From the time he became a presidential candidate in 2015, Donald Trump warned Americans of a government conspiracy working to keep him from winning office and putting America’s interests first. He began using the phrase “deep state” in 2017, and by 2020, he was tweeting the term comfortably, cracking jokes about the “Deep State Department” and insisting that he still had “a long way to go” in fighting “some very bad, sick people in our government.” In other words, Trump vs. the deep state became one of the former president’s own preferred ways of explaining his administration’s triumphs and travails.

Trump’s numerous critics typically responded in two opposite ways. First, they indignantly denied that any such thing existed, claiming that Trump’s fondness for the idea was just another of his outlandish conspiracy theories. Or, second, they said that of course there is (and always has been) a non-sinister deep state, and then they praised it for standing tall against the impositions of the likes of Trump.

To their credit, the authors of Phantoms of a Beleaguered Republic do not take either of those lines. Political scientists Stephen Skowronek, John A. Dearborn, and Desmond King acknowledge that “depth” in the executive branch is an important phenomenon. Defining depth precisely is conceptually tricky and politically fraught, as we shall see. But, in short, depth encompasses the tendency of administrators in the executive branch to develop their own sense of what the government ought to be doing, potentially leading them to resist political direction. At least some of the time, the authors recognize that simply celebrating this feature of contemporary American government is inadequate; they clearly state that “administrators are not innocents” and criticize the “hubris” of government bureaucrats who hold themselves as the final arbiters of the public good.

Their dense, compact book is an exploration of the complicated interplay between ever-increasing depth and the attempt to assert presidential mastery over the system. As they see it, tensions between independence and responsiveness have always been present in the American constitutional order but were historically finessed through pragmatic compromises. For the last half-century or so, however, these accommodations have failed, leading presidents to go to war with depth. Their chief intellectual weapon has been the theory of the unitary executive, which portrays the chief executive as the democratically legitimate agent of the people and casts doubt on any and all resistance by bureaucrats. Skowronek, Dearborn, and King make it clear that they believe this war on depth ill-serves the public and violates the intentions of Congress, which they view as the intentional architect of the deep state.

Much of Phantoms chronicles episodes in which Donald Trump made it impossible to ignore the conflict between presidentialism and depth, and the presentation is not a neutral one: Trump’s actions are usually presented with a sense of abject horror, though also with a grudging respect for the way he intuited the logical endpoint of the unitary executive theory—which they believe is nothing less than personalized government resulting in arbitrary tyranny. But Phantoms goes beyond an indictment of the former president. It also serves as a plea for a “reckoning with depth,” in which America’s political class forces itself to make a “clear-eyed choice” about “whether we value what depth has to offer or not.”

But is that really the right question for our political system to address? Is a categorical answer needed? The authors are clearly right that issues surrounding depth are central to contemporary political conflict. But a decisive reckoning, in which these issues are somehow settled, seems hard to imagine. Politicians might do better to focus on revitalizing non-presidential politics, given the ways in which politics and depth can complement each other in legitimizing a contemporary government.

What is Depth?

We must first try to understand the concept of depth. The authors observe that “any intermediation that depersonalizes control or elevates official duties over incumbent predilections deepens the state.” That includes the presence of an “entrenched officialdom and their extensive support networks.” We usually speak of a president instantly creating his own “administration,” but permanent government persists through presidential transitions. Indeed, the vast majority of the federal government’s personnel are “deep” and owe their positions to processes entirely beyond the control of the president.

The sunny view of this situation is that we have evolved a system of “networked governance,” which has the ability to draw on resources well beyond official staff and is likely to honor professional norms and culture. To give a concrete example, the professionals of the National Weather Service (NWS, a division of the National Oceanic and Atmospheric Administration, itself part of the Department of Commerce) are strongly attached to the norms of the meteorological community and believe in the integrity of their work, which they think of as completely unrelated to America’s political contests. These commitments do, at least potentially, make their agency less useful to a president who sees important political goals their agency might serve. But this particular lack of democratic responsiveness does not leave the American people any worse off; Congress created a weather service to do meteorology, not politics.

The cloudier, more jaded view of depth is that entrenched bureaucrats serve particularized interests which are, to be sure, clever enough to cloak themselves in the mantle of professional respectability, but which nevertheless have agendas that are often directly at odds with the public good. When a president, armed with the plebiscitary legitimacy of an election victory, tries to give the American people what they want, these supposedly disinterested public servants balk. Much as the authors try to keep the book’s discussion of depth studiously nonpartisan, there is no denying that Republican presidents are especially apt to charge that the workers of the permanent government are against them for simple partisan reasons, too.

When the critics of the deep state prescribe the unitary executive as a cure for the ills of American public administration, their move ought not to be understood as some kind of stand against competence. Instead, the unitary executive theory pushes for clear lines of managerial authority.

