Beyond Agamben: Sovereignty,policing and ‘permissive space’ in South Africa,and beyond

Agamben and everyday policing

Agamben’s work has been hugely influential in recent debates on sovereignty, generating a great deal of commentary, critique, and empirical work (see De la Durantaye, 2009). His theoretical starting point is the sovereign decision. To explore this ‘decisionist’ perspective and its implications, we need to grasp two figures whom Agamben saw as having structurally similar relations to the law: the sovereign and homo sacer or ‘bare life’ (Agamben, 1998: 84). Agamben used these figures to challenge Foucault. He argued that if we understood their relationship to one another, and to the law, we would realize that ‘bare life’ was the necessary product of the sovereign decision (Agamben, 1998: 6). For Agamben, this meant that the bio-political had always been bound up in the exercise of sovereignty. Foucault, by contrast, had argued that bio-politics emerged as a defining feature of specifically modern politics and could be theoretically distinguished from sovereignty or, in Foucault’s terms, ‘juridical-institution politics’. Agamben (1998, 2005) held that modern politics simply amplified a bond between bio-politics and sovereignty that was there all along. In the analysis that follows, I begin by exploring the position of the sovereign, and Agamben’s rendering of bare life, before analysing his take on sovereignty in contemporary times.

Agamben’s work drew heavily on Carl Schmitt’s (2006) ‘decisionist’ perspective on sovereignty. Law, Schmitt argued, governed interactions in normal life. The sovereign, he claimed, was the one who could declare that the normal circumstances, within which the law functioned, did not hold. Thus, the sovereign was in a ‘relation of exclusion’ with the law because he belonged to the legal system only in so far as he could decide when and if it needed to be suspended. This ability to decide upon the exception was not just constituted by the sovereign; it was constitutive of him (Agamben, 2005).

The entity produced by this sovereign decision, caught in its own relation of exception to the law, was ‘bare life’. For Agamben, bare life was not natural, unmediated life (zoe) but rather what was left when politically qualified life (bios) had been stripped of its legal-political protection (Agamben, 1998:109). In Ewa Plonowska Ziarek’s (2012) words, it was ‘the remainder of the destroyed political bios’. Abandoned by law, bare life was considered valueless and could be subjected to unmediated sovereign violence. This ‘inclusive exclusion’ of bare life by the sovereign, Agamben argued, was captured in the figure of homo sacer within Roman Law. In Roman times, ‘the sacred man’ was a citizen who had been stripped of his political rights by virtue of being outlawed. He could be killed with impunity by those inside the community but not sacrificed by them because, having been reduced to a bare life, he lacked the value necessary for sacrifice (Agamben, 1998). As Downey (2009: 111) explained, he was, therefore, ‘an outlawed citizen, the exception to the law’ but he was ‘still subject to the penalty of death and therefore still included, in the very act of exclusion, within the law’. Thus, for Agamben, the sovereign and homo sacer were the structural corollary of each other in relation to the law. To the sovereign, everyone was potentially homines sacri. To homo sacer everyone was a potential sovereign (Agamben, 1998).

Although Agamben’s theorizing owes much to Schmitt, the two philosophers differed on their interpretation of the state of exception. For Schmitt, the state of exception offered a means through which the law could claim that which previously lay beyond it: declaring a state of exception was a means of imposing nomos (legal order) on anomie, and ultimately either bringing it back into an existing (but temporarily suspended) constitutional order, or incorporating it within a new constitutional order. In temporary states of exception the ‘normative aspect of law’ had no force, while the ‘force of law’ had no normative basis in law (Agamben, 2005: 87). For Agamben, however, our current situation was one in which ‘the exception has become the rule’. ² What he saw was not a movement between law and anomie, but rather their reaching a point of indistinction. The state of exception, he argued, had opened up a ‘juridical void’ (Agamben, 2005: 43). While law and politics may continue to refer to one another, they were no longer ‘in relation’ and the law could not constrain the violence of the state (Agamben, 2005; see Huysmans, 2008: 173).

For Agamben, the paradigmatic example of this permanent state of exception was the concentration camp at Auschwitz. The emergence of this camp was not an anomaly, he suggested, but revealed a ‘hidden matrix’ ³ that was present in both liberal democracies and totalitarian states (Agamben, 1998: 166). Ultimately, what was of interest to Agamben was not the lives lived within Auschwitz but its juridical-political structure (De la Durantaye, 2009): in Agamben’s mind, the camp was any space in which the state of exception became the rule and bare life was systematically produced. It was because he focused on the juridico-political structure of camps that he placed Guantanamo Bay and the Bari stadium (where Italian police contained Albanians before their deportation) in the same category as Auschwitz. In the context of Guantanamo, he argued that the Patriot Act of 2001 ‘radically erase[d] any legal status of the individual, thus producing a legally unnameable and unclassifiable being […] the object of a pure de facto rule […] entirely removed from the law and from judicial oversight’ (Agamben, 2005: 30).

