1 “History decays into images, not into stories”,2 noted Walter Benjamin in The Arcades Project, and in his theses On the Concept of History he wrote: “The past can be seized only as an image that flashes up at the moment of its recognizability, and is never seen again”.3 Here, Benjamin is not referring to the film images that briefly appear in the light of the projector but rather images as a form of cognition which Benjamin terms “dialectics at a standstill”.4 Sven Kramer demonstrated how Benjamin applied his theory of linguistic images, developed from essays he wrote describing cities in the 1930s, to the writing of history.5 Kramer claimed that Benjamin’s concept of images undermines the distinction Lessing made between painting and poetry in Laokoon. According to Lessing, images must capture the action in its most concise moment, whereas poetry can only represent bodies dissolved in action.6 By contrast, “Benjamin insists on the pictorial element of standstill and coexistence for language and writing as well”.7
In his analysis of Brecht’s epic theater, however, Benjamin notes that dialectical images are by no means solely linguistic images: “The conditions which epic theatre reveals is the dialectic at a standstill”.8 “Epic theatre moves forward […] jerkily, like the images of a film strip”.9 Yet in cinema, Benjamin made no systematic attempt to teach Lessing’s momentary images [Moment-Bilder] the movement of language, even though he emphasized the outstanding role of film as a powerful agent of social transformation in his essay “The Work of Art in the Age of Its Technological Reproducibility”.10 Perhaps this was because he saw images being set in motion in the cinema, but only at the cost of submitting themselves to the temporal succession of the narration. To create dialectical images in film, it would be necessary to bring film images to a standstill that would release their inner movement and thus liberate them from the mechanical progression of time.
Fifty years after Benjamin, Gilles Deleuze classifies films into two broad categories with philosophical intent: the movement-image11 and the time-image.12 He referred to the classical cinema, which Benjamin was familiar with, as the movement-image, characterized by a movement that is carried out from start to finish along a space-time continuum. He also applied this logic of movement to Eisenstein's dialectical montage.13 While the movement-image continued to be used in Hollywood films, Deleuze claims that it was not until after the Second World War that modern cinema succeeded in producing a new time-image that used irrational cuts to break the temporal logic of succession, thereby allowing the direct presentation of time.14 While Deleuze was not looking for a dialectical image, the question nevertheless arises whether the concept of the time-image can be used for the conceptualization of a dialectical image in film in Benjamin’s sense of the term.
Traditional history is not interested in either dialectical or time-images. Positivistic sequences of facts in a progressing history continue to be the prevailing paradigm in historiography. Its structure therefore barely differs from the dramaturgical structure of the movement-image in the average Hollywood production. Nevertheless, film is a medium that most historians deem to be untrustworthy and at most suitable for a popularized dissemination of historical material or a source of historical-cultural information. The conditions for reception in cinemas are too suggestive, the films too replete with fantasy. Historians have been hesitant to acknowledge that films have become an inherent part of today’s world and that they present history in every imaginable format — from TV movies, feature-length documentaries and experimental auteur films to Hollywood’s historical dramas. Only recently did some historians begin to seriously consider how the study of history could benefit from film. Despite a great deal of resistance, three general approaches to studying film and history have, to differing degrees, become established methods for historical research.
Initially, the use of films as a “historical source” was the first generally accepted option. In 1947, Siegfried Kracauer published his ground-breaking study From Caligari to Hitler,15 in which he examined German film production to investigate the collective dispositions that led to Nazism.
Apart from chronicles that present production figures and film descriptions for specific countries, genres, and time periods, “film history” looks at film and society from a cinematographic perspective, seeing the history of film as an ongoing development of different forms and technical possibilities.16
A third approach examines the “representation of history” in film. While these largely literary debates discuss possibilities for representing history, e.g. the representability of the Shoah,17 they usually do not explore how films could become a means of constructing history and thus part of the historical discourse. Eike Wenzel, who studied the analysis of German history in films since 1960, took these approaches even further, applying theoretical methods from media studies and historical theory to discuss the results of his insightful film analyses.18
All three of these approaches attempt to integrate film into the (historical) academic canon that is based on written works. But they have yet to consider the dialectic between the term history and a historical situation or that of the medium of historical research and the representation of history. As early as the 1970s, Hayden White examined historiography in the 19th century as an example of the far-reaching consequences that the necessity of choosing a narrative form to represent history had on the constitution of historical facts and the depicted history.19 This correlation can also be applied to film. In a paper written in 1988, White coined the term “historiophoty” for “the representation of history and our thought about it in visual images and filmic discourse” as the filmic counterpart to “historiography,” the “representation of history in verbal images and written discourse”.20
Joseph Vogl follows a similar argumentation, writing that the history of political economy is based on the fact “that it must first constitute the objects that it uses”.21 He thus recognized “that the emergence of new objects and areas of knowledge correlates with the forms of their representation”.22 To study this correlation, he developed the concept of the “poetology of knowledge.” “That which becomes visible in the poetological dimension is the historicity of this knowledge, the fact that there are no givens beyond its form of representation”.23
White and Vogl go far beyond the issue of representing history. If historical facts must first be constituted in language, and the only thing that is tangible for the study of history is language, then the traditional concept of history is called into question. The reference to the incontrovertible truth of facts that is so important for positivist science thus becomes problematic. In his study on historical referentiality, Hans-Jürgen Goertz concludes that history has become something uncertain whose subject is no longer in the past but is instead the construction of a relationship to the past.24 This relationship is an important aspect for critical history. It cannot be content with the relativizing historization of facts if this process is not dialectically coupled to the social dimension. The historical relationship is contested and must be constantly re-established in human practice. “For it is an irretrievable image of the past which threatens to disappear in any present that does not recognize itself as intended in that image”.25
The concept of the “representation” of history is, at the very least, problematic. It presupposes the notion of a clearly ascertainable previous history that merely needs to be represented in film. This presumption led Rainer Rother — who had otherwise been commendable in his efforts to establish the study of film for historical research in the German-speaking world — to restrict the question of how film affects historiography to its potential as additional sources of material and the problem of representation. In Die Gegenwart der Geschichte [The Present of History], he examines, from a thoroughly dialectical perspective, forms of reflection in historiographic representation in film and literature, including works by Walter Benjamin, Alexander Kluge, and Peter Weiss.26 Nevertheless, he relegates any attempt to relate the form of representation to the constitution of the object to the realm of irrationality and the abandonment of the basis of academic history; only fiction and aesthetic reflection are accorded such a procedure.
