Reading the Supreme Court opinions in Trump v. United States, I was shocked. Not by the majority’s holding that presidents are immune from prosecution for their official acts, nor yet by the dissenting justices’ warning that the Court had placed the president above the law. I was shocked by how totally the majority and the minority seemed to be talking past each other. This mutual misconnection turns on the meaning of a single term: “official acts.”
It is extremely important that future courts interpret that term a particular way, lest the majority be misunderstood and the minority’s fears come to pass. This will be easiest to see if we summarize two different ideas of what constitute official presidential acts and then advert to the opinions.
The first idea, which I will tendentiously call the “correct interpretation,” is that criminal acts by definition cannot be official acts. The president is endued with certain powers by the Constitution; Congress, therefore, may not criminalize the exercise of those powers; and whatever Congress does legitimately criminalize must by definition fall outside the scope of the president’s powers. There is no harm in thinking that presidents have immunity from criminal prosecution for official acts, since this is just to say that they may do what the Constitution (or in some cases the Constitution along with Congress) lets them do; nor will thinking they are thus immune compromise the enforcement of criminal law, since what is criminal by definition cannot be an official act. A president might commit crimes, yes, but if so he does this not as the president but as a private individual, even if his crime is performed simultaneously with a genuine official act, and even if it is the same physical performance under a different description. It is somewhat as if the president knew how to juggle and liked to juggle at cabinet meetings for the pleasure of his advisers while giving instructions to cabinet officials. He does not thereby make juggling part of the official act of giving instructions to cabinet officials; he juggles as a private individual while giving instructions as the president, and Congress is free to regulate juggling for everyone, including the president, however (absurdly) it deems necessary.
The second idea, which I will call the “minority interpretation,” is that it is possible to incorporate crimes into official acts and that this is just what the majority allows presidents henceforth to do. It is as if, when the president juggles while giving instructions, the juggling is part of his official performance; and so if Congress declares it illegal to juggle four balls but not three or five, then the president alone among citizens remains entitled to juggle four at his cabinet meeting (but not, say, as he takes a stroll through the Rose Garden, since this is not an official act into which the new juggling crime may be incorporated).
The principal dissent, by Justice Sotomayor, summarizes the minority interpretation powerfully. “When [the president] uses his official powers in any way,” she writes, “under the majority’s reasoning, he will now be insulated from criminal prosecution. Orders the Navy’s Seal Team 6 to assassinate a political rival? Immune. Organizes a military coup to hold onto power? Immune. Takes a bribe in exchange for a pardon? Immune. Immune, immune, immune.” Justice Jackson reads the majority similarly: “Whether the President will be exempt from legal liability for murder, assault, theft, fraud, or any other reprehensible and outlawed criminal act will turn on whether he committed that act in his official capacity”; “presidents alone are now free to commit crimes when they are on the job, while all other Americans must follow the law in all aspects of their lives, personal or professional.”
Yet the majority opinion by Chief Justice Roberts does not take itself to authorize any such thing. These examples it dismisses as “fear mongering” over “extreme hypotheticals.” It notes that “if the president claims authority to act but in fact exercises mere ‘individual will’ and ‘authority without law,’ the courts may say so” (quoting Nixon v. Fitzgerald); the mere fact that the president claims to have done something in exercise of his official powers thus does not suffice to make it so. The partial concurrence by Justice Barrett rejects the minority’s reading very directly in this sentence: “The Constitution does not insulate Presidents from criminal liability for official acts.” These quotations are less elaborated than what I described above as the “correct interpretation.” But that interpretation is compatible with or implicit in them, and I think something like it will be necessary if the Court’s decision is fully to preclude such outcomes as the minority fears and the majority is at pains to disavow. Let’s focus on one quite clarifying “extreme hypothetical,” from Justice Jackson, to illustrate why.
Removing officials is one of the core presidential powers for whose exercise, the Trump majority says, the president is absolutely immune from prosecution (for the “outer perimeter” of his authority, he is presumptively immune, with the precise extent of immunity to be decided later). Jackson writes: “While the President may have the authority to decide to remove the Attorney General, for example, the question here is whether the President has the option to remove the Attorney General by, say, poisoning him to death.” Jackson thinks the majority says yes: The president cannot be prosecuted for murder because the murder was part of an official act, much as juggling became part of the president’s giving of instructions to cabinet officials. (To make the analogy more exact, we have to make juggling a contrived, pretextual means of giving instructions just as poisoning is a contrived, pretextual means of removing the AG; maybe the president is an especially gifted juggler who bounces balls off the conference table in the rhythms of Morse Code, his real purpose being, again, to entertain or perhaps to annoy his cabinet.)