Agamben concluded that what Foucault saw as the emergence of bio-politics was, in fact, the emergence of the exception as the rule, which had shifted the ‘realm of bare life’ from ‘the margins of the political order’. ‘If today there is no longer any one clear figure of the sacred man’, Agamben (1998: 127) resolved, ‘it is perhaps because we are all virtually homines sacri’, caught within a permanent state of exception of which the concentration camp of Auschwitz was paradigmatic.

Taking exception to Agamben

While Agamben’s theories are thought provoking, they are also flawed. ⁴ These flaws are significant for scholars using Agamben to understand de jure states of emergency in South Africa (Hyslop, 2011; Kihato, 2007; Landau, 2006; Mosselson, 2010; Pillay, 2005) but they pose an insurmountable hurdle for those studying the de facto sovereignty practised in acts of everyday policing in the country, and beyond. Below, I focus on three key criticisms that undermine his case.

First, Agamben’s conceptualization of the ‘state of exception’ as a ‘juristic void’ misconstrued the relationship of law to violence. Guantanamo Bay, the Lindela Repatriation Centre in South Africa, and other camps are not ‘legal no man’s lands’ in which law has no power to mediate the state’s use of force (Agamben, 2005). To understand more accurately how law functions, we need to shift our terminology from one of ‘states of exception’ to ‘states of emergency’ (Neocleous, 2006, 2008). Providing a historical perspective on the evolution of the ‘state of emergency’ in Europe and the United States during the 20th century, Neocleous (2008: 58) demonstrated that ‘the “state of emergency” became perhaps the most common prescription in the pharmacopoeia of statecraft’ in both liberal and fascist states. Moreover, the legal powers of the state were often inflated as each new raft of emergency measures left their legacy to ordinary criminal law (Neocleous, 2006, 2008). It is this process of incorporation - the normalization of the state of emergency - that differentiates Neocleous from Agamben: ‘[f]ar from suspending the law’, Neocleous (2008: 70) argued, ‘violent actions conducted in “emergency conditions” have been legitimated through law on the ground of necessity and in the name of security’. Attempts to avoid legal condemnation are testament to the continued operation of the law in these moments.

This is certainly borne out in the history of South Africa. One of the most remarkable features about apartheid policing, for example, is that ‘not only [was] much of what South African Police [did] perfectly legal, but what they [did] legally [could] be extraordinarily brutal’ (Brogden and Shearing, 1993: 27). The Criminal Procedure Act 51 of 1977 section 49(2) allowed the police to shoot and kill suspects to prevent them from escaping. Suspects did not need to be forewarned, they simply had to be ‘reasonably suspected of having committed a serious offence’ such as petty theft with trespass (Brogden and Shearing, 1993: 27). For collective gatherings, the police could shelter under the Internal Security Act of 1982, which permitted them to use firepower to disband illegal gatherings (Haysom, 1990). A combination of section 50 of the Criminal Procedure Act and section 29 of the Internal Security Act of 1982 gave the police legal cover to detain people for as long as they wished (Brogden and Shearing, 1993: 31). During the Sharpeville Massacre of 1961, the Soweto Uprising of 1976, and the States of Emergency in 1985–1986, the police were protected from civil or criminal proceedings for their actions, providing they had acted in ‘good faith’ (Prior, 1989). In 1986 alone, the South African police killed a recorded 710 people (Foster and Luyt, 1986). Legislation was made all the more accommodating by judges who were usually willing to interpret the law leniently (Gordon, 2006).

Nor is this legal work simply a charade that has, in practice, ‘shed every relation to law’ (Agamben, 2005: 59; see Brännström, 2008; Humphrey, 2004). The importance of the legal parameters set in a state of emergency was demonstrated by George Karekwaivanane in his work on Rhodesia. Karekwaivanane highlighted the work of lawyers who held the state to account when it stepped outside the bounds of its own emergency legislation. Despite the draconian environment of 1970s Rhodesia, Karekwaivanane (2011: 15) argued that the law remained an important ‘mode of rule for the settler state’. David Anderson (2012: 700) has made a similar argument about emergency rule in Kenya. Studying British abuse and torture in Kenya’s uprising between 1952 and 1960, he showed how the British government sought to deny widespread torture, while ‘Kenya’s colonial violence workers’ ‘struggled to present torture as lying within the “rule of law”’. In other words, illegality and legality mattered. What is more, as the recent Mau Mau trials in London have demonstrated, the law has continued to matter, over 50 years later. ⁵ Thus, we should see law as the outcome of political contestation, which can facilitate as well as limit the exercise of state violence (see also Brännström, 2008). Moreover, while the state may sometimes act illegally without repercussion, this does not make the law utterly devoid of its relation to life.

