The triangle of security governance: Sovereignty,discipline and the ‘government of things’ in Olympic Rio de Janeiro

Abstract

This article reconsiders contemporary urban security governance. Conceptually, it revisits Foucault’s governmentality lectures to comprehend how security governance is carried out in places where the use of digital security technologies co-exists with overly lethal and repressive forms of policing. The author advances his analysis by conceptualizing a triangle of security governance in which disciplinary powers of control, apparatuses of security and sovereign/necropower are at work simultaneously, complemented by a fourth dimension that takes into account what Foucault outlined in the lectures as the ‘government of things’, which is the sociotechnical relationship between the agency of humans and machines. Empirically, the article explores the technopolitical turn in urban security policies in the city of Rio de Janeiro in the wake of the 2014 World Cup and the 2016 Olympics. Using discourses that are embedded in globalized mega-event security standards and legacy claims, authorities in Rio promoted a narrative of new material and non-material security measures that were intended both to secure the World Cup and the Olympics and to help overcome permanently entrenched urban conflicts in the city. By critically analysing these two approaches of new material and non-material security measures, the author shows how new security technologies are perfectly integrated into a continuum of death politics in Rio de Janeiro in ways that are conceptually best appreciated by considering how the triangle of security governance works in the digital era.

Introduction

Increasingly, Big Data, algorithms and new technologies are integrated in everyday policing routines. Promising greater efficiency, effectiveness and a reduction in violent conflict, smart and predictive policing strategies, however, co-exist and intermingle with repressive and lethal forms of state violence. This article investigates these antagonistic, but at the same time perfectly compatible, security strategies by conceptually placing central emphasis upon Foucault’s (2007) lectures at the Collège de France between 1977 and 1978, published under the title Security, Territory, Population. In these lectures, Foucault lays out his interpretation of security governance by advancing the idea of a governmentality triangle, contrasting different sovereign and disciplinary powers and his notion of the mechanisms of security apparatuses. In recent years, critical security and surveillance scholars have revisited Foucault’s lectures to thoroughly analyse how security and digital surveillance politics play out in urban security governance (e.g. Fussey, 2013, 2015; Klauser, 2013, 2017). In these contributions resonates the common belief that what Foucault theorized as ‘sovereign power’ has receded into the background, and that discipline and the mechanisms of security apparatuses are much more dominant in and helpful for understanding security arrangements in contemporary societies.

Recently, however, a growing body of postcolonial literature essentially questions whether this shift from spectacular punitive forms of sovereign state power to more subtle practices of discipline, security and forms of societal control is sufficient for understanding the twinning of state violence and digital security technologies in places that are marked by extreme state violence and social discrepancies (e.g. Alves, 2018; Pauschinger, 2020). Even if Foucault himself never really engaged with the postcolonies conceptually, his writings have impacted heavily upon the analysis in and about so-called Global South contexts. I therefore propose two novel theoretical contributions to the debates on whether we can really make sense of the functioning of contemporary security governance without considering sovereign power, disciplinary regimes and security apparatuses all together.

First, I argue for an understanding of Foucauldian security governance arrangements in places of intense state and non-state violence as a triangle, one that simultaneously mobilizes and consists of the three elements: those of disciplinary power, apparatuses of security and modes of sovereign power that are best understood in terms of Achille Mbembe’s (2003) essay Necropolitics, in which he expands his concept of ‘necropower’ that entails socio-spatial patterns in which the unconstrained sovereign deliberately decides who lives and who dies (see also Medeiros, 2018).

Second, I suggest that the Foucauldian triangle of security governance framework must be complemented by a fourth dimension that considers the increased use of digital security technologies and takes into account what Foucault outlined in the lectures as the ‘government of things’ (Lemke, 2015, 2021), which is the sociotechnical relationship between the agency of humans and machines. It is the government of things that helps us essentially to comprehend how the triangle of security governance works in the digital era.

To empirically explore how the triangle of security governance and a government of things are put to work and complement each other, I draw upon research undertaken before, during and after the 2014 World Cup and the 2016 Olympics with security forces in Rio de Janeiro. In barely any other place is the pairing of the digital and the lethality of security governance as explicit as in the former Olympic city. The empirical focus on Rio is particularly interesting since the mega-event period introduced, proliferated and combined digital security technologies and community policing strategies with overly violent and repressive ways of policing, and the city has thus become the ‘perfect’ laboratory for the implementation of global security trends in extreme violent urban environments.

On the one hand, Rio’s security governance strategies for the mega-events focused on both military occupation and community policing strategies like the introduction of the well-studied pacifying police programme in 38 peripheral and favela neighbourhoods (see, for example, Richmond, 2019). On the other hand, the introduction of digital security programmes such as camera surveillance schemes, command and control centres, predictive policing and urban control systems were also prioritized, as these were also meant to integrate the widely dispersed and competing police forces (Cardoso, 2013, 2019). Both of these strategies were intended to interrupt the vicious circle of violence to guarantee a peaceful staging of the mega-events and to leave the city with what in the standardized Olympic jargon is called a long-lasting ‘security legacy’ (Armstrong et al., 2016).

