McLuhanite theorists have argued that the codes of dominant media may have a subtle but profound influence on the perceptual processes or 'world views' of their users. In The Gutenberg Galaxy, Marshall McLuhan himself, following Ivins (1953), emphasized the 'impact' of print (McLuhan 1962). A different technical invention of the early Renaissance which was contemporaneous with Gutenberg's invention of movable type is often cited as having played a role in a profound shift in the Western cultural worldview. The mathematically-based technique of linear perspective was invented in 1425 by Filippo Brunelleschi and codified as perspectiva artificialis (artificial perspective) by Leon Battista Alberti in his treatise, Della pittura (On Painting), published in 1435-6 (Alberti 1966). For the artist it is a rational geometrical technique for the systematic representation of objects in space which mimics the everyday visual illusion that the parallel edges of rectilinear objects converge at what we now call a 'vanishing point' on the horizon.
We need reminding that this 'style of vision' is a historical invention: 'nothing like it appears earlier in medieval painting, suggesting that men and women of earlier ages simply did not see in this fashion' (Romanyshyn 1989, 40). Linear perspective thus constituted a new way of seeing which Samuel Edgerton characterizes as 'the most appropriate convention for the pictorial representation of "truth"' within 'the Renaissance paradigm' (a view of the world which reflected our understanding until the advent of Einstein's theory of relativity) (Edgerton 1975, 162). We have become so accustomed to reading pictures in terms of this illusionistic pictorial code that it now appears 'natural' to us to do so: we are rarely conscious of it as a code at all. In an essay on 'Perspective as Symbolic Form' published in German in the 1920s, the great art historian Erwin Panofsky generated considerable controversy by making the claim that linear perspective was a 'symbolic form' - a historically-situated system of conventions for representing pictorial space which reflected the dominant cultural worldview of the Italian Renaissance (Edgerton 1975, 153ff). Similarly, Herbert Read noted that 'we do not always realize that the theory of perspective developed in the fifteenth century is a scientific convention; it is merely one way of describing space and has no absolute validity' (cited in Wright 1983, 2-3). Critics retorted that strict geometrical perspective is scientifically 'accurate' and accused Panofsky and other heretics of 'relativism' (see Kubovy 1986, 162ff). Certainly, if we discount phenomenal reality, what William Ivins calls 'the grammar of perspective' can be seen as having an indexical character (Ivins 1975, 10). However, it can hardly be doubted that 'to close one eye and hold the head still at a single predetermined point in space is not the normal way of looking at things' (White 1967, 274).
Strict linear perspective does not
reflect phenomenal reality, since we are habituated to the stabilizing psychological mechanism of
'perceptual constancy' which we encountered earlier.
If you are not an artist, try holding a vertical pencil at arm's length in front of you as
a measuring-stick for objects within your field of view.
If you have not tried this before you
may be shocked to discover that some of the things which are close to
you seem implausibly large. This is perhaps most noticeable in relation to the foreshortening of
the human form - protruding feet can seem comically massive.
In tackling the task of depicting the foreshortened body of the dead Christ, even the great Renaissance artist Andrea Mantegna (1431-1506) clearly felt that he needed to compensate for the sacrilege of the distortion introduced by linear perspective, since he both reduced the size of the feet and enlarged the head, showing a 'sense of proportion' appropriate to his subject. Alberti had written that verisimilitude should be tempered by appropriate respect for dignity and decorum (Alberti 1966, 72-7).
Artificial perspective, in short, distorts the familiar size and shape of things. In this sense, Mantegna's representational code is closer to phenomenal reality than a photograph is. It wasn't easy even for Alberti to see things in terms of this code: he found it necessary to place a thin 'veil' or net marked out in parallels 'between the eye and the thing seen' (ibid., 68). Marshall McLuhan asserted that 'far from being a normal mode of human vision, three-dimensional perspective is a conventionally acquired mode of seeing, as much acquired as is the means of recognizing the letters of the alphabet, or of following chronological narrative' (McLuhan 1970, 16). Artificial perspective is a code which everyone with sufficient exposure to it can easily learn to read, although employing it effectively as an artist or architect requires far more deliberate learning. The introduction of 'artificial perspective' both reflected and promoted a preoccupation with verisimilitude; its use became an essential condition of 'realistic' pictorial representation. There is of course no doubt that this technique generates a powerful impression of depth, 'approximating the cues relating to normal perception better than any other strategy until the emergence of photography' (Nichols 1981, 52). But its revolutionary implications were not simply representational. Robert Romanyshyn's reflections on the implications of this invention are subtle and wide-ranging, but the following extract offers some illustrative insights.
