The 'I' of the Beholder
_Projecting Illusion: Film Spectatorship and the Impression of Reality_
Cambridge: Cambridge University Press, 1995
'In projective illusion I experience a pictorial or dramatic representation as if it were a fully realized world of experience and not a representation.' (82)
Legend has it that in 1895 at the first public screening of the films of the Lumiere brothers, audience members reacted to the cinematic image of a train entering a station by screaming, ducking and running away from the screen. I sometimes wonder if the behavior of these early filmgoers was motivated at some unconscious level by a desire to garner for themselves a place in some sort of 'oft-quoted anecdote' hall of fame. Surely, it is an example that is frequently mentioned by cinema theorists as we tackle questions concerning the role the impression of reality plays in the film experience. Although there are certainly a number of ways to explain the behavior of the audience at the Lumieres' film, some would argue that the incident demonstrates the power of the cinematic image to fool the human mind: to confuse us into thinking that an object represented on the screen is actually in our presence. 
Of course, since 1895 millions of film viewers have sat calmly watching movies that seem to show criminal actions without leaving the theater to call the police, movies that take place in the Arctic without reaching for their coats, and so on. Few people seem willing to argue that the average filmgoer ever actually believes that they are directly witnessing the events represented on the screen as they unfold. Yet, there are still occasions during films when viewers cry out and perhaps even duck. Even when we remain quietly in our seats, many of us wonder if there are not moments during the film experience when the fact that we are watching images projected on a screen somehow, in some sense, fades from relevance.
Recently there have been a number of carefully argued attacks against the view that such moments occur.  However, others continue to claim that these moments exist and that they are of vital importance to the film experience. In _Projecting Illusion: Film Spectatorship and the Impression of Reality_, Richard Allen not only defends the significance of such moments, he also attempts to explain the mental processes that make such moments possible. Films, Allen claims, encourage us to voluntarily experience a type of sensory illusion during which we perceive the film's images 'as a fully realized world' (4). This illusion, however, merely involves the ways in which we perceive the images; it does not alter our beliefs about the nature of those images. In other words, we never form the belief that we are present at the film's events, even though, to us, it may *look* like we are. Allen calls this sort of sensory illusion 'projective illusion' and states that it is a species of illusion that can be engendered 'by almost any legible representation', but which is best fostered, at least at present, by cinematic representations (111).
Naturally, the experience of such a sensory illusion is enhanced by, and one might even say depends on, the ability of the cinematic image to create the impression of reality within the minds of its viewers. In fact, Allen introduces the theory of projective illusion as part of a defense of the claim that the impression of reality is of vital importance to the cinematic experience. This is a claim that Allen shares with a group of theorists that he classifies as being a part of 'contemporary film theory'.
Like Noel Carroll, Allen uses the term 'contemporary film theory' to refer only to Althusserian-Lacanian film theory as opposed to, say, all film theory since 1960 (1). One of the theoretical assumptions that unites this contemporary film theory, as Allen defines it, is 'a certain understanding of the impression of reality in cinema' (1). Allen believes that this understanding fails to properly characterize both the nature and the role of the impression of reality in the cinematic experience. However, unlike Carroll, Allen does not believe that this failure justifies the abandonment of all theories that explain our mental involvement with films in terms of illusion. Allen's theory of projective illusion represents an alternative that he believes is not susceptible to the attacks leveled at contemporary film theory.
To make this immunity clear, Allen spends the first two chapters of his book explaining exactly where he believes contemporary film theory went wrong. To do so, he takes the reader on a critical tour of the movement. The tour begins with Louis Althusser's account of ideology, upon which theories of the nature and importance of the cinematic apparatus were based. It continues through to the theories of language and signification developed by Jacques Lacan and Julia Kristeva (amongst others), which were used to explain how the effects of the cinematic apparatus upon the viewer are maintained and augmented by the combination of images for the purposes of storytelling. The tour ends with a critique of the epistemological assumptions underlying contemporary film theory and of the sources of these assumptions: Edmund Husserl's transcendental phenomenology and Jacques Derrida's philosophy of deconstruction. Along the way, Allen manages to attack contemporary film theory from all sides, criticizing both the psychological/philosophical theories at their base, most of which were not developed with cinema in mind, and the ways in which these theories were used by film theorists such as Jean-Louis Baudry, Jean-Pierre Oudart, Daniel Dayan, and Stephen Heath to describe the cinematic experience and the nature of the cinematic illusion.
