Heterodox mediations. Notes on Walid Raad’s The Atlas Group

ABSTRACT The Atlas Group appeals to philosophical thinking in multiple ways—both through its aesthetic figuration and its conceptual references. Presented as a foundation dedicated to the research and the compilation of documents on Lebanese contemporary history and organized in the form of an invented archive, this artistic project deliberately coalesces real and fictitious elements and confronts, subversively, Western views on the socio-political reality of the Middle East with implicit knowledge and experiences from the region. The singular constitution of this curious archive subtly undermines the rational classification to which it alludes, thereby frustrating unilateral appropriations. Moreover, the question of the mediation of subjective experience and factual reality is consistently raised through the hysterical documents. This article aims to deploy how Walid Raadâ€™s project subtly criticises objectivist hegemonic claims by confronting divergent, often incommensurate approaches to the conflictual reality in Lebanon and its appropriation by media, historiography and politics.

Attending a lecture-performance from the artistic project The Atlas Group-for example, the one entitled The Loudest Muttering Is Over: Case Studies from The Atlas Group Archive-is a peculiar experience. The setting is as simple as it is effective: Walid Raad, the inventor and only member of the "group", sits behind an ordinary table equipped with a computer, a lamp and a glass of water, in front of a big screen. With a calm voice, a slight Arab accent 1 and his remarkable rhetorical abilities, he introduces The Atlas Group as a "foundation dedicated to the research and the compilation of documents on Lebanese contemporary history", explains its internal structure, reveals its organization to be an archive and then describes the content of the folders, while a Powerpoint presentation displays organization charts, images from specific files and other related material. His compelling speech reproduces to perfection the elaborate discourses of research structures, notably those concerned with the elucidation of history. Every detail of this accurately adjusted stage act, including minor technical problems occurring from time to time, seems to indicate that we are confronted with the presentation of an actually existing organization by one of its members. Adopting this established format associated with specific standards and scientific validity charges the whole art project with an aura of seriousness and legitimacy.
And indeed, it proves to be very difficult not to believe in the institutional character of The Atlas Group, even if Raad clarifies from the beginning that we are facing an artistic project, a fictional aesthetic construction rather than the presentation of a "real" research structure. "Raad's prefacing words-that the Atlas Group is an 'imaginary foundation' whose task is that of 'producing documents'-subtly but definitely become a lost file in our short-term memory's archive", writes André Lepecki (2006) in this regard. And the artist himself declares: "I also always mention in exhibitions and lectures that the Atlas Group documents are ones that I produced and that I attribute to various imaginary individuals. But even this direct statement fails, in many instances, to make evident for readers or an audience the imaginary nature of the Atlas Group and its documents" (interview with Gilbert 2002). This is all the more surprising as the stories he tells the public appear to be pretty hilarious: historians betting on the delay of the snapshot of the winning horse on the Sunday's horserace during the war, journalists competing to be the first to find the engine (the only preserved remnant) after a car bomb, large photographic prints in different shades of blue emerging from the Mediterranean that mysteriously ensconce small black and white photographs of people who disappeared in the sea-the universe evoked through The Atlas Group documents is undoubtedly quite eccentric.
But also seemingly minor details of Raad's description are worth scrutinizing, for example the fact that he locates his archive in two different cities: New York and Beirut. This is puzzling, not only because it immediately raises the question of the internal distribution of the documents-a genuine material record could not be stored in two venues at the same time-but also because the two cities are paradigmatically opposed, and not only in view of their geopolitical situation and the different ways in which they have been involved in the history at stake. To be sure, in neither of these places does the archive of The Atlas Group exist as a physical institution, in neither of them are the "original" documents stored -they never actually appear as such. Nonetheless, the gesture of identifying two venues as localisation is far from being anodyne. For the mere existence of an archive takes on a totally different meaning in each. In the USA and in Europe, the so-called "Western" countries, national and other archives have been established for a long time, while their political implications remain implicit. However, this does neither mean that they are absent, nor that they are not critically assessed. As Jacques Derrida put it in a conference given at INA in 2002: "There is no archive without a power of capitalization, of monopoly or quasi-monopoly, a power to collect statutory traces recognized as such. In other words, there is no archive without political power" (Derrida et al. 2013, 113-114, my translation). But even if philosophers, historians, archivists and artists have, in the last decades, widely problematised the monopolistic structure, the inherent political significance and the constructedness of the components of archives (see, e.g., Derrida 1996;Thomas, Fowler, and Johnson 2017;Blouin, Francis, and Rosenberg 2011;Burton 2006;Enwezor 2008;Spieker 2008;Merewether 2006), and despite the controversial status of the latters' hegemonic claim for truth, especially with regards to their partiality (as the critical debates on colonial archives show, see, e.g., Hamilton et al. 2002;Istok 2016;Downey 2015), they nonetheless continue to be acknowledged to provide a regularised, albeit incomplete and improvable, access for scientific research. This is related to the fact that they conform to certain conventions, e.g. they apply specific, traceable criteria for the records to be included in the collection and the authentication of the documents, which undergo specific procedures of certification and rational indexation. While those standards for research and archives are continuously criticised, reworked and readapted, their very existence is commonly accepted as a fairly adequate basis for scientific knowledge-production. Hence, the New York basis of The Atlas Group evokes both a distant location allegedly providing the necessary distance for "neutral" research, and a gesture of authority appropriating material concerning an allegedly foreign history.
