Navigazione – Mappa del sito
ontologia del cinema

Digital Technology, Indexicality, and Cinema

David Davies
p. 45-60


My principal concern in this paper is to determine how fundamental are the implications, for the moving image as a medium for works of art, of the loss of indexicality consequent upon the move from a photochemical medium to a computational medium. I begin by examining the bearing of indexicality on the status of still images as photographs, or as photographic art. I then consider whether, where indexicality is lost through the use of digital technology, this has the same significance for still and for moving images. I argue that indexicality does not have the significance for cinema that some have claimed, and that the loss of indexicality does not affect in any essential way the identity of cinema.

Torna su

Testo integrale

The artistic significance of technological change

1Changes in the availability of materials and resources are obviously key factors in the evolution of the various arts and sometimes in the genesis of particular artworks. Only with the development of acrylics, for example, did the “staining” techniques pioneered by Morris Louis become possible. And only with Fox Talbot’s development of the calotype did photographic works become capable of multiple instantiation. In music, the development of synthesisers and contemporary sound engineering technology have enabled the production of both live and recorded acoustic soundscapes impossible with traditional instruments, And, in a broader sense of technological change, the emergence of a sophisticated form of musical notation in the late 18th century was a precondition for the composition of the kinds of “thick” repeatable musical works that we associate with 19th century Romanticism.

  • 1  I follow here the standard distinction between singular and multiple artworks. For a critical exam (...)

2Even from this brief survey, it will be apparent that some impacts of technological developments on the arts are more fundamental than others. Most fundamental are cases where the very nature of artworks in an artform is affected, as with the development of the calotype process in photography. The first enduring images produced photographically, that is, by “writing with light”, were daguerrotypes. In the daguerrotype process, light from a subject directly causes a unique positive image on a highly polished metal surface. The calotype process, on the other hand, is two-staged. Light from a subject produces a negative image that requires re-photographing in order to produce a positive image. As a result, while only a single entity can have the history of making required to be a genuine instance1 of a daguerrotype, a calotype can have a multiplicity of such instances, since the negative can be re-photographed multiple times. Fox Talbot’s technological innovation therefore transformed photography from a singular art – comparable to painting in this respect – into a multiple art – comparable to engraving. While the photographic artwork still articulated its content through an image produced through a process of “writing with light”, it now admitted of a plurality of instances through which this content could be made accessible to receivers. In one significant respect, therefore, the ontological status of photographs, and of photographic artworks, changed.

  • 2  See Steadman 2001, Hockney 2001.
  • 3  Whether and to what extent such matters bear upon the appreciation of artworks depends upon the ex (...)

3In other cases, while technological change does not affect the very nature of works within a given art form, it broadens the range of artistic values realizable in practice in such works. For example, the availability of acrylics augmented the range of intentional visual values realizable in paintings. Sometimes, however, technological change alters neither the ontological status of works nor the range of values realizable in a given art form, but, rather, the ease with which certain kinds of artistic values can be realized. Consider, for example, the effect on works by 15th century and post-15th century European painters of the availability of serviceable forms of optical devices such as the camera obscura, if recent studies of these matters are to be believed2. Or consider the impact of the word processor on late 20th century literary art. In such cases, while the range of available artistic values may not be increased, the changes arguably have a bearing on the appreciation of works. For the use of the camera obscura bears upon the painter’s achievement3.

  • 4  Wallis 2005: 19.
  • 5  Manovich 1995. While Manovich is better known for his 2001, this paper, discussed in Wallis 2005, (...)

4It is widely assumed that the move from photochemical to digital technology in those arts that deal with still or moving images generated by technological means constitutes a fundamental, and not merely a superficial, change in those arts and in the nature of such images. This is a natural assumption to make. For, as we shall see, it has traditionally been held that the basic properties of camera-generated images, and the very possibility of such images as art, are intimately linked to the technological apparatus employed in analogue photography. Clearly, as Holly Wallis maintains, «film-making equipment influences the kinds of images that can be made, as well as the ways in which stories can be told»4. But should we also hold, with Lev Manovich, that «digital media redefines the very identity of cinema»5. He arrives at this conclusion by generalising to moving images William J. Mitchell’s claim that digitally produced still images are closer to painting than to photography. According to Manovich, with the use of digital technology in the production of moving pictures, the essential indexicality of traditional cinema is lost. Wallis agrees with Manovich that digital technology deprives the still image of its indexical relation to the world, a relation taken to be in part constitutive of a photographic image.

5My principal concern in this paper is to determine how fundamental are the implications, for the moving image as a medium for works of art, of the loss of indexicality consequent upon the move from a photochemical medium to a computational medium. I begin by examining the bearing of indexicality on the status of still images as photographs, or as photographic art. I then consider whether, where indexicality is lost through the use of digital technology, this has the same significance for still and for moving images. I argue that indexicality does not have the significance for cinema that some have claimed, and that the loss of indexicality does not affect in any essential way the identity of cinema.

