On the politics of perception in moving image technology

The cognitive critique of the politics of cinema: Walter Benjamin’s ‘Technical Reproducibility’ essay

Benjamin takes his cue from Karl Marx, who, we may note, with his analysis of the class character of production technology throughout history, ‘was the first to unmask the [socio-political] interests behind supposedly technical imperatives’ (Feenberg, 2002: 37). Benjamin regards his essay as a continuation of the prognostic critique that Marx began when capitalism was in its ‘infancy’, but orients himself, in a revolutionary manner, toward analysing the conditions under which capitalism could be abolished and replaced given the new material and cultural conditions developing in the 20th century. Benjamin understands the mechanical reproducibility of art in his time – the phonograph record, the photograph and, perhaps most importantly, cinema – as heralding a momentous change in society that was already having effects on human perception and cognition in the late 19th century and that he thinks has a definite bearing upon the critical consciousness of the working class in the 20th. This change is, for Benjamin, bound up with the development of mass society and the extension of capitalist rationalization. ¹

Benjamin argues that mechanical reproducibility transforms the authenticity of the artwork – there is no original artefact to a photograph in the way an original sculpture may be copied. It makes no sense, Benjamin says, to ask which ‘print’ is the original. As a result, the ‘aura’ that the artwork possessed before mechanical reproduction – that is, the experience of its uniqueness, its mystical existence as such – withers in the face of the experience of mechanical reproducibility. Benjamin argues that this is because aura is associated with the cult value of an artwork and it appears as a historically necessary aspect of its perception before the advent of mechanical reproducibility. The auratic artwork fetishistically ‘distances’ the perceiver due to the artwork’s socio-political function. However, compared with an auratic context, he argues, the technical artwork is a ‘creation with entirely new functions’ (Benjamin, 1986b: 225). The social value of the authentic work of art is reproduced in the aura brought forth through the situational contemplation of artworks in the museum or gallery (the public, formal institutional replacements of the church, in this respect) or in the interiors of aristocratic and bourgeois homes or offices. Mechanical reproducibility, on the other hand, ‘emancipates the work of art from its parasitical dependence on ritual’ and transforms authenticity itself by making it possible for art to be ‘based on another practice – politics’ (Benjamin, 1986b: 223–4). Benjamin is not saying that authenticity is destroyed by mechanical reproducibility, but rather that the destruction of auratic art transforms in a political way the social function, experience and relationships pursuant to art, artistic producers and art’s consumers. Just what this political practice of art is or could be, Benjamin does not explain in any systematic way. But his invitation to explore such possibilities has motivated many. It is not an ideologically charged ‘politicized’ art for which he calls; it is rather a task for art that is ‘far more difficult – that is, to undo the alienation of the corporeal sensorium, to restore the instinctual power of the human bodily senses for the sake of humanity’s self-preservation, and to do this, not by avoiding the new technologies, but by passing through them’ (Buck-Morss, 1992: 5). ²

The aura of the stage actor vanishes in front of the camera. Something of a critical distancing occurs since the lack of unmediated personal contact with the actor ‘permits the audience to take the position of a critic’ via the camera’s constant ‘testing’ of the actor (Benjamin, 1986b: 228). It is the unseeing eye of the camera that is substituted for the ‘naked eye’, which, with the close-up and slow motion, for example, does not simply reveal reality more precisely, but instead ‘it reveals entirely new structural formations of the subject … a different nature opens itself to the camera … The camera introduces us to unconscious optics as does psychoanalysis to unconscious impulses’ (Benjamin, 1986b: 236–7). The ‘different nature’ to be revealed is complex – it is not simply some different claim to reality or objective truth based on a model of true versus false consciousness, since ideology no longer functions simply as a mask. Instead, the ‘different nature’ is that of constitutive relationship itself, which is at a logical level different from that of the visibilities and constituents of the medium of artistic perception (that is, the representative image, plot, devices, technique, etc.). Cinema, Benjamin says, penetrates reality like a surgeon penetrates a patient’s body (the medical metaphor is intended, since it indicates the ‘critical’ nature of what is at stake), yet it does so as the ‘height of artifice’. It is precisely this dialectical tension in cinema, or, more precisely, the mechanical image itself, that Benjamin thinks creates its political potential. ³

