In a previous paper, ‘Digital Cinema: Convergence or Contradiction?’ (Elsaesser 2014), I asked, given the dominance of television and the Internet, how can the cinema hold its ground among the public spaces and private occasions where cinema, television, and digital media compete with each other for audiences and attention? Is there such a thing as ‘digital cinema’, or is it an oxymoron, and how innovative is it as an art form and a technology of sound and image production today? Before attempting to answer such questions, I asked a more fundamental one: do cinema, television, recorded sound, and digital media belong together at all, considering their very different histories? When one compares them, on what basis and by which criteria? If there is a family resemblance (in Ludwig Wittgenstein’s sense), what are the factors bringing them together or what drives them apart? Is the cinema part of a general rush toward ‘media convergence’, or is it set on a divergent course, with diversity the key media-ecological factor? While ‘Digital Cinema: Convergence or Contradiction?’ tried to answer these questions by examining the apparent oxymorons ‘(multimedia) convergence’ and ‘digital cinema’, in this follow-up paper, I continue to approach the same issues across the equally self-contradictory notions ‘virtual reality’ and ‘interactive narrative’. Since these terms have entered everyday language and are no longer felt to be oxymoronic, their constitutive parts must form a unity at a different level, which is why our task will be to find a conceptual register that can reformulate or articulate this unity.

Virtual reality

One of the key differences between the ‘cinematic’ and the ‘digital’ concerns the meaning of the word ‘virtual’ in an ocular–specular system of representation, and in a symbolic–numerical system. ‘Virtual reality’ is one of the terms people most readily associate with the idea of ‘digital cinema’ or digital media, but there may be crucial differences as to what is meant by ‘virtual reality’ as well as between the idea of virtual reality and the philosophical concept of ‘the virtual’. First, the dictionary distinguishes between immersive virtual reality (VR) and non-immersive VR, even though both are based on computer simulations that employ three-dimensional (3D) graphics. For Klaus-Peter Beier (1999), ‘in immersive VR, the user becomes fully immersed in an artificial, three-dimensional world that is completely generated by a computer’. In the first case, wearing devices such as VR helmets and data gloves allow the user to interact with a simulated environment, derived from photographic sources, architectural drawings, or rendered entirely with computer-generated designs. When the term ‘Virtual Reality’ is used for applications that are not fully immersive, other, more external and manually operated input devices are used, such as joysticks or mouse-controlled navigation in order to take the user through a 3D environment on a graphics monitor. Other possibilities are stereo viewing on a monitor via appropriate glasses or stereo projection systems, modelling 3D worlds.

As far as the cinema is concerned, VR might be said to simply continue the long tradition of 3D representations, which began before the ‘invention’ of cinema with the widely popular stereoscopic views in especially the nineteenth century and 3D experiments, such as Oskar Messter’s Alabastra projections from 1903, ever since the early years of cinema. While never establishing itself as a mainstream practice, 3D cinema has thus nonetheless accompanied the history of the cinema throughout. Its relative lack of success (besides the technical difficulties) may be due to the limits of 3D environments based on photographic representations, themselves reproducing Renaissance principles of vision, which in the end, are less obviously what VR is about, even where conceived in terms of 3D environments.1 The concept of VR, in other words, is not covered or indeed recoverable by reference to the ocular–specular mode alone (it is not only about what the eye can see), so that we may have to disentangle VR from this genealogy of the stereoscope, even though head-mounted displays use an optical system that channels the images from the two screens to the eyes in such a way that they present a ‘stereo’ view of a virtual world. Virtual reality at its ‘illusionist’ worst is simply a series of monocular views following each other in quick succession, but this is far from producing the impression of immersive presence. In environments like CAVE (Computer Assisted Virtual Environment), input/output feedback devices other than the eyes and hands support the impression of immersion, by enriching the immersive experience and creating more sensualized interfaces, such as directional sound, tactile devices like wands, voice recognition systems, motion sensors, and trackers. The main characteristics of immersive VR have been summarized as follows:

Head-referenced viewing provides a natural interface for the navigation in three-dimensional space and allows for look-around, walk-around, and fly-through capabilities in virtual environments. Stereoscopic viewing enhances the perception of depth and the sense of space. The virtual world is presented in full scale and relates properly to the human size. Realistic interactions with virtual objects via data glove and similar devices allow for manipulation, operation, and control of virtual worlds. (Beier 1999)

