2 Optimizing Constitutionalism: The Mature Position The last chapter defined and illustrated precautionary constitutionalism, attempting to put that approach in its best light. This chapter turns to criticisms of constitutional precautions offered by early proponents of national power such as Hamilton, Marshall, and Story, and by New Dealers such as Frankfurter and Jackson. These critics argued that precautionary constitutionalism might prove futile for lack of incentive-compatibility; might jeopardize other values, resulting in greater risks overall; might prove perversely self-defeating, if and because the precautions create or exacerbate the very risks they were intended to prevent; and might prove unnecessary, given the availability of after-the-fact remedies against materialized risks. As we will see, these criticisms of precautionary constitutionalism parallel the criticisms of precautionary principles in the theory of regulation. Critics of precautionary regulation are following a path that the critics of precautionary constitutionalism have already traveled. I argue that the modern theory of risk regulation has arrived at a conclusion that is structurally equivalent to Hirschman's mature position.1 On this view, the goal of regulators should simply be to take optimal precautions, according to a calculus that weighs all relevant risks of action and inaction. As it turns out, however, constitutional theorists such as Publius and Story already endorsed the mature position, arguing in their own language for a resolutely optimizing approach to constitutionalism. What's new is, in this case at least, very old. Once the historical and interpretive work is complete, I offer an argument that the mature position is the best approach to constitution-making. I identify a strictly negative but nonetheless valuable function of the mature 1 Albert O. Hirschman, The Rhetoric of Reaction 153-54 (1991). 5^ Optimizing Constitutionalism: The Mature Position 5 3 position, which improves the processes of constitutional design and interpretation by laundering out one-sided arguments and thereby placing all relevant risks before decision makers. The mature position is sometimes criticized on second-order or indirectly consequentialist grounds, the argument being that even if the mature position is the ideal decision procedure, precautionary principles are necessary to compensate for predictable biases or distortions in regulatory decision making. I trace a parallel argument through the constitutional debates and also outline a rejoinder, applicable to both ordinary risk regulation and the regulation of political risks more specifically. Because relevant biases and distortions appear on all sides of the issues, the second-order argument for precautions, based on the capacities of decision makers, neglects tradeoffs and can prove self-defeating in precisely the same fashion as first-order arguments for precaution. In some cases, localized precautions may fall out naturally from a mature, even-handed assessment of all relevant political risks, but the set of constitutional rules will not be generally and systematically precautionary. Even if particular rules end up being precautionary, a mature assessment should govern at the highest level of analysis. Let me caution again that this is a study in the theory of constitutional argumentation, not a treatise on constitutional design or interpretation. I aim to critique precautionary arguments for constitutional rules, not to evaluate the ultimate merits of those rules. Where particular precautionary arguments fail, the rules they were offered to justify may or may not be independently justifiable under the mature position. If we get the analysis right, the set of constitutional rules that will result is not my concern here. COUNTERING PRECAUTIONARY ARGUMENTS If constitutional law, history, and theory provide a wide array of precautionary justifications for legal rules, they also provide a repertoire of counterarguments. In various eras, framers, judges, and other actors have attempted to undermine the arguments for precaution. At the stage of constitutional design, such actors offer arguments to block the formulation of precautionary constraints. At the stage of constitutional interpretation, the point of the arguments is to prevent narrow construction of powers already granted. Although anti-precautionary arguments appear in many eras and in many diverse contexts, I suggest they fall into recurring structural patterns. Adapting a set of categories from Albert Hirschman's analysis of political 54 Part I. Theory rhetoric,2 supplemented by the modern theory of risk regulation, I will sort the arguments into four groups: (1) Futility arguments, in which the opponent argues that a given precautionary principle will fail to attain its ends. (2) Jeopardy arguments, in which the opponent argues that a given precaution will produce net costs in light of countervailing risks on other margins. (3) Perversity arguments, in which the opponent argues that a given precaution will prove self-defeating because of countervailing risks on the same margin - in other words, because it actually exacerbates the very risk that the precaution attempts to prevent. (4) Arguments for ex post remedies, in which the opponent acknowledges the risk but argues that the correct mechanism to address it is not a general ex ante precaution. Instead, the correct mechanism is an ex post remedy applied case-by-case, after the relevant risk has actually materialized. Strictly speaking, this can be described as a special case of jeopardy, but it is an important special case that warrants separate treatment. FUTILITY: "PARCHMENT" PRECAUTIONS AND COMMITMENT PROBLEMS One rejoinder to a proposal for constitutional precautions is that the proposal may fail the criterion of incentive-compatibility.3 The very conditions that make the precaution necessary also ensure that the actors against whom precautions are taken will have incentives to undermine or ignore it, and no other actors will have incentives to enforce the precaution against its violators.4 Where this is so, the benefits of the precaution are zero and the proposal is futile. Checks and balances; standing armies. The locus classicus of futility argument in constitutional theory is Madison's reference to "parchment 2 Id. at 7. 3 For summaries of and citations to the relevant literatures in political economy, see Daryl J. Levinson,Parchment and Politics: The Positive Puzzle of Constitutional Commitment, 124 Harv. L. Rev. 657, 670-80 (2011); Mark Tushnet, Taking the Constitution Away from the Courts 95-128 (2000). A particularly clear treatment is Daron Acemoglu, Why Not a Political Coase Theorem? Social Conflict, Commitment, and Politics, 31 J. Comp. Econ. 620, 639-48 (2003). 