside This State In developing New York’s equity jurisprudence, this Court has looked to courts outside the State for guidance. See, e.g., Russell v. Briggs, 165 N.Y. 500, 504 (1901). Here, cases from other courts further support reversal of the Appellate Division’s decision. Most prominently, in Adams v. Cowen, 177 U.S. 471 (1900), the U.S. Supreme Court upheld the Sixth Circuit’s finding of constructive fraud based on fiduciary deceit by estate administrators-even though the transaction did not - 35 - directly benefit the administrators themselves. See id. at 483-485; see also Cowen v. Adams, 78 F. 536 (6th Cir. 1897). More specifically, the estate administrators procured from one of the decedent’s sons a “release” that amounted to “a surrender …, without any consideration, of practically his whole interest in his father’s estate.” 177 U.S. at 483. The release benefitted the other legatees, by increasing the amounts they inherited, yet the Court affirmed that equitable principles required invalidation of the release: [E]quity … looks with careful scrutiny upon all transactions between trustee and beneficiary, and if it appears that the trustee has taken any advantage of the situation of the beneficiary, and has obtained from him, even for only the benefit of other beneficiaries, large property without consideration, it will refuse to uphold the transaction thus accomplished. Id. at 484 (emphasis added); see also Cowen, 78 F. at 553 (“It makes no difference that the other legatees would ultimately obtain the benefit of the wrong,” because “[h]owever innocent of it they may be, it would come tainted to their hands.”). This holding simply cannot be reconciled with the rule the Appellate Division adopted here. Disputing this, Devon and Steven argued at the leave-to-appeal stage that administrators of a trust (such as a decedent’s estate) are direct parties to a transaction involving the trust. NYCA Opp. 39-40 (discussing Cowen). But that argument depends on a broad definition of “party to a transaction”-a definition the Appellate Division must have rejected because otherwise it would have - 36 - affirmed rather than reversed the Surrogate. Dornbush, after all, was a trustee of the BPT. See Aoki, 985 N.Y.S.2d at 524. He was therefore in effectively the same position as the administrators in Cowen. Likewise inconsistent with the Appellate Division’s rule here is Addis v. Grange, 192 N.E. 774 (Ill. 1934). There, the Illinois Supreme Court held that when considering whether the constructive-fraud doctrine applied, “[i]t is of no consequence by whom the undue influence was exercised-whether by a beneficiary or an outsider.” Id. at 777. Addis concerned the validity of a deed executed by an elderly widow on the advice of her disloyal fiduciary, a local banker. See id. at 774. Because the deed conveyed the widow’s property to a third party rather than the banker, he was not a party to the transaction, and at most had an indirect financial motive. See id. at 777. The court explained that this fact did not preclude a finding of constructive fraud (which the court found), stating that “although [the third party] was the nominal beneficiary of the unfaithful conduct of [the fiduciary banker], she stands upon no better footing than he would stand had the deed been made to him direct.” Id. The court also relatedly held that the fiduciary’s status as a non-party to the transaction did not preclude shifting the burden, explaining that “the [fiduciary] relation being established, the burden is upon the other party to show an absence of undue influence by establishing that the transaction was in good faith upon his part and was equitable and just between the - 37 - parties.” Id. at 776. That is the approach the Surrogate applied here, and that the Appellate Division erred in rejecting. Devon and Steven have argued that Addis’s unfaithful banker had an “interest” in the transaction because the third party owed a debt to the bank and intended to pay it off with the proceeds from the transaction. NYCA Opp. 38-39. That argument fails for essentially the same reason as their attempt to wave away Adams: It requires an expansive definition of “interest,” a definition the Appellate Division cannot have adopted because if it had the court would have ruled the other way. Under Devon’s and Steven’s definition, Dornbush and Shaw