The January 6 Insurrection and the Problem of Constitutional Guardianship

37 Const. Comm. 365

Aziz Z. Huq[1]


The violence of January 6, 2021, embodied one threat to American democracy, and then gave rise to yet others. Rioters that day acted on then-President Donald Trump’s exhortation to “walk down to the Capitol” and then “demand that Congress do the right thing and only count the electors who have been lawfully slated.”[2] What ensued placed Vice President Mike Pence and others at immediate risk of physical harm.[3] The ensuing extra-legal violence came startling close to a disabling or thwarting the counting of Electoral College votes, and so preventing the orderly handover of presidential power after an incumbent had lost at the polls.[4] It would have been the first time in more than two centuries that such a peaceful transfer of power failed to happen.

Rather than shining a clarifying light on the risks to American democracy, however, the events of January 6 have muddied those waters. Like a kaleidoscope, that day showed how fragmented Americans’ shared understanding of the democratic project and its risks have become—and then its aftermath shook those understandings up just a bit more. Since that day, belief in Trump’s false claim of a stolen election has led to “sham ‘audits’” and a wave of “new laws not only restricting the vote but also making it easier to sabotage election results” among the several states, and the primary victory of several prominent advocates of the “big lie.”[5] There is also little sign that belief among Republicans in the falsehood of a stolen presidential election has significantly abated. Indeed, the latter Big Lie, propagating through social media and cable news, is effectively a symptom, as well as a cause, of a broader, more diffuse, and yet more profound distemper—a malady of the popular dispositions necessary for democracy to rule.

Superfically, this discontent might seem to have unexpected positive effects. A July 2022 Pew Research Trusts poll found a “vast majority of Americans (85%) said that the U.S. political system either needs major changes (43%) or needs to be completely reformed (42%).”[6] Under other circumstances, such an appetite for change might be healthy, and might be tapped to push through pro-democracy reforms. But this seems unlikely to happen in 2023. Instead, what will likely flow from such discontent in a season of rampant misinformation about election fraud, when even a public performance of murderous political violence is not enough to dislodge false beliefs, is not so likely be redemptive. January 6, in this fashion, poisons not just democratic practice but also the wellsprings that might feed its reform.

Hence, if the January 6 insurrection was—and still remains—a democratic emergency, it is very far from being a straightforward one. It does not invite a simple diagnosis. Nor does it obviously yield to straightforward remedies. To reflect upon it is instead to uncover a messy plurality of entangled crises. Each compels its own analytics, and its distinctive solutions.

To my mind, there are at least three different ways in which the January 6 insurrection manifests and enables a complex, multifaceted democratic emergency. This term is first an appropriate description of January 6 simply as an event—an eruption of violence wreathed in constitutional slogans and defended, paradoxically in the name of (ordinarily nonviolent) democratic practice.[7] Second, that day also revealed an institutional crisis—an unveiling of the fragility of democracy’s legal foundations—that might also be labeled an emergency. For example, January 6 cast an unflattering light on the creaky apparatus for counting Electoral College votes.[8] And third, as already intimated, it manifested a democratic emergency as in the sense of a sharp splintering of the shared public understandings that are essential for the peaceful back-and-forth of elected office via elections. Divergent understanding of supposedly basic empirical facts (e.g., which candidate won the 2020 election), as well as discordant views about what happened on January 6, constitute a crisis of public knowledge—a sense that Americans no longer share a common ground of facts. This is an epistemic crisis as emergency.

These three emergencies run concurrently into each other. They intertwine, Ouroboros-like. And they interact in ways that mutually reinforce. Yet each also unfolds at a different pace; each has its own history. And each works via a distinctive modality—violence, institutional failure, or misinformation. There is, of course, no single “fix” that can alleviate the ensuing amalgam of challenges. To the contrary, part of the challenge of the moment for American democracy is the interconnectedness and the simultaneity of emergencies flowing at different levels toward the same malign end.

In this Essay, I want to isolate one strand of the ensuing problematic—without, to be clear, claiming to offer a resolution for all the multiple, difficult issues at play.[9] There is a historically pedigreed way of analyzing democracy through what is called the problematic of constitutional guardianship. I aim to consider here whether that lens has anything to teach us now. In a democratic system, this problematic asks, which institution(s) (if, indeed, any at all) should take the lead in responding to threats to the basic institutional structure of popular choice? This is not only the question of which institution (again, if any) should have been understood to have the power to act on an emergency basis to sudden, unanticipated threats to democracy—most obviously through violence of the sort used on January 6, 2021—but which institution can and should decide on the existence of any one of the three enumerated emergencies, so as to facilitate responsive state action?

In its classic form, the question of constitutional guardianship has been understood to demand the identification of a single institution, such as a court or an elected entity, to play this role across a range of different crises. Indeed, the historical debate about constitutional guardianship was very much framed as a choice between judicial power and executive power. In developing a parallel between the earlier debate and contemporary circumstances, I will also focus on that specific binary choice, and largely ignore the legislature. This makes sense since today courts and presidents are the most salient actors in a moment of emergency. But we can use an inquiry about constitution guardianship as a guide without being ultimately constrained by that binary assumption. The constitutional guardianship debate, indeed, is rich and suggestive enough to offer lessons even if that binary is rejected as insufficient or misleading.

To center the question of constitutional guardianship is to move away from the diagnostic question of how to define the threats that democracy faces with which I began, and toward the prescriptive question whether particular political or institutional actors should be understood to have a leading role responding to such threats. A constitutional guardian needs to be identified and designed. This choice implicates a whole slew of interesting constitutional choices. The task of guardianship might, for example, be described an ex ante delimited set of “emergency” powers by statute or constitutional text, triggered only under certain conditions. A specific body might be permitted to decide on the existence of an emergency in the first instance, and then define the scope of actions that another body could take. Or, at another end of the spectrum, the guardian might be akin to the Roman institution of the dictator, and so able to act largely unbounded by law.[10] It might then do whatever it perceived as necessary to preserve the constitutional system, regardless of law. The question of constitutional guardianship is, moreover, embedded within a larger inquiry as to how best to design reactive, pro-democracy public powers: How best should public power be organized to maximize the chances of democratic survival? And to what extent should the latter goal be prioritized in institutional design over other ambitions, such as effective, speedy policy-making or guaranteeing the absolute priority of individual rights perceived to be inalienable? Should emergency powers in general be eschewed on the theory that the downside risk of abuse is greater than the upside gain in moments of crisis? As will become clear, I will confine myself to the first of these questions: It alone is complex enough that we can only just begin to make a start on it here.

In Part I, I set forth the most important historical discussion of constitutional guardianship under conditions of democratic backsliding. This rich debate unfolded found almost a century ago in the febrile context of Weimar Germany. It deserves attention now because of the breadth and complexity of the positions developed by its participants, and also because those positions refract and illuminate our own debates. Yet the Weimar-era debate is hardly known outside jurisprudential circles, and so needs to be reconstructed in some detail. In Part II, I turn to the more familiar, longstanding debate about emergency powers under the U.S. Constitution. While much of the literature has focused on the executive’s emergency powers, I draw attention to the role of the Supreme Court in defining and allocating such emergency powers—and (I argue) in the final analysis acting as a constitutional guardian by deciding on the existence of an emergency, and the scope of consequent new authorities. Because other treatments of emergency powers have (in my view erroneously) focused on executive action to the exclusion of all other actors, I take care to elaborate and detail the judicial power in respect to emergencies. The result, I hope, is a useful counterpoint to prior, more executive-centered work that centers courts, rather than the presidency, in the elaboration of emergency powers. In Part III, I link the Weimar debate to the contemporary guardianship question in the U.S. context. I point to ways in which the earlier debate casts light on (and in turn is illuminated by) the present American dynamics. To be clear, I make no claim to settle the question of who should be the guardian of the Constitution in the contemporary U.S. context. More modestly, I hope to show some profitably counterintuitive and unexpected implications and connections between these two historical moments.

  1. The Weimar Debate on
    Constitutional Guardianship

In the late 1920s, the decade-old experiment in Germany democracy started to come undone at the seams.[11] The Reichstag, or federal parliament, was paralyzed by a plurality of vociferous and antagonistic political parties. Factions on both the far left and the far right repeatedly resorted to street violence and attempted putsches to destabilize democracy. Increasingly, the absence of a stable parliamentary majority meant that the president of the republic de facto ruled though the promulgation of emergency decrees. Debate on whether this was desirable centered on the concept of “constitutional guardianship” arose in this febrile atmosphere. This debate largely unfolded before, it should be noted, the appointment of Adolf Hitler as Chancellor, before the Reichstag fire, and before the vertiginous unravelling of the barest semblance of democratic order and human decency in Germany. Some care must therefore be taken to read this debate in light of what its protagonists knew—as opposed to reading it in light of our own, more fulsome knowledge of the horrors to come.