Beier goes on to note that ‘augmented reality’ combines the real environment with superimposed virtual objects, and telepresence systems immerse a viewer in a real and immediate although distant world, with some systems enabling the viewer to manipulate real objects remotely, using robot arms and manipulators. Eventually, he argues, VR will revolutionize the user–technology interface. He emphasizes that a virtual environment

can represent any three-dimensional world that is either real or abstract. This includes real systems like buildings, landscapes, underwater shipwrecks, spacecrafts, archaeological excavation sites, human anatomy, sculptures, crime scene reconstructions, solar systems, and so on. Of special interest is the visual and sensual representation of abstract systems like magnetic fields, turbulent flow structures, molecular models, mathematical systems, auditorium acoustics, stock market behaviour, population densities, information flows, and any other conceivable system including artistic and creative work of abstract nature. These virtual worlds can be animated, interactive, shared, and can expose behaviour and functionality. Useful applications of VR include training in a variety of areas (military, medical, equipment operation, etc.), education, design evaluation (virtual prototyping), architectural walkthrough, human factors and ergonomic studies, simulation of assembly sequences and maintenance tasks, assistance for the handicapped, study and treatment of phobias (e. g., fear of height), entertainment, and much more. (Beier 1999)

Here, in a very pragmatic way, VR is defined along three complementary parameters: (1) the representation of real worlds/environments/situations, for the purpose of simulation, often with the practical aim of training, education, or therapy (especially dangerous, inaccessible, or remote environments offer themselves for such simulation); (2) VR can be employed in the context of ‘abstract systems’, making visible and representational that which is either invisible to the human eye or does not in situ qualify for visualization, such as statistical or dynamic processes; and (3) artistic works and entertainment objects might benefit from VR, but these are, as it were, afterthoughts to the much more directly ‘useful’ applications in the military field, in architectural design, medicine, and astronomy. More generally, VR comes into its own when it helps modelling real-world systems, whose 3D visualizations serve the purpose of remote control or facilitate ‘action at a distance’.

For instance, the framing documentary of James Cameron’s Titanic (1997), which tells of the underwater action to locate and explore the sunken ocean liner, fits this pragmatic definition of VR in a double sense, since the ‘simulated’ exploration of the shipwreck gives literally ‘rise’ to the VR of the romantic tale, the remoteness in deep underwater space of the sunken ship supplying the appropriate metaphor for the remoteness in memory and time of the love story. It suggests something of the potential of ‘augmented reality’ environments and the use of locative media, in which different ‘worlds’ both actual and virtual coexist in the same physical space with the viewer or user, but where the virtual is defined not in terms of illusion, but rather as a mediated form of presence of something elsewhere in either time or space or both.

There have been two, almost diametrically opposed tendencies in film studies to respond to this dilemma of the virtual not being definable in opposition to realism. One is to revive ‘illusionism’ as itself an aesthetic value in its own right, with a tradition in Western thinking under the name of ‘appearance’ or ‘appearing’ (Seel 2005), rather than seeing it merely as the (negative) verso to the recto of realism. Tom Gunning’s (1989) ‘aesthetics of astonishment’ would fit into this endeavour, but there is also a long line of thinkers from Lessing and Diderot to Kant and Schiller, who considered ‘deception of the senses’ to have potentially educational and therapeutic value, even though they never quite resolved how this could be distinguished from the ‘aesthetic illusion’ per se (was the latter more like the pleasure of ‘knowing self-deception’?), so that we inherited from them the fine line that separates the rhetorical skill of propaganda from the artistic skill of suspending disbelief (for an overview, see Koch and Voss 2006). The difference between sensory deception (seeing something that does not exist) and epistemic deception (believing that because I see it, it must exist) has been invoked,2 in order to rescue illusionism (sensory deception) from the opprobrium of misleading the viewer into unwittingly mistaking something for what it is not (epistemic deception). Instead, one may value illusionism for its affective power and for its ability to render a representation more vivid, authentic, and moving. All these qualities can be usefully examined around the heightened immersive and sensory appeal of VR and digital images, for which these aesthetic debates and philosophical distinctions provide a welcome pedigree. The second line of defence of the virtual has been to look for another way to ground even the digital image in a form of indexicality, by pointing to the enunciative dimension of the ‘index’, and thus to make indexicality an aspect of enunciation and deictics (Doane 2007).

In the present context, the important point is that VR as defined in all these contexts breaks, at the conceptual level, with our traditional definitions of cinematic ‘realism’, because ‘reality’ in VR is no longer identified with index, trace, or reference, but with a total environment: it thus is a function of a coherence theory (of truth), rather than a correspondence theory. Secondly, while there seems a great deal of emphasis on the ‘immersive’, tactile, and ‘haptic’ properties of VR, that is, the body-based nature of the experience, the bodily sensations are distinct both from pictorial realism, however construed, and from illusionism, for example, baroque techniques of trompe l’œil. Visualization in VR refers, in the first instance, to the use of images as a symbolic language, in order to render visible a set of abstract data or processes: what we see is a language of vision, not something that is in any sense actually ‘out there’. It is as if several different perceptual– sensory registers and cognitive systems were employed to render the composite ‘effect’ of reality. Together, their combination does not make for one continuous ‘field’, such as we encounter it in everyday perception. Rather, the amalgamation of heteroclite elements in a geometrically constructed, homogeneous space either repels because of the claustrophobia such a virtual environment engenders, or it fascinates users who feel empowered by the feedback possibilities of ‘controlling’ a space that controls them. The artist Dan Graham explored some of the perceptual ambivalences of similarly closed circuit feedback loops in his Time Delay Room (1974), a sort of meta-critique of VR even before it had become a ‘reality’ and an accepted term, while the philosopher Slavoj Žižek (2009) considers VR ‘a rather miserable idea’, because it is not virtual enough:

