Robert Nelson’s (2010: 167; 169) treatise on The Visual Language of Painting dwells on an analogy between painting and language, an analogy deemed ‘ill-considered’ by Richard Wollheim (1987: 181) for the way in which it ‘foists upon painting something akin to grammar’ with its array of syntactic and semantic requirements. But Nelson (2010: 178) pleads that ‘it is unfair to judge visual language by the prejudicial, logocentric criteria of verbal language,’ conceding nonetheless that this ‘seems like an almost anti-intellectual deflection of dialectic and intelligent responsibility.’ Nelson (2010: 181) insists that ‘the semantics are less important than the consciousness that they scramble for,’ and I would like to cast a sympathetic eye over his book and ask whether there is, after all, something valid to be gained from an analogy between painting and language, or whether he is indeed wading into dangerous territory.
To begin, let us consider his motivation for pursuing this analogy. Nelson (2010: x), familiar with the internal machinations of a fine art academy, opens his book with the premise that painting needs rescuing: ‘The only power that will resuscitate painting and give it long-term sustainability is language: verbal language that recognises visual language, the visual language proper to representational painting.’ Nelson is operating under the assumption that painting is stuck with words: whether it be artist statements, catalogue notes or doctoral theses, the written word clings to painting as a child to its mother’s skirt. In many ways, all this talk obscures painting itself. Our literacy perhaps hinders our visual attentiveness, our perceptiveness. Perhaps we can better learn to approach paintings, surmises Nelson, if we borrow this reassuringly familiar concept of language and describe what it could metaphorically mean in the visual realm of painting. Perhaps we could encourage a comparable ‘visual literacy’ in order to actually liberate painting from words. Words might then accompany painting, rather than smother it.
In no way does Nelson want to establish a strict framework for painting, then; he sees it already constricted by verbal language and he is seeking salvation by means of language. Language literally, in that we will write about painting and discuss it in words, and language metaphorically, in that we will apply the concept of language to the visual realm. This is a vastly different project to an analytical investigation of aesthetics.
Nelson’s metaphor leads him to consider what might be most salient about language, and to then import this trait into painting. Among the many uses and virtues of language, Nelson—like Tolstoy (in Wollheim 1980: 119)—gravitates most strongly towards its communicative possibilities. ‘Visuality,’ he explains of a term he uses interchangeably with ‘the visual language,’ ‘is implicitly a recognition of the visual as being recognisable and capable of transacting communications, a form of language, then, which presents the contemporary world with a certain urgency’ (Nelson, 2010: 167). The visual language is grounded in the intention of the artist and her desire to communicate to others. Not everything that is visible is in a language: rather, it is only when things seen make a ‘purposeful address to the eyes, [that] they become linguistic in character’ (Nelson, 2010: 168). And even when he considers the development and expression of individual thoughts by means of language, he insists on the status of language as a ‘social system’ which above all enables the ‘transaction’ and ‘recognition’ of those thoughts (Nelson, 2010: 176).
At this point we might pose a little interjection from Wollheim, who does in fact entertain the analogy between art and language, though he uses it to demonstrate flaws in what he calls the Ideal Theory of art, which I shall not consider here. First, he very cautiously asks the crucial question, ‘how are we to use the analogy?’ (Wollheim, 1996: 118). Then he firmly states that ‘a point is reached at which the analogy runs out’ (Wollheim, 1980: 137). Wollheim (1980: 137) notes that there is some discomfort at the idea of calling art ‘communicative,’ when it might be pitched precisely against language as ‘expressive.’ That is to say, art and language have different and incomparable functions. But he brushes this objection aside with the simple observation that ‘the theory that language is essentially concerned with communication of ideas is a dogmatic notion, which does not even take account of the variety of ways in which ideas are communicated’ (Wollheim 1980: 137).
There are three things we might say on this. First, Wollheim seems very sensible to caution that the analogy between language and painting might hold, but perhaps not unconditionally. It might only be relevant for demonstrating one point (such as the validity of the Ideal Theory), but we might push it too far if, for example, we demand an actual visual grammar. Should we want to wield this analogy, we must be very precise about why and how we are using it, and upfront about its limitations. Painting is not, after all, literally a language, and metaphors are poetic illustrations and not statements of logical identity.
Second, Wollheim is right to note that there are many important traits of language, possibly co-equal ones, and we might equally consider the way language functions as a medium for private thought, a tool of analysis, a descriptive record of information, a poetic mode of expression and so forth. Then we would need to ask whether it is legitimate to import all of these functions into painting, and why such a correspondence should hold. This motivation might have something to do with finding a lack of generosity in, for example, contemporary painting, in its persistent refusal to visually connect with its viewers, necessitating the dependence on actual text.
