I. Introduction
The law of foreign relations as a constitutional discipline in the United States has grown considerably in recent years.Footnote 1 While people have debated the meaning of the American Constitution's foreign-relations clauses since the beginning of the Republic, in recent years scholars, judges, and politicians have increasingly clashed over what the Constitution says. The acrimonious contenders are the internationalists and the exceptionalists.Footnote 2 Internationalists would like to see the Constitution adapt to the evolving strictures of international law and politics. They generally favor three policies: a robust incorporation of international norms into the American constitutional system;Footnote 3 more congressional and judicial control of the president's conduct of foreign policy;Footnote 4 and a limited role for federalism and the states in international affairs.Footnote 5 Exceptionalists claim either that the original understanding of the U.S. Constitution does not sit well with the internationalist view, or that the United States should follow its own course free of most international constraints, or both. To them, international norms may not be entirely compatible with the Constitution or the values and interests of the United States. According to exceptionalists, constitutional interpretation should reflect those differences. They generally favor policies that are the opposite of those favored by internationalists: limited foreign influence on the American constitutional system;Footnote 6 a strong role for the executive in the conduct of foreign policy and a correspondingly limited role for Congress and the courts;Footnote 7 and a revival of principles of federalism in foreign affairs.Footnote 8 As the newcomers to the foreign policy scene, exceptionalists have challenged the internationalist orthodoxy that dominated the discipline since the end of the Second World War.
The contenders in this debate appeal to constitutional text, history, and structure.Footnote 9 For the exceptionalists, the drafters of the U.S. Constitution were weary of foreign entanglements and set up rules that reflected that concern. Constitutional interpreters should respect that understanding. On the exceptionalist view, the preeminence of the executive in foreign affairs is supported by a combination of historical, textual, and functional arguments. People at the time widely agreed that the executive was supposed to be in charge of foreign relations, and that is why the Constitution remained silent on many of those issues. Functional reasons—reasons of coordination—suggest that the executive must control foreign policy: deliberative bodies are too slow and cumbersome to manage decisions on war and peace. Internationalists challenge these views on several grounds. To them, text and history are not as clear as the exceptionalists claim. The Founders were by and large neophytes in foreign policy, and, in part for that reason, they expected the gaps in the constitutional text to be resolved through custom.Footnote 10 And with respect to constitutional structure, the internationalists dispute the view that an all-powerful executive better serves the foreign policy needs of the Republic. To them, the need to curb an imperial president is as important as the need to have optimal coordination, at least in the long run.
I propose a fresh approach to this foreign policy debate. For the most part, I will set aside the arguments from text, history, and structure, and ask a more abstract question: How should one ideally design the foreign-relations clauses of a liberal constitution? Given liberal values and what we know about the world, what should the constitution say about the conduct of foreign relations? The question is not merely academic, for several reasons. A number of new nations have emerged in the last twenty years or so, some of which purport to embody liberal values. Fresh institutional design is obviously called for in their case, since their emergence is a clean break with their (usually) illiberal traditions. Moreover, focusing on ideal theory may help us to interpret the U.S. Constitution in a philosophically sound way, and to perceive more clearly the differences between what we have and what would be optimal. Last but not least, ideal design has a liberating effect because it steers constitutional debate away from original intent and actual constitutional practice—in particular, judicial precedent. We are forced to think about what the Constitution would look like if it were possible for us to ignore two hundred years of constitutional practice. I hasten to say that I do not mean to imply that standard constitutional arguments are unimportant or irrelevant. On the contrary, a plausible theory of law should give pride of place to such arguments. Yet the goal of this essay is different. I am not interested in the correct interpretation of the foreign-affairs clauses of the U.S. Constitution as a matter of law. Rather, my goal is to imagine what an ideal liberal constitution should do with respect to foreign affairs. Put differently: I will assess the claims of internationalists, exceptionalists, and others, not in terms of constitutional history, text, or structure, as they do, but in terms of political theory and social science. Hopefully I will show that philosophy and social science have something to say about the appropriate foreign policy arrangements for a liberal republic. Having said that, I will often refer to the foreign-relations clauses of the United States Constitution for expository convenience.
I argue that a liberal constitution must enable the republic to implement a morally defensible foreign policy in the light of available knowledge. In Section II, I examine the implications of the first pillar of a liberal foreign policy: the defense of liberty. I claim that the defense of liberty indicates not only national self-defense—defense of persons, territory, and institutions—but also, with due caution, the promotion of liberty abroad. Section III argues that a liberal foreign policy should enable people to pursue prosperity. To that end, the liberal constitution must foster free trade at home and abroad. In Section IV, I apply these principles to institutional design. I argue first that the liberal constitution's war provisions must facilitate the defense of liberty by securing combat readiness above all. This requires a flexible executive to respond promptly to threats. However, in line with most theorists, I recommend a democratic check on war whenever possible. I then address the question of the incorporation of international law into domestic law. I suggest that the approval of treaties should follow the same procedure used for domestic lawmaking, and recommend that treaties be generally self-executing. I then discuss customary law in detail. I propose a test to determine the quality of customary law that, in my view, can make that law eligible for use in domestic courts. I conclude Section IV by proposing that, in the light of well-known pathologies in the political process, the liberal constitution ought to prohibit the government from erecting protectionist barriers. In Section V, I propose some concrete constitutional clauses. Section VI sets forth some conclusions.
II. The Foreign Policy of a Liberal Republic: Liberty
A. What is a liberal constitution?
In this essay, I argue that a constitution should enable the government of a liberal state to defend its society and institutions—to defend the freedom of its people. This argument assumes that the justified constitution is the liberal constitution, that is, the set of basic rules that define the workings of a liberal society. This assumption, in turn, requires that I clarify what I understand by a liberal state. My definition here is simple and quite unsophisticated: a liberal state is a state that has two features. First, it protects individual freedom in the form of a bill of rights or similar legal device, conjoined with an effective (and not just rhetorical) institutional commitment to the rule of law and an independent judiciary. Second, it is ruled by democratically elected leaders who rotate periodically. To anticipate obvious objections, I hasten to say that the first feature is by far the most important. The second condition (the democratic credentials of the government) cannot alone rescue a society from tyranny, as the experiences in Nazi Germany and present-day Venezuela show. But the condition is still important for a number of reasons that have been amply discussed in the literature. In foreign affairs, the government is supposed to represent the people, and universal adult suffrage seems to be the least imperfect tool to establish that representation.
Someone may contest my assumption that the liberal constitution is the only justified constitution. For example, John Rawls has argued that the world contains regimes that, while illiberal, can still be seen as justified: they are “decent.”Footnote 11 If so, I may be unduly biased in my analysis. If there is more than one justified constitution, then the analysis of the optimal foreign-relations clauses should reflect that diversity of (justified) regimes. For reasons I have presented elsewhere,Footnote 12 I disagree with Rawls—I do claim that the liberal constitution is the only one that can be justified philosophically—but here I do not need to substantiate that position. For Rawls's problem was different from the problem I tackle here. Rawls was concerned with devising a system of international law that would take account of the many inconsistent yet reasonable existing political and religious systems of belief around the world. He would accept illiberal regimes (as long as they were not tyrannical) for the purpose of peaceful coexistence, as it were. For Rawls, if illiberal regimes are not tyrannical, liberal governments should treat them as members in good standing of the international community.
My goal is different. I am not trying to propose the best system of international law given the diversity of comprehensive conceptions of the good upheld by nations in the world (although, as I said, I have rejected Rawls's proposal). I am instead trying to imagine what the foreign-relations clauses of a liberal constitution should look like. This does not prejudge the international status of the many illiberal states in the world, nor does it prejudge what the foreign-relation clauses of an illiberal state should look like. I assume the moral superiority of the liberal constitution for all the standard reasons given by political philosophers, yet I need not assume that a system of international law (which I am not discussing here) should or should not welcome illiberal regimes.
