Jack Goldsmith, Legal Ethics, and "the Cover-Your-Ass Syndrome"

Michael Stokes Paulsen

I love a good inside-baseball Washington tell-all book as much as the next guy. But when I'm not playing constitutional law professor, I'm often playing legal ethics professor. Jack Goldsmith's The Terror Presidency, which I read the past two nights and this morning (it's a quick, fairly light read, more memoir than deep legal analysis), raises some questions that have nagged me for a long time concerning the ethical confidentiality obligations of government lawyers and former government lawyers. I am not entirely certain of the answers; but the questions are deeply troubling, and Jack's book is the most troubling of the genre that I have seen.

First, some disclaimers and concessions. I know Jack Goldsmith, and called him before writing this, to express my concerns and ask a few pointed questions. I respect Jack as a lawyer and colleague, though this will not keep me from making points of harsh criticism. I agree with Jack on many issues. I find it indispensable for my own work to consider his academic writings on foreign affairs law issues, even when -- perhaps especially when -- I disagree with them.

In this post, I will, aside from this paragraph, refrain from discussing the "merits" of his critique of Bush Administration OLC legal work on war-on-terror issues, because they are collateral to my concerns here, which lie primarily in the realm of legal ethics. But for what it is worth (and so that readers may know my own perspective, to account for possible bias): I, like Jack, disagree with some of the analysis in the August 2002 so-called "torture" memo, but I agree with the core of its legal conclusions -- as, ironically, Jack Goldsmith does, and did at the time he headed OLC, according to his book. I disagree with Jack Goldsmith on the scope of the Commander-in-Chief Clause power, which appears to be Jack's primary problem with the August 2002 memo. I believe that this constitutional power vests the President with exclusive control of the United States' use of force against enemy powers, forces, and personnel, in any constitutionally authorized war; that, during time of authorized war, this power embraces the direction of policies and practices concerning the capture, detention, interrogation, and possible military punishment of enemy combatants (of both the lawful and unlawful varieties); and that Congress cannot constitutionally limit the exercise of this power by statute, its enumerated powers in this area reaching their outer bounds when they collide with the President's Commander-in-Chief Clause power to direct the use of force against the nation's enemies pursuant to a constitutionally authorized war. (See Michael Stokes Paulsen, The Emancipation Proclamation and the Commander in Chief Power, 40 Georgia L. Rev. 807 (2006)). I regard Jack's position as an entirely legitimate one, however; the issue is one on which reasonable constitutional lawyers may differ, even when they start from common interpretive premises (approximating original-meaning textualism). For precisely that reason, however, I disagree with Jack's assessment that the OLC memo was deeply flawed in this respect; and I find his taking of such a position substantively rather puzzling. Notwithstanding my strong view of the Commander-in-Chief Clause power, I agree in part with Jack's prudential position that it is often smart for an administration to obtain congressional authorization, where possible, in specific support of war-conduct policies. Here again, however, I think this a position on which reasonable people may differ. There is an eminently respectable argument, in principle, that this weakens presidential constitutional power by implying that such further legislative authorization was required, rather than sought out of an abundance of caution or a desire to provide double-support for presidential actions. There is an important argument that there is a danger in seeking such congressional validation or support, because one might not get it, or might not get it in its full appropriate degree. (Indeed, as Jack describes the internal administration disagreements over applicable legal principles, and political strategies about whether to seek further congressional support, it appears to me clear that Jack Goldsmith is Exhibit A for those who hold these concerns: Jack's posture toward congressional statutes like the War Powers Resolution and FISA is to regard them as setting strong presumptive limits on presidential power, rather than Youngstown-Category-I "safe harbors".) At all events, however, these are prudential, policy, political calculations; they are not, of their own force, legitimate legal objections to the constitutional validity of the Bush administration's interpretation of the Commander-in-Chief Clause.

