PRINT February 1980

Art & Language

I. Art, Language, Ideas etc.

The polemical desiderates supposedly satisfied by the meta-artistic text, the magazine-page-as-art-work, the meta-meta text, the reflective-art-talking-about-art-as-art, etc., are grotesquely over-rehearsed.

There are many standard interrelated hypotheses, many of them tendentious. It may be claimed, for example that a critical ‘language-art’ practice involves (somehow) a more or less direct rejection of some extrinsic institutional evaluation which its practitioners see as repressive or on the side of repression or ‘on the side of an alienating hierarchy of values’ etc. This is a claim to a practice of counter-interpretation: the claim is that the meaning of some ‘language art practice’, or some practice of putting-it-in-little-printed-magazines-and-sending-it-through-the-post etc., is that it stands for a rejection of certain dominant meaning structures in favour of ‘a practice of counter-meaning production’.

A further standard hypothesis is that to the extent that a language art, Art & Language, etc., etc., practice repudiates the production and public exhibition of an Art product, it refuses or provides a critical purchase on ‘the moment of objectification itself. What is supposed to follow from this is the possibility that the language art, printed page, etc., work etc., excludes itself from interpretation other than with respect to this public repudiation. (‘To the extent that . . .’ though, as in the previous sentence.)

‘Language art’, etc., is thus supposed to be unilateral intransigent meta-art practice.

In its terrain of contestation ‘language art’, etc., is supposed to pose ‘fundamental’ type questions of the kind ‘Who interprets the interpreters?’

The language art, etc., type practice is supposed or has been supposed to carry out a displacement from the ‘visual’ (erroneously thought to be unmediated) world to the consciously linguistic (mediated) and ratiocinative world in which is the prior moment of construction.

The printed, posted etc.,language art work has often a critical, even subversive self-image. It is sometimes thought to be didactic interpretative text (a sort of non-art). It is sometimes disruptive or disimbricatory of the structure of some possible other art work. Language as a determinate means resists or prevents unreflected interpretation on the part of the bearers of institutional power.

All good stuff. Everyone has read something like it before. The work is alive ‘so long as an atmosphere of questioning is generated’. The trouble is that this atmosphere of questioning is frequently one which favours the talking of rubbish, a constant deferral of substantive enquiry. The Artistic Printed Page (and its Conceptual artistic sibling) has produced a good deal of drivel incapable of satisfying its articulate tendentious claims.

It was a part of this fetishism of Conceptual art to enshrine the printed page— ‘words’ as a new art material. The style of conceptual art has ‘the printed page’ as a major and somehow classical constituent.1

Associated with a kind of art whose primary ‘material’ was ‘the printed page’ or words (tokens?) was the thought that art might catch up with the ‘mass media’. You could have major (not merely ‘graphic’) art sent through the post, cheap or free and easily distributable. This thought was exemplified in the global distributive delusions of (e.g.) the pre-‘Marxist’ Seth Siegelaub.

This was not the work of art in the age of its mechanical reproducibility, not the picture of the cathedral, but xeroxed A4 text, small ads, telegrams. Some of it properly demanded a certain reflected interpretation, most of it demanded interpretation as precious-intaglio-plus-poetry, etc.

There might have been a more substantial historical upshot had the iconoclasm of ‘language art’ been better than a mystification. But its ‘lawlessness’ was the lawlessness not of anarchy but of the unending regress; the lawlessness peculiar to a sort of ‘professionalism’. The most obvious upshot was the opportunity afforded to many conceptual-type artists to be self-curating. Artists could participate overtly and actively (if ambiguously) in the managerial games of the art-press art world. The atmosphere of typewriter and printed page became not so much an atmosphere of cultural contestation as a rat-race of half-truth and intellectual lumpenness.

The ‘explosion’ of artists’ magazines in the USA during the 1970s was, in general, the mediafication of self-promotion by intellectuals: ‘lazzaroni, pickpockets, literati, organ grinders, rag pickers, knife grinders’2, lumpens:. . . ‘The dangerous class . . . the bribed tool of reactionary intrigue’. 3

We have thought of Art-Language work and of Art & Language’s work as representational. Structural or semiological conditions are not the only ones that must be satisfied if a particular artistic problem is to be considered representational. We want to examine some of the possible conditions of a non-idealist sense of representation or signification—a sense that moves towards realism. In trying to do this we will move with no great confidence to some questions at the edge of art’s epistemological problems—particularly the problem of the conditions of its truth or possible truth. It must be clear that these conditions will be such that no art practice, however ‘abstract’, or ‘non-representational’ or ‘meta’-this or -that its products may be claimed to be, will be immune from their operation. In trying to explore these questions we postulate a potential equivalence of representations. That is, all knowledge while it connects to the real world is essentially an active and socially mediated process. All knowledge, visual or verbal, realistic or abstract, is thus representational. We assert that this potential equivalence of representations is a ceteris paribus of enquiry in artistic work. We also assert that a technical change in the way that art is produced is required if artistic production (in any institution) is to be defended in historical materialism.