B. Defense of liberty
An ideal constitution should enable the government to implement a morally justified foreign policy. Thus, I reject the view, anticipated by John Locke and sometimes endorsed by others, that while it is appropriate for constitutions to establish limits on governmental power domestically, foreign policy should largely be left to the government's (usually the executive's) unrestrained discretion.Footnote 13 There are a number of reasons for rejecting this view. First, establishing constitutional restraints on the government's conduct of foreign affairs does not mean depriving the government of the resources to confront foreign threats. Second, in a globalized world it makes sense to design institutions so that the citizens of the liberal state will see their rights protected, not only by their government's internal actions, but by its external actions as well.Footnote 14 If the main reason to have a constitution is to limit the power of government so that our rights will be protected, there is no reason to exempt foreign policy from this stricture. Finally, foreigners have rights too. They possess these rights by being persons. In this sense, basic human rights are not dependent on tradition and culture, and even less on citizenship. Rights come with personhood and are in this sense natural rights; in addition, they have been partially embodied in international law.Footnote 15 The liberal constitution ought not to allow the violation of the basic human rights of foreigners (with some exceptions in cases of urgent threats to domestic liberty).
A precautionary warning is in order: A constitution cannot contemplate everything that might arise in the diplomatic arena. A solution that may seem justified in the abstract may not look so good once non-ideal conditions are specified. Institutional design must be consistent with what we know about the world and must take into account the incentives that constitutional norms create for domestic and foreign actors. So philosophy here is not pure critical philosophy: it is philosophy coupled with the best empirical theories and knowledge available. What's more, constitutional designers should realize that second-best solutions may be all that is possible in an imperfect world, especially because, hard as institutional designers may try, they cannot control future behavior—in particular, the future behavior of foreign governments and their citizens.
With this caveat, I start by outlining the legitimate goals of foreign policy. Assuming a reasonably well-functioning liberal state and the existence of foreign enemies, foreign policy should aim at defending the republic. This goal, in turn, has two related aspects. The more immediate goal of the government is to physically defend its citizens' lives and property and the state's territory against foreign aggression. This is one of the raisons d'être of governments. Moreover, the government must defend liberal institutions against enemies bent on destroying those institutions. The war against fascism and the current war against terrorism are examples of this dual aspect of the defense of liberty. A liberal government is not pacifist: it responds to foreign aggression against its citizens, and it likewise responds to foreign attempts to undermine or destroy its liberal institutions.
Someone may object that foreign policy should aim at defending the nation, its citizens, and its territory, regardless of the quality of its institutions. Let us call this view patriotism. On this view, defending liberty is not as important as defending persons and property. Rulers of unfree societies also have the power to defend their people and territory against aggression, so whether the constitution is liberal or illiberal is irrelevant to this aspect of foreign policy.
I have two answers. First, this essay focuses on the ideal constitution. If one assumes, as I do, that the optimal constitution is the liberal constitution, it follows that the foreign policy of a liberal constitution must enable government to defend the justified institutions of the liberal state, in addition to the government's obvious job of defending its citizens and territory. Determining the justified foreign policy of illiberal states—that part of the self-defense rationale which is equally applicable to liberal and illiberal states—is beyond the scope of this essay. Suffice it to say that a liberal constitution must provide for the defense of persons and the defense of institutions. A liberal government is morally entitled to defend the country against enemies because its institutions are worth defending, not because they are its institutions.
Second, leaving aside the fact that this is an essay about liberal constitutions, I would point out that as a general proposition patriotism is only partially correct. It is true that a dictator has the power to defend the state against an unjustified attack even if he is illegitimate and even if the state's institutions are illiberal. A dictator has caretaker responsibilities notwithstanding the fact that he does not represent anyone, because he is the only person who can in fact command the state's resources and thus prevent a foreign aggressor from subjugating people. However, dictators do not have a right of self-defense against actions directed at ending their tyranny—that is, against justified humanitarian interventions. Thus, for example, Saddam Hussein's Revolutionary Guards were not morally entitled to confront the troops liberating Iraq. This is so, even if the intervention in Iraq was objectionable on some other ground, and even if members of the Revolutionary Guard could be morally excused for bearing arms in defense of a tyrant. So dictators have a provisional right to defend persons and territory against attack, but they do not have a right to defend illiberal institutions, including their own positions of power.
The goal of defending liberty suggests the desirability (and perhaps the obligation) for a liberal republic to secure alliances with other liberal republics. The justification for this is manifold. A liberal alliance secures the gains of cooperation by virtually eliminating the dangers of war among its members. According to the “liberal peace” thesis, war is highly unlikely among liberal nations. This thesis holds that there is a robust causal link between domestic liberal institutions and international peace. It was advanced first by Immanuel Kant, and revived after the Cold War by political scientists.Footnote 16 Kant's original idea was that democracies (“republics,” in Kant's parlance) are less inclined to war because in a democracy the people know they will bear its burdens. Citizens internalize the costs of war. In a dictatorship, in contrast, the despot sends others to fight: his perks and riches remain largely intact.Footnote 17 He externalizes the costs of war. But history has shown that Kant's thesis is too strong, because democracies frequently go to war against nondemocracies. So “democratic peace” scholars have reformulated the Kantian thesis thus: democracies do not go to war with each other; however, they often go to war against illiberal states. And of course, illiberal states fight one another frequently. This reformulation has been dubbed the separate liberal peace, and it better fits the empirical evidence.Footnote 18 Democratic peace theory has provided support for the American (and perhaps European) policy of promoting liberal institutions overseas. The thesis has attracted vigorous criticism, but it is far from being refuted.Footnote 19
Moreover, a liberal alliance is a bulwark against illiberal foreign threats. NATO was formed with this goal in mind (and not just to reap the benefits of liberal cooperation), and it served this function well during the Cold War. The nature of regimes, therefore, is important when forging alliances. Both moral and prudential reasons should incline liberal governments to create robust alliances with fellow democracies. The opposite view, that the moral credentials of allies are unimportant and that a liberal state should be free to secure alliances with whomever helps pursue that state's national interest (a view that has been historically popular, and is popular even today), is objectionable for moral and prudential reasons. Morally, freedom-loving governments should not cooperate with tyrants. Prudentially, there is always the danger that the illiberal ally will defect at crucial moments (precisely because, by definition, the tyrant is not interested in defending liberty), and that sooner or later the illiberal ally will blight the liberal values of the alliance (the Trojan Horse problem).
Finally, liberal governments have an obligation not to cooperate with tyranny.Footnote 20 By forming alliances with tyrants, liberal governments enable them to oppress their subjects. The obligation not to cooperate with tyranny is relatively weak. It does not necessarily entail a positive duty to depose the tyrant. But it does mean that a liberal regime should refrain from gestures toward the tyrant that enable him to persist in tyrannical practices. Of course, this is easier said than done. Often the liberal state has urgent security interestsFootnote 21 that force it to forge provisional alliances with questionable regimes.Footnote 22 In addition, regimes are often hard to classify along this stark liberal-illiberal dichotomy. Latin America, for example, abounds in “kleptocracies”: democratic governments which by-and-large respect fundamental freedoms but which prey on their citizens by stealing from them. These regimes perpetuate the subordination of the citizen to the government and prevent the achievement of general prosperity—usually for the sake of the politicians' own power and enrichment. It is unclear how a well-functioning liberal republic should deal with them.Footnote 23
Not cooperating with tyrants is the right position to take in an ideal world. However, the real world is not so simple. I indicated at the outset that in foreign relations one must sometimes settle for the second-best solution. If making an alliance with a tyrannical regime is the only or the quickest way to defeat an implacable enemy, as it was in World War II, then it is morally permissible for a liberal government (reluctantly) to do so. But notice that this position—that a liberal state may ally itself with a tyrannical state if the alliance is indispensable for the successful defense of freedom (often not only the freedom of its own citizens but the freedom of millions of others)—is very different from the standard realpolitik view. According to the latter, a liberal government should be free to forge alliances with tyrants whenever such alliances advance the liberal state's national interest. This is a much more relaxed standard than the one I accept here, and I believe it should be rejected. Alliances with tyrants should be the exception, not the rule. In normal times, the liberal state should build alliances with fellow democracies, but this requirement is relaxed when survival is at stake.