By contrast, in post-war Lebanon, the problematic political implications of the archive come fully to the fore. The problem reveals itself to be even more complex than that of insufficiency, incompleteness or bias of an existing archive (which could then, eventually, be criticized)-they concern the very possibility of such an institution whose principles are, according to Derrida (1996, 1-3), not only the commencementthe beginning of a knowledge to acquire starting from that which the institution holds ready, but also the commandement, the authority of those who decide about the structure, the elements to include and the rules to follow. Indeed, the complex political conflicts, the heterogeneous constitution of society and the ongoing instability prevented the National archive from being operational in the aftermath of the 1975-1990 civil wars. The amnesty law which was promulgated shortly after the end of the wars blocks every effort to rigorously investigate the events and precludes any attempt to legally pursue the actors responsible for the war, some of which are holding important political, economical and cultural positions even today. Until now, no version of Lebanon's history has been adopted officially-the ongoing polemics regarding a standardized textbook to work with in schools are therefore symptomatic (see, e.g., Salloukh et al. 2006, 46-48). In this context, the idea of an ostensibly impartial archival institution, an independent structure concerned with the very same conflictual situation, but resisting to instrumentalization for political purposes, is far from obvious, 2 as it intervenes in the very centre of the ongoing "cold civil war" (Bilal Khbeiz, quoted by Wright 2002, 14). This is the point of departure of Walid Raad's artistic project. On the one hand, he reacts to an ongoing conflictual political situation, the absence of historical clarification, and a highly aggressive social climate. On the other hand, he delicately problematizes the multiple mediations of this situation through the local and foreign media and institutions, as well as their impact on diverse discourses and representations in different contexts. Instead of upholding the model of an allegedly impartial institution based on supposedly neutral categories, or proposing a counter-archive disposing of alternative data with the aim to criticise a predominant narrative, The Atlas Group mimics the form of an archive so as to establish a seemingly stable framework associated with a certain seriousness, but immediately defies this appearance by contaminating it with the politically overloaded reality of a particular country in the aftermath of a civil war. By appealing concomitantly to both regulated archive structures which bear the danger of invoking a detached, universalist perspective (as well as its imperialist or exotic variants), and a reality in which the very idea of a unified history is superseded by a multitude of particularistic or partisan versions, Walid Raad produces a tension in which they problematise each other mutually.
To be sure, the aim is not to mislead the spectator. As we have seen, The Atlas Group does not conceal its fundamental differences with other archives. Quite the contrary, it undermines their conceptual, structural and methodological foundation explicitly. Hence, Walid Raad openly mentions that most of the characters appearing in the framework of The Atlas Group project are purposely invented, and that the archive documents, far from being traces left from the past, are in fact artefacts specifically produced for the purposes of the project, on the basis of factual events and those "which could have happened". Moreover, he explains, albeit not during the lecture-performance, that the ostensibly stable identity of The Atlas Group constantly transforms and adapts itself to the public of the moment: "In different places and at different times," claims Walid Raad in an interview with Alain Gilbert (2002), "I have called The Atlas Group an imaginary foundation, a foundation I established in 1976 and a foundation established in 1976 by Maha Traboulsi. In Lebanon in 1999, I stated, 'The Atlas Group is a non-profit foundation established in Beirut in 1967.' In New York in 2000and in Beirut in 2002The Atlas Group is an imaginary foundation that I established in 1999.' I say different things at different times and in different places according to personal, historical, cultural, and political considerations with regard to the geographical location and my personal and professional relation with the audience and how much they know about the political, economic, and cultural histories of Lebanon, the wars in Lebanon, the Middle East, and contemporary art." Just as the status of the project, its character and geographical and temporal coordinates change from performance to performance depending on Raad's ideas about those who apprehend it. The spectator (as imagined by the artist) is treated as another variable in the project: always considered as part of a specific public, with specific prejudices, background knowledge and cultural biases, her reaction is anticipated by the artist and integrated as yet another potential declination of the project. Rather than being thought of as a distant, uninvolved observer, or as an active, conscious participant, she takes on a decisive part in the very concretion of The Atlas Group. Hence, instead of proposing a steady object to contemplate or analyse, Raad conceives The Atlas Group as a work in progress which alters its appearance according to the multiple mediations it is subject to: the formal device, the subjectivity that apprehends it, the context through which it appears. Thus, the division between a knowing subject and a contemplated object is suspended on many levels: not only is the constitution of the archive itself contaminated by the figures it is supposed to mediate, but also the reciprocal relation between the artwork and the audience is constitutive of the project as such.
What is at stake is a critical, aesthetic intervention into an established formal device, more precisely a challenge of the idea of disposing a constant, detached form able to comprehend any diversified content. This separation, which constitutes to a certain extent the legitimation of the archive's claim for neutrality, is suspended on many levels in Raad's art project. By introducing foreign subversive elements into the very core of the institution (i.e. by intermingling fictional and factual elements), and by including the potential perceptions its structure and contents generate into the aesthetic construction itself, Walid Raad deliberately produces palpable tensions between different layers of reality (one of which being the archive-structure itself) and blurs the frontiers between subjective experience and objective data. Far from naively imitating archival institutions and their formal devices, The Atlas Group performs a critique of its hegemonic aspirations by multiplying the vectors of meaning. Its subversive character emanates from the particular formation of heterogeneous sensuous and conceptual materials, in which the slightest detail potentially challenges the appearance of the whole. The aesthetic form is here not only itself "sedimented content", as Adorno [1997] 2013, 6) puts it, but also a means to address the very nexus of formal and topical meaning in its perceptual and political dimensions. Hence, by both appealing to and frustrating the regular operation of forms, concepts and signs, The Atlas Group takes on a critical function. How does the very constitution of this artwork intrinsically challenge the notions it deals with and subtly criticise established mediations of form and content? How does it reflect, figuratively, the dissensual encounter between different geopolitical and cultural spaces? How does it undermine both hegemonic objectivist claims and the instrumentalization of data for political means?