The originality of Benjamin’s analysis lies in the links he establishes between the experience of different media and their social and political conditions of possibility. Thus Benjamin speaks of ‘the shock effect of the film’ (1986b: 238) that prevents the concentrated contemplation that one may have before a painting by ‘distracting’ the viewer through its montage construction. The shock effect is not a special effect but a redirection of consciousness achieved dialectically through the film image: it connects cinematic experience with people’s everyday social experience in capitalist society. ‘With its dialectic of continuity and discontinuity, with the rapid succession and tactile thrust of its sounds and images, film rehearses in the realm of reception what the conveyor belt imposes upon human beings in the realm of production’ (Hansen, 1987: 184). Yet, Benjamin thinks, film nevertheless allows an experience from which a new consciousness may be formed about this condition. Benjamin’s ‘mimetic machines’ produce an illumination from ‘the bodily impact of the “dialectical image”’ as a revolutionary potential derived from an engagement with ‘the embodied mind’ in the optical unconscious: ‘Body and image have to interpenetrate so that revolutionary tension becomes bodily innervation’ (Taussig, 1993: 23; Benjamin, 1986a: 192). ⁴ Films are capable of shocking the viewer into a recognition of what lies beyond the dream world of commodified life that constitutes his or her waking life precisely through his or her own dream fetishes. ‘What makes film significant is that it is capable of activating the latent dream imagery of the new collective spaces of the capitalist metropolis’ (McNally, 2001: 215). The homology between the experience of the cinematic image and the imaginary space of the modern city (and, indeed, the political relations constitutive of the city) is revealed through a dialectics of seeing. Advocating such a dialectics of seeing, Buck-Morss contends that the ‘technical apparatus of the camera, incapable of “returning our gaze,” catches the deadness of the eyes that confront the machine-eyes that “have lost their ability to look”’. The aim is no longer to educate the unsophisticated eye to see beauty, ‘but of restoring “perceptibility”’ itself (Buck-Morss, 1992: 18). I wish to emphasize the importance of the embodied cognition that Benjamin analyses and the connection between media technology and the dialectics of seeing that emerges from this analysis. But before delving further into this politics of media technology and perception, let us consider in detail a recent critique of Benjamin from a contrasting philosophical perspective that also treats cognitive and aesthetic questions together. This critique essentially contends that there is no such thing as a politics of technology and so is useful in order to set off and clarify the Benjaminian argument.

Noël Carroll’s philosophy of mass art and cognitivist critique of Benjamin’s essay

Noël Carroll (1998) has challenged Benjamin’s politics of perception in cinema in the context of putting forward a philosophy of mass art of his own that is heavily indebted to the Anglo-American tradition of analytic or ‘ordinary language’ philosophy and that also expresses a certain neo-Aristotelianism that emphasizes Aristotle’s pragmatic and classificatory epistemology. ⁵ Like Benjamin, Carroll prioritizes the cognitive moment – perception and understanding – in mass art production and consumption. He wishes to examine important arguments for and against mass art, arguments that are philosophical because they ‘all rely on some conception of the essential features of mass art’ (Carroll, 1998: 4–5). ⁶ Carroll agrees with Benjamin that mass art is something to celebrate, but he rejects Benjamin’s political theory of perception. Carroll interprets Benjamin as a historical materialist following Marx, which is decisive for each of the arguments he interprets Benjamin using to defend mass art – Carroll calls them ‘the progress argument’ and ‘the new art argument’. The first defence that Benjamin offers is that ‘mass art contributes to historical progress both as an emblem and an exercise in the sort of technical (proletarian), emancipatory vision/consciousness that will release the productive forces’. Second, since mass art is historically and qualitatively new, it requires new standards of judgement and new forms of theorization (ibid.: 127–8). Carroll rejects the first defence of mass art (‘the progress argument’) but is sympathetic to ‘the new art argument’ – at least, to the latter’s conclusion, since Carroll would weaken its premises in order to make it more plausible. The ‘new art argument’ is not a significant issue for Carroll, so we shall focus on his criticisms of Benjamin’s progress argument.

The so-called progress argument relies on several premises, the first of which is that of historical materialism and the second is that any change that contributes to human emancipation is eminently defensible (the second Carroll does not challenge). Leaving aside Carroll’s criticisms of historical materialism itself, which are of less interest for our purposes here, let us discuss his key criticisms of Benjamin’s third and fourth premises and the arguments concerning the politics of mass art. ⁷ Carroll argues that Benjamin assumes in a third premise to the progress argument that art represents and can cause changes in perception – it ‘symbolizes and encourages changes in perception’ – and indeed, as we have seen, this is one of Benjamin’s central claims. However, Carroll is sceptical, since he thinks that perception ‘is not particularly plastic’ and nor would it change as a result of consuming artworks since the ‘human perceptual system evolved over a long period of time, under the influence of natural selection’ and there is not enough time for it to change over the course of the period in question (the Industrial Revolution). Yet, Carroll says, perhaps Benjamin does not mean perception ‘literally’ changes, but only that ‘certain habits of perception change’ as a result of new media technologies – for example, people may begin to notice small, expressive gestures more under the influence of the close-up. Carroll does not think this is the case either, since Benjamin offers no evidence for such changing perceptual habits independent of the changes in art that the latter cites as corresponding to and causing perceptual changes. For Carroll, this is circular and begs the question since art cannot simultaneously correspond to perceptual changes and serve as a cause of them (we shall return to this important epistemological claim). But the presumption that ‘there are special modes of perception that facilitate the expansion of the productive forces and that symbolizing them contributes to the expansion of productive forces’ is plausible, according to Carroll, if we interpret ‘modes of perception’ as ‘habits of perception’ such that becoming habituated to the use of computers (for example, the hand–eye coordination required for video games) trains the workforce in appropriate skills required for automation. Symbolizing these perceptual habits ‘may serve to encourage people rhetorically to adopt them’ and may allow people ‘to understand them in ways that are conducive to making them one’s own’ (1998: 130–3). This argument of Carroll’s begs the question of whether or not such ‘rhetorical’ encouragement of ‘habits of perception’ does harm or not to the individuals who are so trained or to the organization of labour itself. Carroll’s argument assumes (against Benjamin’s overall approach here) that adaptation to the capitalist work process and its increasing productivity requirements is somehow automatically ‘progressive’.

