111 Yale L.J. 231 (2001)

This Article presents a comprehensive textual framework for the allocation of the foreign affairs powers of the United States government. The authors argue that modern scholarship has too hastily given up on the Constitution's text and too quickly concluded that the Constitution contains enormous gaps in foreign affairs that must be filled by contratextual considerations. In particular, modern scholarship incorrectly regards the text as largely unhelpful in resolving three central foreign affairs problems: (1) What is the source of the foreign affairs powers conventionally believed to lie with the President but apparently beyond the President's explicit textual powers? (2) What is the source of Congress' authority to regulate foreign affairs matters that do not seem encompassed by Congress' enumerated powers? (3) How should one allocate foreign affairs powers not specifically mentioned in the text and claimed by both the President and Congress, such as the powers to set foreign policy, to enter into executive agreements, and to terminate treaties?

Contrary to the trend in modern scholarship, this Article arguers that the constitutional text provides a sound guide for resolving these puzzles. The authors derive four basic principles from the Constitution's text (and its history). First, the President has a "residual" foreign affairs power from Article II, Section I's grant of "the executive Power." The executive power, as described by political theorists consulted by the Framers - such as Locke, Monesquieu, and Blackstone - included foreign affairs power. By using a common phrase infused with that meaning, the Constitution establishes a presumption that the President has the foreign affairs powers that were traditionally part of the executive power. Second, the Framers thought the traditional executive had too much authority over foreign affairs, so they specifically allocated many key powers (including war, commerce, and treaty-making). in whole or in part, to other branches. These are allocations away from the President, and thus, despite having "the executive Power," the President cannot claim independent authority in these areas. Third, although Congress lacks a general power over foreign affairs, it has two textual fonts of foreign affairs power: powers specifically given to it (such as war and commerce) and its power to carry into execution powers granted to other branches by the Constitution. The latter is a derivative power, exercisable in conjunction with the President, to give effect to the President's executive power over foreign affairs. Finally, although the President has broad residual power over foreign affairs, that power does not extend to matters not part of the traditional executive power. Hence, the President cannot claim lawmaking or appropriations power in foreign affairs.

111 Yale L.J. 399 (2001)

A veil of ignorance rule (more briefly a "veil rule") is a rule that suppresses self-interested behavior on the part of decisionmakers; it does so by subjecting the decisionmakers to uncertainty about the distribution of benefits and burdens that will result from a decision. A veil rule may produce this distributive uncertainty by either of two methods. One method is to place decisionmakers under a constraint of ignorance about their own identities and attributes. John Rawls coined the phrase "veil of ignorance" to describe a hypothetical original position in which principles of justice are chosen under precisely this constraint. But that is a special case of veil rules generally, indeed a radical case. Rawls's thought experiment introduces uncertainty by allowing the decisionmaker to know the distributive consequences of a decision on future citizens--call them A and B--but denying the decisionmaker the knowledge of whether she herself will occupy A's position or B's position. Where veil of ignorance rules appear under historical rather than hypothetical conditions, however, the relevant decisionmakers will usually know their own identities and interests. Veil rules that appear in actual constitutions, then, more often adopt a second method for introducing uncertainty: Although the decisionmaker knows or can guess whether she will occupy A's or B's position, the rule introduces uncertainty about whether A or B will reap the greater gains from the decision.
 
By speaking of veil rules in constitutions, I mean to pose a very different question than the one pursued in the standard discussions of the veil of ignorance. The constitutional choice literature stemming from James Buchanan, Gordon Tullock, and their successors conceives the uncertainty produced by the veil of ignorance as a mechanism for inducing hypothetical constitutional designers to approach the choice of the constitutional rules themselves in an impartial way. Decisionmaking by legislators and other officials within the framework of the constitutional rules, by contrast, falls in the domain of "ordinary politics," where self-interested individuals and factions struggle for advantage. The constitutional designers' self-interest is constrained by uncertainty; that of ordinary decisionmakers is constrained by voting rules (such as supermajority requirements), by substantive constitutional prohibitions on inefficient legislation, and by institutional competition resulting from the separation of powers. I erase that distinction by asking whether and how constitutional rules might subject in-system decisionmakers to the same uncertainty constraint that governs the hypothetical stage of constitutional choice, and for similar reasons. I also touch upon an important special case, the proposal of constitutional amendments, that shares features of both constitutional choice and ordinary politics.
 