certainly had an “interest” in Rocky signing the release, because their role in inducing him to do so ensured that lucrative legal fees would continue to flow to their firm for years to come. See infra pp.53-57. In other words, they had specific expectations about how the children would use the “proceeds” they would receive from the transaction (their shares of Rocky’s inheritance), just as Devon and Steven say the banker in Addis did. Other cases from outside this State, while not addressing the issue directly, have similarly found constructive fraud despite the fact that the fiduciary was not a formal party to the transaction. For example, in Gardine v. Cottey, 230 S.W.2d 731 (Mo. 1950), the Missouri Supreme Court found constructive fraud where an attorney who represented both spouses in a divorce pressured the wife into - 38 - agreeing to a grossly inequitable property settlement, see id. at 739. The court saw no obstacle in the fact that the attorney was not a party to the contract; it sufficed that the attorney had not divulged to the wife the facts creating a conflict of interest. Id. at 739-740. The Michigan Supreme Court has similarly found constructive fraud where an attorney persuaded his client to deposit money with a savings and loan that was legally ineligible to receive the deposit and later became illiquid, even though the attorney was not a formal party to the transaction. Olitkowski v. St. Casimir Sav. & Loan Ass’n, 4 N.W.2d 664, 666-667, 672 (Mich. 1942); see also Richter v. Richter, 60 So. 880, 883-885 (Ala. 1913) (finding constructive fraud where husband induced wife to surrender her statutory dower rights-to the benefit of his children from an earlier marriage-for virtually nothing). These cases further confirm the Appellate Division’s error here. D. The Appellate Division’s Rule Would Have Far-Reaching Deleterious Ramifications 1. The decision below is infirm not only as a matter of precedent but also as a matter of first principles. Indeed, absent from the Appellate Division’s opinion is any explanation of why its rule is wise or desirable-that is, why even the clearest instances of fraud by a trusted fiduciary should be categorically excluded from the burden-shifting framework, simply because the fiduciary was not a party to the relevant transaction. See Aoki, 985 N.Y.S.2d at 527. There is no sound rationale. - 39 - Disagreeing, Devon and Steven have defended the Appellate Division’s rule on the ground that the contrary rule “would … obviate a client’s obligation to read and understand documents he is signing when drafted by his attorney and instead, make an attorney responsible for explaining every word, permutation and alternative to the satisfaction of the reviewing court years later.” Opp. to Motion for Leave To Appeal at 3 n.1 (App. Div.); see also NYCA Opp. 8 (same). That is meritless. As discussed, a “hallmark of equity” is that an equitable doctrine “remains malleable to address a myriad of facts and circumstances.” Messner Vetere Berger McNamee Schmetterer Euro RSCG Inc. v. Aegis Grp. PLC, 93 N.Y.2d 229, 235 (1999). Hence, rejecting the Appellate Division’s rule would simply ensure that, when circumstances warrant, a court can protect vulnerable clients from the harsh consequences of their fiduciaries’ disloyalty. If instead the circumstances of a case do not so warrant, then the same flexibility that Keiko submits courts should retain (but do not retain under the Appellate Division’s rule) would allow a court to reach the just result. See, e.g., Marx v. McGlynn, 88 N.Y. 357, 372 (1882) (holding the presumption that a transaction was void overcome where it was “clear that [testator] perfectly comprehended the contents of the will, and her subsequent correspondence … shows that it was just as she desired it should be”); id. at 371 (“If fairly made the law does not condemn [a transaction].”). - 40 - 2. In addition to lacking any salutary justification, the Appellate Division’s rule would have serious ramifications for beneficiaries of fiduciary relationships-including attorneys’ clients. The legal profession plays a vital role in New Yorkers’ lives, but “like all other great instrumentalities, it may be potent for evil as well as for good. Hence the importance of keeping it on the high plane it ought to occupy.” Baker v. Humphrey, 101 U.S. 494, 502 (1879); Meinhard v. Salmon, 