Theorizing Fiction in Film Non/Fiction

Some Thoughts on Recent German Film Theory

Mario Slugan
Despite fiction film arguably being the privileged object of film theory the notion of “fiction” has been undertheorised by film scholars in general and those working in German in particular. Perhaps the most important exception to this trend has been Gertrud Koch and Christiane Voss’ (2009) edited volume on fiction on the intersection of philosophy, film, and media studies. This paper tackles two of the most notable film scholarly contributions to the volume – Koch’s and Vinzenz Hediger’s – and their attempts to define fiction in terms of medium properties as well as their efforts to articulate all photographic films as simultaneously fictional and nonfictional. In the first case, I demonstrate that medium under-determines whether something is fictional or not. In the second, I argue that although fiction is a temporally unstable category, it is possible to distinguish between fiction and nonfiction at a given moment in time. I conclude with a call to applying Kendall L. Walton’s (1990) transmedial theory of fiction to film, by listing a number of its advantages over competing proposals and by emphasising its suitability for investigating the change in films’ fictional status over time.
Kendall L. Walton; Gertrud Koch; Vinzenz Hediger; Christian Metz; André Bazin; theory of fiction; fiction as make-believe; fiction film; documentary


Aesthetic Fiction as the Virtual

The Ontological Indeterminacy of Film

Walton’s Theory of Fiction as Mandated Imagining





Suggested Citation


There is undoubtedly a rich tradition of theorising fiction in German-language scholarship. From a philosophical perspective, Hans Vaihinger (1911) has proposed that theoretical models used to describe the world are essentially fictions – we behave as if they fit reality because we have no direct access to reality. Focusing on literary texts, Käte Hamburger (1957) has argued that there are textual markers which define whether something is fiction or not. Wolfgang Iser (1991) has put forth a triad of the fictive, the imaginary, and the real in his pioneering contribution to literary anthropology. Perhaps the clearest sign of the current dynamism in the field is the recent 540-page handbook covering theory, psychology, history, and interdisciplinary aspects of fiction (Tobias Klauk and Tilmann Köppe 2014).

When it comes to aesthetic fictions, most of the work, however, has focused on literary fiction. Even Klauk and Köppe’s reference book – subtitled Ein interdisziplinäres Handbuch (An Interdisciplinary Handbook) – takes the bulk of its examples from literature with film and other arts receiving only scant attention. There is, undeniably, a growing interest in aesthetic fictions beyond literature as evinced by Anne Enderwitz and Irina O. Rajewsky’s 2016 Fiktion im Vergleich der Künste und Medien. But if one is to look for theoretical accounts of fiction in film, even here we find research that is interested more in interpretation rather than theory proper.

This is not specific to German-language scholarship but to film studies in general and film theory in particular. The dearth of monograph-length studies theorizing fiction is even more striking once we consider the fact that fiction film has been and remains the privileged object of film theory. For instance, already in 1916 one of the earliest theorists – Hugo Münsterberg – chose to focus on the film’s dramatic fiction form in his aptly titled study, Photoplay. Although devoting a few pages to experimental film, the focus of both of Béla Balázs (1982, 1984) monographs originally published during the Weimar era, similarly, remains fiction film. The same can be said of Rudolf Arnheim’s work (1933). Victor Perkins (1974) is explicit that his Film as Film is interested in “photographic fiction film”. David Bordwell (1985) also writes about Narration in the Fiction Film. Stanley Cavell, although defining film as “a succession of world projections” (1979: 72) which clearly allows for non-fiction, has virtually nothing to say about it. Over the years there have undeniably been other voices like Bill Nichols (1991) and Annette Michelson (2017) who focus on documentary and experimental film, respectively, and Siegfried Kracauer (1960) who devotes non-negligible portions of his theory to both. But the fact that no writings on non-fiction film are included in Leo Braudy and Marshall Cohen’s widely used Film Theory and Criticism: Introductory Readings (2016) is perhaps the best sign of the relative prevalence of theoretical focus on fiction.

Fiction as a concept even shapes the recent surge of interest in the theory of documentary film, for the latter is primarily understood as non-fictional. Frank Kessler (1998), for instance, titles his piece as an opposition between two concepts – “Fakt oder Fiktion?” (“Fact or Fiction?”). Nichols, a pioneer of documentary studies, frames his definition of the genre along the same lines: “[documentaries] tell stories that, although similar to feature fiction, remain distinct from it” (2017: 4). Another notable scholar writes: “Although the distinction between nonfiction film and documentary cannot bear much theoretical weight, it might be useful to think of the documentary as a subset of nonfiction films, characterised by more aesthetic, social, rhetorical, and/or political ambition than, say, a corporate or instructional film” (Carl Plantinga 2005: 105). However, for all the organising power that the concept of fiction has within the discipline, “fiction” in the notions of “fiction film” and “nonfiction film” rarely deserves a sustained treatment by film scholars. A symptom of this is the fact that an entry for “fiction” cannot be found in either of the two widely read conceptual encyclopaedias of film studies – Susan Hayward’s 2006 Cinema Studies: The Key Concepts and Edward Branigan and Warren Buckland’s 2014 The Routledge Encyclopaedia of Film Theory. In the German-language context, similarly, a chapter devoted to fiction is missing from Bernhard Groß and Thomas Morsch’ 2019 Handbuch Filmtheorie (The Handbook of Film Theory) as well. It is as though fiction is a given, something film scholars need not worry about. Undeniably, some concepts stand as preconditions of the discipline but even the concept of “film” – admittedly absent in the cited volumes as well – has deserved more thorough engagement than fiction over the years (cf. André Bazin 2005; Noël Carroll 2008).