I argue that the Federal Constitution itself contains a number of rules that may usefully be analyzed as veil rules. Provisions, structures, and practices as diverse as the Ex Post Facto and Bill of Attainder Clauses, the Emoluments Clause, the Twenty-Seventh Amendment, Article V's procedures for constitutional amendment, the doctrine of precedent, the original mechanism for selecting senators (by vote of the state legislatures), and the rules governing presidential election and succession may all profitably be considered in this light, although not all of these should count as examples of veil rules rightly understood. The legal literature on these and other topics makes casual references to the veil of ignorance, but there has been very little sustained examination of the subject of veil rules as a general strategy for promoting impartial decisions under actual constitutions. My initial aim, then, is to synthesize and critique these localized literatures in order to obtain an overview of a recurring theme in constitutional design.
 
The payoff from this synthesizing work is that it helps supply an answer to two questions: why the Constitution does not contain more veil rules than it actually does, and why it uses veil rules where it does use them, but not elsewhere. It sounds paradoxical to move from an explication of existing veil rules in some settings to an explanation for their conspicuous absence in others, but that question illuminates the tradeoffs inherent in constitutional design. Having appreciated the power of veil rules to dampen self-interest, we might want to know why the veil technique is not ubiquitous in the Constitution. In particular, it is a striking feature of constitutional law that Congress is subject to more constitutional veil rules, of wider scope, than is the President or the judiciary. Why should that be so, given that it would be perfectly possible to apply a range of veil rules to the latter institutions as well?
 
Some of the literature suggests that the skewed distribution of constitutional veil rules is best explained by the presence or absence of alternative institutional features that suppress self-interested decisionmaking. Federal judges, for example, are not restricted by veil rules requiring prospective and general decisionmaking because life tenure and the design of the adjudicative process independently serve to suppress the decisional bias that veil rules are used to check. While this view gives a plausible account of the paucity of veil rules governing judicial action, its logic suggests that the executive branch should be subject to a far more stringent set of veil rules than it actually is. A second type of explanation applies the insight that the price of reducing bias is to reduce decisionmakers' information. In some settings, the information suppressed by a veil rule is so valuable that its loss might be thought to outweigh even large gains in decisionmaker neutrality. This is true and important, and I shall have recourse to it more than once in explaining the detailed scope of particular veil rules. But paradoxically, the insight is too powerful to be really useful. Any distribution of veil rules across the Constitution, even a distribution much different than the one we see, could be explained by supposing that the costs of foregone information are (or are not) excessive in settings where veil rules do not (or do) apply.
 
I emphasize a third and somewhat different explanation, one that points not to the direct effects of veil rules but to their secondary or indirect consequences. The indirect tradeoff, I argue, is not between information and neutrality, but between information and motivation, or (as the Framers would have put it) institutional "energy." Veil rules not only dampen both information and bias; they also suppress decisionmakers' activity. Removing the spur of self-interest threatens to reduce decisionmakers' activity below acceptable levels, to the point where constitutional designers might plausibly prefer to lift the veil and spur more activity, even if the price is that some fraction of that increased activity is self-regarding. If, like the Framers, we systematically fear excessive congressional activity, on the one hand, and fear insufficient presidential (and even judicial) activity, on the other hand, then something roughly like the current skewed distribution of veil rules suggests itself. The enervating effect of veil rules would amount to a qualified good in the legislative setting and a qualified bad in executive and judicial settings. This is an interpretive explanation or justification of the Constitution and its implicit theoretical commitments. I make no normative claims about how a new constitution should be designed from scratch, nor do I attempt historical analysis of the later development of federal political institutions, such as the (relative) growth of presidential power.
 
The plan of the Essay is as follows. Part I defines terms, distinguishes veil rules from the separation of powers and other types of constitutional rules that restrict self-interested decisionmaking, and sets out a few methodological premises. Part II surveys constitutional veil rules by examining "veil tactics": features of constitutional provisions and doctrines that produce veil-like effects. Examples are constitutional requirements that official decisions be prospective and general, such as the Ex Post Facto and Bill of Attainder Clauses, and constitutional rules that increase the durability of decisions or delay their effective date, such as the doctrine of precedent in constitutional cases, the Twenty-Seventh Amendment, and the Emoluments Clause. I also touch on the (infrequent) use of randomization in constitutional law. Part III examines the direct effects of veil rules on decisionmakers' information and their indirect effects on decisionmakers' motivation, emphasizing that the enervating effect of veil rules helps us toward an account, or a rationalization, of the distribution of veil rules across institutions. Part IV is a brief conclusion.