|
|
Here are the text to base it on:
Kino: The Western Undergraduate Journal of Film Studies
Volume 4 | Issue 1 Article 10 5-20-2013 Realism, Really?: A Closer Look at Theories of Realism in Cinema Timothy John Edwards Follow this and additional works at: http://ir.lib.uwo.ca/kino Part of the Film and Media Studies Commons Recommended Citation Edwards, Timothy John (2013) "Realism, Really?: A Closer Look at Theories of Realism in Cinema," Kino: The Western Undergraduate Journal of Film Studies: Vol. 4: Iss. 1, Article 10. Available at: http://ir.lib.uwo.ca/kino/vol4/iss1/10 Realism, Really?: A Closer Look at Theories of Realism in Cinema Keywords Realism, Film theory, Siegfried Kracauer, Andre Bazin, Cinematic Approach, Cognitive Approach This article is available in Kino: The Western Undergraduate Journal of Film Studies: http://ir.lib.uwo.ca/kino/vol4/iss1/10 Realism, Really?: A Closer Look at Theories of Realism in Cinema by Timothy John Edwards ‘Realism’ has been a theory of art since the days of the ancient Greeks when Plato first put stylus to tablet in an attempt to determine what in fact made ‘art’ art, and furthermore, what made a given work of art a good work of art. A theory of realism in art gained new importance during the Renaissance when the innovation of ‘perspective painting’ was perfected. Then, long after the maturation of Realism as an approach to, and style of, art, came the medium of photography. Influential film theorists like Siegfried Kracauer and Andre Bazin set out to trace this theory to its roots as a human fascination with the artistically mediated reproduction of physical reality. Kracauer and Bazin see the modern medium of cinema as the next step in the evolution of this theory and approach to the creation of art, but, more recently, as the successor to the photographic medium. Kracauer has even gone so far as to posit a prescriptive approach to cinema as an extension of the theory of Realism in the arts, and Bazin also works along similar lines. Kracauer suggests in his writings on Realism in the arts, and Bazin also works along similar lines. Kracauer suggests in his writings on Realism and cinema “that films may claim aesthetic validity if they build from their basis properties; like photographs, that is, they must record and reveal physical reality” (Kracauer 155). While this thesis and Kracauer’s approach to cinema realism more generally – is not without its inherent limitations, it is a theory that carries significant weight even to the present day and one that offers significant and analytical utility for the analysis of films. Siegfried Kracauer purports a theory of cinema in the tradition of artistic Realism. Before delving into his theory, however, it is worth pointing out at the outset that his theory of realism is indeed distinct from the realist tradition in other art forms like painting, and – as will be discussed in the context of Andre Bazin’s ontology of cinema – this is the case by virtue of the very nature of cinema and photography as being means for objectively recording and revealing physical reality. Kracauer’s account begins by outlining the inherent similarities between photography and cinema, pointing out that cinema can be considered to be simply the next step in the evolution of photography. So, it follows that like a photograph, “a film is realistic because it correctly reproduces that part of the real world to which it refers” (Kracauer 299). Kracauer continues to expound his account of cinematic realism by drawing an important distinction between two main tendencies that existed in cinema at the very outset, these tendencies are the ‘realistic’ and the ‘formative’. Kracauer calls upon two early French filmmakers who will serve to exemplify this distinction, Lumiere and Melies. Exemplifying the realistic tendency in early cinema are the Lumiere Brothers. There is perhaps no better way in the mind of this author to characterize this tendency other than to do so by calling upon the testimony of a Lumiere cameraman himself, Mesguich, a testimony first introduced in Kracauer: “As I see it, the Lumiere Brothers had established the true domain of cinema in the right manner. The novel, the theater, suffice for the study of the human heart. The cinema is the dynamism of life, of nature and its manifestations, of the crowd and its eddies. All that asserts itself through movement depends on it. Its lens opens on the world” (Kracauer 292). Mesguich’s words demonstrate the affinity between photography and the approach of the Lumieres in creating films. Many of these films are in every sense ‘moving photographs’. The 1 Edwards: Realism, Really? Produced by The Berkeley Electronic Press, 2013 point of separation between the mediums becomes apparent when one considers the narrative innovation of Lumieres’ films. The Lumiere film Teasing the Gardener (Lumiere, 1895) is – in spite of possible controversy herein – the first fiction film ever to be produced. While Lumieres’ experiments with narrative filmmaking did succeed in providing added entertainment value to members of the paying public, it still left a great deal to be desired. What Lumiere Brothers regarded as merely a “scientific curiosity”, George Melies would come to regard as an unrealized continuum of possibilities. Kracauer claims that if indeed it is the Lumiere Brothers who established the realistic tendency in cinema, it is suiting that Melies be credited as the establisher of the formative tendency in cinema (Kracauer 293). Again according to Kracauer, Melies’ formative aspirations were to become his recipe for success as he went on to pioneer those most essential elements of the modern cinema: masking effects, multiple exposure techniques, as well as superimpositions, the lap-dissolve, and so on (Kracauer 293). It is at this point that Kracauer sets himself up for later contradictions in his theory, because here he acknowledges Melies’ formative approach as having pioneered the techniques that have made cinematic realism possible in the context of the more recent narrative cinemas, yet at the same time he claims that these techniques go against the grain of cinematic realism. Consequently, his distinction seems too dogmatic. Thankfully, it seems that these two tendencies are tenuously reconciled in his Cinematic Approach (Kracauer 297). Kracauer in fact fails to see the distinction between Melies’ use of cinematic illusion, and of presentational modes much in the tradition of the theatre. This failure – while not insurmountable – forms the centre of Gregory Currie’s contemporary debate (Currie). Having now outlined the establishment of these tendencies in early cinema, Kracauer continues to progress and expand his theory by consolidating the parameters of his realistic-formative distinction. What results from this is Kracauer’s prescriptive Cinematic Approach. It is once the camera is used to compose shots via various camera movements that the realistic tendency is realized more fully. This also marks the threshold which photography – with its static images – cannot cross. Kracauer wishes to make a further distinction here, a distinction between the ‘capture of motion’ – as in the use of camera movement – and the staging of shots. It becomes clear here that Kracauer is prepared to elevate the use of camera movement above the use of staging techniques in much the same way that he is prepared to elevate the realistic tendency above the formative. This, again, seems dogmatic and will result in yet another theoretical conflict which he will attempt to reconcile with his more pragmatic Cinematic Approach. What is expressed by this account of staging practices is essentially the observation that the staging of shots – as in a studio or otherwise – can be aesthetically legitimate so long as the staging contributes to the establishment of camera-reality. This is to say that staged shots should create the visual impression of the real world as seen through the lens of the camera (Kracauer 295). Kracauer’s concern with staging is akin to his concern about the formative tendency, which is that when used to fulfill the formative aspirations of the filmmaker, these elements often serve to work against the basic properties of the medium. It is at this point – without further qualification – Kracauer may posit his Cinematic Approach. Kracauer begins, “It follows from what has been said in the preceding chapter that films may 2 Kino: The Western Undergraduate Journal of Film Studies, Vol. 4 [2013], Iss. 1, Art. 10 http://ir.lib.uwo.ca/kino/vol4/iss1/10 claim aesthetic validity if they build from their basic properties; like photographs, that is, they must record and reveal physical reality” (Kracauer 297). Indeed Kracauer’s Cinematic Approach holds true to his thesis, if only initially. In his attempt to reconcile some of the conflicts resulting from his preceding account of cinematic realism, Kracauer offers two considerations: 1. “Favourable response to a genre need not depend upon it’s adequacy to the medium from which it issues.” and 2. “…assume that my definition of aesthetic validity is actually one-sided; that it results from a bias for one in particular, if important, type of cinematic activities and hence is unlikely to take into account, say, the possibility of hybrid genres or the influence of the medium’s non-photographic components” Kracauer 297). It is in this form that Kracauer’s theory of cinematic realism becomes amenable, and thus stands to offer a great deal of analytical utility. This is because his theory is only now able to deal with both favourable and unfavourable qualities in any particular film undergoing analysis, and so with any given film it follows that Kracauer (or any other author employing his theory as an analytical tool) – based on his positive account – may justifiably regard it as being either true to the nature of the cinematic medium, or not. The point is that Kracauer’s theory of cinematic realism may now be treated as a self-contained one. Kracauer finishes here with the following prescription: “Everything depends on the “right” balance between the realistic tendency and the formative tendency; and the two tendencies are well balanced if the latter does not overwhelm the former but eventually follows its lead” (Kracauer 298). With Kracauer’s thesis and theoretical framework now in place, it is fitting that his contemporaries may enter into this discussion. If only brief and elliptical, a treatment of Andre Bazin and Gregory Currie is to follow. Though not technically a ‘contemporary’ in the chronological sense, Andre Bazin’s writings are to be treated here as such for the purposes of identifying certain consistencies between his own ontology and Kracauer’s thesis. In fact, there is – throughout his account – the distinct echo of the ideas presented in Kracauer, if however they are presented in quite a different manner. What is important here is that their theses are in line with one another, that is, Bazin also believes that the basis of the filmic medium is photographic and that films are aesthetically valid if they cultivate cinematic realism. In his own characterization of cinematic realism, Bazin’s distinction lies between true realism and pseudo-realism (Bazin). Indeed, this distinction mirrors Kracauer’s distinction between realist tendencies and formative tendencies. Much of Bazin’s disdain for pseudo-realism is directed at the standardization of techniques pioneered by the Soviet Montagists, but he is also intolerant of the theatrical practices of the German Expressionists (Bazin). Bazin goes a step further than Kracauer in his positive account, however, enthroning the style of Renoir (and also Vittorio de Sica, Robert Bresson, and Orson Welles) as exemplifying the supreme form of cinematic realism (Bazin 315). He bases this assertion on Renoir’s penchant for the “long-take, deep-focus” style (Bazin). Bazin is insistent that this is the most true-to-form style of cinematic realism, and this assertion is entirely coherent with Kracauer’s Cinematic Approach. In fact, the use of ‘long-take, deep focus’ style would constitute the strongest possible application of the Cinematic Approach. There is also contemporary support for this idea in the works of Gregory Currie. Although Currie’s intention is to deconstruct the ideas of classical film theorists, like Kracauer and Bazin, using the empirically based Cognitivist Approach, Currie finds that – to at 3 Edwards: Realism, Really? Produced by The Berkeley Electronic Press, 2013 least some extent – their collective theory of cinematic realism is not entirely unfounded, if however somewhat misguided. For example, Currie points out that the ‘long-take, deep-focus’ style is actually no more realistic in a cognitive sense than are other styles – say those employing soft focus or disjunctive editing techniques. And this is because the ‘long-take, deep-focus’ style requires a much smaller aperture size to render images, meaning the human eye would in fact not be able to perceive these images in the same way as the camera does under the same conditions (Currie 55). It is along these lines that Currie is unlike the two aforementioned theorists. So it follows that if applied loosely (and with empirical support), Kracauer and Bazin’s theories of cinema can still be a theoretical framework of great use to contemporary theorists. While Currie’s Cognitive Approach indeed is the most ‘mature’ theoretical framework of the three discussed here, it will not be taken into account in any significant way within the filmic analysis to follow. Rather, in the interest of respecting the canonical stature and historical weight of these authors, their accounts will be applied charitably in the analysis of specific filmic examples. Furthermore, since there is no significant conflict between their two accounts, their respective accounts will take the form of a theoretical monolith under the heading of the the Cinematic Approach. And so without further qualification, filmic analysis is to proceed with and centre around Jean Renoir’s La Grande Illusion (1937). Bazin is not simply giving an offhand example when he mentions Jean Renoir and La Grande Illusion in his writings on cinematic realism. Upon reviewing the film – or even just one scene – it becomes clear that the film is a veritable template of the sort of approach to realism that Bazin and Kracauer posit. Based on the prescriptive nature of the Cinematic Approach and its exemplification in La Grande Illusion it is conceivable – although never explicitly stated by either REALISM Andrew Kania The term “realism” has been put to almost as many uses in film theory as in philosophy. The basic idea in both areas of study is the same: something is realistic if it bears some sort of veridical relation to reality. Thus, in order to specify a particular sense of “realism” one must specify (i) what is being described as realistic; (ii) what one means by “reality”; and (iii) what relation is being posited between them. In film theory and criticism, one major concern has been with whether particular films, or kinds of film (e.g., film noir, neorealist cinema), veridically represent the true nature of the social or political order, or human nature or consciousness, or interpersonal relations. I largely ignore those questions here. Instead, I address more basic questions about the nature of film in general, and whether it can be said to be a realistic medium at some more fundamental level. I also largely ignore the “classical” film theory of such figures as Arnheim, Bazin, and Panofsky, and contemporary film theory that draws on “continental” theories of psychoanalysis, structuralism, Marxism, and so on. Since these alternative approaches are well served by other chapters in this volume, I focus instead on more recent work in “analytic” philosophy of film and “cognitive” film studies, with their roots in AngloAmerican philosophy of the twentieth century and empirical psychology. (On these distinctions, see Currie 1995: xi–xx; Bordwell and Carroll 1996: xiii–xvii; Bordwell 1996; Carroll 1988a, b, 1996c.) There are a number of distinct ways in which film has been said to be realistic, even at a fundamental level. I address three of these claims: (i) that our experience of motion pictures engenders illusions about the reality of what we are seeing (what I will call “motion picture realism”); (ii) that we literally see the objects captured on film (photographic realism); and (iii) that our experience of film is like our experience of the world (perceptual realism). Finally, I briefly discuss the relation between these metaphysical claims and the aesthetics of film. Motion picture realism Some theorists argue that film is a realistic medium because it engenders an illusion in us that something is real, when in fact it is not. Gregory Currie has usefully divided such theses into two sorts (1995: 28–30). A cognitive-illusionist theory states that film TF17557.indb 237 16/9/08 15:47:30 ANDREW K ANIA 238 engenders a false belief in us, such as that we are literally seeing the fictional events of a film unfold before us. A perceptual illusionist theory states that there is a difference between how film appears to us and how it really is, independently of our beliefs about it. For instance, film images may seem to move, even if we know that such motion is merely apparent and not real. Cognitive-illusionist theories A number of thoroughgoing cognitive illusions have been attributed to film. These claims are not