The idealised journalistic intervention-as-art is no such change. It is no more than a means for artists to prevent themselves from being changed. There are many artists and critics who are interested in ‘intervening’ with one kind of exposée or disimbrication or other. They are usually Structuralism’s lumpens. In particular they are Semiology’s lumpens (vide Art-Language, Volume 4 Number 2, 1977). They engage, for example, in a structuralist extension of Chapter Seven in John Berger’s Ways of Seeing (vide Art-Language, Volume 4 Number 3, 1978). Their disimbrications are exceedingly inflated. The extent to which semiological work of any kind can be pursued beyond an idealist triviality is a matter of conjecture. The drift of the semiological self-image has been towards the view that there is hardly any limit to the world-amenable-to-Semiology. In fact, what Semiology tries to do is to treat non-linguistic culturally realised entities in the same way as structural linguistics treats language. This means that Semiology often provides metaphors. But Semiology started to take itself literally and then to get more imperialistic. It has become a denial of material life and history. Semio-art the more so—since it (quite naturally) conscripts the egregious Althusser as its philosophical mentor, and just learns his fines.

The grounds of semio-art can be shown to be shaky without anything but the most oblique reference to its intellectual masters.

1) The Semiologists’ claim that we are surrounded (etc.) by signs does not imply that we are all ‘ideologically’ transfixed by them, and even if we were it would not follow that we were affected in the same way or as constituents of class monoliths.

2) Semio-art underestimates the complexity and discursivity of the practice of its thematically favoured class.

3) It is very hard to find a way to dissociate the Semiologist and the semio-artist from the belief that people don’t make history.

4) As a consequence of (1 )–(3) it is hard to see how semio-art, etc., can defend itself against the charge that it is anti-working class.

5) As a consequence (etc.), it is hard to see how semio-art is not grotesquely idealistic.

6) As a consequence of (1)–(5) it is hard to see how its ‘interventions’ are not on the side of the ruling class.

7) It is anyway hard to see its ‘interventions’ as anything but opportunist and careerist.

There are many more corollaries and lemmas. It is hard not to get tired of trying to enumerate them.

Lumpens are venal. Semiologists are lumpen idealists. This ‘pair’ is not as strange as it might seem. Semio-artists contribute very little to knowledge. Their work is attached to the world ‘somehow’. Something will have to be said about its genesis. But what do we talk about?

Alison Assiter’s paper ‘Philosophical Materialism or the Materialist Conception of History’4 sorts out some of the problems suggested in its title. Roughly, she points out that Marx was not primarily concerned with materialism as philosophical theory but as intrinsic to historical materialism. And historical materialism is not conspicuously ontological theory nor dearly epistemological theory. (Althusser has Marx producing the science of history and then, wonderfully, Lenin cranking out the philosophy. Marx’s materialism is not to be found in the philosophical tradition of materialism. Lenin’s was a vulgarisation of Marx’s materialism into the philosophical materialist tradition.)

If artistic realism must be compatible with materialism, it must be with historical materialism and not with Lenin’s vulgarisation or with Althusser’s at best half-true (untrue) and confusing claims on behalf of Materialism and Empirio-Criticism.

Some people have ‘taken’ the historical materialist point in a well known, silly, self-refuting way. The result is that they have said that because all ideas (Art, etc.) are determined by the economic base, all ideas are ideological and (by sleight of modus ponens) false.

It can be argued (pace Assiter loc. cit.) that Art, etc., may be determined by the base, etc.—causally generated by the means of subsistence—but this doesn’t mean that we can’t look at it somehow independently of these causal determinations. Does this need justifying? Assiter sets about coping with apparent contradictions in Marx’s work: It has been argued that ‘conscious existence’ and ‘mode of production’ are equivalent when they occur in connection with a description of an individual. They both refer to or pick out producers in the productive process. It has also been argued that the ‘productive life’ determines ‘consciousness’—that ‘consciousness’ is not equivalent to ‘productive life’ or etc. ‘The gougeing careerist with a show at X now’ refers to the same individual as ‘the artist who has disimbricated publicity by such and such a means’. The same individual can be picked-out by both descriptions, but that will not necessarily prohibit the circumstance such that the life of the gougeing careerist will determine the ideas of the artist. Historical materialism doesn’t reduce to something silly simply because of the imputation of an ‘opaque context’.