4 See Adrian Vermeule, Judging Under Uncertainty: An Institutional Theory of Legal Interpretation 118-52 (2006). Optimizing Constitutionalism: The Mature Position 55 barriers" in Federalist 48. The general line of his argument, which extends over Federalist 47, 48, and 51, is that formal specification of electoral accountability and separated powers in the constitution will be inadequate to contain legislative tyranny unless checks-and-balances mechanisms are added as "auxiliary precautions" and unless those mechanisms are made incentive-compatible by tying "[t]he interest of the man" to "the constitutional rights of the place [i.e., the institution]," so that "ambition [can] be made to counteract ambition."5 Yet the claim that parchment precautions are futile first appears in the Federalist Papers in connection with the debate over standing armies, not checks-and-balances, and was made by Hamilton, not Madison. Internal insurrections within Pennsylvania and Massachusetts, Hamilton observed, had compelled both states to raise and maintain standing forces, the general lesson being "how unequal parchment provisions are to a struggle with public necessity."6 Hamilton's futility critique of precautionary prohibitions on standing armies in peacetime thus took the form of an argument in the alternative. One possibility would be that the exception for war or insurrection would be interpreted, in operation, so as to nullify the prohibition, because "the national government, to provide against apprehended danger, might in the first instance raise troops, and might afterwards keep them on foot as long as they supposed the peace or safety of the community was in any degree of jeopardy.... [A] discretion so latitudinary as this would afford ample room for eluding the force of the provision."7 If, alternatively, the prohibition were seriously thought to prevent even the raising of armies in time of peace, it would simply be ignored or violated when apparent risks of invasion or insurrection so required. Whether evaded by interpretation or violated outright, the prohibition would prove futile. None of this is to say that Hamilton's argument was in fact correct. But as noted earlier, the federal constitution did not ultimately embody the stringent precautions against standing armies that the Antifederalists desired. Even without such precautions, standing armies did not become a regular feature of the federal establishment until after the Civil War, suggesting at a minimum that the Antifederalists' preferred precautions were unnecessary. As for the broader issues surrounding the checks and balances of the federal lawmaking system, I take them up at length in Chapter 3. James Bryce, as we will see, argued that both checks and balances, and also The Federalist No. 51, at 290, 322 (James Madison) (Clinton Rossiter ed., 1961). The Federalist No. 25, at 167 (Alexander Hamilton) (Clinton Rossiter ed., 1961). Id. at 165. 56 Part I. Theory the extended federalist republic, amounted to self-defeating constitutional precautions. Free speech. In modern terms, the "parchment barriers" argument offered by Hamilton and then Madison points to the difficulties of commitment. As there is no agent external to society who can enforce the terms of constitutional commitments,8 some indirect incentive-compatible mechanism must be called into play to make such commitments stick, and the existence of such a mechanism cannot be assumed.9 The commitment problem only partially overlaps the issue of constitutional precautions: not all precautions suffer from commitment problems, while, conversely, those problems may also beset constitutional rules and structures not justified in precautionary terms. An example of the area of overlap is the precautionary "pathological perspective" on free speech doctrine, under which "the overriding objective at all times should be to equip the first amendment to do maximum service in those historical periods when intolerance of unorthodox ideas is most prevalent."10 The pathological perspective underwrites doctrines that attempt to create an intertemporal commitment: judges will commit themselves, or their successors, to clear speech-protective rules that will provide a bulwark against majoritarian oppression or other political pathologies. As we have seen, the stock example is the rule of Brandenburg v. Ohio,11 under which government may not ban speech to preserve public order unless there has been advocacy that is intended to produce and likely to produce an imminent violation of the law. However, the pathological perspective by itself provides no mechanism to make doctrinal restrictions of this sort stick when pressing exigencies arise. The predictable result has been that when an impressionistic judicial calculus of expected harms shows the existence of a grave threat to public order that the Brandenburg test would disable government from addressing, it is the test that has given way.12 The main objection to the Q See Jon Elster, Ulysses Unbound: Studies in Rationality, Precommitment, and Constraints 88-174 (2000); Acemoglu, supra note, at 622-23. 9 See Levinson, supra note, at 663. 10 Vincent Blasi, The Pathological Perspective and the First Amendment, 85 Colum. L. Rev. 449,449 (1985)-395 U.S. 444 (1969). 12 See Eric A. Posner & Adrian Vermeule, Terror in the Balance: Security, Liberty, and the Courts 232-34 (2007); Frederick Schauer, A Comment on the Structure of Rights, 27 Ga. L. Rev. 415,417 (1993) (arguing that the pre-Brandenburg test of "clear and present danger" implicitly persists, in modified form, in contexts such as national Optimizing Constitutionalism: The Mature Position 57 "pathological perspective" argument for a precautionary approach to free speech law is simply that precautions will systematically tend to prove futile when they would prevent government from taking action against apparently dangerous threats. Clear statement requirements and emergency powers. Similar problems afflict the idea that presidential action in emergencies can be constrained by a doctrinal requirement of clear statutory authorization. The judges who are charged with enforcing the requirement may themselves be swept away by the sense of crisis, and when they are, they will tend to find the requirement satisfied so long as there is any plausibly relevant statute in the picture. In the words of political scientists Terry Moe and William Howell: The Court can issue rulings favorable to presidents, but justify its decisions by appearing to give due deference to the legislature____Congress's collective action problems, combined with the zillions of statutes already on the books, make it entirely unclear what the institution's "will" is - and this gives the Court tremendous scope for arguing that, almost whatever presidents are doing, it is consistent with the "will of Congress." The upshot is that during conditions of perceived war, crisis or emergency, the judges' track record has been extremely forgiving; judges have frequently found "clear" authorization under statutes whose terms were hardly pellucid.14 The judges do this in part because they have powerful incentives to do so at the time; when a perceived crisis occurs, judges who are aware of the limits of their own knowledge rationally fear the security consequences of holding that an executive measure lacks statutory authorization, and thus are powerfully tempted to read statutes for all they are worth, rather than enforcing a clear-statement requirement. Even if such a requirement is announced in prior legal doctrine, there is little incentive on the part of later judges to enforce it, and the requirement will prove incentive-incompatible. As with other precautionary measures, clear-statement restrictions on presidential emergency powers and war powers suffer from severe commitment problems. security, and allows speech rights to be "outweighed" even where they apply); see also, e.g., United States v. Progressive, Inc., 467 F. Supp. 990 (1979) (granting an injunction against a magazine article that provided instructions for building a hydrogen bomb, with no plausible showing of intent to produce imminent harm). 