usually explicitly linked to the illusion of movement (though see Baudry 2004a [1970], 2004b [1975]; and Panofsky 2004 [1934–47]), but such a link is perhaps implicit, given that the claims are exclusive to film, rather than being applied to other pictorial, dramatic, or narrative media. (Note, though, that not all films include moving images. See, for example, Derek Jarman’s Blue [1993].) In some cases, cognitive-illusionist claims might plausibly be rejected as harmless hyperbole, but in others they seem central to a theory. In any case, as we shall see, they are all false. Film has been claimed to engender the illusion that the real things we see on screen – Jeff Bridges pouring various nonalcoholic liquids into a glass during the shooting of The Big Lebowski (1998), for instance – are present to the viewer in the theater, or that the viewer is present to these things on the set (e.g., Bazin 1967b [1951]; Metz 1974: 1–15, 43). (These claims should not be confused with the thesis of transparency – that film enables us to see things that are not present – discussed below.) Alternatively, film has been claimed to engender the illusion that when the viewer sees fictional things on the screen – The Dude making a White Russian, for instance – she believes that they are really present to her, or she to them (e.g., Balázs 1970: 48). All of these claims are untenable for the same sorts of reasons. First, they do not cohere with viewers’ behavior. Depending on one’s other beliefs, a person might act in various ways if she thought she was in the presence of Jeff Bridges or The Dude. But no one watching the movie asks for Bridges’ signature, or inquires of The Dude whether there’s enough Kahlúa for two. A familiar response to such objections is to weaken the claim. Is the viewer in a state of partial belief, or uncertainty about whether or not, she is in the presence of Bridges or The Dude? Again, no. Though someone might approach a stranger on the street if she suspected he were a famous actor, not even Bridges’ biggest fan would ask for his signature while watching The Big Lebowski. Similarly, no one even considers attempting to intercede to stop the nihilists from dropping their ferret into The Dude’s bath (Currie 1995: 24–5; Walton 1990: 197–200). Second, these claims do not cohere with viewers’ other beliefs about their abilities to move about in and perceive the world. If you believed you were present in the fictional world, or even at the shooting locations, of a film, then you would have to believe that you were able to jump from one place and time to another almost instantaneously with a cut from one scene to another; you would have to believe that you could perceive things in the manner of a zoom-dolly shot, or as if through different colored filters. TF17557.indb 238 16/9/08 15:47:30 REALISM 239 A slightly weaker form of cognitive illusionism holds that the viewer believes that the fictional things represented in the film are real but is not misled about the film’s status as a representation. That is, the viewer knows he is watching a film but believes, at least while the film is playing, that it is documentary footage. One reason to doubt this hypothesis is that we are no more confused or forgetful about the status of the film we are watching (be it fiction or documentary) than we are about the fact that we are watching a film, as opposed to perceiving reality. Moreover, things are commonly presented on screen that we could not coherently believe to have been filmed in reality, such as the nihilists’ threatening The Dude while he takes a bath. More plausible than any of these cognitive-illusionist claims is the idea that the viewer imagines, rather than believes, what is fictionally represented on the screen (Walton 1990). But such a thesis requires no reference to the viewer undergoing any illusion. Perceptual illusionism Most writers on film argue or assume that the apparent motion of film images is an illusion. However, Gregory Currie, who rejects all forms of cognitive illusionism, also rejects perceptual illusionism; that is, he defends the view that film images really do move (Currie 1995: 34–47, 1996: 334–42). Currie thinks the burden of proof in these debates rests on the illusionist, since we should take things at face value unless we have reason to doubt them. The reason most people think that the motion of film images is illusory is that they understand that film projectors project a succession of still images, separated by moments of darkness, in such rapid succession that we seem to see a continuously projected image that moves. Thus there is no one film image that moves, any more than there is one image that moves in a “flipbook” (Kania 2002: 244, 246n8). Currie responds with an analogy to colors. Colors are generally thought to be “response-dependent” properties, that is, there is no way to specify what “being red” is without reference to the way we experience the world. (A property such as “being square,” by contrast, could be specified in purely geometric terms, without any reference to our experience of that property.) Nonetheless, colors are not illusory, since we are not wrong when we say that blood is red. Hence, we can say that the motion of film images is real, albeit response dependent. This argument rests on a false analogy. Colors are response-dependent properties, but motion is not. Motion consists, at least, in a thing’s being in contiguous spatial locations at contiguous moments in time. No film image meets that condition (Kania 2002: 254–7; Gaut 2003: 634–5). Currie can argue that film images have a distinct property that is related to motion – response-dependent motion – but this merely renames the illusion. A stick in a glass of water could be described as “response dependently bent,” but this does nothing to militate against the fact that the stick is really straight, despite appearing bent, plainly speaking. Currie has a number of ancillary arguments against perceptual illusionism: (1) If film images didn’t move, then all we could see in a cinema would be static images (1995: 34–5). (2) If photography is transparent, and we see people moving in films, TF17557.indb 239 16/9/08 15:47:31 ANDREW K ANIA 240 then the film images must be moving (1995: 38). (3) The identity of film images is response dependent, so we should expect their motion to be response dependent, also (1995: 40–2). However, each of these arguments either falsely assumes that there is a continuously existing image on the screen, or fails to acknowledge the vacuity of attributing response-dependent motion to images. It is worth noting that the above arguments apply only to “moving pictures” produced by certain mechanisms (film, video, zoetrope, etc.). It is much more plausible to suppose that images produced by the uninterrupted projection of light can really move. A simple example would be a spotlight sweeping a prison compound. An example from art is the images on the screen of a shadow play. Furthermore, we can imagine a future technology where the projector’s beam is continuously shone through a transparent cell, whose contents move. Nonetheless, it is not indisputable that such images move, since it is not clear what the identity conditions of images and shadows are (Currie 1995: 30–4; Cassati and Varzi 1994: 175–6). Suppose a searchlight is swept across a cloudy night sky. When the beam runs over a gap in the clouds, does the first spot end as the beam hits the gap and a new one begin when it hits the next cloud, or is this the same spot as before? If so, where was it in the meantime? Photographic realism Most people agree that seeing a photograph of something is quite different from seeing a painting or drawing of the same thing, and this difference is sometimes captured by saying that photographs are more realistic than paintings or drawings. Since most, though by no means all, films are photographic, many theorists have argued that film is a realistic art form on this basis. The characteristic of photography that is appealed to in justifying this sort of realism is that photographs are mechanically produced, and thus the appearance of a photograph is “counterfactually dependent” on the appearance of its subject. That is, if the subject of a photograph had looked different, then the photograph would have looked different, no matter whether the photographer noticed the difference. This is to be contrasted with the case of painting or drawing where the image produced depends on the beliefs or intentions of the artist. That is, a change in the appearance of an object will only affect the appearance of a painting, if at all, by way of affecting the artist’s beliefs about its appearance. (Sound recording is mechanical in the same sense, but has been little discussed, partly due to a tendency to focus on film images at the expense of film sound, and partly since sound synthesis – the sonic equivalent of painting or drawing – is relatively recent and postdates recording technology.) Ontological realism Sometimes the nature of photography is used to defend an extreme realism or illusionism of the kind put aside at the beginning of the previous section – that a photograph is identical with its subject, or that it gives the illusion that it is. Though Instructions: edwards_realism-really.pdf choosen as the essential task University on Leicester we have dismissed these claims, the counterfactual dependence of a photograph on its TF17557.indb 240 16/9/08 15:47:31REALISM241subject does justify a related claim that we might call ontological realism, namely thata photograph of something, unlike a painting of it, guarantees the existence of thatthing (Gaut 2003: 634; Walton 1984: 250). Of course, we must continue to distinguish between the real and fictional contents of an image. A photograph of Jeff Bridgesshows that he existed at some time, but not that The Dude existed.TransparencyA more controversial claim is that photographs are transparent, meaning that inlooking at a photograph you literally see its subject (Bazin 1967a [1945]; Walton1984). (If photography is transparent, but film motion is an illusion, then we reallysee the objects filmed but do not really see them move. If photography is transparentand film motion real, then we really see the filmed objects in motion.)Almost everyone agrees that when you see something out of your window youliterally see that thing, rather than seeing an image in the window. But this examplesits at the top of a slippery slope. In which of the following examples do you literallysee something? Seeing something while wearing glasses, in a mirror, in a distortingmirror, through a periscope, through a telescope, through night-vision goggles, onclosed-circuit television, on a live television broadcast, on a delayed televisionbroadcast, on a taped television show, in a mechanically generated drawing, in aperson’s drawing, in a mechanically generated description, in a person’s description.Almost everyone denies that we see things through descriptions – even mechanicallygenerated ones. Opinions differ about where to draw the line between the window andthe description. What one needs is an argument for a principle according to whichphotography is ruled in or out of the class of transparent media.Here are some arguments against photographic transparency that we can dismiss.It is irrelevant to the transparency claim that the object photographed may no longerexist. When we look at the night sky we see many stars that no longer exist (Walton1984: 252). It is also irrelevant to the transparency claim that when we see a photowe may see no object. If you awake in the middle of the night and look around youmay see nothing, though there are objects all around you. This doesn’t show thatyou don’t see things in ordinary vision (Currie 1995: 57). Another argument againsttransparency contends that photography is like painting after all, since photographsdepend on the intentions of the photographer to choose a particular scene, frame it acertain way, expose the film at a certain moment, and so on. None of this is relevantto the key difference between photography and painting, though. A friend mightblindfold you and take you to see a certain sight from a certain angle when the lightis just right, even holding a frame up to the view so that you appreciate it in a certainway. Nonetheless, you still see the view, so the fact that photographers do equivalentsof all these things does not show that we do not see the subjects of their photographs(Scruton 2006 [1983]: 29–30; Walton 1984: 261–2).More recently it has been argued that the reason photographs are not transparent isthat they fail to provide “egocentric information.” That is, you do not know where thesubject of a photograph is in relation to yourself. But the provision of such informationTF17557.indb 241 16/9/08 15:47:31ANDREW K ANIA242is a central function of vision. Thus, photographs are not a way of literally seeingthings (Currie 1995: 65–9; Carroll 1996d: 61–2). However, the relation between thefunction of vision and what counts as vision is not clear. The function of vision is toprovide us with information in our immediate environment, yet we still literally seethe stars. Also, if we see something in a mirror, then it seems we see something in aperiscope, even if we don’t know how the periscope is set up and thus have no ideawhere what we see is in relation to us (Walton 1997: 69–72).Another recent concern in the debate over photographic transparency has beenthe continuity of the transmission of light. When you look at a star with the nakedeye or through an optical telescope, the light that enters your eye comes directly fromthe star, though it may have been refracted through lenses and reflected off mirrors.Similarly, when you see an object in a mirror or through spectacles, the light thatenters your eyes is the very light reflected off the object. When you see something onclosed-circuit television or in a photograph, however, the light that enters your eyesmay be similar to the light from the object that is responsible for the image, but it isnot that very light. Some take this objection to be decisive (Gaut 2003: 637). Othersargue that a transducer that acted just like a window, but collected light on one sideand instantaneously emitted qualitatively similar light on the other, would be transparent (Currie 1995: 60, 70). It is difficult to see how this dispute could be resolvedwithout begging the question.A further kind of example is that of a three-dimensional model that is counterfactually dependent on some part of the world. Like a photographic film, the visualappearance of such a model is counterfactually dependent on its “subject,” yet, alsolike a film, that appearance is not conveyed by the continuous transmission of light.Gregory Currie considers a clock that determines the positions of the hands of a secondclock (1995: 64–5). Berys Gaut considers a model jungle that is richly counterfactuallydependent on a section of real jungle, to which someone sells tickets, advertising theopportunity to “see real gorillas” (2003: 637; see also Noël Carroll’s model rail-yard,1996d: 61). Gaut rightly moves to the second example, since in the case of the clocksit is not clear that the information preserved from one clock to another is rich enough.