Mostly, the authors lean toward the sunny view. At times they suggest that to be against depth is to be against competence. Trump sometimes played into that claim, as when in summer 2019 he warned residents of Alabama, among other states, about Hurricane Dorian. When the NWS in Birmingham quickly tweeted out: “Alabama will NOT see any impacts from #Dorian,” this set off a series of conniptions in the White House and the Commerce Department. As political appointees sought to signal their loyalty to the president, a senior department official said that the contradiction of the president must have been motivated by a desire to embarrass him. The deep state struck back, with the National Academy of Public Administration issuing a report concluding that this reprimand violated the agency’s scientific integrity policy. When Trump doubled down on his position, including presenting a map that had apparently been altered with a Sharpie marker, this episode even earned its own nickname, Sharpiegate, with the president coming in for endless mockery. Nevertheless, Phantoms suggests that the NWS was somehow the loser in this encounter, with the Commerce Secretary “bring[ing] NOAA to heel” in the summer of 2020.

In dissecting clashes like this one, the authors are surely right to argue that deep officials often have legitimate reasons for resisting presidential directives: when the action is illegal, violates norms or established procedures, is in need of further articulation before it can be implemented, is impractical, or will cause some serious unintended consequence.

But for those who call out the deep state’s treachery, these are just excuses, pretexts for illegitimate infidelity to the people’s will. The fight between Trump and the NWS may have been substantively empty, but in other cases, the concerns are much more substantial. Skowronek, Dearborn, and King accuse Trump of stepping up a “war on science” as part of his campaign against depth, and cite his administration’s push for data transparency as an example of “deploy[ing] scientific standards on behalf of unity against depth.” But to frame things that way is simply to take a side. (So is citing lots of New York Times coverage of the matter without any engagement with the administration’s defenders.) When the critics of the deep state prescribe the unitary executive as a cure for the ills of American public administration, their move ought not to be understood as some kind of stand against competence. Instead, the unitary executive theory pushes for clear lines of managerial authority in the executive as a way of ensuring accountability.

How Unitary an Executive?

Phantoms describes the theory of the unitary executive as a late entrant into American politics. Importantly, the authors do not see the Progressives of a century ago as promoters of unity. Although figures like Woodrow Wilson hoped that the chief executive might effectively dictate the agenda to Congress, like a prime minister, they did not seek intra-executive branch unity. Indeed, they were “deepeners” who valued independent executive instrumentalities that were intended to remain cordoned off from politics. While Franklin Roosevelt did attempt a reorganization of the executive along more unitary lines, the compromise he eventually accepted “institutionalized” the presidency rather than “personalizing” it (as Dearborn has previously discussed). The Executive Office of the President, for example, “anchored central direction and coordination in technical expertise and professional judgment.”

It was not until Richard Nixon found himself in charge of a vastly expanded executive branch and faced with a Democratic Congress that “presidents found that they had at least as much to gain by asserting control independently, through administrative direction, as through further collaboration with Congress in the interest of enacting new legislation.” The exigencies of this position created the need for a theoretical justification of an executive acting alone, and scribblers dutifully produced ever-more-elaborate versions of the unitary executive in response (a story that Skowronek previously told in the Harvard Law Review).

For anyone invested in the legal and constitutional claims at stake in the fights over the unitary executive, the theory’s treatment in Phantoms will seem unforgivably dismissive. (The book is also mostly indifferent to the distinctions between a “strong” and “weak” version of the theory, as well as to the distinctions among the theory’s detractors.) The authors assume it is a fool’s errand to try to find in the Constitution any clear prescription for where the president’s authority should end. They are sure that current configurations of power have drawn the theory ever forward, with doctrinal discussions being purely epiphenomenal.

Having gone out of my way not to become a lawyer myself, I have some sympathy with that view, but the authors’ treatment of the unitary executive nevertheless seems inadequate. They come to identify it with the (Trumpian) idea that if the president says jump, the executive branch’s job is simply to answer “how high,” and they argue as if such an emphasis on responsiveness must necessarily come at the expense of all other values.

Indeed, Skowronek, Dearborn, and King often seem to assume that proponents of hierarchy must be skeptics of expertise, at least in the context of the federal government where political appointees are given authority over career civil servants armed with professional expertise. But charging generalists with directing the work of specialists is not at all unique to government. In almost any large enterprise, the leaders who chart the organization’s direction will find themselves instructing subordinates whose specific knowledge is superior to their own. But this is not taken as an argument for a boss-less corporation.

As Trump’s conservative critics would hasten to point out, a unitary executive in which the top job is held by someone contemptuous of the particulars of governing is bound to have many problems.