The most notable exception to this trend, at least in German-language scholarship, is Gertrud Koch and Christiane Voss’ 2009 edited volume “Es ist, als ob”: Fiktionalität in Philosophie, Film- und Medienwissenschaft (“It Is As If”: Fiction in Philosophy, Film and Media Studies) which, although a decade old now, still presents the most relevant book-length discussion of fiction in film theory in recent years. In it a number of proposals are floated on how to grasp film non/fiction, but two of the recurring themes are that the film’s medium properties are crucial for construing film’s non/fictional status and that a given film is fictional and nonfictional at the same time. In what follows I wish to address the contributions by the volume’s most notable film scholars – Gertrud Koch and Vinzenz Hediger – who espouse these theses. In both cases I will argue that the medium under-determines a film’s fictional status but that this does not mean that it is impossible to distinguish between fiction and nonfiction films. Building on Hediger’s perceptive analysis of the historicity of the fiction/nonfiction distinction in film history, I will conclude with a proposal for what constitutes fiction which can accommodate this temporal instability of fiction. Crucially, I will argue that although acquainted with Kendall L. Walton’s work on fiction, contributors to Koch and Voss’ volume miss the opportunity to apply his transmedial theory of fiction to film.1

Aesthetic Fiction as the Virtual

Koch usefully sets the stage for us by posing the key question:

Müssen Objekte als ganze fiktiv sein oder können sie dies nur in bestimmten Hinsichten sein, und wenn Letzteres zutrifft, betrifft es dann nur die Hinsichten oder sind am Objekt ausweisbare Eigenschaften Träger der Unterscheidung von fiktiv/nichtfiktiv? (2009: 140)2

She starts off by distinguishing aesthetic from non-aesthetic forms of fiction. The latter include phenomena such as scientific or legal fictions whereas the former pertain to aesthetic objects such as film and literary works. According to Koch, non-aesthetic fictions such as private property or quarks constitute hypothetical statements about what is possible. One can possess something without actually having it on one’s person or even being in its vicinity, while holding the object remains a distinct possibility. Aesthetic fictions, by contrast, do not afford the possibility of such an engagement. Although we might possess a copy of Alfred Hitchcock’s The Birds (1963, USA), it is impossible either to possess or enter the fictional world represented therein. Even if we were to have visited Bodega Bay in 1963 when and where the events of the film take place, we would never have been able to meet the film’s protagonists and be a part of their world.

Koch’s next step is to propose an account of aesthetic fiction as that which prepares for the “virtual” in C. S. Peirce’s sense of the word. According to Peirce’s definition, there is a clear distinction between the virtual and the potential:

(1) A virtual X (where X is a common noun) is something, not an X, which has the efficiency (virtus) of an X.

This is the proper meaning of the word; but (2) it has been seriously confounded with “potential,” which is almost its contrary. For the potential X is of the nature of X, but is without actual efficiency […] the sun was said to be virtualiter on earth, that is, in its efficiency (C. S. Peirce 1902: 763-764, italics in the original).

This allows Koch to erase the borders between fiction film and documentary and to treat not only fiction as documentary but documentary as fiction as well. In the former case, Koch finds that because there is always something in front of the camera even in fiction film, such film is also a documentary of whatever was in front of the camera – actors, costumes, locations, etc. In the latter case, because documentaries screen images of what has already passed it means that reality always appears as fiction.3 Let us consider the (de)merits of her argument.

Writing in the tradition of documentary theory which has in the wake of Hayden White’s (1978, 1987) work on historical discourse blurred the distinction between fiction and nonfiction, Koch points out that documentaries use the same range of film techniques as fiction films do – camera movement, editing, lighting, use of off-screen space, close-ups, etc. She goes further, however, and proposes two additional reasons why documentaries constitute fictions. The first reason why “das Wirkliche im Film immer als Fiktion auftaucht” (“the real in film always emerges as fiction”) (Koch 2009: 145) is because all the people and objects caught on camera that are screened are no longer where they were when they were recorded. Such objects can only be captured under the Peircean notion of the virtual. Second, fiction also derives from the fact that such worlds are now temporally enclosed and unalterable, much like the aforementioned world of The Birds.

When it comes to White’s ideas, it has been fashionable to claim that because nonfiction uses the same techniques as fiction – e.g. plotting and closure from literary fiction in the case of history and montage, staging, and other devices from film fiction in the case of documentary – then nonfiction must be no different than fiction. But it is well known that stylistic techniques, i.e. textual features cannot determine whether something is fiction or not, precisely because nonfiction can use techniques usually associated with fiction as much as fiction can deploy devices standardly connected to nonfiction. The Blair Witch Project (Daniel Myrick and Eduardo Sánchez, 1999) famously emulates documentary aesthetic but neither it nor any of the members from found footage horror subgenre it has spurred are nonfictional.4 Similarly, if I express how I got lost yesterday in (poor) verse – “I walked about yesterday, I didn’t expect to lose my way” – this does not make my account fictional.

To continue with Koch’s own argument, documentaries undoubtedly present us with images of temporally removed objects. And given Peirce’s example of the sun being virtually on earth in its efficiency, we can easily think about photographed objects as being virtually present in the images. Although strictly speaking not present in the film, the objects’ effects – images – are there for all able-bodied to see. But this is nothing specific to film. Contrary to Koch’s ideas on the subject that “[Musik] nicht im selben Sinne Wiederholung von Vergangenem in der Gegenwart der Aufführung ist wie der Film” (“music is not the repetition of the past in the present of performance in the same sense than film”), recorded music is no different in this regard (ibid.). Like the recorded image, the recorded sound is not the same as the original object being recorded.5 The sound wave that is the original object is only the automatic cause of the recording irrespective of the recording’s medium. It is not the actual vibrations of the air that are stored but only their effects. In other words, if documentaries are fictional due to the virtual character of the images that make them up so is recorded music.

