2 Lefkowitz.indd Democracy, Legitimacy, and Global Governance DAV I D L E F KOW I T Z University of Richmond AbsTrAcT What property (or properties) render international institutions and law legitimate, such that those over whom they claim jurisdiction ought to defer to their directives rather than acting on their own judgment? In this essay I critically examine Tom Christiano’s treatment of two possible answers to this question: global democracy, and an institution’s or legal regime’s capacity to enhance its subjects’ responsiveness to the reasons for action that apply to them. While Christiano rightly rejects the inference from affected interests to global democracy, his argument elides the fundamental reason we ought to do so, namely that at present the degree of cross-border interdependence does not rise to the level where it is possible for citizens of different states to treat one another justly only by submitting to a common legal order that substantially erodes state sovereignty. International law and institutions can enjoy some legitimacy on instrumental grounds, however, even if they are neither democratic nor the product of agreement in free and fair conditions. Keywords: democracy, international law, legitimacy, authority, republicanism INTRODUCTION What property (or properties) render international institutions and law legitimate, such that those over whom they claim jurisdiction ought to defer to their directives rather than acting on their own judgment? In this essay I critically examine Tom Christiano’s treatment of two possible answers to this question: global democracy, and an institution’s or legal regime’s capacity to enhance its subjects’ responsiveness to the reasons for action that apply to them (Christiano 2010, 2011). With respect to the former, I argue that while Christiano rightly rejects the inference from 2 Lefkowitz.indd 200 21/4/17 13:30 Democracy, Legitimacy, and Global Governance 201 LEAP 4 (2016) affected interests to global democracy, his argument elides the fundamental reason we ought to do so, namely that at present the degree of cross-border interdependence does not rise to the level where it is possible for citizens of different states to treat one another justly only by submitting to a common legal order that substantially erodes state sovereignty. With respect to the latter, I argue that international law and institutions can enjoy some legitimacy on instrumental grounds, that their doing so does not depend on their being either democratic or the product of agreement in free and fair conditions, and that we can reliably identify legislative and judicial mechanisms that satisfy the instrumental standard for law’s legitimacy. 1. AGAINST GLOBAL DEMOCRACY A common argument for global democracy infers that from the fact that the conduct of people in one state affects the interests (or, more narrowly, those interests that ground human rights) of those living in other states that the former can treat the latter justly only by submitting to a common legal order whose laws are enacted by a directly elected global parliament (see, e.g. Archibuigi and Held 1995). While Christiano is right to reject this inference, the arguments he offers to support this conclusion mask what I contend is its fundamental error, namely that if it is possible for agents to treat one another justly by limiting their interactions so that they do not threaten to setback one another’s fundamental interests, then they are not morally required to submit to a common set of rules that govern these interactions. Instead, the decision to do so is one over which agents exercise moral discretion. This position is simply the converse of Kant’s well-known argument for the moral necessity of the state, which holds that where agents cannot avoid interacting with one another justice requires that they submit to a common legal order. Consider Christiano’s unequal stakes argument against global democracy. He asserts that a far greater level of interdependence of interests obtains for those who are citizens of the same state than for those who are citizens of different states (Christiano 2010: 132-33; Christiano 2011: 74-5). The former share a common world, while the latter do not. But what exactly should we infer from this, supposing it is true? One possibility is that absent their sharing a common world democratic government will fail to publicly treat all subject to the resulting laws equally. A second possibility, though, is that absent a common world the level of interdependence of interests among a set of agents does not rise to the level where it is impossible for them to treat one another justly except by submitting to the same legal order. Though the two possibilities are not 2 Lefkowitz.indd 201 21/4/17 13:30 202 David Lefkowitz LEAP 4 (2016) mutually exclusive, the second provides the more fundamental objection. This is so because the question of whether agents are morally required to submit to a common set of rules regulating some type of conduct is prior to the question of how the rules of such an order ought to be made if they are to be legitimate. The principle of public equality provides an answer to the latter question, but to answer the former Christiano needs a version of the affected interests principle, namely one that holds that agents have a duty to submit to a common legal order if and only if doing so is necessary to avoid setbacks to their own and/or to others’ fundamental interest in judgment. Christiano’s assessment of the advantages the Fair Democratic Association (FDA) model of global governance has over global democracy lends further support to the claim that it is the possibility of treating others justly without submitting to a common set of rules that blocks