(No matter how accurate your porch-light motion sensor is, you do not see the thingthat set it off through seeing the light come on.)Gaut is wrong to think that the model case is decisive, however. He argues that youcould legitimately ask for your money back on discovering that you were seeing modelsrather than gorillas. There are three problems here. One is that there are differentways in which the term “see” is used. If someone sold you the opportunity to see Marsup close with your own eyes and then showed you to a telescope, you might legitimately ask for a refund. Nonetheless, it is true that you see Mars through a telescope.Another problem I suspect is a failure of imagination. It is difficult to imagine amodel of a section of jungle that looks just like the section of jungle, gorillas and all,as opposed to a clunky animatronic. If a model of such, well, photographic realismcould be achieved, it is not clear one could deny its transparency without begging thequestion. Finally, as Gaut says, our intuitions here are partly driven by “the abidinghuman desire to be in direct perceptual contact with objects” (Gaut 2003: 637). ThisTF17557.indb 242 16/9/08 15:47:31REALISM243muddies the waters, since it would be quite reasonable to believe that one was in thepresence of a gorilla if one came upon a perfect robotic model of one, unlike the caseof being in the presence of a photograph of a gorilla, and hence we are more likely tothink of this case in terms of deception. Moreover, again, there are cases of our beingin indirect perceptual contact with things that count uncontroversially as seeing, suchas the mirror and telescope cases.Given the vagueness or polysemy of “see,” it may be that once we have figured outall the ways in which looking at objects in photographs is like and unlike ordinaryvision, any decision to draw the line between literal and nonliteral seeing will bestipulative. Nonetheless, figuring out these similarities and differences is still of valueif we want to understand the nature of photographic film. For if we can at least placefilm viewing with some precision on the spectrum between simply seeing and seeingin a painting (say), we can appeal to this relative immediacy in explaining one way inwhich photographic film is realistic.Perceptual realismMost films are (apparent) moving pictures, and pictures are one kind of representation. They differ from other kinds of representation, such as language, in beingparasitic on ordinary perception. Unlike our understanding of language, we use thesame capacities to recognize what a picture represents as those we use to recognizeobjects and events in the world (Carroll 1985: 82–8; Currie 1996: 327–30). As aresult, pictures have been described as “perceptually realistic.” One way of seeing thedifference is that if you can understand a few pictures in a given style, say, you canunderstand any other picture in that style, while if you can understand a few wordsor sentences in a given language, it does not follow that you will understand otherwords or sentences in that language. Perceptual realism has something to do withthe fact that we perceive pictures as resembling what they represent (Currie 1996:328–30; Walton 1984: 270–3). For instance, if you are likely to confuse a rhinoceroswith a hippopotamus, then you are likely to confuse a picture of a rhinoceros with apicture of a hippopotamus. But you are unlikely to confuse the word “rhinoceros” withthe word “hippopotamus,” since the words do not resemble each other in the waythe animals (and pictures of them) do. Since resemblance is a matter of degree, andperceptual realism appeals to resemblance, so is perceptual realism a matter of degree.For instance, the stylized appearance of a cartoon donkey may be less realistic than aphotographic film of a donkey, but the cartoon is nonetheless perceptually realistic,as opposed to, say, the word “donkey.” To investigate the nature of pictures further isbeyond the scope of this chapter. (See “Depiction” in this volume.) Instead, I will notea few ways in which perceptual realism relates specifically to film.First, not all the visual elements of a film are pictorial. Even in standard narrativefiction films, titles announcing the time and location of scenes are common. Theseare linguistic rather than pictorial. But motion pictures are perceptually realistic withrespect to more properties than still pictures. Since film is a temporal art, in the sensethat the parts of a film have specific duration and ordering, it can be perceptuallyTF17557.indb 243 16/9/08 15:47:31ANDREW K ANIA244realistic with respect to the temporal properties of what it represents (Currie 1995:92–6). Also, whether or not our perception of movement in film is veridical or illusory,it can be perceptually realistic. (It is perhaps worth noting that much “cognitive” filmtheory has focused on the nature of depiction in general, in reaction to the persistentnotion in previous film theory that film is somehow linguistic. For a selection of suchlinguistic theories, see the first section of Braudy and Cohen [2004]. For argumentsagainst such theories, see Currie [1992] and Prince [1993].)Second, photographs are arguably the most reliably perceptually realistic kind ofpictures we have (Cohen and Meskin 2004), and photographic film inherits thatrealism. Adding the points about film’s perceptually realistic representation of timeand movement, we can understand Currie’s claim that “film is a realistic medium, anddeep-focus, long-take style is an especially realistic style within that medium” (1996:328). Deep focus allows us to perceive more easily or directly the spatial relationsrepresented, and a long take allows us to perceive more easily or directly the temporalrelations represented. More immersive formats, such as IMAX and 3-D film arepresumably still more perceptually realistic.Third, perceptual realism is not restricted to vision. Most films are also highlyaurally realistic, at least with respect to diegetic sound (i.e., sound which “comesfrom within the world of the film”). We recognize the fact that a shot has been firedoff screen by using the same aural perceptual capacities that we use to identify realgunshots (though, as with images, these sounds may be more or less stylized. Think,for instance, of the typical sound effect that accompanies an on-screen punch). Sometheme-park movie rides represent the rocky progress of a spaceship by shaking theaudience’s seats, thus representing the feel of a rocky ride in a perceptually realisticway. There have also been experiments with olfactory realism, such as Smell-O-Visionand “odorama.”Finally, there is a synesthetic perceptual realism that comes from a matching ofsounds and images (and movements, smells, and so on). Perceptually realistic images,coupled with perceptually realistic sounds, will result in a markedly less realistic film ifthe images and sounds do not match than if they do. (Note, however, that this simpleidea calls immediately for qualification, since most film music is “nondiegetic,” thatis, unlike most sound effects, it does not represent sounds in the fictional world of thefilm. See Levinson [1996] for a consideration of such music.)Aesthetic implicationsThe reason realism has been such a hot topic in film theory is twofold. Theorists haveargued, first, that film is uniquely, essentially, or particularly realistic; and, second, thatthis has implications for how films ought to be made. I have investigated three waysin which the first of these claims has been defended, though it must be noted that Ihave focused on the extent to which film might be considered realistic. Whether whatrealism we have found is unique or essential to film is a further question, one to bequite skeptical about. For each kind of realism we have had to limit our discussion tocertain types of film – those using traditional projection apparatus, photographic film,TF17557.indb 244 16/9/08 15:47:31REALISM245films employing a particular style, and so on. No such restricted realism can be takenas essential to film in general. But even if such a claim could be defended, it is not clearwhat implications this would have for what we might call the aesthetics of film – thestudy of what makes a film a good one.Noël Carroll has argued persuasively against “medium-essentialism,” the view thatone ought to exploit effects of one’s art form that are either unique to it, or that itachieves better than any other form (1996a [1984–5], 1996b [1985], 1996d). For onething, it is doubtful that the most interesting effects achievable in film are achievableonly in film. Further, it is difficult to know how we could judge whether films or novels,say, are better at telling stories – something at which they both excel. Supposing sucha decision were made, it would be strange to decide that the inferior form should nolonger attempt to tell stories.So in general we should be suspicious of deriving aesthetic imperatives from claimsabout the nature of the medium, including claims about its realism. I end by considering some aesthetic issues specific to the types of realism we examined above.Motion picture realismSurprisingly few film theorists have focused on film’s ability to produce (apparently)moving images (though see Arnheim 1957: 161–87; and Kracauer 1965: 41–5). Anyway,the idea that film should concentrate on the depiction of motion is questionable forthe general reasons given above. Noël Carroll argues that the motion of film imagesshould affect the aesthetics of film in a different way. He points out that the apparentmotion of images is neither exclusive nor essential to film – there are films withoutmotion and moving images in other media – but he argues that this suggests we shouldreorient the study of these media in an inclusive way. Film, television, video, etc., areall media in which the motion of images is possible, if not necessary. Thus it makessense to study motion pictures or moving images in a broad sense, rather than focusingparochially on film (Carroll 1996d). (Note that it will make no difference in mostcases whether the motion of the images is real or illusory.)Photographic realismThe dispute over the transparency of photographs has been hotly debated in partbecause of its supposed implications for the aesthetics of photography and film. RogerScruton (2006 [1983]) has argued that since a photograph is transparent, it cannotbe a representation in the sense of expressing a thought about its subject. As a result,Scruton claims that a photograph or film cannot in itself be aesthetically valuable,though what we see through it may be aesthetically valuable.If Scruton is wrong about the transparency of photographs, and if this implies thatthey are representations, then his argument will not go through. However, the implication is questionable, since it may be that the mechanical nature of photography issufficient to prevent representation, yet insufficient to achieve transparency (Lopes2003: 441). Nonetheless, Dominic Lopes argues that even given Scruton’s restrictedTF17557.indb 245 16/9/08 15:47:31ANDREW K ANIA246notion of representation, photographs can be aesthetically valuable despite theirtransparency, since there is still a difference between seeing an object face-to-face andseeing it through a photograph. We can take an aesthetic interest in a photograph,then, if we take an interest in (i) the object we see through it; (ii) the way in whichthe photograph enables us to see it; and (iii) the interplay between (i) and (ii) (Lopes2003: 442–6).Another way to defeat Scruton’s argument is to show that transparency does notpreclude a photograph’s being a representation. Walton argues that photographs arerepresentations since they are (usually) props in games of make-believe (1984: 253–4,1990: 88, 330, 1997: 68). Stephen Davies argues that if photographs are transparent,then so too are some paintings and drawings. Both photographs and handmadepictures can be counterfactually dependent on their subjects, though the dependenceis mechanical in one case and intentional in the other. Since paintings and drawingsare paradigmatic representations, transparency cannot preclude representation. Thisdoes not open the door to all seeing being mediated by representation. What we seethrough windows, mirrors, spectacles, and so on is not only counterfactually dependentbut continuously dependent on the thing seen. It is the spatiotemporal separationbetween image and object provided by the camera, the canvas, or whatever, thatmakes pictures representations distinct from the things we see through them (Davies2006: 185–8). Note that cinematography does not violate this condition, despite thefact that it does not capture its object at one moment in time, like still photography.Whether or not you think that film images move, it is still the case that either eachframe or the temporally extended image, once recorded, is no longer sensitive tochanges in its subject.Perceptual realismGregory Currie argues that films with a long-shot, deep-focus style are more perceptually realistic than others, since they represent spatial and temporal relations (betweenthings) by means of spatial and temporal relations (between images). Our experienceof such films is more like our experience of the world than is our experience of filmswith rapid editing, for instance (1996: 327–30). Should more films, then, employthis style? Clearly this is a matter for debate. Some (such as Bazin) will argue for thesuperiority of such a style on the basis of its ability to involve us in the world of thefilm, pointing to the success of films in a realist style. Others will argue that this is adangerous seduction, and that filmmakers should work against it, alienating the viewerin order to force an awareness of the medium upon her, pointing to the success of filmssuch as those of Jean-Luc Godard. It is not clear that one side must win this debate.Film, like every other artistic medium, is capable of employing many different stylesfor many different purposes (Carroll 1996b [1985]), and we are capable of appreciatingmany different kinds of films.TF17557.indb 246 16/9/08 15:47:31REALISM247ConclusionsSome films are surely better than others, and perhaps some kinds of films are superior toothers. However, it seems unlikely that we can discover which ones by measuring thedegree to which they are realistic. Nonetheless, the different kinds of realism we haveinvestigated here are relevant to the study of film for other reasons. Most simply, theyallow us to describe the nature of film more precisely, in terms of its illusory nature,the extent to which it is transparent, and one dimension along which cinematic stylescan be said to be realistic. These descriptions, in turn, may factor into a psychologicalexplanation of the power and popularity of motion pictures in general, or specific filmsand styles (Carroll 1985). We should remain wary, however, of drawing conclusionsabout the value of those films, styles, and motion pictures in general, from premisesabout their power and popularity.See also Consciousness (Chapter 4), Definition of “cinema” (Chapter 5), Depiction(Chapter 6), Film as art (Chapter 11), Formalism (Chapter 12), Medium (Chapter16), Music (Chapter 17), Ontology (Chapter 20), Spectatorship (Chapter 23),Sound (Chapter 24), Style (Chapter 25), Rudolph Arnheim (Chapter 27), BertoltBrecht (Chapter 30), Cognitive theory (Chapter 33), Edgar Morin (Chapter 38), andPhenomenology (Chapter 40), Dogme 95 (Chapter 44).ReferencesArnheim, R. (1957) Film as Art, Berkeley and Los Angeles: University of California Press.Balázs, B. (1970) Theory of the Film: Character and Growth of a New Art, New York: Dover.Baudry, J.-L. (2004a [1970]) “Ideological Effects of the Basic Cinematographic Apparatus,” in L. Braudyand M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: OxfordUniversity Press.—— (2004b [1975]) “The Apparatus: Metapsychological Approaches to the Impression of Reality in theCinema,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed.,New York: Oxford University Press.Bazin, A. (1967a [1945]) “The Ontology of the Photographic Image,” in H. Gray (trans.) What Is Cinema?vol. 1, Berkeley and Los Angeles: University of California Press.—— (1967b [1951]) “Theater and Cinema: Part Two,” in H. Gray (trans.) What Is Cinema? vol. 1,Berkeley and Los Angeles: University of California Press.Bordwell, D. (1996) “Contemporary Film Studies and the Vicissitudes of Grand Theory,” in D. Bordwelland N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press.Bordwell, D., and Carroll, N. (eds.) (1996) Post-Theory: Reconstructing Film Studies, Madison: Universityof Wisconsin Press.Braudy, L., and Cohen, M. (eds.) (2004) Film Theory and Criticism: Introductory Readings, 6th ed., NewYork: Oxford University Press.Carroll, N. (1985) “The Power of Movies,” Daedalus 114: 79–103.—— (1988a) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press.—— (1988b) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: ColumbiaUniversity Press.—— (1996a [1984–5]) “Medium Specificity Arguments and the Self-Consciously Invented Arts: Film,Video, and Photography,” in Theorizing the Moving Image, Cambridge and New York: CambridgeUniversity Press.TF17557.indb 247 16/9/08 15:47:31ANDREW K ANIA248—— (1996b [1985]) “The Specificity of Media in the Arts,” in Theorizing the Moving Image, Cambridgeand New York: Cambridge University Press.—— (1996c) “Prospects for Film Theory: A Personal Assessment,” in D. Bordwell and N. Carroll (eds.)Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press.—— (1996d) “Defining the Moving Image,” in Theorizing the Moving Image, Cambridge and New York:Cambridge University Press.Cassati, R., and Varzi, A. (1994) Holes and Other Superficialities, Cambridge, MA: MIT Press.Cohen, J., and Meskin, A. (2004) “On the Epistemic Value of Photographs,” Journal of Aesthetics and ArtCriticism 62: 197–210.Currie, G. (1992) “The Long Goodbye: The Imaginary Language of Film,” British Journal of Aesthetics 33:207–19.—— (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge and New York: CambridgeUniversity Press.—— (1996) “Film, Reality, and Illusion,” in D. Bordwell and N. Carroll (eds.) Post-Theory: ReconstructingFilm Studies, Madison: University of Wisconsin Press.Davies, S. (2006) The Philosophy of Art, Malden, MA; Oxford: Blackwell.Gaut, B. (2003) “Film,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics, Oxford and New York:Oxford University Press.Kania, A. (2002) “The Illusion of Realism in Film,” British Journal of Aesthetics 42: 243–58.Kracauer, S. (1965) Theory of Film: The Redemption of Physical Reality, London: Oxford University Press.Levinson, J. (1996) “Film Music and Narrative Agency,” in D. Bordwell and N. Carroll (eds.) Post-Theory:Reconstructing Film Studies, Madison: University of Wisconsin Press.Lopes, D. (2003) “The Aesthetics of Photographic Transparency,” Mind 112: 433–48.Metz, C. (1974) Film Language, trans. M. Taylor, New York: Oxford University Press.Panofsky, E. (2004 [1934–47]) “Style and Medium in the Motion Pictures,” in L. Braudy and M. Cohen(eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press.Prince, S. (1993) “The Discourse of Pictures: Iconicity and Film Studies,” Film Quarterly 47: 16–28.Scruton, R. (2006 [1983]) “Photography and Representation,” in N. Carroll and J. Choi (eds.) Philosophyof Film and Motion Pictures: An Anthology, Malden, MA: Blackwell.Walton, K. (1984) “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11:246–76.—— (1990) Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA:Harvard University Press.—— (1997) “On Pictures and Photographs: Objections Answered,” in R. Allen and M. Smith (eds.) FilmTheory and Philosophy, Oxford: Clarendon Press; New York: Oxford University Press.TF17557.indb 248 16/9/08 15:47:31
