I. Introduction
In January 2021, as the Biden administration took office in the United States, many observers may have welcomed promises by top-level U.S. officials to treat “the climate crisis as the urgent national security threat it is.”Footnote 1 It is no doubt comforting to once again have a U.S. administration that purports to take climate change seriously, but what does it mean to count this as a “security” issue? Perhaps the label simply signifies the seriousness of the threat, acting as a call to reinvigorate environmental institutions, such as by rejoining the Paris Agreement or strengthening domestic regulatory authority over greenhouse gas.Footnote 2 But the label could have more sweeping implications. By invoking the concept of security in connection with climate change, the United States could be heading down a path toward invoking emergency powers, perhaps as a basis for imposing economic sanctions on polluting nations or private entities.Footnote 3 More darkly, this policy could signal a new trajectory for the use of force abroad, gesturing toward a future where “climate rogue states” are threatened with military force if they do not comply with environmental commitments.Footnote 4 Alternatively, we might hope that the label will draw attention to the differential impact of climate change on populations that already experience the greatest insecurity, thereby triggering calls for much more far-reaching reforms.Footnote 5 Each of these possibilities suggests a radically different trajectory for foreign policy and international law, yet each is contained within the notion of “security.”
The foregoing example concerns U.S. foreign policy, but similar questions are emerging worldwide and at all levels of the international system. Several states and the UN Security Council have acknowledged the security implications of the climate crisis.Footnote 6 And the fallout from the COVID-19 pandemic is triggering calls to strengthen “global health security,” a concept with a long history but ambiguous implications.Footnote 7 States are describing cyber-threats,Footnote 8 organized crime,Footnote 9 money laundering,Footnote 10 bribery,Footnote 11 human trafficking,Footnote 12 supply-chain issues,Footnote 13 foreign investment,Footnote 14 migration,Footnote 15 national culture,Footnote 16 and cultural propertyFootnote 17 as matters of security.Footnote 18 These developments are emerging alongside, rather than replacing, classical security narratives around great power competition.Footnote 19 At the same time, transnational actors continue to offer alternative visions of security that either complement or contest these state policies. For example, a “second-generation human security” model makes pressing demands for “a very big allocation of resources, a far-reaching reform of global security capabilities, and a transformation in power structures.”Footnote 20 And the transnational social movements declaring that Black Lives Matter point toward alternative imaginations of security based on divestment from policing and carceral systems, and investment in communities.Footnote 21 All of these developments carry a similar demand for security, as well as an ambiguity about how that demand will be met.
This Article develops and applies a novel typology for analyzing the descriptive and normative implications of security claims in international law and politics. Security, as others have noted, is a deeply indeterminate concept, whose power derives not only from its association with particular issues or threats, but from the way that it combines fundamental ambiguity with a sense of heightened urgency.Footnote 22
This Article builds on this insight, arguing that actors take divergent approaches to security by issuing different—often implicit—answers to two questions: first, do experts hold a privileged position in identifying and describing security issues, and, if so, which experts’ views are relevant; and, second, should security be protected through the extraordinary means we ordinarily associate with the national security state, such as emergency powers and secrecy? Traditional militarist approaches to national security offer affirmative answers to both questions: security is a matter of military, intelligence, or diplomatic expertise, and is pursued through military force, surveillance, and similar means.Footnote 23 But any combination of answers to these two questions is possible, and each leads to a radically different approach to security in world politics.Footnote 24 So treating climate change as a security issue might, for example, lead us to think of security policy in terms of scientific knowledge and risk analysis—putting regulators in the driver's seat and triggering, rather than circumventing, the ordinary rule-of-law requirements of the administrative state.Footnote 25
Differing approaches to security, this Article argues, thus reflect a deeper struggle over whose knowledge matters when constructing and responding to the most pressing threats.Footnote 26 By using these variables to construct a typology, we can begin to disentangle the various approaches to security in international law and politics, and assess the desirability of each.
The typology developed here has four implications for the study of security in international law, which are by turns descriptive, normative, critical, and methodological. First, descriptively, this approach shifts our understanding of the security challenge to international institutions from one of conceptual uncertainty to one of epistemic authority and empowerment.Footnote 27 Recent international legal scholarship on security tends to privilege the conceptual dimension, suggesting that the fundamental challenge facing international regimes today comes from the expanding meaning of security and from the resulting uncertainty and scope for disagreement.Footnote 28 From this perspective, it appears critical that the international community achieve consensus on the right conception of security—that is, the conception that focuses our attention on truly the most dangerous threats and prioritizes them appropriately.Footnote 29 Once we reach this consensus, we might hope, then we can decide whether and under what circumstances an issue like climate change is “really” a matter of security rather than simply a matter of politics or policy.Footnote 30
However, if security is understood as a continuing struggle over epistemic authority, this desire for conceptual clarity may undermine, rather than enhance, our ability to accurately describe what is happening with security. To continue with the example at the outset of this Article, the struggle over whether and how to define climate change as a security issue implicates actors with diverse agendas. For instance, militaries may embrace climate security for their own purposes and to shore up their own authority, while resisting more transformative efforts that would place environmental scientists and regulators, or affected populations, at the center of security policy.Footnote 31 Thus we might see militaries embracing the idea that climate change threatens national security, while taking steps to describe that threat narrowly in terms of military readiness and armed conflict.Footnote 32 Approaches that focus on associating climate change with certain security labels rather than others (such as human security versus national security) risk obscuring rather than illuminating the ways that these labels themselves have become a terrain of struggle among various actors and agendas.
Second, as a normative matter, the typology developed here provides an initial guide, though no easy answers, to thinking about where we should position ourselves in these struggles. Judgments about security in international law, this Article argues, are unavoidably context-dependent, strategic, and political. This fact is in large part a consequence of the structure of international law itself. When a demand for security is projected into international politics, it does not encounter a blank slate: the international legal order is to a great degree organized into functionally differentiated subsystems—from human rights to trade to environmental law—which embed diverging assumptions about whose knowledge matters most in formulating policy and about the appropriate logic of policymaking.Footnote 33 Security can be deployed to entrench those existing knowledge practices and defend against challengers, but the power of security today also lies in its capacity for disruption—its demand to substitute a wholly new set of routines. Thus, the same arguments on climate security that might be used to justify and expand the Security Council's power to authorize force against “climate rogue states”Footnote 34 could also be used to disrupt trade rules that treat climate regulations skeptically as potential restraints on trade.Footnote 35 This dynamic is why it is often difficult to formulate a single stance on how security claims should interact with international law: if security is a tool for entrenchment or disruption, then it matters whose knowledge we are privileging, and how. The typology developed here thus provides a vocabulary for assessing the desirability of any given demand for security.
Third, by understanding security in terms of knowledge and authority, this Article raises critical insights concerning the structure of international law and politics. A key theme here is the struggle by non-state actors—whether overpoliced communities, Indigenous groups, small-scale food producers, or communities living near the sites of extractive industry—to have their own knowledge about their security interests recognized and prioritized as authoritative.Footnote 36 Achieving such recognition is particularly difficult in an international legal system that has historically privileged diplomats acting through foreign offices, or, more recently, networked groups of trained and recognized experts on a wide array of global problems.Footnote 37 This disconnect raises critical insights regarding the extent to which international regimes—even those that claim to uphold a humanized vision of security—remain undemocratic, unresponsive, and inaccessible.Footnote 38
Fourth and finally, this Article makes a methodological contribution by suggesting closer engagement between international law and critical security studies. International legal scholarship is steeped in three decades of dialogue with international relations scholars,Footnote 39 and international lawyers have drawn from both realist and constructivist traditions in security studies to interrogate a range of regimes and practices, from the use of force to cybersecurity.Footnote 40 But this interdisciplinary conversation has largely ignored or bypassed a range of critical traditions in security studies,Footnote 41 which for the past three decades have “re-conceptualized what security meant and how it mattered.”Footnote 42 As late as 2015, it was possible to say that most mainstream international law scholarship approaches security “without much reflection on theory or method.”Footnote 43 As international law is once again concerned with security, cross-disciplinary collaboration has the potential to enrich both disciplines by connecting critical security studies’ extensive reflection on concepts, theories, and methods to international lawyers’ deep knowledge of institutional structure and concern for the allocation of power and authority.
The Article proceeds in five parts. Part II situates the problem of security in international law, showing how security makes conflicting demands to either expand or preempt international legal rules. Part III sets out the core descriptive argument, showing how these demands become the terrain for deeper struggles over whose knowledge is relevant to constructing and pursuing security interests. This analysis produces two variables relating to the role of knowledge in any theory of security—that is, the identification of security issues, and the logic by which those issues are addressed—which can be used to categorize and evaluate competing approaches to security. Part IV uses these two variables to construct this Article's core analytical contribution: a four-part typology of approaches to security-knowledge, grounded in current debates in policy, law, and theory. Part V turns to a set of case studies in international economic law, demonstrating how conflicts among these approaches help explain current dynamics in the securitization of economic law. Part VI concludes by drawing out the implications of this study for further research on international law and security.
It should be noted that this Article was substantively complete prior to Russia's invasion of Ukraine in February 2022. The Article does not, and cannot, purport to “make sense” of this unfolding tragedy. But the analysis developed here may offer a preliminary vocabulary for thinking about what we must prioritize—and whom we must empower—in whatever comes next.
II. International Law and Security Claims
Security makes claims on international law, though the precise nature and effect of such claims is almost always ambiguous. In itself, this does not distinguish security from any other deeply held value, but security occupies a special place in today's controversies owing to the intensity of security demands, and their problematic relationship with preexisting legal structures. National security, for example, is frequently described as law's vanishing point, wherein some or all of law's demands simply cannot be observed.Footnote 44 The UN Charter enables the Security Council to “decree” new law to preserve or restore international peace and security, described as the sine qua non of functioning government.Footnote 45 The concept of human security, though closely connected with human rights law and in principle less at odds with legality, is said to have the potential to “become a new organizing principle of international relations,” which reconceptualizes the territorial integrity of states and poses far-reaching challenges to international law.Footnote 46 At a high level of generality, a commonality among all these approaches is a sense of intensity or paramount importance, coupled with a demand to do something with existing legal arrangements—either to extend them or overcome them. It is useful to take each of these dimensions in turn.
