Who is the attorney general's client?Two consecutive presidential administrations have been beset with controversies surrounding decision making in the Department of Justice, frequently arising from issues relating to the war on terrorism, but generally giving rise to accusations that the work of the Department is being unduly politicized. Much recent academic commentary has been devoted to analyzing and, typically, defending various more or less robust versions of "independence" in the Department generally and in the Attorney General in particular. This Article builds from the Supreme Court's recent decision in Free Enterprise Fund v. Public Co. Accounting Oversight Board, in which the Court set forth key principles relating to the role of the President in seeing to it that the laws are faithfully executed. This Article draws upon these principles to construct a model for understanding the Attorney General's role. Focusing on the question, "Who is the Attorney General's client?", the Article presumes that in the most important sense the American people are the Attorney General's client. The Article argues, however, that that client relationship is necessarily a mediated one, with the most important mediating force being the elected head of the executive branch, the President. The argument invokes historical considerations, epistemic concerns, and constitutional structure. Against a trend in recent commentary defending a robustly independent model of executive branch lawyering rooted in the putative ability and obligation of executive branch lawyers to alight upon a "best view" of the law thought to have binding force even over plausible alternatives, the Article defends as legitimate and necessary a greater degree of presidential direction in the setting of legal policy. This position is defended in terms of democratic accountability, epistemic humility in the face of the indeterminacy of law, and historical practice.
The current and previous presidential administrations have served us no end of political-legal controversies, from torture memos penned early in the war against Al Qaeda and its allies, to whether to try Guantanamo detainees before civilian or military tribunals, to how the President ought to be advised about the constitutionality of laws by the Attorney General and the Department of Justice (DOJ). Any number of controversies plucked from the headlines of the last several years of reporting and commentary on the moral, political, and legal challenges inherent in the war against international terrorism refer to episodes that have vexed the United States government generally, and the executive branch in particular, as it struggles to find a way forward in what is frequently uncharted legal terrain. This Article will focus upon the Office of the Attorney General and in particular on the proper understanding of the role of the Attorney General of the United States in serving the President and the American people. Although there is a rich and complicated moral and political context in which these legal questions are situated, for the most part such considerations will operate in the background as we explore the contours of the duties of the Attorney General as a lawyer first and foremost. (1)
I will offer a model of understanding the responsibility and accountability of the Attorney General that I hope can be extended to government lawyering more generally, including to lawyers serving in both the legislative and judicial branches of government. Specifically, I want to focus attention on the Attorney General as a lawyer working for a client--the American people. (2) The model I propose is that of a mediated client relationship. For if it is true--as in some sense it must be--that a lawyer who works for the United States works not for the President alone, nor for a Senator alone, nor for a Judge alone, but for the American people, it must be admitted nevertheless that such a description of the client relationship may raise more questions than it answers. The model of a client relationship mediated by various representatives of the client--here, the President, the Congress, laws, rules, regulations, and customs--is presented as capturing the forces and boundaries that properly ought to shape and constrain the Attorney General in serving all of the American people. A common and appealing alternative approach, emphasizing simply the "independence" of the Attorney General, threatens only to undermine the proper client service of the Attorney General by substituting the judgment, will, and whim of the President with that of the Attorney General. The question of the proper role of the Attorney General cannot be settled by overemphasizing its alleged uniqueness among cabinet posts. Rather, it is best answered by fidelity to the larger constitutional structure, which after all directs the executive branch--headed, of course, by the President--faithfully to execute the laws.
The Supreme Court's recent decision in Free Enterprise Fund v. Public Co. Accounting Oversight Board, (3) handed down on the last day of the October 2009 Term, while not disturbing the broad contours of the modern administrative state (including, in some instances, restrictions on the President's removal power), is highly instructive here. There the Court considered and rejected as unconstitutional what it described as the Sarbanes-Oxley Act's "dual for-cause limitations on removal" of members of the Public Company Accounting Oversight Board, holding that combining layers of removal restrictions "contravene[s] the Constitution's separation of powers." (4) The Court's reasoning in Free Enterprise Fund is consonant in language and logic with the argument that I will make here concerning the Attorney General. Most importantly, the Court made consistent recourse to the fact that democratic accountability in the executive branch rests with the President, not with any of the officers appointed by the President: "The diffusion of power carries with it a diffusion of accountability. The people do not vote for the 'Officers of the United States.' They instead look to the President to guide the 'assistants or deputies subject to his superintendence.'" (5)While it is true that there were no votes among the justices in the majority or dissent in Free Enterprise Fund who indicated a willingness to upset the basic legal assumptions underlying the administrative state, the majority opinion did raise concerns about the dangers of diluting the power of the President as the chief executive: "The growth of the Executive Branch, which now wields vast power and touches almost every aspect of daily life, heightens the concern that it may slip from the Executive's control, and thus from that of the people." (6)
Worries over the control of government power, and especially over the pursuit of justice and the enforcement of law, are nothing new. (7) One suspects that Madison had such in mind in penning his classic formulation in Federalist 51:
If angels were to govern men, neither external nor internal controuls on government would be necessary. In framing a government which is to be administered by men over men, the great difficulty lies in this: You must first enable the government to controul the governed; and in the next place, oblige it to controul itself. (8)
Yet recent years have seen a marked increase in the frequency and intensity of expressions of concern over the independence of the Attorney General from the President in particular and the political activity of the White House in general. The centrality of the DOJ, and especially of the Office of Legal Counsel (OLC), in helping to define the contours of key elements in the war on terrorist groups in the aftermath of 9/11, perhaps more than any other single factor, has ensured that the role of lawyers in our governance will remain a topic of enormous interest to politicians, the academy, and the public for years to come. (9) And while disagreements about just how independent from the President the Attorney General ought to be have not been and will not be limited to issues concerning the war on terror, the urgency and prominence of those issues in our national debate have prompted public officials, journalists, and academic commentators to raise charges of "politicization" of the DOJ's work in both administrations. (10)
This Article examines the separation of powers issues at stake in the operation of the DOJ, focusing in particular on the office of the Attorney General of the United States. To this end, I should clarify three general points about my approach. First, I intend to treat the Office of the Attorney General as a kind of "unitary executive in small" with respect to the DOJ. That is, however much in a given case (as with the executive branch more generally) the Attorney General may be removed from the decisions or work product of other DOJ attorneys, ultimately the Attorney General is responsible for the proper functioning of the Department, and the decisions of the Solicitor General or of the OLC are properly attributable to the Attorney General. To illustrate by way of example, as we will see below, it is at least possible that Attorney General Ashcroft was left out of the loop for at least some of the advising with respect to interrogation policies in the last administration. To the extent this was the case, we should see this as a violation of proper functioning within the department. Thus, both for purposes of simplification in thinking about executive branch lawyering, and for normative reasons relating to institutional accountability, this Article will consider the work of the subordinate units of the DOJ as fairly attributable to the Attorney General. (11) Second, I should emphasize that I am interested in the Attorney General's role as a lawyer and legal advisor to the President here, and have not focused upon the other roles that the Attorney General plays as the head of a large bureaucracy (in which the Attorney General must make decisions about budgets, human resources, and the like). Finally, my discussion will be aimed at helping to clarify the relationship between the Attorney General and her client--the American people--primarily, and only secondarily about the relationship between the Attorney General and the President.
The model that I propose is aimed to respect our constitutional structure and its balancing of powers between especially the Congress and the President. Both the Congress and the President, more than the courts, can claim democratic legitimacy to represent the American people, the Attorney General's client. Both Congress and the President play a role in determining who will hold the office of the Attorney General. Both Congress and the President can play a role in determining the policy priorities of the Attorney General. Thus, if the American people are to be understood as the Attorney General's client, mediation through two often opposing forces will be inevitable in our system. Nevertheless, the Attorney General is an officer of the executive branch, and as such is subject to a far greater degree of control by the President than by Congress.
As I will show below, many of the current calls for greater independence on the part of the Attorney General stem from mistaken notions about the capacity of the Attorney General, qua lawyer, to determine either the best interests of the American people or the "best view" of the law. I reject the first of these contentions on the grounds that legal expertise, the distinctive qualification, one assumes, of any Attorney General, is not training in the best interests of the American people. I reject the second contention on epistemic grounds, for as I will argue, the law just does not admit of sufficient determinacy in all cases to allow the Attorney General to substitute her own judgment for that of the client, as mediated in by the President and the Congress. In the ordinary run of such cases, and against the background of Congress's many powers of oversight and control over the content of the law and the priorities of the executive, we should expect and not fear a high degree of presidential influence over the conduct of the legal affairs of the government--just as we would over the conduct of the affairs of the EPA, or of the Department of State. There may come a time in the course of any executive officer's career when, for reasons of professional integrity or even strongly held policy views about the public interest, a threat of resignation might be an appropriate course of action in the case of disagreement with the President. But there is nothing sufficiently special about the law, nor about the Attorney General as a lawyer, to require a greater degree of independence from the President than other cabinet officials. Indeed, our constitutional structure and history, the role of democratic accountability in promoting the public interest, and the frequent indeterminacy of the law, require the sort of dependency upon the President (and, to a lesser degree, the Congress) if the Attorney General is faithfully to represent the client, the American people.
With those general observations in mind, I turn to a more detailed description of the argument to follow. Part I will examine some recent approaches to the general question of the Attorney General's relationship to the President. Part II will explore some recent controversies from the current and immediately prior administrations that have raised the profile of the independence of the office more than has been the case since the Watergate scandals. Part III will examine some instructive episodes from the history of the office of the Attorney General. (12) Part IV will argue that much confusion arises from the false division of politics from law in a great many cases. Finally, Part V will propose a model of the Attorney General's client relationship that I am hopeful would serve to help understand lawyers working for the government more generally in the executive branch but indeed in all three branches of the federal government. (13)
I. RECENT VIEWS: OBJECTIVITY, NEUTRALITY, AND "THE BEST VIEW OF THE LAW"
In October 2007, the Senate Judiciary Committee held hearings on the nomination of then-Judge Michael Mukasey for the office of Attorney General of the United States. Judge Mukasey had been selected to replace Alberto Gonzalez, who resigned after a stormy tenure that raised questions about the politicization of the DOJ under his stewardship. (14) At those hearings, Senator Patrick Leahy (D-VT), chairman of the committee, honed in on the central question of this Article--whom does the Attorney General serve? Consider the following, from Senator Leahy's opening statement:
There's a good reason why the rule of law requires that we have an attorney general and not merely a secretary of the Department of Justice. This is a different kind of Cabinet position. It's distinct from all others. It requires greater independence. The departing attorney general never understood this. Instead, he saw his role as a facilitator for the White House's overreaching partisan policies and politics. Now, restoring the Department of Justice--and I want to restore it. I have enormous respect for the Department of Justice. I have from my days as a law student here in Washington at Georgetown. It begins by restoring integrity and independence to the position of attorney general of the United States. The attorney general's duty is to uphold the Constitution and the rule of law, not work to circumvent it. Both the President and the nation are best served by an attorney general who gives sound advice, takes responsible action, without regard to political considerations, not one who develops legalistic loopholes to serve the ends of a particular administration. The attorney general cannot interpret our laws to mean whatever the current President, Republican or Democratic, wants them to mean. The attorney general is supposed to represent all the American people, not just one of them. (15)
There is much to this account that is unobjectionable and even obvious. Of course the Attorney General ought not to be interpreting the laws to mean "whatever the current President ... wants them to mean," and of course the job of the Attorney General is to "uphold the rule of law." (16) Finally, and most importantly for purposes of this Article, of course it is true that the Attorney General serves all of the American people, and not just the President. But these statements, true and unobjectionable though they be, do not end the discussion. Indeed, they raise a number of key questions.
To begin with, what can it mean to represent "all of the American people?" Can one meet with them on a Tuesday morning in a conference room? Can one get them on a conference call to determine their desired ends and their preferred means to achieve them? Is it really true that the Attorney General serves "all of the American people" in a way that other cabinet officials do not? Does the Secretary of State or the Secretary of the Treasury serve only one person, or only a subset of the American people? If one cannot meet with the American people on a Tuesday or schedule all of them for a conference call, is there anything one might infer from the structure of the American government about how they are to be served by the Attorney General?
It should be obvious from the questions I have raised that my purpose in this Article is to suggest that while it is clear that the Attorney General serves all of the American people and not merely the President, the meaning of that statement must be unpacked carefully, and in accord with the general structure of the United States government established in the Constitution. It is, therefore, a question touching upon classic separation of powers considerations. In that light, Senator Leahy's sense of the uniqueness of the Attorney General's office cannot be maintained, at least not on the grounds upon which he has proposed it. That is to say, any cabinet official ought properly to understand herself as serving not just one American but all Americans. And no cabinet official ought to be willing to evade the laws of the United States in order to accomplish the President's will. "[W] hen the President does it, that means it's not illegal," was not a good argument when offered by Richard Nixon to David Frost, and it has not improved with age. (17) Any cabinet official must behave with a level of professional integrity that would envision informing the President that some chosen policy is either illegal or at odds with science or otherwise out of the bounds of professional responsibility as defined by that official's understanding and (presumably) expertise. At the same time, that requisite integrity--for the Attorney General or for any cabinet official--ought not to be regarded as a roving commission to implement a given cabinet official's own agenda. While a cabinet official serves all of the American people, his relationship to one American in particular--the President--does have paramount importance.
Much recent academic commentary has tended to focus less on the client relationship of the Attorney General and more upon the attitude of the Attorney General. A prime example in this regard is Nancy Baker's influential study of the Attorney General's office from 1789 through 1990. (18) After reviewing the early history of the office, which we will see began with little structure, staff, or mission, she examines various episodes from history under the rubrics of two models of Attorney General--the "advocate" and the "neutral." As one might guess, the former reflects an Attorney General as a servant and promoter of the President's agenda, whereas the latter might be seen as serving more in a quasi-judicial role, serving the law more generally rather than the President specifically. Baker finally concludes that an absolute choice between the models is probably neither necessary nor possible, but that invariably an Attorney General will embody features of each, even if one tends to predominate over the other. (19) She notes, for example that such "advocate" Attorneys General as Robert Kennedy and Robert Jackson have been looked upon kindly by history, and that advocates generally win the trust of the Presidents whom they serve, which can redound to the greater efficiency and effectiveness of their service. (20)
Roughly similar typologies have been followed by other leading scholars as well. (21) John O. McGinnis's comprehensive study of the opinion writing function of the Attorney General, for instance, employed the categories of "court-centered," "independent authority," and "situational" models of the Attorney General's opinion-writing, the latter two of which tend more toward the "advocacy" side and the former more toward the "neutral" side. (22) For McGinnis, the court-centered advocate will regard the executive branch as being bound by Supreme Court precedent, and will advise in a manner presumably consistent with and predictive of the lead given by the Court. (23) The independent authority model, by contrast, reserves a place for the executive branch to establish its own autonomous interpretations of the Constitution, without rejecting the ultimate authority of the courts on questions properly before them: "A premise of this view is that the text and structure of the Constitution assigns to the executive branch a responsibility to interpret the Constitution independently in carrying out its responsibilities." (24) McGinnis's "situational" model maps more or less onto Baker's advocate--the lawyer who takes his role to be finding a path in the law for the President's aims, even if that path might require a rejection of precedent or an aggressive, creative, or merely "situational" legal argument. (25) McGinnis, like Baker, does not choose from among the models proffered a single "best" model. (26)
More recently, some scholars who formerly served in the OLC, which in modern times has taken on most of the legal advising traditionally represented by the opinion-writing function of the Attorney General, have written important articles laying out their general understanding of that office's approach to legal advising. (27) Randolph Moss, while still serving as the head of OLC in 2000, published an influential account and defense of the "neutral expositor" model of executive branch legal interpretation, wherein he argues that historical, prudential, and constitutional reasons all lead to a requirement that the Attorney General hew to a "best view of the law" approach to legal advising, one that is purported to be marked by "strict objectivity." (28) The reasons he adduces include "[t] he very notion of a system designed to promote consistency and uniformity in legal advice," prudential concerns relating primarily to the prestige and influence that arguably would be undermined by OLC work that did not seem "fair, neutral, and well-reasoned," and the constitutional requirement that the President shall "take Care that the laws be faithfully executed." (29) Each of these reasons has a certain plausibility, and shall have to be considered in turn as we consider whether the argument that the Attorney General (here represented by the OLC) both has access to and owes the American people "the best view" of the law. The power of the "best view" in Moss's view is significant indeed:
The executive branch has no authority to act beyond the authority provided by the Constitution and relevant statutes of the United States, and, if the Constitution and relevant statutes are best construed to preclude a proposed policy or action, it is largely irrelevant whether a reasonable argument might be made in favor of the legality of the proposal. To act beyond the best view of the law is to act beyond those instruments that grant the official the status and authority that he or she seeks to employ. A reasonable argument might diminish the political cost of the contemplated action and it might avoid embarrassment in the courts, but it cannot provide the authority to act. Only the best view of the law can do that. (30)
Note the power of the "best view" here--it trumps even reasonable alternatives and alone can provide "authority to act." Now, in one sense, this may simply be a truism. If one arrives at a view that one thinks clearly superior even to all reasonable alternatives, then one might think oneself obliged to follow that superior view. But what to say of the adviser at the next desk over who holds the reasonable alternative position? That she is lawless? What if that person is not a colleague at OLC, but is the President? Moss does clarify, in addressing concerns about legal indeterminacy, that "[t]his is not to say that the executive branch lawyer should allow his or her personal legal views to dictate the scope of executive branch authority," but must respect both judicial and executive branch precedent. (31) It is not clear that this settles much of anything, however. Presumably, a lawyer must still rely on her own instincts about how far to follow precedent, and so an account of precedent is bound to be part of any given lawyer's own "best view" of the law, and the subjective element does not appear to be eliminated by requiring a "due" respect for precedent. Words such as "best" and "due," then, far from lending the sort of decisiveness and objectivity for which it appears they are employed, in fact end up as question-begging reminders to pay attention to the lawyer behind the curtain. (32)
In the wake of the controversies concerning the DOJ during the Bush Administration, Moss was joined by other like-minded commentators, especially by his colleagues from the Clinton OLC who have made a number of prominent critiques of the lawyering in Bush's OLC. In particular, Martin Lederman took an active public role in calling into question the Bush Administration's legal arguments in the war on terror, especially concerning the question of torture, or socalled "enhanced interrogation." (33) In addition, Dawn Johnsen, President Obama's recently withdrawn nominee to head the OLC, has written an influential article extending some of Moss's points, especially where the best view of the law is concerned. (34) Indeed, she quotes approvingly from the passage quoted above, while allowing that there is some dispute as to the best interpretation of the "best view" of the law. (35) Johnsen's article may fairly be seen as inspired by and an enrichment of a set of Guidelines proffered to Attorney General John Ashcroft in 2004, in a memorandum signed by nineteen former OLC attorneys who served primarily during the Clinton administration. (36) The Guidelines explicitly eschew the "advocacy model of lawyering" and emphasize the need for public disclosure of OLC opinions in the ordinary course. (37) To be sure, while the guidelines implicitly honor a court-centered approach to legal interpretation in the main, they also recognize that the OLC serves the executive branch, and should in its analysis "reflect the institutional traditions and competencies of the executive branch as well as the views of the President who currently holds office." (38) The document, in other words, is hardly hostile to executive power or supine in the face of either congressional or judicial encroachment. It is balanced and moderate in its approach to the work of OLC, and recognizes both the need to give candid advice to the President but also to work within the law to help the executive branch to achieve its policy aims.
