Hostname: page-component-586b7cd67f-dsjbd Total loading time: 0 Render date: 2024-11-25T06:22:30.094Z Has data issue: false hasContentIssue false

The Jus ad Bellum’s Regulatory Form

Published online by Cambridge University Press:  22 April 2018

Monica Hakimi*
Affiliation:
Professor of Law, University of Michigan Law School.
Rights & Permissions [Opens in a new window]

Abstract

This article argues that a form of legal regulation is embodied in decisions at the UN Security Council that condone but do not formally authorize specific military operations. Such decisions sometimes inflect or go beyond what the jus ad bellum permits through its general standards—that is, under the prohibition of cross-border force and small handful of exceptions. Recognizing that this form of regulation is both part of the law and different in kind from regulation through the general standards should change how we think about the jus ad bellum.

Type
Article
Copyright
Copyright © 2018 by The American Society of International Law 

I. Introduction

The international jus ad bellum is a notoriously contentious body of law. By most accounts, it prohibits states from using cross-border forceFootnote 1 except: (1) with the consent of the territorial state,Footnote 2 (2) pursuant to the authorization of the United Nations (UN) Security Council,Footnote 3 or (3) in individual or collective self-defense.Footnote 4 The content of these exceptions is a regular source of friction. In addition, there are longstanding debates about the existence of other exceptions. The most prominent such debate centers on whether states may use force, without the Security Council's authorization, to avert a humanitarian crisis. The dominant view is that they may not,Footnote 5 but some disagree.Footnote 6

For all of the disagreement about the jus ad bellum’s content, however, little attention has been paid, at least in any systematic way, to its regulatory formFootnote 7—meaning the modes for expressing its content as concrete directives that structure legal arguments and decisions. There is a surprising degree of consensus here. The jus ad bellum’s content is widely thought to take the form of a blanket prohibition and a handful of discernible exceptions. Again, the precise content of these standards is unsettled. But few would say that they are either so capacious as to lack objective meaningFootnote 8 or so multiple and fine-grained as to capture all of the factors that might affect whether using force in a given case is good policy.Footnote 9 The conventional account, which almost everyone accepts, defines the jus ad bellum with a small set of generally applicable substantive standards.Footnote 10

These standards are known to be dissociated from some state practice. States periodically conduct and, as a group, endorse operations that cannot plausibly be justified under them. What's significant is that, in these cases, states do not always try to hide their conduct or defend it by advancing controversial interpretations of the standards. Sometimes, they indicate that their positions rest on the facts of a case and are not meant to reflect or inform any general standard of law.

An April 2017 incident is illustrative. The United States struck a Syrian airfield after evidence surfaced that the Syrian regime had again used chemical weapons against its people.Footnote 11 The U.S. operation did not satisfy any of the accepted standards for justifying cross-border force. Syria had not consented to it, the UN Security Council had not authorized it, and it was not taken in self-defense. Further, it did not look like a typical humanitarian intervention. President Trump explained that the United States acted only “to prevent and deter the spread and use of deadly chemical weapons,” not to avert the many other atrocities that were being committed in Syria.Footnote 12 The United States has advanced expansive interpretations of the jus ad bellum’s standards in other contexts,Footnote 13 but it did not do so here. It listed a set of case-specific factors that, in its view, warranted the use of force.Footnote 14 Other states largely followed suit. They overwhelmingly supported the U.S. operation but did not try to justify it under the general standards.Footnote 15 The conventional account, which defines the jus ad bellum entirely through those standards, does not capture and is inadequate to explain the law's operation in this case.

Still, analysts almost invariably use that account to assess such cases. Most commentators claimed that the U.S. action in Syria was unlawful because it could not plausibly be justified under the general standards.Footnote 16 The problem with this claim is that states themselves treated it as if it were lawful. Their broad support meant that the prohibition of the use of force lacked both operational relevance and normative bite. Other commentators picked up on that point to insist that the jus ad bellum contains or is developing a standard on unauthorized humanitarian interventions—one that, with the prohibition, would govern all such interventions and help define when they are lawful.Footnote 17 This position also distorts what happened. States clearly signaled that their support for the U.S. operation was contingent on the facts and not a reliable indicator of how they would respond in the future. They did not purport to apply or establish a standard of general applicability.

There are reasons the conventional account of the jus ad bellum is so resilient, despite its evident limitations. Chief among them is that many consider it to be the only workable option for maintaining a system for collective security rooted in the UN Charter. Things might be different if decisions to conduct or condone deviations from the general standards were made in formal arenas. That way, the decisions might reflect accepted legal principles. But the jus ad bellum’s constitutive structure is not so refined. To a large extent, decisions on the use of force are made by individual states acting in unstructured settings. When states deviate from the standards in these settings, it looks to many like law buckling to power. The worry is that such incidents weaken the jus ad bellum’s salience and push states down a slippery slope toward deregulation. Thus, despite some disagreement about the jus ad bellum’s content in Syria—about whether the U.S. action was lawful—commentators seemed uniformly to agree that the incident betrayed the jus ad bellum’s irrelevance and risked its further deterioration.Footnote 18 If the jus ad bellum regulates force entirely through its general standards, then not invoking or applying those standards is tantamount to ignoring the law.Footnote 19

This article challenges that thinking by examining a dimension of state practice that contradicts it. States that undertake or endorse operations that lack good justification under the general standards sometimes ask the Security Council to deliberate on and, in some way short of authorizing, condone their conduct. As a descriptive matter, the Council's non-authorizing support—as expressed, for example, in a press release or presidential statement—confers authority on the operation. It makes an operation that would otherwise be legally suspect easier to justify and harder to challenge in law. Here, the jus ad bellum regulates states not (or not only) through the general standards but (also) through processes at the Council for approving specific operations that go beyond what the standards permit. The form of regulation is more procedural and particularistic than it is substantive or generalizable.

As shorthand, I call this form of regulation the jus ad bellum’s “informal regulation.” It is “informal” in the sense that it does not reflect and is not meant to affect the black-letter doctrine. The doctrine defines the jus ad bellum as the conventional account does—entirely with the general standards. Nevertheless, the informal regulation is best understood as a feature of, not somehow external or opposed to, the jus ad bellum. It shapes how the law is defined or applied in concrete settings. And it has that effect by trading on the Security Council's authority, which is established in law. Such regulation can be analogized, for example, to common law decisions in equity or by jury nullification. There, like here, decisions that are crafted for specific facts do not reflect a general standard of law but have legal force by virtue of the institutional settings in which they are made.

My claim that such regulation is part of the jus ad bellum does not require a big conceptual leap. It strongly resonates with two existing strands of international legal theory. First, Michael Reisman has described the jus ad bellum as consisting of a “myth system” and an “operational code.”Footnote 20 The myth system is “the official but largely unapplied normative system,” while the operational code reflects “the way things are actually expected to be done.”Footnote 21 Reisman's theory illuminates that there are “discrepancies between certain parts of the formal legal system and the way decisions are actually going to be made.”Footnote 22 Because I also address those discrepancies, my claim can be mapped onto his; the informal regulation can be characterized as an element of the operational code.Footnote 23 Second, my claim draws on work concerning the Security Council's normative power. This work shows that the Council shapes international law in manifold ways and by acting with varying degrees of legal formality.Footnote 24 The informal regulation is an example of it affecting the law's content without clearly satisfying any formal criteria of law.

Although that claim is not itself radical, the article's broader ambitions are. The article aims fundamentally to change how international lawyers and policymakers think about the jus ad bellum. In arguing that the informal regulation is both part of the jus ad bellum and different in kind from regulation through the general standards, it highlights the limits of using the conventional account to do descriptive, analytic, or normative legal work. And it offers an alternative framework that better equips us to understand, strategize about, and appraise the law's operation. The practical payoffs are substantial. For example, the article shows that those who adhere to the conventional account routinely overstate both the doctrinal significance of particular cases and the evidence of the law's irrelevance or erosion.

The article also identifies new possibilities for legal reform. At first glance, the informal regulation might seem simply to devolve decisionmaking from the Security Council to individual states. Clearly, the Council is more peripheral to decisions to use force when states employ the informal regulation than when they strictly apply the general standards and insist that the only option for making an operation lawful is to obtain the Council's authorization. However, that metric for comparison is not particularly useful. Whether the informal regulation ultimately detracts from the Council and loosens the restrictions on unilateral force depends on how and with what effect the jus ad bellum would regulate states in its absence. As I will explain, the best way to fortify the jus ad bellum and preserve the Council's primacy in today's security environment is almost certainly to use more, not less, of the informal regulation. This insight is particularly valuable at the current moment. Although claims about the jus ad bellum’s erosion are often exaggerated, its regulatory purchase has been diminishing. It needs a corrective that is both meaningful and responsive to contemporary realities.

The article proceeds as follows. Part II sets the stage for my argument by describing the jus ad bellum’s constitutive structure—the processes and institutions for making use of force decisions. Part III then explains how the informal regulation differs from regulation through the general standards and is part of the jus ad bellum. Part IV examines incidents in which such regulation is used. These incidents all support my theory of regulation and betray the descriptive and analytic flaws in the conventional account. They show that this account fails to capture and consistently warps how international lawyers analyze this practice. Part V then explains why using the informal regulation is probably preferable to trying to regulate states entirely through the general standards. My goals throughout the piece are not to advance specific positions on the jus ad bellum’s content but to alter how we conceive of this area of international law and thus to create space for new thinking and research on it.

II. A Primer on the Jus ad Bellum’s Constitutive Structure

The jus ad bellum’s constitutive structure derives from the UN Charter but is neither as centralized nor as settled as that text suggests. Article 2(4) of the Charter prohibits states from using force “against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.”Footnote 25 That language has long been understood to prohibit any use of force by one state in another without that latter state's consent.Footnote 26 The Charter identifies only two exceptions. First, Article 51 recognizes that states may use force unilaterally—meaning outside the Security Council—in individual or collective self-defense.Footnote 27 Second, they may decide to use force through the Council. The Council is a collective body that is charged with maintaining international peace and security, so it has the potential to curb nationalistic impulses and advance policies that, on the whole, enhance rather than detract from human security.Footnote 28

The Charter's original design never came to fruition.Footnote 29 The Council was largely paralyzed during the Cold War, and though it has been more active on peace and security issues since then, it still does not exercise as much control over use of force decisions as the Charter text envisions. States never satisfied their commitment, in Article 43, to place under the Council's command a set of standing armed forces.Footnote 30 If they had, the Council would be able to implement its own decisions. Instead, it relies on individual states to do that work; it authorizes them to use force when it determines that such force is appropriate. This creates what is sometimes called “agency slack.” States that act pursuant to a Council authorization decide, at least in the first instance, how to conduct an operation, so they might stray from what the Council itself would have done. Thus, even when the Council authorizes an operation, critical decision functions have devolved to individual states.

Analysts broadly recognize that dynamic,Footnote 31 but many still assume that use of force decisions must be channeled as much as possible through the Council.Footnote 32 In fact, questions of precisely when and how the Council ought to participate in such decisions are themselves the central contest in the jus ad bellum.Footnote 33 This contest is evident in disputes about the jus ad bellum’s content. Consider current debates about whether and, if so, when states may use defensive force against nonstate actors or in anticipation of attacks that are not temporally immediate.Footnote 34 The more Article 51 permits defensive force in those circumstances, the more it shifts decisionmaking from the Council to states acting unilaterally. Debates about the content of Article 51 are thus also about the balance between channeling use of force decisions through the Council and licensing states to make those decisions on their own. The intensity and persistence of these debates show that the balance is contested.

To be clear, unilateral decisions on the use of force do not occur in a vacuum. They feed into a decentralized process in which individual states act or react in concrete incidents. This process is unstructured, so extracting the content of decisions that are made through it—ascertaining the policy message of particular incidents—can be difficult. However, as Jacob Katz Cogan and I recently showed, the decentralized process consistently produces decisions that are more permissive of unilateral force that do international institutions, like the UN Security Council or the International Court of Justice (ICJ).Footnote 35 In other words, when states act outside of those institutions, they routinely conduct and tolerate operations that are incompatible with, or reflect very expansive interpretations of, the standards that the institutions themselves pronounce. Thus, just as the jus ad bellum’s content determines the decisionmaking process that states are expected to use in any given context, the process that they use very often dictates the content of their decision(s). Because states themselves participate in both kinds of processes—institutional and decentralized—neither is necessarily a better barometer of their overall preferences on the use of force. Rather, their practice reveals that their preferences are shifty, contingent, and routinely in play in discrete incidents.

III. Two Theories (and Forms) of Regulation

Because decisions on the jus ad bellum are often conflictual or muddled, identifying its content—determining when it permits or prohibits the use of force, either in general terms or in concrete cases—is more of an art than it is a science.Footnote 36 Almost all of the secondary literature on the jus ad bellum focuses on that question and reflects the conventional account that I intend to challenge. Below, I describe this account in more detail and explain how it differs from my own.

I underscore at the outset that these two theories of regulation have important similarities. First, each is compatible with a broad range of policy positions on the use of force. The theories are not about what the jus ad bellum prescribes but about how—in what regulatory form—it conveys its prescriptions. Second, each is meant to structure our legal thinking on the jus ad bellum. It describes how the jus ad bellum defines its policy content and provides an analytic framework for making decisions in, arguing about, or appraising specific incidents. Third, each theory posits that the jus ad bellum regulates states through a set of generally applicable substantive standards—the Article 2(4) prohibition and a small handful of exceptions.

The key difference between the two theories is that, in the conventional account, the general standards are the only form of jus ad bellum regulation. An operation that is justifiable under one of the standards is lawful, and an operation that is not justifiable under any of them is unlawful. By contrast, I argue that the standards are supplemented by another form of regulation—what I am calling the “informal regulation.” With such regulation, the jus ad bellum’s content is established through processes at the Security Council that operate independently of the general standards and govern discrete cases at a time. Compared to regulation through the general standards, this form of regulation is considerably more: (1) procedural, and (2) particularistic. So, I begin by explaining how the conventional account addresses those two dimensions of regulation.

A. General Standards

1. The procedural dimension

The conventional account of course recognizes that the jus ad bellum has a procedural component. The law plays out through various decisionmaking processes, both institutional and decentralized. Yet this account posits that, in order for a process to be legally relevant, it must breathe life into a general standard. It can shape the law's content only in that regulatory form.

Take the standard that permits force pursuant to the Security Council's authorization. Of all the general standards, this one has the most obvious procedural component.Footnote 37 It is implemented through the process of adopting an authorizing resolution at the Security Council. According to the conventional account, the process is legally significant because it effectuates a general standard. Indeed, an operation that is conducted under color of a Council authorization is almost always assessed by reference to that standard—whether the Council authorized it—not in terms of the process that the Council used to get there. For example, no one seriously contends that the legality of an authorized operation turns on procedural factors, like whether the resolution was adopted unanimously or just barely, or in an open or closed session. So long as the Council authorized it, it is accepted as lawful. By the same token, if the Council had a very productive meeting but never adopted an authorizing resolution, the operation would be depicted as unlawful, unless it could be justified under a different standard. In this account, the Council's process is just a mechanism for operationalizing a general standard. It does not have any legal force that is independent of that standard.

That conception of the relationship between process and substance has doctrinal support. The two principal sources of international law, both generallyFootnote 38 and in the jus ad bellum,Footnote 39 are treaties and customary international law (CIL). A treaty is binding on the states that ratify it.Footnote 40 Because essentially all states have ratified the UN Charter, its substantive standards on the use of force are, for all intents and purposes, universally binding. So, as a matter of doctrine, the Charter prohibition and its exceptions govern all situations involving cross-border force. CIL operates alongside those standards. The process for creating CIL is much more unstructured, but its content is generally thought to take the same regulatory form. In order for a norm to be CIL—and treated as legally relevant—it must attain widespread support in state practice and perceptions of legality (opinio juris).Footnote 41 At that point, it is, like the Charter provisions, a substantive standard of general applicability.

Thus, the doctrine bolsters the conventional account of the jus ad bellum in three distinct but related respects. First, it defines the jus ad bellum as a set of generally applicable substantive standards. The standards derive from the Charter and are further refined or adapted by CIL. Second, in this picture, legal processes affect the law's content only through such standards. They do not have independent legal force. Third, because authoritative bodies, like the ICJ, apply the doctrine, their decisions further reinforce the conventional account.Footnote 42 International lawyers who give great weight to those decisions or who otherwise adhere strictly to the doctrine naturally also gravitate toward that account.

What's striking is that the conventional account animates even the work of scholars who criticize the doctrine or advance theories of international law that are more processual. For example, Sean Murphy has suggested that the doctrine's rigid application has left the jus ad bellum static and dissociated from current expectations.Footnote 43 He offers three ways to regulate force going forward: (1) accept uncertainty about whether and how the Charter standards govern contemporary security situations, (2) reaffirm that the Charter standards are law and insist that they be operative, or (3) adopt new standards that better capture modern sensibilities.Footnote 44 Each of these options is consistent with the conventional account. Each presupposes that the jus ad bellum’s content is established entirely in a set of generally applicable substantive standards.