Like Goertz, Rother distinguishes between the past and history. He defines the past as something distinct from the present that has passed forever, and history as the retrospective reconstruction of the genesis of the past. According to him, the distinction between the past and the present as well as the continuity between the two is required for any kind of historical knowledge.27 But in contrast to Goertz, Rother does not see history as a relationship to the past that is created by research but rather as the subject of historical representation, the part of the past that has essentially entered into the present and thus become recognizable. Historians do not create history; they recognize and represent it. We “cannot come to terms with history by declaring it to be the mere result of historiography, thus eliminating the perception of an object from the ‘perspectivist approach’ to historiography — because if that object was created by the historian, he would no longer have an object”.28 Since Rother denies that historiography has an influence on history, he is forced to assume that historical objects undergo an inevitable development that can be reconstructed from a present that is understood as being the result of this very development.
Consequently, Rother does not subscribe to the dialectic of history that Marx was striving for in “The Eighteenth Brumaire.” “Men make their own history, but they do not make it as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past”.29 Rother fails to understand the relationship Marx described between self-determination in human practices and their double dependency on social circumstances that are both a constitutive condition and the very object human practices wish to change, thus connecting Marx to the world spirit in Hegel’s philosophy of history: “Just like the Hegelian spirit, the capital ratio is both the source and the result of history”.30 Instead of an open process of totalization, society becomes a closed totality. Consequently, Rother understands “truth” as something that is produced by totality and thus discoverable in history, instead of seeking it in the sublation of social alienation, and, like Benjamin, calling for a “[r]esolute refusal of the concept of ‘timeless truth’”.31 Benjamin does not believe that truth “is not […] a merely contingent function of knowing, but is bound to a nucleus of time lying hidden within the knower and the known alike”.32
An intertwining of historical research practice with its given object, which allows historical “truth” to emerge — from a practice both conditioned by and applied to its object, remains alien to Rother. This practice is historical in two senses: first, it is conditioned by and therefore bound to today’s social conditions; second, when it is applied to the traces of past events, a connection is established to these events, thus creating the historical object of its “truth.” These two entangled moments are contrasted with one another in Rother’s argumentation: “If historiography produces ‘truth,’ but that truth is not the truth of an object, irrationality cannot be avoided”.33 This therefore leads to a resistance toward any kind of reflection on the influence of the narrative form on history: “The danger of historiography does not lie in its limitations, which in fact constitute it, but rather in the transferal of the structural characteristics of narrative texts onto the stories they tell”.34
Based on this definition of history, Rother believes that any history that is not an affirmation of the present must be impossible, because the proclaimed “truth” of the objects is proven in their realization in the present or their failure in the past. Consequently, he claims that Benjamin’s philosophical conceptualization of an actualization of the past that is aimed at a critical, defiant history is an endeavor that cannot succeed.35 In his analysis of Alexander Kluge’s film THE PATRIOTIC WOMAN/DIE PATRIOTIN (West Germany, 1979),36 and Peter Weiss’ novel The Aesthetics of Resistance,37 as well, Rother is not interested in the potential of these works for a different kind of historiography. He sees them as an artistic reflection of the conditions of historiography that demonstrate the inevitable failure of any attempt to write anything other than the history of domination. The fictional works
criticize the limitations of the historiographical form, yet for good reasons, do not overcome these very limitations. [...] Nothing can change history except a future that sees its past more rightfully completed in itself than this present. [.] Only in failing can these works remind us of this.38
Rother does not explain how this better future can come about if it requires history to remain unbroken. He thus reveals a naive concept of the future, something Stefan Gandler, following Benjamin, criticizes: “The idea of the future is the result of the rejection of a complete and fully lived present; it is the demolished present”.39 But the future does not exist from an ontological point of view, he argues, and the positivist attempt to reconstruct “today as ‘the future of yesterday’”40 must fail. “The concept of time as an absolute and certain point of reference is the ideological response to a society that is chaotic and full of antagonistic contradictions and deeply irrational structures”.41 According to Rother’s logic, it is only conceivable that history take its own course, whether as the self-fulfillment of Hegel's world spirit or a self-regulating capital ratio. This intellectual self-blockade that prevents Rother from understanding history as a transformative practice thus also becomes a political blockade of emancipation.
Rother’s essentialism is not untypical of history as practiced in Germany, which clings to the idea of an unalterable object called history that precedes all epistemological work and must be recognized and adequately represented in its true state. This is the boundary that must be crossed if one wishes to take advantage of the opportunities offered by film as a medium of historical research. Just as language is the only way a historian “writing” history can “access the past and gain insights from it”,42 film can also be a medium for historians to produce historical knowledge and history.
Examples of the integration of film into the research process can already be found in the “oral history” that became established in the 1970s and 1980s. Eyewitness accounts are a central tool for gaining information in “oral history.” Since the early 1980s, the rapid development of video technology has contributed to the increasingly frequent practice of recording eyewitness interviews as on video as well as on audio tape, often in a collaboration between filmmakers and historians, and using these recordings to produce a documentary ﬁlm.43
This approach uses film as more than simply a source, because the process of acquiring sources has already been carried out — more or less deliberately — in the film. The montage of sections of eyewitness accounts, often containing contradictory statements from different witnesses, and their confrontation with other visual material such as historical footage and photographs, can be seen as a component of a critical assessment of sources and debate. For this reason, more is created than just a narration based on verified facts from external sources; it is an explorative construction of history using film as a medium.