But none of this is my main concern with Jack's book. My concern is with the possible ethical impropriety of a high-level lawyer for the executive branch airing these differences publicly, in order to justify himself in the eyes of some segment of the public (The Harvard Law School faculty?) or to settle old scores. When a non-lawyer does this (a CIA director, an Iraq Ambassador) the tell-all book or NY Times op-ed is often regarded, rightly, as somewhat dishonorable, even by those whose political side profits from it. When a lawyer does it, it raises issues not only of honor but of professional ethics. Since I am uncertain (for now, as I am writing this) what the right answer is, let me formulate the concern as a question: Does it violate the professional ethical standards governing an attorney, for an attorney for an entity publicly to disclose the content of recent, confidential high-level legal advice / discussions / deliberations with the entity client's representative -- including the relevant decision-makers for the entity -- concerning matters that are not generally known, and for which the client has not given informed consent to the disclosures?

As a general proposition, the answer is surely yes, unless one of the standard exceptions to the confidentiality obligation applies (to prevent or rectify a client crime or fruad, to prevent reasonably certain death or substantial bodily harm to a third party, to establish a claim or defense of the lawyer, to comply with other law or a court order). Is the question meaningfully different when the entity in question is not a corporation, but the United States government (executive branch)? I have written (at length, in tedious law review articles) that the U.S. government is a special kind of entity client, but that, in principle, the same ethical rules that apply to entity representation generally apply to representation of the United States government. (The differences concern sometimes-difficult questions of identifying precisely who the "client" is and who speaks on behalf of the client entity, in the government representation context; they do not involve the claim that an attorney owes less of a duty of loyalty and confidentiality to a government-entity client than to a private corporate-entity client.) See generally (for sleeping aid), Michael Stokes Paulsen, Who "Owns" the Government's Attorney-Client Privilege? 83 Minn. L. Rev. 473 (1998).

Did Jack Goldsmith reveal client-confidential information covered by the general ethical duty of confidentiality (which is generally cast in extremely broad trems)? Did his client "consent"? Do such revelations fit within any of the recognized exceptions that exist in most legal ethics rules?

Out of fairness to Jack, I put these questions to him in a phone call. He told me that he had given these issues very serious thought, and resolved them to his satisfaction. I asked Jack if he had had his book pre-cleared for government secrecy requirements -- whether it had been, in my shorthand, "Snepped" for classified information. He said it had. I then asked a more specific question: Had the U.S. government, as an entity client, consented generally to the disclosure of the information contained therein, and the communications revealed, pursuant to attorney-ethics confidentiality rules? (I doubt that pre-clearance for classified information constitutes "consent" to disclosure of information covered by confidentiality rules, but that is a question I raise and pursue below.) Jack told me he did not want to answer that question. He directed me to footnote 2 to the preface of his book, which follows this sentence in the preface text, obviously crafted in contemplation of precisely this concern: "In doing so, this book follows in a long tradition of Justice Department officials and other executive branch lawyers who have felt it important for the American people to understand how and why critical decisions were made during their service in government." The footnote then lists some fifty or sixty books, articles, speeches, or congressional testimony by former AGs, DAGs, OLC AAGs, SGs, and White House counsels. (He might have added an article of mine, in part recounting a story of my own government lawyering experience at OLC; I was careful not to name names, reveal information not already publicly known (except for my own views), or disclose reactions to or discussions of legal advice by high-level client representatives -- but re-reading it again, I wonder if my own minor league storytelling might have been too close to the line, or even just over it. Michael Stokes Paulsen, Hell, Handbaskets, and Government Lawyers: The Duty of Loyalty and its Limits, 61 Law & Contemp. Probs. 83 (1998)). Does the "everybody does it" defense work here? Has the tell-all book phenomenon succeeded in Defining Duty Down, for government lawyers, so that the operative ethical standard is less than the one described in the texts of the legal ethics rules?

I do not know in what jurisdiction Jack Goldsmith is licensed as an attorney, and do not wish to make this a personal Indictment of Jack in any event, so I will frame my questions and discussion in terms of general ethics law principles, as set forth in the ABA Model Rules of Professional Conduct, and the Restatement of the Law Governing Lawyers.

