Charles Caleb Colton once famously observed that “Imitation is the sincerest form of flattery.”1 One place where that maxim seems not to have worked out well (at least in one man’s opinion) is in the interpretation/application of the attorney-client privilege. Given the not so slow but steady erosion in that privilege,2 it has been disheartening to see a New York court recently reach out and imitate the courts of Ohio.

A number of years ago, Ohio took the lead of essentially eviscerating the attorney-client privilege in the insurance coverage context.3 Now, unfortunately, that bad seed has spread to New York and we appear to be following suit.4 As Don Corleone once plaintively asked: “How did things ever get this far?!”

As Ohio Goes, So Goes…

In 2001, the Ohio Supreme Court in Boone v. Vanliner Co.,5 ruled that the mere allegation of bad faith in the denial of insurance coverage meant that the insured plaintiff could discover internal company communications covered by the attorney-client privilege. In Boone, a truck driver had been injured in an accident and sought coverage from his insurer. When the company denied the claim, the driver sued for bad faith and sought a declaratory judgment compelling coverage. In discovery, the driver sought production of the company’s claims file; the company brought on a protective order motion on the ground, inter alia, that a number of documents in that file were covered by the attorney-client privilege.

The trial court, after an in camera inspection, ordered disclosure of certain of the documents. On appeal to Ohio’s Tenth District Court of Appeals that order was reversed on the ground that the documents were privileged. The Ohio Supreme Court thereafter reversed the court of appeals, ruling that the attorney-client communications were not privileged and thus subject to discovery by the plaintiff.

Citing a medical malpractice case (Moskowitz v. Mt. Sinai Med. Ctr.) in which the court had previously held that “documents and other things showing the lack of a good faith effort to settle by a party or the attorney acting on his or her behalf are wholly unworthy of the protections afforded by any claimed privilege,”6 four Ohio Supreme Court justices extended that precedent to the Boone situation. The four justices were unbowed by the policy argument that rejecting the privilege would discourage insurers “from seeking legal advice as to whether a certain claim is covered under a policy of insurance,” and held that pleading a bad faith denial of coverage would enable the insured to discover all attorney-client communications relating “to the issue of coverage that were created prior to the denial of coverage.”7

Three justices of the Ohio Supreme Court lodged a vociferous attack on the majority opinion. Not only did they challenge the majority’s understanding of Moskowitz and its extension to Boone, the dissent made clear that the majority simply did not understand the basic principles underlying the attorney-client privilege, as enunciated by the U.S. Supreme Court.8 Even more troubling was the fact that the majority was “effectively equat[ing] an insurer’s communications with its attorneys prior to a denial of coverage, in any case alleging bad faith, with communications in furtherance of a civil fraud”—i.e., the majority was really invoking the crime-fraud exception to the privilege.

But, as the dissent continued, “bad faith by an insurer is conceptually different from fraud…. [A]n actual claim of fraud required proof of a false statement made with intent to mislead.” Under the majority’s standard, however, only an allegation of bad faith in the complaint would be sufficient to open up the insurer’s privilege file—an extraordinarily low threshold that a large number of states have explicitly rejected.9 Finally, the dissent not only underscored the perverse consequences of the majority’s ruling—that “open and honest discourse between attorney and client” would obviously be impaired—the justices also hinted at the future impact the ruling would have on the insurance industry located in Ohio.

Given the red flags planted by the Boone dissent, the Ohio General Assembly swung into action. Ultimately, it enacted a law to modify/counteract Boone. Ohio’s privilege statute (R.C. 2317.02) was amended to require (i) in camera inspection of privileged materials, and (ii) that the insured make a prima facie showing of bad faith, fraud, or criminal misconduct by the insurer (before privileged communications would be made discoverable). On Jan. 8, 2007, Governor Ted Strickland—in his first day in office—vetoed the statute. But on Aug. 1, 2007, the Ohio Supreme Court ruled that the law had come into effect on Jan. 6, 2007, and thus the governor’s veto two days later was invalid.10

Notwithstanding that action, judicial interpretation of the statute has not been encouraging. In In re Professionals Direct Insurance,11 where there was a bad faith insurance denial claim, a federal magistrate required the insurer, per Boone, to disclose all attorney-client communications pre-dating the denial of coverage. In response to the insurer’s argument that the Assembly’s amendment required a prima facie showing of bad faith prior to disclosure, the magistrate ruled that “[o]n its face [the statutory amendment] applies only to testimony. It does not mention documents.”