Robert Rosenstone is one of the few historians who were actively involved in film production, both as the director of documentaries and co-author and advisor for feature films. In several essays, he compares the structures and possibilities of film texts with other written forms.44 Just as many written texts do not meet the standards of academic research, Rosenstone also points out that only certain films are worthy of closer examination for historical research. He suggests three categories for classifying films with historical content: history as a drama, a document, and an experiment.45 However, this distinction is problematic. Rosenstone himself notes that the categories “document” and “drama” construct history using the same narrative patterns, while documentaries turn fact into fiction during montage, at the latest, by connecting events that were unrelated in reality. For Rosenstone, the inherent distinction between “fact” and “fiction” in the categorization of films as documentaries or narrative films is too simplistic. Rosenstone’s category for “experiment” is largely based on a negative distinction from established Hollywood conventions. He mentions himself that he groups a wide range of films into this category.
Rosenstone developed his categories on the level of the genre-specific construction of films, thus implying films are a neutral medium. They do not allow the dialectic of medium and history to take place. Kay Kirchmann suggests a method for examining this relationship, a concept of media that is in a close, reciprocal relationship with the historical process.46 Kirchmann criticizes the substantialist media concept that is common in media studies and defines and ontologizes the medium from a purely technological and material point of view. Instead, he offers a relational-structural definition that attempts to radicalize the concept of a media dispositif “in order to thus arrive at a fully de-substantiated concept of media [...] and understand media as dense dispositifs of structural relationships in society as a whole”.47 The media and the concept of media are not only subjected to changes over time, they are understood as a structural context that is isomorphically identical to the civilizing process as a process of condensation. For this reason, Kirchmann sees “books, films, or images not as things but as structures, and he therefore replaces the category of the ‘medium’ with the structural concept of ‘mediality’”.48 As human constructs, media remain bound to the sphere of human activity. It is therefore necessary to examine the specific cultural history of the structural conditions under which they were produced. The historical process of distinction in increasingly complex societies leads to the “structural necessity” of using media to ensure a more or less binding construction of reality. “All-encompassing mediatization is thus a sub-process of the modern European civilizing process,” Kirchmann summarizes.49
In order to understand film as a media structure for historical research, this perspective requires us to examine more than the technological and material aspects of film. Accordingly, films are more than just a sequence of recordings captured on celluloid or polyester; they are a dispositif for opening up and appropriating the world, one that has a mutually dependent relationship to historical events. Film as historical research would thus be required to capture historical processes from outside itself and represent them within itself, but also to apply self-reflection to reveal the historical structures within its own filmic structures so that they can be used for academic research. For Benjamin, the changes in human perception that are inherent in the cinema as a space for technically reproducible works of art are an expression of social transformations. “The way in which human perception is organized — the medium in which it occurs — is conditioned not only by nature but by history”.50 Material images are not the only important factor for successfully using film for historical research; how these images are perceived as a film and how they continue to resonate throughout other areas of society also appears to be relevant.
In line with Kirchmann’s definition of mediality, we will now examine the relationship between history and cinematography. In the first step, however, I will refrain from looking at the overall context of film, in order to examine individual shots in more detail. Using the same material carrier, films can represent history in different forms:
a. Film as an “archive”: Footage produced during the historical period under observation, regardless of whether it was produced for documentaries or movies. Here, the relationship to the past is in the history of the film itself.
b. Film as “trace”: Footage that captures the remnants and memories of a historical period, in interviews with eyewitnesses, shots of historical places today, or in scenes that show how characters were marked by their past. Here, the relationship to the past is in the history of what was recorded.
c. Film as a “derivation”: Footage that derives from a relationship to the past and whose object is this very relationship, for example, the re-enactment of a scene or an interview with experts.
In the first case, the film serves as an archive. We see what was recorded on film in another period. The film itself is an artifact from the past and thus refers to a past to which it once belonged and to which it has a special inner relationship. When seen in isolation, the film images are unaltered; but in relation to time, they have changed irrevocably and are thus merely a relationship between the present (its reproducibility or actualization in the projection) and the past (the context of the shot that is contained in the images).
In the second case, film no longer serves as a material carrier across different time periods; but rather, it allows artifacts and memories to become traces. The relationship between the present and the past is not in the history of the footage but in that which was filmed and is, strictly speaking, first established by the film. The film interviews are the occasion for the interviewees to remember; from a historical point of view, a shot of a ruin actualizes its status as a remnant of the past. The same relationship is also depicted in a dramatization of, for example, an encounter between victims and perpetrators of past events. Such footage relates a present object to its past. The trace-image becomes a reflection of the distance to the past. The past and the time that has passed have been imprinted on faces, the countryside, and buildings.
The derived image establishes a relationship between the present and the past in that the footage relates what is depicted to a third event in the past. The relationship between the footage in the present and the event in the past is thus measured using a third depiction and derived from knowledge that lies outside the image. A re-enactment creates images that show how things could have, must have, or should have been through a logical derivation from this knowledge. Experts also derives their opinion from their knowledge of the past. The derived image can condense several past events into a single one or create fictive yet “credible” situations based on deductions. In extreme cases, the derived image depicts a detail that is meaningful for the general narrative without being unfaithful to the historical relationship it embodies, even though it contradicts other confirmed sources.