the inference from affected interests to global democracy. For example, Christiano maintains that a FDA is better able to mitigate the problem of persistent minorities “because states can refuse to enter into negotiations and agreements” (Christiano 2011: 81; Christiano 2010: 136). This claim implies, however, that the model of global democracy fails because it compels groups or states to submit to a particular legal regime when they need not do so in order to treat others justly. While it’s bad enough to be a persistent minority within a governing institution, it is even worse to be needlessly compelled to be a persistent minority within such an institution. With respect to the different stakes states may have in a particular system of international legal rules, e.g. those governing trade, Christiano alleges a FDA will be superior to global democracy because “states with high stakes in an agreement can invest a lot of time and energy in it, while states with lesser stakes presumably will invest less time and energy” (Christiano 2011: 81; Christiano 2010: 136). Yet the focus on time and energy seems misguided for two reasons. First, a global democracy might serve equally well as a mechanism whereby those with greater stakes, such as representatives of districts heavily involved in international trade, invest greater time and energy in the development of, e.g., international trade rules, while those with less at stake (or their representatives) devote less time and energy. Second, what is most important is not how much time and energy different agents devote to the development of international legal norms but how the authority to make those norms is distributed. If votes are equally distributed, despite unequal stakes, then Christiano is committed to the resulting law being illegitimate. The FDA’s true advantage over global democracy vis-à-vis the existence of unequal stakes is that states enjoy moral discretion over whether to enter into international 2 Lefkowitz.indd 202 21/4/17 13:30 Democracy, Legitimacy, and Global Governance 203 LEAP 4 (2016) agreements, and the terms on which they are willing to do so. The FDA model conceives of much new international law not as a set of impartial rules that aim to promote the common good but as mutually advantageous arrangements agreed to under free and fair conditions by parties pursuing their own interests on the basis of their relative bargaining power. The upshot is that those states with greater stakes will likely exercise a greater say in the construction of this kind of international law than will states with lesser stakes, since those are the terms on which it will be rational for both parties to converge. And as long as the resulting legal regimes are both genuinely morally optional (i.e. not required for the just treatment of others) and entered into under free and fair conditions, the resulting norms will be consistent with the publicly equal treatment of all. One final advantage Christiano attributes to the FDA model of global governance over global democracy is that the former is less vulnerable to the problem of citizenship than the latter. Note, first, that even if this is true it is not clear that the FDA model of global governance mitigates the problem of citizenship to a degree sufficient to render the resulting norms legitimate. The extent to which individuals are informed about and take responsibility for the content of international law may still be so slight that it does not warrant the belief that international law publicly treats them all as equals. Second, global democracy may offer avenues for representation the FDA does not that serve to galvanize a more informed and invested citizenry. For example, global democracy might facilitate a greater voice for views that are in the minority domestically, whereas negotiations between democratic states might well present only the views of the domestic majority. In addition, by increasing the number of institutions that might assert a right to govern a particular domain of conduct, global democracy could also foster the kind of forum shopping that can both lead to creative solutions to conf licts over the demands of justice and enhance agents’ belief that the overall system of governance exemplifies a commitment to the equal advancement of interests (see, e.g. Berman 2014). Third, and most importantly, when viewed through the lens of the problem of citizenship the key distinction between a FDA and global democracy appears to be whether international legislators are to be directly elected or indirectly elected; for example, appointed by domestic legislators who are themselves directly elected, as was originally the method for selecting United States Senators. But which of these two models of representation we should adopt is a separate matter from the question of whether states and the individuals they represent have a duty to submit to a common legal order. Thus we might argue that becoming a party to a particular legal regime, e.g. one governing trade in a particular class of goods, is morally optional, while also maintaining that if the regime creates a somewhat 2 Lefkowitz.indd 203 21/4/17 13:30 204 David Lefkowitz LEAP 4 (2016) independent institution charged with developing this body of law its officials ought to be directly elected rather than appointed by the domestic legislatures of its member states. A possible response to the foregoing argument would be to contend that: (1) in certain circumstances, which presently obtain internationally, parties have a moral duty to submit to a common legal order governing some specified type of activity; but (2) because the parties do not have roughly equal stakes in the activity governed by the legal regime in question, its norms