Faculty Bibliography

A system effect arises when the properties of an aggregate differ from the properties of its members, taken one by one. Familiar examples include Condorcet’s paradox and the Prisoners’ Dilemma. Public law is rife with system effects that are more important and less familiar. Although such effects are sometimes recognized in local contexts, they have a common analytic structure and can profitably be analyzed in global terms. The failure to recognize system effects leads to fallacies of division and composition, in which the analyst mistakenly assumes that what is true of the aggregate must also be true of the members, or that what is true of the members must also be true of the aggregate. Examples are (1) the fallacious assumption that if the overall constitutional order is to be democratic, each of its component institutions must be democratic, taken one by one; and (2) the fallacious assumption that if judges are politically biased, courts must issue politically biased rulings. In these cases and many others I will discuss, system effects are an indispensable analytic tool for legal theory. A systemic approach to constitutional theory implies what I will call second-best constitutionalism. Stated abstractly, suppose that at least some of the conditions necessary to produce a given ideal or first-best constitutional order fail to hold. Even if it would be best to achieve full satisfaction of all those conditions, it does not follow that it is best to achieve as many of the conditions as possible, taken one by one. Rather, multiple failures of the ideal can offset one another, producing a closer approximation to the ideal at the level of the overall system. Although the idea is abstract, we will see that problems of second best are chronic in real-world constitutional systems, including our own, because such systems are always partly constrained by technology, economics, and politics.

Tyrannophobia - the fear of dictatorship - is a dominant theme in American political discourse. Yet dictatorship has never existed in the United States or even been likely.The hypothesis that tyrannophobia itself has prevented dictatorship from occurring is implausible; better evidence exists for alternative hypotheses. We conclude that tyrannophobia is an irrational political attitude that has interfered with, and continues to interfere with, needed institutional reform.

This essay compares crisis governance and emergency lawmaking after 9/11 and the financial meltdown of 2008. We argue that the two episodes were broadly similar in outline, but importantly different in detail, and we attempt to explain both the similarities and differences. First, broad political processes and constraints operated in both episodes to create a similar pattern of crisis governance, in which Congress delegated large new powers to the executive. We argue that this pattern is best explained by reference to the account of lawmaking in the administrative state offered by Carl Schmitt, as opposed to the standard Madisonian view. Second, within the broad constraints of crisis politics, the Bush administration asserted its authority more aggressively after 9/11 than in the financial crisis. Rejecting competing explanations based on legal differences, the nature of the threat, or other factors, we attribute the difference to the Bush administration's loss of popularity and credibility over the period between 2001 and 2008 and to the more salient and divisive distributive effects of financial management.

Mechanisms identifies tools of institutional design “writ small” that promote democratic values. Those tools include the veil of uncertainty, submajority voting rules, absolute majority voting rules, rules of delayed and selective transparency, and institutions for promoting legislative deliberation on constitutional questions. In this response, I go beyond the book to consider the complex interaction of democratic mechanisms, both on the demand side (what mechanisms are desirable?) and on the supply side (what mechanisms will political actors have the capacity and incentive to supply?). On the demand side, I consider cases in which mechanisms relate as complements, as substitutes, or in a fashion that is uncertain from the standpoint of the institutional designer. On the supply side, I consider the optimal scope of democratic reforms, the optimal majority for obtaining reform, and problems arising from the general theory of second best.

Chevron, U.S.A. v. Natural Resources Defense Council lays out a two-step process that courts must follow when they review a federal agency's construction of a federal statute. We argue that Chevron, rightly understood, has only one step. The single question is whether the agency's construction is permissible as a matter of statutory interpretation. The two Chevron steps both ask this question, just in different ways, and are thus mutually convertible: any opinion written in terms of one step can be written, without loss of content, in terms of the other step. Chevron's artificial division of a unitary inquiry causes material confusion among commentators and courts, and has no benefits; administrative law should jettison the two-step framework.

