Here is the Tyranny

Most case captions are pretty random. Baker v. Carr? What's that about? The name doesn't really tell you anything. It's just two random names: who knows what Mr. Baker and Mr. Carr were fighting about?

Some captions, though, feel absolutely perfect, an exact distillation of the case they represent. A Quantity of Books v. Kansas? Ah yes, those ancient enemies. (Yes, that's a real case, I did not make it up.) League of United Latin American Citizens v. Rick Perry? Yep, checks out. Bush v. Gore? Sure enough, the question presented there was: Bush or Gore?

And now we have Trump v. United States.

I finally read the damn thing over the weekend, and boy was it infuriating. I had noticed that fairly little of the commentary I had seen really engaged in detail with the reasoning of the opinion, and now I understand why. John Roberts's majority opinion is just rife with sleight of hand and misdirection, conflating things left and right to make nonsense seem sensical. Because of that, I think it's tough trying to construct a critique of his "reasoning" that has any kind of flow or force to it. Practically every paragraph requires a new detour into some slightly different area of law that he's mangled.

For that reason, what I'm going to try to do is articulate the right way to think about the issue of presidential immunity. That, then, will illustrate the many flaws in Roberts's version of things.

The bottom line, by the way: yes, it's really that bad.

The question presented in Trump v. United States is what happens when a sitting president does something violates the terms of one or more federal criminal statutes. The answer to that question is very simple: if the president had actual legal authority to do what they did, then it's not a crime and they cannot be prosecuted. If they did not have actual legal authority, then they're just like any other private citizen and can be prosecuted to the full extent of the law.

This has been the settled understanding of liability for official acts since the Founding. In the nineteenth century, the default model of constitutional litigation was essentially a private law model. To take an archetypal example, suppose that a federal law enforcement officer searches someone's property. That person might then sue the official under the state law of trespass. The official, in turn, would raise their federal authority as a defense: I can't be liable under the state law of trespass because I was authorized under federal law to conduct this search!^[1] Then the plaintiff would say, no, you actually didn't have federal authority, because this search violated the Fourth Amendment. And that's how the Fourth Amendment issue would get litigated.

Implicit in this whole system of constitutional adjudication, you may have noticed, is that if the search did violate the Fourth Amendment then the federal official is exposed to the state law of trespass. Thus, it's actually necessary in resolving the trespass suit to decide whether or not the official violated the Fourth Amendment. This is the "ultra vires" test, the most restrictive test possible for official immunity: if the official was acting outside the bounds of their actual legal authority (i.e. ultra vires), they're not immune.^[2]

Okay, so how do we tell whether the president had actual legal authority for any given action? Simple! We apply Robert Jackson's framework from the Youngstown Steel Seizure Case. Jackson's basic idea is extremely simple: how much authority the president has depends on whether he's working with Congress, against the will of Congress, or against a backdrop of congressional silence:

1. When the President acts pursuant to an express or implied authorization of Congress, his authority is at its maximum, for it includes all that he possesses in his own right plus all that Congress can delegate. In these circumstances, and in these only, may he be said (for what it may be worth) to personify the federal sovereignty. If his act is held unconstitutional under these circumstances, it usually means that the Federal Government, as an undivided whole, lacks power.
2. When the President acts in absence of either a congressional grant or denial of authority, he can only rely upon his own independent powers, but there is a zone of twilight in which he and Congress may have concurrent authority, or in which its distribution is uncertain.
3. When the President takes measures incompatible with the expressed or implied will of Congress, his power is at its lowest ebb, for then he can rely only upon his own constitutional powers minus any constitutional powers of Congress over the matter. Courts can sustain exclusive presidential control in such a case only by disabling the Congress from acting upon the subject.

There's really just nothing to dispute here; Jackson eloquently lays out the essential logic of presidential power, and as a result (even though Jackson was only writing a concurrence, not the opinion of the Court) this has become the standard way of thinking through all such questions.

Okay, so let's return to the question of criminal immunity. The president has done something that violated the terms of a federal criminal statute. It would seem that this necessarily puts us in Jackson's third category! Congress after all has spoken on the subject, through the criminal code. Thus in order to find that the president acted within the scope of their authority, a court would need to find that the act was (1) within their inherent constitutional authority as president; and (2) beyond the constitutional power of Congress. The president's power is, in other words, at its lowest ebb!

Things are not quite that simple, though. Suppose that the president claims authority for their act not under the Constitution directly but under the terms of some other act of Congress. In the absence of the criminal statute, this would put us in Jackson's first category, where the president's authority is at its maximum. So what gives? I think that in this scenario we have an interesting little statutory interpretation puzzle. One statute, by its terms, authorizes the president to do X; the other makes X a crime. One of these laws must necessarily carve out an exception to the other. Either the act, if done pursuant to the president's official authority under the first statute, is not a crime, notwithstanding that it fits the language of the criminal statute. Or the authorizing statute must not be read as authorizing the president to do crimes, even if there are some crimes that fall within the natural meaning of its language.