Similarly, Harold Koh has applied to the jus ad bellum his theory of the transnational legal process. Koh's theory focuses on the myriad ways in which global actors interpret, converge on, and internalize specific prescriptions.Footnote 45 In the context of defending unauthorized humanitarian interventions, he emphasizes that the transnational legal process is critical to the jus ad bellum; it allows the law to adapt to new security challenges.Footnote 46 But he assumes that any normative development would take the form of a general standard. For example, Koh criticizes states, like the United States, that have conducted these interventions without trying to justify them in law—by which he means through the presentation of that kind of standard.Footnote 47 He also proposes his own such standard.Footnote 48 Koh focuses on the transnational legal process because it is a way to create and solidify support for a general standard of law.

Jutta Brunnée and Stephen Toope have likewise applied their theory of international law to the jus ad bellum.Footnote 49 They contend that international law is established and sustained through an ongoing, interactive process in which global actors “collaborate to build shared understandings and uphold a practice of legality.”Footnote 50 This theory breaks with the sources doctrine in certain respects,Footnote 51 but it shares the doctrine's approach to process and substance. Thus, when Brunnée and Toope apply the theory to the jus ad bellum, they conclude that nonconsensual force is lawful only “on exceptional grounds, notably the right to self-defence or Security Council authorization.”Footnote 52 Here again, the jus ad bellum’s processes affect its content entirely through those general standards.

2. The level of generality

Because the standards in the conventional account are so few, and because they are meant to govern all possible scenarios involving cross-border force, they are formulated at high levels of generality.Footnote 53 In other words, they are not so fine-grained as to account for all of the considerations that might, in an ideal world, affect the law's content. Instead, they group together and subject to similar treatment cases that are alike in certain respects but sometimes also have significant differences.

Consider the standard on self-defense. In conjunction with Article 2(4), this standard establishes the law's content for a very broad category of cases. According to the conventional account, it determines the legality of most, if not all, operations that are conducted with neither the Security Council's authorization nor the territorial state's consent. There have been efforts to particularize the standard for narrower subcategories of cases—for example, cases in which states act to protect their own nationals,Footnote 54 repel nonstate actors,Footnote 55 avert anticipated attacks,Footnote 56 or respond to hostile cyber activity.Footnote 57 But each of these subcategories still covers many real-life situations.

Further, the standard for each has not become significantly more fine-tuned over time. There continue to be unresolved, at times heated debates both about how to formulate the standard at the subcategory level and about how to apply a given formulation in discrete cases.Footnote 58 Such debates linger because authoritative institutions, like the ICJ and Security Council, rarely define self-defense with more particularity.Footnote 59 And while states routinely invoke the standard in the decentralized process, the extent to which the decisions in this process establish its content is often ambiguous or contested.Footnote 60 To be clear, I am not saying that the standard on self-defense is imprecise, although it might be. I am saying that, whatever it entails, it governs a very broad category of cases and has hardly, if at all, been particularized for more discrete subcategories. It operates at a high level of generality.Footnote 61

Because the conventional account posits that the jus ad bellum regulates states entirely through such standards, it prioritizes two methods of analysis that are typical of reasoning by generalization. Each method is used to mediate between a general standard, as it is articulated, and its concrete application. The first method is for evaluating specific conduct. It accepts that a given standard is entrenched and controlling, and then “generalizes down” from that standard to appraise the conduct at issue. For example, whatever Article 51 permits, generalizing down from it would mean deciding that conduct that falls within its scope is lawful.

The second method is not for appraising specific conduct but for defining or refining the standard itself. The task here is to identify the proposition for which an incident stands and then to interpret that proposition into a more broadly applicable standard—to “generalize up” from a single incident to a standard that would govern an entire category of cases. This method requires some legal construction. Any incident can stand for multiple propositions, defined at varying levels of generality. However, because the jus ad bellum’s standards are, in the conventional account, highly generalized, analysts tend to operate at that level. Generalizing up usually means interpreting an incident into one of the existing standards or arguing that a new standard of comparable generalizability has emerged.

The two methods can be used in the very same case because states are both the subjects and the objects of international law; they simultaneously are governed by the jus ad bellum (generalizing down) and help to create it (generalizing up). For example, the methods are apparent in the debate on unauthorized humanitarian interventions. Such interventions are difficult to justify under the UN Charter because they are inconsistent with the longstanding interpretation of Article 2(4) and not covered by the Charter exceptions. Further, states have not altogether ratified a new treaty to permit them.Footnote 62 As for CIL, the operational practice and evidence of opinio juris point in different directions.Footnote 63 States have periodically endorsed actions that can be characterized as unilateral humanitarian interventions,Footnote 64 but most states have declined to support a general standard to that effect.Footnote 65

The legal debate centers on whether their behavior nevertheless evinces such a standard. The majority view is that it does not.Footnote 66 Those who take this view insist that the Article 2(4) prohibition remains entrenched and controlling. They generalize down from that prohibition to contend that all unauthorized humanitarian interventions are unlawful. By contrast, analysts who argue that they can at times be lawful typically use the second method. They generalize up from the incidents in which states have widely endorsed these interventions to claim that the jus ad bellum now has a general standard for them—a standard that applies in all humanitarian crises and helps resolve whether any given intervention is lawful.Footnote 67 Thus, although international lawyers disagree about whether the jus ad bellum ever permits unauthorized humanitarian interventions, they widely accept that any prescription would be generalizable.

B. Particularistic Processes

Whereas the conventional account posits that the jus ad bellum’s content is established entirely through the general standards, I contend that it can also be established through the informal regulation—processes at the Security Council for approving operations that go beyond what the standards permit. In the next part of the article, I discuss examples of states using such regulation. Here, I explain how it is both part of the jus ad bellum and meaningfully different from regulation through the general standards.

1. Authoritative processes

Much international legal regulation is procedural in form. States use various reporting, oversight, or review processes to regulate conduct when they could not feasibly prescribe all of their preferences in advance and in generalizable terms.Footnote 68 Some readers might contend, however, that any process for condoning force beyond what the general standards permit is inherently extralegal because it is not established in law. Another way of putting the objection is that, since the UN Charter and CIL define the jus ad bellum entirely through the general standards, decisions that stray from those standards must be unlawful. This objection reflects the black-letter doctrine on the use of force. But the fact that it does just confirms that the doctrine is at times dissociated from and not especially useful for describing or analyzing the law's operation.Footnote 69

It does not follow that decisions that are made through the informal regulation operate outside of and in opposition to the law, or that they are only politically and not legally relevant. These decisions probably qualify as law under certain jurisprudential theories,Footnote 70 but I do not rely on any one such theory here. To sustain my claim, it is enough to show that the jus ad bellum consists of more than just the doctrinal standards. It consists of a whole set of social institutions, processes, expectations, and practices. The informal regulation is part of that mix. It affects the law's content in concrete cases by stripping the Article 2(4) prohibition of some or all of its normative purchase. And it has that effect because it rests on the Security Council's legal primacy.

The Security Council has broad authority both to establish its own working procedures and to “decide what measures shall be taken … to maintain or restore international peace and security.”Footnote 71 As discussed, states do not consistently take a single position on when or how the Council ought to participate in use of force decisions,Footnote 72 but no state seriously contests its pride of place in the jus ad bellum. It retains that position of authority because it continues to be the preeminent collective institution on the use of force.Footnote 73

The critical question is whether the Council affects the law only by satisfying certain formal criteria or also in other ways. That question has already been persuasively answered. The actions and pronouncements of international institutions are routinely treated as legally salient, even when they do not appear in binding legal form.Footnote 74 Such authority might seem squishy, fleeting, or hard to pin down, but it nevertheless is real. It is evident, for example, in the common practice of citing Security Council resolutions as solid, if not dispositive, evidence of the law.Footnote 75 The Council's resolutions are cited this way even when they are not formally binding, and they do not neatly satisfy the sources doctrine.Footnote 76 The Council's normative power over the jus ad bellum is, quite simply, more expansive than the doctrine recognizes.

The Council uses various techniques to communicate its policy positions on the use of force. Sometimes, it adopts a resolution that plainly authorizes force. When it does, it effectuates a general standard, and any operation that is implemented pursuant to its authorization will be widely accepted as lawful. But the Council can also support an operation in other ways—in resolutions that lack authorizing language, presidential statements, press releases, or meeting records.Footnote 77 What matters in these latter cases is not so much the specific contours of the process but the fact that it is institutionalized at the Council. When the Council meets, considers, and condones an operation, it confers authority on the operation, making claims of illegality harder to sustain.

To be sure, the authority that the Council confers through these processes might at times be compromised. For instance, if the Council fails to speak in a unified voice, or if its approval seems tepid, it might not counterbalance the perceptions of illegality that come with deviating from the general standards. In that event, the operation's lawfulness might be a matter of some debate. But this in itself would be unremarkable. With both forms of regulation, there can be cases in which an operation's lawfulness remains uncertain or contested. Just as states do not always prevail on the law when they try to justify an operation under a general standard, neither do they always prevail when they trigger the informal regulation. Even so, the basic point stands: when the Council clearly communicates that it condones an operation, it helps legitimize that operation in law.Footnote 78

This point is to some extent acknowledged in the scholarly literature but remains clouded by the conventional account. Two lines of thought are worth addressing. First, other scholars recognize that the Security Council sometimes legitimizes operations without expressly authorizing them. That recognition is evidence that the Council's normative power extends beyond its interpretation or application of the general standards. Yet efforts to explain the phenomenon in law almost always circle back to the general standards. The most prominent explanation is that the Council's support for an operation constitutes an implicit authorization.Footnote 79 That explanation reflects the conventional account because it squeezes the practice into a general standard. It is unpersuasive. The Council has fairly standard language for authorizing force. It sometimes declines to use that language, even when it appraises and condones an operation. Indeed, in most of the incidents that I discuss, the Council seemed disinclined to authorize operations that states conducted anyway.Footnote 80 Claims of implicit authorization distort what happened in such cases and elide what is really at stake—which is that decisions to use force were made outside, not through, the Council.

Second, my argument is in some respects similar to a position that received considerable attention at the turn of the century, after the North Atlantic Treaty Organization (NATO) intervened in Kosovo, without the Security Council's authorization, to stop a humanitarian crisis. At the time, several prominent international lawyers claimed that the intervention was unlawful but still, in some way, legitimate or excusable.Footnote 81 Thomas Franck's version of this claim is probably the one that comes closest to my own. He contended that incidents like Kosovo present a “confrontation between the strict, literal text of the Charter and a plea of justice and extenuating moral necessity.”Footnote 82 According to Franck, states may use the UN's institutional processes to address that discrepancy and excuse particular operations after the fact. In his words, “the political organs of the UN system, which constitutes something approximating a global jury,” may “weigh considerations of legality against the common public sense of legitimacy” and condone an operation.Footnote 83

Franck's position is like mine because we both recognize that institutional processes can deprive the general standards of their legal effect. But our arguments also differ in important ways. By my account, the processes at the Council do not just confirm or reveal a “common public sense of legitimacy.”Footnote 84 These processes themselves legitimize the operations in law; they do so because the Council's authority is legally salient. Significantly, Franck conceded that there is not a meaningful difference between calling an intervention lawful (as I might) and maintaining that it is unlawful but excused (as he did).Footnote 85 He and others who insisted that the Kosovo operation was unlawful did so because they were profoundly committed to the conventional account and assumed that recognizing the lawfulness of one intervention would require establishing a general standard and opening the door to many others.Footnote 86 I am challenging that account and arguing that the associated assumption is wrong.

2. Particularistic decisions

Unlike the jus ad bellum’s general standards, which govern many different fact patterns, the informal regulation is highly particularistic. The Security Council condones an operation while addressing the specific circumstances of the case. The Council's position is crafted for and meant to apply only in that case, so it is not easily or automatically transferable to other cases. In fact, the Council usually refrains even from articulating a generalizable justification for its position. The member states seem to prefer that opacity either because they disagree on the policy rationale for supporting the operation or because they agree but want to avoid establishing the grounds for a precedent.

Particularistic decisionmaking is not new or controversial for the Council. Council decisions on the use of force routinely address and by their terms govern discrete situations at a time. For example, a Council resolution that authorizes force applies only in the case at hand; it does not extend to other arguably similar cases. Of course, any decision might reflect broader policy considerations or shape expectations. If the Council condones an operation in one case, global actors might anticipate that it or states acting outside of it will do the same in other cases. But precisely because the Council's decisions are particularistic—because they are so intently focused on the situations presented and not rationalized in general terms—their future relevance is at best uncertain and in almost all scenarios dubious. I elaborate on this point and its policy implications for the jus ad bellum throughout the remainder of the article. For now, I emphasize that the informal regulation differs from the general standards in that it is much more particularistic. It helps establish the jus ad bellum’s content for one case, without any indication that it reflects or defines a more generally applicable substantive standard.

IV. Informal Regulation in Practice

I turn to examining the practice that supports my theory of regulation and exposes the flaws in the conventional account. In each of the cases that I discuss, states decided unilaterally to conduct or endorse an operation that was of dubious legality under the general standards. These states did not try to hide their behavior, to defend it by advancing expansive interpretations of the standards, or to sideline the Security Council. Instead, they worked to engage the Council and obtain its backing. To the extent that they succeeded, and the Council condoned their operation, claims about the operation's illegality lost all or most of their traction.

These cases are meant to display the versatility of the informal regulation. The Council gets involved at different stages of the decisionmaking process, addresses diverse security problems, exercises varying degrees of policy control over the situations, and uses assorted procedural devices to communicate its views. Because of this variation, the cases raise interesting follow-up questions about when, why, and how frequently states use such regulation, and the conditions under which it is most effective. My goal is not so much to answer those questions as to argue that they ought to be asked—to show that this form of regulation is and ought to be assessed as part of the jus ad bellum.

Because the conventional account does not recognize as much, it fails to guide those who want to understand, describe, or explain the jus ad bellum in these cases. Worse, it distorts their reasoning. Methods of analysis that are appropriate for regulation through the general standards are unsuitable when the informal regulation is in play. Those who assume the conventional account and use those methods routinely misrepresent what happened in the cases at hand and misdiagnose what these cases portends for the jus ad bellum going forward.

A. Determination of Factual Predicate

I start with what should be a relatively uncontroversial scenario. The Security Council sometimes establishes a factual predicate for the application of a general standard. It thus resolves what would otherwise be an open question under the standard and helps states justify the operation in law. As examples, I use the recent interventions in Mali and Yemen that were conducted with the consent of local authorities.

1. The unresolved legal question

By almost all accounts, a state may use force in another state with that latter state's consent.Footnote 87 Such force is interpreted to be compatible with Article 2(4), not “against the territorial integrity or political independence of any state, or … inconsistent with the Purposes of the United Nations.”Footnote 88 However, in order for an intervention to be justifiable under the consent standard, the person who agrees to it must actually speak for the territorial state. If she does not, the intervention would almost certainly intrude on the state's “territorial integrity or political independence” or contravene the UN's stated purpose of advancing peoples’ self-determination. Thus, the predicate for applying this standard is that consent be given by a rightful representative of the territorial state.

International law does not resolve, in the abstract, who represents a state during an internal armed conflict. In particular, it is not clear at what point, if any, an incumbent leader who has lost considerable authority or control in the territorial state no longer represents it for purposes of consenting to an outside intervention.Footnote 89 The state practice and opinio juris on this question are all over the map.Footnote 90 States do not consistently condemn interventions that are conducted in internal conflicts with a beleaguered leader's consent, but neither do they routinely show that they view such interventions to be lawful. Instead, their reactions tend to be highly contingent and to depend, at least in part, on their perceptions of who within the state best or most legitimately represents it. As a result, states that rely on the consent standard in these circumstances cannot be confident that their interventions will be treated or accepted as lawful.

2. The Security Council's role

In both Mali and Yemen, the Security Council resolved that question for the intervening states. The Council identified the rightful leader of each state—the person who was in a position to consent to an outside intervention on its behalf—and thus established the predicate for justifying the intervention in law.

a. French intervention in Mali

The situation in Mali reached a boiling point in early 2012, when rebel groups seized much of the country's north, and a military coup overthrew the president.Footnote 91 With help from the Economic Community of West African States (ECOWAS), Mali's governmental factions accepted a transitional arrangement under which Dioncounda Traoré would serve as interim president.Footnote 92 But the fighting in Bamako continued, creating a vacuum of authority that allowed armed jihadi groups to consolidate power in the north.Footnote 93 In December 2012, the Security Council authorized what it called an African-led force to help stabilize the transitional government in Mali.Footnote 94 That force had not yet deployed when, in January 2013, the jihadi groups began moving quickly to the south. France at that point intervened to halt their advance.

France initially hinted at multiple justifications for its intervention,Footnote 95 but it ultimately settled on the “request for assistance from the Interim President of the Republic of Mali, Mr. Dioncounda Traoré.”Footnote 96 It presumably invoked Traoré’s invitation because it thought the consent standard gave it the most plausible or palatable basis for using force. While France could have tried to justify its operation under one of the other standards, any such justification would have been strained; it would have meant advancing a controversial position on the law.Footnote 97 But recall that even the consent standard was not a slam dunk. The standard is unclear on whether it permits force in this kind of case.