The horizon... sets the limit for the height of any object to be depicted in the painting and...
it is fixed at the eye level of an observer imagined to be standing on a horizontal
plane and staring straight ahead at the world...
The painter (and the viewer) imagines that he or she is looking at the subject to be painted (the world to be viewed) as if through a window... The condition of the window implies a boundary between the perceiver and the perceived. It establishes as a condition for perception a formal separation between a subject who sees the world and the world that is seen; and in doing so it sets the stage, as it were, for that retreat or withdrawal of the self from the world which characterizes the dawn of the modern age. Ensconced behind the window the self becomes an observing subject, a spectator, as against a world which becomes a spectacle, an object of vision...
In addition to this separation between perceiver and world, the condition of the window also initiates an eclipse of the body. Looked at from behind a window the world is primarily something to be seen. Indeed, a window between me and the world tends not only to emphasize the eye as my means of access to the world but also to de-emphasize the other senses... And with this eclipse of the body fostered by the window, the world on the other side of the window is already set to become a matter of information. As a spectacle, an object of vision, it is analyzable, and readable as a computer print-out, for example, or as a blip on a radar screen...
The Renaissance code of central, one-point linear perspective is thus not simply a technique for indicating depth and relative distance in a two-dimensional medium. It is a pictorial code reflecting the growing humanism of the period, presenting images from a single, subjective, individual and unique visual point-of-view. Mirroring the 'vanishing point' within the picture is the projected 'point of origin' in front of the canvas which was adopted by the artist and left vacant for the 'subject' whose position we adopt when we look at the picture (Nichols 1981, 53). From this position, the represented world is framed as if by a window on a wall. The framed image 'stands in for the world it represents as would an ordinary window if the view beyond it could somehow be imprinted on its surface' (ibid., 159). Alberti himself wrote that 'I inscribe a quadrangle of right angles, as large as I wish, which is considered to be an open window through which I see what I want to paint' (Alberti 1966, 56). Painters should 'seek to present the forms of things seen on this plane as if it were of transparent glass' (ibid, 51). The artist Albrecht Dürer reminded us that 'perspectiva is a Latin word which means a view through something' (Panofsky 1970b, 123).
Yet at the same time as simulating a view of the world through a window, this clearly bounded, static and two-dimensional representation separates the viewer from the represented world. 'Seeing things in perspective' is regarded by Romanyshyn and other theorists as implicated in the development of the subject in its distancing of the knower from the known. William Ivins considered linear perspective to be part of 'the rationalization of sight' which he declared was 'the most important event of the Renaissance' (Ivins 1975, 13). Where linear perspective is very dominant in a work of art it undoubtedly gives it 'a cold effect', as John White observes (White 1967, 274). Marshall McLuhan regarded 'the detached observer' as 'the Renaissance legacy', declaring that 'the viewer of Renaissance art is systematically placed outside the frame of experience' (McLuhan & Fiore 1967, 53). 'The arbitrary selection of a single static position' for the artist and for viewers of the work requires 'a fixed point of view', which McLuhan associates with the 'private stance' of 'separate individuals' and not just with a viewing location (McLuhan 1962, 16, 56; McLuhan & Fiore 1967, 68). Looking at the represented world as through a window confirms us in our sense of ourselves as individuals with our own unique 'outlook on the world'. 'Gaining perspective' reflected 'our deeply embedded habit of regarding all phenomena from a fixed point of view' (McLuhan & Fiore 1967, 68). McLuhan attributed this not only to linear perspective but also to 'the effect of typography': 'inner direction toward remote goals is inseparable from print culture and the perspective and vanishing point organization of space that are part of it' (McLuhan 1962, 125, 214). This apparently purely technical innovation thus had subtle but profound ideological implications. Bill Nichols comments that 'the centering of and upon the subject or ego in Renaissance painting coincides with the first signs of a growing emphasis upon the individual rather than a chain of being, an emphasis that flourished with the emergence of entrepreneurial capitalism' (Nichols 1981, 53). The Renaissance code of artificial perspective constitutes visible testimony to the constitution of the self as subject.