Both these results are plausible, I think. It's actually quite familiar for government officials to be authorized to do crimes: picture undercover police officers participating in drug deals. Even if the relevant criminal statutes do not expressly say "unless you're a cop working undercover," we understand that those officers aren't guilty of drug crimes. But we can also imagine cases were broad statutory language really shouldn't be understood as authorizing the president to do some specific crime, even though it would if read literally. In any event, this inquiry is only about discerning the will of Congress. Once we know that, we can figure out whether the president was acting within the scope of their authority, and therefore whether they're exposed to criminal sanction or not.

This is how I would have understood presidential criminal immunity circa, say, 2015. Immunity applies only where the president had actual legal authority, same as for any other government official. And we tell whether the president acted ultra vires by applying the Youngstown framework. With this in place, how exactly does Roberts manage to mangle things?

The first part of the holding is that the president has absolute immunity for actions taken within his "conclusive and preclusive" constitutional authority. This language comes directly from Youngstown, as Roberts is keen to let you know. Jackson uses the phrase exactly once (Roberts uses it eight times). Here's the full context, which follows immediately after the description of Category Three I quoted above:

Presidential claim to a power at once so conclusive and preclusive must be scrutinized with caution, for what is at stake is the equilibrium established by our constitutional system.

Again the point is that in Category Three, the president's power is at its lowest ebb! Jackson is saying that we should be extremely cautious to conclude that something falls within this zone of absolute presidential authority. Roberts nowhere acknowledges this point even for an instant. And as a result he adopts an extremely broad conception of the immunity that the president enjoys within this zone of unabridged constitutional authority. Per Jackson's logic, the immunity extends only so far as the specific thing the Constitution authorizes the president to do. This is why, for example, you can prosecute the president for taking a bribe in exchange for a pardon, even though pardons are the quintessential example of something within this zone of absolute power. The Constitution authorizes the president to grant pardons, and Congress is powerless to stop the pardon going into effect, even when it was procured through bribery. But this just does not imply that Congress can't impose criminal penalties for the bribery part! It just doesn't!

Roberts really twists Jackson's words throughout the opinion. In the literal first paragraph of the analysis section, Roberts quotes Jackson as saying that "we must not confuse the issue of a power’s validity with the cause it is invoked to promote, but must instead focus on the enduring consequences upon the balanced power structure of our Republic.”^[3] Here's the actual full quote:

The opinions of judges, no less than executives and publicists, often suffer the infirmity of confusing the issue of a power's validity with the cause it is invoked to promote, of confounding the permanent executive office with its temporary occupant. The tendency is strong to emphasize transient results upon policies – such as wages or stabilization – and lose sight of enduring consequences upon the balanced power structure of our Republic.

That is to say, Jackson is warning against a broad construction of presidential power because the cause the power is meant to be advancing is supposedly a noble one, "such as wages or stabilization." Roberts on the other hand is warning against a narrow construction of presidential power, merely because that power was employed in an effort to make Donald Trump the God-Pharaoh. Same thing really!

It's disgusting, and it reminds me of Roberts's infamous work in Parents Involved v. Seattle (2007). That's where his "the way to stop discrimination on the basis of race is to stop discriminating on the basis of race" line comes from; he's cloaking himself in the mantle of Brown v. Board of Education while holding that desegregation is unconstitutional. Similarly here he's using Jackson's concern for the "balanced power structure of our Republic" to justify making the president a king. Just sickening.

Analytically, though, the really big problem with the opinion is that Roberts ditches altogether the ultra vires test in favor of an immunity that attaches whenever the president acts within the outer perimeter of his duties. That is to say, Roberts contemplates immunity for acts that were ultra vires! That were unlawful! On this model the president's immunity from criminal sanction is not actually a function of the authority with which the laws of the United States vest him; rather it is purely a function of their status as President. True, Roberts says this immunity may not be absolute all the way to the outer perimeter, as Trump had claimed. But it's not clear how big a difference this really is: it seems as though the "presumptive" immunity can only be overcome on a showing that prosecution could pose no danger of distorting the internal functioning of the executive branch.

