Challenging the placeless imaginary in digital memories: The performation of place in the work of Forensic Architecture

Abstract

This article discusses the imaginary that digital memories are ‘placeless’, which refers to the notion that, among the changes that memory has experienced under the impact of digital media, one is the loss of a significant link to place. I contend that digital memories are not placeless memories, but, on the contrary, that place is central for digital memory work. Drawing on the concept of ‘performation’, I seek to demonstrate how place is created, executed and staged in the digital ecology. To illustrate this, I analyse two works of the research agency Forensic Architecture, ‘Hannibal in Rafah’ and ‘Saydnaya: Inside A Syrian Torture Prison’, in which ‘place’ plays a central role in the shaping of digital memory. Digital media’s ‘inherently archival’ nature acquires a different meaning in these works: it is the process of collecting and mapping the dispersed fragments of data that transforms the archive into memory.

One of the most persistent imaginaries of digital memories is that they are ‘placeless’, which refers to the notion that, among the changes that memory has experienced under the impact of digital media, one is the loss of a significant link to place. According to this viewpoint, while in the pre-digital era memories had a spatial anchor, which was an essential part of the production of their meaning, nowadays this link dissolves, giving rise to fluid and free-floating entities, fragmented, diffuse, ubiquitous memories, continually in motion, which cannot be assigned to precise spatial coordinates. The loss of a meaningful bond with place connotes, in this imaginary, a loss of meaning, given the centrality that the category of place plays in the shaping, preservation and transmission of collective memory. In this article, I contend that digital memories are not placeless memories, but, on the contrary, that place is central for digital memory work. Following Cresswell (2009), I define place as a ‘combination of materiality, meaning and practice’. Drawing on the concept of ‘performation’, I seek to demonstrate how place is created, executed and staged in the digital ecology – all of which are actions that are encompassed by the term performation. To illustrate this argument, I analyse two works of the research agency Forensic Architecture, ‘Hannibal in Rafah’ and ‘Saydnaya: Inside A Syrian Torture Prison’, in which ‘place’ plays a central role in the shaping of digital memory.

The imaginary of placeless memories figures prominently in the work of the advocates of the ‘connective turn’ (Hoskins, 2011): a paradigmatic shift in the treatment and comprehension of memory and its functions and dysfunctions. For Hoskins (2018b), ‘this turn drives an ontological shift in what memory is and what memory does, paradoxically both arresting and unmooring the past’ (p. 1). Under the impact of digital media, ‘collective memory’ has been replaced by the ‘memory of the multitude’, which marks a fundamental difference in relation to the qualities and the functioning of – what was previously defined as collective – memory:

The connective turn has transformed media-collective relations to shape a new multitude in two defining and related ways. The first is the shift from publics who had no means of replying to media in the broadcast era to today’s participation whereby a new mass constantly snap, post, record, edit, like, link, forward and chat in a digital ecology of media. The second—a direct consequence of the first—is that the memory of the multitude is all over the place, scattered yet simultaneous and searchable: connected, networked, archived (Hoskins, 2018a: 86).

In this account, the transformation from a passive to an active position of the users – what is termed ‘participation’, a defining feature of digital media – appears to provoke a radical destabilisation of memory. The amount of data, its hectic circulation, and the fragmentation that now characterises what used to be a coherent collective entity, produces an inherently volatile memory. Furthermore, collective memory is no longer built on shared understandings of the past but is dispersed in networks and interactions that promote a perpetual ‘becoming’. If ‘sites of memory’ was a central concept in the second phase of memory studies, for proponents of the ‘connective turn’ digital memory would instead be marked by a substantial lack of ‘sites of memory’ – placeless memories.