It is hard to give a brief summary of this critical tour that does it justice. It is clear, however, that Allen praises contemporary film theory for its insistence that cinema creates the illusion of reality -- that the experience of watching cinematic images is, in important ways, deceptively similar to the experience that viewers would have were they directly witnessing the events depicted. It is also clear that Allen condemns contemporary film theory primarily for its description of the viewer as a passive victim of this illusion -- as someone helplessly duped into mistaking the nature of her film-going experience and of her place in it.
For contemporary film theorists, Allen explains, this passive submission to the illusion of cinema is said to be analogous to the passive submission of the human subject to the constructive forces of language and ideology. The subject submits to the manipulation of these forces largely, so the story goes, because she is unaware of their influence on her. This unawareness is said to have its roots in the subject's own sense of self identity, which is built around a conception of herself as a free agent able to know, and often control, a reality that is independent of the processes of signification and conceptualization that she uses to describe it. Rather than reflecting 'how things really are', however, this conception is itself a product the subject's conceptual frameworks, one which masks her radical dependency on these frameworks. Therefore, the subject is said to fundamentally misunderstand the nature of her world and her position in it. This misunderstanding is sustained by society in order to discourage true social change.
According to contemporary film theorists, Allen tells us, cinematic illusion causes filmgoers to undergo a similar misunderstanding; this time concerning the nature of their relationship to the world of a film. There are a number of factors that are said to contribute to the creation of this illusion. Not only do cinematic images look like the objects that they depict, they are also said to reproduce the conception of knowledge that lies at the base of our sense of identity, namely the conception of knowledge as the 'unclouded, transparent perception of [a] real object' (10). By reproducing this conception, the images encourage the viewer to identify with the camera and with its seemingly unrestricted access to the world of the film. This identification is reinforced and developed by techniques of editing and narration that work to insure that the illusion grabs hold of us and keeps a hold until the credits roll.
Due to this manipulation of the viewer, contemporary film theorists identify cinematic illusion with the perpetuation of ideology. Therefore, according to many of these theorists, cinema can become a tool for social change only if we break the hold of cinematic illusion by creating films that challenge the cinematic conventions of realism or that employ non-standard techniques of editing or narration, etc.
In sharp contrast to the contemporary film theorists that he describes, Allen denies that films that encourage the experience of cinematic illusion *necessarily* reinforce prevailing ideology. Allen argues that the film viewer is an active participant in her film-going experience: that she can choose whether or not to experience cinematic illusion, whether or not to identify herself with the position of the camera, and whether or not she experiences the film through the socially accepted, *standard* interpretation that it most outwardly presents. In the first chapter, Allen argues that the interpretation of cinematic illusion offered by contemporary film theory is logically incoherent; that by its own standards it is at best vacuously true; and that it makes claims about the ways in which viewers must interpret cinematic images and editing techniques, without providing adequate reasons why, or evidence that, they must do so. He also attacks the theories of subjectivity and signification that lie at the base of contemporary film theory, denying that the human subject is inescapably at the mercy of ideology. Worried that these arguments will not be convincing to anyone who believes that the human subject is a 'prisoner of language' capable of only the 'illusion' of knowledge (47), Allen devotes all of his second chapter to a critique of this view and of its sources in the work of Husserl and Derrida. Appealing to the later work of Wittgenstein, Allen argues that the claim embodies a false conception of representation, of its relationship to knowledge and, therefore, of the relationship between the mind and reality.
Allen's critical tour of contemporary film theory will be of obvious interest to cinema theorists who have found inspiration in the works that he discusses. I suspect that supporters of contemporary film theory may find that Allen's arguments against the theory and its sources, though numerous, are perhaps a bit too rushed to be thoroughly convincing.  For example, in one of his criticisms of Althusser, Allen argues that, although the individual is surely shaped by social rules and guidelines, this process is not completely out of her control, because 'human agency defines itself through the shaping of that culture, through the application and interpretation of norms and values' (15). This sounds right to me, but Allen does not go into great detail about what it might mean to say that humans can think themselves into new ways of thinking. This lack of detail leaves a lot of questions unaddressed, questions which those sympathetic to Althusser may believe need answering before Allen's statement can be accepted. To be fair to Allen, however, the answers to many of these questions will depend on the nature of the connections between our mind, our concepts, and reality. This relation is something that Allen discusses at length, especially in his second chapter. 