It is, however, the fourth premise of Benjamin’s progress argument that Carroll finds the most problematic. The fourth premise, he says, states that ‘Mass art by its very nature symbolizes and encourages changes in perception that are suitable to the expansion of the productive forces’. Again, for Carroll, this premise faces the difficulty that it seems unlikely that any art literally changes perception, so he thinks we should interpret Benjamin as meaning that habits of perception change rather than biological perception, which is more defensible. Yet Carroll does not think that either ‘shock’ or ‘distraction’ in film viewing will inculcate a ‘critical, analytical, penetrating stance toward mass artworks and what they represent’ since there is no reason to think that the technical aspects of film will engage the audience’s attention any more than a stage play would – indeed, an audience can be more riveted to a film due to its editing and technical aspects and at least equally seduced by acting performances on the screen as the stage. Carroll agrees with Benjamin that the film spectator is active, but there is little evidence, he thinks, that viewers are any better at cognitive and perceptual tasks once they leave the cinema. Benjamin’s notion of the distracted viewer who, as a result, gains critical distance, does not make sense, Carroll argues, since such a distracted condition is more likely the result of that viewer’s preoccupation with something else or the fact that this particular film is boring. Thus, for Carroll, whether or not a viewer is distracted or shocked by film viewing is a matter of a case-by-case analysis of particular films and does not entail any critical consciousness (1998: 128, 133–6). This is, however, not what Benjamin means by the critical aspect of ‘distraction’ since, as Eiland (2003: 57, 60) argues, distraction ‘must itself be understood dialectically – that is to say, beyond the simple opposition of distraction and concentration’ – as ‘a new kind of learning’ in which ‘the metamorphic mechanism of montage [functions] … as a spur to new ways of perceiving [see also the final section below]. In either case, a certain wandering or dispersion makes itself felt.’

Yet Carroll insists that ‘montage as such is not inherently critical socially’ – it may be used this way by particular filmmakers, but there is nothing essential about the nature of mass art that promotes ‘a critical, proletarian class consciousness’. It is here that we get to the nub of Carroll’s position. According to Carroll, the fatal flaw in Benjamin’s approach is the belief that ‘a medium or technology could have a mode of consciousness or a political stance inscribed in it’. Rather, mass art’s moral or political orientation ‘depends on the ways in which it is used, not upon its nature qua mass art … Technologies, in and of themselves, are morally neutral’ (1998: 138–9). Carroll contends that humans had already evolved the perceptual capacities that are brought to the fore in modern society; these capacities are simply being ‘stimulated to a greater extent than previously … taxed by new levels of input’. Benjamin is thus ‘confused about the facts of perception’ since the constant shifting of attention that he analyses is a ‘hard-wired feature of the perceptual apparatus that has been with us for millennia’. The evolutionary basis is the need constantly to ‘look out for predators and prey, or other sources of nourishment’. Such saccadic eye movement has hence always been part of our perceptual make-up (ibid.: 130–1, n. 28, 136, n. 32). ‘Pictorial recognition’, Carroll asserts, ‘does not involve a process of learning over and above object recognition … Anyone with normal perceptual capacities can recognize the referent of a standard motion-picture image … simply by looking’ (ibid.: 192). ⁸ Elsewhere, he says we recognize what a representational picture is and what it is of ‘without any subtending processes of inferring, translating, decoding, or deciphering’ (Carroll, 2004: 103). Carroll thus contends that there is a basic, virtually unchanging nature to the human frame that, to a certain extent, determines culture or at least provides the human stuff to which art addresses itself. We have evolved ‘cognitive, perceptual, and emotive hardware that we share cross-culturally’ and this helps explain the ability of moving pictures to engage culturally diverse mass audiences around the world who possess no special training (ibid.: 99, 103). The faculty of perception has not changed in modernity, according to Carroll – people’s perceptions certainly have, but this is quite a different matter and is not, he thinks, a very interesting claim. It is scientifically possible for the faculty of perception to change, Carroll believes, given the slow drift of evolution (or given sufficient sophistication in genetic engineering), but such evolutionary mutation in perceptual capacities, he thinks, occurs glacially. Carroll hence contends that the ‘pictorial arts have developed in ways that already suit our nearly unchanging … perceptual capacities’ and in this sense the arts are caused by our perceptual nature rather than by that nature’s being causally influenced by the arts (Carroll, 2001: 13, 16). For Carroll, perception has thus very little plasticity: ‘The human sensorium is a biological mechanism.’ To suggest that culture affects perception is to confuse literal perception with conceptual perception – ‘what we literally see is a matter of biology, not culture or history’ (1998: 158–9).

