A cinema of presence and proximity: Gunvor Nelson's collage films and the aesthetics of the signaletic material before the electronic signal

A cinema of presence and proximity: Gunvor Nelson's collage films and the aesthetics of the signaletic material before the electronic signal

John Sundholm*

Film Studies, Karlstad University, Karlstad, Sweden


This essay argues for the importance of an intersubjective and impure film theory in which the signal and the signaletic is considered as figures for approaching film. This in order to make the argument that the signaletic mode indeed enables a novel perspective on moving image history. The aesthetics of the signaletic has thus a history preceding that of electronic media, in particular when it comes to animation and experimental film. When constructing such an archaeology, however, dichotomies into sign and signal should be avoided; otherwise, the complexity of many of the films is reduced. In order to illuminate the latter point, four films by Gunvor Nelson is analyzed (both analog and digital), showing not only how both sign and signal interact but also how the aesthetics of the signal and the signaletic material is not dependant on the electronic as such.

John Sundholm is Professor in Film Studies at Karlstad University. He has published extensively on experimental film and minor cinema and is the co-editor of several books on memory studies. He is also the editor of Historical Dictionary of Scandinavian Cinema (Scarecrow Press, 2012) and Gunvor Nelson and the Avant-Garde (Peter Lang, 2003). He has published work in English in journals such as Canadian Journal of Film Studies, Framework, New Cinemas and Studies in European Cinema. Together with Lars Gustaf Andersson and Astrid Söderbergh Widding he has written the first history on Swedish experimental film: A History of Swedish Experimental Film Culture: From Early Animation to Video Art (National Library of Sweden/John Libbey, 2010). He is also a member of the examination board of the PhD program in Fine Arts at the Academy of Fine Arts in Helsinki, Finland.

Keywords: film theory; experimental film; expanded cinema; Gunvor Nelson; Bruce McClure; Guy Sherwin

Published: 15 June 2012

*Correspondence to: John Sundholm, Film Studies, Karlstad University, SE-65188 Karlstad, Sweden. Email: John.sundholm@kau.se

©2012 J. Sundholm. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Citation: Journal of Aesthetics & Culture, Vol. 4, 2012 http://dx.doi.org/10.3402/jac.v4i0.18150


A major problem of film theory is that most theories are derived from narrative fiction film. This dominant position of feature film has resulted in a recurrent presence of narrative and representation as the primary film analytic tools and frames for understanding, which has led to a self-containing discourse and system in which the feature film and its theoretical counterparts, narrative and representation, feeds one another. However, the emergence of an increasingly heterogeneous audio-visual culture has problematized such established film theories. Today, we are amidst a situation in which different technologies and practices are blended: the film archives that used to be one of the last pockets of resistance against digital technologies utilizes digital technology in order to create digital intermediates when restoring celluloid film;1 the user friendliness of digital technology has included a vast array of different producers into what used to be a quite exclusive sphere of film producers, which in turn has fostered a growing experimentation with rules and conventions transgressing most of the established expectations regarding the current grammar of audio-visual culture. Not to mention YouTube and other online web-communities that have transformed film from being an exclusive object to becoming yet another format for interactive audio-visual interfaces.

Thus, all these recent changes have made the shortcomings of established film theory even more apparent and, therefore, a shift to the “signal” and the “signaletic” is a most welcome reconsideration of the field of contemporary film culture. What if the moving image would be considered within the framework of its effects and sensations that is forcing the viewer to active participation, to sharpen the senses, and to respond in a bodily fashion to film material and technology? Such interfacial dispositions of the moving image material and the viewers are not, however, only a result of a technological transition to the digital, to electronic media, and to the moments of instantaneous indexicality and feedback. Rather, what I want to draw attention to is that a shift to the signaletic, or the perspective of the signal, enables not only a fresh take on the heterogeneous media situation of today but also forces us to reconsider the history of moving image media. While the technological changes have without doubt promoted fresh and novel perspectives on audio-visual culture in general, I am cautious to consider the transition to the signal to be an ontological one or a change caused by technology alone. Many media are still being used according to old paradigms, a reminder of the fact that, although we invent technology, we shape it by using it and thereby assigning meaning to it; the meanings are not inherent in the technology.2 Thus, the signal is as much a potential, a discourse, and a social way of using artifacts, and therefore an archaeology of the signaletic is as important as the investigation into new forms and modalities of electronic media.