In their 1965 groundbreaking article “Wort und Film” (English title: “Word and Film”), filmmakers Edgar Reitz, Alexander Kluge, and Wilfried Reinke discuss whether film is capable of condensed expression. Film is not able to form generic concepts, empty concepts, like literature does. Cinematography must “attempt with great effort, to destroy the superficial sense of precision which film conveys on account of its excessive visual presence [Anschauung]”.51 This superficial sense of precision becomes a particular problem for historical films; it forces producers to invent details in a genre that is especially interested in the truthful rendition of the facts.52 Even if we accept the necessity of “filling in” a scene as truthful or are able to use documentary material from the period in question, the effect is nevertheless distorted. Rosenstone notes that “[a]ll those old photographs and all that newsreel footage are saturated with a prepackaged emotion: nostalgia”.53
There is therefore the danger that the largely mimetic nature of film leads producers to create the illusion of a “window onto history” in the very places where the material appears to be most authentic. The cameraman Günther Hörmann, who, like Kluge, worked at the Institute for Film Design in Ulm, sees the same problem in documentaries concerned with the present: “The problem is that we cannot represent the truth itself but only our relationship to it”.54
Within the context of his editing technique, Kluge also discusses the problem of realism and creating proportions. The basis for montage, he says, is “the immediate, identificational representation in which the object of which I speak is also present in the image”.55 To illustrate a statement that is “self-contained” and “authentic”, he takes a scene from his movie DIE PATRIOTIN in which a bush near Kaliningrad is unaware of the fact that Kaliningrad was once called Königsberg and belonged to Germany. The conditions under which this scene was shot are not relevant for the authenticity of this statement. “If I assume that the bush near Kaliningrad conveys a relationship rather than just a bush, an object, then this relationship can be created in the mind of the spectator independently of where I have shot the bush”.56
This has nothing to do with the traditional self-concept of a historian for whom the critical appraisal of the source is sacred, because the source is the historical referent that guarantees the truth of his or her historical representation. It is not with regard to the correct temporal and spatial (physical) context of representation that the image must be examined, but rather in terms of the relevance of its sign value and its mimetic quality. As its representation, the image does not have an internal relationship to the truth of fact. The relationship between the image as a signifier and fact as the signified is not created by the technical quality of the images; it is produced by a narrative effort to put the image into context in the film. As a “linguistic” sign, the image has an arbitrary relationship to historical fact, but at the same time, its mimetic quality gives it a special relationship to the past. It can only represent something by embodying it.
Georg Otte points out “that in general, Benjamin emancipates reality from language” and “attempts to dissolve the traditional dualism of signifier and signified”.57 And Gandler writes that “signs and images are in permanent conflict with one another; the image completely loses its significance and its truth in the moment it is absorbed by the sign”.58 The mimetic quality of film images thus becomes significant for Benjamin's specific type of materialism, which, according to Otte, surpasses the radicalism of Marxist materialism “in that it attaches great importance to the materiality of the object of knowledge and its corresponding sensory perception”.59
Historiography, especially when it adheres to the standards of academic research, explicitly identifies and documents quotes. In doing so, historiography is integrated into the general historical discourse, disclosing its methods, and allowing them to be verified. In film, this is the exception. The verification of a possible reference, at the least, is extremely difficult considering how quickly an image disappears again. On the other hand, a large part of Alexander Kluge’s film DIE PATRIOTIN, for example, consists of quotes. A surreal talking knee off camera comes from a poem written by Christian Morgenstern, shots of paintings, scenes taken from documentaries, and historical dramas supplement its monolog, accompanied by music composed by Hanns Eisler for Alain Resnais’ NIGHT AND FOG/NUIT ET BROUILLARD (France 1955).60 Almost as a matter of course, Kluge mixes in material from other sources into his own footage. The “quotes” are integrated into the film’s structure, becoming “buried quotes”61 and thus essentially none at all.
By contrast, the structure of Kluge’s montage cancels out this apparent unity. The individual footage, including his own, refuses to integrate into a narrative flow and jut unwieldy out of the film. They have the effect of a quote according to Benjamin's definition of history, like references to content that is outside of the film’s diegesis.
Julia Kristeva was also interested in implicit text refences from a structuralist point of view. Following Ferdinand de Saussure’s studies of anagrams, she coined the term intertextuality. According to Mikhail Iampolski, this type of quoting does not give the texts a linear progression but rather places them in a relation to one another on a vertical axis. The linearity of filmic discourse and the semiotic transparency of its elements are destroyed.62 In this context, intertextuality has the same effect on the individual image as Kluge’s montage technique has between the images.
Meaning is generated between a physically given datum and an image residing in the memory. [...] Meaning is thus situated in this linguistic field between a heightened corporeality and a physically evacuated nonbeing.63
As Iampolski emphasizes, this method is a “theory of reading.” Meaning is created by mediating between the cultural knowledge of the spectator and the world created by the film. The film that is established in the mind of the spectator can thus go beyond the author's intention. Intertextual references can also be established to subsequent texts. Like Kluge’s technique of montage, this type of quotation also corresponds to a concept of history that puts the relationship to the past in the focus of the construction of history. While quotation in academic writing calls up and evaluates a point fixed in the past, intertextuality aims to bring a moment in the past into the present, thereby giving it meaning for the spectator’s current experience.The concept of reading is also key to understanding Benjamin’s dialectical images.
For the historical index of the images not only says that they belong to a particular time; it says, above all, that they attain to legibility only at a particular time. And, indeed, this acceding ‘to legibility’ constitutes a specific critical point in the movement at their interior.64
Otte notes that Benjamin regards “reading as a special kind of cognition and part of an extensive philosophy of history”.65 This is why Benjamin refers to the reading and quoting of historical fragments whereas Kristeva and Iampolski speak of intertextuality.
The decisive question for Benjamin’s sense of the word reading is whether the carrier of meaning, the medium, is constitutive for this reading to be successful or whether it is reduced to a mere intermediary function in the sense of Saussure’s arbitrariness of the sign.66
According to Benjamin, “historicality should regain its rightful position in language — that of words and things”.67 And according to Gandler, the historical image can “only be rightful if the past moment can confront us in a direct manner”.68
Within this context, the categories archival image, trace-image, and derived image, which are derived from the materiality of the historical relationship in film, acquire meaning for the conception of a construction of history in dialectical images. This practice is historical in two senses: first, it is conditioned by and therefore bound to today’s social conditions; second, when it is applied to the traces of past events, a connection is established to these events, thus creating the historical object of its “truth.”
