Restricted sovereignty

In this section, we review extant understandings of sovereignty in International Relations and Political Theory to suggest they offer little purchase on the issue of the failed state. Both fields conceptualize sovereignty as the property of an already constituted entity, the modern territorial state. Insofar as the failed state is the absence of that entity, the failed state lacks sovereignty (or its sovereignty is a fiction). The two fields are united in conceptualizing sovereignty in a theoretically and historically restricted sense, which is best revealed by looking at the failed state.

The term failed state suffers from some definitional ambiguity.Footnote ¹ It appears a polity with a vacuum of authority, and where the central government is non-existent. However, the latter does not necessarily imply the former; even a polity like Somalia, seen to epitomize the failed state, is not necessarily a vacuum of authority because parts of Somalia like Puntland are ruled by tribal groups that serve as de facto governments. So it is more accurate to define a failed state as a polity where the central government is significantly weaker, in material capacity and territorial control, than other actors – warlords, rebel groups, criminal gangs, private security contractors, peace-keepers – within the territory. This definition can be extended to define another category – the ‘weak’ state – where the central government may remain the strongest actor, wielding the most material force and controlling the most territory, but there are other armed actors, possibly controlling territory, that make the central government’s monopoly on violence impossible. Some states are weak but unlikely to see the central government’s advantage usurped – India, where a long-running Maoist insurgency is present in a sixth of all districts exemplifies this – while others may be seemingly stable, but the central government can be overthrown in a short period, an example being Libya. In the failed state, and to a lesser extent the weak state, there are attenuated restrictions on the exercise of violence, and even territorial control, by actors other than the state.

While thinking about sovereignty has a long lineage in Political Theory, in the last decade or so, the theorizations of Carl Schmitt, and its extensions by Giorgio Agamben, have become the focal point to understand the exercise of power in sites as diverse as occupied territories, camps like Guantanomo Bay, and the War on Terror (e.g. Butler Reference Butler2004). We therefore begin with the Schmittian definition of the sovereign as he who declares the exception (Schmitt Reference Schmitt2006, 5–15). Sovereignty is instantiated in the declaration of the state of exception, the suspension of the law in instances of emergency. The state of exception is both the prerogative of the sovereign – the declaration of the exception marks him as sovereign – and the condition in which he can exercise ‘determining force over human life’, that is act sovereign (Edkins Reference Edkins2000; Prozorov Reference Prozorov2005; Doty Reference Doty2007; Passavant Reference Passavant2007, 168, 153–7; Huysmans Reference Huysmans2008). Does the state of exception offer analytical purchase in understanding the failed state?

At first glance, the failed state has a plethora of armed actors who declare their own right to take life; that is a ‘dispersion’ or a proliferation of sovereignty. But this proliferation indicates the limitation of the state of exception framework as a description of the failed state. Insofar as there is no single sovereign, there is no one to declare the exception. Indeed, the state of exception is declared precisely to protect against the threat these other armed actors pose, and preserve the institution of sovereignty beyond the death of a particular sovereign (Bodin Reference Bodin1992; Agamben Reference Agamben2005, 23). Whatever else we can say about ‘ungoverned areas’ like Somalia or the barely policed border areas in Mali, they cannot be described under the rubric of the state of exception: there is no law to suspend in the first place.

But we can go further to make two points. First, if the failed state is not an example of the exception yet is a space where armed actors face few limits on their use of violence – that is a proliferation of sovereigns that transgress the law with impunity – the state of exception cannot be coextensive with sovereignty. We will elaborate on this point in the next section, but we note here that the state of exception marks not sovereignty in general, but a particular, and restricted, manifestation of sovereignty. Second, the state of exception is not, as suggested, an exercise of untrammeled power by the sovereign through the suspension of the law.Footnote ² Rather, the law is suspended so that the sovereign does not transgress it in the course of dealing with the emergency. By suspending the law rather than transgressing it, the sovereign implicitly concedes a limit to his exercise of power. The state of exception is therefore a space where the exercise of sovereignty is limited: the sovereign can suspend, but not transgress the law. By being suspended, the law is preserved in the state of exception. Insofar as the failed state is characterized by the transgression or the dissolution of law, the state of exception framework not only has little application to the failed state, but the failed state reveals that the state of exception is not exhaustive of sovereignty. Sovereignty is not restricted to the exception, hence the theorization of sovereignty in Political Theory is not very helpful as a tool to analyze the failed state.

Do conceptualizations of sovereignty in International Relations theory offer greater purchase on explaining the failed state? Sovereignty in International Relations theory has come to be understood as a norm and legal institution, albeit imperfectly instantiated (Jackson Reference Jackson1999, 431; Lake Reference Lake2003, 305; Holsti Reference Holsti2004). More precisely, sovereignty is a socially constructed status bestowed by the international community to mark legitimate membership (e.g. Bartelson Reference Bartelson1995; Biersteker and Weber Reference Biersteker and Weber1996). But both in terms of who qualifies as sovereign, and what sovereignty is, sovereignty has varied over time (Osiander Reference Osiander2001). Sovereignty has historically been bestowed on a range of polities, such as empires (Nexon Reference Nexon2009; Burbank and Cooper Reference Burbank and Cooper2010). This understanding of sovereignty was expansive, because a range of entities, from the Mughal empire to Native American tribes, were recognized as sovereign, in the sense of being equal partners to the signing of treaties. Dissimilar entities like an Indian tribe and a commercial company functioned as sovereign entities, in that they could make and enforce contracts, for the signing of which their institutional structure had no relevance; only the mutual recognition enshrined in the treaty testified to sovereignty. Since 1945 however, sovereignty has come to be seen as the exclusive prerogative of one particular institution, the nation-state. From being conceptualized as a state’s right to non-interference in domestic affairs (Krasner Reference Krasner2001, 1–2), an absolute right precluding any form of intervention, sovereignty has more recently been rearticulated as a state’s ‘responsibility to protect’ the human rights of its citizens, making intervention possible on condition that a state is either unable or unwilling to safeguard the human rights of its citizens (International Commission on Intervention and State Sovereignty (ICISS) 2001).