Indexicality, photography and the requirements for photographic art

6A sheet of canvas covered with paint marks, an image printed from a photographic negative, and a sequence of images and sounds projected onto a screen may all be the material manifestations of artworks. In each case, we are presented with a perceptual manifold. An artistic interest in such a manifold is an interest in an “artistic content” articulated through it. A work’s “artistic content” comprises all of those lower and higher order representational, expressive, and formal properties rightly ascribable to it. The presented manifold thereby serves as the work’s “artistic vehicle”. Such an artistic interest is “interrogative” in that it seeks to make sense of the manifold in terms of reasons for its being ordered in the way that it is. For example, central to our interrogative exploration and appreciation of a painting like Jan Vermeer’s The Art of Painting is the question: what is the reason for (and not merely the cause of) the marks that make up the visible manifold being ordered in the way that they are? How we answer this question bears crucially on the artistic content we take to be articulated through that manifold. It is because Vermeer deliberately generated this pictorial array in a given art-historical context that we rightly view The Art of Painting as having these representational, expressive, or formal properties.

  • 6  For a survey of early formulations of this kind of argument against the artistic pretensions of ph (...)

7Such reflections on the conditions under which an artifact can serve as an artistic vehicle underlie a much-rehearsed objection to the artistic pretensions of photography. The process whereby a photographic image is produced, it is claimed, is “causal” or “mechanical”. While the photographer sets up the photographic apparatus in a given location so that only certain things fall within the range of capture of the lens, the resulting image is determined, without further human intervention, by a purely causal process. It is, as we might say, nature, rather than any human agent, that “writes with light” upon the light-sensitive surface, hence Fox-Talbot’s famous characterisation of photography as “the pencil of nature”. As a result, the intelligent agency of the photographer does not play a significant role in ordering the elements of which the image is composed. Thus, in contrast to a painting, we cannot see that agency manifested in the design of the resulting image. And only where we can see intelligent agency manifested in the design of a representational image, expressing a cognitive perspective on the imaged subject, can we legitimately take the representational content of the image to be a proper object of artistic interest6.

8We can express this challenge in terms of a widely drawn distinction between what we may term an artwork’s “physical” or “material” medium and its “artistic” medium. As we saw, an artist articulates a particular artistic content by means of an artistic vehicle. In the case of a painting, the artistic vehicle is a visual array displayed on the wall of a gallery. But we may avail ourselves of different vocabularies in describing that array. Described in terms of its material or physical medium, it is a surface with a certain material constitution upon which various marks can be visually discriminated. Here we abstract from any kind of intentional activity that might explain the overall disposition of marks upon the surface. However, when we describe the artistic vehicle of a work of visual art, we normally employ expressions that implicate the intentional agency we take to have been involved in its coming to display this disposition of marks. We talk, for example, of “brushstrokes”, “composition”, “design”, and “impasto”, thereby describing the manifold in terms of an artistic medium. It is through its visible surface apprehended under such a description that we relate the artistic vehicle to the artistic content that we take to be articulated through it. From a description purely in terms of the material medium, nothing follows concerning such an artistic content. However, as noted, to describe a manifold in terms of an artistic medium requires that we apprehend its features in terms of the intentional agency taken to be responsible for them. If, then, the manifold presented to us by a photograph is a result of a mechanical rather than an intentional process, or if we cannot tell which features of the manifold do and which do not result from an intentional process, then we cannot avail ourselves in a principled way of an artistic medium to describe the manifold. Thus we can only describe it in terms of its material medium, and cannot see it as the articulation of an artistic content.

  • 7  See Davies 2008.

9The practice of traditional photographers provides an implicit answer to this criticism7. First, as a direct or indirect function of the distribution of lines and masses in the image, the pictorial composition of a photograph refers the viewer to the agency of the photographer and to the expressive purposes manifest in that composition. Composition indicates how the photographer intended the image to be viewed, the path the eye is intended to follow, the way the subject is meant to be looked at. Thus what Cartier-Bresson termed the “geometry” of a photograph serves both as the necessary stimulus to a detailed scrutiny of the image, and as a necessary guide to the viewer in her engagement in this scrutiny, leading her to see the subject in a particular way. Second, familiarity with a photographer’s style aids our attempts to read, in a photographic image, the expressive intentions of its maker.

