Beyond Compliance: Rethinking Why International Law Really Matters


  • Robert Howse,

  • Ruti Teitel



The conceptual, and more recently empirical, study of compliance has become a central preoccupation, and perhaps the fastest growing subfield, in international legal scholarship. The authors seek to question this trend. They argue that looking at the aspirations of international law through the lens of rule compliance leads to inadequate scrutiny and understanding of the diverse complex purposes and projects that multiple actors impose and transpose on international legality, and especially a tendency to oversimplify if not distort the relation of inter-national law to politics. Citing a range of examples from different areas of international law – ranging widely from international trade and investment to international criminal and humanitarian law – the authors seek to show how the concept of compliance (especially viewed as rule observance) is inadequate for understanding how international law has normative effects. A fundamental flaw of compliance studies is that they abstract from the problem of interpretation: interpretation is pervasively deter-minative of what happens to legal rules when they are out in the world, yet ‘compliance’ studies begin with the notion that there is a stable and agreed meaning to a rule, and we need merely to observe whether it is obeyed.

Policy Implications

  • Looking at the aspirations of international law through the lens of rule compliance leads to inadequate scrutiny and understanding of the diverse complex purposes and projects that multiple actors impose and transpose on international legality, and especially a tendency to oversimplify if not distort the relation of international law to politics.
  • States, as well as other actors – corporations for example – instead of simply ‘complying’ with international legal rules may bargain in light of them, and around them. Given that there are transaction costs of negotiating, the rules will have an effect on the bargain, but one that will not be observed if what one is focused on is rule compliance.
  • In altering the focus and agenda of states and nonstate actors in dealing with conflict and post-conflict transitions, international law may have raised expectations too high that where politics and economics, and for that matter moral idealism, have failed to solve enduring human problems, law will succeed.
  • International law may create benchmarks for a wide range of private decision making, and this even when in the first instance the rules in question have not been explicitly addressed, at least not traditionally to nonstate actors. Such benchmarks may affect to whom firms lend, with whom they deal as suppliers or subcontractors, design specifications for products such as ships and aircraft, the terms of such diverse transactions as the adoption of children, the transportation of hazardous products and the transfer of high technology. Private actors may simply adopt these benchmarks as common terms of commerce regardless of the extent to which they have been ‘implemented’ by states.
  • International law (norms and/or institutions such as courts and tribunals) may shift in whole or in part decision-making, interpretative and/or legitimating power from one set of elite actors to another (for example from diplomats, foreign policy analysts and military planners to legal professionals such as judges, lawyers and law professors).

Traditionally, in most fields of legal study, the question of why, how and to what extent the actors bound by legal rules comply with them has been a secondary, if not marginal, one for scholarly inquiry by jurists. Law and economics and legal sociology, as well as realist approaches such as Critical Legal Studies, have focused greater attention on the actual behavioral effects of rules. But such approaches have a much broader focus than the specific question of rule compliance.

International law is a striking exception: the conceptual, and more recently empirical, study of compliance has become a central preoccupation, and perhaps the fastest growing subfield, in international legal scholarship, engaging some of the best minds in the discipline (see Bradford, 2004). Indeed (and perhaps the most striking or extreme example is Andrew Guzman) there are those who insist that explaining compliance is a central test or criterion for the adequacy of any general theory of international law: ‘[C]ompliance is one of the most central questions in international law. Without a theory of compliance, we cannot examine the role of treaties, customary international law, or other agreements. Nor can we consider how to improve the functioning of the international legal system, or develop a workable theory of international legal and regulatory cooperation’ (Guzman, 2002, p. 1826).

The objective of this essay is to put in question the focus on compliance in much recent international law scholarship. In our view, this focus obfuscates the character of international legal normativity, tending to ignore the centrality of interpretation to the generation of legal meaning, as well as the horizontal relation between diverse norms and regimes (‘fragmentation’). Looking at the aspirations of international law through the lens of rule compliance leads to inadequate scrutiny and understanding of the diverse complex purposes and projects that multiple actors impose and transpose on international legality, and especially a tendency to oversimplify if not distort the relation of international law to politics.

As Benedict Kingsbury has observed, the recent emphasis on compliance in legal scholarship has been accompanied by little critical thought about the meaning of the concept of ‘compliance’ and the work it is supposed to do in any theoretical account of international law. As Kingsbury shows, a ‘compliance-based theory of international law’ is an impossibility since conceptions of the meaning and significance of compliance depend on prior theoretical or conceptual constructs about the nature of international law and – we ourselves would add – of law generally (Kingsbury, 3598).

In the first part of the article, we attempt to explore the question of what thinking about compliance can contribute to our understanding of international law in general and the limits of this contribution. We conclude that although the importance of compliance to an overall understanding or theory of international law is less than is often assumed in the literature, there is nevertheless a strong case for studying the real world effects of international law, for much the same reasons as in other fields as well.