Regarding the inaccessibility of the represented world it is true that we cannot access and/or change the events depicted. But this is not the same as being unable to access the world of The Birds. Undeniably, I cannot talk to the protagonist of Werner Herzog’s Grizzly Man (2005, USA) who has since died but this is not the same as not being able to converse with Melanie Daniels from The Birds. If I visited the Katmai National Park and Preserve in Alaska twenty years ago, I would have been able to engage the bear enthusiast Timothy Treadwell in conversation. By contrast, even if I spoke to Tippi Hedren on the set of The Birds in Bodega Bay I would not have been able to speak to her fictional character. The reason is that fictions, unlike past events, have no spatiotemporal relations to us.

The main problem with Koch’s understanding of recorded images as fictional is that it arbitrarily expands the notion of fiction without giving any reason why we should relate fiction to the virtual in the first place. Theoretical explanations of notions such as fiction should either propose a definition which is (approximately) coextensive with the ordinary meaning of the word or should, in the case they offer a technical definition, explain why this new understanding of fiction is preferable to its ordinary meaning. By defining aesthetic fiction in terms of the virtual, Koch does the latter but without providing any good reason to do so. It is true that Koch sees Peirce’s distinction between the potential and the virtual as fitting nicely with her differentiation between non-aesthetic and aesthetic fiction, but that alone is not a good enough reason for such a radical overhaul of the notion of fiction.

Moreover, the potential and the virtual do not do as good of a job in distinguishing non-aesthetics from aesthetic fiction as she would like in the first place. As such, they further undermine her attempt to introduce a novel account of fiction. Her explanation of property as an example of legal, i.e. non-aesthetic fiction in terms of possibility is misleading.6 It is, indeed, a distinct possibility that I can, for example, wear a watch I own and that even when I do not wear it, I am still its legal owner. But that does not apply to all forms of property. Intellectual property regularly involves subject matter which is not material. For instance, if I were to own the rights to the Spiderman franchise, I would not be able to somehow have the Spiderman franchise on my person. In other words, property is not a matter of having a possibility or potentially holding onto something physically.

Furthermore, although Koch is correct to point out that there are non-aesthetic fictions such as thought experiments, it does not seem that her example of property fits the bill to begin with. As an owner of a watch, it is not that I behave as if I own the watch in the sense that Conrad Veidt behaves as if he was the somnambulist on the set of Robert Wiene’s Das Cabinet des Dr. Caligari / The Cabinet of Dr. Caligari (1920, Germany). I do own the watch. If somebody steals my watch I do not behave as if he stole it. I do not pretend to call the police, the police do not only go through the motions of arresting the perpetrator, the judge does not pass an as-though sentence, the convict is not make-believing serving time. In the case of property, there is a set of societal practices governing behaviour around property with potentially harsh consequences. When Conrad Veidt performs an “as if” murder in Wiene’s film, there are no real-life consequences.

What is more, there are theoretical accounts which, contrary to Koch, subsume all fictions – aesthetic and non-aesthetic alike – under the notion of possibility. In the footsteps of David Lewis (1978) and Saul Kripke (1980), scholars such as Thomas Pavel (1986) and Lubomír Doležel (1998) have proposed that fictional worlds are a sub-class of possible worlds. The problem with such accounts is that fictional worlds can include a priori impossible elements – such as a different value for pi – whereas possible worlds, by definition, cannot. But although possible world theories fail as a general account of fiction, they do demonstrate that both some aesthetic and some non-aesthetic fictions can be treated in terms of possibility, making Koch’s attempt to distinguish between the two in the same terms moot.

When it comes to the option of treating fiction as a documentary, Koch is certainly not the first to point out this possibility. At least as early as 1947 André Bazin (2008) has proposed that all films are social documentaries, i.e. that they provide access to the collective dreams of the society who made them. For Koch the sense in which films are documentaries is different – it is not because they act as symbols for some hidden social reality but because they are recordings of people, objects, locations, etc. that actually existed. She perceptively points out that the digital revolution has not changed the documentary nature of such recordings because even if the way in which images are stored is different, the apparatus for taking images essentially remains the same – light coming from the object still stands at the beginning of the causal chain that produces the image (Koch 2009: 146). In other words, both digital and analogue recordings are documents of whatever was in front of the camera. Although in this section Koch limits her observations about fiction films to what is known in German as “Spielfilm” – roughly, films with live actors – there are still substantial problems with her account.7

First, Koch is explicit that when talking about animation films – regardless of whether they are hand-drawn or computer-generated – we are necessarily speaking of fiction films. This, however, is belied by a number of films including perhaps most notably Vals im Bashir / Waltz with Bashir (Ari Folman, 2008, Israel) which, despite being animated, constitute documentaries. In fact, examples of what now amounts to a veritable genre of animated documentary are nothing new – they can be tracked at least to The Sinking of the Lusitania (Winsor McCay, 1918, USA), most of which consists of hand-drawn representations of the titular atrocity. Put differently, Koch’s account of animation solely in terms of fiction is too conservative.

Second, Koch’s insistence that animation is substantially distinct from live-action recordings when it comes to the documentary nature of the image, implies that for Koch the documentary quality of the image hinges on the automatic causal relationship between the object and its photograph. The above examples demonstrate that documentary status does not hinge on any such relationship. In other words, the properties of the recording cannot determine whether something is fiction or not.

Third, Koch is perfectly correct when she points out that, so long as there was something in front of the camera, the automatic causal relation between the object and its image holds irrespective of whether the camera is digital or analogue. But by the same token, the same holds for hand-drawn animated films as well including experimental animations such as Walter Ruttmann’s Lichtspiel: Opus I (1921, Germany). Given that all the frames in this film were produced by photographing hand-made images, there is no reason why, following Koch’s logic, we could not regard such films as documentaries of Ruttmann’s drawings and/or paintings. Put in yet another way, Koch’s account of documentary is internally inconsistent.