The failed state offers a legal and practical paradox for this state-centric notion of sovereignty. On the one hand, the absolute right of non-interference is seen to prevent members of the international community from intervening in failed states. On the other hand, as the language of ‘failure’ suggests, these polities have fallen short of exercising the functions of ‘sovereignty’. The failed state is simultaneously protected by the institution of sovereignty and undeserving of membership in it. Scholars have suggested that combination of the inability of the government in a failed state to monopolize violence and the unwillingness of the international community to transgress that state’s sovereignty has allowed civil wars to endure (Hironaka Reference Hironaka2005), prevented the emergence of alternative authority structures that might actually provide order (Herbst 1996/19Reference Herbst97), and hinders cooperation with armed actors that are not recognized as sovereign but could offer assistance in dealing with issues like piracy.

The reconceptualization of sovereignty as Responsibility to Protect rather than a right of non-interference responds to this paradox. It sanctions intervention in failed states by making sovereignty conditional on the performance of the central government, but does not extend the status of sovereignty to entities other than the state (Ignatieff Reference Ignatieff2001). Sovereignty is (again) the property of an already constituted entity, the state. The failed state does not deserve its’ sovereign status, and is thus subject to intervention to ‘rebuild’ a state that would deserve the recognition of sovereignty (ICISS 2001). Just like in the state of exception framework, sovereignty in International Relations is not absolute, but conditional. Since the nineteenth century, some polities have conceptualized as lacking the organizational capacity to serve as states, or behave according to civilized norms (Grovogui Reference Grovogui1996; Simpson Reference Simpson2004). As late as the 1850s, Egyptian and Ottoman rulers were invited to European conferences as equals, but by the 1870s, they were seen as despots who had let their people slide into barbarity, and were excluded (Mazower Reference Mazower2012, 71–2). In the case of nineteenth century international law as it applied to non-European despots as with the Responsibility to Protect as it applies to failed states, sovereignty is not untrammeled power: those that actually exercise such power, in the contemporary international system as in the nineteenth century imperial system, are not sovereign.

In both Political Theory and International Relations, sovereignty is the property of an already constituted entity, one that does not exercise untrammeled power. Untrammeled power is in fact a mark of despotism, unworthy of sovereignty (Doyle Reference Doyle1986, 30–2). By contrast, sovereignty in Political Theory and International Relations reflects a modern conception of power, where limits on the sovereign in domestic affairs are necessary for recognition in the international community (Bartelson Reference Bartelson1995; Foucault Reference Foucault2003). In the failed state, there are no such limits and thus no such entity. Hence the rubric of the state of exception does not apply to the failed state, and the failed state is seen as undeserving of sovereignty. Yet, the failed state sees a proliferation of sovereigns, so it is not sovereignty per se, but a certain type of sovereignty that is absent in the failed state. We now turn to theorizing that type of sovereignty, which we will call sovereign power.

Disaggregating sovereignty and sovereign power

The failed state is characterized by the transgression and/or the dissolution of the law. Theorizations of sovereignty in Political Theory and International Relations, by contrast, posit the existence of the law backed by an authority to enforce it. In the state of exception, the sovereign suspends the law precisely to avoid transgressing it. Sovereignty in International Relations theory is conceptualized in relation to the ability to enforce the law; while sovereignty as a right of non-interference protects states that are neither willing nor able to enforce their laws, the reconceptualization of sovereignty contained in the ‘responsibility to protect’ is an effort to rescind that protection, reasserting the link between sovereignty and law. In both instances, the focus on the law suggests that sovereignty is not reducible to domination or extortion by the ruler; indeed, simple domination marks the absence or voiding of sovereignty.

What does the primacy of the law tell us about sovereignty? The law marks the formation of a political community, both a community where the governed respond to the sovereign, but also a community that can be known by the sovereign.Footnote ³ The formation of a community also enables the sacrifice of a part of the community to save the whole, for example in the dispatch of soldiers to a warfront. Equally, the recognition of the community constitutes an individual or an armed group as the sovereign, conferring a unitary, and uniting, quality to a ruling group. In this sense, the sovereign is anthropomorphic, whether the ruler is an individual or a group, because the sovereign is unitary and the relationship between sovereign and governed is disclosed in speech (Wendt and Duvall Reference Wendt and Duvall2008).

Thus conceptualized, sovereign power is not raw, detached power or domination, because the primacy of the law, reflected in its suspension rather than dissolution, always reflects a relationship of reciprocity with the governed (Trainor Reference Trainor2006). In this sense, it can describe a liberal-democratic state as well as a dictatorial regime where power is exercised on behalf of, or in the name of, the governed. The point to be taken about sovereign power in the state of exception is that, contra to freeing sovereign power from limits, the state of exception limits the power of the sovereign and his ability to behave arbitrarily through the law, which both bounds a political community and also enables the sovereign to know that community. Insofar as the exception is designed to preserve the political community, the state of exception does not bring us any nearer theorizing the transgression of the law or the dissolution of sovereign power. To do so, we need to specify a more expansive concept of sovereignty.

Sovereignty, as contrasted to sovereign power in our usage, refers to a more general set of relations, including where the reciprocity between sovereign and governed is unstable to the point of being annulled. Insofar as annulling the reciprocity between sovereign and governed is an act of transgression, sovereignty (henceforth capitalized to distinguish it from extant usage), in the last instance, lies in the capacity and willingness to transgress the law. This potential annulment is not captured either by the concept of sovereignty in International Relations, where sovereignty is a right of recognition due those sovereigns who exercise power within the law, nor by the concept of sovereign power in Political Theory, where sovereign power is bounded by the law, even when the law is suspended. Both these disciplines understand sovereignty as tied to the law, and the transgression of the law as a violation of sovereignty. By contrast, our concept of sovereignty is predicated on the possibility of the sovereign turning away from the governed and dissolving sovereign power itself, as is seen in the failed state. We suggest that if sovereignty means anything, it means that the sovereign can always disregard the rights of subjects under law. Sovereignty is the more universal potential in any relation of rule, of which sovereign power is more delimited, conceptually and historically. Reciprocity between sovereign and governed is not a universal characteristic of relations of rule, which may either not be reciprocal, as in the civilizing mission, where colonial subjects have no ability to contest their subjection, or may in fact annul any relationship with the governed, as in the colony, where the sovereign ruled without jurisdiction (Mbembe Reference Mbembe2001). Sovereignty encompasses all these categories; sovereign power relates to a limited subset of them.

