David Luban

May 13, 2009

Testimony of David Luban
Senate Judiciary Committee,
Subcommittee on Administrative Oversight and the Courts
Hearing: "What Went Wrong: Torture and the Office of Legal Counsel in the Bush Administration"
May 13, 2009

Chairman Whitehouse, ranking member Graham, Chairman Leahy, and members of the subcommittee.
Thank you for inviting me to testify today. You've asked me to talk about the legal ethics of the torture and interrogation memos written by lawyers in the Office of Legal Counsel. Based on the publicly-available sources I've studied, I believe that the memos are an ethical train wreck.

When a lawyer advises a client about what the law requires, there is one basic ethical obligation: to tell it straight, without slanting or skewing. That can be a hard thing to do, if the legal answer isn't the one the client wants. Very few lawyers ever enjoy saying "no" to a client who was hoping for "yes". But the profession's ethical standard is clear: a legal adviser must use independent judgment and give candid, unvarnished advice. In the words of the American Bar Association, "a lawyer should not be deterred from giving candid advice by the prospect that the advice will be unpalatable to the client."

That is the governing standard for all lawyers, in public practice or private. But it is doubly important for lawyers in the Office of Legal Counsel. The mission of OLC is to give the President advice to guide him in fulfilling an awesome constitutional obligation: to take care that the laws are faithfully executed. "Faithful" execution means interpreting the law without stretching it and without looking for loopholes. OLC's job is not to rubber-stamp administration policies, and it is not to provide legal cover for illegal actions.

No lawyer's advice should do that. The rules of professional ethics forbid lawyers from counseling or assisting clients in illegal conduct; they require competence; and they demand that lawyers explain enough that the client can make an informed decision, which surely means explaining the law as it is. Lawyers must not misrepresent the law, because lawyers are prohibited from all "conduct involving dishonesty, fraud, deceit or misrepresentation." These are standards that the entire legal profession recognizes.

There is a common misperception that lawyers are always supposed to spin the law in favor of their clients. That is simply not true. It is true that in a courtroom, lawyers are supposed to argue for the interpretation of law that most favors their client. The lawyer on the other side argues the opposite, and the judge who hears the strongest case from both sides can reach a better decision.

But matters are completely different when a lawyer is giving a client advice about what the law means. Now there is nobody arguing the other side, and no judge to sort it out. Typically, the lawyer-client communication is confidential, and thus the lawyer is the client's only channel of advice about what the law requires. Not only is it important for the client to receive unvarnished advice, it is important for society at large that clients know their legal obligations. The ABA explains the value of lawyer-client confidentiality by pointing out its contribution to law compliance: "Almost without exception, clients come to lawyers in order to determine their rights and what is, in the complex of laws and regulations, deemed to be legal and correct. Based upon experience, lawyers know that almost all clients follow the advice given, and the law is upheld." The ABA's Model Code of Professional Responsibility explains the "essential difference" between advocates and advisors:

Where the bounds of the law are uncertain, the action of a lawyer may depend upon whether he is serving as advocate or advisor. A lawyer may serve simultaneously as both advocate and adviser, but the two roles are essentially different. ... While serving as advocate, a lawyer should resolve in favor of his client doubts as to the bounds of the law. In serving a client as adviser, a lawyer in appropriate circumstances should give his professional opinion as to what the ultimate decisions of the courts would likely be as to the applicable law.

Of course it is likely that the torture memos were exactly what the client wanted; according to a Senate Intelligence Committee report, "On July 17, 2002, according to CIA records, the Director of Central Intelligence (DCI) met with the National Security Adviser, who advised that the CIA could proceed with its proposed interrogation of Abu Zubaydah. This advice, which authorized CIA to proceed as a policy matter, was subject to a determination of legality by OLC." In other words, the "program" had already been approved, pending legal approval by OLC.

However, the requirement of independent judgment in Rule 2.1 does not permit lawyers to shape their opinions to the client's wishes. This is common sense. Otherwise, clients might go to their lawyers to say, "Give me an opinion that says I can do what I want"–and then duck responsibility by saying, "My lawyer told me it was legal." Then we would have a perfect Teflon circle: the lawyer says "I was just doing what my client instructed" and the client says "I was just doing what my lawyer approved." The damage to law and compliance with law would be enormous.