Carroll has also similarly taken to task psychoanalytic and (post)structuralist interpretations of pictorial perspective, which argue that perspective implements a historical ‘subject position’ because perspective is a visual code and therefore arbitrary or conventional. For Carroll, by contrast, perspective is not a replica of vision but instead provides accuracy regarding spatial information – no other mimetic system of representation is as accurate at conveying information about the appearances of the relative dispositions of objects in space. As such, perspective is based on scientific laws of vision and is simply an adaptation to the structure of the world, including human biology (Carroll, 1988: 128–31). I would now like to offer some criticisms of Carroll’s interpretation of Benjamin that lead to more conclusive shortcomings of the former’s general philosophical approach, in order to set the stage in the final section for my recovery of Benjamin’s analysis of the politics of technology.

Carroll contends that saccadic eye movement, founded in the evolutionary development of the species, entails that Benjamin’s regard for distracted viewing simply notes an existing feature of our perceptual apparatus rather than anything specific or new to modernity or mass art. Carroll’s argument assumes that such constant movement and shifting of attention in our normal visual activity is equivalent to the kind of constant visual movement that occurs while watching films. Yet there is no reason to think that the physiological saccadic eye movement associated with normal vision and, according to Carroll, rooted in prehistoric evolutionary needs, produces the same visual experience as similar movement while watching moving images. The context of each visual activity is so different. ⁹ As I ride my bicycle around my city, I am constantly shifting my attention and vision in order to adjust to where I am going and (perhaps most importantly!) to avoid being hit by a car; as I walk this city, I am constantly shifting my attention and vision as I take in the delightful (or occasionally unpleasant) sights and sounds. When I am watching a film, I am drawn along, constantly shifting my attention and perception in order to follow and understand the filmic narrative (or, if there is no narrative, as in some avant-garde film and video, the point or ideas being conveyed). But the experience of the filmic montage – the jump cuts, close-ups, slow motion, extra-long takes, etc. – and the effect of the recorded sound as well as the social and physical space of the cinema on my overall perceptual experience is all profoundly different from my audio-visual (and tactile) experience of biking or walking around the city. Perhaps Carroll would respond to this by saying that he is talking about the distinction between physiological vision and socio-culturally conditioned seeing, which he says we should not confuse, but which Benjamin seems to be saying are interrelated, and it is this latter confusion that Carroll seeks to undermine with his observations concerning saccadic eye movement. My point, however, is to confirm that saccadic eye movement is a physiological feature of the species, but for Benjamin this has no significant bearing on the issues at hand, since his argument concerns a change in seeing as a holistic socio-aesthetic phenomenon and not the sudden arrival in modernity of distracted viewing as a physiological phenomenon. Shortly, I shall challenge Carroll’s notion of vision more acutely when I consider the cognitive science of embodied cognition that conceives of vision as indelibly entwined with the organism’s relationship to its environment and context and not somehow external or independent.

In any case, what Carroll refers to as prehistoric saccadic eye movement (carried over into contemporary settings) occurs primarily at the level of the visual signal rather than the semiotic sign (what Peirce, 1992, distinguishes with the concepts ‘indice’ and ‘symbol’). Movement on the horizon or the rustling of bushes are signals of danger or opportunity that the prehistoric human, just as much as any other animal including the hominids, was required to appreciate without symbolic language. Film, by contrast, functions primarily at the level of the symbolic and its audio-visual experience is, on the face of it, at least categorically different from that of the audio-visual signals we perceive in the aforementioned contexts. This is not to say that the perception of the signal is not also historically conditioned – it is; instead, this is to distinguish the different forms of interpretation required by signal and symbol. Carroll conflates the levels of the symbolic and the signal here while strictly distinguishing physiological from cultural perception. ¹⁰