Rule 1.6 of the Model Rules, entitled "CONFIDENTIALITY OF INFORMATION" provides as follows:

(a) A lawyer shall not reveal information relating to the representation of a client unless the client gives the informed consent, the disclosure is impledly authorized in order to carry out the representation or the disclosure is permitted by paragraph (b).

The general, default duty stated by this rule is quite broad: A lawyer shall not reveal "information relating to the representation of a client . . .". As I teach my students, this is far broader than the evidence-law rule of attorney-client privilege. It says "information relating to the representation." It doesn't say "important" information; it doesn't even say"confidential" information (though I always ask my students, in good socratic fashion, whether the title of the rule implicitly narrows the scope of this prohibition, or -- quite the reverse -- defines as confidential everything concerning the representation). Leaning somewhat in a narrower direction, the Restatement defines "Confidential Client Information" as "information relating to representation of a client, other than information that is generally known."

"Generally known." If the ethical duty of confidentiality contained in Rule 1.6 is thought limited, somehow, implicitly, by the Restatement definition, there is still the question of whether some of Jack Goldsmith's material is "generally known." The fact that his discussion concerns internal executive branch deliberations over legal advice on topics widely discussed in the press, or even concerning specific memos that have been the subject of public debate, probably does not make the discussions that Jack Goldsmith reports "generally known." Comments to the Restatement further distinguish between information that might be known by some third parties and information generally known. Indeed, it is precisely the juicy insider tidbits in Jack's book -- "Then Gonzales said this . . . then Addington did this . . . Ashcroft said this . . . I was shocked when I saw this memo, the contents of which have never been disclosed, because they are classified, but the existence of which has been reported . . . I was receiving pressure from the White House to come out this way . . ." etc. etc. -- that make it so titillating. There's not very much "here's my take, as a Harvard Law Professor, on the legal issues that have become so famous / infamous, some of which were still going on while I was AAG of OLC." Such discussion would, I think, not violate the confidentiality rule. But there's a lot more of the "he said this . . ." discussion than comfortably can be squeezed under the roof of "generally known" information -- if such a roof exists at all, as a matter of legal ethics law.

Exactly how much of what Jack Goldsmith says in The Terror Presidency constitutes "information relating to the representation?" Is any of it subject to an (implied?) exception for information "generally known"? All of it? Is it correct to read such an exception into the duty of confidentiality? These are hard, and disturbing, questions.

Did the client "consent"? Almost certainly not. I know from my experience in the government -- and this sort of statement is clearly not a violation -- that it would be extraordinarily unlikely that the U.S. government would consent, generally, to disclosure of confidential information in order to permit someone to write a book. Why should it? If it granted such consent selectively -- permitting publication of nice, favorable books, but not kiss-and-tell or critical books -- it would open the responsible administration to charges of partisanship, distortion, censorship, etc. What's more, given the type of information revealed in The Terror Presidency -- conversations probably covered not just by the general ethical duty of confidentiality but by attorney-client privilege and executive branch deliberative privilege -- it is almost unthinkable that the government would consent, for doing so would waive the privilege, to some degree. (A fascinating question is presented as to whether allowing the book to proceed -- not seeking an injunction against publication -- is itself a troubling failure to protect the privilege.) In short, I know of no occasion where such blanket consent -- the type of "informed consent" contemplated by the rules -- has been given for a book or article of this type.

Did the client (the government) "consent" by virtue of granting a no-classified-information clearance? One could make this argument -- though Jack did not, when I invited him to do so. I would argue, as a general proposition, that security-clearance for publication simply does not equal client-consent-to-1.6(a)-noncoverage. I would be skeptical of any argument that this constitutes any sort of "implied" consent. For years, legal ethics law generally has been moving in the direction of requiring "informed consent," precisely to prevent lawyers from too-easily making the argument that "I thought the client knew and thought it was okay!" Finally, it seems hard to argue, with a straight face, that the disclosure of the information in question was "impliedly authorized in order to carry out the representation."