The debate on constitutional guardianship hence arose in a wider context of democratic emergency. The latter is not, of course, on all fours with the current American conjunction, and I want to abjure up front any facile comparisons between variants of mild and deep fascism. But there are enough echoes in the mix of deadlock and violent defection from democracy between now and then to make the literature produced by in that earlier crisis moment relevant again today. Hence, by sketching the political context in which the constitutional guardianship debate emerged, and by teasing out the two key positions in the debate in this Part, I hope to provide readers with enough historical context to see both the parallels and the disjunctions between the Weimar situation and today’s democratic emergency. The Weimar debate came to focus on the question of which institution was best understood as the guardian of the constitution—a question that continues to offer a useful framing today, even if there is no easy answer.

The Weimar crisis in democracy was to reach an acme on July 20, 1932. President Paul von Hindenberg’s decision promulgated an emergency decree authorizing Chancellor Franz von Papen to depose the Social Democratic prime minister of Prussia, Otto Braun, and to permanently install instead federal commissioners aligned with the conservative chancellery.[12] While von Hindenberg cited civil unrest created by the Communists, Nazis, and police, the “real goal” of his “Preussenschlag” was to “wrest control of Germany’s largest state from the social democrats and to make Prussia’s executive power available to the conservative federal government.”[13] Braun’s government, however, challenged the Preussenschlag in a special tribunal called the “Staatsgerichtshof,” which was empowered by Article 19 of the Weimar Constitution to adjudicate disputes between the Länder and the national government.[14] The Staatsgerichtshof issued a divided judgment on October 26, 1932, upholding the emergency degree as a temporary response to a violent and volatile public situation, but rejecting von Papen’s permanent ouster of Prussia’s governing bodies.[15]

This mixed judgment elicited criticism from the leading constitutional scholars of the day. In particular, it prompted sharp responses from the German lawyer-scholar Carl Schmitt and the Austrian jurist Hans Kelsen. Schmitt had represented von Papen in the Staatsgerichtshof, and had earlier written influential texts defending the use of emergency powers under the Weimar Constitution. More infamously, he would later join and robustly defend the Nazi party.[16] Kelsen had previously helped draft the 1920 Austrian Constitution, and had served as a judge on its novel Constitutional Court between 1920 and 1929.[17]

The debate between Schmitt and Kelsen had seeds in earlier debates of the late 1920s.[18] And while some of the key texts I’ll discuss date from 1931 (before the Preussenschlag), the Staatsgerichtshof judgment reached to the very heart of their disagreement. This turned on whether in the first instance the role of “constitutional guardian” fell to the executive branch (as Schmitt urged) or the judiciary (as Kelson contended). While their disagreement had larger jurisprudential dimensions, it useful to separate out this one strand of their debate in order to cleanly isolate the arguments for giving the power to decide on the emergency to one institution or another.[19] Their debate focused on the presidency and the courts; this creates a useful parallel to the contemporary American situation discussed below. But it is important to observe that a wider set of institutional possibilities, including other actors, can also be imagined.

In a 1931 essay, “Der Hüter der Vasfassung,” Schmitt set out to condemn and exorcise the “tendency to portray trial-deciding courts as the highest guardian of the constitution.”[20] To substantiate this claim, he started from the observation that judges’ power of constitutional review derives from their obligation to decide on conflicts between hierarchically ordered bodies of law.[21] This “conflict of laws” function, he noted, is not uniquely performed by courts: Officials and ordinary citizens also need to hierarchically order legal norms, so as to decide which to follow.[22] Courts, in other words, do nothing special. He then made a more controversial theoretical claim: All adjudication, he contends, is “bound to norms” and so impossible “as soon as the content of the norms themselves start to get unclear.”[23] Indeed, as soon as a judge departs from the function domain in which they are under “subjection to . . . a norm,” they can no longer be “an independent judge.”[24] Their decisions are hence necessarily political in character.[25] The ensuing risk of a “politicization of adjudication,” for Schmitt, calls for a firm and closely drawn definitional boundary around the tribunal’s sphere of action.[26] Otherwise, judicial independence “loses its basis in constitutional law to the extent to which it distances itself from the uncontroversial content of the provisions of constitutional statutes.”[27]

An obvious objection to this line of reasoning is that Schmitt overstates the distinction between instances in which there is a hierarchical legal norm for a court to apply, and instances in which there is uncertainty or unclarity about the existence of such a supervening norm. In effect, this objection would go, Schmitt fails to understand that there is often (if not always) a degree of discretion and judgment at work in the judicial task, but that this fact has no destabilizing normative implication. I don’t think, however, this objection succeeds. Schmitt is well aware of the “necessary incompleteness and vagueness of every written constitution.”[28] He would demand, nevertheless, an “obvious and indubitable violation” of the Constitution before a court has permission to act.[29] In recognizing the pervasiveness of legal ambiguity, and circumscribing courts’ role to instances of clear constitutional violations, Schmitt thus tracks the thought of one of the first and most famous American theorists of judicial review. In the late nineteenth century, James Bradley Thayer saw a scope for judicial invalidation only “when those who have the right to make laws have not merely made a mistake, but have made a very clear one . . . so clear that it is not open to rational question.”[30] Schmitt can be read as suggesting that judicial invalidations in the absence of a Thayerian “clear” error constitute “political” judgements of a sort ill-suited (or inappropriate) for a court to make.[31] Put in these terms, his intuition is more narrowly bounded, and also quite familiar to American jurists.

In contrast to his jaundiced view of the judiciary, Schmitt posited the president (i.e., his client) as the proper guardian of the constitution. The Reich president, he asserted, is “elected by the people as a whole” and “endowed with competences that make him independent of the legislative authorities.”[32] He is hence “a neutral and arbitrating mediation of conflicts.” In addition, Schmitt suggested, the president’s primacy as guardian of the constitution is “above all” warranted by the fact as a matter of “political technique and harmony” the president is “able to take action.”[33] Schmitt here is palpably reacting to the fissiparous tendencies of Weimar parliamentarism, which made the creation of a stable governing majority impossible.[34]

This paean to executive expedience finds contemporary echoes in two forms. It not only tracks a functionalist, comparative institutional analysis common to American constitutional scholarship in defense of broad executive-branch prerogatives.[35] In an ontological sense, Schmitt further sees executive action as a “source of law on the most important constitutional issues” in a way that has a second striking parallel to modern American thinking about the presidency as the source of “law” in a crisis.[36]

Hans Kelsen disagreed sharply with both Schmitt’s analysis and his conclusion. In a 1931 essay, “Wer soll der Hüter der Vasfassung sein?”, he responded directly to both.[37] The function of a constitution, explained the erstwhile drafter of Austria’s, is to “impose legal limits on the exercise of power.”[38] But Kelsen then concentrated his theoretical fire on Schmitt’s accounting of the judicial power, which he characterized as “arbitrarily defined.”[39] The core of Kelsen’s argument here is that Schmitt erred when he assumed that “there is an essential difference between the function of adjudication and the ‘political’ function.”[40] To the contrary, every judicial act contains “to a higher or lesser degree, an element of decision”—so politics can never be wholly ousted from adjudication.[41] While this is true, Kelsen thought, the “function of a constitutional court is political in character to a much greater degree [than normal].”[42] Rather than aligning with Schmitt (and Thayer) on the possibility of distinguishing clear from murky cases of error, Kelsen insisted that “adjudication usually begins where the content of norms starts to get doubtful and contested.”[43] Lest there be any doubt as to his view of courts, Kelsen labeled them a “negative legislator” because of their ability to “destroy[] a general norm,” much as the Reichstag could by proper parliamentary action negate an earlier law.[44]

On this point, Kelsen’s experience as a judge might well have been formative. A well-known theorem in law-and-economics predicts that litigants will only engage in the expensive, time-consuming, and uncertain labor of adjudication when they genuinely and substantively differ in their estimates of what the law is.[45] On this account, litigants’ anticipation of the expected costs of adjudication mean that only those instances in which the law is susceptible to divergent interpretations are likely to litigated. As a result, a sitting judge (such as Kelsen) is likely to see ambiguity as pervasive rather than occasional. Further, Kelsen’s broader theory of law—in which illegal actions come about solely because an agent of the state issues a norm that contradicts a higher norm—also helps explain his view of courts. What courts confront in hard cases is not so much a conflict of laws, but rather a “noncontradiction” because of the abiding and ousting force of the higher norm.[46]

What of Schmitt’s proposal in respect to the president? Here, Kelsen offered both functional and theoretical objections. On the practical front, Kelsen observed that the immediate instruments of government, including the presidency, are often parties to constitutional disputes in ways that make them inapt referees.[47] The “only question that matters” is “who is more independent and neutral: a constitutional court or a head of state.”[48] For Kelsen, this question almost answered itself.[49] As a matter of constitutional theory, Kelsen contended, Schmitt’s depiction of the presidency as neutral rests upon a dangerous fiction, the “doctrine of the total state.”[50] This is the assertion that the Weimar Constitution presupposes a united sociological entity—the “pouvoir constituant” of the Abbé Sieyès,[51] or a Rousseauian “general will of the unified people.”[52] Kelsen flatly denies that a president selected under the “high pressure of party-political actions” could, or indeed ever did, reflect this sort of unitary, foundational form of popular sovereignty.[53]

What should we make now of this disagreement between then-eminent jurists? An obvious point is the empirical fallacy of Schmitt’s claims about President von Hindenberg, who was most certainly nor “neutral” or above party politics.[54] In his briefing to the Staatsgerichtshof, Schmitt tellingly had not made an argument about neutrality. He had instead contended that a presidential determination of how emergency powers were to be used could not be subject to judicial review.[55]