In many of my interviews with Brazilian mega-event security officials, the legacy was framed as twofold. As the extraordinary secretary for mega-event security said in an interview for my research:

I think the great legacy that remains is the material, the investments, the [security] equipment that I mentioned here. However, [there is also] the non-material legacy [that] stays, which is this junction of efforts of all the security forces. (Interview 1) ¹

In the analysis presented in this article, I investigate how these material and non-material security strategy and legacy narratives played out in Rio de Janeiro during and after the mega-events. I argue that beside these two officially proclaimed material and non-material frames, we must consider a third form of security governance outside of these official claims, one that co-exists and prevails in Rio de Janeiro in material and non-material forms, renewing and intensifying a deadly reality as the ongoing death politics that complements the triangle of security governance and as a legacy of in/security – here understood in terms of how security for some often means in/security for others (Jusionyte and Goldstein, 2016: 4).

For the analysis of this article, I draw upon ethnographic fieldwork carried out with Rio’s security forces over a period of eight months between 2014 and 2019. The methodology is inspired by approaches that consider the socio-spatial production of macro levels of security states (the configurations of state power and governance of security) and micro levels of everyday practices and subjective and bodily experiences of security (the experiential and emotional domain) (Glück and Low, 2017: 285–287). The data gathered consist of 62 qualitative interviews with high-level security officials in public institutions and private organizations and with street-level police officers. The ethnographic material stems from long observations in day and night shifts in police stations, special forces headquarters, command and control centres, and the mega-event security architectures in and around the security perimeters of the many sporting venues. My status as a white Northern European male who speaks Brazilian Portuguese had crucial influence upon my position and granted advantaged access as an ethnographer in the field.

The remainder of the article unfolds with the following structure: After this brief introduction, I elaborate on the entanglement of Rio’s mega-events, the global digital security politics it implies and the city’s long-lasting urban conflict. In the subsequent section, I conceptualize the Foucauldian triangle of security governance and the government of things to then lay out three analytical parts that examine the (1) materiality of security, (2) non-materiality of security and (3) digitalized necropolitical security. The concluding remarks show that what we find in Rio is a triangle of security governance in combination with the government of things and how this sheds light upon the simultaneity of a technology-based modernization of Rio de Janeiro’s public security system and continuous forms of racialized politics based upon repression, violence and the incitement of death.

Digital security politics, mega-events and Rio’s urban conflict

With the intensified digitalization of society, militarized urban security politics now include a wide range of emergent technologies (see, for example, Graham, 2010). Much of the fundamental work in the extensive body of literature on such new digital security technologies discusses the logics and functionalities of software, code, algorithms and artificial intelligence, often in relation to spatial logics (Graham, 2005; Kitchin and Dodge, 2011; Klauser, 2017; Thrift and French, 2002). Other authors explore the ways in which these technologies are applied in different urban settings and how they impact upon specific topics like anticipation (Adey and Anderson, 2012), risk and uncertainty (Ceyhan, 2002), or resilience (Coaffee and Fussey, 2015; Gressgård, 2017). More specifically, scholars have also analysed how the police integrate new technologies in their daily working routines in attempts to make policing ‘predictive’ and ‘smart’ (e.g. Brayne, 2017; Egbert and Leese, 2020; Fussey and Sandhu, 2020; Leese, 2021).

Two discussions within these literatures are helpful for situating the present article. The first concerns how security technologies are being negotiated in the era of digital technologies and how much of the human is still involved in security-related, future-oriented and probability-calculating decisionmaking (Amoore, 2013). Lately, much attention has been paid to the material, such as infrastructures, security objects and artefacts (Aradau, 2010), and how materiality fosters or contests power relations and, thus, raises awareness about the agency of the non-human in the ‘technopolitical’ (Hecht, 2011; Kurban et al., 2017; Mitchell, 2013). Here it is argued, mainly in relation to the post-9/11 era, that harsh state violence is, if possible, today rather applied in the background, and that exceptional measures turn into everyday normalization routines in which algorithmic security practices take on an important role in deciding who, what and in which ways objects are secured and surveilled (Monahan, 2006).

The second discussion that is important for this article looks at digital security technologies specifically applied in the context of the so-called Global South, where policing powers and ‘the right to kill’ are much more diversely distributed across public and private sectors and criminal and non-criminal domains (Denyer Willis, 2015; Diphoorn, 2015). Informed by a postcolonial posture, scholars have shown how a globalized security assemblage of technologies, materialities, knowledge and discourses established the urban settings of the postcolonies as new testing grounds for learning about and developing new security technologies before they are subsequently retransferred to Northern, Eastern and Southern cities (Comaroff and Comaroff, 2006; Grassiani and Müller, 2019; Hönke and Müller, 2012; Müller and Steinke, 2018). In relation to these tendencies towards testing new technologies in settings of intensified state violence, there is also a growing body of literature that contests the dominant narratives of a techno-fix that politically proposes technological solutions to otherwise pressing problems (Bruno et al., 2019; Cardoso, 2012).

Sport mega-events are historical conjunctures of such global security assemblages and can be considered as both the product and producers of contemporary security developments (Klauser, 2008: 75). The organization of the Olympics is accompanied by a standardized usage, testing, development and transference of knowledge to the next host of new security technologies and experiences such as the militarization of public security, the establishment of camera surveillance and the de facto isolation of sport venues from the rest of the city, to name but a few (see, for example, Fussey and Klauser, 2015).

