Printer Friendly

Liberal democracy and cosmopolitan duty.

 A. From Individual to Institutional Duties
 B. The Relevance of Plausibility Constraints
 A. The Source and Significance of Weak Cosmopolitan Sentiments
 B. A More Realistic View of the Democratic Process
 A. Education
 B. Alternatives to Liberal Democracy
 C. A More Realistic Cosmopolitanism
 1. Institutionalism
 2. Instrumentalism
 3. Civil society
 4. National interest
 5. Double standards


The widespread enthusiasm for liberal democracy has many sources. Liberal democracy is the form of government most suited to liberal premises about the moral freedom of the individual. (1) It is the best political mechanism for controlling political agents. (2) It decreases the likelihood of internal armed conflict, and increases protection for other basic human rights. (3) It promotes international security because democracies rarely if ever go to war with one another. (4) It effectuates broader forms of international cooperation. (5) It may even be required by international law. (6)

It is against this background that I want to examine a prevalent form of criticism targeted at the United States--the world's most powerful and, in some respects, most vigorous liberal democracy. The criticism is that the United States acts wrongly, or unjustly, when it fails to take affirmative steps that would help other nations and their peoples. In particular, the United States is frequently criticized for failing to sign or ratify treaties (such as the Rome Statute creating the International Criminal Court) that many believe would increase the welfare of non-Americans, and for failing to intervene in countries (such as Rwanda) to prevent atrocities.

This criticism comes in two forms. One focuses on U.S. national interest, and maintains that the welfare of U.S. citizens would be enhanced in the fairer, safer, and more prosperous world that would result from increasing assistance to others. The basic claim here is that the United States harms itself and its citizens by not ratifying certain treaties and by failing to intervene more frequently and with greater intensity. I have no quibble with this form of argument, which in my view properly focuses on what is best for U.S. citizens, on leaders' information errors, on means-ends rationality (and related issues like unintended consequences), and on democratic-process pathologies such as time inconsistencies and interest group capture.

A second form of criticism focuses on the United States' cosmopolitan duties. It maintains that the United States should ratify global treaties and intervene more vigorously to stop human rights abuses, even if doing so would lower net U.S. welfare. This argument emphasizes that the United States should act to help peoples and nations outside the United States, even when the actions would not survive a U.S.-focused cost-benefit analysis. The argument does not try to clarify the U.S. national interest. It maintains that the United States should focus less on the interests of its own people, and more on the interests of all humanity.

The cosmopolitan duty argument plays a prominent role in recent cosmopolitan thinking in philosophy. (7) It also underlies the charge of U.S. unilateralism, especially when the purported unilateralist action is the United States' refusal to ratify prominent global treaties. More broadly, the cosmopolitan duty argument is frequently made in conjunction with the national interest argument, even though the two arguments are analytically distinct. Often, the national interest argument is motivated by, and is a cover for, cosmopolitan duty concerns.

This essay critiques the cosmopolitan duty argument. My thesis is that underappreciated theoretical, practical, and moral factors limit the duty of liberal democracies to engage in cosmopolitan action. The institutions needed to make liberal democracy work cannot easily generate cosmopolitan action. The problem is not just the absence of democratic support for cosmopolitan policies. Constitutional and collective action hurdles, and other difficulties, constrain cosmopolitan action as well. Cosmopolitan argument must be bounded by institutional and moral constraints that arise in the domestic-democratic sphere. We cannot even have a coherent ideal of liberal democracies' cosmopolitan duties unless we understand these realistic limits on what liberal democracies can do.

I should emphasize at the outset that I do not claim that States are never other-regarding. Such a claim might be wrong, and in any event would be difficult to prove. Nor do I deny that States or other collective entities can be subjects of moral responsibility. My argument is more modest. It is that institutional and related hurdles make it very difficult for liberal democracies to engage in strong (i.e., nontrivial) cosmopolitan action, by which I mean (a) entering into treaties that cannot be justified on local welfare-enhancement grounds, and (b) engaging in humanitarian intervention that is costly to the intervening state in terms of money and lives.

One response to my argument is that if liberal democracies cannot generate strong cosmopolitan action, then we must search for new forms of domestic governance to better serve the ends of international justice. I will examine this claim below. But it is worth noting at the outset that many cosmopolitan theorists avoid this move. One reason why is that plausible superior alternatives to liberal democracy are hard to fathom. Another is that cosmopolitan theorists tend to believe that liberal democracy is the optimal form of domestic governance. Indeed, some claim that there is a conceptual link between liberal democracy and cosmopolitan duty, and that "the domestic aspect of cosmopolitanism is ... liberal democracy." (8) On the whole, cosmopolitan theorists try to have it both ways, maintaining a commitment to liberal democracy and insisting that liberal democracies should act with greater cosmopolitan regard. My argument is that these co-commitments cannot easily be reconciled.

This leaves us in the messy but real world: a world of stark inequality and manifest injustice between nations and peoples that the least bad form of domestic governance (liberal democracy) seems incapable of rectifying in a manner, and time frame, that many would like. The Article does not criticize the cosmopolitan stance per se, however, and does not conclude on this pessimistic note. Rather, the Article suggests ways that cosmopolitan sentiments can be more fully realized by being more realistic.


The legal, policy, and philosophical literature is full of claims that the United States should act with greater cosmopolitan regard. The legal and policy literature rarely examines or defends the ascription of strong cosmopolitan duties to liberal democracies. The philosophical literature does, however.

A. From Individual to Institutional Duties

Cosmopolitan theory begins with the premise that every human being's life is equally valuable, regardless of group or national membership. Cosmopolitanism seeks to enhance attachments and duties to the community of all human beings, regardless of national or local affiliation, and to attenuate attachments and duties to the nation-state, fellow citizens, and local culture.

Some believe that cosmopolitan premises require relatively well-off individuals to assist relatively non-well-off individuals, including noncompatriots. (9) In recent years, however, cosmopolitan theorists have begun to reject (or attenuate) the ascription of strong cosmopolitan duties to individuals. They have begun to argue instead that these duties are best viewed as attaching to domestic institutions (e.g., national governments) and, derivatively, to international institutions. The main reasons for this "institutional turn" are that cosmopolitan duties are too demanding for individuals, and that institutions can better achieve international social justice. In short, cosmopolitan theorists use "plausibility limitations" on individual duties as a basis for ascribing cosmopolitan duties to political institutions.

Martha Nussbaum's recent work provides a clear example. Nussbaum claims that individuals are not capable of generating a "just global order through human psychology alone." (10) Why? Because human beings have an "imperfect and uneven" psychology, "emotions" that "link us to our own sphere of experience," and "critical skills" that are "often undermined from within by fear, haste, and selfishness, and subverted from without by misinformation, competition, and various forms of seduction." (11) In addition, "assigning responsibilities to people one by one is a recipe for a massive collection action problem, and for individual lives that are so consumed with figuring out who owes what to whom that there is no space left for either special duties or the enjoyment of life." (12)

Nussbaum concludes from these limitations on individual human action that "institutions have to be a large part of the solution," particularly "on the global plane." (13) She writes that "[p]olitical institutions that embody a moral ideal can coerce morally adequate results in the absence of even a single perfect human being." (14) Institutions are also better able than individuals to solve collective action problems, and to ensure "fairness in the distribution of burdens." (15) Nussbaum proposes that national institutions create the following international institutions:
 [A] world court that would deal with grave human-rights violations;
 a set of world environmental regulations, plus a tax on the
 industrial nations of the North to support the development of
 pollution controls in the South; a set of global trade regulations
 that would try to harness the juggernaut of globalization to a set
 of moral goals for human development ...; a set of global labor
 standards ...; and, finally, various forms of global taxation that
 would effect transfers of wealth from richer to poorer nations. (16)

Michael Green offers a similar line of argument. (17) Green contends that we cannot properly attribute cosmopolitan duties to individuals. He reaches this conclusion on the basis that "commonsense morality" (18) in the global context is impeded by the "phenomenological features of [individual] agency." (19) Three important features of commonsense morality are that individuals, and not groups, are the "primary bearers of responsibility"; that individuals have greater duties with respect to acts than omissions; and that individuals have "special obligations" and thus give priority to the near over the remote. (20) The commonsense conception of morality is a restrictive one that precludes the ascription of responsibility to individuals for the problems of global injustice.