Both Johnsen's article and the Guidelines seem to share Moss's confidence that there is a best view of the law to be obtained, and that, once having obtained it, the OLC lawyer, and, presumably, the Attorney General, are bound by it as by the law itself. Although Johnsen notes that the OLC's legal interpretations may be overruled by the President or Attorney General, she also notes that this is "exceedingly rare." What is required for such an overruling is either the President or Attorney General's "good faith determination that OLC erred in its legal analysis." (39) The nature of this judgment is not spelled out in detail, however. What if the difference is a difference in the degree of confidence? For example, what if the President were to see the question of law, reasonably, as a much closer question than the OLC? In other words, how much better than the alternatives does a view have to be in order to be, as it were, authoritatively best? If one thinks the courts likeliest to rule one way, but thinks that they may well rule otherwise, and ought to, what is the best view of the law?
Some authors seem to suggest that the oath of office for the Attorney General (and those who work with the Attorney General) creates either the obligation or the ability to provide the sort of "best view" of the law that pervades many contemporary discussions of executive branch legal interpretation. Philip Hamburger recently showed that this tradition has its roots, at a minimum, in English precedents for the office of Attorney General, and colored the approach to the job of the first Attorney General of the United States, Edmund Randolph. (40) Randolph Moss similarly roots his argument in the President's constitutional oath of office. (41) Harold Bruff considers both the President's oath and the separate oath taken by executive branch lawyers and notes that the oath "serves as a daily reminder to those who take it that they must defend the Constitution even when it is not easy to do so." (42) Certainly the oath is an important social practice that adds solemnity to the importance of the job of serving as a lawyer in government. It is not apparent, however, how the oath obliges a lawyer to adopt a particular interpretive stance whether that of "neutral" or "advocate"--in a circumstance where there may be a range of reasonable views of the law, or at least a plausible alternative to the so-called "best view" that has emerged from, say, the OLC. Can an oath reduce the amount of indeterminacy in the law, or does it give guidance to the perplexed in trying to advise the President on a difficult question of law?
These questions and others like them lead me to propose a model here that, though amenable in many respects to the rule of law considerations, the respect for the oath of office, and the need for due professional independence on the part of the Attorney General, envisions and endorses as proper a greater degree of presidential control over executive branch legal determinations than seems permissible under the language and logic of such recent influential commentary as Moss's, Johnsen's, and Bruff's. The model I propose acknowledges that the President is not, in fact, the client of the Attorney General in the traditional sense, but entails that he is that client's--the American people's--chief spokesperson. The model recognizes that the Congress, the courts, and the laws also speak for the client--each represents a mode of mediation in the client relationship that must be taken seriously by the Attorney General. But in difficult cases--and difficult cases in the end are what we must be interested in--the mediating role of the President must be seen to be of paramount importance for the Attorney General, as for any other unelected officer of the executive branch. In short, because I am skeptical that the Attorney General or her subordinates can, in difficult cases, claim access either to a superior knowledge of the client's interests or a "view from nowhere" giving them supreme confidence in the rightness of their own "best view" of the law, I am skeptical that an Attorney General violates her oath of office or her professional duties by serving a President who reasonably disagrees with the Attorney General's own best view of the law. (43)
II. RECENT CONTROVERSIES: TOO MUCH POLITICS IN THE DEPARTMENT OF JUSTICE?
The administrations of President George W. Bush and his successor, President Barack Obama, have both produced episodes of controversy giving rise to allegations of politicization of the work of the DOJ, and resulting in concomitant cries for greater independence on the part of the Attorney General from the political will of the President. In this Part, we will review two episodes from the Bush administration and two from the Obama administration. The selection is not meant to be exhaustive nor, despite the numerical symmetry, to draw any particular equivalence between the administrations. Rather, it is simply to show that, in large and small ways, the question of the proper understanding of the role and functioning of the Attorney General is multifaceted and arises frequently. From among several candidates arising in the eight years of the presidency of George W. Bush, we will consider two episodes: (1) the production of the so-called "torture memos" by the OLC under the direction of Jay s. Bybee and (2) the firing of the U.S. Attorneys (a key precipitating factor in the resignation of Attorney General Gonzalez). In the first two years of the Obama administration, a minor controversy, but one potentially instructive for our purposes, arose last year after reports surfaced that Attorney General Eric Holder had sought advice about the constitutionality of legislation giving a vote in the U.S. House of Representatives to the District of Columbia from beyond the OLC, which had reiterated its standing view that such legislation would be unconstitutional. Of greater significance, undoubtedly, the Obama administration has struggled to formulate a clear policy with respect to the proper forum for prosecuting suspects detained in the war against terrorism, to date not yet having decided whether, for instance, Khalid Sheikh Mohammed will be tried by a civilian or military tribunal. At times, the administration has suggested that such a determination is for the Attorney General to make independent of political pressure, at other times it has suggested that the President will make the decision. Each of these episodes sheds light on different aspects of our central question. In Part IV, after proposing my own model of the Attorney General's client relationship, I will revisit these episodes further to evaluate their significance for understanding the role of the Attorney General.
A. The Torture Memos
In the aftermath of the terrorist attacks of September 11, 2001, the United States undertook an aggressive military campaign against the perpetrators, including the invasion of Afghanistan. In the course of this military campaign, many suspected terrorists have been captured and taken into custody at various locations around the world, most notably at Guantanamo Bay, Cuba, and Bagram Air Force Base in Afghanistan. (44) Many difficult legal questions have naturally arisen in the context of a new kind of war, with ill-defined boundaries of duration, battlefield, victory, and defeat. (45) Urgent among these, early on in the conflict, was the question, "What are the permissible means of interrogating detainees who might have vital information about ongoing terrorist plots threatening American lives and interests at home and abroad?" (46) The operative question became whether we were "forward-leaning" enough in our policies.
In August 2002, OLC issued an opinion answering a CIA request for guidance as to acceptable techniques for interrogation of detainees, especially in light of the 1994 law implementing the Convention Against Torture. (47) That memorandum, principally authored by OLC's John Yoo and signed by Assistant Attorney General Jay S. Bybee, along with at least one other produced in 2003, subsequently came to be known as the "torture memos." (48) The torture memos were called as such for their approval of certain techniques, most notably "waterboarding," which many scholars regard as violations of the torture statute in question. (49)
It is beyond the scope of this Article to delve into detail into the many infirmities that have been noted by commentators reviewing the torture memos, especially the original 2002 memo. (50) These have been well documented by, among many others, Jay Bybee's successor, Jack Goldsmith, who eventually withdrew the memos as indefensible, despite a strong presumption in favor of stare decisis at OLC. (51) It is nonetheless perhaps useful to pause briefly to note the enormity of the errors, which included even the failure to make any mention of the celebrated Steel Seizure case, (52) which is the starting point for any discussions of executive power in the face of Congressional action. Consider the comments of other Bush administration officials evaluating the memos: Attorney General Gonzales's successor, Michael Mukasey, called the memos "slovenly," while Jack Goldsmith's successor as head of OLC, Daniel Levin, recalled that his first reaction to the 2002 memo was to remark, "[t]his is insane, who wrote this?" (53)
In the face of such remarkably widespread and vehement condemnation of the work of highly credentialed, highly skilled lawyers in the OLC, it is natural to ask whether the Attorney General (John Ashcroft at the time) and his top advisers such as Jay Bybee had fundamentally misconceived their roles, as Senator Leahy's opening statement at the confirmation hearing of Michael Mukasey suggested. Had John Ashcroft (and, presumably, at least Mberto Gonzales after him) come to see their role as mere enablers of whatever policy preferences the President expressed? (54) Had they lost sight of their representing not one American but all Americans? Or, in the alternative, were the errors simply attributable to the urgency with which the memos were prepared, and the psychological pressure attendant to the fear of another attack potentially threatening thousands of lives in the wake of the trauma of 9/11? (55) Was this bad lawyering the result of an inadequate model of government lawyering, or was it attributable to other factors?
Before answering this question, one must consider the possibility, indeed the likelihood, that the principal author of the torture memos, John Yoo, certainly appears to have given his advice in good faith. This was the conclusion, significantly, of the DOJ's own internal investigation into the torture memos. (56) Associate Deputy Attorney General David Margolis, reviewing the findings of the DOJ's Office of Professional Responsibility, concluded that John Yoo likely erred because of "loyalty to his own ideology and convictions" in authoring opinions "that reflected his own extreme, albeit sincerely held, views of executive power." (57) Without question, for years before his own service in the Bush Administration's OLC, and before the attacks of September 11, 2001, John Yoo had written extensively about separation of powers and especially about controversies involving war powers; and throughout his oeuvre has advanced a consistently pro-presidential view. (58) ff John Yoo had always been convinced of the truth of the views he expressed in the torture memos (at least as to the broad constitutional questions at stake), and if he was being asked to write about those views in a period of national crisis when the stakes seemed enormous, how ought John Yoo to have approached the question, "what's the best view of the law here?" Ought he to have deferred to precedents he believed to be false when he believed lives were at stake? Even one who believes his views to be mistaken ought to pause before giving an easy answer to that question.
Finally, it ought to be noted that, although Jack Goldsmith did withdraw two memos on the grounds that their legal reasoning was unsustainable, the subsequent memos issued by OLC continued to approve largely the same methods that the original memos had countenanced (albeit on narrower grounds) and even extended approval to the use of such methods in combination. (59) Indeed, to this day, although subsequent executive orders have rendered the OLC memoranda on interrogations inoperative, there is at least no public OLC opinion finding that any of the methods approved by the original or subsequent torture memos in fact amount to violations of the torture statute. (60) If one is to conclude, therefore, that the Yoo and Bybee memoranda were wrong for approving torture, therefore, rather than for their "slovenly" or "insane" reasoning, then many more lawyers than Yoo and Bybee would be implicated in such a conclusion. Any reflection upon the proper model of executive branch lawyering in light of these memos should not stop with the memos whose reasoning was most egregiously flawed when considering whether the approval of techniques amounting to torture was the result of a fundamental misunderstanding of the job of the government lawyer or is to be attributed to some other cause.
B. The U.S. Attorney Firings
The Seal of the Department of Justice, dating to the Nineteenth Century, notoriously features an enigmatic Latin motto: "Qui pro domina justitia sequitur." (61) Over the years, the motto and its proper translation have provided a source of bafflement, particularly to those trained in Latin, for the motto does not reflect proper classical Latin grammar nor is it easily rendered into intelligible English syntax. (62) Happily, however, there is a canonical DOJ explanation of the motto's meaning and import, if not a satisfying historical or linguisitic one. The canonical explanation has it that representatives of the English Crown, appearing in court, would identify themselves as "qui pro domina regina sequitur"--"[one] who prosecutes for our Lady the Queen," and that Attorney General Jeremiah S. Black (in office from 1857 to 1860) thought it fitting to adapt the phrase for this country to reflect that here, the Attorney General appears on behalf of justice itself. (63) This quaint history reflects a deeply held ideal of justice impartially and impersonally meted out by attorneys for the United States of America--similar to the traditional depiction of Lady Justice herself as one blindfolded. The power of the DOJ, after all, includes the awesome power of the state to bring prosecutions that might deprive one of property, liberty, or life itself. The DOJ has a strong institutional interest, therefore, in preserving an image of impartiality in the execution of its mission.
That reputational interest suffered a severe blow in the eyes of many when, in late 2006, nine U.S. Attorneys were instructed to submit their resignations in the middle of the second term of President George W. Bush. (64) In informing the U.S. Attorneys that they ought to "move on" from their posts, it appears that the administration gave them no reasons for their dismissals, and in fact led them to believe that they ought not to regard them as related to the performance of their duties. (65) Nevertheless, as the dismissals became public and both Congress and the press took an interest in understanding the motives for the firings, the DOJ and the White House suggested that they were rooted in performance reviews. (66) This explanation was difficult to defend in light of the apparently outstanding performance reviews of seven of the nine fired U.S. Attorneys, as well as the failure to inform any of them of the nature of the putative performance-related worries, let alone to give them an opportunity to correct them. (67)
Meanwhile, it did not take long to emerge that, in the case of at least one of the fired U.S. Attorneys, David Iglesias of New Mexico, interventions by elected officials--specifically Senator Pete Domenici (R-NM), who was worried about the pace of an investigation into potential wrongdoing by Democratic officials in his home state--played a role in the DOJ's decision making. (68) Senator Domenici acknowledged having communicated both with Iglesias's office and with the DOJ about the progress of the investigations into the Democratic officials--giving rise to speculation that political considerations had invariably clouded any judgment about Iglesias's performance overall. (69)
A joint report of the DOJ's Inspector General (IG) and Office of Professional Responsibility (OPR) concluded that the process to remove the U.S. attorneys was "fundamentally flawed." (70) The report found that "primary responsibility" for the "serious failures" lay with Attorney General Alberto Gonzales and Deputy Attorney General Paul McNulty. (71) Such mid-term removals of a President's own appointees were "unprecedented," and, therefore, likely to raise questions--demanding a serious and responsible process to keep the DOJ above suspicion in the decision to fire such high-ranking officials. (72) So critical were the findings of the report that Gonzales's successor, Michael Mukasey, responded by appointing a special prosecutor to investigate whether anyone committed criminal acts in the process leading to the dismissals of the nine U.S. attorneys. (73) As of this writing, that investigation has not yet been completed.
The firing of nine U.S. attorneys for reasons that remain elusive, and about which the DOJ, including the Attorney General himself, was unable to give a convincing or even coherent account, obviously raises a number of important questions. If we assume that the reasons for the firings were entirely legitimate, then why were the decisions so difficult to explain, and why were apparently false (performance-related) reasons offered? One might imagine reasons for the firings that would be embarrassing but not necessarily illegitimate. An obvious explanation would be that the position of U.S. attorney is a very prestigious one, and among the plumb patronage positions that a President can dole out to friends and political allies. Given the at-will nature of the employment (as the DOJ report noted, presidential appointees, including U.S. attorneys, may be removed "for any reason or no reason"), it would generally not constitute an illegitimate aim for the President to ask some U.S. Attorneys to move on in order to make room for others whom the President might wish to honor with such appointment. (74) Evidence indicates that at least one of the firings was likely made for this reason. (75)
Even if one suspects, though, that the episode arose from such purely political--but not illegitimate--motivations, one cannot deny that the firings raised troubling alternative possibilities. In particular, the plausible speculation that some may have been motivated by a desire to pursue partisan investigations and prosecutions was very damaging indeed, running contrary to the notion of federal prosecutions on behalf of justice itself, rather than the President or the President's political party. It is obvious, therefore, that the miscommunications and apparent (at best) arbitrariness of the process left the DOJ--and especially Attorney General Gonzales--open to the unfortunate but understandable charge of "politicization," here meaning something like "adopting illegal means to political ends" rather than merely politics in the ordinary (and legitimate) sense of the term. (76) Such worries inevitably give rise to questions such as those posed by Senator Leahy of then-Judge Mukasey, about the willingness of the latter to stand independently from the President when the need arises. As our investigation here continues, we shall have to consider whether there is some extra degree of independence required of lawyers working for the President, as opposed to whatever independence might be expected of other cabinet or other executive branch officials.
C. The D.C. Voting Rights Bill in the Obama Administration
According to press reports, early in his tenure as President Obama's Attorney General, Eric Holder learned that the OLC continued to stand by its opinion (expressed in Congressional testimony given during the prior administration in 2007) that a proposal to grant the District of Columbia a vote in the U.S. House of Representatives would be unconstitutional. (77) Holder has been, at least since 2007, a public supporter of such a measure, as has President Obama. (78) In the event, the OLC reportedly affirmed the position it had taken in the prior administration, that the proposed legislation would be unconstitutional. (79) Attorney General Holder sought further advice from the Solicitor General's office, which informed him that it would be comfortable defending such legislation should the President sign it into law. In light of that advice, and his own independent judgment, the Attorney General apparently determined that the proposed law would be constitutional or could plausibly be defended as such.