With Brazil hosting the 2014 World Cup and the 2016 Olympics, something new was established. In an exception to the general trend, the events took place in a societal context in which an already pre-existing urban conflict posed a particular and pressing security issue for the events’ organization as drug-trafficking groups, different police forces, militias and local politicians were engaged in a bloody fight to gain the upper hand in a myriad of neighbourhoods (Leite, 2012). This opens up a unique opportunity for thinking about and bringing together the literatures about Rio de Janeiro’s longstanding public security problems and the postcolonial exploration of global security assemblages.

While an overwhelming proportion of scholars have conducted research that looks at the responsibility of the police for the daily violent dynamics within the city of Rio and its favelas, mostly from the perspectives of the drug traffickers and the community’s residents (e.g. Arias and Barnes, 2017; Richmond, 2015), much less has been written from the perspective of the officers themselves (but see, for example, Medeiros, 2018; Pauschinger, 2020). At the same time, only a marginal amount of research has been conducted into the realm of the digital security technologies and smart urban management systems that have so heavily impacted upon the security organization in Olympic Rio (e.g. Cardoso, 2013, 2019; Edler Duarte, 2019; Gaffney and Robertson, 2016; Luque-Ayala and Marvin, 2016).

Referring to the above streams of literatures, what this article proposes, then, is to contribute, on the one hand, to the scholarship on global security assemblages that examines how security technologies come together in networks of interactions between human and non-human entities, in postcolonial settings and at sport mega-events as part of ‘the renewed contemporary interest in the politics of securing with algorithms’ (Amoore and Raley, 2017: 4). On the other hand, it looks beyond these literatures with a specific focus upon the shadowy aspects of the complex interactions between security technologies and state violence, often exercised by the police, which are here also viewed from the perspective of Rio de Janeiro’s police officers themselves.

Contemporary security governance as a triangle

Sovereign power, discipline and security

Recently, scholars in critical security and surveillance studies have revisited Foucault’s (2007) governmentality lectures at the Collège de France, published under the title Security, Territory, Population, to better comprehend contemporary security governance in the post-9/11 era (e.g. Bigo, 2008). In these lectures, Foucault introduces the ‘apparatus of security’ as a new mechanism, one that adds a novel turn to his governmentality studies that further differentiates between sovereignty, discipline and security. As Foucault says, ‘sovereignty is exercised within the borders of a territory, discipline is exercised on the bodies of individuals, and security is exercised over a whole population’, yet all three modes must be understood as ‘concerned with multiplicities’ (Foucault, 2007: 11–12). While sovereign power as spectacle was often used to demonstrate to the sovereign’s subjects, through such things as military parades and public killings, that the sovereign decides who lives and who dies, and discipline as an individualizing power was directed towards creating new spaces of control (Foucault, 1977), the security apparatus is now concerned with creating a ‘milieu’ through accepting what reality offers and taking into account the possible, uncertain and temporal events within such spaces (Foucault, 2007: 20). As Foucault (2007: 44, 65) expressed it:

Discipline concentrates, focuses, and encloses. . . . Now it seems to me that through the obviously very partial phenomena that I have tried to pick out we see the emergence of a completely different problem that is no longer that of fixing and demarcating the territory, but allowing circulations to take place, of controlling them, sifting the good and the bad, ensuring that things are always in movement, constantly moving around, continually going from one point to another, but in such a way that the inherent dangers of this circulation are cancelled out.

For Foucault, security is no longer a matter of creating new spaces of control, but rather one of accepting the given spaces and existing territories. He suggests that disciplinary spaces, built for the purpose of control, work ‘in an empty, artificial space’, whereas security enables the circulation of what is desired to be in these spaces and tries to minimize any risks. Those who are declared evil by society cannot be erased in full, which means that security works with probabilities and planning to predict the future (Foucault, 2007: 19–20).

Pete Fussey (2013, 2015) and Francisco Klauser (2013, 2017) have powerfully mobilized the Foucault lectures to examine (a) how security is managed in and around mega-event spaces and counter-terrorism projects; (b) how at mega-events venues are, on the one hand, heavily protected and, on the other, designed to enable and filter spectator flows; and (c) how surveillance and control mechanisms are deployed for exactly these purposes. Both thus demonstrate how at sport mega-events and other urban security projects, discipline and security, enclosure and fixity, and openness and flexibility all mark securitized spaces (Fussey, 2013: 358; Klauser, 2013: 292; see also Schuilenburg, 2015: 63).

In these writings, however, it almost seems that Foucault had argued that disciplinary power and the mechanisms of the security apparatuses are more important than forms of sovereign power for understanding the functioning of Western societies, and, in turn, what the aforementioned scholarship understands as a Foucauldian perspective on contemporary security governance. In the following, I add another layer to such a perspective.

Necropolitics in Rio de Janeiro

To make sense of contemporary security governance in postcolonial settings such as Rio de Janeiro, it is not sufficient to simply mobilize Foucauldian governance modes such as discipline and security apparatuses, as there is a conjoining of overly violent forms of policing and the introduction of digital security technologies at the same time. Scholarly attention has therefore been given to the question of whether spectacle as a modality of power has really receded in contemporary society (see, for example, Alves, 2018; Pauschinger, 2020). As Foucault (1977: 217) writes, ‘our society is one not of spectacle but of surveillance’. Yet later he corrects that and suggests that there is not a straightforward displacement of one by the other:

So we should not see things as the replacement of a society of sovereignty by a society of discipline, and then of a society of discipline by a society, say, of government. In fact, we have a triangle: sovereignty, discipline, and governmental management, which has population as its main target and apparatuses of security as its essential mechanisms (Foucault, 2007: 107–108, emphasis added).