Does that mean a client cannot come to a lawyer with the request, "Give me the best argument you can find that I can do X"? As a general proposition, nothing forbids a lawyer from doing so, but it would be deceptive to package one-sided advice as an authentic legal opinion. Emphatically this is not OLC's mission, which is to tender objective advice about matters of law, binding on the executive branch. Nor do Professor Yoo, Judge Bybee, and Mr. Bradbury claim they are simply giving, in a one-sided way, the best arguments they can find for the permissibility of the tactics. The August 1, 2002 "techniques" memo states, "We wish to emphasize that this is our best reading of the law," while Mr. Bradbury describes his May 10, 2005 "techniques" memo in similar terms: "the legal standards we apply in this memorandum?constitute our authoritative view of the legal standards applicable under [the torture statutes]."
Unfortunately, the torture memos fall far short of professional standards of candid advice and independent judgment. They involve a selective and in places deeply eccentric reading of the law. The memos cherry-pick sources of law that back their conclusions, and leave out sources of law that do not. They read as if they were reverse engineered to reach a pre-determined outcome: approval of waterboarding and the other CIA techniques.

Because of time constraints, my oral statement on May 13 discussed only one example of what I am talking about; in this written testimony I include others, beginning with the case my oral statement focused on.

Twenty-six years ago, President Reagan's Justice Department prosecuted law enforcement officers for waterboarding prisoners to make them confess. The case is called United States v. Lee. Four men were convicted and drew hefty sentences that the Court of Appeals upheld.

The Court of Appeals repeatedly referred to the technique as "torture." This is perhaps the single most relevant case in American law to the legality of waterboarding. Any lawyer can find the Lee case in a few seconds on a computer just by typing the words "water torture" into a database. But the authors of the torture memos never mentioned it. They had no trouble finding cases where courts didn't call harsh interrogation techniques "torture." It's hard to avoid the conclusion that Mr. Yoo, Judge Bybee, and Mr. Bradbury chose not to mention the Lee case because it casts doubt on their conclusion that waterboarding is legal.

In past discussion before this Committee, Attorney General Mukasey responded that Lee is not germane, because it is a civil rights denial case, not a torture case. That response misses the point, however, which was not what legal issue the court was addressing in Lee, but the fact that the judges had no hesitation about labeling waterboarding "torture," a label they used at least nine times. They obviously could not reference the Convention Against Torture (CAT) or the torture statutes, 18 U.S.C. §§ 2340-2340A, which did not yet exist. But there is no reason to suppose that they would have reached a different characterization of waterboarding than they did in Lee. That might be the case if CAT and the torture statutes had transformed the meaning of the ordinary-language word 'torture,' making it more technical, and raising the standard of harshness so that waterboarding might not be torture under the new, technical standard.

That simply did not happen. The statutes' definition of torture as severe mental or physical pain or suffering is neither unusual nor technical. Indeed, a standard pre-CAT dictionary definition of torture describes it as "severe or excruciating pain or suffering (of body or mind)?" —a definition so similar to the language of CAT that it seems entirely possible that CAT's drafters modeled the treaty language on the Oxford English Dictionary definition. Other Lee-era dictionaries use formulations that do not in any way suggest that at the time of Lee 'torture' meant something milder than the statutory standard—Webster's Third (1971) says "intense pain"; Webster's Second (1953) says "severe pain" and "extreme pain. "
Other significant omissions include the failure of the August 1, 2002 "torture" memo to discuss or even mention the Steel Seizure Case in its analysis of the President's commander-in-chief power, or the highly significant early decision Little v. Barreme, 6 U.S. (2 Cranch) 170 (1804), which found that President Adams, as commander-in-chief during the "quasi-War" with France, could not authorize the seizure of a ship contrary to an act of Congress. In its discussion of the necessity defense, the Bybee Memo fails to mention the recently-decided United States v. Oakland Cannabis Buyers' Coop, 532 U.S. 483, 490 (2001), which calls into question whether federal criminal law even contains a necessity defense if no statute specifies that there is one. Likewise, the opinion fails to mention that there is no reported case in which a federal court has accepted a necessity defense for a crime of violence—surely a crucial piece of information for a client who might be relying on the OLC's opinion in the momentous decision whether or not to waterboard detainees. In one place, the opinion may fairly be said to falsify what a source says. Discussing whether interrogators accused of torture could plead self-defense, the memo says: "Leading scholarly commentators believe that interrogation of such individuals using methods that might violate [the anti-torture statute] would be justified under the doctrine of self-defense." The opinion refers to a law review article. What the article's author actually says on the page cited is nearly the opposite: "The literal law of self-defense is not available to justify their torture. But the principle uncovered as the moral basis of the defense may be applicable" (emphasis added). Omitting to discuss leading contrary cases, and spinning what cited sources say, is not honest opinion writing, and violates the ethical requirements of candor and independent judgment, and communication to a client of everything reasonably necessary for the client to make an informed decision.