Some former members of the OLC objected rather strenuously to Holder's determination, citing it as evidence of the politicization of the DOJ. Ed Whelan, a former principle deputy at the OLC, penned an opinion piece for the Washington Post entitled "Look Who's Politicizing Justice Now." (80) Whelan's argument is a bit difficult to follow, as he acknowledges that the OLC's opinion-writing function comes as a delegation from the Attorney General, to whom the OLC is responsible, rather than the reverse. Nevertheless, Whelan contends that the Attorney General may only properly override the advice of the OLC after "a full and careful formal review of the legal question." (81) Now, Whelan doesn't specify what the necessary elements of such a process would be, but he makes clear that his view is that a mere litigation position proffered by the Solicitor General's office is not sufficient:
Holder didn't ask for Katyal's best judgment as to whether the D.C. bill was constitutional. He instead asked merely whether his own position that the bill is constitutional was so beyond the pale, so beneath the low level of plausible lawyers' arguments, so legally frivolous, that the Solicitor General's office, under its traditional commitment to defend any federal law for which any reasonable defense can be offered, wouldn't be able to defend it in court. (82)
Whelan, to be sure, cannot cite to a public record of just exactly what the Attorney General asked Mr. Katyal. Nor, significantly, does he cite to any authority for the proposition that the Attorney General, in advising the President, must advise him only to follow the most conservative or most widely accepted assessment of a bill's constitutionality. Doubtless the President is entitled to know the range of views and their relative plausibility among legal experts; but knowing of the range of views and being bound by a particular part of that range are different matters.
Mr. Whelan's concerns have been echoed by John McGinnis, also a former OLC attorney and deputy assistant Attorney General. (83) Professor McGinnis emphasizes that each branch of government owes its own fealty to the Constitution, and has its own concomitant duty of interpretation. But he goes a bit further:
The Framers established a Constitution of separated powers in part to provide citizens multiple protections from unconstitutional acts. The legislature reviews a bill when it passes it, the President when he signs it and the judiciary when it enforces it. In each of those decisions the constitutional actor must give his independent and reasoned judgment on the legislation's constitutionality. (84)
Professor McGinnis, like Mr. Whelan, cites to no authority for the proposition that the President must give his "independent and reasoned judgment" on the constitutionality of any and every act signed into law. Of course, none of the actors may support a law of manifest unconstitutionality--each of them is charged with obeying the Constitution. But where plausible arguments may be maintained both for and against the constitutionality of proposed legislation, what is to guide the President? It is far from clear, in what either Mr. Whelan or Professor McGinnis writes, by what measure one is to determine whether, in the case at hand, Mr. Katyal gave his "best judgment" about the law in question or merely a judgment that the law could be defended. It is possible, of course, for the Solicitor General's office genuinely to believe simultaneously that the position it has adopted is (a) correct ("best judgment"), and (b), unlikely to prevail when tested in court. The latter assessment will often, but not always, influence an attorney's judgment as to the "best view of the law."
This last question is crucial for the advising function of the Attorney General, and is brought into high relief by this episode. One might take "best view of the law" to mean a view of the law that comports best with one's own sense of proper jurisprudence, even if that view has been rejected by the Supreme Court. History makes plain that even the Supreme Court does not hold itself bound, in determining the "best view" of the law, by what prior Supreme Court decisions have said on the matter. (85) Alternatively, one might say that the "best view of the law" ought to be a prediction of what the Supreme Court would likely hold to be true in a given case. An interim position might hold that some sort of consensus of legal practitioners and academics ought to be sought, rather than the best view of the individual Attorney General, who might after all be an outlier. While any of these models for rendering a best view of the law might be defensible, they are not mutually consistent and none is obviously required by the Constitution or by tradition. (86) The question of how to render the best legal advice to the President cannot be avoided, even if the answer is not clearly spelled out by the Constitution or by tradition. (87) At the same time, the question is hardly unique to government lawyers. Lawyers in all manner of practice frequently find themselves in situations where explaining what "the law" requires will involve laying out a range of possibilities--a range of views of the law--that might be measured as more or less plausible by a variety of metrics (acceptance by courts or the academy, comportment with a particular kind of jurisprudence, etc.). A client asking for "the best" view might generally, but not always, be asking for the most conservative--in the non-ideological sense--view of the law. (88) But such is not self-evidently "best" as a constitutional matter, whether the client is a private citizen, a government agency, the White House, or the American people.
D. Khalid Sheikh Mohammed's Expected Trial
A final key episode, also emerging in the current administration and, like the torture memos, arising out of the ongoing struggle against terrorism, concerns the decision of whether and how to prosecute high profile suspects in the war on terror, such as the alleged 9/ 11 mastermind, Khalid Sheikh Mohammed (KSM). The chief question concerned whether the better venue for any trial would be in a federal court or in a military commission. In November of last year, Attorney General Eric Holder, in what was purported to be his own decision made independently of the White House, announced that KSM would be tried in federal court in Manhattan, even as he acknowledged plans to try other suspects before military commissions. (89) The Attorney General's decision was made after a process that included consultation with the Department of Defense, with which the DOJ had jointly issued a prosecution protocol to govern the cases of detainees at Guantanamo Bay. (90)
The decision was met with heavy criticism, especially but not exclusively from Republicans. (91) The critiques touched upon issues from the safety and practicality of holding such a high profile--and highly emotionally engaging--trial so near the epicenter of the 9/11 tragedy, to the propriety of a system in which some suspects would get trials in regular federal courts and others in military commissions, giving rise to the charge, potentially, of forum shopping. (92) In the months since the decision was made, the ferocity of the opposition, and especially the lack of support for it among key officials in New York City such as the mayor and chief of police, appear to have forced the administration to reconsider whether a federal trial for KSM is either possible or desirable. Notably, it has now been suggested that the President himself will be involved in any final decision, rather than leaving it up to the so-called independent judgment of the Attorney General. (93)
Again, it is beyond the scope of this Article to judge the merits of civilian trials as against those of military commissions for persons such as KSM, or to judge the merits of a dual-track system such as that originally envisioned by the Obama DOJ (in which some suspects would have civilian trials, and others trials before military commission). Rather, it is my purpose to note that with regard to an issue of enormous magnitude, and doubtless in light of the Bush Administration's struggle with the torture memos, it has been asserted that the decision of where and how to try KSM ought to be a determination of the Attorney General, rather than of the President.
This suggests a paradigm of strong independence for the DOJ, especially when it comes to the case of criminal prosecutions. Doubtless, as a general matter, for reasons suggested above in the consideration of the firing of nine U.S. attorneys, such independence is generally to be desired--it is justice, and not presidential whim, that the federal prosecutor represents. At the same time, is it realistic or even desirable, in a decision with, at the very least, momentous foreign policy and national security implications, that the Attorney General would be asked to make the decision on the basis of his training and background as a lawyer? Would it be inappropriate to take into account the feelings and preparedness of New Yorkers to host such a trial? Would such considerations be appropriate for the Attorney General, or do they belong to the White House? Should the Secretary of State or the National Security Adviser be consulted by the Attorney General? If so, would the Attorney General be weighing their advice as a legal matter, or under some other rubric? To be sure, both the joint prosecution protocol and the ordinary principles of federal prosecution (which apply along with the protocol) include flexibility in exercising prosecutorial discretion that is not limited to purely legal considerations. (94) Nevertheless, cases of such extraordinary nature as those arising out of the war on terror and the detention facility at Guantanamo (which presents its own political, constitutional, and foreign affairs questions apart from those generally implicated by the war on terror) by their very nature exceed the scope and responsibility of the Attorney General, even when acting in concert with the Secretary of Defense.
As of this writing, it does not appear to be settled whether KSM will be afforded a civilian or a military tribunal. It does appear, however, that the White House is backing away from its earlier claim that this decision is one proper to the discretion of the Attorney General. Indeed, it has been reported that President Obama will now "help" the Attorney General to select the location of the trial. (95)
III. HAS IT EVER BEEN THUS?
In this section, we will briefly examine some episodes from the less recent past to see what light they might shed on the role of the Attorney General (and of government lawyers more generally) in serving the American people. We will examine the early history of the relationship of the Attorney General and the President with federal prosecutors, the history of wartime advice to Presidents, and finally the experience of the independent counsel statute as it played out in the investigation and eventual impeachment of President William Jefferson Clinton. But first, some general discussion of my approach to the history of the office is in order.
To begin with, history does not generally provide iron-clad normative answers to constitutional questions. Although I am sympathetic with originalism as a method of constitutional interpretation, I agree wholeheartedly (though not joyfully) with Henry Monaghan's recent observation that such approaches "cannot account for a good deal of the contemporary constitutional order; an order ... that embodies massive departures from any original understanding of the text." (96) Professor Monaghan's examples of paper money, the rise of the modern administrative state, and the transformation of the presidency strike me as particularly helpful in demonstrating how important some of these non-originalist developments are, and how ineluctable, it would seem, is the acceptance of them as part of our constitutional/legal order. (97)
In the context of this Article, agnosticism about the original constitutional understandings is only the greater, for the office of the Attorney General of the United States was created not by the Constitution but by the Judiciary Act of 1789. (98) As Susan Low Bloch helpfully put it in the title of her history of the office, "[i]n the [b]eginning, [t]here [w]as [p]ragmatism." (99) Thus, while Philip Hamburger recently made a persuasive case that the origins of the office of the Attorney General can be argued to place the role somewhere between that of an ordinary lawyer and a judge, I maintain, and the history of the office as laid out in Bloch's work and elsewhere establishes, that neither statute nor the Constitution countenances a role for the Attorney General other than as an officer of the executive branch. (100)
A. The Early History of Presidential Control over Federal Prosecutions
It is well established that, at its inception, the office of the Attorney General was a far cry from today's instantiation by just about any measure of stature, including salary, staff, and authority. (101) Consider, for instance, that the first Attorney General, Edmund Randolph, had neither a clerk nor a physical office, and was paid such a meager salary (half what other cabinet officers were paid) that he had to maintain an active private practice on the side. (102) Today's Attorney General, by contrast, oversees a DOJ with 111,993 employees and a budget for Fiscal Year 2010 of $27.7 billion. (103) Congress did not provide a clerk or a physical office to the Attorney General until well into the nineteenth century. (104)
Noteworthy, therefore, given the modest stature of the office of the Attorney General at its inception, is the fact that there were, however, district attorneys employed by the national government, who were not subject to the control of the Attorney General. (105) Rather, from the earliest days of the republic, the President himself saw fit to direct and redirect the prosecutorial power of the United States, as has been painstakingly documented by Saikrishna Prakash. (106) As Prakash notes, President Washington took a great personal interest in promising criminal and civil litigation as a general matter, and also personally directed the district attorneys to drop or bring suit in individual cases. (107) Less edifying, but no less instructive, are Prakash's conclusions that President John Adams took a similarly active interest in directing the work of district attorneys, but "ingloriously ... to harass those who hounded him." (108)
As Prakash notes, while the Alien and Sedition Acts were themselves attacked as unconstitutional, the President's decisions to prosecute under them (or to decline to do so) were not similarly challenged. For Prakash, it was simply taken as a given in the founding era that the President had the authority to direct the prosecutorial discretion of the district attorneys. Perhaps the clearest evidence adduced on the point comes from the correspondence of Thomas Jefferson, who unmistakably links the power to control federal prosecutions to the Constitution:
The President is to have the laws executed. He may order an offence then to be prosecuted. If he sees a prosecution put into a train which is not lawful, he may order it to be discontinued and put into a legal train.... There appears to be no weak part in any of these positions or inferences. (109)
As Prakash astutely observes, although the constitutional dimension of Jefferson's argument is unmistakable, it is not unambiguous. That is, it is not clear whether Jefferson is relying on the "vesting clause"--granting the "executive power" to the President--or the "take care" clause--imposing the duty on the President to see that the laws be faithfully executed. (110)
From Prakash's careful study, we may draw some modest but not altogether insignificant conclusions. First, it cannot be argued, in light of the earliest instantiations of the office, that the office of the Attorney General is of its essence less dependent upon the President than other cabinet posts. Indeed, given its relatively humble status among cabinet posts, something like the opposite might more easily be inferred, though for our purposes the former is more than sufficient. Second, apart from the office of the Attorney General itself, given the early Presidents' active role in the exercise of prosecutorial discretion--in the bringing of cases and in refraining therefrom, it cannot seriously be argued that the nature of legal determinations is such that, again, unlike other executive branch functions, these determinations are to be made somehow with less attention paid to the policy judgments of the President. On the contrary, given the President's obligation to see that the laws be faithfully executed, legal determinations--even determinations whether to prosecute--cannot be divorced from the larger context of presidential responsibility and presidential decision making. (111)
Finally, it is not necessary here to examine in detail, much less to settle, the dispute (such as it may remain) between Prakash and other scholars, such as Harold Krent, who have argued that a more diffuse power over prosecution of federal crimes was the historical norm. (112) Rather this brief excursus into the early history of federal prosecutions is focused on the more limited goal of gleaning insights into the nature and structure of the relationship between the President and the Attorney General--not to settle any and all questions falling under the much broader rubric of the theory of the unitary executive. To establish that the earliest understandings of the Attorney General do not suggest an office of greater dignity or independence than that of other cabinet officials does not require a full consideration, much less vindication, of the purest or strongest formalist parsings of, say, the vesting clause. The ample evidence adduced by Prakash demonstrates that, from the start, Presidents have thought that they, not the lawyers working for the executive branch, had the ultimate authority to determine whether to undertake prosecutions.
This subsection relies upon the key facts established by Prakash, as well as the very modest beginnings of the office of the Attorney General, to show that the President's power to interpret and apply the law was envisioned as a matter of course from the very earliest moments of giving flesh to our constitutional structure. It would take much later developments, such as Reconstruction, the rise of the Administrative state, and the twentieth century's burgeoning national security state to result in the vastly increased scope of the Attorney General's responsibilities and presumed competency. But those practical developments, while obviously of enormous importance for the day-to-day exercise of discretion and judgment in the ordinary functioning of government, do not diminish the import of these early episodes for establishing the principles by which we might think through the proper independence of the Attorney General from the President.
B. Attorneys General in Times of War
As noted above, among the deficiencies most often cited in the torture memorandum prepared by John Yoo and Jay Bybee, and ultimately withdrawn by Jack Goldsmith, is its failure to cite the "Steel Seizure" case, Youngstown Sheet & Tube Co. v. Sawyer, (113) in its analysis of the separation of powers issues at stake. Take Professor Stephen Gillers, writing in the The Nation:
Although the OLC memos broadly construed presidential power in foreign affairs, they ignored the Supreme Court's landmark 1952 "steel seizure case," which greatly restricts that power and contradicts the OLC's expansive claims. It would be like advising a client on school desegregation law and ignoring Brown v. Board of Education. (114)
Though one might be hard pressed to produce the language in Youngstown that "greatly restricts" presidential power in foreign affairs, it is surely true that Justice Jackson's famous Youngstown concurrence precludes a grandiose reading of the commander-in-chief power. (115) This Youngstown wrinkle to the torture memo is particularly interesting in light of the following exchange between Justice Jackson and Philip Perlman, the Solicitor General who argued unsuccessfully in defense of the President Truman's order to seize steel mills. Mr. Perlman's briefing relied significantly on arguments advanced by then-Attorney General Jackson in defense of a 1941 seizure of an aviation plant by the Roosevelt administration:
Mr. Justice Jackson: ... I do not believe that unless you amend those points you can make that stand as a precedent for this. I looked it up because I wondered how much of this was laid at my door.
Mr. Perlman: Your Honor, we lay a lot of it at your door.
Mr. Justice Jackson: Perhaps rightly.
Mr. Perlman: I think the statement--
Mr. Justice Jackson: I claimed everything, of course, like every other Attorney General does. It was a custom that did not leave the Department of Justice when I did. (116)
There is some amusing irony in the great Justice Jackson's frank admission that where one stands on issues of executive power depends a great deal on where one sits. To wit, an Attorney General of the United States may be expected to make even extravagant claims on behalf of executive power--not out of a misunderstanding of the job, but out of a proper understanding of it. To be clear, even if that point is correct, it would not excuse an OLC attorney, in rendering a "best view of the law" as the OLC has come to understand its role, from omitting relevant authority in an opinion offering advice to an executive branch "client" such as the CIA. (117) It is one thing for the executive branch, if it wishes, to argue to the Supreme Court that a case such as Youngstown errs in its analysis of separation of powers; it is quite another to offer legal advice as if Youngstown did not exist and did not therefore need to be contended with. (118)
As Jack Goldsmith lays out nicely in his discussion of the opinion of attorneys general in wartime, examining the cases of Abraham Lincoln's Attorney General, Edward Bates, who authored a memo justifying Lincoln's suspension of the writ of habeas corpus, and of Attorney General Robert Jackson's memo in support of Franklin Roosevelt's destroyers-for-bases deal in 1940, very similar statements were made about the work of Bates and Jackson in comparison to the criticisms leveled at the torture memos of the last administration. Perhaps the clearest example Goldsmith adduces is that of constitutional scholar Edward Corwin, who considered the Jackson opinion "an endorsement of unrestrained autocracy in the field of foreign relations" and the most "dangerous opinion ... ever before penned by an Attorney General of the United States." (119) How could it be that the author of so careful and balanced an approach to separation of powers in the realm of foreign and military affairs as Justice Jackson in concurrence in Youngstown had only a few years earlier authored a memo in favor of strong executive action capable of eliciting such derisive criticism? Put in terms of the central question of this inquiry, had the younger Jackson failed to be a lawyer for all of the American people rather than just one American (the President)? How would one determine that?
It is worth pausing to consider that two key players in the torture memo controversy, Jack Goldsmith and John Yoo, have recently written books devoted in large measure to considering executive power in these and other war time episodes. (120) But their aims in doing so are rather different.