Like Nussbaum, Green thinks that "institutional agents do not face the same limitations as individual agents." (21) Institutions are better at collecting and processing information. They have "power" and efficacy, and thus "can alter mass behavior." (22) And they can better spread the costs of action. These differences between the capacities of individuals and institutions justify attributing greater responsibilities to institutions. (23) For example, since "institutional agents are better able to perceive and act on their [sic] consequences of their omissions than individuals are," it makes sense to attribute less significance to the distinction between action and omission when we are attributing responsibility to institutions. (24) "Among other things, this means that "there is more room to hold government[] responsible for taking steps to regulate the harm its citizens collectively cause, even though it does not cause the harm itself." (25)

Iris Young also argues that obligations of social justice are primarily owed by institutions rather than by individuals. The reasons she gives should by now be familiar: "Individuals usually cannot act alone to promote justice; they must act collectively to adjust the terms of their relationships and rectify the unjust consequences of past and present social structures, whether intended or not." (26) Young proposes "a global system of regulatory regimes to which locales and regions relate in a federated system," and she suggests that "reform of the United Nations System is one reasonable goal" toward this end. (27)

Thomas Pogge similarly argues for institutional duties (or, as he calls it, an "institutional" as opposed to "interactional" conception of cosmopolitanism) as a way to redress the problem of imposing unrealistically demanding duties on individuals. (28) More specifically, Pogge proposes a global resource tax--a one percent consumption tax on all national resources that would be:
 used toward the emancipation of the present and future global poor:
 toward assuring that all have access to education, health care,
 means of production (land) and/or jobs to a sufficient extent to
 meet their basic needs with dignity and to represent their rights
 and interests against the rest of humankind: compatriots and
 foreigners. (29)

Other cosmopolitan theorists make similar arguments. (30) These arguments purport to be both realistic (as opposed to utopian) and consequentialist. They are realistic because they define and limit what justice requires by some notion of what is possible; "can" limits "should." In this way, the arguments strive to be relevant to modern policy debates. They are consequentialist in focusing on schemes that can produce good outcomes (from various perspectives of what constitutes goodness). These arguments ascribe duties to institutions with an eye to best achieving certain results (such as eliminating world poverty, ending global warming, etc.).

B. The Relevance of Plausibility Constraints

The cosmopolitan theories summarized above invoke five types of limitations on individual capacities as a basis for ascribing duties to institutions. The first is based on commonsense intuition. In rejecting the ascription of strong cosmopolitan duties to individuals, appeal is made to conceptions of human agency that are informed by our ordinary practices and intuitions. The second concerns limits grounded in human biology or psychology. Certain types of cosmopolitan duties--such as, for example, Peter Singer's version of utilitarianism (31)--make literally superhuman demands of calculation and concern. The third type of limitation is moral. Certain cosmopolitan duties are inconsistent with any reasonable conception of a good life, which must allow space for individuals to flourish without regard to the demands of morality, and especially without regard to the extraordinary demands of some cosmopolitan moral claims. A fourth concern is the problem of noncompliance. Some duties on individuals must be attenuated by the fact that others will not do their fair share. (32) Fifth, and relatedly, individuals often face severe collective action hurdles.

Why is it appropriate to invoke such limits in cosmopolitan argument?

The main answer is that political theory must, in Thomas Nagel's words, be "motivationally reasonable." As Nagel puts it: "If real people find it psychologically very difficult or even impossible to live as the theory requires, or to adopt relevant institutions, that should carry some weight against the ideal." (33) For similar reasons, John Rawls imposes plausibility constraints on the ideal (or full compliance) theory of justice that emerges from the original position. As Rawls puts it, an important consideration for ideal theory is "men's capacity to act on the various conceptions of justice," a consideration that includes "general facts of human psychology and the principles of moral learning." (34) These principles are relevant because in the original position, "[i]f a conception of justice is unlikely to generate its own support, or lacks stability, this fact must not be overlooked," for parties in the original position must suppose that other parties "will adhere to the principles eventually chosen." (35) Even when we consider nonideal (or partial compliance) theory, human frailty remains relevant. As Liam Murphy puts the point, "if our theory has implausible implications for the nonideal case, the theory may have some political interest, but it would fail as a normative political theory." (36)

Something like this reasoning underlies the invocation of human frailty as a basis for ascription of institutional responsibility. Any theory that aims to be realistic and consequentialist in the senses described above must be motivationally reasonable. It must be capable of assent without making extraordinary psychological or physical or moral demands, and it must set forth plausible mechanisms for achieving these ends.

There are at least two significant difficulties in capturing which duties are motivationally reasonable. The first is the danger of thinking that "any radical departure from accustomed patterns is psychologically unrealistic." (37) This is the danger of confounding the familiar with the necessary, with viewing as unalterable that which is merely inconvenient to change. Often, change is not impossible, but rather simply very costly. A second and related difficulty concerns how we identify plausibility limits. Philosophers speak of certain duties as inconsistent with a morally attractive conception of human life; they rely a great deal on intuitions about "commonsense morality," and they often appeal to human biological and psychological limits.

In sum, the institutional turn in cosmopolitan theory ascribes strong cosmopolitan duties to States based on the premise of plausibility constraints on individual cosmopolitan action. The next Part asks whether similar plausibility constraints limit cosmopolitan action by liberal democratic States as well.


This Part describes the theoretical, practical, and moral limitations on the ascription of strong cosmopolitan duties to liberal democratic governments. My claim is that these limitations are akin to the biological, moral, and psychological "plausibility constraints" on individual action that cosmopolitan theorists invoke as a justification for ascribing cosmopolitan duties to political institutions.

This Part's focus on the United States is justified by the theme of this Symposium, and by the fact that the United States is a leading target of cosmopolitan criticism. The United States differs from other liberal democracies in its constitutional details, its citizens' sentiments, its military and economic might, and in many other respects. Nonetheless, as I briefly discuss in Part III.A, I believe the basic points below hold for all liberal democracies.

A. The Source and Significance of Weak Cosmopolitan Sentiments

An individual acts altruistically if she has the goal of benefiting another person, she benefits that person, and she could have done better for herself had she chosen to ignore the effect of her action on the other person.38 The social science literature now confirms what common sense suggests but what some have doubted: Individuals are genuinely altruistic. (39) But if individuals are altruistic, why aren't the liberal States that represent them?

A similar puzzle arises in the corporate context. Individual shareholders are altruistic. But corporations generally are not. (40) The standard explanation is that a corporation furthers the purpose for which its members incorporated, which generally has to do with advancing member welfare, not nonmember welfare. The same logic might apply to the State.

This argument is open to the objection that a corporation (or any group) may consist of cosmopolitan-minded individuals who have organized to pursue cosmopolitan ends. The theories sketched in Part I correctly argue that institutions can (in theory) engage in cosmopolitan action, and that cosmopolitan individuals can act through such institutions more effectively than acting alone. There is power in numbers. Institutions can efficiently gather and transmit the information needed for collective action. They can generate economies of scale in a number of contexts that lower the costs of action. They can monitor individual contributions and punish free-riding, both of which strengthen collective action. They can provide norms and focal points to motivate and coordinate individual participation in group action. And they can solve psychological collective action problems. Cosmopolitan-minded individuals might lack motivation for cosmopolitan action because of a perceived inability to make a difference through individual action alone. An institution with power to effectuate change can motivate such individuals to action by clarifying the causal pathway between individual action and cosmopolitan change, and by helping the individual to envision her action as part of an undertaking involving many others.

These are the basic mechanisms that allow churches, charities, and other nongovernmental organizations to achieve greater collective cosmopolitan ends than group members could achieve acting on their own. But it does not follow that States can commit similar acts of cosmopolitan charity. There are many salient differences between these institutions and States.

First, States are much larger. Their membership does not consist of self-selected members with relatively homogenous and intense cosmopolitan sentiments. Rather, members of pluralistic societies vary significantly in their commitments (if any) to cosmopolitan charity. Many citizens have no cosmopolitan sentiments, or have anticosmopolitan sentiments. Others have weak cosmopolitan sentiments. Even strongly cosmopolitan-minded citizens can differ sharply about the appropriate focus of cosmopolitan charity. Supporters of aid for Israel and supporters of aid for the Palestinians, for example, might cancel one another out in the aggregate.

Heterogeneity of individual preferences related to cosmopolitan action, taken alone, is a reason to be skeptical of the claim that States can perform strong cosmopolitan duties. A major justification for the move to cosmopolitan duties for States is that individuals face collective action problems in performing cosmopolitan duties. (41) If citizens possessed intense and homogenous cosmopolitan sentiments, this argument might, for reasons just canvassed, make sense. But if the bulk of individuals do not have an interest in cosmopolitan charity, or if their interests are wildly varied and uneven, there is no collective action problem at the State level to overcome, and the move to political institutions achieves little.

Another crucial difference between a liberal democratic State and, say, Oxfam International, is that the State does not organize itself for the purpose of engaging in acts of cosmopolitan charity. The dominant purpose of any State is to create a community of mutual benefit for citizens and other members, and more generally to preserve and enhance the welfare of compatriots. The Constitution, for example, was designed to create a more perfect domestic order, and its foreign relations mechanisms were crafted to enhance U.S. welfare. (42) The same is true of liberal democracies generally. In this sense, a liberal democracy is more like IBM than Medecins Sans Frontieres, and skepticism about corporate or institutional altruism makes more sense.

