Democracy, Distrust, and Presidential Immunities

36 Const. Comment. 255

Evan Caminker[1]

Four decades ago in Democracy and Distrust,[2] John Hart Ely masterfully charted a middle path between the two then (and still) dominant methods of constitutional interpretation. The prevailing version of originalism (“interpretivism”), he eloquently expounded, was intrinsically unworkable and unattractive for why-be-bound-by-the-dead-hand-of-the-past reasons; the alternative of norm-based reasoning (“noninterpretivism”) was ungrounded and riven with judicial subjectivity; and neither method was “ultimately reconcilable with the underlying democratic assumptions of our system.”[3] Ely posited and passionately defended a third approach, “process-based constitutionalism,” favoring a “participation-oriented, representation-reinforcing approach to judicial review” that would “ensur[e] broad participation in the processes and distributions of government.”[4] When interpreting “open-ended provisions,” courts should largely confine themselves to promoting the related process values of fair participation in the selection of government officials and fair representation of all by those officials. Although Ely’s attempt was not entirely persuasive on its own terms[5] and the interpretive schools he rebuked dug in their heels rather than capitulate,[6] his was one of the most influential writings on judicial review of his generation and, I dare say, still in ours.

Somewhat ironically for a book focused on promoting fair governance process values, Ely said very little about structural battles among the governmental pillars of our federalist system.[7] Indeed, the book barely even mentions the President. Given that over the last forty years both formal and informal presidential power has significantly expanded and generated many constitutional controversies, this lacuna does reveal the book’s age.

That said, Ely specifically invited broader and different applications of his core thesis: “[t]he elaboration of a representation-reinforcing theory of judicial review could go many ways, and [my proposals] are obviously just one version.”[8] I accept that invitation here and try to extend his core insights into one such important and topical structural issue: whether and to what extent the President of the United States enjoys temporary or permanent immunity from various enforcement efforts to conform her conduct to law.

The Supreme Court most recently addressed this thicket last year in Trump v. Vance,[9] building on previous decisions that subjected sitting Presidents to evidentiary subpoenas in federal criminal cases[10] and federal civil litigation over unofficial conduct[11] but provided sitting and former Presidents immunity from civil damages claims for official misconduct.[12] In Vance, the Court (over two dissents) rejected President Trump’s sweeping claim of temporary immunity from a state grand jury’s subpoena issued to an accounting firm seeking his personal financial and tax records. President Trump asserted “the Supremacy Clause gives a sitting President absolute immunity from state criminal subpoenas because compliance . . . would categorically impair a President’s performance of his Article II functions”[13] by distracting, stigmatizing, or otherwise harassing him. While touting the President’s “independence” and “unrivaled gravity and breadth” of duties,[14] the Court said no: “the President is neither absolutely immune from state criminal subpoenas seeking his private papers nor entitled to a heightened standard of need.”[15] The Court left open, however, questions concerning immunity from further stages of criminal prosecution: whether a sitting President may be criminally indicted, tried, and/or sentenced by federal or state prosecutors, and whether a former President may be criminally prosecuted for her official acts.

Ely’s process-based approach offers a fresh look. Current immunity doctrine highlights notions of presidential efficacy, asking whether various enforcement measures unduly impair the President’s job performance.[16] I suspect Ely would hesitate to frame the problem this way; assessing whether the non-textual value of presidential effectiveness has been “unduly impaired” might feel as ungrounded and slippery as assessing whether a non-textual liberty interest has been “unduly burdened,” a form of open-ended balancing Democracy and Distrust criticized in the context of unenumerated rights.

This Essay sketches how Ely’s representation-reinforcing theory of judicial interpretation might frame presidential immunity doctrines and compares that frame to the Court’s current approach. To what extent might various forms of presidential immunity, or exceptions thereto, be grounded in principles of democratic accountability rather than presidential efficacy? I conclude that a plausibly constructed Elyan paradigm provides an argument for immunity in many settings but also for exceptions to that immunity in narrow but important circumstances.[17] More specifically: immunity can protect the President’s ability to focus on serving her view of the national interest, without being unduly chilled or sidetracked by private burdens imposed by individual actors. On the other hand, certain litigation efforts to constrain presidential misbehavior can enhance presidential accountability in a different way, by deterring Presidents from (1) “clogging the channels of [political] change”[18] through self-entrenching actions (think election interference) or by (2) failing to represent her constituents en masse through self-dealing actions (think steering government contracts for, or basing foreign policy decisions on, personal financial gain). In general these conclusions support, and in some instances helpfully clarify and strengthen, the Court’s current efficacy-based concerns; but the arguments for immunity exceptions are novel and weighty. Overall, in my view, the general Elyan approach and vocabulary offer a useful framework for immunity doctrine that should supplement if not supplant the prevailing approach.

Part I of this Essay briefly canvasses the current doctrinal approach, illustrating how the Court has crafted immunity doctrine out of concern that various forms of civil or criminal process might impair the President’s ability to perform her duties effectively. Part II sketches an alternative Elyan accountability-based approach to presidential immunity, reevaluates the Court’s pro-immunity arguments through this alternative frame, presents the two novel accountability-based reasons to override any erstwhile immunity, and finally sketches options for incorporating the arguments for and exceptions to immunity into doctrine.

My analysis here is admittedly partial; thoroughly considering presidential immunity’s full contours would require deeply engaging other textual, historical, and functional considerations—as well as much more space. But I hope this preliminary exploration suitably celebrates both the analytical insights and staying power of Ely’s masterful book on the occasion of its fortieth anniversary.


Presidential immunity comes in two flavors. Temporary immunity protects a sitting President from judicial process while she remains in office, but no longer. Permanent immunity protects a President after she leaves office as well.

Here I first review the arguments the Supreme Court has offered as favoring temporary immunity, then I review its arguments favoring permanent immunity, and finally I briefly consider the countervailing interests the Court weighs against both types of immunity in a separation-of-powers balancing test.

  1. Temporary Immunity (Civil and Criminal)

In Clinton v. Jones, Paula Jones sued then-President Clinton in federal court on both federal- and state-law claims primarily alleging that Clinton, before becoming President, sexually harassed her. In Vance v. Trump, the New York City prosecutor through a grand jury issued a subpoena to an accounting firm for the private financial papers of then-President Trump as part of a state-law investigation into a potentially large swath of private financial misconduct, which similarly predated Trump’s assuming office.[19] In both cases, the Court rejected the sitting President’s claim that litigation would burden or harass them to the point of unduly impairing their job performance: Clinton as applied to all stages of civil litigation; and Vance as applied only to pre-indictment criminal process, leaving other stages for another day. In this section I’ll briefly sketch the Court’s articulated concerns regarding presidential impairment in these two cases, as these concerns will surely resurface if a sitting President is ever criminally indicted and prosecuted. The Court’s concerns might be grouped in various ways; I’ll group them, as did Vance, into diversion, stigma, and presidential harassment.

  1. Diversion

The Court recognized that both civil litigation and criminal subpoenas could require the President to participate in litigation activities and thereby divert some of her time, energy, and focus away from public duties. A civil defendant might be required to assist her lawyers and participate in discovery and might choose to attend a trial; the owner of subpoenaed papers (even when held by a third-party custodian) might feel compelled to assist with (or, as here, contest) compliance.[20] The Court considered whether such diversion appreciably impairs presidential performance.

Presidents have claimed that “he occupies a unique office with powers and responsibilities so vast and important that the public interest demands that he devote his undivided time and attention to his public duties.”[21] If so, any litigation-caused diversion means less time, energy, and perhaps mental focus for official responsibilities. In Clinton, the Court accepted this premise and assumed arguendo that personal litigation burdens necessarily trade-off against public duties.[22]

In Clinton, the Court suggested a distinction between what I’ll call “time- and space-bound” diversions and “flex-time” diversions. Time- and space-bound diversions occur when litigation requires the President to do something at a particular time in a particular place—for example, if a trial court requires in-court or time-bound appearances—which could obstruct particular high-priority activities.[23] The Court dismissed the concern for such obstruction in civil litigation, noting that presidential discovery or trial testimony may “accommodate his busy schedule” and that the President need not personally attend any trial.[24] In Vance, the Court similarly dismissed the notion that criminal subpoenas could create time- and space-bound diversions.[25] But the concern might well be triggered by criminal trials, as trial attendance is hardly optional in the same way; and post-conviction incarceration obviously triggers these concerns for obstruction.[26]

“Flex-time” litigation burdens may divert Presidents’ attention from public duties generally without obstructing any specific effort. In Clinton the Court recognized though discounted this concern as well: while Presidents “may become distracted or preoccupied by pending litigation,” they “face a variety of demands on their time, however, some private, some political, and some as a result of official duty.”[27] In that context, civil litigation’s “even quite burdensome” flex-time diversions do not “rise to the level of constitutionally forbidden impairment.”[28] Vance similarly concluded based on “two centuries of experience . . . that a properly tailored criminal subpoena will not normally hamper the performance of the President’s constitutional duties.”[29] But the concern again remains valid, if and when triggered by further stages of criminal prosecution.

  1. Stigma

President Trump argued further that “the stigma of being subpoenaed will undermine his leadership at home and abroad.”[30] While the concern is not fanciful, the Court was unmoved having “twice denied absolute immunity claims by Presidents in cases involving allegations of serious misconduct” (obstruction of justice in Nixon and sexual harassment in Clinton).[31] The Court explained “there is nothing inherently stigmatizing about a President . . . ‘furnishing information relevant’ to a criminal investigation.”[32] And given Trump’s concession that criminal investigations are permissible, the President’s reputational injury would not magnify even “if he is the one under investigation”—at least partly because grand jury secrecy rules “aim to prevent the very stigma the President anticipates.”[33] So “even if a tarnished reputation were a cognizable impairment,”[34] subpoena service—without more—failed to trigger appreciable concern.

  1. Presidential harassment

In Vance, President Trump insisted that “subjecting Presidents to state criminal subpoenas will make them ‘easily identifiable target[s]’ for harassment.”[35] The Court rejected Trump’s claim on these facts, but in so doing it described the general concern for presidential harassment in frustratingly opaque and shape-shifting ways.

The Court first treated Trump’s complaint as mundanely focused on frivolousness. The Court had “rejected a nearly identical argument in Clinton, where then-President Clinton argued that permitting civil liability for unofficial acts would ‘generate a large volume of politically motivated harassing and frivolous litigation.’”[36] After Trump objected that state prosecutors “are responsive to local constituencies, local interests, and local prejudices, and might ‘use criminal process to register their dissatisfaction with’ the President,” the Court responded that state as well as federal courts are equipped to “protect against the predicted abuse.”[37] “[G]rand juries are prohibited from engaging in ‘arbitrary fishing expeditions’ and initiating investigations ‘out of malice or an intent to harass,’”[38] and these protections may be enforced by federal courts.[39]

If “harassing” merely means frivolous or vexatious, the Court’s hypothetical concern makes sense. Any frivolous criminal process—whether motivated by a desire to harass or curry favor with local constituents or please the prosecutor’s parents—would both violate procedural rules and also fail any separation of powers balancing test for lack of justification.[40]

Unfortunately the Court confused matters by also observing that harassing subpoenas could “threaten the independence or effectiveness of the Executive.”[41] Here the Court referenced not frivolousness but obstruction: the “Supremacy Clause prohibits state judges and prosecutors from interfering with a President’s official duties.”[42] But the Court left unexplained why a subpoena intended to harass would obstruct presidential performance where a similar but good-faith subpoena would not; impairment seems related to effect, not motive.