Jack Goldsmith, in The Terror Presidency, expressly considers the work of OLC lawyers in the Bush administration against the backdrop of these other historic periods of crisis, especially during the Civil War and World War II. But his purpose, in marked contrast to Yoo's, is to emphasize the willingness of the earlier administrations to work with the Congress to achieve their ends. (121) Goldsmith argues that the key lessons to be learned from the Lincoln and Roosevelt cases are found in a model of "democratic leadership" that involves "soft factors of legitimation" such as consultation and "the appearance of deference." (122) Where there were assertions of executive authority, there were generally efforts to secure (if only after the fact) congressional approval of decisions taken by Lincoln and Roosevelt, contrasting with the "go it alone" approach of the Bush administration, as Goldsmith describes it. (123)
John Yoo's Crisis and Command reads at times like a reply brief to Goldsmith's argument. Yoo builds a case that presidential greatness has correlated with the willingness of executives such as Lincoln and Roosevelt vigorously to press claims of executive power in times of crisis. (124) Consider a typical formulation: "our greatest Presidents have had to act because they have judged their actions necessary to benefit the nation or protect it from harm." (125) Although he specifically disclaims any purpose in writing to justify the policies of the Bush administration, Yoo certainly mounts an argument that the administration's efforts to combat terrorism--including its treatment of prisoners--fall within the ambit of the President's powers as Commander in Chief. (126)
Despite their differences, both authors clearly think that the style and approach to the presidency itself of the sitting President is a necessary component of understanding the work of the President's lawyers. Both authors emphasize that history may well judge the Bush administration more kindly than many current commentators do, in just the way that the contemporaneous vilification of Lincoln and Roosevelt as dictatorial eventually gave way to more generous evaluations of their presidencies. In doing so, they both describe a narrative where future generations' assessment will turn on whether the efforts in the war on terror prove to have been effective and in the nation's best interest. This latter point is key for our search for a model of executive lawyering. Consider, for instance, these questions posed by Professor Gillers in his treatment of the torture memos:
The client deserved better, and that raises another issue, the most troubling. Who was the client? The lawyers told the President what he wanted to hear, but the nation was their client, and its sole interest was in thorough and independent legal analysis. Neither the President's political agenda nor the authors' views of what the law should say can be allowed to slant the OLC's work. So maybe the best and brightest lawyers got it so wrong because they forgot whom they served. Maybe they acted politically, not professionally. If so, we are dealing with a perversion of law and legal duty, a betrayal of the client and professional norms, not mere incompetence, which would be bad enough. (127)
There is, at a minimum, a certain rhetorical appeal to Gillers's case--he diagnoses the problem with the torture memos in much the same way that Senator Leahy diagnosed the problem with the tenure of Attorney General Gonzales. And the Goldsmith and Yoo books both reflect that the personality and philosophy of the President is somehow enormously important in shaping the legal advice that he receives. But is it obvious that Yoo and Bybee or Bates and Jackson were acting in a way that they thought of as contrary to the national interest but somehow important to a narrower interest of the President? Moreover, even if the President himself were understood to be the sole client of the Attorney General, would that mean that he deserved something less than thorough and independent legal analysis? Surely not. Whatever accounts for the quality of the various Attorneys General's analyses, it seems unlikely ultimately to prove to be that the lawyers did not take themselves to be working in the "national" interest. If that is so, and if we wish to avoid such bad lawyering in the future, we may need to find a better analysis than the (admittedly) appealing rhetorical trope that executive branch lawyers work, in the end, for the people and not one person.
C. Revisiting Federal Prosecution: Clinton's Impeachment and Independent Counsels
More recent history brought a new controversy over presidential control of federal prosecutions, especially in the wake of the Supreme Court's decision in Morrison v. Olson, (128) in which the court upheld the Independent Counsel statute in the face of a separation of powers challenge. (129) The lone dissenter on the court, Justice Scalia, wrote a fierce and highly formalist dissent in the case centering upon the "purely executive" character of criminal prosecution. (130) But in addition to the formalist arguments underlying Scalia's dissent, there were powerful practical arguments as well. As the late Justice Harry Blackmun's notes from the Court's Morrison conference indicate, Justice Scalia's thinking in the case was heavily influenced by the amicus brief of three former Attorneys General, led by Edward Levi. (131) Kevin Stack observes that an appendix to the Levi brief contained a famous speech that Justice Robert Jackson, then Attorney General, delivered to U.S. Attorneys in 1940. (132) Justice Scalia quotes two key paragraphs from that speech in his dissent, including the following:
If the prosecutor is obliged to choose his case, it follows that he can choose his defendants. Therein is the most dangerous power of the prosecutor: that he will pick people that he thinks he should get, rather than cases that need to be prosecuted. With the law books filled with a great assortment of crimes, a prosecutor stands a fair chance of finding at least a technical violation of some act on the part of almost anyone. In such a case, it is not a question of discovering the commission of a crime and then looking for the man who has committed it, it is a question of picking the man and then searching the law books, or putting investigators to work, to pin some offense on him. It is in this realm--in which the prosecutor picks some person whom he dislikes or desires to embarrass, or selects some group of unpopular persons and then looks for an offense, that the greatest danger of abuse of prosecuting power lies. It is here that law enforcement becomes personal, and the real crime becomes that of being unpopular with the predominant or governing group, being attached to the wrong political views, or being personally obnoxious to or in the way of the prosecutor himself. (133)
For Justice Scalia, the threat posed by an independent counsel removable only for good cause, setting her own budget and her own agenda, was precisely that such an office made far more likely just the sort of worry that Justice Jackson raised. A roving commission unchecked by electoral accountability, to investigate possible wrongdoing by a single person (or small group of targets), in a high profile case, with virtually no limits to staff and budget, gives enormous incentive to find something. And I think it fair to observe that Justice Jackson's "great assortment of crimes" has not decreased in number since 1940. The incentives, and the opportunity, for the independent counsel to run open-ended, free-wheeling, and costly (both to prosecute and against which to defend) operations seemed to Justice Scalia an enormous practical risk (underscoring, in his mind, the rationale for the more formalist separation of powers argument that is the core of his dissent):
How frightening it must be to have your own independent counsel and staff appointed, with nothing else to do but to investigate you until investigation is no longer worthwhile--with whether it is worthwhile not depending upon what such judgments usually hinge on, competing responsibilities. And to have that counsel and staff decide, with no basis for comparison, whether what you have done is bad enough, willful enough, and provable enough, to warrant an indictment. How admirable the constitutional system that provides the means to avoid such a distortion. And how unfortunate the judicial decision that has permitted it. (134)
One did not have to wait long to see whether Justice Scalia's fears were well founded. Morrison was decided in 1988. In 1994, shortly after President Bill Clinton signed a reauthorization of the independent counsel provision, former D.C. Circuit Judge Ken Starr was named independent counsel to investigate an Arkansas land deal involving President and Mrs. Clinton. (135) The land deal, and the independent counsel inquiry, went by the name of Whitewater.
That investigation would finally conclude--with no indictment of the President or the first lady--six years later. Along the way, it would grow to encompass such various ancillary matters as the billing records of Mrs. Clinton's former law firm in Arkansas, whether the Clintons had pressured allies not to give testimony against them, whether the apparent suicide of White House aide Vince Foster was actually a murder, and, most notoriously, whether President Clinton had in fact had sex "with that woman, Miss Lewinsky." (136)
The investigation cost taxpayers more than $50 million. It doubtless cost the Clintons and their associates millions more in legal fees. It led to President Clinton's impeachment by the House of Representatives, and acquittal by the Senate, on charges stemming entirely from events relating to the Presidents apparent marital indiscretions, having nothing whatsoever to do with the initial matters under inquiry. It is not difficult to see why many have concluded that the independent counsel investigations of the Clinton administration, of which Whitewater was but one of seven, were a vindication of Justice Scalia's dissent, in its practical, if not constitutional dimensions. (137) Thomas Merrill described the independent counsel position as "the worst of all possible worlds":
In other words, the Independent Counsel is an "in-and-outer." He or she has little or no incentive to abide by Justice Department policies, to adhere to traditional prosecutorial norms that have evolved over time, to conserve resources, to wrap-up business expeditiously, or to preserve principles like the confidentiality of grand jury proceedings. In effect then, the Ethics in Government Act manages to combine the worst features of political lawyers and tenured lawyers. It creates an office that lacks political accountability and is prone to inefficiencies--the problems associated with tenured lawyers. Yet it also creates an office having the short time horizons and lack of institutional memory associated with political lawyers. No wonder the Act has been such a disaster for our polity. (138)
In any event, Congress did not--and has not yet since--seen fit to reauthorize the independent counsel provisions. They lapsed in 1999.
For our purposes, the independent counsel episode may well be a cautionary tale in small about the danger--if improperly understood and implemented--of the appealing formulation that the Attorney General (or any executive branch lawyer, for that matter) is not the President's lawyer, but the people's lawyer. If the Attorney General either by statute or tradition (set aside, for the moment, whether any such statute would be constitutional) were to be seen as occupying some place of genuine "independence"--primarily, one supposes, from the President, but must we not also say from the Congress at least as much--the risk of mischief could be great indeed. If it is true that the Attorney General does not represent just one American (the one who appoints him), then it must also be true that the Attorney General does not represent just 535 Americans (the Congress), but all of them. And if this is so, then a genuinely independent Attorney General would presumably have to trust her own wiles, or perhaps the Gallup poll, or her favorite op-ed columnists to understand the will of the people. In any event, if the truism that the Attorney General of the United States represents the United States and not just its President is not to be a recipe for mischief large and small, then we must have more than a slogan to that effect. Some model showing how that client relationship, so worrisome to learned commentators such as Senator Leahy and Professor Gillers, can accommodate accountability will be essential.
IV. A FALSE DICHOTOMY: LAW VS. POLITICS
Earlier, I raised epistemic concerns about the growing tide of calls for greater independence in the legal advising function of the Attorney General. In this Part, I expand upon those concerns in the context of recent discussions of the politicization of the DOJ. I will explore some implications of a noteworthy recent historical study laying out two basic categories of Attorney General from history. (139) In the following Part, I will then offer my own model as a resolution between the false dichotomy set up by imagining a deep rift between law and politics, or idealizing a "depoliticized" DOJ.
Since Watergate, there has been voluminous literature focused on the nature of executive branch lawyering, and much has, at least in a general way, addressed the question of the independence of the Attorney General. Other inquiries have focused on the Office of Legal Counsel, on the Solicitor General, on agency lawyers, or on discreet tasks such as opinion writing or prosecution. A general account of all executive branch lawyering, or of what it means to be "the people's lawyer" is harder to come by. With these considerations in mind, I wish to more carefully consider Nancy Baker's contrasting "advocate" and "neutral" models, discussed above.
The distinguishing characteristics of the "advocate" Attorney General include a partisan background, close ties to the President, and loyalty. (140) It will come as no surprise that an exemplar of the "advocate" tradition is Robert Jackson, especially in light of his (then) much-maligned opinion in the bases-for-destroyers exchange. "As a member of the executive branch, Jackson had a broad view of the law as a means of accomplishing desirable policy ends; in this he was very much an Advocate law officer." (141) As evidence for this view, Baker notes that Jackson seemed willing even to bend on his known commitment to civil rights when asked, approving wiretaps on suspected Nazi agents and a "mail opening" operation to monitor correspondence with foreign countries. (142) Baker notes what nobody would doubt, that despite his clear tendency toward the "advocacy" model as she employs it, Jackson was both highly respected and highly skilled as a lawyer, and had been particularly lauded for his independence and courage as a Nuremburg prosecutor. (143)
Nevertheless, owing in large part to his service for President Roosevelt's wartime agenda, he exemplifies the advocate Attorney General as Baker would have us understand the term. Baker likewise examines as advocates Roger Taney, Caleb Cushing, Robert Kennedy, and Edwin Meese III. She also devotes a chapter to some more "dangerous" examples of advocate attorneys general, such as Mitchell Palmer (who countenanced "Red Raids" in the Wilson administration, but not at the behest of the President) and Harry Daughtry (who was credibly accused of cronyism and corruption, again, not obviously in service of the President or his agenda). These latter of Baker's examples are problematic, however, as examples of "advocacy" at least in terms of this study. That is, Palmer and Daughtry, reprehensible though their cases may be, represent a lawlessness not born of excessive dependence upon the President, but, if anything, excessive devotion to their own private agendas. Surely one can imagine lawlessness in service of the presidency, and I have suggested above that the torture memos come close if they do not cross the line. But these cases are best distinguished, it seems to me, in thinking about models of executive branch lawyering more generally.
A less problematic model, then, for the "advocacy" role is perhaps the last considered by Baker, that of Edwin Meese III. (144) Though, even in the case of Meese, there was the suggestion of scandal. But quite apart from that, it cannot be denied that the Meese Justice Department was active in advancing a particular vision of the law. For example, Baker notes that Meese outlined "policy goals" for the DOJ while he was there, touching upon affirmative action, criminal procedure, abortion, and school prayer. In each of these areas, the agenda outlined reflected policy priorities of President Reagan and shifts from the status quo of the law, even to the point of advocating the overturning of Roe v. Wade. Each of these areas, we may note, continues to occupy a lively space of debate in academia as well as an active life of litigation in the courts. But it is undeniable that the Meese DOJ was not shaping its views of the law merely as a prediction of where the courts were likely to go, but, significantly, following the agenda of the President who had just been elected. As we shall see presently, that model was a far different one from the ideal of the previous administration, which Baker fairly considers as exemplifying the "neutral" stance.
The marks of the "neutral" for Baker are "[p]rofessional eminence, nonpartisanship, and widely recognized integrity." (145) The most recent example examined by Baker is President Carter's first Attorney General, Griffin Bell. President Carter was of course elected in the aftermath of Watergate and of Gerald Ford's pardon of former President Richard Nixon. The Saturday Night Massacre, the perfidy of Nixon's Attorney General John Mitchell, and the general sense of corruption in Washington served as an important backdrop to Carter and Bell's approach to the job of Attorney General. This was so much the case that, at Bell's suggestion during the 1976 campaign, Carter had pledged in a debate with President Ford that he would push to create a statutorily independent Attorney General, with a five-year term that would not be coterminous with the President's. (146) Ultimately, the OLC advised Bell that such a proposal would not pass constitutional muster, and the idea was not pursued. (147)
After leaving office, Griffin Bell gave a speech that explains some of the episodes that mark him out as an Attorney General of rather extraordinary independence. (148) Considering a case where there had been tremendous political pressure on the President from the Mexican-American community to prosecute under federal civil rights laws a police officer who had received an arguably light five-year sentence under Texas state criminal law after shooting a twelve-year old Mexican boy in the head, Bell explained that he had misgivings over the potential double jeopardy problem posed by the case:
[F]rom the Civil Rights Department on up, everyone recommended that we not prosecute the policeman, because he had gotten a substantial sentence and we would probably run across a double jeopardy problem which the Supreme Court would not favor. But the President got very upset with me because I would not prosecute the policeman. He thought that the facts were so bad that we should prosecute it. He told me that I had embarrassed him by refusing to prosecute the case. While I was out of the country, some people in the White House staff asked Ben Civiletti, my deputy, to reconsider my position. Fortunately, Ben ruled in my favor. And that is where the matter ended. The President had a press conference and told the press a great thing. He said, "I appoint the attorney general. The prosecutorial discretion is vested in the attorney general. I can remove the attorney general, but I cannot tell him who [sic] to prosecute. I cannot tell him who [sic] not to prosecute. That is a great thing for this country." He said, "I can remove him. That is all I can do; and I am not prepared to remove the attorney general on account of this case." And that is the way the matter was left. (149)
In addition to that episode, another instructive one that Bell's speech addressed was, he said, the only occasion on which he was overruled by the President. The issue involved a proposal to have public employees working in parochial schools. Bell's advice was that such employees could only perform in "low level positions," and that they could not teach. The President informed Bell that he would overrule him:
So I wrote him and told him that, under the Constitution, he had every right to overrule me. But, I added, he did not have the right to control my ethical obligations under rule 11 of the Federal Rules of Civil Procedure, and I would have to make my own judgment about whether I could support his position in court. We resolved it by appearing and saying that we appeared at the direction of the Executive Department. (150)
Both episodes noted by Bell represent an extraordinary degree of independence as Bell (and Carter) understood the role of Attorney General. There is neither constitutional nor statutory authority (nor historical authority, as the Prakash materials demonstrate) for the proposition that the President's only power over prosecutions is "removal." It might be a very wise thing, most of the time or all of the time, for a President to defer to the Attorney General in such matters. But it is passing strange for a "neutral" Attorney General to commend a legally dubious explanation of what is, at best, a prudential judgment. The second episode likewise raises an issue that places the "neutral" expositor in an odd light. Bell must have concluded that he could make the Establishment Clause argument on behalf of the President without violating Rule 11. But that, apparently, was not enough--he had to let the court know, implicitly, that he was not offering an argument with which he agreed. Apparently, the independence of the Attorney General, for Bell, included a right (an obligation?) to editorialize to the court whenever his own opinion of the law differed from that of the President whom he served.
Some reflection on the text of Rule 11 might give us pause to consider the wisdom of Bell's approach. Bell would have been worried, one presumes, about FRCP 11 (b):
(b) Representations to the Court.