Allen's critical overview of contemporary film theory is also helpful to those of us who, like myself, feel more at home in quite different theoretical pastures. Allen's arguments against the passivity of cinematic illusion are, I believe, convincing and his summaries provide a useful introduction to contemporary film theory and its connections to other cinema theories as well as to wider philosophical traditions. However, at times these summaries can be quite tough going for those who do not have a lot of experience working with the theories that Allen discusses. This difficulty may encourage some readers to skip quickly through the first two chapters and, especially through the second, which Allen explains is not essential to his arguments. Admittedly, for most readers the third chapter will probably be where the book's greatest excitement lies. Nevertheless, the earlier chapters are valuable, not only for the insight that Allen provides to the successes and failures of contemporary film theory, but also because the arguments that Allen makes in these chapters provide insight into the nature of Allen's own more 'active' view of cinematic illusion.
Allen turns to the task of presenting his own theories about cinematic illusion in the third chapter of _Projecting Illusion_. Viewers are not fooled by cinematic images, Allen argues, they use them. Films allow us to voluntarily experience a species of illusion in which we perceive the film 'as if it were a fully realized world of experience and not a representation' (82). We are able to perceive the film in this way, Allen explains, because it can trigger 'a loss of medium awareness' in which we cease to view the film's images *as images* (82). Instead, taking advantage of the image's resemblance to the objects it depicts, we begin to view the events in the film as if we had taken up the perceptual point of view of the camera. The idea here seems to be that projective illusion allows us to *perceptually* leave the theater and enter directly into the world of the film. 'Instead of looking 'from the outside' upon something staged in this world,' Allen writes, '. . . [you] perceive a fully realized though fictional world that has all the perceptual immediacy of your own' (107). 
Presumably, this perceptual occupancy of the world of the film explains why many people describe their film experiences as an 'escape' and why, as many of us note as we emerge from movie theaters to continue with our day or evening, watching movies can be so disorienting to our own sense of time and space. The strictly perceptual nature of this 'escape', Allen claims, also serves to explain why we do not behave like people who have been whisked off to fictional worlds. Throughout the experience of projective illusion and of the loss of medium awareness that it entails *our beliefs remain intact*. No matter what things *look* like, we know that we never really leave the cineplex, living room couch, or wherever we happen to be watching the film.
To explain how a viewer's perceptions can be at such odds with her beliefs, Allen compares projective illusion to other types of visual illusions. In the Muller-Lyer illusion, two lines of equal length appear in the context of an illustration that causes them to look as if they were of different lengths. The effect is quite striking, and when first encountering the illusion most observers form the belief that the lines are of different lengths. Once the illusion is explained, and perhaps after a little work with a ruler, this belief is replaced by the belief that the lines are of the same length. Still, the *appearance* of the lines remains unchanged: the lines continue to look like they are of different lengths.
Illusions such as the Muller-Lyer illusion seem to present us with clear examples of a case in which our perceptions do not match our beliefs. It is not, however, clear exactly what this means. We cannot claim that we can believe 'x' even though we perceive 'not x' (or vice versa), without attributing propositional content to our perceptions. We then owe ourselves an explanation of what it means to attribute such content to perceptions.
We could argue that such attributions betray a notion of perception as being conceptually richer than it is often thought to be. Visual perception is often described as merely a matter of recognizing shapes and colors or, at best, objects. The job of interpreting these acts of recognition is then attributed to the mind's 'higher level', belief systems.  However, it could be argued that (except, perhaps, when we neglect to wear our corrective lenses) perception allows us to recognize quite complex states of affairs or relations (for example, that the lines are of different lengths), without having to engage in an act of interpretation that would involve our belief systems.  This could explain why in the case of the Muller-Lyer illusion our perception of the illusion remains unaffected by our changing beliefs.
It is not clear that Allen adopts such a notion of perception. Nevertheless, he does state that in the case of the Muller-Lyer illusion the 'perception of the lines as being different length cannot be divorced from the *thought* that the lines are of unequal lengths' (99). The distinction between sensory illusion and epistemic illusion is, therefore, said to be similar to the distinction that Noel Carroll makes between believing something and entertaining a thought about something: both involve propositional content, but in the case of belief the proposition is entertained assertively, whereas in the case of thoughts the proposition is entertained non-assertively, namely, 'without commitment to its being the case' (99).
Allen does not claim that the experience of projective illusion is completely analogous to the experience of illusions like the Muller-Lyer illusion. He does claim, however, that projective illusion is a type of sensory illusion in which the viewers choose to perceive a representation as something which they believe it not to be: as a fully realized world. 