A second key critique of Agamben’s work concerns his attempt to ‘correct’ Foucault. In attempting to synthesize the juridico-institutional politics of sovereignty and bio-politics, Agamben ultimately lost sight of bio-politics. Foucault (2002: 138) claimed that ‘bio-politics’ was a productive drive to optimize and multiply life. This is fundamentally incompatible with Agamben’s notion of bare life, which could either be taken or allowed to persist. Bare life fits completely, and exclusively, within the rationale of juridical-institutional politics. Moreover, by arguing that a single decision determined the shift between inclusion proper and the ‘inclusive exclusion’ that created bare life, Agamben depicted ‘bare life’ as undifferentiated. If it is indeed true that ‘everywhere on earth men live today in the ban of a law and a tradition […] that include men within them in the form of a pure relation of abandonment’ (Agamben, 1998: 51) then we are led to question whether the ideas of ‘homo sacer’ and ‘bare life’ hold any analytical significance whatsoever. As Brännström (2008) asked, what do I have in common with those interned in concentration camps, neomorts being kept alive so that their organs might be transplanted, and motorists killed in motorway accidents? And what do they have in common with one another? It would seem that ‘even if all subjects are homines sacri they are so in very different ways’ (Lemke, 2005: 7–8). Agamben’s failure to recognize this adequately, Lemke (2005: 7–8) argued, can be traced back to the fact that he was concerned with thanatopolitics, not bio-politics: he was ‘less interested in life than in its “bareness”’.

What is more, the ‘bareness’ to which he referred was primarily legal: Agamben was not concerned with the resources and strategies that remain when political-legal rights have been stripped away. In contrast, Caroline Humphrey (2004: 420) has argued that people in everyday life ‘do not simply acquiesce to the menace of sovereignty but interpose a solid existence of their own that operates collaterally or against it’. It is, therefore, vital to explore people’s own world-views and life histories, as well as the wider social, political and economic landscape of which they are a part, to understand how sovereignty functions in everyday life, a point to which we return.

In fact, Agamben’s underlying critique of Foucault was largely unwarranted: Foucault did not lose sight of juridico-institutional politics in his analysis of modern states. Rather, he claimed that both were present, creating a ‘demonic combination’ (Ojakangas, 2007: 26). Where he differed from Agamben was in his belief that the two modes of power could not be reconciled (Ojakangas, 2007: 26). The advantage of Foucault’s analysis of power is that he did not believe its exercise was reducible to a singular mode or institution (Jessop, 2006), allowing us to look at the law and the state but also beyond them. ⁶ In contrast, Agamben was transfixed by the state and ignored its reception in society. But, as Barbara Honig (2009) reminds us, even sovereign decisions rely for their effect upon the ‘orientation’ of the ruled towards the ruler. This is borne out by works on the latter years of apartheid in South Africa. In the township of Alexandra, for example, Jonny Steinberg (2008: 74) argued that life was disrupted by violence on all sides, creating a ‘hunger for authority’. Among some township residents this created a conflicted orientation towards the instigators of ‘state violence and depravity’ who offered a brutal, contorted form of the order people desired (see also Dlamini, 2009). Ultimately, struggles over law and social order in apartheid South Africa were infinitely complex and were mediated through a variety factors, including class, race, gender, and knowledge (Bozzoli, 2004). These are all divides that Agamben ‘ontologically erase[d]’ in his theories (Huysman, 2008: 165). In doing so, he closed any theoretical space for understanding political contestation, domination, and resistance in society that does not produce anomic life (Papastergiadis, 2006: 437).

These flaws in Agamben’s work have compromised attempts by scholars to apply his insights to practices of everyday policing in South Africa. Their ideas of ‘civic states of exception’ (Kirsch, 2010) or ‘spaces of exception’ (Buur, 2003) suggest that citizens and state actors are able to act with an impunity that can simply be seized or decided upon. Certainly, in South Africa (as elsewhere), attempts to make the de jure mirror the de facto have often proved either ineffective or unwelcome. Nonetheless, the presence of the law offers the possibility of recourse. Equally as important, the law does not capture the spectrum of ways in which people may seek to defend themselves or gain recompense from others. Employing the language of exception encourages us to ignore (rather than explore) the rights, resources and relationships that people utilize to leverage a response, either inside or outside the state. Ultimately, it is precisely the difference between a state of exception and the negotiation of de facto sovereignty that should occupy scholars of everyday policing: how and why do citizens negotiate (with state officials and others) the ability to exercise violence without repercussion? In what ways and why are these negotiations affirmed, challenged or undermined? We need a subtler approach than Agamben’s to make sense of such questions.

Sovereignty and permissive space

Rather than focusing on the exception, a la Agamben (1998, 2005) or Schmitt (2006), I draw on conceptions of sovereignty that are concerned with everyday life, which contains both norm and exception. In doing so, I ‘question the quest for the ultimate “one” that decides on the exception’. Instead, I cast an analytical spotlight on ‘the multiple layers and practices involved in sovereignty’ (Bottos, 2009: 109). This is best captured, I will argue, in the concept of ‘permissive space’.