Sovereignty comes first

The reciprocal relation between sovereign and governed is not exhaustive of relations of rule. Instead, as the dissolution of authority in much of the world suggests, it is relatively rare, even as it is necessary to the foundation of sovereign power. The question this section poses is ‘are spaces where the reciprocal relation between sovereign and governed does not hold different in kind from spaces where it does?’ Extant approaches answer that question in the affirmative and posit that spaces where reciprocity does not obtain, like the failed state, lack sovereignty. The most intuitive example is to contrast spaces of sovereign power with a Hobbesian state of nature, an asocial, or pre-social, space before the foundation of authority; the failed state being its’ contemporary manifestation.Footnote ⁴

We disagree with such approaches, and instead argue that the dissolution of sovereign power does not mark its’ absence, but instead points to its’ constitution in ‘sovereignty’ (again, we note that we capitalize our usage for ease of reference). We make two significant critiques of extant understandings. First, contra a view where sovereignty is a characteristic of sovereign power, as implied in International Relations, we disaggregate the two and argue that it is sovereignty that is before sovereign power: sovereignty comes first. Second, contra a view where the dissolution of sovereign power is indicative of a lack of sovereignty, or effective sovereignty, we show how the dissolution of sovereign power stems from the logic of sovereignty, namely the capacity and willingness to transgress the law. In sum, sovereignty is a contradictory relation that is both before the foundation of sovereign power and contains the possibility of sovereign power’s dissolution.

To unfold this argument, we will make five interlocking claims. We return to the example of the state of exception, where the sovereign suspends the law in order not to transgress or dissolve it, to ask: what would it mean if the sovereign transgressed or dissolved the law instead of suspending it? The sovereign’s transgression/dissolution of the law implies the repudiation of the reciprocal relationship with the governed (1). What transgression/dissolution reveals is that before the granting of reciprocity, there is always the possibility of not recognizing the governed (2). When we lay out this possibility of non-recognition as a logical and temporal priority, it becomes clear that sovereignty is prior to the establishment of sovereign power (3). As sovereignty precedes sovereign power, the two are not coextensive, and we lay out the differences to substantiate this (4). Finally, the relation between the two reveals a paradox: the moment of repudiation of reciprocity always marks the necessity of the governed. This paradox implies that sovereignty is not a thing to be possessed, but a contradictory relationship rooted in the capacity and willingness to transgress the law (5) (Brown Reference Brown2008, 253–4).

To understand the absence or breakdown of sovereign power, we return to how the state of exception serves to preserve sovereign power. The state of exception is a juridical device that enables the maintenance of the rule of law through its suspension. It is not, as Prozorov puts it, a ‘constitutive transgression’ (Prozorov Reference Prozorov2005, 87, 93). Rather, the sovereign suspends the law in order to avoid transgressing it. In so doing, the sovereign marks a relation of reciprocity with the governed, in order for the society to continue as a corporate body. So the exception preserves rather than constitutes sovereign power; it is not constitutive or generative of sovereign power. Given the effort to declare the exception within the space of the law, what would it mean for the sovereign to transgress or dissolve the law?

If the exception preserves the reciprocal relation between sovereign and governed, to transgress/dissolve the law instead of suspending it is to void that relation. We might say that to transgress/dissolve the law is to repudiate the relation of reciprocity and recognition. This is our first claim (1). This act of transgression/dissolution, by repudiating the relation of reciprocity, appears to follow, temporally, the establishment of sovereign power. The continued existence of sovereign power appears compromised by an irresponsible ruler who refuses to recognize his subjects.Footnote ⁵ Transgression thus appears as a choice of the sovereign that has nothing to do with the nature of sovereign power, which maintains the political community, and is temporally posterior to the establishment of sovereign power. Transgression appears a contingency, an accident almost, not structural to sovereign power. It therefore seems that extant understandings are correct to separate the operation of sovereign power from its dissolution, because the two are essentially unrelated. The dissolution, if anything, is caused by the fact that the sovereign departs from the prior logic of sovereign power, as embodied in the exception.

Yet, if we look at the temporal structure of sovereign power, we find that transgression reveals a paradox. On the one hand, transgression/dissolution is defined as such by the prior operation of the law. One cannot define an act a transgression if the law does not already exist and identify it as such. On the other hand, before law can function, it is instituted through an act of transgression, most often violent transgression. Law does not emerge from anarchy, as in the Hobbesian ideal-type. Rather, the establishment of a particular law replaces and supersedes a prior law. Yet, by the terms of the law being superseded, this act is illegal. This illegal act from which law emerges is prior to the law it establishes, meaning that law is instituted through a transgression (Freud Reference Freud1955).

Take, for example, revolutionary violence that seeks to negate the entirety of the law in order to institute a new law (Fanon Reference Fanon1965). What is crucial is less the violence – there can be and have been non-violent revolutions – than the act of transgression.Footnote ⁶ Yet, precisely through negating the existing law, revolutionary transgression always contaminates the law it founds. This is evidenced by the fact that actors in the future can refer to the revolution itself as legitimizing transgression of the law (Benjamin Reference Benjamin1986; Derrida Reference Derrida2001). The founding transgression compromises the functioning of law because it reveals the law itself to be based on a transgression. This transgression, as it founds a political community, also dissolves another political community, thus introducing the specter of transgression/dissolution into the newly founded community. It is the simultaneity of founding an order and breaking from it that defines transgression because ‘the power of transgression implies in theory an existence outside the rule connected with right’ (Bataille Reference Bataille1993, 125; Wagner-Pacifici Reference Wagner-Pacifici2008, 480). The ability to stand outside the law by explicitly breaking it is both different from the state of exception, and places the reciprocity the state of exception preserves at risk.

In this way, transgression/dissolution attends the moment of the founding of sovereign power. More precisely, transgression is before sovereign power and constitutive of it: sovereign power stems from a ‘founding transgression’.Footnote ⁷ Further, transgression may be repeated in the maintenance of order. The police, for example, may facilitate particular crimes as ‘undercover operatives’ in order to arrest criminals, or engage in extra-legal violence such as torture in the effort to bring order (Benjamin Reference Benjamin1986, 284). In these instances, the law is enforced through transgression (we will expand on this in an examination of extra-judicial killings below). This brings us to our second claim: before the establishment of reciprocity, there is the possibility of reciprocity not being instituted (2). In other words, in the moment of founding transgression, a relation of reciprocity may or may not be created. Because the foundation of order is in transgression (often, but not necessarily involving violence), those transgressing are compelled neither by force nor by law to recognize the governed. They are not compelled by force because they either possess more of it or are willing to disregard the force of the existing order; they are not compelled by law because the law that they are founding does not yet exist and they do not respect the existing law. Indeed, because transgression is a necessary but not sufficient condition for the foundation of law, the law may never come into being at all. This reveals the essentially asymmetric origin of sovereign power.Footnote ⁸

