Digital witnessing and the erasure of the racialized subject

Abstract

Humanitarian agencies are relying more frequently on remote sensing, satellite imagery and social media to produce accounts of violence. Their analysis aims at creating more compelling narratives for the court of law or of public opinion and has contributed towards a forensic turn, thus complicating the already fraught relationship between the practice of witnessing and political subjects. This article explores how digital witnessing allows us to ‘see’ further and deeper into places that are at a distance from us, whilst at the same time creating the conditions that make certain subjects recede from view. I will discuss these issues in relation to a country I am familiar with and one that has been central to the forensic imagination – Pakistan – although the particular geographies within Pakistan that this imagination works with are not mine. Thinking with non-linear temporalities of violence, I explore how the forensic turn may have actually contributed to the erasure of the racialized political subject.

Contemporary visual regimes of modelling and mapping are producing representations of worlds that are all encompassing. From NASA’s Blue Marble, which is a composite image of our planet to computational models produced through data aggregation and the virtual realms of computer games, there is a certain sense that we inhabit a world of totalities, where the horizon beyond which things might remain uncertain or incalculable seems to have disappeared. This deterministic world view can, in many senses, be attributed to digital technologies of sensing, data analysis and visualization. In the context of the witnessing of violence, these technologies are also being deployed more frequently by humanitarian agencies and human rights groups in the form of remote sensing, analysis of satellite data and images from social media to produce evidence on acts of violence. Such analysis aims at creating compelling narratives for the court of law or of public opinion and has contributed towards a forensic turn in the humanities and social sciences (Sharp, 2015; Weizman, 2011). It has also complicated the already fraught relationship between the practice of witnessing and political subjects. This article explores how digital witnessing allows us to ‘see’ further and deeper into places that are at a distance from us, whilst at the same time creating the conditions that make certain racialized and subaltern subjects recede from view. Such seeing operates through a fantasy of a frictionless world that can be accessed by zooming in and out of far-flung places viewed through scopic regimes that render certain aspects hyper visible while obscuring others. It also removes embodied experience while assuming making visible to be an emancipatory act in itself. What forms of counter visuality are possible when the witnessing of violence deploys such technologies and what political possibilities does it produce or foreclose?

As Eray Çaylı points out in the introduction to this themed issue, the post-Holocaust impulse to expand the practice of witnessing beyond the individual subject has led to an interrogation of the relationship between the production of truth and the social, political and environmental contexts within which such truths are embedded. While the need to situate witnessing within this wider context is clear, the question of which acts and what entanglements can become part of a process of witnessing remains open. Here I would like to discuss one aspect of this shift, the emergence of what has been referred to as constellations of witnessing, that is an assemblage of humans and other-than-humans imbricated within the process of witnessing (Kramer and Weigel, 2019; Sheikh, 2018). Such modes have emerged within contemporary art and architectural practice, and their involvement with human rights issues, often working in collaboration with NGOs, activists and community organizations.

The missing witness and the erasure of the subject

This work is influenced by Giorgio Agamben’s account of the paradox of witnessing in Remnants of Auschwitz (2000), which revolves around the figure of the absent witness. Agamben’s critique begins with the identification of the two traditional modalities of witnessing through interrogating the Latin legal terms testis and superstes. The former denotes a third party to an act that may claim an outsider’s position and the latter to someone who has lived through an event, but both parties are in a position to have knowledge of the event and are therefore able to bear witness. In contrast, for Agamben: ‘The “true witnesses”, the “complete witnesses,” are those who did not bear witness and could not bear witness’ (p. 34). This assertion emerges from the specific context of the Holocaust and in particular the figure of the Muselmann, those people who had been completely dehumanized by the atrocities of the concentration camp, and who, according to Agamben’s account, were hovering on the edge of life and non-life, being there but not present and therefore unable to fully witness the events of the concentration camp (did not bear witness); they were also murdered by the Nazis just before their defeat (could not bear witness). For Agamben, these people were ‘those who “touched bottom”: The Muslims, the drowned’ (p. 34) and he goes on to say that those who did survive were bearing witness or giving testimony in their stead. This figure of the missing witness is at the heart of contemporary debates and practices that seek to expand the process of witnessing beyond its traditional focus on the individual testimony and towards viewing testimony as trace found in environments through scientific process or through sifting, uncovering and reading archives against the grain (Schuppli, 2020; Weizman, 2011).

