part 4
Herbert Simon
If we wish to grapple with the intensions of a writer, then we must somehow determine the context that was evoked and present at the time of writing. First, we may focus on the text itself and infer local context by what preceded it and, to a lesser extent, by what followed it. Of course, we cannot assume that context remains constant, even over short segments of text, but we may expect it usually to change in a gradual manner, so that elements of context in one sentence or passage are likely to remain part of the context in closely adjoining parts of the text. Thus, in our quotation from Camus, we are probably safe in assuming that the locale, a bar in a Dutch city, and the bartender will remain part of the context throughout some substantial segment of text. The conditions under which, and the ways in which a writer shifts from one context to another constitute an important element in that writer's style.
A second way in which we may try to infer context is by determining what kinds of ideas and information are in the head of the writer at the time of writing. Here, we may draw on biographical material but also on information about the culture in which the writing took place and the particular social stratum and role in that culture the writer inhabited. We may discover what the writer read or might have read.
To understand Proust, we may wish to supplement his fictional autobiography with other biographical information. We may wish to understand French society at the turn of the century, the Dreyfus trials and their impact on that society, and the special position of upper-class Jewish families. We may seek information about the French aristocracy and the haute bourgoisie and the relations between the Bourbon and Empire aristocrats.
All these bits of information, and many others, are relevant components of the context that was present in some form in Proust's memory, that was evoked from time to time while he was writing and that helped to form his words. It is probably easier to survey the totality of the knowledge to which he had access than to assess what part of his memory store provided the actual context for any passage in his novel.
By this game of inferring contexts are innumerable books and theses made: Joyce's Dublin, Shakespeare's literary sources, poems of chivalry at the time of Cervantes, the social structure of Russian feudal estates in the nineteenth century. And the same processes we use to establish writers' contexts can be used to infer readers' contexts and, thereby, to assess how the same text may receive different readings in various eras and in particular lands. In this way we can come to understand the various meanings that can be attached (by Chinese students, by Chinese peasants, by foreigners of various nationalities) to the Statue of Liberty in Tiananmen Square or to the family intrigues recounted in The Dream of the Red Chamber.
In language, certain kinds of relations among meanings, and corresponding relations among the words and sentences that denote them, occur very frequently. It then becomes convenient to have definite conventions for representing these relations. These conventions constitute the syntax of the language.
Thus, in every language we often have occasion to speak or write about relations between pairs of things (ARB), where A and B are names of the things, and R denotes a relation. So we have particles like prepositions (or postpositions, as in Japanese) that have little semantic meaning associated with them but that signal a relation of some sort between two terms. Every language has such conventions for the most common situations that it needs to represent, and while the conventions themselves are different for different languages, there is great overlap in the sorts of things that possess syntactical conventions in different languages (i.e., agents, patients, temporal relations, instruments, number, determiners).
Thus syntax becomes one of the tools we use to discover meanings, but meaning itself is a matter of semantics, and syntax is (for writer, reader, or critic) no more than an efficient signaling device that sometimes makes the expression of meaning more expeditious.
Many of the meanings, especially of objects, events, and scenes, that are evoked by reading or in other ways take the form of mental images. We seem to "see" what is denoted by the text or evoked by it from memory, although the sight may seldom be as specific or vivid as scenes that actually pass before our eyes.
The first of Wallace Stevens' "Thirteen Ways of Looking at a Blackbird" goes like this:
Among twenty snowy mountains
The only moving thing
Was the eye of the blackbird.
It is hard (at least for me) to read these lines without seeing the high-perched blackbird and its sharp darting eye framed by the range of mountains. It is no more surprising that we experience as mental images our encodings of words we have read or of memories we have recovered than that we experience as mental images our encodings of the quanta of light that fall on our retinas in the presence of an actual scene. In both cases the mental image is neither words nor light but a symbolic coding of the initial stimulus that presents itself for interpretation in the mind's eye.
The most common explanation of mental images in cognitive science today, and the one I shall accept here, is that such images, whether generated from sensations or memories, make use of some of the same neuronal equipment that is used for displaying or representing the images of perceptually recorded scenes--hence the near identity of the ways in which we experience them subjectively. (Kosslyn, 1980). Both actual scenes and remembered scenes are viewed in the mind's eye.
On this hypothesis, a mental picture formed by retrieving some information from memory or by visualizing the meaning of a spoken or written paragraph is stored in the same brain tissue and acted on by the same mental processes as the picture recorded by the eyes. That one can translate memories or words into a mental picture is no more (and no less) mysterious than that one can similarly translate the impact of photons on the retina.
The meanings we have discussed so far are meanings without emotion: they are the stuff of cold cognition rather than hot cognition. But words are quite as potent in evoking feelings as they are in evoking thoughts. It is time (perhaps it is past time) to introduce emotion into the conversation. We do not have to propose a formal definition of emotion; it is enough that we have all amply experienced it. In particular, we have experienced it while reading or listening to words.