Of course, it is not immediately apparent what this means. For illusion or no, there are clearly some perceptual differences between the cinematic image of an event and that event itself. For example, the image lacks depth and is limited by its frame. Actual perceptions of the events depicted would rarely be so constrained. Furthermore, it does not seem likely that such perceptual constraints are things that a perceptually based (as opposed to belief based) illusion could remove. Also, we have no trouble viewing the image as an image of actors, sets and props. Surely, the moments when we view the image this way and moments when we view it as a projective illusion do not correspond to changes in the visible aspects of the image. We seem to be able to switch back and forth between different ways of seeing the same image. How and why are we able to maintain the experience of projective illusion?
For Allen the answers to all these questions depend on understanding exactly the sorts of things that film viewers do and do not think about. While experiencing projective illusion we do not think about the behavior of actors and directors, the edge of the movie screen, the popcorn we are eating, the work we have yet to do that day, etc. Instead, spurred on by the perceptual similarities of the film's images to reality, we (non-assertively) think, for example, of women named Scarlet O'Hara and men named Rhett Butler and of the things that they are doing. To explain how we are able to switch back and forth from looking at the film as a representation of actors and sets to seeing the film as a fully realized world, Allen appeals to the notion of 'seeing as' or 'seeing aspects'. This notion is perhaps best illustrated by the famous duck-rabbit figure, a picture that can be seen *as* a rabbit or *as* a duck, but not as both at the same time.
The medium of film, Allen argues, allows the viewer, for the most part, to choose whether or not to see the image *as an image* or *as a world*. They cannot, however, perceive it both ways at once. The viewer's choice is, of course, constrained because elements in the film experience can call attention to the medium itself. In fact, some of these are purposely employed by directors (inspired, perhaps, by contemporary film theory) who wish to challenge, or at least toy with, film's illusory nature. Conversely, directors can also use techniques that encourage the experience of projective illusion by trying to avoid calling attention to the medium. Allen suggests that Hollywood genres and narrative styles have been developed to do just this.
Of course, we still might wonder why someone would choose to experience projective illusion at all. To answer this question Allen turns, in his fourth and final chapter, to theories of psychoanalysis and, in particular, to its theories of dreams, fantasy and wish fulfillment. Through projective illusion, Allen suggests, we 'experience the events in a film with some of the psychological closeness and somatic intensity of our own dreams' (125). This closeness and intensity in part explains the enjoyment that we derive from films, because through it, Allen writes, film provides 'a forum for the substitute satisfaction of desires in such a way that it tends to leave us in a state of pleasure that intimates a condition in which our desires are actually satisfied' (126). Of course, our film experiences are different from our dreams, since we remain conscious throughout the film experience and we never lose the belief that we are watching projected representations. The biggest question, then, seems to remain: how are we able to sustain projective illusion when the thoughts that make it up are directly contradicted by beliefs that we have?
To answer this question Allen appeals once more to psychoanalysis: this time to the notion of disavowal. In order to sustain the experience of projective illusion and the pleasure that that experience can provide, the spectator must allow herself not only to consider things which she knows not to be the case, but also to consider these things without letting her knowledge of their falsity interfere. She must temporarily 'suspend' or 'circumvent' her beliefs in order to participate in the imaginary world of the film (136, 139). Since the spectator retains these beliefs, however, she can use them to pull in and out of the experience of projective illusion. This ability to enter and exit the world of the film provides us with an additional source of pleasure, which Allen likens to the pleasure a child might derive from toying 'with the experience of presence and absence as in a game of *fort-da* or peekaboo' (140).
Allen's particular interpretations of disavowal and wish-fulfillment illustrate once again the importance he places on the viewer's active role in the creation of the film experience. Although there are many factors external to the mind of the spectator that contribute to her experience of projective illusion (and which may therefore challenge it), the experience itself depends largely on the willingness and propensities of the viewer to engage in the illusion. According to Allen, this control extends into the experience of the illusion itself. Not only can we determine whether we will experience the illusion, we can also control *how* we experience the illusion. Just because we perceptually identify with the camera does not mean that we psychologically identify with some sort of fictional observer who occupies the position of the camera. We can choose instead to identify psychologically with one of the characters or to not make such an identification at all. Using this notion of spectator control, Allen offers an interesting analysis of, and alternative to, contemporary feminist film theory's claim that 'the psychical mechanisms invoked by the cinema, and, in particular, the classical cinema are gender bound' (143).
All in all, I believe that Allen's views on the illusory nature of the cinematic experience represent an important contribution to the philosophy of film and to film theory in general. Although the experience of projective illusion is said to depend on psychological mechanisms of wish fulfillment and disavowal, mechanisms which would seem to function largely on a subconscious level, Allen makes it clear that the conscious mind plays a very important role in the creation of the cinematic experience. I think that many film theories either neglect the role that the conscious mind plays in the cinematic experience or portray it as an obstacle to that experience. Intuitively, I suspect, however, that the mysteries of the film experience will not be solved until we can understand and explain the actual conscious thoughts that movies engender in us. What does it tell us about consciousness and about ourselves that we can think 'Gosh this movie has great special effects', one instant, and 'Oh no, that's Darth Vader in that Tie Fighter' the next, without experiencing any real cognitive dissonance?