By presenting to the court (whether by signing, filing, submitting, or later advocating) a pleading, written motion, or other paper, an attorney or unrepresented party is certifying that to the best of the person's knowledge, information, and belief, formed after an inquiry reasonable under the circumstances--
(1) it is not being presented for any improper purpose, such as to harass or to cause unnecessary delay or needless increase the cost of litigation;
(2) the claims, defenses, and other legal contentions are therein warranted by existing law or by a nonfrivolous argument for the extension, modification, or reversal of existing law or for the establishment of new law;
(3) the allegations and other factual contentions have evidentiary support or, if specifically so identified, are likely to have evidentiary support after a reasonable opportunity for further investigation or discovery; and
(4) the denials of factual contentions are warranted on the evidence or, if specifically so identified, are reasonably based on a lack of information or belief. (151)
To add to the requirements of FRGP 11(b) that the Attorney General also agrees with the legal position she is taking--or ought to inform the court otherwise--is rather remarkable. It suggests that the President, who has been elected by the people, lacks the authority to argue--plausibly--for an extension, modification, reversal, or innovation in the law, or at least is not entitled to ask the Attorney General to do so free from editorial comment to the court. What possible conception of lawyering or independence or client service would warrant such an exalted role for the unelected--indeed, presidentially appointed--Attorney General? Here is Bell's account, as recounted by Baker:
We had some understanding. One was that the Justice Department would be operated on a nonpolitical basis. I think I understand how to do that because I was a judge for fourteen and a half years. Certainly, that was operated on a nonpolitical basis. I intend for the Justice Department to be operated within the strictures of its being a law department which would have nothing to do with politics. Of course, you touch politics because you are advising people, but it will not be a medium of politics; and it will not be used for political purposes. I would rather not be Attorney General than to have it turn out otherwise. (152)
The presuppositions of the Bell model are potentially breathtaking. They suggest that the executive branch ought not to attempt to push the law in any particular direction--that would apparently be politics--but instead must simply "follow" it--where "it" amounts (it would seem) to the Attorney General's prediction of where the current case law stands. Of course, it would be foolish--nay, wicked--to suggest that the executive branch should ever knowingly fail to follow the law. But it would be naive to suggest that the law is generally clear and foreordained, or that where it is relatively so, it is always or generally correct. The language of FRCP 11 itself suggests a mature understanding that the law is frequently, if not typically, in flux. There are plausible arguments that courts ought to hear for modification, reversal, or innovation. But on Bell's view, the executive branch ought only to be permitted, apparently, to make such arguments if the Attorney General, prescinding from politics, agrees. It turns out that the "independence" of the Attorney General, then, is an independence from democratic accountability and a roving commission to follow only her own conscience and call it "law." Christopher Langdell might have approved--if the Attorney General could be shown to be up to the job--but it is hard to see who else would.
To endorse the model of neutrality as proposed by Bell is to embrace a view of the law that is unsustainable. I am aware of no jurisprudence--including the strongest advocacy of textualism and originalism--that claims that all legal problems have some clear answer, the alternative to which would be implausible. If that is so, then even within the (purportedly) least "dynamic" visions of law, there remains room for reasonable, plausible disagreement about the best interpretation of law. The way interpretation develops over time is not uniform. Developments happen in the most obvious ways, legislation and litigation, but also and more often by practice, over time, informed by debate and deliberation. Surely our constitutional structure-which includes a Congress, a President, and a Supreme Court, but not an Attorney General--does not require that the executive branch refrain from playing a role in the development of law. And surely that structure does not require that the constraint be chiefly whatever the Attorney General thinks is best. I shall return to this consideration in proposing my own model below.
V. THE MODEL: A MEDIATED CLIENT RELATIONSHIP
A. One Master, but Many Mediators
To this point I have been skeptical of certain naive conceptions of independence and neutrality. In particular, I have expressed skepticism at the bare assertion that the Attorney General represents "the people" rather than the President--though I acknowledge that in some ways the statement is not only true but obviously so. I hope in this final Part of the Article to draw together some of the disparate observations and questions along the way to suggest a simple answer: the Attorney General's client is the American people, and that is a very complicated reality. By complicated I mean that it cannot be treated as a bromide or a punch line. To some extent, I fear that the understandable revulsion of commentators like Professor Gillers toward the arguments of the torture memos prompts them to note that the obvious answer to these questions is simply that the attorneys involved forgot that they serve the American people is question begging. Certainly one possibility, though, is that the attorneys involved thought that the American people would be better served by allowing the CIA wide latitude for aggressive interrogation. (153) If the Attorney General's job is "to serve the American people" but we can say nothing more, then we have said so little as to unleash more of the very problems that the formulation is doubtless meant to solve.
A great many voices in the legal profession (and outside of it) have argued persuasively that the American people were not served by the torture memos or the torture regime they abetted. But reasonable people have differed, and notable among them, if polls are to be believed, is a majority of the American people. (154) To say that the torture memos would have been avoided if only the relevant DOJ lawyers had remembered the American people is at least overly simplistic if not dangerous. Taken at their word, those lawyers intensely feared for the very safety of millions of the American people and were acting under severe constraints of time and psychological pressure in the wake of the September 11th attacks. While their work was deeply flawed, there seems little reason to doubt that it was undertaken in service--indeed generous service--to the American people. (155) In any event, this Article is not intended to settle the matter of the torture memos themselves--so it should be enough to note that the ways in which they violated the client model I am proposing need to be fleshed out more deeply than with a brandishing of the contrast between serving the President and serving the people.
My proposal is to move precisely beyond such false dichotomies as between "law and politics" or "serving the President rather than the people." These phrases, while reasonably motivated, ultimately run the risk of being empty, and amounting to no more than "do good and avoid evil"--being brandished after the fact, when the speaker thinks that the law has been violated. (156) As Trevor Morrison has observed, "politics has an entirely appropriate role in the executive branch. By politics, I mean discretionary considerations of policy and even ideology, as opposed to the mandatory (though often malleable) constraints of legal rules." (157) Many legal questions are unsettled, and the modes of their evolution are various, including legislation and litigation, but also and especially by custom or practice (formally or informally), which in turn are (one hopes) informed by debate and deliberation. If it is true that, "[i] n the vast majority of cases ... executive branch interpretation is not subjected to judicial review," (158) then, in the vast majority of cases practice and precedent will shape the meaning of the law. That is to say, where the executive branch must make a judgment about a legal matter that is unlikely to be reviewed by the courts, in such areas the legal landscape will be settled only to the extent that executive or Congress choose not to revisit the matter. Setting aside, then, the context of legislation as a mode of moving toward resolution of legal questions (and of course, legislation itself invariably opens new questions even as it aims to answer old ones), in areas of interpretation and litigation--the main work of executive branch lawyers--it is hard to see why, within the realm of the plausible, lawyers should have some obligation to set "politics" aside. Rather, accountability to elected officials would seem to be of the essence of the lawyer's work in a democratic polity. (159)
Such accountability is properly understood as accountability to the real but idealized client that is "the American people." We have seen how this is an "idealized" client, especially invoked in times of perceived mischief on the part of government lawyers. My suggestion that the client is complicated also suggests the model I have in mind for understanding the relationship. Some of the complications of representing "the American people" are obvious but still worth mentioning just so show why the Leahy formulation, however appealing and true, does not tell us enough. A lawyer, indeed the largest of law firms, cannot meet with this client on a Tuesday morning at 10:00 am. The client cannot agree to conference call, either. Nor does the client have a single mind about much of anything. Does this mean that the Leahy formulation is worthless? Not at all. But it requires that we delve deeper into understanding how it can be put into operation than simply employing words like "independence" or "depoliticization." I propose that the government lawyer must keep in mind the American people as client not as an abstraction, or as simply a reason to do good rather than evil (by the lawyer's own "independent" lights), but as a concrete reality concretely mediated.
The American people engage executive branch lawyers by various modes of mediation. The primary, but by no means exclusive, means of mediation for an executive branch lawyer is the President. The President is elected by the American people and vested with the leadership of the executive branch. For this reason, and by virtue of the constitutional structure of separate branches, the executive branch lawyer--whether career or politically appointed--must be responsive and accountable to the duly elected head of the executive branch.
But the mediation does not end with the presidency. The American people as client are represented in a crucial way, of course, by the law itself. In a democratic society, the will of the people is embodied in laws as well as in the agenda of elected leaders. The President has the duty of seeing that the law be faithfully executed, and therefore any lawyer properly serving the American people must always be faithful to the law as a controlling medium of instruction, if you will, from the client. Depending upon the lawyer in question (this Article is focused on the Attorney General, who serves as the paradigmatic case), the law as mediator will include a variety of concrete realities: the organic act of the agency by which the lawyer is employed; the statutes and case law relevant to any given question faced by the lawyer as such; the practices and interests of the institution created by the organic act. This list is not exhaustive, but it is illustrative.
For lawyers in the executive branch, and for the Attorney General, who must be confirmed by the Senate, the American people as client are also mediated by the Congress. Congress possesses, after all, considerable power in relation to the executive branch's interpretation of laws: it has the power to write and rewrite statutes; the power of the purse; powers in some cases to create agencies of relative independence to carry out certain functions, a power that remains intact after Free Enterprise Fund; it shares in the appointment power, giving it considerable leverage in the oversight process. (160) This list of powers, not intended to be exhaustive, demonstrates that Congress may push back with considerable force where it believes that the executive has gotten the law wrong, especially in that range of cases wherein the courts are unlikely to intervene. Such a struggle over interpretation is anticipated by--one might say, even, demanded by--our constitutional structure of government. The contention between the legislative and executive functions, even as to the interpretation of the laws themselves, is part of the mediation of that client relationship to which the Attorney General must attend. But where Congress is feckless, it does not fall to the Attorney General to substitute her own judgment, and the system may well fail at client representation where only the President has weighed in. But that is the system we have.
Having argued that the dichotomy between law and politics is a false one, I should emphasize here that the Attorney General, who has over time come to advise on policy as well on the law itself, ought to be as responsive to Congress as the circumstances warrant for an officer of the executive branch. Congress, especially through its powers of the purse and its sharing in the appointment power, can influence the priorities and strategies of the Department of Justice, and can create an atmosphere of greater or lesser cooperation between itself and the executive. While the removal power, discussed further below, as well as the original power to appoint will always give the President the upper hand in directing the Attorney General, the latter must still be responsive to the will of the client, the people, as expressed by the Congress. Arguments about democratic accountability do not run against that tradition, nor do arguments about epistemic uncertainty as to the law (for this political struggle over the law is just the proper consequence of that uncertainty). Finally, the structural arguments in support of the model fully assume congressional participation in shaping the legal and policy judgments of the Attorney General.
Notably and properly absent from the modes of client mediation are the following few items worthy of consideration here: vague notions of "the public interest"; the lawyer's own sense of what would be good for the country; the lawyer's own preferred reading of the law where a reasonable range of alternatives exists. Now, the Attorney General is primarily an advisor to the President (he is also of course the head of a massive agency, the chief federal prosecutor, the director of federal investigations (ultimately), and so forth) and as such the items on that list are not irrelevant to his job. Just as the private conscience, private sense of the law, and private sense of the good of the client are not irrelevant to the practice of a lawyer in private practice, so too these are going to be important parts of an Attorney General's advice to the President about legal matters. But it would not be correct to suggest that the Attorney General, qua lawyer, is entitled, let alone obliged, to give those aspects of his advising greater weight than other cabinet officials (as Senator Leahy's comments suggested). Moreover, and more to the point of this Article, it would be wrong to confuse those aspects of the Attorney General's independence with his representation of his "real" client, the American people. Certainly, the President and the Congress, by virtue of their direct ties to the electorate, represent the American people to a far greater degree than the appointed Attorney General, and so his personal judgment about the public good, while it may be of value to the President (or, for that matter, the Congress in its oversight capacity) as much as that of any other advisor, is not a component of the mediation of the client relationship argued for here.
With these basic outlines of the model of a mediated client relationship in place, and with the clarification of what such a model does not entail, it is necessary to examine the Supreme Court's recent decision in Free Enterprise Fund v. PCAOB, which helps to flesh out the proper extent of presidential control within the executive. Then, we will have enough of a framework in place briefly to revisit the recent controversies considered earlier to see whether the model can improve our understanding of what went wrong, if anything, in the episodes in question.
B. From Humphrey's Executor to Free Enterprise Fund: The President as Client-Mediator in Chief
As Professors Calabresi and Yoo note in their study of the historical practices of the first forty-three Presidents with respect to the exercise of control over the executive branch, times of crisis in the executive have occasionally resulted in strong assertions of congressional power to constrain the President, such as the Tenure of Office Act of 1867 and the Ethics in Government Act (creating, inter alia, the Independent Counsel) in the wake of Watergate. (161) So too in our own time, as noted above, there is a fairly steady rhythm of accusations by the party out of power to the effect that the current administration is failing to honor the necessary independence of the Attorney General. But while cases such as Humphrey's Executor and Morrison v. Olson preclude a formalistic, pure theory of unitariness in the executive, Free Enterprise Fund teaches that those cases operate at the margins of executive practice, and the Court will continue to play a role in policing and protecting the power of the President to undertake his constitutional duty to see that the laws be faithfully executed. A number of early commentators noted the modesty of the decision, and seized especially on its unwillingness to reexamine precedents such as Humphrey's Executor or morrison. (162) This view, while entirely plausible, may well be mistaken.
What is significant about Free Enterprise Fund is its apparent willingness to cabin those precedents at all, and its awkward embrace, qualified by the reminder that the Court neither was asked to nor chose to revisit them. One might have thought, for instance, that once the President's control over the SEC was limited by for-cause removal, it would make little sense to suggest that he could lose much more control over inferior officers within the department through the "double for-cause" limitation. Yet a majority of the Court thought it significant to emphasize that Congress's power to encroach upon presidential control through limitations on the removal power of the President or of the heads of departments is not without limit, and that such questions go to the heart of the understanding of the executive: "Without such power, the President could not be held fully accountable for discharging his own responsibilities; the buck would stop somewhere else." (163) The logic of the Court's opinion does indeed raise the question of the legitimacy of the presumed for-cause removal limitation of the SEC itself. While nothing in the opinion suggests an eagerness on the part of the Court to take up that question, neither does the opinion preclude such reexamination down the line. (164) Pointedly, the Court emphasizes that it accepts the assumption of the parties that the SEC commissioners may only be removable for cause, and that it has not been invited by the parties to reexamine its precedents, and that it declines to do so. But the language is hardly a ringing endorsement of the modest encroachments on the removal power hitherto approved, and the logic fairly calls them into question, or at least suggests that they remain at the margins of our understanding of the executive.
More importantly, the reasoning of the Court in Free Enterprise Fund seems to ratify what even Justice Brandeis recognized in dissent in the previous high-water mark of unitary executive theory in the Supreme Court, Myers v. United States: "Power to remove, as well as to suspend, a high political officer, might conceivably be deemed indispensable to democratic government and, hence, inherent in the President." (165) Justice Brandeis was emphasizing that no such power need be recognized regarding the removal of inferior officers, but the logic of his contrast seems to be very much the logic of the Court's latest holding in this sphere. Justice Brandeis may well overstate things as a matter of democratic political theory, of course. Many states, for example, have attorneys general elected directly by the people, rather than appointed by the governor, and yet survive as democracies. Indeed, some have suggested that such a model might be fruitfully adopted in the federal Constitution. (166) But in the constitutional and statutory scheme we currently have, the Attorney General is indeed a high officer the scope and centrality of whose office has only increased over time, and his removability at will (and the level of presidential control implicit in that) does indeed seem to be a requirement of democratic accountability.
Free Enterprise Fund comes as a timely reminder, then, in a time of multiple and ongoing political controversies over the Attorney General and the DOJ, of the fundamental structural principles of our constitutional order. The reminder is in the form of a rejoinder to Senator Leahy's bromide that the Attorney General does not serve one American but all of the American people: the executive branch includes only one officer elected by all the American people, and he is the chief spokesman for the Attorney General's client. As I have noted above, the client relationship is not one in which the President is the sole representative of the people--there are other mediating entities that set boundaries upon what the President may claim, request, or instruct. But where the boundaries are not clear, where the law is not yet fixed, it falls to the judgment of the one elected official in the chain of command, and not to his subordinate, to exercise the responsibility for decision making.
The technical expertise and professional judgment of the Attorney General, it is to be hoped, will be eminent and valuable assets for the client. Presumably, no matter how expert in the law a given President may be, the institutional perspective and responsibilities of the Attorney General will ensure that that office remains an indispensable aid to the President in ensuring the rule of law. To this extent, independence of mind, of judgment, and of perspective are all necessary qualities in an Attorney General, and the importance of such should not be gainsaid. But the same can and must be said of any of the top officials of the executive branch. We should not hope for a supine dependence upon the President's own whims from his top military commanders, for instance. Nobody doubts that both the civilian and uniformed officers in the military chain of command owe the President candor, disagreement when necessary, and even the willingness to resign in extreme circumstance if professional judgment and integrity require it. But nobody doubts, when all is said and done, that the Commander in Chief retains the authority to direct and control. The same should of course be said of the Secretary of State, the Secretary of the Interior, and so forth. All of these officials serve not one American, but all Americans. Yet they do so within a constitutional structure that vests the executive power, and rests democratic accountability, with the President. The same must be true for the Attorney General. As the OLC put it when asked by Griffin Bell to consider the constitutionality of President Carter's public suggestion to make an "independent agency" of the DOJ:
The Attorney General is the chief law enforcement officer of the United States. He acts for the President to ensure that the President's constitutional responsibility to enforce the laws is fulfilled. To limit a President in his choice of the officer to carry out this function or to restrict the President's power to remove him would impair the President's ability to execute the laws. Indeed, the President must be held accountable for the actions of the executive branch; to accomplish this he must be free to establish policy and define priorities. Because laws are not self-executing, their enforcement obviously cannot be separated from policy considerations. The Constitution contemplates that the Attorney General should be subject to policy direction from the President. (167)
I hope, however, to have been clear that the Attorney General is of course obliged to help the client's chief representative, the President, to be aware of the full range of the client's interests as expressed by other modes of mediation. A skilled Attorney General must able to identify where the President's will conflicts with law, runs counter to the public interest, or seems otherwise imprudent. The Attorney General indeed owes the President the Attorney General's own "best view" of the law and of legal policy. But no Attorney General, either by training, experience, or oath of office, is gifted with legal, moral, or political omniscience. The best view of the law, especially when it comes to hard questions, is frequently far from pellucid. The nation's top jurists regularly divide 5-4 in settling disputes--frequently with both the majority and the dissent averring that the question is not difficult. Thus, both the fact of human fallibility and the structure of democratic accountability in our constitutional order demand humility of the Attorney General where there is legal room for the executive to maneuver. Something like the equivalent of Rule 11 and an Attorney General's own sense of judgment that perhaps a question is so important that it is worth threatening resignation over--great leverage for any cabinet official--ought to serve as boundaries in the relationship between the President and the Attorney General, but the generalized sense that the Attorney General is somehow situated by training or office better to understand the needs of the client where the law is not fixed is not supportable by logic, tradition, or the Constitution's text and structure.