But it is precisely this distinction in perception that is the issue, Carroll may insist, since he contends that a picture of a bearded man is seen as a picture ‘without any subtending processes of inferring, translating, decoding, or deciphering’ (Carroll, 2004: 103) and thus that film is perceived in a naturally representational way before or at the same time as it is perceived symbolically. My contention is that such a strict separation between the physiological process of vision and socio-culturally conditioned seeing is impossible to make, even though it would appeal to ‘common sense’. By asserting this, I do not wish to deny that an analytical distinction between physiological and socio-cultural processes is possible, since it is. Rather, I mean that it makes no sense to draw such a distinction and make it decisive for comprehending human perception as such. Biology and neuroscience tell us a great many things about vision, but they cannot finally tell us how we see, since seeing, as the cognitive apprehension of visual objects, always mobilizes more than our physiology. Moreover, the apprehension of the visual image appears to involve not just sight but other senses as well and so, according to Bal (2003), vision is more properly called a ‘synaesthetic’ achievement. And vision always involves apprehending an image in social, political, economic and cultural contexts, as Benjamin stressed. I will demonstrate these assertions against cognitivist philosophy of mind in the remainder of the article and seek to show how Benjamin’s analysis remains cogent.

There is still no consensus in the scientific literature concerning the relationship between the visual system and cognition for it would seem that if the entire visual-cognitive process were to count as vision, then there would be no such thing as vision proper, only cognizing. On this issue, cognitive science seems divided between (1) regarding the visual system as a separate and autonomous set of processes that are proprietary to vision and with which cognition then interacts (Pylyshyn, 2003 – this supports Carroll’s argument); and (2) regarding cognition as inherently perceptual such that vision is constructed (not ‘faithful’), active and based in the brain not the eye (Enns, 2004; Findlay and Gilchrist, 2003). My philosophical contention is that the latter approach is likely correct, and recent psychological and neuroscientific research continues to provide evidence for it. We may, as a result, require a ‘hermeneutics of vision’ (Davey, 1999) that includes visuality as a central question for human embodiment and experience. But let us examine the philosophical reasons why vision and seeing must always be considered together.

Embodied cognition and the political

On occasion, Carroll refers to ‘common sense’ and everyday usage when considering the ideas in question. He finds decisive for dismissing his opponents’ views the (im)possibility of empirically testing their claims or the claims’ failure to ‘fit’ or ‘correspond to’ the ‘facts’ of mass art (e.g. 1998: 108, 315). ¹¹ Carroll pursues a ‘classificatory’ approach – that is, he seeks to classify not to criticize mass art, and to determine its ontology or nature. He considers such clarification and ontological inquiry to be the definitive work of philosophy. Carroll’s approach treats ordinary ‘evidence’ as that which can be observed or interpreted in a way that ‘correspond[s] to’ the ‘facts’ of mass art. If it is philosophical evidence that is in question, then arguments are required that can be verified or that can accord with ‘common sense’. If there is deviation from such empirically driven requirements, Carroll is wont to dismiss such ideas as ‘magical’, as he does Benjamin’s politics of technology. The main problem here is that Carroll lacks any kind of adequate social theory or sociology that supplements or coordinates with his philosophy. Philosophical truth in these matters is not best expressed as conceptual clarification in the form of a proposition because to do this begs the question of just how the ‘common sense’ of audiences and the ‘facts’ of mass art are themselves constituted. What cognitive and social processes might lie behind the factual appearance or the apprehension of things does not concern Carroll. Yet there is no outside from which to view the systemic totality or its elements (the latter Carroll calls facts) in a neutral way, since the observer is always already part of the system that he or she observes on account of the mutual determination of meaningfulness – perception in the broadest sense – that structures and organizes the content of every observation. This is what recent cognitive science calls embodied or enacted cognition (Johnson, 2007; Noë, 2004; Thompson, 2007). When considering meaning in the human sense of ‘significant symbols’ (Mead, 1963), comprehension requires not just a background of at least some shared meaning between communicating subjects but also a meeting in symbolic exchange of distinct ‘lifeworld’ contexts. Such contexts extend from the personal to the historical and constitute the subjects’ interpretive ‘horizons’ (see Gadamer, 1975). There is an indelible relation between perception and context for any interpreter of symbols (the hermeneutic circle of interpretive social science). Pace Carroll, there is no meaning inherent in nature, completely independent of the observer – the mind–body dualism behind Carroll’s approach is not philosophically defensible. Indeed, it is precisely the result of Carroll’s practice of unproblematically referring to ‘ordinary’ meanings as the ‘reality’ against which philosophical insight should be judged – as though such ordinary meanings were completely free of the influence of the reified world of objects, instrumental rationality, or capitalist commodification – that he is able to make such claims.

This positivist classificatory method is quite inadequate to the study of culture and society because it cannot account for such dynamic and systematic mutual determinations between subjects and objects that involve internal relations of constitutive formation mediated by power and communication. Such relations of mutual causality must be theoretically reconstructed for social investigation and can almost never be directly observed or made amenable to straightforward empirical investigation in the way that natural systems can be understood scientifically or that artificial intelligence can be mapped mathematically and controlled empirically. Benjamin understood this need for philosophical reconstruction of internal relations of causality, and I would like to suggest that his theoretical approach is consistent with embodied cognitive science, while extending an ethico-political dimension lacking in the latter. An initial link between Benjamin’s dialectics and embodied cognition can be discerned in the sympathies between their respective epistemologies, which can be compared through the concept of mutual causality.