In the second part, we argue that, nonetheless, the notion of rule compliance is much too narrow an angle of vision to comprehend international law’s normative effects. In this part, we seek to sketch a broader framework for understanding these effects, which places considerable emphasis on interpretation and cross-regime impact, elements that are often lost or given inadequate emphasis in compliance studies. Here we build on the work of scholars such as Ryan Goodman, Derick Jinks and Beth Simmons, who are concerned with the effects of international legal rules, but understand these effects in a broader and more subtle way than is connoted by notions of rule compliance or ‘enforcement’ (Goodman, 2001; Jinks and Goodman, 2003; Simmons, 2009). Finally we consider the significance of the expanded discourse of international law once there is a shift in terms of the debate over compliance.

1. Debating the meaning of compliance for international law

One reason that compliance is often seen as a central problem for international legal scholarship is the challenge (by realists and some but not all positivists)1 that law is only really law when accompanied by authoritative interpretation and enforcement (see Morgenthau, 3548). A focus on compliance, or more adequately perhaps obedience (see Henkin, 3568), aims to deflect such a claim by asserting that there is a range of considerations including reputational effects/long-term self-interest that lead to compliance with international law, regardless of the absence of authoritative interpretation and enforcement in most instances.

Such a response at once proves too much and too little. There are many domestic laws that attract widespread noncompliance despite significant sanctions (drug laws and tax laws in many jurisdictions)2 and others that attract widespread compliance even though sanctions and monitoring are relatively lax (seatbelt laws); thus the positivist claim is already weak, if not incoherent, if the heart of the positivist claim is the identification of the existence of authoritative interpretation and enforcement with actual compliance.3

But the positivist claim may at its root be conceptual: a formal, definitional property of law (as opposed to other normative systems, religion, morality, social and familial custom) is that ‘law’ is the binding edict of a sovereign with coercive force, at least in principle, over the subjects to whom the legal commands are addressed. If this is so, then even if actual rates of compliance with international law were superior to those for norms that are commands of a sovereign with a monopoly of violence over the subjects of law, demonstrating this would not itself be an answer to the positivist objection. Indeed, many forms of normativity, including customs and etiquette within particular subpolitical communities, exhibit very high rates of compliance or obedience, and the positivist has a point in that this does not, in ordinary language, induce us to call them ‘laws’. It is worth noting however that one of the most sophisticated positivist accounts of law, that of H. L. A. Hart, rejects the notion that legal obligation implies effective coercive sanctions, specifically informed by a consideration of international law. According to Hart,

To argue that international law is not binding because of its lack of organized sanctions is tacitly to accept the analysis of obligation contained in the theory that law is essentially a matter of orders backed by threats. This theory, as we have seen, identifies ‘having an obligation’ or ‘being bound’ with ‘likely to suffer the sanction or punishment threatened for disobedience’. Yes, as we have argued, this identification distorts the role played in all legal thought and discourse of the ideas of obligation and duty (Hart, 3561, pp. 217–218).

In addition, as Kratochwil has argued, as distinguishing ‘law’ from other forms of social ordering is not a simple matter, it is questionable whether and to what extent the existence of authoritative interpretation and enforcement can be defended as decisive or even the most important formal properties of law (Kratochwil, 3589, pp. 187ff.). Historically, such a view may reflect the function of law as an agent of rationalization, centralization and secularization of authority in the period that could crudely be described as the rise of the modern nation state (Tocqueville, 2001; Unger, 3576; Weber, 3578). Why should international law have to defend itself before the bar of such a definition of law? Instead, might not the very proliferation of norms labeled international law in our own, very different era affect how we approach the meaning of legality?

As early as the 3540s, the French-Russian philosopher Alexandre Kojeve discerned the internal tension between the positivist idea of sovereignty and the positivist conception of law as effective coercion. Granting to positivist or behaviorist approaches that law to be fully law must possess, in principle, irresistible force to compel its subjects, Kojeve noted that these subjects and their activities frequently exceed the territorial bounds of the sovereign; thus, purely domestic law is not fully law, against the positivist’s own criterion, since its (in principle) irresistible force ends at the border; only global law can become fully law, since it is capable of recognition by all sovereigns and thus can become (in principle) irresistible everywhere that the persons and activities that are its subjects may be situated (Kojeve, 2007). Thus, one could actually consider international law in the perspective of the limits of the domestic law paradigm with respect to compliance.

A second, and certainly understandable, reason for focusing on compliance is that there are some international legal norms that have reciprocity as their fundamental normative and/or functional premise. An obvious example is that of peace or disarmament treaties. The nightmare scenario of one party innocently disarming pursuant to the obligations of the treaty while the other secretly builds up its arsenals, acquiring a decisive strategic advantage, has often fuelled skepticism about the capacity of international law to solve the ‘security dilemma’. International trade treaties also have significant reciprocity dimensions. Here an analogy may be drawn to nonsimultaneous contractual performance; general compliance studies may yield some insights into the mechanisms or devices – monitoring, verification, third party guarantees, self-help/retaliation – that are available to address the problem. Yet the current compliance focus in international legal scholarship is not especially preoccupied with these issues; at least as much attention if not more is given to multilateral human rights treaties, which do not pose this kind of challenge or dilemma.

