1

Sensing in/security: An introduction

Nina Klimburg-Witjes, Nikolaus Poechhacker and Geoffrey C. Bowker

There are more automated sensors perceiving our environment and the elements that constitute it than there are living human beings.

Tironi 2017: 2

Almost anything and anyone can become a sensor, gathering and transmitting data about our world. Sensors are omnipresent and increasingly important elements in constituting and controlling contemporary societies in many domains of our lives. Built into (smart) cities, communication devices, and our clothes, attached to our bodies, to drones, satellites and cars, sensors have become our mostly invisible companions. Invested with ideals ranging from ‘invisible computing’, the ‘Internet of Things’, ‘global transparency’, or ‘algorithmic governance’, ‘these automatic electromechanical labourers, at the fringe of our awareness, control the world around us. At times, they even control us. Yet they are now so familiar, so mundane, that we hardly notice’ (Townsend 2014: xi). In/security is one of the domains that we now find equipped, imagined and measured with sensors.

The contributions in this volume bring together science and technology studies (STS) and critical security studies (CSS) to examine in/security, sensors and sensing. By bringing these fields together, we aim to extend long-standing STS concerns with infrastructuring to emergent modes of surveillance and securitization enabled by sensing practices and digital infrastructures. We set out by exploring many by now classical STS issues such as monitoring, registering, representation and visualization (Amoore 2009; Dijstelbloem and Broeders 2015; Vertesi 2014; Dumit 2003; Witjes and Olbrich 2017; Ruivenkamp and Rip 2014); issues of technological mediation and human/non-human networks (Callon and Muniesa 2005; Law 1994; Poechhacker and Nyckel 2020); infrastructures (Larkin 2013); the politics of knowledge and expertise (Ezrahi 2012; Shapin and Schaffer 2011); issues of classification and categorization (Bowker and Star 2000; Star and Ruhleder 1996; Star 1998; Suchman 1994; Barry 2001); group formation and data politics (Edwards et al. 2011; McCosker and Graham 2018; Ruppert 2011); as well as questions concerning the shaping of societies, states and technologies (Bijker and Law 1992; Jasanoff 2004; Felt 2015; Hecht 2009; Scott 2001; Mitchell 2011), with a particular view towards sensors as security infrastructures.

Most sensing activities operate in the background and do not require active or direct registration by those who are monitored (see Andrejevic and Burdon 2014). Sometimes, however, it is deliberately made clear that we are being sensed, or made sense of by devices. Questions about the in/visibility of sensors drives this book: how do sensors shape and how are they being shaped by the environment in which they are placed, and by the processes they (attempt to) render visible (see Frith 2019)? Sensors pick up some data and not others, depending on which data their designers consider relevant. Materially, sensors register only what they are designed to measure (Helmreich 2019). In the case of security-related sensors, sensors pick up data that their designers take to indicate a security threat. Sensor design and deployment, in this way, takes part in constructing and delineating the phenomena that are to be sensed and governed. Sensors actively produce data traces by enacting otherwise contingent realities. Acts of sensing reduce the multiplicity of potential ontologies to a singular reality that the specific sensing regime can register. This translation of reality excludes enactments and actors that escape the sensing regime, making sensing always also a political act (Law 2002; Callon 1986).

Our aim with this volume is to draw attention to the ways in which sensors are integrated into the environment and how they produce different forms of in/security through processes of exclusion and inclusion. STS and CSS alike have observed a shift of security regimes from ‘evidenced-based identification and assessment of danger informed by a causal logic and reliant on empirical analysis’ (Suchman, Follis, and Weber 2017: 2) towards a predictive and risk-based evaluation of potential threats (Amoore 2013). However, the notions of causal logic and empirical evidence have been problematized in STS and neighbouring fields for some time now as emergent qualities of sociotechnical arrangements. Processes of inclusion and exclusion thus produce security and insecurity alike: Security as a performed and shared form of knowledge, insecurity as becoming the subject of security regimes. This distinction can then also be discussed along the lines of becoming visible for someone or becoming visible as someone. In each case, the production of sensory in/visibility creates a dialectic relation between security and insecurity.