In this sense, then, it becomes a matter of what mode of governance in the triangle is predominant, or, rather, in what arrangement to each other and in what hierarchical constellations the different modes are settled. Considering the ways in which policing is carried out in Rio de Janeiro – that is, the employment of highly mediated and spectacular forms of police intervention (Larkins, 2013, 2015; Savell, 2016) – and how during the mega-events the city was occupied by the Brazilian armed forces invites us to reconsider the importance of sovereign power as spectacle in the ways we read security governance in Rio (Pauschinger, 2020). While spectacle is present through military parades and exercised in the city’s everyday life, sovereign power is also exercised through police killings in locations that range from favelas, to the streets, to the insides of police cars and emergency rooms (Alves, 2018: 61).

Recent developments in which far-right former governor Wilson Witzel actively incentivized Rio’s police forces to kill, exempting them from any legal accountability, and made public appearances in which he applauded his own politics of death, however, require us to think a step further. Sovereign power in a Foucauldian sense is here best understood in the terms of what Mbembe (2003: 27) articulated mainly in his essay Necropolitics. Mbembe lays out that governmental techniques are in essence meant to eradicate the lives of those bodies that are governed in specific spaces of death (Medeiros, 2018: 86). Mbembe (2003: 39) expands the idea that sovereignty essentially means ‘the right to kill’ to embrace the concept of ‘necropower’, which entails socio-spatial patterns in which the sovereign decides who lives and who dies, ‘who is disposable and who is not’, as a form of ‘contemporary subjugation of life to the power of death’. What we see in Rio, then, is that contemporary security governance is composed of the three Foucauldian elements of sovereignty – in the very sense of Mbembe’s necropower – discipline and security. Yet there remains a missing element that is needed if we are to comprehend the functioning of the triangle of security governance in the digital age.

The ‘government of things’

There is much discussion in the social sciences about how much agency can be attributed to the non-human, to materiality. In an earlier version of the introduction to this special issue, Frank Müller and Matthew Richmond described ‘the desire to pay closer attention, and credit greater agency, to materiality and non-human actors in the formation of the social’ as what has been designated as the ‘technopolitical’ (see also Hecht, 2011). Such increased interest in the agency of the non-human is particularly important for this article, as in the case of Rio de Janeiro we have to find ways in which we might understand the role of digital technology within the politics of security governance before, during and after the mega-event moment. The Foucauldian security vocabulary introduced earlier in this article is interesting as it complements what Foucault (2007: 96) has written in the same Security, Territory, Population lecture series at the Collège de France about a ‘government of things’ (Lemke, 2015, 2021).

Although there is notable scepticism about Foucault’s account of non-human agency (see, for example, Barad, 2007), here I mainly draw upon Thomas Lemke’s (2015) article New Materialism: Foucault and the “Government of Things”, in which he shows how Foucault actually did account for the non-human. According to Lemke (2015: 9), what Foucault says in his lectures is that governance is not per se something directed solely at humans or departing from humans; rather, ‘the qualification “human” or “thing” and the political and moral distinction between them is itself an instrument and effect of the art of government and does not constitute its origin or point of departure’. In other words, what Foucault advances is that governance – and security governance – can be exercised by both humans and non-humans. As Foucault (2007: 96–97) puts it:

‘Things’ are men in their relationships with things like customs, habits, ways of acting and thinking . . . with accidents, misfortunes, famine, epidemics, and death. . . . [G]overnment is concerned with things understood in this way as the intrication of men and things. . . . [T]o govern means to govern things.

Foucault here puts forth a deeply relational understanding of subjects and objects. Bringing this back to the apparatus of security is here enlightening. Security, as Foucault (2007: 65) has outlined, sorts and filters what can and what cannot enter predefined territories. This sorting of the good and the bad is carried out equally by both humans and machines and can be referred to objects and subjects in what he conceptualized as the ‘milieu’, where – as already outlined above – ‘the specific space of security refers then to a series of possible events; it refers to the temporal and the uncertainty which have to be inserted within given spaces’ (Foucault, 2007: 20).

Yet there is another important point that Lemke (2015, 2021) raises and that helps to shed light onto the important difference between the governance modes Foucault mobilizes. The concept of a ‘government of things’ reveals a mode of power that is different from that of sovereignty. As Foucault (2007: 99) writes:

it is not a matter of imposing a law on men, but of the disposition of things, that is to say, of employing tactics rather than laws, or, of as far as possible employing laws as tactics; arranging things so that this or that end maybe achieved through a certain number of means.

What we can take from this is that the way in which Foucault advances his idea of a government of things consists of distinguishing it from what he thought was and is sovereign power. The great achievement of Lemke (2015, 2021) is to expose that there is, effectively, another mode of power that comes much closer to what Foucault (1982: 790) said elsewhere, namely, that power is a mode of action that ‘structure[s] the possible field of action of others’. Foucault insists that power in itself does not exist. It is not a substance or property of specific actors, but ‘exists only when it is put into action’ (Foucault, 1982: 788).

The key question for Foucault (1982: 786) then becomes how – that is, through what mediating techniques – power is being exercised. Relating this back again to the apparatus of security – which has the filtering of the good and the bad and the enabling of circulation as its core mechanism – it becomes clear that it is the circulation by objects and subjects, along with the filtering exercised by humans and non-humans as actions upon the actions of others, that is meant here. Said differently, instead of seeing the security apparatus as different from the governance of things, I here argue that we have to combine both in order to comprehend the ways in which the human and non-human, the material and non-material, are working together.