A third obstacle is that although individuals are altruistic, their capacity for other-regarding action is not unbounded. Individuals tend to focus their attention, energies, and altruism on members of their community (friends, family, and compatriots) with whom they identify and share a common bond. Many view local attachments, and their cultivation, as central to human flourishing. (43) Others see patriotism and related local-regarding community-building mechanisms as necessary prerequisites to a flourishing State, and especially to a flourishing democracy. (44) I am sympathetic to these normative claims, but for now I need only point to the related positive point that self-sacrifice seems related to solidarity, which in turn is related to (physical, cultural, or familial) proximity. Viewing community from the State level, most citizens are much more likely to sacrifice for a compatriot than a noncompatriot, especially when giving to noncompatriots comes at the expense of needy compatriots. Even within the State community, altruism does not come close to ensuring that the well off adequately care for those who are not well off. State coercion is needed for most in-State welfare transfers. Given this relatively weak altruism toward compatriots, we should not expect individual altruism to extend to people who are physically and culturally much more distant.

None of this is to deny that solidarity is not perfectly coextensive with borders, or that some individuals have strong cosmopolitan commitments, or that many citizens have some regard for, and are willing to sacrifice a little for, noncompatriots. The point is simply that, as some cosmopolitans realize, (45) widespread and intense cosmopolitan sentiments simply do not exist.

To the extent that citizens do in fact have weak or nonexistent cosmopolitan sentiments, political institutions in liberal democracies cannot easily engage in cosmopolitan action. In a liberal democracy, foreign policy must be justified on terms acceptable to voters. (46) The theory of democratic foreign policy is that voters will throw out politicians who deviate too far from their foreign policy preferences. This means that political leaders who care about reelection cannot easily engage in acts of international altruism much beyond what voters or interest groups will support. Because the matter is so important, the U.S. Constitution imposes limits, over and above electoral recall, that reinforce the principal's (i.e., voters') control over the agent (i.e., leaders).

Consider the war power. War is among the most serious and fateful acts a State can undertake. This is one reason why the framers gave Congress the power to declare war. The meaning and scope of this provision is contested, especially in modern times when presidents have asserted independent war powers more aggressively. (47) But at least one idea behind the War Powers Clause was to place an "effectual check to the Dog of war by transferring the power of letting him loose from the Executive to the Legislative body." (48) The Framers aimed to limit the President to wars fought in the interests of, and thus supported by, the people most affected by war--the voters. This agency-cost-reducing justification for a legislative check on the war power is the one that Kant offered as the basis for his predicted democratic peace. (49) And it has become one of the normative cornerstones of the democratic peace thesis. (50)

A similar justification explains the Constitution's involvement of the legislature in the process of ratifying treaties and other legally binding international agreements. (51) Absent a contrary indication in an agreement or some other special circumstance, international law presumes that nations will indefinitely comply with international agreements. (52) Compliance with international agreements is important to secure the reciprocal benefits of the agreement, and to maintain a reputation for compliance for purposes of future negotiations. For these reasons, a State should not enter into an international agreement unless it is sure it can comply with its terms.

Legislative participation promotes compliance through several mechanisms. Some have to do with the conveyance of credible information to treaty partners about the likelihood of compliance. (53) More relevant for present purposes is that legislative consent, like Congress' war declaration power, reduces the agency costs of presidential action. The legislature ensures that the agreement negotiated by the President aligns with the principal whose interests he purports to represent--U.S. voters. Of course, the President (who is also elected) might in some contexts more accurately represent voter preferences than legislators--especially when one factors in the aggregation and related collective action difficulties that attend the legislative process. But this just shows that the Constitution is biased against international agreements, just as it is biased against war. The requirement of dual executive-legislative consent promotes compliance by increasing the likelihood that the United States enters into only those agreements that increase national welfare. But this benefit comes at a cost of not reaching some agreements (i.e., ones that the President failed to negotiate or that the legislature failed to consent to) that would have enhanced national welfare. This is a defendable tradeoff to ensure that treaties promote national welfare, especially since treaty compliance depends on both executive and legislative action.

In these and other ways, the U.S. Constitution--and, with different mechanisms, every liberal democracy--ties foreign policy action to voter preferences. Realists have long decried this tie, for they view the democratic process as an obstacle to a rational and coherent foreign policy. (54) The realist criticism overlooks the many countervailing foreign relations benefits of democratic foreign policy, only some of which are outlined above. (55) The important point for now, however, is not the normative issue, but rather the institutional fact that liberal democratic institutions cannot easily engage in cosmopolitan action unsupported by the people.

Humanitarian intervention provides the best example. Intellectual and policy elites have increasingly urged liberal democratic governments to intervene to prevent human rights atrocities in other states. (56) But despite millions of lives lost as a result of these atrocities in the twentieth century, and despite recent CNN-covered atrocities in Rwanda, Bosnia, Kosovo, and East Timor (among other places), Americans simply are not willing to spend blood and treasure on humanitarian interventions that do not offer a U.S. welfare-enhancement justification.

To be sure, U.S. political leaders and voters sometimes support humanitarian interventions to relieve human suffering, especially starvation. (57) But they do not support these interventions if they are expensive or threaten nontrivial losses of American lives. (58) Politicians understand this and act accordingly. This explains the Bush and Clinton administrations' long delay in intervening to stop the atrocities in Bosnia, and the eventual decision to do so with "pinprick" air attacks rather than ground troops. (59) This is why the otherwise-internationalist Clinton administration pulled out of Somalia when Americans began to suffer casualties. (60) It is one reason why the United States declined to intervene in Rwanda. (61) And it is the lesson of the Kosovo intervention. Even with a mixed strategic-humanitarian justification for intervention, U.S. fighter pilots flew at high altitudes and took other casualty-avoiding steps, and the Clinton administration precommitted not to use high-casualty ground troop operations. (62)

The absence of democratic support is a fundamental check on humanitarian intervention. As David Luban notes:
 In a democracy, the political support of citizens is a morally
 necessary condition for humanitarian intervention, not just a
 regrettable fact of life. If the folks back home reject the idea of
 altruistic wars, and think that wars should be fought only to
 promote a nation's own self-interest, rather narrowly conceived,
 then an otherwise-moral intervention may be politically
 illegitimate. If the folks back home will not tolerate even a single
 casualty in an altruistic war, then avoiding all casualties becomes
 a moral necessity. (63)

These points are overlooked by those who, with increasing fervor, call for humanitarian intervention without regard to its lack of popular support. For example, Samantha Power's prominent critique of the United States's failure to intervene to stop various genocides devotes little attention to the absence of popular support for costly humanitarian interventions. The little attention she does give the issue is devoted to criticizing leaders for deferring to popular opinion. (64) Power fails entirely to grapple with Luban's point, or with constitutional requirements related to the use of force abroad. The democratic deficit for humanitarian intervention is also missed by those who appear to oppose wars that lack congressional authorization except when those wars are fought for humanitarian ends. (65) The requirement for democratic support does not distinguish between wars fought on humanitarian grounds and those fought for national security reasons. If any distinction emerges in practice, it is one that favors welfare-enhancing wars fought for national security reasons, and disfavors humanitarian interventions that lack a national security justification.

In sum, the democratic hurdles to cosmopolitan action should give pause to those who believe that individuals possess limited cosmopolitan sentiments, but who nonetheless ascribe strong cosmopolitan duties to liberal democratic governments. Individuals act through and limit liberal democratic institutions. If there is reason to doubt that individuals lack powerful cosmopolitan motivations, there is reason to believe that this paucity of motivation will be reflected in the output of liberal democratic institutions.

B. A More Realistic View of the Democratic Process

The above analysis is incomplete in at least two important respects. It ignores evidence that U.S. voters might in fact be cosmopolitan-minded. And it assumes that leaders are perfect agents of the voters, which they are not. Even taking into consideration these points, however, it remains doubtful that liberal democracies can engage in strong cosmopolitan action.

We basically have two ways to tell whether and to what extent voters have cosmopolitan sentiments: how their representatives vote, and what opinion polls say. Neither method is foolproof, and tricky issues arise when polls say one thing and representatives act otherwise.

Consider the ICC treaty and the Kyoto Accord, both of which (many believe) might require cosmopolitan action if ratified by the United States. Opinion polls consistently find that a majority of U.S. voters support these treaties. (66) But just as consistently, political representatives from both parties oppose these treaties. For example, by a vote of 97-0, the Senate in 1997 resolved that the United States should not sign a Kyoto-related treaty that (as Kyoto contemplated) did not extend greenhouse gas reduction requirements to developing nations, or that would "result in serious harm to the economy of the United States." (67) Similarly, in 2002 Congress enacted a statute by overwhelming majorities that opposed U.S. participation in the International Criminal Court. (68) To make the puzzle more complex, political leaders and other elites are significantly more committed to internationalism than U.S. voters. (69) This suggests that the disconnect between voters and leaders should run in the opposite direction.

Why would leaders more committed to international engagement than voters oppose ambitious international treaties that voters appear to support? There are several possible explanations, all of which further illustrate the plausibility constraints on U.S. cosmopolitan action.

The first point is that voter support for the ICC and Kyoto treaties is not by itself evidence of cosmopolitan sentiment. As the national interest argument sketched in the Introduction suggests, internationalism is not the same as cosmopolitanism, because in many situations international acts enhance domestic welfare. Many Americans support the treaties on welfare-enhancement grounds, and the surveys do not distinguish the two possibilities. (And, of course, voter motives can be mixed.)

