Criminal Immunity and Schrödinger’s President: A Response to Prosecuting and Punishing Our Presidents

Response - Online Edition - Volume 100


In his article, Prosecuting and Punishing Our Presidents, Saikrishna Prakash argues that sitting presidents have no constitutional protection from being arrested, prosecuted, and incarcerated while in office.[1] He depicts the constitutional case for immunity—which he concedes is “orthodoxy” and “[t]he received wisdom”[2]—as an empty vessel, conjured by people with skewed, wishful readings of the Constitution.[3] As one of the alleged conjurers,[4] I appreciate this opportunity to respond.

Prakash is a tremendous scholar, accomplished and principled,[5] and his article is a useful contribution to this centuries-long constitutional debate. But several of his arguments are flawed and much of his historical evidence is oversold or incomplete. In addition, his response to the pro-immunity argument sidesteps key elements of it.

That said, Prakash and I agree that there are reasonable arguments on both sides of the immunity question.[6] Orthodoxy or no orthodoxy, there is currently no way to say with certainty that sitting presidents are or are not immune. Presidents have broken the law without being prosecuted,[7] but presidents are constrained from engaging in more extensive criminality by the possibility that courts might reject immunity. Until the constitutional question is resolved—not by professors debating it but by prosecutors and presidents litigating it—sitting presidents are like Schrödinger’s cat, simultaneously immune and not immune.[8]

Part I of this response critiques Prakash’s anti-immunity argument. Part II turns to Prakash’s treatment of pro-immunity arguments, critiquing it too but also acknowledging its strengths in places and clarifying my own pro-immunity position in response. Part III considers Prakash’s proposals for legislative reform, explains why Congress likely will not act, and contemplates where this leaves us.

I. Prakash’s Case Against Immunity

Prakash makes an aggressive case that the Constitution offers sitting presidents no immunity from the criminal process. He says that immunity is inconsistent with the Constitution’s text, its historical context, and its structure, and is also belied by actual practice. But each part of his case has significant shortcomings.

A. Constitutional Text: Says Too Little, Proves Too Much

The essence of Prakash’s textual argument is that “executive immunities exist when they are explicitly granted”[9] and that the Constitution contains no explicit grant of immunity to sitting presidents.[10] By contrast, the Constitution grants limited immunity to members of Congress; this makes presidents look unprotected in comparison.[11]

I have written that the Constitution’s silence here is the strongest argument against immunity, compelling in its simplicity.[12] But the textual argument proves too much. Presidents and some other non-legislative officials enjoy absolute immunity from civil liability for their official acts, despite the lack of any direct constitutional textual basis.[13] The Constitution also has no clause granting executive privilege, but executive privilege nevertheless exists.[14] The Supreme Court has found these things woven into the Constitution’s structure, notwithstanding the lack of an explicit textual source and notwithstanding the explicit protection the Constitution gives legislators.[15] As the Court put it in one such case:

Noting that the Speech and Debate Clause provides a textual basis for congressional immunity, respondent argues that the Framers must be assumed to have rejected any similar grant of executive immunity. This argument is unpersuasive. . . . [A] specific textual basis has not been considered a prerequisite to the recognition of immunity.[16]

Even when it rejected President Clinton’s claim that he was immune from personal civil suits while in office, the Court recognized that the text was not the exclusive source of constitutional authority.[17] In any criminal-immunity case, the Court presumably would consider non-textual factors and tests. If it found that the Constitution grants criminal immunity to sitting presidents, the Court would just be adding another item to its existing list of non-textual presidential privileges.

Prakash also points to Founding-era state and foreign constitutions, saying they established a standard that privileges are granted only if they are rendered explicitly in the text.[18] But isolated instances of explicitness cannot bootstrap an explicitness requirement. As Prakash notes, some states explicitly provided that sitting governors could be prosecuted while others explicitly provided that sitting governors had immunities.[19] With explicitness on both sides, the remaining states’ silence does not break obviously in either direction. By the same token, when the federal constitution’s drafters remained silent (notwithstanding a stray comment by James Madison that I believe Prakash overreads[20]), their silence did not rule immunity out or in. Rather, it left the question open for later interpretation, by the Supreme Court among others.

After Prakash admits that his arguments are “in tension with certain Supreme Court pronouncements,” he brushes this off, noting that he is not trying to predict how the Court would rule in an actual case.[21] But the contours of presidential immunity affect what presidents and prosecutors actually do. In a real case, it would matter that Prakash’s textual argument clashes with the Court’s immunity jurisprudence.

B. Originalism: Thin Gruel

Prakash offers an extensive originalist argument. Unlike Burlette Carter—whose historical analysis concluded that the Framers understood presidents to have some criminal immunity[22]—Prakash is confident that the Framers thought sitting presidents have no protection.

But Prakash’s Founding-era evidence is insubstantial and overstated. The Framers did not discuss immunity directly; the material here consists mostly of oblique references and asides during discussions about other things. Moreover, much of Prakash’s evidence is ambiguous, consistent with both anti- and pro-immunity interpretations. And while some of the evidence does clearly favor Prakash’s side, he overlooks evidence on the other side.

1. The Founding

Prakash cites discussions about the limits of the presidency compared to the British monarchy:

While British sovereigns could “do no wrong,” meaning they were immune from all judicial process (both impeachment and ordinary criminal trials), the American President would be different. Seeking to refute the notion that the Constitution established a monarchy, “Americanus,” a New Jerseyan supporter of the Constitution, wrote an essay observing that while the king was “above the reach of all Courts of law” and was “sacred and inviolable,” presidents would not be. “[N]one of [these immunities] are vested in the President.” In Virginia, “A Freeholder” argued that presidents would not be monarchs because they could “be impeached and removed at any time; or . . . be indicted if the case should require it.”[23]

But both sides in the immunity debate agree that presidents are unregally subject to criminal prosecution. The question is simply one of timing. Immunity proponents think presidents can be prosecuted, just not until they have left office—which still makes presidents different from unassailable British monarchs in precisely the way Americanus and the Freeholder said.[24]

In other words, these sources can support either position. Neither Americanus nor the Freeholder said anything specific about timing. Perhaps they believed presidents could be prosecuted at any time. But perhaps not. Had they thought presidents could not only be (1) prosecuted after being bounced from office but also (2) subjected to the entirety of the criminal-law process while in office, that second point would have been a big deal—worthy of separate emphasis.

