The Role of Coercion in Law: An Analysis of International Law

Draft: Please do not cite. THE ROLE OF COERCION IN LAW: AN ANALYSIS OF INTERNATIONAL LAW SANDRA RAPONI ABSTRACT: International legal scholars and lawyers have primarily tried to refute the objection that international law is not really law because it lacks a supranational system of coercive sanctions by demonstrating that international law is in fact coercively enforced albeit it in decentralized ways. I will focus instead on the presumption behind this skeptical view – the idea that law must be coercively enforced. First, I argue that coercive enforcement is not conceptually necessary for law or legal obligations. Second, I consider the claim that coercive enforcement is nonetheless necessary for instrumental reasons. I argue that while physical coercion is instrumentally useful for increasing compliance in the domestic case, this is less effective and more problematic in the international case. In considering what then is essential and distinctive about law and what would increase the effectiveness of international law, I propose that, first, international law needs to be generally accepted as binding by states, officials, and other agents, and second, that international legal rules should be determinatively interpreted and applied by authoritative adjudicative and administrative institutions. In 2009, the American Society of International Law organized a panel at is annual meeting to discuss the question, “is international law law?” While many international legal scholars were frustrated with having to discuss this question and instead wanted to focus on how international law affects state behaviour, Oona Hathaway and Scott Shapiro argue that the question of whether international law is law still matters a great deal.1 I agree. Given that legal systems are morally valuable institutions, whether states and other agents ought to respect, support, or obey international law depends in part on whether it is considered to be real law.2 Reflecting on this question can expand our understanding of the nature of law in the domestic case as well. Those who continue to present the skeptical view that international law is not “law” often take the important features of domestic legal systems within modern states as paradigmatic of law and then apply this conception of law to the global level.3 International 1 Oona Hathaway and Scott J. Shapiro, “Outcasting: Enforcement in Domestic and International Law,” Yale Law Journal 121 (2011): 252-349, at 255. 2 Hathaway and Shapiro, at 255. 3 This model is supported by modern social theory. Thomas Hobbes, John Locke, and Immanuel Kant argue that a lawful condition in which rights can be secured requires a supreme sovereign with centralized legislative power that makes law, a centralized adjudicative power that interprets and applies the law to particular cases, and a centralized 2 law fails to satisfy this model since it does not have a world government that can make law and enforce these laws through a supranational system of sanctions, including an independent international military force. There are important practical and normative difficulties with creating this kind of a coercive world government. First, it seems unlikely that states would agree to relinquish enough of their military resources and personnel to the control of a world government. Second, even if a coercive world government could be created, there are good reasons to doubt whether it would function in democratic and legitimate ways. John Rawls wrote that he agreed with Kant that a world state (“a unified political regime with the legal powers normally exercised by central governments”) would either be a global despotism or would rule over a fragile empire torn by frequent civil strife as different peoples tried to regain their autonomy.4 Given these concerns, it is important to consider how international society can be governed by law without a coercive world government. In this paper, I will focus on the issue of coercive enforcement and the skeptical argument that since international law is not effectively enforced against states by a supranational system of coercive sanctions, it is not real law. I will use the term “coercive enforcement” to refer to the use of force or the threat of sanctions to increase compliance with the law.5 As Hathaway and Shapiro note, this seems to be the principle objection made by critics.6 This objection may seem persuasive since a distinctive feature of legal rules within coercive power that enforces the law through sanctions. Based on this view, some contemporary scholars argue that a world state with these three features is required for a lawful condition at the international level. For example, the following scholars argue for a coercive world state based on Kant’s legal theory: Sharon B. Byrd and Joachim Hruschka “From the State of Nature to the Juridical State of States,” Law and Philosophy: An International Journal for Jurisprudence and Legal Philosophy, (2008) 27.6: 599-641. Sharon B. Byrd and Hruschka, Joachim, Kant’s Doctrine of Right: A Commentary. (Cambridge: Cambridge University Press, 2012); Hoffe, Otfried. Kant's Cosmopolitan Theory of Law and Peace (Cambridge University Press, 2006). 4 John Rawls, The Law of Peoples (Cambridge: Harvard University Press, 1999), at 36. 5 This is the traditional conception of legal coercion in law and political theory. Others have proposed other views of coercion in law. For example, Ekow N. Yankah argues that we should not conflate the coerciveness of law with the idea that law is backed by the threat of sanctions. He argues that “coercion occurs when sufficiently high pressure is applied to compel the adoption of a certain course of action.” Ekow N. Yankah, “The Force of Law: The Role of Coercion in Legal Norms,” The University of Richmond Law Review, 42.5 (2008): 1195-1256 at 1216-17. Scott Anderson argues that the “enforcement approach” to coercion, which focuses on the coercier’s ability to inhibit actions by the use of force or threat of force, is a superior way to understand coercion and a more fundamental way to understand coercion in political, moral and legal philosophy than the “pressure approach.” Scott A. Anderson, “The Enforcement Approach to Coercion,” Journal of Ethics & Social Philosophy 5.1 (2010). 6 Hathaway and Shapiro, at 255-56. They cite Anthony D’Amato’s description of this objection amongst students of law: Many serious students of the law react with a sort of indulgence when they encounter the term "international law," as if to say, "well, we know it isn't really law, but we know that international lawyers and scholars have a vested professional interest in calling it 'law.'" Or they may agree to talk about international law as if it were law, a sort of quasi-law or 3 domestic legal systems appears to be the fact that, unlike moral rules, they can be coercively enforced by political institutions. While most legal philosophers largely agree with H.L.A. Hart’s persuasive rejection of John Austin’s command theory of law, the view that sanctions are a necessary, central, and distinguishing feature of law still persists, particularly outside legal philosophy.7 There are two main responses to the skeptical view. First, one can show that international law is sufficiently enforced without a world government or a supranational system of coercive sanctions, and second, one can reject the claim that coercive enforcement is required for law. Many scholars have defended international law by using the first strategy. While some argue that international law is sufficiently coercive since states can enforce it through military action and economic sanctions, others argue that compliance with international law can be achieved through less coercive means, and even through noncoercive means. I will focus on the second response and I will consider broader conceptual issues regarding the connection between law and coercive enforcement. In the first section of this paper, I will provide an overview of the main positions and arguments on this issue. In the second section, I will challenge the assumption that coercive enforcement is central and necessary for law by supporting H.L.A. Hart’s argument against John Austin’s command theory of law, while also considering some problems with applying Hart’s theory of law to