It is beyond the scope of this article to answer a question so central to the debate on digital memory, that is, whether the digital turn represents a real break with previous cultural frameworks, and, consequently, has a transformative impact on culture – and by extension, on memory. Nevertheless, it is necessary to highlight that the debate about the impact of the digital upon the shaping of collective memory is divided, roughly speaking, into two different positions: those who see a radical shift and those who approach the matter more in terms of continuities between online and offline memory practices or directly challenge the polarisation between two irreconcilable paradigms (see this issue’s editorial). As Bond et al. (2017) contend, ‘there are good reasons to resist this shift, if only because it threatens to erase a number of tensions and distinctions in a way that impoverishes our account of the mobility of memory’ (p. 15). In the same line, Assmann (2017) criticises the narrative of a radical shift as a ‘huge simplification’ and argues for a more complex conceptualisation of change, considering that ‘in a “postscarcity culture” of what José van Dijck calls “automated connectivity” (p. 13), we are perhaps no longer dealing with memory but with data, information, or knowledge - in other words, with concepts that lack the notion of boundedness, constraints, and contours and share the endless extensibility of media’ (p. 70).

The imaginary of placelessness is associated with at least two attributes of the digital ecology, which are closely related. First, the concept of flow, a key term in digital culture (Peters, 2016). The digital environment is a space defined by flows of information, where all the ‘objects’ are in permanent movement and, therefore, everywhere and nowhere. Underlying the association between fluidity and placelessness is the implicit opposition between place and mobility as two irreconcilable entities. If something is fluid – and even more so if that movement has the speed and opacity it acquires in the digital environment – that ‘something’ has no place.

The second attribute to which the idea of placelessness is linked is that of the changing characteristics of the archive and the database as the privileged cultural forms of the digital era. The notion of the traditional archive as static, selective, organised and restricted to a particular place, has been replaced by a dynamic, non-selective and multimedia online archive, whose logic is fluidity and ubiquity, and which is always ‘in becoming’. The fact that the memory of the multitude is ‘inherently archival’ results, again, in an imaginary of ‘placelessness’: as Hoskins (2018a) contends, ‘[i]t is through “shadow archives” that the individual and the social are inescapably entwined and also suspended, in a kind of extratemporal and extraspatial existence’ (p. 86–87) (emphasis added).

In this article, I would like to challenge the imaginary of a radical break by focusing on the issue of place. I argue that digital memories are not placeless memories, even if we accept that fluidity is inherent to digital media. This is because, first, mobility and place do not oppose each other, but are mutually constitutive. Second, and more importantly, digital memory practices actively engage in place-making, rather than succumbing to the logic of hyper-connectivity (Garde-Hansen et al., 2009) and all the attributes upon which it is predicated. I suggest that, in digital memories – encompassing digitally mediated memory practices and the ‘memory objects’ that result from them – the performation of place becomes central.

The notion of ‘performation’, coined by Bianchini (2012), seeks to conceptualise the participatory dimension of digital culture. Jan Baetens, building on Bianchini, suggests that performation is to be understood as the intersection of three other notions, each of them belonging to a different discipline: as automated information management (‘informatique’) and man-machine interaction; as performance in the artistic sense of the word, that is, a unique event or situation, which involves time, space, the performer’s body and the interaction with the audience; and as performativity, as defined by J. L. Austin, who highlights a specific capacity of speech to act, or consummate action, for ‘doing something’ instead of just describing it (Baetens et al., 2020: 126). Drawing loosely on this notion of performation, I apply it to think about the active construction of place in the digital sphere.

The first dimension of performation can be understood as a synonym for execution, accomplishment or actualisation. In terms of man-machine interaction, performing implies an action that follows established patterns and often connotes special skills. Place understood as performance does not pre-exist the mediation of digital technology, but is created and executed through it. When playing an online game – for instance Fortnite or Minecraft – we inhabit a place that is digitally designed and that we execute. As our (virtual) body moves through the different streets, corridors, cities or open spaces of the game – forests, rivers, deserts – we follow patterns that govern the haptic interaction with the remote control, which allow us to make certain movements depending on our expertise. To have the skills and follow the protocols is essential for experiencing the place – to move and to interact with others, as well as to explore and perform the territory.