Moreover, the most comprehensive survey of voter attitudes toward U.S. foreign relations confirms what casual empiricism and other evidence (such as the United States' paltry foreign aid as a percentage of GNP) suggests: "Most altruistic goals of U.S. foreign policy, those primarily concerned with the welfare of people in other countries other than the United States, are not given very high priority by the U.S. public." (70) Polls consistently find that U.S. citizens "much more than foreign policy leaders tend to put a high priority on devoting resources to domestic spending programs rather than to foreign affairs," a tendency that has "grown stronger after the end of the cold war." (71) In this light, cosmopolitan sentiment related to the ICC and Kyoto treaties is probably not deep or intense. This in turn means that well-organized groups with more intense anticosmopolitan preferences--such as business interests who would suffer the main burden of Kyoto's costs--can be more successful in the democratic process. Many environmentalists decry such interest group domination of U.S. international environmental policy as a perversion of the democratic process and the national interest. But whether or not interest group politics are desirable in a democratic polity (many believe they are), it is an inherent feature of democratic process.

Another explanation for the puzzle is that politicians are much more informed than voters about the treaties and in particular about their costs. In many polls finding support for the Kyoto Accord and the ICC, most respondents had never heard of these treaties before being asked about them. (72) Moreover, poll questions are rarely framed in ways that discuss noncompliance by other nations, or the costs of enforcement and noncompliance. When the rare poll asks how much voters would be willing to pay for a treaty regime, support for the regime drops dramatically as the costs increase. (73) As suggested above, polls also show similar cost-sensitivity with respect to humanitarian intervention. (74) Political leaders have powerful reelection incentives to learn about the costs of international action, and the resources to do so. They tend to base their judgments on these facts rather than polling data, for they know they will be accountable to voters when the costs of international action become apparent. Leaders recognize that constituents do not generally support international regimes that are not cost-justified, and they act accordingly.

A related cost of treaty regimes is international noncompliance. National leaders are always uncertain, to some degree, about the information, preferences, and motivations of other nations. As a result, they rightly worry about other nations' noncompliance with norms and agreements. The noncompliance consideration--which takes us from ideal to nonideal theory on the international stage--counsels caution in embracing international regimes that involve national sacrifices and that depend for their efficacy on compliance by other nations. Precisely this concern underlies political opposition in the United States not only to the Kyoto Accord, but also to the Test Ban Treaty, the Landmines Convention, and the Bio-Weapons Convention.

This last point is overlooked by the institutionalist strand in cosmopolitan theory. Even if individual citizens did face a collective action problem in acting on their cosmopolitan sentiments, national institutions cannot necessarily solve the collective action problem. Rather, it changes the level and nature of the collective action problem. Many cosmopolitan proposals (for example, all of Nussbaum's proposals listed above) require international cooperation. Information and power asymmetries, as well as the absence of a centralized enforcement mechanism, make international collective action problems difficult to overcome even when there is a plausible argument that the international regime, if successful, would enhance the welfare of every participating State. The usual collective action hurdles to self-enforcement are all the more severe when the collective action comes at the expense of national welfare, for (among many other reasons) the incentives to cheat are overwhelming in this context, and every State knows it.

These latter few considerations--about intensity of preferences, interest group politics, voter misinformation, aggregation difficulties, and international collective action hurdles--require qualification of the earlier assumption that liberal democratic leaders are simple agents for voters. When voters' anticosmopolitan preferences are clear, informed, intense, and unopposed, and when international collective action problems can be overcome, leaders can act as faithful agents. But often the connection between voter preference and international political action is skewed and complicated. For the reasons already canvassed, these complexities can further raise the bar to cosmopolitan action.

The opposite may be true as well. The slack between voter preferences and leader action can cut in the opposite direction, and theoretically permits leaders to act with cosmopolitan charity beyond what constituents support. An important strand of democratic theory has always held that elected representatives should not be yoked to constituent preferences, especially when constituents are relatively uninformed. (75) Leaders should exercise wisdom and judgment in deciding--subject to electoral recall--what is best for their constituents. They should lead, not follow. They should shape constituent preferences, perhaps to reflect their more cosmopolitan outlook. And their capacity to do so is enhanced by the fact that the public pays relatively little attention to foreign affairs.

This conception of the democratic process does not, in my view, mean that the U.S. government could plausibly engage in more generous acts of cosmopolitan charity. Even political leaders with powerful cosmopolitan sentiments who are unworried about reelection hesitate to engage in costly altruistic acts abroad.

One reason why leaders hesitate is that, whatever their personal sentiments, they have (and perceive themselves to have) a moral duty, in virtue of their election, their oath, and their identity, to promote the welfare of the State and its citizens. The more fluid conception of democracy described above gives leaders discretion to identify what furthers constituents' interests. It does not permit leaders to impose significant local sacrifices for the sake of nonnationals beyond what can be justified in terms of local welfare-enhancement. Political leaders believe this and act accordingly.

Persistent domestic institutional constraints also hinder leader attempts to commit acts of cosmopolitan charity that exceed constituent preferences. It is really the President, and not legislators, who theoretically has the discretion to skirt short-term constituent pressures in this way. The President has broad independent foreign relations powers, and is not burdened by collective action problems to nearly the same degree as Congress. And yet the President cannot act too far beyond the wishes of Congress (or the voters). The President's unilateral discretion is probably at its height with respect to war. But in this context, the President is unambiguously accountable to the people, and, in any event, an uncooperative legislature can still retaliate via legislation, hearings, appointment hold-ups, defunding, and the like. With respect to international agreements, foreign aid, and most other international initiatives, the President's room for unilateral action is more limited because legislative participation, support, and funding is more directly relevant. In addition, any short-term, unilateral, non-welfare-enhancing action the President takes is reversible by the people and their representatives in the medium term. This is precisely how foreign policy in a democracy is designed to work.

This conclusion is consistent with political leaders having wide discretion to emphasize and act upon what they believe enhances U.S. welfare, especially in the short term. For example, the Clinton and Bush administrations interpreted and reacted differently to the Iraqi threat to national welfare, and took different attitudes toward the importance of particular treaty regimes. More broadly, current events are full of examples of liberal democratic leaders departing from apparent constituent foreign policy preferences in the name of promoting a national welfare that leaders believe constituents do not fully appreciate. Nothing in my analysis suggests that these departures are illegitimate. Only time and election returns will tell whether the leaders' assessment of voters' interests was correct. My point is simply that the various mechanisms described above ensure that--at least in the medium term and often in the short term--cosmopolitan action by a liberal democracy is bounded by constituent preferences.


The last Part sketched the plausibility constraints on cosmopolitan action by liberal democracies. If these observations are correct, one should hesitate before claiming that States have duties to engage in strong cosmopolitan action. "Can" limits "should." Just as morality can be too demanding of individuals, it can be too demanding of institutions. At the very least, the ascription of cosmopolitan duties to liberal democratic States requires careful consideration of voter sentiment and institutional reality.

There are many possible objections to this argument. In this limited space, I can address only two. The first is that voters could be educated to be more cosmopolitan, thereby making liberal democratic States more cosmopolitan. The second is that is that liberal democracy is not sacrosanct; alternative forms of governance may better serve the ends of international justice. After addressing these two points, I will consider how liberal democratic States might better achieve the cosmopolitan aim of helping noncompatriots in ways that are consistent with the institutional constraints on liberal democracy.

A. Education

One response to my argument is that individuals' uneven cosmopolitan sentiments are not sacrosanct. Through cosmopolitan education, citizens in democratic nations could become more cosmopolitan-minded. In Nussbaum's prominent formulation, cosmopolitan education can teach individuals to be troubled by world inequality, to understand what is local and nonessential, and to have a greater sense of other cultures and peoples needed for genuine international cooperation. (76) Enhanced cosmopolitan sentiments among individuals will translate into enhanced cosmopolitan actions by their governments.

Similar education arguments, and related assumptions about human perfectibility, have characterized cosmopolitan thinking for centuries. (77) Modern mass communication is the greatest possible educator about distant States, their cultures, and the suffering of their peoples. But despite daily reminders of human suffering around the globe, the peoples and nations of the world have not acted in ways that are progressively more altruistic. In the midst of the global communication transformations during the post-Cold War period (i.e., CNN, the Internet, and the like), foreign aid as a percentage of GNP among the wealthiest States dropped precipitously even though these States enjoyed a "peace dividend" amounting to approximately $450 billion per year. (78) Similarly, increased knowledge about suffering abroad during this period has not led to increased humanitarian interventions. (79)

There are many reasons, in addition to the institutional points already made, why this might be so. Mass communication can in theory enhance sympathy for noncompatriots by increasing knowledge of their suffering. But this effect can be counteracted by increased knowledge of difference, or of countervailing interests, abroad. (80) In addition, the spread of democracy during the past 200 years may have weakened cosmopolitan sentiment among citizens in democratic nations. (81) Many have argued that successful democracies demand a high degree of mutual commitment and solidarity that is inconsistent with strong cosmopolitan sentiment. (82) The types of education appropriate for a liberal democratic culture thus may be in deep tension with Nussbaum's proposed cosmopolitan educational reforms. (83)

One rejoinder to my skepticism about the transformative potential of education is that other liberal democracies are more cosmopolitan than the United States. To take a typically invoked example, Sweden is held out as a State with a cosmopolitan citizenry that supports cosmopolitan action by their government. Sweden is among the world's leaders in foreign aid, and it actively supports international institutions. U.S. citizens, properly educated, might become more like the Swedes, and the U.S. government, in turn, might become more other-regarding in its actions. Implicit in this argument would be the claim that I have confused the characteristics of liberal democracy in the United States with the characteristics of liberal democracy generally.