Learning to read the pictorial code of linear perspective prepared us for the camera. McLuhan observed that 'photography is the mechanization of the perspective painting and of the arrested eye' (McLuhan 1970, 11). Photography offers a powerful illusion of a medium as a transparent 'window on the world'; as with paintings, photographic images are framed (even if only by their edges). In the early nineteenth century the camera obscura was fitted with a rectangular ground glass which showed only a rectangular section of the circular image from the lens (which is blurred at the edges) (Snyder & Allen 1982, 68-9). This made the camera image conform to the dominant form of framing used for paintings. Whilst sharing the single, central viewing point of painterly artificial perspective, photography involves the most remorseless application of this code. Photographs sometimes exhibit even more 'distortion' of phenomenal realism than paintings do: snapshots of tall buildings exhibit a disturbingly pronounced convergence of the vertical lines (we are less accustomed to vertical convergence because even Renaissance artists avoided it). In 35mm photography, the illusion of depth is most striking when 'normal' lenses of about 50mm are used: we become more aware of 'distortion' when a telephoto or wide-angle lens is used (Nichols 1981, 19). 'Photorealism' has nevertheless become the standard by which 'realistic' representations in visual art are subjectively judged (Kress & van Leeuwen 1996, 163-164).
Just as illusionism of Renaissance linear perspective performed the ideological function of 'positioning the subject', so too did the photographic image. 'The installation of the viewer as subject depends upon reserving a singular place for him or her, the reciprocal in front of the image of the vanishing point "behind" it, the point of origin from which the camera "took" its view and where we now take ours' (Nichols 1981, 159). French theorists associated with the journals Tel Quel and Cinéthique argued that since the code of linear perspective is built into the camera, photography and film which, whilst appearing to involve simply a neutral recording of reality, serve to reinforce 'a bourgeois ideology which makes the individual subject the focus and origin of meaning' (Stam 2000, 137). Film and television add a narrative dimension to the positioning of the subject, incorporating not only linear perspective but also dominant narrative devices specific to filmic media. Film theorists refer to the use of 'suture' (surgical stitching) - the 'invisible editing' of shot relationships which seeks to foreground the narrative and mask the ideological processes which shape the subjectivity of viewers. Some Lacanian theorists argue that in the context of conventional narrative (with its possibilities of identification and opposition), the unique character of the cinema (e.g. watching a large bright screen in the dark) offers us the seductive sense of a 'return' to the pre-linguistic 'mirror-phase' of the 'Imaginary' in which the self was constructed (Nichols 1981, 300).
'A sign... addresses somebody,' Charles Peirce declared (Peirce 1931-58, 2.228). Signs 'address' us within particular codes. A genre is a semiotic code within which we are 'positioned' as 'ideal readers' through the use of particular 'modes of address'. Modes of address can be defined as the ways in which relations between addresser and addressee are constructed in a text. In order to communicate, a producer of any text must make some assumptions about an intended audience; reflections of such assumptions may be discerned in the text (advertisements offer particularly clear examples of this).
Rather than a specifically semiotic concept, 'the positioning of the subject' is a structuralist notion - although Stuart Hall notes its absence in early structuralist discourse (Hall 1996, 46); Saussure did not discuss it. It is a concept which has been widely adopted by semioticians and so it needs to be explored in this context. The term 'subject' needs some initial explanation. In 'theories of subjectivity' a distinction is made between 'the subject' and 'the individual'. As Fiske puts it, 'the individual is produced by nature; the subject by culture... The subject... is a social construction, not a natural one' (Fiske 1992, 288; my emphases). Whilst the individual is an actual person, the subject is a set of roles constructed by dominant cultural and ideological values (e.g. in terms of class, age, gender and ethnicity). Ideology turns individuals into subjects. Subjects are not actual people but exist only in relation to the interpretation of texts and are constructed through the use of signs. The psychoanalytical theorist Jacques Lacan undermined the humanist notion of a unified and consistent subject. The individual can occupy multiple subject positions, some of them contradictory. 'Identity' can be seen as 'a matrix of subject-positions' (Belsey 1980, 61). The fluidity and fragmentation of 'identity' is highlighted in the context of the internet, where the 'subject' need have no necessary connection to a supposed referent (a specific individual in the material world); gender, sexual orientation, age, ethnicity or any other demographic marker may be shifted at will (subject to the social competence required to sustain such a virtual identity).