The second dimension of ‘performation’ refers to performance, the artistic movement that originated in the ‘60s and ‘70s. A performance is not just the representation of a work passively contemplated or consumed by the audience but an action in which the body of the artist and the ‘spectator’ are actively interacting in the here and now. Place in the digital realm exhibits the characteristics of a scenario (Taylor, 2016: 134–142) which works through reactivation rather than duplication: ‘[a]s opposed to a copy, the scenario constitutes a once-againness’ (p. 139). When projects reproduce or mimic physical places – for instance, virtual visits to a museum, or the virtual reconstruction of former Clandestine Detention Centres of the Argentinian dictatorship, produced by /IEM/UNGS (Vecchioli, 2018) – they are never just an imitation or a reproduction of the physical place to which they refer. Rather, the physical place is staged and the interaction with the virtual place strongly underscores the bodily dimension of digital memory.

Lastly, Austin’s discussion of the performative use of language can be fruitfully applied to place in the digital era: we do things with places, as the work of Forensic Architecture (discussed below) convincingly shows. The digital does something with place: it enhances it, it alters it, it erases it, it contaminates it, it makes it more meaningful by adding layers of meaning allowed by the specific affordances of digital media.

In what follows, I first briefly review the dichotomies of ‘mobile/place’ and ‘cyber/hybrid space’ in order to show that, in the present digital phase, it is an oversimplification to speak of digital media and the Internet as placeless. Second, I illustrate the performation of place in two works of the research agency Forensic Architecture. Although the work of Forensic Architecture is a paradigmatic example of how technology can be actively engaged in the (re)construction of place at the centre of memory work, I argue that the performation of place is always at stake in digital memory work.

Place and mobility/cyber versus hybrid

Two dichotomies are central to challenging the imaginary that digital memories are placeless memories. The first of these dichotomies is the opposition between ‘place’ and ‘mobility’, a relationship that has been extensively addressed in recent literature in relation to the rise and consolidation of mobile technologies (De Souza e Silva and Sheller, 2015; Hjorth et al., 2012; Wilken and Goggin, 2012a). Cresswell (2009) points out that the relationship between mobility and place is ‘marked by disagreements between those who see mobility and process as antagonistic to place, and those who think of place as created by both internal and external mobilities and processes’ (p. 9). Examples of the former are works such as Place and Placelessness by Relph (1976), who introduced the concept of placelessness (Ek, 2006: 51) or the well-known theorisation of ‘non-place’ developed by Augé (2008).

On the opposite side, those who consider place as created by both internal and external mobilities and processes rely on a process-oriented view of place. From this perspective, place is not something given, which pre-exists the practices we carry out there, and it is not a stable point that dissolves in the dispersion of the flows, but the ‘figure’ that is designed and created through the practices we engage in – practices that are constituted by human and non-human agents. In this view, mobility is not a threat to place but, on the contrary, places are actively constituted by mobility – of people, commodities, thoughts and feelings. Doreen Massey and Nigel Thrift describe how, under the impact of the global transformations of the 1980s and 1990s, the understanding of place became relational:

[A]n imagination of space as constructed out of nets of interconnections began to take shape. [. . .] ‘place’ was now, in this view, understood as being a moment in a wider relational space; its very specificity was constructed out of relations with elsewhere [. . .]; the character of a place was always in construction (cited in Wilken and Goggin, 2012b: 10).

Relational space is thus an outcome of (social) processes, interactive practices and performances. The conception of a relational space is also intimately connected to the performative. Place is performed through mobility rather than being opposed to it. Unlike the traditional notion in which place is a located point in space, now place is the result of a performance of subjects who trace particular connections between several points. The affordances of digital media – especially if we consider mobile and locative media – work not by eroding our attachment to a place but, on the contrary, play a crucial role in constructing and reconstructing senses of place. As Özkul (2015) contends, ‘locational information use can increase one’s attachment to places’ (p. 112) rather than weaken it.

The second dichotomy that helps to scrutinise the imaginary of digital memories as placeless is the opposition between two different ways of conceiving digital space, that roughly reflects the evolution of the Internet since the 90s until the present era. For De Souza e Silva (2006) this evolution is expressed as the transition from ‘cyberspace’ to ‘hybrid space’.

‘Cyberspace’ corresponds to the early days of the digital, when the Internet was seen as an immaterial and disembodied space, that was, most importantly, dissociated from physical space. This static symbolic division between two spaces with opposite characteristics found its correlation in the static mode in which we accessed the Internet: in front of the screen of a fixed computer, we were immobile while we ‘entered’ into that other space – or a space ‘other’, in which we could also ‘be’ others.

