The New York Times recently unearthed a thorough legal memo, prepared twenty years ago for Independent Counsel Kenneth Starr, that advances the view that a sitting president can be indicted while still in office. For those keeping score, this new memo sharpens an internal divide within the Department of Justice on this important question. Two memos authored by the Office of Legal Counsel—one in 1973, in the midst of the Nixon impeachment saga, the other in 2000, on the heels of the Clinton impeachment saga—take the view that a sitting president is immune from indictment. By contrast, two different memos—authored by the Office of Special Counsel investigating Nixon, and the Office of Independent Counsel investigating Clinton—reach the opposite conclusion.
That these different offices have repeatedly disagreed on this central question isn’t really all that surprising. They have different institutional roles, different missions, and different cultures, all of which might impact their respective approaches to the issue. For present purposes, however, the most important practical question is whether the current special counsel, Robert Mueller, is free to exercise his own independent judgment on the immunity issue, or whether he is instead bound to follow OLC’s take. If it’s the latter, then those two OLC memos would together constitute the single greatest shield protecting President Trump from prosecution: No matter how strong the evidence against him may become, if OLC’s memos are binding then the President simply cannot be indicted until after he leaves office—by which point, it bears noting, the statute of limitations for any relevant conduct may well have expired.
But that’s only if OLC’s memos are binding. Jack Goldsmith and Marty Lederman take the view that they “almost certainly” are. The New York Times, by contrast, has twice indicated that the issue may not be so clear cut, each time citing a piece that I wrote expressing some early skepticism on this issue.
That skepticism may well represent a minority view, at least among those with substantial experience working with OLC—experience that Jack and Marty have, and that I do not. Still, even an outsider’s minority view merits an articulation of some of the main points in its favor.
To my mind, there are at least three such points here. First and foremost, the justifications underlying the general practice of treating OLC opinions as binding on executive branch officials do not necessarily apply to the Office of Special Counsel, which is supposed to be insulated from the influence of political appointees when assessing the president’s exposure to criminal liability. Second, the formal regulations setting out the special counsel’s authority do not clearly compel him to follow OLC’s lead. And third, historical practice suggests that he need not do so.
Let’s take these three points in turn.
I. When It Comes to Presidential Immunity, the Normal Reasons for Treating OLC Memos as Binding Do Not Necessarily Apply to the Special Counsel
The notion that OLC opinions bind the Office of Special Counsel draws support from a widely held broader proposition: OLC opinions bind all executive branch officials with respect to whatever legal issues those opinions resolve. As Trevor Morrison explains, that broader proposition is not exactly beyond dispute: “the bindingness of the Attorney General’s (or, in the modern era, OLC’s) legal advice has long been uncertain,” he writes, and is the subject of “almost two hundred years of debate.” That debate, however, can largely be bracketed here. Because even if one accepts OLC’s general authority “to say what the law is” for the rest of the executive branch, it’s not clear that such authority extends to the Office of Special Counsel under the current circumstances.
To see why, just consider OLC’s own explanation for why its opinions are binding, as set out in an opinion it issued back in 1987. According to OLC, that power has two foundations: The first is the attorney general’s “statutory obligation,” delegated to OLC, “to render opinions” when he is requested to do so by one of the heads of the other executive departments regarding any “questions of law arising in the administration of their” duties. The second is the attorney general’s statutory responsibility, “except as otherwise authorized by law,” to “conduct litigation on behalf of the United States,” which “necessarily includes the exclusive and ultimate authority to determine the position of the United States on the proper interpretation” of legal issues.
What stands out most about these two asserted bases of OLC’s authority is that neither clearly supports the claim that the two OLC memos on presidential immunity bind the Office of Special Counsel. As to the first source of authority, neither of those memos was requested by the head of an executive department outside of DOJ, so the attorney general’s duty “to render opinions” to such officials simply doesn’t apply. (Indeed, it’s not altogether clear who requested that these two particular memos be written in the first place.)
That leaves the second source of authority, which by its own terms is subject to an important exception: OLC’s authority to issue binding opinions does not extend to executive branch officials who are “authorized by law” to “conduct litigation on behalf of the United States” without first getting the attorney general’s blessing for the positions they intend to take. Due in part to this exception, independent agencies within the executive branch, to quote former OLC Deputy and now-D.C. Circuit Judge Nina Pillard, are “not…presumptively bound” by OLC’s opinions. Indeed, according to OLC’s own official “Best Practices,” the Office typically won’t even “provide an opinion to an executive agency the head of which does not serve at the pleasure of the president (e.g., an agency head subjected to a ‘for cause’ removal restriction)” unless that agency first agrees to “conform its conduct” to whatever opinion OLC might provide.
