0. Introduction

The presence of graphic or screen violence is often said to have become an everyday phenomenon. (1) I would like to thank, for their comments and criticisms, the receptive audience at the Second NECS conference held in Budapest in July-2008 where an earlier version of the present paper was read. At a couple of times, like in the next footnote, I included my elaboration of the points addressed during discussion. Although some contend that violence has always been there, it would be hard to deny that the extent and intensity to which violent scenes are explicitly shown is significantly greater than it used to be some decades ago. Of course, the reasons for this may be various, they range from slackening morals to the world wide use of visual media. (2) An analoguos example is pornography, which has always been there in the visual media, from ancient times to the rise of film, but the wide availability and the explicitness of its visual representation certainly does not compare with earlier forms. However, the reasons why it is so are exterior to the argument to be proposed in this paper which addresses the theoretical problem of the visibility of violence. But even if it could be argued that, for example, the visual representation of Hell with all its menacing forms of torture and suffering in medieval churches were publicly shown, here the emphasis is put on the conditions of the moving image representation of violence, while painting and photography constitute only a kind of contrast to moving images. There are two forms in which violence is represented visually. On the one hand violence figures in newsreels, reports and documentaries as a common constituent of the world out there. Screen violence is then nothing but a testimony that violence is woven into our everyday life from arms abuse to war scenes, from child abuse to sexual harassment, from street violence to the maffia and social corruption. But violence predominates in another form, in fiction films (and not only but especially in porn and horror movies) and even in animations.

Theoretical explanations for the over-presence of violence in both forms abound. Two general answers surface. As for documentaries and its likes the explanation goes that the media over-represents violence because its basic aim is to capture the attention of the public and - most probably for evolutionary reasons - there are two main channels of attraction, sex and violence, or their combination. What else could grab the viewers' attention more with the increasing rapidity of moving images than violent acts dominating the image, if only for a moment? It only adds further fuel to the argument to say that what grabs the attention most that is (sex and violence), can also be used, and abused, for advertising and political purposes, in one word, for manipulation.

As for violence in fiction, the explanation turns psychological and states that the presence of graphic violence is basically due to the fact that for viewers the pretense of violence results in a heightened level of tension of the alarm system, that is, it keeps it fit for the future. Horror, for one, turns out to be a way of permanent training as well as a psychological discharge of extra energy. Thus, people are driven unconsciously to watch violent scenes, and even more so they are portrayed realistically. Medical surveys confirm the finding that even children are more emotionally responsive to violence if realistically portrayed. Moreover, they are prompted to imitate realistically represented violence as they find violent characters more attractive to identify with. (3) Some relevant surveys are Wilson et alii (1990), Hargrave (2003), Kader (2006) The reason for this may be the fact that the blocking of visceral or sensorimotor reaction is much weaker or undeveloped in children than in adults.

The two explanations for real and fictional representations of screen violence converge in that the realistic rendition of graphic violence leads to a heightened effect in both juvenile and adult viewers: if a violent act is perceived as something that could occur to me instead of being fantastical, it tends to be more scary and evokes a stronger visceral reaction. The emphasis is on realism of representation. It is reported that violence in animation movies may have a negative effect on social behavior in children, which, at the same time, they do not find 'scary'. (Children already seem to be aware of the difference between fact and fiction. (4) See especially Hargrave (2003). But what makes a visual representation realistic? Certainly it has nothing to do with the so-called conventional cues of realism, like the B/W or otherwise degraded or faded quality of the pictures, the use of amateurs, or handy cam, etc. According to the surveys cited there are two criteria: first, pictures have to be distinctively real, that is, non-animation, and second, the portrayed event or situation should be possible and probable in one's own physically lived world, that is, it could happen to anyone.

The present paper is concerned with the first, and not so much with the second criterion. Or rather it takes the second as a more detailed version of the first to the extent that what is perceived as diegetically possible and probable in digital visual media depends on what is called perceptual realism. A trip to the Moon or a flight by Batman may be found unreal or surreal if it is taken to be perceptually unreal. Or, to use the second criterion, it may be found diegetically compelling and realistic just because it is perceived as ecologically realistic. "Ecologically realistic" means - as it is stated in ecological film theory - that viewers see through the moving images the represented scene because (i) they use essentially the same visual apparatus and mechanism which they use for perceiving the existing world; (ii) they use it in the same way; and (iii) with the same assumptions. The key term in an ecologically biased explanation is certainly motion and the perception of motion. (5) Since there are two major visual pathways of which one, the dorsal, carries the fast motion signal, while the other, the ventral, is much slower in processing detailed information about color, form, etc. an ecologically biased explanation would say that perceiving realistic (i.e. continuous, not ragged) motion of blobs (i.e. formless patches) is enough for our system to produce quick emotional and sensorimotor reaction. Motion is, then, perceptually realistic irrespective of the related figure being a Darthwader or a Batman.