13 Terry M. Moe & William G. Howell, The Presidential Power of Unilateral Action, 15 J.L. Econ. & Org. 132,152 (1999). 14 Posner & Vermeule, supra note, at 48. 58 Part I. Theory JEOPARDY: OTHER-RISK TRADEOFFS In many settings, the most forceful argument against precautions is simply that the optimal level of the target risk is not zero, and that some degree of expected harm from the target risk is necessary to obtain other goods. This jeopardy response invokes other-risk tradeoffs. Faced with a precautionary argument aimed at preventing a target risk, the opponent points to a distinct countervailing risk whose expected costs will be increased by the precaution.15 If the opponent instead argues that the precaution will exacerbate the target risk itself, the appeal is to a same-risk tradeoff, and the response is one of perversity. To illustrate the distinction, Hamilton (as Publius) deployed both types of argument against a precautionary rule that would prohibit the raising of standing armies in times of peace. Under such a prohibition, [a] 11 that kind of policy by which nations anticipate distant danger and meet the gathering storm must be abstained from, as contrary to the genuine maxims of a free government. We must expose our property and liberty to the mercy of foreign invaders, and invite them by our weakness to seize the naked and defenseless prey, because we are afraid that rulers, created by our choice, dependent on our will, might endanger that liberty by an abuse of the means necessary to its preservation.16 Here, Hamilton argues both that a prohibition against standing armies in peacetime, justified as a precaution to protect liberty, would create a countervailing risk on a different margin - the seizure of property by foreign invaders - and that it would perversely create a risk to liberty itself, because a foreign invasion would destroy liberty as surely as would domestic despotism. National governmental powers. If there is a main line of argument to the Federalist Papers, jeopardy is it. Although Publius employs memorable futility arguments and perversity arguments as well - we have seen some of the former and will soon see some of the latter - the overall structure of the Federalist Papers frames a large-scale jeopardy argument: the status quo under the Articles of Confederation poses intolerable For the terms "target risk" and "countervailing risk," see, for example, Jonathan B. Wiener, Precaution in a Multirisk World, in Human and Ecological Risk Assessment: Theory and Practice 1509,1520 (Dennis J. Paustenbach ed., 2002). The Federalist No. 25, at 165-66 (Alexander Hamilton) (Clinton Rossiter ed., 1961) (emphasis added). Optimizing Constitutionalism: The Mature Position 59 countervailing risks on multiple dimensions apart from liberty, such as public order, strong national defense, and the security of property. Liberty-protecting precautions against the power of the Union will hamper the strong national government needed to protect against those risks; therefore the ratifiers should be willing to trade off some risks to liberty against other goods. In Federalist 41, Publius offered his most general rebuttal to the general Antifederalist argument for precautions against abuse of power: It cannot have escaped those who have attended with candor to the arguments employed against the extensive powers of the government that the authors of them have very little considered how far these powers were necessary means of attaining a necessary end. They have chosen rather to dwell on the inconveniences which must be unavoidably blended with all political advantages; and on the possible abuses which must be incident to every power or trust of which a beneficial use can be made____[C]ool and candid people will at once reflect, that the purest of human blessings must have a portion of alloy in them; that the choice must always be made, if not of the lesser evil, at least of the greater, not the perfect, good; and that in every political institution, a power to advance the public happiness involves a discretion which may be misapplied and abused.17 Quite remarkably, given the contemporary political context, the passage blandly observes that some abuse of power is the inevitable byproduct of cost-justified grants of governmental discretion; the optimal level of political abuse is therefore greater than zero. Recess appointments. We have seen that the Canning decision, from the federal court of appeals for the District of Columbia Circuit, articulated a precautionary rule against presidential power to make intrasession recess appointments. It can be argued that the decision was myopic, in the sense that the Canning court focused selectively, even to the point of obsession, on a particular target risk, while ignoring countervailing risks, including risks generated by the precautions themselves. The Federalist No. 41, at 255-56 (James Madison) (Clinton Rossiter ed., 1961). For a version of the argument that emphasizes the paranoid or otherwise nonrational cognition of those who worry in general terms about governmental "abuse," see 1 Joseph Story, Commentaries on the Constitution of the United States 408 (Boston, Hilliard, Gray & Co. 1833): A power, given in general terms, is not to be restricted to particular cases, merely because it may be susceptible of abuse, and, if abused, may lead to mischievous consequences. This argument is often used in public debate; and in its common aspect addresses itself so much to popular fears and prejudices, that it insensibly acquires a weight in the public mind, to which it is no wise entitled (emphasis added). 6o Part I. Theory One countervailing risk is that a narrow interpretation of the recess appointment power interferes with the major purpose of recess appointments, the purpose that caused the framers to insert the power in the first place: to "keep important offices filled and the government functioning," as another circuit court put it.18 That purpose has the same constitutional status as the concern the Canning court focused on exclusively, the constitutional provision for a senatorial check on appointments. If Congress as a whole has used its undoubted constitutional powers to create an office and mandates that it be filled; the president has tried to fill it; yet the tug-of-war over appointments within the Senate keeps the office empty for a protracted period, the result is a problem of constitutional stature, not just a policy problem. All provisions of the Constitution, and indeed the document's very existence, implicitly presuppose that a functioning government is a worthy aim of constitutional interpretation. Furthermore, the only reason the recess appointments issue even arises, as a practical matter, is the toxic interaction between appointments and the Senate practice of the filibuster, which requires sixty votes for approval of relevant business, including appointments. If the majority in the Senate -at present, a Democratic majority - could just approve regular appointments under majority rule, no problem would occur. So the Canning court's narrow interpretation of the recess appointments power indirectly promotes the power of a blocking minority in the Senate. Madison assumed in Federalist 10 that the risk of oppression by entrenched minorities was low, because "the republican principle ... enables the majority to defeat [a minority faction's] sinister views by regular vote."19 But if the principle of majority rule is disabled, as it is by the filibuster, then the risk of presidential aggrandizement has to be weighed against the risk of minoritarian factional oppression. (We have learned since Madison's time that it can be just as oppressive to prevent government from operating as to hijack its positive operation for factional ends.) A very clear and narrow interpretation of the recess appointment power minimizes the aggrandizement risk, but also increases the risk and harms of factionalism in the Senate. Precautions on one margin can themselves create new risks on other margins. Reasonable doubt rule. For an example conventionally treated under the rubric of "rights," consider the justification of the reasonable doubt rule as a precautionary principle against a certain type of political 18 Evans v. Stephens, 387 F.3CI 1220,1224 (CA 11 2004). 