Previous Next Up Comments
SEHR, volume 4, issue 1: Bridging the Gap
Updated 8 April 1995

Editor's note: Professor Simon's article appears here in 5 parts. This is the first part.

literary criticism: a cognitive approach

part 1

Herbert Simon

--Je comprends mal ce texte--
--Laissez, laissez! Je trouve de belles choses. Il les tire de moi. . . .
Paul Valery, Instants

In this paper, I will be acting as an unabashed missionary for contemporary cognitive science, which is itself an amalgam of artificial intelligence, cognitive psychology, and linguistics, with a few other trace substances (e.g., anthropology, epistemology) thrown in. I will argue that cognitive science has reached a point in understanding human thinking where it can say a great deal about literary criticism; in particular, that it can cast some light on the theoretical foundations of criticism and even generate useful advice for its practice. But my position is not as asymmetric as these words would make it appear. Written texts, literary and other, provide a rich source of data for understanding cognition. Enormous thought goes into the production of texts and perhaps even more (given the ratio of readers to writers) into interpreting them. These data have not been much mined by cognitive scientists, who therefore have much to learn from literary criticism, which examines the texts in depth. Perhaps what I am attempting here should be viewed as a gesture from the cognitive side to repay a small part of the debt we owe to critics and theorists of criticism for introducing us to literary texts. The paper may also be viewed as an experiment in communication between the two cultures of the humanities and the sciences. I simply take for granted that, pace Leavis, there are two cultures, much as C. P. Snow (1959) described them thirty years ago, and that communication between them is infrequent and then, when it occurs, noisy. I also take for granted that it is important for our society that this communication be improved substantially.

Literary criticism concerns (among other things) the meanings of, in, and evoked by literary texts. Cognitive science concerns thinking, by people and computers, and extracting or evoking meanings while reading and writing requires thinking. Hence, there is a wide expanse of ground common to literary criticism and cognitive science. But a casual examination of leading books from the two domains suggests that each has little awareness of the other, or of the possible relevance of the other to its concerns. With rare exceptions, there is little or no cross-referencing.

It would be too strong to say that literary critics and theorists of literary criticism are ignorant of the social sciences. But although many of them know about Marx and Freud, fewer are acquainted with contemporary cognitive science. Only a few, like Siegfried Schmidt (1968) and Robert de Beaugrande (1980), are well versed in both literature and cognition.

It would also be too strong to say that all cognitive scientists are ignorant of literary criticism, but they certainly do not mention it often in their footnotes. Some researchers, like my colleagues John R. Hayes and Linda Flower (1980) and Patricia Carpenter and Marcel Just (1987), have studied the processes of writing and reading but have not extended their studies to works of literature. Some others, like Jean Mandler (1978) and Wendy Lehnert (1981), have analysed the "grammars" employed in the structures of stories. But the numbers of such scholars are few and the literature of the subject not large.

I am a cognitive scientist, not a literary critic or a theorist of literary criticism. So I have little choice but to start from the psychological side of the gulf in building the bridge between the two domains. I will undertake to sketch the thought processes involved in writing and reading the kinds of texts that we call literary.

But it is not my aspiration to create a new school of critical theory. Rather, I hope to cast some light on the relations among existing doctrines by reinterpreting them in a language that can lend to them a precision that they seldom seem to possess in contemporary literary discussion. Familiar terms like "meaning," "context," "evocation," "recognition," and "image" have gained a clarity from the researches of contemporary cognitive science that they did not have in earlier writing and still do not have in literary criticism and its theory. I will try to introduce some of that precision, divorced as far as possible from technicalities, into the discussion.

That will not be easy, for I will not be using the key terms in their ordinary senses, but in senses dependent upon a theoretical framework and formal language that I can set forth here only in broad outline. Focusing on the term "meaning" and how that term is interpreted in contemporary cognitive science will concentrate most of the technicalities and difficulties in one place. Much of the rest of the conversation can be carried on in ordinary language. If what I say sounds like common sense, so much the better. If it sounds only like common sense, then I have failed.

In recent generations there has been a great spawning of new theories and refurbished old theories of literary criticism. We have the New Criticism, now old, Structuralism, and Deconstruction. We have the New Historicism. We have text-centered theories, reader-centered theories, contextualist theories, and interactionist or constructionist ones. Some theorists place "political correctness" front and center. These do not begin to exhaust the list, but they may serve as examples.

A taxonomy of theories of literary criticism might derive from answers to the questions: How is meaning attributed to the text? Does criticism require us to ask what the author meant, what the text means, or what meaning derives from a reading of the text? Or, as some Deconstructionists would claim, does the text extend (have meanings?) beyond its meanings?

Some contemporary theories of criticism answer these questions by "All of the above." For, assuming that we have a theory of "meaning"--of the meaning of "meaning"--there is no reason why we should not explore what an author intended when in the process of writing down certain words, and explore what interpretations of that sequence of words are consistent with the syntax and semantics of the language (i.e., of the community that uses it), and ask what meanings various readers, with their various histories and experiences, are likely to extract from it. All of these seem to be wholly legitimate, if perhaps difficult, questions.