should not be made by a democratic decision-procedure that accords them all an equal voice. Though Christiano appears to believe that at present the first of the aforementioned conditions is rarely met, Laura Valentini has recently defended this position, arguing explicitly against Christiano that democratic legitimacy does not require an equal say but only a say proportional to the stakes individuals have in the resulting law (Valentini 2014; see also Brighouse and Fleurbaey 2010). As I will now demonstrate, however, the examples she gives to illustrate the joint satisfaction of the two conditions specified above fail to do so. Thus it remains unclear whether once we have limited the scope of democratic authority to important interests (as Valentini maintains we should) it will still be the case that the parties in question have different stakes in the decision. Valentini asserts: “it is unreasonable to deny that, say, the inhabitants of Bangladesh have a greater stake in decisions about how to deal with anthropogenic climate change than the residents of the United Kingdom” (Valentini 2014: 795). Presumably that is because climate change poses a greater threat of harm to them than it does to the lives of UK residents. In the short term that is likely true, but in the long term it is not. British and Bangladeshis have equal interests in the adoption of climate policies that cap the increase in the Earth’s average global temperature, but they may well differ with respect to what the optimal increase is. Claiming that Bangladeshis have a greater interest in a lower peak average global temperature than do residents of the UK is not the same as maintaining that they have a greater interest in the question of what the optimal level of climate emissions are, or what policies ought to be adopted in pursuit of that optimum. Climate policy, then, looks like a matter (one of very few, perhaps) in which all people on Earth have a significant and roughly equal stake. Valentini’s second example concerns laws aimed at facilitating access to public spaces for the disabled. “Legislation about disabled access to public spaces has greater impact on people with disabilities than on the 2 Lefkowitz.indd 204 21/4/17 13:30 Democracy, Legitimacy, and Global Governance 205 LEAP 4 (2016) rest of a country’s population”, which, she implies, entails that the disabled ought to have proportionally greater say in the crafting of such legislation than do the able (Valentini 2014: 795). Like the example of climate policy this argument begins in the wrong place, namely with the effects of the legislation rather than the interests in the activity it regulates. Access to public spaces is something in which all individuals have an equal interest (or so I shall assume), and so all ought to have an equal say in the crafting of legal regulations that specify what those who maintain public spaces must do to advance these interests. Deviation from a procedure that does so is called for only if the disabled turn out to be a persistent minority (as, in fact, has often been the case); that is, where an individual-majoritarian decision-procedure has persistently failed to correct the cognitive biases of the able regarding what the equal advancement of the interest in access to public spaces requires, and left the disabled feeling both alienated from society and not recognized as moral equals. While this argument sanctions an unequal say in making law, the rationale for doing so does not rest on unequal interests in the issue regulated by the law in question but the persistent failure of a process that accords an equal say to all entitled to it to generate just law. Of course, these criticisms of Valentini do not demonstrate the impossibility of satisfying both of the conditions set out above. Absent a successful illustration, however, we have no reason to believe there are any cases in which a given type of conduct affects a set of individuals’ important interests in ways that morally require them to submit to a common legal order governing that conduct, but where those individuals have significantly different stakes in what the content of those legal rules turns out to be. At present, the common legal rules to which all agents, or the political communities of which they are members, have a moral duty to submit are largely those that serve to preserve the independence (or non-domination) of the distinct common worlds that exist and are partly constructed by their domestic legal orders. These are the core rules of Westphalian International Law, e.g. those that ban aggressive war and intervention in the domestic affairs of other states, or that internalize externalities by, for instance, allocating responsibility for cross-border pollution. A world in which such rules were respected would be one in which no individual’s fundamental interests were setback by the conduct of agents who reside in other states. Again, that is not to say that in such a world the activities of individuals in one state would not impact the lives of those living in others, nor does it deny that all might stand to gain by the adoption of a common set of legal norms that eroded sovereignty for the purpose of creating a 2 Lefkowitz.indd 205 21/4/17 13:30 206 David Lefkowitz LEAP 4 (2016) partial common world (e.g. a common market in certain goods). Neither of these facts, however, entails that individuals can treat people in other states justly only by submitting to common legal rules that take the place of or circumscribes existing domestic law. 2. AN INSTRUMENTAL ARGUMENT FOR INTERNATIONAL LAW’S LEGITIMACY Might international law enjoy some legitimacy even if it is the product of a legislative procedure that fails to fully conform to the principle of