the case of international law. While Hart’s argument shows that sanctions are neither central to law, nor sufficient for law, he does not argue that sanctions are not a necessary feature of law. To defend this stronger claim, I will appeal to Joseph Raz’s arguments on this issue in the third section. I will also consider recent objections to Hart and Raz by Frederick Schauer and Ekow N. Yankah. In the fourth and fifth sections, I will consider whether supranational coercive enforcement may nonetheless be required to increase the effectiveness of international law or for normative reasons. In response to the former, I will argue that it is important to keep such instrumental considerations distinct from the question of what is required for rules, norms, practices and institutions to qualify as “law,” and secondly, I will argue that there are significant problems with the use of centralized coercive sanctions in the international case that may not in fact make it as effective as is often assumed. Finally, I will address what is required to have law at the international level. If not coercive enforcement, then what? I propose that the most important elements for regarding international law as law and for developing the rule of law at the global level include the following. First, states and officials that administer and apply international law must near-law. But it cannot be true law, they maintain, because it cannot be enforced: how do you enforce a rule of law against an entire nation, especially a superpower such as the United States…? Anthony D’Amato, “Is International Law Really ‘Law’?” Northwestern University Law Review 79 (1985): 1293 at 1293. 7 Robert C. Hughes notes that while many legal philosophers have questioned the view that coercion is central and necessary for law, this view continues to be presupposed within political philosophy. See Robert C. Hughes, "Law and Coercion," Philosophy Compass 8.3 (2013): 231-240. Two recent articles within legal philosophy that defend the coercive feature of law are Ekow N. Yankah and Frederick Schauer. Frederick Schauer, “Was Austin Right after All? On the Role of Sanctions in a Theory of Law,” Ratio Juris, Vol. 23.1 (2010): 1-21. 4 generally accept that international law is binding on them. This provides reasons to focus more on increasing the perceived legitimacy of international law rather than on developing stronger sanctions. Second, there must be impartial and independent adjudicative and administrative bodies with the recognized authority to determine and apply legal rules. While various kinds of enforcement may be needed to increase the effectiveness of international law in practice, the authoritative and impartial application of law should be seen as having primary importance since it is necessary for determining when and how enforcement can be used. 1. OVERVIEW OF THE MAIN OBJECTIONS AND RESPONSES Allen Buchanan refers to the argument that without a global coercive power to effectively enforce international rules, they are not true law and cannot constitute a true legal system, as the “Legal Nihilist argument.”8 This position begins with a particular view of what features are required for a legal system, it finds that some of these features are lacking in the international case, and it then concludes that international rules are simply not law, as if by definition.9 The Realist view of international relations is even more skeptical about international law. It regards the relation between states as a perpetual Hobbesian state of nature in which states act solely based on their interests. Under the Realist view, while states may agree to be bound by certain treaties and rules of international law, they comply with these rules and treaties only when it serves their interests, and not because they are under any obligation to do so.10 This goes further than the Legal Nihilist view because the Legal Nihilist could still hold that there are obligations between states, but that without external coercive enforcement, these obligations are merely self-binding moral obligations rather than externally binding legal obligations.11 The Realist argues that states are neither bound by 8 Allen Buchanan, Justice, Legitimacy and International Law (Oxford University Press, 2005), at 46. 9 John Austin presents this argument. John Austin, The Province of Jurisprudence Determined. (Cambridge: Cambridge University Press, 1995), at 21 10 Hans J. Morgenthau is the most influential twentieth century proponent of political realism in international relations. Hans J. Morgenthau, Politics Among Nations: The Struggle for Power and Peace, 6th ed., (New York: Alfred A. Knopf, 1985); and Hans J. Morgenthau, In Defense of the National Interest: A Critical Examination of American Foreign Policy, (New York: Alfred A. Knopf, 1951). Other influential realists include E.H. Carr, Reinhold Niebuhr, and Kenneth Waltz (a neorealist). See Kenneth Waltz, Theory of International Politics (Addison-Wesley, Reading, MA, 1979). More recently, this view has been defended by Jack L Goldsmith and Eric A. Posner in The Limits of International Law. (New York: Oxford University Press, 2005). They analyze international politics based on rational choice theory. 11 This view is most clearly presented by John Austin. Jürgen Habermas also makes this argument in his criticism of Kant’s proposal for a voluntary league of states in "Kant's Idea of Perpetual Peace, with the Benefit of Two Hundred Years' Hindsight" in Perpetual Peace: Essays on Kant's Cosmopolitan Ideal, translated by James Bohman, edited by James Bohman and Matthias Lutz-Bachmann (Cambridge: The MIT Press, 1997). 5 legal nor moral obligations to comply with international legal rules and institutions; states only act in accordance with international law when it is in their interest to do so.12 The common response to the coercive enforcement objection by international legal scholars has been to argue that even without a supranational enforcement body, international law is still coercively enforced by states in a decentralized way. Although Kelsen considered law to be a coercive order that must be backed by sanctions, he regarded war and reprisals by states against other states as providing the necessary coercive element for law at the international level.13 He believed that appropriate sanctions for the violation of international law could be determined by just war theory.14 According to the UN Charter, the Security Council can authorize states to use forcible measures to enforce international law and to maintain or restore peace and security.15 While this seems like a good example of coercive enforcement, in practice, the power of the Security Council in this respect has been severely limited since the Council has been deadlocked by the veto power of its five permanent members. International law also includes the use of non-forcible countermeasures such as economic and trade sanctions in response to states that breach their treaty obligations.16 Given the non-uniform way these coercive measures are used, particularly since they rely on the willingness of states to use force or impose economic or trade sanctions, some skeptics are not satisfied with response. Realists point out that states are only willing to take such action 12 See Goldsmith and Posner. The Limits of International Law. Hans Kelsen. Principles of International Law. (New York: Rinehart & Company, Inc. (1952); and Hans Kelsen, "Sanctions in International Law Under the Charter of the United Nations." Iowa Law Review 31 (1945): 499. 14 He defines a sanction as “a coercive act provided for as the consequence of a definite conduct of the state, a forcible interference in the normally protected sphere of interests of the state responsible for this conduct.” Principles of International Law, at 19. In the case individuals, interferences in the normally protected spheres include depriving one of life, freedom and property. In the international case, sanctions include military action, confiscating property, and nonfulfillment of treaty obligations in relation to a state that has violated international law (24). 15 Only the Security Council can authorize the use of coercive force to enforce international law. The Charter prohibits the use of force by states except for self-defence and collective self-defence against attacks subject to the authority of the Security Council to take action to maintain or restore peace and security (Article 51). 