However, this conceptualization of the absent witness is contested by a number of authors who question its exceptional nature both within the context of the camp and a wider historical context. Agamben’s argument is based on the first-person account of the concentration camp by Primo Levi (1996[1958]). Commenting on the differences between Levi’s account and Agamben’s conceptualization, Alexander G Weheliye (2014) writes that, for Levi, the Muselmänner were humanity reduced to an undifferentiated mass through their sheer numbers and the brutal severity of what had been done to them. In contrast, for Agamben they were an individualized category that demonstrated the exceptional nature of the camp. Referring to a study conducted after the defeat of the Nazis that interviewed survivors, Weheliye writes that ‘prisoners in the camps moved in and out of being-Muselmann, and, depending on the time of year and political situation, 50–80 per cent of inmates could be classified as belonging to this category’ (pp. 119–120). Understood through this lens, the Muselmann was not an exceptional figure but a common state of being in the camp.

Additionally, something uncomfortable remains in the use of the term Muselmänner/Muselmann. I am not the first to critique Agamben’s use of the term Muselmann (Muslim) for those who have been completely dehumanized (Jarvis, 2014; Mesnard, 2004; Weheliye, 2014). In a careful and detailed account, Jill Jarvis exposes the discrepancy between the meticulous way in which Agamben analyses juridical categories such as homo sacer, testis, superstes and the taken as read of the odd naming of those at the edge of life as Muselmann, remarking only on its ironic nature. This lacuna also has an historical and colonial context as Jarvis (2014: 709) points out:

Agamben nowhere acknowledges that the use of the epithet ‘muselman’ at Auschwitz coincided with its simultaneous function as a juridical category of exception experimented with by the French in Algeria since at least its 1848 departmentalization – when the constitution of the French Second Republic annexed Algeria to France, carved it into three departments, and drew a juridical distinction between ‘les citoyens français’ (bearers of full citizenship rights) and ‘les sujets français’ (subject to military conscription, forced labour, and a disciplinary system that included concentration camps).

For Jarvis, this inability to connect events in Europe with the colonial contexts that they are entangled with allows the term to go unremarked. It also produces a situation where Agamben can read the events of the Holocaust as a decisive break and the Nazi concentration camps as a complete exception, while mentioning in passing the first camps that were built by the Spanish and the British in their respective colonies to incarcerate the native peoples of those lands. What does it mean then for contemporary art/activist practices, many of whom operate in post-colonial or settler colonial contexts, to be using as their central touchstone a conceptual formulation that elides the history of colonization in order to produce an account that creates a decisive break or a state of exception? Michal Givoni’s (2016) account of witnessing begins to question this understanding of the Holocaust as a singular event (although she does not comment on the naming of the Muselmann either), showing how such an approach ‘downplay[s] the actual manifestations of witnessing and testimony and the mundane political issues they raise because they treat witnessing and testimony as metonyms for epochal and existential crises’ (p. 91). Givoni’s critique makes room for the work of grassroots politics and activism within practices of witnessing rather than seeing the question solely through the lens of crisis or exception in Agamben’s terms, knowing also that in the colonies the exception was in fact the norm.