Fourth, a further insight is to be gained if we distinguish between nonfiction and documentary which in Koch remains conflated. As Plantinga’s citation on the preceding pages reminds us, documentary is only a sub-class of nonfiction. Experimental films such as Peter Kubelka’s Arnulf Rainer (1960, Austria) are not fictions but are not documentaries either. Furthermore, given that Kubelka’s film consists of an alternation of transparent and black filmstock, on the one hand, and of no sound and white noise, on the other, we cannot even say that anything was in front of the camera in the sense there was in Lichtspiel: Opus 1. In other words, Koch is too quick to discuss the documentary as the only relevant non-fiction film genre.

Fifth, it has been argued that there is more to documentary than pure recording. In the case of early cinema, for instance, building on the writings of John Grierson, Tom Gunning (1997) has argued that whereas documentaries proper creatively engage with recorded material, nonfiction films capitalizing on “view” aesthetics are primarily concerned with the mimesis of observation. In other words, although the border between the two is by no means strict, just because unadulterated photographic images carry veridical information about their objects, a temporal sequence of such images does not mean that they constitute a full-fledged documentary film. In other words, we should distinguish between documentary as a film genre and the documentary quality as conveying veridical information secured automatically in the reproduction process, something that appears to be conflated in Koch.8

Sixth, it is true that we can treat recordings which make up fictional films as documentary material of whatever was in front of the camera. For instance, instead of appreciating the fictional story in Das Cabinet des Dr. Caligari I can focus on the film’s images as a primary source for learning about the expressionist acting style. But this does not deny the fact that the default mode of engagement with the film is one of an “as if” attitude towards the people and events represented therein. In other words, distinguishing between fiction and documentary is not the question of whether we can treat something as a fiction or documentary, but rather why by default we do treat some films as fiction and other as documentaries. That is; why, when talking about Das Cabinet des Dr. Caligari, we regularly say that, among other things, we see the fiction of Cesare kidnapping Jane rather than a documentary about Conrad Veidt pretending to be a somnambulist going through the motions of taking Lil Dagover pretending to be in distress?

The Ontological Indeterminacy of Film

In his contribution to the volume Vinzenz Hediger is even more ardent than Koch in his attempt to argue for the ontological indeterminacy of film based on the properties of the medium. Building on Christian Metz (1982), on the one hand, and André Bazin (2005), on the other, he argues that as a medium film is fictional and nonfictional at the same time. Contrary to what Hediger argues, however, the medium underdetermines the ontological status of film.

Metz has famously claimed that “[e]very film is a fiction film” (1982: 44) because of the structure of signification in cinema. For Metz, film, unlike other media, is characterised by an imaginary signifier. It is imaginary because it is present and absent at the same time. It is present as a perceptually rich image, yet it is absent because it is not the object itself that is on the reel. Contrary to objects or actors in the theatre who although standing for something else are present on the stage, the same objects and actors on screen are present only as their images. This duality of presence and absence makes representation fictional.

This duality, however, is not specific to film but is typical of representation in general. Representations, by definition, stand for something else, something that is absent. Arguably, Metz speaks of a specific type of representations which single out specific objects rather than concepts but even then, the cinematic signifier is not the only to denote in the absence of its reference. In linguistic representation, for instance, proper names stand for something apart from themselves. The signifier – the sounds making up the spoken name – are undeniably present whereas the signified can be easily absent if we are talking about a person who is not there. By Metz’s logic all representations that denote a specific object would be fictional – including all the references to actual people in this paper and to Metz in this section. But that is patently false. Moreover, there are also signifiers that, unlike linguistic signs, resemble their signifieds as is the case in figurative drawings and paintings. Like in film, their image is immediately present to the viewer although what they denote is typically absent. Albrecht Dürer’s self-portrait denotes him in his absence, but this does not make the painting fictional.

If the point is that Metz is interested only in signifiers which arise through causally automatic processes, even then the cinematic signifier is not unique because, as I have demonstrated in the previous section, audio recordings have the same structure of signification as visual recordings do. Much like Koch, Hediger believes there is a distinction between these two types of recordings:

Der Unterschied zur Tonaufnahme wird gerne daran festgemacht, das sein Klang, den wir hören, ob er nun aufgezeichnet ist oder ‘live’ gespielt wird, das Objekt selbst ist, wohingegen ein Fotogramm nur ein Bild oder ein Zeichen ist, das für das Objekt steht (2009: 182).9

But to repeat, the sound recording is a distinct object from the original sound. The original vibrations in the air have long since dissipated. What is on the record are merely causal effects of the original sound. That much becomes obvious once we listen to a low-quality recording. All the scratches and noises highlight the fact that what we are dealing here with is a copy which does not share the ontological identity with its source. After all, if something is a copy then it is certainly ontologically distinct from its source. Precisely like in film, the sound that we are playing is just a signifier of the sound that was originally recorded. To my mind, the most likely reason why both Hediger and Koch conflate the two, is because with our current sound recording technology the reproduced sound and the original sound are phenomenologically virtually indistinguishable. So long as we are positioned in space and specific calibrations are made, we can easily confuse the reproduced sound for its live counterpart and think the two are one and the same (not only phenomenologically but ontologically as well). This is also why we could theoretically confuse a hologram of an object for that object. But when it comes to visual recordings, we would hardly ever confuse a photographic image for its object due to the fact such recordings are framed and two-dimensional.