My conceptual proposition, then, is twofold: first, to suggest that we can only analyse contemporary security governance with a Foucauldian perspective if the triangle of security governance with its three elements is considered; and, second, that in the digital era, but not exclusively, the three modes of the Foucauldian triangle of security governance are complemented by a fourth dimension, which Foucault named the ‘government of things’, that grants agency to materiality and, in our contemporary moment, provides greater theoretical understanding of how the exercise of power, discipline and security is delegated to digital technologies.

Materiality of security

Two major frames have fundamentally marked Brazil’s mega-event security strategy. The first referred to investments in visible material and concrete equipment and infrastructures, while the second focused on less palpable and non-material efforts, such as integrating the work of the many different security institutions. Both are interrelated, as the material was meant to technologically enable the non-material.

The Brazilian government tried to overcome both technologically and organizationally the long-lasting fight over the competences of the country’s many different police organizations by implementing the Integrated Command and Control System (Portuguese: SICC) that combines military management philosophies with command and control doctrines. These measures had already been introduced into Brazil’s military planning in 2006 (Cardoso, 2013: 131–133), but now were combined with a new managerial benchmarking philosophy based upon increasing efficiency through the use of statistical data (Cardoso, 2019: 55).

This system, materialized in the Integrated Command and Control Centres (Portuguese: CICC), was envisioned to regroup all of the involved security agencies and to technologically enable the surveillance of both the mega-event security operations and everyday urban security dynamics. Different national, regional, local and mobile centres were connected over a network in the host cities and venues during the World Cup and the Olympics and operated sophisticated surveillance camera systems. In Rio de Janeiro, the CICC was initially planned independently of the mega-events to centralize the state’s emergency services and implement a surveillance camera scheme, but then progressed to become the most sophisticated centre owing to the hosting of the Olympics and the federal investment in the facility (Cardoso, 2013, 2014, 2019).

In Rio, the camera surveillance tools have taken some rather complex turns, as there were both fixed and mobile cameras, operated by at least four different entities: federal, state, municipal and private forces in a ‘surveillance assemblage’ (Haggerty and Ericson, 2000). The surveillance structures and inner life of the CICC do not seem to suggest that it was a panoptical tool for controlling specific urban settings; instead, the surveillance efforts of the CICC resonate with Foucault’s analysis of a security apparatus that highlights surveillance as a way of monitoring from a distance and ‘calculating phenomena in situ’ (Fussey, 2013: 352).

I recall the many hours I spent in the control centre during the Olympics and how important the camera images were for the actors in the control rooms. The images from the cameras served as a way of ensuring that the flow of objects and of people was guaranteed. Consider only what the general CICC coordinator for the World Cup relates in the following quote about how camera surveillance had been an important working tool for the mega-event security:

The whole work was based upon permanent monitoring, right? This is what we call ‘situational awareness’. . . . The more situational awareness, the more elements you have got to decide from . . . [and] you don’t decide in the dark. (Interview 2)

The coordinator states that ubiquitous camera surveillance can serve as a facilitator for decisionmaking and provides a better overview in complicated situations on the ground. It seems as if CCTV, at least in this case, serves as a way of reducing complexities and sheds light on a situational crime prevention philosophy that was salient within many of my interviews. More to the point, ‘situational awareness’ is the expression of a desire to have control over everything, and such an approach transcends the goal of total readiness that proliferated among the authorities (see also Krasmann and Hentschel, 2019).

And, indeed, these efforts to implement surveillance and security systems were not much different from the approach that was adopted at previous events. In relation to EURO 2008, Klauser (2013: 292–296) explored and demarcated three main goals of the variety of surveillance and security measures. Swiss state authorities and private entities applied their catalogue of measures to ensure enclosure and access controls, to manage circulation, and to regulate the internal organization of the monitored zones. The ‘dilemma’ that Klauser (2013: 292) derives from his fieldwork in Switzerland is how to do both: on the one hand controlling space while on the other allowing the flow of people. Klauser’s (2013) analysis was that access and control mechanisms varied in different urban environments like the Fan Zones and the stadium areas at EURO 2008. One major aspect Fussey (2015: 219) found that differed from Klauser’s work was that, at the Olympics in London 2012, access controls varied ‘within the same setting’, as ‘good flows were permitted and assisted in their traversal of borders and boundaries’. In London 2012, protesters and all those that did not fit into the Olympic spirit and marketing strategy were not immediately – and sometimes not at all – banned and/or excluded from the venues, but the flows and categories of ‘good’ and ‘bad’ visitors were constantly watched and monitored.

For Foucault, security relies on a constant movement of the objects and people that circulate within pre-existing territories, ‘but in such a way that the inherent dangers of this circulation are cancelled out’ (Foucault, 2007: 65). In Rio, surveillance and security measures similar to those at previous events like EURO 2008 and the London Olympics were applied, and the space at Rio’s mega-events was equally monitored, controlled and assessed. The mega-event security scheme used already given spaces of discipline – secured enclaves such as stadiums, FIFA Fan Zones and the Olympic Park – among which the flow of people was constantly monitored, and circulation became a priority. Moreover, some of the cameras that were operated from within the different CICCs were able to detect strange objects in the scanned environment and launch an alarm if a bad object were to enter the area. Here, what Foucault (2007: 65) talks about as the ‘sifting of the good and the bad’ to enable a constant flow is put technologically into practice.