This leads back to the question of which acts and what entanglements become part of these witnessing constellations, for whom are such constellations produced, and in what contexts can or do they operate? And particularly for the purposes of this article, what happens to the racialized subject within these composite accounts of the witnessing of violence? These questions do not appear in a neutral world but one where the NGO and development industry that is built on the violence and ruin of war, conflict and structural injustices also produces the abject representations of Black and brown people that are so pervasive. The post-representational turn that accompanied the end of the era of the witness has rightly critiqued such images for producing accounts of certain people and places only through the lens of violence, but perhaps what has been explored less is how this relationship is also affected by an underlying understanding of Black pain and trauma especially as being of a lesser degree. This apathy towards certain kinds of images, I would suggest, as Christina Sharpe (2016) amongst others has commented, is less from a general numbness from seeing too much pain and suffering, and more from certain kinds of pain being illegible to certain audiences. Within the discourse of witnessing constellations then, that are also often working within an evidentiary paradigm, what happens when the audience that such constellations are addressing refuses to see, or to apprehend? The forensic idiom emphasizes investigative approaches precisely by claiming that these negate the effects of certain types of violence being less visible through an appeal to the evidentiary. In the context of social media where information is exchanged in the absence of traditional gatekeepers and fact checkers, facts have become malleable entities for those who could have relied on Western media to produce trustworthy accounts. These problems are only being compounded by the use of sophisticated technologies, such as the algorithms that produce deep fake videos. In such a context, discerning the line between fact and fiction has become in itself a highly complex task leading some commentators to call for the need to protect ‘the authority of established truths’ (Weizman, 2019). That we need some common ground from which to address each other is as crucial now as it was before what has been termed the post-truth era, but the form that this ground takes is a question open to discussion, as is the quality of the truth upon which it may be negotiated. There are also a number of important aspects that remain unaddressed within a simple recourse to the evidentiary, including the temporality of violence. Witnessing violence after the fact, or to put it another way, to consider the temporality of violence through an event and its before and after, that is within the linear logic of causality and of finding perpetrators, often leaves intact and unremarked the underlying conditions that make certain types of violence invisible and so in many cases leaves all the careful work of proving in an investigative manner and through an evidentiary logic impotent. Because the question is not only of not believing, it is also of not caring, something that has been made so painfully obvious in the treatment of refugees and undocumented people at the edges of Europe. Further questions relate to the positionality of those who are producing evidence within open source investigations that often collect and collate first-person accounts from social media. While the original material in the form of digital images and video is posted by those on the ground and close to the event in question, it is verified, edited and narrated by people who are overwhelmingly situated in Europe (Dyer and Ivens, 2020).

Digital witnessing through the modelling of worlds

One influential example of a witnessing practice based on evidentiary claims is Forensic Architecture, a group that has developed a method for using digital technologies to mediate the testimonies of people, objects and environments. In commenting on the role of forensics, Eyal Weizman points to the etymology of the word through the Latin term forensis, which refers to the forum, a place where democracy and arguments are played out in the context of everyday life (Weizman, 2017: 65). In this definition then, witnessing requires the convening of an assembly or a forum, similar to Bruno Latour’s ‘parliament of things’ (Latour and Weibel, 2005). The difference perhaps lies in who or what is allowed or able to enter the forum. For Forensic Architecture, it is the ‘materiality of buildings or of bones’ (Weizman, 2017: 74) that is considered important and Weizman, in narrating the forensic turn that his practice sits in the middle of, gives a number of important examples as precedents. These include the search for the bodies of the disappeared in Argentina during the military dictatorship (1976–1983) and the uncovering of mass graves in the aftermath of the Balkan conflict where body parts were scattered across several sites. Such examples clearly show that, for the inception of this account of the forensic approach, the dead body was central. This is an important point in relation to the earlier discussion on Agamben and the paradox of the missing witness; these are all contexts where there are few or no living testimonies to be sought. In engaging with some of these debates on forensics, the geographer Jo Sharp finds this centrality of the body useful for a feminist approach to geopolitics precisely because of the importance placed on the body and on the materiality of the remains, as this counters the abstract tendency in traditional modes of thinking through geopolitics (Sharp, 2020). But what might be the consequence of deploying a conceptual framework and mode of attending to the aftermath of violence that comes only from an engagement with the dead? That is, what does it mean to keep turning to the dead body when the living are also there to give their testimony? Clearly, in the particular examples cited above there is an important rationale for the focus on the dead body but, as will become clear in the case I turn to later, such an approach is now deployed quite freely in contexts where the living are there to give an account of their lives. Ignoring their narratives is especially concerning when often these are people and communities who are actively countering the many forms of violence they are living through. Could it be that in these contexts the racialized existence of the living makes them recede from view so that Agamben’s paradox can be so easily deployed?

In bringing together such constellations of witnessing, the digital holds an important place, whether it is the DNA analysis of bones or an analysis of satellite imagery for war or environmental crimes, often a key methodology is to create digital models or timelines of particular events in order to unfold and situate the various testimonies being brought together (Brown Moses Media Ltd, 2016; Forensic Architecture, 2014; Kurgan, 2013). The all-encompassing vision of the digital here is different from the familiar god’s eye view that seeks to render this world visible through an objective or distanced perspective. The digital model is not a claim to a view on the real but a simulacrum of a world produced through its own functions of collation, composition, calculation and visualization. While this modelled world can be messy, is incomplete and often highly complex, the efficacy of digital witnessing relies on creating singular narratives of violence. Here, questions of culpability, of bringing the perpetrators to justice and of revealing the truth are central. Within a spatialized understanding of violence, the perspective is not of the victim but a composite that might emerge from the recording and sensing capabilities of many devices, materials and sensors. These are brought together in a composite or a model that might have many holes but is nonetheless constructed to remove doubt, with each point-of-view or each angle providing evidence that negates other possibilities. In this way, a convincing narrative account of an event is produced through investigative forms of analysis. In these worlds created by bringing together a composite set of data from a variety of sources, narrative is used as a key device to help us navigate this complex model. But one important consequence of the removal of embodied experience is the transformation of perspective. Within the horizonless worlds of digital models, what anchors or political tools are available to orientate ourselves?