A third, not unrelated, reason why compliance has often been emphasized in contemporary scholarship may relate to the concern to counter or refute realist or ‘skeptical’ positions that international law does not really matter, because the ultimate causes of state behavior are to be found in self-interest, not legal obligation. Here, emphasizing compliance has at least two drawbacks. The first is that such an emphasis reflects a too quick concession to the way the realists/skeptics frame the test for whether international law matters. The second, and obviously closely related, is that if one concedes the realist/skeptic test, then showing real world normative effects will never be enough to meet it.

Let us use the position of Jack Goldsmith and Eric Posner to illustrate this claim. Goldsmith and Posner argue that most cited examples of obedience to international law are cases where to do so is in the self-interest, either short or long term, of the complying state. They suggest that these cases somehow do not really ‘count’ as a demonstration that international law is an effective legal system. Goldsmith and Posner do acknowledge that international law can provide useful communication and coordination mechanisms where states seek to cooperate on the basis of mutual or reciprocal self-interest, but that it cannot compel obedience against interests.

It is a curious criterion for the validity or even effectiveness of a legal system that it should routinely command obedience from legal subjects against their individual interests. Since lawfulness is often an endogenous preference of individuals and indeed of sovereigns,4 and since as Goldsmith and Posner themselves observe, law itself can give rise to ‘self-interested’ reasons for compliance (e.g. reputational effects), it is unclear how any empirical study could ever prove that compliance to law is exclusively or decisively undertaken contrary to perceived self-interest.

Why should it matter to do so? This question is as pertinent for domestic as for international law. Kant, for one, defined ‘law’ (Recht) in contradistinction to moral virtue as requiring only external obedience, not a habit of mind that compels compliance from duty alone, regardless of heteronomous considerations (Kant, 3596). In some contexts, law may serve an important function in constraining compulsive, passionate or impulsive behavior that will often be contrary to the long-term self-interest of the legal subjects in question, based upon at least their first order preferences (see Sunstein, 3597). In other contexts, even where the law has coercive state apparatus at its disposal (police, prisons, etc.) it has been suggested to be often ineffective in constraining such behavior (e.g. violence by socially disempowered young men) since deterrence through the threat of punishment itself supposes some ability to take into account future consequences.

While compliance cannot solve these kinds of objections to international law and its relevance, understanding compliance is obviously important if we care about whether the law realizes its purposes, that is, on any nonformalist, instrumentalist view of law, whether law and economics or critical legal studies or moral and political theories of law of a perfectionist character.

But here what we are interested in is the impact of the law in all respects, and compliance, unless one stretches the ordinary meaning of the word to breaking point, is much too narrow a lens for conceiving the wide range of effects that can be produced by international law. Why international law matters cannot be tested simply by examining ‘compliance’ in any of its common senses (conformity or obedience to rules or even transformation of state conduct in a manner that conduces to the purposes of the rules).

The effects of norms, including legal norms, have an inherent complexity that, as Ruggie explains, defies the positivist aspiration to link norms causally to discrete behavioral acts:

Norms may ‘guide’ behavior, they may ‘inspire’ behavior, they may ‘rationalize’ or ‘justify’ behavior, they may express ‘mutual expectations’ about behavior, or they may be ignored. But they do not effect cause in the sense that a bullet through the heart causes death or an uncontrolled surge in the money supply causes inflation … The impact of norms within international regimes is not a passive process, which can be ascertained analogously to that of Newtonian laws governing the collision of two bodies. Precisely, because state behavior within regimes is interpreted by other states, the rationales and justifications for behavior that are proffered, together with pleas for understanding or admissions of guilt, as well as the responsiveness of such reasoning on the part of other states, all are absolutely critical component parts of any explanation involving the efficacy of norms. Indeed, such communicative dynamics may tell us far more about how robust a regime is than overt behavior alone (Ruggie, 3598, pp. 97–98, emphasis in original).

2. Moving beyond compliance

The following are some of the possible effects of international law that are not captured by the notion of behavioral ‘compliance’ with a ‘rule’ of law and not tested by studies that look only for rule compliance, or some surrogate for it, such as outcomes shifting in the direction supposedly sought by the rules.

• International law (norms and/or institutions such as courts and tribunals) may shift in whole or in part decision-making, interpretative and/or legitimating power from one set of elite actors to another (for example from diplomats, foreign policy analysts and military planners to legal professionals such as judges, lawyers and law professors). This effect is autonomous from that of compliance: in some cases, legal professionalization may lead to more compliance, and in some cases less. In others, it could even lead to ‘ultracompliance’: effects which go beyond what is desired from the perspective of the objectives of the legal regime, and which may even be perverse.5

• Adhesion to international legal norms and/or perceived compliance with them may be used as a basis for admitting states to a ‘club’ with privileges, as it were; characterizations of certain states as ‘outlaw’, ‘rogue’ or ‘scofflaw’ regimes serve to justify the exclusion of such states from more favorable treatment offered to others, and to legitimate harsher, more forceful diplomatic, political and military strategies in relation to the ‘outsider’ states. This is an effect that is considered in relation to compliance by Goodman and Jinks, for instance. However, it may have significant implications that go beyond whether the prospect of membership can help induce compliance, affecting the capacity of individuals from those states to travel and trade, for example, or determinations of refugee status, or the perceived political risk of foreign direct investment. Goldsmith and Posner acknowledge this effect in referring to the ‘standard of civilization’ function of human rights treaties; but curiously they appear to dismiss this function as an indicator that international human rights law matters, since (contrary to what Jinks and Goodman, 2003, speculate) in deploying their reductive perspective it does not as such lead to ‘compliance’.

• International law can affect the way that policy makers view international problems and conflicts (for example in terms of clashes of rights as opposed to balancing of political or economic interests) and their perception of the constituencies to whom they are accountable in addressing such problems and conflicts. In the Balkans, resolving the conflict, and building post-conflict societies, somehow became identified with the prosecution of crimes against humanity at the International Criminal Tribunal for the former Yugoslavia (ICTY). The role that international criminal law could play in achieving these goals was arguably exaggerated, leading to a relative neglect of other processes, such as local truth commissions, the building of grass-roots democratic institutions and the reconstruction of civil society (Teitel, 3599). In the case of the 2006 Lebanon war, the complex security issues raised by Hezbollah’s entrenchment in Lebanon and the role of Syria were overshadowed in public discourse by claims and counterclaims concerning the commission of war crimes by either or both sides during the conflict. The broad political and economic considerations that might be served by integrating Serbia into the European Union have been almost entirely overshadowed by the concern that Serbia has failed adequately to cooperate with the ICTY prosecution of war criminals. Indeed, compliance has become the central issue with respect to Serbian accession to the EU, where ‘compliance’ with international law is viewed as a surrogate for or symbol of political ‘cooperation’.

Despite the commitment to positivism, the application of pre-existing law according to criminal justice principles and concepts, the tribunals have also reflected ‘teleological goals’6 including broader, noncriminal justice goals such as peace, etc. in the region.7 As the ICTY appellate chamber held in a sentencing ruling, the law applied ‘must serve broader normative purposes in light of its social, political and economic role’.

Consider what compliance might mean in the context of the International Criminal Court (ICC). In this regard, just how important is the fact of signatories turning defendants over to the ICC as indicia of ‘compliance’? Or, might ‘compliance’ be evaluated through analyzing a complex set of behaviors in a longer time frame: commitment to the norms themselves via the act of signing (see Jinks and Goodman, 2003, on joining human rights treaty regimes; see also Simmons, 2009); incorporating these norms via legislating into domestic law, the embarking on prosecutions policy in domestic courts, for example UK trials of their soldiers in Iraq; or cooperating with other countries in their prosecutions via a variety of actions including rendition and extradition, for example Chile’s extradition of Fujimori back to his home country to face rights charges (see Romero, 2007). The normative and positive law innovations triggered by the establishment of and commitment to the ICC might thus lead to prosecutions of a kind never contemplated by the jurisdiction of the Court itself, such as for crimes committed in the distant past. This is a clear instance of ultracompliance. Furthermore, the sense in which there is a layered normativity can be seen in the importance of the behavior of nonsignatories whose behavior may nevertheless be shaped by the legal regime in question. Thus, a compliance-driven theory does not capture US actions in embarking upon prosecutions policies in the orbit of the ICC.

• By transforming policies into legal norms, international law may reinforce those policies in the presence of indeterminacy and controversy concerning their effects and their substantive rationales (the ‘constitutionalization’ of elements of the Washington Consensus in World Trade Organization (WTO) law). Thus instead of changing behavior directly, international legal norms here may make it more possible or less costly for a state to adhere to an existing or recently adopted policy course, despite increasing internal disagreement and dissent concerning the merits of that policy course. Thus Goodman (2002, pp. 540ff.), Moravcsik (2000) and Simmons (2009) have pointed out that acceptance of or creation of international or transnational human rights norms may serve to ‘lock in’ a transition to democracy and the ‘rule of law’, increasing the costs of resistance or reversion. These observations vindicate Jon Elster’s (2000) notion that precommitment is never about tying one’s own hands but rather someone else’s, in this case elements within a transitional society who may seek a return to authoritarianism. Precommitment can fulfill the expressive function of reinforcing the beliefs of citizens that there is no turning back and that as exemplified in international norms the policy choice chosen has a universal force – or at least that ‘world history’ is on the side of the transition. This example points to an internal difficulty within the ‘compliance’ perspective. Goldsmith and Posner (2006, p. 121), for example, maintain that unlike ratification of human rights treaties, ‘democracy, peace and economic development’ have been shown to enhance human rights protection; however, this contrast assumes that democracy, peace and economic development occur entirely exogenously of the effects of international human rights law. If, as just suggested, international human rights law helps to lock in transitions to peaceful democratic conditions, then it may ultimately lead to what is usually conceived of as compliance, but only through a normative effect that is caught in the first instances by focusing on something other than rule compliance.

• International law may create benchmarks for a wide range of private decision making, and this even when in the first instance the rules in question have not been explicitly addressed, at least not traditionally to nonstate actors. Such benchmarks may affect to whom firms lend, with whom they deal as suppliers or subcontractors, design specifications for products such as ships and aircraft, the terms of such diverse transactions as the adoption of children, the transportation of hazardous products and the transfer of high technology. Private actors may simply adopt these benchmarks as common terms of commerce regardless of the extent to which they have been ‘implemented’ by states. Thus, here again, the notion of ‘compliance’ (generally focused on state compliance) does not fully capture the normative effect.

• Conversely, the transnational customs or norms of nonstate actors or institutions that are dominated by nonstate actors may become ‘international law’ through being legalized in a treaty process; the example of the Technical Barriers to Trade Agreement in the WTO, which creates a legal obligation on states to use international standards (often promulgated by nonstate organizations) for mandatory regulations is a telling one. Here, looking at state compliance tells only a small part of the story of the effects of strengthening the norm through ‘legalization’ on the actual operation of markets, consumer choice, etc. (these kinds of effects, that move between as it were public and private transnational behavior and public and private norm setting and interpretation, have begun to receive attention through the Global Administrative Law project, led by Kingsbury, Richard Stewart and their collaborators. See also the analysis of international and transnational securities regulation by Claire Kelly and Roberta Karmel (2009)).

• Quite beyond being formally implemented in domestic legal rules, international law may affect the interpretation of domestic law. Viewing such interpretation simply as part and parcel of ‘compliance’ is inadequate or limiting. The invocation of international law in the interpretative process can serve a variety of jurisprudential purposes which may have little to do with ‘compliance’ with the international rule (see Teitel, 2004). Indeed, using an international legal norm for interpretative purposes may give a reading or effect to such a norm that is different from that which the agents of an international regime may have intended, assuming one could ascertain such an intention. Important normative effects may occur through cross-regime interpretation. Ignorance of such effects may lead to serious underestimation of the influence of particular legal regimes or institutions. For example, international environmental law, including soft law, has shaped the way in which the WTO Appellate Body has interpreted norms of international trade law in a number of cases (Howse, 2008). Koskenniemi observes:

A legal regime such as the European or Inter-American human rights convention makes constant reference to general international law without any act of incorporation. The International Criminal Tribunals on the former Yugoslavia and Rwanda are all the time applying law that comes from beyond their constituent instruments. Tribunals having to deal with State contracts with foreign companies frequently fill gaps and solve inconsistencies by reference to something like a natural law of the transnational commercial system. Even when the law appears to step aside so as to give room for political expediency – as arguably happened in the Lockerbie case in 3592 where the International Court of Justice affirmed that Security Council decisions override the rights of States under particular treaties – this took place as an inference from Article 103 of the UN Charter (Koskenniemi, 2007, p. 8).

While Eric Posner, an international legal scholar heavily focused on compliance, tends to dismiss the influence of the International Court of Justice (ICJ), based on the number of judgments it has emitted and their purportedly distant effects on the controversies decided (Posner, 2004), the jurisprudential acquis of the ICJ on such essential questions as state responsibility, countermeasures and treaty interpretation has been repeatedly invoked, in for example, investor–state arbitrations, characterized by compulsory jurisdiction and a ‘hard law’ remedy, that is, monetary damages that can be enforced in domestic court. Along similar lines, sources such as the European Convention on Human Rights and its interpretation by the European Court of Human Rights have been used by investor–state tribunals.

Another illustration of the growing area reflecting the influence of international law is seen in the contemporary controversy over the importation of international law in interpretations of domestic constitutional law, e.g. Roper v. Simmons, and Lawrence v. Texas where the United States Supreme Court referred to a number of opinions interpreting analogous provisions in international covenants for their ‘confirmatory’ value.

• Rather obviously legal agents bargain in the shadow of the law. We have already mentioned how international law may shape or affect the terms of bargains or transactions between nonstate actors, who are not even directly bound by the rules in question. States, instead of simply ‘complying’ with international legal rules may bargain in light of them, and around them. The rules will shape the bargaining, and thus have an effect, but one that will not be observed if what one is focused on is rule compliance. Once again, ‘compliance-based’ theories of international law do not really have an analytic for understanding such settlements: are they really to be thought of as noncompliance (since the parties have adopted a solution that deviates from the given rule in some respects or may be driven to settle out of uncertainty as to whether even if the legal claim is good the other party will comply)? While under the influence of the Coase theorem, domestic law and economics scholars have examined many contexts where, given transaction costs, background legal rules affect the shape of bargains, the analysis of how international legal rules affect interstate settlements has largely been neglected. (One exception is the work of Busch and Reinhardt (2001) on why countries settle disputes within the WTO legal framework). As is illustrated by the Canada–US softwood lumber dispute, uncertainty about compliance can be a major factor in inducing a settlement (in that case uncertainty concerning whether the US would comply with further WTO and North American Free Trade Agreement (NAFTA) rulings, given its track record of poor compliance up to that point). The settlement in question would be hard to explain if there were no legal rules at all constraining the ability of the United States to impose trade restrictions on imports of Canadian lumber (and certainly inexplicable on realist terms since the US is the far greater power), but equally inexplicable (in its details at least) if Canada had confidence that its legal claims would result in US compliance. Thus international legal rules can produce distinctive effects, and shape behavior, on account of uncertainty about compliance.

In a very different context, that of the international legal duty to punish crimes against humanity, recently reaffirmed by the ICJ in the Bosnia v. Serbia case, the increasing likelihood of such prosecutions given the creation of an international criminal court may well affect peace or regime transition bargains between parties to a conflict, for instance making it more difficult or less plausible to use amnesties as a bargaining chip for acceptance of a peaceful, negotiated transition. On the other hand, with the ICC in the background now, and its ability to enforce international criminal law during an ongoing conflict, more cautious or restrained behavior by some of the participants in the conflict may result, and this could actually make a transitional bargain easier. In other words, the effects of international criminal law on transitional or peace settlements in conflicts are likely to be complex. But these effects do not even come into focus if one centers the analysis on effects on ‘compliance’ with the duty to prosecute and/or the duty to cooperate with the ICC for instance.

• As already noted, international law can produce ‘ultracompliance’, by which we mean it can have normative effects that are greater or more powerful or different than what may be desired or consistent with the values or intentionality that might plausibly be attributed to the ‘creators’ of the norms. Because of interpretative uncertainty and asymmetries of information, governments have been able to invoke international legal rules to justify policy directions, even where alternative policies could be defended under reasonable interpretations of those rules. This has especially been the case with trade liberalization agreements, such as the WTO treaties, where many of the obligations in question are accompanied by exceptions provisions that allow a wide range of legitimate policy interventions. In the case of intellectual property rights, the United States government and the pharmaceutical industry convinced many governments and nongovernmental actors that the WTO Trade-Related Aspects of Intellectual Property Rights (TRIPs) agreement compulsory licensing exception from patent rights was largely unusable, eventually giving rise to the need for a negotiated text to clarify this matter.

Similarly, international human rights and humanitarian law has been deployed to justify to varying degrees and may have actually spurred forceful interventions in, for instance, Kosovo, Afghanistan and Iraq; while at the same time, in the case of Afghanistan and Iraq, such interventions appear themselves to have undermined human rights and humanitarian values and objectives to some extent, especially given the impact on civilian populations. The concept of ‘compliance’ simply does not capture the extremely significant complexity and tension in the countervailing normative effects here: a single act can be seen from the perspective of an attempt to enforce compliance with a given norm (‘humanitarian intervention’) but also, as a violation of it (where there are disproportionate effects on civilians, for instance). This was raised regarding NATO intervention in Kosovo, is being raised now regarding the ongoing casualties in Afghanistan and, unquestionably, is at stake in the deteriorating sequence of events on the ground in Iraq which has further delegitimized the humanitarian/human rights case for intervention.

• During the cold war the Soviet bloc accepted through the Helsinki Accord basic international human rights norms. It came as little surprise to many observers that hardly any compliance with these norms actually followed their acceptance by the eastern countries. But non-compliance had an important effect: it provided a focus for dissident movements and served to further delegitimize the Soviet bloc regimes, arguably hastening the end of communism. Having accepted human rights norms in Helsinki, the states in question could no longer plausibly dismiss such norms as western or bourgeois capitalist ideology (Thomas, 2001).8

3. Thinking about the broader meaning(s) in an expanded international law discourse

Broadening our understanding of the real world effects of international law beyond the notion of ‘compliance’ is a valuable exercise but there is at least one sense in which it will not satisfy the realists/skeptics any more than compliance studies. In the case of Goldsmith and Posner at least, they ask for further proof that the effects in question result from the character of the norms as law. For instance, humanitarianism as a political ideology or morality might have similar effects, even if not packaged as human rights and humanitarian law. But as Goldsmith and Posner recognize, this would then cause us to ask why states invest resources in translating these norms into a legal form. Their answer is that, at least in some contexts, the use of the language of law communicates a level of ‘seriousness’ to a commitment that may have consequences for how other actors respond in their own behavior and the reputational consequences of reneging on these commitments.