I think we partly got the film theories we have because film theory has not been exercised close enough to actual film practice that is always impure and dialogical. This is evident when looking at current filmmaking: today filmmakers may shoot on digital video but print on celluloid; or, shoot on celluloid, edit digitally, and project in whatever format depending on the time and place for the actual show or exhibition; in addition, out of the same primary footage on celluloid, there may exist short films for theatrical distribution, digital copies for loops in galleries and museums, or for online streaming or multi-channel installations. Such a heterogeneous situation and practice raises Edward Branigan's call for an “impure film theory,” which he summarizes into six points:

  1. grammatical (i.e. film theory should be nothing but a grammar of an ensemble of words);
  2. intersubjective (i.e. any film theory should not be conceived as fixed but tied to social practices);
  3. fragmentary (i.e. all film theoretical descriptions become partial and provisional);
  4. figurative (i.e. the abstract language of film theory are metaphorical transcriptions of concrete experiences);
  5. connected (i.e. the theories should not be conceived as hierarchical structures but as “heterarchical,” constituting parts in connection with other systems and structures);
  6. impure (i.e. the theories should not be considered as Theories, being specific to film or to medium specificity).3

My main point with bringing in Branigan's notions is to introduce the idea of the signaletic as a way of establishing a language and a metaphor in order to take a closer look at a particular strand in film practice, which is the experimental film tradition, thereby tracing the archaeology of the signaletic in film. As my material, I will use the Swedish-American experimental film veteran Gunvor Nelson's two collage films or animations, Frame Line (1983) and Natural Features (1990), both on 16-mm, the digital video Tree-Line (1998) and Natural Features Times 3 (2011), a digitally reedited projection of Natural Features divided into three images as a triptych. This also in order to avoid both a technological determinism and the dichotomization into modern and non-modern cinema, or time- and movement-image that many of the Deleuzian followers are all too keen to reproduce.4 Thus, I consider sign and the signaletic to constitute two different practices in making and experiencing film, but they should not be viewed as mutually exclusive practices or modes. Just as we may look at an old photograph in a family album and reflect on what it represents, we may equally consider it in terms of the material sensation it evokes: the design of the album itself, the paper, format and color resolution of the image, its connection with the time and place when it was taken, but also its connection to other similar acts, constituting a part in a network of commemorations included in the vast culture of photographic actions of memory that we become a part of as soon as we open a photograph album. The photograph album as an act is thus a crossroad of both signs and signals in a broad sense. Therefore, I would claim that it is quite counter-productive to stress a Deleuzian dichotomy between classical and post-classical cinema, non-modern or modern cinema, movement- and time-image, something that we should not accuse Deleuze of but some of his all too eager disciples.


One of the impure modes that really challenge established dichotomies of film theory is animation, and as Tom Gunning has noticed, animation seems to be one of Deleuze's blind spots.5 This not only because narrative feature film is Deleuze's primary material and backdrop for his reasoning, but because animation is often beyond all the different taxonomies, bringing the moving image back to its basic attraction and experience: the event of movement and the passage of time, the promise of eternal transformations and modulations, the immersion in the spectacle of the moving image—the signaletic material of the projected film.6 Curiously enough animation becomes the terra incognita in the Deleuzian narrative, perhaps because his concepts are mostly derived from classical cinema, out of which subsequent diverging forms and aesthetic modes are discerned in order to pave the way for the transition to the birth of modern cinema or to a transition from the movement-image to the time-image. Especially, the latter is developed into a philosophy of the art-cinema narrative and the human condition (much in the same way as Stanley Cavell develops his philosophy into a narrative about classical cinema and the human condition during that era).7 I would, therefore, rather suggest a reversion of the main strand in the reception of Deleuze's film philosophy that is too often seen as culminating in the time-image. The notion of the signaletic is thus a mode that would bring us back to the founding moment and event of film, namely movement. Therefore, I will use Sean Cubitt's elaboration of Deleuzian and Piercean philosophy and his taxonomy in “pixel,” “cut,” and “vector” in order to stick to a more “heterarchical” model and merely considering the cubittian principles as three cinematic principles of the filmic event.8 Pixel being the sheer appearance and sensation of the moving image material; cut the organization of the same material; and vector the interpretative maneuvers in relating the moving image material to something else. The event itself constitutes the primary moving image happening; as Deleuze also stresses, film is not photographs in movement; it is not a copy of reality but constitutes a reality as such. This is emphasized by Gunning too, who in his efforts to detach film theory from the grip of photography—and the regime of representation and sign—states that motion and projection is not only the “real thing,” that we in fact not “just see motion,” but “we feel it in our guts or throughout our bodies.”9 It is this presence and proximity that so many filmmakers wanted to explore who worked with expanded cinema performances or in the tradition of structural film, but that also constitute the early film experience, a cinema of attractions.10 Thus, “firstness” and “pixel” are the signaletic moments that are part of any filmic effect, and included in the other principles as well, but which is never reducible to some other principle (just in the same way as firstness is always included in secondness in C. S. Pierce's semiotics, and secondness in thirdness, whereas firstness never may contain secondness or secondness thirdness).

