Abstract: This article sketches a genealogy and typology of the split screen in mainstream film, identifying three distinct phases in the integration of this device since the 1950s, each relating to broader cultural shifts ushered in by media advances and transitions: telephone in the 1950s, television in the 1960s and 1970s, and the computer since the 1990s. I argue that the emphasis upon fragmented and multiplied display relates largely to the cinema’s demonstrated capacity for negotiating the meaning and significance of media change to a wider audience. Through its variegated split screens, the cinema functions as a guide to and user manual of the dangers and possibilities of technological transformation.
With an ancestry that goes nearly as far back as the cinema itself, the technique of the split screen that displays several separate and delineated images within one larger frame, cuts across the different platforms of film, television, and computer-based media. In recent years, this divided and multiplied image has gained visibility and dominance across these different media, yet to date, has not garnered much critical attention. One thinks, for instance, of TV’s multi-layered images (from sports TV through news reporting to Fox Network’s 24 [2001-] series), the ubiquity of the windowed graphical user interface in computer programmes, or the multi-channel video installations of the art world. In this article I want to sketch a genealogy and typology of the multiplied image, usually referred to as ‘split screen’ in mainstream film, identifying three distinct phases in the integration of this device since the 1950s, each relating to broader cultural shifts ushered in by media advances and transitions. In this way, I hope to shed some light on the current predominance of this phenomenon. Film has always been fascinated with technological change and has thereby sought to provide ways of understanding other media forms such as the telephone, television, and computer. I argue that the emphasis upon fragmented and multiplied display relates largely to the cinema’s demonstrated capacity for negotiating the meaning and significance of media change to a wider audience. Through its variegated split screens, the cinema functions as a guide to and user manual of the dangers and possibilities of technological transformation.
Pillow Talk and the Telephone
Between the mid-1950s and the mid-1960s the dominant form of Hollywood comedy was the so-called ‘sex comedy’ in which (heterosexual) romance is exploited for comic purposes. Now, while this brand of comedy has not fared well critically – these films are today routinely charged with conservatism and upholding the status quo of Cold War America – it has also been overlooked that a number of these films employ scenes with split screens, injecting a measure of formal innovation into what often appeared to later observers as stale and predictable comic routines. The most prominent case is surely Pillow Talk (Michael Gordon, 1959), but there are a number of lesser known examples that also employ this technique in interesting ways, among them: Desk Set (Walter Lang, 1957), Indiscreet (Stanley Donen, 1958), The Grass Is Greener (Stanley Donen, 1960), Lover Come Back (Delbert Mann, 1961), That Touch of Mink (Delbert Mann, 1962) and Charade (Stanley Donen, 1963). 
The split screen is often regarded as hard to reconcile with the idea of transparency inherent in continuity editing and a narrative style focused on the seemingly unmediated display of story information because it foregrounds the artificial nature of the image. A frame within a frame draws attention to the act of framing itself by visibly displaying the basic principle that forms the condition of possibility for the image: the frame that draws a distinction between inside and outside, between image and non-image. The sex comedies have a simple solution to this problem of ‘baring the device’ (the formalist term for the visual exhibition of usually invisible techniques) since they only use the split screen for one particular situation: telephone conversations (Wulff 1991a; Wulff 1991b). This type of situation plays out so effortlessly because it echoes the basic properties of the split screen: two spaces are distinct, yet also connected. The division of space refers to the physical space of the callers who are in distant locations, while the acoustic space is shared by both speakers. This paradoxical configuration of (physical) distance and (acoustic) proximity finds a graphic equivalent and match with the split screen. Moreover, the graphic set-up alludes to the playful shifting of distance and proximity in another more oblique way as the split screen allegorises the spectatorial situation in the cinema which is similarly characterised by an oscillation between presence and absence, between proximity and distance.
Since Pillow Talk is the most famous example, and also the film that uses split screen most consistently in the cycle of ‘sex comedies’, I will now provide a brief analysis of this film. The film tells the story of Jan Morrow (Doris Day), an interior designer, who shares a telephone line with composer (and playboy) Brad Allen (Rock Hudson). He constantly blocks the line romancing his countless girlfriends and she gets increasingly angry with the playboy. When they meet face-to-face and he realises who she is, he pretends to be the old-fashioned Texan businessman Rex Stetson. Not surprisingly (for the genre), they fall in love and the split of the male protagonist into a positive love interest and a negative antagonist becomes the pivot of the movie.