The Court also characterized a “harassing” subpoena “as an attempt to influence the performance of [the President’s] official duties,”[43] signaling a worry that a prosecutor might use a subpoena to pressure or punish a President’s policy decisions.[44] Here the Court shifted further away from any obstructive effect on presidential performance, as the Court clearly decried mere attempts to influence presidential behavior, with or without any distortive effect: “[a]ny effort to manipulate a President’s policy decisions or to ‘retaliat[e]’ against a President for official acts through issuance of a subpoena . . . would thus be an unconstitutional attempt to ‘influence’ a superior sovereign ‘exempt’ from such obstacles.”[45] But having framed this concern, the Court never explained why mere efforts to influence the President’s agenda, without more, unduly impair her efficacy.

In the end, if the Court’s hand-wringing over presidential harassment signals more than just “don’t serve frivolous or vexatious subpoenas,” its meaning is elusive, and therefore its application to further stages of criminal prosecution remains hazy. All we know is that this concern fails to justify blanket temporary immunity yet remains theoretically provable on a case-specific basis.

* * *

In sum, the Court has identified (and to a greater or lesser degree explained) several potentially cognizable concerns supporting a finding of temporary immunity. None have yet been meaningfully triggered by either civil litigation or criminal subpoenas; but they will surely be invoked if more aggressive criminal process is deployed against a sitting President.

  1. Permanent Immunity
  2. Civil liability for damages

While Presidents (at least through their subordinate officials) can be restrained by civil injunctions,[46] all government officers enjoy substantial protection from suits seeking civil damages regarding their exercise of official duties—absolute immunity for some actors for some functions, and qualified immunity for the rest.[47] But the “President occupies a unique position in the constitutional scheme” that “distinguishes him from other executive officials”[48] and demands more.

In Nixon v. Fitzgerald, the Supreme Court held that a former President enjoys absolute immunity from damages liability for official acts as a “functionally mandated incident of the President’s unique office, rooted in the constitutional tradition of the separation of powers and supported by our history”[49] (though notably not the constitutional text[50]). Presidents make “the most sensitive and far-reaching decisions,” address “matters likely to ‘arouse the most intense feelings,’” and must retain “‘the maximum ability to deal fearlessly and impartially with’” their duties.[51]

The Court worried that damages liability might make the President “unduly cautious in the discharge of his official duties.”[52] Liability might “diver[t] . . . attention during the decisionmaking process” by causing “needless worry as to the possibility of damages actions.”[53] Even more importantly, the President might ultimately refrain from taking even perfectly legal actions that would serve the national interest but would risk personal liability if a jury, based on available evidence, could misapprehend the true facts (including motive).[54]

Befitting its purposes, immunity from damages extends to “acts within the ‘outer perimeter of [the President’s] official responsibility.’”[55] This perimeter clearly encompasses activity that might be illegal; the Court rejected Fitzgerald’s claim that President “Nixon would have acted outside the outer perimeter of his duties” if his actions actually violated Fitzgerald’s statutory rights.[56] And this immunity is permanent; a former President is protected forever to preserve her in-office freedom to pursue the national interest as she perceives it, constrained only through prospective relief.

  1. Criminal liability

It’s widely assumed that Presidents lack a corresponding permanent immunity from criminal liability and sanctions. The civil‌/criminal distinction finds some support in the constitutional text,[57] the presumed greater public interest in penalizing criminal wrongs than righting civil wrongs,[58] the greater safeguards protecting against overzealous criminal actions than civil ones,[59] and a long history of prosecuting all categories of federal and state officials aside from the President.[60] People debating whether a sitting President enjoys temporary immunity generally assume Presidents are prosecutable after they leave office; this includes framers,[61] scholars,[62] government lawyers,[63] and Ely himself.[64] And in Vance Trump conceded and the Court assumed that prosecutors may “investigate a sitting President with an eye toward charging him after the completion of his term.”[65]

That said, it is not unimaginable that the Court might, if asked, extend Fitzgerald’s broad immunity from civil damages for misconduct “within the ‘outer perimeter’ of [the President’s] official responsibility”[66] (or perhaps at least within the “core” of such responsibility) to criminal liability as well. The question is whether the President’s “unique status . . . distinguish[ing] him from other executive officials”[67] extends to the criminal realm as well. Fitzgerald’s central rationale, that the fear of civil damages might unduly chill otherwise desirable and legal presidential decisions and actions, certainly extends to typically scarier criminal sanctions. Perhaps fewer presidential acts might skirt the boundaries of criminal than civil law, but at least in those instances the chill is likely just as strong if not stronger. Notably, while Justice White’s dissent specifically argued that Presidents lack permanent immunity from criminal prosecution,[68] the Court very pointedly omitted any reference to criminal liability when listing other accountability mechanisms that, taken together, ensure civil damages immunity “will not place the President ‘above the law.’”[69] And the common view that temporary criminal immunity simply “defer[s] litigation and investigations until the President is out of office”[70] may assume that such deferred prosecutions will charge personal rather than official misconduct and thus lie beyond any Fitzgerald-mirroring protection, as was true for the pre-term financial misconduct at issue in Vance.[71]

If and when a former President is indicted for official misconduct in office, I expect she will press this argument.

  1. Balancing countervailing interests

Under current doctrine, the Court considers whether litigation burdens on the general exercise of executive power are “justified by an overriding need to promote [governmental] objectives.”[72]

In Fitzgerald, the Court identified two precedent-based possibilities: first, “[w]hen judicial action is needed to serve broad public interests—as when the Court acts, not in derogation of the separation of powers, but to maintain their proper balance”; and second, “to vindicate the public interest in an ongoing criminal prosecution.”[73] Only the first interest was implicated by civil claims. But rather than further explain what “broad public interests” might justify some measure of executive burdens in civil cases, the Court left this category amorphous and instead merely pointed to “[t]he existence of alternative remedies and deterrents establish[ing] that absolute immunity will not place the President ‘above the law’”[74]—listing impeachment, press and congressional oversight, and the President’s own reputational concerns.[75]

In criminal cases, the Court unsurprisingly emphasized law enforcement objectives. In Nixon, the Court rejected an assertion of unqualified executive privilege because accessing all “relevant and admissible evidence” is critical to courts’ “primary constitutional duty . . . to do justice in criminal prosecutions.”[76] In Vance, the Court extended its concern from protecting the judicial trial function (and parties affected thereby) to promoting law enforcement investigatory interests more generally: “the public interest in fair and effective law enforcement cuts in favor of comprehensive access to evidence.”[77] The Court refused to apply Nixon’s “heightened standard of need” requirement where no executive privilege applies, because that standard “would hobble the grand jury’s ability to acquire ‘all information that might possibly bear on its investigation.’”[78] The Court further noted that delaying compliance would allow incriminating evidence to disappear or grow stale and “prejudice the innocent by depriving the grand jury of exculpatory evidence.”[79] Neither Nixon nor Vance required the Court to consider countervailing interests in prosecuting wrongdoing Presidents specifically, though at times the Court nodded toward an interest in prosecuting wrongdoers generally.[80]

  1. Addressing Further Criminal Process

While the Court has seemingly addressed all relevant civil immunity claims,[81] presidential criminal immunity remains less clear. As discussed above,[82] the Court has yet to consider a presidential claim for permanent immunity from criminal liability. Such immunity would starkly distinguish Presidents from all other government actors, testing just how far their “unique position in the constitutional scheme” might move the Court.[83]

With respect to temporary immunity, Vance was but a dress rehearsal for the day—if it unfortunately ever comes—when a sitting President is criminally indicted and faces further prosecution, trial, and punishment. These next steps might well impose more severe burdens that more strongly support a temporary immunity defense. A criminal trial would likely impose some time- and space-bound diversions; while Clinton downplayed a sitting President’s need personally to participate in a civil trial,[84] surely we expect that all defendants will feel impelled to participate in a criminal trial, both to present themselves before the jury and to avail themselves of various trial-level constitutional rights.[85] Most conventional postconviction sentences would at least appreciably interfere with her duties.[86] Criminal pretrial defense might well divert a sitting President’s time, energy, or mental focus more than defending civil litigation or complying with a criminal subpoena, though that’s likely a context-specific question.[87] Public indictment might stigmatize and mentally preoccupy the President more than mere subpoena service, though perhaps it would remain difficult to demonstrate outright “obstruction.” By contrast, the harassment-based concern for “attempting to influence” presidential decision-making (whatever that entails)[88] might be no weightier for prosecutions than for subpoenas. It’s one thing to imagine a prosecutor subpoenaing a President for partisan advantage or policy leverage; it’s quite another to imagine the prosecutor going further to indict and continue prosecuting the President without a good-faith belief in provable guilt.

To Justice Alito, taking any of these steps against a sitting President would be “farcical”;[89] the Department of Justice agrees, albeit using more measured terms.[90] Among constitutional scholars, many likewise support temporary immunity for criminal indictment and beyond; others would permit indictment (some always and some only when necessary to toll a statute of limitations); still others would permit full prosecution.[91] Given that the Court-articulated presidential burdens and prosecutorial interests are incommensurable, choosing a point along the spectrum at which burdens outweigh interests seems to entail just the sort of open-ended values balancing that, at least in the context of rights-protection, makes Ely uncomfortable.


Before considering how a representation-reinforcing approach might support or reshape the Court’s efficacy-based doctrine, it’s worth considering Ely’s own musings applying the conventional framework. Years before publishing Democracy and Distrust, Ely briefly addressed presidential immunities and, in addition to text and history, referenced “station” and “disruption” objections that resonate with the Court’s impairment focus.[92] In 1973 he sent a letter to Special Prosecutor Archibald Cox, who was then considering “whether [President Nixon] enjoys some immunity from being called before a grand jury” as part of the Watergate investigation.[93] Ely said that such a claim, “let alone [the broader claim] that he is generally immune to the judicial process, seems very faint indeed” and immunity advocates would have their “backs against the wall.”[94] Even more broadly, for unofficial misconduct (which Ely presumed was non-impeachable in accord with the “prevailing view”), a “criminal investigation must proceed unless the president is to be placed above the law.”[95]

Ely also pointedly questioned broader notions of temporary criminal immunity in a never-submitted 1973 draft letter to the New York Times. The draft letter addressed a brief that then-Solicitor General (and friend) Robert Bork had recently filed, taking the position that a sitting President is immune from criminal indictment even though a sitting Vice President is not. Ely responded: “To the extent [the framers] suggested anything on the subject, the debates suggest that the immunities the Constitution explicitly granted members of Congress (which do not, incidentally, include this sort of immunity) were not intended for anyone else. The argument for presidential immunity from indictment is one that must be based on necessity―and perhaps, but only perhaps, the presidency and vice presidency are distinguishable on that score―and not on anything the framers said either in the Constitution itself or during the debates.”[96]

That said, Ely indicated his views were tentative, telling Cox “obviously much research and thinking would need to be done” in a context-sensitive manner.[97] And years later Ely may have warmed to the idea of an ahistorical “necessity” for presidential immunity from indictment. After fellow Symposium participant Professor Akhil Amar coauthored an essay eloquently defending such immunity,[98] Ely wrote Professor Amar a note saying “you may have changed my mind.”[99] So while pre-Democracy and Distrust Ely was strongly inclined to reject presidential criminal immunity,[100] his later views are less clear.[101]

Fortunately given this sparse record of Ely’s own thoughts, my goal here is not to divine Ely’s subjective views but rather to see what his more general theory of judicial review can add to the conversation and how it might shed light on modern-day doctrinal options. Recall his view that “[t]he elaboration of a representation-reinforcing theory of judicial review could go many ways,” and his exemplars “are obviously just one version.”[102] Here I sketch a different version to see where it might support and might diverge from the Court’s current approach to presidential immunities. Section A sketches an Elyan accountability-based approach to presidential immunity. Section B reevaluates the Court’s pro-immunity arguments through this alternative frame, assessing litigation burdens for their degree of interference with an ideal of 24/7 public representation rather than for their degree of interference with an ideal of an adequately (or optimally?) effective President. This section concludes that there are sound (and sometimes better) accountability-based reasons underlying an immunity shield. Section C presents two novel accountability-based reasons to override any erstwhile immunity, providing litigants in certain circumstances with a shield-piercing sword. Section D sketches options for incorporating the arguments for and exceptions to immunity into doctrine.