Interestingly, the ways in which the human and non-human surveillance and control mechanisms are applied in the above examples around Rio’s mega-event venues produce a direct link to the technopolitical turn in Rio de Janeiro and elsewhere (Barry, 2001; Gaffney and Robertson, 2016; Hecht, 2011). In Foucauldian terms, what I have shown with the examples from Rio is not only governance through and with different modes of power like discipline and techniques like security; what I observed in the CICC can also serve as an example of a veritable government of things. Yet what I also take from my observations is that technopolitical governance is about the governance of both humans and non-humans, or, as Foucault (2007: 97) put it, ‘that government is concerned with things understood in this way as the intrication of men and things. . . . [T]o govern means to govern things’. Within the CICCs, it seems, the CCTV camera system is managed by and manages humans in both spaces of enclosure (discipline) and territorial control mechanisms of openness (security apparatus).

Non-materiality of security

Standing in the middle of the mega-event operation room in the CICC in Rio, I watch how helicopter images from Sepp Blatter’s and Angela Merkel’s escorted cars are transmitted onto the flat screens on the video wall. A few hours after the World Cup final, the leaders of the different police institutions give speeches over a microphone to thank everyone present, to thunderous applause. The military police colonel says emotively: ‘The major legacy of this World Cup is the integration of the public security forces.’ One of the civil police delegates echoed his words, elaborating: ‘Integration came to stay and will never vanish.’ These fieldwork vignettes serve as a reminder of how deeply enrooted the non-material security narrative was among Brazil’s security authorities.

To implement and continue this ‘non-materiality’ of security, the Brazilian government founded a Special Secretariat for Mega-Event Security (Portuguese: SESGE). As a subsecretariat of the federal Ministry of Justice, it had responsibilities ‘to plan, define, coordinate, implement and evaluate the actions of the public security [sector] for the mega-events’ (SESGE, 2012: 7). This strategy was important, as the Brazilian public security sector suffers from a lack of integration of the many police forces along with a well-defined but practically complicated division of their responsibilities, which have often resulted in mistrust, miscommunication and fights over competences (see, for example, Soares, 2000). The event organizers knew that they had to better integrate the different police forces, mainly in cities like Rio.

While, overall, the orchestrated mega-event security operations worked out to the extent of ensuring a more or less secure staging of the Games, from the beginning the strategy was not as cooperative as had been intended. Whilst the SESGE organized and oversaw the mega-event security scheme, the Ministry of Defence had also an important role. Within the host cities, the state and municipality administrations had to work together with the federal forces. More importantly, the three-dimensional Brazilian state structure led to a competition between the Ministry of Justice and the Ministry of Defence over funds designated for the security of the mega-event and resulted in tensions in the planning process that produced many infrastructures from different security institutions that served the same purposes but were not necessarily well connected with each other. This was strikingly apparent when I was able to visit the military facilities in which the armed forces had set up their very own command and control centre as a parallel structure to that of the CICC. As a result of these two parallel command and control structures, the government had to create a special commission that would establish communication canals on high-ranked command levels, because the armed forces would not accept a civil policeman commanding a military representative within the CICC in Rio.

At the same time, the integrative working philosophy also did not work out between the public security organs and private organizations like FIFA and the International Olympic Committee and their local organization committees. The problems with the implementation of the integrative working philosophy often entailed different imaginations about the presence of public security forces in privatized urban spaces such as security perimeters in and around stadiums. In Rio, however, the cooperation strategy led to tensions not only among different public and private organizations and between high-ranking ministers, but also among officers at street level.

I remember how, during one of the World Cup games, I took a tour around the stadium within the security perimeter with a civil police agent. When we were almost finished with our walk around the stadium, during which we inspected and discussed the different spatial distributions of the security scheme, we were stopped by military police personnel who did not want to let us pass. A discussion between my civil police companion and the military police officer developed and almost led to a physical fight until a colleague from the nearby mobile police station came to rescue us.

A sentiment expressed by one of the civil police officers in the scene presented above caught my attention: ‘There are too many people commanding here.’ This statement paves the way for understanding how the tensions are shaped among the variety of actors in the field. The various private and public partnerships in the security organization at the mega-events in Rio (e.g. FIFA/police) shed light on debates about the myriad of actors involved in matters of security – private and public – and how the problems in their work relationships are not value-free nor can these working partnerships be understood as simple gatherings of different actors, as the implementing of distinct security strategies ‘is not so much a question of universal principles than of everyday negotiations and micro adjustments between different actors, interests and stipulations’ (Klauser, 2015: 233).

From these scenes follow the fundamental question of whether the so-called non-material legacy – the integration strategy – would actually stay at all, if it already presented such tremendous problems at the planning and operational levels. Another look into the CICC in Rio is helpful. The SESGE had a decisive role in the integrated command and control system strategy. The many different CICCs during both the World Cup and the Olympics were coordinated by the SESGE. The SESGE team saw their main task as gathering information from all of the involved security forces and feeding this into shared data banks. In other words, the integration system worked through the mediation efforts of the SESGE.