Prakash also quotes Cassius’s statement that a criminal President “would be immediately arrested in his career and summoned to answer for his conduct before a federal court.”[25] But again, immunity proponents fully embrace the scenario of a villainous President losing his job and being dragged into court. There is no basis to read Cassius as envisioning those two things happening in reverse order.

Similarly, Prakash cites James Iredell’s statement that, unlike kings, presidents could be impeached, removed, prosecuted, and even executed for capital offenses.[26] But yet again, the point was that presidents’ subordination to the law stood in marked contrast to the complete and permanent impunity enjoyed by kings.[27] If in expressing this sentiment Iredell meant to echo Prakash’s views on timing—that the Constitution would give any local authority the power to arrest, try, convict, and hang a sitting President—Iredell’s comment would qualify as mightily understated.

Prakash does offer Founding-era statements with a less ambiguous anti-immunity tone. Tench Coxe and James Wilson are his two strongest examples.[28] But the defining feature of this debate was how little discussion there was. Stray statements like Coxe’s and Wilson’s stand out because of how lonely they were, not because they exemplify a widely voiced and firmly held position.

Moreover, there are pro-immunity nuggets in the record too. Akhil Reed Amar and Burlette Carter have noted multiple comments made at the Constitutional Convention and during ratification that have pro-immunity implications.[29] More directly, Alexander Hamilton wrote in Federalist 69 that “[t]he President of the United States would be liable to be impeached, tried, and . . . removed from office; and would afterwards be liable to prosecution and punishment in the ordinary course of law.”[30]

Prakash dismisses Hamilton’s “afterwards,” arguing that Hamilton never said a President could only be prosecuted after impeachment.[31] But Hamilton did say “afterwards” when he could have been silent about timing. The point Hamilton drove home in Federalist 69 was how accountable presidents would be. If Hamilton thought presidents were subject to the criminal process while in office, this would have been a logical place to say so clearly. And in Federalist 77, which Prakash does not mention, Hamilton again references timing when he refers to keeping presidents in check via the possibility of impeachment, removal, and “subsequent prosecution in the common course of law.”[32]

Finally, contrary to Prakash’s claim that no Anti-Federalists fretted about presidential criminal immunity,[33] James Monroe seemingly did just that at the Virginia Ratifying Convention:

[The President] is elected for four years and not excluded from re[e]lection. Suppose he violates the laws and Constitution, or commits high crimes. By whom is he to be tried?—By his own council—by those who advise him to commit such violations and crimes? This subverts the principles of justice, as it secures him from punishment.[34]

The “own council,” to which Monroe referred as the trier of criminal presidents, was the Senate.[35] Monroe worried that the President and Senate would have a collusive relationship, making it unlikely that the Senate would punish a lawbreaking President. Coupled with his reference to the President’s re-electability, Monroe’s complaint that the Senate’s complicity would “secure[] [presidents] from punishment” only makes sense if he thought the impeachment process was the exclusive means of pursuing criminality by sitting presidents.

The issue here is not who was right, Coxe/Wilson or Hamilton/Monroe. Rather, the issue is that the Founding-era evidence on immunity is thinner and more two-sided than Prakash portrays it.

2. The Early Republic

Further two-sidedness emerged from Congress in the Republic’s early years. Prakash cites some statements as exemplifying a consensus view against immunity,[36] such as two comments about prosecuting or jailing presidents, and a seemingly skeptical statement about presidential “privileges.”[37]

But this is thin gruel. The first comment was cryptic and made by someone who, elsewhere, strongly supported immunity.[38] The second comment was a tangent in a discussion of an unrelated technical matter.[39]

The third statement, a lengthy quotation from Senator Charles Pinckney, does not relate even fleetingly to presidential immunity; it only appears so because of a redaction (not by Prakash) that both obscures the context and misses Pinckney’s sarcasm.[40] Prakash depicts Pinckney as denigrating presidential privileges when saying: “Let us inquire, why the Constitution should have been so attentive to each branch of Congress, so jealous of their privileges, and have shewn so little to the President of the United States in this respect. . . . No privilege of this kind was intended for your Executive.”[41] But Pinckney was engaged in a debate about the Senate’s unilateral power to punish unruly journalists—this, not criminal immunity, was the “privilege” of which Pinckney was speaking.[42] Pinckney was saying that the President had no power to imprison without trial those who insult him, so a fortiori the Senate had no such power; his “so little to the President” line was sarcastic.[43]

Meanwhile, Prakash notes that a number of other members of Congress favored presidential immunity, but he dismisses their views as misguided.[44] For something to be a consensus view, however, there needs to be general agreement—not just agreement among the subset of people one deems correct.

The most vivid example of the lack of consensus is a chat between some of the pro-immunity figures Prakash dismisses and Senator William Maclay (whose anti-immunity arguments Prakash features in his subpart on constitutional structure).[45] Those disagreeing with Maclay included Vice President Adams and Senator (later Chief Justice) Oliver Ellsworth—no slouches!—and their arguments were more complex than in Prakash’s depiction.[46]

To sum up the evidence from the Founding era: Some people said pro-immunity things; others said anti-immunity things. There was no consensus. Prakash’s confidence in the Founding-era evidence is unwarranted.[47]

C. Constitutional Structure (and Not-Structure)

1. Not-Structural Argument

Prakash roots his section on structure in an anti-immunity comment made in 1789 by Senator Maclay: “Suppose the President committed murder in the street. Impeach him? . . . But [suppose] . . . he runs away. But I will put up another case. Suppose he continues his murders daily, and neither House is sitting to impeach him.”[48]

Maclay and Prakash identify a legitimate downside to immunity. The problem is that this is a policy argument more than a structural one. Structural arguments are based on the Constitution’s overarching designs and recurring themes that transcend individual clauses—things like federalism and the separation of powers.[49] If the Maclay–Prakash policy argument is structural, it is only in a sense too broad to be meaningful: that the Constitution is designed to be effective and sensible, so an interpretation of it that yields bad results must be incorrect. But laying out possible bad results does not tell us how the Constitution is structured; it just shows how well-designed or not that structure is.