Claude Lanzmann demonstrates the relationship between a past event and current historical practice in his films about the Shoah. He rejects the image of extermination itself, any historical illustrations, or re-enactment of the events, and thus the authority of a past fact that is reconstructed or represented in images. Instead, he relies solely on trace-images. He shows the places where the events occurred, the perpetrators, and the few survivors in the present.
Even where Lanzmann does construct settings — in SHOAH (France 1985), for example, he rented a hairdresser’s salon for his interview with Abraham Bomba, who was forced to cut off the hair of victims about to enter the gas chambers in Treblinka — he does not attempt to reconstruct the events themselves but rather to create a situation in which the past is sublated in the present. “If I had asked him to sit in a chair and said: ‘So, tell us about it!’, the result would have been completely different”.69 Instead of a visual reconstruction, Lanzmann uses the concept of embodiment.
Film is an embodiment, a reincarnation. The individuals embody themselves; it isn’t about random historical revelations. [...] Suddenly, he embodies what happened when he starts to cry, when he is unable to speak for almost two minutes [...].70
Gandler relates Lanzmann’s film to Benjamin’s theories of the concept of history and calls it a successful “attempt to halt thinking and time for the duration of nine hours”.71 In the Bomba scene described above, Lanzmann succeeded in “interrupting the continuum of history, directly confronting a moment in the past with the survivor today, beyond the control that an interpretation of the signs would provide”.72 This interruption encompasses the spectators, for whom “an intermediary space is opened up” in which they suddenly see “what they had never before been able to see or perceive with other methods”.73
Despite their differences, Lanzmann’s SHOAH and Kluge’s DIE PATRIOTIN have in common that they do not rely on the reconstruction of the past, they do not attempt to create an illusion of how things once were. The films refuse to give the facts the authority that could only be based on their existence as an inadequate representation. Instead, the images appear to be contemporary material that contains a relationship to the past in which the past fact is sublated. On this basis, they process history while at the same time reflecting on the conditions under which this history is constructed and exists in the film. Even when both directors do not apply the standards of academic research to their films, they demonstrate conditions under which filmic research could be used for historical study. Establishing relationships is one of the strengths of filmic historical construction; it illustrates the momentum of construction. The attempt to justify film as historical research must therefore focus on the form of historical relationships and closely examine the associated nature of history as practice.
In his philosophy of history, Walter Benjamin focused on searching for alternatives through historical “actualizations” instead of presenting an optimistic belief in historical progress, emphasizing the relationship between “what was” and “the now-time” instead of that of the present to the past. Instead of phenomenological “essences”, he refers to images with a “historical index” that says “that they attain to legibility only at a particular time”.74 Knowledge comes only in “flashes”; it appears in a “dialectical image” that prefers to appear at a standstill.
Using this as the basis for a critical historiography leads us to think of history as a form of appropriating the present. Benjamin also places human practice at the center of his conceptualization. “Poised somewhere between philosophy and history, like Foucault, Benjamin put historical practice at the center of both intellectual inquiry and eventual social transformation,” Vanessa R. Schwartz noted when The Arcades Project, the English translation of the Passagen-Werk, was published.75
If a relationship to the past replaces the referent that authenticates the course of history, the goal can no longer be to reproduce the past as realistically as possible. Benjamin’s concept of the dialectical image outlines a historical representation that addresses this problem and no longer aims to present history to us as if we were traveling back in time and watching it unfold. Instead, the aim is to use the past to find a prospect for saving us from the catastrophe of capitalistic progress in the present. Benjamin’s philosophy of history formulates a critique of historiography, which should be taken up in film, in order to achieve a conception of history that corresponds with society in the 21st century.
If the so-called postmodern moment in historiography seems mired in a linguistic dead end, Benjamin's questions, topics, and method can help us take cultural history in a new direction — towards the visual. By this, I mean […] an alternative way to think about historical categories and methods — in some measure what Hayden White referred to as ‘historiophoty’ — the representation of history and our thought about it in visual images, as filmic discourse.76
This view focuses on the importance of the special characteristics of a filmic representation of history. Robert Rosenstone emphasizes that one of the qualities shared by films like Kluge’s DIE PATRIOTIN and Lanzmann’s SHOAH is their refusal to adhere to Hollywood’s code of representation: “All refuse to see the screen as a transparent ‘window’ onto a ‘realistic’ world”.77 Film is not a window onto the past; from the perspective of a criticism of the Cartesian epistemological model78 of the camera obscura, this problem acquires a different meaning: academic research must not be a glance through the window, as every window separates us from reality. The “truth” of historical representation is determined by the ability of film to create an image that expresses the past by embodying it.
Referentiality requires a new method for dealing with past records as material for constructing history. The constructive act of arranging the past into constellations replaces a supposedly realistic reproduction. In contrast to Hegel’s dialectic, Benjamin’s dialectic is at a standstill.79 Benjamin remarks on the dialectical image:
In it lies time. Already with Hegel, time enters into dialectic. But the Hegelian dialectic knows time solely as the properly historical, if not psychological, time of thinking. The time differential [Zeitdifferential] in which alone the dialectical image is real is still unknown to him.80
The sublation of the past in the present relationship must therefore not be understood in Hegel’s sense of a dialectic as part of the ongoing process in which the world spirit realizes itself. In The Arcades Project, Benjamin opposes such a simple idea of progress. He sees history as a “constellation of danger” that must be averted. Revolution is not understood as the completion of the course of history but rather as a messianic escape from it.81
Even if the technical premise of film is to take advantage of the slowness of visual perception to create the illusion of movement through a sequence of individual still pictures, this principle must not be applied to the nature of film images: “[C]inema does not give us an image to which movement is added; it immediately gives us a movement-image”,82 Gilles Deleuze wrote, describing a concept of images whose frame of reference is difficult to grasp.83 At times his concept of images seems to refer to individual shots, for example when he speaks of close-ups as affect-images; at others, his concept of images can only be extrapolated from the film as a whole, for example when he distinguishes between the movement-image and the time-image by means of “rational” or rather “irrational” montage. Just as the film image only exists within the flow of the projection, Deleuze defines his concept of images as part of a whole that is in a constant state of transformation.