16 Under customary international law, states can use a variety of non-forcible countermeasures, such as trade embargoes, freezing of assets, and the suspension of performance of treaty obligations in response to states that breach their treaty obligations. Non-forcible countermeasures to a state’s breach of its treaty obligations are part of customary international law and they have been recognized as such by International Court. There are four recognized conditions on the use of non-forcible countermeasures: (1) they must be intended to obtain redress for the wrong committed; (2) prior notification of the countermeasures and their purposes must be given; (3) they must be proportionate to the violations complained of; and (4) countermeasures which affect individuals are subject to certain limits deriving from human rights standards which form part of general international law. Ian Brownlie, The Rule of Law in International Affairs (M. Nijhoff Publishers, 1998), at 135. 13 6 when it serves their own national interests, such as when there are other reasons for military action or trade sanctions that will benefit them.17 This does not constitute the genuine enforcement of legal rules. Some contemporary international legal scholars have argued that it is a mistake to think of enforcement solely in terms of military action and economic sanctions. They argue that there are non-coercive means of increasing compliance with international law, such as through political pressure and the force of public opinion.18 For example, Scott Shapiro and Oona Hathaway have recently responded to the coercive enforcement objection by arguing that international legal institutions typically deploy “external outcasting” rather than physical force to enforce international legal rules.19 Member states can increase compliance with international law by simply denying the benefits of social cooperation and membership to disobedient states. This kind of response would be stronger if we could first challenge the underlying assumption that coercion is an essential or necessary feature for law. This leads us to the second line of response to the coercive enforcement objection – the argument that coercive sanctions are not a necessary or central feature of law, not even within domestic legal systems.20 2. HART’S REJECTION LEGAL PHILOSOPHY OF THE COMMAND THEORY OF LAW AND HIS CONTRIBUTION TO 2.1 THE COMMAND THEORY OF LAW: HOBBES & AUSTIN The view that effective, centralized enforcement through coercive sanctions is required for law is most strongly defended by Thomas Hobbes and John Austin. According to Hobbes, law is not counsel but commands backed by force: commands of a supreme sovereign addressed to those who are obliged to obey him. For Hobbes, the purpose of law is to provide order, stability and certainty. There is the constant threat of war between individuals in the state of nature because as individuals pursue their own interests, conflicts between them are inevitable, and each person will have her own conception of what is right. Hobbes argues that in order to leave this state of insecurity, a supreme sovereign is needed to conclusively determine what the laws are, to apply the laws to particular cases and settle conflicts, and to enforce these laws through the threat of force. In an important passage, Hobbes writes that there can be no covenants without the sword. Without a sovereign power with the capacity and authority to enforce contracts with the threat of punishment, individuals will have good reason to violate their agreements and the rights of others when it is in their best interest to do so, and when they can get away with it without suffering negative consequences. Hobbes has a prudential account of obligation. He argues that we cannot be obligated to do something 17 For example, Goldsmith and Posner, 28-29. Richard Falk argues that the civil society can provide “normative restraint” and force states to comply with their international obligations. Revitalizing International Law (Iowa State University Press, 1989), at 100. 19 Hathaway and Shapiro. 20 Buchanan and Anthony D'Amato present this argument. Anthony D’Amato, International Law: Prospect and Process (New York: Transnational Publishers, 1987). 18 7 that is contrary to our interests. If you and I make a contract, it is not rational for me to comply with my own contractual promises without any effective assurance that you will respect our agreement, since I would be acting against my own interests.21 The threat of punishment gives everyone a reason to comply, and consequently, it gives each individual effective assurance that others will follow their agreements. The same argument applies to any social contract between people. In order for us to be obligated to follow any rules, these must be first enforced by a sovereign power. John Austin similarly conceives of law as commands by an “uncommanded commander” backed by the threat of sanctions.22 Like Hobbes, he believes that it is the sanction itself that creates the obligation, and that without the sanction, there is no legal obligation. He writes that commands that are not backed by sanctions are not properly called laws.23 Austin concludes that international law is not law but “positive morality” because it lacks coercive sanctions.24 2.2 HART’S REJECTION OF THE COMMAND THEORY CONCEPTION OF LEGAL OBLIGATION OF LAW AND THE NORMATIVE Hart’s persuasive rejection of Austin’s command theory of law can be used to challenge coercion-based views of law that hold that the existence of a central coercive power to enforce 21 Hobbes writes in Chapter 14: “For he that performs first has no assurance the other will perform after; because the bonds of words are too weak to bridle men's ambition, avarice, anger, and other Passions, without the fear of some coercive Power. … But in a civil estate, where there is a Power set up to constrain those that would otherwise violate their faith, that fear is no more reasonable; and for that cause, he which by the Covenant is to perform first, is obliged so to do. Thomas Hobbes, Leviathan (London: Penguin Books, 1968), at 196. 22 Austin argues that a “command is distinguished from other significations of desire, not by the style in which the desire is signified, but by the power and the purpose of the party commanding to inflict an evil or pain in case the desire be disregarded.” The Province of Jurisprudence Determined, 21. 23 Austin, 33, 135-136. 24 Austin, 171. The term “positive morality” refers to the de facto moral rules that happen to exist in international society. A.V. Dicey similarly described international law as consisting of “rules of public ethics, which are miscalled international law.” A.V. Dicey, Lectures Introductory to the Study of the Law of the Constitution (Macmillan, 1885), at 23; cited in Simon Chesterman, “An International Rule of Law?” American Journal of Comparative Law, 56, (2008), 331-361, at FN150. As I noted above, this assessment of international law has also been presented by Habermas in his article on Kant’s “Perpetual Peace.” Based on Kant’s legal theory, Habermas argues that legal obligations must be externally binding. Since international law lacks supranational coercive sanctions, he concludes that it can only create voluntary or self-binding moral obligations and not distinctively legal obligations. 8 the law through sanctions is a necessary and central feature of law and legal obligations.25 Hart argues that while Austin’s view of law as commands backed by threats has some affinities and connections with law, there is a danger of exaggerating this connection and obscuring the special features that distinguish law from other means of social control.26 According to Hart, this theory views law as similar to the situation of a gunman saying to his victim, “give me your money or your life,” except that in this case, the gunman says it to a large number of people who are used to this and habitually surrender their money to such threats. Hart argues that that there is more to law than obeying commands out of fear of punishment.27 While the criminal law consists largely of rules that are like commands that are obeyed or disobeyed, rules that enable individuals to make contracts, wills and trusts are not designed to obstruct antisocial behaviour. Rather than saying: “do this regardless of whether you want to or not” they say “ if you wish to do this, this is the way to do it.” Constitutional legal rules that constrain government action also do not fit the command model of law. For example, when a government’s legislation violates a constitutional law, a court can nullify that legislation, but it does not use a coercive mechanism or the threat of sanctions to force the government to do this. More importantly, the command theory of law distorts the role that the idea of obligation and duty play in legal discourse. Hart argues that it identifies the normative idea of “having an obligation” or “being bound” with the observation that “one is likely to suffer the sanction or punishment threatened for disobedience.” 