A short sensor journey

To illustrate the abundance of sensors built into our everyday practices and experiences, let us take you on a brief journey through sensing infrastructures, each enacting and interacting with the world in its own way. First, switch on your smartphone’s augmented Global Positioning System (GPS). You are no longer alone, and you will no longer get lost, as you are now sensed by apps like Google Maps using a flock of satellites circling the Earth in a series of orbits designed to optimize coverage at any given moment. Each satellite contains an atomic clock, constantly emitting electromagnetic signals carrying an almanack of information about the positions the satellite is supposed to be in. Your device uses these pieces of information to triangulate its position, thereby embedding you in a military-commercial geopolitical infrastructure of ground antennae and data centres with its own (post)colonial legacy (see Oldenziel 2011). While satellites might help you on your way, they also continuously observe, measure and monitor the Earth. In terms of security, they are situated at the intersection of technologies of militarized intelligence, and technologies of human rights, as both are used to reify security threats posed by adversarial countries or groups. For instance, commercial satellite imagery is increasingly used by non-state actors like human-rights activists and think tanks as a tool to hold perpetrators accountable for human rights violations and mass atrocities. At the same time, government agencies are still powerful in determining what is visible and to whom (see Wang et al. 2013; Witjes and Olbrich 2017). Although seen by many as omnipresent surveillance technology from above (Parks 2005; Herrscher 2014; Shim 2016; Hong 2013), the satellite gaze can be hampered by cloud cover as well as by limited windows of observation due to geocentric orbits (Zirker 2013). However, within multi-modal sensing networks, if one sensor is hindered in its function, another is likely to take over.

We have now reached the oceanside, where wave buoys provide local measures that satellites – using scanning radar altimeters, scatterometers, and synthetic aperture radar – cannot (Helmreich 2019: 5). The buoy, as Helmreich suggests, could ‘be read as a symptom of how ocean politics have been enabled by national, military, and corporate infrastructures of measure, with buoys looking like harmless bystanders even as they concretize real relations of territorial domination in ocean space’ (2019: 5). Following anthropology underwater, we encounter multiple sensor networks: Collaboratively, they monitor physical or environmental conditions, such as pressure, sound, temperature, etc. and transmit data to the underwater node. The data are transmitted to a surface buoy via a wired link, and eventually received at an onshore or surface sink via radio communication, thus enabling computation to become environmental (Gabrys 2016) and the environment to become computational (Helmreich 2019). This assemblage can be utilized in many scenarios from environmental monitoring and deep-sea exploration to flood and tsunami alerts, from navigation and communication to underwater warfare (see Starosielski 2015; Oreskes 2003; Mort 2002).

From here, we are bound to the airport, a site where sensors and security-related sensory networks condense, sensing our bodies, belongings and biometrics in multiple ways. At the check-in counter, we are asked to show our passports with now mandatory biometric fingerprint data, detected by a tiny scanner that governs both the mobility and enclosure of bodies (Amoore 2016), turning surveillance into a form of ‘social sorting’ (Lyon 2003a, 2003b; Leese 2016; Cunningham and Heyman 2004). At the smart border, we are likely to go through the procedure of body scanning – shortly after its introduction re-labelled to ‘security scanners’, thus distracting our attention from the vulnerability of human bodies rendered visible with the promise of increased security (Bellanova and Fuster 2013). These security devices ‘illuminate the body with short-wavelength radio waves […] and form an image from the reflected radio waves […] to create a two-dimensional image of the body’ (European Commission 2010: 8) that highlights metallic and non-metallic objects.

This journey has illustrated some of the many instances where sensing devices are employed in the name of security; from satellites to underwater networks, biometric scanners and radars. As with so many sensing technologies, that were first developed for the military (see also Chapter 11 on ‘Drones as political machines’ by Bracken-Roche, this volume), what is being sensed and how we are subjected to different sensing regimes is at least ambivalent, and so are the meanings and the consequences of being sensed; seeing (like) a drone means something different if you are in suburban US house or a village in Pakistan (see Gusterson 2017).

No matter where we go, stories about sensors as actors in techno-societies are complicated, multiple and political. Not surprisingly, then, sensors have come to the foreground in contemporary academic and policy debates about the relations between data, security and politics. Some authors have even postulated that we live in a ‘sensor society that is constituted by the devices we use to work, communicate and play with, and which double as probes capturing the daily lives of people, things, environments, and their interactions’ (Schermer 2008 cited in Andrejevic and Burdon 2014: 6). In STS research, sensors are not new objects, whether in the assembling of controlled experimental setups, the design and implementation of ‘large technical systems’ (Hughes 1987; Summerton 1994) or the production of novel measuring instruments (Gramaglia and Mélard 2019; Gabrys 2016), sensors have been widely studied as ‘lively’ devices that detect, inscribe, capture and record; if not always explicitly as ‘sensors’ (Waller and Witjes 2017; Gabrys 2019, 2009; Gabrys and Pritchart 2018; Helmreich 2019; Suchman, Follis, and Weber 2017; Edwards 2004, Walford 2017; Spencer et al. 2019).