Virtual Reality is a kind of Orwellian misnomer. What is threatened in its rise is the very dimension of virtually consubstantial with the symbolic order. . . . The commonplace according to which the problem with cyberspace is that reality is virtualized, so that instead of the flesh-and-blood presence of the Other we get a digitalized spectral apparition, misses the point: what brings about the ‘loss of reality’ in cyberspace is not its emptiness (the fact that it is lacking with respect to the fullness of the real presence) but, on the contrary, its very excessive fullness (the potential abolition of the dimension of symbolic virtuality). . . . We are thus a long way from bemoaning the loss of contact with a ‘real’, flesh-and-blood other in cyberspace, in which all we encounter are digital phantoms: our point, rather, is that cyberspace is not spectral enough. (200)

Žižek’s comments remind us that we have to distinguish the concept of ‘virtual reality’ from that of the ‘virtual’. A brief look at the semantics of the ‘virtual’ might help to clarify what is at stake in this difference: Gilles Deleuze (2002, 112–115), for instance, has argued that it is the primary state of being (all of the future is virtual, seen from the present), that it is the realm of everything that is possible. We might add that virtual, in the form of contingency, is the ground of our lives: in which case, the actual is not in opposition to the virtual, but merely a specific realization or manifestation of the virtual. Applied to the cinema, one can argue that the virtual is nothing other than the status of objects whose representation is generated by the symbolic code we call ‘image’. If one then wants to accommodate digital images in the history of the cinema, they could be regarded as merely the product of another symbolic code. The difference between analogue and digital images would then not be located in the supposed indexicality and truth-status of the photographic image (its existential link with the real world, and thus its value as document), against the supposedly ‘virtual’ character of the digital image (grounded in nothing but numbers and a set of binary instructions, and thus with no link to the real world and no responsibility to veracity or verification). Instead, the digital image would be amenable to a properly linguistic analysis.

It is an argument which, at the height of the semiological turn and thus within a different vocabulary, was already forcefully put by Umberto Eco (1976) when he deconstructed the indexical level of the photographic image into a dozen or so iconic and symbolic codes. The Czech media historian Vilem Flusser also pointed out, some 30 years ago, that in any photograph, the distribution of the grain already prefigures both the dots of the video image and the numerical grid of the digital image. Other media historians have likewise drawn analogies between the mechanized loom of Jacquart in the eighteenth century and the television image of the de Forester cathode ray tube in the twentieth century, or the Hollerith cards that made the fortune of IBM in the late nineteenth century and the digital computer for which it became known in the 1960s and 1970s.

The difficulty (and confusion) arises because in the case of (photographic, analogue) cinema, we are operating within an ocular–perceptual frame of reference (where technical images only know an ontology of realism/illusionism, real/fake, and whose epistemology is true/false). Within this field, we are bound to see the virtual (as in VR) as also part of the ocular–perceptual register, which corresponds to the logic of photographic cinema, regardless of whether we see its representational mode as ‘perspectival’, ‘flat’, or ‘multi-focal’. This cinema, emulated by the digital in all its properties and effects, adheres to the ocular, but is itself not generated by or determined through the optical. Thus, in the digital, we do not have an ocular–perceptual ontology; instead we have a digital ontology, which is that of the dynamically variable, scalable. This brings us not to the what-if logic of the spreadsheet, as well as to the relational theories of meaning in structural linguistics and its mathematical–logical foundations (in thinkers such as Greimas). The fundamental logic of computation is not: is it real or is it imagined (virtual), but what alternatives/possibilities (virtualities) can we extract or extrapolate from the real – understood as data and information?

These two kinds of virtuality may look the same to the eye – see above – but the logic that produces them is totally different, and thus we need to see once more how we can reconcile the fact of apparent continuity and underlying difference – in other words, it is not good enough to have the binary oppositions, we also need to see how they come together again.