C. Promoting liberty globally
A secondary goal of liberal foreign policy is to promote liberty worldwide. This goal is surely more controversial than the first, but I believe that, properly qualified, it follows from a commitment to liberal principles. The argument starts from the premise of the equal moral worth of all persons. I do not conceive this moral equality in the strong sense that we owe the same duties of justice to all persons around the globe.Footnote 24 Rather, for purposes of this essay, I endorse a weaker form of cosmopolitanism: every human being is entitled to live under just institutions regardless of history and tradition.Footnote 25 This means that every person has basic rights against others, and especially against the government. As I indicated earlier, by “basic rights” I mean the traditional rights recognized by classical liberalism: rights to life, security, speech, property, and the like.Footnote 26 If this assumption is accepted, liberal governments have a conditional duty to promote liberty globally.
One needs to tread with care, however (which is why I use the word “conditional”). A main objection to my suggestion that the liberal state has a (conditional) duty to promote liberty globally is that a liberal society is the result of consent in some form or other, the outcome of a social contract.Footnote 27 The social contract determines the governing principles of a liberal society and empowers the government to protect and implement those principles, and (perhaps) to solve genuine market failures that unrestricted voluntary relations may yield. Like all contracts, the social contract binds only the parties, and the government's fiduciary duty is to protect and implement liberty for its own citizens. On this view, a liberal government has no business promoting the freedoms of persons who are not parties to the contract. It must devote its energies and resources to its domestic mission. Notice that this view (call it “contractarian”) is stronger than it seems at first blush. It coheres with the cosmopolitan assumption of the equal worth of all human beings. Contractarians may agree that every human being is entitled to liberty, but will point out that it is someone else's duty to secure it. A foreigner is entitled to liberty, but the obligated party is his own government. Moreover, the contractarian view sits well with the liberal commitment to limited government. A government has no independent moral standing; it is a mere agent of the people. Just as I expect my lawyer to defend me and not go around implementing justice for everyone, so we expect our government to promote our freedom and not go around promoting the freedom of others to whom it has no fiduciary obligations.
Contractarianism is certainly appealing and correct as far as it goes. However, not all our moral obligations flow from the social contract. This is a foundational issue in moral philosophy, and I cannot address it fully here. But I will say this much. In addition to contractual obligations, we sometimes may have obligations that stem from the role or position we find ourselves in. If my assumption that all individuals around the globe have a right to live under just institutions is true, then I have an obligation (within reason, and conditioned on cost and on the fulfillment of other more pressing obligations) to make my small contribution to realizing that goal. Because isolated individuals cannot (or can hardly) act effectively in this regard, a liberal government seems to be a logical tool to help in the realization of universal freedom. This is true for both big and small governments. In other words, citizens in a liberal state delegate to their liberal government the responsibility to help achieve a more just world, to do the right thing internationally (with due caution and within the limits of the fiduciary relationship). In some extreme situations, one could also argue that the liberal government, which alone commands the resources of the state, has an obligation to stop genocide in a neighboring country, even if there is no evidence that the citizenry may have delegated such a power. This would be an obligation generated by the government's role, not by delegation: the government is the only entity in a position to stop the genocide.
Yet contractarian concerns are powerful enough to refute strong cosmopolitanism. The contractarian argument successfully blocks the view that we owe all persons around the globe the same duties and that there is no moral difference between compatriots and foreigners. It would be unacceptable, for example, to promote the freedom of foreigners at the expense of domestic freedom. However, I do not think the contractarian argument blocks the weaker cosmopolitan approach I endorse here. In most cases, my weak cosmopolitan approach is compatible with the contractarian conception of government, because the duty to promote liberty globally is a secondary duty. It takes a back seat to our duty, and the government's fiduciary duty, to defend liberty at home.
Here we may distinguish three scenarios. The first occurs when liberty at home is endangered. In that case, the liberal government has a quasi-absolute duty to protect liberty at home and defend it against foreign threats, even if doing so threatens the rights or interests of foreigners.Footnote 28 In this case, national defense—concern for the state's own citizens—displaces most (but not all)Footnote 29 concern for foreigners.
The second scenario arises when liberty at home is secure, yet many people abroad are unfree. In that case, the liberal government, I suggest, has a duty to help those persons, provided that the expenditure of resources and the costs to its own citizens are acceptable. Thus, for example, a liberal government may have the obligation to strengthen international institutions that will address the dire situation in Darfur, but perhaps it may not permissibly use military force to liberate the people there if the cost of doing so is prohibitive. Promoting liberty abroad is a conditional duty. It is subject to the primacy of the government's first duty, to defend liberty, and it is conditioned on cost. Under normal circumstances,Footnote 30 liberal governments will discharge their obligation to promote liberty globally by diplomatic and other measures in support of liberal constitutions abroad and liberty-friendly international institutions.
The third scenario arises when domestic liberty is secured and the liberal government attempts to advance, internationally, the interests of its citizens. Surely it can do so, even if the action does not improve the liberty of foreigners. Advancing the interests of its citizens is, of course, a major permissible goal of government. This the liberal state normally does through international competition for resources: competitive harm to foreigners does not give rise to legitimate grievances. But suppose that actions taken in the national interest significantly undermine the freedom of foreigners. Imagine that a liberal government wants to import good, inexpensive coffee in order to improve the welfare of its coffee-loving citizenry. This requires securing arrangements with an oppressive regime that enslaves its own people to produce cheap coffee. While the welfare of the citizens in the liberal state improves, I think the arrangement is morally impermissible. A liberal government should not enable a tyrant and contribute to the oppression of his subjects merely to increase the welfare of its population. This is a corollary of the general obligation not to cooperate with tyranny mentioned above. This may be a genuine difference between the contractarian approach and the cosmopolitan approach suggested in this essay. Contractarians are right that liberal governments should not try to remedy every wrong in the world, but cosmopolitans are right that a liberal government should not cooperate in the violation of the liberty of others unless this is necessary to protect its citizens and institutions against aggression.