Sensor practices – practicing sensing

In a technical sense, sensors are devices that capture and record data which are then transmitted, stored, analysed and linked to other data sets. Oscillating between civilian, police and military domains, sensors are inscription devices (Latour and Woolgar 1986). Inscription devices were originally conceptualized in science studies as crucial elements of laboratory equipment that ‘transforms pieces of matter into written documents’ (Latour and Woolgar 1986: 51), thus creating a reference to the reality in question. Sensors, however, often are no longer part of a confined laboratory space, but are crucial elements in the ‘production of security in ‘laboratory’ conditions’ (Amicelle et al. 2015: 299). As such, sensors enable new forms of interacting with the world at a distance through sociotechnical infrastructures mediating between actors across space (Latour 1999). In short, sensing infrastructures include not only mechanical sensing, but a delicate interplay between humans, artefacts, and discourses (Gabrys and Pritchard 2018). As much work on knowledge infrastructures in STS and beyond has shown, conceptualizing raw data as neutral and objective is a bad idea (Bowker 2008; Gitelman 2013). Because data is always processed, and subject to infrastructures, sensors do not only produce ‘raw data’ but also problematize the relation between epistemic practices and their environment (Waller and Witjes 2017).

This volume aims to explore some of the complex and often invisible political, cultural and ethical processes that contribute to the development of sensors and their data infrastructures (see Bowker et al. 2010; Edwards 2010; Star 1999). By doing so, it shows how sensors reduce complexity and selectively produce a version of the world measured. As such, the power of sensor networks not only ‘work[s] through the sensory capacity of artifacts’ (Kim 2016: 400), but through the embeddedness of sensory capacity in a broader sociotechnical network. While making sensing activities possible in the first place, this embeddedness allows for the sense-making of multiple data traces produced through sensing practices by collecting and combining them in what Latour (1987) called centres of calculation. Sensing traces are thereby not just collected in one centre of calculation – keeping the chains of translations stable – but are compared and calculated in multiple centres, where their meaning is re-interpreted and re-stabilized (see Egbert 2019).

Through the sensors discussed in this volume, we perceive the world like a security regime, producing probabilities and possibilities alike (Amoore 2013). Monitoring and measuring people, processes, and practices, sensors are framed as a means to increase security by diminishing uncertainty and enabling action against perceived, known, and unknown threats and risks. Sensors – as infrastructural actors – thus produce, standardize, and enact a certain notion of security. They transform diffuse ideas of a dangerous and threatening world into an experienceable and graspable entity; we might say, they perform ontological politics (Mol 1999). Yet, the visibility that is produced through sensors, creates also invisibilities, depending who gets included or excluded in the broader sensing regime. Sensors are becoming part of a knowledge/power configuration that is built on the distinction of in/visibility (Foucault 1979; 1991). In this sense, new sensory infrastructures re-materialize already existing social orderings, and are re/generative of dominant cultural, historical, political, and economic relations. Sensors are shaping what type of ‘politics take hold along with these technologies’ (Gabrys 2016: 18), as novel modes of data gathering lead to ‘new configurations of engagement, relationality, sensing, and action’ (Gabrys 2016: 23). For the realm of security this means that novel forms of sensing might not only inform security politics and practices, but enable novel understandings of what security is and ought to be in a specific context: while each sensor is tasked to transmit data that are thought to be relevant for security purposes, the processes of measuring and monitoring render certain issues visible that might have been hidden before, thus co-constructing novel or previously unexpected security issues. Often, the enactment of security rests on prediction through algorithmic means (Suchman et al. 2018). In what Mackenzie (2015) called the production of prediction, machine learning systems and similar applications enact the world so that they can sort, reorder and find patterns (see Dow Schüll 2014). As such, the method of machine learning builds practices that resolve the inherent indexicality of data usage in algorithms, consequently connecting the abstract formulations of computer code to an experienced world (Ziewitz 2017).

Thinking security through and with sensors

To approach security as a social practice of sensing embedded in broader socio-political contexts, critical security studies can provide valuable insights into how security is thought and enacted in different settings, and how it continuously involves constructions of insecurity (Aradau and Van Munster 2008; Buzan et al. 1998; c.a.s.e. collective 2006; Huysmans 2000). Work in this field has done much to show that security fears are not out there to be discovered, but are constructed in the process of securitization (Buzan et al. 1998). Security is here understood as a discourse of power that can be invoked to frame a particular object or subject as a vital threat to society, the state, or public order. This has broader political effects and legitimizes the use of extraordinary measures to tackle the perceived threat. This call to engage with the practices enacted in the name of managing risk and uncertainty is also met by Amoore’ s work on the politics of possibility. Not accepting discourses of a global risk society (Beck 1992) in which we enter an age of uncertainty, she argues that it is not so much a question of whether or how the world is more dangerous but how specific representations of risk, uncertainty, danger, and security are distinctively writing the contours of the world (Amoore 2013: 7). The figure becomes the ground. Security as predictive technoscience, as Suchman et al. (2018: 2) have elaborated, rests on an ‘apparatus of distinction’ (Perugini and Gordon 2017: 2) that turns the suspect/enemy into an anticipatory target with the help of information based on real-time tracking, data mining, and the imagination of an omnipotent sensorium (see Latour and Hermant 2006).