Bracketing its inattention to the contemporaneous facts of the Prussian dispute, Schmitt’s theory might be cleaved into two parts for the purposes of thinking about the present American democratic emergency. There is first his narrow view of what courts should do, which tracked Thayer’s. Second, there is his construal of the presidency as the distinctive vessel of popular sovereignty. We can accept or reject each element separately from the other. In my view, we should reject Schmitt’s view of the executive as motivated by a political theory of sovereignty that most of us now rightly reject; his account of judicial power, however, is less easily set aside.[56]

On the issue of how to characterize the presidency, it seems to me that Kelsen’s argument—that Schmitt intended a “transformation of the parliamentary system into a constitutional monarchy in plebisctarian guise”—is a correct and damning critique.[57] Schmitt’s “merging” of two distinct ideas of “emergency powers and the question of in which charismatic institution sovereignty lies” lies at the heart of his account of the executive.[58] (Having warned against prochronic readings, it is hard still for me to read this sleight of hand without anticipating Schmitt’s subsequent turn, in 1933, to the Nazi party: That regime, of course, rested on the political claim of a single person to represent the authentic and singular voice of a prepolitical people shorn of interest-group conflicts and other divisions.[59]) Schmitt here is vocalizing in high-theoretical terms a kind of “moralistic imagination of politics” as a Manichean confrontation between a morally purified “people” and a corrupt and irremediable “elite” that has animated almost all populist rhetoric for the past century.[60] As such, his approach bodes poorly for constitutional values of equality and individual dignity: Both these values are hard to square with the Schmittian account of popular sovereignty, especially for racial or religious minorities.[61] Kelsen’s view that the guardian protects “the identity of the constitution rather than the identity of the polity” has an obvious, contrasting appeal.[62]

What, though, of their argument about the proper role of the courts? On the one hand, Kelsen was surely correct that Schmitt’s argument proceeds from a contestable premise about what a court is—a body that applies existing law, but that does not itself legislate. On the other hand, Schmitt’s conception of the limited compass of judicial power has clear resonances in American jurisprudence, where it resonates with concerns about counter-majoritarian actors and the rule of law. In contrast, Kelsen’s more frank recognition of the “political” function of judges is likely to produce a frisson of anxiety. There is a long history in American law of criticizing “judicial activism.”[63] This accusation has become even more sharply focused as the federal judiciary has become increasingly polarized in the past decade.[64] The definition of “judicial activism,” of course, varies in meaning and application over time.[65] But the very frequency and promiscuity with which the term is used is some indication that there is a widely held sense that judges should play a circumscribed role in “politics.” That view of the judiciary is much closer to Schmitt’s than to Kelsen’s, whatever the merits of the former’s views about executive power.

In their debate on the guardian of the constitution, both Schmitt and Kelsen saw ways in which these accounts of judicial power bore directly on the question of which body best determined the availability of discretionary emergency authorities. Schmitt’s narrow understanding of the appropriate judicial role operated as a tight leash on its potentiality in situations of emergencies. In contrast, Kelsen’s more frank recognition, and even embrace, of the political nature of judicial action opened the gate to a more capacious judicial role in emergencies. To be sure, nothing in Kelsen’s account suggested that judges would always be front-line actors; rather, he is best read as insisting—against the position that Schmitt took as the president’s lawyer in the Staatsgerichtshof—that courts could and should exercise plenary judicial review in respect to the availability, and hence the legality, of emergency powers.

It seems to me one can hold on to this piece of the clash between Schmitt and Kelsen and ask if it continues to have currency in thinking today about the institutional choice question in democratic emergencies. And one can do so even if one thinks that Kelsen’s critique of Schmitt’s presidentialist account of popular sovereignty is devastating.

  1. Does U.S. Democracy have
    a Constitutional Guardian?

How then does this debate bear on the situation of the contemporary United States? To begin with, it is worth noting that the historical practice of the American high court in the late nineteenth and early twentieth century lurked in the background of the Schmitt-Kelsen debate. In his 1931 essay, Schmitt acknowledged the distinctive (and to him anomolous) role of the U.S. Supreme Court in American politics. But he explained it away by noting that the U.S. Court “reviews the justice and reasonableness of statutes with the help of general principles and fundamental contradictions” that can only be called “‘norms’ if one is willing to abuse the term.”[66] Writing of the Lochner-era federal judiciary, he also characterized the Court as “the guardian of a social and economic order that is undisputed in principle.”[67] That is, on Schmitt’s account, what the U.S. Supreme Court was doing was not adjudication—it wasn’t even an application of “law” per se—rather it was a defense of a social order that was extraneous to the Constitution. Kelsen, in contrast, thought the American high court was “not doing anything different from German courts.”[68] He might have added that Alexander Hamilton anticipated his argument when he spoke of the judiciary’s “duty as faithful guardians of the Constitution,”[69] and not of the social order more generally. In other words, neither one thought that the U.S. Supreme Court had been constrained to the enforcement of legal norms, or that it had kept out of “politics.” Neither, at the same time, extended their analysis to the location or operation of emergency powers in American law.

The Schmitt/Kelsen debate hence stages the question of which institution is the guardian of the constitution in American law, and offers a pair of sharply divergent answers. The question is best approached by asking first what democratic emergency powers exist under U.S. law, and then asking who decides when a given power can be used. Having worked through these details of the American dispensation, we can return to the constitutional guardianship debate with a greater appreciation of its contemporary implications.

Very roughly, there are two kinds of emergency powers in American (federal) law.[70] First, the Constitution supplies positive authorities for the state to act and also imposes negative limits on power. An emergency power might be defined as a positive grant of power. Otherwise it can operate as a carve-out from some otherwise indefeasible negative right. Second, federal statutes supply a wide range of “emergency” powers; some of these are denominated as such, while others work as, in effect, instruments though which an emergency response can be mounted. Of course, claims of constitutional authority often inform the assertion of statutory powers, but the two categories can still be treated separately.

In almost every case of either constitutional or statutory emergency power, it is agents of the federal executive branch that have the power to exercise an emergency power. But this is consistent with both Schmitt’s and Kelsen’s account. Neither of them suggests that judges would be directly administering emergencies. The more pertinent question for the purpose of discerning whose account governs in the United States goes to an anterior matter—i.e., who decides on the legality or permissibility of an emergency measure, whether constitutional or statutory?[71] Of course, this is the question that the Staatsgerichtshof had to answer.

Surprisingly often, it is not the president, but the federal courts that have a final say in respect to whether it is lawful to act in respect to an emergency. This is not always so—and there are important instances (including in respect to the violent event of January 6) where courts have no real say. Yet to many readers, I think even the weaker claim of “often but not always” will sound counterintuitive, or simply wrong. Surely, they will reason, an emergency by its very nature unfolds rapidly. It allows for very little time, let along for running to court. Certainly, this seemed true on January 6—an event that unfolded in a matter of hours; it was equally true on September 11, 2001, another violent catastrophe. Emergencies, in short, seem to offer no space or time for a court to find a foothold for intervention.

But I think this understanding of emergencies is incomplete. To begin with, it assumes that the only sort of emergency is one that unfolds in a very tight timeframe. This is not so. There are many emergencies—around the climate and extreme weather, public health, infrastructure, and more—that play out achingly slowly. Moreover, even a precipitous, seemingly time-limited emergency such as the January 6 insurrection will often have a prehistory—i.e., a sequence of historical predicates that double as missed opportunities for intervention. Similarly, the Preussenschlag occurred on a single day, but cannot be untangled from the longer pathway of legislative gridlock and fragmentation that characterized Weimar governance. By analogy, the event of January 6 may have been narrowly time-bound, but its institutional and epistemic dynamics extended long before and long after the day of the insurrection. Defining that democratic emergency uniquely in terms of a single violent end, in other words, is a little like identifying an iceberg as the material appearing above the water’s surface. In this way, emergencies are not just more varied than the common intuition might suggest, even the canonical examples turn out to be more temporally extended than the news cycle allows. As a result, courts have more opportunities to shape the conditions in which an emergency comes to a head than is commonly appreciated. Further, because emergencies tend to linger (as the institutional and epistemic facts of the January 6 emergency have persisted, and even gotten worse), a judicial response to a violent climax may well shape how other actors respond to an emergency’s aftereffects.

There is another way in which courts can shape emergency responses: They can define ex ante the range of powers that are available. Importantly, I will suggest below, there are certain elements of constitutional jurisprudence—in particular the articulation of exceptions to textually absolute rights—that function in effect as ex ante grants of emergency powers. Because we do not discuss them with this nomenclature, we miss an important element of the judicial authority to define emergency powers embedded in constitutional rights discourse.

When courts do act, they intervene in two different ways. First, judges define the substantive scope of power to authority the emergency override of constitutional rules and determine the substantive scope of statutory emergency powers. In effect, they define the material scope of emergency powers. Second, they decide on the availability of mechanisms to challenge such emergency powers. In this respect, they decide whether the executive has de facto unfettered power to act because of the absence of ex post review, or whether its actions will later be held to account under law. These two judicial powers have been exercised both to say which emergencies count, and when the exercise of emergency powers can be challenged.