According to theorists of textual positioning, understanding the meaning of a text involves taking on an appropriate ideological identity. In order to make sense of the signs in a text the reader is obliged to adopt a 'subject-position' in relation to it. For instance, to understand an advertisement we would have to adopt the identity of a consumer who desired the advertised product. Some theorists argue that this position already exists within the structure and codes of the text. 'Narratives or images always imply or construct a position or positions from which they are to be read or viewed' (Johnson 1996, 101). What Colin MacCabe famously called the 'classic realist text' is orchestrated to effect closure: contradictions are suppressed and the reader is encouraged to adopt a position from which everything seems 'obvious' (MacCabe 1974). This stance assumes both that a text is homogeneous and that it has only one meaning - that which was intended by its makers - whereas contemporary theorists contend that there may be several alternative (even contradictory) subject-positions from which a text may make sense. Whilst these may sometimes be anticipated by the author, they are not necessarily built into the text itself. Not every reader is the 'ideal reader' envisaged by the producer(s) of the text. The phrase, 'the positioning of the subject' implies a 'necessary "subjection" to the text' (Johnson 1996, 101) and is thus problematic since there is always some freedom of interpretation. We may for instance choose to regard a poorly-translated set of instructions for assembling flat-pack furniture as a text constructed purely for our amusement.
The notion that the human subject is 'constituted' (constructed) by pre-given structures is a general feature of structuralism. It constitutes a radical opposition to the liberal humanist (or 'bourgeois') stance which presents society as 'consisting of "free" individuals whose social determination results from their pre-given essences like "talented", "efficient", "lazy", "profligate", etc.' (Coward & Ellis 1977, 2). The French neo-Marxist philosopher Louis Althusser (1918-1990) was the first ideological theorist to give prominence to the notion of the subject. For him, ideology was a system of representations of reality offering individuals certain subject positions which they could occupy. He famously declared that 'what is represented in ideology is... not the system of real relations which govern the existence of individuals, but the imaginary relation of these individuals to the real relations in which they live' (Althusser 1971, 155). He outlined the ideological mechanism of interpellation:
Assuming that the theoretical scene I have imagined takes place in the street, the hailed individual will turn round. By this mere one-hundred-and-eighty-degree physical conversion, he becomes a subject. Why? Because he has recognized that the hail was 'really' addressed to him, and that 'it was really him who was hailed' (and not someone else). (Althusser 1971, 174)
The Althusserian concept of interpellation is used by Marxist media theorists to explain the political function of mass media texts. According to this view, the subject (viewer, listener, reader) is constituted by the text, and the power of the mass media resides in their ability to position the subject in such a way that their representations are taken to be reflections of everyday reality. Such structuralist framings of positioning reflect a stance of textual determinism which has been challenged by contemporary social semioticians who tend to emphasise the 'polysemic' nature of texts (their plurality of meanings) together with the diversity of their use and interpretation by different audiences ('multiaccentuality'). However, a distinction may be appropriate here between message and code. Whilst resistance at the level of the message is always possible, resistance at the level of the code is generally much more difficult when the code is a dominant one. The familiarity of the codes in 'realist' texts (especially photographic and filmic texts) leads us to routinely 'suspend our disbelief' in the form (even if not necessarily in the manifest content). Recognition of the familiar (in the guise of the 'natural') repeatedly confirms our conventional ways of seeing and thus reinforces our sense of self whilst at the same time invisibly contributing to its construction. 'When we say "I see (what the image means)" this act simultaneously installs us in a place of knowledge and slips us into place as subject to this meaning... All the viewer need do is fall into place as subject' (Nichols 1981, 38). Falling into place in a realist text is a pleasurable experience which few would wish to disrupt with reflective analysis (which would throw the security of our sense of self into question). Thus we freely submit to the ideological processes which construct our sense of ourselves as free-thinking individuals.