Therefore, reciprocity stems from a willingness on the part of the sovereign to recognize the governed. This is why reciprocity cannot be understood as a social contract because the governed do not have the legal authority to declare the exception. The legal means to suspend the contract are purely in the hands of the sovereign. The ability of the governed to seek redress under sovereign power is contingent on the extent to which they are recognized as bearing rights in the first place. Even the political agitation through which the governed demand rights must, in the last instance, depend on the sovereign’s recognition of these rights in law, for example, ‘customary rebellions’ where peasants appealed to the Tsar against the infractions of a bureaucrat. This asymmetrical structure in one sense reflects our explication of sovereign power and in another sense contradicts it. Sovereign power is asymmetrical to the extent that it is the sovereign and the sovereign alone who can decide the exception (Schmitt Reference Schmitt2006, 5–6). However, the transgression that establishes sovereign power exceeds the logic of the exception. That is, the asymmetry embodied in the exception stops short of a repudiation of the governed. Transgression, on the other hand, is marked by precisely the possibility that the governed need not be recognized. Therefore it is not necessary for transgression to lead to the establishment of law, as the example of a general strike indicates (Benjamin Reference Benjamin1986, 281–3). Sovereign power cannot be sustained in the absence of reciprocity and recognition, that is through the perpetual repetition of transgression alone; but transgression is before the extension of recognition, and does not necessarily lead to the establishment of law and sovereign power. Thus, transgression is in the odd position of being constitutive of sovereign power, and yet one cannot theorize transgression within sovereign power (for sovereign power is designed to minimize, even disavow, transgression).

If transgression is theorized as constitutive of sovereign power, rather than a departure from it, how would we revise extant concepts of sovereignty? The relation between transgression and sovereign power points towards a broader concept of rule, which includes the possibility of both the recognition and the repudiation of the governed: ‘sovereignty’.Footnote ⁹ As we have suggested, transgression implies, as a necessary possibility, two things that are inconsistent with the logic of sovereign power. First, before the recognition of the governed, there is the possibility that the governed will not be recognized.Footnote ¹⁰ Second, as a continuing possibility, the sovereign may choose to repudiate the governed even after the initial moment of recognition. This simply means that sovereignty haunts the founding and preservation of law with the possibility of its repudiation, and with that, the possibility of the dissolution of the legal order. Footnote ¹¹

Put otherwise, sovereignty haunts sovereign power with both the paradox of its founding and the ever-present possibility of its dissolution. Sovereignty reveals the essentially transgressive and asymmetric origins of sovereign power, a reciprocal relationship that never escapes reference to law. The relationship of sovereign power is not perfectly symmetrical, because only the sovereign can declare the exception. However, this asymmetry is limited by the law, which marks, even in the exception, that the governed have recognition and standing as subjects of the law. In contrast, the asymmetry implied in sovereignty has no limit: if the governed are not granted recognition, the sovereign can discharge his will without any restriction.Footnote ¹² This essentially asymmetric origin of the law always figures as a contradiction, best exemplified when, in revolting against the law, actors point to its illegitimate origin to justify their violence. In such arguments, we see two emphases. First, the law is an effect of a prior transgression that contaminates its functioning. Second, the law is inadequate to address the consequences or costs of this transgression, and must be superseded by a different transgression in the name of another law that does not yet exist. This illustrates our third claim: sovereignty comes first (3). Logically, sovereignty, manifested in transgression, is a necessary but not sufficient condition for the establishment of sovereign power. Temporally, sovereignty precedes sovereign power. In both a logical and temporal sense, sovereignty is before the establishment of sovereign power.

The logical and temporal priority of sovereignty over sovereign power haunts sovereign power and threatens its dissolution. The asymmetry between sovereign and governed is not overcome by the establishment of reciprocity between them. At no time is the divide between sovereign and governed erased, for the sovereign is marked apart by the ability to decide the exception. This divide is exacerbated when the sovereign transgresses or dissolves the law, in a manner that is, if not consistent with the logic of sovereign power, consistent with the logic of sovereignty. Such a transgression negates the juridical relationship of sovereign power and, at the extreme, can lead to the dissolution of sovereign power.

The preceding argument now enables us conceptually to distinguish sovereignty from sovereign power to explicate our fourth claim. Sovereign power is a subset of sovereignty, where the sovereign and the governed recognize each other and the latter can make claims on the former under the law – what we have called reciprocity (4). Sovereignty precedes sovereign power, because law is founded through an originally illegal act by the sovereign. Finally, sovereignty need not necessarily lead to the establishment of law and sovereign power. This is because the asymmetry between sovereign and governed means that the recognition of the governed depends on the will of the sovereign. More explicitly, sovereignty can be understood as the possibility of sovereign power not to be (Agamben Reference Agamben1999, 179, 182).

Extant approaches suggest that sovereignty – in lower case to contrast from our usage – is a characteristic of an already constituted sovereign power, as in a juridical mark such as sovereign statehood. This approach is misleading. It suggests that sovereign power, even if inadequate to the law, is logically and temporally before sovereignty, and not, as we have argued, vice versa. Also, it casts sovereignty purely in a juridical mode (even the exception is a juridical concept), as the opposite of transgression. We have argued instead that transgression is central to understanding sovereignty.

Take the transgression of the prohibition on homicide. Agamben specifies that ‘the sovereign sphere is the sphere in which it is permitted to kill without committing homicide and without celebrating a sacrifice’ (Agamben Reference Agamben1998, 83). A sovereign that commits homicide negates the reciprocal bond with the governed, contradicting the logic of sovereign power. Murder, in violating the right of the governed to life, seems to negate the social bond with the governed. It appears a nihilistic or asocial act, a deviation from the juridically bounded relation of sovereign power. It is therefore necessary to rewrite the law to legalize violence against certain individuals or groups who can be killed without their deaths being seen as homicide. Yet, to murder is to realize sovereignty, to be unencumbered by any limits on the exercise of power. Murder is thus very much within the logic of sovereignty, even as it contradicts the logic of sovereign power. Empirically, sovereign homicides are common, as we will discuss through the example of extra-judicial killings below and in the next section. Sovereign homicides reveal the paradox that sovereignty is simultaneously social, yet sovereignty lies in negating the governed that comprise the social. This paradox is insuperable, because the sovereign never adequately negates and transcends the social, even in the exercise of death over the governed.