Swedes may well be more cosmopolitan than Americans; they certainly are a more homogenous population, and are traditionally more committed to social democracy. But there is little reason to believe that the Swedish government engages in greater cosmopolitan action than the United States. The arguments in Part II suggest that the hurdles to cosmopolitan action in a liberal democracy are structural: that too much cosmopolitan sentiment among a citizenry is inconsistent with democratic statehood; that liberal democratic governments cannot act much beyond what citizens will support; and that liberal democratic processes create multiple hurdles to cosmopolitan action, even assuming individual cosmopolitan sentiments. I believe the evidence from Sweden is consistent with these claims, and suggests broader structural constraints on the transformative potential of cosmopolitan education.

Begin with humanitarian intervention. This is perhaps the best test, for in many cases local costs clearly outweigh local benefits, and unlike foreign aid and certain treaty regimes, we can identify and eliminate mixed-motive cases. If anything, the traditionally neutral Swedes, and Europeans generally, are less cosmopolitan than Americans when it comes to humanitarian intervention. Since World War II, European voters have consistently demanded increases in spending on domestic social programs and decreases in spending on military programs. One result is that Europe's military capacity to intervene for humanitarian reasons has diminished significantly. Even when humanitarian interventions are militarily feasible and close to home--as in Bosnia and Kosovo in the 1990s--Europeans remained skittish and were disinclined to intervene. (84)

As for foreign aid: Sweden is described as the "darling of the Third World" because of its generous foreign aid program. (85) Sweden traditionally gives much more aid than the United States as a percentage of GNP. But there is significant evidence that Swedish aid should not be interpreted as cosmopolitan action. First, while more extensive than most other countries, Swedish foreign aid is still less than one percent of its GNP. Moreover, this aid has been cut in the decade following the end of the Cold War, even though the era was marked by a general peace and a large peace dividend. (86) Taken alone, this suggests that aid was at least in part related to broader national security aims during this period.

In addition, Swedish foreign aid is limited to ideologically similar nations that have significant trade relations with Sweden and where heavy Swedish political and business interests predominate. (87) Although Swedish governments long repudiated any link between aid and economic self-interest, following the Cold War (when the security element of aid had diminished), Sweden began to tie its aid explicitly to the purchase of Swedish goods and services or to favorable financing arrangements. (88) Swedish foreign aid looks even less charitable when one considers that Sweden's domestic agricultural and textile subsidies and other nontariff barriers harm the welfare of poor agricultural states to a significant degree, and possibly enough to offset the effect of its foreign aid. (89) None of this is to deny that many Swedes are motivated by humanitarianism. (90) It is just to point out the reasons why aid by the Swedish government should not be viewed as cosmopolitan action as I have used the term.

More broadly, Sweden's foreign aid and other cosmopolitan-seeming actions must be viewed in the context of Sweden's status as a "middle power." (91) The label refers to nations that exercise political and diplomatic power on the international stage through "soft" mechanisms like food aid, participation in international institutions, international civil service, and similar internationalist mechanisms. Middle powers show a greater devotion to international law and institutions than more powerful nations, because they can exercise power abroad most effectively in this fashion. But here, as before, it is important not to confuse internationalism with cosmopolitanism. Middle powers by definition have relatively little unilateral influence in politico-military issues. They focus their diplomatic and related foreign affairs resources where they can exert the most influence, especially against the major powers. (92) Their commitments to international institutions associated with cosmopolitan charity thus have a structural explanation wholly apart from cosmopolitan sentiment. The more general point is that the welfare of a State's citizens, and thus the structure of the State's foreign policy, varies depending on the power and stature of each State on the international stage.

My aim in this Part has not been to criticize Sweden, or to try to dislodge it and similarly situated middle powers from their exalted status among cosmopolitans. The aim was rather to suggest that Sweden's internationalism is not the same as cosmopolitanism, and that it has a structural explanation consistent with the claim that democratic foreign policy must serve the welfare of local constituents.

B. Alternatives to Liberal Democracy

A second objection is that liberal democracy at the level of the State should not be viewed as sacrosanct. Cosmopolitan theorists are usually quick to deny any desire for "world government" or a "centralized world state," (93) and as mentioned in the Introduction, many are firmly committed to liberal democratic territorial governance. But some cosmopolitan theorists propose an array of global democratic institutions to alleviate international social injustice. (94) These proposals share many common features, including a reverence for the United Nations, and the aim of shifting sovereignty upward toward international institutions. Many believe that the proliferation of international institutions, and the rise of the European Union, evidence moves in the globalist direction.

This is not the place for a comprehensive critique of quasi-world government or global democracy proposals. I will simply mention a few of the more obvious objections.

First are the well-known normative difficulties with global governance schemes. The most obvious difficulty concerns the democratic deficit associated with ever-expanding institutions. A related concern is that large-scale uniformity inherent in global governance schemes comes at the expense of too many unsatisfied individual preferences. Finally, there is the difficulty of human motivation and loyalty with respect to large, impersonal organizations. (95)

Second is the practical problem of how to generate such institutions, assuming they were normatively desirable. I know of no global democracy approach that spells out how or why States--and especially powerful States like the United States (or for that matter the European Union)--would submit to a broader form of genuine global governance. (96) Nations enter into international institutions because they gain more than they lose from doing so. (For powerful nations, this is often because they often reap affirmative benefits; for weaker nations, it is often because not joining the institutions lowers their welfare more than joining does. (97) Most important and effective international institutions (most prominently the World Trade Organization, the World Bank, and the International Monetary Fund) serve the interests of powerful nations, especially powerful Western nations, and most especially the United States. Powerful nations do not join institutions that do not serve their interests.

Successful governance in the domestic realm works differently than this purely instrumental conception of international governance. There are two distinguishing factors in the domestic realm: genuine communal sacrifices (whereby some members sacrifice interests for others), and, perhaps more important, centralized coercion. (98) Neither of these factors can work on a global scale. The standard proposal for international coercion is to strengthen the United Nations. (99) But the United Nations failed in its original ambition of having a free-standing police force, and it has failed to reliably transcend the problem of enforcement ever since. Like all collective security schemes, the United Nations depends wholly on member states' self-interested (and thus uneven) acts for coercion. It is hard to see how or why militarily powerful nations would ever agree to any other scheme.

As for community: There are natural limitations on the size of democratic government. The larger and more ambitious the government becomes, the more varied the governed population becomes (in endowment, culture, language, preferences, and the like), and the more difficult it becomes to maintain social harmony. (100) The European Union is often invoked as a counterexample. But the EU is more like the United States in the eighteenth century and Italy and Germany in the nineteenth--a process of state-building by smaller units with a common heritage and common interests. The EU example shows the difficulties that inhere in such a process even among subunit States that in many respects share a common culture, and that have over two millennia been unified in various ways (for example, the Roman Empire, the Catholic Church, the Holy Roman Empire, and the Concert of Europe). It does not provide a map for global government of peoples of radically different cultures, histories, and endowments.

C. A More Realistic Cosmopolitanism

An alternative to altering constituent preferences or rejecting liberal democratic government is to accept the basic plausibility constraints on the ability of liberal democracies to engage in strong cosmopolitan action, and to work within these limits to better achieve the cosmopolitan end of assisting the less fortunate. Below I sketch some very general elements of a more realistic cosmopolitanism.

1. Institutionalism.

According to a prominent cosmopolitan thinker, "[c]osmopolitanism is a moral outlook, not an institutional prescription." (101) The problem with this claim is that cosmopolitan thinkers actively engage in institutional prescription without attention to the institutions' mechanisms and limitations. As Part I showed, these thinkers take human limitations seriously. But just as human limitations help define and limit what duties are appropriate to impose on individuals, so too must institutional limitations be considered in defining and limiting what duties are appropriate to impose on states. These institutional limitations have special force when the institutions are governments of liberal democracies. Liberal democratic governments are not only guided and limited by their citizens, but also have well-known pathologies, some of which are described above, that further limit the possibilities of cosmopolitan action. And on top of this are the challenges of collective action on the international stage. These and scores of other institutional design issues are not merely relevant to cosmopolitan means. They are also crucial to the identification of appropriate cosmopolitan ends.