This means that the logic of the virtual may well not only absorb the cinematic (a logic of which the latter is henceforth a mere local effect), but it may also extend into the culture at large, as a ‘cultural logic’ whose manifestations have very little to do with digitization per se, and instead indicate a more pervasive but also intangible paradigm shift. The virtual in this sense may even create a modality, a temporality, and a tense, which the visual does not know at all, but which brings the virtual closer to language, rhetoric, and formal logic than one might initially suppose. In addition to the (algorithmic) logic of ‘if . . . then’ conditionals and the hypothetical mode ‘what . . . if’, one would also want to argue that one of the implications of this new cultural logic of the virtual is the prevalence of the tense of the future perfect. The future perfect is a tense that refers to a completed action in the future. When one uses this tense one is projecting oneself forward into the future and looking back at an action that will be completed some time later than now. In other words, the future perfect is a form of anticipated retrospection. The virtual, we might say, has become a tool of ‘looking into the future’, except there is no ‘looking’ involved anymore. What used to be the task of priests, augurs, and shamans, namely, interpreting material signs, in view of their virtuality (i.e. their inherent futures) is now a manipulation of numbers, variables, and their correlation, based on complex (non-linear) models of causality that follow a stochastic (random or indeterminate) order, within parameters of statistic probability (cf. complexity theory, that is, the interaction of parts in a system whose workings are made complex by inherently circular causation, recursiveness, and other feedback effects).

Combining the ‘future perfect’ of precognition with the simultaneous ‘foreclosure of the future’ would be the tragic paradox (i.e. tragic for humans) enacted by ‘the perfect computer’:

the idea of an infinite prediction machine, more or less unable, and constitutively so, to calculate the effect of prediction itself: the problem that, at a certain logical point, the computer will have to compute the effect of a prediction upon itself, the prediction-effect of predicting a prediction. Here is your Moebius strip, the key point being that this algorithm goes on forever, entering into a dreadful, dreamlike mirroring of itself; it is a spiral, a labyrinth, a mise-en-abime. (Daniel 2006)

Yet something else supports the impression or illusion of ‘truth’ emanating from the moving image almost as decisively as the loop or spiral into which it draws us. Roland Barthes (1981), in Camera Lucida, speaks of the way a photograph involves the viewer in a peculiar kind of presence and absence, what he named the sense of ‘having been there’. This sense, as he analyses it, is also a tense, joining a perfectum with a present, in a conjunction of a place and a time. Moreover, with its ‘thereness’ it also embeds a deictic mark, that is, a speaking position and the trace of a subject in discourse: a photograph addresses us by the very act of placing us before it in relation to our existence in time and language. The moving image, by extension, derives its reality-effect similarly not only from its iconic features, but also from the tense-structure within which it holds the viewer, a ‘here and now’ which is, however, always already a ‘there and then’. To this temporality thus correspond distinct spaces, so that the audiovisual experience always involves several space-times, related to each other and yet distinct from each other, a then/here space and a now/there space, which, by at once doubling and displacing the viewer’s own now/here space-time make possible the subject-effects of audiovisual experience as an affective engagement. We might say that vision is only part of what makes the moving image ‘real’ to a viewer, and the question one needs to ask is not so much whether the digitized image severs the link between the material trace of the real in the photograph and its power to resemble the real, but how digitization affects the time–space relations for the viewer, and thereby the ‘tense’ of the image.

Virtual reality, then, when set up in opposition to the experience of cinema (which is still thought of as involving distance, disembodiment, and ocular verification), is the fantasy of tactile, haptic, body-based sensations, rather than its virtual realization, for the relation of VR to the body is still metaphoric, even though the effects are experienced as real, and the relation of VR to reference is via an interface, which functions like a symbolic language, which our mind and body ‘translates’. Once we think of the virtual as a modality, then its importance is neither along the lines of embodied/disembodied, nor true/illusionistic, but syntactic: the virtual is a tense, namely, that of the conditional. Moreover, as a tense, the virtual in VR has its own form of grammar, and – by extension – belongs to the realm of the semiotic as much as the phenomenal or the senses. Its syntax is that of the algorithms of the spreadsheet, the if-then logic of variability and connectivity. On the other hand, one can imagine that the term VR might in time become obsolete, in so far as its pragmatic function (as the simulation of dangerous situations upon which to ‘act at a distance’) will be subsumed under terms such as ‘remote control’ device or ‘closed circuit’ observation, while VR as immersive experience will develop in ways rather similar to the newer 3D technologies: making the prosthetics (head-mounted display, 3D glasses) less and less obtrusive, if not disappear altogether.3 In all other respects, VR would come to be seen as a sub-category of ‘augmented reality’ – the exposure to and interaction with any information-rich environment, whose ‘reality’ is ‘augmented’ by data, accessible or retrievable by an appropriate technology (e.g. smart glasses, smartphones) that serves as a perceptual–experiential interface.