First, security is a generative condition in international law—a basis for the creation or extension of legal institutions. This generative dimension is perhaps as old as the interstate system itself, providing a rationale to cede authority to a sovereign.Footnote 47 As Anthony Anghie points out, this conception of sovereignty emerged in the context of European colonization of the Americas, in a process that involved construing Spanish incursions as benign and Indian resistance as hostile, thereby justifying a limitless war of conquest.Footnote 48 As colonial control deepened, the purported needs of security justified the further extension of sovereign power through borders, blacklisting, detention, dispossession, and warfare.Footnote 49 Following decolonization, many former colonies traded formal sovereignty for more attenuated forms of control, either via economic sanctions or continued military presence justified on security grounds, or via state contracts and economic treaties meant to establish the security of foreign investments.Footnote 50 International legal doctrines sometimes facilitate these forms of indirect control, and at other times exert counter-pressure.Footnote 51
Security's generative capacity is also at the foundation of efforts to create and expand the powers of international institutions. Security lies at the foundation of the modern UN system,Footnote 52 and the UN Security Council enjoys broad authority under the Charter to take binding action for the maintenance or restoration of international peace and security.Footnote 53 This authority is a holy grail for actors seeking to leverage the international system because it enables the Council to take measures up to and including the use of force, which have legally binding effects and supersede all other inconsistent obligations.Footnote 54 Where the Security Council fails to act, regional organizations, such as the Economic Community of West African States, may become sites of innovation in this area.Footnote 55 Security can also provide a policy basis for creating new institutions and endowing them with exceptional powers, such as the power to arrest and prosecute individuals for crimes under international law,Footnote 56 or the power to take emergency action in the name of global health security.Footnote 57 And international human rights law affords a right to security that could support alternative security practices.Footnote 58
Second, “security” can signify the limited reach of a specific international agreement or a set of international rules.Footnote 59 Some international rules may include express carve-outs for matters of security, for situations of emergency, or for other extreme circumstances.Footnote 60 Many human rights treaties, for example, permit states to derogate from obligations in time of national emergency,Footnote 61 and some treaties contain additional carve-outs for security measures,Footnote 62 or impose potentially countervailing individual duties with respect to security.Footnote 63 Exceptions and carve-outs also appear in treaties relating to trade and investment,Footnote 64 digital commerce,Footnote 65 regional integration,Footnote 66 transnational law enforcement,Footnote 67 global health,Footnote 68 the law of the sea,Footnote 69 and nuclear non-proliferation.Footnote 70 At the drafting stage, states may attempt to design rules that provide flexibility for security measures,Footnote 71 and this perceived need for flexibility may generate demands for deference or for restrictive interpretations of the rules even when there is no express exception available.Footnote 72 The customary international law of state responsibility, while it does not speak of “security,” permits states to deviate from their obligations for reasons of self-defense, force majeure, or because doing so is necessary to safeguard an “essential interest.”Footnote 73 The concept of security may also limit international agreements in more indirect ways, such as by constraining the powers of international organizations.Footnote 74
These twin pressures to expand and contract cause considerable anxiety among international lawyers. The possibility that states will escape treaty obligations via security exceptions, or expand the powers of the Security Council by reference to new kinds of security interests, is frequently treated as a threat to the international rule of law.Footnote 75 Considerable scholarly effort is spent on how to limit the potential breadth of security exceptions through conditions on access, standards of judicial review, time limitations, or compensation requirements, either as a matter of treaty design or interpretation.Footnote 76 The extension of Security Council power into generally applicable legislation related to counter-terrorism activities revived an analogous set of inquiries into what, if any, inherent limits exist on the Council's Chapter VII power.Footnote 77
These important legal arguments should not be expected to settle longstanding questions regarding the relationship between international legality and security interests. To be sure, treaties can attempt to offer more precise definitions of security, and treaty drafters and interpreters can impose procedural conditions on the concept's invocation.Footnote 78 But there are strong reasons to think that at least some ambiguity around security is a feature, not a bug, of legal orders.Footnote 79 The label “security” is thus likely to remain a site of struggle, both within and among states, over who gets to define the most important interests of the day and how those interests are to be pursued.Footnote 80 The deliberate open-endedness of security at the conceptual level will also make it difficult to resolve these struggles definitively through interpretation, by, for example, settling on a single “ordinary meaning” of the term in legal disputes.Footnote 81 As the rules of treaty interpretation are themselves open-ended,Footnote 82 conflicts over interpretation often replicate, rather than resolve, political conflicts over who gets to define security and how.Footnote 83
III. Security-Knowledge in International Law and Politics
The foregoing discussion showed how the concept of security places demanding but ambiguous claims on international legal institutions. As a legal matter, such claims will have to be resolved by states, courts, and other actors according to the ordinary rules that apply to the interpretation and performance of international legal obligations.Footnote 84 Nevertheless, as discussed above, bona fide conflicts over interpretation may simply reproduce the ambiguity of security in any given institutional setting, and interpreters will have to make choices among multiple available meanings—choices that have political implications. This Part advances the Article's descriptive argument that, by cashing out and resolving security claims in this way, international legal institutions are implicated in deeper struggles over the role of knowledge in security policy.
International legal institutions shape understandings about whose knowledge matters in security policy in two ways. First, their decisions generate expectations about the kinds of expertise that are qualified to identify security issues. Second, legal instruments and decisions shape understandings about the sort of logic we expect security policy to follow. While this is not the only way to think about the stakes of security decisions, the focus on epistemic authority is particularly helpful for reflecting on the consequences of legal decisions and is likely to lead to more incisive analysis than focusing on competing conceptual definitions of security.
A. Knowledge of Security: Competence to Identify Security Threats
International law implicates the knowledge practices of security, in the first instance, by shaping expectations about whose knowledge matters for identifying potential threats. To illustrate how such choices work, consider President Trump's declaration that migration was creating a national emergency and security crisis at the southern border, which was made to obtain funding for a border wall, justify extreme and inhumane conditions of confinement, and threaten tariffs against Mexico.Footnote 85 If this measure were addressed in an international institution—such as a human rights treaty body or (if tariffs had been imposed) under a trade treaty—what kind of reasons should the United States have been expected to give in support of its policies?
It turns out there are several answers to this question, all of which appeared in the discourse reacting to the administration's policies. One response is to draw on the knowledge of former government officials and other experts in “national security and homeland security issues” to evaluate the supposed threat.Footnote 86 Perhaps this focus on national and homeland security is skewed, and we should instead center expertise on immigration itself.Footnote 87 Maybe our search for specialized knowledge about the supposed crisis should cause us to privilege the knowledge of those specially affected, such as the migrants themselves and local communities.Footnote 88 Or maybe this whole search is silly—any one of us can know that there is no emergency at the southern border, as surely as we can know the meaning of the word itself.Footnote 89 The debate over U.S. immigration policy involved references to all of these forms of expertise.
This slippage—from privileging national security experts to embracing wider forms of expertise—represents a spectrum of approaches to identifying security threats (see Figure 1 below). At one end of this spectrum is a relatively narrow cadre of elite security experts, often composed of current and former government officials who “have held the highest security clearances, and . . . participated in the highest levels of policy deliberations.”Footnote 90 A broader approach would emphasize a wider range of scientific expertise, such as public health or climate science. At the other end of the spectrum is a view that privileges knowledge of laypeople, on the ground that affected persons are best-placed to know their own security interests. Through various interpretive moves, legal institutions likewise shape expectations about which type of knowledge is most valuable in identifying and constructing security interests. The following examples demonstrate these dynamics at work in international law and politics.
Figure 1. Identifying Security Threats
1. Maintaining Closure: Drug Trafficking and Critical Infrastructure
In many regimes, there may be a strong preference against broadening expertise about security, and not merely to protect the special status of supposed security experts. A particularly salient example comes from the international human rights system's confrontations with the securitization of drug trafficking.Footnote 91 Some human rights treaties provide that states may deviate from specific obligations for “compelling reasons of national security,” such as the obligation to permit aliens subject to expulsion proceedings to challenge that expulsion.Footnote 92 One question concerning the expulsion of aliens is whether a state can rely on the security carve-out in cases where the alien in question is accused of drug trafficking.Footnote 93 Today, drug trafficking is widely considered a national security issue in policy statements and international instruments.Footnote 94 Thus, when a treaty uses a term like “national security,” it is difficult to argue that the ordinary meaning of this term per se excludes anti-trafficking efforts.Footnote 95 Whether this security interest ought to be recognized as a legal matter is another issue, and something courts and treaty bodies have been understandably reluctant to embrace.Footnote 96
The European Court of Human Rights (ECtHR) confronted this question in C.G. and Others v. Bulgaria.Footnote 97 In that case, the applicant, a Turkish national who had resided in Bulgaria since 1992, was in 2005 summarily expelled without an opportunity to contest or obtain a lawyer on the ground he posed “a serious threat to national security.”Footnote 98 The government's evidence, provided to a reviewing Bulgarian court in a secret file, purportedly showed that the applicant was “member of a criminal gang dealing in illicit narcotic drugs,” though no specific facts were alleged in the public record.Footnote 99 The ECtHR's analysis focused on the Bulgarian courts’ application of the 1998 Aliens Act, which required the expulsion of aliens whose presence in the country “creates a serious threat to national security or public order.”Footnote 100 In this respect, the Court said:
It can hardly be said, on any reasonable definition of the term, that the acts alleged against the first applicant—as grave as they may be, regard being had to the devastating effects drugs have on people's lives—were capable of impinging on the national security of Bulgaria . . . .Footnote 101
While this statement appears at first glance to limit the conceptual scope of national security in a way that excludes drug trafficking, a deeper look at the case reveals that the decision's focus is instead on what kind of story the state was expected to tell. Indeed, the Court in this case expressly avoided any effort to interpret the term “national security” under Article 8(2) of the European Convention.Footnote 102 Instead, the Court focused all of its analysis on whether the expulsion had been “in accordance with law,” which the ECtHR has previously interpreted to require adherence to basic rule-of-law values like clarity and non-arbitrariness.Footnote 103 In other words, the Court was concerned not with whether drug trafficking fell within national security in the abstract, but with the fact-specific question of whether, in the context of this case, the applicant would have been on notice that the acts alleged by the state could trigger the national security provision in Bulgaria's 1998 Alien Act.Footnote 104 In this respect, the ECtHR emphasized that “the only fact” serving as a basis for the state's national security assessment was C.G.'s alleged involvement with a drug trafficking ring, and this showing fell short of the “specific facts” needed to show that the applicant presented a national security risk.Footnote 105
The judgment does not say exactly what “specific facts” Bulgaria should have been expected to show, but it does offer some clues. First, it will not be sufficient to establish that the person subject to expulsion proceedings broke the law: “run-of-the-mill criminal activities” are distinguished from those that threaten national security.Footnote 106 Second, it is not going to be enough that drugs or drug trafficking have “devastating effects . . . on people's lives.”Footnote 107 The state, in other words, could lard its expulsion orders with reams of testimony by law enforcement, public health professionals, and social scientists attesting to the illegality, dangers, and social impacts of drug trafficking, and it would likely make little difference. The Court indicated, however, that it might afford more deference where there are “serious potential consequences for the safety of the community.”Footnote 108 While the Court does not expressly say so, a state in Bulgaria's position would likely have been on much stronger legal footing if it had connected C.G.'s alleged drug trafficking activity to traditional military narratives about state security by adducing facts and testimony about impacts on regional stability, ties to non-state armed groups, or to international terrorism.Footnote 109 Although the ECtHR itself muddies the waters by talking about the “natural meaning” or “reasonable definition[s]” of security, this reading suggests the decision is less about whether drug trafficking is “in” or “out” of the security box than it is about how drug trafficking should be made to fit inside that box and, thus, who is sufficiently knowledgeable to make that fit.