Here, as elsewhere, it is crucial to be attentive to consequences and incentives. The duty to promote global freedom should not be discharged using counterproductive foreign policy tools. All too often, liberal governments understand their global mission in symbolic terms. Those in government are politicians, and foreign policy rhetoric suffers from the well-known pathologies of political rhetoric. Government officials will often say things that will advance their electoral goals even if doing those things would be ineffectual or counterproductive. Yet symbolic behavior and cheap talk are not good substitutes for actual results. Consider economic sanctions. It is an entirely empirical matter whether or not an economic boycott will promote liberty in the target country or elsewhere. For the politician interested in electoral success, however, it is important to appear to be doing the right thing. For example, when the U.S. president imposes an embargo on a dictatorial regime, he appears to be “tough” on the regime, even if, as an empirical matter, the embargo worsens the situation of the regime's victims.Footnote 31
D. The defense of liberty must not be self-defeating
Because the paramount foreign policy goal of a liberal republic is the defense of liberty, a liberal government may not defend liberty against foreign aggression in a way that destroys or seriously impairs the very liberty that it aims to preserve. This means that restrictions on domestic liberty are justified only for the sake of preserving liberty itself. This may sound obvious; however, the issue has been obscured by recent debates in connection with the war against terrorism. Many people have posed a dichotomy between liberty and security, thus suggesting that these are different, competing values. An increase in security may lead to a decrease in liberty and vice versa, so the solution, it is thought, is to strike the right balance between these competing values. As I have argued elsewhere, this dichotomy is misleading.Footnote 32 The security measures in a liberal republic must themselves be liberty-enhancing. The purpose of increased security measures (such as wiretapping) can only be the protection of our liberty and liberal institutions. It follows that restrictions on liberty that do not make liberty more secure are unjustified. To be sure, we sometimes cannot tell whether a particular security measure is or is not justified as a liberty-enhancing measure. Is the suspension of habeas corpus for Guantanamo detainees justified? Does it really help the government prevent future terrorist attacks? Or are losses of liberty and losses of faith in institutions too great to justify the intelligence gains? The answers to these questions depend entirely on the facts.Footnote 33 From the standpoint of principle, whether or not a particular measure meets the test depends on whether or not there are absolute moral restraints on government action, regardless of consequences. For example, all liberals will quickly agree that the liberal state must never torture prisoners, including war enemies.Footnote 34 Yet even here people disagree about what torture is, not to mention the possibility that in extreme situations, as moral philosophers well know, the absolute prohibition may break down. (For example, is torture prohibited even if not torturing a prisoner will bring about total annihilation?)Footnote 35 So the initial “never” turns out to be not so emphatic. I do not think the disciplines of law or philosophy have an answer for this, and I don't have one either. Perhaps all we can say is that the liberal constitution should not authorize torture, ever, and that whatever departures may occur in times of supreme emergency belong to the realm of moral necessity (or moral tragedy), not the realm of law. Torture, in other words, can never be legally protected. Be that as it may, my point is conceptual: security is liberal security. Only the need to protect liberty justifies restrictions on liberty. If those restrictions reduce liberty overall, then they are impermissible.
III. The Foreign Policy of a Liberal Republic: Prosperity
Another legitimate goal of a liberal foreign policy should be to enable citizens to seek material prosperity. This means that constitutions should have the tools to enable market agents to engage in voluntary transactions for mutual benefit and thus prosper. Domestic economic policy is centrally important here. A liberal constitution should protect robust rights of property and freedom of contract mainly (though not only) for empirical reasons: property rights and freedom of contract are the engine of material prosperity. Reliable economic research shows a strong correlation between robust property rights and the absence of poverty. Conversely, nations where property and contract are unprotected are much more likely to be poor. The ideal constitution must be consistent with these findings and must establish strong protections for property rights and freedom of contract.Footnote 36
In foreign affairs, this means that the constitution must facilitate free trade. The law of comparative advantage is the most reliable analytical tool to ascertain the effects of trade. While not a sufficient condition for material prosperity, free trade is likely to be a necessary condition. Thus, constitutional rules must enable the government to open domestic markets and promote the opening of foreign markets. In particular, constitutional norms should prevent special interest groups from using the political process for protectionist rent-seeking. Each economic agent has an incentive to capture the power of government to shield him from competition, foreign or domestic.Footnote 37 This political pathology impairs prosperity in the name of (at best) false mercantilist assumptions.Footnote 38 It follows that the conventional view that constitutions should be neutral with respect to economic policy is mistaken.Footnote 39 Sound economics shows that a market-friendly economic policy is necessary for the prosperity of the citizens in a liberal democracy. Just as the constitution should protect citizens against government abuses by having a robust bill of rights, so should the constitution protect citizens against this other form of governmental abuse: the capture of the government by special interests seeking a shield against competition. Protectionist policies demonstrably harm domestic consumers, that is, the citizens of the state at large. The constitution should protect consumers because their organizational costs are too high, and thus they are likely to lose systematically in the political process.
My approach avoids the American debate about whether economic policy should abide by the classical-liberal beliefs of the Framers of the Constitution—the so-called Lochner debate.Footnote 40 Classical-liberal constitutional writers emphasize the fact that the United States was founded in accordance with classical-liberal convictions—most notably, commitments to strong property rights and freedom of contract. In their view, the modern administrative state violates this original intent.Footnote 41 Supporters of economic regulation either contest this historical account or claim that it is irrelevant in modern times. I do not pass judgment on the strength of the historical argument. My view here is simpler and (I would hope) less controversial. As such, it should be able to persuade those who deny or eschew original intent, as long as they are sensitive to empirical arguments: a liberal constitution should protect economic liberties because reliable economic theory and empirical studies show that economic liberties are conducive to prosperity. The evidence on this is overwhelming.Footnote 42 Thus, someone who does not care for original intent, yet believes that economic prosperity and the reduction of poverty are worthy goals,Footnote 43 and honestly pays attention to plausible empirical facts and theories, should be sympathetic to my argument. If she is not, she has the burden of challenging the empirical theories on which my argument rests. Moreover, originalism applies by definition to a historical, actual constitution. In this essay, I am concerned instead with constitutional design, that is, with writing a constitution from scratch, given liberal values and what we know about the world.
IV. From Principles to Constitutional Norms: The Law of Foreign Relations
A. War powers
Assuming, then, that the defense of liberty and the promotion of prosperity are the proper goals of foreign policy, what are the consequences for the foreign-relations provisions of the liberal constitution? Given the priority of defending liberty, the constitutional norms of foreign relations should facilitate combat readiness while remaining consistent with basic liberal political principles. The standard position in American constitutional law is that Congress authorizes war, while the president conducts war.Footnote 44 Article I, section 8 of the U.S. Constitution empowers Congress to authorize war,Footnote 45 while Article II, section 2 empowers the president, as commander-in-chief, to conduct war. According to conventional doctrine, this commander-in-chief power comprises both the power to direct the conduct of war and to repel sudden attacks.Footnote 46 There is considerable scholarly controversy on a number of points. First, short of an official declaration of war, it is uncertain what actions undertaken by Congress constitute an authorization of war. Second, it is unclear whether military commitments short of war require congressional authorization. Third, the line between the initiation of military action and the direction of military action is blurred. Finally, it is uncertain whether or not the use of military force to defend an ally who is attacked requires congressional authorization. Here practical reasons of combat readiness are crucial. It stands to reason that the ideal constitution should facilitate the effective use of force, one that is optimally likely to lead to military success. But the constitution cannot decide in advance the merits of any particular military commitment. All it can do is to establish a process by which the government can effectively implement a morally justified foreign policy. Thus, if defending an ally is justified under the defense-of-liberty rationale, the constitution should make that military commitment easier, not harder. If the suddenness of the attack (e.g., against the U.S. or an ally) requires immediate response, then the executive branch should be able to act on its own.