Interactive narrative

Interactive narrative is most commonly understood as the use of digital technology to construct virtual environments within which to present stories interactively, that is, with input and feedback from the spectator. Here, one right away encounters a paradox: one widely held view is that, basically, there is no such thing as an ‘interactive narrative’. The term confuses narratives with games, and interactivity with non-linearity. We normally think of a narrative as a series of events that are linked causally and chronologically, but we are also aware of the fallacy of thinking that because something happened after something else, the two are causally connected. In E.M. Forster’s famous example, ‘the king died and then the queen died’, the causal link is left open. Yet such a link can either be inferred by the reader or suggested by the narrator, which indicates that a narrative is not just a sequence of events but requires a narrating/inferring instance. It also requires one to occupy a particular position in time. As Marie- Laure Ryan (2001) puts it:

When we read a narrative, even one in which the end is presented before the beginning, we adopt the outlook of the characters who are living the plot as their own destiny. Life is lived prospectively and told retrospectively, but its narrative replay is once again lived prospectively.

To add a different frame of reference: what commonly passes for interactivity in narratives is strictly speaking hyperselectivity. The aim is to programme an architecture of multiple choices presented from a pre-arranged menu, and leading to different paths, which among themselves have certain nodal points. When these are carefully or cunningly devised, they can give the illusion of freedom of response: ‘creating a better kind of mousetrap’ as it has been called.4 Yet even if the menu could be extended indefinitely (and with it, the options arising from the nodal points correspondingly multiplied) another, perhaps even graver, difficulty would arise. This difficulty has to do with a story’s ‘point of view’: who tells what to whom, or put differently, who controls the relation between information and inference for the viewer in the process of narration, and organizes the levels and perspectives of narration. In other words, even if cinema were totally ‘digital’ in sound and image, this would not in itself facilitate the films becoming interactive, for the latter implies that the viewer could intervene actively in the progress of the narrative or take over the function of the narrator (for a more detailed discussion, see Bolter 1991).

From what has been said so far, it would seem that narrative interactivity is indeed an impossibility. In fact, the computer does not aid interactivity but leads in the opposite direction: to automated story generation. However, let’s consider this impossibility a little more closely: interactivity is clearly an important aspect of communication, but it may be one that is mistakenly associated with cinema. Certainly, those who consider the cinema an art form (as opposed to a mere medium of communicating information) would argue that art distinguishes itself from other human activities by the distance – the aura – that separates it from the observer, and thus it ‘answers back’ only in virtual ways. Even more crucial in the definition of the traditional artwork is that an artwork ceases to be one if the user/ viewer can materially alter it.

On the other hand, all arts are interactive, if one thereby understands that the viewer, or reader, engages in a constant dialogue with the work and its maker, and that every interpretation of a work is not only an act of appropriation, but also an act of exchange, an invitation to dialogue. If interactive is understood to mean open form or variable access, then indeed, certain modernist works seem to have anticipated what used to be called ‘hypertext’ form, and certain theories of the literary text – those of Roland Barthes, Umberto Eco, or Jacques Derrida for instance – have elaborated on the philosophical or semiotic foundations of textual interactivity.

Nonetheless, there is now a considerable body of literature that has investigated the extent to which games can simulate story worlds and storylines, and films can be plotted to emulate the agonistic structures of games and their multiple levels (Simons 2007; Buckland 2014, etc.). While the promotional ideology of marketing games as stories may suggest open-ended narratives that promise freedom of choice, the options are invariably constrained by the ‘rules’ of the game, which tend to conceal the fact that choice is both limited and predetermined. The user colludes with being a ‘player’, whose freedom can be summed up as: ‘you can go wherever you like, so long as I was there before you’ – which is of course also the strategy of the ‘conventional’ storyteller (or narrational agency) whose skill lies in the ability to suggest an open future at every point of the narrative, while having planned or ‘programmed’ the progress and the resolution in advance.

This casts doubt also on an argument sometimes advanced in favour of the absence of a narrator, of a point of view, of a central perspective: might it not ‘free’ the narrative from its ‘classic realist, bourgeois-patriarchal’ symbolic which film theory in the 1970s always protested against? Reviving Roland Barthes’ term of the ‘writerly text’, theorists such as George Landow and others have argued that Barthes and Derrida’s literary theories ‘anticipate’ hypertext and interactivity and thus provide these practices with their philosophical foundations (Landow 1991). However, this would be attributing an excessive literalness to a theory of excess and indeterminacy! (see Cameron 1996).