The argument to be proposed stems from the hypothesis that the perception of realistic motion is to a great extent responsible for the reality of screen violence. As a test of the hypothesis, a comparison is being made below between the reality effects of the photograph and the moving image. Do we regard the visually moving representation of violence as more real, and if yes, in what sense, vis-à-vis its still version? (6) Here a major qualification should be made. Certainly, fantastic and animated creatures like dinosaurs, giant insects, etc. would not have the same frightening effect that realistically looking beings evoke even though they (the fantastic creatures) are perceived as ecologically realistic when moving. This may be so because they look unlike as any being existing in our real world. That is, their morphology may add considerably to their frightening potential, the ability to move may not be enough. But the argument could be turned around. Such creatures could appear to be very frightening, as it well exemplified by Dartwader, just because they are seen moving realistically (cf. the running and flying of dinosaurs). The still photographic images may well be found lacking in such a power. It should also be added that other non visual, that is, aural elements, like voice or sound, play a considerable role in rendering such imagined creatures frightening.

To answer this question we have to accomplish two things. First, we have to revisit the debate - which is contemporary with film itself - about the aesthetic differences of the photographic image vis-à-vis its moving counterpart. This step is necessary because, intuitively speaking, anything can be shown or represented visually at least from late Romanticism. With the appearance of film, however, the need for a review of the expressive potential of the arts along Lessingian lines was a hot issue among film theorists like Rudolf Arnheim or André Bazin. Second, we have to investigate how violence and death as a special case of it what exactly? are or can be represented visually in a general sense. The accomplishment of the two tasks will undoubtedly result in a metaphysics of screen (and photographic) violence and provide a set of criteria as for when and how and why (extreme) violence can - or cannot - be visually represented. Why does the project become conditional? Yet it will not say anything about how such images affect us, the viewers. Hence it has no appeal for those who, including the present author, think that an explanation of screen violence should contain a cognitive element about the viewers' response to images of extreme violence. This means that the author is not interested in the project - is this likely? Is there anything special in the way we process, and react to, such images? What's more, does our special reaction to such images change in any way their metaphysical aspect? (7) The definition of the sublime within the history and the philosophy of art may be cited here as an analogy to the problem of screen violence. The historical shift from the Kantian quest for the proper object of the sublime to the investigation of the special effect such an object can exert on us leads to the postmodern conception of the sublime effect, although it was Edmund Burk who introduced the psychological element in the definition of the sublime when he foregrounded the awe a sublime scene or object is to evoke in the viewer. That the comparison between screen violence and the sublime is not at all out of place is evidenced by the fact that the object of horror in film is often treated as an example of the sublime. (Cf. Freeland 1999)

To give an example, when Pier Paolo Pasolini, who is [also] known for the saying that editing is a metaphor of death, filmed a real execution in Africa, did he only documented visually the fact of the passing of time (of a man) - which is a metaphysical possibility syntax unclear? Or did he make the passing of time into an experience to be (re-)lived by the viewers? In other words, did he embalm time, as Bazin would say about the photographic image, or did he 'unfold' time by expressing experience by experience as the phenomenologist Vivian Sobchack would put it? Or did he also evoke thereby the consciousness of time as passing, that is, unidirectional and singular, in its viewers? While the first question points toward a metaphysical direction, the second and third appear to be phenomenological and cognitive in their phrasing. There is no place here to discuss each question, let alone their possible interrelationship. I can only draw a sketch of arguments which point to a negative answer to the first and third question and a positive one to the second. Whether Pasolini was right in his portrayal of death aesthetically and morally is one thing, and whether he succeeded in visually representing death is another. (8) Note here that if he failed in his effort and did not represent death visually but he did something else like creating a pretense or make-belief of what African people do, the aesthetic and moral aspects of his work would be changed as well.