19 The Federalist No. 10 (Clinton Rossiter ed. 1961). Optimizing Constitutionalism: The Mature Position 61 risk - erroneous conviction of the innocent. Such justifications usually overlook the countervailing risks of false negatives, when those actually guilty are erroneously set free. If those actually guilty of crime can, as a class, be expected to create more far dangerous risks than the average citizen, then a precautionary perspective might even be invoked in favor of the opposite rule - better to convict ten innocents rather than let one guilty man go free. Thus it has been argued, tongue in cheek, that because "[violence is a problem of public risk and public health [, i]n this context, the precautionary principle would favor earlier and more stringent intervention to prevent the 'future dangerousness' of persons who may, with considerable uncertainty, be forecast to commit violence in the future." This exemplifies a jeopardy argument because the risk against which the reasonable doubt rule takes precautions - unjustified deprivation of liberty through the criminal law - is not the same as the countervailing risk to which the critic of the reasonable doubt rule points - here, "public risk and public health."21 If the criticism is that the violence to be anticipated from letting the guilty go free will itself deprive third-party innocents of their liberty, suitably defined,22 then the liberty of innocents appears on both sides of the balance, and a same-risk tradeoff or perversity argument would arise.23 In light of these countervailing risks, the reasonable doubt rule and the traditional ten to one ratio are abrogated, quite sensibly, when the costs of false negatives seem higher than usual, according to some impressionistic judicial calculus - a pattern that also illustrates the commitment problems with precautionary judicial doctrines, discussed earlier under the heading of futility and parchment barriers. In Hamdi v. Rumsfeld,14 for example, a plurality held that in hearings to determine enemy combatant status, the burden of proof could be placed on the alleged enemy combatant to disprove the government's evidence.25 Here the costs of mistakenly releasing an enemy are high; not a few detainees released from Guantanamo have reappeared as jihadis in Iraq or Afghanistan.26 So the judges relax the Wiener, supra note, at 1517. 21 Id. 22 See Adrian Vermeule, A New Deal for Civil Liberties: An Essay in Honor of Cass R. Sunstein, 43 Tulsa L. Rev. 921,922-28 (2008). 23 Cf. Cass R. Sunstein & Adrian Vermeule, 7s Capital Punishment Morally Required? Acts, Omissions, and Life-Life Tradeoffs, 58 Stan. L. Rev. 703,745 (2005). 24 54211.5.507(2004). ZJ Id. at 534. 2 Alissa J. Rubin, Bomber's Final Messages Exhort Fighters Against U.S., N.Y. Times, May 9, 2008, at A14. 62 Part I. Theory reasonable doubt rule, even though the cost of a mistaken positive finding of combatant status is indefinite preventive detention of the innocent -arguably worse, from the standpoint of a risk-averse innocent, than a term certain with the equivalent expected duration. Decisions of this sort rest on a kind of implicit and unsystematic risk analysis that belies the precautionary principle embodied in the reasonable doubt rule.27 That rule might or might not be optimal in light of all relevant risks, depending on the value of N, but a precautionary perspective does not structure the questions the right way. Free speech. A structurally similar point has been made by Frederick Schauer in the setting of free speech.28 Robust protections for free speech, on Schauer's account, may prevent government from taking precautions against an uncertain possibility or risk of catastrophic harms, such as terrorist attacks. On the other hand, those protections may themselves be seen as precautions against a different harm, "the large-scale restriction of speech."29 The consequence of this view is that free speech analysis should embody a " decision-theoretic approach"30 that takes into account countervailing risks and harms on all sides of speech protections and speech restrictions - in other words, a mature analysis of speech-related problems. The administrative state and the combination of functions. In the protracted rear-guard action fought by various legalists and libertarians against the advance of the administrative state, one of the main arguments has been that combining rulemaking, prosecution, and adjudication in the hands of the same administrative agencies effects a violation of core norms of separation of powers, and thus creates an unacceptable risk of biased agency action. The agency that makes rules and prosecutes violators, the claim runs, cannot possibly adopt an impartial perspective in adjudicating violations of its rules. Accordingly, the Court has periodically been urged to declare the combination of functions in the administrative state a per se violation of due process. The Court, however, has consistently rejected this sort of claim.31 In Withrow v. Larkin,the Court acknowledged that the vast and varied federal 27 See Esmail v. Obama, 639 F.3CI 1075,1077-78 (Silberman, J., concurring). 28 See Frederick Schauer, 7s it Better to be Safe than Sorry? Free Speech and the Precautionary Principle, 36 Pepp. L. Rev. 301 (2009). 29 Id. at 305. 30 Id. at 314. 