  1. Democracy and Accountability

Ely’s general thesis, of course, is that judicial review of open-ended constitutional provisions should “keep the machinery of democratic government running as it should” and “concern itself only with questions of participation, and not with the substantive merits of the political choice under attack.”[103] He proposes two interconnected objectives, both of which generally promote democratic accountability.

First, judicial review should endeavor to “clear[] the channels of political change.”[104] Democratic accountability “malfunction[s]” when “the ins are choking off the channels of political change to ensure that they will stay in and the outs will stay out.”[105] Ely suggests courts should protect against channel clogging by broadly interpreting freedoms of speech and press and the right to vote, encouraging legislative transparency, and discouraging excessive legislative delegation of policymaking.[106] Courts should help enable We the People to decide who we want our leaders to be and prevent those who currently hold power from thwarting our choices.

Second, judicial review should be “representation-reinforcing,” designed to encourage elected “decision-makers . . . to take into account the interests of all those their decisions affect.”[107] Ely’s book focuses primarily on one application of this principle: to facilitate the representation of minorities, who face systemic disadvantaging “out of simple hostility or a prejudiced refusal to recognize commonalities of interest” and are thereby at risk of being denied “the protection afforded other groups by a representative system.”[108] But Ely repeatedly references in other contexts the foundational principle—ensuring that “everyone’s interests will be actually or virtually represented (usually both) at the point of [an official’s] substantive decision.”[109]

These are quite lofty and general objectives, of course; and Ely cashes them out only with broad brushstrokes and only as triggered by congressional acts. But with admittedly equally broad brushstrokes, one can apply these objectives to the debate over presidential immunities. Both aspects of Ely’s proposed process-based judicial review can generate important insights for this timely debate. With respect to presidential burdens triggering judicial scrutiny, a representation-reinforcement framework generally supports (though sometimes for slightly different reasons) the Court’s articulated concerns for both time- and space-bound and flextime diversions that drain the President’s time and mental energy away from her public duties. The framework also supplies a somewhat more cohesive and persuasive explanation for the Court’s recent muddled musings about presidential harassment, though questions remain. And with respect to countervailing interests that might outweigh the presidential burdens just articulated, Ely’s focus on both channel-clearing and public accountability favors exceptions to immunity where a President’s misconduct promotes self-entrenchment or self-dealing.

  1. Elyan Accountability-Based
    Arguments Supporting Immunity

As sketched in Part I, the questions underlying current immunity doctrine—designed to protect some notion of presidential efficacy against impairment[110]—can be distilled as follows:

  • Is the President’s capacity to adequately perform her duties sufficiently unobstructed?
  • Is the President sufficiently physically and mentally available to adequately perform her duties?
  • Is the President’s inclination to serve the public interest sufficiently free from improper chilling or influence?

These questions can be reoriented toward an Elyan focus on presidential representation-reinforcement. Each question considers a different aspect of a President’s willingness and ability to channel her best and entire effort to promote what she perceives as the national interest.[111] “[T]he Framers made the President the most democratic and politically accountable official in Government” who is “‘responsible for the actions of the Executive Branch.’”[112] If litigation thwarts, diverts, or chills her from achieving or pursuing this objective, it precludes her from fulfilling her duty to “exercise” her granted power “in the[ ] interest and behalf” of the entire national public[113] and instead prioritizes the initiating litigant’s agenda.[114]

Obviously there are degrees of injury here, and Ely’s approach wouldn’t avoid the line-drawing challenges faced by current doctrine. But the measuring stick is conceptually different. While again Democracy and Distrust provides only a basic edifice for his approach and he invites further tinkering, in my view Ely’s representation-reinforcement framework is best understood to support judicial review to ensure that elected leaders appropriately serve their constituents’ interests, not to guarantee those leaders secure those interests to the maximum possible degree.[115]

In the following discussion I map these questions onto the various doctrinal concerns traced above.

  1. Chilling-effect distortion of presidential decision-making

The accountability model contemplates that the President will fairly and fully consider the interests of her constituency when formulating her view of the national interest and hence her policymaking agenda. This doesn’t mean each constituent’s views deserve equal weight; but it does mean that each deserves respect in the President’s interest-assessment. If vulnerability to judicial process chills behavior that would serve that interest, the President no longer acts in an accountable fashion; her fear of liability trumps (sorry) her erstwhile agenda.

This accountability concern supports immunity from civil damages claims. As the Court explained in Fitzgerald,[116] a President’s liability for potentially large civil damages awards—whether that liability will be realized while she remains in office or afterwards—might well chill her from presidential actions that are both legal and serve the public interest.[117] As such, this chilling effect constitutes an Elyan cognizable concern.

Indeed, Ely’s model may better explain than does the current efficacy model why the Court held open whether it would matter if Fitzgerald’s suit hadn’t merely asserted “implied” causes of action but instead “Congress expressly had created a damages action against the President.”[118] It’s not clear why the cause of action’s express vs. implied status would differently chill and affect presidential performance—indeed, one could imagine an express law on the books would induce greater chill and thus impairment. But express congressional blessing for a legal restriction on presidential discretion arguably underscores the restriction’s accountability credentials. In that case, at least the risk of chilling (undermining presidential accountability) actually serves a clearly accountable congressional constraint, which might affect the overall calculus.

This Elyan perspective also provides an argument supporting immunity from criminal liability; fear of criminal sanctions is at least as likely to influence presidential decisions as do civil damages. Whether this chilling effect by itself justifies criminal immunity is obviously a more complicated question overall.[119]

  1. Diversion

Here, accountability and efficacy concerns also overlap.

Time- and space-bound diversions

If a President is outright obstructed from performing her public duties by being confined through litigation to a particular place at a particular time, such as by court order to sit for a deposition at a particular time or by fairness-grounded need to attend and participate in her own criminal trial, during that time she is necessarily focused on personal litigation defense and cannot effectively represent her national constituency’s interests. Even though the touchstone is presidential focus rather than efficacy, a non-trivial obstructive diversion of this sort equally favors temporary immunity.

The Court has yet to contemplate a court ordering a President incarcerated after conviction, though Justice Alito stated the fairly obvious when he noted this would threaten to “sacrifice[]” “a functioning government.”[120] An Elyan approach would (at a minimum[121]) highlight the ways in which prison life would almost assuredly impose time- and space-bound diversions; a President cannot advance the national interest while she’s staffing the prison laundry or obeying lights-out. Surely such constraints would require an exceedingly strong countervailing interest, if any could suffice.

Flex-time diversions

Accountability and efficacy concerns also overlap here. If the President’s time, energy, or mental focus is meaningfully diverted to litigation defense, even with flexibility as to when and where, she will be less physically or mentally available to serve the public interest. So here too, significant flex-time diversion concerns would favor immunity under Ely’s approach.[122]

Again, under Ely’s approach as under current doctrine, these burdens need not be measured in an all-or-nothing fashion; one can consider degrees. So the mere fact of some litigation-triggered diversion need not be dispositive; in my view the Court correctly dismissed such concerns as de minimis in both Clinton and Vance. But if criminal indictment or (more persuasively) criminal defense preparation and trial are deemed to significantly divert a President’s attention from her public duties, the accountability framework offers a stronger basis for immunity as the proceedings unfold.

  1. Stigma

Whether stigmatic injuries trigger constitutional scrutiny under an Elyan approach turns on a more nuanced sense of accountability. As captured in the bullet-pointed questions above, thus far I’ve translated accountability into a concern for whether the President is capable of, available for, and inclined toward serving the people. Litigation-induced stigma does not directly implicate these concerns—a President can still devote 24/7 of her attention and effort to serving the public interest to the best of her ability, even if stigma compromises her ability to accomplish her goals. If the Elyan approach is designed to ensure that elected officials serve the “right” interests rather than actually achieve those interests to the utmost degree, then the accountability framework would exclude stigma from the set of doctrinal triggers favoring judicial protection. So stigma would remain a doctrinal concern only if one uses the Elyan approach to supplement rather than entirely supplant the Court’s current focus on protecting presidential efficacy from impairment.

Of course, the line between diverting the sitting President’s attention from her public duties and hindering her ability to accomplish those duties, while to my mind conceptually clear, is admittedly practically thin. Indeed, one might measure stigmatic injury in person-hours; just like sitting for a deposition, overcoming stigmatic injuries can take time and effort. I still see apples and oranges here, processes and effects; I still think it is useful and appropriate to distinguish between “the President has less time‌/energy to represent the public interest” and “the President is less capable of achieving the substance of that public interest.” But I can easily imagine others might flesh out Ely’s account in a broader way to also cognize litigation-triggered impediments to achieving representation-reinforcing goals.[123]

  1. Harassment

As explained above, the Court’s description in Vance of “presidential harassment” is opaque and shape-shifting.[124] The Court’s approach at a minimum reflect concerns for frivolousness, and appears additionally to worry about prosecutorial attempts to use judicial process as a means of influencing presidential decision-making. In my view, Ely’s accountability framework supports the concern about frivolousness and offers a better way to understand the Court’s expressed concern about pressuring the President’s agenda.

Frivolous litigation

Frivolous or vexatious litigation, by definition, serves no public purpose; it instead serves the plaintiff or prosecutor’s personal interests. To the extent that civil or criminal litigation poses any concerns for even flex-time diversions, those accountability costs have no countervailing justification. So the Elyan approach, like the current efficacy approach, would support protecting Presidents from this minimalist harassment concern.[125] As the Court recognized, however, this concern is too weak to justify blanket immunity; case-specific judicial safeguards are sufficient. Nothing in Ely’s model suggests otherwise.

Distortion of presidential decision-making

The Court’s expressed concern for inappropriate influence is unclear in part because it seems to jump from effect to intent. After reminding that the Supremacy Clause prohibits state prosecutors from “interfering with a President’s official duties,” the Court declared that “[a]ny effort to manipulate a President’s policy decisions or to ‘retaliat[e]’ against a President for official acts [through criminal process] would thus be an unconstitutional attempt to ‘influence’ a superior sovereign ‘exempt’ from such obstacles.”[126] How motive alone can “interfer[e]” with or create “obstacles” for presidential performance was left unexplained—perhaps because the Court largely dismissed the concern by assuming that state officers will “observe constitutional limitations.”[127]

That said, one can glean a concern for distortion of presidential decision-making, assuming that counts as a form of interference‌/obstruction. A prosecutor might in theory manipulate the President’s agenda by threatening to subpoena (or indict) her unless she does X or refrains from Y, presumably ready to stand down if she complies with the demand; or a prosecutor might retaliate against the President after the fact for her refusal to do X or not do Y. Again in theory, the mechanism could be a specific quid pro quo threat (“stop bashing sanctuary cities or I’ll start a criminal investigation”) or a generalized message (“stop pushing ideological‌/partisan policies or else”). If the President views the threat as credible, the argument runs, she might well modify a specific policy position or become more cautious and centrist in general.