That question was never really at issue in the Mali incident because the Security Council had already answered it by the time that France intervened. The Council had indicated—in multiple press releases, presidential statements, and resolutions—that Traoré was Mali's rightful leader. During the early stages of the crisis, the Council repeatedly “condemn[ed] the forcible seizure of power from the democratically-elected government of Mali”Footnote 98 and “[e]xpress[ed] its support to the Transitional authorities.”Footnote 99 As Traoré’s grip on power became more attenuated, the Council backed him by name.Footnote 100 It also made clear that the various rebel groups in Mali did not have a legitimate claim to rule.Footnote 101 The Council thus decided that Traoré was Mali's rightful representative and, by extension, that he was in a position to consent to an outside intervention on its behalf. The Council's decisions helped ground France's intervention in the consent standard. And the global reaction was overwhelmingly positive. Some states gave France military or logistical support, while many others publicly endorsed it.Footnote 102

Two aspects of the incident are particularly noteworthy. First, although France decided on its own to use force in Mali, it actively worked to engage the Council. The day before it intervened in Mali, it obtained from the Council a press statement that reiterated the Council's “call to Member States to … provide assistance to the Malian Defence and Security Forces in order to reduce the threat posed by terrorist organizations and associated groups.”Footnote 103 Then, France justified its intervention in terms that the Council itself had endorsed—helping Mali's transitional government—rather than by invoking a more controversial alternative, like the right to use force in anticipatory self-defense against a transnational terrorist group.Footnote 104 A few days after intervening, France reportedly asked for an emergency meeting and received the Council's full support.Footnote 105 Months later, it secured language in a Council resolution that “[w]elcom[ed] the swift action by the French forces, at the request of the transitional authorities of Mali,” and authorized France to continue its operation.Footnote 106 Throughout, France acted independently but in close coordination with the Security Council.

Second, the Council's decisions on Mali were highly fact-specific. The Council was plainly focused on and concerned about the deteriorating security situation in Mali, so its decisions addressed the dynamics in that country. It did not give any indication that it was applying or defining a more general standard on when an incumbent leader represents a state or may invite an outside intervention during an internal conflict.

b. Regional intervention in Yemen

The Yemen incident followed a similar script. In 2011, Yemen's longtime leader agreed to relinquish power through a transitional arrangement that the Gulf Cooperation Council (GCC) had helped negotiate.Footnote 107 Abdo Rabbo Mansour Hadi was elected president but almost immediately confronted violent resistance from the country's Houthi sect. By the end of 2014, Houthi militants controlled Sanaa and much of northern Yemen. In March 2015, with Hadi's government on the verge of collapse, a Saudi-led regional force that included five of the six GCC countries intervened to support him.

Hadi's consent was the only plausible basis for justifying the intervention under the jus ad bellum’s general standards.Footnote 108 But as in Mali, the application of the consent standard would have been tenuous, if the Security Council had not already established its predicate fact. Starting in 2011, the Council endorsed the GCC-led process for the transition of power in Yemen.Footnote 109 And once Hadi was elected, it put its weight fully behind him.Footnote 110 In a series of resolutions,Footnote 111 presidential statements,Footnote 112 and press releases,Footnote 113 the Council consistently supported the GCC-led process and President Hadi, while opposing Houthi and other opposition groups in Yemen.

A few days before the intervention began, Hadi's government and the GCC sought the Security Council's support.Footnote 114 The Council seemed unpersuaded that an outside intervention would be productive,Footnote 115 but having already decided that Hadi was Yemen's rightful leader, it accepted that he had authority to consent to an intervention.Footnote 116 The Council issued a presidential statement reiterating that it viewed Hadi as Yemen's legitimate leader and that it “condemn[ed] the ongoing unilateral actions taken by the Houthis, which undermine the political transition process in Yemen.”Footnote 117 Later, the Council noted in a resolution that the intervention had been conducted at Hadi's request.Footnote 118 These decisions at the Council helped ground the intervention in the consent standard. Indeed, many states endorsed the operation, and very few condemned it.Footnote 119

As with the Mali incident, the jus ad bellum’s content in Yemen was not established entirely through the general standards. The most relevant standard—on consent—is ill-defined for such cases. Its content was also established through processes at the Security Council for condoning specific operations that would otherwise have been legally suspect. By deciding that the factual predicate for the consent standard had been met, the Council made these interventions easier to justify and harder to challenge in law; it conferred authority on them.

3. Analytic pitfalls of the conventional account

The conventional account is inadequate to describe or explain the law's operation in these cases. Further, it warps how analysts examine the cases. Most scholars who have commented on the two interventions recognize that they are difficult to justify under the general standards but conclude from the positive reactions that they must nevertheless have been lawful.Footnote 120 These commentators then look for ways to fit the interventions into the general standards. Their reasoning reflects the conventional account and is unsound.

Those who focus on the consent standard appreciate that this standard has historically been indeterminate for such cases but consider it to be the only plausible legal basis for justifying the interventions. Thus, they assume that the interventions reflect the standard and can be used to clarify its content. In other words, they generalize up from the interventions to define a more broadly applicable standard on when states may intervene in an internal conflict with a beleaguered leader's consent.

For example, Erika de Wet says that the incidents show that states may intervene in an internal conflict so long as they have the incumbent leader's consent.Footnote 121 Karine Bannelier and Theodor Christakis cite the Mali incident for a more restrictive proposition; they contend that states may intervene to help a recognized government fight terrorist groups but not to help it suppress other kinds of opposition movements.Footnote 122 Benjamin Nußberger asserts that any distinction between terrorist and other kinds of groups does not hold up in Yemen. He then offers a different formulation: if “an established transitional process exists, and the request for military assistance aims merely to protect this process,” then the transitional leader may consent to an intervention “by a broad international coalition.”Footnote 123 These scholars disagree about the content of the consent standard—about how and at what level of generality it ought to be articulated—but they all assume that that is the relevant inquiry.

Their approach is flawed in two respects. First, it diminishes the role that the Security Council played in each incident. It treats the Council's decisions as just evidence that the interventions satisfied the consent standard.Footnote 124 In fact, the Council's decisions themselves helped the interventions satisfy the standard. By establishing the factual predicates for the standard's application, the Council contributed to, and did not merely sign off on, the interventions’ lawfulness.Footnote 125 Second, generalizing up from the incidents to define the consent standard exaggerates their future relevance. The Council's decisions on Mali and Yemen were extremely particularistic—focused on the specific circumstances in each country at the time. There is little reason to believe that states will react similarly to any future intervention in an internal conflict at the behest of a beleaguered leader, absent a comparable Council determination that he represents the territorial state. Again, such interventions might still occur, and they might not be vociferously condemned, but neither are they likely to be as widely endorsed and accepted as lawful.

At least one commentator, Ian Johnstone, recognizes that the Security Council was instrumental in making France's intervention in Mali lawful. But he interprets this to mean that the intervention fits within the standard on Council authorizations.Footnote 126 Johnstone's logic is, again, reflective of the conventional account. He acknowledges that, when France intervened in Mali, the Council had expressly authorized only an African-led force. He insists, however, that the best explanation for the positive reaction to the intervention is that the Council implicitly authorized it. He reasons that the December 2012 resolution was ambiguous about whether it permitted states to act independently of the African-led force and that states resolved that ambiguity in France's favor when they supported the intervention; at that point, he says, states clarified that the Council had, in fact, authorized France's action, albeit only implicitly.Footnote 127

This logic obscures more than it illuminates. Johnstone attributes the supposed ambiguity in the Council's resolution to the political dynamics on the Council. He does not explain why the Council was ambiguous about authorizing France but not about authorizing an African-led force, or why the politics that produced any ambiguity did not also produce ambivalence when France intervened. Further, he fails to account for key facts that seem to contradict his story. Why would both the Council and France ground the intervention in Traoré’s invitation, if the Council had authorized it? Why would the Council later grant France authority to act independently in Mali if its intervention fell within the scope of the authorization for the African-led force?Footnote 128 The claim of implicit authorization is insufficiently substantiated. It also is misleading. It suggests that the Council actually authorized what France did, when in fact France acted unilaterally but in close coordination with the Council.Footnote 129 The locus of decisionmaking is meaningful, so it ought to be highlighted and examined, not just papered over.

B. Midstream Support

In other incidents, the Council is even further removed from the decision to use force. It for the first time addresses a security problem when a unilateral and apparently unlawful operation is already underway. The Council's support still works to confer authority on the operation. I discuss below the well-known example of ECOWAS’s 1990–1992 intervention in Liberia. The example is particularly illustrative because many analysts have struggled to make sense of it under the conventional account.

1. The Security Council's role

The Liberian civil war was brutal. By the middle of 1990, rebel groups had overthrown President Samuel Doe, captured almost the entire country, and caused significant human casualties. Doe approached both the Security CouncilFootnote 130 and ECOWASFootnote 131 for assistance. The Council did not take any action, but ECOWAS did. It sent what it described as a peacekeeping force to Liberia.Footnote 132 This force did not have and suggested that it did not need the consent of all of the parties to the conflict.Footnote 133 Soon after arriving, it was attacked by one of the rebel groups and became a major participant in the fighting.Footnote 134 At the same time, ECOWAS tried to negotiate a political solution to the conflict. In November 1990, the key actors in Liberia finally concluded what would be the first of multiple cease-fire arrangements.Footnote 135

Early on, ECOWAS informed the Security Council of its intervention, without seeking authorization or otherwise trying to justify its conduct under the general standards.Footnote 136 It simply noted that it aimed “to stop the senseless killing of innocent civilian nationals and foreigners, and to help the Liberian people to restore their democratic institutions.”Footnote 137 At the time, a few actors expressed tepid support for ECOWAS,Footnote 138 but its action was still evidently lacking in authority. The Liberian representative to the UN himself conceded that

[i]n 1990, at the height of the Liberian civil conflict, international opinion on Liberia was divided between the imperatives for humanitarian intervention, on the one hand, and the value of reaffirming classical conceptions of sovereignty, however anachronistic, on the other.Footnote 139

As the conflict dragged on, ECOWAS actively elicited the Security Council's approval. In January 1991, Nigeria spoke on ECOWAS’s behalf to say that, now that the parties to the conflict had agreed to a cease-fire, “[i]t is important that the Security Council urge them to continue to respect that cease-fire.”Footnote 140 The Council responded with a presidential statement that “commend[ed] the efforts made by the ECOWAS Heads of State and Government to promote peace and normalcy in Liberia” and “call[ed] upon the parties to the conflict in Liberia to continue to respect the cease-fire agreement which they have signed and to co-operate fully with the ECOWAS to restore peace and normalcy in Liberia.”Footnote 141 The Council issued a similar statement in May 1992.Footnote 142

In November 1992, with the situation still deteriorating, ECOWAS pressed harder for the Council's support. It asked the Council to hold an emergency meeting on Liberia.Footnote 143 There, one ECOWAS member state after another emphasized the need for the Council to bolster ECOWAS’s decisions.Footnote 144 The Council responded with Resolution 788, which commended ECOWAS, supported the most recent peace agreement that ECOWAS had helped negotiate, and imposed an arms embargo on the conflicting parties in Liberia.Footnote 145 With time, the Council became more involved in the situation. Between 1993 and 1997, it adopted sixteen resolutions on Liberia. These resolutions repeatedly expressed support for ECOWAS’s actions in Liberia, without using language to authorize force or otherwise fit it into the general standards.Footnote 146 Nevertheless, questions about the legality of the intervention largely dissipated.Footnote 147 Once the Council backed ECOWAS, the claim that its intervention was unlawful became less salient. The Council conferred authority on an operation that had been legally suspect.

2. Analytic pitfalls of the conventional account

Here again, the conventional account is dissociated from state practice and inadequate to describe, explain, or analyze what happened. As commentators have broadly recognized, ECOWAS’s intervention in Liberia did not fit neatly into any of the general standards. However, because it eventually attained such widespread support, most reason that it must somehow have been lawful. They then try to fit it into those standards.

A few scholars cite President Doe's request for assistance as evidence that the intervention satisfied the consent standard. Scholars who make this claim recognize that there are serious doctrinal and policy problems with applying the consent standard in this case.Footnote 148 When ECOWAS intervened in Liberia, Doe's regime had lost control over most of the country, no longer performed basic governmental functions, and had little internal legitimacy.Footnote 149 Nevertheless, David Wippman infers that “the Council's post hoc approval suggests that the Council considered the ECOWAS intervention to be a consent-based peacekeeping operation.”Footnote 150 Similarly, Georg Nolte says that “the Security Council necessarily implied that, in its opinion, the intervention by [ECOWAS] did not require the Council's authorization. The most plausible rationale for this assessment is the assumption that [it] received a valid invitation into the country.”Footnote 151 These inferences are unfounded. Neither the Council nor ECOWAS invoked Doe's invitation as a basis for intervening.Footnote 152 Moreover, as with Mali and Yemen, explaining the Liberia intervention in consensual terms discounts what happened at the Security Council. The Council here contributed to, and did not just accept, the intervention's lawfulness.

Others recognize that the Council helped legitimize ECOWAS’s action but have difficulty explaining that dynamic in law.Footnote 153 For example, Christian Walter reasons that, in adopting Resolution 788, the Council must have retroactively and implicitly authorized the intervention.Footnote 154 Many analysts, including Walter himself, are skeptical of claims of retroactive authorization.Footnote 155 Such claims are misleading because they suggest that a decision to use force was made through the Council, when it actually was made elsewhere and then endorsed at the Council. Again, that distinction is significant, so it ought to be recognized and addressed head on. The claim of retroactive authorization is also unconvincing in this case. The Council had ample opportunity to authorize ECOWAS’s intervention but never did. If anything, it seemed disinterested in authorizing the intervention, at least during the early stages of the conflict.Footnote 156

Finally, some scholars cite the Liberia incident as evidence that the jus ad bellum now has or is developing a standard that permits unilateral humanitarian interventions.Footnote 157 As discussed, states have periodically condoned unilateral operations that, like the one in Liberia, can be labeled humanitarian. But neither in Liberia nor elsewhere has the full group of states endorsed a general standard to that effect. They have instead suggested that they reject such a standard.Footnote 158 Thus, even if states’ decisions on Liberia were ultimately motivated by humanitarian considerations, they were also contingent on the facts. The claim that the incident evinces a general standard on humanitarian interventions overstates its doctrinal significance and misinforms those who might rely on it in future cases. It would be wrong to assume that a comparable future operation would be as widely accepted as lawful, without similar support from the Council.Footnote 159

C. Majoritarian Support

At the outer bounds of the jus ad bellum’s informal regulation, states use the Security Council to appraise and, on the whole, endorse an operation, but the Council, as an institution, does not take a position on it. A display of majoritarian support at the Council appears to confer some authority on the operation. It shows that states have congregated at the institution with legal primacy on the use of force, actively considered an operation, and affirmatively defended or endorsed it. But because the Council's institutional authority is more than the sum of its parts, the hurdle to curing a legal defect under the general standards might be higher in these circumstances than when the Council itself backs an operation. The 2017 U.S. operation against Syria is an example.

1. The Security Council's role

Recall that the United States asserted that its operation was designed “to prevent and deter the spread and use of deadly chemical weapons.”Footnote 160 The day after the United States acted—on April 7—the Security Council held an emergency meeting to address the incident. It did so, even though it had already met on the previous two days to discuss the recent chemical attack in Syria. During those earlier meetings, states vociferously opposed the use of chemical weapons and emphasized that something had to be done, but they did not agree on what to do.Footnote 161 The Council certainly had not authorized the use of force in Syria.

The April 7 meeting was pitched as a referendum on the U.S. operation.Footnote 162 From the beginning, most of the people in the room made clear that they were more concerned about the repeat use of chemical weapons in Syria than they were about the U.S. operation.Footnote 163 Eleven states at the Council affirmatively endorsed the operation.Footnote 164 Another group of states, including China, expressed ambivalence or were silent about it.Footnote 165 Only four states—Bolivia, Iran, Russia, and Syria—claimed that the United States acted unlawfully.Footnote 166 Here, states used the meeting at the Council to communicate their views about the unilateral operation, and most of them publicly defended their decisions not to apply the Article 2(4) prohibition.

2. Analytic pitfalls of the conventional account

As discussed, the conventional account of the jus ad bellum has little descriptive or explanatory purchase in this case. The U.S. operation was not consensual, authorized, or defensive. It might be described as humanitarian, but again, very few states endorse a general standard licensing unilateral force for humanitarian ends. Here, as elsewhere, states made clear that their decisions to conduct, support, or tolerate the operation were contingent on the facts, not intended to reflect or establish a general standard that would automatically license similar actions in the future. Further, the U.S. operation looked less like a typical humanitarian intervention than like a reprisal,Footnote 167 which by almost all accounts would be unlawful.Footnote 168 It aimed to deter only one form of atrocity—the use of chemical weapons—not to avert the ongoing humanitarian crisis in Syria.

Because the jus ad bellum’s general standards appeared to be entirely inoperative, the conventional account is an especially vacuous analytic framework for this case. Most commentators just asserted that the United States violated the jus ad bellum.Footnote 169 These commentators of course recognize that states broadly supported the U.S. operation. However, once one accepts the conventional account and concedes that an operation is not justifiable under the general standards, the only option is to claim that it was unlawful. That claim might ultimately be correct, but it ought to be defended with more than conclusory assertions of illegality. It would mean that the vast majority of states decided to disregard the law, that many then defended their insubordination at the very institution that is legally charged with governing this conduct, and that nothing happened in response. Those who make this claim have not offered an account of why states bothered to go to the Council and endorse the U.S. operation, if they were just ignoring the jus ad bellum, or why the best interpretation of what they were doing at the Council—when they were explaining their decisions not to apply Article 2(4)—is practicing not law but a kind of politics that is antithetical to law.