22
REALISMAndrew KaniaThe term “realism” has been put to almost as many uses in film theory as in philosophy.The basic idea in both areas of study is the same: something is realistic if it bears somesort of veridical relation to reality. Thus, in order to specify a particular sense of“realism” one must specify (i) what is being described as realistic; (ii) what one meansby “reality”; and (iii) what relation is being posited between them. In film theory andcriticism, one major concern has been with whether particular films, or kinds of film(e.g., film noir, neorealist cinema), veridically represent the true nature of the social orpolitical order, or human nature or consciousness, or interpersonal relations. I largelyignore those questions here. Instead, I address more basic questions about the natureof film in general, and whether it can be said to be a realistic medium at some morefundamental level.I also largely ignore the “classical” film theory of such figures as Arnheim, Bazin,and Panofsky, and contemporary film theory that draws on “continental” theories ofpsychoanalysis, structuralism, Marxism, and so on. Since these alternative approachesare well served by other chapters in this volume, I focus instead on more recent workin “analytic” philosophy of film and “cognitive” film studies, with their roots in AngloAmerican philosophy of the twentieth century and empirical psychology. (On thesedistinctions, see Currie 1995: xi–xx; Bordwell and Carroll 1996: xiii–xvii; Bordwell1996; Carroll 1988a, b, 1996c.)There are a number of distinct ways in which film has been said to be realistic,even at a fundamental level. I address three of these claims: (i) that our experienceof motion pictures engenders illusions about the reality of what we are seeing (whatI will call “motion picture realism”); (ii) that we literally see the objects captured onfilm (photographic realism); and (iii) that our experience of film is like our experienceof the world (perceptual realism). Finally, I briefly discuss the relation between thesemetaphysical claims and the aesthetics of film.Motion picture realismSome theorists argue that film is a realistic medium because it engenders an illusionin us that something is real, when in fact it is not. Gregory Currie has usefully dividedsuch theses into two sorts (1995: 28–30). A cognitive-illusionist theory states that filmTF17557.indb 237 16/9/08 15:47:30ANDREW K ANIA238engenders a false belief in us, such as that we are literally seeing the fictional events ofa film unfold before us. A perceptual illusionist theory states that there is a differencebetween how film appears to us and how it really is, independently of our beliefs aboutit. For instance, film images may seem to move, even if we know that such motion ismerely apparent and not real.Cognitive-illusionist theoriesA number of thoroughgoing cognitive illusions have been attributed to film. Theseclaims are not usually explicitly linked to the illusion of movement (though seeBaudry 2004a [1970], 2004b [1975]; and Panofsky 2004 [1934–47]), but such a link isperhaps implicit, given that the claims are exclusive to film, rather than being appliedto other pictorial, dramatic, or narrative media. (Note, though, that not all filmsinclude moving images. See, for example, Derek Jarman’s Blue [1993].) In some cases,cognitive-illusionist claims might plausibly be rejected as harmless hyperbole, but inothers they seem central to a theory. In any case, as we shall see, they are all false.Film has been claimed to engender the illusion that the real things we see on screen– Jeff Bridges pouring various nonalcoholic liquids into a glass during the shooting ofThe Big Lebowski (1998), for instance – are present to the viewer in the theater, or thatthe viewer is present to these things on the set (e.g., Bazin 1967b [1951]; Metz 1974:1–15, 43). (These claims should not be confused with the thesis of transparency – thatfilm enables us to see things that are not present – discussed below.) Alternatively, filmhas been claimed to engender the illusion that when the viewer sees fictional things onthe screen – The Dude making a White Russian, for instance – she believes that theyare really present to her, or she to them (e.g., Balázs 1970: 48).All of these claims are untenable for the same sorts of reasons. First, they do notcohere with viewers’ behavior. Depending on one’s other beliefs, a person might actin various ways if she thought she was in the presence of Jeff Bridges or The Dude.But no one watching the movie asks for Bridges’ signature, or inquires of The Dudewhether there’s enough Kahlúa for two. A familiar response to such objections isto weaken the claim. Is the viewer in a state of partial belief, or uncertainty aboutwhether or not, she is in the presence of Bridges or The Dude? Again, no. Thoughsomeone might approach a stranger on the street if she suspected he were a famousactor, not even Bridges’ biggest fan would ask for his signature while watching The BigLebowski. Similarly, no one even considers attempting to intercede to stop the nihilistsfrom dropping their ferret into The Dude’s bath (Currie 1995: 24–5; Walton 1990:197–200).Second, these claims do not cohere with viewers’ other beliefs about their abilitiesto move about in and perceive the world. If you believed you were present in thefictional world, or even at the shooting locations, of a film, then you would have tobelieve that you were able to jump from one place and time to another almost instantaneously with a cut from one scene to another; you would have to believe that youcould perceive things in the manner of a zoom-dolly shot, or as if through differentcolored filters.TF17557.indb 238 16/9/08 15:47:30REALISM239A slightly weaker form of cognitive illusionism holds that the viewer believes thatthe fictional things represented in the film are real but is not misled about the film’sstatus as a representation. That is, the viewer knows he is watching a film but believes,at least while the film is playing, that it is documentary footage. One reason to doubtthis hypothesis is that we are no more confused or forgetful about the status of thefilm we are watching (be it fiction or documentary) than we are about the fact that weare watching a film, as opposed to perceiving reality. Moreover, things are commonlypresented on screen that we could not coherently believe to have been filmed inreality, such as the nihilists’ threatening The Dude while he takes a bath.More plausible than any of these cognitive-illusionist claims is the idea that theviewer imagines, rather than believes, what is fictionally represented on the screen(Walton 1990). But such a thesis requires no reference to the viewer undergoing anyillusion.Perceptual illusionismMost writers on film argue or assume that the apparent motion of film images is anillusion. However, Gregory Currie, who rejects all forms of cognitive illusionism, alsorejects perceptual illusionism; that is, he defends the view that film images really do move(Currie 1995: 34–47, 1996: 334–42). Currie thinks the burden of proof in these debatesrests on the illusionist, since we should take things at face value unless we have reasonto doubt them. The reason most people think that the motion of film images is illusoryis that they understand that film projectors project a succession of still images, separatedby moments of darkness, in such rapid succession that we seem to see a continuouslyprojected image that moves. Thus there is no one film image that moves, any more thanthere is one image that moves in a “flipbook” (Kania 2002: 244, 246n8).Currie responds with an analogy to colors. Colors are generally thought to be“response-dependent” properties, that is, there is no way to specify what “beingred” is without reference to the way we experience the world. (A property such as“being square,” by contrast, could be specified in purely geometric terms, without anyreference to our experience of that property.) Nonetheless, colors are not illusory,since we are not wrong when we say that blood is red. Hence, we can say that themotion of film images is real, albeit response dependent.This argument rests on a false analogy. Colors are response-dependent properties,but motion is not. Motion consists, at least, in a thing’s being in contiguous spatiallocations at contiguous moments in time. No film image meets that condition (Kania2002: 254–7; Gaut 2003: 634–5). Currie can argue that film images have a distinctproperty that is related to motion – response-dependent motion – but this merelyrenames the illusion. A stick in a glass of water could be described as “responsedependently bent,” but this does nothing to militate against the fact that the stick isreally straight, despite appearing bent, plainly speaking.Currie has a number of ancillary arguments against perceptual illusionism: (1) Iffilm images didn’t move, then all we could see in a cinema would be static images(1995: 34–5). (2) If photography is transparent, and we see people moving in films,TF17557.indb 239 16/9/08 15:47:31ANDREW K ANIA240then the film images must be moving (1995: 38). (3) The identity of film images isresponse dependent, so we should expect their motion to be response dependent, also(1995: 40–2). However, each of these arguments either falsely assumes that there isa continuously existing image on the screen, or fails to acknowledge the vacuity ofattributing response-dependent motion to images.It is worth noting that the above arguments apply only to “moving pictures”produced by certain mechanisms (film, video, zoetrope, etc.). It is much moreplausible to suppose that images produced by the uninterrupted projection of light canreally move. A simple example would be a spotlight sweeping a prison compound. Anexample from art is the images on the screen of a shadow play. Furthermore, we canimagine a future technology where the projector’s beam is continuously shone througha transparent cell, whose contents move. Nonetheless, it is not indisputable that suchimages move, since it is not clear what the identity conditions of images and shadowsare (Currie 1995: 30–4; Cassati and Varzi 1994: 175–6). Suppose a searchlight is sweptacross a cloudy night sky. When the beam runs over a gap in the clouds, does the firstspot end as the beam hits the gap and a new one begin when it hits the next cloud, oris this the same spot as before? If so, where was it in the meantime?Photographic realismMost people agree that seeing a photograph of something is quite different from seeinga painting or drawing of the same thing, and this difference is sometimes captured bysaying that photographs are more realistic than paintings or drawings. Since most,though by no means all, films are photographic, many theorists have argued that filmis a realistic art form on this basis.The characteristic of photography that is appealed to in justifying this sort ofrealism is that photographs are mechanically produced, and thus the appearance of aphotograph is “counterfactually dependent” on the appearance of its subject. That is,if the subject of a photograph had looked different, then the photograph would havelooked different, no matter whether the photographer noticed the difference. Thisis to be contrasted with the case of painting or drawing where the image produceddepends on the beliefs or intentions of the artist. That is, a change in the appearanceof an object will only affect the appearance of a painting, if at all, by way of affectingthe artist’s beliefs about its appearance. (Sound recording is mechanical in the samesense, but has been little discussed, partly due to a tendency to focus on film images atthe expense of film sound, and partly since sound synthesis – the sonic equivalent ofpainting or drawing – is relatively recent and postdates recording technology.)Ontological realismSometimes the nature of photography is used to defend an extreme realism orillusionism of the kind put aside at the beginning of the previous section – that aphotograph is identical with its subject, or that it gives the illusion that it is. Thoughwe have dismissed these claims, the counterfactual dependence of a photograph on itsTF17557.indb 240 16/9/08 15:47:31REALISM241subject does justify a related claim that we might call ontological realism, namely thata photograph of something, unlike a painting of it, guarantees the existence of thatthing (Gaut 2003: 634; Walton 1984: 250). Of course, we must continue to distinguish between the real and fictional contents of an image. A photograph of Jeff Bridgesshows that he existed at some time, but not that The Dude existed.TransparencyA more controversial claim is that photographs are transparent, meaning that inlooking at a photograph you literally see its subject (Bazin 1967a [1945]; Walton1984). (If photography is transparent, but film motion is an illusion, then we reallysee the objects filmed but do not really see them move. If photography is transparentand film motion real, then we really see the filmed objects in motion.)Almost everyone agrees that when you see something out of your window youliterally see that thing, rather than seeing an image in the window. But this examplesits at the top of a slippery slope. In which of the following examples do you literallysee something? Seeing something while wearing glasses, in a mirror, in a distortingmirror, through a periscope, through a telescope, through night-vision goggles, onclosed-circuit television, on a live television broadcast, on a delayed televisionbroadcast, on a taped television show, in a mechanically generated drawing, in aperson’s drawing, in a mechanically generated description, in a person’s description.Almost everyone denies that we see things through descriptions – even mechanicallygenerated ones. Opinions differ about where to draw the line between the window andthe description. What one needs is an argument for a principle according to whichphotography is ruled in or out of the class of transparent media.Here are some arguments against photographic transparency that we can dismiss.It is irrelevant to the transparency claim that the object photographed may no longerexist. When we look at the night sky we see many stars that no longer exist (Walton1984: 252). It is also irrelevant to the transparency claim that when we see a photowe may see no object. If you awake in the middle of the night and look around youmay see nothing, though there are objects all around you. This doesn’t show thatyou don’t see things in ordinary vision (Currie 1995: 57). Another argument againsttransparency contends that photography is like painting after all, since photographsdepend on the intentions of the photographer to choose a particular scene, frame it acertain way, expose the film at a certain moment, and so on. None of this is relevantto the key difference between photography and painting, though. A friend mightblindfold you and take you to see a certain sight from a certain angle when the lightis just right, even holding a frame up to the view so that you appreciate it in a certainway. Nonetheless, you still see the view, so the fact that photographers do equivalentsof all these things does not show that we do not see the subjects of their photographs(Scruton 2006 [1983]: 29–30; Walton 1984: 261–2).More recently it has been argued that the reason photographs are not transparent isthat they fail to provide “egocentric information.” That is, you do not know where thesubject of a photograph is in relation to yourself. But the provision of such informationTF17557.indb 241 16/9/08 15:47:31ANDREW K ANIA242is a central function of vision. Thus, photographs are not a way of literally seeingthings (Currie 1995: 65–9; Carroll 1996d: 61–2). However, the relation between thefunction of vision and what counts as vision is not clear. The function of vision is toprovide us with information in our immediate environment, yet we still literally seethe stars. Also, if we see something in a mirror, then it seems we see something in aperiscope, even if we don’t know how the periscope is set up and thus have no ideawhere what we see is in relation to us (Walton 1997: 69–72).Another recent concern in the debate over photographic transparency has beenthe continuity of the transmission of light. When you look at a star with the nakedeye or through an optical telescope, the light that enters your eye comes directly fromthe star, though it may have been refracted through lenses and reflected off mirrors.Similarly, when you see an object in a mirror or through spectacles, the light thatenters your eyes is the very light reflected off the object. When you see something onclosed-circuit television or in a photograph, however, the light that enters your eyesmay be similar to the light from the object that is responsible for the image, but it isnot that very light. Some take this objection to be decisive (Gaut 2003: 637). Othersargue that a transducer that acted just like a window, but collected light on one sideand instantaneously emitted qualitatively similar light on the other, would be transparent (Currie 1995: 60, 70). It is difficult to see how this dispute could be resolvedwithout begging the question.A further kind of example is that of a three-dimensional model that is counterfactually dependent on some part of the world. Like a photographic film, the visualappearance of such a model is counterfactually dependent on its “subject,” yet, alsolike a film, that appearance is not conveyed by the continuous transmission of light.Gregory Currie considers a clock that determines the positions of the hands of a secondclock (1995: 64–5). Berys Gaut considers a model jungle that is richly counterfactuallydependent on a section of real jungle, to which someone sells tickets, advertising theopportunity to “see real gorillas” (2003: 637; see also Noël Carroll’s model rail-yard,1996d: 61). Gaut rightly moves to the second example, since in the case of the clocksit is not clear that the information preserved from one clock to another is rich enough.