Two things are noteworthy about the minority interpretation as embodied in Jackson’s hypothetical. First, its understanding of official acts is context-free. In particular, it does not look into the relationship between means and ends, nor does it consider whether there are standard as opposed to contrived and pretextual means. There is an official act, removing the attorney general, and that is the act being performed whether the president sends a letter firing him or has him poisoned to death. Never mind that in poisoning the AG, the president is doing something neither essential to removing an official nor an answer anyone would give to the question “How does one remove someone from office?” Removal by letter and removal by murder are just two instances of the official act “removing the attorney general.”
The other thing to note is that by making crimes incorporable into official acts, the minority interpretation yields, in conjunction with the Court’s ruling, the perverse result that the Constitution undermines itself. This is because one of the duties that the Constitution assigns to the president is to “take care that the Laws be faithfully executed.” But if the president may commit any crime in the course of performing his duties, then he is not taking care that the laws be faithfully executed; he is taking care that the criminal law go unexecuted in his case. There is now a contradiction between his duty to execute the laws and the exercise of whatever putative power includes his violation of the law.
Justice Jackson notes just this contradiction and makes it an argument against the majority’s holding. And she is surely right that we should abhor such a result. But given that it is also a result the majority disavows, should we not seek ways consistent with the majority opinion to preclude it?
We do so by taking the following steps. First, rather than ignore the context of putatively official acts, we enrich our analysis by asking whether the means employed were well-tailored to their purpose or went beyond it, and also whether they were consistent with the president’s other powers and duties. The poisoning goes beyond the removal, is not necessary to the removal, should not even be considered an instance of removal. In no ordinary use of language is “poisoning someone to death” equivalent to “firing someone”; why, then, should that equivalence be drawn only for purposes of Article II interpretation?
And second, we must make it an interpretative axiom that the Constitution is not self-undermining. The powers and duties it assigns to the several branches of government must be consistent with one another, and it must not contain any clauses that have the effect of nullifying other clauses. So when the Constitution says that the president shall take care that the laws be faithfully executed, this restricts the president’s official powers to those that are consistent with the law and rules out, by definition, the idea that criminal acts may be official acts.
* * *
As I said, the correct interpretation is not exactly what the majority articulates. But I think it is a reformulation of the essence of Justice Barrett’s reasoning, and that it is consistent with Chief Justice Roberts’s opinion except where Justice Barrett partially dissents from it.
First, let’s expand the quotation from Barrett given above. She writes:
The Constitution does not insulate Presidents from criminal liability for official acts. But any statute regulating the exercise of executive power is subject to a constitutional challenge. A criminal statute is no exception. Thus, a President facing prosecution may challenge the constitutionality of a criminal statute as applied to official acts alleged in the indictment. If that challenge fails, however, he must stand trial.
What is important to see is that, on Barrett’s view, the Constitution’s grants of authority are conceptually prior to Congress’s criminalization of conduct. As I read it, this is simply a tautology: The Constitution is the law before any criminal statutes are; whatever the Constitution says the president may do must therefore be allowed by law; statutory law therefore may not criminalize it. Or if the president’s authority comes from both the Constitution and Congress, Congress will be contradicting itself by criminalizing what it has elsewhere authorized the president to do. So there has been no expansion of “official acts” to newly include what would otherwise be crimes or what Congress may yet declare to be crimes. Congress remains free to criminalize what it will, for the president and everyone else, constrained as it always was by the tautological legality of powers that the Constitution, or the Constitution along with Congress, has independently granted the president. (This way of putting things is slightly inconsistent with Barrett’s expression that the president is not insulated from “criminal liability for official acts,” since on the correct interpretation an official act is by definition not criminal. We should say instead that the president has no immunity from “criminal liability for acts performed in conjunction with official acts” or “criminal liability for acts falsely claimed as official.” But we are justified in taking something like this to be the meaning of Barrett’s shorthand, because nothing else will yield the result that the president, while fully authorized to exercise all his official powers, is never authorized to commit crimes.)