To anticipate the analysis below, I would like to highlight the crucial hypothesis that underlies the argumentation. If we accept that the reality effect of visual violence is closely connected with perceptual realism, and that it is ecologically real to see the continuity of motion even if there is none and we know it, the problem of representing violence, in particular death and in general the passing of time, can be specified by the following conceptual pair of opposites: entropy vis-a-vis the idea of continuity of existence. (9) This is a key point in the argumentation, namely, that watching moving images is accompanied by the latent awareness of the fact that it is a movie, i.e. a sequence of still images. The by now classical distinction between scene and surface is meant to capture that fact of double - a primary and a subsidiary - awareness by which viewers are engaged in projective seeing and cannot switch to seeing the surface unless they stop seeing the scene, that is, the continuity of represented motion. (See for primary and subsidiary awareness Polányi 1958, for scene and surface Wollheim 1987, Walton 1984, and for projective seeing Allen 1995.) It seems that when infants are immersed in watching a puppet show, they disregard the fact that the puppets are moved by hands. It is simply not in their focus of attention. To see the destructive or catastrophic power of violence and death the (infant) viewers must somehow realize that the figure is no longer moved by the respective hand. In other words, he or she must combine both kinds of awareness. That it is difficult for infants is claimed in developmental psychology. That it is generally troublesome for adults is entailed by the idea of accessibility of mainstream movies in ecological film theory for it would mar the entertainment. While entropy is used to characterize the distribution of energy in a given system, that is, its level of order or organization, the continuity of existence is used in ecological physics and ecological film theory and it states that things do not cease to exist when they disappear from our visual field or become overlapped by another object. Psychologists argue, for example, that the appeal of animation films for children (and maybe also for adults) is that they refuse to accept that things can suddenly go out of existence, which is also partly the reason why children like puppet shows so much. Children have to learn that there is entropy and things can suddenly cease to be (cf. Anderson 1996). (10) Although there are two assumptions at issue here which children have to come to know: that things cannot go out of existence for one, and that had they gone out of existence they cannot be reconstituted for another. That there are two different assumption involved in ‘infantile' film viewing, however, is irrelevant to the argument proposed here. Yet the presumption here is that the belief in the continuity of existence may turn out to be a major factor in shaping adult viewers' response to moving images. It is the idea of the continuity of existence that should be partly overcome and overwritten in consciousness in learning that entropy, hence the passing of time, is a reality. Now my hypothesis is in fact the reversal of the previous statement. I state that the processing of moving images is a way to fight the idea of entropy within consciousness, or at least, to play with it. To put it more crudely:

Film viewing triggers, or is triggered by, an underdeveloped - that is, 'infantile' - state of mind; better still, moving images exemplify an infant's mind still unaware of the fact that things once out of existence cannot be reconstituted.

1. A Preliminary Clarification

To make the full argument here would be a very cumbersome task, for it requires the clarification of concepts relevant for the present topic. What I can do instead is to offer a somewhat sketchy definition of these concepts, being aware that each would trigger in itself series of debates which are, however, external to the specific argument proposed here.

1.1 The ontology: the singular vs. the general

Violence and death are prototypically singular events in real life. Hence the representation of violence and death would challenge an important ontological premise. Namely that it is unique events or singularities that are the building blocks of the real world. As such they contrast with the general subject of representation, for singular events resist representation. They can be lived but not represented as a type of action or event. They are not types: they fall under types. It is common to identify spatio-temporal singularities as objects for the senses, briefly as sensorial. (Although many consider inputs to the sense organs as already conceptual.) In Tarnay & Pólya (2005) we called it organic specificity in contrast to the environmental or physical specificity of the world.

1.2 The aesthetic: the documentary vs. the fictional

The singular/general distinction may be projected onto the documentary/fictional divide. Documentaries are often said to represent profilmic material (Carroll's category of physical portrayal), while in fictional films general and narrative elements are dominant (Carroll's depiction and nominal portrayal). In other words, the singularities of the filmed material contrast with the general and typical character of a tellable story or history, something that, for example, Aristotle calls mythos imitating a coherent and structured praxis. (11) Note that Aristotle, in his Poetics, opposes the work of historiographer to the work of the poet in the sense that the first documents the particularities of history while the second grasps its essential characteristics. However, the documentary/fictional distinction seems to depend also on the cognitive attitude of the viewers, that is, the way they take a given film: as real or fictional. However, there is no general agreement among film theorists as to what determines viewers' attitude. Some argue that it is an "evidential or consequential nature" of the film image which elicits an "indignation" in the viewers to intervene (Eitzen 2005). But what is it exactly in the image that prompts the viewers to intervene? Others maintain that in order to take a representation fictional it needs to be framed off from reality. Framing here means that the viewers' attention is guided by some compositionality principle, which may well carry aesthetic value (cf. Maquet 1986). If the image is lacking in such aesthetic value, the viewers' attention goes out to the represented, or rather, documented scene itself, and consequently, if it is violent, it prompts automatic emotional and/or motoric reaction. No doubt this is a negative definition of the realism of the image, but it may be coupled with the positive idea of realism in ecological film theory. As we have seen above, ecological realism means roughly that we see through the image and we see the scene just as we see events and scenes in the real world. Advocators of the theory usually appeal to the principle "less is more" in ecological optics. (12) Certainly, ecological film theory does allow that viewers also see the surface, that is, screen, though they are said to see it secondarily. But since seeing the surface does not alter in any way how alter at all in the way that? the scene is seen through the image surface, the close comparison between seeing the scene in the moving image and seeing it in the real world seems to stand. Accordingly, a scene (of violence) in the moving image is perceived by the viewers as a similar scene witnessed in reality. Let us call this the reality principle (RP).