One of the most original contemporary artists of expanded cinema performances is Bruce McClure who emphasizes the very moment of movement and projection through his use of multiple 16-mm projectors loaded with different film loops of slightly variable lengths. In this way, McClure is able to epitomize movement, creating a cinema of firstness or, of affection or that of the pixel, forcing the viewer to participate in the real event of the projection of moving images. This is further epitomized by his use of sound. He connects guitar effect pedals to the projectors, modulating the optical sound track, thus playing with and “on” the sound track of the celluloid loops. Film is thus brought back to its origin becoming “a-signifying plastic mass” that forces the viewer to experience the projection as a physical event, which is reinforced by the aggressively flickering screen and the loud sound so, both image and sound reverberates in the body of the viewer. As McClure himself expressed it in an interview, one aim is to have the audience feel “the burn” of the performance. This transformation of the moving image experience from that of narrative complementation into sheer physical sensation is also stressed in McClure's performances while he prefers to delay the very act of showing any image whatsoever in the beginning of his performances. At first, there is only a flickering square on the wall after which images begin to appear, thus reminding us of the fact that moving images are the production of the projector and that the images are only one by-product of the interaction between man and machine, perception and projector. The image is merely one dimension in the signaletic event of the performance, stressing that the film event is not the same as experiencing moving images—as the film experience is mostly conceptualized—but that of being immersed in a live audio-visual projection (Figure 1).

Fig 1
Figure 1. Bruce McClure: Cycloptically Through the Windowdisks, performance at Kristinehamn Art Museum, Sweden, 2010. Documentation by Daniel Alegi.


Whereas McClure pushes conventional film technology to its limits in his cult of the cinema of firstness, Gunvor Nelson's work is much more unobtrusive.11 Her career started in the 1960s in the Bay Area experimental film community, and her work from the 1960s and the 1970s intersects with the major trends of experimental film culture at the time, ranging from expressive explorations of sound and image to lyrical films about the facts of life. Especially, the structural film tradition of the late 1960s and the 1970s may be considered as one strong legacy in the archaeology of the signaletic. For example, Paul Sharit's flicker film T, O, U, C, H, I, N, G (1968) shows a man with his tongue in a pair of scissors, as if someone was to cut his tongue off. Due to the viewer's active perception, the moving imagery is seen as a tongue being cut off in a repetitive, flickering motion. Another important strand is the multimedia environment of the 1960s and the early 1970s in the Bay Area that did not only foster experiments with various media but also across media and which gave birth to Stephen Beck's legendary “Beck Direct Video Synthesizer” (1970) and the later “Beck Video Weaver” (1973). Beck's direct transmutations of analog or digital signals had its equivalent and predecessor in the Optical Printer that became a favorite tool among experimental filmmakers for transforming and manipulating the film images that they had shot.12

British artist Guy Sherwin's work addresses similar questions as Sharit's film although sound is here much more to the fore. His work Cycles (1972/1977) consists of sections of paper dots attached to both film leader (transparent film) and the soundtrack, including holes that are punched into the film. When projected, the gaps, or interferences in the filmstrip and its soundtrack, disappear at a certain frequency and creates a coherent audio-visual entity that is reaching up to 72 sound beats per second. Even more radical is Sherwin's Sound Cuts (2007) in which black film is repeatedly cut and rejoined in such a way that the cut is visible, creating a flash of light and a thud of sound whenever the filmstrip is passing through the film gate of the projector. When such loops are projected with several cameras from different angels, a performance of the signaletic material of both the projector and the celluloid film is created.