How then do American courts exercise these powers? As a rough first approximation, the following generalization holds: Courts have exercised these powers so that when the executive responds with physical violence against an emergent threat, their decision will not be constrained ex ante by law, and often will not be subject ex post to judicial second-guessing.[72] In contrast, where the executive responds to an imminent or slowly unfolding emergency with a durable policy decision, the latter will generally be subject to judicial review.

In the ensuing jurisprudence, moreover, different sorts of emergencies receive varying degrees of judicial deference and accommodation. Claims about emergencies linked to national security now get more leeway from courts than other kinds of emergency-based argument. In particular, courts have demonstrated little or no concern about risks to democracy as a species of emergency warranting exceptional action. This is striking given the increasing evidence that the United States’ democracy is subject to increasing strains. Courts have hence created a regime that is Schmittian when it comes to immediate violence, and Kelsenian when it comes to regulatory and democratic policy-making. As such, the American system of emergency powers represents a peculiar blend of the two approaches to emergency actions. Crucially here, this is largely a result of an overarching control exercised by courts in the fashion Kelsen endorsed.

To develop this account of the guardianship role in American law, I want to start with a description of constitutional emergency powers. I then turn to statutory emergency powers. Finally, I consider when judicial review is available and when it is not. The following account, I stress, works at a high level of generality: I aim here to offer a gestalt, and not a fine-grained, picture. In what follows, I will focus on case-law, rather than on federal statutory powers (which have been treated well elsewhere[73]). My aim is to show that there is a larger “jurisprudence of emergency” than is commonly appreciated, and then to reflect on the significance of that fact for the Kelsen/Schmitt debate.

  1. The Constitutional Substance of
    Emergency Powers

With one potential exception, the U.S. Constitution does not spell out any emergency powers in the manner of other nations’ organic documents.[74] Its only “limited grant” of emergency powers is located in the Suspension Clause of Article I, which concerns the availability of the habeas corpus remedy for executive detention.[75] It is not clear this emergency power, though, is terribly important on the ground. In actual practice, the habeas writ has not been a terribly effective remedy for detention outside the criminal justice system because of the ability of executive-branch actors to delay, derail, or otherwise drag their feet.[76] Further, as we will see in a moment, the absence of an explicit constitutional framework for emergency powers has become less important because of the extent and numerosity of statutory emergency provisions.

The Constitution, nevertheless, does contain an array of other individual entitlements to speech, liberty, privacy, and non-discrimination among others. All of these would seem to preclude certain kinds of emergency responses. For example, if a specific political movement presented a risk to democracy, the First Amendment’s protection of speech might be thought to impose an impediment to restrictions on speech. If that movement was defined by its members’ religion or race, the nondiscrimination commitments of the First Amendment and the Equal Protection component of the Fifth Amendment might be expected to bar the state from singling them out for punishment or prescription, at least in certain ways. If that movement was to be addressed with state surveillance or outright state coercion instead, one might think the Fourth Amendment’s protections would be relevant and would stymie government action. And so on.

Given this accent on negative limits on state power, one way to think about the Constitution’s stance on emergency powers is to focus on whether and when these negative limits can be overcome, as opposed to the existence of positive authorities to act. At first blush, the constitutional text looks like this should be a short conversation. On its face, the Constitution describes all these individual interests without carving out any exceptions.[77] So there is no hint in the First Amendment, the Fifth Amendment, or the Fourth Amendment that constitutional entitlements change in strength across different factual scenarios, let alone that there are wholesale exceptions in cases of emergency.

Yet, exceptions exist, and even proliferate. These exceptions to the prohibitory force of constitutional rights are not a textual matter. Instead, they are a result of judicial choices made without textual warrant. The judicial exceptions to constitutional rights take a number of different doctrinal forms, but their effect is largely identical: To excuse state actors from compliance with an individual constitutional entitlement that would otherwise preclude some response to a perceived emergency. The Court, moreover, has not been even-handed when it comes to different flavors of emergency. Exceptions to constitutional prohibitions are often recognized when the executive asserts a public safety or national security interest. In contrast, state and federal executive efforts to respond to public health crises, such as the COVID-19 pandemic, have been rebuffed.

To begin with, almost every constitutional right comes packaged with exceptions as a result of the way in which the Supreme Court formulates them as working legal doctrine. In the Equal Protection context, for example, the Court has demanded that measures containing a racial classification on their face must be “narrowly tailored” to address a “compelling government interest” even in contexts where the state is acting to mitigate the threat of violence.[78] In general, non-white litigants who have challenged federal immigration and security-related actions on racial discrimination grounds tend to lose. The Court uses various doctrinal means to diminish or else to dismiss evidence of impermissible intent.[79] While the Court has recently seemed to repudiate its notorious precedent upholding the Japanese-American internment,[80] the persistence of judicial deference to the government on questions of both law and fact means that there is much less to that disavowal than first appears.[81] It is not hard to imagine a fresh internment policy that would not be invalidated by the Roberts Court, but that would be just as unjust as the internment during World War II. In the immigration context, constitutional review is further diluted by “an amalgam of super-deference, suspect class treatment, and even intermediate scrutiny, depending upon whether immigrants are present in the United States lawfully or not, and whether a state or federal classification is at issue.”[82]

The point to emphasize here is that these exceptions to facially comprehensive constitutional rights are a result of judicial choice. It is the Court that has carved out exceptions to the warrant requirement of the Fourth Amendment without any guidance from the text, and it is Justices who determine what counts as a “compelling state interest” in the application of strict constitutional scrutiny. Such exceptions are created “on the basis of little or no textual inquiry.”[83] Only in the Second Amendment context has the supposedly originalist Roberts Court wholly abandoned the enterprising of teasing out or expelling into the ether compelling state interests as and when it strikes the Justices as appropriate.[84] The difference between the Second Amendment’s treatment (no exceptions) and the treatment of Equal Protection, Free Speech, and Fourth Amendment privacy (all subject to many carve-outs) suggests that the Court exercises broad and unguided discretion as to whether to create exceptions in the first place. Certainly, it stretches credibility to say that the Second Amendment is so important in the constitutional architecture that it requires a different doctrinal articulation from every other constitutional right, including the First and Fourth Amendments. Its differential treatment is a matter of freewheeling judicial discretion, and not the outcome of ex ante legal constraint.

There is a second species of discretion at work when it comes to selecting the kind of state justifications that can warrant emergency action. A feature common to our constitutional law of emergency exceptions is the preferential treatment of national security justifications. For instance, in the most recent case in which a free speech claim was made against a terrorism-related regulation, the Supreme Court underscored that political branch claims of necessary were “entitled to deference,” and so upheld the challenged law.[85] In recent First Amendment cases involving labor association[86] and violent video games,[87] it in contrast explicitly refused to extend the same sort of deferential review. In the Fourth Amendment context, “national security” does not work as a blanket exception to the warrant requirement,[88] but there are gaping exceptions for “public safety”[89] and border control.[90] The exceptional nature of security emergencies is likely to be sharpened by a Court that has turned increasingly to the political right. Justice Thomas, for example, suggested only “national security constitutes a ‘pressing public necessity[]’ that can satisfy strict scrutiny” when it comes to Equal Protection.[91] He has further suggested that strict scrutiny does not apply when a racial classification is used in a carceral context[92]—a conclusion that a fortiori might be thought to extend to other security related contexts. And in a case involving an extensive public record of anti-Muslim sentiment, blithely trashing that historical record as “unpersuasive.”[93] The latter is an important example of how a judge can egregiously contort a factual record to oust a constitutional right of effectual substance for a disfavored group. It is, in effect, a lesson in how to create an emergency power to override constitutional rights through the embrace of misinformation.

Starkly absent from this enumeration of emergencies are instances in which the state acts precipitously for non-security reasons. Recently, for example, the Court has invalidated several states’ pandemic restrictions of collective activity on religious liberty grounds.[94] Public health did not lend the same kind of justificatory heft as national security. Further, and of perhaps greater relevance here, the Court has declined to allow lower courts to use their equitable authorities under conditions of a public-health emergency to facilitate the exercise of the franchise. That is, it has declined to use the federal equitable power that was deployed in the religious liberty cases to enable the exercise of the franchise in the same way that it has enabled the exercise of public religious worship. For example, in April 2020, it vacated a district court injunction issued in Wisconsin to enable absentee voting during a primary election.[95] Because the city of Milwaukee had already closed 175 out of its 180 polling places,[96] voters had to wait in long lines, increasing their chances of catching the virus, or else had to forego voting.[97] In July 2020, the Court also stayed an injunction of Alabama’s witness requirement for absentee ballot, without giving any reasons for doing so.[98] And in other cases, it has repeatedly narrowed lower courts’ power to enjoin election-related practices based on public-health concerns around the pandemic.[99]

These election cases, to be clear, involve the emergency powers of the federal courts rather than the federal executive’s capacity to act in a crisis.[100] I raise them here to underscore the simple point that the Court has resisted the invocation of public health as an emergency. The Wisconsin case nevertheless suggests that the Court does not view the fact that a large number of eligible voters are suddenly deprived de facto of the ability to cast a ballot as an “emergency.” This suggests that a democratic emergency, no less than a public health emergency, will not receive much by way of accommodation from the present federal courts.