None of these moves is particularly controversial in this context, and the point of this exercise is not to undermine the C.G. judgment. The criminalization of drugs is itself a challenge to human rights, and the securitization of the same is more dangerous.Footnote 110 On the contrary, a key point of this investigation is to show that, while opening security to a wider range of inputs and expertise may seem desirable, doing so has variable consequences in a fragmented and multilevel system, and many times closure is to be preferred. Still, the decision effectively leaves the door open to the securitization of drug trafficking by military or internal security officials, and in that way reinforces traditional militarized notions of security.Footnote 111
There are instances, however, where acts of epistemic closure are more troubling, or at least more contentious. In 2013 and 2014, two investors in an Indian multimedia company brought separate cases alleging that India violated its obligations under bilateral investment treaties by causing its state-owned enterprise to annul a contract for a satellite telecommunications spectrum.Footnote 112 For some years, the Indian armed forces had expressed concern that that the excessive lease of this particular spectrum—the “S-Band”—to commercial operators would impede national defense operations.Footnote 113 As an interagency process reviewed and ultimately annulled the satellite contract, the purported need for S-Band was broadened to include not just military and internal security needs, but also other “strategic” and “societal” needs, including railways, disaster management, tele-education, telehealth, and rural communication.Footnote 114 The ultimate decision annulling the contract cited both military and these broader strategic and societal needs.Footnote 115 In defense of its decision, India invoked the “essential security” exceptions in the relevant investment treaties.Footnote 116 Both cases were thus confronted with the question of whether and under what circumstances the protection of critical infrastructure can give rise to concerns under security clauses in investment treaties—an increasingly salient and thorny issue.Footnote 117
The tribunals’ decisions avoided ruling definitively on this question, while seeming to establish a strong presumption against including civilian infrastructure under the treaties’ security provisions. Both tribunals accepted in principle that military and internal security needs could be covered by the treaty provision, and both agreed that the other, more traditionally civilian uses were not covered.Footnote 118 The Deutsche Telekom tribunal, in particular, relied on the above-quoted language from C.G. v. Bulgaria, suggesting that to include issues such as disaster management, rural communication, railways, and tele-health would “distort[ ] the natural meaning” of the term “essential security interests.”Footnote 119 The tribunal added that the security clause could not be interpreted to include just any public interest, lest it swallow the treaty's requirement that states parties pay compensation for takings in the “public interest.”Footnote 120 The tribunal thus determined that the interests at stake must “go to the core . . . of state security” and “protect something of higher value than any ‘public interest’”—a function that it apparently associated with the military and national defense.Footnote 121 Indeed, both decisions are notably solicitous of the military and internal security agencies’ definitions of their own security goals.Footnote 122
The awards thus leave unclear the relationship between investment treaties and state laws designed to protect critical infrastructure and dual-use technology.Footnote 123 There are at least two possible readings of the decisions going forward. First, if the tribunals’ seemingly reflexive exclusion of civilian technology and infrastructure from the scope of “essential security interests” is read with maximum force, then the protection of those features could be entirely excluded from such clauses unless the relevant treaty is amended.Footnote 124 This would spell trouble for the many states that rely on similarly worded clauses to screen and control investments in a wide range of civilian projects deemed to be security-sensitive.Footnote 125 The second option is to focus on the lack of evidentiary support given by India to establish the security-sensitivity of the non-military interests, and to use these cases to suggest that such interests are not per se excluded but are subject to a higher burden of justification than the military's concerns.Footnote 126 Again, as in C.G. v. Bulgaria, the best strategy for triggering the treaties’ security clauses would likely be to show how the protection of such interests goes to the “core” of state security—likely by deploying military or military-adjacent expertise.Footnote 127 A consequence is that the best way to obtain greater flexibility internationally is to increase the authority of the military and defense establishment over industrial policy domestically.Footnote 128 In both cases, the integrity of the treaty regime is preserved at the cost of reaffirming the primacy of military and defense officials in defining what issues have a “higher value than [just] any public interest.”Footnote 129
2. The Struggle for Wider Expertise: Nuclear Weapons
Questions about the proper scope of security-knowledge do not just appear at the margins, as in drug trafficking or infrastructure cases, but plague even the assumed “core” instances of military and defense matters. Consider the question of nuclear weapons and deterrence—perhaps the sine qua non of twentieth century national and international security policy.
The dominant mode for framing nuclear weapons since 1945 has been in adversarial terms, focusing on the threat posed by the actors in possession of those weapons, their incentives and motivations.Footnote 130 As Emmanuel Adler documents, American policy on nuclear deterrence in the 1960s came increasingly under the influence of an “epistemic community” of civilian strategists and nuclear scientists, along with their partners in government, who advocated for arms control.Footnote 131 These experts’ ideas about nuclear policy, premised on concepts drawn from game theory,Footnote 132 shaped U.S. national security policy and international legal regimes for decades.Footnote 133 As understandings about nuclear policy shifted to a greater concern with proliferation and “rogue” actors,Footnote 134 the shape of the legal regime shifted, including with a turn to multilateral sanctions under Security Council authority.Footnote 135 These approaches represented new sets of cultural understandings, and opportunities for new forms of knowledge-creation.Footnote 136
There is, however, another way to describe the danger posed by nuclear weapons, and that is to focus on the explosion itself, rather than the government wielding it. By this logic, it is the very existence of nuclear weapons, rather than the possibility of surprise attack or the supposedly unsteady hand of a rogue state, that is destabilizing.Footnote 137 An accidental detonation of a nuclear device, for example, poses an equivalent danger to the launch of a nuclear weapon.Footnote 138 The relative absence of such concerns in academic and policy circles, as compared with the focus on deterrence and proliferation, suggests a privileging of certain forms of expertise and knowledge-creation, and opens the possibility that other forms of knowledge could be applied to identify and frame the nuclear security threat differently.
This question—who has the competence to identify security threats—emerged as a critical issue in the Legality of the Use by a State of Nuclear Weapons in Armed Conflict case. In that case, the World Health Assembly (WHA) requested an advisory opinion from the International Court of Justice (ICJ) on whether the use of nuclear weapons in armed conflict would violate obligations under “international law, including the [World Health Organization (WHO)] Constitution.”Footnote 139 The WHA is the plenary body of the World Health Organization, which has as its objective “the attainment by all peoples of the highest possible level of health.”Footnote 140 Member states’ delegations “should be chosen from among persons most qualified by their technical competence in the field of health, preferably representing the national health administration.”Footnote 141 The WHA resolution requesting an advisory opinion accordingly stressed the connection between health and the use of nuclear weapons, suggesting these fell within the WHO's competence.Footnote 142 The resolution specifically borrows language from public health to discuss the nuclear weapons threat, stating that “primary prevention is the only appropriate means to deal with the health and environmental effects of the use of nuclear weapons,” and that this required clarification of their legal status.Footnote 143
This request was immediately controversial.Footnote 144 The ICJ in 1995 decided it could not address the substance of the request, finding that the legality of nuclear weapons was not “within the scope” of the WHO's activities.Footnote 145 The Court took particular aim at the “primary prevention” rationale, finding that, while the WHO might assist in averting nuclear war by disseminating information on its adverse health effects, the “political steps by which this threat can be removed” were clearly above the organization's pay grade.Footnote 146 While the WHO has “wide” responsibilities in “sphere of public ‘health,’” there was “no doubt that questions concerning the use of force, the regulation of armaments and disarmament are within the competence of the United Nations and lie outside that of the specialized agencies.”Footnote 147
Judge Weeramantry, in dissent, criticized the Court's conclusion that the “WHO's function is confined to health, pure and simple, and it strays into unauthorized fields when it enters the area of peace and security.”Footnote 148 His dissent emphasizes the overlap between matters of health and international security, and can be read as a strong statement of the broad approach to security that was emerging in international politics at the time.Footnote 149 “I cannot subscribe,” Judge Weeramantry writes, to the view that the preeminent body in global health should sit idly by “for the technical reason that it would be trespassing upon the exclusive preserve of the Security Council, who are the sole custodians of peace and security.”Footnote 150
The divergence between the Court majority and its dissenters thus reveals yet another dimension to this well-known opinion.Footnote 151 There was no doubt that a great deal of legal and diplomatic practice existed with respect to the proliferation and use of nuclear weapons.Footnote 152 The underlying policy question in the WHA case was whether an organization, whose delegations are composed of health ministers and whose secretariat is staffed by public health experts, should have anything to say about how to meet the defining security threat of the century. The Court answered that it did not, suggesting that the WHO's role was limited to addressing the health-related fallout of nuclear war. But an alternative approach was possible, and the WHO's embrace of a concept and legal framework for “global health security” in the 2000s suggests that times may have already changed.Footnote 153
3. The Lay Public: “New Wars” and Food Crises
In many international legal fora where security claims play out, there are only limited opportunities for those most directly affected to define their own security interests. States are ordinarily represented internationally by their executive branches, which have incentives to define security interests in their own terms, sometimes to the exclusion of other domestic authorities. While international parliamentary institutions, consisting of either directly elected representatives or delegates from national legislatures, are relatively common in today's international organizations, these institutions generally lack significant power to shape policy or bind either member states or international organizations themselves.Footnote 154 The representation of non-specialist, non-state interests in global government, to the extent it exists, is most often accomplished through various mechanisms for NGO, civil society, or “multi-stakeholder” participation.Footnote 155 Nevertheless, there are some areas where international legal regimes have enabled lay publics to struggle over and potentially define their own security interests.
First, Christine Chinkin and Mary Kaldor note some halting progress in international law and practice toward prioritizing local knowledge and “local ownership” in the formation of peace agreements.Footnote 156 Historically, peacebuilding had been a top-down affair, which can privilege the perspective of governments, interested third states, and transnational experts to the exclusion of local communities.Footnote 157 As Séverine Autesserre writes, these top-down projects often replicated colonial dynamics, even as the international experts involved in the peace-building process would have emphatically denied doing so.Footnote 158 The early twenty-first century, however, has witnessed a “local turn” in the scholarship and practice of peacebuilding.Footnote 159 This is reflected in several key normative statements at the United Nations, which emphasize the formal or informal participation of civil society groups, and particularly women, in the peacemaking process.Footnote 160 Implementation of the local turn is, in many observers’ views, spotty at best, but Chinkin and Kaldor do note some promising precedents in the 2004–2005 peace process in Darfur and the 2003 Liberian peace deal, where civil society organizations and women's groups played a formal role in the peace process.Footnote 161 These precedents echo struggles in other areas, such as disaster response, where there is both a normative imperative and a series of practical and political obstacles to centering the role of local communities in restoring peace and security following a period of crisis.Footnote 162
A second site for innovation and lay participation can be found in the area of food security.Footnote 163 While “food security” has its own logic and conceptual framework, it also interacts in complex ways with other national, global, and human security discourses.Footnote 164 The Committee on World Food Security, as extensively reformed in 2009 following the global food price spike that began two years earlier, has emerged as a “unique space,” which enables those most directly affected by food insecurity to “have direct influence in global policy coordination.”Footnote 165 Pursuant to the 2009 reforms, civil society organizations and NGOs were empowered to establish their own mechanism for facilitating consultation and participation in the body;Footnote 166 most strikingly, as Michael Fakhri observes, this new “Civil Society Mechanism” has privileged the voices of social movements and peasants’ groups over international NGOs.Footnote 167 While the Committee's place, and civil society's role, in global food policy is contested and far from secure, there appears to be broad agreement that the participation of civil society organizations “served to expand debate, introduce new perspectives and therefore shift the direction of global food security policy.”Footnote 168 The Committee's engagement with civil society thus serves as a “benchmark” for an alternative vision of security-knowledge in global politics.Footnote 169
The foregoing examples, while specialized and partial, suggest alternative mechanisms for producing knowledge about security. These fora are not closely tied to national security establishments, but neither are they embedded in the wider claims to expertise that characterize most other international organizations. Here, instead, is an approach to security that, as in Rana's counterintuitive reading of Hobbes, “is fundamentally egalitarian and thoroughly rejects any distinction between elite and ordinary rationality.”Footnote 170 Notably, however, in international legal settings these sources of lay knowledge must address themselves to, and reconcile themselves with, the still-dominant discourses of international security, diplomacy, and technocratic expertise.Footnote 171 This tension underscores the extent to which security in international politics remains the province of elites, even if the sources of elite knowledge can sometimes be expanded to embrace a wider range of disciplines and expertise from time to time.
