A Reply to Dworkin's New Theory of International Law

Adam S. Chilton

80 U Chi L Rev Dialogue 105 [Essay PDF]


     Scholars have long debated whether states are obligated to follow international law.1 In a posthumously published article, Professor Ronald Dworkin recently contributed to this debate, arguing states have a prima facie obligation to follow international law.2 Professor Dworkin suggests that this obligation arises not because the international legal system is based on consent (as many have suggested), but instead because states are obligated to improve their political legitimacy, and international law can help to do so by correcting the shortcomings of the state-sovereignty system. That is, international law can help provide a check against states that would abuse their own citizens, or can help compensate for the fact that states acting alone cannot solve global problems requiring coordination. Professor Dworkin argues that this theory has the advantage of both justifying the sources of international law—such as customary international laws states cannot opt out of—and providing a principle to guide international law’s interpretation.
     Professor Dworkin’s theory, however, is at best incomplete and at worst fatally flawed; it may provide an account of why international law should be binding over autocratic states that would shirk their obligations to their own citizens and others, but it does not explain why democratic states have a general obligation to comply with laws they disagree with or that are against their interest. Moreover, if the international community took Professor Dworkin’s theory of interpretation seriously, it would result in states being less willing to negotiate deep international agreements in the future.
     In this Essay, I first attempt to accurately and charitably summarize the “new philosophy” of international law advanced by Professor Dworkin.3 Second, I outline three flaws with Professor Dworkin’s argument.