The evoking mechanism, which we have already examined, provides the means for arousing emotions as well as retrieving ideas. Among the memories that we retrieve, some are deeply dyed with happiness or sadness, with pity or envy, with love or hate, with fear or longing. Remembering recovers for us not only facts and events and ideas but also the feelings that have become associated with them. Both Plato and Aristotle regard the evocation of emotion as a central part of esthetic experience.
In the language of neuropsychology, emotion is felt when evoking a memory stimulates the autonomic nervous system in the same way that an experience might stimulate it. Just as visual scenes can be re-displayed in the mind's eye, so emotions can be re-experienced in the "mind's heart," which we are told has its locus in that autonomic system.
There is no good reason to restrict the idea of "meaning" to the cold component of cognition. It makes perfectly good sense, by virtue of the way in which emotion is aroused, to speak of the emotional meaning of a text. And what has already been said here about meanings, the ways in which they are organized in memory and the ways in which they are evoked, applies equally well to emotional meanings.
Absolute music reminds us that emotion is best aroused not by talking about happiness or sadness but by presenting stimuli that produce feelings of happiness or sadness. It is still quite unclear what it is about "sad" music that is sad (or sadness-producing) or about "happy" music that is happy. The nearest approach, perhaps, to a theory of how music produces emotion is the tension-building, tension-release cycle proposed by Hindemith and others.
In literature, as in music, emotion is usually evoked not by using words like "sad" or "happy" but by creating situations to which we respond with sadness or happiness. Some of the basic ways of accomplishing this are obvious enough. Create a character, describe the character's behavior in such a way as to secure readers' empathy with him or her, and expose the character to events of the kind that produce emotions in those experiencing them and those observing them. If this is done skillfully, one may expect readers to respond as they would to the corresponding situations in real life.
The intertwining of emotion with the other dimensions of meaning is already illustrated by the examples I introduced earlier. In the brief passages I quoted from Stendhal, Camus, and Wallace Stevens, we read descriptions of scenes, but descriptions that, even in the first sentences, evoke a mood: the exhilaration of conquest across the Alps, the foreboding gloom of a cheap bar, the loneliness of a vast landscape.
In the discussion that follows, then, we will be concerned with a full-bodied cognition that encompasses emotion as fully as thought.
With this very general explication of the meanings of "meaning" before us, we are ready to look at the relation between an author's meanings and the words he or she writes down. Before an author writes "thunderstorm," something must have occurred to evoke that word from memory. In my case, I paused to look out of my office window, saw a black sky, and thought "thunderstorm." Now in fact, I don't believe a thunderstorm is coming--it's the wrong season--but I was reminded of one. In this case, a rather superficial, even accidental, meaning. "It doesn't mean much," we might say.
Or, if Freudianly inclined, we might ask, "Why does seeing a black sky remind him of a thunderstorm?" The latter question might be difficult or impossible to answer, but it is certainly legitimate to ask it. It is no less than the question of the great tangle of associated meanings that fill my cranium and how they came to be there and to arrange themselves in just the particular organization they now maintain. It is the question of my biography, my life--my "identity," to use that tired word.
A richer meaning might be more complex, yet easier to understand, than an isolated "thunderstorm." In the previous sentence, I used the word "meaning." What did I intend by it? To discover my intension, you don't need to speculate quite as wildly as you did for "thunderstorm." I have been writing about meaning for many pages and have even offered a rather explicit definition of the term. A good guess would be that its meaning in this paragraph corresponds to my definition. We could probably even rephrase this guessing process as a rather reliable heuristic for determining authors' meanings in contexts where they had just defined a term explicitly. We would normally object to an author's defining a term and then using it in a sense that did not fit the definition.
Authors' meanings, then, are to be assigned by discovering what was actually evoked in their minds to form their words. By attributing consistency of meaning to them over intervals of time, we can use context to help us narrow the ambiguity of meanings. But we must not carry the consistency criterion too far. Authors, like all mortals, are fully capable of inconsistency. Since only a small fragment of the potential meaning of a term may be evoked on any occasion of use, different fragments--possibly almost totally different ones--may be evoked on different occasions, with or without an author's awareness or intent. Moreover, the meanings evoked in authors' minds by reading things they have written (even have just written) need not be the same as the meanings that evoked the writing.
On the other side of this coin, as long as we retain the context in which we wrote or spoke some words, it may be difficult for us to recognize the different meanings (derived from other contexts) that others may extract from our words. That is why we must often allow our texts to grow cold before we undertake to revise or edit them. The cold text may evoke a quite different context, hence different meaning, from the same text at the moment of composition. Then we see how we may be misunderstood.