31 See Marcello v. Bonds, 349 U.S. 302 (1955); FTC v. Cement Inst., 333 U.S. 683 (1948). In Wong Yang Sung v. McGrath,-}-}^ U.S. 33 (1950), the Court more or less endorsed the requirements of the Administrative Procedure Act (APA) as constitutionally sufficient. The APA requires separation of functions at the lower level of initial agency adjudication, Optimizing Constitutionalism: The Mature Position 63 administrative state would grind to a halt if the combination of functions -a routine feature of federal administrative agencies - were declared per se unconstitutional on precautionary grounds.32 This is an implicit jeopardy argument: the administrative state supplies so many valued goods that the risk of administrative bias is a constitutionally tolerable byproduct. I will return to this issue later. Takings and public use. We have seen that critics of the Court's 2005 Kelo decision argued for a rule that economic development cannot count as a public use that justifies a taking of private property - a precautionary stance based on the concern that interest groups might cause legislatures to abuse the power of eminent domain for private ends. One of the Court's main responses, along the very lines of Publius's argument in Federalist 41, was simply that all governmental power can be abused: Speaking of the takings power, Justice Iredell observed that '[i]t is not sufficient to urge, that the power may be abused, for, such is the nature of all power - such is the tendency of every human institution: and, it might as fairly be said, that the power of taxation, which is only circumscribed by the discretion of the Body, in which it is vested, ought not to be granted, because the Legislature, disregarding its true objects, might, for visionary and useless projects, impose a tax to the amount of nineteen shillings in the pound. We must be content to limit power where we can, and where we cannot, consistently with its use, we must be content to repose a salutory confidence.'33 Like the taxing power, in other words, the power to take property for purposes of economic development is justified despite the possibility of abuses, because the gains are greater still. By focusing myopically on a particular target risk - the public-choice concern about interest-group influence - the critics of Kelo overlook the possibility that some level of abuse is optimal, in the sense that it is a necessary byproduct of a constitutional grant of power that is cost-justified overall. PERVERSITY: SAME-RISK TRADEOFFS Perversity arguments are particularly useful and attractive to opponents of precautions when there is a dominant constitutional value in the culture but explicitly permits combination of functions at the top level of agency decision making, and combined arrangements are the norm in most federal agencies. 32 See Withrow v. Larkin, 421 U.S. 35,52 (1975) ("The incredible variety of administrative mechanisms in this country will not yield to any single organizing principle"). 33 Kelo v. City of New London, 545 U.S. 469,487 n. 19 (2005). 64 Part I. Theory of the day - a value that has become sacralized, making it unacceptable to argue that the value should be traded off against other goods. Under conditions of that sort, perversity arguments effect a kind of intellectual jiu-jitsu, turning the value against itself and seizing the high ground. Standing armies. We have seen that in the debate over standing armies, Publius offered a straightforward jeopardy argument based on the goods of public order and national security from foreign invasion. Yet Publius also attempted to turn the Antifederalist argument on its head by suggesting that a prohibition on a national standing army would itself endanger liberty. The j ewel of Hamilton's brilliant series of papers on the issue34 was the crushing Federalist No. 8, titled "The Effects of Internal War in Producing Standing Armies and Other Institutions Unfriendly to Liberty,"35 which offered a sustained case that the Antifederalist stance against standing armies would exacerbate the very risk that it sought to prevent. If Antifederalists blocked ratification of the proposed constitution as a precaution against a national standing army, Hamilton warned, the consequence would be to create a European world of warring polities on the North American continent. Such a world would itself inevitably produce an array of standing armies. Moreover, the militarization of the states would result in systematic expansion of executive authority, "in doing which their [i.e., the state] constitutions would acquire a progressive direction towards monarchy.... Thus we should, in a little time, see established in every part of this country the same engines of despotism which have been the scourge of the old world____[OJur liberties would be a prey to the means of defending ourselves against the ambition and jealousy of each other."36 The Antifederalists' error was to focus on the risks of the proposed constitution, while neglecting the countervailing risks of the steadily deteriorating status quo and the accelerating collapse of the Articles of Confederation regime. Executive power and dictatorship. Along similar lines, Hamilton argued throughout the founding era that precautionary restrictions on the power of the executive would have the perverse result of causing the executive to slip off the bonds of constitutionalism altogether. The general problem, which Hamilton's Federalist No. 20 diagnosed by reference to the history of the Netherlands, was that 34 Most directly, see The Federalist Nos. 23-28, at 152-82 (James Madison) (Clinton Rossiter ed., 1961), as well as the somewhat more general diagnoses of the faults of the Articles of Confederation in The Federalist Nos. 2-10,(Clinton Rossiter ed., 1961). 35 The Federalist No. 8, at 66-71 (Alexander Hamilton) (Clinton Rossiter ed., 1961). 36 Id. at 68,71. Optimizing Constitutionalism: The Mature Position 65 [a] weak constitution must necessarily terminate in dissolution for want of proper powers, or the usurpation of powers requisite for the public safety. Whether the usurpation, when once begun, will stop at the salutary point, or go forward to the dangerous extreme, must depend on the contingencies of the moment. Tyranny has perhaps oftener grown out of the assumptions of power called for, on pressing exigencies, by a defective constitution, than out of the full exercise of the largest constitutional authorities.37 Although this passage in isolation might be taken to refer to government generally, Hamilton elsewhere made it pellucid that the dynamic of excessive weakness turning into excessive strength applied especially to the executive. As he warned the Philadelphia Convention, "establish a weak government and you must at times overleap the bounds. Rome was obliged to create dictators."38 Modern Hamiltonians point to similar possibilities. As against the precautionary view that a reinvigorated or reinforced separation of powers is necessary, in the United States today, to prevent a possible presidential or military coup,39 it may be that the separation of legislative and executive powers is itself a risk factor for dictatorship. For one thing, the separation of powers may reduce civilian control of the military by allowing the military to foment or exploit conflicts between civilian institutions40 - a type of divide-and-conquer strategy. Thus one comparative study of civil-military relations finds that civilian control is greater in the United Kingdom than in the United States.41 Furthermore, the separation of powers might increase the stability of the system in normal times while creating a risk or uncertain chance that the system will become radically unstable in abnormal times.42 Suppose that in a system with an independently elected president, constitutional rule-makers set up elaborate vetogates, legislative and judicial oversight, and other checks and balances, all with an eye to minimizing the risks of executive dictatorship. However, these checks and balances create gridlock and 7 The Federalist No. 20, at 136-37 (James Madison) (Clinton Rossiter ed., 1961). 8 i The Records of the Federal Convention of 1787, at 329 (Max Farrand ed., 1911). 9 See Bruce A. Ackerman, The Decline and Fall of the American Republic 141-79 (2010). 0 Samuel P. Huntington, The Soldier and the State: The Theory and Politics of Civil-Military Relations 177-84 (1957). 