2. Instrumentalism.

To take institutions more seriously, cosmopolitan theorists must consider the costs as well as the benefits of various cosmopolitan proposals. Cosmopolitan theorists often commit the nirvana fallacy of comparing the current arrangement of nations, with its manifest evils, to the hypothetically perfect cosmopolitan governance proposals, without consideration of the proposals' pathologies. (102) To take one of many examples, Nussbaum's environmental, tax, trade, and labor proposals described above (103) do not consider ex ante or self-defeating effects. Many believe, for example, that such proposals will adversely affect investment in poorer countries and thus make the peoples of the poorer countries worse off. Issues of this sort are of course contested. The point here is simply that a cosmopolitanism that takes institutions seriously is a cosmopolitanism that does serious cost-benefit analysis.

3. Civil society.

One lesson that institutional analysis teaches is that the institutions most likely to engage in effective cosmopolitan action are ones with relatively homogenous cosmopolitan sentiments among group members. Such groups are most likely to be found in civil society, "the space of uncoerced human association and also the set of relational networks--formed for the sake of family, faith, interest and ideology--that fill this space." (104) Unlike States, civil society can effectively engage in cosmopolitan action, precisely because civil society groups consist of like-minded persons who come together voluntarily to take advantage of the collective action powers that institutions can deliver. Relatedly, cosmopolitan action by such voluntary groups (where costless exit is possible) is, all things equal, more legitimate than cosmopolitan action by liberal democracies, precisely because centralized coercion is not needed in the former case.

4. National interest.

To say that civil society can more effectively engage in cosmopolitan action than liberal democratic governments is not to say that civil society can more effectively assist the world's least well off. Liberal democratic government is more powerful than civil society. But it can generally only act in ways that enhance national welfare. Often, it is possible for a State to enhance its welfare in ways that helps others. By treaty, nations can set terms of cooperation with other nations that can make all better off. Or by intervention, one State can serve its strategic interests in ways that help local populations in other States.

The limiting principle here, however, is national interest as I have defined it in this Article. This means that the best we can hope for is uneven humanitarian intervention that comports with the strategic and security interests that would be furthered by the potentially intervening nations. It also means that sometimes a treaty that makes many nations better off will not be acceptable to other nations, as is often the case when the United States opts out of certain treaty regimes. Sometimes there simply won't be a deal to reach, or deals that seem to promote justice might have to be modified, sometimes dramatically, to generate requisite consent.

5. Double standards.

One consequence of accepting the national interest limitation on other-regarding action is that double standards are sometimes inevitable if the aim is to get nations to bring the most relief to those who are least well off. Double standards are embedded throughout international law--most prominently in the United Nations Charter, which purports to recognize sovereign equality, but which dramatically skews decisionmaking authority to reflect the power asymmetries of the 1940s. Double standards are often a byproduct of the circumstances in which national action is available to help people in other nations.

This point is most obvious in cases of humanitarian intervention, where intervention usually depends on the welfare concerns of the intervening nation. The resulting uneven patterns of enforcement are often derided as hypocritical. Opportunistic interventions are also what give rise to the (not unjustified) concern that many so-called humanitarian interventions are ruses for invasions motivated in large part by strategic ends. A clear-eyed analysis of interventions would realize that such mixed-motive cases are probably the best we can hope for. The presence of mixed motives does not detract from the fact that some such interventions might help local populations, as the Kosovo intervention arguably did. Those who want to eliminate double standards aim to increase intervention to all cases of massive human rights abuses. But the much more likely possibility is that the elimination of double standards would be achieved by eliminating all humanitarian interventions.

A similar point applies to the ICC. The ICC framers rejected the United States's proposal to require Security Council referrals for all prosecutions, and thus to effectively immunize the five veto-wielding members, and their close allies, from prosecution. The ICC framers rejected such double standards as inconsistent with the international rule of law. One consequence is that the United States opposes the ICC. There is a case to be made that an ICC without U.S. support harms global human rights enforcement more than an ICC with a Security Council veto or a world with no ICC at all. (105) The counterargument is that by insisting on the equal application of ICC principles to all nations, the ICC furthers long-term human rights interests even at the short-term expense of U.S. opposition. My aim here is not to resolve this disagreement. My aim is to urge that this is the right kind of debate--a debate that focuses instrumentally on what's best for those who suffer, and not on the moral acceptability of double standards in international relations.


This Article has proceeded in the spirit of being a "corrective to the exuberance of utopianism." (106) The main claim has been that there is a deep tension between liberal democratic governance and the increasing demands that liberal democratic States engage in cosmopolitan action. To be relevant and effective, cosmopolitan thinking must take the institutions and limitations of liberal democracy much more seriously.

(1.) See, e.g., ROBERT A. DAHL, DEMOCRACY AND ITS CRITICS 83-105 (1989).


(3.) See, e.g., Gregory H. Fox & Brad R. Roth, Introduction: The Spread of Liberal Democracy and Its Implications for International Law, in DEMOCRATIC GOVERNANCE AND INTERNATIONAL LAW 1, 6-8 (Gregory H. Fox & Brad R. Roth eds., 2000).

(4.) See, e.g., Michael W. Doyle, Kant, Liberal Legacies, and Foreign Affairs, 12 PHIL. & PUB. AFF. 205, 213 (1983).

(5.) See, e.g., Anne-Marie Slaughter, International Law in a World of Liberal States, 6 EUR. J. INT'L L. 503, 504 (1995).

(6.) See, e.g., Gregory H. Fox, The Right to Political Participation in International Law, 17 YALE J. INT'L L. 539, 544-70 (1992); Thomas M. Franck, The Emerging Right to Democratic Governance, 86 AM. J. INT'L L. 46 (1992).

(7.) See infra Part I.

(8.) Brian Barry, Statism and Nationalism: A Cosmopolitan Critique, in NOMOS XLI: GLOBAL JUSTICE 12, 53 (Ian Shapiro & Lea Brilmayer eds., 1999) (emphasis added).

(9.) See, e.g., Peter Singer, Famine, Affluence, and Morality, 1 PHIL. & PUB. AFF. 229 (1972).

(10.) Martha C. Nussbaum, Toward a Viable Cosmopolitanism, Castle Lecture 4 at Yale University 2 (Mar. 1, 2000) (transcript on file with author).

(11.) Id. at 2-3.

(12.) Id. at 15-16.

(13.) Id. at 3.

(14.) Id.; see also id. ("No real institutions will be perfect either; but they can coerce better behavior than people would likely attain without them.... Institutions both coerce correct conduct and allow people a breathing space, so that their life is not always filled up with the demands of correct conduct."). Nussbaum continues:
 [P]olitical institutions need to be cognizant of the deepest dangers
 in human psychology and to provide bulwarks against those dangers:
 thus they do not simply model us at our best, they also stand
 between us and our worst. (This was a key insight of Madison, and is
 built into the foundations of the U.S. political order.)

Id. at 4.

(15.) Id. at 16.

(16.) Id.

(17.) See Michael J. Green, Institutional Responsibility for Global Problems, 30 PHIL. TOPICS (forthcoming 2003) (manuscript on file with author).

(18.) Id. (manuscript at 3).

(19.) Id. (manuscript at 10).

(20.) Id. (manuscript at 3).

(21.) Id. (manuscript at 13).

(22.) Id.

(23.) "Since institutions have different capacities as agents than individuals, there is less reason to apply the restrictive conception of responsibility to them." Id. (manuscript at 14).

(24.) Id.

(25.) Id. (manuscript at 15).


(27.) Id. at 267, 272.

(28.) See Thomas W. Pogge, Cosmopolitanism and Sovereignty, 103 ETHICS 48, 51 & n.9 (1992).

(29.) Thomas W. Pogge, An Egalitarian Law of Peoples, 23 PHIL. & PUB. AFF. 195, 201 (1994).

(30.) See, e.g., CHARLES R. BEITZ, POLITICAL THEORY AND INTERNATIONAL RELATIONS 153 (1979); Judith Lichtenberg, National Boundaries and Moral Boundaries.' A Cosmopolitan View, in BOUNDARIES: NATIONAL AUTONOMY AND ITS LIMITS 79 (Peter G. Brown & Henry Shue eds., 1981).

(31.) See Singer, supra note 9.


(33.) THOMAS NAGEL, EQUALITY AND PARTIALITY 21 (1991). Nagel continues:
 [S]ince justification in political theory is intended to produce not
 just assent to a proposition, but acceptance of and support for a
 set of institutions and a form of life, the most important facts
 about individuals for these purposes are facts of motivational
 psychology and facts about what individuals have reasons to do and

 [We want principles that] motivate and command respect and that
 will therefore give authority to results which are reached in
 accordance with them ....

Id. at 23.


(35.) Id.

(36.) Liam B. Murphy, Institutions and the Demand for Justice, 27 PHIL. & PUB. AFF. 251 (1999).

(37.) NAGEL, supra note 33, at 22.

(38.) See generally Jane Allyn Piliavin & Hong-Wen Charng, Altruism: .4 Review of Recent Theory and Research, 16 ANN. REV. Soc. 27 (1990). Some definitions drop the requirement that the altruistic action succeed. See id.; Kristen Renwick Monroe, A Fat Lady in a Corset: Altruism and Social Theory, 38 AM. J. POL. SCI. 861, 862-63 (1993). This nuance is not relevant here.