Across both these cases and the broader sweep of emergency jurisprudence, two things are tolerably clear: First, the scope of any emergency power is a matter for the courts to determine. And second, the Supreme Court makes judgments about which emergencies count, and which rights are urgent in ways that cannot be rooted back in the Constitution’s text. (Indeed, there is little or no effort to do so.) This is, at bottom, therefore a regime consistent with Kelsen’s vision of who should be the guardian of the constitution.

  1. Statutory Emergency Powers

Federal statutory law also contains a broad array of emergency powers.[101] Many can be exercised apparently at the relatively uncontrolled discretion of the executive branch.[102] And many turn upon national security,[103] reflecting the view that there is a distinctive category of public policy in which the nature of emergencies varies consistently from other domains. The extensive delegation of statutory emergency powers also makes the absence of any constitutional specification of emergency powers less consequential.[104]

Statutory emergency powers generally vest the executive branch with authority to act, rather than expediting legislative or judicial action. Again, this has the appearance of a Schmittian solution: It is the president or an executive-branch agent who determines whether to act in the first instance, albeit at a legislative invitation. But here again, the ensuing actual exercise of statutory emergency powers is frequently (albeit not invariably) subject to judicial scrutiny. The results of such judicial oversight are often not predictable based on the text of the relevant statute. Instead, the outcomes reflect an exercise of judicial discretion just as much as the palimpsest of constitutional exceptions is a result of judicial choice. Kelsen, not Schmitt, is here again the presiding spirit.

I should note at the start that it is difficult to draw conclusions about the changing pace or nature of judicial review in statutory emergency cases over time. As a result, it is hardly obvious whether there is a time trend in the pace of judicial review of emergency measures between different presidencies. Perhaps the pace of emergencies is changing over time as the security environment and the climate become more disordered.[105] Without understanding how the base rate of emergencies in the world changes, it is hard to say much about time-related changes in the patterns of judicial behavior. But the Trump Administration’s addiction to the dramatic rush of a surprising, destabilizing snap decision, coupled to the genuine (and as of this writing, ongoing) crisis of the COVID-19 pandemic, led to what seemed a sudden squall of litigation challenges to the use of emergency powers.

To be clear, many of the ensuing decisions are not formally denominated “emergency power” cases. Nevertheless, in each instance, the Court flexed its interpretive authority in respect to a government action that expressly purported to address an emergent, often unexpected condition or crisis. The extraordinary use of statutory powers, no less than the ordinary use of extraordinary statutory powers, can hence elicit from the courts a jurisprudence of emergency that is denser and more illuminating than past case-law.

Once again, the Court has been profoundly unsympathetic to state efforts to mitigate the costs of the COVID-19 pandemic. For example, it has invalidated vaccine mandates issued by the Occupational Health and Safety Authority.[106] When the Centers for Disease Control (“CDC”) imposed a nationwide moratorium on evictions in reliance on a decades-old statute that authorizes it to implement measures like fumigation and pest extermination, the Court enjoined it, acerbically stating that it “strain[s] credulity to believe that this statute grants the CDC the sweeping authority that it asserts.”[107] The Court’s views of other kinds of emergencies can be observed in its responses to climate change and migration. In June 2022, it stripped the Environmental Protection Agency (“EPA”) of effectual authority to regulate greenhouse gas emissions to mitigate a climate-related catastrophe.[108] This was particularly striking since the Biden Administration had neither proposed nor promulgated a regulation to that end.[109] The Court’s majority justified the availability of Article III jurisdiction by citing the Government’s failure to categorically disavow any future possibility that it would “reimpose emissions limits.”[110] The same day, however, a challenge to the Trump-era “remain in Mexico” policy was resolved in favor of the government.[111] The other important question of emergency powers[112]—Trump’s exercise of statutory authority to redirect federal funds to the construction of a wall along the Mexican border—was set aside before the Court could intervene.[113] In other instances, the Trump Administration was relatively successful in preventing challengers to federal policies from obtaining interim or final relief from lower federal courts.[114]

The point here is simply that far more questions of statutory emergency powers end up in court—and hence end up being resolved by judicial rather than executive interpretation—than might be assumed. To be clear, I am not suggesting that all statutory emergency powers are amenable to ex ante or in medias res judicial review. In some cases, it is certainly the case that the President who exercises a Schmittian authority to decide on the existence of an emergency, and then to use a consequent legal power. One important example of the latter relates to the violent event around the Electoral College vote count. On January 6, 2021, the arguably most relevant emergency power concerned the deployment of the D.C. National Guard (“DCNG”) to quell the insurrection. The President is by statute the Commander-in-Chief of the DCNG.[115] The authority to decide whether or not to exercise emergency powers to address the violence at the capital, as a result, lay with the person who had arguably catalyzed that violence that day, and stood to gain from it politically. This is not a problem of judicial control; nor is it obviously a problem that the judiciary can address: The risk here was one of abuse through presidential inaction. The latter—unless the defendant is the EPA, apparently—is usually insufficient to create an Article III case or controversy. We hence do not have much evidence of how the courts would exercise its lawmaking power in respect to violent democratic emergencies—but there is little reason to think it will be any more accommodating than it is respecting public health concerns.

In sum, federal courts have been capable and willing to engage in the legal review of a diverse array of statutory powers. Some challenges have been parried by nonjusticiability doctrines—but such decisions seem the exception rather than the rule.[116] Like the judicial articulation of constitutional emergency authorities, so too with statutory emergency powers. The exercise of a Kelsenian power to supervise the latter, once again, has yielded an uneven expression of executive authorities. Some emergencies, it would seem, matter more to the Court than others.

  1. Remedies Against the Misuse
    of Emergency Powers

There is a third way in which judicial action shapes the scope of emergency powers. Whether an emergency action can be challenged turns on the availability of a procedural mechanism for judicial review. If there is no such mechanism granting access to court, the state has a greater degree of freedom to act.

But pathways to review are not equally available to challenge all forms of state action. There is a gap in present case-law between the generally uncluttered path that litigants challenging general policies have, and the difficulty experienced by litigants challenging discrete acts of violence.[117] Hence, polluting firms can challenge the EPA’s power to address climate change even in the absence of any actual regulation.[118] In contrast, the Court has narrowed the circumstances in which individuals can seek damages based on a federal official’s unconstitutional use of violence, first in cases in which national security concerns were manifest,[119] and then when such concerns were, at best, peripheral.[120] The consequence of these decisions is that the federal use of physical coercion—violence or detention—will often not be amenable to ex post remediation. While superficially justified in terms of the absence of a congressionally created right of action,[121] these decisions are in practice read as indicators of the Court’s indifference to the costs of state violence.

As a result, some range of imaginable emergency actions are not de facto amenable to judicial review—as a consequence of the Court’s own decisions. At the core of this zone of emergency discretion is the power to use force in discrete and individual cases. In contrast, where a response to an emergency takes the form of a durable policy, the latter will usually be exposed to judicial review.

  1. Summary

The United States does have a system of emergency powers. While it places the executive at the front line, the scope of authority thereby exercised is only notionally determined by statute or the Constitution. It is instead courts, and in particular the Supreme Court, that usually (but not always or inevitably) regulates the availability of emergency powers. It does this by reading in exceptions to constitutional rights (or refusing to do so), and by alternatively stretching and pinching statutory text. The EPA loses the benefit of the create and broad sweep of the Clean Air Act, on the one hand. The Department of Homeland Security, in contrast, tends to benefit from friendly readings of the immigration statutes.[122] Textual lacuna count against tort plaintiffs seeking to hold federal agents to account for unconstitutional violence; the absence of an actual, existing controversy is no problem when coal companies wish to sue the EPA. And so on.

The point is that the fabric of emergency law is a judicial creation. For all the talk of “Schmittian” administrative law by the jurists,[123] we are living in a world shaped by Kelsen’s constitutional vision.

III. The Weimar Debate Revisited

I have argued that the American resolution to the question of emergency powers looks more like Kelsen’s than Schmitt’s account. At first blush, this might seem reassuring. Schmitt’s theories of constitutional guardianship by the executive were not just powerfully critiqued by Kelsen, but also have been strongly critiqued and broadly rejected by contemporaneous American students of emergency powers.[124] The latter focused instead on the “legislative aspects of emergency powers,” and in particular on the appropriate scope of rule-making delegations and administrative justice.[125] It is fair to ask whether a recent uptick in interest in Schmitt’s thought is motivated by the latter’s integrity and coherence, or rather an antipathy (whether from a left or right vantage point) toward what roughly could be called “legal legalism,” or the “compulsion of legality [as] a permanent feature of advanced liberal democracies.”[126]

Yet I wonder whether the American emergency powers dispensation suggests that there is more to be said about the institutional choice question. That is, it may well be that the flaws of Schmitt’s embrace of the Reich presidency have been well canvassed, but we have yet to ask about the conditions under which Kelsen’s solution works. Kelsen, to be sure, well understood that his preferred solution was no panacea. He noted that “[c]onstitutional guardianship for a democracy becomes irrelevant if democracy has already failed by losing the support of the majority of the people.”[127] I wonder, however, if there is not more to say about when Kelsen’s solution of judicial oversight succeeds, and when it fails. Put crudely, the question is when does the rule of judges, at least in respect to the existence and exercise of emergency powers, conduce to legality and the democratic process that underwrites it?