A primary textual code involved in the construction of the subject is that of genre. Genres are ostensibly 'neutral', functioning to make form (the conventions of the genre) more 'transparent' to those familiar with the genre, foregrounding the distinctive content of individual texts. Certainly genre provides an important frame of reference which helps readers to identify, select and interpret texts (as well as helping writers to compose economically within the medium). However, a genre can also be seen as embodying certain values and ideological assumptions and as seeking to establish a particular worldview. Changes in genre conventions may both reflect and help to shape the dominant ideological climate of the time. Some Marxist commentators see genre as an instrument of social control which reproduces the dominant ideology. Within this perspective, the genre is seen as positioning the audience in order to naturalize the reassuringly conservative ideologies which are typically embedded in the text. Certainly, genres are far from being ideologically neutral. Different genres produce different positionings of the subject which are reflected in their modes of address. Tony Thwaites and his colleagues note that in many television crime dramas in the tradition of The Saint, Hart to Hart, and Murder, She Wrote,
Thus, over and above the specific 'content' of the individual text, generic frameworks can be seen as involved in the construction of their readers.
Saussure emphasized that the language system is a 'given' which precedes its users and is beyond human control. Developing this stance, post-Saussurean structuralist theorists have argued that contrary to the notion that semiotic systems are instrumental tools which are fully subject to the control of 'the individual', the subject is constructed by the semiotic system of language, ideology, and myth. Such structural determinism and autonomy is reflected, for instance, in Lévi-Strauss's declaration that 'I... claim to show, not how men think in myths, but how myths operate in men's minds without their being aware of the fact' ('les myths se pensent dans le hommes, et á leur insu') (Lévi-Strauss 1970, 12). It is similarly evident in Althusser: 'Marx observes that what determines a social formation in the last instance... is not the spirit of an essence or a human nature, not man, not even "men", but a relation, the relation of production' - in other words, as Coward and Ellis put it, 'man is not the origin of society, it is rather that society is the origin of man' (Coward & Ellis 1977, 82, including this citation from Althusser). And the psychoanalyst Jacques Lacan observed that 'man speaks, but it is only that the symbol has made him man' (cited in Coward & Ellis 1977, 107). Whilst providing the key framework from which much of structuralist (and post-structuralist) theory was derived, Saussure did not himself advance the proposition that the subject is constructed by the (language) system. In an astonishingly contemporary observation published in 1868, the co-founder of what we now know as semiotics, the logician Charles Peirce, declared in a quasi-syllogistic form that 'the fact that every thought is a sign, taken in conjunction with the fact that life is a train of thought, proves that man is a sign... Thus my language is the sum total of myself; for the man is the thought' (Peirce 1931-58, 5.314). He went on to note that 'it is hard for man to understand this, because he persists in identifying himself with his will' (ibid., 5.315). As in several other instances, Peirce's notions find their echoes in poststructuralist theory, albeit in more dramatic forms. One hundred years later, the French historian of ideas Michel Foucault declared apocalyptically that 'as the archaeology of our thought easily shows, man is an invention of recent date. And one perhaps nearing its end' (Foucault 1970, 387).
In a famous chapter of his book, The Order of Things, Foucault discusses Las Meninas (The Maids of Honour) painted in 1656 by the Spanish artist Diego Velásquez. Whilst the ostensible subject of the painting is the princess, surrounded by her maids of honour, this is an extraordinarily reflexive painting about painting - or more broadly, about the business of representation. It can indeed be seen as a meditation on the role of the artist, on the depiction of reality and perhaps above all on what Ernst Gombrich refers to as 'the beholder's share' in making sense of the visual world. There is a particular irony here in the fact that it is likely that the primary spectators for the Las Meninas were the models whom the artist is here depicted as painting.