Pillow Talk embeds the split screen in intricate ways in the narrative of the film. In total, the film has ten split-screen scenes that are unevenly distributed over the length of the film. Two thirds of the split-screen scenes are in the first act of the film, while none are in the last act. Far from being coincidental or arbitrary, this uneven distribution functions as a play of foreshadowing and allusion as it provides the couple with a shared (virtual) space before they actually share a (physical) space. Therefore, the device fulfils a double narrative purpose: on the one hand, this technique has an economic function as it enables the film to refrain from clumsy and complicated parallel editing patterns, presenting two separate images in one frame. On the other hand, and this is more important, the spectator can already witness how well the couple fits together as the halves of the split screen correspond to each other in terms of colour, mise-en-scene, montage and internal movement. The spectator sees already the shared communal space, while the narrative has to work through the intricate plot movements in order to get rid of such an unclassical device as the split screen:
We see the split of the party line, represented as a split screen, healed visibly towards the end of the film, as Brad and Jan are splayed not across the split screen but an undivided single screen. In fact, as the film progresses, the split of the screen changes axis; it is no longer split between Brad’s and Jan’s side, but between Brad onscreen and Jan offscreen (McCallum 1999: 94).
Seen from this perspective, the film integrates the split screen into the classical paradigm in which narrative development and character logic overrides technology and technique. Put differently: the price the classical paradigm has to pay for its transparency, unambiguous visibility and distance is to ultimately get rid of the split screen altogether.
Fig. 1. Pillow Talk (Michael Gordon, 1959). Image Source: DVD Universal 2003.
All split-screen scenes in Pillow Talk are telephone scenes, echoing, as argued, the basic properties of the telephone conversation. The paradoxical tension between distance and proximity, between absence and presence is overcome in one scene in particular when the physical division and acoustic closeness are confused as touch complements the visual and aural situation. The split screen shows the protagonists lying in their respective bath tubs, the woman on the left and the man on the right (this placement is consistent throughout the movie) – mise-en-scene, lighting and colour all work to downplay the visual distinction between the two separate images – as if they were in the same bath together (fig. 1). When Brad gently strokes the wall with his toes at the exact point where Jan has put her foot, she pulls it back as though she has been tickled by him (fig. 2). Even though physically impossible (Jan and Brad are in distant places and only talk on the phone), the separating wall becomes semi-permeable. This incident literalises the strange configuration in which the division is at the same time visibly present (both images are in the same frame), yet also visibly negated (we know that we are watching two separate images). A similar example can be found in The Grass Is Greener in which a telephone conversation between Victor Rhyall (Cary Grant) and Charles Delacro (Robert Mitchum) is shown in split-screen. When the call has ended, the graphic configuration of the split screen is upheld. Rhyall flips his fingers drained in champagne in the direction of Delacro who is (in the film’s diegetic logic) a couple of hundred miles away. Delacro reacts just like Jan in Pillow Talk – as if he had been really touched: in disbelief he wipes his cheek and looks in the direction of Rhyall.
The strategy of negotiating one medium (telephone) via another (cinema) is telling, as it provides the audience with a model for making sense of technological shifts. Surely, the ‘sex comedies’ of the 1950s and 1960s influenced each other in turn, yet the fact that they invariably used split screens in depicting telephone conversations shows how this solution proved persuasive to makers and spectators alike. The oscillation of absence and presence, of distance and proximity, that is inherent in the telephone conversation as in the split screen provided a mental model that could explain simultaneously the new spaces and practices of modernity (with its transformed media environments) as well as novel forms of social relations mediated by technologies such as the telephone that play across the presence/absence divide.