‘Scott’ must have been ‘a writer working in such and such a way to live’ as the Author of Waverley. Scott’s consciousness can have been determined by his subsistence practice even though ‘Scott’s consciousness’ and ‘Scott’s subsistence practice’ denote the same individual in the process of production. (There are other ways of looking at this ‘dilemma’—and no doubt the apparent lack of extensionality is a problem for some. This apparent lack is perhaps not so keenly felt when it is remembered that the matter is historical materialism.)

We might accept that there is an interpretation of historical materialism with no tendency to reduce ‘Artist’ to ‘Producer’. It does seem fairly obvious. Rick Rio-Tinto-Zinc-Shareholder could churn out ‘Left-wing’ ideas, quote Marx or Scripture and make ‘true’ statements (though only as oratio obliqua perhaps). The epistemological credentials of this work may well seem to be impeccable. But how intelligible would these ‘themes’ be? Remember, that while there is no tendency to reduce ‘Artist’ (apparently ≠ Producer) to ‘Producer’ (or gouger for that matter) this producer determines or conditions the artist. The Producer’s life (practice) determines how the Artist and his work are produced. The genetic link that the artist’s ideas or pictures have with the world is distinguished from the iconic or descriptive link (see below). This is not to say that the genetic link will not have some explanatory power in relation to the iconic or descriptive ones. Artist’s do not, for example, passively pick-out space-occupying matter. Things are singled out actively, under descriptions. The perceptions are active. But they can be of something that exists or that ‘is the case’ or they can be hallucination or illusion and in particular pictures can be explained or accounted for by reference to what they in fact describe or stand for (as metaphor or etc.), or they can (or must) be accounted for genetically with reference to something else—something other than that which they may be thought to describe.

The fact that historical materialism does not tend (or can be made not to tend) to reduce ‘Artist’ to ‘Producer’ does not remove the dialectical tension between them.

Consider an artist who is also president and major shareholder of a company specialising in the manufacture and sale of police truncheons and riot gear. We would have to give some genetic (i.e., epistemological) weight to his non-artistic work in trying to deal with and explain his ‘art’. This would not necessarily allow us to reduce his actions qua artist to his actions qua managing director etc. But we must ask how he got his artist’s ideas, what determined his artistic actions, on pain of idealising him as artist in terms of what his apparent ‘ideas’ are. Remember, his ideas will be referred to by reference to what they are of—and what they are of is how they are got as well as what they refer to. If what they describe (or involve description of) plays a significant part in how they are got, then carry on regardless. If they are not, what sort of questions do you ask?

We are not saving that a picture is ‘true’ or ‘realist’ exclusively in virtue of the social credentials of the producer. We are saying that the absence of a significant and intelligible coincidence between the genetic and descriptive features of a picture will lead to a hiatus in critical practice. This hiatus will be resolved in the production of something like ‘an informal fallacy’— in fact an unavoidable contextualisation of the ‘descriptive’ content of the picture in respect of what can be defended as its real determining conditions.

What follows is a discussion of some practical considerations for the assessment of realism in art production. This discussion traces some oldie logic by means of which the foregoing assertions may be defended. For a more extended treatment of the same subject, see Art-Language vol. 4 no. 4, February 1980.5

II. On the relation ‘picture of . . .’

The possible distinction between the descriptive and genetic features of pictures is usually left very fuzzy in the higher reaches of realist theory. As a consequence these notions are subjected to an ad hoc synthesis. The problematic and limited character of the How and the What of pictures goes substantially unrecognised. Semiotics clears some of the fog but the cost is a practical epistemological vacuity—a system for picking out descriptive equivalences.

The conceptual problems of the relations of representation are not all over. The earnest hope of some realists that they are all over embraces a flight from reality. We assert that these relations are, crudely, of two sorts: ‘descriptive’ or denotative and genetic. These are two main links that a picture can have with reality. We further assert that the genetic ‘link’ is either neglected or fuzzed into description. This is a pity as it is a very powerful consideration when one is considering the problems of realism. This fast contention is problematic. We need to rummage around between pictures and representations and the world before we can go on to explore the consequences for the higher reaches.

Discussing names in connection with ‘simple’ pictures of one person David Kaplan6 considers a fundamental relation between a name and an object:

"What we are after. . . is a three part relation between Ralph, a name (which I here use in the broad sense of singular term) α, and a person x. For this purpose I will introduce two special notions: that of a name α being of x for Ralph, and that of a name being vivid, both of which I will compare with the notion of a name denoting x.