While subscribing to Metz’s view on the alleged uniqueness of the cinematic signifier to argue for film’s inherent fictional status, Hediger also wants to argue that film is at the same time inherently nonfictional. He does so via Bazin and his claim that the object and its photograph “share a common being” (2005: 15). Hediger rightly points out that Bazin’s photograph is more than what has in the wake of Peter Wollen’s (1969) account of Bazin been referred to as an index. It is not only that there is an automatic causal connection between the photograph and its object, i.e. that the photograph is the object’s trace. Rather, “it [the photograph] is the model [the object]” (ibid.: 14). This transfer of reality from the object onto the photograph is what makes “every film a nonfiction film, insofar the film image is both present and real, i.e. insofar it shares in the being of what it represents” (“jeder Film ein nichtfiktionaler Film, insofern das filmische Bild zugleich vorhanden ist und […] real ist, also am Sein des Dargestellten teilhat.” Hediger 2009: 177).

Now, Hediger is explicit that this does not make every film a documentary of whatever was in front of the camera (ibid.), but just that film as a medium should be understood as nonfiction (a notion that also encompasses the signifier’s documentary quality outlined above). But if that is the case, then all representations are also nonfictions as a medium. For all representations are made of “present and real” entities – phonemes in speech, graphemes in writing, specks of color in painting, traces of carbon in drawing, solid objects in sculpture, bodies in movement in dance, etc. – i.e., all representations have a material reality. Ink on the pages of Kafka’s Der Prozess, for instance, is nonfictional in its physical existence (and tells us, among other things, about the materials and fonts used by a given society at the time of printing) yet it conjures up the fiction of Josef K. Pigments on Sandro Botticelli’s Nascita di Venere / The Birth of Venus are no less fictional in their materiality (and elucidating of paint available to renaissance painters), but they still invite us to make-believe that Venus was born from the sea.

It is true that none of these representations “share a common being” with what they represent – there is no such link between the ink spots spelling out “Josef K.” and Josef K. or between the pigments making up the image of Venus and Venus. But neither do representations in fiction film. As a fiction film, Das Cabinet des Dr. Caligari does not represent Conrad Veidt but Cesare. And whatever we might want to say about the relationship between the film image and Veidt, this relationship does not obtain between the image and Cesare. In other words, the focus on the profilmic is misplaced when it comes to criteria for determining whether something is fiction or not.

Put differently, because fictional status changes the relevant referent (from profilmic to the fictional) the appeal to the medium cannot get us far in determining whether something is fiction. The recourse to medium properties fails to establish anything about whether a representation instantiated in that medium is fiction or nonfiction because the sign and its fictional referent cannot share a common being in the first place. In photographic cinema, more specifically, no approach based on medium properties can establish anything about the non/fictional status of a given representation, because it can only tell us about the presence or absence of the indexical/ontological link between the recording and its profilmic referent. Or to put it in yet another way, recordings cannot record fictions, they can only record actual objects whose images can then be used to signify fictions.

The lesson then is that the medium properties that Koch and Hediger (and Metz and Bazin) focus on cannot help us determine whether something is fiction or not a priori. Much like medium specific theories have failed to tell us what cinema is or to provide us with valid theories of evaluation (cf. Carroll 2008), they also fall short of establishing film’s non/fictional status. To discuss fiction and nonfiction on the level of the medium, moreover, only demonstrates that the structures of signification in film as opposed to other media, recording or otherwise, are not as specific as one would think. What needs to be solved is why photographic films signify fictions more often than not despite being based on recordings of actual objects. In other words, given that fiction is a transmedial phenomenon we first need to ascertain what makes something fiction or not, and only then investigate the media specificities of fictions.

Walton’s Theory of Fiction as Mandated Imagining

Where Hediger is most insightful is in demonstrating that, historically speaking, categories such as “Spielfilm” and “documentary” did not exist during the cinema’s early era up to c. 1915. Given that the main point of his article is to dissuade us from the idea that fiction and nonfiction was always firmly distinguished or at least that it was differentiated in the way in that we do today, it is an important point worth emphasising. In fact, Hediger’s explanation that D. W. Griffith saw his The Birth of a Nation (1915, USA) primarily as an objective representation of the American Civil War and the Reconstruction, i.e., as a sort of a visual correlate to written history, also does away with the theories that non/fictional status can be defined in terms of authorial intentions (Currie 1990; Carroll 1997; Plantinga 2005). I can add that promotional material attests that even films like Un homme de têtes / Four Troublesome Heads (Georges Méliès, 1898, France), which we nowadays treat as fictions, were intended, to use an anachronistic term, as “documentational” recordings of stage tricks by their producers:

One of the most marvelous tricks ever cinematographed. The magician approaches and, after the usual bow, proceeds with the tricks of taking off his head, placing same on a table at his side. […] to show that there is no illusion about the trick, he crawls under the table, upon which is supported his first head. […] [Upon completing the trick] the magician then makes his bow and retreats from the scene. A most surprising and marvelous illusion (“Star” Films Catalogue 1903: 17).

It is not only that the film is explicitly advertised as being “cinematographed”, but the ad also goes to great lengths to demonstrate how the magic is done on stage by emphasising the opening and closing bowing the audience and crawling under the table.

If intentionalist theories were correct, then we would have to treat both The Birth of a Nation and numerous other films like Un homme de têtes as nonfiction rather than fiction. In what follows I would like to sketch out an existing alternative and why this alternative also allows us a better grasp of film history.

The theory which allows for the temporal instability of nonfiction/fiction divide while at the same time allowing us to distinguish between the two at any point in time, is Walton’s account laid out in his seminal Mimesis as Make-Believe. Now the dominant view among analytic philosophers of art (cf. Kathleen Stock 2016; Gregory Currie and Anna Ichino 2016), sees fiction as mandated imagining.