With the rise of mobile technology, Web 2.0 and locative media – technologies that allow for real-time localisation – the idea that physical place was something irrelevant became increasingly untenable. The interfaces of mobile devices enable us to be connected to the Internet while we move across physical space. Mobile technologies literally allow us to ‘carry the digital space’ with us. ‘As a consequence,’ De Souza e Silva affirms (2006), ‘mobility becomes part of the process of connecting to the digital and exploring hybrid spaces’ (p. 268). De Souza e Silva terms ‘hybrid space’ the space we inhabit today as mediated by the ‘social interface’ of our mobile phone, a space that involves a permanent interaction between the physical space through which we move, and the ‘virtual space’ to which we are connected through the mobile device. As De Souza e Silva (2006) puts it, ‘[h]ybrid spaces are mobile spaces, created by the constant movement of users who carry portable devices continuously connected to the Internet and to other users. The possibility of an ‘always-on’ connection when one moves through a city transforms our experience of space by enfolding remote contexts inside the present context’ (p. 262). One strong example of how networked communities interact in hybrid spaces is location-based mobile games, in which players use city space as a game board. While moving through the city, users can, for instance, shoot or target other players depending on the relative position of each player in the city.

Unlike during the immature digital world, when, connected to the desktop computer, being online was a comparatively solitary experience, today we are connected in public spaces, surrounded by other users, interacting simultaneously with individuals who are physically contiguous and others who are virtually co-present. In this process, ‘space is becoming enfolded, so that there is no longer a homogeneous context for a given spatial area, but rather pockets of different contexts in it’ (De Souza e Silva, 2006: 269). In this sense, a new ‘hybrid space’ is performed through the intervention of technology, a place that is richer and more complex than the simple physical place we inhabit without connectivity. To exist, it requires the presence of an embodied subject – as Taylor (2016) contends, ‘embodiment, understood as the politics, awareness, and strategies of living in one’s body, can be distanced from the physical body’ (p. 138).

In summary, the notion of hybrid spaces leads to a redefinition of physical and digital spaces: both are transformed when locative media and mobile technology interact to form a ‘hybrid reality’. In what follows, I explore how FA creates hybrid places in which memory work is at the centre – both as the origin and the endpoint.

Forensic Architecture or the architecture of memory

Forensic Architecture is a research agency founded in 2010 by a group of architects, artists, filmmakers, investigative journalists, scientists and lawyers, based at Goldsmiths University and led by Weizman (2017). They act independently, or at the request of international jurists and NGOs, to investigate state and corporate violence. The agency produces evidence files that include building surveys, animations, video analyses and interactive cartographies, which are presented in forums ranging from international courts and citizen tribunals, to museum exhibitions (p. 9).

Forensic Architecture can be situated in what has been called the ‘forensic turn’ (Anstett and Dreyfus, 2015; Dziuban, 2017), which marked a shift from the ‘era of witness’ – in which knowledge about violence had relied on the epistemic as well as ethical testimony of survivors – to the era of forensic, where the truth is obtained by examining the material traces of violence. But instead of analysing bones, as the traditional forensic method does, Forensic Architecture turns to buildings as the principal objects of inquiry. Buildings – as Weizman (2017) points out – are not static objects that remain the same over time. On the contrary, they continually undergo dynamic transformations. Buildings have a sentient materiality that registers the changes in the environment, changes in humidity, levels of CO₂, heat or shifts in weight distribution – indicative of changes in the way a building is used (p. 51–53). In addition to buildings, larger built environments and territories also act as a surface on which social life is inscribed and registered, thus becoming ‘political sensors to be read’ (Weizman, 2017: 54). For example, plant records – crushed fields around a city’s agricultural outskirts – register the movement of military vehicles on them when they stop photosynthesising, a signal that is captured by remote sensors orbiting above and then beamed back to earth (Weizman, 2017, 58). Thus, the environment has a memory that can be read, although reading this memory is not straightforward. The processes that buildings undergo are chaotic and non-linear, and the evidence they deliver is therefore murky and indeterminate.