To be clear, I am not arguing that the U.S. action in Syria was lawful. I am arguing that assessments that rest on the conventional account ignore an important facet of the practice and are, for this reason, analytically deficient. Using my theory, a key question for assessing the lawfulness of the U.S. operation is whether the support at the Council counterbalanced the perceptions of illegality that come with deviating from the general standards. It might not have. For example, perhaps the United States strayed too far from the standards for this support to cure the legal defect. Perhaps majoritarian support, without an institutional decision by the Council, never suffices. Or perhaps such support pushed the U.S. action somewhere along a spectrum of legality, such that claims of illegality would still circulate but have considerably less traction than they otherwise would. The point is that any legal analysis ought to account for what happened at the Council, rather than just assume that this practice is external and irrelevant to the jus ad bellum.

IV. Policy Implications

I have argued so far that the jus ad bellum’s content is sometimes established through case-specific processes at the Security Council that do not reflect or inform the general standards. This insight is valuable to those who seek to understand or engage with the jus ad bellum. Depending on the circumstances, it can help them better explain specific cases, assess the cases’ future relevance, and craft more effective legal strategies.

Some readers might nevertheless be skeptical of the informal regulation for policy reasons. They might worry that this form of regulation unduly loosens the restrictions on unilateral force and diminishes the Security Council's supervisory role. So, even if they accept that it is part of the jus ad bellum, they might want to avoid recognizing or encouraging it, for fear of inducing states to push the law's boundaries and more often act in legally suspect ways. That worry is understandable and helps explains the almost dogmatic adherence to the conventional account. When the informal regulation is in play, decisions to use force are unilaterally made. The Council is more removed from these decisions than it would be if it were expected to authorize the operations.

Yet whether such regulation ultimately invites more unilateralism and detracts from the Council depends on how and with what effect the jus ad bellum would regulate states in its absence. I argue below that relying entirely on the general standards has real drawbacks and that these drawbacks have become more pronounced over time. Today, the best way to strengthen the jus ad bellum and preserve the Council's primacy is probably to rely less on the general standards and more on the informal regulation.

A. The Jus ad Bellum's Constraining and Legitimizing Effects

As an initial matter, it is worth addressing the common refrain that any deviation from the general standards evinces the jus ad bellum’s inefficacy or irrelevance, and risks its further deterioration. The reality is that states, as a group, sometimes undertake, support, and tolerate forcible operations that do not fit neatly into the general standards. In these cases, the standards do not deter legally dubious conduct. Even so, claims of irrelevance and erosion are overdrawn and need to be circumscribed.

1. The irrelevance claim

Although the modern jus ad bellum is often described an instrument of peace and a constraint on cross-border force,Footnote 170 it has always also facilitated some force.Footnote 171 When it licenses force, it helps legitimize that conduct in law. In these circumstances, the law makes the use of force easier to execute and more difficult to challenge than it otherwise would be.Footnote 172 This means that the jus ad bellum can be relevant and effective, even when it does not inhibit states from taking or supporting legally dubious operations. It can still legitimize their conduct.

That dynamic was evident in each of the cases that I discussed. In each, states sought the Council's support for conduct that would otherwise be difficult to justify in law. The fact that they did suggests that they see value in having or appearing to have the jus ad bellum on their side—that they would rather act with than without the legitimacy that it confers—and that going to the Council was one way to obtain that legitimacy. Indeed, in all but the Syria case, states tried to harness the Council's support both when they initially decided to use force and as their operations progressed. They acted as if having the Council's approval mattered and could affect the contours or efficacy of their ongoing operations.

2. The risk of erosion

A more difficult question is whether the informal regulation risks becoming a loophole that, over time, swallows the Article 2(4) prohibition and loosens the jus ad bellum’s constraints. That question is ultimately an empirical one that goes beyond the scope of this article. It certainly is possible that the more states use the informal regulation, the more they signal that it is a viable option, and the more they galvanize others to do the same—creating a default in which states act unilaterally and then seek the Council's support, rather than channel use of force decisions through the Council and wait for its authorization.Footnote 173

Of course, any such risk would be tempered by what states themselves are willing to do and tolerate, both independently and at the Council. States might have reasons for not overusing the informal regulation. For example, a state that acts first and then seeks the Council's backing cannot be sure of what the Council will do or, therefore, whether its operation will be accepted as lawful. States might have varying degrees of tolerance for that uncertainty or might tolerate it more in certain contexts than in others. Further, using the informal regulation is unlikely to be an effective strategy for curing every kind of legal defect. For instance, it is hard to imagine the Council condoning, rather than at least trying to condemn, a forcible annexation of foreign territory.Footnote 174

Yet even if the risk of deterioration is serious, using some of the informal regulation is probably, on balance, preferable to trying to regulate states entirely through the general standards. This is so for three distinct but related reasons. I outline them briefly here and expand upon them in the following sections. First, although the informal regulation detracts from the Security Council's primacy in some respects, it reinforces that primacy in others. When states use the informal regulation, they deliberately choose to involve the Council in governance decisions. Second, insofar as states broadly support operations that cannot easily be justified under the general standards, relying entirely on those standards to regulate their conduct presents its own slippery slope problem. It helps erode the constraints under the general standards. Third, since the turn of the century, those constraints have become noticeably weaker. States now routinely conduct and tolerate unilateral operations that were once widely assumed to be unlawful. The jus ad bellum would have more regulatory bite if it legitimized some such operations and left others in a legal gray zone than if it tried in vain to constrain all of them through proscriptions of general application.

B. Preserving the Security Council's Primacy

Once states conduct and endorse operations that lack good justification under the general standards, the informal regulation serves to bolster, rather than to diminish, the Security Council's position of authority in the jus ad bellum. This is especially so if the states that use force themselves take the issue to the Council. By trying to involve the Council and obtain its support, these states signal that they are committed to it as an institution and that they accept its primacy, even though they do not always give it as prominent a role as they might. The informal regulation both reflects and reinforces expectations about the Council's preeminence. Indeed, it affirms the Council's standing, even where, as in the Syria case, the Council declines to take an institutional position on the operation. It still serves to engage the Council on the use of force decision.

If the Council's role here seems meager, consider the alternatives. In each of the incidents that I discussed, the most likely alternative to using the informal regulation was not for the acting states to wait for the Council's authorization, to be widely condemned for acting unilaterally, or to forego the operation. The most likely alternative was for states to act entirely outside the Council, without meaningful repercussion. That option would have marginalized the Council even more than the informal regulation did.

The informal regulation also bolsters the Council in at least two ways that go beyond whatever value there is to it opining on particular use of force decisions. First, when states go to the Council, they focus its attention on the situation at hand; they invite it not just to rubberstamp a unilateral decision but to shape the governing policy for that situation. In most of the cases that I discussed, the Council acted on that invitation.Footnote 175 It condoned a unilateral operation as part of a more comprehensive and collective scheme for addressing the underlying security problems in the territorial state. For good or for ill, the Council took ownership of the situation.

Second, even if the informal regulation does not have any operational effect, using it is a way for states to convene at the Council and discuss what happened, rather than just go about their business in the face of an apparent deviation. The Council thus continues to be the principal forum for airing the considerations that are or ought to be at stake in the use of military force, and for subjecting specific operations to external scrutiny. I have argued in other work that fostering those kinds of interactions can help keep states invested in a governance project—here, the one on collective security—even when they disagree about its policy content.Footnote 176 When states congregate at the Council to try to justify and argue about particular operations, they at least reinforce the sense that use of force decisions are matters of collective concern and for the Council's deliberation, not within the exclusive purview of individual states.Footnote 177

C. Limiting an Incident's Precedential Reach

The informal regulation also preserves the Security Council's primacy in another respect. It helps limit the precedential reach of “outlier” incidents. In particular, it allows states to legitimize discrete operations on their facts, without advancing overly permissive interpretations of the general standards. Again, if states sometimes choose to conduct and support such operations, then confining them as much as possible to their facts helps reinforce the Council's preeminence.

To be clear, any forcible incident is a precedent and can shape expectations about the jus ad bellum’s content or salience going forward. When states broadly support an operation that strays from the general standards, they indicate that the standards have changed or are not always operative. The questions then become to what extent and in what circumstances may states rely on the precedent to justify future operations. The conventional account does not offer a sensible answer. Generalizing down from the standards to insist that the precedential action was unlawful, despite the widespread support for it, either is unpersuasive or suggests that the law is irrelevant and need not be followed. Yet generalizing up from the incident means entrenching it in a more broadly applicable standard, such that it automatically extends to other cases. The effect is to empower those who want to use the precedent to justify acting without the Security Council's authorization in the future.

The informal regulation is more cautious and constrained because it is so particularistic. When the Council crafts a policy for one case, without offering a generalizable rationale for its decision, it complicates efforts to use the incident as a precedent.Footnote 178 Of course, states might still invoke it to try to justify a future operation. But precisely because the Council's decision is fact-specific, any such claim would have to be defended on the facts, rather than by rote invocation of a general standard. And unlike when a general standard applies, they cannot be confident that their position will prevail. Indeed, if the Council's backing is ultimately what pushed the precedential action from dubious to lawful, the copycat action is unlikely to elicit the same reaction, without comparable support from the Council.

Perversely, using the conventional account to analyze cases that involve the informal regulation can itself contribute to the law's erosion. Take the efforts to interpret into the consent standard the interventions in Mali and Yemen.Footnote 179 These efforts diminish the role that the Security Council played in each incident and push that standard in a direction that states themselves did not—toward licensing cross-border force in at least some internal conflicts. The result is to bolster the claim that such interventions are lawful, no matter what the Council does. By contrast, highlighting that the informal regulation was in effect underscores that the interventions in Mali and Yemen were lawful at least in part because of the Council's actions. It suggests that future interventions should also be coordinated with the Council.

Or take the claim that the cases in which states have conducted and condoned unauthorized humanitarian interventions reveal a general standard to this effect. Any such standard would extend the precedential reach of those cases beyond what states intended. It would facilitate future interventions by entrenching a general standard that makes some unauthorized humanitarian interventions automatically justifiable. The particularistic nature of the informal regulation is more limiting. The fact that states treat some such interventions as lawful does not mean that they will do the same for others. Future interventions ought to be defended on their own terms and at the Council. Significantly, the informal regulation can restrict an incident's precedential reach, even where, as perhaps in Syria, it does not fully cure the legal defect. It still allows states collectively to underscore that, despite their decision not to apply the Article 2(4) prohibition in one case, they are not interpreting or creating a standard of general applicability. In other words, they preserve the prerogative not to afford the same treatment to any subsequent intervention.

D. Curbing Deregulation

Finally, the informal regulation has the potential to curb a deregulatory trend in the jus ad bellum that is fully underway. As is well known, several states now advance very expansive claims on self-defense. These claims lack good limiting principles, so permitting defensive force in all of the circumstances that they seem to envision would drastically shift decisionmaking authority from the Security Council to individual states. To be clear, the number of states that make such claims is still relatively small. Yet these states are adamant about and routinely act on their positions. Most other states have neither expressly endorsed nor meaningfully resisted them. Thus, the dynamic is one in which a handful of militarily active states are claiming expansive authority for themselves and are regularly acting on that claim, without affirmative support from most other states but also without noticeable repercussion.

Jacob Katz Cogan and I have argued that that dynamic puts unique pressure on the jus ad bellum.Footnote 180 The problem is not so much that the law is unsettled or being violated. The jus ad bellum has long been fluid and contentious but also quite resilient. The problem is that a small group of militarily powerful states appear to be radically remaking the law in their favor by circumventing the available processes for collective legitimization. The informal regulation is an apt corrective because it would reinvigorate those very processes. States can get together and condone specific operations, without formalizing a general standard that would license many other operations. In fact, states have already used the informal regulation in this way in the context of the campaign against the so-called Islamic State in Iraq and Syria (ISIS).Footnote 181 The incident thus provides a template for them to use more of it in the future.

1. The Security Council's legitimization of defensive force

ISIS is a transnational terrorist group that emerged in Syria in 2013, in the midst of the civil war there.Footnote 182 By the summer of 2014, ISIS had crossed into Iraq and controlled large portions of Iraqi and Syrian territory.Footnote 183 Within months, dozens of states were participating in or assisting a U.S.-led military operation against ISIS.Footnote 184 All of these states presumably supported the actions in Iraq and Syria, since combating ISIS in both countries was critical to the campaign's success. But during the early stages of the campaign, several states that were willing to participate in military operations in Iraq, with the Iraqi government's consent, declined to act in Syria.Footnote 185 These states seemed unsure of the proper legal basis for using force in Syria. As the deputy prime minister of the Netherlands explained, “[f]or military operations in Syria, there is currently no international agreement on an internationally legal mandate.”Footnote 186

A few states that were acting in Syria invoked the “unable or unwilling” standard on defensive force against nonstate actors.Footnote 187 Under this standard, a state may use force against nonstate actors in another state, if that other state is unable or unwilling to contain their threat. The standard has been in circulation for decades,Footnote 188 and states increasingly invoke it to justify their counterterrorism operations, but it is potentially very expansive and remains controversial.Footnote 189 Thus, not all states that were participating in the defensive operation in Syria relied on the unable or unwilling standard. Some simply asserted that their conduct was lawful, without articulating a generalizable standard on defensive force against nonstate actors.Footnote 190 Meanwhile, the vast majority of states had not expressed a position one way or another on the operation's legality. So, as of late 2015, many states plainly supported that operation but were unsure how to justify it in law and were unprepared to endorse the unable or unwilling standard.

The legal terrain shifted in November 2015, when the Security Council adopted Resolution 2249. By its terms, this resolution is highly particularistic. It does not articulate or purport to apply any general standard on self-defense; it does not even mention Article 51 or the right to use defensive force. It simply “determin[es]” that ISIS “constitutes a global and unprecedented threat to international peace and security” and “[n]ot[es]” that, according to Iraq, ISIS had “established a safe haven outside Iraq's borders that is a direct threat to the security of the Iraqi people and territory.”Footnote 191 The resolution then “[c]all[s] upon Member States that have the capacity to do so to take all necessary measures, in compliance with international law, in particular with the United Nations Charter, … to prevent and suppress terrorist acts. …”Footnote 192 The resolution is an example of the jus ad bellum’s informal regulation.

As Dapo Akande and Marko Milanović have explained, Resolution 2249 “suggest[ed that] there is Security Council support for the use of force against IS[IS]” and was “clearly designed to provide legitimacy for the measures being taken, and to be taken, against [ISIS].”Footnote 193 Yet Akande and Milanović insist that 2249 “neither adds to, nor subtracts from, whatever existing authority states already have.”Footnote 194 Their position seems to be that 2249 is politically but not legally relevant. That position needs to be defended with more than rote reference to the general standards. In adopting 2249, the Council—the institution with legal primacy on the use of force—communicated that it did not consider the Article 2(4) prohibition to be controlling in this case. The resolution is a decision on the law, even if it does not reflect or inform the general standards. Its apparent purpose and effect were to diminish the claims and concerns about the operation's illegality.

Perhaps the best evidence that it worked is that, after the Council adopted it, a handful of states invoked it, in combination with Article 51, to justify using force in Syria.Footnote 195 Jacob Katz Cogan and I have noted that these states cited 2249 for the propositions that ISIS: (1) posed an unprecedented threat to international peace and security, and (2) was operating from one state to attack another.Footnote 196 In other words, they used the resolution to establish predicate facts for justifying their defensive force—much as the states that intervened in Mali and Yemen used the informal regulation to justify those operations.

2. Effects and prospects

The incident shows how the informal regulation could help preserve the jus ad bellum’s relevance in situations involving claims of self-defense. As in the other incidents that I have examined, the effect of using this form of regulation was to reinforce the Security Council's position of authority in the regime—not relative to what it might ideally have been but relative to what it actually would have been. The alternative here was not for states to seek and obtain the Council's authorization to use force in Syria. The alternative was for them broadly to conduct, support, and tolerate a major military operation that was of questionable legality, without any Council involvement at all. Against that backdrop, using the informal regulation was a way for states to show that they, as a group, still see a role for the Council in the jus ad bellum and value its processes of collective deliberation and legitimization.

Resolution 2249 also helped constrain the incident's precedential reach. The campaign in Syria would have been an important precedent, even without 2249, because so many states had already supported it. For those who might try to use the incident to justify a future action, 2249's added value is limited. The resolution focuses on the ISIS situation and does not articulate or hint at a standard of general applicability. At most, its adoption during the operation supports the claim that defensive force against nonstate actors is in at least some circumstances lawful. Meanwhile, 2249 helped check the most radical implications of the unable or unwilling standard. It gave states a way to resist that standard while still participating in or supporting the operation in Syria. The states that invoked 2249 indicated that they considered the defensive operation to be lawful in part because of what happened at the Council.

The incident is instructive for three reasons. First, it suggests that states, as a group, have not yet accepted the unable or unwilling standard. Multiple states declined to invoke that standard to justify an operation against ISIS that they clearly supported. Second, states had varying degrees of tolerance for conducting a legally dubious operation. While some states acted unilaterally in Syria, others did and would not. The latter states appeared to be constrained by the operation's potential illegality. Third, these states were groping for a way to license a defensive operation, without establishing an expansive right to act unilaterally in a broad category of cases. In Syria, their answer was to use the informal regulation. This form of regulation was appealing precisely because it is particularistic. It allowed for case-specific decisionmaking in the face of ongoing uncertainty about how best to formulate a generally applicable standard.