"We hold these truths to be self-evident. . . ." To understand what Jefferson and his colleagues meant, we would have to know (at least) the extent of their acquaintance with previous writings on political theory (Locke, for example), their beliefs about the philosophical bases for self-evidence, their understanding of and views on the political, economic, and social institutions of their times, their persuasive and rhetorical intent, and their knowledge of the audiences to which they were addressing the Declaration.

To understand how their contemporaries read these words, we would have to answer similar questions about them. The answers might be quite different for different contemporaries. The questions would have still different answers for people in the United States, or in Russia or China, who might read these words today. (Reflect on the meaning of the Statue of Liberty in Tiananmen Square in June of 1989 in this context.)

A literary inquiry into the Declaration of Independence might ask any or all of these questions. Alternatively, it could raise questions of style. Would it have been more or less cogent (or pleasing, or rigorous) had Jefferson begun: "We can safely assume from first principles that. . . ." or "It cannot be denied that. . . ."? These, and many others, would seem to be appropriate inquiries within the domain of literary criticism.

In fact, to a cognitive scientist it is not at all clear why there are schools of literary criticism. There are many things that can be said about a literary work, many standpoints from which it can be read. As long as we identify what we are doing, any one of these is as legitimate an enterprise as any other.[1] Of course, we may have special normative concerns. We might argue that one way of reading has certain personal or social values not possessed by another and might thereby prefer the former to the latter. Leaving aside such normative concerns for the present, we will let a hundred flowers bloom, but without requiring that the hundred schools contend.

To obtain answers to any of the questions I have raised about reading and criticism, we must assign a meaning to "meaning." Of course, it might be argued (perhaps has been argued) that literary criticism merely expresses undiscussable views, or the beliefs, feelings, or values of the critic. Accepting that argument would simply end the discussion without casting any light on the issues. Instead, I will proceed on the premise that critical opinions--at least opinions that do not just assign values but attribute meanings as well--can be more or less reasonable, hence are proper subjects for rational discourse.

To define "meaning," we can turn to philosophy, psychology, or that contemporary blend of those disciplines (seasoned by others as well) called cognitive science. I'll follow the latter course. Meanings flow from the intensions of people (or perhaps people and computers, a controversial issue). The people here might be authors, readers, or those somewhat amorphous clouds called "language communities."

I use "intension" in the broad sense in which this term is used in contemporary philosophy. It implies causality, but not necessarily conscious, deliberate will. In this sense, Oedipus, at least according to Freud, intended to kill his father. In the Greek interpretation, the causality of intension is replaced by fate or cosmic necessity (but a necessity that does not expiate guilt).

We might attribute intensionality to a pun if both meanings of the punning word were clearly stored in the memory of the writer, without arguing that the punning was a conscious act; for we might claim that it was caused by the stirrings of the two meanings in memory even without the author's awareness. But these subtleties of interpretation will not generally concern us here. I mention them simply to show how we can deal with cases where understanding how certain meanings crept into a text requires a knowledge of the author's intensions in this broader sense.

Nowhere has there been more attention to intension than in the interpretation of the American Constitution by the Supreme Court. What did the Framers intend? The huge and apparently irreducible ambiguities that this question discloses illustrate the difficulties in establishing intent or even defining it, but they do not make the question meaningless. Nor should we be misled by the legal aspects, which have a peculiar twist not present in all efforts to attribute intent. Often, the Court is not asking what the Framers did intend--what they had in mind in writing the words they put down--but what they would intend if their words were applied to new situations they had not anticipated and, therefore, did not have in mind at all. Of course we can play this same game with the laws of Hammurabi, but we do not usually need to.

For our purposes, it will be simplest to postulate intenders in the singular and to speak of particular persons who intend that particular strings of words should have particular meanings for particular audiences on particular occasions.[2] An assignment of meaning, then, will postulate a relation between a text and one or more persons. The same text may have (and usually will have) different meanings for different persons or for the same person at different times, and particular portions of a text may have multiple meanings even for a single person.

Meanings are evoked. When a reader attends to words in a text, certain symbols or symbol structures that are stored in that reader's memory come to awareness. (In psychology we might say, more ponderously, "having been noticed, the symbols are activated or transferred from long-term to short-term or immediate memory"). This is the sense in which we will use the term "evoke" throughout this paper. It denotes a specific set of psychological processes that have been much studied in the laboratory and in everyday life: the processes that bring meanings, or components of meaning, into attention.

The process that underlies evocation is recognition. Words in the text serve as cues. Being familiar (if they are not familiar, they will not convey meaning), they are recognized, and the act of recognition gives access to some of the information that has been stored in association with them--their meaning (Feigenbaum and Simon, 1984). Recognizing a word has the same effect as recognizing anything else (a friend on the street). Recognition accesses meaning.

Previous Next Up Comments


1. This point is made very cogently on page 9 of Jerry R. Hobbs's Literature and Cognition, a book to which this essay would be deeply indebted had I been acquainted with it when I delivered and then revised my 1990 Hitchcock Lectures.

2. This simplification does not in any way rule out the influence of the language community on the meanings an individual assigns to a text, but it recognizes that a writer or reader is an individual, not a language community.