Grand Prix, The Boston Strangler and Television
Over the course of the 1950s the American film industry began to feel the combined effects of demographic change, lifestyle transformations (suburbanisation, car culture, teenage youth culture) and the Paramount decree which required the studios to divest themselves of their cinema chains (Lev 2003). Although these developments were felt in Hollywood, prompting experiments with widescreen, colour, and generally, new modes of production, the output of epics, melodramas, westerns and musicals still belonged to this relatively stable period within the studio system. All this changed – slowly, but steadily – over the course of the 1960s. Whereas in the 1950s the split screen was integrated seamlessly into continuity style editing by finally getting rid of it altogether, the late 1960s saw a very different wave of split screen usage. A number of popular films from new or emerging (sub)genres employed the technique in different and innovative ways: the most prominent examples are the racing film with multiple protagonists Grand Prix (John Frankenheimer, 1966), the serial-killer film The Boston Strangler (Richard Fleischer, 1968), the heist film The Thomas Crown Affair (Norman Jewison, 1968) and the rockumentary Woodstock (Michael Wadleigh, 1969). I will look closely at the first two films of this list in order to point out the dominant logic of the split screen during this period. However, this logic also applies to the other two examples as well as to latecomers in the early 1970s such as The Movie Murderer (Boris Sagal, 1970) and Phantom of the Paradise (Brian DePalma, 1974).
By the late 1960s, the dominance of the classical paradigm – based on concepts such as continuity of time and space, transparency, and invisibility of technique and technology – had been largely eroded, making possible new and unexpected uses of an ostentatious device like the split screen. With its star power, above-average length and spectacular production values, Grand Prix turned the 1950s epic into a self-conscious symphony of cars and spectators, using the split screen to underline the multiplicity and seriality of Formula One racing: we see rows of cylinders multiplied, chequered flags and safety goggles, heads turning and wheels flying. Sometimes these repetitive elements are shown in split-screen, sometimes not – and sometimes it is even hard to tell since the very serial nature of the events at a racetrack visually echo the configuration of the split screen. The principles of narrative forward thrust – repetition, alternation, difference – are echoed on the visual plane with near-identical racing cars, motor parts, circuits, spectator masses, seasonal cycles etc. Besides this meta-discursive level, the split screen also visually evokes the sensation of being at a racetrack where the blurring of vision, the sound of the cars and the excitement of the crowd all contribute towards a transformed perception. This overwhelming sensation of speed, noise and chaos is translated into the multiplied images of the split screen.
Another major way in which the split screen was used in the late 1960s could be described as intra-diegetic, whereby the rear-view mirror of a car or a window in a room creates a framed image within the overall widescreen frame. This is not a split screen in the strict sense since it does not entail an obvious post-production intervention into the image, but rather, it entails a specific strategy of composition that emphasises the embedding of one image within another. A striking example occurs about one third of the way into the movie in an empty bistro when driver Pete Aron (James Garner) meets his colleague’s wife Pat Stoddard (Jessica Walter). In the master shot that frames the sequence a complex visual configuration is unfolding within the frame (fig. 3): Pete and Pat sit at the bar, centre screen with their backs turned to the camera. On the left-hand side the framed view through a window allows a glimpse of cars on a crowded country road, while on the right-hand side a mirror gives a front view of the two main characters combined with a partial view of the bar itself. While being a ‘realistic’ shot of an actual space in the sense that no post-production work on the image is visible, the composition nevertheless highlights three separately framed views within the same widescreen shot, thereby underlining the constructed and hence artificial nature of the image.
In fact, Frankenheimer himself claims that his use of the split-screen technique in Grand Prix was at least partly inspired by visiting the World Fair in New York in 1964/65, at which a number of multi-screen experiments were presented: ‘I got the idea for the split screen and the multiple images from Francis Thompson and his film To Be Alive at the New York World’s Fair. It impressed me a great deal as did Charles Eames’ film for IBM’ (Pratley 1969, 156). Indeed, the 1960s had seen an explosion of multi-screen environments that can be related to two distinct developments, one avant-garde and the other more mainstream. On the one hand, the art world began taking an interest in such devices as evidenced for example in works by Andy Warhol such as the multimedia spectacle Exploding Plastic Inevitable (EPI) (1966/67) with The Velvet Underground and the installation Chelsea Girls (1966). The EPI-spectaculars consisted of projections, music, colour schemes and designed environments all linked and coordinated to guarantee maximum impact. Chelsea Girls, on the other hand, prefigures many later video installations in a set-up that combines two images within one frame. Inhabitants of the famous Chelsea Hotel in New York are shown in split screens that are not designed in post-production, but in exhibition, since the piece is actually a multi-channel installation (the film comes with very specific screening instructions) overlaying two separate image sources. The artistic exploration of the multi-screen image culminated briefly around 1970 in the work of Stan VanDerBeek, Anthony McCall and Malcolm LeGrice, to name just a few exponents of the expanded cinema movement. In typical avant-garde fashion, the aim here was to break down the barrier separating art and life, by blowing up and expanding the traditional apparatus of the cinema.
