The Head and Body of Leviathan

A Thought Experiment On Executive Power

I propose a thought experiment: what might be the ultimate legal consequences of the constitutional conception of executive power, as the Supreme Court currently understands it? How far might the logic of that conception go?

In Trump v. United States, the recent decision on presidential immunity, the Court reiterated its view that “the Constitution vests the entirety of the executive power in the President.” The Court here built upon precedent, including Seila Law LLC v. Consumer Finance Protection Bureau, which said that subordinate executive officers “wield executive power on [the President’s] behalf,” and especially the foundation-stone of unitary executive theory, Myers v. United States, which laid down the fundamental proposition that when subordinate officers exercise executive powers granted by statute, they are exercising the President’s own power. As Myers put it, “[t]he highest and most important duties which his subordinates perform are those in which they act for him. In such cases they are exercising not their own but his discretion…. Each head of a department is and must be the President's alter ego in the matters of that department where the President is required by law to exercise authority…. In all such cases, the discretion to be exercised is that of the President in determining the national public interest and in directing the action to be taken by his executive subordinates to protect it” (emphasis added). Writing for the Court in City of Arlington v. FCC in 2013, Justice Scalia laid down a corollary that the Court in Seila Law cited with approval: whether agencies proceed by rulemaking or adjudication, whether by “legislative” or “judicial” forms, those forms “are exercises of — indeed, under our constitutional structure they must be exercises of — the ‘executive Power.’”

Let us take the central proposition established in Myers, Seila Law, and Trump v. United States as a kind of constitutional axiom, together with the corollary established in City of Arlington, in order to elicit the ultimate logic of the unitary executive. Assuming that central proposition for the sake of argument, what follows? I will not defend the proposition itself, beyond noting its grounding in precedent and in the Court’s current conception of executive power. Rather, starting from the constitutional axiom, I want to state a maximalist theory of executive power and explain its legal consequences.

By “maximalist,” I mean that the theory reads Myers, Seila Law, Trump v. United States, City of Arlington and other decisions for the most they could be taken to say, rather than the least they need be taken to say.¹ The theory goes even farther than the standard version of unitary-executive theory, goes to the outer logical limits of presidential power. Indeed a consequence of the maximalist theory is that Myers failed to follow through on its own rationale; as I will explain, Myers admitted limits on executive power that contradicted, and hence compromised, the internal logic of the decision.

If and because “the Constitution vests the entirety of the executive power in the President,” it follows (as City of Arlington says) that when subordinate executive officials, including administrative agencies, exercise delegated discretion under otherwise valid statutory grants of authority,² they are exercising executive power; hence they exercise not their own power, but that of the President. There is no such thing as executive power given to subordinate officers or administrative agencies in their own right; there is only, ever, the executive power of the President, which alone incarnates and gives legal life to the legal authority of all his subordinates. This is true whether or not the subordinate officers are so-called “executive officers,” “executive branch agencies” or “independent agencies,” however defined.

An image may be helpful, adapting the famous frontpiece to Thomas Hobbes’ Leviathan (given above).³ Leviathan is itself composed of many smaller bodies; by the same token, it encompasses and includes them. The citizens are contained within Leviathan, as it were, forming the body of the commonwealth. So too, by analogy (and putting firmly aside the question what use Hobbes himself intended to make of the image), the President as Leviathan encompasses all subordinate executive officials. The President is not only the head of the executive branch, but also its whole body; in contemplation of the law, there is no executive power that lies outside the Presidency. Of the President’s two bodies, his public and legal body subsumes the whole executive establishment, including each and every agency or official exercising executive power.

What concrete legal consequences follow from this picture? Stated most generally, the limitations on the unitary executive indicated in Myers itself represent ad hoc compromises of Myers’ own logic; the maximalist theory would discard them. Let me indicate some examples:

The directive power is the power to command. Myers was ambiguous or ambivalent on whether the directive power of the President is a legal power to command, imperium in the full sense, or something less — a vague power to “supervise” or “guide” the exercise of discretion. Subsequent literature has debated these conceptions of directive power at length, but the maximalist logic of the constitutional axiom settles the issue. If agencies and subordinate officers exercise the President’s own power, he can command them at will within the bounds of his legal authority or even elect to carry out their delegated authorities himself, just as he can exercise his own power, within authorized limits, according to his own will and reason. He can command his subordinates just as he can command himself — emphatically including prosecutors, special counsels, and other subordinates engaged in “investigative and prosecutorial decisionmaking,” functions that the Court expressly described in Trump v. United States as “the special province of the Executive Branch.” A subordinate officer who refuses an authorized presidential command acts incoherently, as though the hand attempted to revolt against the head and the body. And vis-a-vis his own internal subordinates, within the unitary legal body of the Presidency that encompasses the whole executive branch, it is the President alone who ultimately decides whether the command is authorized, although the President may of course have to answer to other branches for the legality of his actions (as discussed below).