So constructed, this argument resonates with the distortion concern underlying Fitzgerald’s absolute immunity from civil damages.[128] There, the chilling effect grew from the fact that the President may not always clearly see the line between legal and liability-generating conduct—or may worry that a jury will not see the line the same way she does—and therefore she will prophylactically refrain from perfectly legal and public-serving behavior to avoid any risk of owing damages to an injured plaintiff seeking redress in good faith. Here, the chilling effect grows from the fear of partisan-motivated prosecutors; the President will prophylactically refrain from perfectly legal and public-serving behavior to avoid any risk of being criminally investigated in retaliation. Notwithstanding the Court’s language of interference and obstruction, then, the real concern is agenda-altering influence.[129]

But this distinction matters. Although it’s old news that state officials cannot interfere with or obstruct federal officials or policies, it’s surprising news that state officials cannot “attempt to influence” federal officials or policies. In many other contexts, no one bats an eye when state or local officials use various forms of leverage to try to influence presidential decision-making. For example, they offer or withhold campaign support and lobby privately or publicly for or against presidential action (e.g., do protect borders, don’t impose tariffs). They bring civil suits to enjoin presidential policies, at least sometimes not just to stop the challenged policies but also to deter similar-but-legal subsequent presidential moves[130] or express displeasure with previous ones.[131] And exercising the “political safeguards of federalism,” they attempt to deter federal policies by refusing to help implement them and thereby undermine their effectiveness or raise their costs.[132]

Relatedly, shifting from Supremacy Clause to general Article II concerns, private individuals also frequently use leverage to try to influence presidential behavior. For example, people frequently offer or withhold campaign support, whether through contributions or attending‌/avoiding political rallies. People also work to deter presidentially preferred policies by making them ineffective (e.g., refusing vaccines, refusing to buy American, exhorting big companies to eschew government contracts[133]). These and similar efforts employ leverage (beyond the mere expression of views and interests) to “manipulate” the President to do X or “retaliate” against her if she refuses.

What’s missing from Vance is any hint why these and other examples of political, economic, and bureaucratic leverage are acceptable efforts to shape presidential decisions, but criminal process is not. Why is it worse for New York officials to attempt to persuade President Trump to soften his disfavored stance on immigration by issuing him a subpoena than by becoming a defiant sanctuary state—again, keeping in mind our assumption that the subpoena is nonfrivolous and it does not actually obstruct the President in any cognizable way? Or from a different angle: why is it worse for District Attorney Vance to issue a nonfrivolous subpoena because he’s upset that Trump’s pandemic policies hurt New York residents than because he’s upset that Trump’s alleged tax-dodging hurt New York taxpayers? From a perspective of protecting the President’s effectiveness from impairment, the answer is not self-evident.[134]

Perhaps here Ely’s accountability model can offer some useful insights. As to the basic concern, if a prosecutor (or private litigant) uses judicial process to pressure the President to deviate from her agenda, the President in a sense becomes captive to and prioritizes the litigant’s interests over the nation’s. So a well-founded threat of actual distortion offers an accountability-based argument for immunity.

But the accountability perspective may also help distinguish criminal process from the conventional methods of leverage mentioned above. Some of them, such as lobbying and campaigning, can be viewed as part of the accountability process itself. These are tools through which people and officials express preferences, and their ability to influence presidential decision-making fits comfortably within the notion of representation-reinforcement.[135] Other methods, such as individuals refusing vaccines or states refusing to implement federal policies, impact the benefit-cost comparison the President faces as she assesses the national interest. For example, the President might oppose pandemic mask mandates when vaccination levels are high and support mask mandates when levels are low; if so, mass vaccination boycotts will lead her to change her position. Such methods of influence may also be said to work within the accountability model: the President is registering preferences and trading off concerns that matter to the public. If a state’s refusal to facilitate mass vaccinations undermines the effort to achieve herd immunity, the public’s view of the optimal mask rules might change along with the relevant facts on the ground. Hence the President’s reaction, while influenced by the state’s refusal to support vaccinations (even if it was motivated by partisan politics rather than sincere policy objections), remains part of the process of fashioning a new accountable position. By contrast, a state’s threat to indict the President if she continues to promote vaccinations pressures her to prioritize purely private concerns—the desires to maintain her reputation and stay out of prison. The public interest doesn’t change; it just gets short-changed by the President’s decision to ignore it in order to save her own hide.[136]

A different sort of accountability concern also may do some work here. For Ely, accountability is multi-faceted: leaders should work on behalf of their constituents’ interests, and constituents should be able to evaluate their leaders and decide whether to vote them out of office. The latter is supported by transparency: constituents may fairly evaluate leaders only if they know what the leaders are doing and why.[137] Here, while some of the conventional methods of leverage described above are fully visible (e.g. individual boycotts and state actions) and others sometimes reside in the shadows (e.g. lobbying and campaigning), presumably prosecutorial threats of the sort contemplated in Vance will be concealed from public view, as will the ways in which the President might be chilled and influenced. So the public may unfairly blame the President for doing or not doing X without knowing that decision was secretly shaped by a prosecutor’s hidden agenda.[138]

These arguments—admittedly but thinly sketched here—at least offer some basis, missing in Vance, for treating criminal process as a peculiarly out-of-bounds mechanism for “attempting to influence” presidential behavior.

All that said, while accountability may outperform efficacy at explaining the Court’s hazy concern, this doesn’t mean the concern should be taken seriously—and the Court, to its credit, does not. I think it’s highly unlikely that a prosecutor would try to distort presidential behavior in a targeted way. To successfully influence a specific presidential decision or policy, the prosecutor would have to communicate a fairly specific threat, such as “stop doing X or I’ll serve a criminal subpoena on you.” That seems farfetched even for a prosecutor motivated by partisanship or reelection prospects. Not only would the violation of prosecutorial ethics be stunningly stark,[139] but if publicly revealed, such a threat would likely be career-damaging if not ending no matter how unpopular the President or her policy. And to successfully influence a President’s general agenda, the prosecutor would have to communicate a much more general and vague threat, such as “stop supporting non-centrist policies ‌/ favor our state more or I’ll come after you.” Here even a worried President wouldn’t really know what or how much to do to defuse the threat, and it seems highly unlikely that she would fundamentally pull back from her goals in response. So even taking the concern at face value as fleshed out through Ely’s eyes, I think the Court rightly held it doesn’t justify absolute immunity and instead merely warrants giving Presidents the opportunity to “challenge any allegedly unconstitutional influence” in a given case.[140] The Elyan approach helps explain the argument, though it doesn’t boost its power.

* * *

In sum, the accountability and efficacy arguments for presidential immunity overlap substantially, but not completely. With respect to permanent immunity, an Elyan account would additionally explain why pro-immunity concerns might have less heft when confronting express congressional authorization for suit. With respect to temporary immunity, an Elyan account would dismiss the stigma concern and provide a sounder foundation for the Court’s concern about distorting a President’s agenda. As to temporary defenses, it’s worth noting that neither concern was invoked in Fitzgerald or moved the needle in Clinton or Vance, and any future indictments or prosecutions will likely be driven by more substantial diversion or obstruction arguments anyway.

By comparison, as the next Section develops, the accountability approach uniquely identifies significant countervailing interests that might outweigh immunity concerns.

  1. Elyan accountability-based arguments supporting immunity overrides

As explained above, the Court professes to “balance the constitutional weight of [pro-prosecution interests] against the dangers of intrusion” on executive functions, and blesses vague aspirations “to maintain the[] proper balance” of powers and “to vindicate the public interest in an ongoing criminal prosecution.”[141] The Court has already recognized important judicial and law enforcement interests in serving subpoenas to support criminal trials and timely investigations.[142] If a sitting President ever faces indictment and prosecution, which clearly threaten to impose greater burdens (whether measured by efficacy or accountability), the Court will have to consider other significant interests as well—for example, the law enforcement interest in indicting a sitting President before the statute of limitations runs.[143]

Whatever Ely might think of the Court’s open-ended balancing test,[144] his focus on democratic accountability offers a new perspective. As explained above, the norm of presidential accountability to her national constituency can support a shield of immunity from litigation. But in certain contexts the very same norm forges a sword to pierce that shield as well. More specifically, an accountability paradigm identifies an “overriding need” to hold the President answerable for misconduct that either (1) “clogs the channels of political change” by entrenching herself (or her cronies) in power, or (2) reflects self-dealing that is not merely indifferent to but injures the public interest.

Entrenchment: Ely’s argument that “unblocking stoppages in the democratic process is what judicial review ought preeminently to be about”[145] directly applies here. The free speech and voting rights doctrines that Ely deploys against congressional efforts to block political change should cover equivalent presidential actions. But the anti-entrenchment norm not only justifies enjoining such actions but provides an argument against letting the President claim immunity for them. If immunity is designed to let the President serve her national constituency rather than be deflected by narrower interests, the rationale dissipates when she herself acts to protect her political power at the expense of democratic fluidity.

One can imagine a President or presidential candidate engaging in many sorts of self-entrenching misconduct. President Trump himself allegedly violated campaign laws by soliciting favorable foreign interference in a domestic election (conditioning security assistance to Ukraine on President Zelensky’s efforts to dig up dirt on opposing candidate Joe Biden’s son); allegedly violated campaign finance rules (for failing to report hush-money payments to women claiming to have had extramarital affairs with him); and allegedly interfered with the right to vote and electoral processes (by inciting supporters to storm the Capitol building to influence Congress’s counting of electoral college votes, and by attempting to influence Georgia’s vote recount).[146] Defined more broadly, self-entrenchment could include misconduct designed to suppress relevant and unfavorable information from electoral or congressional scrutiny.[147]

Self-dealing: The representation-reinforcing justification for an immunity shield also dissipates for presidential misconduct motivated by self-dealing that injures the public interest. Self-dealing captures actions that promote the President’s own interests at the expense of the public she serves, which constitutes a refusal “to take into account the interests of [literally] all those their decisions affect.”[148]

One can also imagine many sorts of self-dealing presidential misconduct. President Trump himself allegedly directed governmental business for personal financial gain by, for example, steering officials, events, and his own entourage to Trump-owned hotels;[149] stopping an FBI Headquarters relocation plan to benefit his hotel business;[150] and pardoning arguably undeserving potential campaign contributors.[151] The definitional boundary might be fuzzy, as some self-dealing might pose limited or perhaps no discernible injuries to the public interest.[152]

Both self-entrenching and self-dealing misconduct might violate criminal and‌/or civil laws, inviting criminal prosecutions or civil damages suits.[153] A sitting President would claim temporary immunity from indictment and further criminal prosecution; a former President might claim permanent criminal immunity with respect to official misconduct; both a sitting and former President would claim permanent civil damages immunity for official misconduct. And, in my view, Ely’s accountability paradigm fails to justify immunity claims in each of these contexts—such presidential actions contravene immunity’s underlying rationale.[154] Permanent immunity would undermine the deterrent effect of anti-entrenching and anti-self-dealing laws; temporary immunity would mute that deterrent effect during the President’s term and perhaps embolden first-term illegal efforts to secure reelection.

Neither the lawsuits in Fitzgerald or Clinton nor the criminal subpoena in Vance appear to have promoted these countervailing interests. But the criminal subpoena enforced in Nixon would qualify, given the underlying self-entrenching goals of Watergate activities.

  1. Incorporating Elyan accountability
    concerns into immunity doctrines
  2. Doctrinal options

Given that Ely’s accountability paradigm supports a shield of presidential immunity in some circumstances and also a sword to defeat that immunity in others, the question remains how one might mesh the countervailing interests into judicial doctrine. Here I preliminarily sketch several options, each worth fuller consideration.

Categorical immunity carve-outs for litigation promoting presidential accountability

As the President highlights cognizable burdens, the prosecutor‌/plaintiff would attempt to demonstrate that the prosecution or suit is designed to punish or seek damages for either self-entrenching or self-dealing misconduct. Upon a motion to dismiss, the prosecutor‌/plaintiff could point to the indictment or pleadings and explain how the misconduct allegedly helped the President secure or stay in power or served her personal interests while injuring the public’s. Perhaps courts might require extra-particular civil pleadings, with extra-plausibility beyond Iqbal requirements,[155] or require the equivalent for indictments, or employ other measures to ensure the litigation’s pro-accountability force. Perhaps this early activity could take place under seal and in camera.

Context-specific balancing of pro- and anti-immunity accountability interests

Under more nuanced options, the fact that a prosecution or suit targets self-entrenching or self-dealing misconduct could place an additional thumb on the scale against immunity rather than trigger a complete override, trumping certain immunity burdens (say, diversion) but not others (say, obstruction). Or perhaps these objectives could pierce certain immunity shields (e.g. civil damages) but not others (criminal sanctions). Perhaps they should carry more weight against official misconduct than against unofficial. Perhaps they should supplement more traditional countervailing interests.[156]

Deeming self-entrenching and self-dealing presidential misconduct to lie beyond the “‘outer perimeter’ of his official responsibility.”[157]

The line between official and unofficial misconduct isn’t always clear, depending on how much weight is placed on general context and how much on specific motives and means. Recall Ely’s comment that obstruction of justice isn’t easily characterized.[158] Self-entrenching or self-dealing aims might automatically, or at least presumptively, qualify the related misconduct as unofficial. This would exclude both Fitzgerald-based civil damages immunity and also permanent criminal immunity if that were otherwise in play.[159]

  1. Concerns with subnational enforcement of national accountability

Of course, to the extent these doctrinal options authorize state‌/local officials or individuals to bring actions to sanction self-entrenching or self-dealing misconduct, one might object that the cure spreads the disease: subnational actors, unaccountable to the nation writ large, will divert the President’s attention from public duties to private defense. If national accountability is prized above all, the part‌/whole objection continues, this norm excludes any non-federal actor from deciding by himself that countervailing interests outweigh the immunity rationale—even if a court would otherwise agree with that assessment, as it did in Clinton and Vance.[160] Only a federal official accountable to the nation, then, should be permitted to wield the shield-piercing sword even when self-entrenchment or self-dealing is alleged.

That might sound fine in theory, but it’s likely feckless in practice. The objection would still permit federal law enforcement officers to prosecute or sue the President, as they are indirectly accountable to the People through their station. But even putting aside case-or-controversy complications[161] and the Department of Justice’s current broad view of immunity,[162] obvious barriers remain. The President’s removal power (even where statutorily constrained), potential self-pardon power (currently contested), and general political authority create at least significant headwinds—if not insuperable barriers—for intra-federal enforcement.[163]

Impeachment offers another national-level check against presidential misconduct: the part‌/whole objection continues that “[n]o single prosecutor, judge, or jury should be able to accomplish what the Constitution assigns to the Congress.”[164] Given bicameralism and the two-thirds supermajority requirement for Senate conviction,[165] in some sense this is a super-whole‌/whole mechanism of accountability.[166] But there is good reason to question the practical availability of this national check as well, given Congresspersons’ strong partisan loyalties to political party (as clearly evidenced in each, and especially the last two, presidential impeachment and trial throughout history).

Whatever one thinks generally of these reasons to suspect that national actors may not vigilantly hold a sitting President accountable, the concern is magnified where the presidential conduct violates Elyan norms in the ways described here. First, to the extent self-entrenching and self-dealing misconduct is deemed unofficial, it might not qualify as an impeachable offense.[167] Second, impeachment might be minimally threatening to a President who is close to leaving office. Third, congressional loyalists of the President’s party might be loathe to challenge or deter self-entrenching acts that, if successful, would help keep their mutual party in power and thus indirectly favor their own political fortunes.

Moreover, while the part‌/whole objection focuses on the part-ness of the prosecutor‌/plaintiff, under my Elyan paradigm at least the cause of action by definition serves the interests of the whole people. It may be one thing for a private plaintiff to burden the President and hence the nation over a personal tort or for a local prosecutor to burden the President over private tax fraud; neither the initiator is nationally accountable nor does the suit directly advance national interests. But if the prosecution‌/suit seeks redress for self-entrenching or self-dealing presidential acts as defined here, we the people do stand to gain—indeed the whole point is that the President is already (allegedly) fighting or ignoring our interests. Put differently, the cure addresses rather than fuels the disease: the Elyan sword may be wielded by various actors, but they are all fighting (in part) to vindicate national interests.[168]

Finally, recall that in Fitzgerald the Court pointedly reserved judgment whether it would balance competing interests differently if “Congress expressly had created a damages action against the President.”[169] Any congressional endorsement of litigation designed to counter presidential entrenchment or self-dealing might further bolster the national representation-reinforcing underpinnings even of part-initiated litigation.


If courts are unfortunately forced to revisit the doctrinal drama of presidential immunity, they will likely face constitutional challenges to criminal indictment and prosecution, which impose more acute burdens than those dismissed in Nixon and Vance. Inevitably the government’s countervailing interests in prosecuting the President (both while sitting and later) will demand more careful consideration.

Whether viewed as a supplement to or a full replacement for the Court’s current efficacy-based rationale, the Elyan accountability paradigm proposed here should prompt the Supreme Court to revisit and refine its articulated concerns for presidential harassment and also to ask new questions about law enforcement objectives. Democracy and Distrust’s theory of judicial review insists that courts give great weight to twin representation-reinforcing interests: (a) thwarting leaders’ channel-choking efforts to grab or stay in power and (b) ensuring leaders represent the interests of all of their constituents. This means criminal prosecutions (and suits for civil damages) that target self-entrenching and self-dealing presidential misconduct directly “serve broad public interests . . . not in derogation of the separation of powers, but to maintain their proper balance.”[170] The Supreme Court’s acknowledgment that such swordbearers promote rather than merely degrade democratic accountability would be a fitting testament to John Hart Ely’s greatest work and legacy.


        [1].    Branch Rickey Collegiate Professor of Law and former Dean, University of Michigan Law School. I am grateful to Jason Mazzone for including me in this wonderful (if socially distanced) Symposium, and to Jonah Rosenbaum and Vik Amar for thought-provoking conversations to which I hope this Essay does justice.

        [2].    John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (1980).

        [3].    Id. at vii.

        [4].    Id. at 87.

        [5].    See, e.g., Gerald E. Lynch, Democracy and Distrust: A Theory of Judicial Review, 80 Colum. L. Rev. 857 (1980) (book review) (praising yet critiquing thesis); Richard A. Posner, Democracy and Distrust Revisited, 77 Va. L. Rev. 641 (1991) (book review) (same).

        [6].    For a complicated set of historical, political, and intellectual reasons, “interpretivism” has largely flourished over the past four decades by better addressing Ely’s general critique.

        [7].    The book “is written against the paradigm of judicial review of a decision ultimately traceable to legislative action.” Ely, supra note 2, at 4 n.*. Ely pokes at federalism by decrying state discrimination against nonresidents and federal entities, id. at 83‒87, and at separation of powers by advocating an invigorated nondelegation doctrine, id. at 131‒34, but otherwise stays in his self-described lane.

        [8].    Id. at 181.

        [9].    140 S. Ct. 2412 (2020).

      [10].    United States v. Burr, 25 F. Cas. 30 (C.C.D. Va. 1807) (No. 14,692d) (enforcing subpoena for President Jefferson’s letter and documents); United States v. Nixon, 418 U.S. 683 (1974) (enforcing subpoena for President Nixon’s tape-recorded Oval Office conversations).

      [11].    Clinton v. Jones, 520 U.S. 681 (1997) (rejecting President Clinton’s claim of temporary immunity from a private suit alleging sexual harassment).

      [12].    Nixon v. Fitzgerald, 457 U.S. 731 (1982) (recognizing permanent immunity from such redress).

      [13].    Vance, 140 S. Ct. at 2425.

      [14].    Id.

      [15].    Id. at 2431.

      [16].    See, e.g., Fitzgerald, 457 U.S. at 751 (“raise unique risks to the effective functioning of government”); Clinton, 520 U.S. at 701–02 (“hamper the performance of his official duties” or “impair the effective performance of his office”); Vance, 140 S. Ct. at 2425 (“ability to perform his vital functions” and “impede the President’s execution of [the] laws”); id. at 2426 (“burden on a President’s ability to perform his Article II functions”); id. at 2427 (“undermine his leadership at home and abroad” with “consequences for a President’s public standing”); id. at 2431 (Kavanaugh, J., concurring) (“interest in performing his or her duties without undue interference”); id. at 2432 (“interferes with the President’s duties”); id. at 2433 (“unduly burden or interfere with a President’s official duties”). See also A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222, 245 (2000) (“prevent the executive from accomplishing its constitutional functions”).

Scholars also generally focus on presidential efficacy. See, e.g., Scott W. Howe, The Prospect of a President Incarcerated, 2 Nexus 86, 87 (1997) (“prevent the President from performing with maximum effectiveness”); Akhil Reed Amar & Brian C. Kalt, The Presidential Privilege Against Prosecution, 2 Nexus 11, 13 (1997) (“impact on the well-being of the nation”). For a nod to accountability norms as well, see id. at 14 (immunity “designed to protect . . . the public interest of the People―all the People of America―to have their chosen leader able to execute his duties ‘for their benefit’”).

      [17].    I say “plausibly constructed” because Ely’s own account of representation reinforcement is itself somewhat thin as well as directed toward answering a different set of questions; I again remind that “[t]he elaboration of a representation-reinforcing theory of judicial review could go many ways.” Ely, supra note 2, at 181.

      [18].    Id. at 103.

      [19].    As of this date the precise scope of the criminal investigation remains confidential, though on remand the Court of Appeals made clear that it might extend to financial matters that go well beyond Trump’s alleged payments to his former lawyer Michael Cohen to use as hush money to quiet women who claimed to have had extramarital affairs with Trump. See Trump v. Vance, 977 F.3d 198, 208‒10 (2d. Cir. 2020).

      [20].    For such purposes, the Court in Vance described the subpoena served on accounting firm Mazars as “functionally . . . issued to the President.” Vance, 140 S. Ct. at 2425 n.5.

This raises an interesting question as to why President Trump sued in his personal rather than official capacity, given that the claimed injuries and legal interests were attached to his office. See Trump v. Vance, 395 F. Supp. 3d 283, 288 n.1 (S.D.N.Y. 2019) (noting but not resolving this issue).

      [21].    Clinton, 520 U.S. at 697.

      [22].    Id. at 699; see id. at 713 (Breyer, J., concurring) (“[T]he President never adjourns.”).

This assumption is obviously indulgent; sitting presidents necessarily attend to some private interests, be they personal, family, or political (fundraising and campaigning), and thus private litigation burdens theoretically could trade off against those private activities rather than public duties. As the Court itself admitted, “we suspect that even in our modern era there remains some truth to Chief Justice Marshall’s suggestion that the duties of the Presidency are not entirely ‘unremitting.’” Id. at 699 (citing United States v. Burr, 25 F. Cas. 30, 34 (C.C.D. Va. 1807) (No. 14,692d)). Indeed, in its extreme form the 24/7-public-duties thesis would raise separation of powers questions about a myriad of general private-capacity obligations that presidents must meet, including filing income tax returns.