Nonetheless, in cases where there is time for deliberation, there are powerful reasons to require the participation of the people's elected representatives. It may be thought that this concurrent war power can be inferred from the general doctrine of separation of powers, an essential component of the liberal constitution. In the oft-quoted words of Montesquieu: “When the legislative and executive powers are united in the same person, or in the same body of magistrates, there can be no liberty; because apprehensions may arise, lest the same monarch or senate should enact tyrannical laws, to execute them in a tyrannical manner.”Footnote 47 The justification for the separation of powers into distinct branches, then, is to maximize liberty by having the person who makes the law be someone other than the person who executes the law. But does this justification carry over to war? Authorizing war is not exactly like legislating, and making war is not exactly like executing the law. In spite of this asymmetry, I think that Montesquieu's point applies to war as well: A tyrannical prince may declare an unjust war and then wage it in an immoral manner. Moreover, the liberal constitution must restraint a president who might lead the nation to war for his own political gain and self-aggrandizement. This adventurism is more likely to occur if there is no legislative control over the decision to wage war. The usual checks-and-balances rationale for the separation of powers applies to war as well, provided the constitution establishes a mechanism to respond to threats when there is no time for deliberation.Footnote 48 As Immanuel Kant suggested, the people should decide whether or not they want to incur the costs of war,Footnote 49 and this is best achieved by placing the decision with the legislature rather than with the executive. To be sure, legislatures will often make mistakes, but the democratic process, imperfect and flawed as it is, is preferable to the unilateral decision of the president. Finally, a successful war requires popular support, as Americans well know, so congressional approval (assuming, again, that events allow the time to seek it) will not only lend legitimacy to the war effort, but will also enhance the chances of victory.
The paramount goal of the constitutional regulation of war and the armed forces, then, is the defense of liberty. This means that this goal takes precedence over other worthy goals. For example, on this view, whether gays or females are allowed in combat is entirely a function of combat readiness. If allowing gays or females in combat enhances combat readiness, then they should be allowed. If not, they should not. To use the armed forces as an arena for equality of treatment in disregard of combat readiness is a serious mistake, as it may lead to military defeat. Conversely, excluding gays or females from combat purely on the basis of prejudice without any attention to combat readiness is irrational and dangerous. The point here is that the proper balance of constitutional considerations should be tipped in favor of military efficiency because in these urgent situations only military efficiency serves liberty well. Once we shift our attention back to civilian life, the standard equal-protection principles will reemerge. The ideal constitution, therefore, should be designed to facilitate combat readiness as a priority over other goals, assuming of course that basic liberties are not undermined.Footnote 50
B. The status of international law in domestic law
Another lively controversy among constitutional scholars concerns the proper status of international law in domestic law. Internationalists advocate a maximal incorporation of international law into the U.S. legal system. Exceptionalists, in contrast, are much more cautious and distrust this wholesale importation of international norms into the domestic system. What should an ideal constitution do? I will discuss treaties and customary law separately, because they raise different questions. On the one hand, treaties are written documents to which governments expressly consent. For that reason, they blend more easily into the constitutional structure. Customary international law, on the other hand, is unwritten and (at least in its classical version) formed over relatively long periods of time by convergent state behavior. Given its relatively amorphous nature, customary international law blends uneasily into the constitutional structure.
1. Who should make treaties? Treaties are an indispensable part of diplomacy. A liberal republic needs treaties for all the obvious reasons. Not even the most rabid isolationist would deny this, and most constitutions have mechanisms to incorporate treaties into domestic law. What should the ideal constitution say about treaties? An initial question is who should be empowered to make treaties. In the United States, the president makes treaties, subject to the approval of two-thirds of the Senate.Footnote 51 It may seem that this procedure is as good as any, but on further inspection this is not so. This peculiar provision means that the treaty-makers are different from the lawmakers. The lawmakers are the two chambers of Congress and the president (including the possibility of Congress's overriding a presidential veto). The treaty-makers are instead the president and the Senate; the House of Representatives does not participate in the treaty-making process. This divergence creates an anomaly: It makes the treaty process vulnerable to the charge that it may be used to bypass the legislative process. To the extent that the lower chamber of the legislature is essential to conferring legitimacy on the law, the American treaty-making process appears somewhat defective. Moreover, it is hard to see the rationale of this provision. There is no compelling reason why an ideal constitution should have a different procedure for treaty approval. If treaties are to be the law of the land, they should be approved by a procedure similar to the one used for ordinary legislation. In that way, debate about a particular treaty will be a debate on the merits, and not a debate about whether or not the government may change domestic law through treaty. The change of domestic law through treaty, then, should be subject to the same strictures as the change of law through legislation.
2. Should treaties be self-executing? American constitutional practice has established a distinction between self-executing and non-self-executing treaties.Footnote 52 A self-executing treaty is one that can be invoked by private parties in court. Such a treaty is operative without the need of legislative implementation. In contrast, when a treaty is non-self-executing, private parties may not invoke it in court. Non-self-executing treaties require further legislative implementation before they can be invoked by private parties in court. The issue is important in the United States because there are a number of treaties to which the United States is a party that are not available to litigants.Footnote 53 This creates an anomaly: the United States is bound internationally by the treaty, but its citizens cannot use it domestically to enforce whatever rights the treaty may grant them. How do we know when a treaty is or is not self-executing? There are some clear cases, as when the Senate declares the treaty to be non-self-executing.Footnote 54 But when the Senate is silent, courts have had some trouble making the distinction.Footnote 55
Fortunately, our analysis here need not be embroiled in this characteristically American debate.Footnote 56 There is no reason why the ideal constitution should require an additional formal legislative implementation of treaties in order to make them the law of the land. In our ideal constitution, treaties are approved by both legislative chambers and thus have the same status as domestic legislation. Legislative approval of a treaty, then, is its implementation. As I have indicated, in the current American system the body that implements a treaty, Congress as a whole, is different from the body that approves the treaty (the Senate) and from the official who makes the treaty (the president). Thus, further implementation represents an affirmation of legislative will beyond the prior act of Senate approval. But if we get rid of the uniquely American method of Senate approval, the issue of self-execution becomes much less important: when Congress approves a treaty, it normally implements it. Now, although I have said that the issue of self-execution becomes less important with the procedure I have proposed, the issue is not dead. For, in my proposed system, the legislature, when approving a treaty, may decide (for whatever reason) that it should not be self-executing. That is, the legislature may want the treaty to bind the state internationally, yet may not want the treaty to be used by private litigants in domestic courts. (The legislature may not want to create a “cause of action,” as lawyers say.) And finally, some treaties by their very nature do not lend themselves to application by courts, and in that sense they may be “naturally” non-self-executing: this may be true, for example, if the treaty, by its very terms, cannot possibly be understood to generate private rights that could be enforced in American courts. But a treaty consented to by a liberal government and approved by its legislature fulfills the condition of legitimacy required for all lawmaking. The treaty has been consented to by the representatives of the people, and there is no reason in principle why the treaty should not be operative as a matter of domestic law (unless the legislature expressly blocks this).Footnote 57
There are some caveats, however. The liberal constitution prevails over any treaty.Footnote 58 Treaties may not violate the constitution. Consider human rights treaties. If the liberal government consents to a treaty that restricts constitutional rights, then the courts should declare the treaty provision unconstitutional. Judicial review is a proper safeguard against inappropriate “importation.” For example, if a treaty obligates the liberal state to restrict speech in violation of the constitution, the courts should declare the treaty unconstitutional.Footnote 59 But there is no objection, in principle, to expanding liberty through treaty. Someone may object that the expansion of the freedom of some is always done at the expense of the freedom of others. Imagine, for example, that the government signs onto a treaty that establishes universal health care for all the parties to the treaty. This would surely be illegitimate, it may be argued, since the government would be bypassing the legislative process. But, again, this objection is entirely parasitic on the fact that in the United States, as we saw, the treaty-makers are the president and the Senate, to the exclusion of the House of Representatives. As I have argued, an ideal constitution establishes treaty approval by the whole legislature, so the law created by treaty would be as legitimate as a law created by legislation. This would avoid the objection we are considering. Objections to universal health care established by treaty would then be based on the merits, not on considerations of process, since the process for treaty approval would be virtually the same as for legislative approval.