1 Deborah D. Avant, Political Institutions and Military Change: Lessons from Peripheral Wars 21-49 (1994)- 2 This paragraph is adapted from Eric A. Posner & Adrian Vermeule, Tyrannophobia, in Comparative Constitutional Design (Tom Ginsburg ed. 2012). 66 Part I. Theory make it difficult to pass necessary reforms. Where the status quo becomes increasingly unacceptable to many, as in times of economic or political crisis, the public demands or at least accepts a dictator who can sweep away the institutional obstacles to reform.43 Here the very elaborateness of the designers' precautions against dictatorship creates pent-up public demand that itself leads to dictatorship. Comparative politics provides (contested) evidence for this story, especially from Latin America.44 The New Deal, a "constitutional moment" of higher lawmaking, can be understood as our brush with the Latin American scenario, in which Roosevelt achieved near-dictatorial stature precisely because he seemed the best hope for overcoming the excessive status quo bias of the Madisonian constitution.45 The comparative evidence on which these mechanisms rest is uncertain. But that very uncertainty, coupled with the severity of the resulting harms if they do materialize, implies that on a precautionary perspective weakening the separation of powers might itself be the best safeguard against dictatorship. At a minimum, there are uncertainties on all sides of the issue; under certain conditions, the separation of powers may represent a self-defeating precaution. Presidential emergency powers. Similar issues arise in recent constitutional theory, a prime example being Ackerman's proposals for constraining presidential emergency powers. Ackerman's "supermajoritarian escalator" proposes a framework statute that requires legislative authorization of emergency powers by successively larger supermajorities. Yet this framework fails to account for the reactions of legislators who themselves are aware of the rules.46 Legislators will know, by virtue of the framework's escalating requirements, that a vote to authorize emergency powers at any See, e.g., Jonathan Hartlyn, Presidentialistn and Colombian Politics, in The Failure of Presidential Democracy 294, 294-96 (Juan J. Linz & Arturo Valenzuela eds., 1994). See id. at 294-96; Juan J. Linz, Presidential or Parliamentary Democracy: Does It Make a Difference? ,inTaE¥AiL\]RE of Presidential Democracy, swpw note 4 3, at 3,6—8; Adam Przeworski et al., What Makes Democracies Endure?,]. Democracy, 39,44-46 (1996). The Latin American evidence is contested in Jose Antonio Cheibub, Presidentialism, Parliamentarism, and Democracy (2007), which argues that the correlation between presidentialism and dictatorship is merely an artifact of selection effects: polities that are less stable to begin with are more likely to have presidential systems. Along similar lines, recent scholarship suggests a roughly one-in-eight chance that executive term limits perversely tend to increase the risk of executive coups, by removing the incentive of strong executives to continue to play within the system (a final-period problem). Tom Ginsburg, James Melton & Zachary Elkins, On the Evasion of Executive Term Limits, 52 Wm. & Mary L. Rev. 1807,1849-50 (2011). See Adrian Vermeule, Self-Defeating Proposals: Ackerman on Emergency Powers, 75 Fordham L. Rev. 631, 641 (2006). Optimizing Constitutionalism: The Mature Position 67 given time will be followed by another vote under even more stringent conditions, and this in effect lowers the cost to legislators of granting such powers in the present. The result is a type of moral hazard: legislators may be more willing to grant emergency powers than they would be in a regime in which, once granted, emergency powers become a new status quo for the indefinite future. In the limit, the result might be perverse; the very mechanism that is intended to constrain presidential emergency powers might cause legislators to grant them more freely. Recess appointments. Recall that the court in Canning barred intrases-sion recess appointments, at least in part as a precautionary measure against presidential aggrandizement. That holding might actually turn out to be perverse, making matters worse on the very same margin the court was worried about. In other words, the court's precaution against presidential aggrandizement might actually increase the overall risk of aggrandizement in the long run. How would this occur? The main mechanism involves the risk of backlash. Suppose that the combination of the filibuster, other obstructionist tactics in the Senate, and decisions like Canning eventually produce so much pent-up public demand for reform of the appointments process that the president offers some radical reinterpretation of the Constitution, one that gives him substantially increased discretion over appointments. Ingenious commentators have already begun to supply such reinterpretations.47 Should the new position stick as a political equilibrium, then - given the Canning court's own concern with safeguards against presidential power -the court might bitterly regret, ex post, that it threw up an obstruction that contributed to creating a backlash in the other direction. The general point is that an enlightened decision maker will do well to consider the systemic, dynamic, and long-run effects of any given precaution, including the long-run risk of backlash resulting in perverse outcomes. True, where information is costly and time is limited, a rational and sophisticated decision maker might decide to ignore all long-run effects, on the theory that the dynamic possibilities are so numerous and varied as to be essentially incalculable. But that would be a different, far more respectable and self-aware sort of myopia than the myopia on display in Canning. The Senate. Antifederalists favored direct and frequent legislative elections and restrictions on the re-eligibility of representatives, as precautions For one possibility, see Matthew C. Stephenson, Can the President Appoint Principal Executive Officers Without a Senate Confirmation Vote?, 122 Yale L.J. 940 (2013). 68 Part I. Theory against elected oligarchy, corruption, and abuse of power. For obvious reasons, the indirect election, long terms, and indefinite re-eligibility of the Senate triggered Antifederalist fears. Madison, writing as Publius in Federalist 63, responded with a perversity argument: In answer to all these arguments, suggested by reason, illustrated by examples, and enforced by our own experience, the jealous adversary of the Constitution will probably content himself with repeating, that a senate appointed not immediately by the people, and for the term of six years, must gradually acquire a dangerous pre-eminence in the government, and finally transform it into a tyrannical aristocracy. To this general answer, the general reply ought to be sufficient, that liberty may be endangered by the abuses of liberty as well as by the abuses of power; that there are numerous instances of the former as well as of the latter; and that the former, rather than the latter, are apparently most to be apprehended by the United States.48 Madison's point rests on the sort of precommitment argument that underpins much of liberal constitutionalism: a Senate "may be sometimes necessary as a defense to the people against their own temporary errors and delusions____What bitter anguish would not the people of Athens have often escaped if their government had contained so provident a safeguard against the tyranny of their own passions?"49 The sting in the final sentence is the word "tyranny," which emphasizes the perverse threat to political liberty posed by unrestricted popular liberty to direct events. Free speech.50 In a liberal legalist culture, the sacrosanct status of free speech principles - understood as precautions against politically motivated abuse of governmental power - implies that such principles are especially likely to become the target of perversity arguments, which attempt to turn such principles against themselves. Two illustrations come from the free speech law of political protest and subversive advocacy. The first is Justice Jackson's dissent in Terminiello v. City of Chicago,51 in which the Court invalidated a conviction of a defrocked Catholic priest, a right-wing demagogue, for breach of the peace. The demagogue had given a speech that caused riotous battles between his supporters and a hostile mob of left-wing activists. Jackson's argument appealed in part to the benefits of 48 The Federalist No. 63, at 387-388 (James Madison) (Clinton Rossiter ed., 1961) (emphasis added). 49 Id. at 384. 50 A helpful treatment of the cases and issues discussed in this sub-section is Eugene Volokh, Freedom of Speech and the Constitutional Tension Method, 3 U. Chi. L. Sch. roundtable 223 (1996). 51 337 U.S. 1 (1949)- Optimizing Constitutionalism: The Mature Position 69 public order, but argued more pointedly in the vein of perversity: liberty -in particular the liberty of free political speech - itself requires public order as a precondition of its existence, so that the Court's short-sighted protection of speech put at risk the very freedom it was intended to protect. As Jackson put it, [i]n the long run, maintenance of free speech will be more endangered if the population can have no protection from the abuses which lead to violence. No liberty is made more secure by holding that its abuses are inseparable from its enjoyment. We must not forget that it is the free democratic communities that ask us to trust them to maintain peace with liberty and that the factions engaged in this battle [Fascists and Communists] are not interested permanently in either.52 Jackson here implicitly points to the classic liberal dilemma whether liberalism requires toleration of the intolerant. In the long run, the argument goes, liberalism may undermine itself by tolerating political speech and participation by groups who would repeal liberal protections if they came to power. This is a large-perversity argument: liberal freedoms, at least if pressed too far, put themselves at risk. That liberal dilemma also underpins the opinion in Dennis v. United States, which upheld a conviction of Communist defendants for conspiring to organize the Communist Party to advocate and teach the overthrow of the U.S. government by violence.53 As we have seen, the majority opinion by Chief Justice Vinson upheld the conviction by adopting Judge Learned Hand's expected-harm test for free speech protection, under which courts "ask whether the gravity of the evil, discounted by its improbability, justifies such invasion of free speech as is necessary to avoid the danger."54 As against Justice Black's dissent, which offered a precautionary argument for free speech protection - "the freedoms [that the First Amendment] guarantees provide the best insurance against destruction of all freedom"55 - the majority replied that if the government fell to Communism, all of Black's freedoms, including free speech, would fall with it: Overthrow of the Government by force and violence is certainly a substantial enough interest for the Government to limit speech. Indeed, this is the ultimate value of any society, for if a society cannot protect its very structure 52 Id. at 36-37. 53 341 U.S.494 (1951). 54 Id. at 510 (internal quotation omitted) (citing United States v. Dennis, 183 F.2d. 201,212 (2d Cir. 1950)). 55 Id. at 580 (Black, J., dissenting). 70 Part I. Theory from armed internal attack, it must follow that no subordinate value can be protected.5*5 This sort of perversity argument boils down to the claim that liberty itself depends on strong and stable government - the obverse of Benjamin Franklin's civil-libertarian claim that "[t]hose who would give up essential [l]iberty to purchase a little temporary [s]afety, deserve neither [ljiberty nor [s]afety."57 Judicial Review. The perversity critique of a precautionary approach to free speech cases can be generalized into a critique of the larger precautionary justification for judicial review. Conditional on certain theories of rights that recognize affirmative claims to governmental aid or support, judicial review can be argued to block legislative or executive measures that are necessary to implement rights or to protect rights against private violation. On this view, the perverse result is that judicial review might increase the overall incidence of rights-violations. "Inserting an additional veto point into the process of obtaining effective legislation threatens erroneous underprotection of fundamental rights; it does not provide a 'hedge' against legislative underprotection of fundamental rights."58 Judicial review, interest groups, and property rights. Along similar lines, perversity underlies a general critique of interest-group justifications for judicial review.59 Such theories posit that courts should exercise closer scrutiny of governmental action where there is a risk of interest-group "capture" - a risk that narrow, well-organized groups will exert disproportionate influence over legislatures or agencies, according to some normative conception of proper influence. As we have seen, the capture concern underlies many criticisms of the Kelo decision, which allowed governmental takings of private property for purposes of economic development. The critics believe that interest groups will cause government to abuse the takings power in order to confer benefits on private groups, reducing social welfare overall and undermining the security of property rights. In Chapter 4,1 will discuss some of the affirmative "benefits of capture,"60 which can provide decision makers with specialized information 56 Id. at 509 (majority opinion). 57 Benjamin Franklin, Pennsylvania Assembly: Reply to the Governor (Nov. 11,175 5), in 6 The Papers of Benjamin Franklin 242 (Leonard W. Labaree ed., 1963). 58 Mark Lushnet, How Different are Waldron's And Fallon's Core Cases For and Against Judicial Review?, 30 Oxford J. Legal Stud. 49, 61 (2010) (emphasis in original). 59 For an overview and critique of such theories, see Einer R. Elhauge, Does Interest Group Theory Justify More Intrusive Judicial Review?, 101 Yale L.J. 31 (1991). 60 See Dorit Rubinstein Reiss, The Benefits of Capture, 47 Wake Forest L. Rev. 569 (2012). Optimizing Constitutionalism: The Mature Position 71 or spur decision makers to take decisions. Even apart from such benefits, however, it has been argued that enhancing judicial scrutiny in order to raise the costs of interest-group capture may have perverse effects. In the words of a leading critic of interest-group justifications for judicial review, even if more intrusive judicial review does increase the transaction costs of capture, that can perversely encourage interest group activity by making successful capture harder to undo. [Furthermore], because increasing transaction costs also increases the costs facing large diffuse groups, it may increase the relative advantage of small intense groups and thus increase their success.61 In other words, the interest-group justifications for judicial review focus myopically on one sort of response to the threat of capture, overlooking that the precautions they advocate may themselves exacerbate that very threat. Maximin constitutional design.62- As we have seen, Federalist 41 gives the most general form of the jeopardy argument: precautions that restrict governmental discretion to provide other goods may reduce social welfare overall. The most general form of the perversity argument holds that designing a constitution on worst-case assumptions - maximin constitutional design, loosely speaking - may itself bring about the worst case. In light of that risk, systematic constitutional caution may prove self-defeating. An example is a critique of Hume's knavery principle, discussed in Part LA. Anti-Humean critics object that the expectation of knavery is self-fulfilling: presuming officials to be knaves will tend to make officials into knaves.63 One possibility is that official motivations are partly endogenous to the constitutional rules. On this view, "a constitution for knaves crowds out civic virtues,"64 a possibility that in turn can rest on one of several mechanisms.65 Constitutional sanctions for self-interested behavior 1 Elhauge, supra note, at 8 8. 2 For an earlier stab at these issues, see Adrian Vermeule, Hume's Second-Best Constitutionalism, 70 U. Chi. L. Rev. 421 (2003). 3 An excellent treatment of these issues is Lewis A. Kornhauser, Virtue and Self-Interest in the Design of Constitutional Institutions, 3 Theoretical Inq. L. 21 (2002). 4 See Bruno S. Frey, A Constitution for Knaves Crowds Out Civic Virtues, 107 Econ. J. 1043,1044-45 (i997)- 5 For a model of conditions under which laws (here meaning legal sanctions or incentives) and norms act as either substitutes or complements, see Roland Benabou & Jean Tirole, Laws and Norms (Nat'l Bureau of Econ. Research, Working Paper No. 17 5 79, November 2011), available at http://ssrn.com/abstract= 1954505. 72 Part I. Theory might undermine social norms that would constrain the same behavior; the net effect may then be an increase in self-interested action by officials. Alternatively, sanctions for self-interested behavior might (unintentionally) convey a signal that many other officials are engaged in the behavior that the sanctions aim to eliminate. If so, the net result may be an increase in noncompliance by office-holders, either because they are conformists who adjust their behavior to track what the majority does,66 or because they are "reciprocal altruists" who would comply with public-regarding norms if others were complying also, but who are afraid of being chumps and are thus unwilling to comply unilaterally.67 Even if officials' motivations are not endogenous to the constitutional rules, the Humean presumption of knavery may have selection effects that perversely tend to filter self-interested actors into office while filtering out public-spirited actors. If public-spirited actors experience a cost from being subjected to elaborate monitoring devices based on a presumption of knavery - the cost of frustration, or of operating under a cloud of suspicion - then the presence of such devices will tend, at the margin, to filter out such actors while filtering in actors for whom the presumption of knavery is, in fact, accurate. Although this is merely one possible effect of such devices, the overall result may be that designing a constitution for knaves perversely tends to select knaves into the public sphere. EX ANTE PRECAUTIONS VS. EX POST REMEDIES: "NOT WHILE THIS COURT SITS" By their nature, precautions are taken ex ante the materialization of the relevant risk. Accordingly, another argument against precautions is that ex ante safeguards are unnecessary in light of the availability of ex post remedies, such as suits for damages against officers who execute an unconstitutional policy. Technically, this is a special case of jeopardy and could have been covered under that heading. The core of the concern is simply that ex ante precautions that are unnecessary will imposes costs greater than their benefits. Because the temporal dimension is distinctive, however, separate treatment is warranted. Dirk Sliwka, Trust as a Signal of a Social Norm and the Hidden Costs of Incentive Schemes, 97 Am. Econ. Rev. 999,1000 (2007). Joel J. Van der Weele, The Signaling Power of Sanctions in Social Dilemmas, 28 J.L. Econ. & Org. 103-26 (2012). Optimizing Constitutionalism: The Mature Position 73 There is a standard conceptual issue about the distinction between ex ante precautions and ex post remedies, one that crops up in this setting as well. Clearly, the anticipation of an ex post sanction will itself produce ex ante deterrent effects, if the law can make a credible commitment to providing remedies after the fact. Yet it is wrong (here and elsewhere) to assume that there is no difference between ex ante precautionary regulation and a system of ex post sanctions. For present purposes, a key difference is whether the constitutional rule is formulated to ward off an uncertain harm, or instead is formulated to require that the complaining party demonstrate that a harm has already materialized. The latter approach places the burdens of production and proof on the complainant and requires case-by-case assessment of evidence before the tribunal. Taxation of federal instrumentalities and contractors. As we have seen, Chief Justice Marshall's opinion in McCulloch formulated a rule against state taxation of federal instrumentalities, based on the precautionary principle that "the power to tax involves the power to destroy."68 The rule of McCulloch - a "prophylactic per se rule"69 that still holds today - is that states may not regulate or tax federal instrumentalities without express congressional authorization. Later cases extended the precautionary zone to bar state taxation of private parties who do business with the federal government.70 The governing doctrine was that any state regulation or tax that indirectly regulated the federal government's activity was invalid; the point of this doctrine was to create a precautionary buffer to protect the freedom of otherwise valid federal operations. Justice Holmes, by contrast, articulated a competing position: ex post, case-by-case assessment of the destructive effect of state taxation on federal contractors would be enough to protect vital federal interests, without overprotecting those interests to such a degree as to squash the legitimate taxing power of the states. In the most famous of Holmes's opinions on this issue, a dissent in Panhandle Oil Co. v. Mississippi ex rel. Knox,71 Holmes voted to uphold a state sales tax on oil sold to the United States for the use of the Coast Guard. "The power to tax," Holmes argued, "is not the power to destroy while this Court sits."72 The Court's ability to review interference with federal operations at retail thus undermined the need for a wholesale 68 1711.8.316,431(1819). 69 Laurence H. Tribe, American Constitutional Law 1223 (3