On appeal to the U.S. Court of Appeals for the Sixth Circuit, that court declined to vacate the magistrate’s order, ruling that it was not clearly erroneous. The court went on to determine that the amended statute, which became effective on Oct. 31, 2007, did not apply retroactively (and thus not to that case); finding the statute thus inapplicable, the court did not feel required to “interpret its scope.” And while some subsequent courts have suggested that Boone did not survive the Assembly’s action,12 there has not been a lot of bright line clarity for Ohio insurers of late.

…So Goes New York

Into this state of affairs, the Supreme Court of Nassau County recently made its presence known. In Melworm v. Encompass Indemnity,13 a married couple filed suit against their insurance company to compel it to pay for damage done to their boat. During discovery, the insurance company declined to produce materials covered by the attorney-client privilege (counsel had been specifically retained to advise the company with respect to its rights and obligations vis-à-vis the plaintiffs’ insurance policy); the plaintiffs litigated the refusal on the ground that the insurance company’s lawyers’ work and related communications were not privileged—and the Supreme Court agreed with that position!

Citing seminal New York Court of Appeals decisions on the privilege (e.g., Rossi14 and Spectrum Systems15) as a prologue, the Supreme Court went on to rule (i) that insurer had clearly retained counsel for the purpose of rendering legal advice on the matter in dispute, but (ii) that if the work by counsel was “primarily” done to investigate the plaintiffs’ claims, then any communication therewith would not be covered by the attorney-client privilege. For its second ruling, the Supreme Court relied upon a trilogy of prior Second Department precedent: Landmark Ins. v. Beau Rivage Restaurant,16 Bertalo’s Restaurant v. Exchange Ins.,17 and Bombard v. Amica Mutual Ins.18

The starting point for this judicial odyssey is Landmark Insurance. There, the issue was whether the investigative report of an arson expert prior to the denial of a claim was prepared in anticipation of litigation (and thus covered by the attorney work-product doctrine). Because the court found that “it is readily apparent that the plaintiff employed…the arson expert to conduct an investigation for the purpose of aiding it to decide whether to accept or reject the defendant’s claim and not solely for the purpose of preparing for possible litigation,” the court ruled that the expert’s work was discoverable. Whether or not this reflected an appropriate analysis and application of the work product doctrine is one thing (see below); another thing is clear—this case (and opinion) have nothing to do with the attorney-client privilege.

Next came Bertalo’s Restaurant. As with Landmark Insurance, the case involved a claim that was denied on the ground that the restaurant owner was a “procuring cause” of the fire that destroyed the restaurant. At issue was whether reports by lawyers made before the decision to deny coverage were materials prepared in anticipation of litigation. Following its prior decision, the Second Department held that any materials pre-dating the coverage decision were fair game for discovery by the plaintiff.

Next came Bombard. Again faced with a denial of coverage, the Second Department focused on whether the attorney materials relating to the decision to deny coverage were prepared “exclusively” in anticipation of litigation. In rejecting the insurance company’s attempt to protect the attorney materials, the court opined in language the Supreme Court would pull up verbatim to justify its decision in Melworm:

[T]he payment or rejection of claims is a part of the regular business of an insurance company. Consequently, reports which aid it in the process of deciding which of the two indicated actions to pursue are made in the regular course of its business” [citing Landmark Insurance]. Reports prepared by insurance investigators, adjusters, or attorneys [emphasis by the court] before the decision is made to pay or reject a claim are thus not privileged and are discoverable [citing, inter alia, Landmark Insurance and Bertalo's Restaurant], even when those reports are “mixed/multi-purpose” reports, motivated in part by the potential for litigation with the insured [citing, inter alia, Landmark Insurance].19

What Is Going On?

So, what is wrong with this line of cases in the Second Department? Plenty.