28 Even in domestic law, where there are effective organized sanctions, Hart argues that we have to distinguish the variety of reasons given for the obligations created by law. He emphasizes the importance of the internal perspective by which legal rules are accepted as guiding standards of behaviour.29 He recognizes that rules that give rise to obligations or duties generally call for some sacrifice of private interests and are generally supported by serious demands for conformity and insistent criticisms of deviations. However, he argues that once we reject the conception of law as essentially an order backed by threat (“the gunman situation writ large”), there seems no good reason for limiting the normative idea of obligation to rules supported by organized coercive sanctions.30 While the command theory of law emphasizes obedience towards law based on the fear of sanctions, Hart emphasizes the importance of the normative aspect of law from the internal perspective: the idea that individuals and officials are bound by legal rules because they accept these rules as valid legal rules and hence, as binding.31 They take themselves to 25 As Buchanan argues, to deny international law the title of “law” because it lacks a Hobbesian enforcement agent is to assume a now discredited Austinian conception of law and to ignore the realities of systems that certainly deserve the title of legal system (47). 26 Hart, The Concept of Law, 213. 27 H.L.A Hart, “Positivism and the Separation of Law and Morals” Harvard Law Review (1958): 593-629. 28 Hart, 218. 29 Hart, 218. 30 Hart, 218. 31 According to Hart, it is not necessary to answer the question of why people ought to obey the law in a foundational way. The motives for voluntarily supporting the rules of a legal system may be extremely diverse. While a legal order may be at its healthiest when there is a 9 be under an obligation to obey laws that are created in the manner that is recognized to be authoritative. This provides a better account of legal obligations because it is the acceptance of valid legal rules as binding that justifies the enforcement of law, and not the other way around.32 While sanctions are normally used to reinforce legal duties, the mistake that Hobbes and Austin make is to view the very existence of legal duties as dependent on sanctions.33 As Leslie Green argues: “…the normal function of sanctions in the law is to reinforce duties, not to constitute them. It is true that one reason people are interested in knowing their legal duties is to avoid sanctions, but this is not the only reason nor is it, contrary to what Oliver Wendell Holmes supposed, a theoretically primary one. Subjects also want to be guided by their duties — whether in order to fulfil them or deliberately to infringe them — and officials invoke them as reasons for, and not merely consequences of, their decisions.”34 The fact that domestic laws are enforced provides one reason why individuals obey the law, but it is not what makes them law, and it is not the reason they are binding. According to Hart’s view then, the idea that a law is binding on us means that the legal rule is valid and that we can be said to have some obligation under it. The issue then is whether the rules of international law can meaningfully be said to give rise to obligations.35 What matters is that states and other agents take themselves to be under obligations of international law, and that officials, such as judges on domestic and international courts and officials of international institutions, take states and themselves to be bound by international law. As Hart states, “the proof that ‘binding’ rules in any society exist, is simply that they are thought of, spoken of, and function as such.” 36 generally diffused sense that it is morally obligatory to conform to it, nonetheless, adherence to law may not be motivated by it, but by calculations of long-term interest, or by a desire to continue a tradition, or by disinterested concern for others. Hart argues that there is no good reason for identifying any of these as a necessary condition of the existence of law among individuals or states. Hart, 231-232. 32 According to Hobbes’ prudential account of obligation, we are only under an obligation to obey laws when they are externally enforced against all because we are not obligated to act contrary to our own interests.. 33 Hart, 217-218. 34 Leslie Green, “Legal Obligation.” Stanford Encyclopedia of Philosophy. 35 Questions about the binding character of international law express a doubt about the general legal status of international law, not its applicability. He argues that a better way to formulate the question is “can such rules as these be meaningfully and truthfully said ever to give rise to obligations?” Hart, 216. 36 Hart, 211. Hart suggests that an important feature of a legal system is that its subjects and administrators regard legal rules as binding on them, even though their reasons and motives for this may differ. Hart points to various elements in the relations of states that support the statement that there are rules among states that impose obligations upon them. Hart notes that rules could not exist or function in the relations between states unless a significant majority of states accepted the rules and voluntarily co-operated in maintaining them. While he 10 Many theorists of international law similarly argue that what makes international law “law” is not whether or not it is analogous to domestic law, and not whether or not it uses sanctions to enforce international legal rules, but whether states themselves accept international law as binding.37 As Hart notes, in the practice of states, certain rules are regularly respected at the cost of certain sacrifices, claims are formulated by reference to them, and breaches of the rules expose the offender to serious criticism and are held to justify claims for compensation or retaliation. An often-cited observation of this is Louis Henkin’s statement that "almost all nations observe almost all principles of international law and almost all of their obligations almost all of the time.”38 There is a complex web of international legal rules that receive a high degree of compliance, such as international laws concerning telecommunication, postal services, banking, aerial navigation, trade, and the law of the sea. Unfortunately, we of course get quite a different impression when we see states violating fundamental rules of international law, such as aggressive action against other states, torture, and genocide. However, even when states oppose interference in their internal affairs, they often appeal to basic principles in international law, such as the internal sovereignty of states. Even when states take aggressive against other states, they try to justify their actions as in accordance with international principles regarding the right to self-defense. As in the domestic case, the fact that laws are generally recognized as binding does not guarantee that actors will always comply with all particular laws. The fact that some agents violate laws does not prove that they are not generally recognized as binding. Based on Hart’s account, there is a second factor that makes the rules of international law legally binding: they must come into being in the manner accepted and recognized as authoritative. 