I.  Dworkin’s “New Philosophy for International Law”

     According to Professor Dworkin, in the middle of the twentieth century, theorists of international law were primarily interested in a single existential question—is there international law?4 The question was not whether there were documents that legal academics or diplomats referred to as international law—such as multilateral treaties—but instead, whether these sources of law placed binding obligations upon states. Although in the eyes of many the moment for this debate may have passed, Professor Dworkin’s essay was motivated by the view that this existential challenge to international law remains. Professor Dworkin argues, “Even though almost everyone agrees that ‘international law’ is really law, . . . the question of why these documents constitute some kind of legal system is crucial because how these rules and principles should be interpreted hinges on it.”5
     To situate this debate, Professor Dworkin outlines why he thinks many serious legal scholars doubted that “there was any such thing as international law.6 This doubt arose not because the empirical facts were radically different in 1950; in fact, many of today’s leading international institutions and international agreements were already in place. Instead, Professor Dworkin suggests that the reason skepticism of international law was pervasive within the academy was because of the popularity of legal positivism. Simply put, legal positivists argue that whether law on a given topic exists is a historical fact.7 In Professor John Austin’s early articulation of this theory, law is the command of a sovereign over a given territory. That is, law exists when the king or parliament with control of a state lays down a rule.8 Given that there is not an international sovereign, it is of course easy to see that the logical implication of Professor Austin’s theory was that there is not international “law.”9 A more recent version of positivism, championed by Professor H.L.A. Hart, instead argues that law exists when a given political community has accepted two kinds of rules: secondary rules, establishing how laws are created; and primary rules, created by following the secondary rules.10 Although this theory does not require the presence of an absolute sovereign over a territory, it remains committed to Professor Austin’s view that law is a historical—and not natural—fact.
     Although Professor Hart did not try to extend his theory from the domestic to the international sphere, many international lawyers have taken Professor Hart’s version of legal positivism as a starting point to build a doctrinal account of international law.11 In Professor Dworkin’s account, these scholars take the view that whether a particular rule or restriction is “international law” is a matter of history. The question that they have then tried to tackle is: What are the secondary rules that have been agreed upon to determine whether a primary rule is in fact law? The answer that has been settled upon is consent. Sovereign states are bound by international legal obligations when they have previously consented to them. States consent to be bound by treaties when they sign them, and states consent to customary international law by assuming in their practices that they are bound to follow a given rule.12 The advantage of this argument is simple: it solves the paradox of how a state can both be sovereign and be subject to laws.13
     Professor Dworkin strongly disagrees with this position and instead argues, “we cannot take the self-limiting consent of sovereign nations to be the basic ground of international law.”14 This is because, Professor Dworkin persuasively argues, there are several fatal flaws in viewing consent as the secondary rule establishing international law. First, Professor Dworkin points out, customary international law does not clearly establish how many states must adhere to a particular practice before it becomes customary.15 Second, there are “preemptory norms” of customary international law—such as prohibition of slavery—that states are widely viewed as being unable to consent out of.16 Third, when states sign treaties or engage in particular practices, it is unclear exactly what it can be said they are consenting to. After all, the meaning of treaties is often vague and must be expounded over time.17 Fourth, the logic of consent providing the secondary rules of international law is necessarily circular because, regardless of whether a given state has consented to that as the foundational principle, consent is nonetheless viewed as the basis of the international system.18 Finally, Professor Dworkin views it as fundamentally unfair that people could be bound by obligations that previous generations consented to through different political processes.19
     If international law is real and consent is not the basis for it placing obligations upon states, then what could be?  Professor Dworkin argues that what is needed is a more abstract view of the nature of law that is grounded in principles of political morality. To lay the foundation of this argument, Professor Dworkin advances a doctrinal concept of law that is interpretive and not criterial.20 That is to say, in a political community, we share a doctrinal concept of the law “not by agreeing about tests for application but by agreeing that something important turns on its application and then disagreeing, sometimes dramatically, about what tests are therefore appropriate to its use, given that its application has those consequences.”21
     Once Professor Dworkin asserts that we need an interpretive conception of what is doctrinally required by the law, he argues that the relevant question is: How can we understand international law as part of what morality requires of states?22 What is important to recognize, Professor Dworkin argues, is that unrestricted sovereignty hurts the citizens of states in a number of ways.23 First, unrestricted sovereignty may allow states to violate the rights of their own citizens.24 Second, respecting unlimited sovereignty may lead states not to intervene to help the citizens of other states when the most reprehensible crimes, such as genocide, are committed.25 Finally, unrestricted sovereignty allows states to subject citizens to risk in a more subtle way—by preventing those states from coordinating with other states to avoid prisoner’s dilemmas.26 That is, by failing to limit their sovereignty, states are unable to coordinate with other states in a way that eliminates the risk of potential environmental disasters like global warming.
     Given the risks that unmitigated sovereignty poses, Professor Dworkin argues states have an obligation to their citizens to take steps to “mitigate” the failures of the state sovereignty system.27 Professor Dworkin argues that the concept of mitigation thus provides the “most general structural principle and interpretive background of international law.”28 Once Professor Dworkin has slowly built the case for this general concept, he clarifies that a more specific articulation is the principle of “salience.”29 In Professor Dworkin’s words, the salience principle states:

If a significant number of states, encompassing a significant population, has developed an agreed code of practice, either by treaty or by other form of coordination, then other states have at least a prima facie duty to subscribe to that practice as well, with the important proviso that this duty holds only if a more general practice to that effect, expanded in that way, would improve the legitimacy of the subscribing state and the international order as a whole.30

     Professor Dworkin argues that the principle of salience not only provides a persuasive account of the moral foundations of international law, but also has the advantages of offering a compelling justification for the conventional sources of international law while additionally providing guidance on the interpretation of international law.31 In other words, the reason that customary international law and treaties provide the sources of international law is not just because countries have consented to them (because they have not always consented), but instead because these sources have become salient; and the sources that meet the standard of salience should then be interpreted “so as to advance the imputed purpose of mitigating the flaws and dangers of the [state-sovereignty] system.”32