Let us begin by distinguishing the descriptive content of a name from the genetic character of the name as used by Ralph. The first goes to user-independent features of the name, the second to features of a particular user’s acquisition of certain beliefs involving the name. It is perhaps easiest to make the distinction in terms not of names but of pictures, with consideration limited to pictures which show a single person. Those features of a picture, in virtue of which we say it resembles or is a likeness of a particular person, comprise the picture’s descriptive content. The genetic character of a picture is determined by the causal chain of events leading to its production. In the case of photographs and portraits we say that the picture is of the person who was photographed or who sat for the portrait. The same relation presumably holds between a perception and the perceived object. This relation between picture and person clearly depends entirely on the genetic character of the picture. Without attempting a definition, we can say that for a picture to be of a person, the person must serve significantly in the causal chain leading to the picture’s production and also serve as object for the picture. The second clause is to prevent all of an artist’s paintings from being of the artist. will shortly say a bit more about how I understand this relation, which I designate with the italicized ‘of’.

The ‘user-independence’ of the descriptive content of a picture lies in the fact that ‘identical’ pictures, such as two prints made front a single negative. will resemble all the same persons. In this sense, the descriptive content of a picture is a function of what we might call the picture-type rather than the picture-token. The ‘user-dependent’ nature of the genetic character of a picture lies in the fact that ‘identical’ paintings can be such that they are of different persons (e.g. twins sitting separately for portraits). Thus the genetic character of a picture is a function only of the picture-token. In order to accommodate genesis. I use ‘picture’ throughout in the sense of ‘picture-token’."

Those features of a picture according to which we are able (under certain conditions) to see it ‘resembling’ a person or, etc., comprise (under certain conditions) the descriptive content of the picture, although these features are in general neither necessary nor sufficient for descriptive or representational content. ( For some people these pictures will be seen as comprising the ‘representational content’.)

It seems that resemblance comprises the descriptive content of the picture but only with considerable modifications, under some conditions. Resemblance is relative to practice and to other resemblances and cultures, and it is variable in time, etc. A judgement of ‘high descriptive content’, say, will not necessarily follow upon normal judgements of resemblance. What we could perhaps say is that among pictures that represent actual objects (or real objects) and within certain cultural constraints (and for pictures, though possibly not for texts, notably within a legacy of nineteenth-century cultural constraints), those features in virtue of which we might say that a picture resembles or is in some sense iconically tied to an object or scene comprise its descriptive content.

The convention (pace Kaplan) concerning the genetic sense of ‘of’ entails that irrespective of questions of what it is iconically connected to, a picture is of what it is genetically connected to. This may be full of intuitive fissures, but it is still very powerful as an instrument for keeping your eye on the material character of pictures as produced. It may be no more than that. It seems intuitively obvious that the genetic history of a picture can determine its iconic (etc.) features and not vice versa. What is not very clear is what are to be counted as the constituents of this genetic history—exactly how a picture’s genetic link with reality is to be circumscribed.

A picture that is an icon of Attilla the Hun may not have been generated by Attilla the Hun in the genetic sense of of. Considerations of type and token aside, this statement is fraught with difficulty on an almost theological scale. A picture that was generated by Attilla the Hun may have been generated via hearsay, fantasy and decay and, as a consequence or partial consequence, be iconically connected to Duane Bobick. How feeble could the causal chain become before Duane Bobick started to get genetically or at least quasi-genetically significant without abandoning or seriously mutilating our present conventions concerning genetic of? There are a lot of homework-type problems associated with this and similar or contrasting questions.

It seems you can’t sort out ‘description’ without reference both to cause and to use. A description characterises something to some end. Because the causal chains between object and representation can be very complex and highly mediated by interests, concepts of use, etc., specification of the object of the characterisation in terms of a given representation of that object will need to involve a sense of the initial contingency of attributions of cause and of iconicity.

While it would be absurd to over-inflate ofness—to make a genetic object a unitary cause (agent or condition) of a picture—the notion of genesis directs our attention to the world, to the problem of material causation and not to the patrician intricacies of an idealised cultural coherence. Of course, the prospects for correspondence may be dim, but the concepts lurking in its logical penumbra are illuminating. When Chomsky smashed Skinner to bits he did not suggest that speech production was uncaused. The lack of glue between a sign and the world does not make the world (which is stratified and differentiated) causally insignificant; does not make causal connections an unimportant historical and epistemological problem. Of course, causation and a ‘correspondence theory’ are not necessarily connected; it’s just that some people who have been concerned with ‘correspondence’ have sustained some of the dialectical problems of genesis—not buried them in a bitty cultural algebra.