Thus, states that believe that defensive force against nonstate actors is sometimes justifiable but are concerned about the overly expansive implications of the unable or unwilling standard have a tool in their arsenal for countering it. These states might, for example, decline to partake in future operations without the Council's buy-in. Or they might insist that certain operations—perhaps those involving sustained military force, rather than one-off strikes—be channeled through the Council. Of course, there is no guarantee that such efforts would alter the conduct of states that are intent on advancing and acting on highly disagreeable interpretations of Article 51. Still, the informal regulation could help distinguish the defensive operations that are widely accepted as lawful from the ones that remain legally tenuous. It could make some such operations easier to execute or harder to challenge in law. And it would give states more opportunities to engage together on their collective security project. It would, in other words, help revitalize the jus ad bellum.

V. Conclusion

Although the modern jus ad bellum has never worked as the text of the UN Charter suggests, it has been fairly resilient over time. Despite its limits, it remains foundational to the global order, largely because states have found ways to keep it relevant as their sensibilities and geopolitical dynamics have changed. For those who remain committed to this system for collective security, the current regulatory challenge is to find ways to adapt it yet again—and establish new parameters to regulate the security problems that states now view and treat as the most pressing. These problems are not new; they relate mostly to transnational terrorism, the commission of mass atrocities, and the proliferation of weapons of mass destruction. But in today's security climate, they are evidently not amenable to regulation through the general standards. The informal regulation offers a promising alternative. It allows states to balance, on a case-by-case basis, their competing demands for dispatch, flexibility, and collective legitimization. Of course, such regulation can be effective only if states themselves decide to use it, but their past practice suggests that they might. They would almost certainly strengthen the jus ad bellum if they do.

References

1 UN Charter Art. 2, para. 4.

2 Military and Paramilitary Activities in and Against Nicaragua (Nicar. v. U.S.), Judgment, 1986 ICJ Rep. 14, para. 246 (June 27) (recognizing that an outside state's intervention “is already allowable at the request of the government of a State”).

3 UN Charter Art. 39.

4 Id. Art. 51.

5 See, e.g., Military and Paramilitary Activities in and Against Nicaragua, supra note 2, para. 268 (“[T]he use of force could not be the appropriate method to monitor or ensure … respect [for human rights].”); Group of 77, Declaration of the South Summit, para. 54 (Apr. 14, 2000), at http://www.g77.org/summit/Declaration_G77Summit.htm (“We reject the so-called ‘right’ of humanitarian intervention, which has no legal basis in the United Nations Charter or in the general principles of international law.”); Randelzhofer, Albrecht & Dörr, Oliver, Article 2(4), in The Charter of the United Nations: A Commentary 200, 223 (Simma, Bruno, Khan, Daniel-Erasmus, Nolte, Georg & Paulus, Andreas eds., 3d ed. 2012Google Scholar) [hereinafter Charter Commentary] (“[T]here is no room for the concept of [unilateral] humanitarian intervention being deduced from the UN Charter … .”); Gray, Christine, The Use of Force for Humanitarian Purposes, in Research Handbook on International Conflict & Security Law 229, 253 (White, Nigel D. & Henderson, Christian eds., 2013)Google Scholar [hereinafter Research Handbook] (“It is difficult, if not impossible, to make a legal case for the existence of a right of unilateral humanitarian intervention today.”).

6 E.g. U.K. Prime Minister's Office, Chemical Weapon Use by Syrian Regime: UK Government Legal Position (Policy Paper, Aug. 29, 2013), at https://www.gov.uk/government/publications/chemical-weapon-use-by-syrian-regime-uk-government-legal-position/chemical-weapon-use-by-syrian-regime-uk-government-legal-position-html-version; UN SCOR, 54th Sess., 4011th mtg., at 12–13, UN Doc. S/PV.4011 (June 10, 1999) (statement of Netherlands).

7 To say that little attention has been paid is not to say that there has been none. See infra notes 10, 20–23, and accompanying text.

8 Some states have made claims on anticipatory self-defense that are extremely elastic, but these claims have not been widely endorsed, and even they would not swallow the blanket prohibition. As an extreme example, they would not permit forcible annexations of foreign territory. For more on them, see Monica Hakimi & Cogan, Jacob Katz, The Two Codes on the Use of Force, 27 Eur. J. Int'l L. 257, 283–86, 288–90 (2016)Google Scholar.

9 A few international lawyers have suggested that the jus ad bellum is highly contextually variable; their views remain outside mainstream legal thinking. As an example, see Myres S. McDougal & Florentino P. Feliciano, Law and Minimum World Public Order: The Legal Regulation of International Coercion 151–52 (1961).

10 Two recent articles address the jus ad bellum’s regulatory form, and both reflect this conventional account. First, Matthew Waxman has analyzed various debates between those who define the jus ad bellum with relatively “clear and rigid rules that admit little case by case discretion” and those who prefer more “flexible standards that call for weighing contextual factors.” Waxman, Matthew C., Regulating Resort to Force: Form and Substance of the UN Charter Regime, 24 Eur. J. Int'l L. 151, 152 (2013)CrossRefGoogle Scholar. Waxman notes that, although international lawyers disagree about whether the jus ad bellum’s directives operate more like precise rules or like elastic standards, very few claim that they fall at an extreme end of the rule-standard spectrum. Id. at 158–59. I use the word “standard” in this article not in contradistinction to “rule” but to refer to the range of possibilities in the middle of that spectrum. Waxman's study confirms that, despite debates about their pliability, almost everyone accepts that the jus ad bellum consists of a handful of generally applicable substantive standards. See id. at 167. Second, Ashley Deeks has analyzed the multi-part standards that some states and scholars have advanced to justify or appraise certain kinds of cases. Deeks, Ashley, Commentary: Multi-part Tests in the Jus ad Bellum, 53 Hous. L. Rev. 1035 (2016)Google Scholar. For example, Deeks discusses efforts to specify the standard on self-defense for situations in which force is used in anticipation of an armed attack, in response to a cyberattack, against nonstate actors, or to protect a state's nationals. Id. at 1050–57. Compared to the generic standard on self-defense, these sub-standards would govern narrower categories of cases. Still, they would define the jus ad bellum by reference to certain, generally applicable substantive criteria. So, they would increase the number of general standards that comprise the jus ad bellum but would not fundamentally change that regulatory form. I return to this point and its implications for my argument in Section III.A.2 of the article.

11 See Michael R. Gordon, Helene Cooper & Michael D. Shear, Dozens of U.S. Missiles Hit Air Base in Syria, N.Y. Times (Apr. 6, 2017), at https://www.nytimes.com/2017/04/06/world/middleeast/us-said-to-weigh-military-responses-to-syrian-chemical-attack.html?emc=edit_na_20170406&nl=breaking-news&nlid=47276260&ref=cta&_r=1.

12 Text: Trump Statement on U.S. Military Strikes in Syria, Reuters (Apr. 6, 2017), at http://www.reuters.com/article/us-mideast-crisis-syria-text-idUSKBN1790A1.

13 The current U.S. position on self-defense is a good example. See Hakimi & Cogan, supra note 8, at 278–86 (2016).

14 See Daugirdas, Kristina & Mortenson, Julian Davis, Contemporary Practice of the United States Relating to International Law, 111 AJIL 781, 783–84 (2017)Google Scholar.

15 See infra notes 162–166 and accompanying text.

16 E.g., Ryan Goodman, What Do Top Legal Experts Say About the Syria Strikes?, Just Security (Apr. 7, 2017), at https://www.justsecurity.org/39712/top-legal-experts-syria-strikes (quoting Michael Glennon & Derek Jinks); Julian Ku, Trump's Syria Strike Clearly Broke International Law—and No One Seems to Care, Vox (Apr. 19, 2017), at https://www.vox.com/the-big-idea/2017/4/19/15345686/syria-un-strike-illegal-un-humanitarian-law (“In the view of most international lawyers, the U.S. strike on Syria is a crystal-clear violation of the UN Charter.”); Marko Milanović, The Clearly Illegal US Missile Strike in Syria (Apr. 7, 2017), EJIL:Talk!, at https://www.ejiltalk.org/the-clearly-illegal-us-missile-strike-in-syria; Nancy Simons, The Legality Surrounding the US Strikes in Syria, Opinio Juris (Apr. 25, 2017), at http://opiniojuris.org/2017/04/25/the-legality-surrounding-the-us-strikes-in-syria (“The only conclusion from an international law standpoint is that the missile attack that occurred on the Syrian airfield is illegal.”).

17 E.g., Harold Hongju Koh, Not Illegal: But Now the Hard Part Begins, Just Security (Apr. 7, 2017), at https://www.justsecurity.org/39695/illegal-hard-part-begins; Jens David Ohlin, I Agree with Harold Koh, Opinio Juris (Apr. 8, 2017), at http://opiniojuris.org/2017/04/08/i-agree-with-harold-ko; Michael Schmitt & Chris Ford, The Use of Force in Response to Syrian Chemical Attacks: Emergence of a New Norm?, Just Security (Apr. 8, 2017), at https://www.justsecurity.org/39805/force-response-syrian-chemical-attacks-emergence-norm; cf. Goodman, supra note 16 (“The strikes might constitute an important precedent—and perhaps a watershed moment—driving the emergence of a new rule permitting humanitarian intervention in some circumstances … [, but] the Trump administration would need to provide a fully articulated defense of the strikes on humanitarian grounds … [and] some non-trivial fraction of the international community would need to react to the strikes in such way as to make clear its endorsement or acceptance of this defense.”) (quoting Derek Jinks).

18 E.g., Goodman, supra note 16 (“The use of force in Syria underscores the ongoing erosion of UN Charter norms on the lawful use of force under international law.”) (quoting Fionnuala Ni Aolain); id. (“The real question that's raised—yet again—is whether the Charter restraints haven't been violated so many times by so many nations that they no longer constitute good law.”) (quoting Michael Glennon); Ku, supra note 16 (“The uncomfortable answer seems to be that, at least with respect to this question—can a state use military force against a regime that uses banned weaponry against citizens?—international law simply doesn't matter very much.”).

19 E.g., John Bellinger, What Was the Legal Basis for the U.S. Air Strikes Against Syria?, Lawfare (Apr. 6, 2017), at https://www.lawfareblog.com/what-was-legal-basis-us-air-strikes-against-syria (“[I]t would be better if the Administration attempts to explain its actions as lawful, or at least justified, under international law.”); Goodman, supra note 16 (“It is essential that the United States articulate its international law justification for the strikes … .”) (quoting Brian Egan); id. (“Does the United States really want to convey to other international stakeholders is that there is an undefined, extra-legal grey zone where they can use force without regard to Charter principles?”) (quoting Stephen Pomper); Rebecca Ingber, International Law Is Failing Us in Syria, Just Security (Apr. 12, 2017), at https://www.justsecurity.org/39895/international-law-failing-syria (“[By] ask[ing] states to disregard international law because it fails to meet our sense of what is legitimate … we risk eroding these legal rules.”); Shane Reeves, The Problem of Morally Justifying the United States Strike in Syria, Lawfare (Apr. 11, 2017), at https://www.lawfareblog.com/problem-morally-justifying-united-states-strike-syria (“Here lies the problem with these types of moral-based use of force decisions: they are inherently subjective and, consequently, easily abused.”); Anthea Roberts, Syrian Strikes: A Singular Exception or a Pattern and a Precedent?, EJIL:Talk! (Apr. 10, 2017), at https://www.ejiltalk.org/syrian-strikes-a-singular-exception-or-a-pattern-and-a-precedent (“[T]he lesson that other great powers are likely [to] take from this action is that, if the United States doesn't think international law applies to its own actions, neither should they.”); Ben Saul, US Missile Strikes Expose the Untenable Status Quo in International Law, Chatham House (Apr. 26, 2017), at https://www.chathamhouse.org/expert/comment/us-missile-strikes-expose-untenable-status-quo-international-law (“The US strikes were remarkable for the conspicuous absence of any serious effort by the US or supporting states to reconcile them with international law.”).

20 Reisman, W. Michael, The Quest for World Order and Human Dignity in the Twenty-First Century: Constitutive Process and Individual Commitment, 351 Recueil des Cours 1, 91, 377403 (2012)Google Scholar [hereinafter Reisman, The Quest for World Order and Human Dignity].

21 Id. at 90.

22 Id. at 89.

23 There are two possible differences between Reisman's theory and my own. First, in his account, the myth system is available to and governs most actors, while the operational code is known to and used by only key decisionmakers. Id. at 303–22. I am not sure to what extent that distinction holds for the phenomenon that I am describing. The two forms of jus ad bellum regulation do not appear to be directed at different audiences, and though the informal regulation is regularly discounted or overlooked, it is more or less hiding in plain sight. Second, to the extent that Reisman defines the content of the operational code, he does so in generally applicable substantive terms. Id. at 377–403. I emphasize that the law's content sometimes manifests in a more procedural and particularistic form.

24 For excellent overviews, see José E. Alvarez, The Impact of International Organizations on International Law 53–141, 366–69 (2017); Johnstone, Ian, The Security Council and International Law, in The UN Security Council in the Twenty-First Century 771 (von Einsiedel, Sebastian, Malone, David M. & Ugarte, Bruno Stagno eds., 2016)Google Scholar.

25 UN Charter Art. 2, para. 4.

26 E.g., GA Res. 2131 (XX), at 11–12 (Dec. 21, 1965) (positing that armed interventions “for any reason” are unlawful); Randelzhofer & Dörr, supra note 5, at 211.

27 UN Charter Art. 51.

28 Id. Arts. 24, 39.

29 For more details, see Christian Henderson, The Centrality of the United Nations Security Council in the Legal Regime Governing the Use of Force, in Research Handbook, supra note 5, at 120.

30 UN Charter Art. 43.

31 See generally Henderson, Christian, International Measures for the Protection of Civilians in Libya and Côte D'Ivoire, 60 Int'l & Comp. L. Q. 767, 778 (2011)CrossRefGoogle Scholar (“[T]he problem of ambiguity, intentional or otherwise, in the mandates of Chapter VII resolutions of the UN Security Council, along with the possibilities for unilateral and potential controversial interpretations that this provides, is one that still persists.”); Lobel, Jules & Ratner, Michael, Bypassing the Security Council: Ambiguous Authorizations to Use Force, Cease-Fires and the Iraqi Inspection Regime, 93 AJIL 124, 125 (1999)CrossRefGoogle Scholar (“This ‘contracting out’ mode leaves individual states with wide discretion to use ambiguous, open-textured resolutions to exercise control over the initiation, conduct and termination of hostilities.”).

32 E.g., Lobel & Ratner, supra note 31, at 125 (arguing that fidelity to the Charter “require[s] that the Security Council retain strict control over the initiation, duration, and objectives of the use of force.”); Nico Krisch, Article 42, in Charter Commentary, supra note 5, 1330, 1342 (“The Charter seeks centralization of the use of force… . If this warrants a general rule in favour of restrictive interpretation, it also supports the proposition that a resolution has to give clear indications if it is to be interpreted as mandating the use of force.”).

33 For a fuller defense of this claim, see Hakimi & Cogan, supra note 8.

34 Compare, e.g., Bethlehem, Daniel, Self-Defense Against an Imminent or Actual Armed Attack by Nonstate Actors, 106 AJIL 770 (2012)CrossRefGoogle Scholar, with, e.g., Akande, Dapo & Liefländer, Thomas, Clarifying Necessity, Imminence, and Proportionality in the Law of Self-Defense, 107 AJIL 563 (2013)CrossRefGoogle Scholar, and Tladi, Dire, The Nonconsenting Innocent State: The Problem with Bethlehem's Principle 12, 107 AJIL 570 (2013)CrossRefGoogle Scholar.

35 Hakimi & Cogan, supra note 8; cf. Reisman, The Quest for World Order and Human Dignity, supra note 20, at 105 (“[U]norganized and non-hierarchical systems tend to mirror the power process, in which the quintessential grundnorm is Thucydides’ ‘The strong do what they will and the weak suffer what they must.’”).

36 See, e.g., Olivier Corten, The Law Against War: The Prohibition on the Use of Force in Contemporary International Law 5–29 (2012); Christine Gray, International Law and the Use of Force 6–24 (3d ed. 2008).

37 To be clear, the other standards also have procedural components, but the processes for making decisions under them are more decentralized.

38 See Statute of the International Court of Justice, Art. 38, para. 1, annexed to UN Charter; Thirlway, Hugh, The Sources of International Law, in International Law 91, 9293 (Evans, Malcolm D. ed., 4th ed. 2014)Google Scholar (“The generally recognized formal sources are identified in Article 38 … but the two most important sources in practice are treaties and international custom.”).

39 Military and Paramilitary Activities in and Against Nicaragua, supra note 2, para. 34 (“There can be no doubt that the issues of the use of force and collective self-defence … are issues which are regulated both by customary international law and by treaties, in particular the United Nations Charter.”).

40 Vienna Convention on the Law of Treaties, Arts. 11, 14, May 23, 1969, 1155 UNTS 331.

41 See, e.g., Military and Paramilitary Activities in and Against Nicaragua, supra note 2, para. 207 (“[F]or a new customary rule to be formed, not only must the acts concerned ‘amount to a settled practice,’ but they must be accompanied by the opinio juris sive necessitatis.”).