Recent cases, more logical than Myers in this regard, have strongly pushed in the direction of seeing the directive power as full imperium. Thus Seila Law said that the authority of individual executive officials is subject to the President’s “supervision and control” (emphasis added). Nothing in this logic supports any distinction between executive branch agencies and putatively independent agencies — unsurprisingly, given that Seila Law invalidated restrictions on presidential removal power as to independent agencies. Many have suggested that in light of the broad rationale of the Court’s recent decisions, including not only Seila Law but successor cases — a rationale resting on a conception of executive power that sweeps well beyond the narrower holdings of the decisions — the Court has either de facto overruled, or will inevitably overrule, the decisions that legitimated independent agencies, and will preclude the very possibility and existence of such agencies.

All delegations are delegations to the President. Myers indicated in Delphic terms that “there may be duties so peculiarly and specifically committed to the discretion of a particular officer as to raise a question whether the President may overrule or revise the officer's interpretation of his statutory duty in a particular instance.” On the maximalist theory, there is no basis for this limitation, which merely flinches on the logic of Myers. No matter how clearly Congress speaks, statutory duties committed to the discretion of subordinate officers or agencies are only ever grants of executive power, which is the President’s own power and which can be directed as the President sees fit — as Myers itself elsewhere indicated, saying that “in this field [the President’s] cabinet officers must do his will.” Any grant of statutory authority to a subordinate executive officer is a grant of statutory authority to the President, who alone holds the executive power that inheres in and gives life to such authority. It is not that Congress acts unconstitutionally when it attempts to confer statutory authority on a subordinate officer other than the President, but that by force of the constitutional conception of executive power, the attempt necessarily directs ultimate authority to the President himself.

Formal adjudication by executive officers is no exception. Another somewhat mysterious limitation on the directive power sketched, but not explained, in Myers was that “there may be duties of a quasi-judicial character imposed on executive officers and members of executive tribunals whose decisions after hearing affect interests of individuals, the discharge of which the President can not in a particular case properly influence or control” (although Myers did indicate that even in such cases, the President could remove the adjudicating officials on the ground that in the President’s view “the discretion regularly entrusted to that officer by statute has not been on the whole intelligently or wisely exercised”). On the maximalist logic of the constitutional axiom, however, there can be no such limit as to the directive power, even as to formal adjudication within the executive branch. Per City of Arlington, the adjudicator exercises executive power, which is not his own power, but that of the President. Hence the President may either decide to exercise such power himself, or to command the adjudicator to rule one way or another by applying the relevant law as the President thinks warranted under that law.⁴

A corollary is that a lower court erred when it held that the President could not intervene in a formal adjudication within an executive body, reasoning that because the President counted as an “interested person outside the agency” within the terms of the Government in the Sunshine Act, the intervention counted as a forbidden ex parte contact with the adjudicator. The problem is that the President is not external to or distinct from the agency in either a constitutional or statutory sense, any more than the legal body of the President can be separate from his own limbs. Rather, the agency exercises the President’s own power, even when adjudicating. Because the President is both the head and the body of Leviathan, the agency is, if anything, inside the President. The court indicated that its interpretation of the relevant statute was affected by background norms of due process, but the invocation of due process begged the question. The issue in the case was not whether the neutrality of the adjudicator had been compromised, but who exactly the adjudicator should be taken to be. (Although the president is an elected official, there neither is nor can be any general constitutional rule that elected officials are inherently incapable of serving as neutral adjudicators for due process purposes. Such a rule would mean that the elected judiciaries of many states are automatically incompatible with due process).

The President may remove not only principal officers but also inferior officers at will. The final limitation indicated in Myers itself, drawing upon an earlier decision in United States v. Perkins, was that Congress could limit the President’s power of removal of inferior officers, at least where (but perhaps only where) Congress had vested the appointment of such officers in a department head, as opposed to the President himself. The subsequent history of this limitation is tortuous, running through the discredited decision in Morrison v. Olson and through Seila Law among other cases, and is or ought to be familiar to all students of constitutional law. It would take me too far afield to recount that history. Instead, I will content myself with the following observations: (1) in a recent and comprehensive review, two eminent scholars of removal jurisprudence take the view that the law as to presidential power to remove inferior officers at will is in a confused and unsettled state, and that the question is open to a through re-examination by the Court; (2) the maximalist logic underwrites the vision of removal at will that Madison articulated in the First Congress: in virtue of the presidential power of removal at will, “the chain of dependence [will] be preserved; the lowest officers, the middle grade, and the highest, will depend, as they ought, on the president” (emphasis added). Inferior officers hold the same type of power as principal officers, a share of the executive power that is vested in the President alone; the only difference is the scope of that power. If the President may both remove and direct principal executive officers at will, because they exercise his own power, so too with inferior officers.