The principle that independent officers are not presumptively bound to follow OLC’s opinions is important in the present context because the Office of Special Counsel bears many of the traditional hallmarks of such independence. The special counsel himself can be removed from office only for “good cause,” such as “misconduct” or “dereliction of duty.” Moreover, he has “independent authority” to conduct litigation on behalf of the United States, in the trial courts and on appeal, without being “subject to the day-to-day supervision of any official” in the Department of Justice, including the attorney general. Similarly, he has no obligation to seek preapproval from any such official prior to taking a given course of action.
To be sure, the special counsel is required to keep the attorney general apprised of significant developments in the investigation. And the attorney general does have some authority to block the special counsel from taking a “prosecutorial step” like filing an indictment. That authority, however, is quite limited: the attorney general must “give great weight to the views of the Special Counsel” on the matter, and can only overrule the special counsel if he determines “that the action is so inappropriate or unwarranted under established Departmental practices that it should not be pursued.” Moreover, if the attorney general overrules the special counsel, he must explain that decision to Congress.
In view of these limitations, Deputy Attorney General Rosenstein—the only official with any authority over Special Counsel Mueller, given Attorney General Sessions’s recusal—has made clear that, in his view, Mueller has “full independence” to conduct the pending investigation. One can parse Rosenstein’s words. But the sentiment they express shouldn’t really be too surprising. After all, independence from political actors is the very raison d’être for having an Office of Special Counsel in the first place. Absent such independence, criminal investigations of high-level executive branch officials, like the president, simply could not escape the inherent “conflicts and potential conflicts of interest” that would arise if they were overseen by the president’s own political appointees. Indeed, it’s hard to imagine circumstances in which improper structural biases would be more stark or more problematic than a case in which a lawyer is asked to decide whether his boss—the person who picked him for the job and who can dismiss him from it—can be sent to prison.
And yet, that is precisely the question that was put to the Office of Legal Counsel in 1973 and in 2000, as the clouds of indictment drew near to the two presidents then in office.* Crucially, however, unlike the Office of Special Counsel, the Office of Legal Counsel is not insulated from presidential control. On the contrary, as Judge Pillard observes, OLC is subject to a greater degree of political control than many other DOJ offices: not only is it headed by the President’s political appointee, but “all of the OLC deputies are politically appointed as well,” unlike, say, the Office of the Solicitor General, where “three out of the four deputies are career employees.”
As Judge Pillard goes on to observe, “a more politically led office seems less likely to make impartial, arms-length constitutional decisions.” That seems especially true when the constitutional decisions at issue are guided by “neither the text nor the history of the Constitution,” but turn instead on how one balances the president’s interests “as the sole head of the executive branch” against the broader societal interests in ensuring that no person is above the criminal law—which is precisely how OLC describes the nature of the presidential-immunity analysis.
In conducting that interest-balancing analysis, it would hardly be unusual for OLC to place special—and perhaps outsized—emphasis on the president’s side of the ledger. Indeed, to hear Yale Law School’s Bruce Ackerman tell it, OLC “almost always concludes that the president can do what he wants.” One need not endorse so broad a critique, however, to acknowledge the more basic fact that OLC not only works for the president but also tries to “facilitate the objectives of the President” where possible. On the contrary, that is precisely how OLC itself describes its role in its “Best Practices” memo.
That memo goes on to observe that, in addition to facilitating the president’s interests, OLC also typically “keeps the Office of the Counsel to the president appropriately apprised of its work.” Imagine, however, a White House Counsel calling up the head of OLC for an update on the “Can we indict the Boss?” memo.
The very prospect that such a phone call might occur is why the Office of Special Counsel exists—to avoid such inherent conflicts of interest. And no matter how earnest or upright OLC’s attorneys may be, there is simply no avoiding the central concern: binding the Special Counsel to the judgments of the president’s political appointees, on the very question of the president’s own criminal liability, would significantly undermine the independence that the Office of Special Counsel was built to provide—and that, in the current situation, it has been promised. Given that independence, it’s simply not clear that OLC’s memos, even if generally binding on other executive branch officials, are binding on the Special Counsel here.