In proposing a model of a mediated client relationship, I hope to have improved upon the usual approach to considering the conduct of the Attorney General, which is typically modeled, as in the case of Baker's study, along the lines of the Attorney General's attitude toward the President, rather than upon the question "who speaks for the client." It is both worthwhile and necessary to attend to the structure of the system and the ways in which the client's wishes and interests may be identified to understand how the Office of the Attorney General ought to be conducted. Of course, the character and attitude of the occupant will always be crucial in determining the quality of the service rendered. But those attitudes must be informed by a clear sense of what our constitutional order envisions, rather than by a confused or exalted sense of the Attorney General, qua lawyer, as somehow more exalted, more independent, less political (in the sense discussed above) than other cabinet officials. We should expect of all high officers that they serve the nation--and the nation's President-with candor, courage, and independence of mind. But we should also expect that they respect to the maximum extent possible the role of the President as the elected and accountable head of the executive branch, duly entitled, and indeed required, to make the decisions necessary to see to the faithful execution of the law.
C. The Model Applied: Revisiting the Recent Controversies
1. The Torture Memos
Beginning with the torture memos, some of what I have to say will already be clear. There is strong evidence that the chief author, John Yoo, believed in the arguments that he made--especially the constitutional arguments about the powers of the President in his capacity as Commander in Chief. There is also evidence (slightly more ambiguous) that the lawyers involved believed what they were doing was for the good of the American people, perhaps as embodied by the will of the President, but in any event in union with it (as opposed to serving the President's will over the American people's). From the point of view of the model I propose, we can concretely identify the ways in which the client was ill-served by these memos. To wit, the body of written laws and case law were very badly attended to, in two ways. The statutory arguments, as has been established especially well by Jeremy Waldron (among many), were in some instances frivolous and in any event unavailing. Perhaps this was motivated by the genuine belief of John Yoo in his constitutional position that the Commander in Chief has powers at war upon which such statutes may not encroach. (168) If that were so, Yoo and his superiors still had an obligation to get the statutory analysis correct, and in this instance appear to have failed. But if the failure was made in good faith, then it was a failure of lawyering for routine reasons. The President was entitled (if not obliged) to ask for clarity in establishing a permissible and effective regime of interrogation. Taking that question and its motivation seriously was a necessary part of genuine client service here. So, too, to be sure, was careful attention to the body of laws governing the field, and the failures in that regard seem to me well established. But, unless they were made in bad faith, they were simply a matter of bad lawyering in service of the right project. And it is important to note that the other mediating actor, the Congress, was at best silent and in the dark on these questions at the time, and at worst silent and complicit. The public record is incomplete, but it is bereft of evidence that Congress was sending messages to the executive that the interrogation regime was either forbidden or wrong. (169)
If the errors were made in bad faith, it is hard to see how any model of executive branch lawyering would have prevented them. Bad actors will act badly even if reminded that they are actually not supposed to serve the President as if his will were law. But if only a gauzy notion of serving the people rather than the President were to be the best model of client service for an executive branch lawyer, it would be easy to imagine that the torture memos might have been even likelier to be produced by the lawyers in question. For while John Yoo has at times said that the policy choices were for the President and the CIA when it came to interrogations, and he was only answering legal questions, (170) at other times he has remarked, "I think it would be very difficult to be a Kantian and to have any responsibility in the government." (171) In any event, there is no need to speculate about Yoo in particular--surely it is not difficult to see the dangers inherent in authorizing lawyers to substitute their own moral judgments for those embodied in laws like those absolutely prohibiting torture. It would be nice to be able to say that my model (or some model) of being a government lawyer would, of itself, prevent either bad character or bad lawyering from producing the torture memos, but I do not believe it. Whether Yoo, Bybee, and ultimately Ashcroft acted in good faith in these matters is for God to judge, unless some other tribunal intervenes. But it would be a mistake to assume, as some have, that because they are smart lawyers then only bad character can account for the bad lawyering they did here. This is why it is possible to imagine that a smart lawyer, taking account of the various modes of mediation and aiming to do so with integrity, might still go wrong. The reasons for that might be the very ones suggested by Yoo, Bybee, and their critics: time pressure, a lawyer's own ideological presuppositions, the psychological pressure of working on questions of national security in a time of widespread, intense fear. In any event, no model will be foolproof in eliminating the humanity from human institutions.
But beyond the mediating role played by statutes and case law, the client interests of the American people might also have been represented here by a wider array of government actors than the OLC (contrary to its own best practices) appears to have consulted. This would include both other executive branch agencies with relevant expertise and interests, especially the Department of State, and the Congress. As Goldsmith is at pains to show in his treatment of the historical examples of Presidents and attorneys general facing similar crises, the client is best represented when all conceivable appropriate modes of mediation in that client relationship are attended to. (172)
2. The U.S. Attorney Firings
Alberto Gonzales's decision to replace nine U.S. attorneys for reasons that remain elusive in some cases, that look bad in others, and that in any event cannot be squared with the initial public justification that the removals had come pursuant to performance reviews present a different set of problems from those posed by the torture memo episode. As an initial matter, this does appear to be an episode where at least some of the firings came at the behest of political masters for political motives (the rewarding of allies), which might have proven politically acceptable (recall the DOJ investigation's formulation that U.S. Attorneys may be removed for any reason or no reason, but not an illegitimate reason) if it had been done in an above board fashion. But the dodgy testimony and, at best, arbitrary fashion in which the selections were made suggests that important client duties in this instance were either downplayed or ignored, as I have modeled things. If I am correct that the law itself as mediator includes not just statutes, regulations, and case law, but also and very significantly the (more or less) settled practices of agencies and other actors, then it seems quite clear that little attention was paid to these either by the Attorney General or his subordinates. The DOJ investigation makes clear that the manner and timing of these firings were unprecedented. And the at best incompetent (if not dishonest) explanations rolled out once the firings became known to the public and especially the Congress make crystal clear that neither the Attorney General nor his staff had given adequate reflection to the precedent (or lack thereof) for their actions in weighing whether and how to undertake these firings. Note that presidential direction about the personnel in question is undoubtedly appropriate--but the mediated client relationship requires due respect for the client--appropriate transparency, honest communication, and the like are a part of the Attorney General's duties in such cases.
Unlike in the case of the torture memos--where the countervailing statutes and treaties were at least systematically attended to, however wrongly--here there appears to have been a process that was entirely responsive to only one aspect of the mediated client relationship, the President (or at least the executive office of the President). The fact that the subject matter of the decisions to be taken dealt with federal prosecutors makes that failure to attend to the variety of actors even more damning. One would also expect that, where no impropriety was at work, and where a lively respect for the various forms of client mediation characterized the Attorney General's work, the unprecedented nature of the actions contemplated would have spurred some consultation--formal or informal--with the Senate. Here a return to then-Attorney General Jackson's 1940 speech is helpful:
Because of this immense power to strike at citizens, not with mere individual strength, but with all the force of government itself, the post of federal district attorney from the very beginning has been safeguarded [sic] by presidential appointment, requiring confirmation of the Senate of the United States. You are thus required to win an expression of confidence in your character by both the legislative and the executive branches of the government before assuming the responsibilities of a federal prosecutor. Your responsibility in your several districts for law enforcement and for its methods cannot be wholly surrendered to Washington, and ought not to be assumed by a centralized department of justice. (173)
Jackson's remarks reflect very well the sense in which the client of the executive branch lawyer relates to the lawyer by various modes of mediation, which include the DOJ (and the White House, presumably) but do not begin and end there. How dangerous and unfortunate, in dealing with offices so reliant upon the confidence of the citizenry, to have attended so little to history and precedent, and to the Senate as well, each of which represented important mediating factors that constitute, in our system, client directions to the Attorney General just as much as a call from the President.
The model I have proposed, then, adds necessary substance and structure to Senator Leahy's justified worries about how this particular episode reflected very poorly indeed on Attorney General Gonzales's conception of his relationship to his client, the American people, and how that relationship is instantiated in reality.
3. Did Eric Holder Inappropriately "Forum Shop" in the D.C. Voting Rights Matter?
Once we have come to understand that the dichotomy between law and politics represents a false choice in government lawyering, the controversy over Attorney General Holder's consultation of the Solicitor General's office in addition to the OLC in considering how to advise the President about the D.C. Voting Rights bill appears to be without foundation. The OLC performs a number of enormously important functions in helping the Attorney General to acquit his role as the President's omnibus legal adviser. In that role, it traditionally tends to take a conservative approach to understanding the "best view" of the law (with the caveat, noted by many commentators, that it has, appropriately, traditionally taken a robust view of executive branch prerogatives in separation of powers matters). At the same time, while OLC's determinations of law are, as a matter of executive branch policy, binding law within the executive, that power flows, as it were, downhill; the Attorney General and the President are not bound by the OLC, to whom certain of the Attorney Generals duties are delegated. Nor is the Attorney General, when advising the President, bound only to offer the President the narrowest range of views on a particular legal question.
In private practice, an attorney is well within her rights to advise a client that there are a range of legitimate views on a given legal question. While some approaches may seem likelier than others to be vindicated by a given tribunal, it is not a breach of ethics for an attorney to be willing to take a non-frivolous (if long-shot) position on behalf of a client. Where there are open questions involving the government as a party, or where there are open questions of the constitutionality of legislation, there is no constitutional, statutory, or other obligation for the President only to be offered the most conservative approach by the Attorney General.
In the case of the D.C. Voting Rights Act, I think the bill is clearly unconstitutional (not that anyone has asked!). But there are competent lawyers who take a different view, including lawyers whose political commitments might suggest they would oppose the law. (174) Thus, even if the OLC's standing position is the right one here, it must be conceded that the bill can be supported by non-frivolous arguments that would not come close to violating FRCP Rule 11 (to return to Griffin Bell's formulation of things). If that is so, and if the bill accords with the President's policy preferences, then it seems to me entirely appropriate that a range of views on its constitutionality ought to be presented for the President's consideration. After all, the question only becomes significant for the President if Congress itself has already passed the bill--an important communication from the client to any government lawyer--and if the Attorney General in good faith believes that the law may be defended, then advising the President so seems faithful to the model of client service I have laid out here. Consulting with the OLC, with the Solicitor General, and relying upon the Attorney General's own legal judgment in the face of congressional action seems to me to be a process that has attended to the various mediated forms of client to lawyer communication.
The American people are well served by the Attorney General as their lawyer when he can, in good faith, defend the constitutionality of a bill passed by both houses of Congress and desired by the elected President, even if the OLC happens to believe the bill is unconstitutional. Surely the OLC is not an infallible prognosticator of outcomes at the Supreme Court, and the country would not be well served if the executive branch tied its own hands (by following only the most cautious sort of legal advice) in the mix of debate, deliberation, evolution of practice, and passage of legislation that work to shape and reshape the law.
4. The KSM Trial: An Attorney General's Decision to Make?
Finally, we come to the very complicated situation presented when the Attorney General's advice has been asked in the matter of the best manner of prosecuting alleged terrorists such as Khalid Sheikh Mohammed. For our purposes, let us take it as a given that both the federal courts and the military commissions represent legitimate tribunals available for such trials. How is the client relationship mediated in such a case? What are the factors necessary to serve the client best? Clearly, high-profile trials in the ongoing war against al Qaeda and allied terrorists are not run of the mill prosecutions. The Attorney General's own discussion of these cases in public suggests that the decision implicates a variety of client interests that are rarely involved in other kinds of prosecutions: America's image in the world; the potential of terrorist attacks upon the location of the trial itself; the thorny question of whether to free an enemy combatant properly captured in battle after an acquittal before whatever tribunal, to name but three. Answers to these questions will not be found by the study of statutes or case law. They almost certainly require mediation of the client's interests not merely by lawyers in the DOJ but by national security experts, experts in domestic security, and the people's elected representatives in Congress. Given that, it is hard to see how the Obama administration's efforts to portray this decision, in the first instance, as one properly committed to the discretion of the Attorney General, rather than to the President himself, passes muster.
As an historical matter, the Prakash materials strongly suggest an appropriate role for the elected executive in managing certain exercises of prosecutorial discretion. It would be hard to imagine a case where that was more appropriate than here, nor one where something like President Carter's claim that his only recourse in disagreeing with the Attorney General about bringing a prosecution was to demand his resignation. In the case of the KSM trial, published reports suggest that the Attorney General's initial decision to proceed with a trial in federal court in Manhattan was made without consultation of the mayor of the city of New York or of its police chief. The Attorney General did state that he consulted closely with the defense secretary in determining the proper venue for prosecuting KSM and other similarly situated detainees, but the subsequent uproar from Congress and from the officials of the City of New York, as well as the apparent decision by the Attorney General to revisit his decision and to seek the "help" of the President in making it, suggest that the accountability to the client--the American people--was not adequately attended to in this case in the first go round.
My views on this episode, and more generally, stand in distinct contrast to those laid out by Peter Strauss in his recent, typically careful study of whether the President is best viewed as "overseer" or "decider." (175) Strauss considers a variety of cases from history illustrative of the relationship between the Chief Executive and other members of the executive branch to whom Congress has delegated significant authority. He gives prominent attention, for example, to Andrew Jackson's well-known struggle to find a Secretary of the Treasury who would remove the federal funds from the Bank of the United States. (176) Until Roger Taney agreed to execute the President's wishes, they were frustrated by two prior treasury secretaries who refused and whom Jackson removed. In the modern era, he considers a Ninth Circuit decision in Portland Audubon Society v. Endangered Species Committee, which held that the President was subject to the APA's prohibition on ex parte communications with respect to the decisions of the Endangered Species Committee. (177) Strauss sees in these and in similar episodes support for an understanding that the ultimate decision authority rests with the Secretary, rather than with the President, whose job is not to execute the laws but to oversee their faithful execution. As he concludes: "Oversight, and not decision, is his responsibility." (178) This principle governs, for Strauss, "the mixed questions of law and politics that are the everyday focus of administrative law and of judicial review." (179)
My own conclusion from the Taney episode (and others like it) is something like the opposite: the power to remove is more or less the power to decide. (180) Consider again the Court's decision this past term in Free Enterprise Fund. There, the Court clearly considered the removal power as crucial to control. Rejecting the notion that the SEC's control over the PCAOB's budget, its power to amend the Board's sanctions, or the like, were enough to cure any constitutional defect of the removal provisions at stake in the case, the Court noted that "altering the budget or powers of an agency as a whole is a problematic way to control an inferior officer. The Commission cannot wield a free hand to supervise individual members if it must destroy the Board in order to fix it." (181)
Certainly, in the ordinary run of events, given the vast array of executive branch decision making, the President is indeed an overseer (at best) rather than a decider. But when, for whatever reason in a given case, a President's attention is drawn to an issue where an executive officer must exercise discretion, the structural reality is that the President gets to decide. That such an officer--like Taney's predecessors-can raise the transaction cost for the President does not, at the end of the day, alter the reality that Jackson, and not Taney's predecessors, decided the issue. After all, many actors in our system can raise the transaction cost of a presidential decision--talk radio hosts, former Presidents, even former vice-presidential nominees with Facebook pages--but we would not say that the ability to exert political pressure on the President constitutes decision-making authority. Strauss does not contest that Jackson was perfectly within his rights to remove the Secretary of the Treasury for refusing to carry out his will, but sees in the very fact that removal was necessary a distinction between the authority to take the decision (belonging, on Strauss's view, to the Secretary) and the removal power. (182) I think this distinction cannot bear the weight that Strauss places upon it. The fact that the President is not empowered to undertake the actions of executive branch officials in their stead complicates, but does not displace, the President's ultimate responsibility for the substance of the decision, and not merely the process that produced it.
I should note, however, that my view in no way suggests that, for example, Taney's predecessors were under an obligation blindly to follow the President's will. If an executive branch official determines that the law she is executing requires a judgment different from the President's, it is the law, and not the presidential will, that binds. As Strauss puts it when considering the line between politics and law: "Save in some inconceivable cyber-age, we could never have a government purely of laws; and we surely do not wish a government just of men." (183) Where the professional integrity of an executive branch officer leads her to defy the President's will even to the point of insisting upon resignation or removal as the only alternative, then such insistence is entirely appropriate. Whether considering the Attorney General, the Secretary of the Treasury, or any other official answerable to the President, our system is not such as to ask that humans behave as automata. Structurally, however, political accountability in the executive rests with the one individual with the power to remove--Taney's predecessors could exact a price from the President for sticking by his decision, but ultimately they could not thwart it. But Strauss, to my mind, undermines his thesis by conceding that most decisions we are interested in when considering this problem of who decides are mixed questions of politics and law. However complicated the political reality of exercising the power may be, the power to remove in many cases implies the power to decide.