1.3 The Reality Principle

The Reality Principle (RP) certainly overrides the principle of compositionality of the image often appealed to by aestheticians like Maquet (1986), for compositonality belongs to the image surface and not to the scene seen through it. The RP, however, explains why and how viewers get involved and even implicated in the film narrative (i.e. the represented scenes). It also seems to equate Carroll's physical and nominal levels of photographic representation making a way for "documentarizing" the moving image. This special effect of the moving image is confirmed by findings in developmental psychology. Briefly, it can be summarized in that the more realistic the portrayal of violence looks like, the more emotionally children react to it, and it becomes more likely that they try to imitate it. Thus, as social psychologists say, violence in animation for example does not produce strong emotional reaction and a desire to imitate in small children. According to ecological film theory it is motion and the perception of motion that is responsible for the sense of realism of the images. But clearly this is not enough for someone concerned with distinguishing documentaries from fiction as seen in the images. Both rigger motion perception and seeing the scene. Now, if we restate the original question "what cues realism and the viewer's disposition to intervene in the film?" it seems what should be examined more closely is the nature of the filmic object and the possible role it may play in the definition of the realism of the film image.

2. Realism and the Nature of the Filmic Object

Since the conception of film theory there is an ongoing debate about the nature of the filmic object. Apparently there are two models. In one, the filmic object is directly related to physical reality, for it bears its sign or trace (wears it on its sleeve): it is said to be a picture of something but it does not have to be recognized/recognizable as such: even if the picture is dim, non figural, etc. it is (a form of) the optic array that was obtained right at the moment when and where the camera was turned on (the shutters opened and closed). Be it ontological or instrumental, in this view the filmic object is the record of passing time in as much as it conserves the variation/ modulation of the optic array, i.e. the movement of light. This means that film images are not only representations of movement but they are "images that move", as Danto (1979) puts it. In short, they exist in time.

In the other model the filmic object is a representation (of motion or movement) as it is processed and interpreted by the human brain or cognitive system. It is essentially the model proposed in ecological film theory. Since in perceiving the (representation of) moving images the system works essentially analoguously to the way it perceives the real world, the filmic object appears to be an illusion or more precisely a surrogate (of reality) (Cf. Anderson 1996). In this respect it does not matter if the modulation of the optic array is the result of a simulation or partial simulation process in which the images are generated, produced, edited, etc. that is it bears no connection or trace of the reality it represents. Neither does it matter how much it resembles actual reality, that is how much the degree of visual resemblance (see early cinema technique) influence the grip of the reality effect the film can exert on its viewers. In this model the filmic object as a representation is temporal not because it exists in time but rather because its perception, like musical perception, is a temporal process. It represents a temporal world or "reality" on the analogy with the temporality of its perception. When compared with pictorial or sculptural representation, it can be seen that this analogy between representation and its perception is specific of the filmic object: although the perception of a painting is a temporal process, it does not follow that the painting represents time based on the analogy with the time of its perception. If it does, it is by means of conventions or pictorial tools like compositionality (see e.g. the interpretations of paintings like Giotto's Arena frescoes in Imdahl 1980 or Pousssin's Manna Harvest in Imdahl 1987), its materiality or seriality (see e.g. the Villa dei Misteri frescoes near Pompei) etc.

The two models are by no means exclusive. As Danto observed, the filmic object both exemplifies and represents motion or movement. But in line with the present investigation the crucial question is how the temporality of the moving image relates to the singular and unique nature of the filmic object.

3.1 VISUAL REPRESENTATION: the Nature of Photographic Representation

3.1.1 Two Basically Different Concepts of Visual Representation

Let us start with the following distinction within visual representation elaborated in Alpers (1983) but as applied to the moving image in Lastra (1995): a photograph or still image is either a momentary image of a pre-existent world passing out of the frames, or it is a world cut at and framed within the frames. In the first case the viewer is situated within the world as a casual observer who makes herself insignificant as far as the representing image and its representation is concerned. "In this world the individual captures images as they happen to him or her without intervening" (Lastra 1995: 274). In the second case the key element is to make the fleeting and fragmentary character of the photographic image subservient to a hierarchical order of the narrative diegesis as a whole. That is, the author's aim is to linearize the shots of fleeting vision in an order where subsequent shots appear to link up with the previous ones in a representation of an entire and coherent event as a whole. However, without the immediacy of the single images the narrative sequence would be far from realistic. A very important result of the 'immediate' images of unfolding events en passant is that they drag the camera and/or onlooker into the scene itself as witnesses. In other words, they produce subjective vision.