At its root, sovereignty can be understood as a historically situated claim to authority over life, death, and everyday conduct that is always ‘grounded in violence’ but which may or may not be grounded in state law (Hansen and Stepputat, 2006). Although the capacity for violence is a central component of sovereign power, the preceding analysis has demonstrated that focusing purely on the sovereign decision and the exercise of violence does not provide us with a full account of how sovereignty operates in practice. To do so, we must explore the social, political and economic forces that shape it. For, as Robert Latham (2000: 3) argued, what ‘makes paramountcy possible are the bodies of relations that effectively structure practices and agency in any given area of social life’. These are not all reducible to force and fear. The acceptance, adaptation or rejection of sovereignty cannot be understood without exploring the structures, relationships, and ideas that make up everyday life in all their political, economic and socio-cultural complexity (Honig, 2009; Humphrey, 2004).

All of this points to the fact that sovereignty is not simply a capability that can be possessed. Rather, it is a claim that is under constant negotiation against the claims of others and with those over whom it would be practised (Hansen and Stepputat, 2006; Humphrey, 2004). It follows that de jure state sovereignty may not represent a meaningful claim to sovereignty in everyday life. In South Africa, however, I would argue that de jure state sovereignty continues to play an important, if inconsistent, role in everyday policing. The constitution, as Loren Landau (2010: 217) has convincingly argued, only has a ‘partial hegemony’ over its citizens and ‘irregularly determines the nature of socio-political interactions’. Nonetheless, this partial hegemony remains important: it guides everyday negotiations over who can legitimately act as the ultimate political authority and, relatedly, who is allowed to exercise force, how, and against whom. Any analysis of sovereignty in South Africa needs to incorporate this insight.

Caroline Humphrey’s (2004) concept of ‘nested sovereignty’ comes close to capturing this perspective. In Siberia, she argued, ‘localized forms of sovereignty’ were contained within other forms of sovereignty, such as national sovereignty, while ‘retain[ing] a domain in which control over life and death [was] operational’ (2004: 420). ‘Nested sovereignty’ is useful in this context because it allows us to recognize the fact that, in South Africa, the state is often able to exert a hierarchical claim to sovereignty on the basis of the legitimacy it holds in law and its coercive power. ⁷ The situation, however, is less ordered and rigid than Humphrey’s term would suggest, and we cannot always assume that hierarchical claims made by the state are considered feasible or desirable by state officials or others. Given this shortcoming, I believe that we need a new conceptual language to explore the negotiation of de facto sovereignty. I use the remainder of this article to introduce the term ‘permissive space’, explain its parameters, and use examples from my own research in Durban to illustrate its applicability.

The parameters of permissive space

The notion of permissive space captures the idea that those who act illegally must negotiate their impunity with other citizens and the state, so as to avoid repercussions for their actions either within or beyond the criminal justice system. Within permissive space, illegal practices are allowed to continue unabated and may be conferred with legitimacy. Negotiations over permissive space shape the imaginaries, boundaries and practices of states and societies across the globe (Heald, 2005: 279, 281; Sieder, 2011).

The concept of ‘permissive space’ captures a double notion of permission and permissiveness: it allows for the possibility of constraint and excess to be simultaneously at stake within the same act. For George Bataille, sovereignty was defined by excess for its own sake. He believed that ‘life beyond utility is the realm of sovereignty’ (Aksoy, 2011: 219). Consequently, the sovereign actor was only concerned with the present, beholden to neither the past nor the future. In contrast, the idea of permissive space re-situates even the most excessive sovereign performance in the fray of life, to be shaped by the web of relationships in which it is enmeshed and executed with an eye to how these relations will develop over time.

The notion of permissive space is a flexible one. Those within permissive spaces can act as sovereigns but this sovereignty is as fragile as the negotiated space itself. The tone of the negotiations that create permissive spaces range from amicable to antagonistic. Negotiation, in this context, simply emphasizes that permissive spaces are not unilaterally imposed, nor are they immutable. Instead, they are established in an ongoing exchange. This exchange need not be symmetrical and the ‘consensus’ reached in such negotiations may be fraught and shallow. In some cases, the borderline between ‘decision’ and ‘negotiation’ may appear thin. However, to some degree, impunity is always unstable, being open to the possibility of (re)negotiation at different times, scales and spaces.

Using the concept of permissive space, we expose ‘negotiation[s] over the logics of authorization and subjection that stitch together the polity’ in South Africa, and those elsewhere (Smith, 2011: 416). Such negotiations are present across all sections of society. The negotiation of permissive spaces, for example, is a necessity for poorer South Africans who rely on the fluidity and flexibility it provides to make their lives and livelihoods tenable. However, negotiation also suits the desires of the affluent, who can utilize this same fluidity to press their privilege further. Therefore, unless one believes that legitimacy and legality are inherently bound, the process of negotiation in and of itself is amoral.