public equality? Though he sometimes appears to think otherwise, in this final section I argue that Christiano ought to give an affirmative answer to this question. Recall that to characterize law as legitimate is to maintain that when they deliberate its subjects have a moral reason to defer to its judgment regarding what they may, must, or must not do even in a range of cases in which the law’s judgment conf licts with their own. For Christiano law is fully (and inherently) legitimate if and only if it is the product of a law- making process that satisfies the principle of public equality: specifically, a democratic decision-making process in the case of a common world, and free and fair agreement where submission to a common rule is morally optional. At least in circumstances characterized by the facts of judgment, it is only by guiding their conduct according to law made in a way that satisfies the principle of public equality that individuals can advance or honor one another’s fundamental interests in judgment. These include the interest in correcting for others’ cognitive biases, the interest in being at home in the world, and the interest in being treated by one’s fellows as a person with equal moral standing. The value of the first of these interests appears to be largely instrumental; that is, the ability and opportunity to correct others’ cognitive biases is valuable primarily because – and to the extent that – it makes it more likely that individuals will treat one another justly. Or perhaps the point would be better put in terms of reducing the incidence and severity of unjust treatment. Regardless, suppose we concede arguendo that law made democratically or agreed to in free and fair conditions best serves the aim of advancing justice (or reducing injustice) by combatting cognitive bias. Nevertheless, other legislative procedures that perform less well in this respect may perform well enough that their subjects do better at treating one another justly by obeying the law than by acting on their own judgment, even in a range of cases where they think it substantively mistaken. If so, the law produced by such procedures will enjoy some legitimacy in virtue of its advancing individuals’ 2 Lefkowitz.indd 206 21/4/17 13:30 Democracy, Legitimacy, and Global Governance 207 LEAP 4 (2016) fundamental interests in having others, and their own, cognitive biases corrected. The form of the foregoing argument is most closely associated with Joseph Raz’s service conception of legitimate authority. Raz maintains that A enjoys legitimate authority over B if the following two conditions are met (Raz 2006: 1014): (1) The Normal Justification Condition (NJC): The subject would better conform to reasons that apply to him anyway (that is, to rea- sons other than the directives of the authority) if he intends to be guided by the authority’s directives than if he does not. (2) The Independence Condition (IC): The matters regarding which the first condition is met are such that with respect to them it is better to conform to reason than to decide for oneself, unaided by authority. Where the NJC and IC are met, B has a duty to defer to A’s judgment; that is, to act as A directs him to act even if B believes he has an undefeated reason to act otherwise. As I understand it, the service conception provides a formal analysis of legitimate authority, by which I mean it tells us the kind of argument we must offer if we are to substantiate or successfully contest the claim that a putative authority, such as international law generally or WTO law in particular, enjoys legitimacy. But the service conception itself tells us neither what reasons apply to agents independently of the law, nor how obedience to law serves to advance our conformity to those reasons. For example, the service conception does not rule out the possibility that individuals have fundamental interests in being at home in an egalitarian world and in recognition as a moral equal that all agents have reason to advance, nor the possibility that obedience to law can facilitate their doing so by constituting the advancement of those interests.1 Theorists may dispute the existence or the importance of these interests, or the ways in which law may enhance its subjects’ responsiveness to them, without disagreeing over the general account of legitimate authority provided by the service conception.2 Our concern, however, is whether law can enjoy legitimacy solely in virtue of its content ref lecting less cognitive bias than does the judgment of its subjects. It seems obvious to me that it can, and that where the law’s judgment regarding the demands of justice suffer from less cognitive bias 1 On Christiano’s account, law serves to advance those interests in this way if it is the product of a process that satisfies the principle of public equality. 2 Christiano acknowledges as much; see Christiano (2008: 55). 