10The objection levelled at traditional photographic images fails, therefore, to acknowledge that we apprehend an exhibited photograph as the product of a process as deeply intentional as it is causal. Our interest in such photographs as photographs is properly artistic because it is an interest in how the photographer has chosen to represent what we may term the “bare subject” of the photograph, and in the way in which she has employed the photographic medium for these purposes. And the object of such an interest is what might be termed an “intentionally rendered trace”, a product of both agency and mechanism. A photographic image is a trace to the extent that the direct product of the photographic event has a non-intentional counterfactual dependence upon the thing photographed given the intentional activity that precedes the taking of the picture. The photographer selects her subject, decides how it will be shot, and takes the picture. The direct product of this action – for example the negative – would have differed, in counterfactual situations, in ways that systematically, and independently of the photographer’s intentional states, track differences in those features of the thing photographed falling within the scope and focus of the lens of the camera.

  • 8  Manovich 1995.

11It is in virtue of being mechanical recordings in this sense that photographic images are indexical. Just as bear tracks in the sand indicate the recent passing of a bear, so the distribution of marks on the surface of a photograph indicates certain aspects of pro-filmic reality at the time when it was taken. It is this fact about the making of a photographic image that leads Manovich to describe traditional cinema as grounded in “deposits of reality” and as «an attempt to make art out of a footprint»8. But, as we have seen, such an attempt can succeed, at least in the case of still photographic images of the sort generated by Cartier-Bresson, For, while the bare subject of a photograph is determined mechanically, the photographer can make manifest the role of agency by her way of rendering the particular trace presented to us. Thus we have a way of describing the photographic manifold in terms of an artistic medium, thereby answering the earlier objection.

12Indeed, only when such an interest is possible do we have a work of genuinely photographic art. This follows from more general considerations about the ways in which artworks are classified by reference to their media. Something is a work of x-art, where “x” designates an artistic medium, when an x-manifold or structure serves as the artistic vehicle for the work. It is through interrogative engagement with this x-manifold or structure, appropriately contextualised, that we determine the work’s artistic content. A work of photographic art, then, is an artwork for which a photograph serves as the artistic vehicle. Photographic art thus requires a delicate balance between mechanism and agency. On the one hand, the image must originate in a trace-making process of “writing with light” if it is to be photographic. On the other hand, we must be able to refer details of the image to the agency of its maker – bring those details under an artistic medium – in determining the work’s content if it is to serve as the artistic vehicle of a work. We saw above how this balance has been struck in traditional analogue photography.

Indexicality and the digitally generated image

  • 9  Mitchell 1992.

13Indexicality is central to a still image’s being photographic because it is a consequence of the trace-making process whereby photographs are generated. A moving image that is generated by a photographic process will inherit this indexicality. As noted above, however, a number of authors have maintained that one loses indexicality, and consequently the photographic status of the image, when an image is generated digitally. An influential proponent of such a view of still images is William J. Mitchell9.

  • 10Ivi: 7.

14Mitchell argues that the digital image is non-photographic because it is shorn of those features traditionally held to distinguish photography from painting. Whereas the traditional photographic process allows the photographer relatively little opportunity to change the resulting image, the digital image is “inherently mutable”: «The essential characteristic of digital information is that it can be manipulated easily and very rapidly by computer. It is simply a matter of substituting new digits for old […]. Computational tools for transforming, combining, altering, and analyzing images are as essential to the digital artist as brushes and pigments to a painter»10.

  • 11Ivi: 31.
  • 12Ivi: 225.
  • 13Ivi: 7.
  • 14Ivi: 4.

15Furthermore, because we are aware of this mutability, we cannot with any confidence draw conclusions about the nature of pro-filmic reality from features of the digital image. In looking at the latter, we have no way of determining which features of the image indicate aspects of the pro-filmic world and which reflect the intentional activity of the photographer. This calls into question the traditional way of marking the difference between (indexical) photographs and (non-indexical) paintings: «The distinction between the causal process of the camera and the intentional process of the artist can no longer be drawn so confidently and categorically. Potentially, a digital “photograph” stands at any point along the spectrum from algorithmic to intentional»11. Traditional photographs «were comfortably regarded as causally generated truthful reports about things in the real world, unlike more traditionally crafted images, which were notoriously ambiguous and uncertain human constructions […]. But the emergence of digital imaging has irrevocably subverted these certainties, forcing us to adopt a far more wary and more vigilant interpretive stance»12. The digital image is a kind of «electrobricollage»13 which may draw on various kinds of digital processes in a way that is inscrutable for the viewer. Thus, Mitchell maintains, «although a digital image may look just like a photograph when it is published in a newspaper, it actually differs as profoundly from a traditional photograph as does a photograph from a painting»14. The inherent mutablity of the digitally generated image thus undermines its ability to function as a trace or index, and thus its status as photograph.

  • 15  Manovich 1995.
  • 16Ivi:.
  • 17  Wallis 2005: 33.