Unlike the decisionist concepts explored above, the concept of permissive space provides us with the conceptual room to avoid state-centric approaches, without assuming that the state is unimportant. It allows us to recognize the potential significance of actors inside, outside and on the constructed borders of the state. It also enables us to explore the multiple forms of power each may hold. In so doing, the concept of permissive space acknowledges that the location of sovereignty is ‘never self-evident’ (Kelly and Shah, 2006: 253). This is not just because it is ‘dispersed’, as Kelly and Shah (2006) argue, but also because claims and counter-claims can be concentrated within the same, conflicted individual. The police who are charged with making state sovereignty a reality, for example, have complex motivations as ‘civilian/citizens’ and state officials. Their interactions reveal a ‘play of force and sentiment […] social contract and individual conviction’ (Chalfin, 2008: 520, 2011). At some moments they may seek to create or reinforce permissive space and at others they make work to destroy it. Over the course of my research in Durban, I witnessed several situations in which the police oscillated between these two positions – their willingness to support legal infringements waning as citizens pushed for action to be taken or increasing in response to normative strictures, political pressure or economic reward. The concept of permissive space allows us to explore all of these possibilities.

The theorising above has a conceptual affinity with nodal policing, which also insists that the state should not have ‘conceptual priority’ (Shearing and Wood, 2003: 404) in our analysis of everyday policing (Johnston and Shearing, 2009; Marks et al., 2009). This body of work has concentrated on the mapping emergent policing nodes (spanning state institutions, the private sector, and society) and delineating their interactions with one another both in South Africa, and beyond. My work on permissive spaces can contribute to this broader literature by providing a conceptual tool with which to explore the negotiation of policing that breaks state laws. ⁸ While the application of permissive space is focused on policing here, the concept can also be applied to other forms of law breaking in South Africa, and beyond.

In the sections that remain, I will first explore how we can conceptualize the parameters of permissive space shifting in the midst of negotiations that can establish, extend, contract, close or collapse these spaces. My final section delves deeper in to the notion of space, exploring the relationship of permissive space to physical space in South Africa.

Permissive space: Establishing, contracting, closing and collapsing

Permissive spaces can be relatively broad in the acts they allow and relatively stable over time. Conversely, they may also be narrow, fleeting and unstable. Much depends on the degree of permission granted and the power of those granting it: both of which are mutable. Take private road closures in Durban, for example. The degree to which road closures or ‘boom-gates’ are opposed, facilitated, permitted or tolerated varies greatly from municipality to municipality. Durban is located within eThekwini municipality. At the turn of the century, eThekwini’s opposition to road closures was fierce, driven by then City Manager Michael Sutcliffe. Echoing the findings of the South African Human Rights Commission, eThekwini municipality stressed that road closures contravened people’s freedom of movement. Strong rhetoric and legal action faced anyone who infringed this ban. When I interviewed the city manager 10 years later, his opposition to road closure remained. In the interim, however, the situation on the ground had changed. Sutcliffe’s attention drifted onto other matters, not least Durban’s hosting of the 2010 World Cup. He had, in his words, ‘other headaches to deal with’.

As the City Manager’s anti-closure campaign waned, those who lived on side streets and dead-ends had the opportunity to negotiate a relatively stable permissive space for action with their local police and councillors. In diverse areas across Durban—from former townships to former suburbs—residents had notable success in this regard. In public, the police and councillors would oppose these exclusionary practices but in private, several had agreed with their constituencies not to act unless they received public complaints on the matter. Given that private security companies had effectively normalized the contravention of people’ s constitutional right to free movement in these areas (see Berg, 2007; Kempa and Singh, 2008), such complaints were not forthcoming. Therefore, residents and councillors alike acknowledged that their negotiated compromise had created a stable permissive space. This was why several individuals that I spoke to referred to the practice as ‘illegal in inverted commas’: road closures had become a ‘public secret’ (Taussig, 1992), slowly proliferating across the municipality.

As the case above demonstrates, the breadth and stability of permissive space often shifts meaningfully over time. In this example, permissive space had expanded and solidified. The converse, however, was also possible: permissive space could contract, covering fewer acts or less visible excess than it had before. In the formerly black township of KwaMashu, for example, many interviewees had witnessed violence by the police against criminal suspects. Their recourse against these officers was limited, not least because any case rested on the word of a complainant against that of a police official. The advent of mobile phones capable of capturing video, however, had brought important (although incomplete) shifts in this power dynamic. In recent years, residents argued, police officials had increasingly moved their violence and abuse into less visible spaces. Sometimes, they had shifted into people’s houses, where they could be more certain that witnesses endorsed their violence (Steinberg, 2008). At other points they moved into police vans and cells where their actions remained largely impervious to the advent of new technologies. Even in a country that has a very poor track record of punishing ‘police vigilantism’ (Bruce, 2002), these state actors felt their permissive space constrict.