On the other hand, the 1960s saw large companies, government institutions and research centres investing considerable attention to the development of immersive surround experiences that were presented at science expositions, trade fairs and general audience shows. Amongst the pioneers of these scientific-economic-educational environments, special mention must go to two pairs of innovators: Charles and Ray Eames, and Alexander Hammid and Francis Thompson. Without going into detail, it suffices to say that their environments and installations were designed to overwhelm and envelop the spectator in order to win them for some (political, ideological, or economic) goal. As Beatriz Colomina (2001) has demonstrated in a study of the Eames’ multi-screen work Glimpses of the USA (1959), these experiments connect the artistic exploration of multi-screen with scientific advances in data processing and perception.
Just like Frankenheimer, Richard Fleischer and his team for The Boston Strangler have acknowledged that the multi-screen experiments of large industrial exhibitions had a decisive influence on their appropriation of such techniques. In their case, it was Expo 1967 in Montreal which proved to have a lasting effect on their brand of mainstream filmmaking. Visual designer Fred Harpman states in interview: ‘We went up together to EXPO 67 to see the unusual film presentations and we got all excited over them’ (Harpman 1969). Director Fleischer claims ‘I realised the potential of this technique for the cinema due to my visit to the Universal Exposition of 1967 in Montreal.’ (Bourguin 1986, 99; my trans.) In The Boston Strangler, the split screen is put to encyclopaedic use in scope and variation, but it remains nevertheless within the confines of the continuity system. Thus, the split screen substitutes for patterns of shot-reverse shot (when the strangler communicates with a prospective victim via an intercom), eyeline matches (when a opened door lights up a dark room to reveal a corpse at the same time that we see frightened faces in another frame) and montage sequences (when the murderer is terrorising the city and the reactions of ordinary citizens are depicted as a montage of up to five split screens).
While the direct inspiration for the split screens of this era lies with the experimental innovations of world fairs and large expositions, the effective force of this technique is found elsewhere. The Boston Strangler constantly highlights the difference between cinema and its strongest rival at the time: television. The film’s narrative is framed temporally by two public happenings which were experienced at the time as live television events: the homecoming parade of the astronauts of an Apollo mission and the funeral of president John F. Kennedy. Consequently, the film often displays TV monitors running in the flats where the strangler is going about his business. Also, within montages of the fearful atmosphere in Boston at the time, television sets are regularly displayed within the widescreen image. The effect of this persistent display of television within the cinematic film is clearly to underline the aesthetic distance in image size and quality between the two different media. The situation of economic competition between cinema and television is mirrored inside the film as a situation of aesthetic competition: the widescreen image of cinema is able to display multiple TV images as well as demonstrating its superior quality in terms of colour and resolution. Here, the split screen becomes a marker of cinema’s aesthetic and technological otherness, functioning in a similar vein to other post-war additions such as colour and widescreen, also introduced as a means of product differentiation vis-à-vis television. Yet again, as TV adopted colour and grew in size (culminating in today’s flatscreen revolution), these differences slowly vanished and the split screen was put to other uses.
Hulk, Timecode, Ocean’s 13 and the Digital
From the mid-70s to the mid-90s split screens were rarely used within mainstream filmmaking. An exception to this rule is the work of Brian DePalma who has employed split screens consistently throughout his oeuvre. In DePalma’s films the split screen is embedded within a (neo-)baroque universe of elaborate camera movements, intricate framings, radical montage and narrative constructions characterised by repetitions, mirror images and picture puzzles – the cinema and its means of expression form a universe closed unto itself. DePalma employed split screens in the 1980s and early 1990s in films such as Blow Up (1983) and The Bonfires of the Vanities (1990) when it was definitely not agrave; la mode. Despite this dearth of the split screen for fifteen to twenty years, there is a noticeable increase in the use of this particular device from the mid-1990s onwards, not just in theatrical motion pictures, but also in television (news and sports), music videos and advertisements.