II.  Limitations of Dworkin’s Theory

     It is undeniable that Professor Dworkin’s new theory of international law has many attractive features. Foremost among them, his theory outlines a justification for the sources of international law that scholars and diplomats point to as legitimate—even when they are not entirely consented to—while also providing an account of how those sources should be interpreted that is consistent with the way that many academics and policymakers think.33 In other words, Professor Dworkin’s theory provides an account for why, although consent is important, international law should still allow state action to prevent violations of rights that occur within sovereign states even in the absence of consent. But despite these appealing features of Professor Dworkin’s theory, it suffers from major shortcomings. In this Essay, I discuss three.
     First, Professor Dworkin does not provide an explanation of what, if anything, should give when domestic preferences and international obligations are in tension. Professor Dworkin reasonably argues that, for a state to be legitimate, citizens must play some “role in their own government.”34 Professor Dworkin specifically argues that “[p]olitical theorists disagree about what kind of participation is essential . . . but it is generally understood . . . that some form of widespread suffrage in the election of officials is both necessary and sufficient within a distinct political community.”35 Given that Professor Dworkin argues that political legitimacy requires giving people a voice in their government,36 what is unclear in Professor Dworkin’s theory is whether elected officials are still required to comply with international law when it is against the preferences of their citizens to do so.
     For example, one problem Professor Dworkin argues that more robust international law is required to solve is climate change.37 It is far from self-evident, however, that taking steps to address climate change right now is rational. Even if one holds the beliefs that climate change is occurring, its causes are manmade, the potential harms are high, and steps could be taken to curb it, it may still be rational hold the view that it would be a mistake to pass any legislation (domestic or international) aimed at addressing the problem. For example, Professors Eric Posner and Cass Sunstein have argued that addressing climate change requires making a payment now for a reward that will be enjoyed by those living in the future;38 and if one holds the reasonable belief that people in the future will be wealthier and more technologically advanced, then it might be unwise to make what is essentially a wealth transfer to future generations.39 If the majority of citizens of a state were to hold this view, it is quite plausible that their elected officials would prefer not to act to slow global warming for reasons other than simple political gridlock or shortsightedness.
     The question, of course, is whether those elected officials would still have an obligation to comply with international agreements aimed at addressing climate change. There are two answers to that question that may be permissible under Professor Dworkin’s theory. The first answer would be that legitimate governments—which Professor Dworkin claims must give citizens a voice—are not obligated to obey international law that their citizens do not support. The obvious problem with this answer is that, if legitimate governments are only obligated to comply with international laws they support, it is difficult to see how there would be much daylight between Professor Dworkin’s theory and the theories of international legal obligation based on consent that he so forcefully argues against. The second answer would be that elected officials are still required to comply, even if their citizens do not support compliance. For example, one might suggest that a legitimate government is obligated to provide aid to severely impoverished states even if this were not popular with the government’s citizens to do so. Having this obligation rest with the elected officials and not the citizens, however, requires an argument for cosmopolitanism at the expense of democracy that may do more to undermine political legitimacy—one of the principles motivating Professor Dworkin’s theory—than the principle of salience he supports.
     Second, Professor Dworkin’s theory does not provide an account of why states should be bound by international law when the coordination problem that states face is not a prisoner’s dilemma. Professor Dworkin argues that it is difficult for states to address many problems—like climate change or overfishing40—because they face a classic prisoner’s dilemma.41 That is, the states would be better off if they all took a certain action, but the dominant strategy is to defect and not to cooperate unless coordination is possible. The shortcoming with Professor Dworkin’s argument is that not all coordination problems are prisoner’s dilemmas. Instead, there are many coordination problems where all parties would not be better off if a coordinated response occurred.
     This point can also be illustrated by the climate change example. There are certainly many states that would likely benefit from coordination on climate change and some states that may face catastrophic consequences if that coordination does not occur (for example, Pacific island nations). There are other states, however, that may stand to gain from global warming (for example, Russia and Canada).42 The fact that there are opposite expected payouts for coordination, instead of evenly distributed payouts, suggests that addressing climate change is one of the many examples of international coordination problems that is not a prisoner’s dilemma subject to Professor Dworkin’s theory.
     The natural question, of course, is whether compliance should still be demanded in cases where a given state stands to lose from coordination. As with the last objection I raised, the options available to Professor Dworkin’s theory similarly appear to be either to allow states not to comply when it would not be in their interest to do so, or to demand that states comply even if it were against their interests. For example, it is possible to argue that states are obligated to comply with individual laws that are against their interest because they are still better off from the international system as a whole.43 Once again, this obviously would require elected officials in democratic states to act against the interests of their citizens in a specific case to generate a larger benefit to the international system, and it is unclear how this is consistent with Professor Dworkin’s argument that states must increase their political legitimacy.
     Third, if Professor Dworkin’s theory of how sources of international law should be interpreted was taken seriously by international courts and organizations, it would run the very serious threat of causing states to be unwilling to negotiate robust agreements in the future. This is because Professor Dworkin’s theory fails to account for the fact that generating new sources of international law is a repeat game. According to Professor Dworkin, sources of international law should be interpreted to provide robust checks against state sovereignty, even if there was not initial consent to that agreement.44 The risk that this approach would run, however, is that, if states begin to be held to more demanding standards than they thought had previously been agreed upon, in future negotiations those states would have strong reasons to block even weak language in international agreements to avoid it being held against those states later on.
     For example, if the International Court of Justice (ICJ) were to take the broad reading of treaties that Professor Dworkin suggests, with the right case the ICJ may decide to hold that the United States is violating international agreements that it has signed by holding prisoners in solitary confinement.45 If this were to happen, it is very possible that the United States would respond with some combination of watering down future agreements, withdrawing from current agreements, and trying to curb the authority and funding of the ICJ.46
     The reason that this result would be particularly problematic is that international law may be more effective when used as a tool by domestic political actors than when used by international courts and organizations. For example, the weight of the empirical evidence suggests that the reason international law helps to improve human rights is not because of international interpretations or enforcement, but instead because domestic political actors are able to use prior international commitments as powerful political tools when lobbying their government for change.47 In other words, if we actually were concerned with states violating the rights of their citizens, we should not encourage the ICJ and other international institutions to start reading human rights treaties more broadly. Instead, we should find ways to make sure that the international community does not take steps to discourage states from signing onto strongly worded agreements, so that those agreements can later help in domestic political struggles to improve human rights chances.