2.3 THE ROLE OF SECONDARY RULES IN HART’S LEGAL THEORY What does it mean for a legal rule to be valid? For Hart, the special features of law that distinguish it from other means of social control are best understood through the idea of the union of primary and secondary rules.39 Primary rules are rules of conduct that confer acknowledges that the pressure exercised on those who break or threaten to break the rules is often relatively weak, and has usually been decentralized and unorganized, he argues that, as in the case of individuals who voluntarily accept the far more coercive system of domestic law, the motives for voluntarily supporting such a system may be extremely diverse: adherence to a particular law may be motivated by a general moral obligation to act in accordance with the law, by calculations of long-term interest, or by the desire to continue a tradition or by concern for others, and some may be motivated by the fear of punishment.36 37 Malcolm N. Shaw, International Law (Cambridge: Cambridge University Press, 2003) and Martin Dixon, Textbook on International Law (Oxford, 2005) 38 Louis Henkin, How Nations Behave: Law and Foreign Policy (Columbia University Press, 1979), at 47. 39 In his analysis of law, Hart contrasts a developed legal system with the union of primary and secondary rules against a more primitive society of individuals that only has primary rules of obligations. He treats the existence of this characteristic union of rules as a sufficient condition for the application of the expression “legal system,” but he does not claim that the 11 obligations on individuals (for example, the rules of criminal law). Secondary rules are power-conferring rules that are addressed to officials. They set out rules for the creation, recognition, change and adjudication of primary rules. The most important kind of secondary rule is the rule of recognition – the most fundamental, basic rule of a legal system that is accepted at least by the officials who administer the legal system as specifying the sources of law and the criteria for determining whether a rule has legal validity. Hart makes an important contribution to legal and political philosophy by proposing the idea of a foundational rule of legal validity. Against the modern idea of the supreme sovereign that is outside the law and against Austin’s idea of the “uncommanded commander,” Hart argues that legislators are constrained by foundational legal rules that specify law-making procedures. 40 As David Dyzenhaus argues, this distinguishes legitimate legal authority from arbitrary political power, and it places restrictions on law-makers and legal administrators concerning their creation and application of the law.41 The ultimate source of law then is not the sovereign’s will but foundational legal rules. Rather than the rule of man, or the rule of a sovereign that is above the law, we have the rule of law as the ultimate source of political authority.42 One of the reasons why Hobbes argues for a supreme sovereign that is not subject to any other authority is to conclusively resolve issues of indeterminacy that arise from conflicts and from the application of law to particular cases. For Hobbes, the sovereign’s authority must be supreme because if another body could question the sovereign’s actions, then the problem of indeterminacy arises again.43 This is part of Hobbes reasons for rejecting a right to rebel: we need a sovereign that has the final say. For Hart, the source of law is not the sovereign’s will but the secondary rules, and if a government violates these in its actions, then these actions do not have the normative force of law; they are brute force. This idea is particularly important in the international case. Instead of resting the ultimate source of law with a sovereign global legislative and executive body, the legitimacy and authority of international law can rest on its own foundational principles. These foundational secondary rules also provide a determinate and authoritative way by which international legal rules are created, identified, and applied. 2.4 PROBLEMS WITH HART’S VIEW OF INTERNATIONAL LAW word “law” must be defined in these terms. Hart offers an elucidation of the concept of law, rather than a definition of “law” which could provide rules for the use of this term. Hart, 213. 40 Hans Kelsen introduced a similar idea before Hart. He argued that the legal order was based on a Grundnorm or basic norm. 41 David Dyzenhaus, “The Justice of the Common Law: Judges, Democracy and the Limits of the Rule of Law.” In Cheryl Saunders and Katherine Le Roy (eds.) The Rule of Law (Sydney: The Federation Press, 2003). 42 Hart would probably take issue with this analysis of the implications of his theory since he argues that his conception of law is a merely descriptive and has no such normative implications. I am grateful to Lars Vinx for raising this point. Regardless of what Hart intended, I think these are important normative implications of his theory. 43 Kant presents a similar view. 12 One of the challenges with using Hart’s theory to make the case that international law is indeed law is that Hart believed that international law lacked secondary rules of change and adjudication which provide for legislature and courts in international law, as well as a unifying rule of recognition which specifies the sources of law and provides general criteria for the identification of its rules.44 Based on Hart’s conception of a domestic legal system as a union of primary and secondary rules, this seems to lead to the conclusion that international law is not a legal system since it lacks secondary rules. Instead, it is a simple or primitive form of social structure that consists only of primary rules of obligation. Hart tries to avoid this conclusion. Since he criticizes those who take domestic law to be paradigmatic of law and then evaluate international law through an adverse comparison with domestic law, he tries to avoid making this mistake himself.45 Consequently, since he develops his conception of a legal system by analyzing domestic legal systems, he admits that there may be problems with applying his conception of law as the union of primary and secondary rules to decide the issue of whether international law is really law. I disagree with Hart on this issue. I think the development and clarification of such foundational rules of legal validity is particularly important for a more decentralized model of international law. If we had a world parliament, then the test of validity for international laws could simply be whatever the world legislature passes according to recognized procedures. But if the creation of legal rules is more decentralized, then it is all the more important to have clear foundational rules regarding the sources and validity of international legal rules and obligations. Significant secondary rules of change and adjudication already exist in international law. For example, the Vienna Convention on Treaties codifies pre-existing international customary norms that govern the formation and effect of treaties.46 This provides secondary rules of change since it defines what is required to make a treaty valid. Article 39 of the treaty that created the International Court of Justice provides secondary rules of adjudication and also lists various sources of law.47 Article 39 states that the Court should apply international conventions, international customary law, the general principles of law recognized by states, and juristic writings in settling disputes between states. Such secondary rules need to be developed further but what exists so far is sufficient for a workable system of law.48 This issue points to another limitation with appealing to Hart’s theory to defend a decentralized model of international law against the enforcement-based objection. Hart doubts that secondary rules exist in international law because his conception of secondary rules is too centralized and too hierarchical, particularly his suggestion that there is an ultimate rule of recognition that unifies a legal system. Even in domestic legal systems, it is difficult to determine a single ultimate rule of recognition. Instead, there are various interrelated practices, institutions, rules and agreements regarding the creation and application 44 Hart, 214. Hart, 216. 46 Signed in 1969. 47 Established in 1946. 48 Samantha Besson considers how customary international law can also be regarded as part of international law’s secondary rules. Samantha Besson, "Theorizing the Sources of International Law” in Samantha Besson and John Tasioulas (eds.) The Philosophy of International Law. (Oxford University Press, 2010. 