Both Sharit's and Sherwin's work show how the film material is considered as a signal, a transmission, which may be forever modulated, manipulated in order to create different experiences in which our physiological capacities are tested and in which our aural and visual perception is taken for a ride. In Gunvor Nelson's work, however, all these material imprints are integrated into a vast assemblage of techniques, covering pixel, cut, and vector, thus not leading exclusively to a signaletic aesthetic. Instead, the signaletic becomes one important path and mode, demonstrating that all those technical features that Gene Youngblood, Peter Weibel, and Steina and Woody Wasulka deemed essential for the shift to the code were already there in traditional film.

In their essay “Cinema and the Code,” published for the first time in 1989 in Youngblood's name, they distinguish four new features of electronic moving image media:

  1. image transformation (in contrast to transition in cinema);
  2. parallel event-streams;
  3. temporal perspective;
  4. the image as object.13

Youngblood et al. do state that these characteristics are not exclusive to electronic media and the signal, but that what is new is that they are now within reach of manipulation and modulation by a simple push of a button, whereas to accomplish the same thing with celluloid film was highly time-consuming. What is important for my point in this context, however, is to stress that the devices described by Youngblood may be used both in traditional film and electronic media. This is in order to emphasize the point that it is important to approach the signal and the signaletic as a language for addressing another mode for audio-visual aesthetics and not restrict the language to a certain technological feature.

Gunvor Nelson's first collage film, Frame Line, is a 16-mm film shot in black and white. The film follows no narrative except for that it is based on Nelson's experience of returning back to Sweden and Stockholm in the early 1980s. Her encounter with Sweden and Stockholm as places is mostly displayed in the film as an encounter with images of Stockholm and of Sweden (while she was planning the film and applied for using the facilities of the Stockholm film workshop, Filmverkstan, she described the film as a series of glimpses).14 After the initial intersecting shots that show someone kneading dough and a camera that is moving out on the Stockholm waterfront, follows an animated sequence displaying how Nelson is painting over some of the imagery that she has taken of Stockholm. Soon, a set of animated lines appears, superimposed on the footage of the hands that are kneading the dough and the footage of the waterfront (Figure 2).

Fig 2
Figure 2. Gunvor Nelson: Frame Line (1983), 16 mm.

Already during this first minute of the film Nelson introduces image transformation (by using animation and superimposition), parallel event-streams (cutting back and forth between the dough and the movements out on the sea), temporal perspective (liberating the viewer from a specific position in both time and place, moving freely around as the camera is seeking its way out on the water), and the image as object (inserting animated sequences of whatever visual imagery Nelson picks up during her 22 minutes long journey in the film, manipulating and changing the images, painting, cutting, and reworking the material). These opening moments carry a highly ambivalent and disharmonious soundtrack that consists of various sounds and musical themes, singing voices, and a voice saying “yes” and “no” in Swedish. Especially prominent is a repetitive machine-like sound that is similar to the noise from a turntable when the pick-up has finished the last track but the turntable keeps spinning and the stylus is still picking up the sound.

Already with this beginning, Nelson moves from the regime of the sign to that of the signal, modulating, interfering, and playing with the means of the cinematic, creating a cinematic event that is not transferrable into a pattern of signs and corresponding referents. Rather, by taking a close look at the material at hand and interfering, she creates a signaletic material that the viewer is free to explore, thus supporting the dictum of Deleuze that “[i]t is at the level of the interference of many practices that things happen.”15 But, she is not reducing the film to an “a-signifying mass” either, as she continuously cut and develop the work, but without creating a story line or establishing some other modus that would play down the signaletic experience, that of audio-visual matter moving and vibrating out of the screen.