II. The Special Counsel’s Governing Regulations Do Not Clearly Compel Him to Follow OLC Opinions
Even if one thinks that the Special Counsel has no freestanding obligation to follow these two OLC memos, he is unquestionably bound by his own authorizing regulations, which require him to “comply with the rules, regulations, procedures, practices and policies of the Department of Justice.” Does this regulatory text require the special counsel to follow OLC’s conclusions?
As I’ve previously noted, probably not, given that this “regulatory text seems to focus more on administrative protocols and procedures than on” the “legal analyses” that OLC produces. That textual argument, moreover, grows only stronger once the special counsel regulations are read in the context of their adjacent regulatory provisions, which define the authority of all of the other component offices of the Department of Justice. As those other regulations make clear, there are a host of offices in the Department—ranging from the Office of the Assistant Attorney General for Administration, to the Advisory Committee of U.S. Attorneys, to the aptly named Office of Legal Policy—that are expressly authorized to promulgate “rules,” “regulations,” “procedures,” or “policies” for the Department, and thus to bind the special counsel.
The Office of Legal Counsel, however, is conspicuously not granted such authority. Rather, according to OLC’s authorizing regulation, its central charge is to render “opinions and legal advice to the various agencies of the government.” Indeed, the sole mention of “policies,” “rules,” “regulations,” or “procedures” in OLC’s authorizing regulation is telling. A seemingly peripheral paragraph in OLC’s mandate tasks the Office with “consulting with the Director of the Office of Government Ethics regarding the development of policies, rules, regulations, procedures and forms” necessary to implement the rather mundane provisions of “section 402 of the Ethics in Government Act of 1978.” That narrow statutory provision, however, has nothing to do with OLC’s core opinion-writing duties, let alone with establishing departmental policy concerning immunity from prosecution.**
In short, while departmental regulations repeatedly and expressly grant other offices within DOJ the power to promulgate “rules, regulations, procedures, and policies” on a wide range of issues—and thus to bind the Office of Special Counsel on those issues—OLC is simply not granted any such authority, except with respect to an unrelated administrative function not relevant here.
III. Historical Practice Supports Allowing the Special Counsel to Assess the Presidential-Immunity Issue for Himself
Perhaps in recognition of the points raised above, the only other special prosecutors who have ever investigated a sitting president for criminal misconduct both performed their own independent assessments of the presidential-immunity issue, without simply taking OLC’s opinions as binding authority. Neither of those special prosecutors was subject to the exact same regulatory authorization that establishes Mueller’s office. But they were subject to similar provisions defining the scope of their authority. Specifically, the regulation governing Watergate special prosecutor Leon Jaworski said that he was “subject to the administrative regulations and policies of the Department of Justice.” And the statute establishing Kenneth Starr’s Office of Independent Counsel similarly required him to “comply with the written or other established policies of the Department of Justice respecting enforcement of the criminal laws,” unless doing so “would be inconsistent with the purposes” of the statute.
And yet, notwithstanding these express commands, Jaworski and Starr both sought their own independent analyses of the presidential-immunity question, clearly indicating that they did not consider themselves bound by OLC’s prior conclusions. Indeed, the Starr memo twice acknowledged that the “Office of Independent Counsel is, in general, required to follow Department of Justice regulations governing other federal prosecutors,” an obligation that the memo did not treat as at all inconsistent with its ultimate conclusion, contra OLC, that the president can be indicted. And the Jaworski memo went so far as to expressly acknowledge OLC’s contrary view, before proceeding to reject it.
In short, both of Mueller’s predecessors were bound to follow DOJ policy, but neither considered himself bound by OLC’s memos concerning presidential immunity. Mueller is governed by a different set of regulations, which have scant independent history of their own. And the fact that his predecessors took a close, independent look at the presidential-immunity question does not conclusively determine what Mueller’s obligations are today. But it does offer some persuasive evidence as to the course he is permitted to take.
* * *
In sum, unique features of the specific OLC memos at issue here may well cut against the claim that Mueller is bound by them, and neither the regulatory text nor the history of the special counsel’s office requires a contrary conclusion.