Of course, there is no way a liberal constitution can prevent a government from importing bad treaties, just as it cannot guarantee that the legislature will not enact bad laws. All we can hope is that the constitution will prevent the importation of some bad treaties and the enactment of some bad laws. The constitution can achieve this by establishing a robust bill of rights (including strong protection of property rights) and limits to governmental powers. In terms of importing international law, the doctrine of constitutional supremacy will hopefully block bad treaties in many cases.
3. Customary law. Customary international law is a different story. According to classical doctrine, international customary law has two components: state practice and opinio juris. State practice is the convergent behavior of governments over time. Repeated acts by governments that meet with acquiescence or approval by other governments may generate rules of behavior. After a certain time, the conduct in question (that is, the act performed plus the acceptance by others) may become customary law. However, not all practice ripens into law. According to traditional doctrine, state practice is a necessary but not a sufficient condition for the existence of a customary legal rule. There are certain practices that are considered merely optional, such as the rules of international etiquette. In order to distinguish nonbinding international practice from binding customary international law, lawyers have long required that, in order to be law, the practice must be accompanied by a belief by governments that the practice is required by law. This requirement, called opinio juris, has puzzled many commentators for a number of reasons, of which I shall mention only one. Must governments have as a reason for their behavior the belief that the behavior is required by law? This seems implausible, for governments act for all kinds of self-interested reasons—reasons of national interest—unrelated to their perception of the legal status of the act in question.Footnote 60 More plausible is the idea that at some point in the life of the practice, participants begin to insist on compliance. Practice becomes law when supported by social pressure.Footnote 61 This is a better view because it does not require that governments act on a sense of legal obligation but simply that governments publicly articulate the rule as a legal rule, thus insisting that all governments must comply with the rule.Footnote 62
This is the classical definition of customary law, under which state practice is a necessary condition for the formation of a legal norm. According to some scholars, however, modern customary law includes, in addition to the classical rules, rules derived from “soft” materials such as United Nations resolutions, provisions in unratified treaties, judicial decisions by national courts, statements made by governments at international conferences, and the writings of law professors.Footnote 63 On this view, state practice is not a necessary condition for the existence of a customary international rule: sometimes, depending on context, importance, and other factors, a norm may become part of international law even if it is unsupported by state practice. Typically—but not always—such norms will be supported by a wide international consensus.
Internationalists favor the incorporation of all customary law (classical and modern) into domestic law. To them, domestic courts may use customary law both to decide cases and to interpret domestic law.Footnote 64 Exceptionalists challenge this view. To them, customary law suffers from a variety of infirmities that should preclude its use by domestic courts in the absence of formal incorporation by the legislature.Footnote 65 John McGinnis and Ilya Somin have offered a forceful defense of the exceptionalist position.Footnote 66 They argue that U.S. courts should not treat “raw international law” as the law of the land. By “raw international law” they mean international law that has not been formally incorporated into domestic law by the authorities designated by the Constitution. A properly ratified treaty meets the condition of formal incorporation, because the Constitution prescribes a procedure for domestic assimilation. Likewise, Congress can enact laws implementing parts of customary law, as it did with the Torture Victim Protection Act (1992).Footnote 67 However, McGinnis and Somin argue that, aside from these cases of formal assimilation, customary international law should be kept out of domestic courts because it suffers from a crucial legitimacy deficit, which in turn causes a host of other problems. In particular, customary law, because it is unwritten, is particularly vulnerable to manipulation by advocates and other agents of special interests, such as law professors. Customary law is made, not by the people's representatives, but by others who do not represent the individuals against whom the law is domestically enforced. Customary law does not, therefore, satisfy a minimal pedigree condition: that the law, to be binding, must be created by the people in accordance with democratic constitutional procedures. McGinnis and Somin intend their argument to apply to the United States, but it can be generalized. The law in a liberal state is created by the people, who alone can delegate lawmaking power. To be binding, a law must not be merely imposed by persons who lack delegated lawmaking powers (whether they are domestic dictators or foreigners). Customary international law does not satisfy this condition; therefore, it is not binding on citizens.
The exceptionalist position is buttressed by a sobering fact: The world does not consist only of liberal states. The participation of many illiberal states in the international lawmaking process blights customary law. Not only do the lawmakers often lack proper democratic credentials, but also some customary rules are unfriendly to liberal principles.Footnote 68 Even if we think that the liberal state may have to live with some of those bad rules internationally, it would be an affront to justice for domestic courts to foist those rules on the liberal state's citizens. This is another reason—a substantive reason—to be wary of importing customary law. Customary law is always undemocratic and sometimes morally objectionable on the merits. It is procedurally and substantively deficient, and for that reason, exceptionalists argue, it should not become the law of the land.
There is much that is right about McGinnis and Somin's argument. If customary law were little more than an invention of partisan advocates and biased academics, then McGinnis and Somin's argument would be a powerful indictment of the attempt to import it into domestic law. I agree with them that much of what passes as customary law should be unmasked for what it is: manipulative advocacy. However, I think that with the appropriate guarantees, some rules of customary law can be imported by the courts of a liberal state in a way that assuages these authors' concerns.
I start by noting that what international lawyers present as customary law varies widely. One distinction, already anticipated, is between classical and modern customary law.Footnote 69 Classical customary law, as we saw, is supported by robust international practice. Diplomatic interaction has yielded a number of rules over a relatively long period of time. Those rules have emerged as the result of behavior aimed at solving various coordination problems.Footnote 70 Once a practice has emerged, we may assume that it is irrational for participants to deviate from it. Opportunistic deviation is punished with criticism, ostracism, or stronger action. This pressure to conform characterizes opinio juris. Classical customary law, then, is characterized by robust interaction over relatively long periods of time, supported by pressure to conform.Footnote 71 Modern customary law, as I indicated, is characterized by the use of soft materials such as United Nations resolutions, provisions in unratified treaties, pronouncements by officials and nongovernmental organizations, and scholarly writings. Notably, the rules thus advocated lack the support of state practice. As a result, they are less certain and more vulnerable to manipulation than classical customary law. We might say that modern customary law is strong in the opinio juris department, but weak in the practice department. Norms of modern customary law may have strong support in the opinion of international actors, even if this support is not accompanied by a long history of practice. Some human rights norms belong to this group.Footnote 72
McGinnis and Somin claim that neither kind of customary law should be eligible for judicial importation. To them, domestic courts should not use modern customary law because it is not supported by state practice. For that reason, and because (like any customary law) it is unwritten, it cannot be determined objectively and is especially subject to manipulation. While classical international law fares better (because it is supported by state practice), it still cannot be determined objectively because it is unwritten. As a result, law professors and others with no authority get to say what the law is. In either case, customary law suffers from a fatal legitimacy deficit, and domestic courts should not use it to determine their decisions.Footnote 73
However, I think the classical/modern dichotomy, while helpful and accurate as far as it goes, still does not capture the diversity of what international lawyers consider customary law. I propose to classify (what international lawyers call) “customary law” into genuine customary law and fake customary law. Genuine customary law comprises two subcategories. First, genuine customary law includes classical international law as already defined: a robust practice supported by strong pressure to conform (opinio juris). Second, genuine customary law includes a real, strong international consensus, even if it is not supported by a robust practice.Footnote 74 This second subcategory corresponds to a genuine, nonmanipulated version of modern customary law. Genuine international law includes, then, classical international law plus those rules that have been universally agreed to by nations, that is, a subset of modern customary law.