Such models are constructed in conversation with Harun Farocki’s influential thesis on the operative nature of images, which was based on an analysis of the images emerging out of the Gulf War and the workings of cruise missiles (Hoelzl, 2014). These were images whose central function was not to represent something, meaning that they were not depictions, but were instead involved in operations such as tracking, navigation and other forms of control, in this case particularly in relation to the military. Farocki’s later work (the film series, Parallel I–IV) interrogated the modelled worlds of computer games and their technologies and techniques that practices like Forensic Architecture mobilize; for example, the gaming engine Unity and Adobe After Effects are key software in this kind of work. In thinking the political dimension of such software and the modelled worlds they produce, Farocki emphasized the importance of navigation – that is, once you have constructed this world you have to be able to navigate it in order to produce certain narratives. The narrative should at this point bring back a certain perspective and a point-of-view, but unfortunately and perhaps because of the evidentiary character of the forensic approach, this now becomes a voice of authority guiding us through the world that has been constructed and so we arrive back to the god’s eye (Weizman, 2019). Farocki’s film series seemed to be highlighting navigation as an incredibly important political tool in understanding and resisting state and corporate power, that is as we move through a world of burgeoning data towards what seems like the complete entwining of our lives with algorithmic production and personalization, a key form of resistance might be situated in finding modes of counter-navigation that allow other ways of moving through what are increasingly ubiquitous and all-encompassing worlds (Awan, 2020; Mende, 2017). I am unsure whether the production of modelled worlds that aim to reproduce complete and unchallengeable narratives necessarily help in this regard.

An important case for the development of some of these techniques of spatial modelling and narrative in the early work of Forensic Architecture is the US-led drone bombing of Pakistan and Afghanistan in the context of the so-called war on terror. It would seem that Pakistan, or more precisely what were then the Federally Administered Tribal Areas (FATA) ¹ of the country, constituted a practical case study for working at the limits of what could be seen from the remote sensed view of the satellite. Additionally, the technique of ‘situated testimony’ seems to have been developed when interviewing a survivor of a drone attack in North Waziristan (Forensic Architecture, nd). This technique involved interviewing a witness using a three-dimensional model of the location as an aid to memory in order to establish the facts of an event. In contrast to this practical definition, the film scholar Janet Walker defines situated testimony as ‘figuring the relationship among the body of the individual, the ground from which s/he speaks, and the past events that transpired of a time but are, at the same time, brought into being by the testimonial act’ (Walker, 2009: 85). For Walker, then, situated testimony is a negotiation between time, place and memory that is invested in what she calls ‘the materiality of testimony’, where situatedness means being at the location of an event rather than its proxy as is the case with the digital model. Despite the importance of being located in situ, for Walker, place is not reified but is helpful for revealing the contingency of testimony, based as it is on a set of social and environmental relations, a complexity that is unable to emerge through the simulated space and distanced perspective of the digital model. Commenting on the scopic dimension of their approach, Weizman (2017: 13) writes of the ‘threshold of detectability’ in relation to the resolution of satellite images. He describes in particular how the size of the hole made in buildings by the missiles used in places like Waziristan made them almost impossible to detect in publicly available satellite data. Unlike other missiles that would detonate on impact, these were specifically designed to pierce through roofs and explode a few seconds later leaving the building largely intact while killing those inside. The undetectability in satellite images of the damage caused to buildings by drone warfare is described as a ‘forensic-architectural problem’, one that forced an examination of ‘the relation between an architectural detail, the media in which it could be captured, a general policy of killing, and its acts of denial’ (p. 27). It also prompted the use of a modelled environment with a drone survivor to reconstruct her home in order to reveal the damage caused. While it is true that, described from the perspective of the physical damage caused by the drone strike, the forensic description holds true, this definition also serves to obscure other forms of violence perpetuated by drone warfare, and their associated trauma, that worked across different temporal registers and spatialities other than that of the building as object.