(No matter how accurate your porch-light motion sensor is, you do not see the thingthat set it off through seeing the light come on.)Gaut is wrong to think that the model case is decisive, however. He argues that youcould legitimately ask for your money back on discovering that you were seeing modelsrather than gorillas. There are three problems here. One is that there are differentways in which the term “see” is used. If someone sold you the opportunity to see Marsup close with your own eyes and then showed you to a telescope, you might legitimately ask for a refund. Nonetheless, it is true that you see Mars through a telescope.Another problem I suspect is a failure of imagination. It is difficult to imagine amodel of a section of jungle that looks just like the section of jungle, gorillas and all,as opposed to a clunky animatronic. If a model of such, well, photographic realismcould be achieved, it is not clear one could deny its transparency without begging thequestion. Finally, as Gaut says, our intuitions here are partly driven by “the abidinghuman desire to be in direct perceptual contact with objects” (Gaut 2003: 637). ThisTF17557.indb 242 16/9/08 15:47:31REALISM243muddies the waters, since it would be quite reasonable to believe that one was in thepresence of a gorilla if one came upon a perfect robotic model of one, unlike the caseof being in the presence of a photograph of a gorilla, and hence we are more likely tothink of this case in terms of deception. Moreover, again, there are cases of our beingin indirect perceptual contact with things that count uncontroversially as seeing, suchas the mirror and telescope cases.Given the vagueness or polysemy of “see,” it may be that once we have figured outall the ways in which looking at objects in photographs is like and unlike ordinaryvision, any decision to draw the line between literal and nonliteral seeing will bestipulative. Nonetheless, figuring out these similarities and differences is still of valueif we want to understand the nature of photographic film. For if we can at least placefilm viewing with some precision on the spectrum between simply seeing and seeingin a painting (say), we can appeal to this relative immediacy in explaining one way inwhich photographic film is realistic.Perceptual realismMost films are (apparent) moving pictures, and pictures are one kind of representation. They differ from other kinds of representation, such as language, in beingparasitic on ordinary perception. Unlike our understanding of language, we use thesame capacities to recognize what a picture represents as those we use to recognizeobjects and events in the world (Carroll 1985: 82–8; Currie 1996: 327–30). As aresult, pictures have been described as “perceptually realistic.” One way of seeing thedifference is that if you can understand a few pictures in a given style, say, you canunderstand any other picture in that style, while if you can understand a few wordsor sentences in a given language, it does not follow that you will understand otherwords or sentences in that language. Perceptual realism has something to do withthe fact that we perceive pictures as resembling what they represent (Currie 1996:328–30; Walton 1984: 270–3). For instance, if you are likely to confuse a rhinoceroswith a hippopotamus, then you are likely to confuse a picture of a rhinoceros with apicture of a hippopotamus. But you are unlikely to confuse the word “rhinoceros” withthe word “hippopotamus,” since the words do not resemble each other in the waythe animals (and pictures of them) do. Since resemblance is a matter of degree, andperceptual realism appeals to resemblance, so is perceptual realism a matter of degree.For instance, the stylized appearance of a cartoon donkey may be less realistic than aphotographic film of a donkey, but the cartoon is nonetheless perceptually realistic,as opposed to, say, the word “donkey.” To investigate the nature of pictures further isbeyond the scope of this chapter. (See “Depiction” in this volume.) Instead, I will notea few ways in which perceptual realism relates specifically to film.First, not all the visual elements of a film are pictorial. Even in standard narrativefiction films, titles announcing the time and location of scenes are common. Theseare linguistic rather than pictorial. But motion pictures are perceptually realistic withrespect to more properties than still pictures. Since film is a temporal art, in the sensethat the parts of a film have specific duration and ordering, it can be perceptuallyTF17557.indb 243 16/9/08 15:47:31ANDREW K ANIA244realistic with respect to the temporal properties of what it represents (Currie 1995:92–6). Also, whether or not our perception of movement in film is veridical or illusory,it can be perceptually realistic. (It is perhaps worth noting that much “cognitive” filmtheory has focused on the nature of depiction in general, in reaction to the persistentnotion in previous film theory that film is somehow linguistic. For a selection of suchlinguistic theories, see the first section of Braudy and Cohen [2004]. For argumentsagainst such theories, see Currie [1992] and Prince [1993].)Second, photographs are arguably the most reliably perceptually realistic kind ofpictures we have (Cohen and Meskin 2004), and photographic film inherits thatrealism. Adding the points about film’s perceptually realistic representation of timeand movement, we can understand Currie’s claim that “film is a realistic medium, anddeep-focus, long-take style is an especially realistic style within that medium” (1996:328). Deep focus allows us to perceive more easily or directly the spatial relationsrepresented, and a long take allows us to perceive more easily or directly the temporalrelations represented. More immersive formats, such as IMAX and 3-D film arepresumably still more perceptually realistic.Third, perceptual realism is not restricted to vision. Most films are also highlyaurally realistic, at least with respect to diegetic sound (i.e., sound which “comesfrom within the world of the film”). We recognize the fact that a shot has been firedoff screen by using the same aural perceptual capacities that we use to identify realgunshots (though, as with images, these sounds may be more or less stylized. Think,for instance, of the typical sound effect that accompanies an on-screen punch). Sometheme-park movie rides represent the rocky progress of a spaceship by shaking theaudience’s seats, thus representing the feel of a rocky ride in a perceptually realisticway. There have also been experiments with olfactory realism, such as Smell-O-Visionand “odorama.”Finally, there is a synesthetic perceptual realism that comes from a matching ofsounds and images (and movements, smells, and so on). Perceptually realistic images,coupled with perceptually realistic sounds, will result in a markedly less realistic film ifthe images and sounds do not match than if they do. (Note, however, that this simpleidea calls immediately for qualification, since most film music is “nondiegetic,” thatis, unlike most sound effects, it does not represent sounds in the fictional world of thefilm. See Levinson [1996] for a consideration of such music.)Aesthetic implicationsThe reason realism has been such a hot topic in film theory is twofold. Theorists haveargued, first, that film is uniquely, essentially, or particularly realistic; and, second, thatthis has implications for how films ought to be made. I have investigated three waysin which the first of these claims has been defended, though it must be noted that Ihave focused on the extent to which film might be considered realistic. Whether whatrealism we have found is unique or essential to film is a further question, one to bequite skeptical about. For each kind of realism we have had to limit our discussion tocertain types of film – those using traditional projection apparatus, photographic film,TF17557.indb 244 16/9/08 15:47:31REALISM245films employing a particular style, and so on. No such restricted realism can be takenas essential to film in general. But even if such a claim could be defended, it is not clearwhat implications this would have for what we might call the aesthetics of film – thestudy of what makes a film a good one.Noël Carroll has argued persuasively against “medium-essentialism,” the view thatone ought to exploit effects of one’s art form that are either unique to it, or that itachieves better than any other form (1996a [1984–5], 1996b [1985], 1996d). For onething, it is doubtful that the most interesting effects achievable in film are achievableonly in film. Further, it is difficult to know how we could judge whether films or novels,say, are better at telling stories – something at which they both excel. Supposing sucha decision were made, it would be strange to decide that the inferior form should nolonger attempt to tell stories.So in general we should be suspicious of deriving aesthetic imperatives from claimsabout the nature of the medium, including claims about its realism. I end by considering some aesthetic issues specific to the types of realism we examined above.Motion picture realismSurprisingly few film theorists have focused on film’s ability to produce (apparently)moving images (though see Arnheim 1957: 161–87; and Kracauer 1965: 41–5). Anyway,the idea that film should concentrate on the depiction of motion is questionable forthe general reasons given above. Noël Carroll argues that the motion of film imagesshould affect the aesthetics of film in a different way. He points out that the apparentmotion of images is neither exclusive nor essential to film – there are films withoutmotion and moving images in other media – but he argues that this suggests we shouldreorient the study of these media in an inclusive way. Film, television, video, etc., areall media in which the motion of images is possible, if not necessary. Thus it makessense to study motion pictures or moving images in a broad sense, rather than focusingparochially on film (Carroll 1996d). (Note that it will make no difference in mostcases whether the motion of the images is real or illusory.)Photographic realismThe dispute over the transparency of photographs has been hotly debated in partbecause of its supposed implications for the aesthetics of photography and film. RogerScruton (2006 [1983]) has argued that since a photograph is transparent, it cannotbe a representation in the sense of expressing a thought about its subject. As a result,Scruton claims that a photograph or film cannot in itself be aesthetically valuable,though what we see through it may be aesthetically valuable.If Scruton is wrong about the transparency of photographs, and if this implies thatthey are representations, then his argument will not go through. However, the implication is questionable, since it may be that the mechanical nature of photography issufficient to prevent representation, yet insufficient to achieve transparency (Lopes2003: 441). Nonetheless, Dominic Lopes argues that even given Scruton’s restrictedTF17557.indb 245 16/9/08 15:47:31ANDREW K ANIA246notion of representation, photographs can be aesthetically valuable despite theirtransparency, since there is still a difference between seeing an object face-to-face andseeing it through a photograph. We can take an aesthetic interest in a photograph,then, if we take an interest in (i) the object we see through it; (ii) the way in whichthe photograph enables us to see it; and (iii) the interplay between (i) and (ii) (Lopes2003: 442–6).Another way to defeat Scruton’s argument is to show that transparency does notpreclude a photograph’s being a representation. Walton argues that photographs arerepresentations since they are (usually) props in games of make-believe (1984: 253–4,1990: 88, 330, 1997: 68). Stephen Davies argues that if photographs are transparent,then so too are some paintings and drawings. Both photographs and handmadepictures can be counterfactually dependent on their subjects, though the dependenceis mechanical in one case and intentional in the other. Since paintings and drawingsare paradigmatic representations, transparency cannot preclude representation. Thisdoes not open the door to all seeing being mediated by representation. What we seethrough windows, mirrors, spectacles, and so on is not only counterfactually dependentbut continuously dependent on the thing seen. It is the spatiotemporal separationbetween image and object provided by the camera, the canvas, or whatever, thatmakes pictures representations distinct from the things we see through them (Davies2006: 185–8). Note that cinematography does not violate this condition, despite thefact that it does not capture its object at one moment in time, like still photography.Whether or not you think that film images move, it is still the case that either eachframe or the temporally extended image, once recorded, is no longer sensitive tochanges in its subject.Perceptual realismGregory Currie argues that films with a long-shot, deep-focus style are more perceptually realistic than others, since they represent spatial and temporal relations (betweenthings) by means of spatial and temporal relations (between images). Our experienceof such films is more like our experience of the world than is our experience of filmswith rapid editing, for instance (1996: 327–30). Should more films, then, employthis style? Clearly this is a matter for debate. Some (such as Bazin) will argue for thesuperiority of such a style on the basis of its ability to involve us in the world of thefilm, pointing to the success of films in a realist style. Others will argue that this is adangerous seduction, and that filmmakers should work against it, alienating the viewerin order to force an awareness of the medium upon her, pointing to the success of filmssuch as those of Jean-Luc Godard. It is not clear that one side must win this debate.Film, like every other artistic medium, is capable of employing many different stylesfor many different purposes (Carroll 1996b [1985]), and we are capable of appreciatingmany different kinds of films.TF17557.indb 246 16/9/08 15:47:31REALISM247ConclusionsSome films are surely better than others, and perhaps some kinds of films are superior toothers. However, it seems unlikely that we can discover which ones by measuring thedegree to which they are realistic. Nonetheless, the different kinds of realism we haveinvestigated here are relevant to the study of film for other reasons. Most simply, theyallow us to describe the nature of film more precisely, in terms of its illusory nature,the extent to which it is transparent, and one dimension along which cinematic stylescan be said to be realistic. These descriptions, in turn, may factor into a psychologicalexplanation of the power and popularity of motion pictures in general, or specific filmsand styles (Carroll 1985). We should remain wary, however, of drawing conclusionsabout the value of those films, styles, and motion pictures in general, from premisesabout their power and popularity.See also Consciousness (Chapter 4), Definition of “cinema” (Chapter 5), Depiction(Chapter 6), Film as art (Chapter 11), Formalism (Chapter 12), Medium (Chapter16), Music (Chapter 17), Ontology (Chapter 20), Spectatorship (Chapter 23),Sound (Chapter 24), Style (Chapter 25), Rudolph Arnheim (Chapter 27), BertoltBrecht (Chapter 30), Cognitive theory (Chapter 33), Edgar Morin (Chapter 38), andPhenomenology (Chapter 40), Dogme 95 (Chapter 44).ReferencesArnheim, R. (1957) Film as Art, Berkeley and Los Angeles: University of California Press.Balázs, B. (1970) Theory of the Film: Character and Growth of a New Art, New York: Dover.Baudry, J.-L. (2004a [1970]) “Ideological Effects of the Basic Cinematographic Apparatus,” in L. Braudyand M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: OxfordUniversity Press.—— (2004b [1975]) “The Apparatus: Metapsychological Approaches to the Impression of Reality in theCinema,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed.,New York: Oxford University Press.Bazin, A. (1967a [1945]) “The Ontology of the Photographic Image,” in H. Gray (trans.) What Is Cinema?vol. 1, Berkeley and Los Angeles: University of California Press.—— (1967b [1951]) “Theater and Cinema: Part Two,” in H. Gray (trans.) What Is Cinema? vol. 1,Berkeley and Los Angeles: University of California Press.Bordwell, D. (1996) “Contemporary Film Studies and the Vicissitudes of Grand Theory,” in D. Bordwelland N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press.Bordwell, D., and Carroll, N. (eds.) (1996) Post-Theory: Reconstructing Film Studies, Madison: Universityof Wisconsin Press.Braudy, L., and Cohen, M. (eds.) (2004) Film Theory and Criticism: Introductory Readings, 6th ed., NewYork: Oxford University Press.Carroll, N. (1985) “The Power of Movies,” Daedalus 114: 79–103.—— (1988a) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press.—— (1988b) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: ColumbiaUniversity Press.