249 N.Y. 458, 464 (1928) (Cardozo, C.J.) (fiduciaries owe beneficiaries “[n]ot honesty alone, but the punctilio of an honor the most sensitive”). A robust constructive-fraud doctrine-free of bright-line rules that unscrupulous attorneys could easily circumvent-gives attorneys a strong incentive to avoid conflicts of interest or other conduct detrimental to their clients. It also correspondingly assures clients that they can rely on the loyalty of their counsel, without fear that they will be deceived and left without a remedy. That is consistent with this Court’s decisions adopting rules to encourage attorneys to practice with integrity. See, e.g., Amalfitano v. Rosenberg, 12 N.Y.3d 8, 14 (2009) (construing attorney- deceit statute broadly in light of “an attorney’s special obligation to protect the integrity of the courts and foster their truth-seeking function”). A robust doctrine is particularly important at a time when “there is a strong sense among many non- - 41 - lawyers that dishonesty is prevalent in the practice of law.” Long, Attorney Deceit Statutes, 44 U.C. Davis L. Rev. 413, 472 (2010).6 E. Properly Applied, The Constructive-Fraud Doctrine Requires That The Release Be Invalidated Once the Appellate Division’s error in adopting its novel rule is recognized, there is no doubt that the Surrogate’s invalidation of the release should be upheld. The Surrogate found as fact that Rocky did not understand the release was irrevocable, that he signed it only because of the misrepresentations, omissions, or concealment of his fiduciaries, and that had he known the truth he would not have signed. See supra pp.17-23; see also, e.g., Barnard v. Gantz, 140 N.Y. 249, 256 (1893) (ignorance that a document was irrevocable is “equivalent” to ignorance of the very “nature and effect of the instrument”). Under a proper application of the constructive-fraud doctrine-under which the children have the burden to adduce “clear evidence … that the transaction was understood, and that there was no fraud, mistake or undue influence,” Matter of Gordon v. Bialystoker Center & Bikur Cholim, Inc., 45 N.Y.2d 692, 698 (1978)-those findings require, as the Surrogate concluded, that the release be invalidated. 6 The foregoing argument obviously sounds in public-policy considerations. This Court regularly considers the policy implications of its decisions when formulating rules for the application of equitable doctrines. See, e.g., In re Thelen LLP, 24 N.Y.3d 16, 31-33 (2014) (addressing public-policy concerns implicated by rules for partnership dissolution); Kirschner v. KPMG LLP, 15 N.Y.3d 446, 475- 477 (2010) (addressing public-policy consequences of in pare delicto doctrine). - 42 - To be sure, the release could be sustained if the Surrogate’s findings were held to be incorrect. But they cannot be. To the contrary, the findings are supported by the wealth of record evidence detailed in the Statement, including contemporaneous documents (some written by the disloyal fiduciaries themselves), live testimony that the fact-finder credited-again some of it from the attorneys- and various undisputed facts. The Surrogate’s opinion exhaustively catalogued that evidence, and offered a coherent and credible narrative based on it. There was of course contrary evidence, but none of it remotely warrants setting the Surrogate’s findings aside. For example, Dornbush testified that “Rocky had the Releases fully explained to him and read them[] on several occasions.” RA 42. The Surrogate, however, “mostly discredited” that testimony, based on, among other things, the lawyers’ failure to disclose to Rocky the crucial facts about the nature of the release. RA 35-36, 42; see also RA 42 n.51 (“Dornbush may-or may not-have met with Rocky on the 23rd and may have ‘explained’ the Release to him, but there is no evidence that he told Rocky it was ‘irrevocable.’ And I have found it more likely than not that he did not.”). That rejection was amply justified, given the attorneys’ strong incentive to provide self- serving testimony that would obscure their conflicts of interest. See RA 35-37. Another piece of contrary evidence was the fact, discussed earlier, that Rocky did not himself sue to invalidate the release before he died. The Surrogate - 43 - considered that fact, but explained why it did not mandate a finding that Rocky understood the release’s irrevocability and had signed voluntarily. See supra pp.20-21; cf. Gardine, 230 S.W.2d at 736, 751 (invalidating transaction due to constructive fraud even though the defrauded party took “no steps to set aside the contract and deed [that unfairly favored her husband] until after her husband’s death”). That explanation is entirely reasonable. Indeed, Rocky’s reluctance to oppose the release by suing during his lifetime is easy to understand: It would have involved years of litigation centered around the fact that he was duped by someone in whom he had placed his trust. In any event, while another fact-finder might have been able to take a different view on this point if viewed in isolation, the evidence must be taken as a whole. Once that is done, i.e., once all the other evidence detailed above (and in the Surrogate’s opinion) about the conduct of the attorneys and the children is considered together, it is clear the Surrogate’s findings were correct. Indeed, even the Appellate Division did not declare any of the Surrogate’s findings erroneous. Instead, it simply ignored them. For example, the court averred that “[t]here is nothing to indicate that the attorneys either concealed from or did not affirmatively provide Rocky with any information he needed to make an informed decision.” Aoki, 985 N.Y.S.2d at 527. The Surrogate reasonably found, to the contrary, that Dornbush’s explanation to Rocky of the release-that Rocky - 44 - could no longer leave his interest in Benihana to “the ASPCA”-was either “deliberately obfuscatory or accidental,” but in no event “relevant to what the Releases were actually intended by Dornbush and Shaw to do: keep Keiko … from ultimately taking the BPT assets.” RA 36 n.43. In fact, the Surrogate found, “[a]lthough the loss of the ability to leave the stock to Keiko was what Rocky would undoubtedly have cared most about, Dornbush never told him that this would be a result of signing the release[.]” Id. (emphasis added). The Surrogate also found, as explained, that Dornbush misled Rocky by failing to tell him the release was irrevocable. E.g., RA 36 (“[It] was never said[] that once the release is signed, Rocky was giving up forever the right to change his mind, and execute a further release or subsequent document.”). There is no way to reconcile these clear and fully supported findings with the Appellate Division’s declaration quoted above. And as noted, under this Court’s precedent the Surrogate’s credibility- based findings, made after she had the opportunity to observe the witnesses’ testimony, are entitled to “the greatest weight.” Amend v. Hurley, 293 N.Y. 587, 594 (1944). The Appellate Division also blamed Rocky for not having read the release before he signed it. See Aoki, 985 N.Y.S.2d at 527. Citing only Appellate Division precedent, the court stated that “a party who signs a document without any valid excuse for having failed to read it is conclusively bound by its terms.” - 45 - Id. But see Albany City Sav. Inst. v. Burdick, 87 N.Y. 40, 46-47 (1881) (“It has certainly never been announced as the law in this State that the mere omission to read … a written instrument should bar any relief … on account of mistake or fraud.”). The Appellate Division went on to assert that “[t]he record is devoid of any excuse, let alone a valid excuse, for failing to read the release prior to signing it.” Aoki, 985 N.Y.S.2d at 527. That too is contrary to the Surrogate’s findings. As the Surrogate thoroughly recounted, the “excuse” was that Rocky trusted that Dornbush, who had been his attorney for three decades, would not provide an explanation for the release that was misleading or omitted critical details-such as its irrevocability. To the extent the Appellate Division was suggesting that invalidation was not warranted despite the Surrogate’s findings that there was a good excuse for Rocky’s failure to read the release, that is meritless. Indeed, facing comparable circumstances, this Court held in Smith v. Smith, 134 N.Y. 62 (1892), that a deed had to be canceled because the plaintiffs’ aunt and her “respected” attorney misrepresented its terms, id. at 65, 66. The Court acknowledged that plaintiffs “had intelligence enough to understand the effect of the deed if they had read it,” but excused them from not having done so because “the circumstances under which they were