To answer this question, of course, we will need a better understanding of the relationship between our thoughts, our beliefs and our perceptions. I believe that Allen has provided us with an excellent way to approach these questions through the study of our experiences with movies. However, I think that a lot remains to be said. For example, Allen is not completely clear exactly what sorts of thoughts he feels we are entertaining during the experience of projective illusion. At several points Allen suggests that projective illusion allows viewers to imagine that they *are* seeing the film's events directly (e.g., 100, 110). If this implies that viewers entertain such thoughts as, for example, '*I* am witnessing the Battle of Waterloo', then, as Gregory Currie has argued, we face a slew of problems trying to explain why this thought doesn't trigger a series of related thoughts like 'What am I doing on a battlefield in the 19th century?', 'How did I just manage to hear what Napoleon was thinking?', 'Why wasn't I hurt by that explosion?' and 'How am I able to change positions so quickly?' The process of disavowal seems plausible enough when we are said to be ignoring the belief that what we are seeing is an image, but how about when we are also said to be disavowing the beliefs that we cannot travel in time, that we cannot travel across continents in one twenty-fourth of a second, that we cannot hear the thoughts of others, etc.
It seems more accurate to say that as we watch movies we sometimes simply stop thinking about ourselves and our position in the world (whether actual or fictional). Does this suggest that we must disavow the belief that we exist or, say, the belief that we are not Indiana Jones, etc.? Or, does it tell us something about the role that the notion of self plays in perception and in our conscious thought processes? How does my sense of self figure in my thoughts as I sit people-watching in a cafe, or as I read a philosophy article? More generally, what role does one's identity play in the seemingly conscious activity of gathering information from and interpreting the world around us? Currie, I believe, would say that the notion of self is always there, because our perceptual mechanisms are designed specifically to allow us to position ourselves within the world. Perhaps, inspired by Allen's work on film, we can argue that our sense of self can play different roles as we fantasize, engage in play, or simply watch a movie.
Of course, there is still much work to be done. Allen does a good job of criticizing the connection that contemporary film theorists describe between the individual's sense of self and her film experience. Yet, I do not think that he says enough about the role of self-identity in his own account. In the end, it is still unclear to what extent the illusion of cinema involves the 'I' of the beholder. Still, I recommend the book to anyone interested in these questions and look forward to any comments and suggestions from other theorists working on these issues.
McGill University, Montreal, Canada
1. Richard Allen mentions this incident (85) and refers the reader, in particular, to Tom Gunning's 'An Aesthetic of Astonishment: Early Film and the (In)credulous Spectator', _Art & Text_ 34 (1989): 31-45.
2. Gregory Currie's _Image and Mind: Film, Philosophy, and Cognitive Science_ (Cambridge: Cambridge University Press, 1995) presents one such denial.
3. Allen does, however, cite the work of a number of theorists who present more detailed arguments against particular aspects of contemporary film theory.
4. Some readers may challenge the accuracy of Allen's summaries. Unfortunately, I am not familiar enough with contemporary film theory to challenge Allen's interpretations. I would be interested to hear from others who are more qualified to do so.
5. Note that Allen is not committed to the idea that the term 'illusion' is completely appropriate as a description of the cinematic experience, however he believes we will better understand that experience by comparing it to experiences that we *would* classify as illusory (82).
6. This, for example, seems to be Currie's view (Currie, 1995).
7. Here I have in mind a view based, perhaps, on the work of John McDowell. See _Mind and World_ (Cambridge, Mass.: Harvard University Press, 1994).
8. Allen subscribes to Kendall Walton's views on the transparency of the image. These views are surely at work here, although Allen states that transparency does not guarantee the experience of the illusion.
Karen Bardsley, 'The 'I' of the Beholder', _Film-Philosophy_, vol. 2 no. 7, March 1998 <http://www.film-philosophy.com/vol2-1998/n7bardsley>.
Copyright © _Film-Philosophy_ 1998
Save as Plain Text Document...Print...Read...Recycle
Join the Film-Philosophy salon,
and receive the journal articles via email as they are published. here
Film-Philosophy (ISSN 1466-4615)
PO Box 26161, London SW8 4WD, England
Back to the Film-Philosophy homepage