The problem can also be more succinctly phrased as one between ‘narration’ and ‘navigation’: there is a category shift between random access in media that offers some resistance in the form of an irreversible temporal flow, like the novel or a narrative film, and those that do not.5 If a medium does not offer this resistance – such as, for instance, an ordinary electronic database – what exactly does random access bring about or add, in respect of this dialectic of freedom and constraint that seems so fundamental to storytelling? To what extent does navigating rely on a ‘map’, and what would a map in the case of such a narrative look like? No longer would a story be the exploration of a world thanks to a narrative (a quest, an initiation). The interactive video game is more like the exploration of a narrative (an itinerary, a journey) thanks to a ‘world’. The setting or set, the fiction, the collection of information functions then as background to a new kind of (serial) activity: a sampling or sorting of information, making connections in the form of montage, or more likely, following the connexions laid by others. In other words, with an interactive story accessed via a game platform or on the Internet we may be invited to ‘enter a world’, but besides exploring its story-paths, we also experience its narrative architecture: its paths and its detours, its branching and its multiple choices. We enter a territory in order to explore its map, rather than use a map to explore a territory. What we have as our mode of transport – to move, surf, connect – is a device (console, mouse, track pad): a static vehicle par excellence. Shooting an enemy, ascending a tower, or descending into a dungeon are actions translated into clicks, which in turn initiate a movement at one remove, while the locations we thus reach conceal their own motion vehicles: the hyperlinks and hot spots, transporting us elsewhere. It is the point at which narrative becomes, in Lev Manovich’s words, ‘tele-action’ and we are in what would seem related but nonetheless distinct paradigms (of action, rather than representation) (see Manovich 1998). The digital is in this case the more convenient or efficient means to this end, though not the end itself. The end has to do with the spatialization of time: providing mobility, conveyance, ‘transport’, action. The ‘motion picture’ narrative as a form of travel while never leaving home.

Alternatively, the space-time coordinates of interactive cinema have to be constructed in order to convey an impression of movement while allowing for the ‘illusion’ of freedom and choice to arise. Grahame Weinbren seems to have pondered these questions harder than most, partly by commenting on his own interactive cinema. Once again, the issues arise, irrespective of digitization, and Weinbren conceived and developed his interactive films prior to the advent of digital images, since he initially used analogue laser discs, connected to a computer software programme (see Weinbren 1989, 1995). His work is a most intriguing example of interactive/multiple choice cinema, and in its terms, extremely successful: with The Erl King and Sonata he has made film installations which create a sort of multiple perspectivism by first of all telling the story from different vantage points or ‘narrators’ (à la Rashomon or the films of Alain Resnais), all of which are accessible to the viewer in the form of hyperselectivity. However, he also allows the viewer-as-user to ‘enter’ into and explore ‘pockets’ of the narrative world, which are not exactly parallel universes, yet have a high degree of autonomy, while nonetheless belonging thematically or in their visual or aural motifs to the underlying story world. What adds to the suspense, however, is that while one is ‘inside’ these pockets, the story ‘elsewhere’ is nonetheless progressing in linear fashion towards its predetermined conclusion. The viewer-user is therefore structurally split between his desire to explore the pockets and his anxiety not to lose the thread of the narrative, which is an intriguing compromise between (the ‘laws’ of) narrative and (the possibilities of) interactivity, making one ‘police’ the other, so to speak, while creating a number of unusual and unusually powerful subject-effects.

Weinbren’s early experiments with interactive cinema were installation pieces: not so much a prototype as a meta-type and a philosophical object, rather than a game. In the nature of such work, his interactive cinema did not have a mass-media future. While video games – some of which, though by no means the most popular, are based on successful movies – have proven to have major commercial potential, one would hesitate to apply to them the term interactive narrative. Nor would one call them interactive cinema: Manovich has proposed the idea of ‘soft cinema’, and has produced a prototype, which – rather like Weinbren’s work – has been taken note of in the art world, but not among gamers. The difficulties of interactive cinema are not primarily technological, and they may not even be (solely) conceptual. What they highlight is the social character of the cinema experience, that is, its event status, and the temporalities it implies. Two of the most respected theorists of the cinema, Raymond Bellour and Jacques Aumont, have been quite categorical on this point:

The live projection of a film indoors, in the dark, in the prescribed time of a more or less collective performance, became and remains the condition for a unique experience of perception and memory, defining spectatorship, which any other viewing situation would more or less alter and impair. And this alone is worth being called ‘cinema’. (Raymond Bellour, quoted in Leroy 2013)

And: ‘This is a crucial point today, one I think we can use as a good criterion of what is cinema: any film presentation that leaves me free to interrupt or modulate this experience [of irreversible flow] is not cinema’ (Jacques Aumont).6

Andy Cameron (1996), a pioneer of interactive media, has pointed out, in ‘Dissimulations, the Illusions of Interactivity’, that the difficulties of ‘interactivity’ are too often euphemized by fine art circumlocutions, such as ‘interactive media environment’, ‘interactive installation’, or ‘transformational space’, where by rights the more obvious word ‘game’ should be used, though because of its low-culture associations it is avoided, even repressed: ‘if the repressed reading of interactivity is that of the game, the preferred reading of interactivity is that of postmodernism’. What is satisfying about Weinbren’s pieces and Manovich’s Soft Cinema is that they are precisely neither ‘games’ nor ‘postmodern’ narratives, but instead, feel ‘readerly’, in the way they give a strong sense of an underlying, continuous time management, in other words, suggest the presence of a storyteller, in the form of a coordinating presence, while still fully engaging – in fact, enabling – the spectator.