The US falsely claimed that the drone campaign produced very little collateral damage to other people or property due to the precision of the bombings. It is beyond the scope of this article to refute that narrative but detailed investigations by the Bureau of Investigative Journalism (2004) and Amnesty International (2013) have shown that many innocent people were killed going about their everyday lives. What is perhaps more salient to note is that just as the US argument around collateral damage focused on falsely reporting the number of deaths, or not counting them at all, a concomitant problem was the denial of the wider impacts of these bombings. Anyone from or living in Pakistan would find it difficult to speak of drone bombings without mentioning suicide bombings. The US and its allies have continued to deny the contention that the impunity of drone warfare was directly responsible for radicalizing people in FATA who reacted to the threat from above by becoming suicide bombers, targeting cities across Pakistan for the state’s complicity in the attacks (Akbar, 2015; Bashir and Crews, 2012; Pilkington and MacAskill, 2015). ² The link between these two acts of violence is crucial as it shows that the bombings cannot only be understood as singular acts, even if we remain at the level of apprehending violence solely through spectacular events that occur within the logic of a linear and causal temporality. What kinds of witnessing constellations should we convene to gain an understanding of the complex entanglements of these disjointed yet connected events?

Accounting for slow and disjointed forms of violence

Thinking with the figure of the suicide bomber, an investigative approach might lead to an understanding of the type of bomb used, who the person was and where they came from, that is, a factual account of what occurred. Or to put it in terms coined by Bruno Latour (2004), this would mean approaching these events as ‘matters of fact’ rather than viewing them as ‘matters of concern’, which would necessarily entail thinking about the value placed on certain lives over others and about forms of violence as they concatenate across lives and spaces. During the mid-2000s, Pakistani towns and cities were being regularly hit by suicide bombings as brown people paid the price for the so-called war on terror. In 2015, three quarters of all suicide attacks occurred in just three countries – Afghanistan, Pakistan and Iraq – and out of those only Pakistan was not officially at war. The discourse around these events in the Pakistani national media revolved around the question of who the suicide bomber was because their motives were already assumed to be related to the brainwashing potential of so-called Islamist ideology promoted by the many variants of the Taliban and Al-Qaeda operating in the region. ³ It also served the interests of the Pakistani state to suppress any discussion on whether the drone attacks were radicalizing these young men. Instead, they promoted a narrative of the ‘terrorist-monster’, as Jasbir Puar and Amit Rai (2002) describe the disturbing conflation of gender and sexuality within the discourse on terrorism. At this time, a rather gruesome phenomenon emerged in relation to the bombers that both fascinated and repelled the Pakistani public in equal measure. ⁴ At some point, in the weekly, often daily bombings that were occurring, the belts used by the bombers were designed in such a way that when they detonated the force of the bomb would push up so that the face of the person wearing the belt would be lifted clean off their head in one piece. While their body would be scattered in small parts their face would essentially remain intact. This was happening with such regularity that these came to be known as ‘face masks’ and they allowed the authorities to determine very quickly who the person was until the potential bombers realized this and made sure to look down when detonating the bomb.

At first glance, the story of the face masks seems to fit into the narrative of ‘terrorist-monsters’ especially due to the gruesome nature of these artefacts, but seeing the face of a suicide bomber prompted – at least in some parts of the Pakistani public – deeper questions about who they were and what led them to this final act. Finding out the identity of a person is not so difficult within the militarized context of Pakistan, especially as many of the bombers were from FATA and it is not possible to travel into or out of that region without a Shenakht (ID card) and a Rahdaari (a kind of temporary pass) that allow passage through the numerous military checkpoints. A young man from FATA, that is areas of the country that were heavily affected by the US drone surveillance and bombing campaign, would need an ID card and one of these passes to be able to travel even a small distance from their home, let alone to another province in the country. As Puar and Rai (2002: 119) state, ‘the monster is not merely an other; it is one category through which a multiform power operates. As such, discourses that would mobilize monstrosity as a screen for otherness are always also involved in circuits of normalizing power.’ Seeing the face of the person who was the suicide bomber revealed something of the circuits of power that refused to make the link between the hi-tech bombs dropped by drones and the lo-tech bombs of those who committed suicide. It also said something about the nature of the violence at the heart of these bombings, which are of course extreme and spectacular events, but the longer temporality and wider context that led to their emergence is a story of what Elizabeth Povinelli (2016: 40) calls “quasi-events” ‘. . . a form of occurring that never punctures the horizon of the here and now and there and then and yet forms the basis of forms of existence to stay in place or alter their place.’