Deleuze “resists a conceptualization of film that would lead to it being considered solely in linguistic or semiotic terms”.84 Film semiotics frequently applies linguistic models to images, leading to “a strange circle here, because syntagmatics assumes that the image can in fact be assimilated to an utterance, but it is also what makes the image by right assimilable to the utterance”.85 Deleuze thus helps to realize Rosenstone’s wish that film be measured by its own standards.
Deleuze distinguishes between two fundamentally distinct film images, the movement-image and the time-image, each of which has its own taxonomy. According to Deleuze, rational cuts, which should be understood in the mathematical sense, are characteristic for the movement-image. The montage creates a continuum of the cinematographic space; every interval is at once the beginning of one part and the end of the other. Movement is identified with action, which ensures the continuity of space. The movement-image thus only indirectly represents time as the measure of a movement.86
“The terms movement or movement-image refer to a filmic form in which the perception and thinking of filmic individuals is directed toward targeted action (a happy end or a show down)”,87 Eike Wenzel summarizes the quintessence of Deleuze’s category. The most pronounced form of the movement-image is the action-image, which is characteristic of Hollywood films. Deleuze differentiates between a “large form” and a “small form” of the action-image. The “large form” follows the S-A-S' pattern: the protagonist reacts to the initial situation S with an action A, which leads to a modified situation S'. The “small form” reverses this pattern to A-S-A': the protagonist’s action A leads to the situation S, which provokes a new action A'. For the representation of history, Deleuze uses the example of the monumental film for the “large” form and the costume drama for the “small” form.88
Deleuze insists on defining film language as a reaction to a signaletic material that precedes it. This may seem fastidious, but it is important to prevent us from understanding film as always being structured by the narrative. And this is precisely what opens up new possibilities for the construction of history in film. The historical facts, too, initially appear to be parts of a constantly changing whole, but at the same time, they are coupled to a historical event at their specific positions. Following Deleuze, the historical field can be understood as already pre-linguistically structured, even if the narration has always taken over when history appears in the movement-image. The time-image promises to transcend the movement-image in this exact respect.
Deleuze developed the time-image as a contrast to the movement-image. Historically speaking, it emerged in film history at a later point, about the end of the Second World War in Italian neorealism, the French New Wave, and the New German Cinema as well as in Japanese cinema, the cinematography of the Third World, and in independent American productions. Yet the time-image is not simply the opposite of the movement-image: “The movement-image has not disappeared, but now exists only as the first dimension of an image that never stops growing in dimensions”.89 The time-image therefore sublates the movement-image.90 As Wenzel writes, the movement-image is
surpassed toward releasing time/history. [...] Characters and spaces diverge; and for the film — that is the central theory — the helpless silence and apathy of people after the war was a unique opportunity to perceive reality in a new way.91
A typical feature of the time-image is the dissolution of the sensory-motor link, i.e. the temporal-spatial continuity that is held together by the movement of action. The cut thus becomes irrational in the mathematical sense. The interval that divides space becomes autonomous and irreducible; it is neither the end of the one segment nor the beginning of the other.92 Images and sounds also gain relative autonomy; even if they refer to one another, they can no longer be united in an organic whole. The spectators will have to ask themselves: “‘What is there to see in the image?’ (and not now: ‘What are we going to see in the next image?’)”.93
Chris Marker’s LA JETEE (France 1962) illustrates this change. In an experiment, a prisoner of war is forced to travel through time to get help for humanity after a nuclear attack. The individual shots are frozen in still pictures. In order to travel through time, the prisoner is given drugs that activate his memories. A sequence of images appears, a park, a cat, a street, a cemetery, a woman. There are no connections between the shots, it is not even possible to decide what is “real” and what are dreams; the chronology of the events cannot be determined. This montage does not follow a series of movements; it is only through the disparate images that a sequence is established. The perception of the protagonist does not establish a logical sequence of actions.
In a way, the time-image is a doubling of the image that Deleuze describes as a crystal-image: “The crystal-image may well have many distinct elements, but its irreducibility consists in the indivisible unity of an actual image and ‘its’ virtual image”.94 The actual image and the virtual image are distinct yet indiscernible. Distinct in that the real and the imaginary cannot be blended together. Indiscernible because “it does not suppress the distinction between the two sides [of the crystal-image], but makes it unattributable, each side taking the other’s role in a relation which we must describe as reciprocal presupposition, or reversibility”.95 Deleuze illustrates the reciprocity of the actual image and virtual image using the example of a mirror image: “The mirror-image is virtual in relation to the actual character that the mirror catches, but it is actual in the mirror, which now leaves the character with only a virtuality and pushes him back out-of-field”.96
Wenzel established three aspects of the time-image that are relevant for representing history in film.97 First, “a historization of audio-visual material”.98 A temporal panorama in the form of an unorganized archive replaces the person-centered psychological memory in a flashback. Second, a break in the “truth model of chronological and organic narratives” and the associated identity formation.99 Third, the images and sounds “are deprived of their logically expected and conventionalized rules of combination”.100 The time-image therefore no longer attempts to represent reality and aims instead for a new readability. Film reception becomes an ambiguous cognitive act, enabling a “perception that conjures up the potential of unrealized counter-images and the images of memories and the past in the current (visible) image”.101
Deleuze's concepts of the movement-image and time-image can more accurately describe what Rosenstone seeks to capture with the distinction between history in film as a drama and as an experiment, because they are justified on the same level as the signs that emerge in the film. According to Wenzel, Deleuze believes that the movement-image of Hollywood films “embodies an image of history that was decisive for the 19th century”.102 The experimental break with the conventions of classical Hollywood drama appears as a break with the space-time continuity of the movement-image. A filmic place is created where the past and the present overlap; a place that has a great deal in common with the history constructed by a historian, which can only exist in the present yet still refers to events that have long since passed. In history, current events and the virtual past enter into a constellation for which the distinct nature of the past and the present as well as their indiscernibility in an interdependent context are constitutive. Because the past can only be understood from the present as its constitutive pre-history; the present is thus inextricably written into the resulting history as its constitutive condition.