—— (1996a [1984–5]) “Medium Specificity Arguments and the Self-Consciously Invented Arts: Film,Video, and Photography,” in Theorizing the Moving Image, Cambridge and New York: CambridgeUniversity Press.TF17557.indb 247 16/9/08 15:47:31ANDREW K ANIA248—— (1996b [1985]) “The Specificity of Media in the Arts,” in Theorizing the Moving Image, Cambridgeand New York: Cambridge University Press.—— (1996c) “Prospects for Film Theory: A Personal Assessment,” in D. Bordwell and N. Carroll (eds.)Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press.—— (1996d) “Defining the Moving Image,” in Theorizing the Moving Image, Cambridge and New York:Cambridge University Press.Cassati, R., and Varzi, A. (1994) Holes and Other Superficialities, Cambridge, MA: MIT Press.Cohen, J., and Meskin, A. (2004) “On the Epistemic Value of Photographs,” Journal of Aesthetics and ArtCriticism 62: 197–210.Currie, G. (1992) “The Long Goodbye: The Imaginary Language of Film,” British Journal of Aesthetics 33:207–19.—— (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge and New York: CambridgeUniversity Press.—— (1996) “Film, Reality, and Illusion,” in D. Bordwell and N. Carroll (eds.) Post-Theory: ReconstructingFilm Studies, Madison: University of Wisconsin Press.Davies, S. (2006) The Philosophy of Art, Malden, MA; Oxford: Blackwell.Gaut, B. (2003) “Film,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics, Oxford and New York:Oxford University Press.Kania, A. (2002) “The Illusion of Realism in Film,” British Journal of Aesthetics 42: 243–58.Kracauer, S. (1965) Theory of Film: The Redemption of Physical Reality, London: Oxford University Press.Levinson, J. (1996) “Film Music and Narrative Agency,” in D. Bordwell and N. Carroll (eds.) Post-Theory:Reconstructing Film Studies, Madison: University of Wisconsin Press.Lopes, D. (2003) “The Aesthetics of Photographic Transparency,” Mind 112: 433–48.Metz, C. (1974) Film Language, trans. M. Taylor, New York: Oxford University Press.Panofsky, E. (2004 [1934–47]) “Style and Medium in the Motion Pictures,” in L. Braudy and M. Cohen(eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press.Prince, S. (1993) “The Discourse of Pictures: Iconicity and Film Studies,” Film Quarterly 47: 16–28.Scruton, R. (2006 [1983]) “Photography and Representation,” in N. Carroll and J. Choi (eds.) Philosophyof Film and Motion Pictures: An Anthology, Malden, MA: Blackwell.Walton, K. (1984) “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11:246–76.—— (1990) Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA:Harvard University Press.—— (1997) “On Pictures and Photographs: Objections Answered,” in R. Allen and M. Smith (eds.) FilmTheory and Philosophy, Oxford: Clarendon Press; New York: Oxford University Press.TF17557.indb 248 16/9/08 15:47:31
22
REALISMAndrew KaniaThe term “realism” has been put to almost as many uses in film theory as in philosophy.The basic idea in both areas of study is the same: something is realistic if it bears somesort of veridical relation to reality. Thus, in order to specify a particular sense of“realism” one must specify (i) what is being described as realistic; (ii) what one meansby “reality”; and (iii) what relation is being posited between them. In film theory andcriticism, one major concern has been with whether particular films, or kinds of film(e.g., film noir, neorealist cinema), veridically represent the true nature of the social orpolitical order, or human nature or consciousness, or interpersonal relations. I largelyignore those questions here. Instead, I address more basic questions about the natureof film in general, and whether it can be said to be a realistic medium at some morefundamental level.I also largely ignore the “classical” film theory of such figures as Arnheim, Bazin,and Panofsky, and contemporary film theory that draws on “continental” theories ofpsychoanalysis, structuralism, Marxism, and so on. Since these alternative approachesare well served by other chapters in this volume, I focus instead on more recent workin “analytic” philosophy of film and “cognitive” film studies, with their roots in AngloAmerican philosophy of the twentieth century and empirical psychology. (On thesedistinctions, see Currie 1995: xi–xx; Bordwell and Carroll 1996: xiii–xvii; Bordwell1996; Carroll 1988a, b, 1996c.)There are a number of distinct ways in which film has been said to be realistic,even at a fundamental level. I address three of these claims: (i) that our experienceof motion pictures engenders illusions about the reality of what we are seeing (whatI will call “motion picture realism”); (ii) that we literally see the objects captured onfilm (photographic realism); and (iii) that our experience of film is like our experienceof the world (perceptual realism). Finally, I briefly discuss the relation between thesemetaphysical claims and the aesthetics of film.Motion picture realismSome theorists argue that film is a realistic medium because it engenders an illusionin us that something is real, when in fact it is not. Gregory Currie has usefully dividedsuch theses into two sorts (1995: 28–30). A cognitive-illusionist theory states that filmTF17557.indb 237 16/9/08 15:47:30ANDREW K ANIA238engenders a false belief in us, such as that we are literally seeing the fictional events ofa film unfold before us. A perceptual illusionist theory states that there is a differencebetween how film appears to us and how it really is, independently of our beliefs aboutit. For instance, film images may seem to move, even if we know that such motion ismerely apparent and not real.Cognitive-illusionist theoriesA number of thoroughgoing cognitive illusions have been attributed to film. Theseclaims are not usually explicitly linked to the illusion of movement (though seeBaudry 2004a [1970], 2004b [1975]; and Panofsky 2004 [1934–47]), but such a link isperhaps implicit, given that the claims are exclusive to film, rather than being appliedto other pictorial, dramatic, or narrative media. (Note, though, that not all filmsinclude moving images. See, for example, Derek Jarman’s Blue [1993].) In some cases,cognitive-illusionist claims might plausibly be rejected as harmless hyperbole, but inothers they seem central to a theory. In any case, as we shall see, they are all false.Film has been claimed to engender the illusion that the real things we see on screen– Jeff Bridges pouring various nonalcoholic liquids into a glass during the shooting ofThe Big Lebowski (1998), for instance – are present to the viewer in the theater, or thatthe viewer is present to these things on the set (e.g., Bazin 1967b [1951]; Metz 1974:1–15, 43). (These claims should not be confused with the thesis of transparency – thatfilm enables us to see things that are not present – discussed below.) Alternatively, filmhas been claimed to engender the illusion that when the viewer sees fictional things onthe screen – The Dude making a White Russian, for instance – she believes that theyare really present to her, or she to them (e.g., Balázs 1970: 48).All of these claims are untenable for the same sorts of reasons. First, they do notcohere with viewers’ behavior. Depending on one’s other beliefs, a person might actin various ways if she thought she was in the presence of Jeff Bridges or The Dude.But no one watching the movie asks for Bridges’ signature, or inquires of The Dudewhether there’s enough Kahlúa for two. A familiar response to such objections isto weaken the claim. Is the viewer in a state of partial belief, or uncertainty aboutwhether or not, she is in the presence of Bridges or The Dude? Again, no. Thoughsomeone might approach a stranger on the street if she suspected he were a famousactor, not even Bridges’ biggest fan would ask for his signature while watching The BigLebowski. Similarly, no one even considers attempting to intercede to stop the nihilistsfrom dropping their ferret into The Dude’s bath (Currie 1995: 24–5; Walton 1990:197–200).Second, these claims do not cohere with viewers’ other beliefs about their abilitiesto move about in and perceive the world. If you believed you were present in thefictional world, or even at the shooting locations, of a film, then you would have tobelieve that you were able to jump from one place and time to another almost instantaneously with a cut from one scene to another; you would have to believe that youcould perceive things in the manner of a zoom-dolly shot, or as if through differentcolored filters.TF17557.indb 238 16/9/08 15:47:30REALISM239A slightly weaker form of cognitive illusionism holds that the viewer believes thatthe fictional things represented in the film are real but is not misled about the film’sstatus as a representation. That is, the viewer knows he is watching a film but believes,at least while the film is playing, that it is documentary footage. One reason to doubtthis hypothesis is that we are no more confused or forgetful about the status of thefilm we are watching (be it fiction or documentary) than we are about the fact that weare watching a film, as opposed to perceiving reality. Moreover, things are commonlypresented on screen that we could not coherently believe to have been filmed inreality, such as the nihilists’ threatening The Dude while he takes a bath.More plausible than any of these cognitive-illusionist claims is the idea that theviewer imagines, rather than believes, what is fictionally represented on the screen(Walton 1990). But such a thesis requires no reference to the viewer undergoing anyillusion.Perceptual illusionismMost writers on film argue or assume that the apparent motion of film images is anillusion. However, Gregory Currie, who rejects all forms of cognitive illusionism, alsorejects perceptual illusionism; that is, he defends the view that film images really do move(Currie 1995: 34–47, 1996: 334–42). Currie thinks the burden of proof in these debatesrests on the illusionist, since we should take things at face value unless we have reasonto doubt them. The reason most people think that the motion of film images is illusoryis that they understand that film projectors project a succession of still images, separatedby moments of darkness, in such rapid succession that we seem to see a continuouslyprojected image that moves. Thus there is no one film image that moves, any more thanthere is one image that moves in a “flipbook” (Kania 2002: 244, 246n8).Currie responds with an analogy to colors. Colors are generally thought to be“response-dependent” properties, that is, there is no way to specify what “beingred” is without reference to the way we experience the world. (A property such as“being square,” by contrast, could be specified in purely geometric terms, without anyreference to our experience of that property.) Nonetheless, colors are not illusory,since we are not wrong when we say that blood is red. Hence, we can say that themotion of film images is real, albeit response dependent.This argument rests on a false analogy. Colors are response-dependent properties,but motion is not. Motion consists, at least, in a thing’s being in contiguous spatiallocations at contiguous moments in time. No film image meets that condition (Kania2002: 254–7; Gaut 2003: 634–5). Currie can argue that film images have a distinctproperty that is related to motion – response-dependent motion – but this merelyrenames the illusion. A stick in a glass of water could be described as “responsedependently bent,” but this does nothing to militate against the fact that the stick isreally straight, despite appearing bent, plainly speaking.Currie has a number of ancillary arguments against perceptual illusionism: (1) Iffilm images didn’t move, then all we could see in a cinema would be static images(1995: 34–5). (2) If photography is transparent, and we see people moving in films,TF17557.indb 239 16/9/08 15:47:31ANDREW K ANIA240then the film images must be moving (1995: 38). (3) The identity of film images isresponse dependent, so we should expect their motion to be response dependent, also(1995: 40–2). However, each of these arguments either falsely assumes that there isa continuously existing image on the screen, or fails to acknowledge the vacuity ofattributing response-dependent motion to images.It is worth noting that the above arguments apply only to “moving pictures”produced by certain mechanisms (film, video, zoetrope, etc.). It is much moreplausible to suppose that images produced by the uninterrupted projection of light canreally move. A simple example would be a spotlight sweeping a prison compound. Anexample from art is the images on the screen of a shadow play. Furthermore, we canimagine a future technology where the projector’s beam is continuously shone througha transparent cell, whose contents move. Nonetheless, it is not indisputable that suchimages move, since it is not clear what the identity conditions of images and shadowsare (Currie 1995: 30–4; Cassati and Varzi 1994: 175–6). Suppose a searchlight is sweptacross a cloudy night sky. When the beam runs over a gap in the clouds, does the firstspot end as the beam hits the gap and a new one begin when it hits the next cloud, oris this the same spot as before? If so, where was it in the meantime?Photographic realismMost people agree that seeing a photograph of something is quite different from seeinga painting or drawing of the same thing, and this difference is sometimes captured bysaying that photographs are more realistic than paintings or drawings. Since most,though by no means all, films are photographic, many theorists have argued that filmis a realistic art form on this basis.The characteristic of photography that is appealed to in justifying this sort ofrealism is that photographs are mechanically produced, and thus the appearance of aphotograph is “counterfactually dependent” on the appearance of its subject. That is,if the subject of a photograph had looked different, then the photograph would havelooked different, no matter whether the photographer noticed the difference. Thisis to be contrasted with the case of painting or drawing where the image produceddepends on the beliefs or intentions of the artist. That is, a change in the appearanceof an object will only affect the appearance of a painting, if at all, by way of affectingthe artist’s beliefs about its appearance. (Sound recording is mechanical in the samesense, but has been little discussed, partly due to a tendency to focus on film images atthe expense of film sound, and partly since sound synthesis – the sonic equivalent ofpainting or drawing – is relatively recent and postdates recording technology.)Ontological realismSometimes the nature of photography is used to defend an extreme realism orillusionism of the kind put aside at the beginning of the previous section – that aphotograph is identical with its subject, or that it gives the illusion that it is. Thoughwe have dismissed these claims, the counterfactual dependence of a photograph on itsTF17557.indb 240 16/9/08 15:47:31REALISM241subject does justify a related claim that we might call ontological realism, namely thata photograph of something, unlike a painting of it, guarantees the existence of thatthing (Gaut 2003: 634; Walton 1984: 250). Of course, we must continue to distinguish between the real and fictional contents of an image. A photograph of Jeff Bridgesshows that he existed at some time, but not that The Dude existed.TransparencyA more controversial claim is that photographs are transparent, meaning that inlooking at a photograph you literally see its subject (Bazin 1967a [1945]; Walton1984). (If photography is transparent, but film motion is an illusion, then we reallysee the objects filmed but do not really see them move. If photography is transparentand film motion real, then we really see the filmed objects in motion.)Almost everyone agrees that when you see something out of your window youliterally see that thing, rather than seeing an image in the window. But this examplesits at the top of a slippery slope. In which of the following examples do you literallysee something? Seeing something while wearing glasses, in a mirror, in a distortingmirror, through a periscope, through a telescope, through night-vision goggles, onclosed-circuit television, on a live television broadcast, on a delayed televisionbroadcast, on a taped television show, in a mechanically generated drawing, in aperson’s drawing, in a mechanically generated description, in a person’s description.Almost everyone denies that we see things through descriptions – even mechanicallygenerated ones. Opinions differ about where to draw the line between the window andthe description. What one needs is an argument for a principle according to whichphotography is ruled in or out of the class of transparent media.Here are some arguments against photographic transparency that we can dismiss.It is irrelevant to the transparency claim that the object photographed may no longerexist. When we look at the night sky we see many stars that no longer exist (Walton1984: 252). It is also irrelevant to the transparency claim that when we see a photowe may see no object. If you awake in the middle of the night and look around youmay see nothing, though there are objects all around you. This doesn’t show thatyou don’t see things in ordinary vision (Currie 1995: 57). Another argument againsttransparency contends that photography is like painting after all, since photographsdepend on the intentions of the photographer to choose a particular scene, frame it acertain way, expose the film at a certain moment, and so on. None of this