If we were to read the events of the bombings in this longer timeline it would include the long struggle of the Pashtun people against the governance of their lands by the draconian Frontier Crimes Regulation (FCR), which meant that until recently the area was under the direct control of the federal government through executive powers, rather than having representation in the provincial assembly as all other areas of the country. The colonial law meant that during British colonization there was a degree of self-rule but the same law also stipulated collective punishment across tribal structures. Following independence, FCR meant that the people of FATA had no recourse to the Supreme Court of Pakistan. The frontier was thus constructed as a spatial, political and administrative category that in recent times has meant that FATA has had its own administrative system that was largely unaccountable and resembled a series of fiefdoms with individual political agents in charge answering direct to the federal government. In May 2018, FATA was merged with the adjoining Khyber-Pakhtunkhwa province marking the formal end of FCR, although the dissolution of the powers of political agents in everyday legal, political and bureaucratic processes remains a challenge. The Pashtun people’s struggle for equal representation is the backdrop for the drone bombings in addition to their struggle against the foothold that the Taliban gained in the area, as well as the Pakistan army’s operations against these groups from 2004 onwards. A combination of all these factors has not only killed people but destroyed centuries-old forms of life, poisoning water sources and orchards, and destroying markets and homes (Wazir, 2018). Other more diffuse forms of violence were also unleashed by the drones, such as the extreme anxiety and stress caused by the continual presence of the machines hovering in the sky above and the incessant noise they generated; this produced the kind of slow violence that Rob Nixon (2013) writes of in relation to environmental catastrophes. How can such diffuse forms whose temporality exceeds and disrupts the causal logic of violence considered in the forensic idiom be accounted for? What kinds of testimonies or witnessing constellations would be required to account for this violence that continues to haunt its living victims?

Mapping and the production of a stable ground

The techniques of visual and spatial analysis deployed in Forensic Architecture’s investigation of the drone bombings relied heavily on remote sensing techniques due to the stated difficulty of gaining access to the region.

This violence of secrecy – practised not only by the government in the region, but also by the Taliban – necessitates, for the most part, remote sensing in the reporting of what happens in FATA. Of course, many people still live there, and movement is not entirely restricted, but most journalists are now dependent on information provided by local sources as opposed to what they can see themselves. (Burns, 2014: 405)

In thinking about who or what counts as a witness it is useful to pause for a moment on the distinction being made between local sources and journalists. What does it mean to assume that those who have lived through these atrocities cannot bear witness to their own traumas? It is a violent act in itself to discount the stories told by local people of lives that have not only been lost but have been subjected to protracted violence that continues today in the form of internal displacement, anxiety and exhaustion.

The remote sensed mapping of drone strikes produced by Forensic Architecture was based on data collected by the Bureau of Investigative Journalists and collated in an interactive map that through spatial visualization created a pattern analysis of drone strikes, the types of destruction caused, numbers of deaths and injuries. This mapping used as its ground a map of the area produced by the CIA, which the group acknowledge must have been the basis for the drone war being waged and thus constituted a problematic shared point-of-view. They note that this shared viewpoint ends at the moment of the strike because they do not have access to the video footage made by the drones of each attack. While having access to the drone footage would likely have given definitive proof of individual bombings, it does seem strange to lament the lack of a shared perspective with those who perpetrated these extrajudicial killings. Of course, what the counter-investigation could have had access to, had they not discounted them already, were the local sources whose accounts were perhaps not seen as reliable enough. As Madiha Tahir (2017: 6) writes in relation to what stories get heard: ‘How stories travel has a lot to do with their truth-value; some travel a long mile to the center but remain unheard, some embody centrality wherever they are.’ And herein lies the problem, Forensic Architecture’s location in the global North gives their narrative a certain credence over the stories that come from the region and, through discounting such narratives, they leave intact the underlying conditions that make certain accounts of violence invisible. Forensic Architecture acknowledged their investigation as being very limited due to the exceptional legal status of FATA, proposing instead ‘that one way to develop an investigation of drone strikes, and to overcome such an investigation’s limits, is to start to incorporate within it a call for change in local conditions’ (Forensic Architecture, 2014: 401). What I find difficult here is that there is a long, committed and extremely difficult history of the fight to end FCR in the region, which encompasses complicated issues of local policing, the role of tribal leaders and how traditional forms of decision making such as jirgas are incorporated into national and provincial law (Alam, 2018). This complexity extends far beyond the headline of a lack of appeal to the Supreme Court of Pakistan, extremely important though that is. To have not referred to this ongoing and complex set of issues and struggles is, I would suggest, to follow the colonial logic of frontier governmentality as Benjamin Hopkins (2015) has named it, to think of the people in these areas as being absolutely subsumed within a state of exception with no agency. The recent rise of the Pashtun Tahufuz Movement, a very strong grassroots political movement, shows just how wrong this conception is. It would seem that a materially-led forensic approach is often unsuitable to be applied by investigators located far away, in this instance in the UK, because it relies on an intimate scale that is not possible. This intimacy can come through a traditional route of gathering the stories of those who were close to the events, but importantly it also requires a much stronger understanding of the place and its culture that might emerge through longer commitments and in collaboration with those whose lives are affected.