approached, and the representations made, and their relations to their aunt and confidence in her, put them off their guard; if indeed it could have - 46 - occurred to them to be on their guard against her.” Id. at 65. Plaintiffs’ failure to read was all the more excusable, the Court added, because the aunt “was supported by the attorney whom [plaintiffs] respected, and whose presence and participation were the assurance that it was legally right.” Id. at 65-66.7 Smith’s reasoning applies fully here. Finally, the Appellate Division suggested, and the children have argued, that the record does not support a finding that Rocky lacked the English skills to understand the import of the release. See Aoki, 985 N.Y.S.2d at 527; NYCA Opp. 14, 22 n.12, 54. But Keiko’s submission that the release is invalid does not depend on concluding that Rocky lacked any particular command of English. It rests instead on the fact that Rocky placed his trust in his attorneys not to have him sign a legal document that irrevocably limited his ability to dispose of the fruits of his life’s labor unless they were sure he understood what he was doing-let alone mislead him into doing so. Because the record supports the Surrogate’s findings 7 Accord Matter of Smith, 276 N.Y.S. 646, 652-654 (4th Dep’t 1935) (the “general rule that a party … cannot escape the legal effect of [signing a document] by a plea that he did not know the contents of the paper” is “not applicable” where a litigant was reassured by a fiduciary “in whose honesty and integrity she had a right to rely, that the nature of the instrument which she was asked to sign was a partnership agreement, and not a relinquishment of her rights to his property”); Matter of Paul, 482 N.Y.S.2d 121, 123 (3d Dep’t 1984) (question of fact as to whether wife’s release consenting to the settlement of her husband’s estate was invalid due to attorney’s questionable representations regarding the settlement’s terms). - 47 - that the attorneys did precisely that, she was correct in concluding that the doctrine of constructive fraud (unencumbered by the Appellate Division’s new rule) requires the invalidation of the release. II. THE FACTS HERE SATISFY THE APPELLATE DIVISION’S RULE Even if the Appellate Division’s rule regarding the scope of the constructive- fraud doctrine were correct, that rule was satisfied here. Contrary to the court’s view, Dornbush and Shaw both were effectively parties to and “ha[d] an interest in” the release. Aoki, 985 N.Y.S.2d at 527. The attorney-client relationship that Dornbush and Shaw had formed with Kevin and Kana made the lawyers-as the two children’s agents-effective parties to the release. And the attorneys had their own personal interest in the release, namely that it would help them secure future business from the children, who would be in a position to control Benihana, Inc. The Appellate Division reached the contrary conclusion by implicitly adopting definitions of “party” and “interest” so narrow that they exclude much of the class of self-interested, deceitful fiduciary conduct that the constructive-fraud doctrine is designed to redress. In particular, the Appellate Division’s ruling means that two forms of attorney disloyalty present here can never constitute a qualifying “interest” in a transaction: (1) acting simultaneously in the service of separate clients who are parties to an agreement, while concealing that conflict from one of them; and (2) acting to secure potential future financial gain that might - 48 - result from a client’s entering into a contract with a third party. Those exclusions are illogical and inequitable, and will have far-reaching consequences, injecting significant uncertainty into fiduciary relationships. A. The Attorney-Client Relationship Between The Lawyers And The Children Made The Lawyers Effective Parties To The Release 1. In holding that its new constructive-fraud rule was not satisfied, the Appellate Division overlooked the fact that Dornbush and Shaw, as agents of Kevin and Kana, were effective parties to the release. An attorney-client relation- ship had formed when the children approached Dornbush to ask him to prevent Keiko from “tak[ing] dad’s wealth,” RA 305, and the attorneys drafted the release for Rocky’s signature to effectuate that goal. The lawyers were therefore in an attorney-client relationship with Kevin