In a Foucauldian sense, then, the SESGE was responsible for guaranteeing the flow and enabling the circulation of information, filtering the important (good) and unimportant (bad) objects both human and non-human (Foucault, 2007: 65). The SESGE here has the position of letting things happen. As Foucault (2007: 46–47) puts it:

It seems to me that in the apparatus of security . . . what is involved is precisely not taking either the point of what is prevented or the point of view of what is obligatory, but standing back sufficiently so that one can grasp the point at which things are taking place, whether or not they are desirable. This means grasping them at their level of nature . . . grasping them at the level of their effective reality. The mechanism of security works on the basis of this reality, by trying to use it as a support and make it function, make its components function in relation to each other.

Foucault (2007: 47) sees the mechanism of security as different from the law that ‘prohibits’ and the discipline that ‘prescribes’, though security possibly makes use of both, the law and discipline, to cancel out the reality to which security responds – ‘nullifies it, or limits, checks, or regulates it’. This regulation of reality mentioned by Foucault is organized and translated in my case study of Rio de Janeiro by the SESGE, an organization that acts as a mediator in the CICC to ensure that the other institutions involved in the CICC relate to and communicate with each other. Much as Foucault says in the above quote, the SESGE ‘grasped’ the work of the involved institutions, took them at their very ‘nature’, to ensure that they produced ‘security’ together.

As the SESGE was an agency that was dissolved after the Olympics, the natural question is therefore what happens when the facilitating agency is not present to mediate anymore. The SESGE was mainly equipped with federal police officers who also played only a subordinated role within the daily running of the control centres and Rio’s quotidian public security. Many of the involved officers I spoke to were worried about the legacy of integration when the SESGE stopped exercising its facilitating role. As one of them said, ‘tomorrow they change the teams – we start all over again’.

Digitalized necropolitical security

In the two previous analytical parts of this article, I have shown how the material and non-material security strategies employed in Rio de Janeiro have played out. In this third part, I will show that if we really want to grasp how contemporary urban security governance is organized, and specifically in Rio de Janeiro before, during and after the Olympics, we have to consider a third element. Instead of only grounding mega-event and urban security analysis upon Foucault’s discipline and security, what I propose is to mobilize the Foucauldian triangle of security governance in its entirety, including the element of sovereign power (through spectacle) and the related Mbembian concept of necropolitics. This provides an opportunity to show both how these three elements go hand in hand and how the government of things allows us to understand how the triangle works in the digital era.

In material terms, the functioning of Rio’s public security sector spectacle as a mode of sovereign power has not receded at all and was extensively used in the preparation, staging and in the aftermath of the Olympics. During the mega-events, this materiality of a spectacular security strategy was enacted by military parades such as the occupation of the city by the armed forces, the ostensive patrolling of the various policing institutions in strategically important neighbourhoods, the deployment of warships in front of Copacabana beach and the many roaring military helicopters above the venues. For the Olympics, more than 88,000 security personnel were deployed to secure the Games, 41,000 alone from the military. ² The security measures here had the clear function of conveying a feeling of security to the visitors at the events, suggesting that the state forces had everything under control. Thus, the performance of Olympic ‘spectacular security’ (Boyle and Haggerty, 2009) clearly acted upon the Olympic audience.

However, as much as spectacle intends to make certain aspects of security visible, it also aims to hide others (Goldstein, 2004: 16). It is a process that I have explored elsewhere in terms of a ‘security of camouflage’ (Pauschinger, 2020), demonstrating that sovereign power through spectacle also worked during the mega-events in territorialities other than those designated for the Olympics, acting on different audiences at the city’s edges and in the favelas. Police work in Rio de Janeiro has long been embedded in ‘highly visible performances of state power’ (Larkins, 2013: 565), targeting populations racialized as black, to whom the state demonstrates through deliberate, almost public killings in a process that Mbembe (2003: 39–40) has conceptualized as necropower, within which entire populations in certain territories are deemed to live in ‘death worlds’ where sovereignty is exercised by deliberately deciding who deserves to die, and who to live. These killings are organized as a form of ‘police terror’ that ‘becomes a tool for the spatial arrangement of racial difference’ (Alves, 2018: 61) in Brazilian cities such as Rio de Janeiro. Within the triangle of security governance in which I have argued that Foucauldian sovereign power must be understood as being as equally important as discipline and the security apparatus, the element of sovereign power is best understood as necropower that exercises control through killings (Medeiros, 2018: 86).

In non-material terms, conversely, the roots to this material order of death lies in the non-materiality of the necropolitical that is engrained in Rio’s urban conflict. Consider the following vignette. In late 2016, I was writing up my PhD thesis when I got news from Rio that one of the special forces personnel I had interviewed for my research had been brutally assassinated while on his way home with his girlfriend after a visit to a restaurant. I immediately tried to contact one of his colleagues in the special forces but could not get hold of him. I reached the colleague’s girlfriend, who also works in the police, who told me: ‘You know he is up the hills [synonym for a favela]. They are looking for the guys with blood in their eyes. It’s a war. It is a war.’ Countless times during my research, I came across scenes in which I heard state agents arguing for the ‘right to kill’, justifying warlike scenarios without the need for any legal accounting. On numerous occasions, I participated in conversations with police officers who cheered the deaths of suspects and witnessed state violence unfold (Pauschinger, 2019).