The limits of such policy arguments are evident by analogy to Nixon v. Fitzgerald.[50] Under Nixon, presidents can employ their powers to do illegal things to people without ever facing civil liability for it. And Nixon immunity is permanent; even impeachment does not strip a President of his protection.[51] Is it problematic that a lawless President could run amok under cover of Nixon? Yes. Did that policy problem prevent the Supreme Court from finding civil immunity embedded in the Constitution’s structure? No.

As a matter of policy, Maclay and Prakash are right that impeachment might work too slowly (if at all) to incapacitate an actively criminal President, while arresting him might impede his crime spree. But there are policy problems with denying immunity too. The most obvious: it would embolden the nation’s least restrained local officials—imagine a “constitutional sheriff”[52] like Maricopa County’s former sheriff Joe Arpaio and a like-minded prosecutor—to apply the full force of the criminal law to sitting presidents. Whatever the Constitution provides, immunity or no immunity, would be imperfect.

Prakash floats a more extreme hypothetical: a President might order the arrest of enough members of Congress to deprive it of a quorum, thereby preventing his impeachment and removal.[53] But an attack on Congress like this would be a problem even if the President was ordering it from a jail cell (where, Prakash elsewhere suggests,[54] a President might still be able to wield his powers).

Going further still, pondering a President leading a violent insurrection, Prakash asks, “How could an unhurried impeachment process help arrest or foil an ongoing, possibly bloody coup?”[55] But neither impeachment nor criminal liability would be a fail-safe bulwark in the face of such a raging plot by the Commander in Chief and head of the executive branch. The constitutional order will always be vulnerable to collapse if those in power abuse its mechanisms to destroy its foundations. That weakness is built into any democratic republic and does not lend itself to interpretive solutions.

Underlying this policy dispute (“Which presents a bigger problem, preventing prosecutions of sitting presidents, or allowing them?”) is a bona fide structural issue. To the extent that the Constitution structures our system of government—the presidency, checks and balances, the separation of powers, federalism, and everything else—so that the President can be incapacitated, whom does it empower to do so? This inquiry, “Who decides?”, is the core of the structural argument for immunity,[56] but Prakash does not engage it here.

2. Structural Argument

At the tail end of his structure section, Prakash offers a truly structural argument when he (rightly) criticizes Founding-era immunity proponents for employing a monarchical view of the presidency that is contrary to the structure of the presidency that the Constitution actually wrought.[57] But Prakash also suggests that contemporary pro-immunity arguments echo that misguided monarchism.[58] Perhaps some do, but one key element of an important contemporary argument for immunity is the Constitution’s structure of popular sovereignty—the polar opposite of monarchism.[59]

Prakash’s best structural argument appears outside of his section on structure. Later, addressing the pro-immunity argument that prosecution could incapacitate the presidency and with it the executive branch, Prakash responds that Article II of the Constitution anticipates and solves this problem when it provides for the Vice President to step in when the President is incapacitated.[60]

But here again, this begs the question: Who decides? Whom does the Constitution empower to incapacitate the President in the first place? When the Constitution provides for presidential succession, it is not thereby declaring open season on presidents.

Consider this analogy. Suppose the President is resisting a hostile House committee investigation by broadly asserting executive privilege. In response, the House decides to make aggressive use of its “inherent contempt” power.[61] The conflict gets heated, and the House votes not only to hold the President in contempt of the House but also to arrest and jail him until he cures the contempt.[62] This confluence of inherent contempt, executive privilege, and the separation of powers would present a difficult constitutional-law puzzle. But the question would be whether the House has the constitutional authority to do this. Succession—the fact that, if the House can do this, the Vice President could step in—does not strip presidents of whatever constitutional protection they have from the House’s inherent-contempt power.

Another problem with Prakash’s vice-presidential argument is that it assumes too readily that the Vice President would take the reins. Article II specifies that the Vice President takes over if the President suffers an “Inability to discharge the Powers and Duties of [his] office.”[63] But Article II provided no standards or procedures for determining inability and was never used.[64] Sections 3 and 4 of the Twenty-Fifth Amendment were designed to fill this gap.[65] They did not define inability either, but they did designate decision makers and provide a process.[66]

Prakash notes that the Amendment’s standard (“unable”) can be interpreted broadly enough to include a President being arrested, prosecuted, or imprisoned.[67] But the Amendment’s drafting history indicates that Congress was targeting more extensive incapacitation than a President being prosecuted would represent.[68]

More importantly, that limited intention is reflected in the Amendment’s structure, which would be a poor fit for such a case. A President facing prosecution might well invoke Section 3 (declaring himself unable),[69] just as he might waive his immunity. But if he were to insist on clinging to power, things could get messy. Under Section 4, the Vice President and Cabinet could declare the President unable without his consent and strip him of his powers, but if the President denied that he was unable and tried to retake his powers, he would succeed unless two-thirds majorities in both houses lined up against him too.[70] If there was not a simple majority in the House with the political will to impeach the President over his alleged crime, it is unlikely that there would be a two-thirds House majority—plus the rest of the required ducks in a row—interested in forcing him to stand down.[71] In other words, arrest, prosecution, and incarceration might incapacitate the President enough to present a problem for the country, but not incapacitate him enough for the Twenty-Fifth Amendment to present a solution.

Even in the subset of cases where Section 4 might work well,[72] the same thing is true of Section 4 as Article II. If individual prosecutors, judges, and juries did not already have the constitutional authority to incapacitate the President, Section 4’s succession provisions do not give it to them.

D. Precedent: Speedy President Grant

Prakash highlights President Grant’s reckless carriage driving and resulting brush with the law, arguing that it represents “a precedent for presidential amenability to the criminal law.”[73] But there is more—and less—to the story than Prakash relates. Ultimately, it poses no difficulty to immunity proponents.