One cannot know such things for sure, and Jack wasn't saying, but I would wager that the government affirmatively denied any general consent to publication, or, more gently, strongly urged Jack not to publish the book -- and that Jack went ahead and published it anyway, in the teeth of the administration's moderate-to-vigorous objections. Here's the tell-tale clue: As reported in Jeff Rosen's New York Times Magazine article, Jack has directed that he receive no profits from the book. All net proceeds go to charity. This is interesting. On first reading, I was inclined to view this as some combination of a simple act of generosity, and a desire not to be accused of having profited financially from betraying his client -- showing his integrity and purity of purpose by throwing away the thirty pieces of silver in advance. But in looking at the Restatement, which prohibits disclosure of confidential client information (section 60), I noticed that one approved remedy contemplated for a violation of that command is that any pecuniary gain derived from such misconduct go to the client: "Except as stated in section 62 [client consent], a lawyer who uses confidential information of a client for the lawyer's pecuniary gain other than in the practice of law must account to the client for any profits made." National Security and First Amendment lawyers will recognize this as the same remedy enforced in Snepp for unauthorized disclosure of classified information by a former government employee. My guess then, is that Jack was told "no," but received (legal?) advice or himself concluded that the government's only effective remedy would be to seize any pecuniary gain Jack received. (This highlights another problem with books of this type. The government truly has no good remedy that does not make the aggrieved administration look even worse in the public eye. An injunction against publication of a book critical of the administration? Forget about it. So, a prospective writer unworried about financial profit might think, what are they going to do?)

Is there any softening of the general prohibition on disclosing confidential information when the situation is one involving a former client? Perhaps, but only very slightly (if at all). Rule 1.9, entitled "Duties to Former Clients" prohibits a lawyer for a former client from "reveal[ing] information relating to the representation except as these Rules would permit or require with respect to a client." Rule 1.9(c)(2). The immediate preceding sub-section prohibits lawyers from using "information relating to the representation to the disadvantage of the former client except as these Rules would permit or require with respect to a client, or when the information has become generally known" Rule 1.9(c)(1)(emphasis added) -- that old "generally known" potential escape hatch again. But the better reading, probably, is that the specific prohibition on revealing information is not subject to the "generally known" escape hatch for using information. I am uncertain of this, and so I leave this, too, posed as a question: Is this the right escape hatch?

Finally, as to the general duty of confidentiality, what of Jack's everyone-does-it defense? Does practice change the content of the governing legal rule? As a good formalist, I must insist that the answer is no. Perhaps a committed legal-process-school common-law-lawyer would argue that consistent violation of a rule impliedly repeals the rule, or softens it substantially. But it's not at all clear to me that everybody does do this. I certainly have not read every book in Jack's footnote 2, but a quick perusal leads me to be deeply skeptical of Jack's defense. Francis Biddle's memoir -- which does appear to have revealed client confidences -- was published nearly twenty years after the events in question, which at least puts the disclosures in a different light. So too, surely, Robert Jackson's insider biography of FDR, published posthumously in 2004, nearly sixty years after the fact. (And if I'm correct about this, didn't folks find it in Jackson's papers? Had he perhaps decided against publishing it after all?)

John Dean's book -- which I read as a high school kid, when it came out, but barely remember -- was tell-all, but is either a bad precedent or fits under a rectify-fraud exception (discussed below), depending on one's point of view. Jack Goldsmith also cites John Ashcroft's 2006 "Never Again: Securing America and Restoring Justice," which I've perused in a bookstore but not read, for the reason that many of the books listed in footnote 2 aren't worth buying: They don't really tell you anything! Because they can't.

I have read John Yoo's "War By Other Means," but it is, ironically, less memoir and more analysis of all the hot-button issues he addressed at OLC. Where Yoo discusses these issues, I don't recall him ever disclosing the substance of insider-conversations, confidential legal advice, or a client's (or client's agent's) reaction to it. I wasn't reading Yoo's book for legal-ethics problems (and was not reading the NYTimes Mag article about Goldsmith for such purposes, either, at least not at first), but I recall being struck, and mildly amused, at the verbal circumlocutions that Yoo had to use in order scrupulously to avoid any statement that might be thought to rest on a non-public source. If Yoo made some mistakes in that regard, I don't recall anything popping out at me. Jack Goldsmith's book uses that approach . . . sometimes. But where it doesn't, Goldsmith's narrative leaps off the page with legal ethics problems.