and Kana when the release was signed. The Appellate Division itself recognized this, stating that “Kana and Kevin met with Dornbush to express their concern that their father did not have a prenuptial agreement” and “Dornbush advised them that a postnuptial agreement would resolve their concerns.” Aoki, 985 N.Y.S.2d at 524-525 (emphasis added). Under New York law, “formality is not essential to the formulation of an attorney-client relationship; rather, it is necessary to look at the words and actions of the parties to ascertain if such a relationship was formed.” McLenithan v. McLenithan, 710 N.Y.S.2d 674, 675 (3d Dep’t 2000) (internal quotation marks omitted); see also, e.g., Tropp v. Lumer, 806 N.Y.S.2d 599, 600 (2d Dep’t 2005). - 49 - All that is required to create the relationship is “an explicit undertaking to perform a specific task.” Sucese v. Kirsch, 606 N.Y.S.2d 60, 62 (3d Dep’t 1993). Here, the Surrogate found that “Kevin and Kana … approached Dornbush and Shaw about protecting the interests of Rocky’s children and grandchildren as potential beneficiaries of the BPT,” RA 723-724, and further found that Dornbush and Shaw pursued that task by drafting the release for Rocky’s signature, RA 724- 725. In fact, the BPT trustees, including Kevin and Kana, conceded before the Surrogate that “ ‘Kana and Kevin sought legal advice [from Dornbush] individually, as potential beneficiaries of BPT (and on behalf of Rocky’s other children and grandchildren as potential beneficiaries of BPT).’ ” RA 724 n.6; see also Respondents-Appellants’ App. Div. Br. 52 (conceding that Dornbush “may have incidentally acted in the children’s potential interest”); NYCA Opp. 47 (acknowledging that Dornbush and Shaw “may have incidentally conferred a potential benefit on [Rocky’s] children”). The Surrogate thus concluded at summary judgment that there was “no question that Kana and Kevin had sought Dornbush’s legal advice and assistance for the purpose of limiting or denying Rocky’s ability to provide for his new wife,” RA 726, and that Dornbush “had an impermissible conflict of interest” because of his “ ‘simultaneous representation’ ” of Rocky and Kevin and Kana, RA 726, 727 n.14. This conclusion is supported by, among other things, memos that the lawyers drafted after the release was - 50 - signed, which make clear that their loyalties were divided between two sets of clients. See RA 728-729 & nn.16-17. One memo, for example, expressed dismay that Rocky might bequeath his fortune to Keiko, “ ‘leaving the children out in the cold with respect to the Aoki empire.’ ” RA 728 n.16. As the Surrogate found, “[i]t is difficult to read this memo as consistent with an undivided loyalty to … Rocky.” Id. The Surrogate reaffirmed these conclusions after trial, where as noted she had the opportunity to directly “assess[] … Dornbush’s (and, to a lesser extent, Shaw’s) credibility.” RA 36. She found that the evidence “reinforce[d] the charge of conflict of interest and violation of their professional responsibility to Rocky.” Id.; see also RA 43 (“Dornbush believed he was acting in Rocky’s-and the children’s-best interest” (emphasis added)). And here too, that evidence included the attorneys’ own words: In 2003, Shaw wrote a detailed memorandum explaining that he drafted the release in response to the “concern … communicated to me as well as to Darwin [Dornbush]” by Kevin that “Rocky might be influenced to exercise his testamentary power of appointment under the BPT in favor of Keiko, thereby … leaving Kevin and the other children out in the cold with respect to the Aoki empire.” RA 628. Shaw wrote that he was “asked, definitely by Kevin … if there were some way that Rocky could put in place an arrangement that - 51 - would provide the necessary assurance” that the children would receive the BPT assets. Id. (emphasis added). Devon and Steven have argued that Kevin and Kana did not become clients simply because “a testamentary document potentially benefit[ed] them.” NYCA Opp. 47. That is incorrect. The two children entered into an attorney-client relationship with Dornbush and Shaw when they asked the attorneys to do something (protect their financial interests) and the attorneys did it (by drafting the release). See Sucese, 606 N.Y.S.2d at 62. Devon and Steven have offered no response to the Surrogate’s conclusion that Dornbush engaged