16Manovich draws stronger conclusions. First, where Mitchell claims that digital photography departs from analogue photography as much as the latter departs from painting, Manovich claims that the inherent mutability of the digital image «erases the difference between a photograph and a painting»15. Second, given the further assumption that what goes for digitally generated still images applies equally to moving images so generated, he maintains that «since a film is a series of photographs, it is appropriate to extend Mitchell’s argument to digital film. With an artist being able to easily manipulate digitized footage either as a whole or frame by frame, a film in a general sense becomes a series of paintings»16. For Manovich, the difference between traditional cinema and digital cinema resides in the former’s indexicality. Digital cinema «redefines the very identity of cinema» because the essentially indexical nature of traditional cinema is lost due to the role that digital cinema accords to “manual” manipulation of the image. Wallis follows Manovich in extending Mitchell’s conclusions from still to moving images: «Whereas film stock registers an indexical connection to the world through the impact of light striking, and altering, emulsion, thereby forming a physical and ontological connection to the world, [digital video’s] transformation of light into digital information severs that connection»17.

  • 18  Gaut 2010: 47-49.
  • 19Ivi: 16.

17A number of issues arise out of Mitchell’s original claims and the extensions of those claims by Manovich and Wallis. First, as Berys Gaut has recently argued18, while digital technology makes possible the generation of images in ways that either subvert or exclude indexicality, this still allows for digitally generated images that qualify as photographic. In such cases, the digital techniques employed are restricted to ones that preserve the trace status of the resulting image. Gaut distinguishes three ways in which digital technology can enter into the generation of an image19. First, it may play a part in encoding the results of a trace-making process of “writing with light”. Most obviously, this occurs when pro-filmic events are captured with a digital camera. Such a camera may differ from a camera employing photochemical technology only in how it records the light-rays stemming from the pro-filmic event. While the analogue camera registers light-rays entering through its lens as photochemical changes on a strip of film, the digital camera contains a charge-coupled device that «records the light levels as voltages, stores them, and then the values are digitised». While the trace produced photochemically by a standard analogue camera is a negative, the trace produced by a digital camera is a bitmap in which «the integer […] stored at each pixel […] encodes information about the light emanating from the part of the object that the pixel represents». Second, digital technology can enter into the generation of an image through “painting”, the alteration, by means of a software editing tool, of the information encoded by the pixels in a bitmap. Finally, an image can be directly generated by a computer through the execution of an algorithm, often by means of a computer generated 3D model to which the algorithm is applied.

  • 20  See, for example, Savedoff 1997.

18Gaut characterises the product of such digital techniques – Mitchell’s “electrobricollage” – as a “mélange image” because the results of these different modes of digital generation can be seamlessly combined in a single image. He agrees that, because of this, we are more reticent in ascribing indexically-based properties to digital images. Some, indeed, have seen this as having significant implications for the ways in which we relate to camera-generated images in general20. But whether the image is indexical depends upon how it was produced, not upon our ability to tell how it was produced from looking at the image. Thus some digital images will be indexical, and qualify as photographic, because, while they involve digital capture, the resulting trace has not been digitally altered after capture. For capture by a digital camera is no less indexical and no less productive of a trace that depends counter-factually on the nature of a pro-filmic event than capture by an analogue camera.

19Furthermore, Gaut points out, even if there is digital manipulation of the trace result of a digital capture through the use of digital “painting”, this by itself is not sufficient to compromise the status of the image as photographic, as indexical, and as a trace. For some forms of digital “painting” simply rehearse, in a digital key, well-established techniques whereby analogue photographic images can and have been modified. From its inception, serious analogue print photography has usually involved post-trace manipulation of the information traced on the negative. Analogue photographers have control, through cropping and developing, over the tone, size, contrast, and composition of the image, and this may affect the image’s formal, expressive, and higher order representational properties. And portrait photographers have long known how to use devices such as air-brushing, dodging, and burning to control the details of the image as well. As long as digital “painting” modifies only these kinds of features of the image, consistency seems to require that we view the resulting digital images as photographic and as indexical.

  • 21  This will, at least, be the case unless we hold that information about the process whereby individ (...)

20Our inability to tell, in looking at a digital image, the extent of the digital modifications it has undergone, and whether it originated in a trace-making process, does not, as noted above, call into question the indexical and photographic status of the image. But it does call into question whether the image can serve as the vehicle of a work of photographic art. For, as we saw, photographic art requires not merely that an image has trace status but also that we can apprehend it under an artistic medium as an intentionally rendered trace who details can be apprehended as the products of intentional agency. Our capacity to do this, however, seems to be challenged by the availability of digital technology, since we have no way of telling from the image which kinds of intentional processes conferred upon it the details that it manifests. Thus, while digital still images may indeed be indexical and photographic, they may function as artworks not as intentionally rendered traces but as images over the details of which the maker is taken to have complete control. In other words, they may function as artworks in the way that paintings do, rather than in the way that traditional photographs do. Abstention from the use of more advanced “painting” or computational manipulations will then be treated as one more creative choice on the part of the artist who generates the digital image21. This, perhaps, is the underlying truth in Mitchell’s claims that digital images differ from traditional photographs to the same extent that the latter differ from paintings.