     Professor Dworkin’s theory may justify the sources of international law that are currently generally accepted, but it falls short of establishing a moral argument for an obligation of compliance. This is because the theory ignores the crosscutting obligations that domestic political demands put on states and the potential that democratic political processes have to use international law as an instrument of change. Both of these concerns should perhaps lead us to be skeptical of claims of prima facie obligations to comply with international law. As a result, instead of trying to develop a unified theory of the moral obligations created by international law, it would perhaps be more productive to continue the discussion that Professor Dworkin has started while being mindful of the emerging empirical evidence on the conditions under which international law can help to alleviate the excesses of the state sovereignty system.

           Bigelow Teaching Fellow and Lecturer in Law, The University of Chicago Law School; PhD in Political Science, Harvard University Department of Government; JD, Harvard Law School. I am grateful to Britt Cramer, Ryan W. Davis, Richard Fallon, and Greg Reilly for helpful conversations and advice.

       1     For a sense of this debate, compare Jack L. Goldsmith and Eric A. Posner, The Limits of International Law 105 (Oxford 2005) (arguing states are not morally obligated to comply with international law when it would be efficient to breach), with Liam Murphy, The Normative Force of Law: Individuals and States (NYU Law School Working Paper 2012) (on file with author) (arguing states are obligated to follow international law because failure to do so would threaten the international system).

      2     See generally Ronald Dworkin, A New Philosophy for International Law, 41 Phil & Pub Aff 2 (2013).

      3     There are two reasons for providing a detailed and careful account of Professor Dworkin’s theory. The first is that this theory is both new and dense, so there is value in undertaking the exercise of trying to distill it into an easily understandable form. The second is that only by providing a detailed account of Professor Dworkin’s theory will my later criticisms be clear.

      4     See Dworkin, 41 Phil & Pub Aff at 2 (cited in note 2).

      5     Id at 3.

      6     Id.

      7     Id.

      8     See John Austin, Lectures on Jurisprudence 182–85 (Spottiswoode 1869) (Robert Campbell, ed).

      9     See id.

      10    See H.L.A. Hart, The Concept of Law 80–91 (Oxford 1961). See also Dworkin, 41 Phil & Pub Aff at 4 (cited in note 2).