2 Lefkowitz.indd 207 21/4/17 13:30 208 David Lefkowitz LEAP 4 (2016) than do the judgments of its subjects, they have a duty to obey it.3 But Christiano’s admonition that justice must not only be done, but be seen to be done, rings true even where we are concerned only with law as a means to realizing justice (or mitigating injustice). While the service conception tells us when, as an objective matter, a legal subject has a duty to obey the law, we also need an account of how we are to identify when a putative authority satisfies the service conception. In the case of democratically enacted law, the right to an equal say serves both to correct cognitive bias and to provide subjects with reason to believe that the resulting law ref lects a good faith effort to equally advance the interests of all (enfranchised citizens) even where the content of the law strikes some subjects as at odds with that aim. Might legislative processes that are neither democratic nor voluntary agreements reached under free and fair conditions satisfy this evidential demand as well, so that they not only satisfy the NJC by, at a minimum, reducing some of the injustice individuals’ cognitive biases can cause, but can be reliably identified as doing so? I think the answer is yes, and at least with respect to the law of international organizations such as the WTO, Allen Buchanan and Robert Keohane’s standard of complex legitimacy provides one example of an institutional design that would do so. The complex standard of legitimacy consists of a set of substantive and procedural requirements that, when met, provide evidence for the legitimacy of a global governance institution’s attempt to rule (Buchanan and Keohane 2006). The former include not persistently violating the least controversial human rights, and not intentionally or knowingly engaging in conduct at odds with the global governance institutions’ purported aims and commitments. The latter include mechanisms for holding global governance institutions accountable for meeting the aforementioned substantive requirements, as well as mechanisms for contesting the terms of accountability. To be effective, these mechanisms must be broadly transparent; e.g. information about how the institution works must be not only available but also accessible to both internal and external actors, such as inspectors general and non-governmental organizations. What unifies the various elements of the complex standard is that they all provide the legal subjects of global governance institutions with reason to believe that officials in these institutions are making a good faith effort to determine what justice requires. In the absence of one or more elements of the complex standard those subject to a global governance institution’s rule may (rightly) suspect that governance is being exercised in pursuit of 3 For descriptions of some of the ways in which international law can serve as a check on judgments of justice distorted by international actors’ predictable cognitive biases, see Tasioulas (2010); Lef kowitz (2016). 2 Lefkowitz.indd 208 21/4/17 13:30 Democracy, Legitimacy, and Global Governance 209 LEAP 4 (2016) other goals, such as the national interests of powerful states. Consider, for example, the substantive elements of the complex standard: no attempt at international governance by either global governance institutions or by states that persistently violated “the least controversial human rights”, or that systematically discriminated in the application and enforcement of international legal norms, could plausibly claim to be making a good faith effort to enhance its subjects’ conformity to the demands of justice. Similarly, the procedural elements that compose the complex standard evidence a good faith effort to determine what right reason requires because they militate against efforts to deploy international law for private interest rather than the public good. Christiano acknowledges the value of reforming global governance institutions so that they satisfy the complex standard of legitimacy, but denies that such reforms could render their rule legitimate. Something like complex legitimacy, he writes, may give us reason to think that the institutions will produce minimally desirable outcomes. We may often have reason, therefore, to go along with those outcomes. But it does not give us the kind of moral legitimacy that implies reasons to go along with them even when we disagree with the outcomes (Christiano 2011: 94). It seems to me that Christiano makes the perfect enemy of the good, and in doing so downplays two crucial considerations. The first is that our own judgments regarding the justice of the outcomes of global governance institutions that satisfy the complex standard necessarily, and predictably, ref lect our biases and fallibility. In acting on those judgments, therefore, we may be less likely to treat others justly than if we obey the law. The second is that the law’s legitimacy requires only that its subjects be more likely to “get it right” by deferring to it than by acting on their own judgment. In circumstances where domestic political officeholders generally know very little about the interests of people living in other states and act within an institutional structure that provides them with a strong incentive to be unjustifiably biased toward the interests of citizens and against the interests of foreigners, the bar for international law’s legitimacy may be set quite low. Indeed the complex standard suggests as much. Thus I maintain that satisfaction of the complex standard of legitimacy does provide those subject to the resulting law with a duty to defer to it, a presumption in favor of doing so sufficiently weighty to warrant conformity to the law even in some range of cases in which agents believe the law is mistaken on its merits. In The Constitution of Equality Christiano maintains that the mere fact that one is more likely to act as one has most reason to act by obeying the 2 Lefkowitz.indd 209 21/4/17 13:30 210 David Lefkowitz LEAP 4 (2016) law than by acting on one’s own judgment cannot provide a sufficient condition for the legitimacy of domestic law.4 If it did then individuals could have a duty to obey (some of ) the law of deeply unjust states, but that is absurd. Such states often “implicitly threaten morally terrible consequences if their subjects do not comply with commands that require them to participate in evil activities” (Christiano 2008: 234). Christiano maintains, however, that: “even if complying without question is the right thing to do, the authority that issues