Gilles Deleuze’s taxonomy of filmic images shows surprising parallels to Benjamin’s philosophy of history. Benjamin’s criticism of historicism’s additive procedure of filling homogenous and empty time will remind readers of Deleuze’s criticism of seeing film as a succession of individual images that were subsequently brought to motion103 and his assertion that “the whole is no more an addition than time is a succession of presents”.104 The dialectical image that brings the movement to a standstill and “wherein what has been comes together in a flash with the now to form a constellation”105 is close to the time-image in which “the actual image must enter into relation with its own virtual image as such”.106 And just as history sublates the past in its relationship to it, the time-image sublates the movement-image as its first dimension.
In the medium of film, society’s structural contexts are condensed into a time-image that can place the historical relationship of the present to the past into the constellation of a dialectical image in order to serve as a meaningful matrix of action for human practice. “Cinema [is] becoming, no longer an undertaking of recognition [reconnaissance], but of knowledge [connaissance]”.107 In order to understand film as a method of historical research, we must take on the practice perspective; it is from this perspective that history is the appropriation of the present. Film is thus faced not with the task of representing the (imaginary) view of the historian on her or his (past) object; film must be used as a means of shaping the practice of appropriation.
Benjamin’s now-time, which hits the present and its past like a bolt of lightning reminds us of the opsigns and sonsigns that, according to Deleuze, transcend the movement-image towards the time-image. Just as the new signs no longer require the movement-image as a representation of the whole but rather are constitutive of a transparent material that they specify themselves,108 the now-time blasts open the continuum of time. The dialectical image maintains the ambivalence between the irrevocability of the past and the index of actuality that historical images carry with them. Similarly, the time-image maintains the ambivalence of the actual image and the virtual image, which become indiscernible without relinquishing their distinctiveness. They are constantly replacing one another and are thus part of a constant practice, one for which the time-image appears to be particularly well suited, of shaping the form of history that is understood as appropriation.
Like the flashing dialectical images, the time-images alternate between actuality and virtuality, rejecting a fixation without becoming arbitrary, they establish a referentiality without fixing on a referent. They are images of practice or, better yet, practices of images that, each in their own way, reject the discourse of consciousness of the modern age.
Benjamin’s theses on the philosophy of history, in which dialectical images play an important role, are still groundbreaking for a critical history that does not affirm the prevailing conditions. Deleuze’s study of cinema, and especially his concept of the time-image, can be used to conceptualize a dialectical image for film in Benjamin’s sense of the term. This gives critical history a new prospect for integrating film not only as a source but also as a “tool” for historical research. It is becoming apparent that film enables a type of historical relationship in “dialectical time-images” that, at the least, allows a different order of historical facts than the usual narrative models of “the writing of history.” Through the practice of images, the historical relationships embedded in archival images, trace-images, and derived images directly confront spectators. The associative power of the images creates dialectical images in the intervals of irrational cuts. These images, in turn, enable a reading of the film and encourage critical thinking, just as Benjamin described it for Brecht’s epic theater:
It basically operates through repeated shocks, as the sharply defined situations of the play collide. [...] This constantly creates intervals which undermine the audience’s illusion; these intervals are reserved for the audience’s critical judgements, its moments of reflection.109
Despite considerable overlap between Benjamin and Deleuze, the appropriation of Deleuze’s concepts for a critical theory of historical study is nevertheless problematic. By reintroducing Deleuze’s concepts into dialectical thinking, such appropriation must systematically “misunderstand” him; an approach Deleuze himself took in dealing with other authors.110 Deleuze’s anti-dialectical attitude, as Jan Rehmann describes, is also based on two basic methodical decisions that result in his criticism missing its target,
namely on the one hand, identifying the dialectic with the speculative form that Hegel especially gave it in his logic; on the other hand, contrasting the speculative principle of negation, opposition, and contradiction with the no less speculative ‘principle’ of distinction.111
Thus, there are good reasons for understanding Deleuze’s movement-image and time-image as a dialectic in Benjamin’s sense of the term. Unlike Hegel’s philosophy, Benjamin’s dialectic does not contain a progression of the spirit’s self-realization from beginning to end. Instead, the dialectical tension unfolds anew in each moment, starting from the brief flash of the dialectical image. The synthesis “is not a movement towards resolution”112 but rather the intersection of the axes of contradictory concepts. The goal is not progress but rather redemption, a revolutionary “tiger’s leap” that would be possible at any moment.
An appropriation of Deleuze’s film concepts has proven to be beneficial for examining the dialectic of the presentation medium and the representation of history in film. The movement-image organizes historical material into a chronological succession. The image is subjected to the narration, which tends to affirm the present in its linearity. The future becomes the mere extension of the past, the result of a succession. In contrast, the time-image achieves a break in the chronological succession, a dialectical image, that is loaded with now-time. In the moment of a fleeting present, it organizes the remnants of the past into a history and directs them toward the hope of creating a better present, or the possibility of doing so. Both the past and the possibility are only virtual; but they are indiscernible from the actuality of the practice to which they are connected.
The qualities of the time-image that Deleuze emphasizes offer new perspectives for a critical history in the sense of a practice of appropriating the present in the form of history. Whether and how such a construction of history can succeed is something that further research, including filmic research, will have to determine. Success does not depend solely on these academic practices but also on the network of other social practices in which they must be integrated. What Kluge says about film, is also true for history: it occurs in the mind of the spectator.