If a permissive space constricts completely, we can say that it has closed. In such cases, those who were previously covered by a permissive space will find themselves subject to punishment if they act illegally in the future. This process of closure can occur relatively quickly, as it did when people came to the aid of suspects being targeted by ‘street justice’, ⁹ changing the calculations of those who would have continued assaulting the suspect in question. To understand this process in more depth, let us focus on street justice in KwaMashu.

In KwaMashu, street justice is often imagined (in the media and by local residents) to be the work of an anonymous, sovereign mob attacking an alienated, evil criminal. In practice, however, street justice is more likely to be the work of an ‘intimate crowd’ (Cooper-Knock, 2014): a collection of residents who are bound by multiple, interpersonal relationships, which may tie them to others in the crowd and to their victims. Consequently, the permissive space in which street justice can occur is the product of contestation and compromise with other residents and the state.

In some instances, permissive space was closed by the police who were called by observers or by participants in the assault. As I have argued elsewhere, narratives of ‘mob justice’ depict the police as being effectively absent from street justice in South Africa’s townships, but this is often untrue (Cooper-Knock, 2014). In fact, people frequently made recourse to the bureaucratic and coercive power of the police, on the understanding that police action would be limited to removing the subject of an assault and that they would not attempt to prosecute those involved in the act. In other words, the police were called to close a permissive space but not to collapse it—a move that would have jeopardized the impunity of those who had already acted. To avoid collapse was to show the ‘necessary respect to the crowd’, in a context where many claimed they were acting in lieu of a state that seemed unable or unwilling to do so. As Sizamile, a middle-aged resident reflected on one recent instance of street justice:

What happened? The usual. Once you fall into the hands of the community, you are given a hiding and the police are called in to take you away. And you are fortunate if the police are quick enough to come and take you away.

In such cases, the closure of permissive space marked a point in time at which an illegal act was brought to an end. However, closure could also have more wide-reaching consequences for the feasibility of similar acts in the future, for the impunity of particular actors, or for the use of specific spaces.

Of course, to argue that people hoped or expected the police to close rather than collapse permissive space is not to claim that this always occurred. Police officials are often keenly aware of the tacit arrangements that evolve with certain groups or in certain spaces, which shape the authority they can meaningfully exercise or the acquiescence that they can expect (Steinberg, 2008). But that is not to say that they always adhere to these arrangements. For a mixture of institutional and personal reasons they may take actions that thwart the efforts of local residents to negotiate permissive space, earning their ire and sometimes their active resistance. In the formerly white suburb of Berea, for example, residents had hired a private security guard called Innocent to protect their road. One afternoon, Innocent confronted two men who were attempting to break into cars on the street. The men responded by attacking him. In reaction, two residents attempted to defend Innocent and punish the would-be car thieves by assaulting them with golf clubs. Afterwards, when a neighbour drove Innocent to the police station to report the attempted theft, they were berated by the police and informed that a case had already been opened against all involved by those they had attacked. As my research drew to a close, they were still facing criminal charges. Back in KwaMashu, interviewees recounted similar legal action (see also Cooper-Knock, 2016).

Given the dominant narrative of police inefficiency in South Africa, the acknowledgement that police can close permissive space—and were regularly called to do so—is theoretically and empirically significant. Similarly, it is important to recognize that the police had the capacity (albeit more limited and rarely attempted) to collapse permissive space. But, of course, the police were not the only operatives who could perform these manoeuvres: permissive space could also be closed by actors and processes based primarily outside the criminal justice system. ¹⁰ In KwaMashu, for example, the family and friends of those being assaulted could close permissive space by threatening the moral or physical integrity of those committing the assault should they continue.

Alternatively, the suspect under threat might have a sufficiently violent reputation to protect them. This was certainly the case for a drug dealer in Chatsworth called Anjeet, who utilized extreme violence (bolstered by selective philanthropy) to secure his permissive space. Anjeet had emerged from the loosely formed gangs in his area (Glaser, 2000, 2008; Pinnock, 1984). Over multiple decades he had bought off, intimidated or killed anyone who might threaten him or his drugs network. Otherwise, he had developed a reputation for philanthropy in the sector, holding feeding stations, distributing Christmas hampers and paying the electricity bills of those who stored his drugs. The most potent of these was Sugars—an impure opiate—that had taken hold of the lives of many young residents, known locally as ‘Sugar Boys’. While there were other drug dealers in the area, none were as dominant or operated in such a stable permissive space as Anjeet and he played a crucial role in mediating the permissive space of others.

Alvin, for example, was a drug addict whose family stored drugs for Anjeet. These two factors played an important role in shaping the permissive space that he could negotiate. Alvin’s capacity to secure permissive space for himself or constrain the permissive space of others was largely compromised by his status as a ‘Sugar Boy’. The archetypal Sugar Boy was, at once, an object of disdain and fear in Chatsworth: ¹¹ their desperation for drugs had placed them on a path towards physical decline and destitution but this same drive could increase their propensity to engage in acts of violence.