in “ ‘simultaneous representation’ ” that should have been disclosed. RA 726, 727 n.14. The Appellate Division did not expressly disturb any of the Surrogate’s findings regarding the existence of an attorney-client relationship between the children and the lawyers. Nor could it plausibly have done so, given that “[t]he determination of the Surrogate, who presided at the trial and heard all of the testimony, is entitled to great weight” and generally “should not be set aside on appeal.” Matter of Feinberg, 543 N.Y.S.2d 300, 300 (2d Dep’t 1989); see also Matter of Poggemeyer, 449 N.Y.S.2d 12, 12 (2d Dep’t 1982) (“The decision of a fact-finding court … should not usually be disturbed on appeal[.]”). Rather, as noted the Appellate Division’s opinion itself indicates that there was an attorney- client relationship. See Aoki, 985 N.Y.S.2d at 524-525, quoted supra p.49. In - 52 - short, Kevin and Kana engaged Dornbush as their attorney in order to prevent Rocky from giving his assets to Keiko-even if it was his clear desire to do so. 2. By forming an attorney-client relationship, the children and lawyers took on the roles of principals and agents. See, e.g., Matter of Coop. Law Co., 198 N.Y. 479, 483 (1910) (“The relation of attorney and client is that of master and servant in a limited and dignified sense[.]”); Poucher v. Blanchard, 86 N.Y. 256, 260 (1881) (“The law which regulates the relation between attorney and client is that of agency[.]”). For purposes of the Appellate Division’s new constructive- fraud rule, then, Dornbush and Shaw should have been treated as parties to the release. Their participation in the drafting and execution of the release, coupled with their fiduciary relationship with Rocky, raises the very same possibility for mischief that courts of equity sought to address in fashioning the constructive- fraud doctrine in the first place. Consistent with that commonsense proposition, numerous jurisdictions have construed the constructive-fraud doctrine to encompass the actions of attorneys who have violated their fiduciary duties. See supra pp.38-39. The Appellate Division erred in not doing likewise here. B. Dornbush And Shaw Had Their Own “Interest In The Transaction” The facts here also satisfy the second prong of the Appellate Division’s rule (although the rule’s disjunctive phrasing, see Aoki, 985 N.Y.S.2d at 527, shows that only one prong need be met). That is because Dornbush and Shaw had their - 53 - own interest in the transaction: protecting a steady source of future lucrative work by ensuring that their patrons would have control of the Benihana purse strings. The definition of “interest” that the Appellate Division implicitly adopted in order to reverse the Surrogate is so narrow that it would exclude a wide array of self- dealing fiduciary behavior from the protection of the constructive-fraud doctrine. 1. During the relevant period, it was clear to Dornbush and Shaw that a split in the family was forming, with Rocky and Keiko on one side and the children on the other. RA 18-19, 305-306; see also RA 25-26 (section of opinion entitled “[w]orsening relations between Rocky and BPT Trustees”); RA 631 (memo drafted by Shaw considering the release’s effect from the perspective of the “Rocky/Keiko Side” and the “Current Trustees’ Side” (i.e., Dornbush and the children) (italics omitted)). By drafting a release that ensured the children would maintain control over the Benihana empire for years if not decades to come, Dornbush and Shaw chose a side in that split, cementing their roles as trusted legal advisors to the children and thereby ensuring profitable business and a continued trusted advisory role for themselves. And the first rewards for the lawyers’ efforts were not long in coming: When the BPT trustees (including Kevin and Kana) wanted a legal opinion on whether the release effectively nullified the subsequent codicil to Rocky’s will, they turned to Dornbush’s firm, and to Shaw in particular, for the analysis. RA 636-637 (letter written to BPT trustees “in response to your - 54 - request … for our opinion as to whether a Codicil signed by your father, Rocky … would be valid to dispose of the assets of the BPT”); see also RA 445-446. Devon and Steven have suggested that the release did not benefit Kevin and Kana (and thus did not increase the likelihood that they would retain