Fake customary law, in contrast, consists of those putative rules that are put forth by international lawyers and others but lack the pedigree of genuine custom just described. They are rules that international lawyers would like to be genuine rules of law but that in reality are not. Lawyers (in good or bad faith) assert these rules as law, but they are just pieces of advocacy. A rule of fake customary law is characterized, not by state practice or strong consensus, but by advocacy, partisan rhetoric, and unsupported assertion. Unfortunately, this kind of bad legal reasoning is quite prevalent in international law circles. Practitioners of fake customary law label as “customary law” any rule they favor. Typically, they cite “soft” materials in support of the desired rule, in an effort to look like they are asserting a genuine rule. Moreover, they do not merely attempt to justify the rule on policy or moral grounds, while recognizing that it is not yet law. (This would be honest, but would deprive them of the rhetorical force inherent in legal language.) Fake customary law lacks the appeal of genuine law. Genuine customary law reflects the aggregate sense of right and wrong in the community that has been elaborated and developed over a relatively long period or embodied in a strong consensus. Genuine customary law embodies equilibrium of some sort. In contrast, the practice of fake customary law misses all that. It misses what is most important about custom: its ability to reflect institutional history and filtered state expectations about what is lawful, mutually beneficial, ethical, or efficient. The practice of fake customary law has unfortunately degraded international law in the eyes of the general public, the courts, and the conventional legal community. To make things worse, some international courts have spread fake customary law by accepting it in their decisions.Footnote 75 While it is unclear how this trend can be modified in the international arena, for our purposes fake customary law lacks the necessary pedigree to be considered international law at all—let alone to become part of domestic law in a liberal state. The domestic courts of the liberal state should firmly reject this practice.
My initial proposal, then, is that the courts of the liberal state should only import genuine customary law. Yet even genuine customary law might be problematic. The international communityFootnote 76 may strongly agree on rules that are unacceptable to the liberal state, so the principle for importing customary law should be restricted further. My amended proposal is that the liberal constitution may allow courts to use rules of customary law as the law of the land, provided (i) that they are genuine and not fake, (ii) that they are consistent with the liberal constitution, and (iii) that they do not displace duly promulgated domestic laws. It is my hope that provisos (ii) and (iii) will assuage some of McGinnis and Somin's concerns. Customary international law would be operative domestically only where there is no constitutional, statutory, or precedential rule to the contrary—just as Justice Horace Gray said in the 1900 Paquete Habana case.Footnote 77 Moreover, these provisos will weed out genuine customary rules that are inconsistent with constitutional guarantees of individual rights. The imported rules of customary law cannot be illiberal rules; they cannot be rules unsupported by state practice or strong international consensus; and they cannot prevail against domestic legislation.
But someone may still ask why the courts of the liberal state should accept any customary rule, genuine or fake, that has not been formally incorporated. In other words, I have not yet given a reason in favor of domestic assimilation. I think, however, that with the foregoing guarantees in place, there are good reasons for treating genuine customary international law as part of the domestic law of the liberal state. Good customary rules—that is, those that are genuine and consistent with liberal principles and domestic legislation—permit international coexistence and thus enable the liberal state to prosper in the global arena. A necessary condition to preserve a liberal society is having a reasonably peaceful coexistence with other nations of the world. This coexistence increases the chances that the liberal state will preserve and protect its liberal institutions. Genuine customary law is an important part of international law that contributes to this coexistence, precisely because the rules have withstood the test of time or consensus. Now, some of those rules are not the kind of rules that can be naturally assimilated into domestic law. Others, however, can be naturally assimilated. The courts in the liberal state can use well-developed doctrines of standing, cause-of-action, and the like, to determine which customary rules can naturally be used as rules of decision in domestic cases. Courts can, for example, allow customary rules that specify rights and duties of individuals, and disallow those that do not do this.Footnote 78
Consider the customary rule of international law that exempts diplomats from local taxes. Imagine that the government of a liberal state tries to collect a tax and the diplomat argues in court that he is immune to taxes under customary international law.Footnote 79 McGinnis and Somin would deny this immunity, since the rule invoked is “raw” international law. It has not been enacted by the legitimate legislative process or by the process of treaty implementation. However, I think that a liberal state can accept the importation of the rule of diplomatic immunity, even if the rule lacks formal democratic credentials. This is the price that citizens pay to secure the coexistence that makes the flourishing of the liberal state possible in the global arena. A citizen in a liberal state understands that, in order to coexist peacefully, the state's government should not disregard genuine customary rules pertaining to diplomatic immunity. In this way, importation of genuine customary law is indirectly linked to the preservation of liberal institutions and thus substantively justified, even if not formally enacted by the legitimate domestic lawmaking bodies. In the example discussed, a customary rule affording diplomatic immunity requires domestic importation if it is to be effective. The alternative is for the liberal government to turn around and say to other nations that it cannot honor the rule because it has not been duly incorporated into domestic law. But doing this defeats the purpose of the rule and thus creates unnecessary costs for the liberal state in its international relations. Designers of the liberal constitution can anticipate that it will be in the interest of the liberal state to assimilate those rules of genuine customary law that are consistent with liberal principles and facilitate international interaction. Another way of looking at it is this: a rule's democratic pedigree, important as it is, is not the only reason for accepting the rule. Importing a limited number of customary rules that meet the stringent conditions mentioned above does not do great violence to the principle of democracy and contributes to the health of the liberal state. Finally, if we look at the experience in actual liberal democracies, the incorporation of customary law does not seem to have disrupted or threatened liberal institutions.
A 2004 decision of the United States Supreme Court, Sosa v. Alvarez-Machain, illustrates the approach I suggest here.Footnote 80 In that case, the Court wisely held that any use of the law of nations (i.e., customary international law) for purposes of domestic litigation under the Alien Tort Statute must meet two conditions. First, the law invoked must be sufficiently determinate; that is, it should not consist of vague generalities but should instead clearly specify the rights and duties asserted by the litigants (the Court requires “definite content”). Second, the invoked rule must be universally agreed upon by the nations of the world (in the Court's words, “acceptance by civilized nations”). Notice that the Court does not require that domestically acceptable customary law should be supported by practice, but rather that it should be widely agreed upon by the nations of the world. The Sosa Court thus accepted not only classical customary law, but also genuine modern customary international law, as I have suggested here. This important holding about customary law (if extended beyond the Alien Tort Statute) establishes a kind of quality control (in the Court's words, “vigilant door-keeping”) for the domestic use of customary international law—a level of quality control that would exclude both fake customary law and genuine customary law inconsistent with domestic law or liberal principles.