This reasoning is easiest to apply to the president’s core powers, for the exercise of which the majority says he has absolute immunity, because such powers are mentioned or clearly implied by the text of the Constitution and so it is fairly simple to say whether a statute conflicts with them. Congress could not, for example, make it a crime for the president to negotiate a treaty, since the Constitution grants him that power.
As Barrett notes, “the Constitution does not vest every exercise of executive power in the President’s sole discretion. Congress has concurrent authority over many government functions, and it may sometimes use that authority to regulate the President’s official conduct, including by criminal statute. Article II poses no barrier to prosecution in such cases.” Barrett would analyze these cases — which extend to the “outer perimeter” of executive authority, where the majority presumes but is not sure that the president is immune — in two steps. First we are to ask “whether the relevant criminal statute reaches the President’s official conduct.” If so, we next ask “whether applying [the statute] in the circumstances poses no “dange[r] of intrusion on the authority and function of the Executive Branch” (Nixon v. Fitzgerald, internal quotation marks omitted). This also amounts to the correct interpretation. The question whether the president may be prosecuted reduces to the question whether the criminal statute conflicts with “the authority and function of the Executive Branch,” and this can make sense only if, again, the definition of that authority and function is conceptually prior to the criminal statute. If Barrett had wanted the president to be able to incorporate crimes into official acts, as the minority has it, then there would be no need for her to ask whether a criminal statute conflicts with his authority: It would not matter because the president would be immune anyway. There are, Barrett observes, “unsettled questions about the scope of Article II power,” but the settling of those questions — the defining of Article II powers — still must be prior to our asking whether a criminal statute conflicts with the answer. Every sentence of the analysis assumes that there are definite answers, already or yet to be found, to questions about the extent of presidential power and that the criminal law must be consistent with them — not that it simply ceases to apply in case of conflict.
Because Barrett is so clear about this, she is able to answer correctly certain questions that the majority, spoken for by Roberts, returns to lower courts for further analysis. “For example,” Barrett writes, “the indictment [against Trump] alleges that the President ‘asked the Arizona House Speaker to call the legislature into session to hold a hearing’ about election fraud claims.” Roberts is unsure whether the president is immune from prosecution for this conduct, but Barrett writes that “the President has no authority over state legislatures or their leadership, so it is hard to see how prosecuting him for crimes committed when dealing with the Arizona House Speaker would unconstitutionally intrude on executive power.” The point is obvious enough that one wonders why a lower court needs to study it.
Barrett also departs from the majority in thinking that a president’s official conduct may be admitted as evidence in a criminal prosecution, citing the example of bribery: “The Constitution, of course, does not authorize the President to seek or accept bribes, so the Government may prosecute him if he does so. Yet excluding from trial any mention of the official act connected to the bribe would hamstring the prosecution.” Therefore it cannot be correct to hold, as Roberts and the majority do, that official acts are inadmissible as evidence in lawful prosecutions of a president. In mentioning the official act connected to the unofficial bribe, a prosecutor would not be impinging on legitimate executive authority but would simply be presenting evidence of a crime. It is hard to see how this would constrain a president from doing anything except accepting a bribe.
* * *
The Roberts opinion, like the Federalist Papers on which it relies, reads as a piece of consequentialist political reasoning concerned above all with likely effects. Simplistically but not unjustly, the main argument could be summarized thus: We have Article II, which establishes presidential powers; the Framers intended those powers to be exercised boldly and decisively; the president is unlikely to exercise them boldly and decisively if he has to fear frivolous prosecution for exercising them; therefore, he must have immunity from prosecution for their exercise. This is rooted in the Constitution in that Roberts wants to make sure Article II is not encroached upon, but the way of encroachment passes through the president’s psychology.
There is less need to psychologize the matter if, like Barrett, we just (a) attend carefully to what the Constitution and the statutes do and do not authorize the president to do and (b) decline to reinterpret definitions of his official powers in ways that abuse the meanings of words (as by saying that poisoning the AG is an instance of removing the AG). The president and the rest of us can look directly at the Constitution and the statutes and see how far executive powers extend.