To study security critically thus requires a focus on practices and the modes of governing they shape and promote (Amichelle et al. 2015; Huysmans 2006). Recent work in CSS linked to the ‘material turn’ of the field has shifted the focus from discourse to technologies and materialities; and from conceiving ‘security’ in terms of performative constructions to networks and associations. In this line of work the ‘technologization’ of security (Ceyhan 2008) and the logics and rationale that are undergirding security practices has received increased attention (see Amichelle et al. 2015: 295). This shared interest in the materiality and ontology of security issues and the mutual influences of technological devices and security practices is precisely what has spurred an inspiring and engaged conversation between STS and CSS (see Valkenburg and van der Ploeg 2015; Bellanova and Duez 2012; Jeandesboz 2016; Schouten 2014; Klimburg-Witjes and Wentland 2021; Leese 2016).

As a contribution to this exchange, this volume is a joint effort of scholars at the intersection of STS and CSS to come to terms with the messy and complicated properties of sensors as important and powerful elements1 of security infrastructures. The following chapters can be read as attempts to exemplify the processes and practices of sensing in/security visible. Engaging with the multiple entanglements of sensing practices, data infrastructures and in/security in different parts of the world, they empirically explore the contingencies of sensory knowledge, standardization process of security infrastructures and transgressions of boundaries between civilian and military spheres. They address the question of how sensors shape, shift, and constitute domains of national and international security policy and by this, explore the role of sensor infrastructures in the constitution of and mediation between state and non-state actors.

Coming from various academic lineages, the authors in this volume speak to these themes from multiple perspectives using a variety of case studies from various regions. In jointly presenting their views on sensing in/security, the authors seek to illuminate some of the shared concerns from different fields about surveillance, control, social sorting, border practices and social exclusion and envisioned security futures as enabling and enabled by sensing infrastructures.

Making sense of Sensing In/Security: Introducing the chapters

The issue of in/visibility is particularly relevant in the chapters that explore the ways in which sensors and their data infrastructures are either deliberately kept out of sight, physically hidden, underground, in remote areas, hidden from attention, behind technical terms, or powerfully deployed to create climates of in/security among those being or assuming to be sensed. Martin Tironi and Matthias Valderrama’s account of aerial surveillance in Chile addresses the latter. The authors show how the skies over modern cities have been increasingly occupied by new flying monitoring and datafication devices. Over the past ten years, a climate of fear and insecurity has developed in Chile, a feeling that is also widespread in Las Condes, one of the country’s wealthiest municipalities. Inspired by the techno-imaginary of Smart Cities, the local government has introduced a series of innovative and dynamic surveillance technologies as part of its efforts to manage and secure urban spaces and wage war on crime. However, residents and local organizations have protested the use of these technologies, citing profound over-surveillance and questioning the use of these security devices.

Drawing on qualitative interviews and participant observation, Tironi and Valderrama propose that vertical surveillance capacities must be analysed not only in terms of the surveillance and control they generate but also the affective atmospheres that they deploy in urban space and the ways in which these atmospheres are activated or resisted by residents. Reflecting on aerial sensing technologies, they show how these open up an affective mode of governance by air in an effort to establish atmospheres or micro-climates in which one experiences (un)expected sensations such as safety, disgust or indifference. The air, they argue, emerges as an ambience that must be controlled and securitized by the use of a series of aerial sensors and technologies that generate a vertical distancing between control rooms and the experiences of entities that coexist with/under the aerial gaze of such technologies of sensing in/security (see Adey 2010; Graham and Hewitt 2013; Klauser 2010; Weizman 2002).

Sensing infrastructures are increasingly disseminating and performing across the urban space techniques that are specific to borders, and especially to smart borders, such as algorithmic profiling, biometrics recognition, scanning, and screening. Drawing on fieldwork in New Town Kolkata in India, Ilia Antenucci explores how, in contrast with popular narratives of smart cities as seamless inter-connected spaces, the processes of urban digitization entail bordering practices. These work through the sensing networks and devices that are becoming more and more embedded in everyday life – bus shelters, water and electricity meters, garbage bins, home automation, apps etc. She discusses the political effects of ubiquitous sensing networks from two perspectives. First, it is suggested that sensing infrastructures introduce a new distribution of the sensible (Rancière 2000), setting boundaries between the different aspects of reality and perception, and measuring them incessantly; in this sense, the border operates at an ontological and epistemic level. Second, the chapter goes beyond the paradigm of surveillance/dataveillance to look at the nexus between algorithmic modelling, preemption and security decisions (Amoore 2013; De Goede et al. 2014) in the government of digital cities. This chapter contributes to an understanding of algorithms as creating new regimes of visibility that are politically charged.