Walton models fictions on games of make-believe children play with their toys, dolls, or found objects as props for generating certain imaginings. While playing in the woods, for instance, children can agree that when encountering a stump, they should make-believe a bear in its place. In this game stumps are props. The specific imaginings – imagining a bear where stumps are – generate fictional truths. The collection of such fictional truths – that there are as many bears as there are stump – amounts to the fictional world. Representational works, on this view, are props which mandate us to imagine specific imaginings. The main difference is that whereas in the above example the mandate to generate fictional truths is explicitly agreed upon by participating children, in the case of art mandates are implicit.

That make-believe is mandated also means that although one is always free to imagine anything or not to imagine anything at all, a work is fictional only so long as one is supposed to make-believe something specific while regarding the work. As such, mandate is a normative category and not a guarantee of actual reception. When watching Das Cabinet des Dr. Caligari we are supposed to imagine that the protagonist telling the story is institutionalised. We are free not to do so and imagine that he is sane but that would not be playing the game of make-believe appropriately. Crucially, because, as we have seen above, the mandate may change over time it cannot hinge either on textual features and/or authorial intentions. Rather, it arises from a negotiation of institutional factors including production, promotion, exhibition and reception.10 In one such case, whereas present-day audiences treat Méliès’ trick films as fictions, c. 1900 even a film like La lune à un mètre / The Astronomer’s Dream, or the Man in the Moon (Méliès, 1898, France) was both advertised and received by the audiences as a recording of stage magic, i.e. as “a life motion picture reproduction of a celebrated French spectacular piece” (Philadelphia Inquirer, 4 September 1899: 10). In other words, whereas presently the film is seen as authorising a game of make-believe – i.e., as a fiction –at the turn of the last century it was not – i.e., it constituted nonfiction.

What is the other component in Walton’s definition? Imagining is an as-if stance which may include actions (imagining running), objects (imagining x), experiences (imagining seeing), propositions (imagining that p), etc. As such, imagining may but need not include mental imagery. Unlike perception and memory, imagining does not imply the existence of its objects. Unlike perception, the objects of imagining are voluntary. Unlike belief (connoted by notions of “willing suspension of disbelief” or “diegetic illusion” prevalent in film studies), lastly, imagining is neither true nor false, but “mandated” or not. More positively, then, imagining is a generally controllable off-line simulation mostly decoupled from behavioural reactions.

There are a number of advantages to Walton’s theory when compared to alternatives. First, unlike Koch’s and Hediger’s proposals, Walton clearly specifies the necessary condition for what it means for something to be fiction.11 Both the notion of the virtual and the play of absence and presence as defining traits of fiction suffer from numerous counterexamples.12 Second, Walton allows for both aesthetic and non-aesthetic fictions. But contrary to Koch he easily excludes examples such as the notion of property from the latter. Third, Walton’s definition is not medium specific as Hediger’s is. As such it is not hampered by problematic accounts of what the consequences the indexical/ontological bond between the signifier and the signified has for the non/fictional status of representation. Fourth, because Walton does not privilege any medium of fiction, he can easily accommodate fictions in which no assertions are made (e.g. painting, sculpture, silent cinema) unlike theories modelled on literature and speech acts (John Searle 1975; Currie 1990; Peter Lamarque and Stein Haugom Olsen 1994; David Davies 2007). Fictions are generated by the help of props which need not only be assertions in a story, but can also be images in films, colours in painting, sounds in music, etc. Fifth, Walton’s proposal easily accommodates impossible worlds in which pi has a different value and as such is superior to possible world theories of fiction. It is perfectly easy to imagine a fiction in which the value of pi is, say, 4. Sixth, Walton identifies an activity which precedes other fiction-making practices like story-telling both phylogenetically and ontogenetically as a model for fiction – children’s games of make-believe. They easily use all kinds of props such as toys towards which they espouse an as if attitude of, for example, having a tea-party with their furry friends. And last, because mandates are defined neither by intentions nor by textual features, Walton’s theory explains why it is possible that fictional status may change over time. This is not only of importance for aesthetics fictions but for non-aesthetic ones as well given that cultural texts such as mythologies of various peoples (Greek, Egyptian, etc.) have migrated from non-fiction to fiction over the years.13 In other words, although both Koch and Hediger (as well as numerous other contributors to the volume) briefly refer to Walton, they do not engage what to this day remains the most robust theory of fiction. This is an opportunity which German-language film scholarship has hitherto missed but which, as this article has hopefully demonstrated, is worth pursuing.

Mario Slugan

Ghent University


This project has received funding from the European Union’s Horizon 2020 research and innovation programme programme under the Marie Skłodowska-Curie grant agreement No 746619.


1 While deserving little attention among film scholars, Walton’s work has been influential in German-language scholarship on literature with Frank Zipfel (2001) and J. Alexander Bareis (2014) as his most notable proponents. Bareis (2016) has also recently argued for applying Walton’s theory to a range of arts including film.

2 “Do objects need to be fictitious as a whole, or can they be such only in certain respects and, if the latter is the case, does it concern only these aspects or do the objects’ identifiable properties determine the distinction between fiction and nonfiction?” Translations mine unless noted otherwise.

3 Here, “screening” encompasses both traditional projection and more contemporary screen technologies (smartphones, tablets, etc.)

4 For an overview White’s influence on the discussions of documentary see Carroll (2003).

5 For an argument that some photographic images are partially identical with their objects see Mario Slugan (2017).

6 It should be noted that Koch is not discussing “legal fiction” as defined by legal professionals. There it is understood as an assumed fact that helps courts reach a decision. For instance, in matters of inheritance, when two people die simultaneously the courts in England and Wales assume that the older of the two died first. Cf. Hugh Chisholm (1910).

7 For an equally problematic discussion of documentary see Koch (2016).

8 Although Koch is clearly indebted to Bazin (2005) in her emphasis on the profilmic in fiction film, it seems that for her one can speak of documentary only if the image resembles the object. Bazin, however, has famously claimed that resemblance is irrelevant to the nature of ontology.