Genetic of-ness is followed by lemmas. Iconicity, of-ness and their family are notions which have their feet in empiricism. When we’ve tried a bit longer to deal with some of the puzzles consequent upon the attempt to preserve a schematic or crystalline account of relations (the lacunae) between genetic of-ness and iconicity it will be necessary for us or someone to examine them in relation to some of the powerful arguments of historical materialism. Among these upshots will be conditions of closure on genesis in iconic and non-iconic, symbolic and non-symbolic contexts and so on. A result of this examination should be the raising of such questions as to what extent the boundaries of icons and produced images and produced things of other kinds must collapse or to what extent must iconicity be in the margins of all discussion of genesis, and so forth. This can not all be done at once. It will be appropriate to consider the matter of historical materialism and other closure conditions, other explanatory functions, against these schematic rehearsings in order to use these rehearsings as instruments to discover how it is that some closures on some genetic-type arguments and considerations and some closures on iconic and iconic-type arguments are to be defended or undefended within a broad realist proposal.

There is a dialectical relation between identification of genesis and identification of descriptive content. A strong sense of how a picture is of something can thus be expected to generate lemmas or anomalies (in a strict sense) with respect to identifications of or suppositions about its descriptive content. This, because of the comparatively greater material force and functionality of information about a picture’s genesis. This may seem to go against those intuitions which are secured in and by ‘on the street’ identifications of a picture’s iconic resemblance (to something). Nevertheless the nomological priority of genetic explanation and the notion of a dialectical relation between that and identification of descriptive content should be preserved as restraining conditions upon the operation and autonomy of idealist concepts of representation; and by this we mean that it is an important condition of any realistic criticism of representation that genesis be recognised in general as a more powerfully explanatory concept than resemblance.

With regard to Kaplan’s second ‘special notion’, that of ‘vividness’, it is highly likely that pictures vivid for someone or for some class or interest have a high plausibility for them. (Class naturalism? Bourgeois naturalism?) And vivid pictures might be expected to involve an ontological claim or be thought to involve some sort of ontological commitment (but not knowledge as such).7

It would be hard to think of a picture as vivid unless its being vivid involved some sort of ontological commitment to what it is thought to describe for those for whom it is vivid. This may be hard to take—just sort of stipulative. We may seem to be restricting vividness to an artificially small class of pictures—pictures that seem to involve an existence claim in respect of what they describe on the part of those for whom they are vivid.

Bearing in mind ‘vivid’, we return to of. Consider the following:

A hyper-realist artist H is going to have a show at a downtown gallery. The owner of the gallery, an entrepreneur of hyper-realist painting has informed the artist that the portrait he exhibited in the gallery artists’ Summer Show was the subject of enormous interest on the part of the illiterate plutocrats who patronise his gallery. Could he possibly do 10 more, just like it? H agrees. He photographs Bob, Carol, Ted, Alice, Jules, Jim, Tex, etc. with his perfectly functioning Hasselblad. His brilliant technician girlfriend develops the film and produces perfect 10“ x 8” colour prints. H sticks the prints in his epidiascope which enlarges the image to 10’ x 8’ and proceeds to airbrush and fudge a hyper-realist picture complete with enlarged pores, blackheads, eyelashes and hair-roots.

H’s ’picture of Bob’ etc., satisfies Kaplan’s criteria of of-ness, resemblance and vividness. At the same time it involves us in no undue metaphorical strain to say that H’s ’picture of Bob’ etc., is of no one, that it is only marginally iconic and that it is utterly lacking in vividness. The metaphor is not strained. We could say that it is almost to be taken literally.

We may be able to accept—in a commonsensical or culturally bound phenomenalistic way (?) the argument that H’s picture of Bob is a picture such that Bob played an important part in ‘the causal chain leading to its production’. We are nevertheless struck by the vacuity of the argument. In fact it seems to be the poor historiographical brother of ‘the First World War was caused by the assassination of the Arch Duke Ferdinand at Sarajevo’. It is a vastly inferior argument to account for the picture when compared with the statement that the picture was produced—materially caused—by other hyper-realist pictures, by the pressure of the hyper-realist market, and that these material causes do not involve reference to Bob in any significant sense.

The point about the notion of genesis (not the only point, but it must be important) must be the question of the explanatory power of an account of the production of something. For Kaplan, at first sight, explanatory power may not seem to be of the essence. This is a pity. If you are considering how you got a name you could very well be explaining to yourself how you acquired it.