Labels, Ray Abrahams (2008: 433) once argued, ‘often have a lethal quality’. That was certainly true in this sector, where labelling someone a Sugar Boy was a powerful speech act (Jensen, 2008, 2010). Undoubtedly, some residents were afraid to take any action against the Sugar Boys who stole from them. As Nileen, a resident in her 50s insisted, ‘they would kill me, my baby, they are raw’. The numerous residents who were willing to take action, however, did so in the knowledge that the police would likely share their disdain for Sugar Boys. Stories were rife in the sector of such individuals getting ‘slaps’ from the police who had ‘no respect for smokers’. Consequently, attempts by drug addicts to open cases against their assailants could often be easily undermined.

Ronald, for example, was a resident in his 50s. When he tried to move on Sugar Boys who were smoking on his stairs, they attempted to fight back physically. In response, Ronald assaulted them. When one of the young men attempted to open a case against Ronald, he countered by using the claimant’s addiction to denigrate his character: ‘I told the policeman I never handled him, he is drugged up, he is telling lies, he is trying to get away.’ Whether this policeman believed Ronald’s account, sympathized with him, or simply wanted to avoid mediating a messy conflict, he supported Ronald’s attempt to secure a permissive space: the case was never opened.

But if Alvin’s status as a ‘Sugar Boy’ weakened his hand in negotiating permissive space his place within Anjeet’s business empire potentially strengthened his leverage. Generally speaking, the degree to which Anjeet was willing to lend his weight to battles over permissive space depended on the degree to which he lent upon those involved for economic, social or political traction. He was not, for example, willing to jeopardize the tacit communal support he had accrued through his selective philanthropy for those on the fringes of his network who were egregiously undermining law and social order in the area. In fact, he was willing to actively punish some of his customers and runners personally when they were seen as stepping out of line with laws and local norms. As Simeon, a resident in his 30s, claimed:

If maybe your son tend to fight, or tend not to be controllable in the house and if he was breaking things, if you report, he will send his gangs, they will sort out your child, yeah really they will give him [slaps hands] that discipline, you know he try to maintain order.

Alvin was far from indispensable to Anjeet but he was useful. As such, the leverage he accrued via Anjeet in negotiations over permissive space was meaningful but limited. This leverage became evident in an altercation with Ronald. In this instance, Alvin’s family could not keep the permissive space that they desired open, but they could determine how it collapsed, and with what consequences. On the day in question, Alvin had snatched the phone of Ronald’s son, Neville. Recognizing that it was best to minimize any conflict with those linked to Anjeet, Ronald peaceably appealed to Alvin’s wife to return his phone. Unfortunately for Ronald, the phone had already been sold on to the local pawnbroker, who enjoyed even closer relations to Anjeet. Opening a case against the pawnbroker could threaten Anjeet’s business network and would certainly provoke a response from ‘Anjeet’s boys’. However, Ronald felt he could risk opening a case against Alvin. When the police were ‘taking their time’ in investigating his claims, Ronald personally visited the station and demanded that a police officer accompany him to accost Alvin. Once they had been mobilized, the police were reportedly keen to prosecute this Sugar Boy who had a string of previous allegations against his name. At this point, however, Alvin’s wife re-emerged, reportedly stating, ‘Please, we know you, you also know my… family, drop this charge and I will make sure that I will buy a phone and give it to you.’ To ‘know’ a family in Chatsworth was to understand the ways in which their socio-economic circles were imbricated with Anjeet’s networks. Recognizing this statement as the veiled threat that it was, Ronald dropped the case in return for a replacement phone: he realized that he could secure the collapse of Alvin’s permissive space, but not on the terms of his choosing.

In sum, this section has demonstrated the ways in which permissive space could be established, expanded, contracted, closed and collapsed. Ultimately, these processes could involve a mix of actors, including state officials and those who traversed and blurred the constructed border between state and society. The repertoires on which people drew were also varied, including discourses of legality, morality, politics, and identity. Negotiations were shaped by the broad contours of post-apartheid South Africa, but they were also deeply rooted in the idiosyncratic socio-cultural, political and economic landscapes of the local areas in which they emerged. Over time, people had forged expectations about what should and could be achieved in these negotiations. Such ideas were important but they were always subject to revision as different actors sought to shift their terms of engagement. Ultimately, who was punished, for what, by whom, was always subject to renegotiation.

Space and sovereignty

Permissive space operates in dialectic with physical space. As geographers like Derek and Urry (1985) have highlighted, we need to incorporate the analysis of space into our understandings of everyday life. On one level, space is an ‘arena in which social life unfolds’: physical space is the stage on which illegal acts occur and permissive spaces for those acts are contested but it is also a ‘medium’ through which relationships and realities are (re)forged (1985: 3).