9 “The difference to the sound recording is based on the fact that the sound that we hear, whether it is recorded or played ‘live’, is the object itself, whereas a photogram is just an image or a pure sign representing the image.”

10 Bareis (2016) speaks of paratextual information.

11 In his original formulation (Walton 1990) the condition is both necessary and sufficient. In his most recent account, however, he states it is only necessary (Walton 2015).

12 This is not to say that Walton has no detractors (cf. Stacie Friend 2012; Derek Matravers 2014) or that Walton himself has not changed his position (Walton 2015) but rather that Koch and Hediger have failed to address a prominent theory.

13 One disadvantage of Walton’s view which, ironically enough, should be an advantage for those like Koch and Hediger who are interested in doing away with the distinction between fiction and non-fiction is that Walton thinks that all photographic images, and by extension all photographic films, are fictions insofar they mandated imagining whatever they depict. This is an aspect of the theory can be criticised from many fronts (cf. Mario Slugan 2019).


Mario Slugan is a Marie Skłodowska-Curie Actions Fellow at the Department of Communication Studies, Ghent University, working on a European Union's Horizon 2020 funded project “Fiction, Imagination, and Early Cinema”. He has authored three monographs – Montage as Perceptual Experience: Berlin Alexanderplatz from Döblin to Fassbinder (Camden House, 2017), Noël Carroll on Film: A Philosophy of Art and Popular Culture (Bloomsbury, 2019), and Fiction and Imagination in Early Cinema: A Philosophical Approach to Film History (Bloomsbury, 2019, forthcoming).


Arnheim, Rudolf. 1932. Film als Kunst. Berlin.

Bazin, André. 2004. “The Ontology of the Photographic Image.” In What is Cinema? Vol. 1, edited by Hugh Gray, 9-16. Berkeley.

Bazin, André. 2008. “Every Film is a Social Documentary.” Film Comment 44 (6): 40-41.

Balázs, Béla. 1982. Schriften zum Film. Band 1. Edited by Helmut H. Diederichs, Wolfgang Gersch, and Madga Nagy. Munich.

Balázs, Béla. 1984. Schriften zum Film. Band 2. Edited by Helmut H. Diederichs, Wolfgang Gersch, and Madga Nagy. Munich.

Bareis, J. Alexander. 2014. “Fiktionen als Make-Believe.” In Fiktionalität: Ein interdisziplinäres Handbuch, edited by Tobias Klauk and Tilmann Köppe, 50-67. Berlin.

Bareis, J. Alexander. 2016. “Randbereiche und Grenzüberschreitungen.” In Fiktion im Vergleich der Künste und Medien, edited by Anne Enderwitz and Irina O. Rajewsky, 45-62. Berlin.

Bordwell, David. 1985. Narration in the Fiction Film. Madison.

Branigan, Edward, and Buckland, Warren (eds). 2014. The Routledge Encyclopedia of Film Theory. London.

Braudy, Leo, and Cohen, Marshall (eds). 2016. Film Theory and Criticism: Introductory Readings. 8th Ed. Oxford.

Carroll, Noël. 1997. “Fiction, Non-fiction, and the Film of Presumptive Assertion: A Conceptual Analysis.” In Film Theory and Philosophy, edited by Richard Allen and Murray Smith, 173-202. Oxford.

Carroll, Noël. 2003. “Postmodernist Skepticism and the Nonfiction Film.” In Engaging the Moving Image, 165–192. New Haven.

Carroll, Noël. 2008. The Philosophy of Motion Pictures. Malden.

Cavell, Stanley. 1979. The World Viewed: Reflections on the Ontology of Film. Harvard.

Chisholm, Hugh, ed. 1911. “Fictions.” Encyclopædia Britannica. 11th Ed. Vol. 10. Cambridge, 319–320.

Currie, Gregory. 1990. The Nature of Fiction. Cambridge.

Currie, Gregory and Ichino, Anna. 2016. “Imagination and Make-Believe.” In The Routledge Companion to Aesthetics. 3rd Edition, edited by Berys Gaut and Dominic McIver Lopes, 320-329. New York.

Davies, David. 2007. Aesthetics and Literature. Basingstoke.

Doložel, Lubomír. 1998. “Possible Worlds of Fiction and History.” New Literary History 29 (4): 785-809.

Enderwitz, Anne and Rajewsky, Irina O. (eds.) 2016. Fiktion im Vergleich der Künste und Medien. Berlin.

Friend, Stacie. 2012. “Fiction as a Genre.” Proceedings of the Aristotelian Society 112 (2): 179-209.

Groß, Bernhard, and Morsch, Thomas (eds.). 2019. Handbuch Filmtheorie. Frankfurt [forthcoming].

Gunning, Tom. 1997. “Before Documentary: Early Nonfiction Films and the ‘View’ Aesthetic.” In Uncharted Territory: Essays on Early Nonfiction Film, edited by Daan Hertogs and Nico de Klerk, 9–24. Amsterdam.

Hayward, Susan. 2006. Cinema Studies: The Key Concepts. 3rd Ed. Abingdon.

Hämburger, Kate. 1957. Die Logik der Dichtung. Stuttgart.

Hediger, Vinzenz. 2009. “Von Überhandnehmen der Fiktion: Über die ontologische Unterbestimmtheit filmischer Darstellung.” In “Es ist, als ob”: Fiktionalität in Philosophie, Film- und Medienwissenschaft, edited by Gertrud Koch and Christiane Voss, 163-183. Munich.

Iser, Wolfgang. 1991. Das Fiktive und das Imaginäre: Perspektiven literarischer Anthropologie. Frankfurt am Main.

Klauk, Tobias and Köppe, Tilmann. 2014. Fiktionalität: Ein interdisziplinäres Handbuch. Berlin.