Granting that normally only games should be called ‘interactive’, these examples highlight differences between a narrative scenario (where the user/viewer confronts a narrator as his Other) and a ‘game’ scenario (where the user/player confronts another player as antagonist, and the designer as the super-narrator). Important, again, is the time–space dimension. Games may seem to have a different way of representing time, but one can relate it to narrative devices: video games/interactive programmes usually create strict time constraints which force decisions, choices. This means that the immersive potential of film-narrative time (made up of suspense, surprise, the uneven distribution of knowledge between protagonists and spectator, and the flow of information from one plot situation to the next) is replaced by the temporality of the countdown, by levels of difficulty (usually measured in the time it takes to complete tasks), or of branching points and the containment of hyperselectivity or ‘branching explosion’. On the other hand, classical narratives are also often constructed around deadline scenarios that act as countdowns.

It is for reasons such as these that both narratologists and game theorists have tried to narrow the gap, and map more precisely the common ground that might exist between narratives and games. The question is whether one does so by splitting the difference, as it were, or by seeking an encompassing conceptual framework that shows how both games and narratives are special cases of a more general phenomenon, or constitute specific answers to a common problem. For instance, a distinction that refers to film spectators as ‘external observers’ and to gamers as ‘involved players’ will not do. The same goes for comparing and contrasting ‘cinematic identification’ with ‘role-playing’ in gaming. The similarities are too slight, the differences major. On the other hand, there is no doubt that – whatever the precise causal link – a certain ‘classical’ conception of narrative in mainstream cinema has come under pressure since the emergence of the Internet and digital media. The fact that we now speak of modular narratives, network narratives, forking path narratives, mosaic narratives, multi-protagonist narratives, multiple draft narratives, complex narratives, puzzle films, mind-game films, fractal films, hyperlink films, twist films, ensemble films, etc. is ample proof that non-linearity, distributed agency, branching structures, and a proliferation of protagonists with competing agendas have become staple features of mainstream cinema, and to this extent, these features do indeed have their correspondences in gaming environments. However, there are theorists, most notably David Bordwell, who forcefully argue that all of these new features can be explained within the existing parameters of ‘classical’ narrative and traditional Hollywood storytelling.

The term ‘interactive’ has also turned up in connection with documentaries. One such example is The Last Hijack (2014). Using classical techniques of both documentary and feature films by showing the same events from different perspectives and points of view, it reconstructs (partly through re-enactment) the hijack of a cargo ship by Somali pirates:

Exploring both sides of the story, Last Hijack Interactive allows you to uncover the complex realities behind piracy in Somalia though the eyes of a pirate and a captain, an advocate and a journalist, a parent and a wife. Combining live-action video shot on location and animation, the interactive experience gives the user the opportunity to navigate the real stories of these people, building to the hijack itself and the resulting aftermath. Here, the interactive version is a digital extension (via random access and interrupting the flow) of a classical linear narrative, with the usual paratexts that come with a DVD (behind the scenes material, background information, interviews etc) accessible at any point during the playing of the film. (Wolting 2014)

Would it therefore make sense to speak of ‘digital narrative’, instead of interactive narrative, and if so, what are the main characteristics of a digital narrative? Warren Buckland has identified a number of features, including serialized repetition of actions (to accumulate points and master the rules), multiple levels of adventure, space-time warps, magical transformations and disguises, immediate rewards and punishment (which act as feedback loops), as well as interactivity. He notes that:

the structure of a few contemporary narrative films, such as The Fifth Element (Besson 1997) . . . and Inception (Nolan 2010), as well as Source Code, are in part structured using the abstract rules of video games. These films are shaped by both narrative and video game logics . . . (Buckland 2014, 185)

Conclusion: from play station to workplace, and from narrative to database

We seem to have come full circle. If at the cognitive level, the common basis for both games and narrative would appear to be certain fundamental human traits about ‘mastering’ life situations by proxy, that is, through testing, vicarious experience, and simulation – a mastering that can be analysed via mathematical game theory, evolutionary psychology as well as cultural theory – then one answer to our contradictions and oxymorons would indeed centre on the meaning we give to the terms ‘virtual’ and ‘interactive’. Both terms can belong to the semantics of the hypothetical and the conditional, to the realm of the ‘as-if’ and the ‘thought experiment’. Both narratives and games would ‘rehearse’ life at the same time as they replay it, which would open up to a broader framework for understanding the social function of the digital media, in the way Walter Benjamin (1969) understood the cinema as ‘rehearsing’ the distracted attention/decentred perception of the metropolitan subject of ‘modernity’ (222–223). While ‘multi-tasking’, ‘short attention span’, ‘parallel processing’, and the like have become the default values of perceptual participation in our consumer society, the call for ‘interactivity’ may nonetheless herald a further regime of rehearsing and disciplining new modes of behaviour and agency. Consider the many ways that ‘interaction’ with so-called ‘intelligent’ machines has come to determine our daily lives, whether it is asking an ATM for money or answering an automated phone when making a call for services. The human–machine interface is in place at work and in the office, as well as in situations of leisure, requiring special skills of cognition, reaction–response, and embodied sensory coordination of vision and audition. If the digital in the realm of images frees representation from its bonds to the index and in the realm of narrative minimizes the differences between narration and navigation, and between the plot and the puzzle, between ‘restoring a lack’ and ‘solving a problem’, then by the same token (though not primarily for technological reasons) the digital privileges a new cognitive coordination of ‘seeing’, ‘feeling’, ‘believing’, ‘knowing’, and ‘acting’: although, as one might add: ‘not necessarily in that order’.