It is perhaps interesting to consider Kaplan’s of in relation to W. B. Gallie’s ‘genesis.’ According to William Dray:

‘In his general account of genetic explanation, Professor W. B. Gallie says . . . if a historian explains the rapid rise of Christianity by referring to its possession of the proselytizing platform of the Jewish synagogue . . . this does not commit him to arguing that the development was either necessary or probable. The force of the explanation is rather to show how Christianity got its opportunity. An everyday example of the same pattern would be the explanation of an angry retort by reference to the taunt which evoked it. Once again, there is no claim to show the retort to be deducible or predictable. It is rather that but for the taunt, the statement would remain unintelligible in the sense of lacking an appropriate historical context.’

Now, it may seem that we could substitute ‘Bob’ for ‘the taunt’ and ‘Hyper-realist picture’ for ‘the retort’ in the passage above. This will not however lead to a satisfactory result. We would argue that in order to get a satisfactory result what would have to be substituted for ‘the taunt’ in the above passage is a lengthy sentence concerning the interests of art gallery, art money and art market. Thus:

‘An everyday example of the same pattern would be the explanation of a hyper-realist picture by reference to the art market (etc.) which evoked it. Once again, there is no claim to show the Hyper-realist picture to be deducible or predictable. It is rather that but for the art market (etc.) the hyper-realist picture would remain unintelligible in the sense of lacking an appropriate historical context.’

What would artistic pictures be like if a Kaplan sense of the genesis of such pictures were not only true but important as an explanation—contributed something to a sense of material cause?

A possible criticism is that anyone who conscripts Kaplan’s of-ness into art discourse is trying to make some sort of glue with which to stick (some) signs to the world and vice-versa. Not many of these objections are easily met. The problems come thick and fast when it is remembered that of-ness is being considered not because of its psychological interest but primarily because of its historical and historiographical explanatory interest. It has to do with how something was produced.

We are suggesting, in adapting of-ness, that it is a special sort of restriction in arguments concerning the production of certain types of pictures or representations. This restriction is genetic. Of-ness has genetic significance but is not genetically sufficient. Genetic considerations are, as we have earlier mentioned, in some clear and vulgar ways distinct from considerations of (e.g.) iconicity and isomorphicness. The defensibility of a particular sort of genetic argument in analysis distinguishes some kinds of art production from other kinds. (And to put what we keep saying another way, we pursue this conjecture on the near conviction that the semiological self-restriction of many earnests of ‘realism’ has them worrying about whether pictures are like something or other or nothing, whereas what they should be worrying about or asking is whether pictures are of something or other or nothing. If you ask the latter question the genetic debate is brought to the forefront and the anxious squeaks of those whose interests are served by the high-minded suppression of informal fallacies are stifled.)

The quasi-experimental character of some picture production is dependent upon the fact that the producer produces a sort of closure, that this closure is intelligible, that its intelligibility is rooted in some reality. Of-ness proposes and can be defended as a condition of such a closure.

A closure of a possibly ‘stipulative’ but not antecedent kind is defensibly associated with of in instances of realistic pictures. That is, there may be a way to regard realistic pictures as, among other things, pictures whose genetic systems contain a ‘closed’, or rather partially closed system. We may be able to regard a realistic picture as a picture whose termini are that picture and that which it is of. Such a system is not really closed, rather it is differentiated from other non-realist art practice in that it is limited or restricted in some way that is historically or cognitively significant, and riot just structurally or because of an investment in dominance.

This is not to presume that human agency is not an important genetic question. What of-ness does is introduce a closure in the form of a conjecture and a set of lemmas as to what you can go on and say about agency—what that agency is like. (A greedy painter’s painting of something is perhaps not the same sort of genetic object as a greedy painter’s painting of nothing.) Because of the ‘realist’ conjecture above and its lemmas, questions concerning agency may well differ in cognitively dramatic ways.

Whether or not of-ness distinguishes a particular kind of ‘mechanism’, as we appear to be suggesting, is far from clear. If we are suggesting such a mechanism, to say that a picture p is of X is not to say that no other presumably contrastive predicates may be simultaneously applicable. Of-ness does not suggest some variant of action by contact thought not to require explanation. The causal agent in the genetic sequence is the artist or artists, men and women as producers of produced production. The notion of producer is better understood than the notion of a producer of pictures of something.