the directives is clearly not legitimate” (Christiano 2008: 234). Might a version of this argument apply to international law, either in general or vis-à-vis specific international legal regimes? Note, first, that the complex standard of legitimacy may well satisfy Christiano’s demand that a political institution “have some reasonable degree of justice” in order to be legitimate. But second and more importantly, as I argued above, the significance for a political institution’s legitimacy of its satisfying certain minimal demands of justice is partly epistemic. Where the institution fails to do so, its subjects have no reason to believe that it meets the NJC. As Christiano notes, they may still judge that they will do best by conforming to the unjust state’s laws, or even treating its laws as if they were authoritative. Their operative reason for doing so, however, likely will not (and should not) be the belief that the unjust state is more likely than they to determine what justice truly requires of them. Moreover, this conclusion holds even in those cases where, as a matter of fact, the unjust state is more likely than its subjects to discern what justice truly requires of them.5 Where the NJC is satisfied those whose just treatment is at issue have a claim against the law’s subjects that they obey it rather than act on their own judgment. Those who are subject to the rule of a deeply unjust state are unlikely to be at fault for failing to discharge this duty, however; after all, they have little or no reason to believe they have it. As noted above, the foregoing argument rests on a distinction between what it is for A to enjoy legitimate authority over B, namely that B ought to act as A directs rather than on her own judgment, and the reasons that justify A’s legitimate authority over B; that is, the reasons why B ought to 4 Christiano develops this argument as an objection to the NJC, but in light of the earlier discussion in the text I think it better to construe it as an objection to a specific way in which law can serve to enhance its subjects just conduct (or at least reduce the injustice they commit). 5 One source of hesitation to embrace this conclusion may be the thought that no one can owe obedience to a political institution that perpetrates grave injustices. W here law’s legitimacy is a matter of it increasing the likelihood that its subjects will act justly, however, the duty to obey is owed not to the law (or legal officials) but to those the law’s subjects are more likely to treat justly by obeying the law than by acting on their own judgment. See Lef kowitz (2016) for discussion of this point. 2 Lefkowitz.indd 210 21/4/17 13:30 Democracy, Legitimacy, and Global Governance 211 LEAP 4 (2016) act as A directs. One advantage to drawing this distinction is that it allows us to focus on the most prominent substantive divide among theorists of legitimate authority, i.e. whether the exercise of moral judgment warrants respect per se, independent of its veracity, rather than getting bogged down in definitional battles.6 A second advantage to foregrounding the distinction drawn above is that it enables the concept of legitimacy to play a role in both ideal and non-ideal theories of global governance. Christiano’s FDA may model legitimate authority in an ideal moral community, and as such it may provide a lodestone for long-term reforms to the global political order. In the near and medium-term, however, the extent to which the current world order deviates from that ideal may render the purely instrumental accounts of international law’s legitimacy more important, both for rebutting those who deny that international law enjoys any legitimate authority and as a guide to feasible reforms that can begin to mitigate the extent to which international law and institutions serve merely as tools for the powerful. BIBLIOGRAPHY Archibugi, D. and Held, D., 1995: Cosmopolitan Democracy: An Agenda for a New World Order, Oxford: Blackwell Publishers. Berman, P., 2014: Global Legal Pluralism, New York: Cambridge University Press. Brighouse, H. and Fleurbaey, M., 2010: "Democracy and Proportionality”, Journal of Political Philosophy 18: 137-155. Buchanan, A. and Keohane, R., 2006: “The Legitimacy of Global Governance Institutions”, Ethics and International Affairs 20: 405-437. Christiano, Th., 2008: The Constitution of Equality: Democratic Authority and Its Limits, New York: Oxford University Press. —2010: “Democratic Legitimacy and International Institutions”, in The Philosophy of International Law, eds. S. Besson and J. Tasioulas, 119-138, New York: Oxford University Press. —2011: “Is Democratic Legitimacy Possible for International Institutions?”, in Global Democracy: Normative and Empirical Perspectives, eds. D. Archibugi et al., 69-95, New York: Cambridge University Press. —2013: “Authority”, The Stanford Encyclopedia of Philosophy, ed. E. N. Zalta, URL = . Lef kowitz, D., 2016: “The Legitimacy of International Law”, in Global Political Theory, eds. D. Held and P. Maffettone, London: Polity Press. Raz, J., 2006: “The Problem of Authority: Revisiting the Service Conception”, Minnesota Law Review 90: 1003-1044. 6 “It is not a useful aim of philosophers or political thinkers to determine which one of these conceptual accounts of political authority is the right one” (Christiano 2013). 2 Lefkowitz.indd 211 21/4/17 13:30 212 David Lefkowitz LEAP 4 (2016) Tasioulas, J., 2010: “The Legitimacy of International Law”, in The Philosophy of International Law, eds. S. Besson and J. Tasioulas, 97-116, New York: Oxford University Press. Valentini, L., 2014: “No Global Demos, No Global Democracy? A Systematization and Critique”, Perspectives on Politics 12 789-807. 2 Lefkowitz.indd 212 21/4/17 13:30