There are a few ways of thinking about why Kelsen’s argument for courts is inapplicable, or at least an imperfect fit, in the American context. Given the limited scope of this article, I will offer here only one such argument (and the most obvious one at that).

Kelsen’s view that only “parliament and government tend to be parties in disputes” over the constitution, whereas the court is a “third institution” sitting at a removal, may not be a sound description of the U.S. case at present.[128] Scholars have long observed the political party system has “tied the power and political fortunes of government officials to issues and elections,” which in turn fostered “a set of incentives that rendered these officials largely indifferent to the powers and interests of the branches per se.”[129] It is not clear why this observation about “separation of parties, not powers,” should be applied only to legislative/executive relations, and not the judiciary. It has long been clear to the informed observer that there is a close relationship between national politics and federal judicial action.[130] There is no reason this nexus will abate soon. To the contrary, the Supreme Court in the past two years has moved sharply away from the preferences of the average American, and expressed preferences closely aligned with the average Republican.[131] The connection between this shift and the increased number of Republican appointees needs no elaboration. Under these conditions, it is not surprising that the Court has become less sensitive to the weight of democracy as a normative value in the constitutional order—especially when doing so would disadvantage elected officials with whom the Court’s majority is aligned. It should also not be a surprise that the Court, in respect to electoral management during the pandemic, has undermined rather than advanced the effective operation of democracy at a moment of (public-health) crisis when doing so had expected partisan effects.[132]

Kelsen, while acknowledging the possibility of an elected court, did not reflect on the way in which a consistent alignment between a court and one national faction might change the way in which the Court behaved. Constitutional guardianship under these conditions will tend to under-supply certain forms of democracy-related prophylaxis.

One of Kelsen’s observations that startles the modern ear with its precision is that the “function of a constitutional court is political in character,” and to “a much greater extent” than that of other courts.[133] So long as constitutional design maintains a tolerable level of insulation between the judiciary and partisan forces (embodied, typically, in elected offices), the “political . . . character” of this function may not be very consequential. But when the judiciary becomes polarized, as a consequence of more general patterns of partisan polarization among political elites, new dynamics—unexpected by Kelsen yet implicit in his adage—might well emerge. The temptation to use emergency powers to suppress one side’s votes, to torque electoral frameworks in favor of one side over the other, and to facilitate the violent stanching of dissenting speech, may become overwhelming. The allocation of the guardianship role to the courts—which otherwise may have been quite sensible—suddenly becomes a source of instability.


Where then does this leave us? I have used this essay to recall the Kelsen-Schmitt debate to American readers’ attention, and to suggest that it can be used as a frame through which to reconsider our current democratic emergencies, including the events of January 6. While the scholarly consensus has long been that Kelsen made the better argument, in fact there are important kernels of truth in Schmitt’s critique of judicial power. Particularly under conditions of partisan stress, a judicial guardian of the constitution poses distinctive threats.

That is not to say, however, that a comprehensively Schmittian solution would be any better. To the contrary, it may well be that there are only bad and worse solutions to the problem of emergency powers under conditions of relatively extreme polarization given our current institutional configuration. To pick up an idea I mentioned at the beginning of this Essay, perhaps we’re better thinking beyond the options supplied by our present constitutional architecture. Indeed, at some point, it may be that our inherited institutions are simply not up to the task of offering an effective a guardian of the constitution—and there is a desperate need to innovate under political conditions in which innovation is very difficult indeed.

Still, even if it is too late for us, it is perhaps not too late for others to draw lessons from our hasty and presumptuous belief that judicial power would somehow magically lead to the preservation of legality and democracy.

        [1].     Frank and Bernice J. Greenberg Professor of Law, University of Chicago Law School; the Frank J. Cicero Fund supported this research. Thanks to Jill Hasday for incisive comments. All errors are mine.

        [2].    Brian Naylor, Read Trump’s Jan. 6 Speech, A Key Part Of Impeachment Trial, NPR (Feb. 10, 2021, 2:43 PM), Trump had also set the ground for the riot by aggressively spreading false claims of fraud. See, e.g., Jim Rutenberg, Nick Corasaniti & Alan Feuer, Trump’s Fraud Claims Died in Court, but the Myth of Stolen Elections Lives On, N.Y. Times (Oct. 11, 2021), politics/republicans-voter-fraud.html.

        [3].    Brett Samuels, Dramatic Testimony: Pence Security Detail Feared for Lives During Riot, The Hill, July 21, 2022.

        [4].    Deirdre Walsh, Congress’ Electoral Count To Resume After Violent Protests Halt Process, NPR (Jan. 6, 2021, 8:00 AM),

        [5].    Richard L. Hasen, Identifying and Minimizing the Risk of Election Subversion and Stolen Elections in the Contemporary United States, 135 Harv. L. Rev. F. 265, 265–66 (2022); see also Jennifer Medina, Reid J. Epstein, & Nick Corasaniti, In Four Swing States, G.O.P. Election Deniers Could Supervise Elections, N.Y. Times (Aug. 3, 2022),

        [6].    Katherine Schaeffer, On July Fourth, How Americans See Their Country and Their Democracy, Pew Research Trust (Jun. 30, 2022),

        [7].    See Aziz Z. Huq, On the Origins of Republican Violence, Brennan Ctr. for Justice (Jun. 29, 2021),

        [8].    For a useful history and analysis of various reform possibilities, see John D. Feerick, The Electoral College: Time for a Change?, 90 Fordham L. Rev. 395 (2021).

        [9].    In other word, Tom Ginsburg and I have offered a set of institutional prescriptions for mitigating systemic risk to democratic institutions. See Tom Ginsburg and Aziz Z. Huq, How to Save a Constitutional Democracy (2018). And in a forthcoming article, we consider how such legal and constitutional reforms might be sequenced. See Tom Ginsburg & Aziz Z. Huq, The Pragmatics of Democratic “Front-Sliding,36 Ethics & Int’l Aff. 437 (2023). The question of institutional leadership I address here is distinct from those inquiries.

      [10].    The temporary dictators appointed by the Roman Senate in fact were constrained both “formally and informally,” albeit in imperfect ways. Nomi Claire Lazar, Making Emergencies Safe for Democracy: The Roman Dictatorship and the Rule of Law in the Study of Crisis Government, 13 Constellations 506, 510–12 (2006) (describing the process of appointment and powers of the office).

      [11].    For an excellent recent account, see Benjamin Carter Hett, The Death of Democracy: Hitler’s Rise to Power and the Downfall of the Weimar Republic (2018).

      [12].    Lars Vinx, The Guardian of the Constitution: Hans Kelsen and Carl Schmitt on the Limits of Constitutional Law 1, 3–4 (2015). Vinx notes that von Papen had drafted the emergency decree. Id. at 1.

      [13].    Id. at 1.

      [14].    Id. at 4.

      [15].    Id.

      [16].    Id. at 9–10; see also Carl Schmitt, Dictatorship (Michael Hoelzl & Graham Ward trans., 2013).

      [17].    Vinx, supra note 12, at 7; see also Hans Kelsen, Judicial Review of Legislation: A Comparative Study of the Austrian and the American Constitution, 4 J. Pol. 183, 188 (1942) (defending the Austrian Constitutional Court after its dissolution).

      [18].    Vinx, supra note 12, at 7.

      [19].    There is a scholarly debate over whether, and to what extent, Schmitt was ever an earnest conservative defender of the Weimar Republic, and whether his turn to fascism was a minor or a major change. Certainly, his views on the scope of emergency powers to be exercised by the executive changed dramatically over time. See John P. McCormick, The Dilemma of Dictatorship: Carl Schmitt and Constitutional Emergency Powers, 10 Canadian J. L. and Juris. 163 (1997) (extensively documenting the shift from the early 1920s to the 1930s).

      [20].    Carl Schmitt, The Guardian of the Constitution: Schmitt’s Argument Against Constitutional Review, in Vinx, supra note 12, at 79 [hereinafter “Schmitt, Guardian of the Constitution”].

      [21].    Id. at 83 (noting that judges do not engage in the “denial of the validity of the ordinary statute. It is only a non-application of the ordinary statute to the concrete case at hand that occurs by virtue of an application of the constitutional statute.”).

      [22].    Id. at 89.

      [23].    Id. at 87.

      [24].    Id.

      [25].    Id.

      [26].    Id. at 91; id. at 101–02.

      [27].    Id. at 103; id. at 107–08 (noting the “boundaries that the rule of law imposes on legitimate adjudication”).

      [28].    Id. at 101.

      [29].    Id. at 116–17 (adding that “doubt about the content [may be] so well founded” that it is “impossible to speak of a violation”).

      [30].    James B. Thayer, The Origin and Scope of the American Doctrine of Constitutional Law, 7 Harv. L. Rev. 129, 144 (1893).

      [31].    Id. at 135. As Judge Posner notes in his account of Thayerian review, “questions relating to the power of the different branches of government are inescapably political, and so courts have perforce to use political, rather than just legal, criteria in answering them.” Richard A. Posner, The Rise and Fall of Judicial Self-Restraint, 100 Cal. L. Rev. 519, 524 (2012).

      [32].    Schmitt, Guardian of the Constitution, in Vinx, supra note 12, at 125, 157.

      [33].    Id. at 158.

      [34].    Vinx, supra note 12, at 12.

      [35].    See, e.g., Harold Hongju Koh, The National Security Constitution: Sharing Power After the Iran-Contra Affair 118–19 (1990) (lauding the president as “institutionally best suited to initiate government action,” because its “decision-making processes can take on degrees of speed, secrecy, flexibility, and efficiency that no other governmental institution can match”).

      [36].    David Dyzenhaus, The Long Arc of Legality: Hobbes, Kelsen, Hart 443 (2022).

      [37].    Hans Kelsen, Who Ought to be the Guardian of the Constitution? Kelsen’s Reply to Schmitt, in Vinx, supra note 12, at 174.

      [38].    Id. at 175.

      [39].    Id. at 183.

      [40].    Id.

      [41].    Id. at 184.

      [42].    Id. at 185.

      [43].    Id. at 186.

      [44].    Id. at 194.

      [45].    George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J. Legal Stud. 1, 6–30, 55 (1984).

      [46].    David Dyzenhaus, “Now the Machine Runs Itself”: Carl Schmitt on Hobbes and Kelsen, 16 Cardozo L. Rev. 1, 10–11 (1994).

      [47].    Kelsen, supra note 37, at 203.

      [48].    Id. at 210; see also id. at 2156 (noting that a constitutional court might be elected, making it as democratic as a president).

      [49].    That said, Kelsen is clear that a constitutional court “can only form a part” of the necessary guarantees of a constitutional order. Id. at 212.

      [50].    Id. at 203–04. Kelsen also notes that Schmitt’s argument rests on an “illicit equivocation” by which he simply asserts that the Weimar Constitution has already made the president the constitution’s guardian, so the court simply cannot play this role. Id. at 217. That is, Schmitt takes his conclusion as a premise of his argument.

      [51].    For an account of Schmitt’s reliance on Sieyès, see Jan Muller, Carl Schmitt and the Constitution of Europe, 21 Cardozo L. Rev. 1777, 1783 (2000).

      [52].    Kelsen, supra note 37, at 209; see also McCormick, supra note 19, at 179 (noting the link to Rousseau). As McCormick notes, Schmitt’s “description of the source of the president’s legitimacy . . . increasingly sound as though they were mandated not by the constitutional order but something like a sovereign will that is itself prior to that order.” McCormick, supra note 19, at 176.

      [53].    Kelsen, supra note 37, at 209.

      [54].    Vinx, supra note 12, at 1–2.

      [55].    Id. at 16.

      [56].    Here, I follow McCormick in trying to distinguish what is useful from what is profoundly awry in Schmitt’s thought. McCormick, supra note 19, at 181–82.

      [57].    Vinx, supra note 12, at 19.

      [58].    McCormick, supra note 19, at 177.

      [59].    On Schmitt’s “trashing” of the rule of law, see Bill Scheuerman, The Rule of Law Under Siege: Carl Schmitt and the Death of the Weimar Republic, 14 Hist. of Pol. Thought 256, 258 (1993).

      [60].    Jan-Werner Müller, What Is Populism? 19–20 (2016).

      [61].    Cf. Aziz Z. Huq, The People Against the Constitution, 116 Mich. L. Rev. 1123, 1142 (2018) (“It is a logical consequence of populism to identify groups (whether political, racial, ethnic, or class-based) as the enemy and to attack them on the basis of false accusations.”); see also McCormick, supra note 19, at 185 (arguing that on Schmitt’s view of presidential power, “the constitution is inviting its own disposability”).

      [62].    Or Bassok, The Schmitelsen Court: The Question of Legitimacy, 21 German L.J. 131, 139 (2020).

      [63].    Craig Green, An Intellectual History of Judicial Activism, 58 Emory L.J. 1195, 1200, 1206–16 (2009) (arguing that “the concept of judicial activism . . . has older foundations,” and canvassing prior periods when the judiciary played a controversial role, including the Lochner era, the decades after the Civil War, the Dred Scott decision, and the Marshall Court).

      [64].    Jane S. Schacter, Putting the Politics of “Judicial Activism” in Historical Perspective, 2017 Sup. Ct. Rev. 209, 246–52 (presenting empirical data to show the increasing prevalence of questions regarding judicial activism being asked of nominees during confirmation hearings). On the increasing polarization of the federal judiciary, see Adam Bonica & Maya Sen, Estimating Judicial Ideology, 35 J. Econ. Persp. 97, 97
(2021). For a very recent snapshot, see Aziz Huq, An Ominous Debate
about “Trump Judges,”
Politico (Sep. 9, 2022), news/magazine/2022/09/09/the-ominous-debate-over-trump-judges-00055808.

      [65].    James Andrew Wynn, When Judges and Justices Throw Out Tools: Judicial Activism in Rucho v. Common Cause, 96 N.Y.U. L. Rev. 607, 610–26 (2021) (examining several definitions).

      [66].    Schmitt, Guardian of the Constitution, supra note 20, at 81.

      [67].    Id.

      [68].    Kelsen, supra note 37, at 182.

      [69].    The Federalist No. 78 (Alexander Hamilton).

      [70].    This much is conventional wisdom. Amy L. Stein, Energy Emergencies, 115 Nw. U. L. Rev. 799, 806 (2020) (“[E]mergency powers only come in two flavors: (1) constitutional and (2) statutory.”).

      [71].    I distinguish “legal” from “permissible” because it is possible to imagine a regime of extralegal emergency powers. Lincoln’s suspension of habeas corpus, followed by his appeal to Congress is one such model. See James F. Simon, Lincoln and Chief Justice Taney: Slavery, Secession, and the President’s War Powers 186–87 (2006).

      [72].    I suspect that the common assumption that only precipitously violent emergencies count as such is both supported by, and justifies, this judicial move. That is, the assumption that law has no role in violent emergencies drives the institutional practice of withdrawing law during violent emergencies.

      [73].    The leading account is still Jules Lobel, Emergency Power and the Decline of Liberalism, 98 Yale L.J. 1385 (1989). For a recent account of how statutory emergency powers have developed since Lobel’s work, see Jonathan W. Ellison, Note, Trust the Process? Rethinking Procedural Due Process and the President’s Emergency Powers over the Digital Economy, 71 Duke L.J. 499, 506–13 (2021).

      [74].    Even the “political constitution” of emergency” responses is fashioned though institutional channels “only a small part of which flows from the Constitution’s texts or judicial precedent.” Mark Tushnet, The Political Constitution of Emergency Powers: Some Lessons from Hamdan, 91 Minn. L. Rev. 1451, 1458 (2007).

      [75].    John Ferejohn & Pasquale Pasquino, The Law of the Exception: A Typology of Emergency Powers, 2 Int’l J. Const’l L. 210, 214 (2004).

      [76].    Aziz Z. Huq, The President and the Detainees, 165 U. Pa. L. Rev. 499, 507 (2017) (concluding, based on an empirical study of release patterns that “post-Boumediene federal courts . . . arguably did more to entrench rather than to dissolve the prison doors at Guantánamo”).

      [77].    There are, of course, exceptions to that rule. See, e.g., U.S. Const. amend. XIII § 1 (prohibiting “slavery” and “involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted”).

      [78].    Johnson v. California, 543 U.S. 499, 505 (2005) (quoting Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 227 (1995)); accord Miller v. Johnson, 515 U.S. 900, 920 (1995) (“To satisfy strict scrutiny, the State must demonstrate that [the] legislation is narrowly tailored to achieve a compelling interest.”).

      [79].    See, e.g., Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891, 1915 (2020) (rejecting equal protection challenge to DACA rescission as lacking any support); Trump v. Hawaii, 138 S. Ct. 2392, 2422 (2018) (rejecting claim that President Trump’s travel ban was motivated by antireligious sentiment and citing instead the plausible national security basis for the order). For a forceful, and more general, critique along these lines, see Ian Haney-López, Intentional Blindness, 87 N.Y.U. L. Rev. 1779, 1828 (2012).

      [80].    Trump v. Hawaii, 138 S. Ct. at 2423 (2018) (overruling Korematsu v. United States, 323 U.S. 214 (1944)).

      [81].    Aziz Z. Huq, Article II and Antidiscrimination Norms, 118 Mich. L. Rev. 47, 114 (2019) (“[B]ecause the government has a wide variety of close substitutes that can mimic many (although perhaps not quite all) of the internment’s effects while evading any formal parallelism, it has a broad capacity to circumvent any collision with the law.”).

      [82].    Jenny-Brooke Condon, Equal Protection Exceptionalism, 69 Rutgers Univ. L. Rev. 563, 563 (2017); accord Gabriel J. Chin, Segregation’s Last Stronghold: Race Discrimination and the Constitutional Law of Immigration, 46 UCLA L. Rev. 1, 2 (1998).