21We might assume that what goes for still images also goes for moving images. Thus Wallis and Manovich are wrong in thinking that the very use of digital technology in the generation of a moving image renders it non-indexical. Against Wallis, it can be argued that the use of a digital video camera in creating a moving image does not entail that the image lacks indexicality, since the resulting image may be digital only in respect of its manner of capture, or in being modified post-capture in ways analogous to those embraced by traditional photography. And Manovich is incorrect in thinking that the use of digital technology necessarily renders a work of digital cinema a series of paintings rather than a series of photographs. Again, what needs to be determined is the kinds of digital interventions that went into the generation of the moving image.

22However, while we may have preserved the photographic status of some digitally generated still or moving images, many such images involve uses of digital technology that do compromise their indexical and photographic status. For example, some photographs are generated entirely computationally, and many contemporary movies employ a degree of computational manipulation that renders absurd the suggestion that they stand, as whole images, in any relation of indexicality to a pro-filmic reality. Does the loss of indexicality in such cases, then, justify Manovich’s claims that what results is a radically different kind of cinema? And does this give further significance to the point conceded above concerning the nature of the still digital image as art. Surprisingly, I think that the answer to both questions is negative. To see why, we must first look more closely at the significance of indexicality in traditional cinema.

Indexicality and cinema

  • 22  Manovich 1995.
  • 23  Bazin 1967.

23How significant is the indexicality of traditional cinema for its functioning as cinema and for its status as cinematic art? Manovich, as noted above, takes indexicality to be traditional cinema’s defining feature. He claims that the latter «pretends to be a simple recording of an already existing reality» and «works hard to erase any traces of its own production process, including any indication that the images which we see could have been constructed rather than recorded». It thus tries to minimize the attention accorded to «rear projection and blue screen photography, matte paintings and glass shots, mirrors and miniatures, push development, optical effects and other techniques which allowed filmmakers to construct and alter the moving images»22. This is puzzling if intended as a claim about the makers of traditional non-documentary cinema, who, even if not overly forthcoming about the use of the particular manipulative techniques cited by Manovich, surely never presented their product as «a simple recording of an already existing reality». Indeed, even if we bracket such obvious exceptions as directors who, like Eisenstein, put their faith in a “montage of attractions”, those directors like Renoir and Welles who, in Bazin’s sense23, put their trust in reality rather than in the image surely intended viewers to be aware of the artifice that went into the construction of their films, if not of the specific devices used in that artifice.

  • 24  Currie 1995: 166-179.

24But we need not dwell on this. Out interest is rather in the significance of the indexical element in traditional cinema to its functioning and to the claims of at least some of its products to be artworks, and thus in the significance of the loss of indexicality in much digitally generated cinema. One way in which the indexicality of traditional cinema might be thought to be significant is if the receiver’s possibly false impression that the presented images are, at least as individual shots, recordings of pro-filmic reality plays a crucial role in the reception of traditional cinema and thus in the way in which it is able to accomplish its goals. There is of course a considerable literature on the general question of how viewers experience cinematically presented narratives. Gregory Currie among others has argued against what he terms the «imagined observer hypothesis», the idea that viewers imagine that they are observing the narrated events24. If the imagined observer hypothesis were true, then it might be claimed that viewers of traditional cinema imaginatively engage with a cinematically presented narrative as if they were watching «a simple recording of an already existing reality», and indeed that directors trade on the indexicality of the cinematic image to achieve this effect.

