Herbert Simon's "Literary Criticism: A Cognitive Approach" is a courageous foray into interdisciplinary research. He discusses complex literary texts with an obvious ease and cogent breadth of culture. I believe, however, that Professor Simon is unduly optimistic about the possibility of bridging the gulf between the culture of science and the culture of art. This gulf is not a mere side effect of over-specialization or of different human temperaments, however much these factors may contribute to it. I would go so far as to say that the sciences and the arts, while sharing certain generalized methodological conceptions, have developed incommensurable, even mutually canceling, domains.
There is, it appears, a need in cognitive science which something like literary criticism might fill, but I can only say that what Simon calls for is not very much like literary criticism. Although literary critics can, and sometimes, coincidentally, have engaged the questions which the research program of cognitive science requires, these questions lead quickly into sterile formalizing. Moreover, Professor Simon's unexamined assumption that Cognitive Science provides a foundation for literary study seems to me simply mistaken.
Although there are certain shared grounds, relating most notably to theories of representation and formal systems of symbol manipulation, on which the philosopher of cognitive science and the literary theorist are both comfortable, the scientist and the poet are figures of an unresolvable dualism. For one, meaning has to do with symbolic exchange, the return of a symbol for a symbol, and for the other, meaning has to do with the destruction of the symbol system altogether and its replacement with the experience of value. Literary art is only incidentally representational; its processes are only incidentally involved with information processing.
Until after the beginning of the twentieth century, logic texts commonly began with a distinction between existential judgment and spiritual judgment, the one having to do with the determination of fact and the other having to do with the determination of value. The words in a poem and the same words in a scientific report may belong to the same language, but they have utterly different significances. Their identity is an illusion. Wordsworth, for example, notes a puddle of water three feet long by four feet wide. In the poem, he is not making an existential judgment. The passage is not about a puddle, real or imagined, it is proposing the value of a particular mode of attention which is interested in particularity rather than generality. Its meaning has not to do with an isolatable and specific existential origin but with its workings in the whole of a human imagination in a cosmic context. To put the matter differently, a poem communicates no information; it does not reduce uncertainty. Indeed, by taking a creation of imagination seriously, it substitutes a world where things are important or unimportant for a world where things are true or false.
Popular culture has developed to provide an all-receptive, alogical domain where rumors of what exists and what is important can be cooked up and decorated for general consumption, and it is the prevalence of this domain which makes an easy accommodation of literary criticism and cognitive science seem possible. The distinction between value and existence has broken down, and the homogeneous statistical world, with its three-cornered structure, closing on the popular media, the shopping mall, and the university, has been substituted for the prior dualism. This is the sublation of the most troublesome philosophic problem of the post-Cartesian West. History reaches its end not with the hoped-for apocalyptic fireworks, but with a chilling hub-bub of universal but empty exchange.
Intellectually, we are left stranded by this turn of events. Literary criticism seems muddled to Simon, in part because many of its practitioners are muddled, but more fundamentally because he does not understand the question which literary critics are attempting to answer. The confusion has to do with issues deep in literary criticism itself, not with the failure of literary criticism to make the neat distinctions of cognitive science.
Simon proposes that modes of literary criticism be classified according to how they answer the question, "How is meaning attributed to the text?" The fact is that literary criticism rarely asks about the meaning of the texts its studies, and, if it does, it makes the same category mistake that Simon is making. When the new criticism appeared to challenge the old New Critical dictum, "the poem does not mean but be," it challenged not so much the denial of meaning as the assertion of being. The poem is not an object but an event. The poem does not mean but happens. The critic, following the event of the poem, is not an interpreter but a performer, following lines of intensity not because of their exchange value but for their inherent excitement, their interest in themselves, and the possible revelation of this excitement to others. The reading is not the consumption of an expression but an event of life value. To read a literary work is neither to posit a blank, known as an author, to which the reader attributes beliefs, nor to posit a subject or inner-reader, to which meanings are revealed (or by which meanings as mysterious, originless entities are constructed), but to enter concretely a community in which author, reader, text, other authors, readers, texts, and other things of every conceivable kind are constantly interacting physically and logically, valuing, mutually impinging, and so forth, and this totality is categorically different from the information which it helps to make manifest.
To be sure, meanings are exchanged in this process: programmed interpreters which exchange symbol for symbol are variously employed. But they are as trivial as any other isolated
element in this intimidatingly complex event. There is not a text as such but an assemblage. The organization is better described by reference to modern music than modern logic. Ezra Pound writes in his remarkable Treatise on Harmony,
A sound of any pitch, or any combination of such sounds, may be followed by a sound of any other pitch, or any other combination of such sounds, providing the time interval between them is properly gauged; and this is true for any series of sounds, chords or arpeggios.
Now we know that the same is true not only of sounds, chords, or arpeggios but also of facts, concepts, or logical categories. The organization of knowledge as value is constituted as arrays of events in time. The dead linguistic tokens are removed from their conceptual mechanisms and enter into the unique destiny of the reader where events are unrepeatable.
The scientific enterprise makes us responsible for what we believe. For this purpose, knowing what we mean is crucial. The other great human function is divided between ethics and aesthetics: it makes us responsible for what we feel. In this domain judgment addresses senses of moving totalities and sustainable acts of mind. One speaks of flows, vectors, dispositions, drifts, rhythms, arrays, dynamisms, vibrations, or measures more readily than meanings. These quantities are utterly unique, expressions of life moments of an individual's actual experience. They are not relevant to the computer programmer attempting to simulate generalized human intelligence. Human intelligence and the human consciousness which is its medium are not simulations. The argument between Simon and John Searle over "whether ISAAC understands what it is doing" is wrongly posed: the question is not whether consciousness is a necessary posit of intelligence. Consciousness explains nothing, consciousness is the thing which requires explanation. Given that we are conscious, it is not strange that we can play chess or do mathematics or even write poems. Whoever does these things can explain how it is done. Even writing poems, as Jonathan Swift observed, is a simple matter: one finds the right words and puts them in the right order. There will be, no doubt, statistically reliable algorithms for this kind of order (whether humans will find their products engaging is another matter). The peculiar thing about humans is that they are conscious in the first place.
If we are to develop a useful, interdisciplinary relationship between those working with literary forms and those working with computer simulation, it will be necessary to begin with the recognition that the language does not broadly translate from one discipline to the other. It will be necessary to find a common ground outside of both disciplines.1
1. In this connection, I might mention my recent book, The Poetics of the Common Knowledge (Byrd, 1993).