42 E.g., id., paras. 175–76; Oil Platforms (Islamic Rep. of Iran v. U.S.), Judgment, 2003 ICJ Rep. 161, para. 42 (Nov. 6).

43 Murphy, Sean D., Protean Jus ad Bellum, 27 Berk. J. Int'l L. 22, 2223 (2009)Google Scholar.

44 Id. at 51–52.

45 See Koh, Harold Hongju, Why Do Nations Obey International Law?, 106 Yale L.J. 2599, 2645–58 (1997)Google Scholar; Koh, Harold Hongju, The 1998 Frankel Lecture: Bringing International Law Home, 35 Hous. L. Rev. 623, 626 (1998)Google Scholar (“By transnational legal process, I mean the process whereby an international law rule is interpreted through the interaction of transnational actors in a variety of law-declaring fora, then internalized into a nation's domestic legal system.”).

46 Koh, Harold Hongju, Address, The War Powers and Humanitarian Intervention, 53 Hous. L. Rev. 971, 1012 (2016)Google Scholar [hereinafter Koh, The War Powers and Humanitarian Intervention].

47 Id. at 971, 980, 1011–16.

48 Id. at 1011.

49 Jutta Brunnée & Stephen J. Toope, Legitimacy and Legality in International Law: An Interactional Account (2010).

50 Id. at 7.

51 Id. at 7, 75.

52 Id. at 349.

53 This section draws on insights in Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule-Based Decision-Making in Law and in Life (1991).

54 Compare, e.g., John Dugard, Special Rapporteur on Diplomatic Protection, First Rep. on Diplomatic Protection, para. 46, UN Doc. A/CN.4/506 (Mar. 7, 2000) (articulating a standard on when states may use force to protect nationals abroad), and Yoram Dinstein, War, Aggression and Self-Defence 231–34 (4th ed. 2005) (claiming that such force is sometimes permissible), with, e.g., Gray, supra note 36, at 157 (claiming that “few states accept[] a legal right to protect nationals abroad”).

55 See infra Part V.D.1.

56 Compare Reisman, W. Michael & Armstrong, Andrea, The Past and Future of the Claim of Preemptive Self-Defense, 100 AJIL 525 (2006)CrossRefGoogle Scholar (documenting that several states have advanced expansive claims on anticipatory self-defense), with Albrecht Randelzhofer & Georg Nolte, Article 51, in Charter Commentary, supra note 5, at 1397, 1424 (“[T]he object and purpose of the prohibition against the use of force can … be fulfilled by a narrowly circumscribed right of anticipatory self-defence.”).

57 See Oona A. Hathaway, Rebecca Crootof, Philip Levitz, Haley Nix, Aileen Nowlan, William Purdue & Julia Spiegel, The Law of Cyber-Attack, 100 Cal. L. Rev. 817, 845–48 (discussing the most prominent positions on defensive force in this context).

58 For evidence of the debates, see supra notes 34, 54–57 and accompanying text.

59 This is so for two reasons. First, these institutions infrequently apply the standard on self-defense. Second, when they apply it, they often avoid taking clear positions on its content.

60 See supra note 36 and accompanying text.

61 To appreciate the difference between precision and generality, imagine a standard that prohibits all unilateral force. This standard would be precise. It would clearly establish that such force is unlawful. The standard would also operate at a very high level of generality. It would subject to similar treatment many different fact patterns. Now imagine a standard that consists of fifty sub-standards, each delineating, with considerable specificity, when unilateral force is lawful. The sub-standards would still be precise so long as they establish whether particular operations are lawful. But compared to the first standard, they would operate at lower levels of generality. Each would govern a narrower set of cases.

62 However, African states have agreed to allow certain regional organizations to conduct such interventions. See Constitutive Act of the African Union, Art. 4(h); African Union, Protocol Relating to the Establishment of the Peace and Security Council of the African Union, Art. 4(j); Economic Community of West African States, Protocol Relating to the Mechanism for Conflict Prevention, Management, Resolution, Peacekeeping and Security, Art. 25, ECOWAS Doc. A/P10/12/99 (Dec. 10, 1999). The proper relationship between these regional instruments and Article 53 of the UN Charter, which provides for the Council's primacy over regional organizations, is a matter of some dispute. See UN Charter Art. 53; Christian Walter, Article 53, in Charter Commentary, supra note 5, 1478, 1491–92.

63 See, e.g., Michael Byers, War Law: International Law and Armed Conflict 92 (2005) (reviewing incidents and concluding that there is an “absence of opinio juris”); Simon Chesterman, Just War or Just Peace?: Humanitarian Intervention and International Law 84, 87 (2002) (same).

64 See generally Nicholas J. Wheeler, Saving Strangers: Humanitarian Intervention in International Society (2000).

65 E.g., GA Res. 60/1, para. 139, 2005 World Summit Outcome (Sept. 16, 2005) (“[W]e are prepared to take collective action, in a timely and decisive manner, through the Security Council, in accordance with the Charter, including Chapter VII, on a case-by-case basis … should peaceful means be inadequate and national authorities are manifestly failing to protect their populations from genocide, war crimes, ethnic cleansing and crimes against humanity.”) (emphasis added); Group of 77, supra note 5, para. 54 (“We reject the so-called ‘right’ of humanitarian intervention, which has no legal basis in the United Nations Charter or in the general principles of international law.”).

66 See, e.g., sources at supra note 5.

67 E.g., Koh, The War Powers and Humanitarian Intervention, supra note 46; Richard B. Lillich, Humanitarian Intervention: A Reply to Ian Brownlie and a Plea for Constructive Alternatives, in Law and Civil War in the Modern World 229 (John Norton Moore ed., 1974); see also Jane Stromseth, Rethinking Humanitarian Intervention: The Case for Incremental Change, in Humanitarian Intervention: Ethical, Legal, and Political Dilemmas 232, 241–45 (J.L. Holzgrefe & Robert O. Keohane eds., 2003) (describing the dominant positions in this debate).

68 See, e.g., Trachtman, Joel P., The Economic Structure of the Law of International Organizations, 15 Chi. J. Int'l L. 162, 185–86 (2014)Google Scholar.

69 This phenomenon is not particular to the jus ad bellum. See Harlan Grant Cohen, Finding International Law: Rethinking the Doctrine of Sources, 93 Ia. L. Rev. 65, 70 (2007) (“[There] is a disconnect between the rules identified as law by the doctrine of sources and the rules actually treated as law by the actors in the international system.”); Hugh Thirlway, The Sources of International Law, in International Law 115, 117 (Malcolm D. Evans ed., 2006) (recognizing that the doctrine “has attracted enormous amounts of … criticism” and “presents some anomalies and difficulties”).

70 The New Haven school of jurisprudence is particularly on point. This school defines international law not, like the sources doctrine, as a set of standards, but in processual terms—as a process for communicating policies that are controlling and authoritative. Michael Reisman, the longstanding “dean” of this school, see Koh, Harold Hongju, Michael Reisman, Dean of the New Haven School of International Law, 34 Yale J. Int'l L. 501 (2009)Google Scholar, has emphasized that a policy can be law, even if it is not articulated in legal language, even if it is highly context-specific, and even if it deviates from the “law on the books.” See Reisman, W. Michael, Address, International Lawmaking: A Process of Communication, 75 ASIL Proc. 101 (1981)Google Scholar [hereinafter Reisman, International Lawmaking]; Reisman, The Quest for World Order and Human Dignity, supra note 20, at 95–97. The key, again, is that politically relevant actors communicate to one another that a policy decision is both controlling and authoritative. With the jus ad bellum’s informal regulation, a decision to use force is controlling in the sense that states broadly conduct, support, and tolerate it. However, because the decision lacks good justification under the general standards, it is on its own deficient in authority. The Council's stamp of approval helps cure that deficiency, making the decision not just controlling but also authoritative—or, under the New Haven approach, lawful.

71 UN Charter Art. 39; see also id. Art. 30 (“The Security Council shall adopt its own rules of procedure… .”); Certain Expenses of the United Nations (Article 17, Paragraph 2 of the Charter), Advisory Opinion, 1962 ICJ Rep. 151, 168 (July 20) (“[E]ach organ [of the United Nations] must, in the first place at least, determine its own jurisdiction.”).

72 Supra Part II.

73 See, e.g., Brunnée & Toope, supra note 49, at 317 (“[T]here is no plausible alternative to collective legitimation of the use of force through the Security Council… .”). I leave open the question of whether other institutions could ever occupy this position. The idea in the Uniting for Peace Resolution that the General Assembly may in certain circumstances authorize force, as a substitute for the Security Council, fell quickly out of favor when the Council became active at the end of the Cold War. See Uniting for Peace Resolution, GA Res 377 A (V), UN GAOR, 5th Sess., Supp. No. 20, at 10, UN Doc. A/1775 (Nov. 3, 1950); see also Christina Binder, Uniting for Peace Resolution (1950), in Max Planck Encyclopedia of Public International Law, para. 35, available at http://opil.ouplaw.com/home/epil (last updated June 2013) (“After the end of the Cold War, the … increased activity of the Council led to a certain loss of the resolution's importance.”). But one can imagine the Council again losing its standing and other institutions, like the General Assembly or regional organizations, filling the vacuum.

74 E.g., sources at supra note 24; cf. Reisman, International Lawmaking, supra note 70, at 110 (“It is the audience, whether or not its members realize it, that endows the prescriber with the authority that renders his communications prescription.”).

75 E.g., Corten, supra note 36, at 9 (“It is essential to take into account the decisions of [the Council], but also its silences, if we are to specify the meaning of the rule of law … .”); Gray, supra note 36, at 20 (“It may be argued that condemnation of a particular use of force by the Security Council … is conclusive or at least persuasive as to illegality.”).

76 Here is a good example: the day after the September 11, 2001, terrorist attacks, the Security Council adopted Resolution 1368, which did not purport to be binding but recognized “the inherent right of individual or collective self-defence in accordance with the Charter,” expressed the Security Council's “readiness to take all necessary steps to respond to the terrorist attacks,” and described all acts of international terrorism as “threats to international peace and security.” SC Res. 1368 (Sept. 12, 2001). There has been an extensive scholarly debate about whether the resolution reflected the Council's interpretation of Article 51 and, if it did, what that interpretation was. The entire debate assumes that any Council interpretation would be highly persuasive, if not conclusive, evidence of Article 51's content. See Tom Ruys, ‘Armed Attack’ and Article 51 of the UN Charter: Evolutions in Customary Law and Practice 433–43 (2010) (reviewing the literature).

77 As a matter of practice, statements by the Security Council president are made either by consensus in informal consultations or through a “no objection” procedure; Council press statements are made by consensus. See UN Security Council: Working Methods Handbook, at 90, UN Sales No. 11.VII.1 (2011).

78 I focus in this article on a legal legitimacy that is mostly sociological. Such legitimacy is different from, but ultimately related to, two other kinds of legal legitimacy: (1) jurisprudential legitimacy, which attaches to conduct that satisfies certain formal criteria of law; and (2) normative legitimacy, which goes to when or why the law ought to be followed. On these dimensions of legitimacy, see Roughan, Nicole, Mind the Gaps: Authority and Legality in International Law, 27 Eur. J. Int'l L. 329, 340–41 (2016)CrossRefGoogle Scholar. On the more general point that deliberation at the Security Council can legitimize decisions, see Johnstone, Ian, Legislation and Adjudication in the UN Security Council: Bringing Down the Deliberative Deficit, 102 AJIL 275 (2008)CrossRefGoogle Scholar.

79 E.g., Ian Johnstone, When the Security Council is Divided: Imprecise Authorizations, Implied Mandates, and the ‘Unreasonable Veto,’ in The Oxford Handbook of The Use of Force in International Law 227, 242–43 (Marc Weller ed., 2015) [hereinafter Johnstone, When the Security Council is Divided]; Scott, Craig, Interpreting Intervention, 39 Can. Y.B. Int'l L. 333, 348 (2001)Google Scholar.

80 The only possible exception is the French intervention in Mali. See infra Section IV.A.2.a.

81 E.g., Independent International Commission on Kosovo 186 (2000); Johnstone, Ian, The Plea of “Necessity” in International Legal Discourse: Humanitarian Intervention and Counter-terrorism, 43 Colum. J. Transnat'l L. 337, 339 (2005)Google Scholar [hereinafter Johnstone, The Plea of “Necessity”]; Michael Byers & Simon Chesterman, Changing the Rules about Rules? Unilateral Humanitarian Intervention and the Future of International Law, in Humanitarian Intervention: Ethical, Legal, and Political Dilemmas, supra note 67, at 177, 198–201; Simma, Bruno, NATO, the UN and the Use of Force: Legal Aspects, 10 Eur. J. Int'l L. 1, 22 (1999)CrossRefGoogle Scholar.

82 Thomas M. Franck, Recourse to Force: State Action against Threats and Armed Attacks 186 (2002).

83 Id. at 186, 187.

84 Id. at 187.

85 Id. at 191 (“[T]he distinction between what is justified (exculpated) and what is excusable (mitigated) is so fine as to be of pure (yet also considerable) theoretical interest.”); see also Roberts, Anthea, Legality vs Legitimacy: Can Uses of Force Be Illegal but Justified?, in Human Rights, Intervention, and the Use of Force 179 (Alston, Philip & Macdonald, Euan eds., 2008)Google Scholar (arguing that, in the absence of a court or other institution to declare the action unlawful, the distinction is immaterial) [hereinafter Roberts, Legality vs Legitimacy]; Simma, supra note 81, at 22 (“[O]nly a thin red line separates NATO's action on Kosovo from international legality.”); but see Johnstone, The Plea of “Necessity,” supra note 81, at 385 (“[A]n interpretive community exists … [and] is capable of drawing precisely the distinction that the necessity excuse requires.”).

86 Franck, supra note 82, at 171–72 (arguing that recognizing a new exception to Article 2(4) “could launch the international system down the slippery slope into an abyss of anarchy.”); see also, e.g., Chesterman, supra note 63, at 231 (“[I]t is more dangerous to hand states a ‘right’—even of such a limited nature—than simply to assert the cardinal principle of the prohibition of the use of force and let states seek a political justification for a particular action if they find themselves in breach of that norm.”); Johnstone, The Plea of “Necessity,” supra note 81, at 387 (arguing that, rather than establish an exception that would justify such interventions in law and undercut the blanket prohibition, the interventions should be unlawful but excusable through case-specific pleas of necessity); Simma, supra note 81, at 22 (“To resort to illegality as an explicit ultima ratio for reasons as convincing as those put forward in the Kosovo case is one thing. To turn such an exception into a general policy is quite another.”); cf. Roberts, Legality vs Legitimacy, supra note 85, at 190 (“If there is no clear statement that unilateral humanitarian intervention is illegal, that opens the door to claims that there is or should be a legal exception to the prohibition on the use of force in extreme humanitarian crises.”).

87 See, e.g., Military and Paramilitary Activities in and Against Nicaragua, supra note 2, para. 246 (recognizing that an outside state's intervention “is already allowable at the request of the government of a State”); GA Res. 3314 (XXIX), Declaration on the Definition of Aggression, Annex, Art. 3(e), UN Doc. A/9631 (Dec. 14, 1974) (defining as unlawful “[t]he use of armed forces of one State which are within the territory of another State with the agreement of the receiving State, in contravention of the conditions provided for in the agreement … ”).

88 UN Charter Art. 2, para. 4.

89 For extensive discussions of this issue, see Doswald-Beck, Louise, The Legal Validity of Military Intervention by Invitation of the Government, 56 Brit. Y.B. Int'l L. 189 (1985)Google Scholar; Gregory H. Fox, Intervention by Invitation, in The Oxford Handbook of The Use of Force in International Law, supra note 79, at 816.

90 See Brad R. Roth, Governmental Illegitimacy in International Law 4 (1999) (“The multilateral treaties, authoritative declarations, and ad hoc responses to crisis that constitute the source material of international legal doctrine exhibit tensions, paradoxes, and even contradictions with respect to the question of governmental illegitimacy.”); Nußberger, Benjamin, Military Strikes in Yemen in 2015: Intervention by Invitation and Self-defence in the Course of Yemen's ‘Model Transitional Process,’ 4 J. Use of Force Int'l L. 110, 131, 143 (2017)Google Scholar (asserting that, “in principle states have refrained from intervening in civil war situations,” but recognizing that “state practice within the field of the doctrine of intervention by invitation sometimes appears to be ambiguous, arbitrary and motivated by political reasons or, to quote Kolb, ‘chaotique.’”) (quoting Robert Kolb, Ius Contra Bellum. Le Droit International Relative au Maintien de la Paix 328 (2d ed. 2009)).

91 For background on this incident, see UN Human Rights Council, Rep. of the Indep. Expert on the Situation of Human Rights in Mali, at paras. 8–26, UN Doc. A/HRC/25/72 (Jan. 10, 2014); Bannelier, Karine & Christakis, Theodore, Under the UN Security Council's Watchful Eyes: Military Intervention by Invitation in the Malian Conflict, 26 Leiden J. Int'l L. 855 (2013)CrossRefGoogle Scholar; Maslen, Stuart Casey, Armed Conflict in Mali in 2013, in The War Report: Armed Conflict in 2013, at 147 (Casey-Maslen, Stuart ed., 2014)Google Scholar.

92 See Mali Coup Leaders to Stand Down as Part of ECOWAS Deal, BBC News (Apr. 7, 2012), at http://www.bbc.com/news/world-africa-17642276.