With Natural Features, Nelson ended her series of five collage films. This work contains no conventional image material, as if shot in a window-like manner using the camera as a registering, representing machine. Instead, the film consists of image and sound-events or happenings, in which objects and images clash with each other creating an endless journey into image and sound. In this way, the film is made to be inherently present; there is no tension between the bygone past of the camera image and the present presence of the animated transformations as in the previous collage films. The animation techniques used are meticulous but raw, including moments of randomness and physical challenges. For example, several of the complex takes in which images are reflected—and often distorted—were shot while Nelson was holding a mirror in one hand and painting a picture that she had attached to herself in the other, thus not being in full control of what was happening. In Natural Features, literally everything is moving and constantly changing, though not in accordance with any predetermined schedule, but in accordance with the principles of animation whereby we as viewers watch the transformation and entanglements of lines and figures. This live-action and live-painting, her way of working with animation was, as Youngblood et al. expressed in their essay, intensely time-consuming and complicated; therefore, it was no coincidence that Natural Features became her last collage film. The film was also so saturated with image information that the audience was simply baffled (Figure 3).

Fig 3
Figure 3. Gunvor Nelson: Natural Features (1990), 16 mm.

It was quite logical that Nelson should return to Natural Features reworking it digitally as it already had a “digital look,” being a work of infinite image-transformations that simply ignored traditional parameters of time and space, perspective, and viewer positions, denying the tradition of photography and film as constituting a window out to the world. In Natural Features, the surface of the screen is the place where the cinematic events unfold, we are so to speak beyond the logic of the camera and instead immersed in audio-visual events.

In Nelson's first video work Tree-Line, the idea of the camera is, however, reintroduced, making the video into a highly interesting reflection on the transition from celluloid to digital. The video is foremost made out of footage that accompanied Premier's software at the time and that shows a train that is passing by. Besides this sequence, Nelson has added another shot, an image of a tree that she has taken from a book. Thus, the video is completely created out of already existing material (as also the sound of the moving train that is taken from a sound archive). In Tree-Line, Nelson places the viewer on this side of the train so it is only possible to see the tree as it occasionally appears in between the wagons as the train passes by. After this primary set up, Nelson starts to play with sound and image, changing perspectives, changing colors, shapes, and the size of the train and the tree, but always returning to the perspective in the beginning: someone watching a train passing by and getting sudden glimpses of a solitary tree. Thus, although Nelson works with software and an electronic signal, there is an ongoing dialectics between sign and signal, representation, and production, in which we move between the two different registers. Thus, Tree-Line both simulates the capturing camera of conventional film-representation and is an advanced journey into image transformation, a playing with the software. In fact, in comparison with Natural Features on 16-mm, Tree-Line looks more like a film when it comes to the language used, but the pixelation of the image reveals the medium that is being used (Figure 4).

Fig 4
Figure 4. Gunvor Nelson: Tree-Line (1998) Digital video.

With Natural Features Times 3, Nelson returns to her signaletic project of creating endless image transformations. The form of the triptych enables her to triple the image information that existed already as a potentiality in the original film, but she has also chosen to slow down the pace of the film. These cuts are, however, not a way of organizing space in the film but merely a device for adjusting the rhythm. The celluloid version of Natural Features is, therefore, much more an event of movement whereas the triptych has spatialized the movement due to the projection that shows three distinct images. The images are related to each other as visual image transformations and not as much as an event of movement.


As I have shown, sign and signal, when considered as language-games in the social field of film theory, need not to be held as mutually exclusive but instead constituting vital elements in any moving image act or event. Experimental film has a long tradition of exploring the signaletic material of film, in particular when it comes to different forms of expanded cinema actions. But, there are also more subtle ways of using the notion of the signal and the signaletic in order to establish a more profound relation to some moving image work, as in this case, the work of Gunvor Nelson and her collage films in particular. I consider Nelson to be one of the most interesting interventors between celluloid and the digital, the sign, and the signal, exploring the whole range of cinematic devices, pixel, cut, and vector and not restricting any of them to a specific medium. In this sense, she is an impure artist in the history of experimental filmmaking.