(39.) See Piliavin & Charng, supra note 38.

(40.) See id. at 57. To be clear, corporations give to charity, but (the claim is) they do so because "giving is good for business." Id.

(41.) See supra Part I.A.


(43.) This is a central theme of the literature on liberal nationalism. See, e.g., DAVID MILLER, ON NATIONALITY (1997); NAGEL, supra note 33, at 169-79; YAEL TAMIR, LIBERAL NATIONALISM (1993).

(44.) See, e.g., Robert Post, Between Philosophy and Law: Sovereignty and the Design of Democratic Institutions, in NOMOS XLII: DESIGNING DEMOCRATIC INSTITUTIONS 209, 211, 216-17 (Ian Shapiro & Stephen Macedo eds., 2000); Charles Taylor, Why Democracy Needs Patriotism, in MARTHA CRAVEN NUSSBAUM, FOR LOVE OF COUNTRY: DEBATING THE LIMITS OF PATRIOTISM 119 (Joshua Cohen ed., 1996).

(45.) See Nussbaum, supra note 10.

(46.) Cf HENRY SIDGWICK, THE ELEMENTS OF POLITICS 309 (4th ed. 1919) (defending "the national ideals of political organisation," under which foreign policy should "promote the interests of a determinate group of human beings, bound together by the tie of a common nationality").


(48.) Letter from Thomas Jefferson to James Madison (Sept. 6, 1789), in 15 THE PAPERS OF THOMAS JEFFERSON 392 (Julian P. Boyd ed., 1958). For similar sentiments, see Debates in the Convention of the State of North Carolina on the Adoption of the Federal Constitution, in 4 JONATHAN ELLIOT, THE DEBATES IN THE SEVERAL STATE CONVENTIONS ON THE ADOPTION OF THE FEDERAL CONSTITUTION 107-08 (1863) (remarks of James Iredell); The Pennsylvania Convention, Tues., 11 December 1787, in 2 THE DOCUMENTARY HISTORY OF RATIFICATION OF THE CONSTITUTION: RATIFICATION OF THE CONSTITUTION BY THE STATES 583 (Merrill Jensen ed., 1976) (remarks of James Wilson). But see John C. Yoo, The Continuation of Politics by Other Means: The Original Understanding of War Powers, 84 CAL. L. REV. 167 (1996) (arguing that Framers did not intend Congress to have a legal check on presidential war power).

(49.) IMMANUEL KANT, To Perpetual Peace (1795), in PERPETUAL PEACE AND OTHER ESSAYS ON POLITICS, HISTORY AND MORALS 341 (Ted Humphrey trans., 1983).

(50.) See, e.g., BRUCE RUSSETT, GRASPING THE DEMOCRATIC PEACE: PRINCIPLES FOR A POST-COLD WAR WORLD 38-42 (1993). It is worth noting here that the agency-cost-reduction idea applies to all war-related activities by democracies, and cannot by itself explain the democratic peace thesis, which is limited to wars between democracies.

(51.) There are essentially two such processes: the treatymaking process, which requires the consent of two-thirds of the Senate, and the process of making congressional-executive agreements, which requires the consent of the majorities in both Houses. The President can enter into "pure" executive agreements on his authority alone, but the scope of the power is limited, and the fact is that there are relatively few pure executive agreements.

(52.) See Vienna Convention on the Law of Treaties, May 23, 1969, art. 56, 1155 U.N.T.S. 331, 345.

(53.) See LISA L. MARTIN, DEMOCRATIC COMMITMENTS: LEGISLATURES AND INTERNATIONAL COOPERATION (2000); Jack Goldsmith & Eric A. Posner, International Agreements: A Rational Choice Approach, 43 VA. J. INT'L L. (forthcoming 2003) (manuscript on file with author).

(54.) See HANS J. MORGENTHAU, POLITICS AMONG NATIONS: THE STRUGGLE FOR POWER AND PEACE 187-89 (1948); George F. Kennan, American Democracy and Foreign Policy, in GEORGE F. KENNAN, AT A CENTURY'S ENDING: REFLECTIONS 1982-1995, at 127, 137 (1996); cf. ALEXIS DE TOQUEVILLE, DEMOCRACY IN AMERICA 219 (Harvey C. Mansfield ed., Delba

Winthrop trans., 2000) (noting that democratic governments are "decidedly inferior" when it comes to pursuing consistent foreign policy).



(57.) Even in this context, national interest matters. For example, in 1992, National Security Advisor Scowcroft justified the ostensibly humanitarian intervention in Somalia as necessary to demonstrate, in the wake of the recent failure to intervene in Bosnia, that the decision not to intervene in Bosnia was unrelated to the Bosnian victims' Muslim faith, and that the United States was not afraid to intervene abroad. See POWER, supra note 56, at 293.

(58.) Americans support various types of "passive" (i.e., relatively noncostly) intervention in the face of atrocities or starvation, but this support drops precipitously with the introduction of ground troops or the possibility of casualties. See Richard Sobel, The Polls--Trends: United States Intervention in Bosnia, 62 PUB. OPINION Q. 250 (1998). Sobel also indicates that opposition to intervention in Bosnia was more intense than support for intervention, a crucial factor for any democratic leader. For a review and qualification of the evidence supporting the "casualties" hypothesis, see James Burk, Public Support for Peacekeeping in Lebanon and Somalia: Assessing the Casualties Hypothesis, 114 POE. Sci. Q. 53 (1999).

(59.) See POWER, supra note 56, at 283-88, 304-05.

(60.) See id. at 317.

(61.) See id. at 364-84.

(62.) See id. at 454-58; David Luban, Intervention and Civilization: Some Unhappy Lessons of the Kosovo War, in GLOBAL JUSTICE AND TRANSNATIONAL POLITICS 79 (Pablo De Greiff & Ciaran Cronin eds., 2002).

(63.) Luban, supra note 62, at 85-86.

(64.) Power's most extended treatment of the issue can be found in POWER, supra note 56, at 305-06.

(65.) See John Yoo, The Dogs That Didn't Bark: Why Were International Legal Scholars MIA on Kosovo?, 1 CHI. J. INT'L L. 149 (2000).

(66.) See, e.g., CHI. COUNCIL ON FOREIGN REL., WORLDVIEW 2002: AMERICAN PUBLIC OPINION & FOREIGN POLICY 34 (2002) [hereinafter CCFR 2002 REPORT] (finding that 71% of Americans supported U.S. participation in the ICC and 64% supported U.S. participation in the Kyoto agreement), available at detailreports/usreport/index.htm.

(67.) Byrd-Hagel Resolution, S. Res. 98, 105th Cong. (1997). The Senate added that any treaty that satisfied these criteria would also need to be accompanied by a detailed cost analysis of its implementation.

(68.) See American Servicemen's Protection Act, Pub. L. No. 107-206, 116 Stat. 899 (2002).

(69.) See CCFR 2002 REPORT, supra note 66, at 63-73 (comparing leader and public attitudes to various foreign policy issues, and showing that, by wide margins, leaders are more supportive of active internationalism, and the public places a higher priority on domestic as opposed to international programs); see also ERIC ALTERMAN, WHO SPEAKS FOR AMERICA? WHY DEMOCRACY MATTERS IN FOREIGN POLICY 14-15 (1998) (surveying public opinion evidence in support of similar conclusion); Benjamin I. Page & Jason Barabas, Foreign Policy Gaps Between Citizens and Leaders, 44 INT'L STUD. Q. 339 (2000) (analyzing similar results based on twenty-four years of Chicago Council of Foreign Relations surveys).

(70.) CCFR 2002 REPORT, supra note 66, at 20; see also CHI. COUNCIL ON FOREIGN REL., AMERICAN PUBLIC OPINION & U.S. FOREIGN POLICY 1999, at 18 (John E. Rielly ed., 1999) (noting that Americans possess "limited altruism," and that "[n]otably absent from ... first-tier and second-tier priorities are goals that might be associated with altruistic internationalism, or goals that would primarily benefit others"), available at

(71.) Page & Barabas, supra note 69, at 347. The most recent Chicago Council on Foreign Relations survey shows a spike in support for giving a higher priority to foreign affairs following the September 11 attacks, but as the CCFR Report emphasizes, the reasons for this have to do with local security enhancement. See CCFR 2002 REPORT, supra note 66. But see STEVEN KULL & I.M. DESTLER, MISREADING THE PUBLIC: THE MYTH OF A NEW ISOLATIONALISM (1999) (arguing that, despite polling data and political branch behavior, the U.S. public supports more active international engagement).