45 13 of law that emerge over time and become generally accepted and recognized by officials and even citizens to some extent. It is tempting to believe that a legal system’s ultimate rule of recognition is its constitution or founding document, but the legal validity of the constitution must rest on something else. Because Hart bases his theory of law on secondary rules and the role of government officials, he has been criticized for still viewing law as a one-way projection of authority; instead of grounding law on a supreme sovereign authority (as Hobbes and Austin do), law is grounded on a hierarchy of rules.49 For example, against the conception of law as an imposition of authority from above, Lon Fuller provides a less hierarchical and a more interactive or reciprocal conception of law. For Fuller, law depends neither on force, nor the exercise of authority, nor a hierarchy of rules; instead, law depends on the effective cooperation between citizens, and lawmaking and law-applying officials.50 This interactive conception of law appears to be present in Hart’s theory of law, but only to a limited extent. We see this feature in his consideration of what makes international legal rules binding. An important factor for him is that the subjects of international law (states, individuals, and other agents) recognize international law as binding on them. However, those who support more interactive and constructivist theories of law criticize Hart’s focus on rules instead of recognizing the important role of legal process and dialogue. They also criticize his focus on the perspective of officials rather than on the perspective of the subjects of law. Instead of focusing on whether there is an ultimate rule of recognition that can provide an ultimate ground for law’s validity, it is more important to consider the recognition and acceptance of law by its subjects. In order for international law to provide determinacy so that it can guide and coordinate the actions and interactions of various agents in ways that are less susceptible to problems of bias and power politics, it is important to develop authoritative and impartial adjudicative and administrative institutions. Ultimately, however, the parties and subjects of international law (states, individuals, corporations, global and transnational organizations) must recognize international law as binding, and they must accept the authority of judicial and administrative institutions to determine what international law requires. 3. GOING BEYOND HART: IS COERCIVE ENFORCEMENT A NECESSARY FEATURE OF LAW? There is an important objection that can be raised against my use of Hart to address the coercive enforcement view against international law. One can argue that while Hart was correct to criticize Austin’s and Hobbes’ reductive view of law as commands by an “uncommanded commander” backed by the threat of sanctions, this does not support the conclusion that sanctions are not conceptually necessary for law; it only proves that they are not sufficient and that other features are required as well. One may agree with Hart that a system of commands backed by threat alone does not constitute a legal system and that Austin’s focus on sanctions leads to an inadequate understanding of legal obligations. 49 Lon Fuller, “Positivism and Fidelity to Law – A reply to Professor Hart” Harvard Law Review 71.4 (1958): 630–672. Jutta Brunnée and Stephen J. Toope develop Fuller’s criticism of Hart in their analysis of international law. Legitimacy and Legality in International Law: An Interactional Account (Cambridge: Cambridge University Press, 2010). 50 Brunnée and Toope, Legitimacy and Legality, at 15 [check page]. 14 However, one can argue that the claim that coercive sanctions are not conceptually necessary for law goes too far. One can agree that there is more to law than Austin’s “commands backed by sanctions” but still hold that coercive sanctions and other forms of coercion are required for law and for legally binding obligations; law may still be inherently coercive.51 This objection seems quite persuasive since we tend to think of legal rules as rules that are enforced through public coercion, and that can be so enforced. This is a significant feature that distinguishes legal rules from other moral and social norms. Also, all legal systems seem to depend on the widespread use of sanctions. However, while the use of coercive sanctions to enforce law is common to all modern legal systems, is it conceptually necessary for law? Is it possible to have law or a legal system without coercive sanctions? Can one imagine a legal system without coercion? If so, what would be the purpose of law in such a system? 3.1 RAZ AND THE COMMUNITY OF SAINTS THOUGHT EXPERIMENT I want to go further than Hart and argue that while coercive sanctions are commonly used in modern legal systems, they are not conceptually necessary for law. Joseph Raz provides a good argument for this view. Raz imagines a society of saints in which all members of that society act according to what they think is right. They pursue their self-interests when they think it right to do so (they are not self-denying), but they may be wrong about what is right.52 We see this idea in the discussion of the state of nature in Locke and Hobbes: no matter how morally good and honourable individuals may be, there will still be disagreement about what is right. Even if all members of a society are morally good, even if they all want to honour their agreements and respect the rights of others, they may nonetheless unintentionally harm others or violate the rights of others by accident or due to ignorance. Conflicts will still arise and there may be more than one way to settle certain conflicts. Law is required then to set down general rules for all, to determine people’s remedial rights and duties, and to settle conflicts by applying these general rules to particular cases. This highlights the central role of adjudication in a legal system. There needs to be some authority to create, interpret, and apply general laws. If all members of this ideal society are motivated by these general rules and the authority of certain institutions to adjudicate particular conflicts, and if these are regarded as normative obligations that bind them, then punitive, coercive sanctions are not necessary. When members of this society harm others by mistake, they will still have to pay compensation to set things right, but this differs from punitive sanctions that are intended to deter law-breakers.53 Punitive sanctions are only needed when individuals refuse to comply with what the law requires or what a judge determines in a particular case. In such circumstances, the threat of coercion both motivates compliance and provides assurance to all that others will also comply. In the case of human beings (who are not saints), being arrested, imprisoned, and compelled to pay fines works quite well to fulfill this purpose, even though these are not effective enough to achieve full compliance. 51 This argument is presented by Schauer and Yankah. Joseph Raz, Practical Reason and Norms (Oxford: Oxford University Press, 1999), 157-62. 53 Raz, 160. 