The painter is standing a little back from his canvas. He is glancing at his model... He is
staring at a point to which, even though it is invisible, we, the spectators, can easily
assign an object, since it is we, ourselves, who are at that point... The spectacle he is
observing is thus doubly invisible: first, because it is not represented within the space of
the painting, and, second, because it is situated precisely in that blind point, in that
essential hiding-place into which our gaze disappears from ourselves at the moment of our
actual looking... In appearance, this locus in a simple one: a matter of pure reciprocity:
we are looking at a picture in which the painter is in turn looking out at us... And yet...
the painter is turning his eyes towards us only insofar as we happen to occupy the same
position as his subject. We, the spectators, are an additional factor. Though greeted by
that gaze, we are also dismissed by it, replaced by that which was always there before we
were: the model itself... The great canvas with its back to us on the extreme left of the
picture... prevents the relation of these gazes ever being discoverable or definitely
established... Because we can see only that reverse side, we do not know who we are, or
what we are doing. Seen or seeing?... We are observing ourselves being observed by the painter,
and made visible to his eyes by the same light that enables us to see him...
Now as it happens, exactly opposite the spectators - ourselves - on the wall forming the far end of the room, Velásquez has represented a series of pictures; and we see that among all those those hanging canvases there is one that shines with particular brightness. Its frame is wider and darker than those of the others... But it isn't a picture: it is a mirror. It offers us at last that enchantment of the double that until now has been denied to us... In fact, it shows nothing of what is represented in the picture itself... What it is reflecting is that which all the figures within the painting are looking at so fixedly, or at least those who are looking straight ahead; it is therefore what the spectator would be able to see if the painting extended further forward, if its bottom edge was brought lower until it included the figures the painter is using as models... At the far end of the room, ignored by all, the unexpected mirror holds in its glow the figures that the painter is looking at (the painter in his represented, objective reality, the reality of the painter at his work); but also the figures that are looking at the painter (in that material reality which the lines and the colours have laid out upon the canvas)... King Philip IV and his wife, Mariana... What all the figures in the picture are looking at are the two figures to whose eyes they too present a scene to be observed. The entire picture is looking out at a scene for which it itself is a scene...
Inversely, as far as they stand outside the picture and are therefore withdrawn from it in an essential invisibility, they [the king and his wife] provide the centre around which the entire representation is ordered; it is they who are being faced, it is towards them that everyone is turned... In the realm of the anecdote, this centre is symbolically sovereign since it is occupied by King Philip IV and his wife. But it is so above all because of the triple function it fulfils in relation to the picture. For in it there occurs an exact superimposition of the model's gaze as it is being painted, of the spectator's as he contemplates the painting, and of the painter's as he is composing the picture (not the one represented, but the one in front of us which we are discussing). These three 'observing' functions come together in a point exterior to the picture; that is, an ideal point in relation to what is represented, but a perfectly real one too, since it is also the starting-point that makes the representation possible...
Perhaps there exists, in this painting by Velásquez, the representation as it were, of Classical representation, and the definition of the space it opens up to us.
The modes of address employed by texts within a code are influenced primarily by three inter-related factors:
In this context it may be useful to consider a basic typology of modes of communication in terms of synchronicity - whether or not the participants can communicate 'in real time' - without significant delays. This feature ties together the presence or absence of the producer(s) and the technical features of the medium. The obvious options are:
Note that this framework seems to posit an apparently empty category of synchronous mass communication which is hard to imagine. Clearly such features of the mode of communication also relate to the relative numbers of participants involved, which are sometimes categorised in terms of: one-to-one; one-to-many; many-to-one (e.g. petitions and requests for information); and many-to-many (e.g. internet discussion lists and newsgroups). Once again, the limitations of such a framing should be noted - this one tends to overlook the importance of communication in small groups (which consist of neither 'one' nor 'many'). Whatever the shortcomings of any particular typology, however, all of the factors referred to here have the potential to influence the mode of address employed.
Modes of address differ in their directness, their formality and their narrative point-of-view. In relation to literature narrative point-of-view has received exemplary treatment in Wayne Booth's book, The Rhetoric of Fiction, originally published in 1961 (Booth 1983; see also Genette 1972).