The complexity of reading and interpreting the traces recorded in the built environment is furthered by the fact that the built environments in which we live are not only constituted by physical objects, but, equally important, by a complex network of information. The urban environment is a dense media environment, ‘highly sentient in both material, analogue, and digital terms’ (Weizman, 2017: 57): cameras (satellites, security cameras and smartphones) take photographs and videos, air-quality sensors pick up increases in traffic, people use their camera phones to record what happens around them, from the most banal to crucial evidence of violence. When investigating a case, the work of Forensic Architecture consists in reading all these fragmented data, cross referencing them and pulling them together. In other words, the agency creates ‘evidence assemblages’ by establishing relations between actors, infrastructures, digital data and human testimony, in which place is always at the centre of the scene (Weizman, 2017: 58). According to Weizman, the ‘architecture’ in forensic architecture has distinct meanings, being at the same time the object of the investigation – buildings that function as primary evidence – a mode of research – locating bits of data and constructing models out of this to reconstruct the event – and, finally, a mode of presentation – the architectural models Weizman’s (2017) team composed together with victims of violent events, helping them to recall what had been obscured by trauma (p. 58–59). ‘Architecture in this context becomes a mnemonic device’ (Weizman, 2017: 58).

At this point, it is fair to ask: what memory work does Forensic Architecture engage in? Why can their creation and presentation of ‘evidence assemblages’ be considered as digital mnemonic objects? The status of the fragments or the raw data they collect is the point in question. Are these fragments – photos, short blurry videos that partially record a street that is under attack or people fleeing, a post on social media, to give just a few examples – ‘memories’ in their own right? Or, on the contrary, are they only the material out of which a story, an event, a significant link with the past, has yet to be constructed? The answer to these questions helps to shed light on some of the claims that inform the debate on digital memory. The fact that digital technologies archive an incredible amount of information by default – even, and especially, what was previously ephemeral – does not necessarily mean that this material has the status of ‘memory’. Their status as memory depends on how we engage with these objects. If while searching on the Internet we accidentally encounter a photograph of someone being abducted during the 1970s in the framework of the Argentinean dictatorship, it is undoubtedly a trace of the past. Nevertheless, if we cannot establish any significant link of this trace with the present it is not properly speaking a memory object. Memory is a process in which we, as individuals or collectively, establish a significant link with the past. In the absence of such a significant relationship, there is no memory but just raw data. If we agree that collective memory is, by definition, narrational (Neiger et al., 2011; Rigney, 2016) then the photos or videos we can accidentally find while navigating the web are traces of the past but not, strictly speaking, memory. Along these lines, we can interrogate, for instance, whether the collapse of past and present argued by some critics (Garde-Hansen et al., 2009:7) to be a defining feature of digital memory relies on a conflation between ‘traces of the past’ and ‘memory’. If past and present collapse, if we can no longer distinguish between them, as memory scholars contend when describing the features of the present era, to what extent are we still talking about memory?

Coming back to our initial question, in the work of Forensic Architecture it is only when the multifarious bits of data are meaningfully assembled into a narrative that we can speak properly of (digital) memory. What is distinctive in the work of Forensic Architecture is their use of digital technology as an instrument to build a story that would not have been possible without the technological affordances. Moreover, their reconstruction of the ‘event’ puts place at the centre. In other words, they create digital memories whose main trait is to be composed by and on place.

Hannibal in Rafah

This is clear, for example, in the investigation ‘Hannibal in Rafah’. Friday August 1, 2014 – later known as ‘Black Friday’ – was the deadliest day of the 2014 Israel-Gaza conflict. That day was not only exceptional because of the high number of victims, but also because of the ‘logic of fratricide’ that determined the conflict. On ‘Black Friday’, Hamas soldiers captured an Israeli soldier and took him to one of the tunnels in the hope of exchanging him later for their prisoners. That capture prompted the Israeli military to launch a secret order – the Hannibal Directive – which consisted of attacking the entire area with massive fire to ‘eliminate’ the Israeli soldier. The Hannibal Directive ‘solved’ the problem that arose from the capture of one of their own soldiers, avoiding potential problems during negotiations by choosing instead to kill him (Weizman, 2017: 165–80). The investigation conducted by Forensic Architecture sets out to demonstrate that what happened on ‘Black Friday’ was the result of a ‘hunt’, and an attack carried out under the Hannibal Directive, which the Israeli army had previously denied. Moreover, regarding the controversy of whether the attack had occurred during a time of ceasefire, they could demonstrate that the violation ‘took place in space: an Israeli unit proceeded into an area controlled by Hamas during a time when they claimed there was a ceasefire’ (Weizman, 2017: 174).