Subordinate officers enjoy the President’s own immunity from civil and criminal process. Moving beyond Myers, there is the question of the immunity of subordinate executive officers from personal liability to civil and criminal process, a subject related to the issue of presidential immunity in Trump v. United States. On the maximalist theory, subordinate executive officials, who only ever exercise the President’s own power, partake of the President’s own immunity for acts in the President’s official capacity (whether that immunity is absolute or presumptive, depending on the nature of the presidential power at issue under the distinctions the Court recently drew). The maximalist logic rejects the idea that even if the President can do no legal wrong, his subordinates can.⁵ Just as the central incoherence of independent agencies is to create a fourth branch in which executive power is “headless,” so conversely, the view that the President enjoys an immunity that does not extend to subordinates exercising his own power would make the executive power bodiless. Those subordinates only ever exercise the President’s discretion, not their own; they are no more than extensions of the President, limbs of the President’s legal body. The President’s immunity includes and subsumes that of his subordinates.

So much for the direct legal consequences of the maximalist logic of the unitary executive. On this logic, what limits the Presidency? The law does. Nothing in what I have discussed raises any issues of extraordinary prerogative or of “dictatorship,” whether in the classical sense or other senses. The President is always subject to the directive power of the law, even if not its coercive power. Nothing in the maximalist view of the unitary executive suggests that the President is entitled to act ultra vires, beyond the boundaries of his constitutional and statutory authority, a constraint established long ago and repeatedly reaffirmed. To be sure, courts may for various reasons interpret the boundaries of presidential authority deferentially, and the immunity rules established in Trump v. United States will, on the maximalist theory, shield both the President and his subordinates from personal liability, civil or criminal, for official acts within the outer perimeter of their legal powers and duties; but those subordinates, like the President, have in principle no entitlement to act ultra vires.⁶ Furthermore, because the President’s public body is a legal body, it must move and act the way law moves and acts, through procedures and norms that are intrinsically constitutive of legality, and through law’s intrinsic ordering to the common good or (as Myers put it) the public interest - an ordering reflected in the law’s interpretive presumptions for determining the boundaries of executive authority. As I have argued elsewhere, drawing on the work of Lon Fuller, Leviathan is redeemed by law’s internal morality.

As to the question what utility this thought experiment may have, I will leave that to the judgment of the reader. Suffice it to say that there is a maximalist version of unitary executive theory that lies just beyond the holdings of current law, but that is implied by the logic and rationale of the Court’s recent decisions, and that is distinctly visible over the near horizon.

1. For a counterpoint, reading the precedents for the least they need to say, let me commend some excellent recent work by Zachary Price. See, e.g., Zachary S. Price, “Congress’ Power Over Military Offices,” 91 Texas L. Rev. 491 (2020); “Even if the President is Immune, His Subordinates Are Not,” Notice and Comment (July 11, 2024).

2. By “otherwise valid,” I mean to bracket any questions involving the so-called constitutional “nondelegation doctrine.” Skeptical treatments of that doctrine can be found here and here, but nothing in the argument in text assumes any view, one way or another, about the precise boundaries and force of the nondelegation doctrine.

3. It should probably go without saying, but in our age of bad-faith readers does not, that the Leviathan image merely supplies a convenient and widely known analogy, for purposes of exposition. Nothing in the legal arguments I present here depends upon Hobbes’ metaphysical assumptions or legal or political theory. Indeed various versions of the image of the ruler, prince, or executive as either the head or the body of the state long predate Hobbes.

4. Whether and how a presidential decision in formal adjudication (or for that matter generally) would be reviewable by an Article III court presents separate questions, as to which the background law is complex and unsettled. Needless to say, in the overwhelming bulk of cases, the President leaves both adjudication and rulemaking to subordinate officers, whose acts are judicially reviewable, at least presumptively, under the Administrative Procedure Act.

5. Again, I recommend the excellent work of Zachary Price on these issues, although his perspective is contrary to my own.

6. On the (presumptive) reviewability of the acts of subordinates for statutory and constitutional authority, see note 4 above.