is relevantto the key difference between photography and painting, though. A friend mightblindfold you and take you to see a certain sight from a certain angle when the lightis just right, even holding a frame up to the view so that you appreciate it in a certainway. Nonetheless, you still see the view, so the fact that photographers do equivalentsof all these things does not show that we do not see the subjects of their photographs(Scruton 2006 [1983]: 29–30; Walton 1984: 261–2).More recently it has been argued that the reason photographs are not transparent isthat they fail to provide “egocentric information.” That is, you do not know where thesubject of a photograph is in relation to yourself. But the provision of such informationTF17557.indb 241 16/9/08 15:47:31ANDREW K ANIA242is a central function of vision. Thus, photographs are not a way of literally seeingthings (Currie 1995: 65–9; Carroll 1996d: 61–2). However, the relation between thefunction of vision and what counts as vision is not clear. The function of vision is toprovide us with information in our immediate environment, yet we still literally seethe stars. Also, if we see something in a mirror, then it seems we see something in aperiscope, even if we don’t know how the periscope is set up and thus have no ideawhere what we see is in relation to us (Walton 1997: 69–72).Another recent concern in the debate over photographic transparency has beenthe continuity of the transmission of light. When you look at a star with the nakedeye or through an optical telescope, the light that enters your eye comes directly fromthe star, though it may have been refracted through lenses and reflected off mirrors.Similarly, when you see an object in a mirror or through spectacles, the light thatenters your eyes is the very light reflected off the object. When you see something onclosed-circuit television or in a photograph, however, the light that enters your eyesmay be similar to the light from the object that is responsible for the image, but it isnot that very light. Some take this objection to be decisive (Gaut 2003: 637). Othersargue that a transducer that acted just like a window, but collected light on one sideand instantaneously emitted qualitatively similar light on the other, would be transparent (Currie 1995: 60, 70). It is difficult to see how this dispute could be resolvedwithout begging the question.A further kind of example is that of a three-dimensional model that is counterfactually dependent on some part of the world. Like a photographic film, the visualappearance of such a model is counterfactually dependent on its “subject,” yet, alsolike a film, that appearance is not conveyed by the continuous transmission of light.Gregory Currie considers a clock that determines the positions of the hands of a secondclock (1995: 64–5). Berys Gaut considers a model jungle that is richly counterfactuallydependent on a section of real jungle, to which someone sells tickets, advertising theopportunity to “see real gorillas” (2003: 637; see also Noël Carroll’s model rail-yard,1996d: 61). Gaut rightly moves to the second example, since in the case of the clocksit is not clear that the information preserved from one clock to another is rich enough.(No matter how accurate your porch-light motion sensor is, you do not see the thingthat set it off through seeing the light come on.)Gaut is wrong to think that the model case is decisive, however. He argues that youcould legitimately ask for your money back on discovering that you were seeing modelsrather than gorillas. There are three problems here. One is that there are differentways in which the term “see” is used. If someone sold you the opportunity to see Marsup close with your own eyes and then showed you to a telescope, you might legitimately ask for a refund. Nonetheless, it is true that you see Mars through a telescope.Another problem I suspect is a failure of imagination. It is difficult to imagine amodel of a section of jungle that looks just like the section of jungle, gorillas and all,as opposed to a clunky animatronic. If a model of such, well, photographic realismcould be achieved, it is not clear one could deny its transparency without begging thequestion. Finally, as Gaut says, our intuitions here are partly driven by “the abidinghuman desire to be in direct perceptual contact with objects” (Gaut 2003: 637). ThisTF17557.indb 242 16/9/08 15:47:31REALISM243muddies the waters, since it would be quite reasonable to believe that one was in thepresence of a gorilla if one came upon a perfect robotic model of one, unlike the caseof being in the presence of a photograph of a gorilla, and hence we are more likely tothink of this case in terms of deception. Moreover, again, there are cases of our beingin indirect perceptual contact with things that count uncontroversially as seeing, suchas the mirror and telescope cases.Given the vagueness or polysemy of “see,” it may be that once we have figured outall the ways in which looking at objects in photographs is like and unlike ordinaryvision, any decision to draw the line between literal and nonliteral seeing will bestipulative. Nonetheless, figuring out these similarities and differences is still of valueif we want to understand the nature of photographic film. For if we can at least placefilm viewing with some precision on the spectrum between simply seeing and seeingin a painting (say), we can appeal to this relative immediacy in explaining one way inwhich photographic film is realistic.Perceptual realismMost films are (apparent) moving pictures, and pictures are one kind of representation. They differ from other kinds of representation, such as language, in beingparasitic on ordinary perception. Unlike our understanding of language, we use thesame capacities to recognize what a picture represents as those we use to recognizeobjects and events in the world (Carroll 1985: 82–8; Currie 1996: 327–30). As aresult, pictures have been described as “perceptually realistic.” One way of seeing thedifference is that if you can understand a few pictures in a given style, say, you canunderstand any other picture in that style, while if you can understand a few wordsor sentences in a given language, it does not follow that you will understand otherwords or sentences in that language. Perceptual realism has something to do withthe fact that we perceive pictures as resembling what they represent (Currie 1996:328–30; Walton 1984: 270–3). For instance, if you are likely to confuse a rhinoceroswith a hippopotamus, then you are likely to confuse a picture of a rhinoceros with apicture of a hippopotamus. But you are unlikely to confuse the word “rhinoceros” withthe word “hippopotamus,” since the words do not resemble each other in the waythe animals (and pictures of them) do. Since resemblance is a matter of degree, andperceptual realism appeals to resemblance, so is perceptual realism a matter of degree.For instance, the stylized appearance of a cartoon donkey may be less realistic than aphotographic film of a donkey, but the cartoon is nonetheless perceptually realistic,as opposed to, say, the word “donkey.” To investigate the nature of pictures further isbeyond the scope of this chapter. (See “Depiction” in this volume.) Instead, I will notea few ways in which perceptual realism relates specifically to film.First, not all the visual elements of a film are pictorial. Even in standard narrativefiction films, titles announcing the time and location of scenes are common. Theseare linguistic rather than pictorial. But motion pictures are perceptually realistic withrespect to more properties than still pictures. Since film is a temporal art, in the sensethat the parts of a film have specific duration and ordering, it can be perceptuallyTF17557.indb 243 16/9/08 15:47:31ANDREW K ANIA244realistic with respect to the temporal properties of what it represents (Currie 1995:92–6). Also, whether or not our perception of movement in film is veridical or illusory,it can be perceptually realistic. (It is perhaps worth noting that much “cognitive” filmtheory has focused on the nature of depiction in general, in reaction to the persistentnotion in previous film theory that film is somehow linguistic. For a selection of suchlinguistic theories, see the first section of Braudy and Cohen [2004]. For argumentsagainst such theories, see Currie [1992] and Prince [1993].)Second, photographs are arguably the most reliably perceptually realistic kind ofpictures we have (Cohen and Meskin 2004), and photographic film inherits thatrealism. Adding the points about film’s perceptually realistic representation of timeand movement, we can understand Currie’s claim that “film is a realistic medium, anddeep-focus, long-take style is an especially realistic style within that medium” (1996:328). Deep focus allows us to perceive more easily or directly the spatial relationsrepresented, and a long take allows us to perceive more easily or directly the temporalrelations represented. More immersive formats, such as IMAX and 3-D film arepresumably still more perceptually realistic.Third, perceptual realism is not restricted to vision. Most films are also highlyaurally realistic, at least with respect to diegetic sound (i.e., sound which “comesfrom within the world of the film”). We recognize the fact that a shot has been firedoff screen by using the same aural perceptual capacities that we use to identify realgunshots (though, as with images, these sounds may be more or less stylized. Think,for instance, of the typical sound effect that accompanies an on-screen punch). Sometheme-park movie rides represent the rocky progress of a spaceship by shaking theaudience’s seats, thus representing the feel of a rocky ride in a perceptually realisticway. There have also been experiments with olfactory realism, such as Smell-O-Visionand “odorama.”Finally, there is a synesthetic perceptual realism that comes from a matching ofsounds and images (and movements, smells, and so on). Perceptually realistic images,coupled with perceptually realistic sounds, will result in a markedly less realistic film ifthe images and sounds do not match than if they do. (Note, however, that this simpleidea calls immediately for qualification, since most film music is “nondiegetic,” thatis, unlike most sound effects, it does not represent sounds in the fictional world of thefilm. See Levinson [1996] for a consideration of such music.)Aesthetic implicationsThe reason realism has been such a hot topic in film theory is twofold. Theorists haveargued, first, that film is uniquely, essentially, or particularly realistic; and, second, thatthis has implications for how films ought to be made. I have investigated three waysin which the first of these claims has been defended, though it must be noted that Ihave focused on the extent to which film might be considered realistic. Whether whatrealism we have found is unique or essential to film is a further question, one to bequite skeptical about. For each kind of realism we have had to limit our discussion tocertain types of film – those using traditional projection apparatus, photographic film,TF17557.indb 244 16/9/08 15:47:31REALISM245films employing a particular style, and so on. No such restricted realism can be takenas essential to film in general. But even if such a claim could be defended, it is not clearwhat implications this would have for what we might call the aesthetics of film – thestudy of what makes a film a good one.Noël Carroll has argued persuasively against “medium-essentialism,” the view thatone ought to exploit effects of one’s art form that are either unique to it, or that itachieves better than any other form (1996a [1984–5], 1996b [1985], 1996d). For onething, it is doubtful that the most interesting effects achievable in film are achievableonly in film. Further, it is difficult to know how we could judge whether films or novels,say, are better at telling stories – something at which they both excel. Supposing sucha decision were made, it would be strange to decide that the inferior form should nolonger attempt to tell stories.So in general we should be suspicious of deriving aesthetic imperatives from claimsabout the nature of the medium, including claims about its realism. I end by considering some aesthetic issues specific to the types of realism we examined above.Motion picture realismSurprisingly few film theorists have focused on film’s ability to produce (apparently)moving images (though see Arnheim 1957: 161–87; and Kracauer 1965: 41–5). Anyway,the idea that film should concentrate on the depiction of motion is questionable forthe general reasons given above. Noël Carroll argues that the motion of film imagesshould affect the aesthetics of film in a different way. He points out that the apparentmotion of images is neither exclusive nor essential to film – there are films withoutmotion and moving images in other media – but he argues that this suggests we shouldreorient the study of these media in an inclusive way. Film, television, video, etc., areall media in which the motion of images is possible, if not necessary. Thus it makessense to study motion pictures or moving images in a broad sense, rather than focusingparochially on film (Carroll 1996d). (Note that it will make no difference in mostcases whether the motion of the images is real or illusory.)Photographic realismThe dispute over the transparency of photographs has been hotly debated in partbecause of its supposed implications for the aesthetics of photography and film. RogerScruton (2006 [1983]) has argued that since a photograph is transparent, it cannotbe a representation in the sense of expressing a thought about its subject. As a result,Scruton claims that a photograph or film cannot in itself be aesthetically valuable,though what we see through it may be aesthetically valuable.If Scruton is wrong about the transparency of photographs, and if this implies thatthey are representations, then his argument will not go through. However, the implication is questionable, since it may be that the mechanical nature of photography issufficient to prevent representation, yet insufficient to achieve transparency (Lopes2003: 441). Nonetheless, Dominic Lopes argues that even given Scruton’s restrictedTF17557.indb 245 16/9/08 15:47:31ANDREW K ANIA246notion of representation, photographs can be aesthetically valuable despite theirtransparency, since there is still a difference between seeing an object face-to-face andseeing it through a photograph. We can take an aesthetic interest in a photograph,then, if we take an interest in (i) the object we see through it; (ii) the way in whichthe photograph enables us to see it; and (iii) the interplay between (i) and (ii) (Lopes2003: 442–6).Another way to defeat Scruton’s argument is to show that transparency does notpreclude a photograph’s being a representation. Walton argues that photographs arerepresentations since they are (usually) props in games of make-believe (1984: 253–4,1990: 88, 330, 1997: 68). Stephen Davies argues that if photographs are transparent,then so too are some paintings and drawings. Both photographs and handmadepictures can be counterfactually dependent on their subjects, though the dependenceis mechanical in one case and intentional in the other. Since paintings and drawingsare paradigmatic representations, transparency cannot preclude representation. Thisdoes not open the door to all seeing being mediated by representation. What we seethrough windows, mirrors, spectacles, and so on is not only counterfactually dependentbut continuously dependent on the thing seen. It is the spatiotemporal separationbetween image and object provided by the camera, the canvas, or whatever, thatmakes pictures representations distinct from the things we see through them (Davies2006: 185–8). Note that cinematography does not violate this condition, despite thefact that it does not capture its object at one moment in time, like still photography.Whether or not you think that film images move, it is still the case that either eachframe or the temporally extended image, once recorded, is no longer sensitive tochanges in its subject.Perceptual realismGregory Currie argues that films with a long-shot, deep-focus style are more perceptually realistic than others, since they represent spatial and temporal relations (betweenthings) by means of spatial and temporal relations (between images). Our experienceof such films is more like our experience of the world than is our experience of filmswith rapid editing, for instance (1996: 327–30). Should more films, then, employthis style? Clearly this is a matter for debate. Some (such as Bazin) will argue for thesuperiority of such a style on the basis of its ability to involve us in the world of thefilm, pointing to the success of films in a realist style. Others will argue that this is adangerous seduction, and that filmmakers should work against it, alienating the viewerin order to force an awareness of the medium upon her, pointing to the success of filmssuch as those of Jean-Luc Godard. It is not clear that one side must win this debate.Film, like every other artistic medium, is capable of employing many different stylesfor many different purposes (Carroll 1996b [1985]), and we are capable of appreciatingmany different kinds of films.TF17557.indb 246 16/9/08 15:47:31REALISM247ConclusionsSome films are surely better than others, and perhaps some kinds of films are superior toothers. However, it seems unlikely that we can discover which ones by measuring thedegree to which they are realistic. Nonetheless, the different kinds of realism we haveinvestigated here are relevant to the study of film for other reasons. Most simply, theyallow us to describe the nature of film more precisely, in terms of its illusory nature,the extent to which it is transparent, and one dimension along which cinematic stylescan be said to be realistic. These descriptions, in turn, may factor into a psychologicalexplanation of the power and popularity of motion pictures in general, or specific filmsand styles (Carroll 1985). We should remain wary, however, of drawing conclusionsabout the value of those films, styles, and motion pictures in general, from premisesabout their power and popularity.See also Consciousness (Chapter 4), Definition of “cinema” (Chapter 5), Depiction(Chapter 6), Film as art (Chapter 11), Formalism (Chapter 12), Medium (Chapter16), Music (Chapter 17), Ontology (Chapter 20), Spectatorship (Chapter 23),Sound (Chapter 24), Style (Chapter 25), Rudolph Arnheim (Chapter 27), BertoltBrecht (Chapter 30), Cognitive theory (Chapter 33), Edgar Morin (Chapter 38), andPhenomenology (Chapter 40), Dogme 95 (Chapter 44).ReferencesArnheim, R. (1957) Film as Art, Berkeley and Los Angeles: University of California Press.Balázs, B. (1970) Theory of the Film: Character and Growth of a New Art, New York: Dover.Baudry, J.