B. Knowledge in Security: The Logic of Security Policy
Once a security issue is identified, the question turns to how expert or lay knowledge will be deployed to address it. This question, which concerns the logic of security policy, is a central concern, particularly in judicialized international systems today.Footnote 172 At one end are doctrines of “public reason”—devices like proportionality, reason-giving, means-ends rationality, and publicness, which force states to justify any policies that burden human rights, trade, or property.Footnote 173 At the other extreme is the idea that legal institutions, whether national courts or international tribunals, have no place second-guessing a national executive's action to protect the country's security interests. Rather than belonging to the realm of reviewable and intelligible public reason, such judgments are, in words recently resurrected by the U.S. Supreme Court, “‘delicate, complex, and involve large elements of prophecy.’”Footnote 174 This tension, between public reason and prophesy, is well-known in the study of security exceptions and international law.Footnote 175 But these doctrines also have implications for whose knowledge matters in security policy, which have gone undertheorized.Footnote 176
One of the major, field-spanning developments in post-Cold War international law has been to subject state policymaking in various domains to requirements of rationality and reason-giving.Footnote 177 International legal regimes from trade to human rights have developed robust requirements of consistency, ends-means proportionality, scientific rationality, and publicity in their review of national decision making.Footnote 178 For example, when a state adopts a public-health measure that negatively affects foreign investors—such restrictions on marketing for tobacco products—a tribunal will ask whether and to what degree the measure is rationally related to the pursuit of its health objective.Footnote 179 Even where a treaty appears to preclude this kind of inquiry by reserving substantial discretion to the state, courts, tribunals, and other bodies might still review the procedure by which those aims are addressed.Footnote 180 As the level of required scrutiny increases, these forms of process and rationality review shift power from elected officials and political appointees, and toward the scientists, experts, and lawyers that staff government bureaucracies.Footnote 181
Security decision making, the traditional view goes, resists the demands of public reason and tends toward prophecy. Security is said to demand a degree of “decision, activity, secrecy, and dispatch” that is inconsistent with the broad, open-ended debates of ordinary politics.Footnote 182 Security decision making is also resistant to the demands of scientific and procedural rationality: it is recognized as being ad hoc, subject to demands of expediency, and not necessarily amenable to the requirements of reasoned consistency and publicness that attend ordinary government regulation.Footnote 183 Expertise is still required—indeed, contemporary security policy is replete with recognized and self-appointed experts clamoring to be heard on everything from migration to climate change.Footnote 184 But on one standard account, security expertise is an amalgam of judgments about what is prudent, expedient, or possible, and in that respect it is not easily compared to the kind of expert knowledge that is imagined to support ordinary, science-based regulation. It was this conception of security, for example, that appeared to animate the U.S. Supreme Court's 2018 decision upholding a travel ban against claims of anti-Muslim bias, reasoning that any rule “‘that would inhibit the flexibility’ of the President ‘to respond to changing world conditions should be adopted only with the greatest caution,’ and our inquiry into matters of entry and national security is highly constrained.”Footnote 185
In international law, the tension between expertise and prophecy can be framed, but not fully resolved, by the text of the relevant legal instruments. Some treaties, for example, may offer guidance as to how tightly a particular policy must be related to an articulated security objective, and some also indicate the level of deference to be afforded by a reviewing court.Footnote 186 Still, these texts do not resolve the question, and under each of these regimes security demands still face competing pressure toward public reason and prophecy. Even where the law does not explicitly demand deference at all, institutions often feel pressure to implicitly soften their standards of review when it comes to security measures.Footnote 187 Standards like “necessary,” or even the “only way” requirement of customary international law, can be ratcheted up or down through legal interpretation.Footnote 188 Where treaty law appears to grant wide discretion, interpreters have relied on customary international law,Footnote 189 general principles of law,Footnote 190 jus cogens norms,Footnote 191 or the law of the reviewing tribunalFootnote 192 to ratchet up the level of scrutiny.
These varying degrees of scrutiny have implications for whose knowledge matters in security policy and how that knowledge is deployed. The capitulatory approach, where legal institutions abstain in the face of security measures, amplifies the significant power that executive agencies enjoy in the national security state. Stricter forms of review, such as “hard look” review or strict necessity and proportionality, seek the “progressive submission of power to reason.”Footnote 193 This view shifts power from political to legal, scientific, and technocratic forms of expertise, promoting evidence-based policy but raising questions about the suitability of having technocrats choose between competing values.Footnote 194 A middle approach—sometimes referred to as a “suitability” or “rational basis” test—is limited to ensuring a relationship exists between the policy goal and the measure pursued.Footnote 195 This approach expands the space in which legislators and political appointees can work, giving them broad scope to define and pursue policies, while ensuring that policies pursue publicly defined goals subject to political accountability.Footnote 196 The implications of these approaches are depicted below in Figure 2.
Figure 2. Responding to Security Threats
By framing these techniques in terms of epistemic authority, we can see how different approaches might be valued in different circumstances. In some circumstances, such as C.G. v. Bulgaria, it may make sense to push toward rationalization in a way that empowers lawyers to tailor security policies to minimize the burden on civil liberties.Footnote 197 But, more generally, it is open to question whether and to what extent international law should empower bureaucratic experts over political actors. In many cases, it may make sense to ratchet review downward to give full scope for political debate, struggle, and compromise—thus empowering lay forms of knowledge—even if the results cannot be fully justified on rationalist terms.Footnote 198 Even minimal demands for rationality can be troubling when they are used to entrench elite forms of knowledge or outdated notions of security against possible contenders whose politics may appear unruly. For example, some tribunals have shown an unexamined distrust of “politically motivated” policies, treating these as non-rational.Footnote 199 In such circumstances, a turn toward strong, even self-judging exceptions may be a useful means to disrupt established routines and press for transformative change.Footnote 200
IV. Four Approaches to Security, Knowledge, and International Law
The foregoing analysis indicates why, when international lawyers speak about “security,” we are often talking past each other. To take the example from the outset of this Article the characterization of climate change as a security threat potentially suggests a wide range of meanings and approaches to security, including the elevation of science and environmental expertise, or the encroachment of the military into climate policy.Footnote 201 Though potentially clarifying, it does not fully answer the question to suggest that climate policy is a human or international security issue as opposed to a national one, because these vocabularies can also be associated with a range of approaches from community empowerment to military humanitarianism.Footnote 202 A more robust discussion of the merits and demerits of securitizing climate change, or any other issue, demands that we take apart the constituent parts of security itself, to see how they might be scrambled and reassembled. The previous part made the case that the power of security lies to a large degree in the way in which it vindicates competing claims to epistemic authority. This Part uses that insight to build a framework for understanding and analyzing security claims and their impact on international law.
To that end, this Part sets out a four-part typology of approaches to security. These types are developed from actual historical and social practices of security but are designed to be sufficiently abstracted to enable comparisons and analysis across particular situations.Footnote 203 Such a typology thus exists at some remove from the actual, far messier reality of history, and few actual instances of security are likely to easily conform to any one particular type. It is hoped that whatever is lost in terms of strict fidelity to history will be gained by enabling comparative analysis across time, space, and regime, but the usefulness of this classification can in the end “only be judged by its results in promoting systematic analysis.”Footnote 204 Note that this typology is not meant to provide either a comprehensive or exhaustive discussion of all possible approaches to security,Footnote 205 nor are the citations here meant to firmly associate the scholar cited with the ideal-typical frame being put forth.
Figure 3, above, introduces each of the types developed in this Part, and compares their views of how knowledge about security is developed and put into practice. The top row refers to approaches that rely on some form of specialized knowledge to identify security threats: for realist security, as I define it, this means primarily military knowledge; for widened security, threat identification is open to a wider range of expertise. The bottom row of Figure 3 contains theories that embrace, at least in principle, a despecialized approach to framing security issues: discursive security, drawing on the Copenhagen School of security studies, accepts in principle that anyone can attempt to securitize an issue, though in practice political elites are more readily able to do so; pluralist security expects that threats will be constructed on the basis of shared and emergent identities, rather than according to some expert logic. Realist and discursive security, in the left-hand column, are united in their view that security is fundamentally associated with extraordinary measures, such as emergency power, secrecy, or violence.Footnote 206 Widened and pluralist security are distinguished, on the righthand side, as having no necessary connection to extraordinary measures, though these are not strictly ruled out either.
Figure 3. Four Views of Security-Knowledge
Each of these types is designed to be primarily descriptive—a way to identify, categorize, and analyze existing statements about and practices of security—but these types have both normative and forward-looking implications. First, by suggesting how security policy is likely to unfold, each type is likely to provoke in us a normative intuition about how widely the security frame should be deployed. For example, if security policy empowers military officials to classify the world and pursue threats through force and emergency powers, then many of us would likely want its deployment to be strictly limited.Footnote 207 Second, as the following discussion emphasizes, each type is inherently dynamic, in the sense that it contains elements that, if rigorously applied, open doors to the alternative approaches to security reflected in the remaining types.Footnote 208 The four types described here are thus not stable equilibria but rather more like quantum states, in which each type contains the potential for the others. This dynamism is reflected in actual practice,Footnote 209 and underscores the extent to which our understanding of the grammar and logic of security can be only a first step toward formulating a normative position in any concrete case.
A. Realist Security
The first approach identified here identifies security primarily with the use or non-use of military force. The term “realism” is used here to reflect the affinity with Kenneth Waltz's argument that security, defined as state survival, “is the highest end” among the infinitely varied world of possible national interests.Footnote 210 Survival, moreover, is in the (neo-)realist tradition frequently associated with use, threat, and non-use of military force.Footnote 211 On this view, security is used primarily to signify the defense of states against destruction or destabilization by force, either coming from outside or within.Footnote 212 This close connection between force, statehood, and security remains influential on interdisciplinary scholarship in international law and politics.Footnote 213
The focus on the use of force here centers military knowledge, alongside shifting sets of civilian expertise that support the security establishment, in the definition of security threats and issues. The U.S. security establishment, as it developed during and immediately after the Cold War, offers an example of this approach to security knowledge. Prior to World War II in the United States, military affairs were largely the province of professionals, and civilian involvement in strategy and military planning were discouraged.Footnote 214 This changed with the emergence of the concept of “national security” during the war, which, as Dexter Fergie points out, was from the beginning bound up with civilian expertise—exemplified by the “hordes of social scientists who contributed their expertise to the war effort.”Footnote 215 The emergence of nuclear deterrence as a primary security concern during the Cold War led for a time to the ascendance of game theory and quasi-economic modeling as an adjunct to security policy.Footnote 216 As U.S. security concerns shifted in the 1990s to concerns about internal strife and “resource wars,” new forms of expertise were incorporated into the security and defense establishment.Footnote 217 A similar shift happened again after 2001, as U.S. policy refocused on international terrorism and state-building.Footnote 218 In each of these examples, a broad range of expertise is attached to the security and defense establishment for the purpose of aiding the military and intelligence agencies in identifying and framing potential security threats. The military and security establishment thus remains at the center of the process, with other forms of knowledge performing a critical but adjunct role.