I haven't looked at every source Jack cites, and it could be that there is a broader pattern of ethics-law violations of this sort by former high-level government attorneys. But if so, that is exactly what should be said about it: there is a pattern of ethics-law violations of this sort by former high-level government attorneys. It should not be said that this sort of conduct is, therefore, an exception to or an implied repeal of Rule 1.6(a). (Fluffy academic confabs gathering former White House counsels or SG's may contain minor violations, at times. But without reading all of them, I venture to guess they tend to fall in the category of no error or harmless error.)

Rule 1.6(b) is a laundry list of exceptions to the general rule of confidentialtiy. I will quote the rule, then discuss each exception briefly. My short preview is that none of them seems to apply to justify the disclosures made in The Terror Presidency -- with one potential exception.

Rule 1.6(b) provides:

A lawyer may reveal information relating to the representation of a client to the extent the lawyer reasonably believes necessary:
(1) to prevent reasonably certain death or substantial bodily harm;
(2) to prevent the client from committing a crime or fraud that is reasonably certain to result in substantial injury to the financial interests or property of another and in furtherance of which the client has used or is using the lawyer's services;
(3) to prevent, mitigate or rectify substantial injury to the financial interests or property of another that is reasonably certain to result or has resulted from the client's commission of a crime or fraud in furtherance of which the client has used the lawyer's services;
(4) to secure legal advice about the lawyer's compliance with these Rules;
(5) to establish a claim or defense on behalf of the lawyer in a controversy between the lawyer and the client, to establish a defense to a criminal charge or civil claim against the lawyer based upon conduct in which the client was involved, or to respond to allegations in any proceeding concerning the lawyer's representation of the client; or
(6) to comply with other law or a court order.

Do any of these exceptions to the baseline rule of confidentiality apply here? Some vehement opponents of Bush administration war policies no doubt argue that administration lawyers, like Jack Goldsmith, should have "blown the whistle" for reason (1) -- "to prevent reasonably certain death or substantial bodily harm." If one felt that way about the policies Jack discusses (the exceptions are permissive ; a lawyer "may" reveal information in such circumstances), then that exception might well apply. But, interestingly, as his book makes clear, Jack Goldsmith did not feel that way. One of the odder things about The Terror Presidency's critique of certain OLC memos is that Jack disagreed with certain arguments made in earlier memos, not with the ultimate conclusions. To be sure, part of Jack's narrative is that he felt that the arguments themselves, contained in extant OLC opinions, posed a certain danger. That is why, he says, that he worked to withdraw certain opinions. But that is what he did. To the extent that Jack felt the opinions were unsound and (to ratchet things up a notch, merely for the sake of discussion) would result in "reasonably certain death or substantial bodily harm," his action was what was required; and by his account his actions were successful. It is hard to see what further value is served by writing a book about the events in question. With all due respect, one would have to make the argument that David Addington's comments and views themselves needed to be disclosed in order to prevent reasonably certain death or substantial bodily harm.

(2) Was publishing this book necessary to "prevent" a "crime or fraud" likely to result in substantial injury to the financial interests of another, and in furtherance of which the client has used the lawyer's services? If this is the justification, Jack does not say so. It seems even more strained to resort to this exception than to resort to exception (1). So too with exception (3), disclosure to prevent, mitigate, or rectify financial injury to another that made use of a lawyer's services. Both exceptions are geared toward non-disclosure of financial crimes, known to the lawyer, in which the lawyer was played for the fool and unwittingly assisted the client's financial crime or fraud. Neither exception seems to apply here.

(4) The Terror Presidency is not disclosing client confidences in order for Jack to seek legal advice. And, if I may go out of order, exception (6) does not apply either. The publishing of this book, and the revealing of client confidences, was not done in order to comply with other law or a court order.