3.1.2 The Photo/ Painting Debate: the Incidental Quality as Aesthetic

Now let us review the relationship between three different media: the painting, the photograph and the moving image. Is their difference a question of degree (cf. Walton against Bazin) or of kind (cf. Cavell's idea: a photo shows a world, whereas a painting is a world)? The question was originally phrased as a search for what makes a photo aesthetically valuable. At its birth photographing was heralded precisely for its accidental nature, that is, the secondariness of the human hand vis-à-vis the mechanical structure of the camera. The incidental and fragmentary character of the image and its textural density pointed to a lack of centre of significance or signifying compositionality. There were two consequences: (i) first, the image or picture was thought to imitate vision itself with its incidental and semantically rich character, and (ii) second, emphasis in visual art shifted from compositionality to the way of selecting the viewpoint to capture the ‘picturesque' which is the fleeting image of the world. It was in this sense that photographers called themselves ‘image hunters' (Cf. Lastra 1995).

However, it needs no arguing today that an image, photographic or painted, always bears the mark of the human hand. In other words, it is essentially intentional, not simply incidental, be it in its compositional (as Maquet wants it) or sequential (as Lastra sees it) character. But again the two qualities, the incidental and intentional character of the visual image do not necessarily exclude each other. Moreover, when Lastra (1995) argues that in the latter case the hiatus between single images is sacrificed or ‘filled in' for the sake and in favour of a ‘higher-order' diegetic/ narrative unity, he appeals to the incidental character of the photo as a kind of authority which is now assigned to a character within the represented world. This is how Lastra thinks the classical idea of point of view (POV) is introduced into film making: the incidental quality of the photo as still image indicated that the seen event existed independently of the viewer. In classical narrative film, i.e., in narratively edited moving images, the POV shot indicates that the event was being witnessed by someone, and consequently existed independently of both the camera and the witness (i.e. character). But the shift is there: while the incidental quality of the photo as still image bears the trace of the external world existing independently, the subjective shots of a given event are being processed or constructed by the human mind as the trace of an external world. Note that the only mechanical difference between photo and film is the latter's sequential - that is, moving - character, while with films a new compositional quality appears, the intentionality of POV. It is more than probable that it is the combination of the two qualities that underly the working of the Reality Principle so much emphasized in ecological film theory. It is not only that images move but that they provide an analogy of subjective vision, or in Anderson's term, a proxy of the world out there. If people engage in film viewing as they do in watching the real world, it must be the effect of the combination of the two qualities. But this does not mean that they see films as documentaries, and rightly so, because what they in fact watch is subjectively told narratives, and we know fairly well from experience that narratives are not perceived but rather told. Narratives are re-tellable units, whereas documentaries are singular events, at least this is how they are conceptualized here. So, although we cannot exclude at this point that moving images can represent such events, we know at least the following: the singular quality of the world cannot be represented or reproduced by producing an analogy of normal vision to tell a narrative for narratives are types and moving images appear to be second order vision in that they mimick the way one perceives (sees and hears), but it does not constitute an act of vision and hearing in itself. The difference is like the difference between telling how to cure a disease and actually curing someone. However, the question still lingers if the difference between singularity and generality, documentary and fiction is not a difference that lies in the spatio-temporal structure of the image, still and moving.

3.2 A Short Historical Survey of the Photographical Representation of Time

Once again my survey here can only be very schematic. What we would need for the proposed argument is a threefold difference associated with four different theorists. Taking them one by one choronologically, Bazin proposed that film is embalmed time, while Barthes contrasted the presence of the filmic object with the past of the photograph. Barthes went on to contrast photogram to film, but not as two exclusive mediums, but rather as what are "weaved together". Closed, that is repeatable, filmic presence is opposed to the unrepeatable past of the photograph. It was a practicing theorist, Pier Paolo Pasolini, who added a further twist by saying that editing brings back unrepeatability to the moving image, for the cut or the cutting of the filmic object is analoguous to the act of passing out of time, that is, death. With this he accomplished a change of perspective from active to passive role. Cutting resembles creating a ‘section', a photograph of motion. It invests the film with a kind of second order stillness akin to that of the photograph. And it is important that it is second order - clearly, the cut pertains to the authorial hand, it is through and through intentional and individual. It is singular precisely in this respect and not in the sense that it represents, for the cut does not represent in itself anything; rather it produces a unique object which then becomes what it is: a closed and repeatable representation of time as at most an embalmed present.