      11    See Dworkin, 41 Phil & Pub Aff at 5 (cited in note 2).

      12    See id at 6.

      13    See id.

      14    Id at 10.

      15    See Dworkin, 41 Phil & Pub Aff at 7 (cited in note 2).

      16    See id.

      17    See id at 7–8.

      18    Id at 9.

      19    See Dworkin, 41 Phil & Pub Aff at 9–10 (cited in note 2).

      20    See id at 11.

      21    Id.

      22    Id at 13.

      23    See Dworkin, 41 Phil & Pub Aff at 17 (cited in note 2).

      24    See id.

      25    See id at 17–18.

      26    See id at 18.

      27    Dworkin, 41 Phil & Pub Aff at 19 (cited in note 2).

      28    Id.

      29    Id.

      30    See id at 19.

      31    See Dworkin, 41 Phil & Pub Aff at 20–22 (cited in note 2).

      32    Id at 22.

      33    These sources are listed in Article 38(1) of the Statute of the International Court of Justice. They include: (a) international conventions, (b) international custom, (c) the general principles of laws recognized by civilized nations, and (d) judicial decisions and writings of “highly qualified publicists” of international law. Statute of the International Court of Justice, Art 38(1), 59 Stat 1055, 1060, Treaty Ser No 933 (1945). For a discussion on this issue, see Dworkin, 41 Phil & Pub Aff at 5–6 (cited in note 2).

      34    Dworkin, 41 Phil & Pub Aff at 18 (cited in note 2).

      35    Id.

      36    See id. It should be noted that some may argue that this is not a logical requirement. It might be possible to hold the position that governments can be legitimate as long as they make consistently “just” decisions, regardless of whether there is input from the citizens.

      37    See id at 15, 18, 27.

      38    Of course, this could be either our future selves or those who have not yet been born.

      39    See Eric A. Posner and Cass R. Sunstein, Climate Change Justice, 96 Georgetown L J 1565, 1583–86 (2008).

      40    See Dworkin, 41 Phil & Pub Aff at 18 (cited in note 2).

      41    See note 26and accompanying text.

      42    See Laurence C. Smith, The World in 2050: Four Forces Shaping Civilization’s Northern Future 254–55 (Dutton 2010). See also Eric A. Posner and David Weisbach, Climate Change Justice 22–25 (Princeton 2010).

      43    For an articulation of this argument, see Murphy, The Normative Force of Law (cited in note 1).

      44    See Dworkin, 41 Phil & Pub Aff at 22 (cited in note 2).

      45    The argument that has previously been advanced is that solitary confinement constitutes cruel punishment that violates the Universal Declaration of Human Rights, the International Covenant on Civil and Political Rights, and the Convention against Torture. See Tracy Hresko, In the Cellars of the Hollow Men: Use of Solitary Confinement in U.S. Prisons and Its Implications under International Laws against Torture, 18 Pace Intl L Rev 1, 17–20 (2006). It’s worth noting that this is not a widely held position, but instead simply a conclusion that could be reached if an international tribunal decided to take the kind of expansive reading of international agreements that Professor Dworkin advocates.

      46    This scenario should not sound implausible to anyone familiar with the factual background of Medellin v Texas, 552 US 491 (2008). After an adverse ruling at the ICJ holding that the United States must allow foreign nationals that are arrested in the United States access to consular services from their country of citizenship, the United States withdrew from the Optional Protocol to the Vienna Convention that gave ICJ jurisdiction over the agreement. Id at 500.

      47    See Beth A. Simmons, Mobilizing for Human Rights: International Law in Domestic Politics 127–35 (Cambridge 2009) (using quantitative evidence to argue that prior ratification of human rights agreements influences human rights practices through domestic politics and not external enforcement); Xinyuan Dai, International Institutions and National Policies 69, 84–90 (Cambridge 2007) (making a similar argument about the environment).