In Prakash’s version of the story, based on a recent Washington Post piece recounting a 1908 article in the Sunday Star, D.C. police officer William West insisted on arresting Grant for speeding.[74] Grant posted a $20 bond, which he forfeited (effectively pleading guilty) when he failed to show up for court the next day.[75]

The Star story’s accuracy is questionable, as the Post article concedes.[76] There appear to be no contemporaneous reports of Grant being arrested or convicted (which seemingly would have been newsworthy).[77] Another version of the story, from 1903, predates the Sunday Star article.[78] In this account, the police officer arrested Grant without recognizing him.[79] On the way to the police station, Grant identified himself to the officer, who was so horrified that he had arrested the President that he was unwilling to proceed.[80] But Grant insisted on continuing to the police station with the officer and “paying the proper fine.”[81] The version of the Grant story I learned in the 1980s is similar.[82]

Yet another variation emerged in a 2012 radio interview with D.C. Police Chief Cathy Lanier.[83] In Lanier’s version, Grant was arrested but the police at the station “were unsure if they could charge a sitting president if he had not been impeached” and they let him pay a fine and leave.[84]

Waiver is an important part of the structure of immunity, but one that Prakash downplays here.[85] Even if presidents are immune, they can always waive that immunity—and they would have good reason to do so for a case as minor as a traffic violation.[86] As such, no version of the Grant story is inconsistent with immunity. Most depict Grant as consenting to, or even insisting on, being punished. Notably, the one version that does not specify whether Grant consented is Lanier’s, and she said that the police thought Grant might be immune.

The nature of the fine is relevant too. Prakash suggests that Grant paid the $20 as a bond, which he forfeited by failing to appear in court the next day.[87] If Grant skipped bail without any consequences, it suggests that the authorities deferred to the bounds of his consent. More plausibly, the other accounts depict this not as a bond but as a fine, which Grant chose simply to consent to pay and not contest. That this was a fine and not a bond is evident from the statute Grant violated.[88]

It is unclear whether Grant’s offense was even a crime. The modern taxonomy of civil infractions versus crimes was not well-formed in 1872, but it is notable that the D.C. law Grant violated was in Part I, Title XII (“Of the internal police and municipal regulations”), not Part IV, Title II (“Of crimes and punishments”).[89] And as Prakash notes elsewhere, arrests were not limited to the criminal context in the nineteenth century.[90]

Regardless of whether this was a criminal conviction or a speeding ticket, Grant waived whatever immunity he might have had. As such, this quirky case says much less about immunity than more recent, significant cases like those of Nixon, Clinton, and Trump—who most decidedly did not consent to be subjected to the criminal process while in office.

II. Prakash’s Treatment of the Case for Immunity

In the next part of his article, Prakash presents his negative case: a response to pro-immunity constitutional arguments. He lands some blows here. Some of his points are ones I already subscribed to, but others warrant clarifications and concessions on my part. Disappointingly though, Prakash does not engage some of the most important pro-immunity arguments.

A. Impeachment’s Role: The Real Argument

Prakash’s first attack is an example of his hitting an easy mark but leaving a tougher one alone. He quotes the Impeachment Judgment Clause, which says that people who are impeached and removed “shall nevertheless be liable and subject to Indictment, Trial, Judgment and Punishment, according to Law.”[91] He says that some people take this to mean that prosecution can only occur after a successful impeachment and removal, but he explains in detail why those people are wrong.[92] Prakash is correct; I have previously dismissed the argument he dismantles as “specious.”[93]

But Akhil Reed Amar and I said something quite different about the preemptive role of impeachment. As noted above, a central element of our argument was to ask whom the Constitution empowers to deal with criminal presidents.[94] We first discussed the unique constitutional implications of pursuing a sitting President (all other officials unquestionably can be prosecuted in office), which demands a unique level of accountability.[95] After noting additional constitutional problems with having state or federal prosecutors prosecute a President, we explained how impeachment—a carefully wrought, purpose-built constitutional procedure—avoids those problems.[96] Taking this all together, we concluded that the Constitution makes impeachment the sole mechanism to move against a criminal sitting President.[97]

It would be unfair to expect Prakash to be in perfect agreement with me over which arguments deserve his foremost attention. But Amar’s and my structural argument for the preemptive role of impeachment was the core of our article and seems at least as worthy of engagement as the Judgment Clause argument.

B. Disgrace, Distraction, and Incapacitation

Prakash next turns to the argument that prosecuting sitting presidents would unacceptably “discredit, distract, and disable” them.[98] Regarding the first element, shame, he argues persuasively that the Constitution does not protect the President from it.[99]

Citing Clinton v. Jones,[100] Prakash says that distraction is the same—that “[i]ncumbents do not enjoy a constitutional right to be free from distractions.”[101] Clinton v. Jones was decided shortly after Amar and I published our article, and I concede that it took some wind out of our sails with regard to the distraction issue.

But this still leaves incapacitation (distraction to the extreme), which Prakash acknowledges is “a weightier concern.”[102] His answer, as discussed above,[103] is that incapacitation is no problem at all—the Constitution provides for vice presidents to step into power when presidents are incapacitated, which renders it constitutionally unproblematic for an arrest, prosecution, and/or incarceration to incapacitate a President.[104] But, as also discussed above,[105] providing for a backup when the President is incapacitated does not empower anyone and everyone to incapacitate the President. It does not answer the question of whom the Constitution empowers to take down a President in the first place.

Prakash might criticize this as “assum[ing] the conclusion.”[106] But it follows our conclusion.[107] If the Constitution allows individual prosecutors, judges, and juries to incapacitate the president, then having the Vice President take over would be a natural part of the process. But if prosecutions are unconstitutional, having a Vice President take power away from a President who is incapacitated because he is incarcerated would represent part of the problem, not part of the solution.

Separately, Prakash notes the possibility of treating different stages of the criminal-justice process differently, and crimes of different severity differently.[108] Without conceding that there should be any immunity, he says that if there is immunity it might well be limited to more serious crimes and to more consequential stages in the process.[109]

This is a fair point. The constitutional case for sitting presidents’ immunity is stronger for serious crimes—the sort of thing for which both imprisonment and impeachment are more likely to be on the table—than for pettier offenses.[110] By the same token, the case for immunity is stronger with regard to imprisonment than for trial, and stronger for trial than for mere indictment. My personal view is that indicting a President but delaying the trial might well be constitutional. Arrest is much less likely so; I do not share Prakash’s optimism about a President’s ability to function completely as President while in custody.[111]

That said, there is a problem with such nuanced immunity. As Prakash and I have both suggested, it is hard to argue plausibly that the Constitution requires something so complex and specific if the text is silent and one is relying on implications from the Constitution’s structure.[112] Note, though, that partial immunity would present that same problem for immunity’s opponents as well as its proponents.