(72.) See, e.g., ROPERASW, AMERICANS' ATTITUDES TOWARD AN INTERNATIONAL CRIMINAL COURT (2002) (finding that 61% of Americans have never heard or read about the International Criminal Court), available at roper_report.pdf; Steven Kull, Americans on the Global Warming Treaty, Program on Int'l Pol'y Attitudes (citing a September 1998 Wirthlin poll in which 86% of Americans had never heard of the Kyoto Protocol on global warming, and in which, after details of the conference were given, 69% said they had not heard of it) [hereinafter PIPA, Americans Survey], at (last modified Feb. 4, 2000). Similarly, one week after the Senate declined to consent to the Test Ban Treaty, 50% of American voters had not heard about the vote, and 61% had not heard of the reasons why some Senators voted for the treaty and others against it. Pew Research Center for the People and the Press, Survey Reports: Senate Test Ban Little Noticed, Less Understood (Oct. 21, 1999), at

(73.) See PIPA, Americans Survey, supra note 72 (finding that 76% are willing to pay $10 per month to significantly reduce global warming, 63% are willing to pay $25 per month, 36% are willing to pay $50 per month, 19% are willing to pay $75 per month, and 11% are willing to pay $100 per month). The inconsistency of polling data is made clear by a P1PA poll conducted one month earlier, in which 63% of Americans agreed that "[p]rotecting the environment is so important that requirements and standards cannot be too high and continuing environmental improvements must be made regardless of cost." Id. (emphasis added).

(74.) See supra note 58; see also Richard Sobel, U.S. and European Attitudes Toward Intervention in the Former Yugoslavia: Mourir pour la Bosnie?, in THE WORLD AND YUGOSLAVIA'S WARS ch. 6 (Richard H. Ullman ed., 1998) (noting that 67% of Americans would support a U.S. deployment in Bosnia if no American soldiers were killed, but only 31% would continue to approve if 100 American soldiers were killed), at

(75.) See, e.g., THE FEDERALIST NOS. 10, 37, 49, 51, 63 (James Madison).

(76.) Martha Craven Nussbaum, Patriotism and Cosmopolitanism, in NUSSBAUM, supra note 44, at 3, 11-15.

(77.) For an earlier summary and criticism, see HANS J. MORGENTHAU, SCIENTIFIC MAN VERSUS POWER POLITICS 17-19, 172-75, 209-11 (1946).

(78.) See THOMAS POGGE, WORLD POVERTY AND HUMAN RIGHTS: COSMOPOLITAN RESPONSIBILITIES AND REFORMS 8, 232 n.143 (2002) (relying on various U.N. Developmental Progress Reports).

(79.) For recent examples, see POWER, supra note 57.


(81.) Cf. Hans J. Morgenthau, The Twilight of International Morality, 58 ETHICS 79, 8897 (1948) (arguing that the rise of democracy and nationalism undermines the moral limitations of international politics).

(82.) See sources cited supra note 44.

(83.) Cf. Michael W. McConnell, Don't Neglect the Little Platoons, in NUSSBAUM, supra note 44, at 78 (arguing that cosmopolitan education of the type proposed by Nussbaum may be destructive, and that the best way to promote concern for others is to first foster affection for the local).

(84.) See, e.g., ROBERT KAGAN, OF PARADISE AND POWER: AMERICA VS. EUROPE IN THE NEW WORLD ORDER (2003). On Sweden in particular, see Bjorn Moller, The Nordic Countries: Whither the West's Conscience?, in KOSOVO AND THE CHALLENGE OF HUMANITARIAN INTERVENTION: SELECTIVE INDIGNATION, COLLECTIVE ACTION, AND INTERNATIONAL CITIZENSHIP 151 (Albrecht Schnabel & Ramesh Thakur eds., 2001).

(85.) Peter J. Schraeder, Stephen W. Hook & Bruce Taylor, Clarifying the Foreign Aid Puzzle: A Comparison of American, Japanese, French, and Swedish Aid Flows, 50 WORLD POL. 294, 295 (1998), available at

(86.) See U.N. DEV. PROGRAMME, HUMAN DEVELOPMENT REPORT 2002: DEEPENING DEMOCRACY IN A FRAGMENTED WORLD 216 (2002) (noting Swedish aid dropped from 0.91% of GNP to 0.8% of GNP between 1990 and 2000), available at

(87.) Katie Laatikainen, The Disillusionment of Nordic Aid, in FOREIGN AID TOWARD THE MILLENNIUM 109 (Steven W. Hook ed., 1996); Schraeder et al., supra note 85, at 316-17.

(88.) Schraeder et al., supra note 85, at 316-17.

(89.) Although the harmful effects of subsidies on developing countries are hard to quantify, they are clearly significant. See ACTION AID, FARMGATE: THE DEVELOPMENTAL IMPACT OF AGRICULTURAL SUBSIDIES (2002) (noting that governments in rich countries are paying over $300 billion each year to subsidize their agricultural sectors, which is six times the total amount of aid to developing countries), available at I have been unable to determine the specific effects of Swedish agricultural (and textile) subsidies, but again, the subsidies are significant. See Magnus Blomstrom, Sweden's Trade and Investment Policies vis-a-vis the Third World, in THE OTHER SIDE OF INTERNATIONAL DEVELOPMENT: NON-AID ECONOMIC RELATIONS WITH DEVELOPING COUNTRIES OF CANADA, DENMARK, THE NETHERLANDS, NORWAY, AND SWEDEN 167, 173, 175 (1990); Carl Hamilton, Agricultural Protectionism in Sweden, 1970-1980, 13 EUR. REP. AGRIC. ECOLOGY 75 (1986).

(90.) Cf. DAVID HALLORAN LUMSDAINE, MORAL VISION IN INTERNATIONAL POLITICS: THE FOREIGN AID REGIME, 1949-1989 (1993) (arguing that foreign aid has a large humanitarian component). Lumsdaine's thesis about the moral basis for foreign aid was based on Cold War data and is not borne out in the post-Cold War world. See Schraeder et al., supra note 85.

(91.) On middle power internationalism, see generally MIDDLE POWER INTERNATIONALISM: THE NORTH-SOUTH DIMENSION (Cranford Pratt ed., 1990).

(92.) Cf. Robert O. Keohane, Lilliputians' Dilemmas: Small States in International Politics, 23 INT'L ORG. 291,296 (1969) (describing a middle power as "a state whose leaders consider that it cannot act alone effectively but may be able to have a systemic impact in a small group or through an international institution") (emphasis deleted).

(93.) See, e.g., POGGE, supra note 78, at 178.

(94.) Prominent examples include YOUNG, supra note 26; David Held, Democracy and Globalization, in RE-IMAGINING POLITICAL COMMUNITY: STUDIES IN COSMOPOLITAN DEMOCRACY 11 (Daniele Archibugi, David Held & Martin Kohler eds., 1998); Pogge, supra note 28.

(95.) See generally NAGEL, supra note 33, at 169-79; Michael Walzer, Governing the Globe: What Is the Best We Can Do?, DISSENT, Fall 2000, at 44.

(96.) Even Walzer's effort, which is among the best and most sensible analyses of the tradeoffs of international governance schemes, does not explain how his preferred international governance scheme would actually come about. See Walzer, supra note 95.



(99.) See YOUNG, supra note 26, at 271-75; Held, supra note 94.

(100.) See Walzer, supra note 95.

(101.) Barry, supra note 8, at 35.

(102.) A counterexample is Walzer, supra note 95, who makes the tradeoffs clear.

(103.) See supra notes 10-16 and accompanying text.

(104.) Michael Walzer, The Civil Society Argument, in THE CITIZENSHIP DEBATES: A READER 291, 291-92 (Gershon Shafir ed., 1998).

(105.) See Jack L. Goldsmith, The Self-Defeating International Criminal Court, 70 U. CHI. L. REV. 89 (2003).

(106.) CARR, supra note 98, at 10; see also supra text accompanying note 3.

Jack Goldsmith, Professor of Law, University of Chicago (on leave); Special Counsel to the General Counsel, U.S. Department of Defense. The views expressed here are the author's alone, and do not reflect the views of the Department of Defense or the U.S. government. Thanks to Curtis Bradley, Ryan Goodman, Stephen Krasner, Daryl Levinson, Sebastian Mallaby, Derek Jinks, Jim Madigan, Eric Posner, Duncan Snidal, Paul Stephan, David Strauss, Cass Sunstein, Adrian Vermeule, John Yoo, and workshop participants at the University of Chicago and NYU law schools for valuable comments, and Bryan Dayton for valuable research assistance.
COPYRIGHT 2003 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Symposium on Treaties, Enforcement, and U.S. Sovereignty
Author:Goldsmith, Jack L.
Publication:Stanford Law Review
Date:May 1, 2003
Previous Article:Navigating law and politics: the prosecutor of the International Criminal Court and the independent counsel.
Next Article:The new confederalism: treaty delegations of legislative, executive, and judicial authority.

Related Articles
Reconceptualizing sovereignty in the Americas: historical precursors and current practices.
Why sovereignty matters: the erosion of democracy.
Foreword: on American Exceptionalism.
Democracy in Aotearoa/New Zealand.
Myth vs. fact: in their myth-busting attempts, the architects of the Security and Prosperity Partnership (SPP) use half-truths and lies to conceal...
What kind of foreign policy under "Prime Minister Gordon Brown"?
Democracy and international human rights law.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters