These are trying times for international organizations (IOs). After an extended period of expansion in their functions and autonomy, we may be witnessing a backlash. Not long ago, the main focus of scholarship about IOs was on accountability, a reflection of their growing influence. For some, the demand for accountability follows naturally from the principle that with power comes responsibility.Footnote ¹ For others, it reinforces a long-held belief that IO autonomy is a threat to democracy, the work of a global elite bent on empowering global governance bodies at the expense of national prerogatives and individual rights.Footnote ² This anti-globalist narrative has taken on new life with the rise of populism in the United States, Europe, and elsewhere. Its call for policies of economic protectionism and ethnic nationalism are captured by the slogan “America first” in the United States. Populist politicians claim to represent “ordinary people” and are suspicious of self-serving elites and so-called experts who occupy establishment institutions. While often cast as an appeal for governance based on the popular will, populism tends to be hostile to democratic constraints on strong leadership. Brexit and the victory of Donald Trump are the most obvious manifestations of this, but one can also point to the rise of populist parties—of the right and left—in other European countries,Footnote ³ and to the grip of strong leaders in Russia, Turkey, Venezuela, the Philippines, Egypt, and elsewhere.

The most full-throated “sovereigntist” response to this “globalist” threat is to tear down global governance institutions.Footnote ⁴ A more moderate response is to devise mechanisms for making these bodies more responsive to the constituents they serve, and more accountable for wrongs they commit. José Alvarez, professor of law at New York University School of Law and co-editor-in-chief of the American Journal of International Law, clearly falls on the side of the globalists but is attuned to the claims of sovereigntists on the left and right. In The Impact of International Organizations on International Law, he provides a nuanced account of the power of international organizations without celebrating them. He asks but does not directly answer whether we have too little or too much global governance (pp. 45–46). A course he delivered to students at the Xiamen Academy of International Law evolved into this sophisticated work of scholarship that is both a primer on legal theory and a highly perceptive review of the functioning of key IOs, building on his influential earlier book International Organizations as Lawmakers. Framed as a sustained argument about the inadequacies of legal positivism, he examines the United Nations (UN), World Health Organization (WHO), and various international tribunals to describe the post-positivist international legal order in which we now live.

Alvarez's central argument is that positivism as a legal theory fails to capture the current state of international law. He uses the impact of the United Nations (and IOs generally) to prove this point (p. 47), starting from the premise that the legal output of international organizations does not fall squarely within any of the traditional sources of law listed in Article 38 of the International Court of Justice (ICJ) Statute: treaties, custom, and general principles. Yet he argues that no single paradigm has replaced legal positivism and none seems be on the horizon (pp. 48, 412). Natural law does not hold much interest for Alvarez (it is never mentioned in the book), nor does a return to “formalism” as advocated by Jean d'Aspremont, Marti Koskenniemi, and others (pp. 412–13). In his view, the “constitutionalization” of the international legal system is premature, if not wishful thinking. While global administrative law may someday be the foundation of a “global administrative state,” we are not there yet. Indeed, one of the most appealing features of the book is that Alvarez seems not to be in a rush to find a successor to positivism. A pluralist conception of what counts as law and a more fluid picture of how law is made, interpreted, and applied can be liberating. As Alvarez states, it frees the mind to consider the characteristics of the law produced by international institutions, without “a preconceived idea of what those obligations need to be” (p. 347).

His method of argument is straightforward. In both the first and final chapters, he presents an idealized version of positivism: law emerges from state consent; it is distinct from morality and politics; the only sources of law are those articulated in Article 38 of the International Court of Justice Statute; and the international legal system is closed, meaning there is no need to draw on other disciplines to understand how it works (pp. 1, 344). From these core tenets, various propositions follow, which are introduced in the first chapter and challenged throughout the book: there is no soft to hard law spectrum (p. 7); judging is a “science,” an almost mechanical exercise of identifying rules and applying them to facts; treaties are like contracts and therefore should be interpreted narrowly, with an eye to original intent (pp. 29–32); customary law is not easily made because it requires near universal practice and a clear sense of obligation (pp. 39–42); and “general principles” are a thin source of law given how hard it is to find a consensual rationale for them (pp. 15–16).

This positivist ideal never existed in such stark terms. Alvarez himself suggests that the “golden age of legal positivism” is a myth (p. 413). His purpose in describing it that way is as a baseline for demonstrating how far the practices of international organizations are from that ideal. The essence of his claim is that many of the core assumptions of positivism do not hold up when measured against what IOs actually do.

In building his argument that we are living in a post-positivist legal order, he tends to downplay the counterarguments. The first chapter ends with a list of caveats, but these are mainly normative, i.e. about whether IOs have made the world a better place as opposed to whether they wield as much power (for good or bad) as he claims. They do not fully capture the backlash against “mission creep” by IOs in recent years. In addition to the pushback against the investment regime and the good governance agenda of the international financial institutions (pp. 397–98), the travails of the International Criminal Court, growing doubts about ambitious peace operations with nation-building mandates, and resistance to giving the WHO more authority to deal with infectious disease—coupled with the populist impulses described above—suggest that we may be entering a period of retrenchment. Alvarez is undoubtedly aware of all this, but it comes as an afterthought rather than a fully integrated part of his argument. Readers will forgive him for not dwelling on the ups and downs of the EU as the book is about global organizations, but his case would have been stronger if he had addressed reassertions of sovereignty such as Brexit head on.

That being said, Alvarez makes a persuasive case that a strictly positivist conception of international law cannot account for many innovations that have emerged from IOs. He adopts a capacious view of the law as embodying not just rules but also institutions and processes (pp. 21, 36–42). He observes that multilateral treaties and IO constitutions are not “the positivist contracts of old” (p. 38) but rather living documents that are (and should be) interpreted dynamically in light of changing circumstances (pp. 353, 366–69). He insists that there is a spectrum from non- to soft to hard law (pp. 42 and 359). He sees value (as well as dangers) in interdisciplinary approaches to understanding the law and how it functions (pp. 42–45). He notes that state consent is often attenuated (pp. 346–50). He boldly claims that IOs challenge the very notion of anarchy in international affairs, a treasured concept of realists and neoliberal institutionalists alike (pp. 346–47). And he backs up these theoretical propositions with a meticulous review of the actual practice of international organizations.

In his chapter on the UN Security Council, Alvarez points to the Council's unprecedented actions in dealing with Iraq between 1991–2013; its response to the Lockerbie bombing of 1988; its creation of ad hoc criminal tribunals for the former Yugoslavia, Rwanda, and Sierra Leone, as well as its referrals to the International Criminal Court. He includes a section on the Council's promotion of a “right to democracy,” which focuses on the intervention in Haiti and could have been expanded to include the cases of Sierra Leone, Cote d'Ivoire, and the Gambia, as well the democracy-promoting activities of consent-based multidimensional peacekeeping operations. He references the move to “smart sanctions” that target individuals rather than the citizenry of a country as a whole. He has a good section on the Security Council as a counterterrorism legislator, in which he makes the interesting point that Resolution 2178Footnote ⁵ on the recruitment of foreign terrorist fighters goes further than the better-known resolutions, 1373 and 1540, in that the obligations imposed are not an extension of treaty obligations. As Alvarez states, Resolution 2178 is “an even more definitive step by the Council toward ignoring or bypassing positivists' most revered and legitimate method for generating new international obligations: the time-consuming negotiation, ratification and implementation of a new treaty” (p. 125). Alvarez does not claim that all of these innovations had a positive outcome—indeed he is critical of many of them. Nor does he claim that any single action was a radical departure by the Security Council. But the cumulative effect is striking. Designed as a body to manage responses to security crises, the Council has become a creator, adjudicator, and enforcer of international law—taking on the functions of all three branches of domestic government—legislative, judicial, and executive.Footnote ⁶

Unfortunately, he devotes relatively little attention to the Responsibility to Protect (R2P) norm. R2P stipulates that governments have a responsibility to protect their own populations from genocide, crimes against humanity, war crimes, and ethnic cleansing. If they fail, the international community—acting through the UN Security Council—has the right (though not legal duty) to intervene. It is surprising that Alvarez does not spend more time on R2P because it is a bellwether for the Council's post-positivist agenda. The Libya intervention in 2011 put the concept to the test, with some claiming it was a vindication of R2P and others calling it the nail in the coffin. While Alvarez lays out the implications of that intervention effectively, he does not consider the only other case where R2P was explicitly invoked (Resolution 1975 on Côte d'Ivoire) or the failure to invoke it for Syria. Nor does he consider the relationship between the R2P concept and protection of civilian mandates in peace operations. In my view, the result of such an analysis is that R2P is struggling but not dead yet. From the perspective of legal positivism, it has not crystallized as hard law, but may yet have “normative ripples”—to borrow a term Alvarez uses in his earlier work International Organizations as Lawmakers. Footnote ⁷ If nothing else, it is an advocacy tool in a discursive process that can stimulate action when appropriate and constrain it when not.

That Alvarez devotes a chapter of close to fifty pages to the UN General Assembly is a welcome corrective to the tendency in writing on the UN to focus entirely on the Security Council. He details the diffuse ways in which the General Assembly functions as initiator, interpreter, and enforcer of the law. Evoking constitutional principles, he claims that it provides institutional checks and balances, and serves as a “gatekeeper” to statehood—for example by upgrading the status of Palestine from “permanent observer entity” to “non-member observer state” status (emphasis mine).

In his chapter on the World Health Organization, two features stand out: first is the power of expertise in IOs; and second, the normative role of the WHO. The expertise theme is important because a source of IO autonomy and influence is a secretariat's ability to generate new knowledge about global public policy challenges. Epistemic communities tend to coalesce in and around IOs like the WHO, and these communities often shape policy and legal debate (a point I return to below). Sometimes their efforts help crystallize hard law, such as the International Health Regulations.Footnote ⁸ Often they impact compliance, not through coercive enforcement but rather softer “naming and shaming” tools, persuasion, and socialization (p. 255). This relates to the normative function of the WHO. The only treaty negotiated and adopted there, the Framework Convention on Tobacco Control,Footnote ⁹ is described by Alvarez as a “managerial multilateral treaty” (p. 236). The commitments it creates are flexible and relatively non-threatening to sovereignty. It does not impose hard penalties for non-compliance or even judgments on whether a violation has occurred. Instead, fulfillment of the treaty is achieved through guidelines, periodic meetings, and the “the persuasive role of repeated discourse among participants” (p. 236). Alvarez does not elaborate on what I have elsewhere called a “discursive theory” of compliance,Footnote ¹⁰ but it is an interesting thread that runs through the WHO chapter and reappears in his conclusion (pp. 418, 420). Jon Elster speaks of the “civilizing force of hypocrisy”:Footnote ¹¹ the notion that when an actor expresses rhetorical commitment to a norm, it comes under pressure to match deeds with words, even if the original commitment is not sincere. In the sphere of international law, the pressure results from a process in which legal claims are subject to scrutiny by other actors who have the ability to impose reputational and social costs.Footnote ¹²

If the international legal system were more centralized, with most disputes settled in courts, then there would be little need for this discursive process as a compliance mechanism. But the much-celebrated “judicialization” of the international system tends to exaggerate the power of courts as the guardians, implementers, and enforcers of international law.Footnote ¹³ In his chapter on international adjudication, Alvarez rightly points out that binding dispute settlement is just one function of adjudication, and even that function is more complex than the term would suggest (p. 266). Courts are also fact-finders, lawmakers, and suppliers of what he calls rather cryptically “governance.” I say cryptically because he uses the term to capture everything from regulation, to “resolv[ing] conflicts between values,” to inducing states to modify domestic laws, to “steering” state behavior, to having global constitutional effects (pp. 310–11, 314, 317, 320). “Purveyors of public values” (p. 346) is the clearest articulation of what he has in mind. Surprisingly, he does not connect this more directly to his earlier scholarship on an “expressive mode” of judicial review. Commenting on the ICJ Lockerbie Footnote ¹⁴ and Bosnia Genocide Footnote ¹⁵ cases, he argued then that the Court did not, and probably never will, engage in direct judicial review of the Security Council, but some of its judgments and dissenting opinions sent “signals” or “warnings” to the Council to take care not to exceed the limits of its powers.Footnote ¹⁶ Similarly, while the European Court of Justice and national courts cannot engage in judicial review of the UN Security Council, they did so indirectly in Kadi Footnote ¹⁷ and similar cases of individuals being put on the Taliban/Al-Qaeda sanctions list (p. 323). For its part, the Council responded indirectly by providing more procedural protections to those listed, without accepting that courts could second-guess its decisions. In this dialogue between the Council and courts, judicial oversight occurs through more subtle mechanisms than a strict positivist view of law would suggest is necessary. Moreover, that “governance function” is not performed by courts alone. Other organs of the UN often comment on and criticize without purporting to formally strike down decisions of the Security Council: the General Assembly, the secretary-general, and the high commissioners for refugees and human rights, to name a few. Judicial dialogue is also a feature of the EU and some national systems.Footnote ¹⁸

This brings me to Alvarez's ambivalent treatment of “principal-agent” theory as it applies to international courts and IOs generally. He introduces the concept of “delegation” from state principals to IO agents early in the book (pp. 25, 50), expands upon it in Chapter V on courts, and returns to it in the Conclusion (p. 397). He expresses skepticism about its explanatory power in some places, yet invokes it uncritically in others.Footnote ¹⁹ Ultimately, the skeptical approach is more satisfying and, to my mind, more consistent with Alvarez's overarching thesis about the limits of positivism.

The simplest kind of delegation is when a state or states explicitly grant authority to an international organization. Principal-agent theory can be helpful to understand certain features of IO behavior as long as it is easy to identify the principals, the agent, and the powers conferred on the agent by the principals. It is less useful to explain acts of implied delegation, which are more common in the realm of IOs. Consider the UN Security Council. Can one reasonably assume that in establishing the UN, the founding members (the principals) intended for the Security Council (the agent) to engage in intrusive state-building, create international criminal tribunals, or legislate on terrorism? Is this best understood as falling within the agent's margin of discretion or as exceeding its authority? Does is it even make sense to speak of 193 member states as a collective principal?

More difficult still is what I have called “attenuated delegation,”Footnote ²⁰ a term Alvarez uses somewhat differently. International relations theorists speak of the unintended consequences of delegation. IOs may acquire their own preferences, separate from those of the member states, leading to policy “drift,” “slippage,” or “shirking.”Footnote ²¹ In principal-agent terms, this is undesirable from the principal's point of view, something to be controlled. From other perspectives, IO autonomy is both necessary and desirable. This is true not only of courts but also the executive heads of IOs, such as the secretary-general of the UN, who is expected to act somewhat independently. While some of the powers he exercises are beyond direct member-state control, they are not unintended. The growing role of non-governmental actors in lawmaking by international organizations is also revealing.Footnote ²² Other than in the International Labour Organization, they are not delegated lawmaking powers per se, but do impact the development of international law. In these cases of attenuated delegation, the principal and agent may be so far removed from each other that it is misleading to suggest they are in a relationship at all. Thus, as Alvarez points out, while principal-agent theory can help to illuminate some features of international adjudication (pp. 269–71), much of what happens in IOs is not well-captured by that frame of analysis.

As noted above, a great strength of The Impact of International Organizations is that it presents a nuanced picture of how law is made, interpreted, and applied. In this more pluralistic, fluid conception, many IO tasks impact the development of international law. This includes operational activities, as well as more explicitly normative behavior like the negotiation and adoption of treaties, or GA resolutions that provide evidence of customary law, or the recommendations adopted by specialized agencies.Footnote ²³ A good deal of international law builds up as a body of practice that sets precedents and ultimately becomes a more generalized rule. The traditional place to look for that practice is states; increasingly it is the practices of IOs. They do not fall neatly into any of the sources of law identified in Article 38 of the ICJ Statute, nor are they entirely inconsistent with those sources. Some of these practices may be seen as authoritative interpretations of a treaty, for example the impact of electoral assistance programs on the right to political participation embodied in the International Convention on Civil and Political Rights.Footnote ²⁴ Some may contribute to customary law formation, like the guiding principles on internal displacement.Footnote ²⁵ The “general principles” of international law may include transparency, participation, and the giving of reasons in decision-making within IOs.Footnote ²⁶ Indirectly, all of this is driven by states, but the practices are one step removed—or many steps when delegation is highly attenuated—from state consent.