At the same time, a new regime of invisibility is created – one of the code strings and operative systems that process urban data, crucial components remain largely inaccessible not only to citizens but also to the city agencies that are expected to act upon data. In their essay, A.R.E. Taylor and Julia Velkova show how data centres facilitate and make possible the work of sensing media, the tracking and collection of data and the production of metric cultures while remaining curiously absent in discussions of digital security infrastructures. Their piece introduces readers to the sterile technological spaces where sensor data is secured. As a critical intervention in recent scholarship that understands data centres as striving to remain invisible (see Holt and Vonderau 2015: 75), Taylor and Velkova draw on empirical work inside the buildings that store the vast volumes of sensor data now produced on a daily basis. They show how data is persistently imagined in terms of ‘flows’, like a constantly moving and circulatory form that never stays still – an imaginary that overlooks its situatedness, and the static, unmoving sites of digital information storage and accumulation where different technologies of sensing – human, mechanical and digital – intersect. Following Taylor and Velkova into the data centre, we understand how data centres are not just enablers of new sensor-based security regimes, but the sensory mirrors of the quantified, metrified societies that they infrastructurally help produce.

These chapters are in conversation with two of the three Visual Vignettes that both invite the observer to explore the cities’ hidden, invisible and secretive sensing infrastructure. The Visual Vignettes in this volume are a method by which sensing technologies can be differently seen, accessed, and understood, both by analysts and those with whom we as scholars might wish to share our work. Making Visual Vignettes for sensor stories brings novel forms of research communication into conversation with novel forms of sensing. Finding ways to communicate about our wired, and wireless world is a task of demonstrating the mutual co-constitution of security and insecurity.

The first vignette by Evan Light, Fenwick Mackelvey and Simon Hackbarth explores how International Mobile Subscriber Identity (IMSI) catchers, commonly known as StingRays allow users to determine which mobile phones are being used in a given location, to intercept phone calls, text messages and Internet traffic, and to send fake text messages. The past ten years, the authors argue, have seen a rise in the use of IMSI catchers by police departments, intelligence agencies and any number of non-state actors to monitor mobile phones. More recently, both commercial and non-commercial systems and products have emerged that aim to detect the use of IMSI catchers – so-called IMSI catcher catchers. IMSI catchers repurpose mobile telephone infrastructure as a surveillance device. Rather than embedding surveillance in mobile standards, IMSI catchers are technically a hack, collecting data not meant to be technically shared by our phones with anybody but a legitimate network provider. Drawing on the concept of infrastructural parasitism (Gehl and McKelvey 2019), they approach IMSI catchers as a parasitic surveillance device wherein the vulnerabilities and weaknesses in infrastructure might entice intelligence agencies and others. They argue that infrastructural weaknesses become opportunities for spying and surveillance as IMSI catchers feed on vulnerabilities in wireless code just as the Edward Snowden disclosures revealed how the 5 Eyes exploited vulnerabilities and the interconnection points of the global Internet (see Musiani 2015). Rather than seeing infrastructure as one coherent system, the parasite invites thinking of infrastructure as a plurality of technical projects that coexist with each other in a parasitic chain (see Serres 1980). Inspired by Anna Tsing’s work on the Matsutake mushrooms and their pickers that prototypes a landscape story which ‘requires getting to know the inhabitants, human and not human’, they look for IMSI catchers within this urban environment, as transient objects that, only by getting to know their enabling environment and human contact points, become possible to discover.

Chris Wood then invites us to walk with satellites and explore the meanings held within the GPS satellite network (typically hidden behind the hegemony of user interfaces). He contends that rather than being concentrated in the ways an individual interacts with technical objects and interfaces, an experience of space is held by the multiple human and non-human objects which form GPS infrastructures. Wood uses walking workshops which leverage GPS diagnostic tools to speculate on themes and phenomenologies across such networks. In doing so, this Visual Vignette brings our attention back to the infrastructure by leveraging architecture to create an experience where GPS fails, thereby inspiring reflection on how meaning emerges across the entire network, rather than being concentrated in the hands of the user. To make GPS infrastructure visible, Wood chooses architectural sites that have the potential to disrupt its usually smooth operation, such as spaces with limited lines of sight with the sky (e.g. narrow streets or building complexes with covered walkways and underpasses). During the walk, each person was given an android smartphone running an app which reverse-engineers the process of locating to show participants where the satellites are in relation to them. After walking around the site individually for some time, the attendees reconvened and drew and wrote responses to the experience around perceptions of infrastructure and surveillance. By gaining insights into how a hidden but essential technology operates, we are enabled to reflect on that technology’s implications.