The proper assumption here is that presidents must be available 24/7 for public duties: “Constitutionally speaking, the President never sleeps.” Akhil Reed Amar & Neal Kumar Katyal, Executive Privileges and Immunities: The Nixon and Clinton Cases, 108 Harv. L. Rev. 701, 713 (1995).

      [23].    Clinton, 520 U.S. at 691‒92.

      [24].    Id. at 691‒92.

      [25].    Vance, 140 S. Ct. at 2426.

      [26].    A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222, 247‒49, 251‒53 (2000).

      [27].    Clinton, 520 U.S. at 705 n.40.

      [28].    Id. at 702; id. at 705 n.40 (“[L]itigation distractions may be vexing” but “do not ordinarily implicate constitutional separation-of-powers concerns.”). By comparison, the Court opined, private litigation defense “surely cannot be considered as onerous as the direct burden imposed by judicial review and the occasional invalidation of his officials actions.” Id. at 705.

      [29].    Vance, 140 S. Ct. at 2426; see id. (“[T]he burden [of subpoena compliance] will ordinarily be lighter than the burden of defending against a civil suit.”).

      [30].    Id. at 2427.

      [31].    Id. The Court, while withholding comment on President Trump’s other sources of self-inflicted wounds, hinted that any stigmatic injury here was fueled in part by “the current suit [which] has cast the Mazars subpoena into the spotlight.” Id.

      [32].    Id. (citing Branzburg v. Hayes, 408 U.S. 665, 691 (1972)).

      [33].    Id.

      [34].    Id.

      [35].    Id. (citing Nixon v. Fitzgerald, 457 U.S. 731, 753 (1982)).

      [36].    Id. (citing Clinton v. Jones, 540 U.S. 681, 708 (1997)). The Court responded that most frivolous civil litigation “is terminated at the pleading stage or on summary judgment, with little if any personal involvement by the defendant,” Clinton, 520 U.S. at 708, and Rule 11 sanctions significantly deter suits against the president “for purposes of political gain or harassment.” Id. at 708‒09.

Notably, the Court has waffled on these claims since Clinton. Cf. Cheney v. U.S. Dist. Ct., 542 U.S. 367, 386 (2004) (protecting Vice President Cheney from broad discovery orders in part because “[a]lthough under [Rule 11], sanctions are available, and private attorneys also owe an obligation of candor to the judicial tribunal, these safeguards have proved insufficient to discourage the filing of meritless claims against the Executive Branch”).

      [37].    Vance, 140 S. Ct. at 2428. Both criminal venue requirements, U.S. Const. amend. VI, and ethical rules prohibiting indictment “that the prosecutor knows is not supported by probable cause,” Model Rules of Pro. Conduct r. 3.8(a) (Am. Bar Ass’n 2020) (adopted by almost every state), further minimize the risk of vexatious prosecutions.

      [38].    Vance, 140 S. Ct. at 2428 (quoting United States v. R. Enter., Inc., 498 U.S. 292, 299 (1991) and then referencing New York state law to the same effect).

      [39].    Vance, 140 S. Ct. at 2428 (endorsing President Trump’s effort to enjoin the state subpoena via federal court injunction, such that “a President would be entitled to the protection of federal courts” when challenging state criminal process).

      [40].    Given these safeguards, the harassment-qua-frivolous concern at most supports case-specific scrutiny, not blanket immunity. Id. at 2429.

      [41].    Id. at 2428.

      [42].    Id.

      [43].    Id. at 2430.

      [44].    Id. at 2428; id. (president may “challenge any allegedly unconstitutional influence”).

      [45].    Id. (citing McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 427 (1819)) (emphasis added). Later the Court specifically separated results from effort, rejecting heightened federalism concerns “if the state subpoena is not issued to manipulate . . . and the Executive is not impaired.” Vance, 140 S. Ct. at 2429‒30 (internal references omitted) (emphasis added). And after rejecting Trump’s pleas for absolute immunity, the Court specifically invited two distinct subpoena-specific challenges: unmask the subpoena “as an attempt to influence the performance of his official duties, in violation of the Supremacy Clause,” or show compliance would “‘significantly interfere with his efforts to carry out’ [his] duties.” Id. at 2430–31 (citing Clinton v. Jones, 520 U.S. 681, 714 (1997)) (emphasis added). Upon remand, Trump did neither, and so the court of appeals offered no clarity. Trump v. Vance, 977 F.3d 198, 208 (2d. Cir. 2020).

For similarly unexplained separation of attempts and result, see Vance, 140 S. Ct. at 2433 (Kavanaugh, J. concurring) (noting state prosecutor may not issue subpoena “to manipulate, influence, or retaliate against a President’s official acts or policy decisions, or in a way that would impede, conflict with, or interfere with a President’s official duties”) (emphasis added); id. at 2447 (Alito, J., dissenting) (“[Doctrinal] rule should take into account both the effect of subpoenas on the functioning of the Presidency and the risk that they will be used for harassment.”) (emphasis added).

      [46].    See Franklin v. Massachusetts, 505 U.S. 788, 827 (1992) (Scalia, J., concurring) (arguing the “President and the Congress (as opposed to their agents)―may not be ordered to perform particular executive or legislative acts at the behest of the Judiciary,” a constraint that precludes “issu[ing] a declaratory judgment against the President”).

      [47].    See Nixon v. Fitzgerald, 457 U.S. 731, 744‒48 (1982) (describing tiers of immunity for various federal and state officers and legislators).

Interestingly, this regime rests in superficial tension with the ascendant “unitary executive theory” according to which the president has ubiquitous control over executive branch activities, as it suggests that a subordinate official might be liable for violating a clear legal obligation (thus vitiating qualified immunity) even if he acts on direct orders from the president.

      [48].    Id. at 749–50.

      [49].    Id. at 749.

      [50].    Id. at 750 n.31 (acknowledging lack of textual basis for this and all other executive and judicial immunities).

      [51].    Id. at 752 (citing first Pierson v. Ray, 386 U.S. 547, 554 (1967), then Ferri v. Ackerman, 444 U.S. 193, 203 (1979)).

      [52].    Id. at 752 n.32.

      [53].    Clinton v. Jones, 520 U.S. 681, 694 n.19; see also Fitzgerald, 457 U.S. at 751.

      [54].    See Clinton, 520 U.S. at 693 (permanent immunity protects officials from “fear that a particular decision may give rise to personal liability”); Trump v. Vance, 140 S. Ct. 2412, 2426 (2020) (“[T]he ‘dominant concern’ in Fitzgerald was . . . the distortion of the Executive’s ‘decisionmaking process’ with respect to official acts that would stem from ‘worry as to the possibility of damages’” (citing Fitzgerald, 520 U.S. at 694 n.19)). The Court in Fitzgerald further worried that, absent broad damages immunity, a president would be “subject . . . to trial on virtually every allegation that an action was unlawful, or was taken for a forbidden purpose,” which “would deprive absolute immunity of its intended effect.” Fitzgerald, 457 U.S. at 756.

      [55].    Fitzgerald, 457 U.S. at 756; see Clinton, 520 U.S. at 694 (chilling effect concern “provides no support for an immunity for unofficial conduct”). That’s why permanent immunity was not at issue in Clinton: with the possible exception of a state-law defamation charge, “the alleged misconduct of [the president] was unrelated to any of his official duties” and “occurred before he was elected to that office.” Id. at 686.

      [56].    The Court never addressed Fitzgerald’s claim that he was unlawfully dismissed in retaliation for truthful congressional testimony, even though the Civil Service Commission ruled that Fitzgerald’s dismissal violated civil service regulations. Fitzgerald, 457 U.S. at 737‒38; id. at 762 (Burger, C.J., concurring) (“Although the individual who claims wrongful conduct may indeed have sustained some injury, the need to prevent large-scale invasion of the Executive function by the Judiciary far outweighs the need to vindicate the private claims.”).

      [57].    The Impeachment Judgment Clause states that an impeachable officer (elsewhere defined to include the “President, Vice President and all civil Officers of the United States,” U.S. Const. art. II, § 4) who is removed from office upon impeachment and conviction “shall nevertheless be liable and subject to Indictment, Trial, Judgment, and Punishment, according to Law.” U.S. Const. art. I, § 3, cl. 7. The Office of Legal Counsel has opined generally that this clause extends to the President. See Whether a Former President May Be Indicted and Tried for the Same Offenses for Which He was Impeached by the House and Acquitted by the Senate, 24 Op. O.L.C. 110, 114 (2000).

      [58].    See, e.g., Fitzgerald, 457 U.S. at 754 n.37 (“The Court has recognized before that there is a lesser public interest in actions for civil damages than, for example, in criminal prosecutions.”); Cheney v. U.S. Dist. Ct., 542 U.S. 367, 384 (2004) (as against claims of privilege, “the need for information in the criminal context is much weightier” than in civil cases).

      [59].    See, e.g., Cheney, 542 U.S. at 386 (describing various criminal procedure safeguards including budgetary and ethical considerations that “filter out insubstantial legal claims” more robustly than do safeguards against meritless civil lawsuits).

      [60].    See, e.g., Eric M. Freedman, Achieving Political Adulthood, 2 Nexus 67, 75 (1997) (“[T]he existing legal structure for control of virtually all federal and state policymaking officers combines broad civil immunity with no criminal immunity.”); Imbler v. Pachtman, 424 U.S. 409, 429 (1976) (“This Court has never suggested that the policy considerations which compel civil immunity for certain governmental officials also place them beyond the reach of the criminal law.”); United States v. Gillock, 445 U.S. 360, 372 (1980) (“[T]he cases in this Court which have recognized an immunity from civil suit for state officials have presumed the existence of federal criminal liability as a restraining factor on the conduct of state officials.”). As a result, “any justification for granting the President an immunity from criminal prosecution must rest on theoretical or practical considerations unique to that office.” Freedman, supra, at 73.

      [61].    For example, Alexander Hamilton claimed that the president could be removed from office through impeachment and conviction “and would afterwards be liable to prosecution and punishment in the ordinary course of law.” The Federalist No. 69, at 416 (Alexander Hamilton) (Clinton Rossiter ed., 1961); see also The Federalist No. 77, at 464 (Alexander Hamilton) (Clinton Rossiter ed., 1961).

      [62].    The scholarly debate over whether sitting presidents enjoy temporary immunity would be immaterial if presidents enjoy absolute and permanent immunity. See, e.g., Amar & Kalt, supra note 16, at 11 (supporting temporary immunity but noting presidents “can be held accountable for their actions after they leave office, and they can be impeached to hasten this”); Freedman, supra note 60, at 84 (rejecting temporary immunity and adding that “the President is amenable to prosecution, and has no ‘generalized’ criminal immunity”).

      [63].    A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222, 255 (2000) (“Recognizing an immunity from prosecution for a sitting President would not preclude such prosecution once the President’s term is over or he is otherwise removed from office by resignation or impeachment.”).

      [64].    See infra note 100.

      [65].    Trump v. Vance, 140 S. Ct. 2412, 2426‒27 (2020); see also (now-Justice) Brett M. Kavanaugh, Separation of Powers During the Forty-Fourth Presidency and Beyond, 93 Minn. L. Rev. 1454, 1462 (2009) (“Moreover, an impeached and removed President is still subject to criminal prosecution afterwards.”).

      [66].    Nixon v. Fitzgerald, 457 U.S. 731, 756 (1982).

      [67].    Id. at 750.