IO practice alone cannot make law; it must be accompanied by the functional equivalent of opinio juris. Always hard to gauge, the best way of knowing whether that sense of legal obligation exists is to follow the paper trail. What are states saying about what the IO is doing? How are other states reacting? Building on the pioneering work of Oscar Schachter, Abram and Antonia Chayes describe a process of justificatory discourse when claims are made, challenged, defended, and critiqued in a sustained “diplomatic conversation.”Footnote ²⁷ I have argued that this discourse occurs within “interpretive communities” that coalesce around IOs.Footnote ²⁸ Courts are one venue for this, but not the only or necessarily most important one. And even judicial opinions are often no more (or less) than contributions to a dialogue among international actors.

Let me conclude by connecting the notion of interpretive communities to two other concepts that have gained traction in the international relations literature: epistemic communitiesFootnote ²⁹ and communities of practice.Footnote ³⁰ All three are associated with what has been called the “pragmatist” turn in IR scholarship, which sees knowledge, learning, and shared understandings as emerging from social practices, not through deduction from abstract principles. These communities exist not because of who they are or where they work, but what they do. In an early definition of a community of practice, the organizational behavior theorist Etienne Wenger describes it as a group of people who interact with each other in a joint enterprise, who acquire competence in a shared repertoire of routines, concepts, words, tools and ways of doing things.Footnote ³¹ In international relations, Emanuel Adler and Vincent Pouliot have written about communities of diplomats, traders, environmentalists, and human rights activists.Footnote ³²

The concept of interpretive communities was first introduced to international legal theory as a way of explaining treaty interpretation in the security realm, where court decisions are rare.Footnote ³³ Legal interpretive communities are a special type of community of practice because they are engaged in a highly structured professional enterprise (the practice of law) that is characterized by a distinctive form of discourse, with its own set of conventions, categories of understanding, and stipulations of what is relevant and irrelevant. One can speak of a community of diplomats, but the enterprise of diplomacy is far less bounded than that of law (in fact engaging in legal discourse is but one of many “practices” that characterize diplomacy). Moreover, legal interpretive communities tend to coalesce in and around international organizations, not only courts but also political bodies like the Security Council or secretariats like the European Commission. In that way, the practices become institutionalized, though not formally.

From the point of view of international legal order, the bounded nature of legal discourse can be good because it gives the law some bite even when formal dispute settlement and coercive enforcement are lacking. But it does raise legitimacy questions. Is the interpretive community hegemonic—merely a reflection of material power in the international system, with the dominant states dominating the discourse? Is it technocratic, privileging legal expertise over more democratic decision-making? Is there ever a truly global interpretive community coalescing around an international legal norm, or do we have to think in terms of multiple communities dispersed across regions and legal systems?Footnote ³⁴ If international organizations are places where legal interpretive communities coalesce, we need to look at who these “practitioners” are and how they operate. That points to both a research and a reform agenda. The research agenda is to explore how cohesive interpretive communities are in different areas of law. One can imagine a fairly cohesive community of trade lawyers, or international criminal lawyers. What about the law on the use of force? Is interpreting that law really a shared enterprise? Just as there are varying commitments to positivism in different parts of the world (pp. 18, 49), there are varying views on the legality of preemptive military action or the applicability of international law to cyberattacks. The reform agenda relates to the potential for hegemonic and/or overly technocratic IOs. How can IOs be designed to account for the value of power and expertise without marginalizing key stakeholders and silencing important voices? Alvarez's comprehensive account of the “state of the art” of international law and organization provides a rich theoretical and empirical basis for pursuing both agendas.