Lovers of poetry, including those who enter into the fascinating arguments of literary criticism, should be neither startled nor alarmed by Professor Simon's conclusion that "criticism can be viewed simply as a branch of cognitive science." Such a hope is at least as old as Aristotle's Poetics. More recently, in the early 20th century, the literary critic I. A. Richards took such an approach. (Indeed, Professor Simon might well have borrowed for the title of his talk the title used by Richards and the linguist C. K. Ogden for their collaborative book: The Meaning of Meaning: A Study of the Influence of Language Upon Thought and of the Science of Symbolism.) Like Professor Simon, Richards wished to bring together the two cultures of science and art. Like Simon, Richards wished to reach an understanding of poetry through an understanding of the neurological mechanisms of mind. Richards too was frustrated with the clamor of competing schools of literary criticism and put forward a cognitive vocabulary to quiet the noise. And no doubt the chief objection to Richards' approach-that the stuff of poetry is ultimately too delicate, too human, for such a mechanical approach-will be voiced against Simon as well, but with luck this will not be the end of our response to Simon's call for a cognitive approach to literature. Though Professor Simon perhaps ventures down one cul-de-sac, he also suggests many useful avenues for future thought.
Given the critical concerns of the last twenty years, it is not surprising that Professor Simon (unlike Richards, who was chiefly concerned with the value of poetry) focuses his thoughts on the referentiality of language. The cognitive vocabulary Professor Simon develops-intention, context, recognition, evocation, schema-does much to dispel the fog that has come to enshroud language referentiality in recent literary criticism. In demonstrating the importance of both intention and context for recognition and evocation, for instance, Professor Simon's vocabulary can help lay to rest the old dogmas of the New Criticism most often associated with Wimsatt and Beardsley, dogmas that dismissed both the author (by enjoining against the "intentional fallacy") and the reader (by likewise enjoining against the "affective fallacy") from consideration of poetic meaning. Indeed, Professor Simon's cognitive emphasis on context, recognition, and evocation provides important theoretical support not only for Reader-Response and New Historical approaches to literature, but also for such old-fashioned approaches as biographical criticism and the literary essay. More scientifically, his demonstration of the differing nature of context at the beginning of Stendahl's The Charterhouse of Parma, where context is provided by the reader's knowledge of the world, and at the end of the novel, where it is provided by the words of the novel themselves, renews Vygotsky's observation of the change in meaning of the title words of Dead Souls between the beginning of Gogol's story and its end, a change effected by the words of the story taken altogether (Vygotsky, 1962: 147-148). Professor Simon's notion of the world as a text in our heads both rescues and explodes Harold Bloom's insistence that poems can only be about other poems (Bloom, 1975: 70). His observation that different cultural backgrounds produce different mental schemas, which provide contexts for both writing and reading, drains much of the ideological fury from the debate on various "centrisms," while at the same time pointing to the importance of the debate itself. And perhaps most needfully, Professor Simon's description of ISAAC's success in translating words into non-verbal pictorial representations should help put an end to the post-structuralist/deconstructive frettings over the relationship of the word "tree" (for instance) to all those things in the world we call trees.
I am not convinced, however, that the artificial mind of the ISAAC program is necessary for this demonstration. (I've often argued, for example, that studies in the referentiality of language should begin not with Derrida's Of Grammatology or Saussure's Course in General Linguistics but with a cookbook or telephone book-real cakes do get made from verbal instructions; actual individuals are rung up.) Neither am I convinced that computer embodied artificial minds-AI-can tell us very much about how emotions are evoked by words, much less about how these emotions are transformed by the formal properties of poetic language into texts essential to the preservation of our humanity. Though the day may come when a computer, without being programmed specifically to do so, will nevertheless produce poetry in response to its own internal needs, even despite programming to the contrary (poetry eagerly sought out by some fellow members of its computer community, though perhaps ignored or rejected by others), such a day seems yet far off. But cognitive science is not solely, or even primarily, based on AI and its machines; there are already some five billion or so of the wet machines which do produce poetry out walking around in the world and thus available to provide insight into the workings of the human mind. Neurology, physiology, evolutionary biology, non-Freudian psychology, linguistics, anthropology, archaeology, comparative zoology-all these cognitive fields have much to tell us about the human capacity (and need) for poetry and art and music. The development of a cognitive approach to literary criticism-the project of Aristotle and of I. A. Richards and of Herbert Simon-has much yet to accomplish. We needn't wait for artificial minds to come into being (though imagining them and experimenting with their construction is always fun and often instructive) for the work to proceed.