To the extent that McGinnis and Somin's argument applies specifically to the United States, it has only a limited bearing on the topic of this essay, since my purpose here is to design ideal foreign-relations clauses. For example, McGinnis and Somin observe that the American public is rationally ignorant of international law, a fact which contributes to the democratic deficit. Yet that is an entirely contingent fact. The public in other liberal democracies seems to be better informed, and we should not assume ignorance of customary law in the design of ideal institutions. Moreover, the public is also rationally ignorant of domestic law, yet obviously this fact should not influence the courts' acceptance of that law. Also, McGinnis and Somin claim that American law might serve people (citizens and foreigners alike) better than international law, given the special position of the United States as a superpower. Yet that special position, important as it is, is also a matter of contingent fact. It is an open question whether or not the ideal constitutional rules should take into account superpower status. We want rules for a liberal republic that will endure regardless of its relative position of power in the world. The liberal state has an obligation to defend and protect its liberal institutions and to promote similar institutions elsewhere, regardless of its relative power.Footnote 81
Finally, I have suggested that customary rules that meet the proposed tests (i.e., that are genuine and consistent with domestic legislation and liberal principles) can safely be used by domestic courts. Does this mean that those customary rules should displace state laws in a federal system? Recall that my purpose is to explore ideal constitutional design for foreign relations. It is an open question whether liberal principles require a unitary constitution or a federal constitution. The arguments for and against federalism are well known: On the one hand, by diluting power geographically, the federal constitution maximizes liberty. On the other hand, local rulers can be oppressive, and if so, their victims could have a hard time finding relief in a federal state. At any rate, it seems to me that liberal principles do not mandate federalism. However, if the benefits of federalism are thought to outweigh its costs, then the issue of whether customary international law should displace state law will arise. It is well accepted in liberal states that properly ratified treaties prevail over state law. The reason for this is to ensure proper coordination in the conduct of foreign affairs. The same reasoning suggests that customary international law that meets the tests suggested here should generally preempt state law, unless the state law in question is securing a fundamental individual liberty in accordance with the federal structure.Footnote 82
C. Foreign commerce
Virtually all liberal constitutions grant broad governmental powers to regulate commerce with other nations.Footnote 83 The U.S. Constitution is no exception: Article I, section 8, clause 3 grants Congress the power to “regulate Commerce with foreign Nations.” It is undisputed that this constitutional provision allows Congress to erect trade barriers. This protectionist function was very much on the minds of the Founders. For example, Alexander Hamilton argued that the federal power to regulate commerce would allow the United States to use access to U.S. markets as “bargaining chips” to exact favorable commercial conditions. He wrote: “By prohibitory regulations, extending, at the same time, through the States, we may oblige foreign countries to bid against each other, for the privileges of our markets.”Footnote 84 The United States Supreme Court has held that this power is very broad: “[N]o individual has a vested power to trade with foreign nations which is so broad in character as to limit and restrict the power of Congress to determine what articles of merchandise may be imported into this country and the terms upon which a right to import may be exercised.”Footnote 85 Thus, in the modern post-Lochner era, free trade has suffered an even worse fate than property and contract.Footnote 86 On the view of the Supreme Court and mainstream constitutional scholars, just as in domestic cases “[t]he Fourteenth Amendment does not enact Mr. Herbert Spencer's Social Statics,”Footnote 87 so, it seems, the Constitution does not enact Mr. David Ricardo's theory of comparative advantage.Footnote 88
U.S. citizens have a de facto right to trade freely across states, that is, within the United States. The constitutional structure generally favors domestic competition.Footnote 89 Europe has adopted a similar structure (domestic free trade and foreign protectionism) but has done so more directly. The European Union Treaty expressly recognizes freedom to trade within the Union; that is, neither governments nor European agencies may impede the freedom of trade among Members.Footnote 90 But the Union has an almost unlimited power to enact protectionist laws regarding trade with non–European Union members (witness the European Union's extensive agricultural protectionism). Both in the United States and in Europe, these protectionist powers are checked only by these countries' obligations toward the World Trade Organization. Citizens do not have a right against their governments to trade freely with outsiders.
Allowing the government to erect trade barriers is a major constitutional flaw. Luckily, this anti-trade precedent does not bind the ideal liberal constitution. Constitutional design must incorporate reliable economics. If we do so, we learn two things. First, we learn that, because of the principle of comparative advantage, free trade increases prosperity in the long run, nationally and globally (even though those persons who would be favored by the segmentation of markets will lose out). Second, we learn that governments, including (and, some would say, especially) liberal governments, are vulnerable to capture by special-interest groups who do not wish to compete with foreign producers. Governments harm their own citizens when they succumb to this pressure, because they deny citizens access to better and cheaper goods. Moreover, protectionist measures are morally objectionable because they redistribute wealth for private, not public, purposes. Yet political mechanisms make it quite easy for protectionist interests to succeed. In view of this grim reality, I propose that the liberal constitution prohibit government from erecting trade barriers. The writers of the constitution should anticipate the pathology just described and should ban the government from harming citizens in this way. An analogy with individual rights may help us see this. Why does the liberal constitution prohibit the government from censoring speech? There are, of course, several reasons, but one reason is the knowledge that governments, even decent ones, are prone to silence their critics. No one likes criticism, and politicians like it even less, because criticism interferes with the exercise of power. Similarly, constitutional designers should anticipate that governments are likely to be captured by protectionist interests if the constitution grants government the power to enact protectionist policies. Therefore, rulers should not have that power.Footnote 91
V. Concrete Suggestions
Can the proposals put forth in this essay be codified in a written constitution? Because my concern is with the design of an effective liberal constitution, I am neutral about the best way to implement the foreign-relations provisions that I recommend in this essay. Depending on history and context, perhaps it might sometimes be best to draft only a brief constitutional list of the various powers and rights, and to allow the courts to interpret them in the way I recommend. However, experience has shown that words alone, even words solemnly enshrined in revered constitutional documents, have failed to deter the enemies of freedom and those who seek to enlist the coercive power of the government for private gains. Time and again, governments and courts have ignored or mangled clear constitutional language, even in countries with the best constitutional traditions. I can offer no satisfactory solution to this,Footnote 92 except to say that mere pieces of paper, even venerable ones, cannot preserve free institutions: Men and women must be prepared to defend them.
With that caveat, here are my suggestions:
1. The Legislature shall have the power to authorize all large-scale uses of force, subject to the following provisions.
2. The President shall be the Commander-in-Chief of the armed forces. This includes the power to direct combat and the power to repel attacks against the state's territory, the state's citizens, the state's institutions, and the state's foreign allies. In conducting hostilities, the President shall observe the relevant rules of domestic and international law pertaining to combat.
3> The President shall also have the authority to use force to stop a massive humanitarian catastrophe in another country, where there is no time to seek legislative approval.
War Powers
1. This Constitution, the laws made in pursuance thereof, and the treaties made by the President with approval by the Legislature by simple majority, are the law of the land.
a. Treaties shall be self-executing unless: (i) the Legislature, when approving a treaty, expressly indicates that it is not self-executing; or (ii) by its very nature, the treaty does not recognize individual rights or claims that can be enforced in a court of law.
2. The law of nations is also the law of the land, provided:
a. That it is consistent with the Constitution, laws, judicial precedents where applicable, treaties in force, and the principles of free government;
b. That it is sufficiently precise; and
c. That it is universally agreed upon by the community of nations.
Treaties and Customary Law
1. Everyone has the right to private property and the right to contract freely with others, including the right to trade freely with foreigners.
2. The government shall not interfere with these rights except:
a. In the interests of national security; or
b. When their exercise significantly harms others. For the purposes of this provision, “harm” does not include competitive harm.
Economic Liberties
VI. Conclusion
Many have seen the foreign-relations clauses of the U.S. Constitution as primarily concerned with practical functionality: what is the distribution of powers that will assure an effective, smooth conduct of foreign relations? I concede that this is very important, and that we should always learn from the historical experience of the functioning of liberal constitutions. However, in this essay I have tried to show that the content and structure of the foreign-relations provisions of a constitution are connected to the philosophical principles that underlie a liberal society. I have suggested that, over a broad range of issues, the foreign-relations provisions of a constitution can channel the implementation of liberal ideals in different ways, and that some ways are better than others. My project here is tentative, to be sure, and there are many avenues for future research. One of them is to connect the principles for the conduct of foreign relations by the liberal state with the principles of global justice. For example, analyzing the war powers of a liberal state presupposes a theory of just war; and proposing a particular design for the regulation of foreign commerce presupposes a theory of economic justice across borders. Another avenue for future research is to connect this project with the literature on the incentives created by constitutional design. In this essay, I have made no more than an educated guess about those incentives. Finally, the discussion about incorporation of international law into domestic law must be connected to a general theory of international law.