The various narrative points-of-view in literature are as follows:
In television and film, omniscient narrative is dominant, although narrative point-of-view tends to shift. Camera treatment is called 'subjective' when the camera shows us events as if from a particular participant's visual point of view (encouraging viewers to identify with that person's way of seeing events or even to feel like an eye-witness to the events themselves). This first-person style in filmic media is rarely sustained, however (or we would never see that character). The point-of-view is selective when we are mainly concerned with a single character but the camerawork is not subjective. Voice-overs are sometimes used for first-person narration by a character in a drama; they are also common as a third-person narrative mode in genres such as documentary. Where first-person commentary shifts from person to person within a text, this produces 'polyvocality' (multiple voices) - constrasting strongly with the interpretative omniscience of 'univocal' narrative which offers a single reading of an event (Stern 1998, 63). Where the agency of a narrator is backgrounded, events or facts deceptively seem to 'speak for themselves'.
Modes of address also differ in their directness (Tolson 1996, 56-65). In linguistic codes, this is related to whether 'you' are explicitly addressed, which in literary modes is quite rare. In Laurence Sterne's highly 'unconventional' novel Tristram Shandy (1760), one chapter begins thus: 'How could you, Madam, be so inattentive in reading the last chapter?' (vol. 1, ch. 20). 'Realist' fiction avoids such 'alienatory' strategies. In representational visual codes directness is related to whether or not a depicted person appears to look directly at the viewer (in the case of television, film and photography, via the camera lens). A direct gaze simulates interaction with each individual viewer (an impossibility, of course, outside one-to-one communicative media, but a feature of 'cam-to-cam' communication on the internet or in video-conferences). In film and television, directness of address is reflected in linguistic codes as well as camerawork. Films and (especially) television programmes within the documentary genre frequently employ a disembodied voice-over which directly addresses the audience, as do television commercials. On television, directness of address is also a matter of the extent to which participants look directly into the camera lens. In this way too, commercials frequently include direct address. As for programmes, in a book on The Grammar of Television, an industry professional warned: 'Never let a performer look straight into the lens of a camera unless it is necessary to give the impression that he is speaking directly to the viewer personally' (Davis 1960, 54). In television programmes, a direct mode of address is largely confined to newsreaders, weather forecasters, presenters and interviewers - which is why it seems so strange on the rare occasions when we notice an interviewee glancing at the camera lens. In short, people from outside the television industry are seldom allowed to talk to us directly on television. The head of state or the leader of a political party are amongst the few outsiders allowed to look directly at the viewer, and then typically only within special genres such as a party political broadcast or an 'address to the nation'. Direct address reflects the power of the addresser and the use of this signifier typically signifies 'authority'. Direct address is rare in the cinema, and when it is used it tends to be for comic effect. Indirect address is the principal mode employed in conventional narrative, masking authorial agency in the interests of foregrounding the story. Conventional film and television drama depends on 'the illusion that the represented participants do not know they are being looked at, and in which the represented participants must pretend that they are not being watched' (Kress & van Leeuwen 1996, 126).
Additionally, the mode of address varies in its formality or social distance. Kress and van Leeuwen distinguish between 'intimate', 'personal', 'social' and 'public' (or 'impersonal') modes of address (Kress & van Leeuwen 1996, 130-135). In relation to language, formality is quite closely tied to explicitness, so that intimate language tends to be minimally explicit and maximally dependent on non-verbal cues, whilst public language tends to reverse these features (especially in print). In usage related also to directness of address, social distance can also established through the use of loaded quasi-synonyms to reflect ideological distinctions of 'us' from 'them', as in 'I am a patriot; you are a nationalist; they are xenophobes'.