Recent controversies over two presidential administrations have raised afresh perennial questions concerning the proper degree of independence to be maintained by the Attorney General of the United States. Most such discussion has erred in two significant ways: first, by attending too little to the question of who really is the Attorney General's client; and, second, by undervaluing the role that the executive branch may and ought to play in shaping, and not merely discovering, the meaning of the law. In particular, the right of the people, through their elected President, to enjoy the fruits of the vigorous contestation of law and policy among the three branches of government envisioned by the Framers, is threatened by a notion that the Attorney General, by virtue of his training or his oath, ought to exercise a uniquely independent role among high officers of the United States. I have argued for a model that, while assuming a high degree of professional integrity and independence on the part of the Attorney General, and especially emphasizing that the President is but one among several mediators of the Attorney General's client, the American people, assumes that the Attorney General is meant to serve an agenda that is communicated strongly by the executive who nominates and can remove her. In arguing for this model, I have especially noted the weaknesses of arguments that the Attorney General has special access to a "best view of the law" that is somehow binding on the President over against even other plausible views of the law. While prudential considerations will frequently counsel that the President defer to the Attorney General's views, and while those views will often hew to a court-centered, conservative (in the non-ideological sense of the term) view of the law, such prudential considerations ought not to be confused with constitutional or otherwise legally binding authority. Arguments to the contrary threaten dangerously to increase the power of the unelected Attorney General to impose debatable legal and political preferences over against the will of the people as represented by the elected and accountable head of the executive branch. If there is promise in the model I propose, much work remains to be done, of course, in exploring and filling out the modes of mediation outside of the executive branch, in particular the role of Congress, in order to build a more complete understanding of the Attorney General's proper client relationship.
(1) For an excellent recent discussion of the ways in which the legal, moral, and political became intertwined in the case of the torture memos, see generally Jeremy Waldron, Torture and Positive Law: Jurisprudence for the White House, 105 COLUM. L. REV. 1681 (2005).
(2) For a brief, though ultimately inconclusive, survey of views of the Attorney General's client relationship, including the suggestion that the notion of the American people as the client is "incoherent as a guideline," see HAROLD H. BRUFF, BAD ADVICE, 73-74 (2009). See also Robert P. Lawry, Who is the Client of the Federal Government Lawyer? An Analysis of the Wrong Question, 37 FED. B.J. 61 (1978) (arguing that, for questions of determining the ethical obligations of government lawyers, at least, no answer to the question I address here will be definitive enough to settle key ethical questions). I intend to address the ethical obligations of government lawyers, about which there is a rich and recently growing literature, in a forthcoming article.
(3) 130 S. Ct. 3138 (2010). For an excellent general discussion of the issues raised, those resolved, and those left for another day in Free Enterprise Fund, see Peter L. Strauss, On the Difficulties of Generalization--PCAOB in the Footsteps of Myers, Humphrey's Executor, Morrison and Freytag (Columbia Law Sch. Pub. Law & Legal Theory Working Paper Grp., Paper No. 10-253, 2010). Strauss emphasizes the mildness of the Court's practical line-drawing in the case, while also noting that the case creates significant unresolved issues with respect to presidential control by its more sweeping general discussion of first principles.
(4) Free Enterprise Fund, 130 S. Ct. at 3143. The "dual for-cause" language refers to the fact that the members of the PCAOB were, under the statutory provision at issue, only able to be removed by the SEC commissioners for cause, while in turn the SEC commissioners themselves may only be removed by the President for cause. (The Court assumed this latter point without deciding it.)
(5) Id. at 3155 (citations omitted).
(6) Id. at 3156.
(7) Almost invariably, contemporary discussions of the separation of powers in general and of the executive branch in particular draw upon discussions of the theory of the "unitary executive" branch. Much of the public discourse surrounding this issue has been confused and confusing, alternating between discussions of whether the executive is united under the control of the President and discussions of whether the executive branch is somehow superior to the other branches. Then-Judge Samuel Alito, in an exchange with Senator Edward M. Kennedy during the former's confirmation hearings as a nominee for the Supreme Court, perspicuously addressed such confusion:
I think it is important to draw a distinction between two very different ideas. One is the scope of executive power. Often Presidents--or occasionally Presidents--have asserted inherent executive powers not set out in the Constitution .... The second question is: When you have the power that is within the prerogative of the executive, who controls the executive? Those are separate questions. The issue of, to my mind, the concept of the unitary executive, does not have to do with the scope of executive power. It has to do with who within the executive branch controls the exercise of executive power. The theory is the Constitution says the executive power is conferred on the President.
U.S. Senate Judiciary Committee Hearing on Judge Samuel Alito's Nomination to the Supreme Court, WASH. POST (Jan. 10, 2006, 12:49 PM), http://www.washingtonpost.com/wpdyn/content/article/2006/01/10/ AR2006011000781.html. Judge Alito's reply to Senator Kennedy crisply captures the approach to the unitary executive theory adopted in this Article. That is, the problem to be treated is one of the nature of the executive branch ad intra, as it were, not of its relations ad extra. To the extent this Article discusses the executive as unitary, then, it should not be taken to be endorsing anything like a view of executive supremacy over Congress or the courts, executive exemption from the rule of law, or any of the other parade of horribles frequently-wrongly--associated with the notion of a unitary executive, at least as I shall employ the term. Indeed, far from endorsing such a notion of executive supremacy, this Article presumes that a properly unitary executive requires more robust engagement by the other branches properly exercising their own powers. The Court in Free Enterprise Fund declined to challenge the basic presuppositions that frequently bother the strongest advocates of a unitary executive--the ability of Congress, for example, to place restrictions on the President's removal power--yet, even so, it ratified the importance of Presidential control over the key officers of the executive branch as an essential part of our constitutional structure. See generally Steven G. Calabresi & Kevin H. Rhodes, The Structural Constitution: Unitary Executive, Plural Judiciary, 105 HARV. L. REV. 1153 (1992) (positing that "only a holistic approach" will provide a satisfactory understanding of Congress's relationship to the executive).
(8) THE FEDERALIST No. 51, at 347-53 (James Madison) (Bantam Classic 1982).
(9) See generally JACK GOLDSMITH, THE TERROR PRESIDENCY (2009) (outlining the role of the DOJ in dealing with terrorist groups).
(10) See, e.g., Edward Whelan, Look Who's Politicizing Justice Now, WASH. POST, Apr. 5, 2009, at B3 (discussing Attorney General Holder's decision to decline the opinion of OLC on the legality of a District of Columbia Voting Rights Bill pending in Congress).
(11) Treating the Attorney General as the unitary executive of the DOJ is certainly a simplification--the DOJ's enormous size and bureaucratic complexity inevitably results in significant decision making occurring beneath the attention of the Attorney General. Even in the traditional role of the Attorney General as the chief legal advisor of the executive branch, the picture has been complicated at least by the rise of the OLC as the Attorney General's own chief legal advisor. For a clear discussion of these issues, see generally Douglas W. Kmiec, OLC'S Opinion Writing Function: The Legal Adhesive for a Unitary Executive, 15 CARDOZO L. REV. 337 (1993). For arguments that the Solicitor General in particular should be seen as having an independence from the Attorney General, see Drew S. Days, In Search of the Solicitor General's Clients: A Drama with Many Characters, 83 KY. LJ. 485 (1995). Finally, a more general discussion of the historical complications attendant to considering the Attorney General as a unitary executive can be found in Griffin B. Bell, The Attorney General." The Federal Government's Chief Lawyer and Chief Litigator, or One Among Many?, 46 FORDHAM L. REV. 1049 (1978). Most of the reasons adduced for independence from the Attorney General tend to be pragmatic in nature--especially that a certain independence is inevitable given the enormity of the Attorney General's responsibilities, and also that pragmatic reasons (credibility as a repeat player before the Court, for instance) strongly encourage a distance from day to day politics. These considerations do not seem to me significant for my argument, given that the Attorney General does ultimately retain responsibility for the Solicitor General's decisions. See, e.g., Days, supra, at 502-03 (discussing role of Attorney General in informing Congress of the Solicitor General's decision to decline to defend constitutionality of a law).
(12) I will not argue that there is an Archimedean point in the history of the office or of the text of the Constitution that rules in or rules out a particular position for the functioning of today's DOJ. History rarely works that way in the field of separation of powers, or more generally. At the same time, history can provide an invaluable negative check against certain kinds of "essentialism" in argument: if history shows that the republic got along for a substantial period of time while rejecting x as a feature of its constitutional structure, at a minimum we may say that x is not essential to that structure, even if the feature might be arguably desirable as improvement upon the status quo. See generally Susan Low Bloch, The Early Role of the Attorney General in Our Constitutional Scheme: In the Beginning There Was Pragmatism, 1989 DUKE L.J. 561 (analyzing the history of the office of the Attorney General).
(13) I would note that, if this is so, then my treatment, partially for purposes of simplification, of the Attorney General as a unitary executive should be less potentially problematic.
(14) See Philip Shenon, Attorney General Choice Treads Careful Line at Senate Hearing, N.Y. TIMES, Oct. 17, 2007, at A1.
(15) Senate Judiciary Committee Hearing for Nomination of Judge Mukasey as Attorney General, WASH. Post (Oct. 17, 2007), http://www.washingtonpost.com/wp-srv/politics/ documents/attorney_general_hearing_101707.html.
(17) Excerpts from Interview with Nixon About Domestic Effects of Indochina War, N.Y. TIMES, May 20, 1977, at A16. Indeed, the lawlessness of the Nixon Administration is so well known and, resulting as it did in criminal convictions, that it will not be a focus of this Article. As Justice Stevens suggested in his unusual concurrence in Burnham v. Superior Court, 495 U.S. 604, 640 (1990), sometimes easy cases can make bad law. In any event, the trauma of the Nixon administration spawned a great deal of commentary on the office of the Attorney General. See, e.g., Removing Politics from the Administration of Justice, Hearings on S. 2803 and S. 2978 Before the Subcommittee on Separation of Powers, Senate Committee on the Judiciary, 93d Cong. 18 (1974) (discussing the inherently political nature of attorney general appointments). But see Henry P. Monaghan, The Protective Power of the Presidency, 93 COLUM. L. REV. 1, 7-8 (1993) (arguing against a too easy dismissal of Nixon's underlying national security argument).
(18) NANCY V. BAKER, CONFLICTING LOYALTIES (1992). I will consider Baker's analysis in more detail below.
(19) Id. at 166-79.
(20) Id. at 172-73.
(21) For an intriguing and far less conventional discussion of independence in the office of the Attorney General, invoking Emerson and Weber and especially the history of Reconstruction, see Norman W. Spaulding, Professional Independence in the Office of the Attorney General, 60 STAN. L. REV. 1931 (2008).
(22) John O. McGinnis, Models of the Opinion Function of the Attorney General: A Normative, Descriptive, and Historical Prolegomenon, 15 CARDOZO L. REV. 375, 377 (1994) (briefly introducing the models).
(23) Id. at 382-89.
(24) Id. at 389-90.
(25) Id. at 402. McGinnis likens this "case by case" lawyering to the "bad man's" view of the law from Holmes's classic formulation.
(26) Id. at 377 n.5.
(27) Though the authors I will mostly consider here served in Democratic administrations, in the main their views would be shared by former OLC attorneys of both parties. See Trevor W. Morrison, Stare Decisis in the Office of Legal Counsel, 110 COLUM. L. REV. (2010) (manuscript at 5 n.15) (on file with author) (noting general agreement of Bush Administration officials with Clinton administration OLC officials' jointly proposed OLC guidelines).
(28) Randolph D. Moss, Executive Branch Legal Interpretation: A Perspective From the Office of Legal Counsel, 52 ADMIN. L. REV. 1303, 1306-16 (2000).
(29) Id. (quoting U.S. CONST. art. II, [section] 3).
(30) Id. at 1316.
(31) Id. at 1323.
(32) Trevor Morrison argues for a weaker version of the "best view of the law," namely one that emphasizes OLC's institutional function, and so is less susceptible than Moss's view (which as noted seems for Moss to be generally applicable to the Attorney General and the executive branch more broadly) to the accusation that it implies a "view from nowhere." Morrison, supra note 27 (manuscript at 52) (on file with author). I take it that Morrison's view envisions OLC as occupying the role of "neutral," leaving to others in the DOJ the role of advocate, a role that he recognizes is not "constitutionally inevitable." Id. (manuscript at 8 n.30) (on file with author).
(33) A collection of Professor Lederman's internet commentaries may be found at Marty Lederman, The Anti-Torture Memos: Balkinization Posts on Torture, Interrogation, Detention, War Powers, Executive Authority, DOJ and OLC, BALKINIZATION (July 8, 2007), http://balkin.blogspot.com/2005/09/anti-torture-memos-balkinization- posts.html.
(34) Dawn E. Johnsen, Faithfully Executing the Laws: Internal Legal Constraints on Executive Power, 54 UCLA L. REV. 1559 (2007).
(35) Id. at 1582 n.99. Johnsen's suggestion is that the "best view" is ambiguous as between the court-centered view or the more independent view, as in McGinnis's first two models. In my own view, even when one selects an interpretive stance at that level, problems remain for defenders of the legal force of any putative "best view." It is not as if predicting what courts will decide, or choosing to hew to an executive branch centered jurisprudence will eliminate the uncertainty, ambiguity, and thorniness of legal questions.
(36) Id. at 1579 and app. 2, 1603.
(37) Id. at 1604, 1607.
(38) Id. at 1606 (quoting WALTER E. DELLINGER ET AL., PRINCIPLES TO GUIDE THE OFFICE OF LECAL COUNSEL (2004)).
(39) Id. at 1577.
(40) See PHILIP HAMBURGER, LAW AND JUDICIAL DUTY 320-21 (2008).
(41) See Moss, supra note 28, at 1312-13.
(42) BRUFF, supra note 2, at 77-78.
(43) For a discussion of the problem with "views from nowhere," see THOMAS NAGEL, THE VIEW FROM NOWHERE (1986). Randolph Moss anticipates and attempts to address the sorts of concerns I raise here. See Moss, supra note 28, at 1316-26. The heart of his reply is that it is not the individual view of the lawyer, but of the office the lawyer holds, and in particular that the traditions of executive branch lawyering require a strong respect for both judicial and executive branch precedent. But it is hard to see how an argument that the law is frequently indeterminate and may be reasonably contested can be effectively rebutted by noting that there is a vast body of precedent to work with. As Moss must acknowledge, precedent is not binding, and of course precedents are frequently the very source of legal indeterminacy, rather than the cure.
(44) See Tim Golden & Eric Schmitt, A Growing Afghan Prison Rivals Bleak Guantanamo, N.Y. TIMES, Feb. 26, 2006, at A1.
(45) See generally GOLDSMITH, supra note 9 (outlining the OLC's role in the controversy).
(46) R. Jeffrey Smith & Dan Eggen, Gonzales Helped Set the Course for Detainees, WASH. POST, Jan. 5, 2005, at A1.
(47) See GOLDSMITH, supra note 9, at 141-76.
(48) See, e.g., R. Jeffrey Smith, Slim Legal Grounds for Torture Memos; Most Scholars Reject Broad View of Executive's Power, WASH. POST, July 4, 2004, at A12.
(49) For a detailed and persuasive critique of many of the infirmities of the Torture Memos, see Johnsen, supra note 34, at 1567-73; Daniel Kanstroom, On "Waterboarding": Legal Interpretation and the Continuing Struggle for Human Rights, 32 B.C. INT'L & COMP. L. REV. 203 (2009); Waldron, supra note 1.
(50) Though, to be clear, I am in general agreement with Waldron in both his detailed critique of the logical flaws of the memorandum as well as his general paradigm for interpreting laws such as the torture statute.
(51) GOLDSMITH, supra note 9, at 141-76.
(52) Youngstown Sheet & Tube Co. v. Sawyer, 349 U.S. 579 (1952).
(53) These and other reactions are collected in David Cole, They Did Authorize Torture, But...., N.Y. REV. Books, Apr. 8, 2010, at 42.
(54) One complicating factor here is whether Attorney General Ashcroft was properly informed of the work of his subordinates. See GOLDSMITH, supra note 9, at 24.
(55) See id. at 165-71.
(56) Memorandum from David Margolis, Assoc. Deputy Attorney Gen., to the Attorney Gen. 64 (Jan. 5, 2010) [hereinafter Margolis Memorandum].
(57) Id. at 67.
(58) See, e.g., John C. Yoo, The Continuation of Politics by Other Means: The Original Understanding of War Powers, 84 CALIF. L. REV. 167 (1996); see also GOLDSMITH, supra note 9, at 167-71 (noting Yoo's academic expertise prior to his OLC service, and his likely good faith in authoring the torture memos); Spaulding, supra note 21, at 1975 n.222 (collecting Yoo's relevant academic positions both prior to and subsequent to Yoo's time at the OLC).
(59) See Scott Shane et al., Secret U.S. Endorsement of Severe Interrogations, N.Y. TIMES, Oct. 4, 2007, at A1; see also Cole, supra note 53 (discussing the harsh interrogation tactics endorsed by the Justice Department under Attorney General Alberto Gonzalez).
(60) See Shane et al., supra note 59; see also Morrison, supra note 27 (manuscript at 65-66).
(61) HOMER CUMMINGS & CARL MCFARLAND, FEDERAL JUSTINE 522-23 (1937).
(62) DOJ SEAL--HISTORY AND MOTTO, U.S. DEP'T JUSTINE, http://www.justice.gov/ jmd/ls/dojseal.htm (last visited Nov. 25, 2011) (reproducing the DOJ letter explaining seal and motto).
(63) Id.; see CUMMINCS & MCFARLAND, supra note 61, at 522-23.
(64) See John McKay, Train Wreck at the Justice Department: An Eyewitness Account, 31 SEATTLE U. L. REV. 265 (2007) (detailing a former United States Attorney's insider view into the politics of the Justice Department during the Bush Administration); see also Dan Eggen, Gonzales Ready to Leave the Stage, WASH. POST, Sept. 14, 2007, at A11 (discussing the political nature of the dismissal of those U.S. Attorneys).
(65) See DEP'T OF JUSTICE, AN INVESTIGATION INTO THE REMOVAL OF NINE U.S. ATTORNEYS IN 2006 22, 47 (2008) [hereinafter DOJ Investigation]; McKay, supra note 64, at 265.