There are a number of explanations for this phenomenon: the availability of digital editing equipment which simplified the complicated process of constructing split screens in optical printing (Talen 2008), an awakening consciousness of style and layered visuality in television production (Caldwell 1995), and the proliferation of the overlapping windows in computer applications (Friedberg 2006). Whatever the reasons behind it, one can draw a number of conclusions about this phenomenon from some recent films which employ split screens in a variety of ways. Primarily, they show how the classical construction of time, space and narrative (transparent, unambiguous, well-balanced) has given way to a multi-dimensional and layered post-classical cinema. The most important lesson to be learnt from the use of the split screen since the mid-1990s is that the unambiguous time and space favoured by classical continuity seems no longer a central feature of contemporary film. The image can no longer be seen naively as a transparent depiction of an independently existing reality, but rather, has morphed into an information tablet, a display monitor or an instrument panel. Therefore, the strict division between realism and formalism, so long a mainstay of film theory, is giving way to a more fluid and malleable configuration in which the image shows signs of both paradigms simultaneously: it is an image that approximates our visual experience of the outside world, but it also foregrounds markers of difference and otherness.
A fitting example of this transformation is Ang Lee’s adaptation of Marvel Comic’s Hulk (2002) in which the split screen is used extensively, albeit in a different fashion from the previously discussed examples. Firstly, the split screen quite obviously adapts the comic format to the screen in which normally a number of panels are displayed side by side on one page, thus giving a simultaneous survey of what is usually viewed sequentially. Yet, the film stages its multiplied panels in more complicated ways as it refuses to be bound by the conventional wisdom of the split screen which dictates that the separate images take place at the same time. In a scene when the Hulk is transferred to a secret underground research centre under heavy security measures, this trip is shown in multiple windowed shots overlapping spatially (within the visible frame, but also within the filmic space represented) and temporally (they emerge and vanish at different moments, while also depicting overlapping timespans). Constantly changing in size and position, the shots are presented in rough chronological order yet nevertheless sometimes overlap temporally, creating an impression similar to scratching in music. This effect, which is not easily detectable on first viewing, points out the basically arbitrary and manipulable nature of the filmic images. In the same sequence, an even stranger configuration takes place when the photographic image of the Hulk being transported down a diagonal shaft transforms during a zoom-out (like the ones familiar from google earth) into a graphic representation of the same shaft as part of the wider cave system (fig. 4). A couple of shots later, this schematic overview of the underground research laboratory returns as the graphic display on a huge screen in that self-same centre (fig.5).
Apart from the computer-generated and hyperreal surface of the images, this sequence reveals another characteristic feature of current trends in film style: the position of the spectator in relation to the image becomes unstable and ambiguous. It is no longer clearly discernible how the spectator is located and positioned through point-of-view shots constructed by the film (through camera and mise-en-scene): from a more or less realistic representation of the captured super-hero the shot morphs seamlessly into a graphic schema of the cave system which then in turn becomes a worker’s field of vision from his monitor. Whereas classical cinema tended to establish clear spatial parameters for the spectator, often sharing a character’s point-of-view, the contemporary aesthetics of displays disorient the spectator’s anchoring in narrative space: are we watching the super-hero being transported; are we inside the computer or are we sharing the guard’s vision – are we outside or inside the spaces presented in the film? This dissolution of a fixed spectatorial position also effects the end of off-screen space. The rules of classical cinema, dictated that off-screen space was always just around the corner, being uncovered with every new cut to reveal further aspects of the film’s logical spatial dimensions. A spectator would anticipate an object glanced at in an eyeline match, a reframing in a match-on-action-cut or a reverse shot in a conversation. In many contemporary films however space is multiplied, layered and highly ambiguous, effectively erasing the traditional category of off-screen (ors champ).