The question of the ground is brought up by Tahir in a discussion of everyday life in Waziristan under the shadow of drones. She writes of how the ground cannot be trusted because it quite literally moves due to the mines buried below but also of the fluidity of the ground due to the way that laws and rules are applied so haphazardly and are subject to the whims of various law enforcement agencies. She describes her own struggle in working out how people managed to move in and out of this heavily controlled area and that every answer she got was somehow contingent; that movement depends on what road you take, the time of day, who is manning the check posts, the risk you are willing to take etc. (Tahir, 2017). In contrast, in a section entitled ‘The Importance of the Ground’, Burns (2014) writes of the need to end FATA and to make the ground stable, an easy causal assumption that emerges from a lack of understanding of the place and how it functions. The ground was not unstable because of FCR only as that regulation was very concrete, but because of the way it was applied in practice at the whim of security personnel and political agents. We can see this in the current situation where formally FCR has ended but the ground remains unstable due to the mismatch between what is written in formal law and its everyday instantiation in people’s lives.

Making room for unstable testimonies situated elsewhere

Forensic Architecture recount the situated testimony of the one person affected by drone bombs that they did speak to, a woman who had made it back to Europe from Waziristan. Weizman (2017: 44) writes about the unstable nature of what constituted an account of her experience, explaining: ‘Victims might remember what happened before a traumatic incident or after it, but the closer one gets to the essence of a testimony, to the heart of the most violent incidents, the more elusive memory can become.’ For Weizman, it is these lacunas that should be mobilized in a forensic-architectural account and here we see the notion of the missing witness return in a different guise. While not wanting to downplay the importance of the woman’s testimony or the efficacy of situated testimony as technique, problematic questions around the temporality of violence remain, as well as the crucial question of where one has to be situated in order to give a situated testimony. What was it about the move from Pakistan to Germany that made this woman more than a ‘local source’ and thus allowed her testimony to become admissible within the definition of witnessing operating in the global North? At that time in 2010 when this testimony was taken, an estimated 254,000 registered internally displaced people from South Waziristan were living in camps in Dera Ismail Khan outside FATA in Khyber-Pakhtunkhwa province (BBC Urdu Service, 2010; UNHCR, 2010). While I do not want to contend that evidence of drone warfare was so easily available, it may have helped to turn attention to places where those who had fled the bombings were situated. The elsewheres of a Western gaze are too many to name but DI Khan would be one of those places, and those very same displaced people continue to endure the long arc of slow violence (Nixon, 2013: 8).

The question not only revolves around the missing witness or parts of a testimony that are obscured due to the trauma of violence, but what is also at stake is the instability of testimonies that are not always enunciated in the tense of Western juridical practices. Following Povinelli’s cautionary remark that ‘the tense of the other is not always at hand’, it seems imperative to argue not only for understanding violence beyond causal temporalities, but to also acknowledge the impossibility of a stable ground from which to produce evidential narratives on violence (Povinelli, 2011: 52). As Walker (2009) writes, the place of trauma is never a given, even if it happened to be someone’s home and a space of belonging; the very fact of violence produces an othered space. In this sense, this is not a call for a return to the fetishization of the individual testimony, but an appeal for witnessing constellations that make room for the racialized, gendered and environmental relations that are embedded within the experience of violence and its aftermath. At a time when evidentiary practices are being positioned as an antidote to a post-truth world, we still need other ways of knowing that can encompass a politics beyond the logic of pre-emption and violence understood only in the idiom of the spectacular.