93 Adam Nossiter, Jihadists’ Fierce Justice Drives Thousands to Flee Mali, N.Y. Times (July 17, 2012), at http://www.nytimes.com/2012/07/18/world/africa/jidhadists-fierce-justice-drives-thousands-to-flee-mali.html.

94 SC Res. 2085 (Dec. 20, 2012).

95 See Tzeng, Peter, Humanitarian Intervention at the Margins: An Examination of Recent Incidents, 50 Vand. J. Transnat'l L. 415, 446 (2017)Google Scholar.

96 Permanent Rep. of France to the UN, Identical Letters Dated 11 January 2013 from the Permanent Representative of France to the UN Addressed to the Secretary-General and the President of the Security Council, UN Doc. S/2013/17 (Jan. 14, 2013).

97 On self-defense, see infra note 104 and accompanying text. On Security Council authorization, see infra notes 126–129 and accompanying text.

98 UN Security Council Press Release, Security Council Press Statement on Mali Crisis, UN Doc. SC/10590 (Mar. 22, 2012), at http://www.un.org/press/en/2012/sc10590.doc.htm; see also UNSC, Statement by the President of the Security Council, at 1, UN Doc. S/PRST/2012/7 (Mar. 26, 2012) (same, with modest grammatical changes); SC Pres. Statement 2012/9 (Apr. 4, 2012) (same); SC Res 2056, pmbl. (July 5, 2012) (same).

99 SC Res 2056, supra note 98, para. 8 (July 5, 2012).

100 E.g., SC Res. 2071, para. 1 (Oct. 12, 2012); UN Security Council Press Release, Security Council Press Statement on Mali, UN Doc. SC/10741 (Aug. 10, 2012), at http://www.un.org/press/en/2012/sc10741.doc.htm; UN Security Council Press Release, Security Council Press Statement on Mali and Sahel, UN Doc. SC/10772 (Sept. 21, 2012), at http://www.un.org/press/en/2012/sc10772.doc.htm; UN Security Council Press Release, Security Council Press Statement on Mali, UN Doc. SC/10851 (Dec. 11, 2012), at http://www.un.org/press/en/2012/sc10851.doc.htm.

101 E.g., SC Res. 2085, supra note 94, at para. 4; Security Council Press Statement on Mali and Sahel, supra note 100.

102 See Tzeng, supra note 95, at 447 n. 192 (noting that Belgium, Canada, Chad, Denmark, Germany, the Netherlands, Spain, Sweden, the United Arab Emirates, the United Kingdom, and the United States all provided logistical or military support for the French operation, and that Israel, Colombia, and Chile publicly endorsed it); France Confirms Mali Military Intervention, BBC News (Jan. 11, 2013), at http://www.bbc.com/news/world-africa-20991719 (“British Foreign Secretary William Hague said … that the UK supported the French decision to help Mali's government against northern rebels. The US and African Union have also expressed support for the mission.”).

103 UN Security Council Press Release, Security Council Press Statement on Mali, UN Doc. SC/10878 (Jan. 10, 2013).

104 In an earlier article, I claimed that, notwithstanding what France and the Security Council said about consent, the intervention seemed more defensive than consensual. After all, Dioncounda Traoré’s consent was evidently imperfect, and France was almost certainly motivated, at least in part, by its concern that the jihadi groups in Mali would affect its own national security. See Afua Hirsch & Kim Willsher, Mali Conflict: France Has Opened Gates of Hell, Says Rebels, Guardian (Jan. 14, 2013), at http://www.theguardian.com/world/2013/jan/14/mali-conflict-france-gates-hell. But in discounting the claims on consent, I myself overlooked the informal regulation and thus the role that the Security Council played in this incident. See Hakimi, Monica, Defensive Force Against Non-state Actors: The State of Play, 91 Int'l L. Stud. 1, 1112 (2015)Google Scholar [hereinafter Hakimi, Defensive Force Against Non-state Actors]. Fortunately, this error does not detract from the article's main point, which was that the law on defensive force against nonstate actors was in flux, with multiple legal positions in play.

105 Mark Doyle, Mali Conflict: UN Backs France's Military Intervention, BBC News (Jan. 15, 2013), at http://www.bbc.com/news/world-africa-21021132.

106 SC Res. 2100, pmbl. (Apr. 25, 2013).

107 For detailed accounts of the situation, see International Crisis Group, Yemen at War, Middle East Briefing No. 45 (Mar. 27, 2015); Nußberger, supra note 90, at 112–18.

108 The UN Security Council had not authorized it, and despite some loose rhetoric, the intervening states did not try to advance a credible theory on self-defense. See Permanent Rep. of Qatar to the UN, Identical Letters Dated 26 March 2015 from the Permanent Representative of Qatar to the UN addressed to the Secretary-General and the President of the Security Council, UN Doc. S/2015/217 (Mar. 27, 2015) (quoting Hadi's invitation and transmitting the explanation for the intervention on behalf of five GCC states). The most that was said was that Houthi militants were active “near the Saudi Arabia border,” that they “carried out a bare-faced and unjustified attack on the territory of Saudi Arabia, in November 2009, and [that] their current actions make it clear that they intend to do so again.” Id. at 5; cf. 26th Arab League Summit, Final Communique of the 26th Arab League Summit (Mar. 29, 2015) (invoking Hadi's invitation and self-defense). If those grounds suffice to justify defensive force, it would be under an expansive application of an already very controversial interpretation of Article 51. For example, even the United States claims that an attack must be, in some way, imminent in order to trigger the right to use defensive force anticipatorily. See U.S. White House, Report on the Legal and Policy Frameworks Guiding the United States’ Use of Military Force and Related National Security Operations, at 9 (Dec. 2016). No claim or showing of imminence was made in the Yemen case. Further, even if Article 51 licensed some defensive force against Houthi militants in Yemen, it probably would not have licensed an intervention of this scale.

109 SC Res. 2014, pmbl. & para. 7 (Oct. 21, 2011).

110 SC Res. 2051, paras. 4, 6 (June 12, 2012) (recognizing Hadi as president and “demand[ing] the cessation of all actions aimed at undermining” his government); SC Pres. Statement 2012/8 (Mar. 29, 2012) (“welcom[ing] the Yemeni-led peaceful transition process … [and noting] the 25 February transfer of power to President Abd Rabbuh Mansour Hadi, as per the Gulf Cooperation Council [process]”).

111 SC Res. 2140, pmbl. & para. 1 (Feb. 26, 2014) (“Commending the engagement of the Gulf Cooperation Council … [and] the leadership of President Abd Rabbo Mansour Hadi” and “Reaffirm[ing] the need for the full and timely implementation of the political transition … in line with the GCC [process].”); SC Res. 2201, paras. 1, 4 (Feb. 15, 2015) (“Strongly deplor[ing] actions taken by the Houthis to dissolve parliament and take over Yemen's government institutions, including acts of violence,” and “[s]trongly call[ing] upon all parties, in particular the Houthis, to abide” by the [GCC-led transition mechanism].”).

112 SC Pres. Statement 2013/3 (Feb. 15, 2013) (“The Security Council welcomes President Abed Rabbo Mansour Hadi's announcement of the launch of the National Dialogue Conference” and “expresses concern over reports of interference in the transition by individuals in Yemen representing the former regime, the former opposition, and others who do not adhere to the … transition process.”); SC Pres. Statement 2014/18 (Aug. 29, 2014) (“The Security Council welcomes the recent progress in Yemen's political transition, in line with the Gulf Cooperation Council [process,] … supports President Abd Rabbo Mansour Hadi in his efforts… . [and] note[s] with concern that the Houthis and others continue to stoke the conflict in the north in an attempt to obstruct the political transition.”).

113 UN Security Council Press Release, Security Council Press Statement on Fighting in Yemen, UN Doc. SC/11470 (July 11, 2014) (“express[ing] support to President Abdo Rabbo Mansour Hadi as he leads the political transition” and “demand[ing] that the Houthis, all armed groups and parties involved in the violence withdraw and relinquish control of Amran and hand over weapons and ammunition pillaged in Amran”); UN Security Council Press Release, Security Council Press Statement on Yemen, UN Doc. SC/11578 (Sept. 23, 2014) (“The members of the Security Council underline that President [Abdrabuh Mansour] Hadi is the legitimate authority … . All parties and political actors in Yemen must unite behind President Hadi to keep the country on track to stability and security.”) (modification in original); UN Security Council Press Release, Security Council Press Statement on Yemen, UN Doc. SC/11743 (Jan. 20, 2015) (same); Security Council Press Release, Security Council Press Statement on Yemen, UN Doc. SC/11826 (Mar. 20, 2015) (same).

114 UN SCOR, 70th Sess., 7411th mtg., UN Doc. S/PV.7411 (Mar. 22, 2015).

115 E.g., id. at 6 (“[T]he solution to the situation in Yemen is through a peaceful, inclusive, orderly and Yemeni-led political transition process … .”); UN SCOR, 70th Sess., 7426th mtg., UN Doc. S/PV.7426 (Apr. 14, 2015) (statements of multiple countries underscoring the need for a political and/or consensual solution to the crisis).

116 See also SC Res. 2216, pmbl., para. 1 (Apr. 14, 2015) (“[r]eaffirming its support for the legitimacy of the President of Yemen, Abdo Rabbo Mansour Hadi,” and “demand[ing] that all Yemeni parties, in particular the Houthis, … refrain from further unilateral actions that could undermine the political transition in Yemen”) (emphasis in original).

117 SC Pres. Statement 2015/8 (Mar. 22, 2015).

118 SC Res. 2216, supra note 116.

119 See Nußberger, supra note 90, at 118, 121–23; Ruys, Tom & Ferro, Luca, Weathering the Storm: Legality and Legal Implications of the Saudi-Led Military Intervention in Yemen, 65 Int'l & Comp. L. Q. 61, 6768 (2016)CrossRefGoogle Scholar.

120 But cf. Ruys & Ferro, supra note 119, at 98 (concluding that the incident illustrates “the risk of abuse” and that the intervening states and their supporters “have undermined the primary role of the UN Security Council for the maintenance of international peace and security, and set a dangerous precedent”).

121 Wet, Erika de, Reinterpreting Exceptions to the Use of Force in the Interest of Security: Forcible Intervention by Invitation and the Demise of the Negative Equality Principle, 111 AJIL Unbound 307 (2017)CrossRefGoogle Scholar (citing the interventions in Mali and Yemen for support); Wet, Erika de, The Modern Practice of Intervention by Invitation in Africa and Its Implications for the Prohibition of the Use of Force, 26 Eur. J. Int'l L. 979, 996–97 (2015)Google Scholar (citing the intervention in Mali).

122 Bannelier & Christakis, supra note 91, at 873–74.

123 Nußberger, supra note 90, at 147.

124 E.g., Bannelier & Christakis, supra note 91, at 873 (“The attitude of the UNSC during these interventions clearly demonstrates that the Council accepted the validity of the legal basis of intervention by invitation … .”); Nußberger, supra note 90, at 147 (explaining that Council decisions endorsing a transitional process are “[o]f special relevance” to determining whether the “process and its leader are representative of the people's will” and thus whether the intervention satisfies the consent standard).

125 One might claim that the Council's decisions went only to the recognition of each country's government and did not have any bearing on the jus ad bellum. But in answering the recognition question, the Council plainly shaped the jus ad bellum’s content in these cases.

126 Johnstone, When the Security Council is Divided, supra note 79, at 242–43; see also ECOWAS Press Release, Statement of the President of the ECOWAS Commission on the Situation in Mali (Jan. 12, 2013), available at http://reliefweb.int/report/mali/statement-president-ecowas-commission-situation-mali (“welcom[ing] UN Security Council Press Release of 10th January 2013 authorising immediate intervention in Mali to stabilise the situation”); Fox, supra note 89, at 826 (using the Mali incident to suggest an analytic distinction between internal conflicts that are subject to the default rules on interventions by invitation and those in which the Council authorizes force).

127 Johnstone, When the Security Council is Divided, supra note 79, at 242–43.

128 See SC Res. 2100, supra note 106, para. 18.

129 The closest the Council came to authorizing France's action was in “[u]rg[ing] Member States, regional and international organizations to provide coordinated support to” the African-led force. SC Res. 2085, supra note 94, para. 14. In the event, France did not provide support to an African-led force. No such force had been deployed.

130 See UN SCOR, 45th Sess., 2974th mtg., UN Doc. S/PV.2974 (Jan. 22, 1991) (statement of Liberia “recall[ing] that seven months ago we made efforts to have the Council seized with the deteriorating situation in Liberia, which efforts were not approved”).

131 Letter from Samuel K. Doe to the Chairman and Members of the Ministerial Meeting of the ECOWAS Standing Mediation Committee (July 14, 1990), reprinted in Regional Peace-Keeping and International Enforcement: The Liberian Crisis 60–61 (M. Weller ed., 1994) [hereinafter The Liberian Crisis].

132 ECOWAS Standing Mediation Committee, Decision on the Cease-fire and Establishment of an ECOWAS Cease-fire Monitoring Group for Liberia, Banjul, Republic of Gambia, A/Dec.1/8/90 (Aug. 7, 1990), reprinted in The Liberian Crisis, supra note 131, 67–68.

133 BBC Monitoring Report: ECOWAS Peace-keeping Force to be Sent to Liberia; Foreigners Released by INPFL (Aug. 7, 1990), reprinted in The Liberian Crisis, supra note 131, at 66 (reporting that, upon leaving the ECOWAS meeting, the Guinean president claimed that the ECOWAS force “do[es] not need the permission of any party involved in the conflict” and would deploy “with or without the agreement of any of the parties”).

134 Ofodile, Anthony Chukwuka, The Legality of ECOWAS Intervention in Liberia, 32 Colum. J. Transnat'l L. 381, 413 (1994)Google Scholar; Renda, Luca, Ending Civil Wars: The Case of Liberia, 23 Fletcher F. World Aff. 59, 6970 (1999)Google Scholar.

135 ECOWAS Authority of Heads of State and Government, Decision Relating to the Approval of the Decision of the Community Standing Mediation Committee Taken During its First Session from 6 to 7 August 1990, Bamako, Republic of Mali, A/DEC.1/11/90 (Nov. 28, 1990), reprinted in The Liberian Crisis, supra note 131, at 111.

136 See Permanent Rep. of Nigeria to the UN, Letter Dated Aug. 9, 1990 from the Permanent Representative of Nigeria to the UN Addressed to the Secretary-General, UN Doc. S/21485 (Aug. 10, 1990) (informing Council of action); Permanent Rep. of Gambia to the UN Letter Dated Dec. 14, 1990 from the Permanent Representative of Gambia to the UN Addressed to the Secretary-General, UN Doc. S/22025 (Dec. 20, 1990) (updating Council on situation).

137 Permanent Rep. of Nigeria to the UN, supra note 136.

138 E.g., Neil Henry, Doctors’ Group Criticizes U.S. for Not Intervening in Liberia, Wash. Post, Aug. 16, 1990, at A17 (United States); The Massacres Continue in Liberia, Globe & Mail, Sept. 1, 1990, at D6 (“[T]he Secretary-General of the Organization of African Unity has given the current peace mission his blessing.”).

139 UN SCOR, 47th Sess., 3138th mtg., at 13, UN Doc. S/PV.3138 (Nov. 19, 1992).

140 UN SCOR, 45th Sess., 2974th mtg., supra note 130, at 8.

141 SC Pres. Statement S/22133 (Jan. 22, 1991).

142 SC Pres. Statement S/23886 (May 7, 1992).

143 See Permanent Rep. of Benin to the UN, Letter Dated Oct. 28, 1992 from the Permanent Representative of Benin to the UN Addressed to the President of the Security Council, UN Doc. S/24735 (Oct. 29, 1992) (“I have been instructed [by ECOWAS] to ask you to call an emergency meeting of the Security Council … in order to consider the Liberian crisis… .”).

144 See, e.g., UN SCOR, 47th Sess., 3138th mtg., supra note 139, at 11 (statement of Benin) (“This disastrous situation requires urgent effective action, which our Governments earnestly hope will be adopted and implemented by the Security Council.”); id. at 18 (statement of Liberian representative) (“call[ing] here for the Council's support of these measures” and insisting that the “Council must act decisively to buttress, support and sustain the leaders in West Africa”); id. at 23 (statement of Senegal) (“It was in order to receive the Council's help that we asked for this meeting, convinced that at the present stage that is the best way to contribute to a future of peace in Liberia.”); id. at 31 (statement of Côte d'Ivoire) (“[I]t is imperative that, with the support of the Security Council, an effective cease-fire be put into effect speedily in Liberia … .”); id. at 37 (statement of Gambia) (“The purpose of our presence here today is to seek further support and assistance from the Security Council in resolving the Liberian conflict.”).