And third, Wollheim does something quite spectacular when he says language does not hold a monopoly over communication. And he is correct: ideas are conveyed in many ways, though the types of ideas may vary by medium. Much is conveyed through body language, for example, or diagrams, or music, or the extremely controlled movement of the body that we call dance. Certainly, a ballet does not communicate the same thoughts as a scientific report. But it can wordlessly transmit other ideas about the human condition. In fact, Wollheim’s dismissive observation invites us to think of many things as being at least partially analogous with language. Though perhaps what he really wants to emphasise is that communication and language are not identical, and that painting might be more closely analogous to some other mode of communication.
The way that Nelson uses the analogy between language and painting is rather loose and imprecise. He senses that there are limits, and thus centres all his comparisons on his chosen principle trait of communication, which is arguably the most obvious trait of language, without necessarily excluding the nuance of a perspective like Wollheim’s. As for why, Nelson is motivated by painting’s existing bondage to verbal language. He wants to defeat language with language.
Let us look at how Nelson conceives of the visual language of painting—how he applies the analogy. He is very quick to emphasise that the visual language of painting does not consist purely in technique, though technique turns out to be a fruitful way of investigating it. He argues that a ‘visual perspicacity and mental agility’ necessarily merge with the material when the visual language of painting is used effectively (Nelson, 2010: xi). This makes his position psychological rather in the way that Wollheim’s (1987: 22) is, because he necessarily incorporates the inner life of the artist—particularly her thoughts and intentions—into the manipulation of the medium. Keeping this attitude in mind, Nelson turns to the complicated and interlocking technical components of painting—each chapter dealing with the nebulous themes of colour, drawing, structure and composition, tone and plasticity, gesture, edge control and atmosphere, detail and weight, and layering and luminosity—in an effort to build up a rich tapestry of the visual language. Music, he concedes, lends itself much more easily to parallels with language, mostly because it is easier to compare them since it is easier to identify their ‘quantum units’ (Nelson, 2010: 170). It is easier to point to a note, a word, and compare—for example—their symbolic possibilities. Painting offers no obvious quantum unit; only a gently fluctuating integration of such elements as those listed above, and quite probably more. As such, painting ‘is harder to recognise as a language’ (Nelson, 2010: 172).
Here the analogy becomes a little hazy. Nelson (2010: xi; 170) starts dropping words like ‘symbol;’ right from the beginning he talks about the thoughts of the artist being ‘pictorially encoded’ in the medium. Again, we might defer to the thorough Wollheim (1980: 132): ‘The analogy… is one between art and language. The insistence is necessary: for there is another analogy, which bears a superficial resemblance to mine, and which may, deliberately or in error, be substituted for it. That is the analogy between art and a code.’ Wollheim (1980: 132) identifies two corresponding streams of thought, both of which he says lead into error: the first, that the more apt analogy holds between art and code than art and language; the second, that language and code ‘become so confused or transposed’ that the analogy slips, and ‘in point of fact it is to a code, not to language, that art is assimilated.’ In Robert Nelson’s case, the confusion is only slight, and his resistance of syntax likely saves him from labouring the idea of a code too far. But in the case of Nelson Goodman, the analogy undeniably shifts to code, and, as I see it, leaves the question of language and painting unresolved, and certainly not refuted.
Goodman, in Languages of Art, is cautious in his approach, starting with the theme of pictorial representation. He tentatively proposes an ‘analogy between pictorial representation and verbal description,’ that is, between one aspect of painting and one aspect of language (Goodman, 1976: 40). ‘The temptation is to call a system of depiction a language; but here I stop short,’ says Goodman (1976: 41), with immense restraint. And from here he embarks on a long, technical investigation of ‘what distinguishes representational from linguistic systems’ (Goodman, 1976: 41). He makes a fundamental division in the arts which leads him to focus on notation, and that division stems from the puzzle of forgeries: ‘in music, unlike painting, there is no such thing as a forgery of a known work’ (Goodman, 1976: 112).
Goodman identifies that there is something importantly different about ‘single’ and ‘multiple’ arts, and, explains Wollheim (in a supplementary essay to Art and its Objects, 1980: 167), ‘thinks that the more fundamental division within works of art is between the ‘autographic’ and the ‘allographic.’’ For an autographic work of art, the original and the copy are importantly distinct (Goodman, 1976: 113). It matters very much how the work came into being: its history of production—for whether it was or was not painted by, say, Van Dyck makes a spectacular difference. Each autographic work is a one-off. But we may perform a Chopin Nocturne without compromising the integrity of that work of art. ‘Thus painting is autographic, music nonautographic, or allographic’ (Goodman, 1976: 113). Wollheim’s (1980: 168) terms are a little less imposing: he simply speaks of ‘individuals’ (paintings), and ‘types’ and their ‘tokens’ (plays, books, musical pieces and their instances). Yet their categories are drawn up a little differently: Wollheim (1980: 167; 170) considers history of production across all art forms to be essential. Goodman (1976: 122), however, considers it only relevant to one-off works, simply as the means of identifying them. Allographic works, he argues, are severed from their creator and freed of their history of production, and because of that we need an alternative way to identify them: a notation.