I would like to briefly discuss other ways that the torture memos twisted and distorted the law, even though doing so requires getting even further into technicalities that, quite frankly, only a lawyer could love. The first Bybee memo advances a startlingly broad theory of executive power, according to which the President as commander-in-chief can override criminal laws such as the torture statute. This was a theory that Jack Goldsmith, who headed the OLC after Judge Bybee's departure, described as an "extreme conclusion," reached through "cursory and one-sided legal arguments"—a conclusion that "has no foundation in prior OLC opinions, or in judicial decisions, or in any other source of law." It comes very close to President Nixon's notorious statement that "when the President does it, that means it is not illegal"—except that Mr. Nixon was speaking off the cuff in a high pressure interview, not a written opinion by the Office of Legal Counsel. The Youngstown case I mentioned previously found that President Truman could not seize steel mills during the Korean War because doing so impinged on Congress's powers. It is a case limiting the commander-in-chief power, and it is known to every law student who has taken constitutional law.
Professor Yoo has explained that he and Judge Bybee did not discuss the Steel Seizure case because of a long-standing OLC tradition of upholding the President's commander-in-chief powers, central among which is the power to interrogate captives. Suffice it to say, however, that nothing in either U.S. law or U.S. military tradition suggests that authority to torture captives belongs among the commander-in-chief's historical powers, any more than the authority to execute captives as a way of inducing other captives to reveal information is part of the traditional commander-in-chief power. It is perhaps for this reason that the TJAGs of the Army, Navy, Marines, and Air Force all protested the torture memos when they learned of them months after they were issued. MG Jack Rives, TJAG of the Air Force, objected that "the use of the more extreme interrogation techniques simply is not how the U.S. armed forces have operated in recent history. We have taken the legal and moral 'high ground'?." And BG Kevin Sankuhler, the Marine TJAG, noted sharply that "OLC does not represent the services; thus, understandably, concern for servicemembers is not reflected in their opinion."
I believe Professor Goldsmith's view that no source of law supports the Bybee Memo's proposition that the commander-in-chief power can override the criminal law on torture is correct; surely Professor Goldsmith, a Bush appointee, a conservative, and an intellectual ally of Professor Yoo, cannot have lightly decided to withdraw the memos. The same conclusion is reached in the definitive study of the commander-in-chief's power, the nearly 300-page articles by Professors David Barron and Martin Lederman, who conclude that "[t]here is a radical disjuncture between the approach to constitutional war powers the current President [George W. Bush] has asserted and the one that prevailed at the moment of ratification and for much of our history that followed."
This is not simply a matter of scholarly disagreement; and, obviously, I am not saying that taking one side of a contested and complex constitutional issue is unethical. It is not. But omitting the leading case on the commander-in-chief power "at lowest ebb" (that is, in the face of a contrary statute) is a different matter. A lawyer writing an appellate brief on whether the torture statute encroaches on the President's constitutional authority who failed to cite or discuss Youngstown Sheet & Tube would be committing legal malpractice, and might face professional discipline for failing to cite directly contrary authority if, improbably, the adversary also failed to cite Youngstown. Briefs have more, not less leeway to present a one-sided view of the law than advisory opinions for clients, and an omission that would be malpractice in a brief is a fortiori unacceptable in an opinion.
The first Bybee memo also wrenches language from a Medicare statute to explain the legal definition of torture. The Medicare statute lists "severe pain" as a symptom that might indicate a medical emergency. Mr. Yoo flips the statute and announces that only pain equivalent in intensity to "organ failure, impairment of bodily function, or even death" can be "severe." This definition was so bizarre that the OLC itself disowned it a few months after it became public. It is unusual for one OLC opinion to disown an earlier one, and it shows just how far out of the mainstream Professor Yoo and Judge Bybee had wandered. The memo's authors were obviously looking for a standard of torture so high that none of the enhanced interrogation techniques would count. But legal ethics does not permit lawyers to make frivolous arguments merely because it gets them the results they wanted. I should note that on January 15 of this year, Mr. Bradbury found it necessary to withdraw six additional OLC opinions by Professor Yoo or Judge Bybee.
Of course, it is well-known that the 2004 Levin memorandum that replaced the Bybee Memo stated, "While we have identified various disagreements with the August 2002 Memorandum, we have reviewed this Office's prior opinions addressing issues involving treatment of detainees and do not believe that any of their conclusions would be different under the standards set forth in this memorandum." However, Mr. Levin stated in testimony to the House Judiciary Committee that he "did not mean, as some have interpreted—and . . . this is my fault, no doubt, in drafting—that we had concluded that we would have reached the same conclusions as those earlier opinions did. We were in fact analyzing that at the time and we never completed that analysis." Rather, he meant that his predecessors, Professor Yoo and Judge Bybee, would have reached the same conclusions based on his standards.