The dissenters emphasize with alarm Roberts’s pronouncement that “in dividing official from unofficial conduct, courts may not inquire into the President’s motives,” since “such an inquiry would risk exposing even the most obvious instances of official conduct to judicial examination on the mere allegation of improper purpose.” An example from later in the opinion may suggest the kind of thing Roberts has in mind:
Section 371 — which has been charged in this case — is a broadly worded criminal statute that can cover “any conspiracy for the purpose of impairing, obstructing or defeating the lawful function of any department of Government.” Virtually every President is criticized for insufficiently enforcing some aspect of federal law (such as drug, gun, immigration, or environmental laws). An enterprising prosecutor in a new administration may assert that a previous President violated that broad statute. [Internal quotation marks omitted.]
One would not want the president to fear when, say, allocating executive resources or interpreting a regulation that he will be accused of doling out political favors and so hesitate to follow what seemed the best course. But it is hard to see how Roberts’s requirement would make it easier for the president to get away with poisoning the attorney general. To say that poisoning is not removing, and that the Constitution authorizes removing but not poisoning, does not require us to assume anything about motives, which is why we can say it in that perfectly general way, perfectly consistent with the idea of an official act (one that is recognizable as the exercise of a government function in the manner and sort of circumstance for which it was intended, all of this being confirmable against written laws and standard dictionaries).
We should not be so naïve as to deny ambiguity. It could be unclear whether a president ordered investigations of bogus claims of election fraud, for example, because he believed them or because he was trying to subvert an election outcome. When the AG was poisoned, the president did something flat-out other than fire him, but now the issue is whether the president has done something superficially compliant with his powers but for a hidden and possibly illegal purpose.
To me this is mostly a knowledge problem: Ordering an investigation is an official act and ordering one fraudulently and illegally is not; but because we may not see motive clearly, it may not be possible to say which act the president performed and so whether that act was official or unofficial. We could even get scholastic and distinguish unconfirmable “motive” from demonstrated “intent”: If there were a recording of the president saying, “Please order this sham investigation now because I would like to illegally subvert the election,” I think we could say that ordering the investigation did not qualify as an exercise of official powers, because official powers cannot be used subvert the law and the Constitution without becoming, to quote Roberts quoting Fitzgerald again, exercises of “individual will” and “authority without law.”
So I don’t know that it was wise for the majority to forbid looking at motives when separating “official” from “unofficial.” It might have been enough that when intent is indeterminate, it is indeterminate, and that the president would be entitled (under Barrett’s view) to pre-trial interlocutory review: One assumes the executive branch would swat down frivolous charges quickly and that prosecutors would soon learn their futility. If that is too idealistic, it nonetheless seems correct to say that seeking charges against a president is not obviously a fine occasion for prosecutorial creativity and that, in cases of grave political offenses (high crimes and misdemeanors) that are only ambiguously crimes according to the judicial system, the surest remedy lies with Congress.
* * *
Not that Congress can be trusted to do its job, especially when it is most needed. In a demagogic time, there will be a tendency both for the president to abuse his authority and for Congress to shrink from reining him in. This is because both the demagogue and the enablers are riding a wave of popular anger and those who would resist them must go against it.
For this reason, it is very important that future rulings disavow the minority interpretation of “official acts” and develop reasoning that excludes it. Roberts says that, in their focus on the “extreme hypotheticals,” the dissenting justices “overlook the more likely prospect of an Executive Branch that cannibalizes itself, with each successive President free to prosecute his predecessors, yet unable to boldly and fearlessly carry out his duties for fear that he may be next.” But more and less “likely” is not good enough. We need the Constitution, if it can, to absolutely rule out poisoning the AG, and if it doesn’t we need to amend it — not just because we don’t want the AG poisoned but also, and fundamentally, because a Constitution that lets the president poison the AG does not make any sense.
If learned and fair-minded justices of the Supreme Court can find in the majority opinion the idea that the president may also be a supervillain, how much more readily might cynics serving an aspiring tyrant embrace it — at any future time? The dissenting justices are mistaken; Trump v. United States means, at root, that the president’s true powers are not criminal and must not be treated as so, not that his true powers include the commission of crimes. But the dissenting justices deserve our thanks for illuminating so clearly the path we needn’t and mustn’t tread.