When it comes to the logic of decision making, the default mode for realist security is raison d’état. On this view, requirements of legality, publicity, and rationality may yet be appropriate for a wide range of issue areas or circumstances. But where (national) security is threatened—as in an emergency—courts, legislators, and other actors have “no real choice but to hand the reins to the executive and hope for the best.”Footnote 219 The typical legal controls for non-arbitrariness, reason-giving, and factfinding will be suspended, or dialed so far down as to be meaningless.Footnote 220 This is because the national security bureaucracy, headed by the executive, is alone thought to possess the secret knowledge, capacity for efficient action, or political legitimacy to address urgent and existential threats.Footnote 221 The connection between this rationalist view of international affairs and the treatment of security policy as ad hoc, improvisational, and not amenable to review or publicity is historically contingent rather than logically necessary.Footnote 222 Indeed, while the justifications for such deference are often premised on the expertise of the national security establishment, this flexibility also creates substantial space for the influence of professional norms, ideology, bias, and racial or religious animus.Footnote 223 Nevertheless, this view of security logic remained popular among those who came to theorize security after the attacks of September 11, 2001.Footnote 224
We can see the union of these two variables—military-centered knowledge and raison d’état—in arguments that claim wide latitude under international law for national security policies. For example, in a recent case brought by Iran under a 1955 commercial and consular treaty, the United States defended its reimposition of economic sanctions against Iran on the ground that the sanctions were necessary to protect its essential security interests.Footnote 225 The United States’ arguments in this pending case have emphasized the fact that this “core national security decision was made at the highest levels of the U.S. Government, following a National-Security-Council-led review of the United States’ policy toward Iran.”Footnote 226 The United States has further emphasized that the treaty's security exception “provides wide discretion for a State to evaluate and determine what measures are necessary to protect those essential interests,” and that its corresponding burden to justify those measures internationally is low.Footnote 227 This position reflects a view that “core national security tools” are subject to limited oversight by the economic treaties at issue, and affording maximum flexibility to the national security establishment is preferable to close judicial scrutiny.Footnote 228 As of this writing, the ICJ has yet to rule on the merits of this defense.Footnote 229
As this example suggests, those adopting something close to the ideal-typical vision of security knowledge—including various strands of realist thought and so-called traditional security studies—are likely to counsel modesty for both security and for law. As to the former, realist security recognizes that the pursuit and defense of security is bloody business not to be widely extended.Footnote 230 By limiting international security to “core” matters of military affairs, realism arguably can be deployed to reject the conceptual apparatus that has justified the expansion of military affairs into nearly every corner of domestic and international life.Footnote 231 In international law, this sense of modesty is reflected in calls to refocus international institutions like the Security Council on narrower conceptions of “international peace and security” that are centered on interstate conflict.Footnote 232 Such approaches are suspicious of the mobilization of military force for humanitarian intervention, regime change, or democracy promotion, emphasizing the folly of these military adventures and their cost in terms of human life and security.Footnote 233 This caution, however, should not be confused with a radical commitment to peace.Footnote 234
Equally, we should expect modesty when it comes to what law might accomplish with respect to “real” security interests.Footnote 235 Military force will sometimes appear necessary, and it is unlikely that law will do much to constrain that possibility.Footnote 236 We should therefore recognize that security—in its core, military-focused sense—is likely to work as an implied, de facto limitation on the effectiveness of international legal regimes.Footnote 237 Realism may also be congenial to narrowly framed treaty-based exceptions that generally embed states in a relatively thick set of international rules vis-à-vis allies, but offer states a free hand to pursue security competition with adversaries.Footnote 238 At the same time, realist security would be skeptical of attempts, whether by neoconservatives or human security advocates, to expand the notion of security to encompass issues like human rights violations, and potentially to justify military responses thereto.Footnote 239 Realist thinking, however, offers fewer critical remarks about the expansion of security concepts to extend hegemonic power through international organizations, such as the use of Security Council authority to extend the reach and effect of U.S. sanctions, with minimal legal oversight.Footnote 240
This ideal type, however, contains internal tensions, which tend to drive it toward any one of the other three types discussed below. Even in its most military-focused sense, security as an academic discipline and political practice has always been inherently interdisciplinary, inviting other forms of expertise to perform adjunct roles.Footnote 241 This interdisciplinarity raises the possibility that those other forms of civilian and political expertise might become dominant, shifting the relevant knowledge and logic of security away from military affairs or from exceptionalism.Footnote 242 More radically, as Rana points out, realism's exceptionalism suggests that, in truth, “no science or expertise of security exists,” pointing the way to a radically democratic approach that supplants experts with the lay public.Footnote 243 These possibilities are explored in the following Sections.
B. Widened Security
Set against realist security are various approaches that seek to expand security into new realms and decenter the role of military affairs. Many of these approaches find their roots in an intellectual and policy-oriented push in the 1980s, which accelerated after the end of the Cold War, to “broaden” and “deepen” the concept of security.Footnote 244 A key theoretical development in this period was the rise of feminist security studies, which, among many insights, showed how threats to women's security can come from their own states and state security forces, as well as from non-military threats such as inadequate access to health care or birth control.Footnote 245 These approaches scored some political successes, including the adoption at national, regional, and international levels of documents relating to human security,Footnote 246 global health security,Footnote 247 and women, peace, and security,Footnote 248 among others. At the same time, states reformulated their national security policies over the first two decades of the twentieth century to increasingly address non-military threats, though not always with results that human security advocates would have endorsed.Footnote 249 The “broadening and deepening” debates have produced a complex field with a wide range of approaches to security, which cannot be reduced to a single conception or ideal type.Footnote 250 Each of the remaining three types discussed here owes a debt to those debates.
One type that emerges from these debates, which has gained traction in mainstream security circles and even in some state policies, is referred to here as widened security. This view has sought to dislodge military affairs from their central role in security policy, and argues that today's most deadly threats, such as climate change and pandemics, do not come from any foreign or domestic adversary,Footnote 251 and instead are made intelligible through the application of scientific expertise.Footnote 252 Nevertheless, the widened-security view, as described here, remains reliant on state-based structures to ensure security, and is likely to count diplomacy, lawmaking, and regulation high among its list of tools.Footnote 253 This distinguishes the approaches described here from those that focus on “deepening” security by going beyond the emphasis on states in the international system.Footnote 254
This widening view finds echoes in the Obama administration's national security strategies,Footnote 255 and has enjoyed a revival of mainstream interest since the COVID-19 pandemic. For example, at the height of the pandemic, Oona Hathaway wrote:
[W]e should broaden the lens of national security to think about all serious global threats to human life. Terrorism should be a part of the conversation, but it should be considered next to other, more pressing threats to American lives, including pandemics, other public health threats, and climate change. The assessment of threats should be based on scientific assessments of real global threats that require serious global solutions. That's what “national security” must mean in the post-COVID-19 world.Footnote 256
While this comment explicitly refers only to knowledge of security threats, its assertion that threat assessment must be based on scientific evidence also has implications for the logic of security policy. By yoking threat-assessment to the full range of available scientific expertise, this approach narrows if not eliminates the distance between security policy and the ordinary rationality requirements of the administrative state.Footnote 257 There is also the possibility that a more rationalized and reviewable security policy could emerge even in more “classical” domains of war-fighting and counter-terrorism.Footnote 258 Perhaps, for example, the securitization of the environment has promoted alternative “security practices based on prevention, risk management and resilience,” while challenging the close association between security, emergency, and exception.Footnote 259 Likewise, the close association between human security-style approaches and human rights law suggests a preference for proportionality and rule of law values over exceptionalism.Footnote 260
Despite its limited institutional successes, this type, too, is unstable. First, widened security can collapse back into forms of realism if these new security interests like climate and health become militarized.Footnote 261 For example, Diane Otto has argued that the “women, peace, and security” frame has in a sense become a victim of its own success, as it has “become captive to the militarized security frame” that typically characterizes Security Council action.Footnote 262 Second, as actors turn to exercising extraordinary legal powers to address these new security threats, widened security's claim to preserve a place for legality and rationality in security policy becomes increasingly tenuous.Footnote 263 Here, climate change is a relevant example: where it was once possible to say that the climate policy was still “designed and developed in the realm of ordinary policy debate,”Footnote 264 this assumption arguably no longer holds as actors are increasingly considering the use of emergency powers to pursue climate policy.Footnote 265 Finally, the increasing pluralization of expertise in determining security threats suggests, consistent with many “deepening” approaches to security, that lay knowledge rather than expertise does, and should, drive security policy.Footnote 266 The latter two possibilities are taken up below.