C. Popular Sovereignty, Not Popular Election

Prakash next considers a 2000 memo from the Office of Legal Counsel that stressed the significance of the President’s election by a national popular election.[113] Prakash rightly notes several reasons why presidents are not necessarily the people’s choice, and also that to the extent presidents are the people’s choice, vice presidents are too.[114] In a footnote, Prakash notes that Amar and I “briefly touch upon” the same point.[115]

But while Amar and I did mention the typical President’s national popular election, it was in service of a different point: popular sovereignty, not popular voting.[116] The key is not how many people choose the President but rather that the President serves all the people. As the head of the only federal branch whose power is vested in one person, the President is the only figure who combines this kind of individual responsibility with this kind of national constituency.[117] Prosecuting him is not like prosecuting anyone else.

This connects with our “Who decides?” argument. As we said, “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.”[118] Given the President’s unique position, we argued, pursuing him requires a special level of national accountability—a level that Congress (via the impeachment process) has, but that county prosecutors and federal special prosecutors do not. While Prakash makes some good points against the popular-voting argument, he does not confront this popular-sovereignty argument.

D. Federalism: Check. Special Prosecutors: ???

That last point about local officials provides an apt segue into the last part of Prakash’s negative case: federalism. The pro-immunity argument is that it is constitutionally awkward for the part to dominate the whole, which is what would happen if local officials put the President (who “embodi[es] the continuity and indestructibility” of the nation) in the dock.[119]

Prakash is untroubled by the constitutional implications of local officials subjecting sitting presidents to the full force of the criminal law.[120] But he concedes that the federalism argument adds something to the case for immunity, and says that a pro-immunity argument limited to state prosecutions is superior to an argument for complete immunity.[121]

Just as the argument for state immunity adds a unique element, though, so too does the argument for federal immunity. This is another important element of the pro-immunity argument.[122] Because prosecution is a core executive function and the President sits atop the executive branch, prosecuting a President in federal court is constitutionally awkward.[123] The prosecutor would either be truly independent of the President’s control (which presents constitutional problems[124]) or the prosecutor would be under the President’s sway (which presents an unacceptable conflict of interest[125]).

Some might say that this problem is solved by the President’s power to fire the U.S. Attorney or special prosecutor who dared to pursue him (as President Nixon did).[126] If a President is unwilling to pay the political price for doing so (as President Nixon was after the political fallout from firing the first special prosecutor), one could argue he is choosing to allow his own prosecution, thereby waiving his immunity. But even this would not resolve the conflict-of-interest problem.

It would be interesting to get Prakash’s take on all of this, as he has written important scholarship on presidential control of federal prosecutions.[127] Unfortunately, his critique of the case for immunity does not cover this part of the argument.

III. Statutory Reform and Schrödinger’s President

While Prakash and I disagree deeply about whether the Constitution makes sitting presidents temporarily immune from the criminal process, we agree that Congress has an opportunity to make a helpful difference. Prakash floats ideas like conferring statutory immunity on sitting presidents but simultaneously tolling statutes of limitations.[128] Another is for Congress to protect sitting presidents from state prosecution, respecting the structure of federalism and defanging local partisan prosecutors.[129] I have made similar suggestions, among others.[130]

But currently, it is hard to imagine Congress passing any legislation regarding immunity. Presidents face seemingly continuous allegations of corruption and criminality from their opponents. Regardless of who holds power, one side will perceive any pro-immunity legislation as corruptly aiding and abetting a perfidious President. The other side will perceive any anti-immunity legislation as viciously attacking a blameless President. Add in the filibuster and any immunity legislation is dead on arrival.

That leaves us with the status quo. Today, presidents can conduct themselves as if they are likely immune while in office, which is a problem. But presidents have to draw the line somewhere. They need to avoid conduct that is so heinous that it forces a prosecutor’s hand, which would put the immunity issue before the Supreme Court in a case with facts very unsympathetic to the President.

On the other side, prosecutors can still investigate presidents. To the extent that the statute of limitations is not a problem, they can build their cases patiently, waiting until the President has left office and they have the unquestioned ability to proceed.[131] And prosecutors can stand at the same line as the President, ready to take the immunity issue to court if the President does something bad enough both to require immediate action and to present a compelling anti-immunity fact pattern to the Supreme Court. But the constitutional case against immunity is not a slam dunk and the stakes would be high. If the prosecutors come at the President, they best not miss.

Prakash concludes his article by depicting immunity proponents as creating results-oriented constitutional interpretations, motivated by their “hope” of protecting a powerful presidency.[132] Speaking for myself, I reject this characterization. When I first wrote about presidential immunity in 1997 with Akhil Reed Amar it was because I thought we had the right answer. It was not because I was happy about the increasingly imperial presidency, let alone happy about the uses to which Richard Nixon or Bill Clinton had put their temporary immunity. Nor, decades later, was I happy about the uses to which Donald Trump put his.

Presidential immunity is only growing more troubling. In recent decades, one core of the theory of constitutional immunity—reliance on the impeachment process—has eroded. We have moved from a “two-party” system into a hyper-polarized “two-reality” system in which the possibility of a Senate conviction has become much more remote.[133]

If I have any hope here it is that the status quo described above is resolved. When the opportunity presents itself, we hopefully will open Schrödinger’s box and see whether presidents are immune, not immune, or somewhere in between. To the extent that the President is not immune, future criminal presidents could be more readily held responsible, and Congress would have the clarity and the motivation needed to fine-tune immunity by statute. To the extent that the President is immune, Congress would have sharpened incentives to take its impeachment role more seriously, and to take other steps like tolling the statute of limitations.

But until and unless the box is opened, Prakash will believe the Constitution says one thing, and I will believe it says something else. Until and unless the box is opened, we will both be right and we will both be wrong.