52 15 Based on this hypothetical example, Raz concludes that a sanctionless legal system is logically possible but humanly impossible.54 While it may seem strange to address this issue by appealing to the legal system of an imaginary society of saints since this will not work for a real society of human beings, this example suggests that some other explanation of the normativity of laws is needed. As Raz argues, the sanction-based attempt to explain the normativity of the law leads to a dead end; it explains one way in which laws provide reasons for action, but it fails to explain in what way they are norms.55 If law is not necessarily or inherently coercive even though coercive public sanctions are needed as a practical matter to deal with the problem of non-compliance, this alters the way we think of law and legal obligations. It is important to keep these practical and instrumental considerations distinct from the conceptual question of what is law and what is required for a legal obligation. Schauer and Yankah criticize Raz’s approach of determining the essential features of law. Instead of considering what is theoretically possible, they argue that legal theory should aim at illuminating law as it exists and as it is experienced. In their view, coercion is of central importance to existing modern systems of law and to people’s experience of law.56 Their approach is problematic if one takes these dominant features of modern domestic legal systems to be necessary for any system of law and if one then concludes based on this that international law is not really “law.” This is the mistake that is made in the coercive enforcement objection against international law. Their approach also limits the possibilities for law. Yes, it is important to recognize that coercion is widespread and central in existing modern systems of law and people’s experience of law. But rather than basing our conception of law on practices that are predominant in our current domestic legal systems, it is important to be able to ask if this is the best way for law to achieve the purposes that we think law ought to achieve. Given the high personal and social costs of mass incarceration, it is important to question the connection between law and coercion and to ask whether there are less coercive ways and non-coercive ways of increasing compliance with the law. 3.2 WHAT THEN DISTINGUISHES LEGAL OBLIGATIONS OBLIGATIONS? FROM MORAL AND OTHER Schauer and Yankah also argue that in understanding the concept of law, it is important to consider what distinguishes law from other normative systems, and what distinguishes legal obligation from other kinds of obligations, such as moral and religious obligations. Since the right to use coercive sanctions is an important distinctive feature of law, they argue that a coercion-free account of law or legal obligation is defective.57 This view is appealing because in a liberal society, using public coercion to enforce legal obligations is justifiable, but using public coercion to enforce moral or religious obligations is not. If one thinks that international law is not sufficiently enforced through coercive sanctions, then why is international law not simply a system of morality, as Austin and Habermas have claimed? 54 Raz, 158. Raz, 162. 56 Yankah, at 1240. Schauer at 17. 57 Yankah at 1240. Schauer at 17. 55 16 Why not conceive of treaty obligations simply as a kind of moral obligation, such as promisekeeping? While focusing on features of modern domestic legal systems that are pervasive and distinctive is illuminating in understanding domestic law, this approach does not conclusively answer the question of whether coercive sanctions are necessary in order for international law tor any other system of rules and obligations to be “law.” First, non-legal and nongovernmental institutions and organizations (such as religious institutions, academic institutions, employers, and social clubs) can also enforce their rules and increase compliance through various kinds of sanctions short of imprisonment, such as the threat of fines, disciplinary action, and having one’s membership or employment terminated or suspended. Therefore, the use of punitive sanctions is not unique to law. Second, even though it is true that the use of stronger coercive sanctions, such as imprisonment, is something that distinguishes domestic legal rules from other social rules, this is only true for some domestic laws. As discussed above, many laws are not enforced through coercive sanctions, such as wills, contracts, and constitutional law. Third, we can distinguish between moral obligations and legal obligations on other grounds. For example, according to Hart, whether we have a legal obligation in a particular case is determined by whether there is a valid legal rule. This is determined by secondary rules. While individuals can make promises to each other that give rise to moral obligations, in order to create legal obligations, a promise or agreement between two people must accord with legal rules about is required for a valid contract. This can be determined by a court of law. In the case of international law, rules, practices, and procedures have been developed to determine this, and international and domestic courts have the authority to determine this. There are important differences between the obligations created by international law and moral obligations that correspond to the distinction between law and morality in domestic law and that do not concern coercion. When states reproach each other for immoral conduct or praise themselves or others for living up to the standard of international morality, this moral appraisal is recognizably different from the formulation of claims, demands and the acknowledgements of rights and obligations under the rules of international law.58 What is predominate in the arguments states address to each other over disputed matters of international law are references to treaties, court judgments, and juristic writings. They are often very technical, and often, there is no mention of morally right or wrong, good or bad. For example, we can distinguish between whether NATO’s intervention in Kosovo was the morally right thing to do, and whether it nonetheless violated existing rules of international law. If we think that NATO violated the existing rules of international law, we may also think that the rules, procedures, and institutions of international law should be changed to better address such cases of humanitarian intervention. Why does maintaining this distinction matter? Law allows us to settle conflicts in a legitimate and authoritative way. It provides certainty, predictability, and order, and it allows individuals in a society to coordinate their behaviour. These goals are important in the international case as well. Rather than settling conflicts by mere brute force and the threat of force, states can settle their disputes according to accepted legal rules and before courts of law. The distinction between law and morality, between one’s legal and moral obligations, is 58 Hart, 228. 17 particularly important to the extent that there is moral disagreement between individuals or societies that have different moral views. 4. IS COERCIVE ENFORCEMENT NONETHELESS NECESSARY FOR INSTRUMENTAL REASONS? So far I have argued that coercive enforcement is not conceptually necessary for law or legal obligations. Sanctions may however be necessary as a practical matter to give individuals, states, and other actors additional incentives to comply with what are independently binding legal rules. I will briefly consider whether a centralized system of coercive sanctions is instrumentally necessary to increase compliance with international law. One problem with the objection that a decentralized model of international law is not effective enough to count as law is that it is not clear what would count as sufficient effectiveness. It cannot be complete effectiveness or complete compliance since many people continue to break the law in the United States and countries with strong legal systems, and many escape legal punishment. Much of domestic law is not coercively enforced in practice. Second, there are significant factual differences between the use of coercive enforcement in the domestic case and the use of coercive enforcement in the international case. It is easier to see why physical sanctions are possible and necessary in the case of individuals who are approximately equal in physical strength and vulnerability. They ensure that those who voluntarily submit to the restraints of law do not become victims of those who would, in the absence of sanctions, take advantage of others’ respect for the law while not respecting it themselves. Among individuals living in close proximity to each other, the opportunities for injuring others are great, as are the chances of escape, and consequently, natural deterrents and other reasons would not be adequate to restrain people from disobeying the law.59 In the international case, the use of force against states is more complicated, less efficacious, and comes at a very high cost, especially with respect to human lives. First, would a supranational coercive body be able to enforce law in a safe and effective way? In the domestic case, the police can use force to arrest an individual with little risk of harm to others. However, the use of force against states is always public. When the violator of international law is a state, it is difficult if not impossible, to direct sanctions solely against those who are responsible for violating international law. While this is most clear with the use of military action, economic sanctions can also cause serious harm against the poor and vulnerable populations of a state, including citizens who may in fact oppose their government’s violation of international law. As Hart argues, since the organization and use of sanctions internationally involves great risks, the threat of them adds little to other deterrents and reasons for compliance.60 Second, natural deterrents and other reasons for compliance seem stronger in the international case, so the need for problematic coercive sanctions may not be as important as in the case of individuals. In the case of states, there are a limited number of actors and their actions are public in nature. To the extent that they must interact with each other in our globalized world and are dependent on future good relations with other states, this provides 59 60 Hart, 218-9. Hart, 219. 