Unlike the attacks that occurred in 2008–2009, the 2014 attacks were abundantly documented by the material that Gaza inhabitants themselves produced and uploaded to social networks. While they were attacked, they photographed or filmed the attacks, putting their lives at risk. However, although invaluable, these scattered fragments only present partial and incomplete versions of what happened (Weizman, 2017: 166).

Forensic Architecture worked from a corpus of about 7000 photographs and video clips obtained from websites and social networks, which showed civilians under attack, clouds of smoke, buildings in ruins and other remnants of violence. Since the images lacked the metadata that could have helped to establish the time and location of these sources, the agency had to find another way to relocate the images, re-assign the coordinates and reconstruct the event in its entirety. This ‘other way’ meant choosing the clouds produced by bomb explosions as the object of analysis. The cloud that is formed from a bomb explosion lasts approximately ten minutes. During this time, it goes through several different phases. The cloud is formed from the destruction of the objects on which it impacts, and these remnants in turn become the ‘materiality’ of the cloud, a dust composed of sand and soil from under the foundations, the concrete of their structure, plastic, wood, glass and remains of human bodies. The continual transformation of a bomb cloud and its unique form enable them to function as a form of metadata – indicators of an image’s time/space coordinates with which they could synchronise and sequence many of the images of the day, thereby building a narrative (Weizman, 2017: 191–96). The changing – gaseous – state of the bomb cloud transforms it into a kind of ‘negative’ of the solidity and stability of buildings, and yet, as Weizman (2017) points out, ‘it is architecture nevertheless, a temporary, gaseous architecture with a life span of seven to ten minutes’; if the starting point of Forensic Architecture is the fact that all architectural works always change, clouds represent ‘the extreme condition of architecture, its cycle of emergence and decay played in fast-forward mode’ (p. 193).

Through a classification process according to stage – explosions, ascending columns, mushrooms and dissipations – and a comparison of the intersections between the different images of each cloud, the team was able to determine where each bomb had fallen and the time that passed between one bomb and another. Subsequently, the ‘cloud atlas’ was connected to the surface with the testimonies of those who had suffered the bombings, in order to draw the full map of the attack on August 1, in ‘a forensic assemblage was at once the product of media, memory, and material reality’ (Weizman, 2017: 196).

Drawing on the concept of performation, ‘Hannibal in Rafah’ can be read as performing place in three ways. The first relates to the specific affordances provided by digital media: the user-generated content – particularly photographs and videos – uploaded to social media platforms, which represented the raw material out of which the image space was constructed; but also the affordances provided by the digital tools used by the team in order to classify, compare and make assemblages of the data in an image space, an optical device to establish and view images and the relations between them. Second, ‘Hannibal in Rafah’ can be regarded as a performance, in the sense that the clouds are not static objects that are located in a given space – the sky – but, on the contrary, they are in constant transformation, each of them constituting a unique event – the bomb exploding. In an extreme form, bomb clouds reflect the conclusion that place is not an object but an event (Ek, 2006). Moreover, the artistic notion of performance resonates in the way Forensic Architecture describes its own work, in which the production of evidence is as important as its presentation (Bois et al., 2016: 131–35). Finally, the reconstruction of what happened in Rafah on ‘Black Friday’ is not just a description of events. The digital image space does something: blame the Israeli army for the attack. Although only a judge at a trial has the power to declare someone guilty – as a performative discourse – the evidence constructed here goes in the same direction, created with the specific purpose of producing a political effect.