A different answer, more pertinent in their view to explaining international human rights ‘law’, is that states tend to want to justify their actions in universalist terms, and the language of law is particularly amenable to this, given its ‘formal’ character (Goldsmith and Posner, 2006, pp. 182–184). States need to water down their rhetoric to appeal to more and more audiences and ‘law’ that does not (in principle) have such a content tied to particular religious, moral or civilizational outlooks serves this purpose well.

Perhaps Goldsmith and Posner are right that the resort to ‘law’ has significance in defining what is common, a set of norms that is public and capable of expression and rationalization in general terms. Beyond these formal qualities of ‘law’, however, the legacy of the Second World War and more recently of the cold war has associated the ‘rule of law’ with a set of thicker, more substantive values: limited government, objective and impartial treatment of individuals by the state, lack of arbitrariness generally, the rejection of open-ended struggle and violence, and decent and bearable order.9 In many societies, belief in the capacity of politics and/or economics to guide solutions to basic global problems has eroded. In these circumstances, ‘law’ has enough positive resonance, especially as public law, to have become a preferred vocabulary for social order and for legitimating decisions of governance.

Unlike Goldsmith and Posner on the one hand, and international law ‘utopians’ on the other, we worry not so much that international law boils down to ineffective or largely meaningless (in real world terms) rhetoric but rather that instead it has, in a range of contexts, been all too effective. In altering the focus and agenda of states and nonstate actors in dealing with conflict and post-conflict transitions, international law may have raised expectations too high that where politics and economics, and for that matter moral idealism, have failed to solve enduring human problems, law will succeed. We have already given one very vivid example of such arguably excessive expectations, namely the hopes pinned on international criminal justice for successful post-conflict transitions and, even more so, for bringing about a world free of dehumanizing conflicts. The messy and contingent business of brokering political deals between groups and factions and of economic reconstruction10 has become less glamorous than that of trying and punishing the ‘villains’ (however worthy and justified in itself).

In sum, it may well be true as an abstract matter that it is not only or exclusively legal normativity that would be capable of producing effects of the kind we have identified in this article, and that at other times and places other forms of normativity might serve similar functions in international order (religious, moral, political norms, for instance). But such an essentializing, timeless claim on behalf of law as a distinctive type of normativity is hardly necessary to defend the proposition that international law matters in all kinds of ways for us, here and now.11 Further, as Liam Murphy has argued in a recent article, a careful examination of debates about the ‘concept’ of law shows the extreme difficulty or perhaps impossibility of drawing stable unambiguous boundaries between ‘law’ and other kinds of normativity (Murphy, 2008, p. 1088). The extraordinary range of normative effects generated by international law may well be a product of our distinctive historical situation, just as realist or national interest-based approaches to the right conduct of international relations may understandably have been a dominant frame at other times, such as during the cold war.

Nevertheless, there are those, such as Goldsmith and Posner, who assume that there is a timeless framework for international relations whereby states always turn to morality and/or law to justify or explain their actions, but it is self-interest that motivates these actions. In fact this is an even more radical challenge to the notion that international law matters, or at least matters a lot (of course, Goldsmith and Posner have to admit that rhetoric performs some function, otherwise states would not invest in it). On this view, the basic facts of war and peace, who is powerful and who is not in the international system, are determined or predetermined by self-interested behavior, and so whatever justificatory rhetoric may exist at a given time, it does not alter or transform those facts. Indeed, the very structure of this argument further indicates why studies of compliance, however methodologically sound, cannot put to bed, as it were, skepticism about whether international law matters. Even a very well-designed regression analysis cannot test for the internal motivation that generates a given behavior.12 The question of whether and to what extent particular state behaviors are generated by considerations of justice or interest, or some mix thereof, has been a matter of moral-philosophical speculation at least since Thucydides.

Analytically, however, we emphasize a difficulty with the role of self-interest in the skeptical theory of Goldsmith and Posner, to which we alluded earlier: it is incoherent on their rational choice view of domestic politics to define a state’s self-interest as exogenous to the preferences of its citizens. Yet, citizens may care for international law for different reasons than those adduced by Goldsmith and Posner in the case of states, in reality political leaders of states – namely justificatory rhetoric. Indeed, if citizens did not care about justice, it would be unclear why states would need to invest in such rhetoric: who would they hope to persuade? For if states are aware that other states like themselves act only from self-interest, not from the considerations advanced in their justificatory rhetoric, they would certainly not be investing in the rhetoric in the hope of persuading other states or, more precisely, their political leaders.