Oh, right, I suppose I should mention: that's why Roberts says the president needs this broad presumptive immunity. Otherwise the fear of criminal prosecution might prevent the president from carrying out his duties "fearlessly." I really cannot stress enough that the explanation of why this would be bad is just absolute gibberish. It's true and correct that we should expect our presidents to take some ultra vires actions! A government that never exceeds the scope of its powers is probably not making adequately bold use of them in service of the public interest. But something's being ultra vires does not of itself make it a crime! Thus Roberts consistently exaggerates the scope of criminal laws, like when he suggests that 18 U.S.C. § 371, which has been interpreted to cover "any conspiracy for the purpose of impairing, obstructing or defeating the lawful function of any department of Government," would apply when the president is not sufficiently zealous in executing any given law. That's just risible, and I have absolutely no doubt that John Roberts would say so if such a prosecution were ever brought.^[4] He adopts this whole "a future president could argue..." tone in this section, to which I cannot help but reply in the voice of Oliver Wendell Holmes: "Not while this Court sits."^[5]

But honestly it barely feels worth engaging with Roberts's reasoning here, because he never really acknowledges ultra vires as a possible standard. Throughout the opinion he aggressively conflates "official capacity" with "actual legal authority;" he never clearly addresses the relevant category, which is acts taken in an official capacity but without actual legal authority. The whole opinion is therefore addressing a strawman: the notion that the president could be prosecuted for doing something for which he had authority! But literally no one is making that claim, because it's nonsensical.

To reinforce this nonsense, Roberts brings in two adjacent areas of law: civil immunity and executive privilege. In Nixon v. Fitzgerald (1982), the Supreme Court held that presidents do enjoy an absolute immunity from civil liability for anything done within the outer perimeter of their official duties.^[6] And there's a well-recognized privilege for the executive branch to withhold records of its internal deliberations from compulsory process from the other branches. Both doctrines are in one way or another about protecting the deliberative process of the executive branch. But Roberts simply ignores the reasons why the same reasoning doesn't apply to criminal immunity.

The point of civil immunity is that civil liability would give the president a personal financial motive, which (supposedly) would tempt him away from pursuing the public interest. I should hardly, ahhh, need to spell out why that doesn't apply to criminal liability, but apparently I do! Crimes are manifestly against the public interest! For the president to be worried he might go to jail if he acts beyond the scope of his authority and in so doing commits a crime seems entirely to the good! Also part of the concern about civil liability is that any idiot can bring a civil suit, which really does raise a concern that the president might become swamped by vexatious claims. Criminal prosecutions are not like this; they are necessarily brought by the Department of Justice. But for some reason, in this opinion that's all about the respect that must be paid to the president's authority, Roberts consistently treats prosecution of a former president as something that would be ginned up for bad-faith, political reasons. It's almost like he's applying one standard for Republicans presidents and another for Democrats!

As for executive privilege, it's meant to enable the president to have frank and open discussion with his advisors, considering ideas that they ultimately discard as unwise without fear of public rebuke for even having considered them. It is notably NOT meant to shield criminality; that was the entire point of the Nixon Tapes Case, which as Quinta Jurecic points out would seem to have been overruled even though Roberts literally cites and relies upon it. Also, and this is important, executive privilege belongs to the federal executive as an institution. That is, it belongs to the current president. If Jimmy Carter decides to waive executive privilege as to Richard Nixon's papers, he can do that! Nixon has, or at least should have, no possible claim to the contrary.^[7] If we're reasoning by analogy to executive privilege, then, there's absolutely no problem with the federal executive prosecuting somebody who used to be the federal executive.

Like I said, there's just almost no point engaging on any of this. It's all sloppy, deliberately so I suspect. At one point, in addition to the grand conflation of "official capacity" with "official authority," in at least one place Roberts seems to confuse the issue of immunity from liability with that of immunity from process while in office.^[8] The whole thing is just a mess, and honestly if this were all there was to criticze the case for, I might almost say that it's not quite as legendarily bad as the Discourse has made it out to be.

It is that bad, though. The reason why is hidden in the details of the Court's "tentative" discussion of which of the acts charged in the indictment against Trump counted as "official acts." That discussion is long and arduous, and frankly I haven't gone through it myself especially closely. After all, on the correct analysis as I've outlined it, the question of "official capacity" isn't a meaningful one: all that matters is official authority. But there's a common theme to what Roberts has to say on this matter, and it's an unbelievably disturbing one: acts taken by a public official with the express purpose of transgressing the limits of their authority are to count as official acts (and hence are presumptively immune, at least for the president). I'm not just talking about actions that are beyond the scope of their authority, here; we know "official acts" includes those that are ultra vires. No, I'm talking about efforts to insulate oneself against the very possibility of legal rebuke. I am talking very specifically about an effort to seize power in defiance of law, to transform the Office of the Presidency from one defined by law to one defined by force. I am talking, in other words, about usurpation.

Because that's what Trump v. United States is really about. Donald Trump strove to usurp the powers of the federal government, to claim as his own even the right of the People to choose their own leaders. John Roberts, surveying these events, tells us that usurpation is part of the duties of the President of the United States, and therefore that Trump cannot be held to account for it. There is no other word for such a system of government but tyranny.