C. Discursive Security
In contrast to each of the above frames, which privilege expertise and specialized knowledge for identifying security threats, an alternative view emphasizes the irreducibly political character of such determinations. One such view is readily captured for our purposes by what is known as securitization theory.Footnote 267 This approach, developed by the so-called Copenhagen School of Security Studies, emphasizes the fundamental and irreducibly political character of claims to know something about security.Footnote 268 Security, on this view, is not necessarily associated with any particular threat (e.g., military invasion), or any particular object (e.g., securing the state).Footnote 269 Rather, security is a discursive practice wherein some actor identifies something as an existential threat, and argues that the threat requires an extraordinary, and perhaps extra-legal, response.Footnote 270 As these moves gain acceptance, an issue is “securitized,” meaning it is transferred to the realm of extraordinary measures and left outside ordinary politics.Footnote 271 This association with exceptionalism makes the theory inherently wary of security, arguing that “desecuritization is the optimal long-range option,” because it moves issues “out of th[e] threat-defense sequence and into the ordinary public sphere.”Footnote 272
When framed this way, the insights of securitization theory are compatible with a set of intuitions about security-knowledge that are likely to be held by many international lawyers.Footnote 273 First, as noted above, security is an indeterminate concept, whose meaning and content at any given time is socially and politically constructed.Footnote 274 On this view, we can acknowledge the privileged position that elites—such as politicians, bureaucrats, officials, lobbyists, pressure groups, and, increasingly, scientists—hold over the definition of security issues, just as the aforementioned types do.Footnote 275 But at the same time we can recognize that the moment of securitization itself is not determined by that expertise, and instead reflects a move that belongs to politics and narrative.Footnote 276 Second, despite being radically open to redefinition, lawyers might intuitively agree that security has a troubling historical connection to emergency powers and exceptionalism.Footnote 277 Third, this link to exceptionalism suggests a normative orientation that, like many public lawyers, is skeptical of security claims, and prefers the long-term desecuritization of issues.Footnote 278 The likelihood that security claims will be fused with demands for secrecy, emergency power, extra-legalism, military force, etc., places a thumb on the scale against such claims, even if the desirability of using the security label can never be fully resolved in the abstract.Footnote 279 Together, these insights are broadly compatible with a liberal-constitutionalist spirit that seeks to limit the abuse of security claims and provide offramps back to normal politics, even if that liberal spirit was not the intention of the theory's progenitors.Footnote 280
This set of insights is broadly compatible with the ways that international lawyers have addressed derogations from internationally recognized human rights. Many human rights treaties include derogations provisions allowing states to “suspend certain rights during emergencies while subjecting such measures to international notification and monitoring.”Footnote 281 The substantive scope of what might constitute an “emergency” is open ended, and in practice these provisions have been applied to war, insurrection, terrorism, economic crises, natural disasters, and COVID-19.Footnote 282 Once this threshold is reached, a public emergency can justify suspensions of rights otherwise ordinarily guaranteed in a liberal democratic society, subject to reporting and monitoring requirements, as well as substantive requirements of proportionality.Footnote 283 By encouraging states to channel their exercises of exceptional power through the derogations system's procedural framework, human rights treaties force officials to make “public assertions about the nature of the crisis and the scope and duration of emergency restrictions,” which can provide benchmarks for public pressure if those are later exceeded.Footnote 284
This dynamic, though it specifically relates to emergencies, reflects more generally both the perils and hopes of the discursive security frame.Footnote 285 Under the derogations regime, a wide (though not unlimited) range of issues can be “securitized” through declarations of public emergency. Such declarations provide a rationale for extraordinary power in the form of rights suspensions. From the perspective of many human rights lawyers, the role of law is to help desecuritize—to end the exception and restore an ordinary, rights-respecting politics.Footnote 286 From this perspective, the derogations regime fails insofar as it allows “permanent” emergencies that normalize such exceptional measures, and it succeeds insofar as it enables opposition to those regimes.Footnote 287
Discursive security's focus on exceptionalism, however, raises difficult questions that pull toward our final ideal type. The drama of exception and emergency, for instance, arguably ignores the extent to which security practices are bureaucratized and routinized in national, transnational, international, and other institutions.Footnote 288 The consequent normative preference for desecuritization, moreover, raises questions about what “ordinary” politics looks like, and how such a politics exists apart from security practices.Footnote 289 The derogations example above offers a helpful illustration: while liberal-legalist critiques of the derogations regime seek technical fixes that would limit emergencies and facilitate the return to “normal” politics, more radical critiques seek to undermine this normal/abnormal dichotomy and show how racialized and oppressive security practices are entrenched in, and are even constitutive of, “ordinary” politics in modern liberal states.Footnote 290 Notably, in 2020, Alison Howell and Melanie Richter-Montpetit published an article titled Is Securitization Theory Racist?—a question which they answered in the affirmative.Footnote 291 The article's publication was met with a trenchant defense from securitization theory's architects, and the range of disputes between the authors goes well beyond what can be covered here.Footnote 292 Most relevant for the present purposes is Howell and Richter-Montpetit's observation that a normative orientation against securitization appears to privilege the status quo and leaves unexamined “the racial violence of normal (liberal) politics.”Footnote 293 Implicit in this critique is a concern that the security claims of “racialized, indigenous, and poor communities”—whose enslavement, displacement, dispossession, and detention helped construct existing political ordersFootnote 294—would be reflexively dismissed out of a fear of exceptionalism and a desire to maintain the established order.
This argument, regardless of how well it identifies problems internal to an academic discipline like securitization theory, is useful in delineating a fourth and final approach to security-knowledge. Each of the three approaches thus far could be invoked, in many contexts, to resist racist or oppressive invocations of security. Even so, each does so from its own perspective of security-knowledge, either that of military affairs (realist security), wider expertise (widened security), or from the perspective of a set of knowledge practices grounded in respect for the rule of law and “ordinary” politics (discursive security). Howell and Richter-Montpetit's intervention points toward the conclusion an alternative frame is needed if the security interests of the colonized, marginalized, racialized, and subaltern are to be taken seriously on their own terms. The final Section attempts to sketch such a frame.
D. Pluralist Security
A fourth approach to the production of security-knowledge would strip away the last of the necessary connections, presenting security as fundamentally political and open to contestation. The previous view, informed by securitization theory, already did this with respect the identification of security threats and issues, treating the construction of security issues as a matter of politics rather than expertise. This final approach likewise eschews the logic of exception, recognizing that, at the very least, “[t]he exclusionary and violent meanings that have been attached to security are themselves the result of social and historical processes, and can thus be changed.”Footnote 295 This view has a further consequence: once any necessary conceptual connection to emergency power and institutionalized expertise is sheared away, the state is radically decentered in the security analysis. Security is defined and executed politically, and we should thus expect to find it articulated in the first instance not by states themselves but by political groupings, which both fragment and extend beyond state boundaries.Footnote 296 Moreover, the institutions of the state may themselves be sources of insecurity, against which demands for security are made.Footnote 297 Security-knowledge is in this sense “pluralist”—it emerges from the claims of overlapping and contesting social groups, and is pursued via a range of exceptionalist and routine logics both inside and outside of the state.Footnote 298
Pluralism's descriptive purchase is illustrated by a recent study of private security companies in the Kenyan oil sector.Footnote 299 As Charis Enns, Nathan Andrews, and Andrew Grant explain, the discovery of oil in the Lake Albert Basin in 2006 led to an explosion of development of the oil sector in northern Kenya, which already faced “regional insecurity, a limited state presence, and a proliferation of small arms.”Footnote 300 The oil companies and local community both expressed an interest in securing the region, though they articulated the idea in vastly different terms.Footnote 301 The oil companies, for their part, involved a complex assemblage of actors that included the Kenya Police Reserves, a force of “untrained civilians armed by the government to provide communities with protection in the absence of a formal police presence,” who are the “main visible state-sanctioned security” in northern Kenya.Footnote 302 The reserve forces were established to deter livestock raids and intercommunal conflict, but their local knowledge and language expertise proved valuable to the oil companies interested in facility security.Footnote 303 Researchers found that this practice caused military-age men to leave their communities to work for the reserves, contributing to “new risks and vulnerabilities at home, such as . . . cattle raiding, banditry, and other forms of violence.”Footnote 304 As these stories, first articulated by community members, were taken up by international human rights groups, many companies discontinued their reliance on the reserves.Footnote 305 But young men “continue to be drawn towards oil exploration and drilling sites, . . . putting aside their [Reserve] uniforms and weapons in exchange for positions with private companies.”Footnote 306
This example is a useful portrait of security-knowledge production in a pluralist perspective. The two entities whose security is most visible in this picture are those of the foreign oil companies’ operations in northern Kenya, and that of the rural communities, often dozens of kilometers away, whose pre-existing security practices are further threatened by the oil operations. These two entities share an interest in regional security but articulate that interest differently. For the oil companies, security primarily refers to the stability of extractive operations, and for this purpose the companies relied on a complex assembly of state security forces, private security companies, risk analytic consultants, and subcontracted reserve police.Footnote 307 For the local communities, security was articulated in terms of community resilience, as well as the absence of theft and violence, and on that view the drilling activities themselves constituted a potential for further insecurity. The state here is clearly instrumentalized in the pursuit of both community and corporate security—a fact made readily visible by its sparse presence in the region and by the ability of the companies to subcontract the state's deputized police. And the case study shows how overlapping interests in security can at once be united, in the sense of desiring more regional stability, and conflictual.
Once we have elucidated these dynamics, we can see them everywhere, even in highly developed states with an overwhelming national security apparatus. For example, the Movement for Black Lives frames its demands in terms of ending “the war against Black people,” arguing effectively that in the United States the security of predominately white neighborhoods has been achieved by the overpolicing, militarization, surveillance, and incarceration of Black communities.Footnote 308 The state, in this view, can be redescribed as an entity that reliably puts its security forces in the service of protecting white lives and property, while rendering non-white communities insecure.Footnote 309
This picture of conflicting and overlapping security claims also has radical implications for the knowledge and logic of security issues. On this view, knowledge about security threats emerges from communities, rather than from any purportedly objective field of expertise. To carry on the above example, the Movement for Black Lives platform suggests that the knowledge needed to identify security threats, and the logic appropriate to addressing them, emerges from the affected communities themselves, rather than from technocratic discourses about law or policing.Footnote 310 Likewise, the security interests of the dominant communities are understood to be constructed not out of objective determinations informed by technical expertise, but from an anti-Black bias that is pervasive both in everyday social practice and in legal and political institutions.Footnote 311 In both cases, the ideas about appropriate knowledge practices emerge from the felt needs of the community making the security claim.
Security pluralism thus leads to a much more contingent, and even tactical, orientation toward exceptionalism.Footnote 312 The same orientation toward non-state centers of security knowledge may also suggest alternative, non-exceptionalist security logics.Footnote 313 Monica Bell, for example, suggests that a “sense of security” in Black communities can be multifaceted and not primarily about the presence of armed police, embracing “economic integration, racial integration, greening, increased community engagement, activities for youth, employment opportunities for youth, social activities for adults, and widely available healthy food.”Footnote 314
But exceptionalism is not always ruled out. Movement campaigns for divestment, abolition, and other “non-reformist reforms” are, in their own way, demands for exceptional measures that disrupt ordinary categories and legal routines.Footnote 315 In international politics, non-state entities may also take exceptional measures, including the use of force, to protect what are framed as security interests.Footnote 316 This view was reflected in the long-running efforts to extend international legal protection to non-state forces fighting “wars of national liberation”—a movement that ultimately led to the inclusion of such recognition in Additional Protocol I to the Geneva Conventions.Footnote 317 And smaller, post-colonial states have defended their sovereignty and their rights under security exceptions as a means of preventing the emergence of a “one-way international law which lacks mutuality in its observance and therefore becomes an instrument of oppression.”Footnote 318 In all, the portrait that emerges from a pluralist perspective is far more mixed with respect to the types of knowledge deployed to define security, and potentially strategic in its orientation to exceptionalism.
V. Security-Knowledge in International Economic Law
Once we have elucidated these divergent approaches to security, we can identify these ideal types in operation throughout international law. As actors within different legal regimes take up a particular approach, they shape expectations about who is entitled to speak about security and how security policy is expected to unfold.Footnote 319 International legal institutions, in this way, do not passively receive ideas about security that were generated elsewhere. Instead these institutions actively participate in reshaping how we think about the way security is defined and pursued. The following examples, drawn from current debates in international economic law, demonstrate how the four ideal types described above are deployed on the international stage in ways that empower certain actors, frame policy alternatives, and shape understandings of security itself. These examples also demonstrate how the above types can be used analytically to diagnose tensions in existing approaches to security, predict how one vision of security might lapse into another, and thus expand the range of alternatives.
International economic law provides a useful object of study because it exists today at the center of so many contests about the meaning of security. Since the nineteenth century, the ideal of “desecuritizing the international economic realm” has been central to the ideology of capitalism and to the project of constructing a liberal international economic order.Footnote 320 This project, of course, was never entirely complete, as desecuritized economic relations frequently depended on military and political alliances.Footnote 321 Today even that partial success seems to be eroding: the universalization of the liberal economic order—symbolized most dramatically by the accession of China and Russia to the WTO—has, ironically, contributed to resecuritizing economic relations within that system.Footnote 322 Alongside these great power rivalries, international trade and investment law, by virtue of their expansion, are now colliding with renewed debates over the securitization of other matters such as cybersecurity, migration, pandemic disease, and climate change.