(5) This one is interesting. Is there a sense in which Jack's revealing what he has revealed might be thought reasonably necessary "to establish a claim or defense on behalf of the lawyer in a controversy between the lawyer and the client, to establish a defense to a criminal charge or civil claim against the lawyer based upon conduct in which the client was involved, or to respond to allegations in any proceeding concerning the lawyer's representation of the client"? To my knowledge, there is no controversy between Jack Goldsmith and the United States government, akin to the run-of-the-mill disputes that sometimes arise between lawyers and former clients over fees, services rendered, and the like. Even if there were, it is unclear how the book's revelations tend to establish a claim or defense relevant to such hypothesized dispute.

But might there be an argument that Jack's revelations -- which are almost uniformly in the direction of self-serving, blame-shifting, "not-me-but-some-other-guy-and-I-tried-to-stop-them" insider accounts of events -- are designed as self-defense? They are literally that, of course. And it is not crazy -- as Jack's book discusses at length -- to imagine charges being filed in some forum, national or international, by someone, charging Jack Goldsmith with being a war criminal based on his legal work. Of course, such charges might well be thought ridiculous. But I have had faculty colleagues who take such nonsense ridiculously seriously, and so too do some European nations. The international law legal academy is full of people who take such things seriously. Some of them read -- perhaps some of them write for! -- Balkinization. As Jack writes, some of them are on the Harvard Law School faculty, and made his arrival there a less than perfectly welcome one.

But put Harvard aside (please). The simple point here is that one might plausibly argue that a government lawyer handling sensitive, politically-charged national security issues involving war, interrogation, international law, and the like, is privileged to disclose confidential client information to the extent he or she thinks it reasonably necessary to establish, even prospectively, a "defense" to a "criminal charge or civil claim," or "to respond to allegations in any proceeding concerning the lawyer's representation of the client. (When I teach this provision in Legal Ethics, I argue (contrary to the rule's comments) that there need not be an extant charge in order for the right to disclose otherwise confidential information "to establish" a defense to exist.) Self-justifying disclosure of client confidences, to protect one's reputation against wild accusations of criminal lawyering, might be ethically permissible.

Is that what Jack Goldsmith was thinking? That he needed to tell his tale, even if it meant disclosing confidential client communications and information concerning his representation of the government that he otherwise would have been thought obliged to keep quiet, in order to protect his own hindquarters?

A central thesis of the book is that these issues have become overlawyered, and that great harm is often produced by creating an atmosphere of excessive caution in government officials. He writes eloquently of "intelligence officers spooked by cautious lawyers" who "failed to take actions that might have prevented the 9/11 attacks." (p. 95). He writes about how lawyers are "by nature and training a cautious bunch" and quotes Senator Bob Graham (with approval) noting how CIA lawyers had "displayed a risk aversion in the advice they give their clients." (92). (Ironically, this seems to have been, according to Jack's own narrative, the defining feature of his own representation of the government during his brief tenure as head of OLC.) Jack's argument continues: The consequent urgent need for legal authorization enhances the urgent need for, and importance of, OLC legal opinions justifying urgently needed actions. "These cycles of timidity and aggression," Jack Goldsmith writes, "are the bane of the intelligence community, and are a terrible problem for our national security." (p. 164). And one of the "pathologies" from which this problem flows is "the blame game, and the cover-your-ass syndrome."

That last phrase might have made an appropriate alternative title for Jack Goldsmith's book. The Terror Presidency contains some interesting legal and policy analysis, and some interesting historical discussion. But it's main appeal -- literally it's selling point -- is the insider high-level attorney's disclosure of the content of legal advice, deliberations, and client conduct in response thereto, involving some of the most important war-time legal judgments made during the Bush Administration. If such disclosures are not ethically improper -- as they certainly appear to be -- it seems that the best argument for such a conclusion is that Jack Goldsmith was, after all, merely doing what he felt he needed to do to cover his own butt.

Michael Stokes Paulsen
University of St. Thomas School of Law
Posted 12:08 PM by Michael Stokes Paulsen [link]