The theme of sensor-based knowledge and related processes of infrastructuring is the focal point of Francis Lee’s and Erik Aarden’s chapter, respectively, that both focus on the different enactments of health security, legitimizing political actions on the grounds of contingent knowledge production. A fact that is not new in STS or CSS, but one that deserves special attention when it comes to the analysis of security regimes. Written during the global COVID-19 pandemic, these two chapters are timelier than we hoped they were. Drawing from post-ANT sensitivities and fieldwork at the European Center for Disease Control (ECDC), Lee discusses how different practices of sensing and making sense of the world have been used to argue for or the distribution of responsibility in the case of a salmonella outbreak. By utilizing the method of genetic sequencing and finding genetic similarities between geographically distributed mutations of the bacteria, the team at the ECDC concluded that the disease had its origin in a specific country. Yet, this mode of sensing had been contested on the grounds of another sensing practice, following the bacterium through transport routes and logistical infrastructure. Applying what Lee calls shoe-leather epidemiology, the opposition argued that there is no identifiable causal link connecting the outbreak and the country. Thus, different sensing practices and infrastructures had been brought in place to support different political claims in global health security regimes.

Similarly, Aarden uses the case of the Million Death Study (MDS) in India to show how human sensors are deployed and sensitized in order to create new forms of national health statistics. This new form of infrastructuring, he argues, brings into opposition two distinct matters of concern within the existing health security regime: first, the increasingly prevalent discourse on global health security with its focus on ‘exceptional events that may be anticipated with jointly developed digital sensing methods’ (Aarden, this volume). This marks an interesting shift in governance practices, as it means a transition from classical biopolitical governance towards the effort to prepare for singular and unpredictable events (Collier and Lakoff 2008). Thus, the MDS marks an attempt to contest a security regime that is built towards ‘politics of possibility’ (Amoore 2013). Second, the MDS applies a distinct form of sensing in contestation to the established clinical system. Combining interviewer skills for, what the study calls, verbal autopsies with standards for interpretation and machine learning applications, MDS is hoped to ‘access data on causes of death closer to the source and interpret that data more accurately’ (Aarden, this volume). More accurately here means also overcoming the bias towards urban regions inherent to the clinical system at the time. MDS sensing infrastructure contests not only the way of sensing but also what is managed within the health security regime, highlighting health issues of the rural households and those of low socioeconomic status.

In the case of ECDC, as well as the Million Death Study, the different sensing infrastructures are becoming each other’s brick wall and object of demolition. Providing different enactments as socio-political arguments for or against something, the focus shifts to the interplay of these diverse assemblages as infrastructures of contestation, where different enactments must be managed through negotiations (Mol 2002).

Discursive visions and perceptions of the world, entangled with the usage of technologies, are equally important to understanding how social orders are established. The idea of visions of an (un-)foreseeable future often drives the reordering of security regimes, as Jutta Weber explores in her essay on ‘wild cards’ as challenging traditional security doctrines. By focusing on highly unlikely, but potentially devastating events, a shift of orientation towards risky futures becomes the new mode of ordering in regard to thinkable interventions – also reflected in national security programmes. With that, another boundary is renegotiated: the way the (vision of a) future influences contemporary security orders. ‘Thinking the unthinkable’ creates future risks that call for action in the present. This dystopic performance of a potential future as a mode of establishing a social order has been reflected in STS research for some time now (Jasanoff and Kim 2009). With her contribution, Weber points at a specific form of reordering the present – not only by probable or possible events but also towards highly unlikely ones through the description of these wild cards. Sensing and making sense of the future and the present, in this case, works fundamentally different than in algorithmic or calculative forms of knowledge production – challenging our assumptions of what a sensor is and can be. Enacting security risks through wild-cards goes beyond the notion of probability and realizes non-calculative politics of possibility (Amoore 2009).

This question of how the future is being made sense of through sensors is also one that drives the Visual Vignette by Katja Mayer and El Iblis Shah. They explore the notion of human sensors and an interesting genealogy of prediction within security domains, based on the practice of consulting occult seers during the cold war to create predictions in a politically tense and potentially unforeseeable situation. Prediction, aside from risk calculation, became a fascinating element of security order, as they argue. The ways in which they provocatively put the spiritual human and an alternative construction of the future and security side by side, questions the dominance and the apparent objectivity of predictions, thereby creating a space to reflect on often implicit assumptions about practices of future taming and future-making.