In the administrative state, how should expert opinions be aggregated and used? If a panel of experts is unanimous on a question of fact, causation, or prediction, can an administrative agency rationally disagree, and on what grounds? If experts are split into a majority view and a minority view, must the agency follow the majority? Should reviewing courts limit agency discretion to select among the conflicting views of experts, or to depart from expert consensus? I argue that voting by expert panels is likely, on average, to be epistemically superior to the substantive judgment of agency heads, in determining questions of fact, causation, or prediction. Nose counting of expert panels should generally be an acceptable basis for decision under the arbitrary and capricious or substantial evidence tests. Moreover, agencies should be obliged to follow the (super)majority view of an expert panel, even if the agency's own judgment is to the contrary, unless the agency can give an epistemically valid second-order reason for rejecting the panel majority's view.

Our administrative law contains, built right into its structure, a series of legal “black holes” and “grey holes” – domains in which statutes, judicial decisions, and institutional practice either explicitly or implicitly exempt the executive from legal constraints. Legal black holes and grey holes are best understood by drawing upon the thought of Carl Schmitt, in particular his account of the relationship between legality and emergencies. In this sense, American administrative law is Schmittian. Moreover, it is inevitably so. Extending legality to eliminate these black and grey holes is impracticable; any aspiration to eliminate the Schmittian elements of our administrative law is utopian.

Many-minds arguments claim that in some way or another, groups of decisionmakers tend to make better decisions than individuals. This essay identifies five general and recurring problems with such arguments, as follows: (1) Whose minds? The group or population whose minds are at issue is often equivocal or ill-defined. (2) Many minds, worse minds. The number of minds endogenously influences their quality, often for the worse. More minds can be systematically worse than fewer because of selection effects, incentives for epistemic free-riding, and emotional and social influences. (3) Epistemic bottlenecks. The epistemic benefits of many minds are often diluted or eliminated because the structure of institutions funnels decisions through an individual decision-maker, or a small group of decision-makers, who occupy an epistemic bottleneck or chokepoint. (4) Many minds vs. many minds. The institutional comparisons that pervade legal theory are typically many-to-many comparisons rather than one-to-many. (5) Many minds vs. other values. Epistemic considerations systematically trade off against other goods, such as the costs of decision-making and the expression of moral norms. The epistemic quality of the laws is a good to be optimized, not maximized.

Human reason is limited. What are the consequences of this fact for the contested lawmaking claims between courts, legislatures and the executive branch? In light of the limits of reason, how should legal institutions be designed? In Law and the Limits of Reason, Adrian Vermeule criticizes the view that the limits of reason counsel in favor of judicial lawmaking in the style of the common law. He argues that there is no logical connection between the limits of reason, on the one hand, and the superiority of common law or of judge-made constitutional law on the other. The relatively small number of judges on relevant courts, their limited informational base and generalist rather than specialized skills, ensure that judicial reason is itself sharply limited and that the argument to judicial lawmaking from the limits of reason outruns the logical, causal, and evidentiary support. Instead, Adrian Vermeule proposes and defends a "codified constitution" - a regime in which legislatures have the primary authority to develop constitutional law over time, through statutes and constitutional amendments. Precisely because of the limits of human reason, large modern legislatures, with their numerous membership, complex internal structures for processing information and their abundant informational resources, are the most effective lawmaking institutions.

It seems odd that despite the torrent of writing on emergencies and the law after 9/11, no one has systematically examined the view of emergencies held by our greatest judge. Perhaps the problem is that Justice Holmes has so often been subdivided along doctrinal lines. There is the Holmes of free speech law, represented by the majority opinion in Schenck v. United States and by the dissents in Abrams v. United States and Gitlow v. New York. There is the Holmes of property and takings law, represented by the majority opinion in Pennsylvania Coal Co. v. McMahon. There is the Holmes of due process law, represented by the dissents in Lochner v. New York and Tyson & Bro. v. Banton. And no one much talks about the Holmes opinions first upholding and then invalidating emergency rent control, Block v. Hirsh and Chastleton Corp. v. Sinclair, or about the opinion upholding emergency executive detention in Moyer v. Peabody. In what follows, part of my aim is to suggest that what doctrine has put asunder, a focus on emergencies can reunite. Emergencies are a central theme of Holmes's jurisprudence, one that cuts across doctrinal categories and clarifies theoretical puzzles. My central suggestion is that Holmes's judicial and extrajudicial writings, in their best light, implicitly suggest a coherent account of emergencies, law, and constitutional adjudication. I will call this account the epistemic theory of emergencies, with the caveat that I use "theory" not in any rigorous way but just to indicate that Holmes tended to approach questions of emergency powers with a distinctive set of prejudices. We will see that, quite characteristically, Holmes was suggestive but not systematic about his theoretical premises. Despite the ambiguities, however, it is possible to reconstruct a Holmesian account of emergencies that is both plausible and (I hope) theoretically fresh.

A central, organizing motif of Cass Sunstein's work is the effort to spell out the consequences of the New Deal for American law. I suggest that anyone who shares Sunstein's premises can and should go even farther in this direction. The logical consequence of Sunstein's views is a New Deal for all civil liberties and personal liberties. Criminal law and procedure, and cases growing out of the Global War on Terror, should be approached through New Deal lenses. All civil and personal liberties will ultimately have to justify themselves at the bar of cost-benefit analysis. This essay is prepared for a symposium on "The Scholarship of Cass R. Sunstein" to be published by the Tulsa Law Review.

This essay offers case studies of three emergency statutes, all dealing with terrorism and all enacted within less than a year after a major terrorist attack: the September 14, 2001 Authorization to Use Military Force; the USA PATRIOT Act; and the U.K. Terrorism Act 2006. A standard worry about such cases is that the circumstances of emergency lawmaking produce blank-check delegations to the executive. The fog of uncertainty, emotions such as urgency and visceral fear, and the tendency of legislators and the public to rally 'round the flag, all cause legislators to vote the executive massive new powers, regardless of whether those powers are rationally justifiable. This view is descriptively and theoretically flawed. Descriptively, executives in all three episodes lost control of the political dynamics, faced bipartisan resistance or rebellion in the legislature, and ended up obtaining far less than they asked for or desired. Theoretically, emergency conditions have cross-cutting political effects on legislators. The mechanisms and forces operative during emergency lawmaking cut both ways, constraining as well as empowering the executive, with unpredictable net results in particular cases. Although executives usually receive new powers in emergencies, there is no reason to think that they systematically tend to receive more new authority than a rational legislature would provide.

A constitutional showdown is a disagreement between branches of government about their constitutional powers that ends in the total or partial acquiescence by one branch in the views of the other and that creates a constitutional precedent. Standard examples of showdowns include disputes over executive privilege, war-making and -funding, and court-packing. Showdowns are costly because they interfere with the normal operation of government, but they also produce important and overlooked benefits. They are an important mechanism of constitutional development, and, among other things, clarify the lines of constitutional authority. We tote up the costs and benefits of constitutional showdowns, show their continuity with other legal phenomena, and criticize the prevailing wisdom that government agents should avoid them as much as possible.

Positive and normative legal theory often seem to have little to do with one another. I suggest that the disconnect arises from two sources: the gap between fact and value, and the gap between external and internal perspectives on law. I then lay out a repertoire of strategies and mechanisms for connecting positive and normative legal theory. In some cases, positive theory can serve as a direct source of normative arguments. In other cases, positive theory serves as an indirect constraint on normative decisionmaking, thereby narrowing the set of normative arguments that must be considered. Finally, I ask: in light of our best positive theories, to what audiences can normative scholarship coherently be addressed?

In Massachusetts v. Environmental Protection Agency (2007), the Supreme Court held, among other things, that the EPA has statutory authority to regulate greenhouse gases under the Clean Air Act, and that the agency cannot decline to do so on political grounds. We analyze the logic of MA v. EPA and its broader implications for administrative law and regulatory policy. We locate MA v. EPA in the context of the Justices' increasing worries about the politicization of administrative expertise, particularly under the Bush administration. The majority's solution for this worry, we suggest, is a kind of expertise-forcing: the Court attempts to ensure that agencies actually do exercise expert judgment, and that they do so free from outside political pressures, even or especially political pressures emanating from the White House or political appointees in the agencies. Whereas a line of caselaw and commentary stemming from Chevron USA Inc. v. Natural Resources Defense Council sees presidential politics and expertise as complementary, expertise-forcing has its roots in an older vision of administrative law, one in which presidential politics and expertise are fundamentally antagonistic. Because the Court subjects the denial of a rulemaking petition to hard look review, we suggest that MA v. EPA is State Farm for a new generation.

This article considers absolute majority rules, which require the affirmative vote of a majority of all those eligible to vote in the institution. I compare absolute majority rules to simple majority rules under which only those present and voting are counted, and to simple supermajority rules. Under plausible conditions, absolute majority rules prove superior. Absolute majority rules insure majorities against strategic behaviour by minorities and combine supermajoritarian effects with majoritarian symbolism.

In recent years, the central claim of common-law constitutionalism has been that precedent and tradition embody some form of latent wisdom. Judges will generally do best by deferring to the wisdom embodied in precedent and tradition, rather than trusting to their unaided reason. In what follows, I offer a critical analysis of this family of claims. Drawing throughout on Jeremy Bentham's critique of the subconstitutional common law, I conclude that the constitutional common law is not plausibly seen as a repository of latent wisdom, at least not to any greater extent than statutes and other nonjudicial sources of law. The mechanisms advanced by common-law constitutionalists suffer from infirmities of internal logic and from a failure to make the necessary institutional comparisons between and among precedent and tradition, on the one hand, and the outputs of legislatures, executive officials, and constitutional framers on the other.

What institutional arrangements should a well-functioning constitutional democracy have? Most of the relevant literatures in law, political science, political theory, and economics address this question by discussing institutional design writ large. In this book, Adrian Vermeule moves beyond these debates, changing the focus to institutional design writ small. In established constitutional polities, Vermeule argues that law can and should - and to some extent already does - provide mechanisms of democracy: a repertoire of small-scale institutional devices and innovations that can have surprisingly large effects, promoting democratic values of impartial, accountable and deliberative government. Examples include legal rules that promote impartiality by depriving officials of the information they need to act in self-interested ways; voting rules that create the right kind and amount of accountability for political officials and judges; and legislative rules that structure deliberation, in part by adjusting the conditions under which deliberation occurs transparently or instead secretly. Drawing upon a range of social science tools from economics, political science, and other disciplines, Vermeule carefully describes the mechanisms of democracy and indicates the conditions under which they can succeed.

Legal and constitutional theory has focused chiefly on the risk that voters and legislators will trust an ill-motivated executive. This paper addresses the risk that voters and legislators will fail to trust a well-motivated executive. Absent some credible signal of benign motivations, voters will be unable to distinguish good from bad executives and will thus withhold discretion that they would have preferred to grant, making all concerned worse off. We suggest several mechanisms with which a well-motivated executive can credibly signal his type, including independent commissions within the executive branch; bipartisanship in appointments to the executive branch, or more broadly the creation of domestic coalitions of the willing; the related tactic of counter-partisanship, or choosing policies that run against the preferences of the president's own party; commitments to multilateral action in foreign policy; increasing the transparency of the executive's decisionmaking processes; and a regime of strict liability for executive abuses. We explain the conditions under which these mechanisms succeed or fail, with historical examples.

Professor Gary Lawson has written an illuminating response to our book, Terror in the Balance: "Security, Liberty and the Courts" (Oxford University Press 2007). Lawson's main thesis is that the original understanding of the Constitution supports what we call the judicial deference thesis - that courts should defer to the executive and legislative branches during emergencies. In this brief reply, we offer three claims. First, we express skepticism about whether there can be an overlapping consensus between originalists and nonoriginalists about judicial deference in times of crisis. Second, we affirmatively argue that if originalist adjudication is at all justified by reference to its consequences, then judges should be less originalist in emergencies than in normal times. Third, judges should also be less Burkean or traditionalist during emergencies than in normal times.

By "lay Justices" I mean Justices of the Supreme Court of the United States who are not accredited lawyers. Currently the number of lay Justices is zero, although there is no constitutional or statutory rule that requires this. Commentators who urge that the Supreme Court should be diverse on all sorts of margins—methodological diversity, ideological diversity, and racial or ethnic or gender diversity—say little or nothing about professional diversity on the Court. I shall suggest that the optimal number of lay Justices is greater than zero, under specified empirical conditions. I do not know whether those conditions actually hold, but on the other hand no one knows that they do not. It is very plausible that the conditions do hold, in which case the status quo of zero lay Justices is an implausible extreme. In the strong form of the argument, it would be a good idea (whether or not it is a politically feasible one) to appoint a historian, economist, doctor, accountant, soldier, or some other nonlawyer professional to the Court. In a weaker form of the argument, I also suggest that at a minimum, we should appoint more dual-competent Justices—lawyers who also have a degree or some other real expertise in another body of knowledge or skill...

In Terror in the Balance, Posner and Vermeule take on civil libertarians of both the left and the right, arguing that the government should be given wide latitude to adjust policy and liberties in the times of emergency. They emphasize the virtues of unilateral executive actions and argue for making extensive powers available to the executive as warranted. The judiciary should neither second-guess security policy nor interfere on constitutional grounds. In order to protect citizens, government can and should use any legal instrument that is warranted under ordinary cost-benefit analysis. The value gained from the increase in security will exceed the losses from the decrease in liberty. At a time when the 'struggle against violent extremism' dominates the United States' agenda, this important and controversial work will spark discussion in the classroom and intellectual press alike.

In Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., the Supreme Court created a new framework for judicial deference to agency interpretations of law: courts should defer to an agency interpretation unless the relevant statute is clear or the agency interpretation is unreasonable. In the past two decades, however, the doctrinal Chevron framework has come under increasing strain. We suggest an alternative, which is to cast Chevron as a judicial voting rule, thereby institutionalizing deference to administrative agencies. Our thesis is that a voting rule of this sort would capture the benefits of the doctrinal version of Chevron while generating fewer costs. The principal advantage of institutionalizing Chevron as a voting rule is that it makes agency deference an aggregate property that arises from a set of votes, rather than an internal component of the decision rules used by individual judges. A voting-rule version of Chevron would also allow more precise calibration of the level of judicial deference over time, and holding the level of deference constant, a voting rule would produce less variance in deference across courts and over time, yielding a lower level of legal uncertainty than does the doctrinal version of Chevron. We consider and respond to various objections.

My basic suggestion, in Part I, is that there is no such thing as “instrumentalism.” There is only a variety of instrumentalisms, offered in different theoretical contexts for different purposes. The merits of these different instrumentalisms must be evaluated locally rather than globally. Furthermore – this is a separate point, but a complementary one – there are several antonyms for legal instrumentalism that are materially different. It is no more coherent to praise all of them, just because they are not instrumentalism, than it would be to praise all of anarchism, fascism, and communism because they are alternatives to liberal democracy. Subsequently, in Part II, I ask what prescriptions for the legal system follow from a critique of legal instrumentalism. I suggest that in a legal culture pervaded by instrumentalism (in all of its possible senses), there are powerful discursive pressures to justify an anti-instrumental view by reference to the beneficial effects that holding such a view will produce – by reference, that is, to the instrumental benefits of anti-instrumentalism. When combined with the claim that anti-instrumentalism requires certain beliefs, not merely certain actions, this is an intrinsically paradoxical stance; it leads, perhaps unavoidably, to a type of esoteric legalism, under which the theorist is quite willing to promote a false belief in the truth of anti-instrumentalism in order to secure the benefits of that belief. Unfortunately, however, there are well-known paradoxes of esotericism that make views of this sort self-defeating. In the Conclusion, I suggest that despite the theoretical puzzles underlying LME, it possesses a thematic and emotional unity as a kind of legal dystopia. As such, its contributions should be assessed by literary as well as theoretical criteria.

Critics of emergency measures such as the U.S. government's response to 9/11 invoke the Carolene Products framework, which directs courts to apply strict scrutiny to laws and executive actions that target political or ethnic minorities. The critics suggest that such laws and actions are usually the product of democratic failure, and are especially likely to be so during emergencies. However, the application of the Carolene Products framework to emergencies is questionable. Democratic failure is no more likely during emergencies than during normal times, and courts are in a worse position to correct democratic failures during emergencies than during normal times. The related arguments that during emergencies courts should protect aliens, and should be more skeptical of unilateral executive actions than of actions that are authorized by statutes, are also of doubtful validity.

This essay describes and explains the political constraints that limit structural reform of the Supreme Court, using the failure of Roosevelt's 1937 court-packing plan as a running example. The thesis is that movements for structural reform of the Court have a self-negating tendency. The very conditions that produce demand for structural reform of the Court also tend to produce counterforces that block reform.

Most academics who have written on coercive interrogation believe that its use is justified in extreme or catastrophic scenarios but that nonetheless it should be illegal. They argue that formal illegality will not prevent justified use of coercive interrogation because government agents will be willing to risk criminal liability and are likely to be pardoned, acquitted, or otherwise forgiven if their behavior is morally justified. This outlaw and forgive approach to coercive interrogation is supposed to prevent coercive interrogation from being applied in inappropriate settings, to be symbolically important, and nonetheless to permit justified coercive interrogation. We argue that the outlaw and forgive approach rests on questionable premises. If coercive interrogation is ever justified, and the benefits outweigh the risks of error and unintended consequences, it should be legal, albeit strictly regulated. The standard institutional justifications for outlaw and forgive - rules/standards problems, slippery slopes, and symbolism - are unpersuasive.

How should judges, in America and elsewhere, interpret statutes and the Constitution? Previous work on these fundamental questions has typically started from abstract views about the nature of democracy or constitutionalism, or the nature of legal language, or the essence of the rule of law. From these conceptual premises, theorists typically deduce an ambitious role for judges, particularly in striking down statutes on constitutional grounds. In this book, Adrian Vermeule breaks new ground by rejecting both the conceptual approach and the judge-centered conclusions of older theorists. Vermeule shows that any approach to legal interpretation rests on institutional and empirical premises about the capacities of judges and the systemic effects of their rulings. Drawing upon a range of social science tools from political science, economics, decision theory, and other disciplines, he argues that legal interpretation is above all an exercise in decisionmaking under severe empirical uncertainty. In view of their limited information and competence, judges should adopt a restrictive, unambitious set of tools for interpreting statutory and constitutional provisions, deferring to administrative agencies where statutes are unclear and deferring to legislatures where constitutional language is unclear or states general aspirations.

Replying to Matthew C. Stephenson, Legislative Allocation of Delegated Power: Uncertainty, Risk and the Choice Between Agencies and Courts, 119 Harv. L. Rev. 1035 (2006). Matthew Stephenson models "the decision calculus faced by a rational, risk-averse legislator who must choose between delegation to an agency and delegation to a court." On the assumption that delegation to agencies tends to produce interpretive consistency across issues while delegation to courts tends to produce interpretive consistency over time, a risk-averse legislator interested in reducing variance along either dimension will face a tradeoff between intertemporal risk diversification and interissue risk diversification. From this basic tradeoff, Stephenson derives comparative statics about the rational, risk-averse legislator's choice of delegates under various conditions. I suggest that Stephenson's legislative-delegation model rests on excessively artificial assumptions and is unable to yield significant predictions - in either the political or statistical sense. In particular, it is unsatisfying to model legislators as entering a "policy lottery" by enacting ambiguous delegating statutes, but then also to picture them as clearly specifying the identity of the delegate. The same institutional and political factors that tend to produce a first-order policy lottery over statutory substance also tend to produce a second-order "delegation lottery" over the question whether agencies or courts have ultimate interpretive authority. Moreover, the factors the model includes are, at best, second-decimal considerations relative to the factors it excludes.

Many people believe that the death penalty should be abolished even if, as recent evidence seems to suggest, it has a significant deterrent effect. But if such an effect can be established, capital punishment requires a life-life tradeoff, and a serious commitment to the sanctity of human life may well compel, rather than forbid, that form of punishment. The familiar problems with capital punishment--potential error, irreversibility, arbitrariness, and racial skew--do not require abolition because the realm of homicide suffers from those same problems in even more acute form. Moral objections to the death penalty frequently depend on a sharp distinction between acts and omissions, but that distinction is misleading in this context because government is a special kind of moral agent. The widespread failure to appreciate the life-life tradeoffs potentially involved in capital punishment may depend in part on cognitive processes that fail to treat “statistical lives” with the seriousness that they deserve. The objection to the act/omission distinction, as applied to government, has implications for many questions in civil and criminal law.

A fully specified voting rule must state both a multiplier (for example, a majority or a supermajority) and a multiplicand (for example, a majority of the votes cast or a majority of all members eligible to vote in the institution). The theory of voting rules developed in law, political science, and economics typically compares simple majority rule with alternatives such as supermajority rule. This sort of comparison picks up variation in the multiplier alone. In this paper, by contrast, I consider variation in the multiplicand. The focus is on absolute voting rules, whose multiplicand is all members eligible to vote in the institution. I compare absolute voting rules to voting rules that use a standard multiplicand, under which only those present and voting are counted. The thesis is that under a range of circumstances, absolute voting rules prove normatively superior. Absolute voting rules can insure majorities against strategic behavior by minorities, combine supermajoritarian effects with majoritarian symbolism, and liberate voters from accountability when it is socially desirable to do so.

The standard consequentialist analysis of constitutional law focuses on the incentives that shape the behavior of government officials and other constitutional actors. Incentive-based accounts justify elections as a means of constraining officials to promote the public welfare, or at least the welfare of the median voter; justify the separation of powers as a means of making "ambition counteract ambition"; justify negative liberties, such as free speech and free association, as a necessary corrective to incumbent officials' incentives to suppress political opposition; and so forth. In this experimental Essay, I offer a preliminary sketch of a different way of looking at constitutional law generally and constitutional structure in particular: through the lens of "selection effects." Constitutional rules, on this account, should focus not only on the creation of optimal incentives for those who happen to occupy official posts at any given time, but also on the question which (potential) officials are selected to occupy those posts over time. Where an incentive analysis is short-term and static, asking only how legal rules affect the behavior of a given set of officeholders, selection analysis is long-term and dynamic, asking how legal rules themselves produce feedback effects that, over time, bring new types of government officials into power. This turn to selection-based analysis yields fresh insight into the dynamics of constitutionalism. Because constitutional rules affect the pool of potential and actual officeholders, as well as the behavior of current officeholders, focusing on selection effects shows that some constitutional rules prove "self-stabilizing": the rules tend to select a corps of officeholders who will act to uphold and stabilize the rules themselves. Other constitutional rules, by contrast, prove "self-negating": the rules tend to select a corps of officeholders who work to undermine or destabilize the rules themselves. This framework supplies insights into diverse areas of constitutional law and theory, ranging from governmental structure, campaign finance, and voting rights to criminal sentencing, free speech, and affirmative action.

A federal statute prohibits marine terminal operators from discriminating against terminal users. The Federal Maritime Commission is authorized to enforce the law. A company claimed that the South Carolina State Ports Authority had discriminatorily refused berthing space to a cruise ship, and brought an adjudicative complaint before the commission. The commission rejected the Port Authority's claim that state sovereign immunity extends to proceedings before federal administrative agencies. The Supreme Court held that state sovereign immunity bars a federal agency from adjudicating a private party's complaint against a nonconsenting state. The text of the Eleventh Amendment restricts only the “judicial Power” of the United States, while federal administrative agencies exercise executive power. But the majority opinion candidly acknowledges that the Court's previous decisions have rendered the amendment's text essentially irrelevant. The Court currently takes state sovereign immunity to be a background structural principle that applies to federal‐law claims brought by private parties against their own states or other states, in either state court or federal court. Given those textually ungrounded decisions, the further extension of state immunity to administrative proceedings was a predictable step. The most impressive objection to the Court's decision is not textual, but structural. State immunity does not extend to suits against states brought by the United States, rather than a private party; in such cases, the Court had previously reasoned, the suit against an unconsenting state requires an exercise of political responsibility by elected officials. The same is true of proceedings before federal administrative agencies, whose officers are appointed by the president, and who are subject to congressional oversight.

Legal and political theory have paid a great deal of attention to supermajority rules, which require a fraction of votes greater than 1/2+1 to reach a decision, and thus empower a minority to block change. In this paper I consider the opposite deviation from simple majority rule: submajority rules, under which a voting minority is granted the affirmative power to change the status quo. Among the examples I will consider are: - The Journal Clause, which allows 1/5 of the legislators present in either House to force a roll-call vote; - The discharge rule in the House, which (at various points, although not today) has permitted a specified minority of legislators to force bills out of committee for consideration on the floor; - Senate Rule XXII, under which a cloture petition is valid when signed by sixteen Senators; - The Seven Member Rule, under which a minority of designated committees in the House and Senate can require the executive branch to divulge information; - House Rule XI, which entitles committee minorities to call witnesses at hearings; - The famous Rule of Four that allows four Justices to grant a writ of certiorari and thereby put a case on the Supreme Court's agenda; - Rules governing direct democracy that permit a defined minority of a state's electorate to place a question on the ballot, or to force a recall election; - Rules governing international organizations, which frequently allow a defined minority to call an emergency session or to force a roll-call vote. Submajority rules are rarely discussed, either because they are assumed not to exist, or because they are assumed to lack any institutional virtues, or because submajoritarian decisions are assumed to be chronically unstable in light of the risk that subsequent majorities will reverse the submajority's decision. I will dispute all three assumptions. Submajority rules have important procedural and deliberative virtues: in a range of situations they enable a minority to force public accountability upon a majority, to the benefit of the institution as a whole. The reversibility problem can be, and is, dampened by other institutional rules and norms that protect submajoritarian decisions, or by the simpler expedient of adopting submajority rules only for decisions that are inherently irreversible or costly to reverse, such as decisions that release information into the public domain.

Symposium: Free Speech in War Time Conference. In a standard analysis, the history of civil liberties is characterized by a series of security panics. A range of mechanisms - cognitive heuristics and biases, various forms of cascading and herding, conformity and preference falsification, and so on - cause periodic panics in which aroused publics demand repressive measures to curtail the civil liberties of perceived enemies of the nation, particularly noncitizens or other outsiders. Government officials may themselves panic, or will at least supply the panicky measures that constituents demand. The standard remedy is to urge changes to legal doctrine or institutions, in order to curtail government's power to repress civil liberties in response to security panics. The standard model of security panics has been criticized on several grounds. Sometimes security panics are justified, even if produced by disreputable mechanisms; fear can motivate beneficial action as well as detrimental action. In any event, legal doctrines, and perhaps even institutional design, will prove incapable of constraining a genuinely panicked public. In what follows I will sketch a different criticism of the standard model. Even if that model is right as far as it goes, it is fatally incomplete. My central claim is that the mechanisms underlying security panics have no necessary or inherent pro-security valence. The very same mechanisms are equally capable of producing libertarian panics: episodes in which aroused publics become irrationally convinced that justified security measures represent unjustified attempts to curtail civil liberties. I will suggest that libertarian panics have been a regular occurrence in American history, and that we may be living through one now, in the form of a widespread and thoroughly irrational, even hysterical, reaction to small legal changes adopted after 9/11. Indeed, the tendency to diagnose the existence of a security panic can itself be symptomatic of a libertarian panic. The existence of libertarian panics undermines the institutional reforms urged by the advocates of the standard model. The very reforms that would minimize the risks and harms of security panics will maximize the risks and harms of libertarian panics. The institutional-design problem, then, is to optimize in light of these offsetting risks; whatever legal and institutional arrangements turn out to be optimal, they will necessarily prove less protective of civil liberties than the arrangements favored by advocates of the standard model.