Dornbush and Shaw in the future) because the document “in no way ensured that Kana and Kevin would become beneficiaries of the BPT.” NYCA Opp. 49. That claim strains credulity. Before Rocky signed the irrevocable release, he could leave his assets to virtually any person or entity (including Keiko). After the release was signed, by contrast, he could leave those assets only to a handful of people-including Kevin and Kana. See RA 541. The notion that Kevin and Kana did not benefit from the elimination of virtually all of their competitors as potential beneficiaries is absurd. And the related notion that Dornbush and Shaw were not anticipating the boon that would result from conveying this benefit to Kevin and Kana only heightens the absurdity. 2. The release also solidified Dornbush’s and his firm’s roles as corporate advisors to Benihana. As the Delaware Chancery Court explained, Dornbush served as a director and corporate secretary of the public company, Benihana, Inc., and, “in effect, act[ed] as its general counsel”-including in connection with the company’s efforts to dilute Rocky’s stake in Benihana, Inc. Benihana of Tokyo, Inc. v. Benihana, Inc., 891 A.2d 150, 157 (Del. Ch. 2005), - 55 - aff’d, 906 A.2d 114 (Del. 2006).8 This was lucrative work: In 2003, the year after Dornbush persuaded Rocky to sign the release, Dornbush’s firm earned more than $1 million in legal fees from Benihana, while Dornbush himself received an additional $5,000 a month from Benihana for consulting services. Id. at 187. Dornbush feared that these emoluments would disappear if Keiko gained control of sufficient Benihana stock. As the Chancery Court put it, Dornbush “ ‘shared’ a ‘concern’ that, upon obtaining control of Benihana, Keiko Aoki, would ‘remove all of the people who were there for 20 years of service.’ ” Benihana, 891 A.2d at 186 (quoting Dornbush’s testimony); see also id. at 158 n.20 (Dornbush telling Benihana CEO that “if Keiko were hostile to management, ‘you … have employment agreements in place, and you’ll have to weather the storm or do what you want to do at that point in time’ ”); RA 331-332 (discussing Dornbush’s role in the struggle to prevent Keiko from obtaining control of Benihana: “THE COURT: ‘You are all going to get fired, Keiko is going to take over everything.’ ”). The irrevocable release, which prevented Rocky from naming Keiko as a beneficiary of the BPT, secured Dornbush’s personal roles as a director, officer, and paid consultant to Benihana, Inc., as well as his firm’s work for the company. 8 Dornbush also provided legal services to Benihana of Tokyo, for which he was a director as well. RA 298, 355-356. - 56 - That concrete benefit easily satisfies the Appellate Division’s new rule and therefore shifts the burden to the proponents of the irrevocable release. Excluding such blatant self-interest would unjustifiably constrain the doctrine’s ability to protect trusting clients from fiduciary self-dealing. Indeed, Devon and Steven themselves state that the “interest in the subject transaction” requirement should be construed “ ‘flexib[ly].” NYCA Opp. 43 n.28; see also id. at 6-7 (suggesting that the “interest” requirement “allow[s] for flexibility in a doctrine that seeks to prevent fiduciaries from benefitting themselves by using their position to take advantage of parties to whom they owe duties”). In short, while this Court should reject the Appellate Division’s rule and thereby reaffirm the imperative that attorneys and other fiduciaries adhere to the highest standards in regard to their beneficiaries, the facts of this case satisfy the Appellate Division’s rule. Reversal of the Surrogate’s decree was thus improper. CONCLUSION The Appellate Division’s decision should be reversed and the case remanded with instructions to reinstate the Surrogate’s decree invalidating the irrevocable release. Date: May 11, 2015 - 57 - Respectfully submitted. /s/ Alan E. Schoenfeld Seth P. Waxman (pro hac vice) Alan E. Schoenfeld Daniel S. Volchok (pro hac vice) WILMER CUTLER PICKERING Adam I. Klein HALE AND DORR LLP Thomas G. Sprankling (pro hac vice) 7 World Trade Center WILMER CUTLER PICKERING 250 Greenwich Street HALE AND DORR LLP New York, New York 10007 1875 Pennsylvania Avenue N.W. (212) 230-8800 Washington, D.C. 20006 (202) 663-6000 seth.waxman@wilmerhale.com Attorneys for Appellant Keiko Ono Aoki