And if that all still feels strange, let me close by offering a hypothetical twist to the ongoing Russia investigation that will hopefully make this all feel a bit more concrete. Imagine that, a few months from now, as indictment chatter builds, a newly installed attorney general (Jeff Sessions’ replacement) issues a formal “Opinion of the Attorney General” that opens with the following statement:
Upon careful review of the Constitution’s text, history, and structure, it is the opinion of this Office and of the Department of Justice that close family relatives of the President of the United States—including his spouse, his children, and his children’s spouses—are immune from criminal prosecution for so long as the President is in office. Moreover, it is the opinion of this office that a President can never be prosecuted, whether during his term of office or thereafter, for any conduct related either to his seeking the presidency or to the performance of his official acts while President. 44. Op. Att’y Gen. 1 (2018)
I may be wrong, but I suspect that were a future attorney general to issue such an opinion, in the midst of the Mueller investigation, he or she would be met with fierce opposition—including, perhaps, from some who now contend that Mueller is bound to follow “the formal opinions of the Attorney General” as articulated by OLC. Note, however, that the substantive merits of this hypothetical opinion are not exactly absurd. Prosecuting the President’s children, after all, would surely “interfere with the President’s ability to carry out his constitutionally assigned functions,” perhaps even more so than prosecuting the President himself. (I’ve represented over 120 people accused of crimes, and most had parents who would have quickly opted to take their child’s place.) Similarly, the notion that the Constitution might bar a prosecution premised on the President’s official acts or his campaign activities has already been raised by some legal scholars.
To be sure, the attorney general opinion suggested above would certainly draw criticism on the merits. But still, I suspect a significant portion of any potential backlash would stem from the fact that the opinion was issued by the President’s own political appointee—and that even though it articulates principles of general applicability, it has a practical effect much like an anticipatory pardon, shorn of the political complications that an actual pardon would entail.
Of course, like most law professor hypotheticals, this one is in some sense an exaggeration—or at least so one hopes. But it serves to illustrate the central point: The existing OLC memos on presidential immunity, like the hypothetical one above, also function to some degree as de facto anticipatory pardons of potential criminal activity committed by a president, especially given the serious practical challenges that arise when a prosecution is delayed for too long. And the existing OLC memos, like the hypothetical one above, were also crafted by presidential appointees, at the behest of presidents facing genuine prospects of indictment—not by the special counsels whose entire purpose is to insulate presidential charging decisions from precisely such political influence.
Perhaps the current special counsel, exercising the independent judgment that his predecessors embraced, would conclude that OLC is right and that a sitting president is in fact immune from indictment. There is a big and important difference, however, between agreeing with such a view and being forced to adopt it even if you think it’s wrong. The whole point of having a special counsel is to benefit from his independent judgment on precisely this sort of issue.
* It bears noting that the timing of OLC’s two immunity memos is somewhat curious. According to the “Best Practices” memos written by the heads of OLC under two different administrations, “OLC generally avoids opining on questions likely to arise in pending or imminent litigation involving the United States as a party.” As Nelson Lund explains, that practice reflects the reality that “OLC does not serve as the mouthpiece for the Solicitor General or the litigating divisions” of the Department of Justice, which “will often defend” or advance a proposition in court “even if OLC would have advised against it.” When the ball is in the litigating divisions’ court, in other words, OLC typically defers to their authority to articulate the position of the United States. And when the litigation at issue is a potential criminal prosecution of the president of the United States, the relevant “litigating division” is the Office of Special Counsel.
Yet, in an apparent deviation from its articulated best practices, OLC issued the presidential-immunity memos in the thick of two separate pending cases. Indeed, the first memo actually contradicted the litigating position that Special Counsel Jaworski soon articulated on behalf of the United States to the Supreme Court. And the second opinion was prepared as the Office of Independent Counsel was considering a draft indictment of President Clinton, in a process that had also already generated litigation. The fact that OLC may have deviated from its ordinary best practices to issue these memos—thereby perhaps usurping the special counsels’ rightful authority to articulate the government’s litigating position—may be yet another reason not to treat the memos as binding. [Editor's Note: An earlier version of this post suggested that the Special Counsel’s Reply Brief in United States v. Nixon was filed in July of 1973. The brief was filed in July of 1974.]
** Coincidentally, a separate section of the Ethics in Government Act of 1978 established the now-defunct Office of Independent Counsel, the predecessor to the Office of Special Counsel currently established by 28 C.F.R. §600 and occupied by Mueller. OLC’s regulations, however, do not assign it any responsibilities with respect to “policies, rules, regulations, or procedures” connected to that now-expired office, which was created by Section 601 of the Ethics in Government Act, not the Section 402 that is referenced in OLC’s authorizing regulation.