25But such a claim is open to at least two objections. First, Currie has argued persuasively that the imagined observer hypothesis is an inadequate account of viewers’ engagements with cinematically presented narratives. He argues that viewers are in fact quite aware that what they are viewing is a sequence of images presented as a means of narrating a story. They make believe the content of the story they take to be narrated through these images, but not that they are observing the narrated events. Second, and more crucially in the present context, even if Currie is wrong in his criticisms of the imagined observer hypothesis, there is no reason to think that the viewer’s experience of a moving image in a fictional film is affected either by the indexicality or non-indexicality of the image, or by her beliefs about the indexicality or non-indexicality of the image. In particular, there is no reason to think that, in this respect, the viewer’s experience in viewing a digitally generated moving image, or what she takes to be a digitally generated moving image, differs from her experience in viewing a moving image generated analogically, or what she takes to be such an image. For viewers have surely always assumed that, if what they are viewing violates how they take the world to work, then some kind of technological trickery has been employed in generating the image. Digitally generated flying figures in contemporary Superman films look more lifelike than those in traditional photochemically generated narratives about the “man of steel”, but in neither case is it likely that viewers are under any illusions as to the constructed nature of the images they are watching. Or, if there is any such illusion, it is more likely in the case of digital cinema because it is so much more “natural”. On the other hand, where what is cinematically presented is something that might indeed occur in reality, viewers are no more likely to suspect technical manipulation of a trace in the digital case than they are in the analogue case. Of course, technically sophisticated viewers might occupy themselves trying to work out which kinds of techniques have been used in a given scene in a digitally generated fictional film. But technically sophisticated viewers of traditional cinema are equally aware – and indeed, much better able to detect – the use of such devices as rear projection and matte screens. So it is difficult to see how the indexicality, or believed indexicality, of traditional cinema establishes a significant difference in the viewer’s experience, or why it would be reasonable to ascribe different goals in this respect to the makers of traditional and digital films.

26If the indexicality of traditional cinema doesn’t differentiate in this way between the viewer’s experience of traditional and digital cinemas, does indexicality nonethless contribute significantly to the functioning of traditional cinema in other ways? Indexicality is certainly crucial to the idea of the still or the moving image as a means for documenting actual events, as Mitchell points out. This is significant in the case of still images, where fictional representation has been a very minor concern, especially in the case of traditional photography. As seen earlier, this is reflected in the focus, in discussions of the possibility of the still photograph as artwork, on the importance of being able to see in the image how the “indicated” bare subject of a photograph has been represented. But fictional narration has been central to the cinema, and to the cinema as an art form, and this may be an important difference. For in fictional cinema, indexicality relates the screened image to actors and film sets, rather than to the characters and events within the fiction. And it is through the manner of representing the latter rather than the former that fictional films have the artistic content that they do. The fictional content articulated through the presented images may therefore be unaffected by the presence or absence of indexicality. The images can have precisely the same properties bearing upon the articulation of such content whether they are indexically related to the doings of actual actors or not.

27The manner of generation of the images that make up fictional films thus stands in a different kind of relation to the artistic content of such films than is the case with photography. In photographic art, as we have seen, the artistic vehicle must be an intentionally rendered trace. It is crucial that we be able to subsume the manifold presented by the photographic image under an artistic medium and thereby determine the expressive roles played by various constitutive elements. Where we have digitally generated still images, our ability to determine this is called into question, along with our ability to bring the image – as possible intentionally rendered trace – under an artistic medium. Because of the ways in which photographs articulate their artistic contents, details of the provenance of an image are crucial to the manner in which the viewer should respond to it as vehicle and to the content thereby articulated, and thus crucial to the identity of the artistic vehicle.

  • 25  Arnheim 1957.

28But, in the case of fictional film, it isn’t clear that content rests in any crucial way upon whether the individual shots were generated analogically or digitally. So the injection of a measure of uncertainty on the audience’s part as to the provenance of individual shots doesn’t obviously affect the viewer’s ability to ascribe content to the film. In grasping the artistic content articulated through the images and shots in a fictional film, we assume that those responsible for the film have control over the details of the image. This is not only through the kinds of strategies upon which Arnheim dwells25 – for example, zoom, lateral panning, and editing. It also involves being able to shoot on a prepared set and being able to retake a shot. Thus, as noted above, we experience a fictional film as a construction rather than as a representation of an independently existing reality. Indeed, we are surprised when we detect something in a fictional film that seems to indicate a failure on the part of the film-maker to exercise such complete control. When we catch sight of a wayward microphone, for example, this surprises us not because we were under the illusion that we were watching reality, but because we assume that the film-maker is always able to control for such things. We rightly assume that the shots that make up a traditional fictional film are traces, but we also assume that they have been assembled for a purpose in a way that almost certainly does not reflect the order in which the shots were derived from the shoot. Since we take the shot-composition and shot-ordering of a film to be a matter of the intentional agency of the director, whether or not the individual shots are traces, our ascription of artistic content to a fictional film does not depend upon our beliefs about its digital or analogue nature. The additional possibility that the image is the product of digital manipulation does not, therefore, change our relationship to the image. It is simply one more device, like mattes and backscreen projection, that permits the film-makers to construct the film. Thus the very thing that raises at least the impression of a significant difference between analogically and digitally generated still images plays no real part in the articulation of the content of cinematic works.