We can explicate this by revealing the paradoxically social logic at work even in extreme instances of transgression, for example, sovereign violence conducted in secret, as in Stalin’s purges or the ‘disappearances’ in Latin America. Here, transgression appears to dissolve a society in silence. The space of speech and sociality seems to end at the point at which sovereignty begins. As Arendt put it, ‘real power begins where secrecy begins’ (quoted in Schmitt Reference Schmitt2003, 336; also see Derrida Reference Derrida1978, 266). Yet, even this secrecy has a social, communicative logic. It indicates to others that their relation with the sovereign is equally likely to be negated and it is thereby constitutive of a political community, albeit limited in social extent. In both these instances, the negation of reciprocity for one section of the population marks the precarious status of recognition for others, inducing or compelling those others to identify with the sovereign. That is, even the purges or disappearances mark the ultimately social character of sovereignty. This is not just for the obvious reason that they were conducted by organized state agents who kept records, which are now being used in criminal proceedings and truth commissions (suggesting that the force used by the sovereign in excess or violation of the law can never be truly silent). More consequentially, even in silence and secrecy, the disappearances had a social character. As a public secret, something everyone knew but could not speak to others about, the disappearances kept individuals from associating with each other to challenge the sovereign (Figes Reference Figes2008). The silence constituted a particular sort of political community, and prevented others, such as anti-statist associations, from forming. Then, as small groups, like the Mothers of the Disappeared in Latin America, began to publicize the events and call for justice and information, the events catalyzed opposition to the sovereign (Taussig Reference Taussig1991). Even in its silence and secrecy, the sovereign communicates its power to transgress to its subjects, whether by killing opponents or conveying the threat to potential opponents, in order to be sovereign. These transgressions by the sovereign are therefore not nihilistic, even as they negate the very basis of sovereign power, the reciprocity owed to the governed.Footnote ¹³

Therefore, the fact of homicide (or any other transgression) does not constitute sovereignty – if so, any criminal would be ‘sovereign’ – but its communicative quality does. An example will bear this out. A drug trafficker may murder a police officer in the course of transporting drugs, but this homicide does not instantiate sovereignty unless the drug trafficker murders the officer in order to signal to others that his writ overrides that of the police. The symbolic and communicative aspects of these homicides often drive them beyond a utilitarian calculus to involve mutilations and gruesome acts.Footnote ¹⁴ Such spectacular transgression is a declaration of sovereignty; it conveys the force of the sovereign beyond those who are killed, to make the killings themselves a communicative act.Footnote ¹⁵ In the extreme (or ideal) situation, the police respond by conducting operations wearing masks to conceal their identities, as occurs in Mexico. Obviously, if the police wear masks – even at press conferences announcing successful operations – they are not confident in their ability to protect themselves from retribution, much less protect the governed from the depredations of the drug traffickers.

We can now formalize the paradox that is sovereignty. While sovereignty lies in repudiating the governed, its instantiation in transgression marks both the priority of the governed and the resolute sociality, for want of a better word, of sovereignty.Footnote ¹⁶ Efforts to repudiate the governed mark all the more the sovereign’s dependence on her, even when the sovereign does not recognize her. To recognize the claims or rights of others is to avow a dependence on them and render the self servile because it is not really the source of its own authority, not really sovereign (Derrida Reference Derrida2005, 12).Footnote ¹⁷ This recognition, and by implication any belonging to a political community, limits the self from becoming sovereign. As Bataille put it, ‘the fact is that solidarity keeps man from occupying the place that is indicated by the word “sovereignty”: human beings’ respect for one another draws them into a cycle of servitude where subordinate moments are all that remains’ (Bataille Reference Bataille1993, 178–9). But at the same time, to be sovereign implies the existence of others with respect to whom one is sovereign. Maurice Blanchot notes this paradox in the Marquis de Sade, a thinker of sovereignty beyond the law and exception

Sade’s characters, Blanchot notes, negate others through various acts of cruelty in order to be sovereign. Yet, even these acts of negation mark the priority of and need for these others (Blanchot Reference Blanchot1965, 58). What this reveals is that sovereignty is not a possession or property, but a contradictory relationship, our fifth claim (5). Sovereignty implies the presence of others in relation to the sovereign, and the recognition of that relationship, but to be sovereign means to deny that presence and those relations. Put otherwise, sovereignty is a social relation that can be realized only in a negation of the social. Sovereignty thus becomes impossible to grasp, hence Bataille writes that ‘sovereignty is NOTHING’ (Bataille Reference Bataille1993, 256, emphasis in original, 430).

If sovereignty is ‘nothing’ and a negation of the social, how are we to understand the relation that is sovereign power? In sovereign power’s relative stability and duration, don’t we see the opposite of the Sadean sovereign? Our answer is ‘no’, because the unconditional nature of sovereignty, and the asymmetry of power between sovereign and subject, imply that there is no need for the sovereign to recognize the subject, much less grant reciprocity to her and found sovereign power. Over history, stable regimes of sovereign power have been the anomaly, and seemingly powerful states like the Soviet Union have unraveled swiftly (North et al. Reference North, Wallis and Weingast2009; Acemoglu and Robinson Reference Acemoglu and Robinson2011). While other scholars attribute this rarity to the decisions of elites seeking to preserve their economic power by limiting the property rights of their subjects, economic relations between rulers and subjects can also be understood under our concept of sovereignty. Indeed, our concept does a better job explaining the excessive actions of sovereigns that detract from the economic value of their subjects, treating them as animals, objects, or even nothing, as in the slave camp, and the colony.Footnote ¹⁸ In the colony, the subject is denied the status of a rights-bearing subject, but can still serve as labor (Mbembe Reference Mbembe2001, 183). The colony, however, reveals precisely the impossibility of sovereignty. On the one hand, the sovereign rules without jurisdiction, unlimited by any rights of the governed. On the other hand, this rule produces its own dissolution because it does not recognize the governed, who in turn do not recognize the sovereign and disclose information to him. Sovereignty is in the paradoxical position of being both prior to the recognition of subjects – the sovereign chooses not to recognize subjects – and yet dependent on that recognition for its existence, and continuation thereof. Plantation slavery reveals this paradox even more sharply. Slavery functions by sundering the slave from her community, exemplified by the process of renaming the slave after the master. Through this ‘social death’, the master exercises power over the slave (Patterson Reference Patterson1982). The master is sovereign, to the extent that ‘without the master the slave does not exist’ (Tuareg, quoted in Patterson Reference Patterson1982, 4). But by the same token, the existence of the slave is necessary for the master to exercise sovereignty. And so, as in Toni Morrison’s Beloved (based on the true story of the slave Margaret Garner), where the slave Sethe kills her daughter rather than let her be enslaved again, the master’s sovereignty is, in the last instance, in the hands of the slave (Morrison Reference Morrison1987).