Another objection to of-ness is that it can only go to features of a very restricted class of pictures. Think of this as a theoretical argument in the sense that what is being objected to is the casual implication that there is some intelligibly equivalent Markovian-ness to be found in very long and attenuated genetic chains and in very short and apparently simple ones. That is, it may be thoroughly objectionable to suggest that a relation of of-ness established between a photographic print and a depicted object (where e.g., mediation and so forth are not all that complicated) is significantly like are relation of of-ness thought to be reconstructable in a very long, complex and highly mediated chain. Such an objection to of-ness amounts to saying that while it may be sort of commonsense to think of some very simple pictures as epiphenomena of the ‘actual objects’ of the world, it makes no sense to think this way of most pictures which are genetically the result of human activity, competences, cognitivity etc. It’s not obvious whether this objection concerns some sort of implicit phenomenalism of of-ness. To accuse of-ness of phenomenalism may not be to say that it is genetically insignificant—although it would be if you thought that phenomenalism was incapable of accounting for anything. The objection may not be so much an accusation of phenomenalism but a suggestion that of-ness can’t account for anything non-phenomenalistic— which is different.

Of is a link in stipulative association of a picture and some part of the world. Of does not split up as a notion just because some pictures are of some part of the world though this stipulative association does not seem to express one of their links with reality, while other pictures are of some part of the world and this stipulative association does seem to express one of their links with reality. Indeed, the relevance of this very distinction is a pointer to the critical explanatory potential of our concept of of.

Of is always potentially marginal or genetically vacuous. What distinguishes some representational picturers from others is the genetic significance of of-type relations in accounting for their character.

Of handles the possibility that two formally and, at a certain level, psychologically identical pictures could have different genetic links with reality. Of course, pictures of the same thing can have different genetic character, etc., pictures of nothing could differ genetically, etc. Of-ness is a link that a picture can have with reality. It can be suggested that it is a link that any realist picture must have. In order to point out the difference between a descriptive or iconic link of of-ness and a genetic link of of-ness diagrammatic examples might well be used. Such diagrammatic examples would no doubt make ‘a picture of’ seem like ‘based on a perception of’. This is fraught with anxieties about practical redundancy and logical collapse (as we have seen). Of’s apparent reducability to some form of phenomenalism is the problem.

Picture production is not passive perception-copying. It must be clear that perception of or mechanical conjunction of object and picture (or picturer) is by no means genetically sufficient for any recognisable pictures. It was never supposed to be. Perception of is a sort of genetic closure. Of can comprehend that virtual absence of the ‘object’ which is involved in a defensible sense of picture genesis. By ‘virtual absence’ we mean the following: If Courbet’s The Stonebreakers is of actual people then that is epistemologically significant. The painting could have been based on a few sketches, a verbal description full of perceptions of the class character of society, produced by his auntie. If his auntie’s sketches and description are of the two men (that is if the two men and Auntie are in some epistemologically significant perceptual (and ratiocinative) connection, and Gustave’s picture is based on them, then it is of the two men. If not, the picture can’t do any better than Auntie’s fantasy, hallucination, fiction or mistake. Gustave’s picture derives its link with reality front the genetic credentials of Auntie’s sketch and description. This implies you can have a sense of of-ness highly mediated.

The ambiguity of the more or less conventional phrase ‘plays a significant part in the (causal) genesis of the picture’ can be thought to permit some sense of this ‘virtual absence’. By virtual absence we mean to emphasise the fact that pictures have a mediated genesis. We mean to bury the idea that the epistemological significance of of denotes the epiphenomenal-or-worse picture. ‘Mediation’ is not here supposed to throw fancy intellectual Lukacsian (or etc.) idealist notions into the discussion. By talking of a ‘virtual absence’ of the object we mean to point to the fact that pictures are produced production and that within the of-ness relation between a picture and an object other pictures and descriptions can be accommodated.

We have tended to dwell on the problem of a very schematic of. The example of Courbet’s Auntie above proposes that of can be sustained in highly mediated genetic circumstances. Genetic credentials and the derivation of genetic credentials are of great epistemological significance. If The Stonebreakers was transported without comment to Southern California it could denote two Californian artistic persons collecting stones for their crib of a Richard Long. The painting’s realistic status as having something to do with knowledge can only be extracted from its genetic attachment to some real peasants—whether it naturalistically denotes them or not.