Let's take a closer look at The Atlas Group's constitution as an archive and research structure. As mentioned above, it appears to be quite conventional at first sight. The records are divided into three different types, each of which is further subdivided according to the provenance of the documents. Yet, a thorough inspection quickly reveals the internal inconsistencies of this classification. The three main categories are the following: A (authored), for those records to which an author is attributed, FD (found documents), for those described as anonymous, and AGP (Atlas Group Productions), for those explicitly produced by The Atlas Group. However, it is also said in the same description that the artist actually produced all the documents himself. We are thus confronted with two dissimilar layers on the same site, one of which subsuming all the documents under a generalized fiction, the other differentiating the nature of the documents according to their alleged origin. Furthermore, the files sometimes change their position inside this classification, and the same documents appear under different names, while keeping the same description-all operations in the purpose of nullifying the very logic of categorization supposed to assign a specific identity to each object. This intricate classification resembles in some respects the Certain Chinese Encyclopedia in Jorge Luis Borges' short story The Analytical Language of John Wilkins, which Michel Foucault used in order to illustrate the collision of dissimilar epistemes. 3 The passage in question features an enumeration of different kinds of animals-real and imaginary, particular specimens or simply those that are "included in the present classification"-which reveals itself to be unthinkable under the proposed classification system despite the latter's descriptive accuracy. The problem lies not so much in the absurdity of the categories themselves-each one on its own is, in fact, conceivable-but in the incompatibility of their community with the idea of a rational classification. Even though each part of the system seems to be coherent in itself, their togetherness eludes rational comprehension. However, Borges' fictional story is exactly that: the site on which they coexist in an apparently coherent way. In Foucault's terminology, such sites are heterotopic, 4 for they are "capable of juxtaposing in a single real place several spaces, several sites that are in themselves incompatible" (1986,25). Heterotopias "make it impossible to name this and that, […] they shatter or tangle common names, […] they destroy "syntax" in advance and not only the syntax with which we construct sentences but also that less apparent syntax which causes words and things (next to and also opposite one another) to 'hold together'" (2002, XIX). By concomitantly appealing to and frustrating the subsumption under specific logical categories, heterotopias dilute the hegemony of an epistemic order, thus pointing to the fact that the very basis of our knowledge is far from naturally given, and that its apparently natural character always hides in its folds some or another element of power and domination.
It is certainly not anodyne that in Borges' story, the heterotopic system is located in a truly existing, but-seen from Borges' or Foucault's standpointvery distant and to a large extent impenetrable location: China, a place that has been saturated with projections, common preconceptions and attempts to appropriate this exotic Other. Borges' Certain Chinese Encyclopaedia deliberately thwarts such intentions, by appearing both ostensibly consistent, yet inconceivable to "our" understanding. It makes us sense that the "mode of being of order" on which our very epistemic position is grounded prevents us from grasping its logical constitution, because we are unable to abstract from the fundamental principles of our own thought. As Foucault puts it: "In every culture, between the use of what one might call the ordering codes and reflections upon order itself, there is the pure experience of order and of its modes of being" (Foucault 1970, XXIII). Hence, the very presence of heterotopias disrupts, perceptually, the apparent obviousness of our way of conceiving and the act of identifying as such, and reveals that it is historically developed and socially produced, and interrelated with hegemonic interests. Heterotopias not only disturb the grounding of our knowledge and the way it is logically organized, but they also appeal to a critical assessment of our own position. Thus, Foucault evokes the image of the mirror in this context: heterotopias not only lead to a projection in a space where we are not present, but also reflect our image, from the outside, onto the position we actually occupy (1986,24).
In the case of The Atlas Group, the shock of the encounter with an utterly different system of knowledge is certainly more minor than Michel Foucault's profound unease with Borges' Certain Chinese Encyclopaedia. The discordances are not as obvious. They are subtly insinuated: rather than profoundly unhinging the very ground of our knowledge, they worm themselves into our perception, destabilize delicately our position and produce a diffuse confusion. Lebanon is perhaps not a place as exotic and strange as China for "Western" observers, not only because of the multiple economic, cultural and diplomatic relations, but also because Lebanon has been affected by the French Mandate  and other international interventions and occupations for a long time, and partly integrated not only different languages such as English and French, but also their respective educational systems, cultural references and specific values. However, those appropriations, as dissimilar "ways of operating", as Michel de Certeau (1988, XI) names "the innumerable practices by means of which users reappropriate the space organized by techniques of sociocultural production" (XIV) and which are not just pure adoptions, but adaptations that transform the forms and contents through a process of "secondary production" (XIII). Thus, just like in Borges' story, the encounter with a slightly deviating regime of concepts and signs generates disorientation and make us sense a discordance. By disturbing our cognitive and perceptual habits, The Atlas Group appeals to a dialectical back and forth between the system supposed to subsume groups of things under an abstract category and their very constitution as particular entities. It does so through a challenge of a recognizable form that becomes questionable as such through a subtle estrangement, thereby revealing that, far from being a neutral platform, such forms are themselves conveyers of meaning.
Throughout the documents, this unsystematic logic of uncertainty operates on different levels. Sometimes, it is subtly embedded in a more complex narrative. I Only Wish that I Could Weep-a video file attributed to a certain Operator #17 and classified sometimes in type A, sometimes in type FD-is constructed around the story of a Lebanese Army intelligence officer. In the late 1990s, his task was to monitor the Corniche in West Beirut for suspect individuals. However, instead of focussing on suspicious persons, he filmed the sun setting over the sea -a dissident act that he explains by the fact that during the civil war, he had been living in East Beirut (mostly Christian contrary to the predominantly Muslim West Beirut), a sector of the city which did not have leisurely access to the Mediterranean Sea at the time. On a first level, this document expresses the persisting latent atmosphere of danger prevailing in Lebanon's capital in the aftermath of the wars and the ensuing generalized climate of suspicion, as well as the sense of lassitude wrought by the situation, and illustrated by the romantic attempt to circumvent the state-mandated directives and simply enjoy the sunset.