And state violence continued in intensified forms after the Olympics and also raises fundamental questions about the exercise of such state violence in our current political climate. The example put forth in the introduction to this special issue is telling. It features a scene in which Rio de Janeiro’s far-right former governor Wilson Witzel was filmed in August 2019, three years after the Olympics, descending from a helicopter on the famous Rio-Niterói Bridge, running towards the sniper who had killed a hijacker, cheering as if he had just scored the winning goal in a football game. During his time in office, Witzel actively encouraged police forces to kill, which in 2019 led to 1810 people being killed by state forces – the highest number since measurements began in 1998 (BBC, 2020). The composition of the materiality – the spectacular performance – and the non-materiality – the discursive encouragement of killing, authorizing death and deciding who deserves to die and who does not, best illustrates how it is indeed the Foucauldian triangle that is at work in Rio de Janeiro. Here, sovereign power, discipline and the security apparatus are working hand in hand. In contemporary Rio, however, sovereign power is exercised as necropolitics, in which killing becomes a form of security governance.

The technopolitical turn in Rio de Janeiro and the induction of a techno-fixated belief in technology can be perfectly combined with the triangle of security governance by considering the government of things. If we comprehend the Foucauldian government of things as the governance of both the human and non-human, and if we want to take seriously my proposition about the need to consider the three elements of the triangle as essential to the understanding of contemporary security governance in places like Rio de Janeiro, then the triangle and the government of things must be understood together.

The camera surveillance system and the command and control centres are examples of how new technologies now work hand in hand with already established ways of producing security. Viewed as the ‘essential mechanism’ (Foucault, 2007: 108) of a government of things, the camera system allows circulations while ‘controlling them, sifting the good and the bad’ (Foucault, 2007: 65) within and around both artificially created spaces of control and in already given territories. Here, the government of things then means nothing else than the filtering of and by humans and machines.

Whereas digital technologies may be fundamentally new to the equation, the techniques as such are not. The machines serve to leverage an exclusionary politics that has been commonplace in Rio de Janeiro since the beginning of its urbanization and long before democratization, and that has only proliferated since then (see, for example, Fischer, 2008). The system filters those who are wanted from those people who are not wanted in the spaces where the cameras are installed. Stated differently, the camera surveillance scheme fortifies and follows old traditional patterns of Rio de Janeiro’s exclusion and death politics.

Conclusion

In this article, I have joined conceptual debates that ask how we can possibly understand security governance in times in which digital technologies find ever more entry into the organization, management and deployment of urban security tactics both at sport mega-events and – more generally speaking – in cities of postcolonial settings and the parallel increased use of state violence by considering forms of sovereign power as still in place and active in these spheres.

I have investigated how the official material and non-material security strategies and legacy claims of Brazil’s mega-event organizers have been implemented in the context of Rio de Janeiro. Materiality has been present in the emphasis upon security technologies such as surveillance camera schemes and the mobilization of military forces. The non-materiality was meant to be the integration strategy intended to better unite and coordinate the myriad public and private security organizations.

The empirical details I have presented invite the reiteration of two main conceptual points. First, the analysis brings to the fore how contemporary security arrangements in highly violent places like Rio de Janeiro cannot be analysed from Foucauldian perspectives that set out from the assumption that discipline and the apparatuses of security have substituted or pushed into the background forms of sovereign power. Much to the contrary, it is a question of the hierarchical positioning of these different security governance techniques that are more or less strongly alive depending on the societal context. To repeat Foucault’s (2007: 107) own words: ‘In fact, we have a triangle.’ I have conceptualized this as the triangle of security governance, which is composed of sovereign power, discipline and security apparatuses, which all operate together simultaneously.

Second, I argue that to properly appreciate the imbrication of the widespread deployment of digital security technologies and repressive and deadly state violence, the triangle is best understood in combination with the agency of materiality in what Foucault conceptualized as the ‘government of things’. With the example of the material legacy of the camera surveillance system that was built for the mega-events, and that has remained in place since, I have shown how there is a continuation of necropolitical conceptions of security in technological terms. A Foucauldian government of things must therefore be thought together not only with how it technologically sophisticates disciplinary spaces and the functioning of security apparatuses, but also with how it is intrinsically interwoven with how power is delegated to technology to continue longstanding exclusionary politics.

Security governance in Rio can therefore not just be considered as enabling circulation and controlling access to distinct territories by and with the help of human and non-human entities. It is meant to bring necropolitics to the next level. Consider only the military occupations and the technological control mechanisms set up in and around favela neighbourhoods before, during and after the mega-events. Here, the filtering works upside down, as the access control is not meant to define who can enter the favela; rather, it is about who can exit it. More than that, military occupation in this sense is also about the control within these territories. And, here, a different kind of filtering takes place, in which the non-material has material consequences. The killings by both Rio’s police forces and the army in the favelas are not isolated incidents, but are part of a long-lasting death politics that clearly filters out who is ‘good’ from who is ‘bad’, and thus who is disposable and who deserves to live, along lines of race and social inequality, with the help of digital security technologies.

From the killing of George Floyd in Trump’s USA to Rio de Janeiro’s deadly police forces in Bolsonaro’s Brazil, the implementation of digital security technologies, the use of state violence and the rise of far-right populism merge in worrying ways. Whereas these developments have traditionally been studied separately, much more attention should be paid to how these phenomena coalesce, influence each other, and are essentially lived, negotiated and contested. There is an urgent need to better explore how material and non-material forms of the technopolitical intermingle with necropower and how power is ever more delegated to digital technologies.