  1. .Saikrishna Bangalore Prakash, Prosecuting and Punishing Our Presidents, 100 Texas L. Rev. ‌55, 60–61 (2021).

  2. .Id. at 58–59.

  3. .See id. at 135–36.

  4. .See generally Akhil Reed Amar & Brian C. Kalt, The Presidential Privilege Against Prosecution, Nexus, Spring 1997, at 11. Prakash cites and discusses our article in various places.

  5. .Prakash emphasizes that his analysis is not about the particular case of Donald Trump but rather is about what the Constitution has meant “from its inception.” Prakash, supra note 1, at 62. He can say this very credibly; his scholarly record speaks for itself. I appreciate Prakash underlining this point and I echo it. I have been writing about presidential immunity for decades—long enough to be accused of pro-incumbent bias by partisans on both sides.

  6. .See Brian C. Kalt, Constitutional Cliffhangers: A Legal Guide for Presidents and Their Enemies 11–38 (2012) (conducting a neutral survey of presidential immunity); Prakash, supra note 1, at 58 (quoting Amar & Kalt, supra note 4). To be clear, my personal opinion—in 1997, 2012, and today—has consistently favored the pro-immunity interpretation.

  7. .After leaving office, President Ford famously pardoned President Nixon, which would not have been necessary had prosecution been completely off the table. See Proclamation No. 4311, 39 Fed. Reg. 32,601–02 (Sept. 10, 1974). President Clinton might have been prosecuted had he not cut a deal with the independent counsel the day before his term ended. See Kalt, supra note 6, at 13.

  8. .Physicist Erwin Schrödinger famously hypothesized about a cat placed in a special box designed with a 50% probability of killing it. Following the concept of “quantum superposition,” until the box was opened and the cat observed to be either alive or dead, the cat would be both alive and dead. See John D. Trimmer, The Present Situation in Quantum Mechanics: A Translation of Schrödinger’s “Cat Paradox” Paper, 124 Proc. Am. Phil. Soc’y 323–38 (1980), http://‌‌‌/‌f4Aa2012c2‌/‌files‌/‌2012‌/‌08‌/‌Schrod‌_cat.pdf [https://‌‌‌/‌8DNB-FWLX]. Schrödinger intended by this example to show the absurdity of quantum superposition, but it is a useful metaphor and (non-lethal) experiments have confirmed superposition’s validity. See Quantum Superposition, Wikipedia, https://‌‌‌/‌wiki‌/‌Quantum‌_superposition [https://‌‌/JH3R-PHRH].

  9. .Prakash, supra note 1, at 62.

  10. .Id. at 63–68.

  11. .Id. at 66–67.

  12. .Kalt, supra note 6, at 25–26.

  13. .See Nixon v. Fitzgerald, 457 U.S. 731, 749 (1982) (recognizing presidents’ immunity from damages actions for their official acts); id. at 750 n.31 (surveying similar immunity for judges and prosecutors).

  14. .See United States v. Nixon, 418 U.S. 683, 713 (1974) (recognizing a limited privilege by presidents to shield their communications from disclosure).

  15. .See id. at 708 (declaring executive privilege “fundamental to the operation of Government and inextricably rooted in the separation of powers under the Constitution”); id. at 705 n.16 (rejecting notion that explicit protection for legislators rules out implicit protection for presidents); Fitzgerald, 457 U.S. at 754 (declaring presidential immunity from suits for damages arising out of official acts and rooting this immunity in the Constitution’s separation-of-powers structure); id. at 750 n.31 (rejecting argument that explicit protection for legislators rules out implicit protection for others).

  16. .Fitzgerald, 457 U.S. at 750 n.31.

  17. .See Clinton v. Jones, 520 U.S. 681, 697–706 (1997) (performing structural analysis).

  18. .Prakash, supra note 1, at 68–70.

  19. .Id. at 68–69.

  20. .Prakash discusses Madison’s attempt to get Convention delegates to discuss presidential privileges; he takes this to mean that Madison thought presidents had no privileges and wanted to add some. See id. at 71. But Madison’s comment was vague and was tacked onto a long discussion about other things—an insignificant tangent at the end of a long day, to which nobody paid any regard. See 2 The Records of the Federal Convention of 1787, at 503 (Max Farrand ed., 1911) [hereinafter Records] (showing the stray nature of Madison’s comment, immediately after which the convention adjourned for the day).

  21. .Prakash, supra note 1, at 60 n.28.

  22. .See W. Burlette Carter, Can a Sitting President Be Federally Prosecuted? The Founders’ Answer, 62 How. L.J. 331, 389 (2019). Prakash does not address Carter’s analysis.

  23. .Prakash, supra note 1, at 72 (alteration in original) (citations omitted).

  24. .King Charles I was tried and executed in 1649, but this was in the context of a (temporary) abolition of the monarchy itself—an exception that helps prove the rule. See Josh Chafetz, Impeachment and Assassination, 95 Minn. L. Rev. 347, 385–87 (2010).

  25. .Prakash, supra note 1, at 72 (quoting James Sullivan, Cassius, X, Mass. Gazette, Dec. 21, 1787, reprinted in Essays on the Constitution of the United States 38 (Paul Leicester Ford ed., N.Y. Hist. Printing Club 1892)).

  26. .Id. at 73 (quoting Debates in the Convention of the State of North Carolina on the Adoption of the Federal Constitution (Jul. 28, 1788) (statement of James Iredell, N.C. Delegate to the Const. Convention), in 4 The Debates in the Several State Conventions on the Adoption of the Federal Constitution 109 (Jonathan Elliot ed., Washington, 1836) [hereinafter Debates]).

  27. .See 4 Debates, supra note 26, at 109 (statement of James Iredell).

  28. .Prakash, supra note 1, at 72–74.

  29. .Akhil Reed Amar, America’s Unwritten Constitution: The Precedents and Principles We Live By 43 (2012) (noting comments by Gouverneur Morris and Samuel Johnston); Carter, supra note 22, at 374–75 (noting comments by Benjamin Franklin and Gouverneur Morris).

  30. .The Federalist No. 69, at 416 (Alexander Hamilton) (Clinton Rossiter ed., 1961) (emphasis added).

  31. .Prakash, supra note 1, at 79.