      [68].    Id. at 780 (White, J., dissenting) (“[T]here is no contention that the President is immune from criminal prosecution in the courts under the criminal laws enacted by Congress or by the States for that matter. Nor would such a claim be credible.”).

      [69].    Id. at 758; see id. at 757 (listing impeachment, press scrutiny, congressional oversight, and the president’s concerns for reelection, prestige, and historical stature). This omission is particularly notable because in refuting the “above the law” concern the Court specifically cited Imbler v. Pachtman, 424 U.S. 409, 428‒29 (1976) to assure that civil damages immunity “does not leave the public powerless to deter misconduct or to punish that which occurs,” id. at 757 n.38. But Imbler specifically relied first and foremost on criminal liability to support that proposition. 424 U.S. at 429.

      [70].    Kavanaugh, supra note 65, at 1462.

      [71].    Few of the historical, scholarly, or governmental sources referenced above specifically discuss a possible distinction between official and unofficial misconduct; perhaps most have in mind criminal misconduct that would be labeled “unofficial” and thus beyond any Fitzgerald-mimicking coverage. Indeed, much of the debate over temporary immunity was occasioned by President Nixon’s and President Clinton’s alleged obstruction of justice, plausibly viewed on their facts as unofficial misconduct. But some do recognize and honor the distinction, including Ely. See John Hart Ely, On Constitutional Ground 139 (1996) (recognizing that “some [potentially criminal misconduct] will be fairly classifiable as not performed in an official capacity . . . [i]ndeed, it is not altogether clear where conspiracy to obstruct justice should be placed”). With respect to whether official misconduct would be subject to permanent immunity, compare Benjamin G. Davis, United or Untied: On Confronting Presidential Criminality in the Savage Wars of Peace, 84 Tenn. L. Rev. 671, 676 (2017) (noting prosecutor charging official misconduct “would have to address the kinds of well-recognized functional immunity jurisprudence leading to dismissal of the federal prosecution”) with Howe, supra note 16, at 94 (doubting permanent prosecutorial immunity for “actions related to certain official functions” because it “raises far greater concerns than for temporary immunity”).

      [72].    Nixon v. Adm’r of Gen. Servs., 433 U.S. 425, 443 (1977); Fitzgerald, 457 U.S. at 754 (“[A court] must balance the constitutional weight of the interest to be served against the dangers of intrusion on the authority and functions of the Executive Branch.”).

      [73].    Fitzgerald, 457 U.S. at 754.

      [74].    Id. at 758.

      [75].    Id. at 757. In Clinton, the Court never reached the interest-balancing stage because it found no cognizable presidential burden. Clinton v. Jones, 520 U.S. 681, 701‒06 (1997).

      [76].    United States v. Nixon, 418 U.S. 683, 711, 707 (1974); see id. at 709 (“The ends of criminal justice would be defeated if judgments were to be made on a partial or speculative presentation of the facts.”). Nixon’s focus on “relevant and admissible” evidence significantly qualifies this countervailing interest; many constitutional and evidentiary rules constrain admissibility.

      [77].    Trump v. Vance, 140 S. Ct. 2412, 2430 (2020).

      [78].    Id. (citing United States v. R. Enters., Inc., 498 U.S. 292, 297 (1991)). Cf. id. at 2432 (Kavanaugh and Gorsuch, JJ., concurring) (favoring Nixon test “to balance the State’s interests and the Article II interests”); id. at 2448‒49 (Alito, J., dissenting) (advocating more nuanced balancing standard).

      [79].    Id. at 2430.

      [80].    See Nixon, 418 U.S. at 708‒09 (“[O]ur historic commitment to the rule of law . . . is nowhere more profoundly manifest than in our view that ‘the twofold aim [of criminal justice] is that guilt shall not escape or innocence suffer.’”) (citing Berger v. United States, 295 U.S. 78, 88 (1935)); id. at 713 (“Without access to specific facts a criminal prosecution may be totally frustrated.”); Vance, 140 S. Ct. at 2430 (subpoena delay “could frustrate the identification, investigation, and indictment of third parties”).

      [81].    To recap: permanent immunity from civil damages claims for official misconduct; no immunity (at least for her subordinate officials) for injunctive relief; and no permanent or temporary immunity from lawsuits alleging unofficial misconduct, assuming courts accommodate presidential schedules to avoid time- and space-bound diversions.

      [82].    See supra text accompanying notes 57‒65.

      [83].    Nixon v. Fitzgerald, 457 U.S. 731, 749 (1982).

      [84].    See supra text accompanying note 24.

      [85].    A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222, 251‒52 (2000).

      [86].    Depending on the crime the president’s sentence could be limited to fines and/or probation, or a prison sentence could be stayed until the president leaves office (whether at the end of a term or via earlier impeachment). But wonders of modern communications technology notwithstanding, prison bars would seem quintessentially obstructive.

      [87].    See A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. at 254 (emphasizing the criminal/civil distinction, though rejecting a case-by-case assessment in favor of categorically assuming that the “physical and mental burdens imposed by an indictment and criminal prosecution . . . are of an entirely different magnitude” than subpoena defense or civil litigation).

      [88].    See supra text accompanying notes 35‒45.

      [89].    Trump v. Vance, 140 S. Ct. 2412, 2445 (2020) (Alito, J., dissenting).

      [90].    A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222.

      [91].    For a small sampling of relevant scholarship, see Davis, supra note 71, at 1‒2 n.2.

      [92].    Ely, supra note 71, at 139 & 416 n.190.

      [93].    Id. at 137.

      [94].    Id. at 138, 140.

      [95].    Id. at 139, 140.

      [96].    Id. at 141.

      [97].    Id. at 140.

      [98].    Amar & Kalt, supra note 16.

      [99].    Letter from John Hart Ely to Akhil Amar (on file with Akhil Amar).

    [100].    It’s also apparent from context that Ely dismissed, or at least never contemplated, the idea of a permanent immunity from criminal prosecution for official misconduct. See Ely, supra note 71, at 139 (assuming sitting president would be prosecutable at least post-impeachment if not earlier for official misconduct and supporting prosecution immediately for unofficial misconduct: “one thing that is clear is that the president does not stand above the law”).

    [101].    Ely added a 1996 aside opposing temporary immunity for civil litigation. Referencing Paula Jones’s then-pending suit against President Clinton, he wrote “justice delayed often is justice denied. We’ve come far enough toward the sort of regal presidency the framers abhorred without effectively immunizing behavior of the sort alleged . . . .” Id. at 415 n.183.

    [102].    Ely, supra note 2, at 181.

    [103].    Id. at 76, 181.

    [104].    Id. at 105.

    [105].    Id. at 103.

    [106].    Id. at 105‒34.

    [107].    Id. at 87, 100.

    [108].    Id. at 103; see id. at 84 (such minorities “might just as well be disenfranchised”).

    [109].    Id. at 101. See, e.g., id. at 78 (decision-makers should “live under the regime of the laws they passed and not exempt themselves from their operation”); id. at 82 (democratic principles “preclude a refusal to represent . . . groups that constitute minorities of the population”); id. at 135 (officials “are in a position to vote themselves advantages at the expense of the others, or otherwise to refuse to take their interests into account”).

    [110].    See supra note 16 and accompanying text.

    [111].    Obviously the “national interest” is a contestable construct with contestable content. And the President’s definition of national interest is constrained by valid law, enforceable through injunctive relief. See supra text accompanying note 46.

    [112].    Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2203 (2020) (citing Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 496–97 (2010)).

    [113].    Ely, supra note 2, at 77 (“It is a principle of general application that the exercise of a granted power to act in behalf of others involves the assumption toward them of a duty to exercise the power in their interest and behalf . . . .” (quoting Steele v. Louisville & Nashville R.R. Co., 323 U.S. 192, 202 (1944))).

    [114].    While Professors Amar and Kalt focus primarily on efficacy concerns, they also connect these accountability dots. Amar & Kalt, supra note 16, at 14 (stating temporary immunity is designed to protect the “public interest of the People―all the People of America―to have their chosen leader able to execute his duties ‘for their benefit.’ This right of all the People to a functioning government trumps the right of only a few of them to have an instant prosecution.”).

    [115].    See supra text accompanying notes 103–109; see also Ely, supra note 2, at 82 (noting that representation-reinforcement “cannot mean that groups that constitute minorities of the population can never be treated less favorably than the rest, but it does preclude a refusal to represent them”).

    [116].    See supra text accompanying notes 49–56.

    [117].    If they are not legal, then suits for prospective relief (whether aimed at her or her subordinates) can surgically deter them. This constraint on presidential discretion doesn’t itself undermine her accountability to the people so long as the legal restriction itself reflects an appropriately accountability process (such as fair and responsive congressional lawmaking).

    [118].    Nixon v. Fitzgerald, 457 U.S. 731, 748 n.27 (1982).

    [119].    See supra text accompanying notes 57–71.

    [120].    Trump v. Vance, 140 S. Ct. 2412, 2445 (2020) (Alito, J., dissenting).

    [121].    See supra note 86.

    [122].    One might ask whether impeachment raises a similar accountability concern; as Ely himself noted, impeachment triggers the same presidential diversions. Ely, supra note 71, at 139 and 416 n.190. Some suggest the diversion is different in kind, as the president is more accustomed to dealing with Congress than with opposing prosecutors. See Amar & Kalt, supra note 16, at 19 (“[T]he President is already institutionally equipped to deal with Congress; while impeachment is a rare event, it is much closer to the regular business of Presidents than is a criminal prosecution.”). But it’s worth recognizing that the president is still diverted to the same degree from her accountability to the people (as well as hampered to the same degree in her job performance), no matter the source of the diversion. See Clinton v. Jones, 520 U.S. 681, 705 n.40 (1997) (lumping together for diversion purposes “a variety of demands on their time, however, some private, some political, and some as a result of official duty”). An alternative accountability-based distinction is that prosecution‌/litigation diverts the president’s attention to an individual actor’s agenda, whereas impeachment diverts the president’s focus merely from one version of national accountability (her unhindered view of the national interest driven by the people’s unmediated concerns) to another version (the people’s concerns as filtered through their impeaching congressional representatives). Of course the clearest distinction between litigation and impeachment process is that the latter is explicitly authorized by the Constitution and clearly applies to the president. U.S. Const. art. II, § 4.

    [123].    Of course no amount of time and effort might overcome certain types or degrees of stigmatic injury, just as no amount of time and effort can overcome certain obvious and significant disabilities imposed by incarceration. To my mind, these types of burdens don’t map well onto an accountability framework even under this broader perspective though they remain prominent and potentially dispositive in an efficacy-based approach.

    [124].    See supra text accompanying notes 35–45.

    [125].    See supra text accompanying note 40.

    [126].    Trump v. Vance, 140 S. Ct. 2412, 2428 (2020) (citing McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 427 (1819)).

    [127].    Id.

    [128].    See supra text accompanying notes 51–54.

    [129].    In this respect, the Court’s argument in Vance logically supports a claim of permanent and not just temporary immunity. A prosecutor can threaten to subpoena or prosecute a president after she leaves office if she fails to support the prosecutor’s agenda while in office; courts could thwart this threat only by protecting the president permanently. So at minimum the Court appears to make a category error here, even though admittedly Trump didn’t take the argument this far.

    [130].    Defending against civil suits seeking injunctive relief can be costly in both monetary and other ways (for example, litigation might galvanize opposition, undermine public relations, and invite potentially embarrassing discovery).