145 SC Res. 788 (Nov. 19, 1992).

146 SC Res. 1116, pmbl. (June 27, 1997) (expressing appreciation to ECOWAS and commending the states that contributed to the ECOWAS force); SC Res. 1100, pmbl. (Mar. 27, 1997) (same), SC Res. 1083, pmbl. (Nov. 27, 1996) (same); SC Res. 1071, pmbl. (Aug. 30, 1996) (commending ECOWAS); SC Res. 1059, pmbl. (May 31, 1996) (same); SC Res. 1041, pmbl. (Jan. 29, 1996) (same); SC Res. 1020, pmbl. (Nov. 10, 1995) (same); SC Res. 1014, pmbl. (Sept. 15, 1991) (same); SC Res. 1001, pmbl. (June 30, 1995) (same); SC Res. 985 (Apr. 13, 1995) (not commending or expressing appreciation to ECOWAS); SC Res. 972, pmbl. (Jan. 13, 1995) (commending ECOWAS); SC Res. 950, pmbl. (Oct. 21, 1994) (same); SC Res. 911, pmbl. (Apr. 21, 1994) (same); SC Res. 866, pmbl. (Sept. 22, 1993) (same); SC Res. 856, pmbl. (Aug. 10, 1993) (same); SC Res. 813, pmbl. (Mar. 26, 1993) (same).

147 See infra Section IV.B.2. But cf. Ofodile, supra note 134, at 418 (“The ECOWAS intervention in the Liberian civil war does not have any solid anchor in international law.”).

148 E.g., Nolte, Georg, Restoring Peace by Regional Action: International Legal Aspects of the Liberian Conflict, 53 Zestschrift für Ausländisches Offentliches Recht und Völkkerrecht 603, 621–26 (1993)Google Scholar; Wippman, David, Military Intervention, Regional Organizations, and Host-State Consent, 7 Duke J. Comp. & Int'l L. 209, 223, 227 (1996)Google Scholar.

149 Id. at 224.

150 Id. at 226.

151 Nolte, supra note 148, at 633–34.

152 See supra notes 133, 136–137 and accompanying text; see also Wippman, supra note 148, at 226–27; Borgen, Christopher J., The Theory and Practice of Regional Organization Intervention in Civil Wars, 26 N.Y.U. J. Int'l L. & Pol. 797, 818 (1994)Google Scholar (explaining that ECOWAS “marginalized the Doe government and forged a future for Liberia that did not envision Doe at all”).

153 Franck, supra note 82, at 156 (asserting that the incident “seemed to signal that the Council, in appropriate circumstances, could retroactively sanitize an action that may have been of doubtful legality at the time it was taken”); Levitt, Jeremy, Humanitarian Intervention by Regional Actors in Internal Conflicts: The Cases of ECOWAS in Liberia and Sierra Leone, 12 Temp. J. Int'l & Comp. L.J. 333, 347 (1998)Google Scholar (“[I]t can be said that Resolutions 788 and 866 placed a retroactive de jure seal on the ECOWAS intervention.”).

154 Walter, supra note 62, at 1501 (“Res. 788 (1992) may be interpreted as a subsequent (implicit) authorization.”).

155 Id.; see also Reisman, W. Michael, Unilateral Action and the Transformations of the World Constitutive Process: The Special Problem of Humanitarian Intervention, 11 Eur. J. Int'l L. 3, 16 (2000)CrossRefGoogle Scholar (“One of the functions of organized and institutionalized decisionmaking is to insure that due deliberation precede action thereby minimizing the inevitable tendencies to impulsiveness and arbitrariness … .”); Krisch, supra note 32, at 1343 (“The possibility of ex post authorizations would … undermine the constraining effect of having to seek prior authorization by removing the only clear identifiable condition for military measures that go beyond self-defence.”); but cf. UN Secretary-General, A More Secure World: Our Shared Responsibility, para. 272(a), UN Doc. A/59/565 (Dec. 2, 2004) (“Authorization from the Security Council should in all cases be sought for regional peace operations, recognizing that in some urgent situations that authorization may be sought after such operations have commenced.”).

156 See UN SCOR, 45th Sess., 2974th mtg., supra note 130 (statement of Liberia “recall[ing] that seven months ago we made efforts to have the Council seized with the deteriorating situation in Liberia, which efforts were not approved”); Kathleen Best, U.N. Moves to Halt Liberian Arms, St. Louis Post-Dispatch, Nov. 20, 1992, at A13 (“Western diplomats, including those in the United States, have pushed to limit U.N. involvement to non-military support … .”).

157 E.g., Franck, supra note 82, at 162 (describing the Liberia incident as a “purely humanitarian” operation and suggesting that it stands for the general proposition that regional organizations may “use force, even absent specific prior Security Council authorization, when that seemed the only way to respond to impending humanitarian disasters”); Levitt, supra note 153, at 375 (“ECOWAS has strongly contributed to spurring what appears to be new norms of customary international law … permitting unilateral humanitarian intervention by groups of states and regional actors in internal conflicts.”).

158 See supra note 65 and accompanying text.

159 Indeed, a few years later, the dominant view of NATO's unauthorized humanitarian intervention in Kosovo was that it was unlawful. See Roberts, Legality vs. Legitimacy, supra note 85, at 182 (“Many commentators ended up adopting the ambivalent position that NATO's use of force was formally illegal but morally justified.”); Stahn, Carsten, Enforcement of the Collective Will After Iraq, 97 AJIL 804, 814 (2003)CrossRefGoogle Scholar (“The prevailing opinion on Kosovo continues to maintain that the intervention was illegal under the Charter.”).

160 Text: Trump Statement on U.S. Military Strikes in Syria, Reuters (Apr. 6, 2017), at http://www.reuters.com/article/us-mideast-crisis-syria-text-idUSKBN1790A1.

161 See UN SCOR, 71st Sess., 7915th mtg., UN Doc. S/PV.7915 (Apr. 5, 2017); Richard Roth, Inside the Tense Closed-Door UN Security Council Deliberations on Syria, CNN (Apr. 10, 2017), at http://www.cnn.com/2017/04/10/politics/un-security-council-syria (reporting on the “‘exceptionally poisonous’” closed-door session that ended minutes before the U.S. strikes).

162 E.g., Russia Calls for Emergency U.N. Meeting After U.S. Strikes on Syria, Reuters (Apr. 7, 2017), at http://www.reuters.com/article/us-mideast-crisis-syria-russia-un-idUSKBN179112 (“Russia wants an emergency meeting of the U.N. Security Council to discuss U.S. missile strikes on Syria, the Russian Foreign Ministry said … .”).

163 The UN under-secretary general for political affairs set the tone of the meeting by focusing on the military strife and use of chemical weapons in Syria. UN SCOR, 71st Sess., 7919th mtg., at 2–3, UN Doc. S/PV.7919 (Apr. 7, 2017).

164 Id. (Australia, France, Germany, Italy, Japan, Netherlands, New Zealand, Saudi Arabia, Turkey, United Kingdom, and Ukraine). Several others and the European Union supported it in alternative arenas. See Syria War: World Reaction to US Missile Attack, BBC News (Apr. 7, 2017), at http://www.bbc.com/news/world-us-canada-39526089 (reporting on support from Canada, Israel, Poland, and the European Union); U.S. Allies Show Support for Strikes on Syria, Reuters (Apr. 6, 2017), at http://www.reuters.com/article/us-mideast-crisis-syria-reaction-idUSKBN1790M4 (reporting on support from Qatar, UAE, Bahrain, and Kuwait, and ambivalence from Indonesia); Madison Park, Who's with the US on Syria Strike and Who Isn't, CNN (Apr. 8, 2017), at http://www.cnn.com/2017/04/07/world/syria-us-strike-world-reaction/index.html (reporting on support from Jordan and Spain, and opposition from North Korea).

165 UN SCOR, 71st Sess., 7919th mtg., supra note 163 (Egypt, China, Uruguay, Senegal, Kazakhstan, Sweden, and Ethiopia).

166 Id.

167 See Shane Darcy, Retaliation and Reprisal, in Oxford Handbook of the Use of Force, supra note 79, at 879, 880–83.

168 See Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996 ICJ Rep. 226, para. 46 (July 8); Draft Article on Responsibility of States for Internationally Wrongful Acts with Commentaries, in Int'l Law Comm'n, Rep. on the Work of Its Fifty-Third Session, [2001] 2 Y.B. Int'l L. Comm'n 30, Art. 50(1)(a) & Art. 50, cmt. (4), UN Doc. A/CN.4/SER.A/2001/Add.1 (Part 2); GA Res. 2625 (XXV) (Oct. 24, 1970).

169 See supra note 16 and accompanying text.

170 E.g., Corten, supra note 36, at 550 (asserting that the jus ad bellum “institute[s] a ‘law against war’ principle”); Steven R. Ratner, The Thin Justice of International Law 70 (2015) (“[I]t is also clear that states bind themselves through international rules in part in order to prevent war and solidify peace.”); Bothe, Michael, Terrorism and the Legality of Pre-emptive Force, 14 Eur. J. Int'l L. 227, 238 (2003)CrossRefGoogle Scholar (describing “international law as a restraint on the use of military force”); see also García-Salmones, Mónica, Walther Schücking and the Pacifist Traditions of International Law, 22 Eur. J. Int'l L. 755, 756 (2011)CrossRefGoogle Scholar (“[I]nternational law has always been permeated with ideals of peace, … [so] war always constitutes a failure.”).

171 For a further defense of this claim, see Hakimi, Monica, The Work of International Law, 58 Harv. Int'l L. Rev. 1, 4345 (2017)Google Scholar.

172 On the point that the jus ad bellum sometimes legitimizes force and makes it easier to implement, see Charlotte Peevers, The Politics of Justifying Force: The Suez Crisis, the Iraq War, and International Law (2013); Hurd, Ian, The Permissive Power of the Ban on War, 2 Eur. J. Int'l Sec. 1 (Aug. 9, 2016)Google Scholar, available at https://doi.org/10.1017/eis.2016.13.

173 Cf. Thomas Franck, The Power of Legitimacy Among Nations 24 (1990) (explaining that noncompliance undermines a norm's “compliance pull” and make it less likely that other states will comply).

174 Cf. UN Doc. S/2014/189 (Mar. 15, 2014) (vetoed Security Council resolution that, in the context of Russia's activities in the Crimea region of Ukraine, “reaffirm[ed] that no territorial acquisition resulting from the threat or use of force shall be recognized as legal”).

175 The exception is the Syria case. See supra Section IV.C.

176 Hakimi, Monica, Constructing an International Community, 111 AJIL 317 (2017)CrossRefGoogle Scholar.

177 For a similar argument in the context of the 2003 Iraq war, see Stahn, supra note 159, at 808.

178 Schauer, supra note 53, at 183.

179 See supra notes 121–125 and accompanying text.

180 See Hakimi & Cogan, supra note 8, at 288–89.

181 My analysis here builds on three earlier works: id.; Hakimi, Defensive Force Against Non-state Actors, supra note 104; and Monica Hakimi & Jacob Katz Cogan, A Role for the Security Council on Defensive Force?, EJIL:Talk! (Oct. 21, 2016), at https://www.ejiltalk.org/a-role-for-the-security-council-on-defensive-force.

182 See Peter Baker & Eric Schmitt, Many Missteps in Assessment of ISIS Threat, N.Y. Times (Sept. 29, 2014), at http://www.nytimes.com./2014/09/30/world/middleeast/obama-fault-is-shared-in-misjudging-of-isis-threat.html?_r=0.

183 Id.; Zeina Karam, A Look at the Islamic State Militants in Syria, Seattle Times (Aug. 28, 2014), at https://www.seattletimes.com/nation-world/a-look-at-the-islamic-state-militants-in-syria.

184 See David Brunnstrom & Randy Fabi, U.S. Seeks More from China, Asian Muslim Nations, on Islamic State, Reuters (Oct. 16, 2014), at http://www.reuters.com/article/us-mideast-crisis-kerry-asia/u-s-seeks-more-from-china-asian-muslim-nations-on-islamic-state-idUSKCN0I527920141016.

185 Eric Schmitt, Obstacles Limit Targets and Pace of Strikes on ISIS, N.Y. Times, Nov. 10, 2014, at A1.

186 Emma Rapport, Dutch Parliament Commits Soldiers, F-16s to Fight ISIS in Iraq, NL Times (Sept. 24, 2014), at www.nltimes.nl/2014/09/24/parliament-commits-troops-isis-fight (Neth.).

187 E.g., Permanent Rep. of the United States of America to the UN, Letter Dated Sept. 23, 2014 from the Permanent Rep. of the United States of America to the UN Addressed to the Secretary-General, UN Doc. S/2014/695 (Sept. 23, 2014); Permanent Rep. of Australia to the UN, Letter Dated Sept. 9, 2015 from the Permanent Rep. of Australia to the UN Addressed to the President of the Security Council, UN Doc. S/2015/693 (Sept. 9, 2015); Chargé d'affaires a.i. of the Permanent Mission of Canada to the UN, Letter Dated Mar. 31, 2015 from the Chargé d'affaires a.i. of the Permanent Mission of Canada to the UN Addressed to the President of the Security Council, UN Doc. S/2015/221 (Mar. 31, 2015); Chargé d'affaires a.i. of the Permanent Mission of Turkey to the UN, Letter Dated July 24, 2015 from the Chargé d'affaires a.i. of the Permanent Mission of Turkey to the UN Addressed to the President of the Security Council, UN Doc. S/2015/563 (July 24, 2015).

188 See, e.g., Chargé d'affaires a.i. of the Permanent Mission of Turkey to the UN, Identical Letters Dated June 27, 1996 from the Chargé d'affaires a.i. of the Permanent Mission of Turkey to the UN Addressed to the Secretary-General and to the President of the Security Council, UN Doc. S/1996/479 (July 2, 1996); UN SCOR, 36th Sess., 2292nd mtg., at para. 54, UN Doc. S/PV.2292 (July 17, 1981) (Israel).

189 For a recent survey of state positions on the unable or unwilling standard, see Brunnée, Jutta & Toope, Stephen J., Self-Defence Against Non-state Actors: Are Powerful States Willing but Unable to Change International Law?, 67 Int'l & Comp. L. Q. (forthcoming 2018)CrossRefGoogle Scholar.

190 E.g., Permanent Rep. of France to the UN, Identical Letters Dated Sept. 8, 2015 from the Permanent Rep. of France to the UN Addressed to the Secretary-General and the President of the Security Council, UN Doc. S/2015/745 (Sept. 9, 2015); Permanent Rep. of the United Kingdom of Great Britain and Northern Ireland to the UN, Letter Dated Sept. 7, 2015 from the Permanent Rep. of the United Kingdom of Great Britain and Northern Ireland to the UN Addressed to the President of the Security Council, UN Doc. S/2015/688 (Sept. 8, 2015); see also Flasch, Olivia, The Exercise of Self-Defence Against ISIL in Syria: New Insights on the Extraterritorial Use of Force Against Non-state Actors, 3 J. Use of Force & Int'l L. 37, 64 (2016)Google Scholar (reviewing the opaque statements of Arab states that supported or participated in the strikes).

191 SC Res. 2249, pmbl. (Nov. 20, 2015).

192 Id., para. 5.

193 Dapo Akande & Marko Milanović, The Constructive Ambiguity of the Security Council's ISIS Resolution, EJIL:Talk! (Nov. 21, 2015), at https://www.ejiltalk.org/the-constructive-ambiguity-of-the-security-councils-isis-resolution.

194 Id.; see also Gross, Oren, Unresolved Legal Questions Concerning Operation Inherent Resolve, 52 Tex. Int'l L.J. 221, 233 (2017)Google Scholar (“For those who reject the position that the resolution authorizes member States to use force against ISIS, the resolution is nothing more than a restatement that member states may only use force against ISIS if, and only if, they are already permitted to do so legally, namely in exercising their inherent right of individual or collective self-defense in accordance with Article 51 of the Charter.”); Ashley Deeks, Threading the Needle in Security Council Resolution 2249, Lawfare (Nov. 23, 2015), at https://www.lawfareblog.com/threading-needle-security-council-resolution-2249 (“If UNSCR 2249 does not authorize force, then why did the French pursue it? The answer is politics.”).

195 Permanent Rep. of Denmark to the UN, Letter Dated Jan. 11, 2016 from the Permanent Rep. of Denmark to the UN Addressed to the President of the Security Council, UN Doc. S/2016/34 (Jan. 13, 2016); Permanent Rep. of Norway to the UN, Letter Dated June 3, 2016 from the Permanent Rep. of Norway to the UN Addressed to the President of the Security Council, UN Doc. S/2016/513 (June 3, 2016); Permanent Rep. of Belgium to the UN, Letter Dated June 7, 2016 from the Permanent Representative of Belgium to the UN Addressed to the President of the Security Council, UN Doc. S/2016/523 (June 9, 2016); Chargé d'affaires a.i. of the Permanent Mission of the Netherlands to the UN, Letter Dated Feb. 10, 2016 from the Chargé d'affaires a.i. of the Permanent Mission of the Netherlands to the UN Addressed to the President of the Security Council, UN Doc. S/2016/132 (Feb. 10, 2016); Letter Dated Dec. 10, 2015 from the Chargé d'affaires a.i. of the Permanent Mission of Germany to the UN Addressed to the President of the Security Council, UN Doc. S/2015/946 (Dec. 10, 2015); see also Permanent Rep. of the United Kingdom of Great Britain and Northern Ireland to the UN, Letter Dated Dec. 3, 2015 from the Permanent Rep. of the United Kingdom of Great Britain and Northern Ireland to the UN Addressed to the President of the Security Council, UN Doc. S/2015/928 (Dec. 3, 2015) (invoking Resolution 2249 to supplement earlier notifications of the use of force in Syria); Flasch, supra note 190, at 62 (“Following the unanimous adoption of UNSC Resolution 2249 in late 2015, multiple states made remarks endorsing and approving the strong, unified message the resolution sent.”).

196 Hakimi & Cogan, supra note 8.