In their chapter Visual Vignette as a format, Mascha Gugganig and Rachel Douglas-Jones situate it within the shifting grounds of STS’s knowledge infrastructures and discuss its affordances for work in STS. While their project originates in the anthropological embrace of multimodal, imaginative work (Collins, Durrington and Gill 2017, Elliott and Culhane 2017), the authors put their experimental engagements with analysis and communication of research in conversation with efforts to work across media that are simultaneously gaining prominence in STS (Ballestero and Winthereik, 2021; Dumit 2017; Jungnickel 2020; Le Bot and Noel 2016). Gugganing and Douglas-Jones then review the capacities of the Sensing In/Security Visual Vignettes to bring forward critical aspects of our sociotechnical world, and offer a guide for those who might be inspired to experiment with the format and its potentials of working with images alongside text, and to stay with the dissonance produced when a conventional tool (Powerpoint) is pressed into alternative, imaginative use.

The (supposed) invisibility of sensors and sensing infrastructures in the making of security issues and politics has provoked us to engage with the issue of representation in research and the form, normativity, and power of written words in more experimental ways. The three Visual Vignettes in this book all aim at breaking up the ‘division of labour’ of text as content and image as its illustrator as they engage the reader/viewer to critically reflect and rethink the dialectic between visuals and text. The genre of Visual Vignettes considers research, data analysis and dissemination tools as methodological infrastructure. It challenges us to reconsider the norms of common research, writing and communication practices that have defined STS, often borrowed and readapted from neighbouring disciplines. Methodological infrastructures, like all infrastructures are made and remade, leak and break and get fixed and repurposed. As such, this format allows us to make sense of sensors by creating new forms of visibility and tangibility, reflecting the multi-modal data that sensors capture, transmit and are part of.

The third major theme of this volume, sensors as boundary infrastructures and bordering practices, is addressed by Annalisa Pelizza and Wouter Van Rossem, who take up the question of reordering security and its boundary infrastructures by focusing on a network of migration hotspots. In this fascinating account, the authors combine empirical insights and a textual experiment to explore how ‘architectures of sensor networks and trans-national security orders’ can influence each other. First, the hotspots are what the authors call nodes of equivalence, where standards and procedures are homogenized, creating a space of comparability, connecting diverse national and transnational actors. Second, new forms of boundaries of responsibilities are drawn, and new forms of labour divisions between sensors at the periphery, i.e. migration hotspots, and centres of calculation are established. The double movement of renegotiating borders within the system of border security infrastructures and, at the same time, the blurring of boundaries between national security regimes shows the potential impact of sensor networks on social order(s).

In her chapter, Bracken-Roche contributes to the discussion of sensors and the renegotiation of boundaries and borders by showing how drones do not obey traditional bounds of state and security. The transgression of traditional boundaries between different spatial and political spaces is thereby the result of the economic interests of industrial actors. Drones, as sensing devices, transitioned from the military domain into the realm of civic applications – performing a securitization of risk and publics through technologies constructed for military needs. Bracken-Roche argues that domestic drones are commonly framed by industry groups as benign sensing technologies as compared to militarized drones, while at the same time security professionals deploy particular narratives about drones to suit economic and political agendas. The chapter highlights how drones in Canada, in both civilian and military applications, represent a technological zone (Barry 2001, 2006) and how these sensing machines dramatically shape public spaces and impact individuals across various contexts.

Aiming towards an at least temporary demolition of disciplinary borders, the experimental chapter by Jan-Hendrik Passoth, Geoffrey C. Bowker, Nina Klimburg-Witjes and Godert Jan van Manen addresses questions of, and experiments with, possible forms of engagements between social science, hacking and security policy through a conversation on ‘Infrastructures, Security and Care’ over the course of two years. Their aim is (at least) two-fold: First, they explore novel ways of listening to, and discussing and engaging with people who are experts on sensors outside of academia – yet explicitly not in a sense of an extraction of knowledge and information, which almost always creates the risk of patronizing or exploiting the expert engineer, but as a form of mutual exchange of perspectives, questions, and issues. Second, the contribution is an experiment with novel formats, looking for ways to integrate these engagements into an academic, edited volume while being sensible to the different work logics as well as the different disciplinary logics of crediting (academic) work and the challenges that bear for traditional processes of academic peer-review.