But there is more to find in the file than this rather evident dimension. A closer look at the specification of those who are considered dubious individuals discloses another layer. The small introduction of the video enumerates as suspects, without further differentiation, political experts, spies, double agents, fortune-tellers and phrenologists. On the face of it, this listing appears to be simply absurd. A mission targeting fortune-tellers, usually associated with superstition and shady esoteric beliefs and operating in a personal, apolitical sphere seems to be quite remote from state preoccupations. Targeting phrenologists seems not only strange, but also anachronistic, as their popularity, influence and scientific claim date back to the 19 th century. As for political experts, they are at best as efficient as football experts, analysing events after they occur or trying to evaluate current events in order to predict a possible outcome, but they are usually not considered a threat to the state as such. The only obvious targets of Operator #17's mission seem to be the spies and double agents as they constitute by definition a threat to the state.
But what is to be done with the putting together of all these categories by subsuming them all under the same idea of suspicious persons? There is surely a performative jolt whereby the person confronted with this file feels completely disoriented, as if the usual political grid is suddenly suspended. Strange and vague associations surface in her mind without ever solidifying into a system. For example, resorting to fortune-tellers can be interpreted as a far-fetched allusion to the fact that due to the complexity and lack of transparency of political events in Lebanon, the subject experiences them in a fatalistic manner, as if the sphere of politics was regulated by a kind of esoteric force, to such an extent that fortunetellers appear to be at least as efficient as political analysts. The same occurs as to the recourse to phrenologists, which relate ideas and behaviours to physical properties, i.e. the physical properties of this or that ethnical group in Lebanon. Here too, there is a far-fetched allusion to the reality of ID-killings during the war, whereby people would be assassinated depending on the sectarian group to which they belong, or to ongoing racist currents persisting in the country. But any attempt to squeeze out of these vague associations any kind of system of positive knowledge is doomed to fail. Rather than positively elucidating the reality at stake, the apparatus itself constitutes a negative critique. It puts the viewer in the uncertainty of her own standpoint. The apparently purely formal description of the document covers a multitude of potential signs and insinuations. And it is exactly here that The Atlas Group unfolds its performative force: rather than revealing something topical about Lebanese reality, the file ropes its audience into a diffuse atmosphere of uncertainty, vague ideas and intuitions, in which every sign is considered a possible clue, thereby involving the spectator by transforming her into yet another suspicious person. Instead of understanding the situation by following the supposed injunctions and finally sublating the paradox, we entangle ourselves in the very speculations that are ascribed to Lebanese post-war-society by the file. Incidentally, the video itself seems to adhere to this tense climate. Far from identifying anyone on the rushes, the persons we see appear ghostly, similar to one another in their movements, the only contrasting element being the very slow, natural movement of the setting sun.
Another example of the use of heterotopic principles can be found in an intricate document attributed to a certain Dr. Fadl Fakhouri, presented as the "most renowned historian" of his time. His file consists of several dissimilar documents. One of them, his notebook entitled Missing Lebanese Wars, is introduced as follows: "It is little known that the major historians of the Lebanese wars were avid gamblers. It is said that they met every Sunday at the racetrack-Marxists [Maronites in the French description] and Islamists bet on races one to seven, Maronite nationalists and socialists on races 8 through 15. Race after race, the historians stood behind the track photographer whose job was to image the winning horse as it crossed the finish line, to record the photo-finish. It is also said that they convinced (some say bribed) the photographer to snap only one picture as the winning horse arrived. Each historian wagered on precisely when-how many fractions of a second before or after the horse crossed the finish line-the photographer would expose his frame" (The Atlas Group and Walid Raad 2004). Once again, one can sense the paradoxical interweaving of dissimilar layers even before looking at the document itself.
What is striking in the first place is that the "most renowned historian" is not presenting the results of his research, but telling the story of one of his regular leisure activities during his free time along with other representatives of his profession in the midst of a civil war. Instead of elucidating the past, they gamble one against the other. They not only turn a blind eye to the historico-political situation, but also to the horse race as such, as they are only interested in the gap between the actual event of the horse's arrival on the finish line and its photographic proof to be discovered the next day in the newspaper (which raises the question why they were even present at the hippodrome). The visual aspect of the notebook, however, seems to confirm the alleged importance of the operation. It consists of several green single pages of a common block-note. Loosely arranged on each is a cut-out newspaper photograph of the official picture of the winning horse, hand-written annotations in English and Arabic and some peculiar calculations. It seems that we are confronted with complex notes for an ongoing research, which could only be accurately read by the author himself. But what is perhaps more disturbing in his account is that Dr. Fakhouri seems to be explicitly concerned with the sphere of rumours rather than with historical facts. 5 Once again, we are immersed in an atmosphere where hearsay, suspicion and rumour make their way into factual reality.
On a more analytical level, it seems to be rather odd to divide historians-allegedly sharing first and foremost the same professional ethos of impartialityaccording to their religious, communitarian or political affiliation. However, in the context of war and post-war Lebanon, this division is not so preposterous. The civic identity in Lebanon is, until today, largely determined by sectarian affiliation. For instance, the distribution of political positions is regulated according to such affiliations, and until today, there is no possibility of civil marriage. "Since independence, sectarianism was institutionalized in the form of multiple corporate consociational power-sharing arrangements, namely the 1943 National Pact and the 1989 Ta'if Accord, in the context of a centralized but institutionally weak state. Control of state institutions and revenues by an overlapping alliance of sectarian/political and economic elite consecrates a sectarian institutional set-up and lubricates sophisticated clientelist networks that co-opt large segments of the population, thus ensuring that the Lebanese remain unequal sectarian subjects compartmentalized in self-managed communities, rather than citizens with inalienable rights" (Salloukh et al. 2006, 2). This structural communitarian division is also reflected in Lebanese historiography: divided into divergent sectarian versions of events and their impacts, none acquires the status of an official version. Historians in Lebanon are thus generally associated, even against their own conviction, with their community (see, e.g., Raymond 2013;Havelmann 2002;Beydoun 1984;Haugbolle 2010).