Mutual causality is a notion developed in modern systems theory that identifies the internal communicative relations of elements within a dynamic system and the relations of these elements to the system as a whole through the mediation of feedback communication (Macy, 1991; Maruyama, 1968; Laszlo, 1996). As such, it is a causal relation that may be used to understand non-living as well as living systems, depending on the mediation. The dialectical approach to social analysis developed initially by Hegel and transformed materialistically by Marx also expresses this notion of mutual causality (Hegel himself being the first great systems theorist). One cannot understand the operation of any individual element without understanding how each relates to the system as a whole as it operates in time and space, with individual elements necessarily existing at a different logical level from that of the whole. Since every system is the temporal realization of a dynamic process, the constitutive communication of feedback (internal to any system) is always changing and developing and must always situate itself within the larger context that the system has with its environment. This means that the units of communication are always produced in that relationship and do not originate with any particular element (see Maturana and Varela, 1980, 1992). Art is such a communicative phenomenon, which is why art can simultaneously correspond to changes in perception and serve as a cause of them.

In the case of symbols, meaning is always generated out of the relationship that the mind forges with its environment, but always in a ‘doubly’ mediated fashion that combines apperception and perception. The interpretation of symbols can be more or less mixed with instinctual signals and stimuli, but symbolic interpretation relies constitutively on a socially produced communicative competence in lexicon and grammar along with the ability to contextualize against a background of shared knowledge and expectations. ‘Mind’ is the term we commonly use for the experience of cognition as such, but this should not deceive us into thinking that cognitive activities can be neatly separated from the totality of the experiencing mind–body. The mind–body always exists in a relationship to environmental context or ‘lifeworld’ (in the phenomenological sense) that, for humans, consists of at least three intertwined levels, namely, the objective world of externality in which things appear, the social world of interrelations among people along with their constructed social environment, and the subjective world of one’s inner life (this tripartite concept of the lifeworld is developed by Habermas, 1987). The mediation of each of these mind–body–lifeworld relations is heavily symbolic but it is also non-linguistic – that is, aesthetic – in many decisive ways.

It is hence important to recognize that human social systems are living systems in an ontologically distinct way compared with non-human biological systems due to the primary mediation of symbolic meaning and the new lifeworld domains that are thereby opened up. The production of meaning, of symbolic artefacts, cannot be assimilated to the myriad forms of non-symbolic communication in organic life. A qualitatively different form of interpretation is required for humans acting in symbolic domains compared with that required for the interpretation of the stimulations, signals and gestures realized by living systems and non-linguistic animals (as I indicated earlier). The interpretation of symbols, however, still involves the individual in an ongoing relationship to context and to the special higher-level system–environment couplings in which the symbols make sense. Systems theory tends to conceive of these relationships as functional, as oriented toward the goal of systemic equilibrium (which is apt to lend systems theory a politically conservative bent), but this is to neglect in human societies the unique and complex processes of interpretive action through which meaning is realized and for which systems theory on its own cannot seem to account adequately. ¹² The bringing-forth of meaning within the symbolic domain is a cognitive achievement involving the bodily perceptual apparatus in conjunction with the symbolic and the social. It is a cognitive achievement of uniqueness, complexity and depth. The bringing-forth of meaning and the world context in which meaning makes sense cannot be reduced to any one element in this complex – for example, meaning-making is not merely a functional operation of social system maintenance, for it is not a purely social achievement (as in social constructivism or certain kinds of systems theory), and nor is it a matter of empirically observable physiological coordination (as in the behaviourism of B. F. Skinner’s psychology or, today, what Brothers [2001] calls ‘neuroism’ in neuroscience). ¹³ Interdisciplinary cognitive science that is philosophically informed hence gravitates toward the approach of ‘embodied cognition’ (Varela, Thompson and Rosch, 1991; Thompson, 2007; Noë, 2004; Stern, 2004; Johnson, 1987, 2007) – a recognition of the deep entwinement of sensual bodily processes of perception with the cognitive achievement of meaning in the mind and the coordination of the mind–body with its world.