I have said little about the three May 2005 opinions, beyond the point I have already noted that they approve waterboarding without citing or discussing Lee. (Nor do they acknowledge earlier cases where the U.S. has condemned water torture—the Glenn court-martial from the U.S. Philippines campaign in the early twentieth century, and the Sawada case, in which a Japanese general was condemned for forms of cruelty that included water torture. ) The 2005 memos are not as conspicuously one-sided as the August 1, 2002 torture memo which—again quoting Professor Goldsmith—"lacked the tenor of detachment and caution that usually characterizes OLC work, and that is so central to the legitimacy of OLC." Mr. Bradbury's memos are more cautious, and contain repeated reminders that reasonable people could reach the opposite conclusion. But they too contain troubling features.

To take one example, the May 30, 2005 memo states twice that courts might reach the opposite conclusion in their interpretation of whether the CIA techniques "shock the conscience." This is an important warning, and I believe that it is perfectly ethical for a lawyer to offer a non-standard interpretation of the law in an advisory opinion, provided that the lawyer flags—as Mr. Bradbury does—that it may indeed be non-standard. However, in both places he immediately adds that the interpretation is "unlikely to be subject to judicial inquiry." This is uncomfortably close to a lawyer telling the client, "it's likely to be found illegal, but don't worry—you probably won't be caught."

Other features of the memos are likewise troubling. To reach the conclusion that waterboarding does not cause "severe physical suffering," the memos rely on a specious finding from the 2004 memo, namely that to qualify as severe, suffering must be prolonged. There is no such requirement in the torture statute—and indeed, there is strong reason to believe that no such requirement was intended. Congress did stipulate that severe mental suffering must be "prolonged" (18 U.S.C. § 2340(2)). Ordinary canons of statutory construction would lead virtually all competent lawyers to conclude that if Congress omitted the word "prolonged" in connection with physical suffering, but included it in the definition of mental suffering in the same statute, it does not exist in connection with physical suffering. In Mr. Bradbury's memo, the requirement of duration is crucial in finding that waterboarding does not induce severe physical suffering, because it does not last long enough. But, to repeat, the law itself contains no duration requirement for severe physical suffering—and it is wildly implausible that the overwhelming sensation of drowning, which is surely a form of physical suffering, is not severe.
Equally troubling is the manner in which the May 30 memo responds to bodies of law strongly indicating that the United States Government condemns the very techniques the memo is approving, which would indicate that these techniques are not "traditional executive behavior" or "customary practice"—as, hopefully, they are not! For example, the memo notes that our own State Department 's annual Country Reports routinely condemn several of the practices the CIA used dousing people with cold water, food and sleep deprivation, waterboarding, stripping and blindfolding them. The memo responds that "The condemned conduct is often undertaken for reasons totally unlike the CIA's." But of course these countries often undertake the conduct for reasons very similar to the CIA's: learning information about terrorists. We still condemn it. In any event, the response does not even speak to the question of whether these practices represent U.S. custom or traditional executive behavior—as they surely do not. The memo goes on with an argument that is absurd on its face: the fact that the U.S. offers SERE training shows that these SERE-derived interrogation tactics are indeed traditional executive behavior. It is obvious that a method of training SEALs to resist torture and cruelty is hardly traditional executive behavior in dealing with captives. In any case, the May 10 "techniques" memo notes explicitly that SERE is quite different from the CIA's program, and that the detainees were waterboarded far more often than in SERE (dozens of times instead of two).
These arguments are so implausible that it seems clear that Mr. Bradbury was straining to reach a result. There are other difficulties with these three memos, which I do not wish to belabor here. While I find these memoranda deeply troubling—and their conclusions are even more troubling than the Bybee memos, because the 2005 memos discuss the techniques both singly and in combination and conclude that they do not violate a lower standard than the definition of torture—they do not exhibit the one-sided and manipulative use of law to the same extent as the August 1, 2002 memos. And that is where the main problems of legal ethics in these memos lie.

Recent news reports have said that the Justice Department's internal ethics watchdog, the Office of Professional Responsibility, has completed a five-year investigation of the torture memos. OPR has the power to refer lawyers to their state bar disciplinary authorities, and news reports say they will do so.

I have no personal knowledge about what OPR has found. Presumably, investigators were looking either for evidence of incompetence, evidence that the lawyers knew their memos don't accurately reflect the law, or evidence that process was short-circuited.

This morning I have called the torture memos an ethical train wreck. I believe it's impossible that lawyers of such great talent and intelligence could have written these memos in the good faith belief that they accurately state the law. But what I or anyone else believes is irrelevant. Ethics violations must be proved, by clear and convincing evidence, not just asserted. That sets a high bar, and it should be a high bar. Obviously, proving that lawyers were not candid in their advice, when they continue to assert publicly that they believe it is legally correct, is not easy.

In closing, I would like to emphasize to this Committee that when OLC lawyers write opinions, especially secret opinions, the stakes are high. Their advice governs the executive branch, and officials must be told frankly when they are on legal thin ice. They and the American people deserve the highest level of professionalism and independent judgment, and I am sorry to say that they did not get it here.