But still, the denomination of the different groups as Marxists, Maronites, Islamists, Maronite nationalists and Socialists remains ambiguous. To whom do such categories refer? The term "Maronite" denotes one of the 18 state-recognized communities in Lebanon, the largest Christian one, which has been supported by the French ever since the nineteenth century. What then is a Maronite historian? A historian who writes history for the Maronites? From the perspective of the Maronites? Weren't historians supposed to be as objective as possible and not simply propagandists? And how can the Maronites be distinguished from Maronite nationalists? Doesn't the category of the Maronites comprise that of Maronite nationalists? But also on another level, the subdivision Maronite nationalists is puzzling, not because there were no right-wing nationalists among the Maronites -quite on the contrary-but because this nationalism is sectarian, and thus based on the exclusion of other groups from national identity. 6 Thus, while listing different kinds of historians usually refers to different schools of thought, here the identitarian and sectarian dimension is infused to pervert categories. The same is true, although not directly apparent for a non-Lebanese viewer, for the term "Socialists", which in fact designates the "Progressive Socialist Party", another name for the party of the Druze community, as this party constitutes its major political organ. This is so true that in Lebanon, Socialist in Arabic, came to simply signify Druze in the language of everyday life. This could not be said of the "Marxists"-there were at least two communist parties, both multiconfessional, in Lebanon at that time (see, e.g., Maasri 2009, 27-28). However, this did not prevent its members from acting in the civil wars in the same way as all the other militia groups, gaining influence in certain areas, and attacking other militias representing specific communities, etc. The term "Islamists" is perhaps the most ambiguous, for even if there is an association with the Shiite Hezbollah, which has close ties to the postrevolutionary politics of Iran, it is mostly used in a pejorative way to denounce Islamic fundamentalism of any kind. Is it thus Dr. Fakhouri's own Christian bias operating here, for whom all Muslims-Shia, Sunni and the other denominations alike-whether they claim to be or not, are in fact Islamists, that is, seeking to reinstate the Sharia and realigning the political with the religious sphere? Once again, efforts to understand the mentioned classifications necessarily remain speculation.
What seems to be a coherent list of identifiable groups contains a multitude of subtle antagonisms, revealing themselves through further examination of the historical context in which they occur. Their heterotopic constitution points to a whole lot of problems inherent in the apprehension of designations from the outside: in this particular historical context, a strict distinction between religious communities and politics, or between current politics and history, is not operable, as they intertwine in many ways. The very ground on which one could unilaterally apprehend Doctor Fakhouri's file vacillates and dissipates into vague assumptions, implicit knowledge accessible only to insiders familiar with the particular historical moment. This impression becomes even more striking with regard to the short comments that Fakhouri adds to each page-annotations which do not concern the horserace, the historical situation or the political distribution, but the very personality of the winning historian. Those sometimes sarcastic or ironic, sometimes sly or jaundiced, sometimes puzzling comments appear as laden of a meaning which remains elusive without further knowledge of the person in question. What becomes palpable is that approaching political organizations according to normative ideas without taking the underlying identitarian ascriptions, the personal bias, the historical context into consideration proves to be not only reductive, but also manifestly distortive. On the other hand, the file alludes to the fact that order there is, even if this order drags the one who assesses it into his mill. Apparently, we are either too close, that is, interlaced with the logic in question, or too far, which means that our categorical thinking is too rigid, too inflexible as to conceptually grasp what is meant. Raad's complex project shines a light back on the idea that such a distanced, detached look on a political reality is ever coherent.
The Atlas Group is full of such subtle references to the problem of knowledge transfer from one reality to another. Under its seemingly uniform facade, its ostensibly coherent forms and the allegedly universally understandable concepts, frictions, ambiguities and inherent tensions emerge in manifold ways and undermine the consistent appearance. In some of the files, this transpires directly through the incongruity of the translations. Most of the documents are indeed accompanied by explanatory or referential textual parts in two or more languages. Sometimes the same reference is rendered in different languages, thereby making the ability to ascertain the "original" unattainable. Sometimes, as in Missing Lebanese Wars, some of the information is written in one language, while the rest is expressed in another, thereby impeding many spectators from accessing the content as a whole. This alludes once more to the Lebanese situation: while Arabic is the official language, French and English (and, to a lesser extent, Armenian) are much used in everyday life as well as in official, cultural and educational institutions. However, in the documents of The Atlas Group, we are confronted with more than a mere coexistence of multiple languages. Most of the time, subtle but relevant differences insinuate themselves in the passage from one language to another. In Already been in a Lake of Fire, another file attributed to Dr. Fadl Fakhouri, this even becomes visually perceivable. The notebook consists of cut-out photographs of cars, arbitrarily pasted on white pages and accompanied by Arabic handwritten texts, artfully integrated in the design of the page. The cover and back pages (depending on the reading direction-left to right in English, right to left in Arabic) contain the title in English and an Arabic "Appendix" explaining what it is about. As the story goes, Dr. Fakhouri sought to photograph vehicles whose make, model, and colour corresponded exactly to those that have been used as car bombs during the civil wars. Alluding to the fact that it was a common habit of journalists of that time to replace elaborate elucidations of the political motifs and impact of such attacks by a symptomatic fetishization of the car itself, this file is once more concerned with the generalized climate of insecurity and paranoia. All the photographed cars with blackened license plates are in fact reproductions of real cars captured in their particular everyday environment, discernible through the reflection in the windows. None of these cars has been effectively used as a car bomb, but their reproduction in the framework of this document alludes to the fact that they could have been or still may be, and that the danger is imminent, unpredictable and very close. Next to the images, a type-written list in English enumerates allegedly factual details: the make, model and colour of the car, the date and location of the incident, the number of victims and the size of the crater left by the explosion, as well as specifications concerning the explosives that were used. Even if this data deflects the attention from the event as such, and even if it is sometimes evident that it is only approximate (e.g. the make could be "Toyota or Subaru", and the colour, "red or blue"), it suggests a certain neutrality and objectivity. In contrast, the Arabic text is not only handwritten and integrated in the image, but also differently structured. Instead of enumerating cold facts, it describes what happened, adding further elements like the exact or approximate location, thereby helping those familiar with the local geography deduce the targeted community. Furthermore, it sometimes quotes its sources and reports the event in a more dramatic way (e.g. "a new carnage", "the series of horrors", "the war of the car bombs"). While the English text follows a logic of tabularization, listing pre-specified, supposedly relevant characteristics, the Arabic version links disparate information through narration, constructs a more complex story and alludes to implicit local knowledge beyond the effectively reported facts. Both avoid a thorough analysis of the situation, and both provide information without an actual connection with the photographed car as such. Rather, one could say that the images connected to the texts serve as a kind of visual argument or illustration of what the words say.