First of all, while the Second Department courts have not made the same mistake that the Ohio Supreme Court made (equating allegations of “bad faith” in the denial of coverage with “fraud,” and thus invoking the crime/fraud exception), the courts have made a worse one: They have fundamentally confused the attorney-client privilege and the attorney work product doctrine.

The privilege, of course, has nothing whatsoever to do with litigation or the anticipation thereof.20 Instead, it relates to when a client seeks out the professional advice of an attorney and does so in a fashion to ensure that the client’s interactions with the attorney are done in a confidential fashion.21 The fact that the Landmark Insurance, Bertalo’s Restaurant, and Bombard courts took a wayward turn (perhaps because the insurance companies may have litigated the wrong privilege) does not mean that the Supreme Court in Melworm should have imposed those precedents upon an insurance carrier which litigated the correct privilege to block disclosure.

Second, the Melworm court, in its confusion, clearly misconstrued/misapplied (or perhaps did not read) the New York State Court of Appeals’ decisions in Rossi and Spectrum Systems.22 Not only are those decisions privilege decisions, but in both cases the court protected against disclosure of the attorney-client privileged materials: Rossi (internal corporate memorandum authored by an in-house lawyer was privileged where the attorney-client communication was “primarily or predominately of a legal character”; the privilege was not lost even where the memorandum referred to “certain nonlegal matters”); Spectrum Systems (report by outside law firm after its investigation into conduct at a bank was privileged because “it was made in order to render legal advice or services to the client”). Unfortunately, this is neither the first instance in which courts have confused the privilege and work product, nor is it the first time courts have misconstrued/misapplied Rossi and/or Spectrum Systems.23

Finally, even if they had been right in analyzing the line of insurance cases as work product cases, the Second Department courts were not doing that analysis correctly. As set forth above, those courts were utilizing an “exclusively,” “solely,” or “primarily” test in determining whether the attorneys’ work should be protected from disclosure. But as Judge Pierre Leval made clear 14 years ago in United States v. Adlman,24 such words have no place as a gloss or add-on to the work product doctrine. To determine whether materials are prepared “in anticipation of litigation,” the relevant criteria is to ascertain whether the materials are created “because of” litigation or its prospect. In 1998, Leval’s opinion was hailed as bringing sanity and certainty to an area of the law that had been a litigation “gotcha” trap.25 It looks like the Second Department never got the news.


One thing is clear: Being a lawyer in the insurance space is a dangerous place to ensure the confidentiality of your advice or work product, at least in Ohio and New York. Another thing is clear: Courts frequently misunderstand the attorney-client privilege and work product doctrine; thus, litigating those issues with any degree of predictive certainty can be problematic, at best. We have come a long way since Leval (and the U.S. Supreme Court before him) brought light and clarity to these subjects26; unfortunately, we now seem to be going in the wrong direction.

C. Evan Stewart is a partner at Zuckerman Spaeder. He is an adjunct professor at Fordham Law School and a visiting professor at Cornell University.


1. Lacon, vol. I, no. 217 (1820-22). It seems that Colton’s observation has become pretty widespread. See K. Raustiale and C. Sprigman, “In Praise of Copycats: Far from Killing Creativity, Imitation Spurs Invention in Industries From Fashion to Finance to Football,” Wall Street Journal C3 (Aug. 11, 2012). Colton is less well known for having dispensed advice that mothers everywhere have passed on to their children (which has most often gone unheeded): “When you have nothing to say, say nothing.” Id. at no. 183.

2. I have been writing about this state of affairs for more than 20 years. See, e.g., “Whither the Attorney-Client Privilege?” New York Law Journal (Oct. 11, 1990); “The Corporate Attorney-Client Privilege: Is Nothing Sacred?” The Corp. Crim. & Const. L.R. (April 5, 1991); “Corporate Counsel and Privileges: Going, Going…,” New York Law Journal (July 11, 1996); “The Attorney-Client Privilege: The Best of Times, the Worst of Times,” The Professional Lawyer (1999); “The Attorney-Client Privilege and Email: Strange Bedfellows?” The Computer & Internet Lawyer (March 2007); “Will Waiving the Privilege Save It?” New York Business Law Journal (Spring 2007); “Pandora’s Box and the Bank of America,” New York Law Journal (Nov. 4, 2009).