In visual representation, social distance is related in part to apparent proximity. In camerawork, degrees of formality are reflected in shot sizes - close-ups signifying intimate or personal modes, medium shots a social mode and long shots an impersonal mode (Kress & van Leeuwen 1996, 130-135; see also Deacon et al. 1999, 190-94 and Tuchman 1978, 116-20). In visual media, the represented physical distance between the observed and the observer often reflects attempts to encourage feelings of emotional involvement or critical detachment in the viewer. The cultural variability of the degree of formality signified by different zones of proximity was highlighted in relation to face-to-face interaction in an influential book by Edward T Hall - The Hidden Dimension (Hall 1966). Proximity is not the only marker of social distance in the visual media: angles of view are also significant. High angles (looking down on a depicted person from above) are widely interpreted as making that person look small and insignificant, and low angles (looking up at them from below) are said to make them look powerful and superior (Messaris 1997, 34-5; Messaris 1994, 158; Kress & van Leeuwen 1996, 146). The interplay of these techniques is important. In the three photographs shown here of Michelangelo's David (1501-4, Accademia, Florence), whilst all of the shots are taken from below this gigantic figure, the close-up from below seems to me to emphasize the power of the figure in contrast to the mid-shot, in which - despite the musculature - David seems somewhat softer and more vulnerable. The closer we are the more we look upwards. Power is signified most strongly by a low angle which is also a close-up - as if, as we get closer, we become more vulnerable.
Note that whilst the significations such as those listed in relation to photographic and filmic modes of address may represent the currently dominant, conventional or 'default' linkages of signifiers and signifieds, no programmatic decoding based on a 'dictionary' of one-to-one correspondences is possible - in analogue codes in particular there is a sliding relationship between signifiers and signifieds which the codes of the particular textual systems in which they are employed may function to anchor in various ways (Nichols 1981, 108).
Textual codes construct possible reading positions for the addresser and addressee. Building upon Jakobson's model Thwaites et al. define 'the functions of address' in terms of the construction of such subjects and of relationships between them.
A textual code can be defined as a set of ways of reading which its producers and readers share. Not everyone has access to the relevant codes for reading (or writing) a text. The phatic function excludes as well as includes certain readers. Those who share the code are members of the same 'interpretative community' (Fish 1980, 167ff, 335-6, 338). David Morley demonstrated differential access to the textual codes of a programme in the television 'news magazine' genre (Morley 1980). Familiarity with particular codes is related to social position, in terms of such factors as class, ethnicity, nationality, education, occupation, political affiliation, age, gender and sexuality. This argument need not reflect social determinism, since one may argue that there is still scope for variety in the ways in which individuals engage with such codes.
Some codes are more widespread and accessible than others. Those which are widely distributed and which are learned at an early age may seem 'natural' rather than constructed (Hall 1980, 132). John Fiske distinguishes between broadcast codes, which are shared by member of a mass audience, and narrowcast codes which are aimed at a more limited audience; pop music is a broadcast code; ballet is a narrowcast code (Fiske 1982, 78ff). Broadcast codes are learned through experience; narrowcast codes often involve more deliberate learning (Fiske 1989, 315). Following the controversial sociolinguistic theories of Basil Bernstein, what Fiske refers to as broadcast codes are described by some media theorists as 'restricted codes', with Fiske's narrowcast codes being described as 'elaborated codes' (Bernstein 1971). 'Restricted' codes are described as structurally simpler and more repetitive ('overcoded'), having what information theorists call a high degree of redundancy. In such codes several elements serve to emphasise and reinforce preferred meanings. In contrast, literary writing - in particular poetry - has a minimum of redundancy (Lotman 1976). The distinction between 'restricted' and 'elaborated' codes serves to stress the difference between an élite ('highbrows') and the majority ('lowbrows'). Michael Real claims that the 'most popular' culture of 'the mass market' is marked by a high degree of redundancy (notably in the use of standard conventions and 'formulas'), whilst 'higher, élite or avante-garde art' employs 'elaborated codes' which are held to involve 'more originality and unpredictability' (Real 1996, 136). Similarly, Fiske suggests that narrowcast (elaborated) codes have the potential to be more subtle; broadcast (restricted) codes can lead to cliché. Jonathan Culler suggests that 'literature continually undermines, parodies, and escapes anything which threatens to become a rigid code or explicit rules for interpretation... literary works never lie wholly within the codes that define them' (Culler 1985, 105). Insofar as such positions suggest that broadcast codes restrict expressive possibilities this arguments has affinities with Whorfianism. The dangers of élitism inherent in such stances make it particularly important that the evidence is closely examined in the context of the particular code under study.