And yet, the form alludes to a different device common in media and research documents-a device linking an indexical image with a caption which grounds it in a specific context or meaning, thereby producing an effect of immediacy. In this widespread device, the image comes first. It is the indexical character that produces an allegedly incontestable (past) relation to the reality it represents-the famous "ça a été" ("That-has-been") dear to Barthes [1981Barthes [ ] 2000. However, the indexical relation of a photograph with the represented reality is never sufficient for the understanding of its meaning. As Charles S. Peirce, who introduced not only semiotics, but also the idea of photography as an index (cf. Peirce 1955, 106), strikingly states: "The index asserts nothing; it only says "There!"" (Peirce 1993, 379) Just as the notion of the purely factual, which is not only problematical because it appears as uncontestable historical truth (see, e.g., Horkheimer 1947, 56-59), but also because it does not suffice to explain any situation, indexical images require further knowledge so as to become comprehensible. Therefore, the associated text-the caption or the article-is needed in order to provide a complementary information. Already Been in a Lake of Fire, however, confuses this order, by concomitantly appealing formally to it but undermining its apprehension by breaking the images' direct association with the given information through explication in the "Appendix".
The problematization of the device associating a caption with an image is even more striking in another file entitled, Sweet Talk: A Photographic Document of Beirut. It consists of a collection of round photographs, and is described as a commission that The Atlas Group gave to diverse photographers, asking them to capture various places in the city of Beirut, and indicating one date, but three possible locations, only one of which corresponds to the referent in the picture. Here, the attention is explicitly drawn to the device itself, which is frequently used but rarely examined (while criticism concerning the content of legends or the framing of the image is frequent). In Sweet Talk, the space taken up by the caption is still present; its function is thus maintained. But its content, due to the proliferation of multiple meanings, is immediately eliminated, as if what remained of the sign after the elimination of the signifier was nothing but a signified waiting to be complemented. Due to this suspension, to this deferral to a subsequent moment, the relation between the image and the associated meaning, or between something signified and its motif, can no longer be perceived as immediate. The dialectic between image and attributed meaning is triggered but immediately voided, or, more precisely, suspended. This, in turn, opens up a space for multiple projections, or to constructions in perpetual movement which cannot be stabilized in a given. What remains is an image relating to a signification that is not simply absent but explicitly missing, in suspense, yet to be disclosed. These photographs, with their foreshadowed legends, cannot be brought down to a specific historical meaning, which is why they appear themselves to be fuzzy: the dialectic implodes in the images, which become thus saturated with the entire tenor of the device.
Also in Already been in a Lake of Fire, the formal allusion to this device and the coexistence of two different ways of textually mediating the same event points to the problem of mediation and the performative agency generated by its standardization. Despite their photographic nature, the images do not represent the historical situation described. They are at best clumsy "copies" of the original cars. Neither of the two texts associated directly with the images relates to the effectively represented cars, but both also elude the event to which they refer. Moreover, the file opposes, through one and the same image, two different ways of relating the photographs to a past reality: the English account apparently documents objective facts, thereby producing an effect of evidence, and the Arabic narration charges the event with numerous implicit meanings, raises suspicion and generates speculation. This intricate composition creates a space of tension, where each element undermines the other without one taking the upper hand, thereby challenging the device linking image to text, text to translation, and an event to its mediation.
One of the characters appearing in the framework of The Atlas Group clearly expresses his refusal of accurate translations, albeit not through the file attributed to him. In a fictive interview with Walid Raad, Souheil Bachar, an imaginary co-author with The Atlas Group for the file bearing his name, declares having himself translated the video from Arabic to English, to add that he did not wish to comment on the fact that the English version is sometimes opposite, sometimes completely unrelated to the Arabic original (cf. Raad 2002, 125). Thus, the distortions are presented as deliberate decisions, with an implicit critical tenor. They suggest that the relation between an object or event and the words supposed to grasp it is not simply transferable without consideration of the historical situation, the geopolitical location and the socio-cultural environment. This is especially the case in a media landscape which often describes events and actors in a simplified manner, abstracting from the field of meaningful resonances in which they are embedded. This is the case, for example, with the mediation of events in the Middle East by the American press, where the designation of someone as an Arab or Muslim already charges the affair with specific connotations, as, for example, Edward Said has comprehensively analyzed (cf. Said 1997; see also, e.g., for an artistic assessment of this subject, Elia Suleiman's and Jayce Salloum's film Introduction to the End of an Argument/Speaking for Oneself … Speaking for Others, 1990). The very dense and complex Bachar Tapes, of which only two are available, 7 are extensively concerned with the question of the perception and assessment of a fact, its conceptual framing and medial representation.