My contention is that visual experience is always seeing, in the sense that perception and understanding always appear together in the active relationship between the organism and its world that is required for this. Recent research in philosophy of mind and in visual studies appears to confirm this. Alva Noë (2004) argues that visual experience is the result of an activity he calls ‘skilful looking’. Not only is vision connected to other senses – Noë considers vision to be more like touch than anything else – it does not require or use visual representations or models in the brain or in the physiological visual system. For many decades, scientists have thought that the brain somehow fills in the visual field to a greater or lesser extent using an internal representation, since we experience an unbroken field of vision despite having a ‘blind spot’ – the small region on the retina where there are no photoreceptors because in this spot axons must come together to form the optic nerve. We nevertheless experience a high-resolution, full-color visual field despite there being physical obstructions in front of the receptors of the retina and despite the uneven distribution of rods and cones across the surface of the retina. There are increasingly less cones outside the foveal (central) region, which means that the eye is nearly color-blind in its para-foveal region, yet we experience a high-resolution and uniformly colorful world despite these limitations (ibid.: 37). It is a mistake, Noë argues, to think that the brain or visual system somehow compensates for these limitations by constructing the better image that we actually experience from the inferior retinal stimulation on which it is based. There is no filling-in, he contends, because that would require something inside that is used as the basis or model for the filling in – an internal representation or picture in the mind – as well as something, some entity, to do the filling-in (this is the so-called ‘homunculus fallacy'). There is no evidence of ‘internal neural structures that are spatially or topographically isomorphic to that which they represent’ (ibid.: 48) – that is, the physiological visual system does not use images or representations of the world, only the mind does. Indeed, it is also a mistake to think, as much of the history of philosophical reflection on vision has done, that even the retina receives an image – Noë does not believe that what is projected onto the retina is an image at all – for one thing, if it existed, it would be upside down as a result of the eye’s lens and so would require something to turn it right side up in order for us to see the world properly (the homunculus fallacy again). Instead, he argues, we see by moving our eyes about, by fixing our gaze upon something, perhaps reaching for it – the whole scene is never given to us in a flash all at once, like a snapshot, but is acquired through an active experience of the world. Visual experience ‘depends on more than merely optical processes’, since any perceiver ‘must also make use of sensorimotor knowledge’ such that normal vision appears to require not only movement in relation to the environment but ‘self-actuated movement’ (ibid.: 13, 17, 26, 72–3; original emphases). Noë’s ‘skilful looking’ hence indicates that vision is enacted cognition, it always involves the body’s relation to its world, the accumulated knowledges gained from such relations and the enactment or bringing-forth of the world through action. This world is also – although Noë never comments on this – historical, social and political.

Indeed, Noë, for all his fine phenomenological investigations, somehow does not include social or political considerations, even though his analysis of the involvement of the body-in-the-world would seem also to involve the body in the realm of the political, since for humans being-in-the-world is also always already politically constituted. Under dominant conditions of capitalist accumulation, cultural experience, like work, is alienated and it is precisely the subtle violence of this alienation and all the systematic deceptions, suppressions and seductive compensations associated with it that makes the being-in-the-world of capitalism political in this critical sense. For Benjamin and the Frankfurt School, however, there is a dialectic to the domination of perception that generates reaction from subjects – psychological and somatic – and that manifests not just in pathologies of perception but in ‘lines of flight’, to use Deleuze’s felicitous phrase. While Adorno may have leaned too far towards critical despair in the face of ‘mass deception as enlightenment’, Benjamin finds significant potential for critical resistance within the form of the production and consumption of mass communication technologies. Such a politicization of perception complicates things significantly. Noë’s insights seem to neglect that ‘skilful looking’ – perception based on the active relations of the mind–body to its environment – can be significantly damaged or distorted when such relations are mediated by the requirements of capitalist industrialization.

The critical problem is that the damage and distortion are not experienced as such but are instead enjoyed or at least parried through mass cultural consumption. The politics of perception reaches beyond consciousness and as such is a factor and the constitution and experience of reality itself. ‘One could say’, Buck-Morss argues

… that the dynamics of capitalist industrialism had caused a curious reversal in which ‘reality’ and ‘art’ had switched places. Reality had become artifice, a phantasmagoria of commodities and architectural construction made possible by new industrial processes… Mass media … could now replicate this commodity world endlessly as the mere image of an illusion … But the critical, cognitive function in which a politicised art might participate was precisely the opposite: not to duplicate illusion as real, but to interpret reality as itself illusion. (Buck-Morss, 1983: 214)