For the same reasons, the unitary executive theory can promote competent government as much as any alternative. Indeed, its proponents pointedly argue that a more diffuse conception of competence (like the one the authors prefer) is likely to be counterfeit, relying on self-interested metrics untethered from (or simply threatening to) the common good. Real ability to deliver what the people want is likely to result only from a clean line of accountability.

Must the proponents of the unitary executive then believe that fealty to this structuring idea is all that is needed to ensure good governance? Phantoms seems to assume they do, in which case it can treat the outcomes of Donald Trump’s presidency as a decisive test-case for the theory.  

But believers in the unitary executive theory need not think that having the right theory is itself sufficient to produce good government. The very recalcitrance of the deep state, which the theory warns of, may thwart the president’s ability to take responsibility, whatever his willingness. Certainly, this is Trump’s preferred explanation for his administration’s shortcomings; Phantoms may talk of Trump “routing” his deep state opposition, but the former president’s message is that he did not manage to vanquish the deep state, indeed, that it took its foul revenge on him.

More fundamentally, as Trump’s conservative critics would hasten to point out, a unitary executive in which the top job is held by someone contemptuous of the particulars of governing is bound to have many problems. Indifferent leadership will lead to unfocused policymaking, yes, but why should that detract from the potential of a well-functioning unitary executive?

As Skowronek, Dearborn, and King see it, it is because the theory of the unitary executive is destined to be used as a rhetorical mantra rather than a genuinely structuring principle of government. Trump’s team, and especially his third chief of staff, Mick Mulvaney, frequently lashed out at his critics by intoning the phrase, “Elections have consequences.” That impulse to discredit opposition is the real essence of the unitary executive, Phantoms argues, and the concept of an “ideal” unitary executive just a pipe dream—all the more so because the plebiscitary style of democracy it assumes fits so awkwardly with our underlying constitutional structure and the deep, layered reality of our contemporary federal government.

What’s the Alternative?

If the Unitary Executive is a phantom, what is the reality we should strive for? Skowronek, Dearborn, and King offer an alternative way of thinking about authority, which they confusingly label “republican,” and which features insulation of decision-makers, collaborative governance, mediated authority, a pragmatic rather than formalistic sensibility, and “prudential” legitimacy. “A republic,” they assert, “will sacrifice a good measure of unity in the executive to set political support for the administration of government on an inter-branch footing.”

In practice, that means that Congress should be devising “pathway[s] back to cooperation” with the executive, such that the deep elements of the state are drawn into working closely with the political branch. Such an effort would not entail a disorganized campaign to reassert Congress’s prerogatives in hard opposition to the presidency, which they dismiss as “republicanism as whack-a-mole.” If episodic conflict is all Congress can devise to rein in the executive, the legislature “will never again be much more than a watchdog.” Throughout, Skowronek, Dearborn, and King suggest that separation of powers and checks and balances are concepts in some tension with each other, and they favor deemphasizing the latter so that informal accommodations between branches can turn the former into a useful division of labor. In the New York Times, they flesh out this prescription with some specific suggestions, including for a congressional regulation office (which Kevin Kosar and I have proposed).

There is a good deal to be said for this vision. But it is precisely its lack of “republicanness” that hinders it. “Any redesign will have to begin by breaking our obsession with the presidency,” they tell us, and amen to that. But if skepticism of personalized presidential government is indeed one republican characteristic, it is not enough to make a vital republic.

Many people, including several generations of political scientists, thought “responsible party government” was the best way to secure a robust political debate. With differentiated parties giving citizens a clear choice at the polls, presidents would follow as much as lead, playing their role in the constitutional system rather than dominating the whole political process. But if parties do undoubtedly possess some ability to shape a president’s agenda, American politics in the 21st century has mostly demonstrated how readily a president can reorient his party and how dependably it will rally to his defense, notwithstanding the imperatives of interbranch struggle. Rather than injecting real political responsibility, then, contemporary partisanship seems to collapse into fights over particular presidents.

To escape the gravitational pull of presidentialism, what we need is a politics grounded in the activity and concerns of the citizenry. Such a vital politics cannot be about depth vs. responsiveness—and, at least in 2021, perhaps it cannot be about Democrats vs. Republicans if we hope for it to get beyond the bitterly personal. Instead, it must be, first and foremost, about meeting the substantive challenges that face the nation. Politics, as Michael Oakeshott and Bernard Crick have taught us, is the business of figuring out how to live well together. Once common ends are identified, questions of governmental organization do inevitably arise.

The key to breaking away from all-consuming presidentialism, though, must be for our representatives in Congress to realize that, collectively, they are better qualified than the president to work out a sense of common purpose. They will not achieve this resuscitation by “reckoning with depth,” but by reckoning with each other as something more than electoral adversaries. Without that, Congress is an institution adrift, and it is far from certain that it can function even as a mere watchdog. As in another republic beset by phantom menaces, American citizens may soon find themselves effectively under a different form of government.