Souheil Bachar is described as a low-level employee in the Kuwaiti Embassy in Beirut. Played by a famous Lebanese actor, Fadi Abi Samra, his fictional character is immediately evident to the Lebanese audience, but not to foreign spectators. Bachar is said having been abducted by "Islamist" terrorists and held captive in the same room as five American hostages. The names he mentions-Terry Anderson, Thomas Sutherland, Benjamin Weir, Martin Jenco and David Jacobsenare all known for having effectively been held hostages in Lebanon in the 1980s. While their individual cases and those of other Western hostages have often made the headline news and were sometimes related to huge political scandals (like the Iran-Contra affair), the hundreds of Arab citizens that have been held captive at the same time have fallen, to a large extent, into oblivion. Thus, Bachar's visual presence directly responds to this ideological imbalance resulting from the bias and double standards of the media, which contributed as well to a monolithic perception of the conflicts. But the video is much more than a mere attempt to balance this state of affairs through counter-information or by granting some visibility to one of the concerned victims: besides the representation of a former Arab hostage reporting about his detention, the file also consists of a subjective appropriation of local discourses and images about the event, and a subversive, critical assessment of their media treatment.
Bachar Tape #17 is composed of an intricate montage of different visual and audio materials. The first images after a short "Prologue" already mark the tone: they show Bachar himself sitting in front of a white sheet, switching on the camera with a remote control. The first impression is already an ambiguous one. The setting immediately evokes the numerous videos with hostages, made by the kidnappers to show that they are still alive, and to make their demands. In Bachar's video, however, it is the kidnapped himself who is in control of the camera image, and who reappropriated this recognizable device in order to problematize the associations it triggers. Furthermore, the power-on in the direction of the lens alludes to a shooting, thereby sensibly confusing Bachar's role in the story. In the following film, various kinds of images are edited together: low quality rushes focussing frontally on Bachar while he is speaking and archival material showing (real) television news from the period alternate, cross and overlap. The soundtrack is an intertwining of Bachar's monologue in Arabic and its simultaneous translation into English, spoken by a "neutral female voice", as he says in the prologue. Further to the above, there is also another male voiceover reporting general information, as well as abstract noises and sounds from the included film excerpts. Hence, the video appears as an overloaded constellation which correlates the divergent images and sounds without hierarchy. Even if the personal account of Souheil Bachar somehow serves as a frame. For it is he who introduces the subject and clarifies details about his shared captivity with the American hostages, and he as well who allegedly composed the video, chose the fragments and edited them. However, he does not reveal much about the way he experienced his captivity. Instead, Bachar's monologue focuses on his conditions and on his co-detainees, the aforementioned "Americans", who had taken an ambiguous stance towards him. Despite the fact that they were all forced to share the same conditions as hostages, they would gang up on him since he shared the same origins as the perpetrators. Not only did they exclude him from the community of victims they built, but they were also visibly both attracted and repulsed by his presence, keeping him at a distance while secretly seeking physical contact. Rather than denouncing this behaviour, 8 Bachar addresses this attitude critically by associating it to the way the Americans dealt with the situation at the time and afterwards. Contrary to Bachar, who relates their (and his own) abduction to the political context and the historical situation, the Americans would perceive their kidnapping as something fateful, extraordinary and tragic, according to Bachar. This evaluation is echoed by the fact that all the books they wrote about their detention begin with a description of the meteorological conditions at the moment of capture: just like a thunderstorm, they considered their abduction a fatalistic coincidence and themselves its accidental victims. These mostly bestselling memoirs, concentrating on the dramatic personal experiences rather than on their embedding into a highly conflictual political situation, dissociate the subjective sphere from the complex reality. Instead of considering the different layers that link the objective facts to the subjective experiences and their mutual interferences, they depoliticize the situation by considering it as a merely personal catastrophe. Bachar's film, by contrast, complexifies the angle instead of reducing it to its individual dimension. Rather than dissociating the representation of personal experience and that of factual truth, he constantly relates one to the other. Thus, he confronts the subjective ways of dealing with the situation with both, the television news of the time and the allegedly sober male voiceover, which appears, on the contrary, as a completely impersonal account of cold facts. The authority of these devices is further backed up by the extensive use of indexical images, which seemingly attest their close and authentic connection to the reality in question. Just as the English list in Already Been in a Lake of Fire, they appear as neutral reports with a hegemonic claim to truth. What transpires through these seemingly complementary perspectives-one claiming to concern only the most individual, personal, subjective experience, the other allegedly concerned with the merely factual, objective reality-is that both position themselves as utterly detached from the other, suspending their mutual mediation.
Bachar's video, in contrast, interrelates subjective and objective elements in heterogeneous ways. It breaks the appearance of objectivity of the reported facts through the multiplication of divergent angles, for instance by diversifying the sources and doubling the voiceovers, and by alternating from the television image to his own. As a first-hand witness of the facts, this confrontation reveals the partiality of the information divulged through the news. In contrast to those allegedly neutral attempts to grasp reality, Bachar's speech is an openly auto-reflexive appropriation of events, not only because he is concerned with emotional or inward experiences, but because he approaches the facts through a reflection on his own subjective position. Instead of accepting his speech to become yet another neutral-sounding account, he sticks to his dialectical position by presenting the complicated mediation in a constant movement, thereby questioning the respective attributions of seriousness and triviality, and intervening as a questioning subject. His attitude can be approximated not only to the critical stance that transpires through the projects of artists such as John Akomfrah or Trinh T. Minh Ha (who also strikingly conceptualizes her critical approach, see for example, Minh Ha, 1990), but also to Theodor W. Adorno's critique of the established definition of both spheres as featured by, among others, mass media, the positivist sciences and the culture industry. In Minima Moralia Adorno states: "The notions of subjective and objective have been completely reversed. Objective means the non-controversial aspect of things, their unquestioned impression, the façade made up of classified data, that is, the subjective; and they call subjective anything which breaches that façade, engages the specific experience of a matter, casts off all ready-made judgments and substitutes relatedness to the object for the majority consensus of those who do not even look at it, let alone think about it-that is, the objective" (Adorno [1974] 2005, 69-70).