Another film which clearly illustrates this transformation – albeit rather art house than mainstream (if these categories still make sense after all) – is Timecode (Mike Figgis, 2000). The film tells its story in real-time: 97 minutes of uninterrupted action without any cut. Yet, the whole time, the screen is divided into four quadrants, each split screen displaying the continuous take of one single camera. The digital cameras employed are extremely mobile, following characters around in hand-held fashion and reacting to new developments with reframings, zooms or pans, thereby evoking a sense of documentary realism or popular websites such as Wikileaks or YouTube with their spontaneous, raw and unedited look. Even though the four cameras are independent of each other, they are nevertheless concerted and coordinated as they share the same space in Los Angeles around the headquarter of the fictional production company Red Mullet (after Figgis’ own production company of the same name) and the same time – one-and-a-half hours on a Friday afternoon. Two features in the film point to this shared time and space: on the one hand, two minor earthquakes punctuate the action unfolding on the screen, demonstrating that all events take place simultaneously and in the same location: we notice the quake in all four quadrants simultaneously. On the other hand, as characters move about the premises of the company, they leave and enter the visual field of the cameras, even moving from one quadrant to the other, thereby demonstrating the contiguity of the spaces depicted, that is, in the first place, not logically discernible for the spectator.
This sense of overlap and layering is increased by narrative devices involving a series of phone calls, and people eavesdropping and spying on each other – underlining the complex relationships among the cast of characters. One example (fig. 6) occurs when Lauren (Jeanne Tripplehorn) placed in the foreground of the top left image listens in on the events of her bugged partner Rose (Selma Hayek) who occupies the bottom left image. At the same time Alex (Stellan Skarsgard), depicted bottom right, is on the phone to Bunny (Kyle MacLachlan) who can be seen twice: top left, through the window of the car in the background of the image in which Lauren appears, as well as top right behind Emma (Saffron Burrows). In the classical film system, a relation such as a phone call would tend to be translated into a string of shots alternating between the two speakers (a, b, a, b), whereas in Timecode the spectator is able to watch both participants at the same time, while deciding for herself whom to watch at any given moment. Again, it is not anticipation (provided by continuity rules) and expectation (of the reverse shot) that propels the film forward. Rather, the spectator makes these cuts as she shifts her attention from quadrant to quadrant. Yet, this heightened freedom on the part of the audience is tied back to the diegesis through the use of sound, at least in the theatrically released version. Here, sound is regulated in such a way that vital story information is more pronounced and louder (in the mix) than redundant conversations, so as to allow first-time spectators to understand the intricate maze of relationships as the film unfolds. This shift from image to sound as the prime carrier of vital narrative information is highly relevant as it reconfigures the classical convention for sounds to be subordinated to visuals. Yet again, this device undercuts the formal radicalness of the film as the spectator is guided by the design of formal parameters (sound volume, close-ups, framings) to concentrate on particular moments in the film.
A somewhat different take can be found in the popular blockbuster cean’s 13 (Steven Soderbergh, 2007) in which space is again subjected to many transformations. On one occasion, the split screen is employed less as a means to advance storytelling, but rather to flesh out a different, non-classical and ambiguous space. During several scenes in the casino, a triptych shows three adjacent views (fig. 7). Yet, the three images do not show different angles or even separate views, such as the action itself and a reaction shot of spectators, as traditional split screens would do, but rather, minimally dislocated views of the same scene. The effect is similar to the minimally overlapping time montage in the split screens of Hulk discussed above – the act of perception itself is foregrounded as one has to question these three adjacent images showing the same event while not quite fitting together visually. The patched overlaps between the three images foreground the endless possibilities of recombination that the digital image allows. It thus also reflects on the flexibility, modularity and potential transformations of the image in the computer – its overall unstable nature.
An even more extreme example occurs during the heist sequence at the core of the film when the computer system operating the gambling machines is switched off. Suddenly numerous players across the huge gaming room win large sums of money. The film depicts this by displaying numbers (indicating the sums of money) over the heads of players that appear like pop-up menus in bitmapped images familiar from navigating websites and CD-ROMs. Here, the film image is layered with meta-data – additional information that relates to the image, but that is not necessarily visible all the time (Manovich 2005). Moreover, the division between image, text and numerical data breaks down while traditionally these categories are divided and distinct. There is always something additional inside the image or beyond its visible borders, creating a layering effect that is not necessarily visual, incorporating data that is hidden like in a bitmap (highlights when the cursor crosses over).