      [83].    Richard H. Fallon, Jr., Strict Judicial Scrutiny, 54 UCLA L. Rev. 1267, 1322 (2007); accord Jed Rubenfeld, On the Legal Status of the Proposition That “Life Begins at Conception,” 43 Stan. L. Rev. 599, 604 (1991) (“[T]here are no set formulas defining compelling state interests.”).

      [84].    New York State Rifle & Pistol Ass’n, Inc. v. Bruen, 142 S. Ct. 2111, 2131 (2022) (rejecting balancing tests for the Second Amendment).

      [85].    Holder v. Humanitarian L. Project, 561 U.S. 1, 33 (2010).

      [86].    Janus v. Am. Fed’n of State, Cnty. & Mun. Emps., Council 31, 138 S. Ct. 2448, 2472 (2018).

      [87].    Brown v. Ent. Merchants Ass’n, 564 U.S. 786, 855 (2011) (Breyer, J., dissenting) (noting the absence of deference in the majority ruling); see also Aziz Z. Huq, Preserving Political Speech from Ourselves and Others, 112 Colum. L. Rev. Sidebar 16, 29 (2012) (finding “a striking divergence between the Court’s magnanimous gestures of broad deference to elected actors in the national security domain and its beady-eyed skepticism in the campaign finance context”).

      [88].    United States v. U.S. District Court (Keith), 407 U.S. 297, 313 (1972).

      [89].    See, e.g., Nat’l Treasury Emps. Union v. Von Raab, 489 U.S. 656, 659–60, 677 (1989) (concluding, in a Fourth Amendment case involving drug testing of U.S. Customs Service agents, that “the Government has demonstrated that its compelling interests in safeguarding our borders and the public safety outweigh the privacy expectations of employees”).

      [90].    See, e.g., United States v. Flores-Montano, 541 U.S. 149, 155 (2004) (suspicionless inspections at the border).

      [91].    Grutter v. Bollinger, 539 U.S. 306, 351 (2003) (Thomas, J., concurring).

      [92].    Johnson, 543 U.S. at 540 (2005) (Thomas, J., dissenting).

      [93].    Trump v. Hawaii, 138 S. Ct. 2392, 2424 (2018) (Thomas, J., concurring).

      [94].    See, e.g., Tandon v. Newsom, 141 S. Ct. 1294, 1297 (2021); Roman Cath. Diocese of Brooklyn v. Cuomo, 141 S. Ct. 63, 67 (2020); Harvest Rock Church v. Newsom, 141 S. Ct. 889 (2020); see also Lee Epstein & Eric A. Posner. The Roberts Court and the Transformation of Constitutional Protections for Religion: A Statistical Portrait, 2022 Sup. Ct. Rev. 315, 319–20 (documenting changing patterns in the cases).

      [95].    Republican Nat’l Comm. v. Democratic Nat’l Comm., 140 S. Ct. 1205, 1207 (2020).

      [96].    Henry Redman, Wisconsin’s Closed Polls, Wis. Examiner (Apr. 6, 2020),

      [97].    Elise Viebeck et al., Long Lines, Anger and Fear of Infection: Wisconsin Proceeds with Elections Under Court Order, Wash. Post (Apr. 7, 2020, 6:40 PM),

      [98].    Merrill v. People First of Ala., 141 S. Ct. 190 (2020) (mem.).

      [99].    See, e.g., Clarno v. People Not Politicians Or., 141 S. Ct. 206 (2020) (mem.) (staying an injunction of Oregon’s requirement to gather signatures to place initiatives on the general election ballot); Raysor v. DeSantis, 140 S. Ct. 2600 (2020) (mem.) (staying an injunction of Florida’s law forcing reconstituted felons to pay their fines in order to regain voting rights under the Purcell principle); Tex. Democratic Party v. Abbott, 140 S. Ct. 2015 (2020) (mem.) (denying a request to vacate a stay of an injunction of Texas’s law automatically granting absentee ballots to citizens 65 or older).

    [100].    Formally, they turn on the so-called “Purcell principle,” a judge-made doctrine concerning the timing of election litigation. Purcell v. Gonzalez, 549 U.S. 1, 1–2 (2006).

    [101].    See A Guide to Emergency Powers and Their Use, Brennan Ctr.
for Just. (2019), Emergency%20Powers_Printv2.pdf (identifying 136 statutory emergency powers).

    [102].    Id. (suggesting that 96 statutory emergency powers can be used by the President without any restrictions or constraints).

    [103].    Amy L. Stein, A Statutory National Security President, 70 Fla. L. Rev. 1183, 1193 (2018) (“Almost 400 statutes discuss national security authority provided to the President, as opposed to other agents of the government, and over sixty provide the President with explicit power to act in the name of national security.”).

    [104].    Oren Gross, Chaos and Rules: Should Responses to Violent Crises Always Be Constitutional?, 112 Yale L.J. 1011, 1111 n.426 (2003) (“The upsurge in the number and scope of statutory delegations of powers from Congress to the President—including a wide array of emergency powers—all but made superfluous the need for any claims of extralegal emergency powers.”).

    [105].    For a brilliant recent account of the interconnected crisis in energy, democracy, and geopolitics, see Helen Thompson, Disorder: Hard Times in the 21st Century (2022). It is not hard to read back Thompson’s diagnosis into the squall of financial and public health emergencies of the past five years.

    [106].    Nat’l Fed’n of Indep. Bus. v. Dep’t of Lab., Occupational Safety & Health Admin., 142 S. Ct. 661, 664–65 (2022).

    [107].    Alabama Ass’n of Realtors v. Dep’t of Health & Hum. Servs., 141 S. Ct. 2485, 2486 (2021).

    [108].    On the effects of climate change, and the importance of policy change, see Will Steffen et al., Trajectories of the Earth System in the Anthropocene, 115 Proc. Nat’l Acad. Sci. 8252, 8257 (2018).

    [109].    W. Virginia v. Env’t Prot. Agency, 142 S. Ct. 2587, 2606 (2022) (“EPA informed the Court of Appeals that it does not intend to enforce the Clean Power Plan because it has decided to promulgate a new Section 111(d) rule.”).

    [110].    Id. The majority also noted that the Solicitor General “‘vigorously defend[ed]’ the legality” of new emissions limits. Id.

    [111].    Biden v. Texas, 142 S. Ct. 2528, 2530 (2022).

    [112].    Proclamation No. 9844, 84 Fed. Reg. 4949 (Feb. 15, 2019).

    [113].    Sierra Club v. Trump, 977 F.3d 853, 890 (9th Cir. 2020) (invalidating presidential action in part), cert. granted, judgment vacated sub nom. Biden v. Sierra Club, 142 S. Ct. 56 (2021).

    [114].    Stephen I. Vladeck, The Solicitor General and the Shadow Docket, 133 Harv. L. Rev. 123, 126 (2019).

    [115].    D.C. Code Ann. § 49-409.

    [116].    But see Ctr. for Biological Diversity v. Trump, 453 F. Supp. 3d 11, 30–34 (D.D.C. 2020) (holding that challenge to border emergency declaration presented a nonjusticiable political question); cf. Washington v. Trump, 441 F. Supp. 3d 1101, 1124–25 (W.D. Wash. 2020) (rejecting invocation of political question doctrine).

    [117].    See generally Aziz Z. Huq, The Collapse of Constitutional Remedies (2021).

    [118].    W. Virginia v. Env’t Prot. Agency, 142 S. Ct. 2587 (Feb. 28, 2022).

    [119].    Ziglar v. Abbasi, 37 S. Ct. 1843 (2017); see also Hernández v. Mesa, 140 S. Ct. 735 (2020).

    [120].    Egbert v. Boule, 142 S. Ct. 1793, 1800 (2022).

    [121].    Id.

    [122].    Shalini Bhargava Ray, The Demise of Rights-Protective Statutory Interpretation for Detained Immigrants and the Rise of “Piecemeal” Textualism, SCOTUS Blog (Jun. 14, 2022),

    [123].    Adrian Vermeule, Our Schmittian Administrative Law, 122 Harv. L. Rev. 1095 (2009).

    [124].    Joel Isaac, Constitutional Dictatorship in Twentieth-Century American Political Thought, in States of Exception in American History 225, 237 (G. Gerstle & J. Isaac, eds., 2020) [hereinafter “States of Exception”].

    [125].    Id. at 238–39.

    [126].    David Dyzenhaus, Beyond the Exception, in States of Exception, supra note 124, at 69, 75.

    [127].    Vinx, supra note 12, at 21.

    [128].    Kelsen, supra note 37, at 203.

    [129].    Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2311, 2323 (2006).

    [130].    For a classic statement to this effect, see Robert A. Dahl, Decision-making in a Democracy: The Supreme Court as a National Policy-maker, 6 J. Pub. L. 279, 285 (1957).

    [131].    Stephen Jessee, Neil Malhotra, & Maya Sen, A Decade-Long Longitudinal Survey Shows That the Supreme Court is Now Much More Conservative Than the Public, 119 Proc. Nat’l Acad. Sci. e2120284119 (2022).

    [132].    See supra text accompanying notes 95–99.

    [133].    Kelsen, supra note 37, at 85.