‘Why,’ Goodman (1976: 121) asks, is the use of notation appropriate in some arts but not in others?’ In short, because some can already be identified by their history of production. The purpose of devising a notation is to ensure we are in fact encountering this specific work of art in one of its instances. And, further, they are necessary for the kinds of works of art that would overtax an individual: notations enable us ‘to transcend the limitations of time and the individual’ (Goodman, 1976: 121). By means of a score, a script, a manuscript, we can both reproduce and identify a work of art, divorced from its author (Goodman, 1976: 122).
And thus, on Goodman’s (1976: 121) terms, painting does not qualify as amenable to notation. But nor, he goes on to explain, do any of our natural languages qualify as notational systems (Goodman, 1976: 178). Goodman’s (1976: 225) analysis of musical scores (an arguably arbitrary choice which sets up the framework for his entire book) leads him to submit five semantic and syntactic requirements for a notational system; language is only able to meet the first two syntactic requirements. His analogy remains firmly between painting and code, and in rejecting any congruence between painting and notation, he does not reject possible parallels between painting and language. And I am quite happy to abandon this conception of a painterly notation (in the very literal sense of encoding meaning into a strict painterly grammar). As, I am sure, is Nelson, who takes great pains to describe the fluidity and unpredictability of his proposed visual language, while nevertheless insisting that painters hold fast to their communicative responsibility. And Wollheim (1980: 83) is unimpressed when he entertains the idea of notation permeating the entire range of the arts: ‘With such a notation there would no longer be any executant arts: the whole of the execution would have been anticipated in the notation.’ Painting, along with the other art forms, would collapse into a display of ‘mechanical skills’ (Wollheim, 1980: 84). Painting as notation is not an attractive analogy and is not the idea being put forward.
Yet Goodman (1976: 192) perseveres: ‘A sketch,’ (and by extension other autographic works of art like paintings) ‘is not in a language.’ No, he considers these works to be in entirely nonlinguistic systems, which ‘differ from languages … primarily through lack of differentiation—indeed through density (and consequent total absence of articulation)—of the symbol system’ (Goodman, 1976: 226). Painting would belong to a representational system, which must be dense, which one might imagine as continuous like an analogue gauge. Language does not represent, but describes; and descriptions, by contrast, are articulate, like discrete digital measures (Goodman, 1976: 230). Painting, Goodman (1976: 234) is trying to get us to understand, is too flexible, too nuanced, too direct (where it actually exemplifies the colour, shape, or feeling represented) to be subsumed under a structured, differentiated and abstracted system like language.
But to insist on the articulacy of language would be to discredit its qualities of subtlety and expression. Wollheim’s (1980: 135) level-headed remark brings that back into focus: ‘The elements or alphabet of a code are denumerable, whereas no precise limit can be set to the vocabulary of a language.’ Language itself has enough fluidity to perhaps rival the density of painting. And perhaps painting would profit from finding limits to its unbounded physical possibilities—perhaps this very limitlessness is what leads painting into incomprehensibility. Perhaps language teaches us (only by analogy) that if we want to capture meaning, to even make ourselves understood, we need to find common ground with our audience.
Language thus may still be nebulous enough a concept to import into painting, but we must ask ourselves seriously what for, and how far we are committed to drawing that analogy—and at what point it runs out. If Nelson’s reason for seizing a metaphorical visual language as a way to escape the tyranny of a limited verbal language in the current practice of painting is compelling, I do not see a reason against it, at least not from Goodman’s notation-focussed perspective. It only prompts us to consider whether other, better, analogies might exist—between, perhaps, painting and music, or painting and dance—and forces us to examine our obsessive preoccupation with language and our tendency to view it as the key to unlock all our problems. Literacy has swollen into the panacea of our age, but perhaps illegitimately so.
Goodman, Nelson. 1976. Languages of Art: An Approach to a Theory of Symbols. 2. ed., [Nachdr.]. Indianapolis, Ind.: Hackett.
Nelson, Robert. 2010. The Visual Language of Painting: An aesthetic analysis of representational technique. Melbourne: Australian Scholarly Publishing.
Wollheim, Richard. 1987. Painting as an Art. 1. publ. London: Thames and Hudson.
Wollheim, Richard. 1980. Art and Its Objects: With Six Supplementary Essays. Reprinted 2. ed. Cambridge: Cambridge University.