In other words, sovereignty as a non-relation is the prior condition for the reciprocal relation that is sovereign power. Sovereign power is a relationship of reciprocity that, by binding the would-be sovereign, forms a political community that can aggrandize the power available to the sovereign. But the priority of sovereignty means that the bounding effect of sovereign power is always partial, and can be voided. This is contrary to understandings of sovereign power as the prior relationship which becomes atrophied because of tyranny and ultimately dissolves. In this view, there is nothing structural to sovereign power that brings about sovereign dissolution, and therefore such dissolution is a contingent outcome, an aberration from the logic of sovereign power. The implication of this, however, is that spaces of sovereign dissolution are also lacking or outside of sovereignty, understood as juridical characteristic. Yet, these spaces are hardly alien to the logic of sovereignty, even if they cannot be understood as spaces of sovereign power. What they reveal is the paradoxical nature of sovereignty, that it both repudiates and requires the governed. In so doing, sovereignty makes the establishment of sovereign power contingent and precarious to the point of dissolution. Before the establishment of rule on a reciprocal basis, sovereignty as an unconditional, transgressive, act always implies the possibility of the repudiation of the governed. It is this possibility that continues to inhabit institutions of sovereign power, introducing the possibility of arbitrariness and dissolution to even the most stable polities.

Sovereignty in most of the world

The question remains, however, as to the value of this abstract formulation for empirical questions in International Relations. In the course of this section, we will suggest that our theorization can better illuminate politics in failed and weak states – how sovereignty works both in service of and to compromise order, whether that order is imposed by the state or foreign powers or a combination thereof – than extant approaches that posit these polities lack sovereign power or their sovereignty is undeserving. Failed states are empirically rare – at any given time, less than 10 states in the international system can be said to have failed – so we focus our analysis on the weak state, a polity where the central government does not have a monopoly on violence, but remains the most materially powerful actor. We ask: how do sovereignty and sovereign power interact and coexist? The stakes of this question are significant: as weak states account for the majority of states in the international system, we claim to provide a theorization of sovereignty as it operates in much of the world, instead of claiming that sovereignty is absent (the failed state) or deficient (the weak state).Footnote ¹⁹

One of the overlaps between conceptualizations of sovereign power in Political Theory and International Relations is the regulation of transgressive sovereign violence so that it is wielded under specified conditions, for example outside the body politic on an external enemy (or by formally excluding a group that was a part of the body politic). This does not shed light on cases where violence occurs inside the body politic, without being legalized, to preserve sovereign power. Consider, for example, the prevalence of extra-judicial killings in institutionalized democracies like India and Brazil. These killings are homicides, because their victims are not formally expelled from the status of rights-bearing individuals. The victims are not, therefore, homines sacri whose death is neither sacrifice nor homicide (Agamben Reference Agamben1998, 83). There are two aspects to extra-judicial killings worth noting: first, they are resorted to because the judicial process is seen to be ineffective in the prosecution of criminals; second, they often have the implicit support of wealthier groups, who are usually associated with rights-based activism. Extra-judicial killings therefore are paradoxical in that they simultaneously repudiate the reciprocal recognition granted certain members of the community (without formally expelling them from that recognition), but that repudiation enables the perpetuation of sovereign power. Sovereign power is perpetuated not through the exception which suspends the law in order not to transgress it, but through the transgression of the law by state functionaries.

Extra-judicial killings need not be a solely domestic issue. We would describe the targeted killings, including but not restricted to drone strikes, used to eliminate suspected terrorists in Pakistan as extra-judicial killings because they do not occur in a war zone against recognized enemy combatants (Amnesty International 2013, 43–4). The Pakistani state’s role in these deaths is murky. On the one hand, the Peshawar High Court has found drone strikes, in particular, to be illegal and a violation of Pakistan’s ‘sovereignty’ and Pakistani politicians have been publicly critical. On the other hand, elements of the Pakistani security services are alleged to be providing the information on which targeting decisions are made (Amnesty International 2013, 53–5; Mazzetti Reference Mazzetti2013). It seems reasonable to assume that targeted killings and drone strikes are politically unpopular and not legal under Pakistani (and probably international) law, yet the Pakistani and US governments are collaborating to eliminate common enemies. Threats to sovereign power – to the United States and Pakistan – are thus eliminated by transgressing international and Pakistani law. Further, the collaboration between the United States and Pakistan is effectively a hierarchical relationship, where the weaker Pakistani state contracts aspects of its own security – the elimination of suspected terrorists – to the stronger state (Lake Reference Lake2011). David Lake suggests that such ‘social contracts’ introduce stable hierarchical relationships within the anarchical international system. However, in contrast to Lake’s explication of hierarchy as mutually beneficial arrangements – for example, the US–Japanese relationship after 1945 which was described as a ‘voluntary continuation of the occupation’ (Quoted in Lake Reference Lake2011, 55) – the Pakistan–US hierarchy, by transgressing the rights of Pakistani citizens, is seen to foment opposition to the Pakistani state, and destabilize it. Hierarchies in international politics where a foreign power supports a ruler in and through transgressing the rights of his subjects are examples of sovereignty as we have conceptualized it, which have deleterious effects on domestic and regional stability.

The paradox of state operatives preserving sovereign power by transgressing the law that underpins it leads to another. One would think that in a place like Brazil or Pakistan, it is the strong that would appropriate the power of the law, while the weak either suffer its depredations, rebel against it, or flee. But, faced with state agents breaking the law, their victims often appeal to domestic law and international human rights law, no matter how ineffectual. In the aforementioned court judgment against drone strikes in Peshawar, the judge cited both the UN Charter and the Geneva Conventions.Footnote ²⁰ In Brazilian prisons, incarcerated gang members promulgate constitutions in order to remain united and demand prison reform (Caldiera Reference Caldiera2006, 110–1). Here, again, sovereignty is not simple domination: the private, transgressive use of force by the strong is brought into the public realm by the weak, with the irony of criminal gangs assuming vestiges of legality in contrast to the state’s illegal use of force. The effect is what the anthropologists Jean and John Comaroff call ‘lawfare’, a proliferation of legal claims and legalisms because the law is frequently broken (Comaroff and Comaroff Reference Comaroff and Comaroff2006, 29–31). This indicates that generalized or frequent transgression, even when it involves illegal acts by state functionaries, should not always be understood as the absence of the law, or lead to its dissolution. Indeed, the former may lead to a proliferation, even a fetishization, of the latter.