  32. .The Federalist No. 77 (Alexander Hamilton), supra note 30, at 464 (emphasis added).

  33. .See Prakash, supra note 1, at 74.

  34. .3 Debates, supra note 26, at 220; see Arthur Scherr, “The Confidence of His Country”: James Monroe on Impeachment, 44 The Midwest Q. 27, 36 (2002) (discussing this passage).

  35. .See 3 Debates, supra note 26, at 489 (statement of James Monroe) (referring to the Senate as the President’s “own council”).

  36. .Prakash, supra note 1, at 74–75.

  37. .Id.

  38. .See The Notes of John Adams (July 15, 1789) reprinted in 9 Documentary History of the First Federal Congress of the United States of America: The Diary of William Maclay and Other Notes on Senate Debates 446 (Kenneth R. Bowling & Helen E. Veit eds., 1988) [hereinafter Diary of William Maclay] (quoting Senator Ellsworth) (“The President it is Said, may be put to Gaol for Debt.”); compare with infra note 46 and accompanying text (discussing Senator Ellsworth’s pro-immunity arguments).

  39. .See Diary of William Maclay, supra note 38, at 446; Prakash, supra note 1, at 120 (situating the comment by William Grayson in a debate about the proper form of judicial writs).

  40. .The redaction was by Max Farrand. See 3 Records, supra note 20, at 385.

  41. .Prakash, supra note 1, at 75 (omission in original) (quoting 3 Records, supra note 20, at 385).

  42. .See 10 Annals of Cong. 69–84 (1800).

  43. .See id. at 74.

  44. .See Prakash, supra note 1, at 75–76, 79.

  45. .See id. at 75–80.

  46. .See Diary of William Maclay, supra note 38, at 168; Amar & Kalt, supra note 4, at 17 (noting Adams and Ellsworth’s argument rooted in federalism or the separation of powers).

  47. .See Kalt, supra note 6, at 33 (concluding, after briefly discussing Founding-era evidence, that “the only thing that was clear about presidential prosecutability was that it was unclear”); Eric M. Freedman, The Law as King and the King as Law: Is a President Immune from Criminal Prosecution Before Impeachment?, 20 Hastings Const. L.Q. 7, 15–21 (1992) (finding, in an anti-immunity article, Founding-era evidence to be inconclusive).

  48. .Prakash, supra note 1, at 76 (alteration in original) (quoting Diary of William Maclay, supra note 38, at 167).

  49. .See Philip Bobbitt, Constitutional Fate: Theory of the Constitution 7 (1982) (defining structural arguments).

  50. .457 U.S. 731 (1982).

  51. . Id. at 749.

  52. .See Robert L. Tsai, The Troubling Sheriffs’ Movement That Joe Arpaio Supports, Politico (Sept. 1, 2017), https://‌‌‌/‌magazine‌/‌story‌/‌2017‌/‌09‌/‌01‌/‌joe-arpaio-pardon-sheriffs-movement-215566 [https://‌‌‌/‌88VZ-SX5N] (describing the burgeoning “constitutional sheriff” movement, which posits that local law-enforcement officials have authority that trumps federal officials’).

  53. .Prakash, supra note 1, at 77.

  54. .Id. at 102.

  55. .Id. at 58.

  56. .See, e.g., Amar & Kalt, supra note 4, at 14–15, 17–18.

  57. .See Prakash, supra note 1, at 79; see also Kalt, supra note 6, at 33 (describing these monarchical pro-immunity sentiments as “jarring . . . to modern ears”).

  58. .Prakash, supra note 1, at 80.

  59. .See infra subpart 0).

  60. .Prakash, supra note 1, at 92

  61. .See Josh Chafetz, Executive Branch Contempt of Congress, 76 U. Chi. L. Rev. 1083, 1091, 1145–46 (2009).

  62. .E.g., id. at 1146–53 (discussing scope of the House’s powers to enforce contempt citations against executive branch officials).

  63. .U.S. Const. art. II, § 1, cl. 6.

  64. .See Brian C. Kalt, Unable: The Law, Politics, and Limits of Section 4 of the Twenty-Fifth Amendment 29–36 (2019) (recounting history of non-use).

  65. .See id. at 43–46.

  66. .See U.S. Const. amend. XXV, §§ 3–4.

  67. .Prakash, supra note 1, at 92–94.

  68. .See Kalt, supra note 64, at 54–55.

  69. .See U.S. Const. amend. XXV, § 3.

  70. .Id. § 4.

  71. .See Kalt, supra note 64, at 124–27 (discussing the application of Section 4 to a President facing criminal charges); Amar & Kalt, supra note 4, at 23 n.15.

  72. .A President facing both a criminal prosecution and a serious impeachment effort would be a suitable subject for Section 4, effectively suspending him while slow-but-sure impeachment deliberations proceeded. See Kalt, supra note 64, at 126.

  73. .Prakash, supra note 1, at 82–84.

  74. .Id. at 83; Michael S. Rosenwald, The Police Officer Who Arrested a President, Wash. Post (Dec. 16, 2018, 7:00 PM), https://‌‌‌/‌history‌/‌2018‌/‌12‌/‌16‌/‌police-officer-who-arrested-president [https://‌‌‌/‌23LY-C346]; Only Policeman Who Ever Arrested a President, Sunday Star, Sept. 27, 1908, pt. 4, at 2, https://‌‌‌/‌lccn‌/‌sn83045462‌/‌1908-09-27‌/‌ed-1‌/‌seq-46 [https://‌‌/4HE7-RUS3].

  75. .Prakash, supra note 1, at 83.

  76. .See Rosenwald, supra note 74 (noting that “standards of journalism, particularly with quotations, were not as rigorous back then as today, so it’s nearly impossible to know if this is the whole truth and nothing but the truth”). The Post article refers to “other historical references” to the story but offers only one example: a website citing the Star article. See Ben Kemp, The Thin Blue Line, Grant Cottage (May 18, 2018), https://‌‌‌/‌blog‌/‌2018‌/‌5‌/‌18‌/‌the-thin-blue-line [https://‌‌/KPM2-LFBC].