The title of Nelson's first collage film is telling, Frame Line. It suggests that she is liberating film from the idea of the frame as constituting a necessary window out to the world and instead discovers the line, which is one of the main devices in animated image-transformations transgressing the established conventions of time and place in film grammar.16 The transition from frame to line also entails the transition from that of moving images to movement as such, stressing the point that the projected audio-visual movement is a reality in itself. Film is not a medium for representations but for interventions and experiences. But, Nelson's way is not either to liberate film in full, in order to abandon the regime of the sign or representation. As I have shown, Tree-Line is also a return to the act of image representation. Nelson's most signaletic work, Natural Features, is made on celluloid and is a film that pushes animation toward its limits, demonstrating that animation plays a major part in the archaeology of the aesthetics of the signaletic and the signal.


1. For an excellent overview, see Giovanna Fossati, From Grain to Pixel: The Archival Life of Film in Transition (Amsterdam: Amsterdam University Press, 2009).

2. For a discussion of technological determinism, see Does Technology Drive History?—The Dilemma of Technological Determinism, ed. Merritt Roe Smith and Leo Marx (Cambridge, MA: MIT Press, 1994).

3. Edward Branigan, Projecting a Camera. Language Games in Film Theory (London: Routledge, 2006), 19-20.

4. This is especially reproduced in numerous essays or book chapters. Deleuze is not always crystal clear on this point either. Damian Sutton, for example, argues that time and movement-image should be considered as two different modes that are not in opposition to each other, whereas, Jacques Rancière claims that we are rather passing “from one side to the other of the same images” and not “from one family of images to another”—thus time- and movement-image are two different aspects, not two different images as such. For Deleuze, the movement-image is the prerequisite for the time-image. Damian Sutton, Photography, Cinema, Memory: The Crystal Image of Time (Minnesota: University of Minnesota Press, 2009), 40; Jacques Rancière, ‘From One Image to Another? Deleuze and the Ages of Cinema’, in Film Fables, ed. Jacques Rancière (Oxford: Berg, 2006), 113.

5. Tom Gunning, ‘Moving Away from the Index: Cinema and the Impression of Reality’, Differences 18, no. 1 (2007): 29-52.

6. There is also, I would say, in Deleuze's arguments, a recurrent conflation of plot with film material or the various types of images, i.e. narrative structure or development is often used as the foundation for analyzing or defining different types of images, not the film material as such.

7. Stanley Cavell, The World Viewed. Reflections on the Ontology of Film. Enlarged edition (Cambridge, MA: Harvard University Press, 1979).

8. Sean Cubitt, The Cinema Effect (Cambridge, MA: MIT Press, 2004).

9. Gunning, ‘Moving Away from the Index’, 39.

10. For expanded cinema, see also Ulla Angkjaer Jorgensen's essay in this collection and the seminal catalogue X-Screen: Film Installations and Actions in the 1960s and 1970s, ed. Matthias Mihalka (Köln: Walther König, 2004).

11. I want to stress that my thoughts presented throughout this essay have not primarily risen out of reading theory but are based on my own experience of the collage films by Nelson and of the obvious shortcomings of the existing theoretical work that has addressed her films (including my own).

12. For an overview of the period see, for example: Looking for Mushrooms: Beat Poets, Hippies, Funk, Minimal Art, San Francisco 1955-68, ed. Barbara Engelbach, Friederike Wappler and Hans Winkler (Köln: Walther König, 2008) and Radical light: Alternative Film & Video in the San Francisco Bay Area, 1945-2000, ed. Steve Anker, Kathy Geritz and Steve Seid (Berkeley: University of California Press, 2010), in particular the essay by Tony Reveaux, ‘A Legacy of Light’. ibid., 104–10.

13. Gene Youngblood, ‘Cinema and the Code’, in Future Cinema: The Cinematic Imaginary after Film, ed. Jeffrey Shaw and Peter Weibel (Cambridge, MA: MIT Press, 2003), 156–61, originally published in Leonardo, Computer Art in Context Supplemental Issue (Oxford: Pergamon Press, 1989), 27–30.

14. Letter to Filmverkstan, May 22, 1980, Archive of the Swedish Film Institute, Filmverkstan-collection.

15. Gilles Deleuze, Cinema 2. The Time-Image (Minneapolis, MN: University of Minnesota Press, 1989), 280.

16. Frame line is of course also a technical term signifying the unused space in-between two frames on the film strip. Thus, “Frame Line” may also be interpreted as suggesting a film that is in-between the established film grammar.

About The Author

John Sundholm


Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Related Content