The proliferation of the split screen since the mid-1990s highlights other transformations that cinema is currently undergoing in the digital age. In particular, attention is drawn to three inter-related and inter-dependent features. Firstly, the manner in which contemporary images seem to mix formerly distinct categories and registries, combining signs, text and data, for instance. Today, these different formats are often layered and interlaced within a single composite image. Secondly, the image is often presented, or appears to be presented, in real-time, following the control-tower logic of video surveillance and closed-circuit television (CCTV). This change has far-reaching consequences when one contemplates traditional interpretations of the photographic image as indexical and displaying a past presence. And thirdly, the image looses depth and space beyond the frame – the convention of off-screen space entering the image with every cut is vanishing into a perpetual present and presence. Thus, certain features that has been considered by various observers as specific to the medium of film, even though the exact specificity was open to debate (realism for André Bazin, for instance, or the capacity to display dialectical relations for Sergei Eisenstein), is now converging with other forms of audiovisual media.
The split screen – and this can be said across history, from the first examples in early cinema until today – finds a graphical solution for a paradox that lies at the heart of the cinema: the film image evokes a sense presence and yet what we see is absent. A film therefore connotes distance as well as proximity, and if we get involved with a film as spectators, a resulting confusion of self and other often ensues. This is also the foundation of spectator positioning in the cinema where one witnesses events that do not really take place (at least at that moment and at that place), where the sense of presence often overwhelms the knowledge of absence, even though sometimes the formal properties of the medium are foregrounded in the very act of reception. While this basic assumption is valid for all of cinema that functions according to the dominant ‘as if’ logic in which the spectator temporarily suspends her disbelief, the use of the split screen also emphasises the intricacies of media transitions. The three distinct episodes from the history of the split screen discussed in this article highlight three corresponding moments of media transformation: the everyday proliferation of the telephone in the 1950s, the diffusion of television in the 1960s and the explosion of digital media culture (the internet and the computer) in the 1990s. The films discussed above address these changes in terms of temporal and spatial mutations that act as allegorical configurations of changing media practices.
Bellour, R. 2000. The Analysis of Film. Ed. Constance Penley. Bloomington: Indiana University Press.
Bolter, J.D. and R.Grusin.1999. Remediation: Understanding New Media. Cambridge: MIT Press.
Bourgoin, S. 1986. Richard Fleischer. Paris: Edilig.
Bukatman, S. 1998. Zooming Out: The End of Offscreen Space. In The New American Cinema, ed. Jon Lewis, 248-71. Durham: Duke University Press 1998.
Caldwell, J.T. 1995. Televisuality: Style, Crisis and Authority in American Television. New Brunswick: Rutgers University Press.
Colomina, B. 2001. Enclosed by Images. The Eameses Multimedia Architecture. Grey Room 2 (Winter): 6-29.
Elsaesser, T. and W. Buckland. 2002. Studying Contemporary American Cinema: A Guide to Movie Analysis. London: Arnold.
Friedberg, A. 2006. The Virtual Window: From Alberti to Microsoft. Cambridge: MIT Press.
Hagener, M. 2008. Geteilte Bilder, getrennte Betten. Zur Verwendung von Splitscreen in US-amerikanischen ‘Sex Comedies’, 1955-1965. In Die Erotik des Blicks. Studien zur Film Aesthetik und Unterhaltungskultur, ed. W. Faulstich, N. Dable; M. Hagener, K. Rothemund, 25-37. Paderborn: Wilhelm Fink.
Harpman, F. 1969. Multiple-Image Technique for ‘The Boston Strangler’. merican Cinematographer 50 (2): 205.
Jeffers McDonald, T. 2007. Romantic Comedy: Boy Meets Girl Meets Genre. London, New York: Wallflower.
Lev, P. 2003. The Fifties: Transforming the Screen: 1950-1959. New York: Scribner.
Manovich, L. 2005. Metadata, mon amour. In Black Box – White Cube. Berlin: Merve.