Kessler, Frank. 1998. “Fakt oder Fiktion? Zum pragmatischen Status dokumentarischer Bilder.” montage AV. Zeitschrift für Theorie und Geschichte audiovisueller Kommunikation 7 (2): 63-78.

Koch, Gertrud. 2009. “Tun oder so tun als ob? – alternative Strategien des Filmischen.” In “Es ist, als ob”: Fiktionalität in Philosophie, Film- und Medienwissenschaft, edited by Gertrud Koch and Christiane Voss, 139-150. Munich.

Koch, Gertrud. 2016. “‘Madagascar, Nisko, Theresienstadt, Auschwitz.’ On the Visibility of Sites in Claude Lanzmann’s The Last of the Unjust (2013).” Ghetto Films and their Afterlife (ed. by Natascha Drubek). Special Double Issue of Apparatus. Film, Media and Digital Cultures in Central and Eastern Europe 2-3. DOI:

Koch, Gertrud and Christiane, Voss, eds. 2009. “Es ist, als ob”: Fiktionalität in Philosophie, Film- und Medienwissenschaft. Munich.

Kracauer, Siegfried. 1960. Theory of Film: The Redemption of Physical Reality. Oxford.

Kripke, Saul. 1980. Naming and Necessity. Cambridge, MA.

Lamarque, Peter, and Olsen, Stein Haugom. 1994. Truth, Fiction, and Literature: A Philosophical Perspective. Oxford.

Lewis, David. 1978. “Truth in Fiction.” American Philosophical Quarterly 15 (1): 37-46.

Matravers, Derek. 2014. Fiction and Narrative. Oxford.

Metz, Christian. 1982. The Imaginary Signifier: Psychoanalysis and the Cinema. Bloomington.

Michelson, Annette. 2017. On the Eve of the Future: Selected Writings on Film. Boston.

Nichols, Bill. 1991. Representing Reality: Issues and Concepts in Documentary: Bloomington.

Nichols, Bill. 2017. Introduction to Documentary. 3rd Ed. Bloomington.

Pavel, Thomas. 1986. Fictional Worlds. Cambridge, MA.

Peirce, C. S.  1902. “Virtual.” In Dictionary of Philosophy and Psychology, Vol. II, edited by James Mark Baldwin, 763-764. London,

Plantinga, Carl. 2005. “What a Documentary Is, After All.” Journal of Aesthetics and Art Criticism 63 (2): 105-117.

Perkins, Victor, 1972. Film as Film: Understanding and Judging Movies. Hardmondsworth.

Searle, John. 1975. “The Logical Status of Fictional Discourse.” New Literary History 6: 319-332.

Slugan, Mario. 2017. “Taking Bazin Literally.” Projections: The Journal of Movies and Mind 11 (1): 63-82.

Slugan, Mario. 2019. Fiction and Imagination in Early Cinema: A Philosophical Approach to Film History. Bloomsbury [forthcoming].

Stock, Kathleen. 2016. “Imagination and Fiction.” In The Routledge Handbook of Philosophy of Imagination, edited by Amy Kind, 204-217. New York: Routledge.

Vaihinger, Hans. 1911. Die Philosophie des als ob: System der theoretischen, praktischen und religiösen Fiktionem der Menschheit auf Grund eines idealistichen Positivismus. Mit einem Anhang über Kant und Nietzsche. Berlin.

Walton, Kendall L. 1990. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, MA.

Walton, Kendall L. 2015. “Fictionality and Imagination: Mind the Gap.” In In Other Shoes, 17-35. New York.

White, Hayden. 1978. Tropics of Discourse: Essays in Cultural Criticism. Baltimore.

White, Hayden. 1987. The Content of Form. Baltimore.

Wollen, Peter. 1969. Signs and Meaning in the Cinema. Bloomington.

Zipfel, Frank. 2001. Fiktion, Fiktivität, Fiktionalität: Analysen zur Fiktion in der Literatur und zum Fiktionsbegriff in der Literaturwissenschaft. Berlin.


Folman, Ari. 2008. Vals im Bashir / Waltz with Bashir. Bridgit Folman Film, Gang Les Films d’Ici, Razor Film Produktion GmbH.

Griffith, D. W. 1915. The Birth of a Nation. David W. Griffith Corp., Epoch Producing Corporation.

Herzog, Werner. 2004. Grizzly Man. Lions Gate Films, Discovery Docs, Real Big Production.

Hitchcock, Alfred. The Birds. 1963. Alfred J. Hitchcock Productions.

Kubelka, Peter. 1960. Arnulf Rainer.

McCay, Winsor. 1918. The Sinking of the “Lusitania”. Universal Film Manufacturing Company.

Méliès, Georges. 1898. Un homme de têtes / Four Troublesome Heads. Star-Film.

Méliès, Georges. 1898. La lune à un mètre / The Astronomer’s Dream, or the Man in the Moon. Star-Film.

Myrick, Daniel and Sánchez, Eduardo. 1999. Blair Witch Project. Haxan Films.

Ruttmann, Walter. 1921. Lichtspiel: Opus I. Ruttmann-Film.

Wiene, Robert. 1920. Das Cabinet des Dr. Caligari / The Cabinet of Dr. Caligari. Decla-Bioscop AG.

Suggested Citation

Slugan, Mario. 2019. “Theorizing Fiction in Film Non/Fiction Some Thoughts on Recent German Film Theory.” Fiction in Central and Eastern European Film Theory and Practice (ed. by J. Alexander Bareis and Mario Slugan). Special Issue of Apparatus. Film, Media and Digital Cultures in Central and Eastern Europe 8. DOI:


Copyright: The text of this article has been published under This license does not apply to the media referenced in the article, which are subject to the individual rights owner's terms.