    [131].    It may be fair to put into this category, for example, Texas’s and other states’ persistent legal efforts to frustrate the Affordable Care Act, which seem to have broader political and policy goals than merely enjoining assertedly illegal provisions. See, e.g., AG Paxton and Wisconsin AG File 20-State Lawsuit to End the Grip fo Obamacare on Texas and the Nation, Tex. Att’y Gen. (Feb. 26, 2018),‌/news‌/releases‌/ag-paxton-and-wisconsin-ag-file-20-state-lawsuit-end-grip-obamacare-texas-and-nation (Texas Attorney General Paxton claiming that “[t]hrough our multi-state lawsuit, we hope to effectively repeal Obamacare, which will then give President Trump and Congress an opportunity to replace that failed experiment”).

    [132].    State‌/local governments may resist presidential policies by exercising their own sovereign powers or federally-delegated powers. For the former, think sanctuary cities. For many examples of the latter, see Jessica Bulman-Pozen & Heather K. Gerken, Uncooperative Federalism, 118 Yale L.J. 1256, 1307 (2009) (describing and defending how states “use their power as federal servants to resist, challenge, and even dissent from federal policy”).

    [133].    See, e.g., Shirin Ghaffary, Google Employees are Demanding an End to the Company’s Work with Agencies like CBP and ICE, Vox (Aug. 14, 2019),‌/2019‌/8‌/14‌/20805562‌/human-rights-concerns-google-employees-petition-cbp-ice (employees petition Google to eschew bidding on government cloud computing contract to protest immigration border policies).

    [134].    Perhaps we share a cultural norm that criminal legal process is simply “off limits” whenever motivated by factors extraneous to criminal justice. This strikes me as intuitively plausible, akin to our shared (I hope) belief that kidnapping-as-leverage is off-limits. But such a norm seems orthogonal to a focus on presidential efficacy, which is perhaps why it is not articulated in Vance.

    [135].    At least putting aside concerns that certain individuals or companies might have inordinate lobbying or campaigning power due to connections or wealth, factors that Ely might view as anathema to a fair system.

    [136].    I easily reject the suggestion, voiced in Trump’s defense during his first impeachment trial, that saving the president’s own hide by definition counts as serving the public interest because the president believes he is better than his replacement. See Vikram David Amar & Evan Caminker, The Real Insidious Part of Dershowitz’s Impeachment Defense, Justia: Verdict (Feb. 4, 2020),‌/2020‌/02‌/04‌/the-real-insidious-part-of-dershowitzs-impeachment-defense.

    [137].    Ely briefly develops this theme when promoting a more rigorous nondelegation doctrine: “by refusing to legislate, our legislators are escaping the sort of accountability that is crucial to the intelligible functioning of a democratic republic.” Ely, supra note 2, at 132.

    [138].    State and local prosecutorial threats may also raise an Elyan concern within the state polity itself. To the extent a prosecutor serves a nonfrivolous subpoena on the president because he views state citizens to be victims of her bad pandemic policy rather than victims of her tax-dodging, he likely acts ultra vires. That’s a way of saying the prosecutor evades accountability to his own state or local constituency, which chose and empowered him to promote criminal justice rather than social policy.

    [139].    See supra note 37; it’s worth remembering the strict geographical limits on prosecutorial jurisdiction mentioned there as well.

    [140].    A case-specific challenge also faces obvious hurdles. If a prosecutor subpoenas or indicts the president, presumably that means the president refused to comply with the prosecutor’s earlier threat; had the president succumbed, the prosecutor would have had no reason to “retaliate.” So the most the president could claim in defense is “the prosecutor tried to influence me; I stood firm; and now he’s retaliating.” But we must also assume the prosecutor has a good-faith basis for his subpoena or indictment notwithstanding his retaliatory motive; otherwise this would fall into the frivolousness category discussed above. If so, what’s the problem? Now we have a failed attempt to influence married to a subpoena or indictment satisfying conventional evidentiary standards. In this context the “harassment” concern seems to have little purchase from either an efficacy-based or Elyan perspective.

    [141].    Nixon v. Fitzgerald, 457 U.S. 731, 754 (1982).

    [142].    See supra text accompanying notes 76–80.

    [143].    For some, the Court’s balancing test supports “indictment but no more.” The Department of Justice disagrees: “the better view of the Constitution accords a sitting President immunity from indictment by itself.” A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222, 259 (2000). Addressing the concern that the relevant statutes of limitations might run, OLC’s opinion maintains that “[a]t most . . . prosecution would be delayed rather than denied, id. at 256, for either of two reasons. First, for the “most serious criminal wrongdoing” the president might face immediate impeachment “permitting criminal prosecution at that point.” Id. And second, “whether or not it would be appropriate for a court to hold that the statute of limitations was tolled while the President remained in office (either as a constitutional implication of temporary immunity or under equitable principles), Congress could overcome any such obstacle by imposing its own tolling rule.” Id. Although I participated in drafting this opinion as a Deputy Assistant Attorney General, I find persuasive Professor Jed Shugerman’s recent challenge to the opinion’s noncommittal view on whether judicial equitable tolling might be available. Professor Shugerman observes that lower courts have not embraced equitable tolling in criminal cases over the past two decades, and he further explains why tolling’s “equity” might fall short due to investigation-specific facts. Jed Shugerman, The Single Fatal Flaw in the Legal Argument Against Indicting a Sitting President, Slate (Dec. 11, 2018),‌/news-and-politics‌/2018‌/12‌/trump-indict-sitting-president-statute-of-limitations.html. Absent a tolling option, the government interest in indicting within the limitations period strengthens considerably (even if filed under seal and with further proceedings stayed).

Professor Shugerman does not similarly question the point on which the OLC opinion actually rests, that Congress could impose its own tolling rule. But the OLC Memo addresses only federal and not state prosecution, A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. at 223 n.2, and thus it does not directly consider the more complicated question whether Congress may toll or otherwise override the normal operation of state statutes of limitation.

    [144].    Ely’s rights-focused doctrinal discussions suggest that he generally favors categorical over case-by-case reasonableness balancing. E.g., Ely, supra note 2, at 105‒16 (free speech).

    [145].    Id. at 117.

    [146].    The Ukraine allegation triggered President Trump’s impeachment, of which he was acquitted. See Seung Min Kim, In Historic Vote, Trump Acquitted of Impeachment Charges, Wash. Post (Feb. 5, 2020),‌/politics‌/in-historic-vote-trump-acquitted-of-impeachment-charges‌/2020‌/02‌/05‌/8b7ea90e-4832-11ea-ab15-b5df3261b710‌_story.html. The Capitol storming allegation triggered Trump’s second impeachment, of which he was also acquitted. See Amy Gardner, Mike DeBonis, Seung Min Kim & Karoun Demirjian, Trump Acquitted on Impeachment Charge of
Inciting Deadly Attack on the Capitol
, Wash. Post (Feb. 13, 2021),‌/politics‌/trump-acquitted-impeachment-riot‌/2021‌/02‌/13‌/dbf6b172-6e12-11eb-ba56-d7e2c8defa31‌_story.html. Several of these allegations are currently the subject of pending civil litigation and‌/or state or local criminal investigations. See Karl Mihm, Jacob Apkon & Sruthi Venkatachalam, Litigation Tracker: Pending Criminal and Civil Cases Against Donald Trump, Just Sec. (updated July 6, 2021),‌/75032‌/litigation-tracker-pending-criminal-and-civil-cases-against-donald-trump‌/.

    [147].    Trump’s first impeachment included obstruction of justice charges for refusing to comply with congressional subpoenas. See supra note 146. Trump’s admitted hush-money payments to women alleging affairs, see supra note 19, were designed to suppress unfavorable information prior to the 2016 election.

    [148].    Ely, supra note 2, at 100 (emphasis added).

    [149].    See, e.g., David A. Fahrenthold, Josh Dawsey & Joshua Partlow, Room Rentals, Resort Fees and Furniture Removal: How Trump’s Company Charged the U.S. Government More than $900,000, Wash. Post (Aug. 27, 2020),‌/politics‌/trump-company-secret-service-spending‌/2020‌/08‌/27‌/9331bd86-de36-11ea-8051-d5f887d73381‌_story.html; Robert Costa, David A. Fahrenthold & John Wagner, Trump Encouraged Pence to Stay at His Golf Resort in Ireland, Wash. Post (Sept. 3, 2019),‌/politics‌/trump-encouraged-pence-to-stay-at-his-golf-resort-in-ireland‌/2019‌/09‌/03‌/a2dc63c4-ce3f-11e9-b29b-a528dc82154a‌_story.html.

    [150].    See, e.g., Lauren Fox, Trump More Involved in Stopping FBI HQ Move than Previously Known, Emails Show, CNN (Oct. 19, 2018),‌/2018‌/10‌/18‌/politics‌/trump-fbi-building‌/index.html.

    [151].    See Michael Biesecker, Trump Picks Pardon Requests from Wealthy Pals and GOP Donors, AP News (Feb. 19, 2020),‌/article‌/illinois-tx-state-wire-election-2020-nfl-nv-state-wire-5f5462f4b8a41a63e0235e7dff33295a.

    [152].    For example, personal tax evasion by a sitting president indirectly injures the public fisc; steering government business may self-enrich without costing additional public monies.

    [153].    See, e.g., Mihm, Apkon & Venkatachalam, supra note 146.

    [154].    Courts should not countenance the tendentious argument, advanced during Trump’s first impeachment trial, that his solicitation of foreign election assistance actually served the national interest because his reelection would serve the national interest. See Amar & Caminker, supra note 136.

    [155].    Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (federal pleading must “state a claim to relief that is plausible on its face”).

    [156].    See supra text accompanying notes 76–80.

    [157].    Nixon v. Fitzgerald, 457 U.S. 731, 756 (1982).

    [158].    See supra note 71.

    [159].    See supra text accompanying notes 57–71 (discussing whether Supreme Court might conclude a president enjoys such immunity). Vance suggests the official‌/unofficial distinction doesn’t drive temporary criminal immunity doctrine, as the effects on presidential performance are similar.

    [160].    See Amar & Kalt, supra note 16, at 12 (“If the President were prosecuted, the steward of all the People would be hijacked from his duties by an official of few (or none) of them.”). This objection is typically phrased in efficacy concerns, e.g., id. at 14 (“right of all the People to a functioning government”), but it remains apt here.

    [161].    See United States v. Nixon, 418 U.S. 683, 692‒97 (1974) (finding the subpoena litigation justiciable despite being “an intra-branch dispute between a subordinate and superior officer of the Executive Branch”).

    [162].    See A Sitting President’s Amenability to Indictment and Criminal Prosecution, 24 Op. O.L.C. 222 (2000).

    [163].    See Jed Shugerman, The President’s Lawyers Are Making a Dangerous Argument for Presidential Immunity, The Atlantic (Oct. 26, 2019),‌/ideas‌/archive‌/2019‌/10‌/dangerous-arguments-presidential-immunity‌/600727‌/ (DOJ lawyers unlikely to be inclined or permitted to investigate or prosecute sitting presidents).

    [164].    Kavanaugh, supra note 65, at 1462.

    [165].    U.S. Const. art. I, § 3, cl. 6.

    [166].    Though note that Senators representing the least-populated 34 states, meaning 31.64% of the national population, can control the conviction decision (per 2020 Census data).

    [167].    See, e.g., Ely, supra note 71, at 139 (under “prevailing” view, “impeachment is limited to acts performed in an official capacity”).

    [168].    Even if a part‌/whole objector remains unpersuaded that this recognition supports full override of immunity concerns, perhaps it still supports giving the pro-prosecution interests substantial weight in a context-specific balancing. See supra text accompanying note 156.

    [169].    457 U.S. 731, 748 n.27 (1982) (case arose under “implied” causes of action).

    [170].    Id. at 754.