-L. (2004a [1970]) “Ideological Effects of the Basic Cinematographic Apparatus,” in L. Braudyand M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: OxfordUniversity Press.—— (2004b [1975]) “The Apparatus: Metapsychological Approaches to the Impression of Reality in theCinema,” in L. Braudy and M. Cohen (eds.) Film Theory and Criticism: Introductory Readings, 6th ed.,New York: Oxford University Press.Bazin, A. (1967a [1945]) “The Ontology of the Photographic Image,” in H. Gray (trans.) What Is Cinema?vol. 1, Berkeley and Los Angeles: University of California Press.—— (1967b [1951]) “Theater and Cinema: Part Two,” in H. Gray (trans.) What Is Cinema? vol. 1,Berkeley and Los Angeles: University of California Press.Bordwell, D. (1996) “Contemporary Film Studies and the Vicissitudes of Grand Theory,” in D. Bordwelland N. Carroll (eds.) Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press.Bordwell, D., and Carroll, N. (eds.) (1996) Post-Theory: Reconstructing Film Studies, Madison: Universityof Wisconsin Press.Braudy, L., and Cohen, M. (eds.) (2004) Film Theory and Criticism: Introductory Readings, 6th ed., NewYork: Oxford University Press.Carroll, N. (1985) “The Power of Movies,” Daedalus 114: 79–103.—— (1988a) Philosophical Problems of Classical Film Theory, Princeton, NJ: Princeton University Press.—— (1988b) Mystifying Movies: Fads and Fallacies in Contemporary Film Theory, New York: ColumbiaUniversity Press.—— (1996a [1984–5]) “Medium Specificity Arguments and the Self-Consciously Invented Arts: Film,Video, and Photography,” in Theorizing the Moving Image, Cambridge and New York: CambridgeUniversity Press.TF17557.indb 247 16/9/08 15:47:31ANDREW K ANIA248—— (1996b [1985]) “The Specificity of Media in the Arts,” in Theorizing the Moving Image, Cambridgeand New York: Cambridge University Press.—— (1996c) “Prospects for Film Theory: A Personal Assessment,” in D. Bordwell and N. Carroll (eds.)Post-Theory: Reconstructing Film Studies, Madison: University of Wisconsin Press.—— (1996d) “Defining the Moving Image,” in Theorizing the Moving Image, Cambridge and New York:Cambridge University Press.Cassati, R., and Varzi, A. (1994) Holes and Other Superficialities, Cambridge, MA: MIT Press.Cohen, J., and Meskin, A. (2004) “On the Epistemic Value of Photographs,” Journal of Aesthetics and ArtCriticism 62: 197–210.Currie, G. (1992) “The Long Goodbye: The Imaginary Language of Film,” British Journal of Aesthetics 33:207–19.—— (1995) Image and Mind: Film, Philosophy, and Cognitive Science, Cambridge and New York: CambridgeUniversity Press.—— (1996) “Film, Reality, and Illusion,” in D. Bordwell and N. Carroll (eds.) Post-Theory: ReconstructingFilm Studies, Madison: University of Wisconsin Press.Davies, S. (2006) The Philosophy of Art, Malden, MA; Oxford: Blackwell.Gaut, B. (2003) “Film,” in J. Levinson (ed.) The Oxford Handbook of Aesthetics, Oxford and New York:Oxford University Press.Kania, A. (2002) “The Illusion of Realism in Film,” British Journal of Aesthetics 42: 243–58.Kracauer, S. (1965) Theory of Film: The Redemption of Physical Reality, London: Oxford University Press.Levinson, J. (1996) “Film Music and Narrative Agency,” in D. Bordwell and N. Carroll (eds.) Post-Theory:Reconstructing Film Studies, Madison: University of Wisconsin Press.Lopes, D. (2003) “The Aesthetics of Photographic Transparency,” Mind 112: 433–48.Metz, C. (1974) Film Language, trans. M. Taylor, New York: Oxford University Press.Panofsky, E. (2004 [1934–47]) “Style and Medium in the Motion Pictures,” in L. Braudy and M. Cohen(eds.) Film Theory and Criticism: Introductory Readings, 6th ed., New York: Oxford University Press.Prince, S. (1993) “The Discourse of Pictures: Iconicity and Film Studies,” Film Quarterly 47: 16–28.Scruton, R. (2006 [1983]) “Photography and Representation,” in N. Carroll and J. Choi (eds.) Philosophyof Film and Motion Pictures: An Anthology, Malden, MA: Blackwell.Walton, K. (1984) “Transparent Pictures: On the Nature of Photographic Realism,” Critical Inquiry 11:246–76.—— (1990) Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA:Harvard University Press.—— (1997) “On Pictures and Photographs: Objections Answered,” in R. Allen and M. Smith (eds.) FilmTheory and Philosophy, Oxford: Clarendon Press; New York: Oxford University Press.TF17557.indb 248 16/9/08 15:47:31 – that the style of this film would do well to be adopted by each and every maker of films henceforth. While Renoir was clearly as concerned with the content of the film as with its form(the film is even today considered by many to be an unsurpassed cinematic masterpiece), a studyof its content will indeed be more or less excluded from the analysis of the following sequences inthe interest of addressing Kracauer’s initial thesis about the nature of the filmic medium. As wasjust alluded to, just about any scene in La Grande Illusion would serve to exemplify the CinematicApproach, but of course, what better place to begin than with the film’s opening shot.The opening shot of the film is unconventional in that it begins with a close-up rather than anestablishing shot from a distance – this would be the norm in any Hollywood film, for example. Arecord spins on a gramophone accompanied by music, the camera then pans upward toaccommodate Marechal, who is perched over the record, seemingly overcome with nostalgia of hispre-war life. Marechal’s attention is then drawn to his comrade Halphen as he calls out to him,which might otherwise mean an alternating shot would be required to follow the action, but inkeeping with the Cinematic Approach the camera instead tracks Marechal as he moves close toHalphen to engage him in conversation. The two are some distance from the camera on the otherside of the room, but the deep-focusing of the camera allows the viewer to now see, simultaneously,other men conversing at a table in the foreground, as well as the performance of bar-tendingbanalities by a barkeep to the rear of Marechal and Halphen. The camera remains fixed on thedoorway as the two men part and Marechal moves back toward the gramophone and out of frameas Ringis approaches. The camera then tracks back to the gramophone once more to frame the two4Kino: The Western Undergraduate Journal of Film Studies, Vol. 4 [2013], Iss. 1, Art. 10http://ir.lib.uwo.ca/kino/vol4/iss1/10men. As Ringis and Marechal move toward the doorway, the camera again tracks the action asthey exit. This shot lasts for approximately one minute. While this is among the longer takes ofthe film it is by no means an anomaly; Julian Jackson notes in his book on La Grande Illusion thatRenoir’s films have a consistently higher average shot length than other films. Jackson offers theexample of another French film, Julian Duvivier’s Pepe le Moko (1937), which at only one hourand forty minutes has 452 shots while La Grande Illusionhas only 352 shots at a run time of twohours (Jackson 38). Comparing this opening shot to the entire opening sequence of The Matrix(Wachowski, 1999), for example, there is a stark contrast in terms of approach. And yet, apartfrom its Science Fiction plot, modern special effects, and computer-generated imagery, the liveaction shots in the opening sequence are in many ways against the grain of the CinematicApproach. There are many cuts and little camera movement, and the individual shots utilize a(relatively) soft-focal range. In the example of The Matrix- -and there are surely other, possiblybetter, examples – the formative tendency overpowers the realistic tendency, and so it follows thatthis film is not ‘aesthetically valid’ as per Kracauer’s initial thesis. But, of course, part of theCinematic Approach also entails a sparing use of techniques that are associated with the formativetendency. Renoir in fact does so brilliantly with La Grande Illusion. Take just one example, hisuse of shot-reverse-shot cutting as opposed to the more predominant use of long-take, deep-focuswithin the film.There are only a few instances in which conventional shot-reverse-shot practices are used in LaGrande Illusion, and as per the Cinematic Approach, these instances only serve a punctuativepurpose in that they stand out starkly from the rest of the film rather than blending in and beingotherwise imperceptible. Martin O’Shaughnessy discusses one particular sequence in a chapter ofhis book on the film that deals with the film’s style in the context of its “group dynamics”(O’Shaughnessy 59). During the exchange in which the engineer asks Marechal whether or not deBoeldieu can be trusted, shot-reverse-shot cross-cutting emphasizes their difference. WhenMarechal finally convinces his comrade of de Boeldieu’s trustworthiness, they are then framedtogether in close-up. The engineer then begins to discuss escape plans, and with their differenceresolved the cross-cutting ceases. This pattern recurs in the scene in von Rauffenstein’sWintersborn chamber and again during the successful escape. In the latter example, shot lengthsprogressively shorten as more cuts are used to capture the action from less ambiguous perspectivesthan the more common shot lengths and focal-depths used in the rest of the film. Then, when deBoeldieu is shot by von Rauffenstein this more rapid editing quickly grinds to a near halt until deBoeldieu dies in a hospital bed alongside von Rauffenstein in what is one of the more lengthy takesof the film. The usage of techniques belonging to the formative tendency are in full adherence tothe Cinematic Approach because they are used so infrequently and with such a great sense ofpurpose within the scheme of the film. The inverse of this usage of techniques belonging to theformative tendency can be found in Magnolia (P.T. Anderson 1999). The majority of this filmutilizes standardized Hollywood conventions that are generally contra the Cinematic Approach,however, there are several instances where long-take or deep-focus techniques are used. Thesetechniques belonging to the realistic tendency serve a similar punctuative role to the use ofshot-reverse-shot cross-cutting in La Grande Illusion. The difference being that in Magnolia, theuse of techniques belonging to the realistic tendency far outweighs the use of techniques belongingto the formative tendency. So, the film fails to satisfy the requirements of the Cinematic Approach5Edwards: Realism, Really?Produced by The Berkeley Electronic Press, 2013and thus would not be considered aesthetically valid according to Kracauer.What should be apparent in the discussion contained with this essay is that with boldtheoretical prescriptions comes a greater potential for self-contradiction and dogma. It is for thisreason that no grand theory such as ‘Cinematic Realism’ should ever be considered as purportingto be definitive and all encompassing – even if the author impels this conclusion. Rather, anytheory should be used as an analytical tool for a task which it is best suited. If thought of in thisway, any self-contained theoretical framework can contribute to significant enlightenment in agiven area of study. Thus, Kracauer’s initial thesis is not incorrect, but it must be considered inlight of its inherent limitations.6Kino: The Western Undergraduate Journal of Film Studies, Vol. 4 [2013], Iss. 1, Art. 10http://ir.lib.uwo.ca/kino/vol4/iss1/10Works CitedCarroll, Noel. The Philosophy of Motion Pictures. Malden, MA: Blackwell Pub., 2008. Print.Corrigan, Timothy, Patricia White, and Meta Mazaj. "Andre Bazin From What Is Cinema?"Critical Visions in Film Theory: Classic and Contemporary Readings. Boston: Bedford/St. Martin's, 2011. N. pag. Print.Corrigan, Timothy, Patricia White, and Meta Mazaj. "Siegfried Kracauer From Theory of Film."Critical Visions in Film Theory: Classic and Contemporary Readings. Boston: Bedford/St. Martin's, 2011. N. pag. Print.Jackson, Julian. "Variations on Realism: 'Interior Truth' and 'Exterior Truth'" La Grande Illusion.Basingstroke: Palgrave Macmillian, 2009. N. pag. Print.O' Shaugnessy, Martin. "Film Style and Group Dynamics." La Grande Illusion (Jean Renoir,1937). London: I.B. Tauris, 2009. N. pag. Print.Renoir, Jean. La Grande Illusion. London: Lorrimer, 1968. Print.Currie, Gregory. Image and Mind: Film, Philosophy and Cognitive Science. Cambridge[England]: Cambridge UP, 1995. Print.Kracauer, Siegfried. Theory of Film. London: Oxford UP, 1974. Print.Films CitedGrande Illusion, La. Dir. Jean Renoir. Compagnie Jean Renoir, 1937.Magnolia. Dir. Paul Thomas Anderson. Perf. Julianne Moore and Tom Cruise. New LineCinema, 1999.Matrix, The. Dir. Andy Wachowski and Lana Wachowski. Perf. Keanu Reeves. Warner Bros.,1999.Pépé Le Moko. Dir. Julien Duvivier. DisCina, 1937.7Edwards: Realism, Really?Produced by The Berkeley Electronic Press, 2013 |
|
||
|
|
The
Use of Cognitive Illusionism in Predestination
|
|
||
|
|
|
by
|
|
|
|
|
<Your
Name>
|
|
||
|
|
|
<Date>
|
|
|
|
|
<Lecturer’s
Name and Course Number>
|
|
||
|
|
|
|
||
|
|
|
|
|
|
|
Cognitive and perceptual illusionism
while of interest as separate topics in literature, are complimentary in film
making. This is especially true in sci-fi movies where the audience have to
believe what they are seeing is real to perceive it as real.
|
||||
|
|
|
|
|
|
Predestination is a 2014 film in which the
protagonist is stuck in time loop. The plot relies heavily on time travel,
which is the main way used to advance the plot. Cognitive illusionism is
applied effectively in the film. For one, it is assumed that the protagonist
moves both himself and his later self when using the coordinate transformer,
the audience will most of the time actually believe that time travelling to the
past was actually achieved.
For the technique to actually work, it is
not always important for images to be used though they are an integral part of
achieving the illusion. In the film, this was achieved using dialogue between
the protagonist and his younger self as well as monologue on the part of the
younger self. The film is a great example of how other aspects of film making
can be harnessed to achieve the desired effect on the audience by the film
maker.
It would be impossible for predestination
to hold water as a time travel sci-fi thriller without the use of perceptual
illusionism. There is no way to actually time travel and the film maker chose
to simply make characters ‘disappear’ when they time travelled. There is also
no way to actually change one person to another person during the shooting of a
film, yet in predestination there were more than three people representing the
same character.
To
create the desired effect of cognitive illusionism, perceptual illusionism was
employed. When time travelling, objects are pushed away from the point of
departure and at the destination. Since no one actually knows what happens when
one time travels, this creates a plausible reaction in the real world of what
would happen in such a scenario. Obviously, how this was achieved during
filming and the illusion it creates to the audience are quite different.
Another application of perceptual realism
is in the transition between characters depicting the protagonist. In
particular, the uncanny valley effect was effectively used to further separate
the other people depicting the protagonist and the Unmarried Mother by changing
the Unmarried Mother’s facial features. This helped in further cementing the
different mental models of the different characters in the audience’s mind,
further promoting cognitive illusionism. At the same time, perceptual
illusionism was employed in order to reduce the effects or realism
inconsistency theory (Elsevier, 2015).
In conclusion, it can be said that
cognitive illusionism works really well in films that requires the viewer to
imagine beyond reality to create an alternative reality such as Predestination
and The Matrix. The use of perceptual illusionism and other aspects of film are
great in aiding achieve the desired cognitive illusionism.
Reference List
Bibliography
Elsevier, B.V. (2015) Reducing
consistency in human realism increases the uncanny valley effect; Increasing
category uncertainty does not. Available at:
http://www.sciencedirect.com/science/article/pii/S0010027715300755 (Accessed:
01 February 2017).
No comments:
Post a Comment