If Courbet heard snatches of a report about a burial at Ornans and that’s all he had to go on then (some of) the genesis of Burial at Ornans could be traced back from mouth to mouth or from ‘mouth to eye’ to the actual burial (eventually). This is not at all surprising. This tracing may seem to have very little to do with the significance of Burial at Ornans. But if Burial at Ornans is supposed to have anything to do with the class struggle in France it will have to be linked to it by more than resemblance to people at a funeral plus the presumption that the descriptive features of the picture are somehow discursive enough to produce or to match knowledge. Of course, the picture may not be linked genetically to people at a funeral—it may be linked to a different event in the life of the country bourgeoisie, and the burial may be a kind of metaphor. There is nothing genetically odd about that. The picture is of what it is of. Alternatively, it may be ‘about’ the rural proletariat in a metaphor that ‘describes’ the country bourgeoisie. If it’s of the rural proletariat it has to satisfy of-ness conditions. ‘Serve as subject’ is not that simple. It is no less of what it is of when what it is of is not the subject of the picture. Title and subject do not have to coincide. Artist’s subject and subject may well be a matter of genetic investigation, etc. (We do not need to say that what a picture is of must also be the subject of the picture in order to stop pictures just being of the artist. We can either stipulate that they shall not be of the artist or else include the artist in the list of those—usually many—things which the picture is of, stipulating that there are no realist pictures which are genetically restricted to being of the artist.)

Someone might object that the suggestion that Courbet’s Stonebreakers is of artistic persons making a silly Richard Long sculpture is itself so silly, so lacking in plausibility, as to be able to mark no boundary between iconic ‘naturalism’ and of-ness. The objection has a certain merit—but not much. It is perhaps more genetically significant to say that Courbet was French, that he lived at a certain time and that the genesis of the picture is what was happening at the time he lived and his perception of it and his perception of truths about it. Maybe. But the production of certain resemblances by Courbet will not link his picture with reality genetically. We cannot infer realism from resemblance. Neither can we infer realism from the extruded heroic personality. The assertion that Courbet held such and such a social position or positions, that he had such and such a grasp on die character of society, and that these views were true and were knowledge as a sort of consequence, cannot itself be true unless these views and the pictures which formed and were formed by them are genetically linked to a world which is independent of Courbet and his pictures.

Unless ‘realistic’ pictures have of-ness as at least one of their properties, no decisive move can be made against a theory of mere pictorial coherence. To fail to pay attention to the genetic credentials of Courbet’s pictures is to extrude a mysterious hero. To extrude the hero is to minimise or to obliterate the genetic (epistemological) credentials of his pictures.

A picture of E.P. Thompson is genetically tied to E.P. Thompson. This genetic tie is with a specific actual entity. For a picture to be tied thus to E.P. Thompson is not have to have this or that genetic history predictable of it or glued to it by necessity. It is, rather, to say that it has a link with the actual E.P. Thompson such that the well rehearsed distinction between what is produced and how it is produced is weakened: E.P. Thompson stands as a significant contributor to How? ‘How’ must conceptually penetrate ‘What’ if representation is to be intelligible at all beyond the vagaries of coherence theory. Indeed, it might be argued that unless the genetic link with reality is supplied initially by something like of-ness then all realistic projects must be unintelligible. If of-ness were not a powerful link between a picture and reality, if it were merely a fifth-rate genetic explanatory device, then realist projects would, it might be conjectured, degenerate as modernist art projects have degenerated, into super-mediated art-world projects susceptible to no intelligible closures, susceptible only to the arbitrary closures which are functions of the instrumental apologetics of the rest of the bourgeois cultural creation.

Art & Language 1979


1. Hypothetical cases raised, generally in a spirit of irony, in earlier Art & Language writings seem too often to have been interpreted as claims (to some kind of wonderful radical art practice perhaps). Some curatorial ‘mistakes’ are self-serving no doubt. See e.g. the Introduction to Art-Language Volume I Number 1, May 1959 and ‘De Legibus Naturae’ in Studio International, April 1971.

2. Marx, The Eighteenth Brumaire, in Selected Works of Marx and Engels, London 1968, p 138.

3. Marx and Engels Manifesto of the Communist Party. Peking 1965, p 46.

4. Read at the British Sociological Association conference in April 1978 and printed in Radical Philosophy no. 23, Winter 1979.

5. Current and some back issues of Art-Language may be ordered from Printed Matter, 7 Lispenard Street, New York. N.Y., 10013 and Arts Bibliographic, 37 Great Russell Street, London WCI.

6. ‘Quantifying In’, first published in Words & Objections: essays on the work of W. V. Quine, Davidson D. & Hintikka J. (eds.), 1969, Dordrecht-Holland, D. Reidel Publishing Co. pp 178–214. Reprinted in Reference and Modality, Linsky L. (ed.) 1971, Oxford Readings in Philosophy Series, London O.U.P. pp 112–144. References are to the latter edition.

7. Kaplan loc. cit. p 136.

8. W. B. Gallic ‘Explanation in History and the Genetic Sciences’ in Theories of History, Glencoe III. 1959; See William Dray, ‘The Historical Explanation of Actions Reconsidered’ in The Philosophy of History, Oxford Readings in Philosophy, London 1974.