Finally it was Gilles Deleuze who completed the shift from the general representation of time to a concept of time as singularity. He thought that film can become the direct representation of time when the sensorimotor (action/reaction) scheme breaks down in them. There are two ways for it to happen. Either within the image or in its sequential relations with other images. In the first case the image becomes spatially "empty" or "closed" to spaces beyond the frame itself. In the second case, it opens up an almost infinite number of relations to other images with which it can also be intercut (irrational gap). And the more closed or two-dimensional an image becomes, the more it opens itself to the forth dimension of time, or Thought or Spirit. It is clear from the analysis he gives of films that he conceives of time as a direct representation of space. That is, for him time is still spatial if only in the sense of Foucault's heterotopy, a kind of irregular, incoherent ‘any-where' or ‘non-place'. His idea of time is a physical-mathematical construct which results from the singular and incidental relationships of the actual and virtual spaces he calls regions. However, it must be clear that Death as a singular act of passing out of time cannot be represented in such a conceptual system, simply because image relations are virtual; the vectorial directions among them are by definition neither privileged, nor unidirectional. The singular event happens, it is singular in as much as it is not to be compared, but for the same reason it is not a representation of passing out of time, but rather a passing among equivalent regions of time. (Cf. Deleuze's use of the time pyramid.) Paradoxically, for him death as a vectorial movement in time would remain a proper topic for films constructed on the basis of the sensorimotor scheme, essentially narrative and indirectly represented as concept or type. Deleuze's conception does not fall very far from Pasolini's in that he bases his thought on the cut that is unique in making possible an infinite variety of spatio-temporal connections among the images. It is also second order in that it is not the image itself that represents time, nor a specific sequence of images; what constitutes time in this case is rather a sequencing that keeps modulating. And its working is akin to the working of the mind (herein lies his Bergsonian heritage) verb missing? Syntax unclear not a specific act of perception but a second order perception of the variability of matter.

Concluding, neither Bazin's repeatable presence, nor Pasolini's act of cutting, nor Deleuze's nominalization of the cut can be identified as a proper form of representing time, or what comes down to the same within the present argument, the entropy of death. At most, we could say that viewers watch and take pleasure in the supposed entropic representations in film like violent scenes or death as intrinsically non-temporal, i.e. repeatable or re-tellable. When confronted with an extreme scene of violence like the execution in Pasolini's documentary, viewers may take refuge in the principle of continuity of existence, not in the sense of falsely believing that the executed man can be revived as they believe watching a 'real' scene and they know in the external world there is entropy, but in the sense that what they see is an enaction of a world which is very much like the world they believe from infancy to be without entropy.

4 Uniqueness and Singularity Revisited

Let us return now to the original question that prompted the investigation of the present paper: How can a unique and singular event be meant, signified or represented? We have seen that it cannot be represented by a concept or type like a narrative, it can at most be named or referred to in and by that. We have also seen that it cannot be represented by higher order acts or events of cognition, for they operate at a level much removed from the world where entropy obtains: second order thought is a form of abstraction which long surpassed the immediacy of a direct representation of singular events like the passing of time would require. There is another case of signifying to be considered here, however. Perhaps unique and singular event or form could still stand for another unique and singular event or form by analogy of substance. What Bazin says about the ontology of the photographic image is close to what ecological optics say about the causal similarity of the optic array captured in the image to the optic array as perceived by the camera lenses at the moment of shooting. It is in the same vein that Roland Barthes talks about the punctum by which he means that the photo renders the singular event 'timeless' or heterochronic (cf. Barthes 1981 cited in Pethő 2003: 44). The singularity of the photographic object is preserved and signified through the singularity of the photographic image, although Barthes does not refer explicitly to its singular substance like the optic array. The idea of the singular as punctum, however, re-appears in his musings on the filmic image in the form of third or obtuse meaning (see. Barthes 1970). By this term he refers to a visual component of the filmed scene which resists categorical interpretation. For the latter he introduces the forms of meaning as symbolic and denotative-narrative interpretation. Obtuse meaning in turn is based on a unique and singular trait of the photographic object like Ivan's beard or the old lady's hairstyle, which at most can only become an excess in signification, that is a signifiance. (13) It is worth noting that the term ‘signifiance' recurs with an ethical connotation in many texts of the New French Thought including philosophers from Lévinas and Derrida to Barthes and Lyotard. Generally, ethical refers to the singular and unique aspect of meaning and it is opposed to signification or significance the content of which is clearly conceptual and categorical. This unique quality of the photograph as causal or incidental content is lost - at least this was my argument here - as soon as movement or sequentiality is introduced within the content of meaning or representation. One could argue at this point that sequentiality is in fact an analogical form of chronological time, so it is akin to the analogy of substance between the photographic object and image. But at closer scrutiny it turns out not to be the case, for while chronological time is irreversible and is a form of entropy in physics, the sequence of moving images can be replayed at any time. That is, it could mimick temporal processes like the workings of the mind, but it remains to be an analogy of form and never becomes an analogy within the substances of moving images and entropic processes in the world.

To sum up: generally, the uniqueness of the ‘real' scene, or the pro-filmic matter is signified by its analogy with the uniqueness of the photo itself. Hence it is a token-reflexive relation. It is a relation also known from anthropology of religion and ritual analysis that establishes connections, for example, between different performances or tokens of a particular rite or even of the Christian mass. Scheffler (1981) calls it mention-selection when a specific performance of a ritual scenario relates to previous performances of the 'same' scenario. We can also think of the different performances of a given musical score.