Assuming we grant that cinema is a form of symbolic action, combing features of storytelling with those of a social ritual, then the more general question of interactivity would again pose itself not so much around narrative per se, but around the different types of (actual and virtual) feedback, typical of any act of viewing/participation. Such feedback loops (positive: self-reinforcing; negative: self-regulating) are cognitive and affective for the viewing situation of cinema, and more embodied and dependent on physical response in the video game environment. However, feedback is also crucial in religious ceremonies (‘trance’), at public spectacles (‘contagion’), online life (‘going viral’), as well as in therapeutic situations (‘transference’). Feedback phenomena may have become more pervasive in contemporary life, but they are also becoming more invisible, when one factors in the feedback loops created by our online presence (cookies) and via other, even more intrusive and hard-to-detect surveillance devices, which now constitute a major form of ‘interaction’ that the state has with its citizens, and the corporation with its customers.

It suggests that if such digitally interactive situations are becoming the (social) norm, then we would do well to separate ‘interactive’ from ‘narrative’ after all, mainly by reinterpreting what we understand by narrative. Over its more than 5000 years of documented history, storytelling has had many functions in human lives – as collective memory, as problem-solving mechanism, as imaginary resolution to real contradictions, as ordering principle of contingent events, as therapy for life’s traumata, as consolation and agent of redemption. However, narrative is also one of the most efficacious ways human beings have devised for storing and retrieving information, for organizing perceptual data, actions, and events in a comprehensible and easily communicable way. By modelling stories on the human experience of time as succession in sequence, and thus following the logic of the ‘post-hoc ergo propter hoc’, while also taking as its dramatic arc the life cycle of beginning, middle, and end, narrative has made itself seem natural and eternal, mutable but impervious to fundamental change.

Confronted by digital archives, the vastly expanded availability of machine memory, and the ease of electronic retrieval, narrative may find itself having to compete with other types of databases in general and other storage-and-retrieval modes in particular. Interactive narrative in this sense – and here its oxymoronic conjuncture may act as a salutary irritant – obliges us to make a clearer distinction between storage mode and retrieval medium, where narrative refers to the particular mode, while interactivity would point to the medium.

If the cinematograph is a recording and storing instrument complementing or superseding language and writing, their relation to each other is governed by a calculus of gain and loss, with different variables intervening. On the one side, they are just two of the many material supports and technologies that humans have devised to record experience and store knowledge – from architectural monuments to scrolls, from codices and manuscripts to printed and illustrated books, from canvass to photographic plates, from celluloid to magnetic tape, from tape to hard disk, from hard disk to cloud storage. On the other side, there are the different modes (or genres) that have evolved for perceptually accessing and cognitively processing the data thus stored. With respect to the medium, some of the variables that come into play are compactness of storage, physical access and speed of transmission, durability of the material support, and fidelity of the signal. Concerning the modes or interfaces, they vary: in literature, it would be the different genres – poetry, drama, novel – each of them with a different ratio between density of data or information and ease of access or apperception. Horace famously predicted that his poetry would outlast a bronze sculpture, but poetry, while linguistically more dense than a novel, is less accessible without special decoding; a dramatic performance of a play is more immediate as a sensory experience, but it has to be witnessed in situ. The same goes for imaging: the pencil sketch is quicker than the engraving, but less durable; the fresco is larger in size, but fixed and in situ, the easel painting is smaller, but mobile and portable; the photograph is instantaneous, but on a perishable support; a film is more portable than a play, and it makes a virtue out of its immateriality; the digital image is variable in size, malleable, and rich in information, but requires special hardware and dedicated software to be perceptible or intelligible at all. In short, narrative stands in a complex relation to other storage media and storage modes, each not only with a distinct history, but also with competing ratios of information density and decodable accessibility. These, then might be some of the parameters to guide our thinking when trying to go beyond the apparent contradictions of narrative and interactivity. Faced with ‘interactivity’ as the new disciplinary norm of the ‘control society’ (Deleuze), the digital archive and the database would be the ambiguous ground of both compliance and resistance: whether accessed through unstructured random access, the rules of game scenarios, or structured narratives, the challenge in each case is to shape an actual out of the virtual and to retain or regain agency from interactivity.