Johnson (1993, 2007) has made explicit connections between embodied cognition, moral imagination and society but not yet with politics. Like Benjamin, Johnson is at pains to reject the Cartesian mind–body dualism and the empiricism that corresponds to it as well as to reject ‘moral objectivity’ and its attendant ‘Universal Reason’. Johnson believes that humans ‘are fundamentally imaginative creatures whose understanding of experience is built up with the imaginative materials of cognition’. These imaginative cognitive materials include the prototype structure of concepts, frame semantics, metaphor and narrative – none of which can support traditional Kantian ‘Moral Law’ theory. Such imaginative cognitive materials indicate that moral thinking always requires ‘imaginative rationality that is that once insightful, critical, exploratory, and transformative’. This indelibly connects moral understanding with aesthetics – the ‘aesthetic’, for Johnson, is construed broadly as ‘the imaginative structures, activities, orientations and transformations by means of which we are able to have coherent experience’ (Johnson, 1993: 3, 187, 208). Johnson (2007) deepens this perspective by contending that all human understanding has an aesthetic – that is, embodied – basis such that reason, meaning, imagination and, indeed, spirituality are all inextricably tied up with embodied processes. Drawing on American pragmatism and recent embodied cognition research, Johnson is in accord with Noë’s emphasis on the importance of action for perception such that ‘meaning’ must be understood to reach beyond conceptually conscious feeling and thought. The experience of meaning is first prepared by our ‘bodily connections with things … via our sensorimotor engagements’, which is clear in infant-development studies – but we do not shed infant-learned meanings when we grow up, we appropriate, recruit and refine them in more abstract modes of understanding (Johnson, 2007: 25, 51). What is needed to complete (but also to complicate) Johnson’s picture of imaginative cognition is the acknowledgement of the political dimension of embodied cognition.

Benjamin and the politics of perception

Noë, Johnson and Thompson offer philosophical reflections and accounts of experimental evidence that support my view that visual perception is essentially bound up with other senses and with a general internal relationship between mind–body and environment. The Benjaminian ‘different nature’ revealed by the camera, the ‘unconscious optics’ accessed through the viewing experience made possible by the camera – these experiences indicate, in the first instance, the awareness of this contradiction in visual experience – but historical content is then available to consciousness if reflection continues from this momentum. What is crucial to recognize as philosophical ‘truth-content’ in such an event is not the specific content itself but the experience of the contradiction. Such a moment of perception does not, in other words, primarily involve the apprehension of a representation at all, but rather the moment of contradiction itself. Such perception in moving image technology is mediated by the visual, audio and text but this specific awareness cannot be reduced to the semiotic elements that are its vehicle. Such a ‘dialectical image’ offers the possibility, it seems, of an awareness of the creative trace of perceptual world-making itself, yet without the need to define its form and content that, for example, a conventional political theory or a political ideology would demand for remaking the world. What this experience of moving image, sound and context indicates are processes and powers beyond the given in everyday visual representation. Such perception accesses and expresses the intellectual power of imagination, but my Benjaminian focus on the dialectics of seeing demands a specific kind of imagination: it involves the kind of perception that is sensitive to the distance between concept and thing, object and name, signifier and signified. Since overlooking this distance or having this distance systematically suppressed contributes to the reproduction of unofficial power relations in late capitalist society precisely by concealing them, the experience of such ‘distance' and ‘distraction' has an inherently consciousness-raising potential. But such critical experience is not limited to the potential for consciousness-raising, for the ‘spur to new ways of perceiving’ also, then, trains new habits of perceptual experience that are mediated non-linguistically by the mind–body and its relations with others. Such training thus must be understood quite differently from Carroll's notion of `habits of perception' because of its critical nature–Benjamin's is incompatible with the demands of capitalist rationalization whereas Carroll's is entirely compatible because the latter's notion of `habit' is merely operational.

What then does ‘the politics of perception’ indicate for media societies of the 21st century? It does not primarily refer to ideological struggle over the content or accessibility of media – such questions of organizing socio-political values, political openness, accessibility and representation are important and central to the politics of the media, but these issues are at a different (albeit intertwined) level from that of the politics of perception. The politics of perception concerns the mutual relations of the sensorium with its constitutive environments of sensation (including cognition) that are structured and organized by environmental, social, political and technological conditions into unique life histories–here, such mutual relations are understood as a totality, which Benjamin called the `collective body'. This indicates a politics that prepares the reception of media or technology itself and thus it requires special attention from political philosophy. I suggest, as a result of these considerations, that the recognition of embodied cognition provides a hopeful source of Benjaminian resistance. Since social coordination, ordering and reproduction take place increasingly via technically mediated communication in postmodern societies, Benjamin-inspired resistance against the oppressions, suppressions and distortions imposed by the capitalist requirements for such coordination, ordering and reproduction, which are themselves reproduced in the communication technologies corresponding to these processes, will continue to find inspiration precisely in and through technologies that are capable of generating dialectical images. A politicized art, for Benjamin, through its use of dialectical images, shocks the audience and distracts them from the reality of illusion. The structural conditions that generate this dialectic of experience are not ideological because they are internal to the sensory and cognitive experiences of technically reproduced art itself. The potential of Benjaminian politicized art is hence reproduced in technology that is central to the reproduction of late capitalism. Such a politics of perception expresses a dialectical principle that applies to all technically reproduced art – modern mass art as well as the new ‘demassified’ consumption increasingly associated with postmodernity. For both Marxist and post-Marxist thinkers, Walter Benjamin’s insights rightly continue to guide critical-political investigations into contemporary communication technologies.