Moving beyond ‘compliance’ as a central concept of international legal theory and inquiry opens up new horizons, or at least suggests new emphases in international legal scholarship and a recasting of certain of the predominant debates. First of all, empirical inquiry or theoretical speculation as to how much ‘compliance’ there is with respect to international law, and how and why it happens, cannot as such play much of a role in the debate about whether international law is ‘law’ or what it contributes to global order that some other non-legal discourse – cosmopolitan moralism, for example – would not contribute. What one needs is much more reflection on those properties of ‘law’ that it possesses which make international law distinctive as a mode of discourse in international order, and then to see the effects of international law through such an understanding. This could give new purchase on some old puzzles such as the meaning and rationale of ‘soft law’; it may be that ‘soft law’ is effective, for example, because it possesses the relevant or desirable ‘law’ characteristics for the purposes in question (transparency, generality, connection to common or widely shared norms and practices), while not itself possessing the ‘bindingness’ characteristic (see Kelly and Karmel, 2009). Furthermore, an implication of going beyond compliance is a great deal more attention to interpretation as the manner in which international law creates effects in the world. Interpretation is pervasively determinative of what happens to legal rules when they are out in the world; and yet ‘compliance’ studies begin with the notion that to look at effects, we start with an assumed stable and agreed meaning to a rule, and whether it is complied with or obeyed, so understood. Standard search techniques reveal a large and burgeoning literature on ‘compliance’ and a general dearth of literature on ‘interpretation’ as central for international legal inquiry: to understand properly the working of international law in the world would probably entail reversing that emphasis.


  • We had the opportunity to present versions of this essay at the Université de Paris/NYU/Cardozo conference on constitutionalism and globalization, the International Studies Association annual meeting 2008, the IILJ International Legal Theory Colloquium and Hauser Scholars Forum at NYU, and Fordham Law School. Our thanks to participants on those occasions for helpful comments, and especially to Benedict Kingsbury, Joseph Weiler, Tom Lee and Cathy Powell, and also to Claire Kelly for very useful suggestions on a later draft.

  • 1

     See the case of H. L. A. Hart, discussed below.

  • 2

     See for a recent inventory of laws that are underenforced and often uncomplied with in the US, Wu, 2007.

  • 3

     It would also be at odds with sophisticated social science about why people actually do obey the law; see for example, Tyler, 2006. Tyler found that a sense of legitimacy of the law is a more important factor than fear of punishment in most people’s law-abiding behavior, and that there is a strong link between procedural justice and legitimacy.

  • 4

     See the socialization theory of Jinks and Goodman, 2003.

  • 5

     With respect to enhanced legalization in the WTO, Joseph Weiler notes: ‘Whether the shift in legal paradigm has been a victory for the Rule of Law or merely a victory for the rule of lawyers is a very serious matter on which the jury is still out. There are some very thoughtful actors and observers who are seriously wondering whether the “historical deal” has truly been beneficial to some of the deeper objectives of the WTO such as establishing stability and “peaceful economic relations” ’ (Weiler, 2000, p. 4).

  • 6

     See Prosecutor v. Tadic, Case No. IT-94-1, Appeals Judgment, International Criminal Tribunal for the Former Yugoslavia, 15 July 3599.

  • 7

     Joint and Separate Opinion of Judge McDonald and Judge Vohrah, Erdemovic, Case No. IT-96-22-A, at para. 75.

  • 8

     Thanks to Claire Kelly for useful thoughts on the distinction between acceptance of a norm and compliance with it.

  • 9

     Some of these elements are present in Kingsbury’s understanding of ‘publicness’ as a key feature of law, of central significance for the meaning of law in the context of global administrative law. See Kingsbury, 2009.

  • 10

     In these particular respects, we see merit in David Kennedy’s concerns about the apolitical or purportedly transpolitical outlook of humanitarian jurists. See Kennedy, 2004.

  • 11

     Interestingly, in the Minos, Plato has Socrates define the relevant question that begins the dialogue as ‘What is law for us?’ This evokes the possibility that the very character of ‘law’ is such that the ‘what is’ question cannot be posed outside the context of actual practices of law, or the givenness of law as a historical practice (and/or, perhaps, divine revelation) is the phenomenon to which analysis must be addressed.

  • 12

     Thus, in the domestic law context, studies of what motivates people to obey the law need to rely on survey tools. See Tyler, 2006. Since especially today international law ‘compliance’ engages in fact a wide range of actors – a fact, as we observe in this article, often ignored in the literature, which still in large measure focuses on behavior of ‘states’ as if they were autonomous, unique agents – such studies of motivation in the international law context would need to survey the various actors in question, both within the state apparatus formally responsible for determining compliance-related decisions on behalf of the state, and others to whom the rules are addressed in part directly (soldiers who may be personally criminally liable for war crimes, etc.). Some of the best literature, such as Janet Levit (2004; examining rules on concessional trade finance), does, at least anecdotally, attempt to grasp the complexity with which the motivations of different actors lead to compliance outcomes with respect to international legal norms.

Author Information

Robert Howse, Lloyd C. Nelson Professor of International Law and Faculty Director, Institute for International Law and Justice, New York University Law School.

Ruti Teitel, Ernst C. Stiefel Professor of Comparative Law, New York Law School and Visiting Professor, London School of Economics and Political Science, LSE Global Governance.