  77. .There are plenty of routine “police blotter” reports to be found in Washington newspapers of the 1870s (the Library of Congress has an extensive, searchable database at https://‌‌‌/‌newspapers), but I have found nothing that mentions Grant’s case.

  78. .When President Grant Was Arrested, 50 Unity 337, 337–38 (1903).

  79. .Id.

  80. .Id.

  81. .Id.

  82. .See Irving Wallace, David WAllechinsky & Amy Wallace, Significa 119 (1983). In Significa’s version, Grant insisted that West “do [his] duty.” Id. Grant walked home as West impounded the carriage, which was later brought back to the White House. See id. Prakash says that pro-immunity advocates were unaware of the Grant case. See Prakash, supra note 1, at 128–29. I plead not guilty. Books like Significa form an important part of my nerdly origin story. I read and re-read Significa innumerable times during my childhood, including the story about Grant.

  83. .D.C. Police Once Arrested a U.S. President for Speeding, WTOP (Oct. 6, 2012, 5:05 AM), https://‌‌/news‌/2012‌/10‌/dc-police-once-arrested-a-us-president-for-speeding [https://‌‌/RV5Z-FFPG].

  84. .Id.

  85. .See Prakash, supra note 1, at 64. Prakash suggests that waivability is a matter of debate and cites Terry Eastland, The Power to Control Prosecution, Nexus, Spring 1997, at 43, 49, as though Eastland is the only one who has favored waivability. Eastland definitely is not. See infra note 86. More to the point, I am unaware of any sort of immunity that is unwaivable.

  86. .See Kalt, supra note 6, at 22 (discussing waiver); Amar & Kalt, supra note 4, at 15 (same); cf. Carter, supra note 22, at 334, 389 (positing that immunity would not even apply to an offense as piddling as Grant’s).

  87. .See Prakash, supra note 1, at 82–83.

  88. .Statutes in Force in the District of Columbia, 1872, H.R. Misc. Doc. No. 42-25, at 247 (1872) (“It shall not be lawful for any person or persons to ride or drive any horse . . . at a pace faster than a moderate trot or gallop . . . under a penalty of not less than twenty dollars for each and every such offense.”).

  89. .Id. at 53–55, 591, 606; see also Stuart P. Green, Why It’s a Crime to Tear the Tag Off a Mattress: Overcriminalization and the Moral Content of Regulatory Offenses, 46 Emory L.J. 1533, 1542–43 (1997) (noting taxonomies, including that in the Model Penal Code, that treat “public welfare” offenses like this one as separate from the criminal law). A violation of the D.C. speeding law that caused an injury could subject the violator to 30–90 days imprisonment in the workhouse, a more clearly criminal punishment. See Statutes in Force in The District of Columbia, 1872, supra note 88, at 247.

  90. .Prakash, supra note 1, at 88–89.

  91. .Id. at 85 (quoting U.S. Const. art. I, § 3, cl. 7).

  92. .Id. at 85–87.

  93. .Kalt, supra note 6, at 26; cf. Amar & Kalt, supra note 4, at 18–19 (noting that the Impeachment Judgment clause does not prevent unimpeached, unremoved officers from being prosecuted, but does illustrate how impeachment can hasten criminal prosecution of an immune sitting President).

  94. .Supra text accompanying note 56.

  95. .Amar & Kalt, supra note 4, at 11–13.

  96. .Id. at 13–20.

  97. .Id. at 20.

  98. .Prakash, supra note 1, at 90.

  99. .Id. at 90–91. This was not an argument that Amar and I included in our case, see Amar & Kalt, supra note 4, or that I thought worthy of including in my general survey of pro-immunity arguments, see Kalt, supra note 6, at 11–38.

  100. .520 U.S. 681 (1997).

  101. .Prakash, supra note 1, at 91; see Clinton, 520 U.S. at 705 n.40.

  102. .Prakash, supra note 1, at 92.

  103. .See supra section 0).

  104. .Prakash, supra note 1, at 92–93.

  105. .See supra section 0).

  106. .See Prakash, supra note 1, at 94.

  107. .See supra text accompanying notes 94–97.

  108. .See Prakash, supra note 1, at 95.

  109. .See id.

  110. .Cf. Carter, supra note 22, at 334, 386 (defining immunity as having this sort of limit).

  111. .See Prakash, supra note 1, at 101–102.

  112. .See Kalt, supra note 6, at 28; Prakash supra note 1, at 99 (criticizing my and Amar’s notion about tolling the statute of limitations).

  113. .Prakash, supra note 1, at 105–07.

  114. .Id. at 106.

  115. .Id. at 105 n.249.

  116. .See Amar & Kalt, supra note 4, at 12. Earlier in his article, Prakash briefly recognizes (but quickly dismisses) this understanding. See Prakash, supra note 1, at 64. I confess that in my book, I spoke too much about elections and not enough about sovereignty. See, e.g., Kalt, supra note 6, at 18.

  117. .See Amar & Kalt, supra note 4, at 12.

  118. .Id.

  119. .See id. at 12 (quoting Alexander Bickel, The Constitutional Tangle, The New Republic, Oct. 6, 1973, at 14, 15).

  120. .See Prakash, supra note 1, at 107–08.

  121. .Id. at 109.

  122. .See Amar & Kalt, supra note 4, at 18.

  123. .See Kalt, supra note 6, at 19–20; Amar & Kalt, supra note 4, at 18.

  124. .See Kalt, supra note 6, at 19–20.

  125. .Id. at 185 n.15.

  126. .Id. at 19–20; Amar & Kalt, supra note 4, at 18.

  127. .See, e.g., Saikrishna Prakash, The Chief Prosecutor, 73 Geo. Wash. L. Rev. 521 (2005).

  128. .Prakash, supra note 1, at 110.

  129. .See id. at 111.

  130. .Kalt, supra note 6, at 36–37.

  131. .This is what was facing President Clinton as he prepared to leave office on January 20, 2001; on January 19 he reached a settlement agreement with the prosecutor. Id. at 13.

  132. .Prakash, supra note 1, at 113.

  133. .See generally Brian C. Kalt, Presidential Impeachment and Removal: From the Two-Party System to the Two-Reality System, 27 Geo. Mason L. Rev. 1 (2019).