The exercise of sovereignty to preserve sovereign power creates its own problems for sovereign power. As we have argued, sovereign homicides are essentially communicative, because they convey the force of the sovereign. But the fact of the homicide is communicative in another way. By repudiating the reciprocity owed to some members of the community, the homicide reconfigures relationships between sovereign power and the governed, and between members of the governed. By placing some members of the community outside the protection of the law, extra-judicial killings remove sovereignty, the repudiation of reciprocity, from the public realm of sovereign power where it can be legally contested. Recognition becomes contingent on the essentially private decisions of actors, which increases indeterminacy and uncertainty for both sovereign power and the governed. The latter react to this indeterminacy by, at the very least, not disclosing information to the agencies of sovereign power. Faced with unrelenting pressure to make unrealistic grain targets during the Great Leap Forward, local cadres regularly falsified statistics about grain production in their areas. By the time these reports reached Mao, it appeared that China had a massive grain surplus, leading to the following exchange in 1958:

• Mao: how are you going to eat so much grain? What are you going to do with the surplus?

• Zhang Guozhing (leader of Xushai Province): we can exchange it for machinery.

• Mao: but you are not the only one to have a surplus, others too have too much grain! Nobody will want your grain! (quoted in Dikotter Reference Dikotter2011, 41)

The Great Leap Forward led to 45 million deaths (of which at least 2.5 million were tortured to death or executed) and the destruction of 40% of available housing (Dikotter Reference Dikotter2011, x–xii). But a significant portion of the death and destruction came about due to perverse behavior by cadres and the public themselves that was not directly ordered by the state. Indeed, as the above exchange shows, Mao was often not even aware of what his cadres were doing, leading Dikotter to wryly conclude ‘to say that knowledge is power is a truism, and one that does not go very far in explaining why the more absolute power was, the less truth it managed to produce’ (Dikotter Reference Dikotter2011, 327). Here, the exercise of sovereignty ran counter to the logic of a sovereign power that would know its population in order to aggrandize its power. Consequently, while the Chinese state was responsible for mass deaths because of the environment of misinformation and repression it created, the scale of destruction of the Great Leap Forward exceeded the goals of its architects, and almost led to the disintegration of the state itself.Footnote ²¹

The governed can go further than denying information, to actively constitute a new community through transgressing and dissolving the existing law. A revolution, for example, both transgresses the existing law and, through that transgression, constitutes a new community. It is the constitution of a community through transgression that creates a tension between sovereignty and sovereign power in a revolutionary situation. More subtly, but insidiously, as transgression is constitutive of the new sovereign power, it is hard to excise transgressive modes of mobilization from politics as normal in future. More precisely, those who have participated in a revolution may have done so in order to maintain a condition where ‘everything is permitted’ as Saint-Just said of the French Revolution (Arendt Reference Arendt1963, 87). Such a polity, as France did in the nineteenth century, might fluctuate between insurrection and autocracy. To use a contemporary coinage, polities founded through transgression – the instantiation of sovereignty – may fluctuate from ‘state failure event’ to ‘state failure event’ as the establishment of sovereign power is constantly transgressed in the name of the revolution that established it in the first place.Footnote ²²

There are more mundane possibilities as well. After decolonization, postcolonial leaders stressed to their compatriots that the transgressive modes of political protest characteristic of the independence movements were no longer required. Jawaharlal Nehru, for example, declared the strike and other Gandhian tactics of civil disobedience as outdated practices (Chakrabarty Reference Chakrabarty2007, 37). The specter of a leader jailed for civil disobedience decrying it was ironic, of course, and has been overtaken by events in postcolonial India where strikes and riots are part of the everyday repertoire of political parties, and 162 of 543 (30%) Members of the lower House of Parliament in 2009 had one or more criminal cases pending against them (522 cases total), of whom about half (76) were alleged to have committed serious crimes like murder and kidnapping.Footnote ²³ Nehru was concerned about precisely this, that transgression would become politics as normal, and disrupt sovereign power’s functioning. But insofar as sovereign power in postcolonial India had come into being through transgression, and transgressive modes of political mobilization, these have become normal politics, and are impossible to dismiss as aberrations (Chatterjee Reference Chatterjee2004).

That transgression cannot be dismissed as aberrations or treated as pathologies – as the language of ‘failure’ encourages us to do – require us to understand how sovereign power can work alongside, and sometimes through, transgression. As we have already suggested in the case of extra-judicial killings, transgression can allow state actors to eliminate threats, sometimes in collusion with foreign powers, when the judicial system is slow or ineffectual. This can be extended to situations of fiscal constraint. Faced with low or shrinking budgets, states often reduce the salaries of bureaucrats and soldiers, who then supplement their incomes through bribes or instituting checkpoints. In the extreme case, as in Uganda under Idi Amin or Zaire under Mobutu, this privatization of the decision to recognize or not recognize the rights of the governed leads to state failure, or the dissolution of sovereign power (Bates Reference Bates2008). More commonly, it leads to an (uneasy) coexistence between sovereignty and sovereign power by repudiating and then renegotiating the relations of reciprocity between rulers and governed. Bribe-taking, for example, may not be an illegal departure from the prior functioning of state institutions, but the essential prior for the state to function, with paradoxical results on the law. A poorly paid policeman may not show up for work at all if he does not expect to be bribed. Once, however, he expects to be bribed, he has an interest in creating not fewer, but more laws, which he can then use to extract more bribes. Ironically, this thicket of laws, so long as they can be circumvented through paying a bribe, is not unacceptable to some citizens, who may be better off paying a bribe than paying for every regulatory permit and rule. This creates uncertainty and arbitrariness in the operation of the law. It also discriminates against the poor, who are unable to pay the bribe that would get them the services they are in principle entitled to (Gupta Reference Gupta2012, 19–26).