A. Security Realism Under Pressure at the WTO
In the international trading system, there is no more visible marker of the boundary between security and economy than the General Agreement on Tariffs and Trade (GATT) security exception. Article XXI of the 1947 GATT recognizes the right of any state to take any measure “it considers necessary for the protection of its essential security interests” relating to nuclear materials, arms, military supplies, war, or “other emergency in international relations.”Footnote 323 Under the GATT system, this exception set up a boundary between ordinary liberal economic relations and matters of security, and the phrase “it considers” in the exception was often understood to render the provision “self-judging,” meaning that it was for states themselves to decide on the precise location of this boundary.Footnote 324 For almost seventy years, well into the era of the WTO, the meaning and scope of this exception, while sometimes hotly contested, was never tried before a dispute-settlement panel.Footnote 325
This picture suddenly changed in the late 2010s, as the WTO dispute settlement system saw a range of disputes that potentially turned on the security exception.Footnote 326 The most high-profile and dramatic of these involved complaints by several countries lodged against the Trump administration's tariffs on steel and aluminum, which had been imposed using statutory national-security authority and justified internationally by invoking Article XXI.Footnote 327 This case was important conceptually as well as politically, because the United States’ justification so brazenly blurred the line between economic industrial policy and national security that there was a felt need to draw that line ever more sharply. The first case to reach a decision, however, would concern a more traditional security issue: trade restrictions imposed by Russia as part of its ongoing conflict with Ukraine.Footnote 328 A second dispute, concerning intellectual-property measures imposed by Saudi Arabia as part of a broader diplomatic rupture with Qatar, would follow in short order.Footnote 329 In this context, these two panels faced the difficult task of resolving the relatively straightforward cases at hand, with the knowledge that their decisions would go on to shape outcomes in the more difficult cases to come.
The panels’ decisions, whether consciously or not, carefully deploy security realism to protect against the threat that more expansive approaches pose to the liberal economic order. Interpreting GATT Article XXI, the Russia–Transit panel explained that each state may define its own essential security interests, and that the state also retains the discretion to determine whether a measure is “necessary” for the protection of those interests.Footnote 330 This comes close to a fully exceptionalist approach to security, though the panel did impose some legal controls, including by requiring the invoking state to “articulate” its essential security interests “sufficiently enough to demonstrate their veracity.”Footnote 331 In so doing, the state is expected to defend its conception of security in terms of “defence or military interests, or maintenance of law and public order interests.”Footnote 332 These terms are flexible, and the decision leaves panels ample room to expand what counts as a security interest in future cases.Footnote 333 But the panel is expressly anchoring the concept of security in realist terms, defined as involving military affairs abroad and law enforcement domestically.Footnote 334
The resolution of these first two cases, under this framework, was mostly straightforward. The border restrictions in Russia–Transit were imposed vis-à-vis Ukraine during a period of deteriorating diplomatic relations and outright armed conflict, wherein the existence of a strong security interest on Russia's part was, in the panel's view, nearly self-evident, even if it was also self-serving.Footnote 335 In the subsequent Saudi Arabia–IP Rights case, the panel found that Saudi Arabia's articulated interest in “protecting itself ‘from the dangers of terrorism and extremism’” justified wide-ranging and draconian measures that were designed to “to end or prevent any direct or indirect interaction or contact between Saudi Arabian and Qatari nationals.”Footnote 336 Saudi Arabia could therefore take measures that had the effect of preventing a Qatar-based entertainment company from enforcing its intellectual property (IP) rights against a pirate broadcaster operating in the territory of Saudi Arabia.Footnote 337 These decisions reflect the basic shape of realist security: a wide scope for action within a zone defined by military and defense forms of knowledge-production, relating to border security, armed conflict, and terrorism.Footnote 338
The Saudi Arabia–IP panel, however, took a further step toward defining security's boundaries by identifying actions that fell outside the scope of the exception. In addition to preventing the Qatari company from using civil law to enforce its IP rights, Saudi Arabia's own authorities had failed to apply criminal penalties against the pirate broadcaster, which had been engaged in infringement on a commercial scale.Footnote 339 The panel struggled to understand how the failure to enforce criminal laws could be in Saudi Arabia's security interests, particularly when the pirate broadcasts also infringed the rights of companies from third-party countries.Footnote 340 In this respect, the panel cited favorably Brazil's third-party statement, which questioned whether “any country's essential security interests” could be protected by failing to prosecute a commercial-scale copyright pirate.Footnote 341 In the panel's view, Saudi Arabia could have commenced the prosecutions without undermining its stated security goal of avoiding contact with Qatari officials and nationals.Footnote 342
Together, the panel decisions reflect both the promise and perils of realist security. The panels’ adoption of this frame for security is, given the context, almost self-evidently an effort to ring-fence the concept against the background of the Trump administration's combative assertion that “economic security is national security.”Footnote 343 By policing the boundaries of security, but imposing only minimal plausibility requirements on its logic, the panels have crafted an approach that prevents the security exception from swallowing trade law whole by limiting it to situations relating to armed conflict, breakdowns in diplomatic relations, and terrorism. At the same time, the realist approach largely avoids putting panels in the unenviable position of having to second-guess the security judgments of major players in the trading system.Footnote 344
This is nevertheless a relatively traditional approach to security, with all the drawbacks that may suggest. The panels’ approach privileges actors who are capable of coding persons and ideas—such as terrorism, border instability, and “extremism”Footnote 345—as threats to national defense and public order. And this frame can validate even extreme controls on the bodies and activities of the persons so coded. The idea of security presented here is also one-sided: the Saudi Arabia–IP panel decision, for example, was unable to process even the possibility that a state's security interests might be served by refraining from criminal prosecutions.
These aspects of the rulings place the panels’ realist security frame under continuing pressure. The COVID-19 pandemic and the continuing reality of climate change are likely to cause actors to seize on the open-textured elements of the treaty provisions and panel decisions, to push toward a wider view of security in international trade.Footnote 346 This, in turn, is generating pressure to demand a more rationalized logic of security in the trading system, rendering security matters subject to increasing levels of scrutiny and narrowing the distance between security and exceptions for other types of public-policy measures.Footnote 347 The Saudi Araba–IP panel judgment, for its part, raises the interesting possibility that the rights of IP holders are themselves being securitized through requirements of criminalization and international legal limits on prosecutorial discretion, and that these interests are being set against the security claims of the Saudi authorities.Footnote 348
B. “Global Health Security” as Widened Security
In contrast to the GATT security exception, the 2005 International Health Regulations offer a paradigmatic example of widened security in practice at the international level. Substantially overhauled following the World Health Organization's success in combatting SARS in 2003,Footnote 349 the Regulations are a binding legal instrument that provide, in one commentator's words, “the only international rules governing global health security.”Footnote 350
Although it may appear strange to treat this regime in a discussion of economic law, the Regulations are in many ways a trade treaty, even if not usually discussed as such.Footnote 351 The Regulations’ objective is not to enable a forceful and overwhelming response to disease outbreaks, but rather a disciplined and rationalized one, which avoids “unnecessary interference with international traffic and trade.”Footnote 352 The Regulations thus provide a unique framework for “global health security” that imagines a rationalized and proportionate response to health emergencies, preserving as far as possible the status quo of liberalized commerce.
The conception of security at work here is reflected most powerfully in the Regulations’ framework for emergency powers.Footnote 353 The Regulations empower the WHO director-general to issue an emergency declarations on the basis of expert advice and in accord with scientific evidence and principles.Footnote 354 Such an emergency declaration, as in analogous constitutional systems, enables the WHO director-general to exercise special powers, but with a widened-security twist. That is, rather than opening up a zone of exception, WHO emergency declarations are arguably designed to create a legalized and rationalized space, in which all states’ responses to a pandemic are based on available science, rationally related and proportional to appropriate objectives, and no more restrictive on travel and trade than necessary.Footnote 355 In this way, a “public health emergency of international concern,” rather than being a normless space, is designed to be a norm-governed and rationalized one, in which cool-headed scientific expertise prevails over panic.Footnote 356
As is well-known following the COVID-19 pandemic, these ideals do not work well in practice.Footnote 357 In particular, there is a widespread belief, with some supporting anecdotal evidence,Footnote 358 that the very issuance of WHO emergency declarations spurs states to take extraordinary measures to restrict travel and trade, against WHO advice and in contrast with the Regulations’ objectives.Footnote 359 Following a 2018 outbreak of Ebola in the Democratic Republic of the Congo, the WHO appeared to recognize that emergency declarations might have this perverse effect on developing countries and initially refrained from exercising its emergency authority, finding that doing so would produce “no added benefit.”Footnote 360 This approach, which may have represented an attempt to desecuritize health emergencies, produced an outcry from public health experts, and the WHO eventually relented.Footnote 361 Even so, many experts today see the benefit in de-escalation, and are considering formally or informally dialing back the WHO's emergency system, replacing it with a graded system of “alerts” that lacks the same dramatic flair.Footnote 362
These controversies over the Regulations suggest that the widened-security vision remains unstable. The perverse effects on trade and travel suggest the discursive power of security, whereby WHO emergency declarations lure authorities into an “emergency trap” that justifies increasingly restrictive measures.Footnote 363 Alternatively, current proposals to de-escalate the WHO's emergency powers could reflect a realist view that global health security both mislabels the threat and overloads the WHO, suggesting the need to return to a more lower-profile, technocratic form of health governance.Footnote 364 Finally, the concern about perverse effects in the DRC case suggests a pluralist view: the existing system protects some states while destabilizing others, compounding the economic and political hardships caused by outbreaks.Footnote 365 In this context, a de-escalated WHO framework could reflect an attempt to accommodate all security interests at stake.Footnote 366
C. Speaking “Security” at the Trade-Climate Nexus
Climate change, today, poses one of the most pressing policy and security challenges for international economic institutions. The extent to which liberalized trade and investment can both contribute to, and help alleviate, climate change is a matter of extensive debate, leading to a wave of thinking on how to redesign economic law to promote sustainability and environmental justice.Footnote 367 In the meantime, some policy tools to combat climate change, such as border carbon adjustments (BCAs) designed to address the carbon leakage that global supply chains enable, are potentially in tension with existing WTO rules on non-discrimination.Footnote 368 Environmental advocates and WTO critics frequently point to the fact that environmental measures tend to fare poorly in trade disputes, though defenders argue that the system shows substantial flexibility.Footnote 369 Regardless, the lack of legal certainty around climate measures is leading some actors to seek tools that afford a wider scope for action, and in that context “climate security” provides a helpful frame.