(66) See, e.g., Adam Zagorin, Why Were These U.S. Attorneys Fired?, TIME, Mar. 7, 2007 (claiming that the U.S. Attorneys were fired for speaking out against the administration).
(67) See DOJ Investigation, supra note 65, at 357; Zagorin, supra note 66.
(68) See DOJ Investigation, supra note 65, at 42-43.
(69) See id.
(70) Id. at 356.
(71) Id. at 357.
(73) See Press Release, Dep't of Justice, Statement by Attorney Gen. Michael B. Mukasey on the Report of an Investigation into the Removal of Nine U.S. Attorneys in 2006 (Sept. 29, 2008), available at http://www.justice.gov/opa/pr/2008/September/ 08-opa-859.html.
(74) See DOJ Report, supra note 65, at 356-57.
(75) See Kevin Johnson, Prosecutor Fired So Ex-Rove Aide Could Get His Job, USA TODAY, Feb. 6, 2007, available at http://www.usatoday.com/news/washington/ 2007-02-06-prosecutor-rove-aide_x.htm.
(76) Eggen, supra note 64.
(77) See John P. Elwood, Constitutionality of D.C. Voting Rights Act of 2007, May 23, 2007, http://www.justice.gov/olc/2007/dcvotingrights-act-2007.pdf; Carrie Johnson, A Split at Justice on D.C. Vote Bill: Holder Overrode Ruling That Measure Is Unconstitutional, WASH. POST, Apr. 1, 2009, at A1.
(78) See Johnson, supra note 77.
(79) See id.
(80) Whelan, supra note 10.
(83) See John McGinnis, An End Run Around the Rule of Law, EXECUTIVE WATCH, http://executivewatch.net/2009/04/06/an-end-run-around-the-rule-of-law/ (last visited Nov. 25, 2011).
(85) See Planned Parenthood v. Casey, 505 U.S. 833, 854 (1992).
(86) See McGinnis, supra note 22, at 380.
(87) Cf. id. at 402; Morrison, supra note 27 (manuscript at 47).
(88) Here by "conservative" I mean something like the view of the law that would involve the least risk of being rejected by courts, or would garner the most widespread acceptance among learned commentators, etc.
(89) See Jeremy Pelofsky & James Vicini, 9/11 Suspects to Be Tried in New York, REUTERS (Nov. 13, 2009, 5:28 PM), http://www.reuters.com/article/idUSTRE5AC1S7 20091113.
(90) DEP'T OF JUSTICE & DEP'T OF DEFENSE, DETERMINATION OF GUANTANAMO CASES REFERRED FOR PROSECUTION (2009), http://www.justice.gov/opa/documents/tabaprel-rpt-dpff-072009.pdf.
(91) See, e.g., Jane Mayer, The Trial: Eric Holder and the Battle over Khalid Sheikh Mohammed, NEW YORKER, Feb. 15, 2010, at 52; Khalid Sheikh Mohammed's Manhattan Show Trial, GOP.GOV (Nov. 17, 2009), http://www.gop.gov/policy-news/09/11/17/ khalid-sheikh-mohammed-s-manhattan.
(92) See, e.g., Glenn Greenwald, The Administration Guts Its Own Argument for 9/11 Trials, SALON (Nov. 19, 2009, 8:20 AM), http://www.salon.com/news/opinion/glenn_greenwald/2009/11/19/obama/.
(93) See Kenneth R. Bazinet et al., Shaikh it up! Bam Gives in on Trial, DAILY NEWS, Jan. 28, 2010, at 9; Anne E. Kornblut & Carrie Johnson, Obama to Help Pick Location of Terror Trial, WASH. POST, Feb. 12, 2010, at A1, A18.
(94) See U.S. DEP'T OF JUSTICE, UNITED STATES ATTORNEYS' MANUAL [section] 9-27.220 (1997), available at http://www.justice.gov/usao/eousa/foia_reading_room/usam/ title9/27mcrm.htm#9-27.220.
(95) Kornblut & Johnson, supra note 93, at A1, A18.
(96) Henry Paul Monaghan, Supremacy Clause Textualism, 110 COLUM. L. REV. 731, 788 (2010).
(97) See id.
(98) See Bloch, supra note 12, at 561.
(100) See HAMBURGER, supra note 40, at 320 ("Attorneys general and solicitors general shared in the judges' duty."). Hamburger's discussion focuses on the oaths of attorneys and solicitors general and upon some episodes from the career of Edmund Randolph. While I am persuaded that Edmund Randolph understood himself to have something like a quasi-judicial duty as Attorney General, the structure of the office and of our constitutional order--indeed, the fact that Randolph himself argued to courts at all--undermine the logic of his position, however much it may have led him to restraint in his advocacy. See id. at 490-92.
(101) See LUTHER A. HUSTON, THE DEPARTMENT OF JUSTICE 7 (1967) ("As prescribed by the Judiciary Act [of 1789], the duties and powers of the Office of the Attorney General were few and vaguely defined and reflected the legislators' concern lest the office become a center of federal power that would infringe upon the prerogatives of the states."); Bloch, supra note 12, at 561-66.
(102) See Spaulding, supra note 21, at 1953.
(103) U.S. DEP'T OF JUSTICE, U.S. DEP'T OF JUSTICE OVERVIEW (2011), http://www.justice.gov/jmd/2011summary/pdf/overview.pdf.
(104) See Spaulding, supra note 21, at 1953 n.132.
(105) See Saikrishna Prakash, The Chief Prosecutor, 73 GEO. Wash. L. REV. 521, 521-25 (2005).
(106) See id. at 537.
(107) See id. at 553-58.
(108) Id. at 560.
(109) Id. at 561 (quoting Letter from Thomas Jefferson to Edward Livingston (Nov. 1, 1801), reprinted in 8 THE WRITINGS OF THOMAS JEFFERSON 57 (Paul Leicester Ford ed., 1897)).
(110) See id. at 580.
(111) See id. at 580.
(112) See generally Harold J. Krent, Executive Control over Criminal Law Enforcement: Some Lessons from History, 38 AM. U. L. REV. 275 (1989) (arguing that there were historical, substantial checks on executive prosecutorial discretion).
(113) 343 U.S. 579 (1952).
(114) Stephen Gillers, The Torture Memo, THE NATION, Apr. 28, 2008, at 6.
(115) See Youngstown, 343 U.S. at 634-55 (Jackson, J., concurring).
(116) Transcript of Oral Argument, Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579 (1952), reprinted in 48 LANDMARK BRIEFS AND ARGUMENTS OF THE SUPREME COURT OF THE UNITED STATES 920 (P.B. Kurland & G. Casper eds., 1975).
(117) Cf. Memorandum from Steven G. Bradbury, Principal Deputy Assistant Attorney Gen., U.S. Dep't of Justice Office of Legal Counsel, to Attorneys of the Office 2 (May 16, 2005), http://www.justice.gov/olc/best-practices-memo.pdf ("Decisions of the Supreme Court and courts of appeal directly on point ... should be thoroughly addressed....").
(118) In this regard, Professor Charles Fried's testimony at the confirmation hearings of then-Judge Samuel Alito is illuminating. Fried discussed a memo authored for him by Justice Mitt when the former was Solicitor General and the latter was an assistant solicitor general. The Alito memo apparently took the view that while Roe v. Wade had been wrongly decided, it would be imprudent for the Solicitor General to ask the Court to revisit the decision. In the event, Fried rejected Alito's advice and did so urge the Court. As Fried testified, Alito in his memo correctly predicted that the Court would react with hostility to such an argument. (My recent reading of the Blackmun papers in the Library of Congress showed me just how hostile that reaction had been.) See The Confirmation Hearing on the Nomination of Judge Samuel Alito, Jr. to Be an Associate Justice of the Supreme Court of the United States, 109th Cong. 713-14 (2006) (statement of Charles Fried, Former Solicitor General of the United States and Benefical Professor of Law, Harvard Law School). Surely there is an analogue here for the torture memo. It may well be that John Yoo's genuine good faith belief is that Youngstown's analysis errs; that argument might be made in an advisory opinion, but simply ignoring Youngstown altogether, rather than assessing its ongoing vitality and applicability, surely does not comport with the OLC's aim to provide a "balanced presentation of arguments." Memorandum from Stephen G. Bradbury, supra note 117, at 3.
(119) GOLDSMITH, supra note 9, at 168.
(120) See id.; JOHN YOO, CRISIS AND COMMAND (2009).
(121) GOLDSMITH, supra note 9, at 177-216.
(122) Id. at 215.
(123) Id. at 213-15.
(124) Yoo, supra note 120, at viii-xx.
(125) Id. at 427.
(126) Id. at 421 ("Commanders have long set the standards for the capture and treatment of enemy prisoners."). Curiously, Professor Yoo omits any mention of Congress's power under the Constitution to make rules concerning capture on land and water.
(127) Gillers, supra note 114, at 8.
(128) 487 U.S. 654 (1988).
(129) Id. at 655-56.
(130) Id. at 697-734 (Scalia, J., dissenting).
(131) See Kevin M. Stack, The Story of Morrison v. Olson : The Independent Counsel and Independent Agencies in Watergate's Wake, in PRESIDENTIAL POWER STORIES 401, 428 (Christopher H. Schroeder & Curtis A. Bradley eds., 2009).
(133) Robert H. Jackson, The Federal Prosecutor, 24 J. AM. JUDICATURE SOC'Y 18, 19 (1940).
(134) Morrison, 487 U.S. at 732 (1988).
(135) See Dan Froomkin, Untangling Whitewater, WASH. POST (2000), http://www. washingtonpost.com/wp-srv/politics/special/whitewater/whitewater.htm.
(137) See Yoo, supra note 120, at 380. But see Stack, supra note 131, at 445 (arguing that Morrison was responding to Watergate-style independent counsel and suggesting that while a facial challenge failed, an as-applied challenge may have prevailed in the Whitewater case).
(138) Thomas W. Merrill, High Level, "Tenured" Lawyers, 61 LAW & CONTEMP. PROBS. 83, 107 (1998).
(139) Any number of models of independence have been proffered. See, e.g., Spaulding, supra note 21, at 1970-79 (examining four different conceptions of independence).
(140) BAYER, supra note 18, at 67-68.
(141) Id. at 80.
(143) Id. at 79.
(144) See id. at 102-05.
(145) Id. at 126.
(146) See GRIFFIN B. BELL & RONALD J. OSTROW, TAKING CARE OF THE LAW 28 (1982).
(147) See id.
(148) Griffin B. Bell, Office of Attorney General's Client Relationship, 36 BUS. LAW. 791 (1981).
(149) Id. at 795-96.
(150) Id. at 796; see also Merrill, supra note 138, at 90-92 (discussing similar episode from Reagan Administration where the acting Solicitor General disclaimed government views in his own Supreme Court Brief).
(151) FED. R. Cry. P. 11(b).
(152) BAKER, supra note 18, at 155.
(153) See Margolis Memorandum, supra note 56, at 67-69.
(154) See, e.g., ERIC A. POSNER & ADRIAN VERMEULE, TERROR IN THE BALANCE (2007); 58% Favor Waterboarding of Plane Terrorist to Get Information, RASMUSSEN REPORTS (Dec. 31, 2009), http://www.rasmussenreports.com/public_content/politics/general_politics/ december_2009/58_favor_waterboarding_of_plane_terrorist_to_get_informa-tion. But see Poll Results: Waterboarding Is Torture, CNN (Nov. 6, 2007), http://articles. cnn.com/2007-11-06/politics/waterboard.poll_1_waterboarding-michael- mukaseyinterrogation?_s=PM:POLITICS.
(155) See Margolis Memorandum, supra note 56, at 67.
(156) Cf. Catherine J. Lanctot, The Duty of Zealous Advocacy and the Ethics of the Federal Government Lawyer: The Three Hardest Questions, 64 S. CAL. L. REV. 951 (1991) (discussing the complexity of ethical dilemmas faced by government lawyers).
(157) Trevor W. Morrison, Constitutional Avoidance in the Executive Branch, 106 COLUM. L. REV. 1189, 1194-95 (2006).
(158) Moss, supra note 28, at 1304.
(159) See generally W. Bradley Wendel, Government Lawyers, Democracy, and the Rule of Law, 77 FORDHAM L. REV. 1333, 1337 (2009) (arguing that government lawyers have an obligation of fidelity to the law that exceeds any improper political influence).
(160) For a discussion of these powers and their interrelationship, see, e.g., Charles Tiefer, Congressional Oversight of the Clinton Administration and Congressional Procedure, 50 ADMIN. L. REV. 199 (1998).
(161) GUIDO CALABRESI & CHRISTOPHER YOO, THE UNITARY EXECUTIVE 426-27 (2008).
(162) See, e.g., Rick Pildes, The Free Enterprise Decision: A Symbolic Victory for the "Unitary Executive Branch" Vision of the Presidency, but of Limited Practical Consequence, BALKINIZATION (June 28, 2010), http://balkin.blogspot.com/2010/06/free-enterprise-decision-symbolic.html (arguing that Free Enterprise Fund was a pyrrhic victory for unitary executive theorists).
(163) Free Enter. Fund v. Pub. Co. Accounting Oversight Bd., 130 S. Ct. 3138, 3164 (2010).
(164) This may be an overstatement, depending especially upon the interpretation of the Court's suggestion that Congress would be free to pursue other options to cure the defect found by the Court here, including making PCAOB members "removable by the President, for good cause or at will." Id. at 3162. It seems clear, however, that such a passage constitutes dicta, and that the Court's off-handed suggestion of potential remedies more appropriate for Congress than the Court to decide is not necessarily a reliable indication that each such remedy would be constitutional. The passage does indicate as strongly as any that the Court has come to embrace at least a fair bit of Humphrey's Executor and Morrison. It seems highly unlikely, therefore, that the Court would revisit Humphrey's Executor even if invited, though it does not seem eager to extend that precedent.
(165) 272 U.S. 52, 247 (1926) (Brandeis, J., dissenting).
(166) See generally William P. Marshall, Break Up the Presidency--Governors, State Attorneys General, and Lessons from the Divided Executive, 115 YALE L.J. 2446 (2006) (arguing that an independent Attorney General would provide a much-needed check on executive power).
(167) CALABRESI & YOO, supra note 161, at 364.
(168) See Waldron, supra note 1, at 1706-09 (discussing torture memo's treatment of the statutory term "severe pain").
(169) See Pelosi's Tortured Denials, FACTCHECK.ORG (May 21, 2009), http://www.factcheck.org/2009/05/pelosis-tortured-denials/.
(170) Yoo's Testimony with Addington Before House Judiciary: From the Department of Justice to Guantanamo Bay: Administration Lawyers and Administration Interrogation Rules (Part III), 108th Cong. 7-18 (2008) (statement of John Yoo).
(171) See Marty Lederman, Waldron/Yoo Debate on Torture, BALKINIZATION (April 26, 2005), http://balkin.blogspot.com/2005/04/waldronyoo-debate-on-torture.html (internal quotation marks omitted).
(172) See supra notes 121-23 and accompanying text.
(173) Jackson, supra note 133, at 18.
(174) See generally Richard L. Hasen, Let Them into the House, SLATE (Jan. 28, 2009, 1:50 PM), http://www.slate.com/articles/news_and_politics/jurisprudence/2009/ 01/let_them_into the house.html (noting some division among scholars on constitutionality of D. C. Voting Rights Act).
(175) Peter L. Strauss, Overseer, or "The Decider"? The President in Administrative Law, 75 GEO. WASH. L. REV. 696 (2007). My own views of this key episode more closely accord with the discussion in CALABRESI & Yoo, supra note 161, at 105-22, 420-21.
(176) Strauss, supra note 175, at 706-07.
(177) Portland Audubon Soc'y v. Endangered Species Comm., 984 F.2d 1534, 1550 (1993); see Strauss, supra note 175, at 710-11.
(178) Strauss, supra note 175, at 760.
(179) Id. at 709.
(180) To be sure, this is so only for prospective decisions in most instances. One can certainly envision an Attorney General being removed after taking a decision that cannot be reversed by his successor--in such circumstance, the President's removal power will have proven inadequate to control the Attorney General. I do not view this as undermining my overall model, however. Certainly I do not think that the level of deference I am suggesting an Attorney General owes the President as the chief representative of the Attorney General's client requires supine or robotic obedience in undertaking the duties of the office. I am indebted to Henry Monaghan for this observation.
(181) Free Enter. Fund v. Pub. Co. Accounting Oversight Bd. 130 S. Ct. 3138, 3144 (2010). Some might note that the Court's reasoning should apply as well to the member of the SEC. As noted at the outset, the Court declined to revisit its precedents approving of limitations on the President's removal power. Neither, however, did it forswear revisiting those precedents down the line. Rather, the Court noted that "the parties do not ask us to reexamine any of these precedents, and we do not do so." Id. at 3147.
(182) Strauss, supra note 175, at 707 & n.57.
(183) Id. at 713.
William R. Dailey, CSC, Visiting Associate Professor, Notre Dame Law School. Thanks to Howard Anglin, Joseph Blanco, Sam Bray, Neil Delaney, Richard Garnett, Abbe Gluck, Kent Greenawalt, Jamal Greene, Philip Hamburger, Bert Huang, William Kelley, Harold Krent, Tom Merrill, Gillian Metzger, Henry Monaghan, Michael Moreland, Trevor Morrison, John O'Callaghan, Graham O'Donoghue, Anthony O'Rourke, Nate Persily, Peter Strauss, and Jay Tidmarsh, as well as the participants in faculty colloquia at Chicago-Kent, Toledo, and St. Thomas Law Schools for their valuable comments at various stages of the project. I am also indebted to Adam Ramsey, Jason Littrell, Alex Day and Thomas Waugh for invaluable research assistance and support. Finally, the editors of the Notre Dame Law Review provided thorough and thoughtful editorial support for which I am grateful.