Regarding the dichotomy of place and mobility, ‘Hannibal in Rafah’ inverts the allegedly mobile character of the digital versus the static physical place. Bomb clouds are in constant transformation, in continuous movement, whereas it is the digital model that fixes the movement, locating each cloud at a point and establishing the relations between these points, producing a bomb-cloud atlas. By mapping the clouds, the work spatialises the violence, turning the fragmented and dispersed data into a meaningful event. Importantly, the fact that digital media are ‘inherently archival’ acquires a different meaning in the work of Forensic Architecture: even though the data are there, automatically stored by media, they do not of themselves constitute ‘memory’, but require a process to render them meaningful. It is the process of collecting and mapping the dispersed fragments of data which transforms the archive into memory.

Locating the memory of victims through digital modelling

In ‘Saydnaya: Inside A Syrian Torture Prison’, an investigation carried out by Forensic Architecture in 2016, digital memory engages with place in a different manner. The investigation involved building a digital model of the detention centre in which former detainees from Syria had been held captive.

In this clandestine Syrian prison, the prisoners’ experience of the space was limited by the conditions of their captivity: they were blindfolded, the cells were dark and prisoners were forbidden to emit any sound. Therefore, they were only able to perceive the place by perceiving changes in temperature, humidity, vibrations or sounds. The relation between the architectural model and memory in this case has two dimensions: on the one hand, the model is a ‘product of memory’, following the description provided by witnesses; on the other hand, the process of building the model helped survivors to remember things that had been obscured by the trauma they had experienced. The possibility of navigating through the virtual prison, adding or changing objects, measuring and intersecting their – scant – visual memories with the acoustic ones and with the virtual space created by the Forensic Architecture team, led not only to the reconstruction of the place as potential evidence in a trial, but also to the reconstruction of the victims’ experience in prison (Barenblit et al., 2017: 60–73).

The digital model thus allows for visualising a place that until now only existed ‘virtually’ in the minds of witnesses. By cross-referencing data it was possible to specify certain details. For example, by analysing witnesses’ recollection of the echo they could hear, the distance between different parts of the prison could be reconstructed. Conversely, parallel to the objective of reconstructing the physical space as accurately as possible, Weizman insisted that errors, indeterminacies, contradictions and blank spaces were also recorded in the model because these signs of trauma, though they did not reflect the architecture of the prison, revealed the experience of survivors in it. Weizman (2017) gives the example of a significant ‘memory object’: a circular corridor that one of the witnesses describes, but which cannot have existed as such. The perception that the corridor was circular and not straight, as the other testimonies indicate, can be explained by the torture to which the prisoner was subjected at the time when he experienced the corridor (p. 93). There is a tension in the practice of locating the memory of victims through digital modelling, a difficult negotiation between a positivistic synthesis and respect for the subjective experiences of the witnesses. On one hand, the model aims to ‘correct’ mistakes and contradictions in the testimonies in order to produce an accurate description of the prison – one that could eventually be used as evidence in a trial. Weizman responds to the concern raised by Hal Foster – ‘Could the imperative to resolve contradictions into a story be a problem in its own way’? – with a categorical response: ‘whenever, from that great mess of contradictions and unknowables, we are able to put together, with great efforts, a faint fragment of a narrative—not a grand narrative, but something fragile, a construction that is cognizant of the very problem of truth-telling—and say this is what happened here, the last thing we need is a poststructuralist to kick it in the name of some kind of relativism’ (Bois et al., 2016: 129–130). On the other hand, the model is built to incorporate the inaccuracies of remembrance, because contradictions and mistakes are a record of the very effect of violence, ‘the presence of trauma and thus the ultimate truth of the event’ (Bois et al., 2016: 130). In registering the contradictions and mistakes, this digital memory aims to recompose not only the physical place – the clandestine Syrian prison – but also the sense of place victims felt during their incarceration.

As in ‘Hannibal in Rafah’, the investigation builds a digital place that ‘replaces’ absent, non-existent, or partially perceived physical places: clouds which have disappeared, and a prison that is inaccessible because the Syrian government prohibits entrance and is remembered by those incarcerated there only as sensory fragments.