Conclusion

Sensors and sensing infrastructures are neither neutral nor innocent but imbricated in politics on all levels, from international migration to sensing (genetic) evidence for disease outbreaks, from biometric to aerial surveillance, from huge data centres to satellites to tiny cell phone sensors eavesdropping on our conversations. Sensors often do invisible work, while simultaneously making (perceived) threats experienceable. We might thus say that where there are sensors, there is also governance. But then, where are the control rooms, and how are agencies arranged between people, things and politics in sensing security infrastructures? Building on and linking work from science and technology studies, security studies, critical data studies, sociology, and anthropology, this edited volume tackles these questions as it seeks to understand the role of sensors in the making of transnational security infrastructures. Sensors contribute to the production of in/security in manifold ways, producing in/visibilities and modes of in- and exclusion. Sensing realties raises questions of what is being sensed in which way, and visible to whom. Sensing therefore draws boundaries on different levels, sorting actors into sensed populations, regulating access to sense-making tools, or producing discipline through the visibility of sensing processes. The relation between in/visibility and in/security is thereby not always straightforward. Becoming the target of a weaponized security system creates insecurity without a dash. However, being excluded from sensing regimes based on lack of health insurance creates – as the current pandemic painfully demonstrates – vast insecurities. In/securities thereby are the result of in- and exclusion processes of at least three different dimensions, which are reflected in the collection of this book. In/visibility of sensing, sensing as knowledge production, and the construction of (new) borders.

First, the in/visibility of sensing devices and possible processes of infrastructural inversion. Here, we bring together work in STS on the (in)visibility of infrastructures with studies interested in security and surveillance. Research in STS and adjacent fields on the nexus of visualization and materiality has continuously engaged with questions of how ‘things are made visible’ and ‘which things are made visible’, and investigates ‘the politics of visible objects’ (Kuchinskaya 2014; Rose and Tolia-Kelly 2012: 4). The emergence of sensors is connected to the social orders they co-constitute. Yet, STS has not only attended to the tendencies of infrastructures to fade into the background, but indeed also shown that there is movement, a process in which some (parts of) infrastructures become visible whereas other (parts) become invisible. Thus, an important question to raise is: when do we make these infrastructures of sensing visible and to what end? Sensors tend to become invisible or so much part of our daily life that the enactment of in/security only becomes visible to certain stakeholders, while others are only included as objects of inquiry, but excluded from the sensor data informed security discourse. Visibility thus becomes not an effect, but an issue, as surveillance ‘has become increasingly unaccountable and less and less visible to ordinary people’ (Lyon 2015).

Second, the collection contributes to work interested in the social construction of sensor-based knowledge and related processes of infrastructuring. As Star (2002: 116) put it: ‘One person’s infrastructure is another’s brick wall, or in some cases, one person’s brick wall is another’s object of demolition’. Through different sensing practices, different versions of the sensed world are created, including or excluding issues, people, sensations, and geographical places, creating the basis for different argumentations and rationales. As such, sensing infrastructures are always political, as they enact varying matters of concern (Latour 1999). Taking up on this observation, the contributions of this volume exemplify how the different ways of sensing become the basis for making or contesting political arguments on security issues. This dynamic is illustrated by health infrastructures and the question of sensing health incidents. Political and health care systems have a tremendous impact on how the tests are distributed and how the distribution of the virus is made visible.

Lastly, the book engages with sensors as boundary infrastructures and bordering practices. Information streams and communication structures are often integral elements of the way a state or other big institutional setting is organized (Mukerji 2011). Sensor infrastructures are no exception. They play an important role in the production of political entities, social orders and the production of manifold boundaries by moments of performative integration of actors. This integration – and with it also always moments of exclusion – can be explored from at least two different perspectives. Starting from the idea of infrastructuring (Pipek et al. 2017), the spread of trans/national networks defines moments of connect-ability and the forms of possible interactions between different elements within these networks. Are you using the same protocols, the same standards (Bowker and Star 1999), and is the distribution of tasks compatible with the broader systemic practices? With the production of transnational sensor infrastructures, national boundaries seem to be pierced and weakened while other boundaries are produced.

This collection contributes to a growing literature on the diverse processes of both securitization and normalization as integral to these infrastructures, along with their performativity in the making of boundaries and borders. Instead of solely focusing on the specific sensory devices and their consequences, the book engages with the emergence of sensing infrastructures and networks, and how sensing devices become invested with socio-political significance. By paying attention to sensors as an important part of the material equipment of in/security practices, this collection unpacks sensing as situated practices of constructing, reconfiguring, stabilizing and disrupting in/security. As such, it encourages us to be both critical and hopeful that networks of in/security withstand drives to build all-encompassing surveillance regimes. There are always modes of contingency and practice which exceed the panopticon – which is necessarily always incomplete, but whose power is multiplied by beliefs that it is all-encompassing. Securing our futures entails living joyfully with insecurity.