There are at least three key ways in which physical space forms a dialectic with permissive space. First, space can produce as well as reflect interpersonal connections between groups and individuals, shaping the ‘bodies of relations’ on which sovereignty rests (Latham, 2000). At the broadest level, the sectors that I studied were still inflected in important ways by the apartheid government’s remapping of race and place. The fact that these areas were racially zoned under the National Party shaped ideas of order and disorder, security and insecurity. The changes that have happened since the fall of apartheid have variously amended and amplified these associations. All this has important implications for who tends to be targeted by acts of everyday policing, and by whom. Narrowing our focus down further, the forms of housing in which people live, from high-rises to single-storey buildings, from gated estates to informal settlements, also shapes the parameters of the possible when it comes to permissive space.

Second, authority is always exercised in space, and delineated by it. Within the urban landscape of eThekwini, there were spaces in which certain people could exercise a stronger claim to authority. In some cases, these claims were relatively narrow, and tied to an individual’s profession or position: the head of a Resident’s Association in her block’s yard, for example, or a police officer inside his police station. In other cases, people could draw on broader claims to authority that were made possible through historical memory or metaphorical association. The historical image of the ‘sovereign mob’ on the streets of a township during apartheid, for example, remained a powerful one (see Smith, 2012). It strengthened the hand of residents negotiating permissive space in the present day, even though their negotiations were simultaneously constrained by the socio-political realities of everyday life.

Finally, space was important in the sense that urban form created spaces of visibility and invisibility, which have long shaped practices of statehood and sovereignty (Baviskar, 2004; Zukin, 1991). People are far more likely to be able to act outside the law if their actions are hidden because they do less to destabilize the performance of state power, even if they destabilize its practice. In Berea, Chatsworth, and KwaMashu, this realization shaped the behaviour of neighbourhood watch patrols who were seeking state support. Knowing which state officials and citizens would be most offended by, and likely to respond to, visible transgressions of state sovereignty required an intimate knowledge of the players involved and their shifting perspectives, which was beyond many patrol members. Therefore, many of those on patrol attempted to perform the role of law-abiding citizens in visible, public spaces through a range of tactics, such as donning uniforms, writing constitutions, wearing badges, and declaring that they would only ‘hold suspects until the police came’. Conversely, they sought to contain any transgressions to less visible spaces where witnesses were few or known to be sympathetic. Where this was not possible, they realized that their ability to negotiate a permissive space for any transgressions would be more challenging.

A similar picture emerges from Lars Buur’s (2009) work in Port Elizabeth, South Africa. He focuses his attentions on an organization called Amadlozi who sought to police and punish criminal suspects in the township. Some of the police and citizens in their area, however, rejected any obvious displays of illegal violence. Consequently, members of Amadlozi used the local landscape to hide the punishments that they administered, making them palatable to state actors, and less likely to provoke a legal response.

Conclusion

Despite its popularity in the literature on policing, I have argued that Agamben’s decisionist approach to sovereignty and his concept of the ‘state of exception’ are too uni-dimensional, binary and legalistic to be of use. At its root, sovereignty can be understood as a historically situated claim to authority over life, death and everyday conduct that is always ‘grounded in violence’ but which may or may not be grounded in state law (Hansen and Stepputat, 2006). To understand the operation of sovereignty in daily life, we must comprehend the ‘bodies of relations’ on which claims to authority are made (Latham, 2000), and the counter-claims with which they must contend. In order to do so we need a new conceptual language that facilitates us in exploring the negotiation of de facto sovereignty. To this end, I have introduced the notion of ‘permissive space’.

Permissive space is space of impunity, which can be expanded, contracted, closed or collapsed in negotiation with those inside, outside, and in the shadows of the state. Such negotiations can be antagonistic or amicable, and at times they may barely seem like negotiations in any meaningful sense. However, any concept tied to de facto sovereignty must allow for mutability over time and space, and between different audiences. This is what the concept of permissive space provides.

Analysing why different actors engage in particular negotiations, as well as the terms and consequences of their exchange, tells us a great deal about the factors that shape the operation of law and social order in South Africa, and beyond: how do people understand their own (in)security? What are the archetypal criminals that haunt them? What forms of protection do they seek, why, and with what success? What ideas about authority have traction and why? Which relationships are most important in shaping everyday policing? The concept of permissive space enables us to search for concrete answers to these questions while also allowing us to capture all that is ambiguous and amorphous at the heart of everyday policing.

Although this article has focused in on South Africa, the concept of permissive space is applicable to the study of everyday policing elsewhere. The work of Philippe Bourgois (1996) in the USA; Didier Fassin (2011) in France; Denyer Willis (2015) in Brazil, and Beatrice Jauregui (2016) in India all highlight the importance of illegal policing practices within and beyond the state. The diversity of their collective work confirms that the parameters of permissive space cannot be assumed. We can only map these parameters as they emerge from the history of a particular space, and the socio-cultural, political and economic realities of the present. Going forward, comparative ethnographies of permissive spaces can help us to delineate the different registers of authority and modes of contestation that operate across the globe, fundamentally shaping everyday policing and the relationship of law to life.