By constantly disturbing the smooth construction of events, by slipping elements which elude ready-made objectivist appropriation or purely emotional reactions, the intricate Bachar Tapes draw the attention on the impact of diverse mediations-be it through media, subjectivity, an institution, an established form-for the recognition of a historical truth, a geopolitical situation or a social phenomenon. They disclose the multiple presuppositions on which hegemonic claims to truth are erected, alienate supposedly direct links between indexical images or factual data and the reality to which they refer, and restitute the ideological parts conceived by established forms of mediation, thereby opening an access to its other, that which resists: subjective voices in their political dimension, perceptions difficult to grasp, but omnipresent in a society and constitutive for the relations between its people, or the highly conflictual charge of words.
In the interview with Alain Gilbert already mentioned above, Walid Raad claimed that the documents of The Atlas Group should be considered "hysterical documents" (Raad/Gilbert) rather than historical ones, "in the sense that they are not based on any one person's actual memories but on 'fantasies erected from the material of collective memories.'" He thereby stresses the fact that The Atlas Group is not a "real" research structure; rather than being concerned with the factual elucidation of events, a rectification of data or the erection of a comprehensive counter-narrative, it sensitizes for the latent imageries and forces beyond the sphere of the factual, and introduces frictions in that which appears as objectively real. But the reference to a pathological phenomenon leads also to a critique of the societal order, and as such, its political foundation and respective institutions. For implicit in the term "hysterical" is the contrast with that which is considered normal or healthy, a standard that is acknowledged in a certain context and linked to a certain power. By naming the documents hysterical and considering them as symptoms, Raad adapts his discourse to the prevailing one, where everything that is not rigorously assigned a fixed identity seems to become non-existent. Hysterical is a positive term given to the sphere that cannot be grasped by positivism. That which is repressed, excluded or simply not admitted constantly reappears in intricate, paradoxical ways. For this designation refers also to the complex figure of the symptom-an overloaded, intrinsically conflictual figure, in which heterogeneous temporalities, multiple signs and antagonistic forces cross (see for example Didi-Huberman 2017). As Georges Didi-Huberman writes: "the symptom plays with the antithesis: it creates 'incomprehensible situations' because it knows how to impart to the most complex workings of contradictory simultaneity a plastic intensity-that is, a phenomenal evidence presented in its entirety to the spectator, like a sculpture. Here, conflict and compromise, Reaktionsbildungen ('reaction formations') and Ersatzbildungen ('substitute formations') coexist and respond to one another. Here, representations that are repressed coexist and exchange with representations that repress" (2001,636). In Raad's discourse, what is designated hysterical are the documents themselves: instead of conforming to the established rules, of being subsumable under coherent ideas or categories, of exhausting themselves in topical content, they are a complex constellation in which form and content interlace on many levels. They expose the frames of the conventional against the impression of conformity they evoke, without however subletting the profound antagonisms and tensions which permeate them. And just as symptoms of madness, they reflect back on the norm, on the system of inclusion and exclusion.
In the framework of this peculiar art project, the persistence of rumours, the multiplicity of diffuse signs and the heterogeneous constitution of the Lebanese reality are figuratively interwoven with the problem of the construction of events and their acknowledgement as facts. The documents draw the distant spectator into their climate of suspicion and uncertainty: caught in the paradoxical logic of this invented archive, but also by the persuasive rhetoric of Raad's compelling discourse and the convincing presentation of the documents, we lose the ground necessary for a distant exploration of its content. Herein lies the performative force of the project: rather than representing a historical reality, a socio-political situation, or even the institution supposed to hold its traces, The Atlas Group worms its enigmatic logic into our perception and makes us sense the instability and uncertainty it is concerned with. Rather than producing a positive knowledge presenting yet another hegemonic claim to the truth about the Lebanese history and present or formulate a philosophical critique, Raad destabilizes our certainties and incites us to reflect, first of all, on our own position. Notes 1. André Lepecki writes with regards to this subject: "Accent, the grain in the historian's voice, was the first indicator of the game of mirrors structuring that evening's lecture-a lecture that veered slowly, hesitantly, yet never quite fully openly, towards performance. As I discovered later on by attending. 2. However, there are some archival organizations, mostly NGOs with a clear political objective, which aim to collect heterogeneous documents and make them available for the public, e.g. the controversial UMAM Center for Documentation and Research. Established in 2005, its explicit purpose is to critically engage with Lebanon's complicated history and to trigger public discussions. 3. This classification says that « animals are divided into: (a) belonging to the Emperor, (b) embalmed, (c) tame, (d) sucking pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush, (l) et cetera, (m) having just broken the water pitcher, (n) that from a long way off look like flies' (see Borges 1993, 101-105;or Foucault 1970, XVI). 4. Foucault applied his concept of heterotopia both to linguistic spaces (in The Order of Things) and geographical places (in Of Other Spaces), with slightly different connotations. 5. It might be important to mention that, due to the lack of official information and the very nature of these civil wars, the sphere of rumours took on a very important, and sometimes performative, role during the Lebanese wars (see, for example, Kovacs and Nassif 1998). 6. The most evident reference would be the Kataeb Party (Phalangists)-a political party founded in 1936 and inspired by Nazi-Germany, and which militarized in 1975 (see, for example, Fisk 2001, 65); or the Lebanese Forces, which was the militarized branch of the Phalanges until 1982 before becoming a separate party; however, it could also refer to any other splinter group. 7. According to the description given in video #17, Souheil Bachar allegedly produced 53 videotapes about his captivity, but only authorized the public diffusion of #17 and #31 outside Lebanon. However, in the interview quoted above, he claims having shown the video publically in Lebanon, Syria, Iran, Iraq, Egypt, Libya, Palestine, Sudan and Morocco. 8. There is a clear resonance with Edward Said's analysis of 19 th century orientalism, consisting of "Western" projections of vulgarity and eroticism onto the "Orientals"; see Said (1978); see also Walid Raad's doctoral thesis (1996) which took the media representation of Western hostages as its subject, and includes numerous references to Said's book.

Disclosure statement
No potential conflict of interest was reported by the author.