Marchessault, J. and S. Lord, eds. 2007. Fluid Screens, Expanded Cinemas. Toronto: University of Toronto Press.
McCallum, E.L. 1999. Mother Talk: Maternal Masquerade and the Problem of the Single Girl. Camera Obscura 42: 70-95.
McLuhan, M. 2001. Understanding Media. London, New York: Routledge. (Orig. pub. 1964.)
O’Connor, Cara. 2004. When You Cut Up the Frame: Interview with Julie Talen. Senses of Cinema 30, http://www.sensesofcinema.com/contents/04/30/julie_talen.html (accessed May 15, 2008).
Peretz, E. 2008. Becoming Visionary: Brian de Palma’s Cinematic Education of the Senses. Stanford, CA: Stanford University Press.
Pratley, G. 1969. he Cinema of John Frankenheimer. London: Zwemmer / New York: Barnes.
Talen, J. 2002. 4: Split Screen’s Big Comeback. GlimpseCulture: Essays and Films by Julie Talen. http://www.glimpseculture.com/essays/24_splitscreen_comeback.html (accessed May 5, 2008).
Wulff, H.J. 1991a. Split Screen: Erste Uberlegungen zur semantischen Analyse des filmischen Mehrfachbildes. Kodikas/Code 14 (3-4): 281-90.
—–. 1991b. Ikonographie, Szenentransition, Narration: Zur Analyse der Beziehungen zwischen filmischer Form und filmischem Telefonat. In elefon und Kultur: Das Telefon im Spielfilm, ed. B. Debatin and H.J. Wulff, 127-65. Berlin: Spiess.
Youngblood, G. 1970. Expanded Cinema. Toronto: Clarke, Irwin &Co.
Zoller, M. 2007. laces of Projection: Re-Contextualising the European Experimental Film Canon. PhD diss., University of London.
1. I will stick to the term ‘split screen’ as it is most widely used, even though I agree with writer and filmmaker Julie Talen who states: ‘I don’t like ‘split-screen’ because it seems to imply only two’
2. The capacity of media to integrate other (older as well as newer) media has often been noted, for example in such classic works of media studies as MacLuhan (2001) or Bolter and
3. For a good introduction to the intricacies of the sex comedy see Jeffers McDonald (2007).
4. For a more thorough discussion of the sex comedy cycle in relation to the split screen see Hagener (2008).
5. For reasons of length I will not go into a detailed analysis of the sound track of the examples discussed here; surely, a more thorough investigation into the employment of sound would render similarly interesting results.
6. See Raymond Bellour’s (2000) seminal analyses of classical films such as The Big Sleep (Howard Hawkes, 1946), The Birds (Alfred Hitchcock, 1963) or orth by Northwest (Alfred Hitchcock,
7. For the origin of the term, see the contemporary text by Youngblood (1970). For a productive reappraisal, see Zoller (2007). See also the anthology by Marchessault and Lord (2007).
8. For the particular philosophy of the image in De Palma’s films, see Peretz
9. Here, I am following the distinction made by Elsaesser and Buckland (2002) between the ‘classical’ and the ‘post-classical’ as the ‘classical-plus’.
10.Psychoanalytical film theory has developed the concept of ‘suture’ for this relation between on-screen and off-screen space; no matter if one wants to follow this contested notion, the tension between on and off is not limited to psychoanalytically inflected theories.
11. For similar results in response to a different problem and in a different perspective, see Bukatman (1998).
12. The DVD edition of Timecode (London: Optimum Releasing, 2000) allows for choosing the sound of each of the quadrants, as well as an ‘unbiased’ mix which gives all four soundtracks the same volume.
Malte Hagener is Associate Professor in Media Studies at Leuphana-Universität Lüneburg and Visiting Professor at Ruhr-Universität Bochum. He is the author of Moving Forward, Looking Back. The European Avant-garde and the Invention of Film Culture, 1919-1939 (Amsterdam University Press, 2007) and (with Thomas Elsaesser) of Filmtheorie zur Einführung (Hamburg: Junius 2007) published in English by Routledge as Film Theory. An Introduction
through the Senses in 2009.
Contact Email: firstname.lastname@example.org