The digital model of Saydnaya performs place in three ways. First, by allowing the witnesses to navigate through the model, the digital enhances their memory, enabling them to remember things that were obscured by trauma. The digital permits them to recuperate traces that had been lost, following the ars memoriae – a series of techniques created to improve or ‘train’ memory – that goes back to the well-known story of the Greek poet Simonides, as related by Cicero in his De oratore, who was able to identify the corpses of those killed by a building collapse because he remembered the places at which they had been sitting, and thus realised that orderly arrangement was essential for good memory. The intrinsic relationship between memory and locus continued to be reflected in the multiple spatial images that can be traced throughout the history of metaphors of memory – from the wonderful palaces, treasure houses and caverns of Augustine, to the theatre of Fludd, including the romantic labyrinth of Carus (Draaisma, 2000). Adding the digital as a technique continues this tradition, and, at the same time, improves the ars memoriae. As Weizman describes it: ‘[t]his is a kind of ‘art of memory’ for the digital age’ (Bois et al., 2016: 129).

Second, this memory comes into being through a performance that includes the four basic elements of its definition: time, space, the performer’s body or presence in a medium, and a relationship between performer and the audience. The witnesses’ memories are embodied – they remember bodily experiences which are intensified by the limited conditions of perception imposed in the detention centre, and which intensify even more while they navigate through the virtual model. The performance of the witness is conducted in front of a particular audience, namely the Forensic Architecture team, and the interaction between both leads to the construction of memory. Third, ‘Saydnaya: Inside A Syrian Torture Prison’ is performative because it makes the prison appear in the public sphere. In addition to the assemblage of data and the construction of models and interactive platforms, the work of Forensic Architecture has a fundamental performatic dimension that involves presenting their work in particular forums, such as museums, courts and trials. Displaying and ‘staging’ the work is as central in the production of truth as building the evidence. In this sense, it is interesting to note how the search for a ‘positivistic’ truth, supported by scientific evidence, goes hand in hand with an aesthetic truth, in which objects are made to speak: ‘[f]orensic is the mode by which the present theatre of horrors is performed by objects in front of a public’ (Weizman, 2017: 68).

In materialising the witnesses’ memory in a spatial model, the work turns these individual memories into collective memories. It socialises them, rendering them part of a public memory of human rights violations that, until now, because of their clandestine character, had been obfuscated from the public sphere. Put differently, the shift from the private memories of the witnesses to the public digital memory of the platform performs the shift from individual to collective memory.

Conclusion

The imaginary of digital memories as placeless relies on features assigned to the digital space which are either a relic of the pre-mobile Internet, or an exaggeration of the changes undergone by memory under the impact of digital media. Today, in a time of locative media, when our devices permanently relate our access to the Internet to our physical location, and with the continuous hybridisation of digital and physical places through mobile media, it is impossible to detach the digital experience from place. This is valid for every digital practice, including memory practices. The claim that the digital has radically altered the production, circulation and reception of memory, to the point of approaching the end of collective memory, or memory as we have known it until now, emphasises the loss of all traits that previously defined remembrance – loss of a clear separation between past and present, loss of the meaning of the ‘collective’, and also loss of place. Yet if we depart from this apocalyptic discourse and examine the continuities between the old and the new, as well as the affordances of new technologies for place-making and territorialising memories rather than delocating them, new possibilities emerge. In order to capture these possibilities, I have proposed drawing on the concept of ‘performation’ for describing the different dimensions involved in the creation, execution and usage of place in the shaping of digital memories.

In this article I have demonstrated how digital memory work performs place. By analysing two works of Forensic Architecture, I have outlined the different dimensions contained in the term ‘performation’: first, how digital tools serve to construct spatial assemblages, both material and informational; second, how place, as the model of the bomb-cloud illustrates, turns into an event, a performance that is captured and staged as a result of this memory work; and third, how the digital image space does something: blame the authors of the crime. Importantly, the fact that digital media are ‘inherently archival’ acquires a different meaning in the work of Forensic Architecture: it is the process of collecting and mapping the dispersed fragments of data that transforms the archive into memory. The digital affords new configurations – or new ‘performations’ – of place, in manners that enhance and improve our attachment to places rather than debilitating them. Not only does digital technology allow us to construct more emplaced memories, but it also opens up new venues for exploring the way in which space and place continue to play a central role in how we understand, shape and relate to the past.