29Even if, in viewing a traditional fictional film, we believe the images to be a true recording of what the actors did, this bears only indirectly on our appreciation of the film. It speaks to the issue of authorship – the credit for the work that should be given to the actors – rather than to the issue of what was in fact authored. It thus bears upon our assessment of the artistic achievement involved in articulating a particular content through a given vehicle. To the extent that we think that the artist’s achievement bears upon the appreciation of an artwork, the ways in which digital technology may have played a part in the generation of a moving image do indeed bear upon our appreciation of the film as artwork. For example, certain kinds of shots in “spatial realist” films such as Renoir’s Rules of the Game and Welles’ A Touch of Evil contribute significantly to the films’ achievements as artworks. But identical “shots” in digitally generated films might represent very different, and probably lesser, achievements – they might not involve such imaginative use of mise en scene and cinematography, for example. But this is a difference in the technological resources whereby given elements in the artistic vehicle have been realized, not a difference in the manner in which the moving image serves as the artistic vehicle through which the content of the film is articulated. In this sense, the situation is similar to one rehearsed in the opening section of this paper concerning the significance of the use of the camera obscura in generating familiar images of Renaissance visual art.

Is digital cinema a new art form?

  • 26 I leave unaddressed here the question whether we should think of digitally generated documentary c (...)

30I have argued that whether or not the images making up a cinematic fiction were generated analogically or digitally makes a difference not to the artistic content articulated through those images, as artistic vehicle, but to the achievement involved in articulating this content in this way. This suggests that we should resist the suggestion that the loss of indexicality entails that traditional and digital fictional cinema differ essentially as art forms. For we have defined the artistic vehicle of an artwork as the means whereby a work articulates its content in virtue of an artistic medium under which the artistic vehicle is apprehended. And we have defined being an x-artwork in terms of the kind of thing that serves as the artistic vehicle. I have questioned whether there is any basis for saying that the loss of indexicality consequent upon the use of digital technology in fictional cinematic narratives entails a change in the nature of the artistic vehicle, and thus a change in art form. Insofar as most cinematic artworks are indeed fictional cinematic narratives, this suggests that the move to digital technology does not have the same kinds of general implications for the artistic status of the resulting images as we saw it to have in the case of still images produced by digital photography26.

  • 27  Gaut 2010: 305.

31But it would be wrong to overlook ways in which digital technology does have more significant implications for cinematic art. For one thing, like the development of acrylics in painting, it changes in certain ways the range of accessible artistic values, something stressed by Gaut27. He notes, first, that the use of “painting” allows the film-maker to directly create expressive content without needing to record some actual object. This not only allows for the easier creation of certain kinds of expressive content but also for expressive content that would not otherwise be creatable. As Gaut puts this, anything that is visually imaginable is now probably filmable. Furthermore, there are styles possible in digital cinema that are in practice unachievable in traditional cinema – for example, photoreal animation and the blending together of aspects of animation and photography.

  • 28  Wallis 2005: 38-39.

32Gaut focuses here on possibilities that arise through digital manipulation of captured content and through computer generation of images. Holly Wallis, in contrast, stresses ways in which the availability of digital technology has revolutionized the very process of capturing pro-filmic reality. The digital video camera, for example, is much more portable than analogue movie cameras and the cheapness of the technology makes it easy to film an event from multiple perspectives. It is also possible to film high-quality extended single takes, as in the film Russian Ark. Perhaps most significant if we are looking for fundamental changes in digital cinema, she notes that digital technology makes possible what she terms a “desktop aesthetic” where, as on a computer, we can work with different frames at the same time. Such an aesthetic «emphasise[s] the role of the image frame, even referencing the frame as the site for the collision, layering, interpenetration and general orchestration of disparate elements»28. Mike Figgis (Timecode) and Peter Greenaway (The Pillow Book, The Tulse Luper Suitcases) are directors who have reveled in these kinds of possibilities.

  • 29  Gaut 2010: 289.
  • 30Ivi: 300.

33Does this justify the claim that digital cinema is indeed a new artistic medium, and a new art form? Gaut claims that it does. He endorses Dominic Lopes’ characterisation of an art medium as «a set of practices for working with some materials, whether symbolic […] or physical»29 for artistic purposes – more specifically, presumably, to articulate or convey a content of some sort in the way that artworks articulate their contents. «For a medium to constitute an art form», Gaut maintains, «it must instantiate artistic properties that are distinct from those that are instantiated by other media»30. He claims that digital cinema meets this requirement in virtue of the ways in which digital technology enlarges the range of cinematically realizable artistic values as illustrated above.

34It falls beyond the purview of this paper to assess Gaut’s claim. Obviously any such assessment would have to critically examine his proposed criterion for individuating art forms, and also determine how this criterion relates to the one to which I have appealed in this paper. In the present context, however, my concern is only to argue that a loss of indexicality consequent upon the use of digital technology does not in itself justify the claim that digital cinema differs fundamentally from its traditional counterpart.

Torna su


Arnheim, R.