Translated by Jessica Wallace
Suggested Citation: Berg, Olaf: Benjamin and Deleuze: Approaches to a Critical History in Film Images. In: Research in Film and History. Sources – Meaning – Experience (2021-01-28), No. 3, pp. 1–34. DOI: https://doi.org/10.25969/mediarep/15453.
Benjamin, Walter. The Arcades Project. Translated by Howard Eiland and Kevin McLaughlin. Cambridge, Massachusetts, and London: Belknap Press, 1999.
Benjamin, Walter. Selected Writings, Volume 3: 1935–1938. Translated by Edmund Jephcott, Howard Eiland, and Others. Edited by Howard Eiland and Michael W. Jennings. Cambridge, Massachusetts, and London: Belknap Press, 2002.
Benjamin, Walter. Selected Writings, Volume 4: 1938–1940. Translated by Edmund Jephcott and Others, ed. Howard Eiland and Michael W. Jennings. Cambridge, Massachusetts and London: Belknap Press, 2004ff.
Benjamin, Walter. Understanding Brecht. Translated by Anna Bostock. London: Verso, 1998.
Buck-Morss, Susan. The Dialectics of Seeing. Walter Benjamin and the Arcades Project. Cambridge, London: MIT Press, 1989.
Deleuze, Gilles. Cinema 1. The Movement-Image. Translated by Hugh Tomlinson and Barbara Habberjam. Minneapolis: University of Minnesota Press, 1986.
Deleuze, Gilles. Cinema 2. The Time-Image. Translated by Hugh Tomlinson and Robert Galeta. Minneapolis: University of Minnesota Press, 1989.
Deleuze, Gilles. Negotiations. 1972–1990. New York: Columbia University Press, 1995.
Goertz, Hans-Jürgen. Unsichere Geschichte. Zur Theorie historischer Referentialität. Stuttgart: Reclam, 2001.
Iampolski, Mikhail. The Memory of Tiresias. Intertextuality and Film. Berkeley: University of California Press, 1998.
Kirchmann, Kay. Verdichtung, Weltverlust und Zeitdruck. Grundzüge einer Theorie der Interdependenz von Medien, Zeit und Geschwindigkeit im neuzeitlichen Zivilisationsprozeß. Opladen: Leske+Budrich, 1998.
Kracauer, Siegfried. From Caligari to Hitler. A Psychological History of the German Film. Princeton: Princeton University Press, 1947.
Kramer, Sven. “Stillstellung oder Verflüssigung. Schrift-Bild-Konstellationen bei Walter Benjamin und Peter Weiss.” Zeitschrift für kritische Theorie 10, no. 18–19, (2004): 99–115.
Lanzmann, Claude, and Heike Hurst. “Eine befreiende Wirkung. Gespräch mit Claude Lanzmann.” In Shoah. Edited by Claude Lanzmann. Düsseldorf: Claassen, 1986.
Lenger, Hans-Joachim. Fragen an Deleuze. Fragen an Wenders. Das Digitale, Schriftenreihe zur Theorie. Hamburg: material-Verlag, 2003.
Marx, Karl. “The Eighteenth Brumaire of Louis Bonaparte (1852).” In Marx and Engels Collected Works, Volume 11, 99–198. Moscow: Progress, 1979.
Otte, Georg. “Zitieren und Antizipieren. Die Frage der Lesbarkeit der Welt im Passagen-Werk Walter Benjamins.” Zeitschrift für kritische Theorie 10, no. 18–19 (2004): 133–149.
Reitz, Edgar, Alexander Kluge, and Wilfried Reinke. “Wort und Film.” In Ulmer Dramaturgien. Reibungsverluste (Arbeitshefte Film 2/3). Edited by Klaus Eder, Alexander Kluge, 9–27. München: Hanser, 1980.
Rehmann, Jan. Postmoderner Links-Nietzscheanismus. Deleuze & Foucault. Eine Dekonstruktion. Hamburg: Argument Verlag, 2004.
Rodowick, David Norman. Gilles Deleuze's Time Machine. Durham, London: Duke University Press, 1997.
Rosenstone, Robert A. “Does a Filmic Writing of History Exist?” History and Theory 1, no. 41 (2002): 134–144.
Rosenstone, Robert A. “History in Images/History in Words: Reflections on the Possibility of Really Putting History onto Film.” American Historical Review 93, no. 5 (1988): 1173–1185.
Rosenstone, Robert A. Visions of the Past. The Challenge of Film to Our Idea of History. Cambridge, London: Harvard University Press, 1995.
Rother, Rainer. Die Gegenwart der Geschichte. Ein Versuch über Film und zeitgenössische Literatur. Stuttgart: J.B. Metzler, 1990.
Schwartz, Vanessa. “Walter Benjamin for Historians.” American Historical Review 106, no. 5 (2001): 1721–1743.
Schweppenhäuser, Hermann. “Dialektischer Bildbegriff und dialektisches Bild in der Kritischen Theorie.” Zeitschrift für kritische Theorie 9, no. 16–19 (2003): 7–46.
Vogl, Joseph. Kalkül und Leidenschaft. Poetik des ökonomischen Menschen. München: Sequenzia, 2002.
Weiss, Peter. Die Ästhetik des Widerstands, 3 Bände. Berlin: Henschelverlag, 1987.
Wenzel, Eike. Gedächtnisraum Film. Die Arbeit an der deutschen Geschichte in Filmen seit den sechziger Jahren. Stuttgart, Weimar: Metzler, 2000.
White, Hayden. “Historiography and Historiophoty.” American Historical Review 93, no. 5 (1988): 1193–1199.
White, Hayden. Metahistory. The Historical Imagination in Nineteenth-Century Europe. Baltimore: JHU Press, 1973.