18 strong reasons to comply with their legal obligations, particularly their treaty obligations. If I sell a defective product to someone, that person will probably not want to interact with me again, but I can find new customers. If word gets out, I can sell it under a different name or move to a new area. This is much more difficult for states to do. In addition, given the public nature of state actions, the condemnation of the violation of international law by international institutions, non-governmental organizations, other states, and individual citizens has a stronger role in guiding behaviour than in the case of individuals in large societies where shame and social condemnation may have little or no effect.61 For example, as Hathaway and Shapiro show, international legal institutions have been able to increase compliance by denying the benefits of membership to disobedient states (“outcasting”).62 Against this different factual background, international law has developed in a form different from that of domestic law. As Hart points out, given the large populations of modern states, if there were no organized repression and punishment of crime, violence and theft would be frequent occurrences. However, for states, long years of peace have intervened between disastrous wars. This is to be expected given the risks and stakes of war and the mutual needs of states.63 Even in the absence of a central enforcement body above states, what the rules of international law require is still thought and spoken of as “obligatory,” and there is still a general pressure for conformity to the rules.64 5. NORMATIVE REASONS FOR A SUPRANATIONAL SYSTEM OF ENFORCEMENT I would like to briefly consider a final objection that also applies to the first line of response. Regardless of whether coercive enforcement is necessary for law, to the extent that international law is enforced in various decentralized ways, this raises a normative problem. Whether or not a state attempts to enforce international legal rules against another state may be based on a state’s own interests and the respective power of the states involved. If State A is powerful and State B is not, then State A can enforce international law against B, but B would not be able to enforce it against A. These factors undermine important norms that are associated with the rule of law, such as impartiality and universality. International legal rules do not seem to apply to all states equally. While these is an important concern that needs to be addressed, it is not a good reason for denying that international law is “law” or denying that the rule of law can exist internationally in the absence of a supranational coercive body. Instead, the rule of law ideal 61 Unfortunately such reasons may be less compelling for very powerful states and states that are less interdependent and less concerned with being part of an international community of states. 62 “Outcasting: Enforcement in Domestic and International Law,” 63 Hart, 220. 64 Hart, 220. When international rules are disregarded, it is usually not on the ground that international law is generally not “binding.” Instead, efforts are made to conceal the facts that these rules were broken, or efforts are made to argue against the applicability of a particular rule of international law. For example, the United States has tried to defend itself against the charge that the war on Iraq is a violation of international law by arguing that the right to selfdefense in international law should be extended to include preemptive self-defense. 19 that the law should apply equally and impartially to all should guide the development of international legal rules and its institutions. I will briefly suggest a few ways that international law can begin to address these concerns without a coercive world government. First, and perhaps most importantly, international law cannot be said to apply equally to all states when five states have veto power over Security Council resolutions that authorize of the use of force, and when this body is often paralyzed by this. This needs to be reformed. Second, in order to have the rule of law, sanctions and other means of enforcement ought only to be used in accordance with international legal rules. For example, the UN Charter defines when states can use force, such as in cases of self-defence, and international customary law sets out and limits the use of non-forcible countermeasures for the breach of treaties. While the International Court of Justice (ICJ) has adjudicated both kinds of cases, a major weakness of our current system of international law is that the ICJ does not have compulsory jurisdiction; instead, a state must voluntarily authorize the ICJ to settle a case involving itself. This illustrates the importance of developing and strengthening adjudicative institutions that will apply the rules of international law in a fair and impartial way. Impartial adjudicative institutions are needed to authorize and constrain any attempts by states or other agents to enforce the law. 6. CONCLUSION The main aim of this paper has been to respond to the skeptical argument that without a centralized, supranational system of coercive enforcement that includes effective punitive sanctions, we cannot have law or legal obligations at the international level; at best, we can only have voluntary moral obligations. In my response, I first challenged the view that sanctions are a central and necessary feature of law (Sections 2-3). Secondly, in response to the argument that coercive enforcement is nonetheless necessary for the effectiveness of law in practice, I argued that there are important factual differences between the case of individuals within states and the relationship between states at international level that make the use of coercion less effective and more problematic in the international case (Section 4). While others emphasize the importance of a strong supranational system of coercive enforcement for the development of a law-governed condition at the global level, my view is that the following are the most important elements for a viable system of law at the global level. First, states, international bodies, officials, and other agents must regard international legal rules as generally binding on them. This point was supported by Hart’s theory of law and his criticism of the command theory of law in Section 2.2. Second, the criteria for the creation, adjudication and application of international legal rules must be clearly specified. I argued for this point in my criticism of Hart’s claim that international law lacks secondary rules (Section 2.4). Third, the rules of international law must be determined and applied impartially by adjudicative bodies or other institutions with the accepted authority to do so, such as the International Court of Justice, the International Criminal Court, the monitoring bodies of particular treaties, as well as domestic courts. I demonstrated the importance of this feature for law and the rule of law in my discussion of Raz’s argument (Section 3.1), in my discussion of the distinction between moral and legal obligations (3.2), and in response to the normative concerns that are raised against decentralized means of enforcement (Section 5).