– 1957, Film as Art, Berkeley, University of California Press Bazin, A.

– 1967, What is Cinema?, v. 1, trans. H. Gray, Berkeley, University of California Press Carroll, N.

– 1988, Philosophical Problems of Classical Film Theory, Princeton, Princeton University Press

Currie, G.

– 1995, Image and Mind: Film Philosophy and Cognitive Science, Cambridge (Mass.), Cambridge University Press

Davies, D.

– 2004, Art as Performance, Oxford, Blackwell.

– 2008, How Photographs “Signify”: Cartier-Bresson’s “Answer” to Scruton, S. Walden (ed.), Photography and Philosophy: Essays on the Pencil of Nature, Oxford, Blackwell, pp. 167-186

– 2009, Scruton on the Inscrutability of Photographs, “British Journal of Aesthetics”, 49, 4, pp. 341-355.

– 2010, Multiple Instances and Multiple “Instances”, “British Journal of Aesthetics”, 50, 4, pp. 411-426.

Gaut, B.

– 2010, A Philosophy of Cinematic Art, Cambridge (Mass.), Cambridge University Press Hockney, D.

– 2001, Secret Knowledge, New York, Thames and Hudson Manovich, L.

– 1995, What is Digital Cinema?, published online 1995, accessed at on 1st November 2010.

– 2001, The Language of New Media, Cambridge (Mass.), MIT Press Maynard, P.

– 1997, The Engine of Visualization: Thinking through Photography, Ithaca, Cornell University Press

Mitchell, W.J.

– 1992, The Reconfigured Eye: Visual Truth in the Post-Photographic Era, Cambridge (Mass.), MIT Press

Savedoff, B.E.

– 1997, Escaping Reality: Digital Imagery and the Resources of Photography, “Journal of Aesthetics and Art Criticism”, 55, 2, pp. 201-214

Scruton, R.

– 1983, Photography and Representation, in The Aesthetic Understanding, London, Methuen, pp. 102-126

Steadman, P.

– 2001, Vermeer’s Camera, Oxford, Oxford University Press Wallis, H.

– 2005, New Digital Cinema: Reinventing the Moving Image, London, Wallflower

Torna su


1  I follow here the standard distinction between singular and multiple artworks. For a critical examination of the notion of an “instance” of a work, see Davies 2010.

2  See Steadman 2001, Hockney 2001.

3  Whether and to what extent such matters bear upon the appreciation of artworks depends upon the extent to which one subscribes to a broadly empiricist epistemology of art. See Davies 2004: chapters 2 and 3.

4  Wallis 2005: 19.

5  Manovich 1995. While Manovich is better known for his 2001, this paper, discussed in Wallis 2005, is a singularly forthright presentation of the view I want to critically assess in this paper.

6  For a survey of early formulations of this kind of argument against the artistic pretensions of photography, see Maynard 1997, Carroll 1988: 20-29. Such arguments are a primary target of Arnheim 1957. The contemporary locus classicus of this kind of argument is Scruton 1983. For critical discussions of Scruton, see Davies 2008, Davies 2009.

7  See Davies 2008.

8  Manovich 1995.

9  Mitchell 1992.

10Ivi: 7.

11Ivi: 31.

12Ivi: 225.

13Ivi: 7.

14Ivi: 4.

15  Manovich 1995.


17  Wallis 2005: 33.

18  Gaut 2010: 47-49.

19Ivi: 16.

20  See, for example, Savedoff 1997.

21  This will, at least, be the case unless we hold that information about the process whereby individual digital images are generated is required for their proper appreciation.

22  Manovich 1995.

23  Bazin 1967.

24  Currie 1995: 166-179.

25  Arnheim 1957.

26 I leave unaddressed here the question whether we should think of digitally generated documentary cinema as having a different kind of artistic vehicle, and as therefore belonging to a different art form, from its traditional counterpart. It is fictional film that has for the most part concerned those theorists whose claims about indexicality I have critically examined.

27  Gaut 2010: 305.

28  Wallis 2005: 38-39.

29  Gaut 2010: 289.

30Ivi: 300.

Torna su

Per citare questo articolo

Notizia bibliografica

David Davies, « Digital Technology, Indexicality, and Cinema », Rivista di estetica, 46 | 2011, 45-60.

Notizia bibliografica digitale

David Davies, « Digital Technology, Indexicality, and Cinema », Rivista di estetica [Online], 46 | 2011, online dal 30 novembre 2015, consultato il 27 luglio 2017. URL : ; DOI : 10.4000/estetica.1655

Torna su

Diritti d'autore

Licenza Creative Commons
Rivista di Estetica è distribuita con Licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 4.0 Internazionale.

Torna su
  • Logo Rosenberg & Sellier