We can now provide some insight into the functioning of a state where state agents engage in illegalities and weaker sections demand more laws, but this diminishes rather than strengthens the rule of law.Footnote ²⁴ Here transgression coexists with a proliferation of the law: weak states like India, Brazil, and South Africa have constitutions that grant a range of rights to their citizens. Those who die because they did not receive what was owed them, such as medical help or crop insurance, or are directly killed by state agents cannot be defined as homines sacri because they are included, not excluded, by the law. The law in such places emerged not just to limit the excesses of a previous sovereign, such as a colonial power or military dictatorship, but to reverse a history of exploitation. Hence its appeal to the poor, who demand not just negative rights, but also positive rights to food and education, for example. But these rights either limit the state, or are beyond its ability to provide, so we see the exercise of sovereignty in transgressions to kill ‘undesirables’ or arbitrarily deny or grant rights, creating privatized relationships between public officials, sometimes in collaboration with foreign powers, and some of the governed, and the exclusion of others, with dire direct and indirect consequences on their life chances. This is the ‘weak state’: where the expansion of the law and appeals to it do not necessarily strengthen the rule of law or aggrandize the power of the state. Rather, sovereignty is privatized or proliferates, and the relationship between sovereign and governed is constantly negotiated and dispersed.Footnote ²⁵

Given this, we can revisit Charles Tilly’s comparison of state-making to organized crime (Tilly Reference Tilly1985, 169). Tilly meant that at the core of the state is an extortion racket, where an armed gang extracts taxes from producers to protect them from others and the gang itself.Footnote ²⁶ But clearly not all armed gangs – no matter how legalistic their internal proceedings – become states. The presence of some armed gangs and their exercise of sovereignty is seen, in fact, to be a violation of sovereign power. As an example, consider the ‘narco-state’, a sovereign state like Guinea-Bissau which is unwilling or unable to regulate the transit of drugs through its territory to North America or Europe.Footnote ²⁷ The drug traffickers exercise the capacity to take life, and by definition, transgress the law. Given the poverty of Guinea-Bissau, the state is unable to pay its bureaucrats and police forces anything like what a drug gang can, or protect them from the latter’s firepower. It would be in the self-interest of state functionaries to look the other way or actively participate in drug trafficking.Footnote ²⁸ Insofar as its officials ignore or abet drug trafficking, Guinea-Bissau is unable to fulfill the functions of a recognized sovereign; it is a ‘failed state’.Footnote ²⁹ But if they combated drug trafficking forcefully, the polity might descend into civil conflict, another type of ‘state failure’. Either way, Guinea-Bissau shows an absence of sovereign power compared with the drug gangs that either exceed its writ or can undermine it in a contest of arms.

But Guinea-Bissau’s ‘failure’ can be understood differently. The drug gangs in Guinea-Bissau are transnational commercial concerns that deploy violence, and the legality of whose actions are frequently in question. In this, they are not dissimilar to the East India Company in the nineteenth century, which was involved in trading opium, controlling territory, and signing treaties with Asian rulers. The heteronomous character of the East Indian Company was structural to the functioning of empires, which recognized and incorporated a variety of legal codes within treaty negotiations (Burbank and Cooper Reference Burbank and Cooper2010). Yet, the legality of these heteronomous arrangements was consistently questioned: Edmund Burke had excoriated the treaty-making of the East Indian Company in India as ‘geographical morality’ and Gladstone had decried the opium trade as ‘infamous contraband traffic’ in the House of Commons.Footnote ³⁰ In 1858, during the Second Opium War, Palmerston declared direct rule over India, but specified that the actual system of rule in India was not changing, just the ‘administrative organization at home’.Footnote ³¹ That is, Britain took over the control of India from the East India Company, at the same time that the Company was a participant in the Opium War. We are not making a relativistic equivalence between the British Empire and Guinea-Bissau but making a theoretical point: even the most established sovereign powers have origins in or work through the transgressions characteristic of sovereignty, which they disavow and encompass. But, as Guinea-Bissau suggests, even if every state has organized crime/sovereignty at its core, not all organized crime/sovereignty leads to a state/sovereign power.

As other scholars have pointed out, and our example bears out, a range of polities have historically been seen as sovereign (e.g. Ruggie Reference Ruggie1993; Nexon Reference Nexon2009). But does that mean that the empirical prevalence of weak states and the rise of non-state wielders of sovereignty mark a return to archaic forms of rule like despotism? If so, one would be justified in calling for a renewal of imperial practices to deal with heteronomous, ‘traditional’ spaces. We answer in the negative, because as we pointed out above, the emergence of modern states has involved now disavowed ties to criminals, warlords, and such figures: sovereignty is hardly pre-modern.Footnote ³² But we should also emphasize that trangressive acts by historically subjugated groups invoke modernity too, and hence we should be wary of resurrected imperial practices.

The transgression through which sovereignty is asserted is different from pre-modern forms of protest that either appealed to traditional authorities such as the king for redressal of grievances or consisted of localized acts of rebellion or marked an exit from the state. The first form of protest strengthened the vertical relationship between ruler and governed rather than transgressed the ruler’s writ, the second could be dealt with internally as criminal acts, the third marked a realm that refused existing institutions and was thus exterior to them, such as pirates or bandits. Such localized transgressions certainly exist, but it is the phenomenon we discussed earlier of transgression as normal politics that suggests contemporary modes of protest are qualitatively different from customary rebellions, local protests, or exit.

Transgression need not be exterior to the functioning of the state or a momentary irruption of protest that is repressed or bought off. Rather, by transgressing, historically subjugated groups mark their own originary exclusion from the law within the law: this is a quintessentially modern gesture. Transgression as normal politics involves a perpetual contestation of sovereign power in its ability to declare and enforce the law. By transgressing, such groups declare their own sovereignty, but this is not exterior to the law for two reasons. First, the self-understanding of subjugation is often itself an artifact of the law: a poor peasant in India is likely to mobilize as a member of his caste rather than as a member of the poor or as a peasant because there are constitutional protections and entitlements granted members of the lower castes, and caste-based political parties to pursue these claims. Second, and following, acts of transgression such as electoral violence are means of registering claims with the state. Both the idiom of transgressive mobilization and its goals are in relation to the modern state, indeed, shaped by the state. This means the category of ‘resistance’ as the opposite or exterior to sovereign power may be a misnomer as a description of at least some of the transgressions committed by subjugated groups (Scott Reference Scott1985). We would suggest that these be understood as declarations of sovereignty, performed periodically to mobilize a community which can then make claims on sovereign power.

Our more expansive concept of sovereignty – the always-present possibility of the repudiation of the reciprocity owed subjects – offers some purchase on understanding the exercise of power in a majority of states. It allows us to theorize transgression, both as it disrupts sovereign power to the point of dissolution in the failed state, and, more paradoxically, how it perpetuates sovereign power in the weak state. The relationship between state-making and organized crime is a complex one; the link can be constitutive, as Tilly argued, but it can also lead to an uneasy coexistence of law and transgression, or detract from state-making altogether.