One such proposal suggests that the U.S. executive use extraordinary statutory authority over national security to impose BCAs.Footnote 370 This proposal, raised by Tim Meyer and Todd Tucker, argues that the U.S. president has authority under Section 232 of the Trade Expansion Act to impose “security-motivated restrictions on carbon-intensive imports,” on the ground that “climate change poses a threat to national security.”Footnote 371 Meyer and Tucker argue that Section 232, as applied by the executive and interpreted by U.S. courts, gives the president discretion to regulate commerce “based on an expansive conception of security,” with “virtually no administrative guardrails” in terms of process controls.Footnote 372 Elsewhere, Meyer and Tucker argue that a properly designed climate-security BCA should be justifiable under the WTO's environmental exception (Article XX(g)), obviating any need to invoke GATT Article XXI, but that the best reading of the text and jurisprudence suggests that the security exception is also available as a backstop.Footnote 373 Reading both pieces together, the proposal uses security for a kind of climate brinksmanship on at least two levels: the dispute settlement mechanism is pressured to resolve the legality of BCAs under Article XX(g) in order to avoid opening the floodgates under Article XXI, while the persistence of these tariffs under domestic national security authority “can motivate allies and competitors alike to come to the table” to achieve lasting agreement on trade-related climate measures like BCAs.Footnote 374
This proposal is an almost paradigmatic example of the discursive approach to security at work.Footnote 375 Both the U.S. statute and the GATT security provision are “exceptionalist” in the terms described here, in that they are capable of giving rise to expansive visions of security and are accompanied by only limited procedural controls.Footnote 376 Meyer and Tucker are not unaware of the dangers posed by exceptional power. Rather, they appear to share the goal of desecuritization as the “optimal long-range option,”Footnote 377 presenting this use of extraordinary power not as an end in itself, but as a means to re-striking the balance between trade liberalization and climate protection. Securitizing climate change, in this way, might be characterized as a necessary but hopefully temporary disruption, oriented toward “saving the political consensus in favor of free trade”Footnote 378 in a time of global warming. It arguably reflects an attempt to adjust the liberal economic order, not to fundamentally challenge it.
This approach, not surprisingly, is contestable in ways that evoke the other three security frames discussed above. In the United States, proponents of liberalized trade have argued for transferring competence for national-security investigations to the Department of Defense,Footnote 379 thereby restoring the realist emphasis on military knowledge.Footnote 380 Another view is not to deny the urgency of climate change, but rather to argue that it should be addressed through more ordinary logics of administrative action and multilateral cooperation.Footnote 381 Finally, from a critical perspective, the imposition of tariffs appears as a fundamentally nationalistic solution, which reifies the very national borders that enhance insecurity and impede solidarity in the face of a truly global threat.Footnote 382
D. Investment Treaties and Pluralist Security
The fourth—pluralist—approach to security illuminates rarely discussed dynamics in international law, and specifically in investment law. While there is an extensive literature on the role of security in investor protection, this literature is almost solely focused on the scope and effect of security exceptions in investment treaties, and the related issue of screening investments for national security risks.Footnote 383 A pluralist frame, however, reveals how investment treaties prioritize the security of foreign investors in ways that potentially conflict with state security interests, and how the security interests of rural, Indigenous, or other marginalized communities are easily rendered invisible in the context of investor-state disputes.
Investment law can be understood as a regime of security for foreign investors.Footnote 384 In particular, investment treaties frequently guarantee investors “full protection and security,” which is understood, at a minimum, to require that the state provide an appropriate level of police protection for investors and their property.Footnote 385 This means that investors whose property is damaged during protests, strikes, demonstrations, or riots can bring claims against the state, arguing that the police were too restrained and failed to prevent property damage.Footnote 386 This aspect of the regime is rarely controversial today and is widely accepted by both claimants and respondent states in arbitral proceedings. Nonetheless, the suggestion that investors enjoy a special right to demand police activity—beyond what is afforded under some states’ domestic lawsFootnote 387—is striking at a time when many are critiquing, and actively seeking to pare back, the role of police in protecting property ownership and in destabilizing communities.Footnote 388
The security interests of investors can also clash with, and sometimes prevail over, the felt security needs of the state and its armed forces.Footnote 389 This clash is exemplified by the first-ever investment treaty arbitration, AAPL v. Sri Lanka.Footnote 390 In this case, a Hong Kong-based investor sought to recover for damages sustained by its prawn farm during a counter-insurgency operation by government security forces, which took place during a “civil war between Tamil separatists and the Sri Lankan Government.”Footnote 391 The tribunal closely scrutinized the government's decision to retake the farm—located in an area that was “practically out of the Government's control”—and decided that its decision to proceed with the operation without first trying more peaceful means violated its obligation to accord the investor full protection and security.Footnote 392 The tribunal's decision, as James Gathii notes, effects in many ways the opposite of GATT Article XXI: rather than providing states with “safe harbor” for their national security measures, the AAPL case suggests that a state's decision to deal forcibly with a threat of internal rebellion is reviewable by a tribunal of investment lawyers if the state's action “conflicted with its treaty obligations to protect [the investor's] commercial rights.”Footnote 393
The AAPL case also reveals how other relevant security interests are effectively erased in the legal conflict between the investor and the state's security apparatus. The dispute between AAPL and Sri Lanka took place against the backdrop of a civil war, in which both government forces and rebel groups were accused of large-scale human rights abuses.Footnote 394 The experiences of the communities struggling for security in this context lack legal relevance under the investment treaty, and thus do not appear in the award or in much of the related commentary.Footnote 395 Decades later, it remains the case that, to have their security claims fully heard in the context of investor-state disputes, affected communities often must have their narratives seized upon by one of the parties.Footnote 396
In this way, international investment law validates the competence of specific actors—investors and state security forces—to speak about security concerns while obscuring others. The destabilizing effects of foreign investment on local communities are frequently obscured, and, even if raised in the context of a proceeding, are difficult to recognize under the applicable law.Footnote 397 Even ongoing reform efforts, which are said to range from reformist to radical, rarely contemplate mechanisms to enable such communities to shape understandings about the relationship between investment and security.Footnote 398 The failure to recognize the relationship between foreign investment and the insecurity of affected communities, in favor of privileging investor and (to a lesser extent) state security, remains what Gathii calls the “dark underbelly” of the international investment system,Footnote 399 and the pluralist perspective highlights the urgency of much more far-reaching reform.
Security pluralism thus points the way for a radically different approach to investment law, which goes beyond full security for investors and essential security for host states. A vital long-term project is to consider how investment law might be reformed to ensure the security of local communities, at every stage of an investment project, and enable them to speak for and define their own essential interests.Footnote 400 In the meantime, it is important to recognize that the security interests of local and Indigenous communities are often raised and recognized in investment disputes, if at all, through the arguments of the host state, after the communities have managed to obtain the government's attention.Footnote 401 It will thus be necessary to build in and enhance mechanisms for those security claims to be heard and given legal weight, including through the invocation of treaty-based exceptions when necessary.
VI. Conclusion
The foregoing analysis offers no easy conclusions about the relationship between security and international law. International legal regimes always empower the knowledge practices of some actors over others. Security can entrench or expand the existing practices of a regime, or it can disrupt the established order, empowering other forms of knowledge and allowing actors to operate according to a different logic. But we cannot know whether and how we want to disrupt an existing set of rules and institutions until we are faced with a particular regime and a concrete set of circumstances. To be sure, there are reasons to be cautious about security: the concept has longstanding associations with the sharp edge of state power, and there is no clear way to tell how even the most desirable security interests, once vindicated, will reverberate across international law's “spread-out web of normativity.”Footnote 402 But security may also supply a much-needed disruption where international legal systems entrench outmoded forms of expertise, or where international law actually contributes to the creation of insecurity. The analytical framework developed in this Article contributes to addressing these dilemmas, enabling actors in the international system to take apart security claims and see how they work or might be made to work differently.
In addition to informing the work of legal actors, the framework also suggests, by way of conclusion, a systemic point about the position of the state in the international legal order. Security is closely associated with the rise of the state and the problematic notion of sovereignty, and in the international system security interests continue to be vindicated largely by states, acting either individually or collectively.Footnote 403 If security talk is indeed on the rise in contemporary politics, we may naturally ask whether this signals a return of the state, and potentially a thinning out of global ordering or, as we are seeing in Europe as this Article goes to press, a slide toward war.Footnote 404
The analysis presented here suggests the picture could be more complex. The first three of the four types of security—realist, widened, and discursive—are closely associated with the military, bureaucratic, and executive apparatuses of the state.Footnote 405 But by distinguishing between these types, we can see how the state's relationship to security can be remarkably plastic. A focus on security interests can lead to militarization and the dominance of defensive thinking. But other vocabularies aim to domesticate security in the service of expertise and rational problem-solving, or attempt to reappropriate the power of the security state to address emerging threats like climate change or pandemic disease. Whether either of these alternative visions is obtainable under the pressure and constraint of today's politics is a far more difficult question. But it is critical to think deeply about the possibilities for remaking the security state, lest that thinking be done only by those with far more dangerous goals.Footnote 406 To aid this rethinking, this Article provides a conceptual apparatus that can contribute to ongoing interest in the “topography” of the security state in order to identify possible pathways of influence and change.Footnote 407
Nevertheless, this Article's final frame—security pluralism—suggests that even these questions are insufficient to grasp the realities of security today. The insight that people can know their own security needs just as well as appointed experts sits poorly with the structures of the modern state,Footnote 408 particularly where those needs are embedded in associative ties that do not neatly conform to national boundaries. The security interests of the most vulnerable are thus often re-articulated and distorted by state officials, international civil servants, non-state experts, NGOs, or “well-intentioned intellectuals,” burying their security claims under “layers of representation and thus silenc[ing] them further.”Footnote 409 Forceful assertions of state sovereignty may nevertheless be the best available means of protecting those interests, at least under current conditions, and this Article considers when such assertions can be a critical tool to “change the rules of the game.”Footnote 410 The analysis developed here is thus in sympathy with recent efforts to revisit moments such as Bandung or the New International Economic Order that offered alternative visions of effective state sovereignty, and recover in them lessons for the future of international law.Footnote 411
The pluralist frame urges us to further broaden our horizons to imagine what security practices might look like in a system that more radically decenters the state. In this respect, we can follow history even further back to the early colonial period, where notions of security and safety, sovereignty and statehood, were commingled. As Lauren Benton writes, this early period was characterized not by the full extension and dominance of European sovereignty, but by a robust “jurisdictional politics,” in which both the colonizer and colonized were sophisticated in interpreting and manipulating the plurality of normative orders to secure their interests.Footnote 412 The events of the early twentieth century, in particular the United States’ response to the 9/11 attacks, have reproduced a similar kind of jurisdictional plurality, as an “international state of emergency” has justified the creation of novel forms of administrative power and legal subjecthood.Footnote 413 This new reality, like that under colonialism, is far from a benign or congenial pluralism.Footnote 414 Yet elsewhere we see resurgent interest in other vocabularies—such as food sovereignty and Indigenous sovereignty—that offer alternative visions of safety and security that emerge from associative ties other than statehood.Footnote 415
This Article explores the extent to which these vocabularies create alternative demands for security and survival that are often rendered invisible by the international legal system, but are no less important than those of the dominant actors. The struggle for visibility is a long-running theme of strands of scholarship that are often left out of mainstream discussions of international law and security.Footnote 416 The framework developed here attempts to build a bridge across those divides in effort to foster a richer, more inclusive, and more authentic discussion of what it means to be secure in an uncertain, contentious, and warming world.
Target article
Making Sense of Security
Related commentaries (6)
Climate Change and Global Security: Framing an Existential Threat
Filters and Fragments: Making Feminist Sense of Security
Introduction to the Symposium on J. Benton Heath, “Making Sense of Security”
Making Sense of Violent Extremism
Race in Security
Whose Security Matters?