3. See Boone v. Vanliner Ins., 91 Ohio St. 3d 209, 744 N.E.2d 134 (2001).

4. See Melworm v. Encompass Indemnity Company and Encompass Insurance F/K/A CNA Personal Insurance, 2012 NY Slip Op 22193 (Nassau Cty. Sup. Ct. July 16, 2012).

5. See supra note 3.

6. See Moskowitz v. Mt. Sinai Med. Ctr., 69 Ohio St. 638, 661, 635 N.E.2d 331, 349 (1994).

7. By drawing the line at “prior to,” the four justices were satisfied that attorney work product would not be at risk of discovery. And as if that curious analysis were not bad enough (see Garg v. State Automotive Mutual Insurance, No. 2003 CA 12 (Ohio Ct. App. 2d Dist. Nov. 7, 2003)), they then added: “Of course, if the trial court finds that the release of this information will inhibit the insurer’s ability to defend the underlying claim, it may issue a stay of the bad faith claim and related production of discovery pending the outcome of the underlying claim.” Huh?!

8. E.g., Swidler & Berlin v. United States, 524 U.S. 399 (1998); Upjohn v. United States, 449 U.S. 383 (1981); Clark v. United States, 289 U.S. 1 (1933).

9. E.g., Massachusetts, Louisiana, Delaware, California, Indiana, Maryland, Texas, Florida, Pennsylvania, West Virginia, South Carolina.

10. State ex rel. Ohio General Assembly v. Brunner, 114 Ohio St. 3d 386 (2007).

11. 578 F.3d 432 (6th Cir. 2009).

12. See, e.g., Cobb v. Shipman, 2012 WL 1269128 (Ohio App. 11 Dist. April 16, 2012).

13. See supra note 4.

14. Rossi v. Blue Cross & Blue Shield, 73 N.Y.2d 588 (1989).

15. Spectrum Sys. Intl. v. Chemical Bank, 78 N.Y.2d 371 (1991).

16. 121 A.D.2d 98 (2d Dept. 1986).

17. 240 A.D.2d 452 (2d Dept. 1997).

18. 11 A.D.3d 647 (2d Dept. 2004).

19. 11 A.D.3d at 648.

20. See United States v. Upjohn, 449 U.S. 383 (1981); see also C.E. Stewart, “The Attorney-Client Privilege: The Best of Times, the Worst of Times,” The Professional Lawyer (1999).

21. Id.

22. See supra nn. 14 and 15. To those who would argue that New York takes a unique approach to the privilege, it should be noted that both Rossi and Spectrum Systems are clearly aligned with and derived from the teachings of Upjohn (see supra n. 20).

23. See C.E. Stewart, “Corporate Counsel and Privilege: Going, Going,…,” New York Law Journal (July 11, 1996). Even more recently the First Department bollixed up the work product doctrine in the expert context. See Beach v. Touradji Capital Management, 603611/08 (1st Dept. Aug. 21, 2012) (reported in New York Law Journal (Aug. 22, 2012)) (expert report prepared at the request of counsel held not to be work product; expert’s review of report before his deposition did not waive the work product privilege; both rulings are wide of the mark).

24. 134 F.3d 1194 (2d Cir. 1998). See also Hickman v. Taylor, 329 U.S. 497 (1947).

25. See, e.g., In re Sealed Case, 146 F.3d 881 (D.C. Cir. 1998); In re Grand Jury Subpoena, 357 F.3d 900 (9th Cir. 2009); United States v. Roxworthy, 457 F.3d 590 (6th Cir. 2006). See also C.E. Stewart, “‘Hickman v. Taylor’ Reinvigorated by the Second Circuit, with Important Benefits for Litigants,” ABA Pretrial Practice and Discovery (July 1998).

26. See supra nn. 20 and 24. But see C.E. Stewart, “Caveat Corporate Litigator: The First Circuit Sets Back the Attorney Work Product Doctrine,” New York Business Law Journal (Summer 2010).