Our problem here is still more specific: can (unique and singular) time as passing be represented? Violence and death is a singular event within time, an act of moving out of time. Is there any analogical event - for there cannot be any analogical material substance - it could be represented with? Is there any token-sensitive event or act? Maybe Pasolini's idea of the cut could be such an act. But surely, the film as product, as something to be seen, with all its cuts has already become a type. Maybe it is the author's unique ‘handwriting' that could serve as an analogical event. So the question still lingers: Can film represent time as a unique and singular event? Or to put it methodologically: How should we address the question whether the effect, the passing of time can be the proper object of either photographic or filmic representation?

5. The Phenomenology of the Image: to Express Experience with Experience

Vivian Sobchack came very close to this conception at the beginning of her book The address of the eye when she claimed that film is an experience expressed by an experience. According to her semiotic phenomenology is tantamount to perceiving the expression of someone else's perception. Each act of viewing is the viewer's singular and unique personal experience of someone else's singular and unique experience. Note that Sobchack does not speak about representing in this respect.

Even summing up the theoretical background of Sobchack approach to film viewing is beyond the scopes of the present paper. Interestingly and notably, hers is not the phenomenology predicated by Metz in that vision is equal to its very meaning so that the signifier and the signified have the same domain (probably they share substance, they participate in each other), although a Metzian conception would look to be much closer to the idea of the photo expressing singularity of meaning through an analogy of substance re-phrase and break up sentence. Contrary to what one would expect, Sobchack's theoretical framework is communicational or semiotical phenomenology as laid down by Richard Lanigan: vision as film partakes of the same communication the viewer is engaged in. Thus, it is ongoing experience and it does not represent as, say, a photograph does. In one word, the phenomenology of film is nothing but to express experience with experience or rather motion by motion. In the terms of the previous section, film viewing for Sobchack is a unique and singular acting out of the human communicational situation in which the confronting partners are the film itself and the viewer. And both bring their subjective bodily experiences into the pragmatics of the situation. That the film has a body is not simply a terminological extravaganza: it points to the very essence of viewing. Namely that each instance of viewing is unique and unrepeatable and as such it engages the entire subjectivity of the viewer as a singularity and not simply as a member of a class. It becomes part of her innermost personal history. Intuitively, it exemplifies the common saying that we as viewers of a film never remain the same person we were before the viewing. Understandably, as long as the filmic experience is considered to be intrinsically pragmatic and deeply personal, it cannot be re-told and repeated. I would even venture to say that Sobchack's film phenomenology comes very close to Ricoeur's existential reading of the Bible. When Ricoeur emphasizes that an interpretation of the parables is in the end basically existential he alludes to the making of the reader's personal history: no-one remains the same person after having read the Good Samaritan.

Now short-circuiting my interpretation of Sobchack's phenomenology I would say that film does not represent but maybe express time as a form of entropy but it expresses it for the personal, hence unique and singular, viewer. It expresses it through the changes it brings about hic et nunc, i.e. pragmatically, within his or her personal history. To demonstrate this conclusion in terms of this paper I would need to show that there is some analogy of substance between the film's body (in Sobchack's term) and the existential body of the viewer. Although there are many points and allusions in Sobchack concerning such an analogy, she never goes as far as claiming that moving images, despite their inherent sequentiality, can accomplish something akin to the punctum or the third meaning attributed to the photograph. Such a demonstration must be the topic of another and much longer paper. As a conclusion of this paper I repeat the basic tenets of my ramblings for future discussion.

6. Conclusion

My basic tenets in this paper for which I tried to argue were:

• violence or death is not a proper object of representation
• realism inheres in the mind and the senses, rather than in the image (cultural codes follow, rather than precede, the recognition of 'reality' in the image)
• the reality effect in the image is that it drags the viewer into the recognized, rather than ‘represented' world
• singular events can at most be expressed, rather than represented

And I would like to add a final one for which I did not argue but which could be the conclusion of a sequel to this paper:

• expressing experience, motion or time - in the context of pragmatic embeddedness - cannot step outside experience, motion or time if not by naming or referring to experience, motion or time

That is, although an existential and semiotic phenomenology of film viewing could demonstrate that film viewing is an expression of experience by experience for and also in terms of the viewer, the experience of film viewing can never be a first order or immediate experience of another individual, say, the author of the film: experiencing his or her experience will necessarily become second order for the viewer. And it is a truism of the hierarchy of levels that analogical relations cannot obtain through different levels. At a higher level the objects of the lower level can only be named or referred to but not exemplified.

László Tarnay: On the Metaphysics of Screen Violence and Beyond

Apertúra. Filmelméleti és filmtörténeti szakfolyóirat , 2008. nyár.

http://apertura.hu/2008/nyar/tarnay

Lábjegyzetek

[1] I would like to thank, for their comments and criticisms, the receptive audience at the Second NECS conference held in Budapest in July-2008 where an earlier version of the present paper was read. At a couple of times, like in the next footnote, I included my elaboration of the points addressed during discussion.

[2] An analoguos example is pornography, which has always been there in the visual media, from ancient times to the rise of film, but the wide availability and the explicitness of its visual representation certainly does not compare with earlier forms. However, the reasons why it is so are exterior to the argument to be proposed in this paper which addresses the theoretical problem of the visibility of violence. But even if it could be argued that, for example, the visual representation of Hell with all its menacing forms of torture and suffering in medieval churches were publicly shown, here the emphasis is put on the conditions of the moving image representation of violence, while painting and photography constitute only a kind of contrast to moving images.

[3] Some relevant surveys are Wilson et alii (1990), Hargrave (2003), Kader (2006)

[4] See especially Hargrave (2003).

[5] Since there are two major visual pathways of which one, the dorsal, carries the fast motion signal, while the other, the ventral, is much slower in processing detailed information about color, form, etc. an ecologically biased explanation would say that perceiving realistic (i.e. continuous, not ragged) motion of blobs (i.e. formless patches) is enough for our system to produce quick emotional and sensorimotor reaction. Motion is, then, perceptually realistic irrespective of the related figure being a Dartwader or a Batman.

[6] Here a major qualification should be made. Certainly, fantastic and animated creatures like dinosaurs, giant insects, etc. would not have the same frightening effect that realistically looking beings evoke even though they (the fantastic creatures) are perceived as ecologically realistic when moving. This may be so because they look unlike as any being existing in our real world. That is, their morphology may add considerably to their frightening potential, the ability to move may not be enough. But the argument could be turned around. Such creatures could appear to be very frightening, as it well exemplified by Dartwader, just because they are seen moving realistically (cf. the running and flying of dinosaurs). The still photographic images may well be found lacking in such a power. It should also be added that other non visual, that is, aural elements, like voice or sound, play a considerable role in rendering such imagined creatures frightening.

[7] The definition of the sublime within the history and the philosophy of art may be cited here as an analogy to the problem of screen violence. The historical shift from the Kantian quest for the proper object of the sublime to the investigation of the special effect such an object can exert on us leads to the postmodern conception of the sublime effect, although it was Edmund Burk who introduced the psychological element in the definition of the sublime when he foregrounded the awe a sublime scene or object is to evoke in the viewer. That the comparison between screen violence and the sublime is not at all out of place is evidenced by the fact that the object of horror in film is often treated as an example of the sublime. (Cf. Freeland 1999)

[8] Note here that if he failed in his effort and did not represent death visually but he did something else like creating a pretense or make-belief of what African people do, the aesthetic and moral aspects of his work would be changed as well.

[9] This is a key point in the argumentation, namely, that watching moving images is accompanied by the latent awareness of the fact that it is a movie, i.e. a sequence of still images. The by now classical distinction between scene and surface is meant to capture that fact of double - a primary and a subsidiary - awareness by which viewers are engaged in projective seeing and cannot switch to seeing the surface unless they stop seeing the scene, that is, the continuity of represented motion. (See for primary and subsidiary awareness Polányi 1958, for scene and surface Wollheim 1987, Walton 1984, and for projective seeing Allen 1995.) It seems that when infants are immersed in watching a puppet show, they disregard the fact that the puppets are moved by hands. It is simply not in their focus of attention. To see the destructive or catastrophic power of violence and death the (infant) viewers must somehow realize that the figure is no longer moved by the respective hand. In other words, he or she must combine both kinds of awareness. That it is difficult for infants is claimed in developmental psychology. That it is generally troublesome for adults is entailed by the idea of accessibility of mainstream movies in ecological film theory for it would mar the entertainment.

[10] Although there are two assumptions at issue here which children have to come to know: that things cannot go out of existence for one, and that had they gone out of existence they cannot be reconstituted for another. That there are two different assumption involved in ‘infantile' film viewing, however, is irrelevant to the argument proposed here.

[11] Note that Aristotle, in his Poetics, opposes the work of historiographer to the work of the poet in the sense that the first documents the particularities of history while the second grasps its essential characteristics.

[12] Certainly, ecological film theory does allow that viewers also see the surface, that is, screen, though they are said to see it secondarily. But since seeing the surface does not alter in any way how alter at all in the way that? the scene is seen through the image surface, the close comparison between seeing the scene in the moving image and seeing it in the real world seems to stand.

[13] It is worth noting that the term ‘signifiance' recurs with an ethical connotation in many texts of the New French Thought including philosophers from Lévinas and Derrida to Barthes and Lyotard. Generally, ethical refers to the singular and unique aspect of meaning and it is opposed to signification or significance the content of which is clearly conceptual and categorical.