Structuralism and Literary Theory

So What’s all this Stuff About Structuralism and its Effects on Literary Theory?

John Webster

Note: Barthes' "The Death of the Author" was first published in 1967 (in English) and then again (in French) in 1968. It appeared a third time in a collection of Barthes' essays in 1977. It owes much to his efforts to extend the insights of structuralism to literary theory.

"There are only two or three human stories and they go on repeating themselves as fiercely as if they had never happened before."—Willa Cather

I. If "Structuralism" has had an effect on what we do in the reading of literature, what is it? What happens when one reads structurally? Basically, it means we do our best to transfer the principles of structuralist analysis in other areas to our work with literary texts. That means that we will want to look for “structures” in literary discourse by looking beyond (or below) surface features to find instead deeper and more abstract elements that underlie surfaces, and to establish both what those basic elements are and what the rules for their combination are.

What’s that mean? It all starts with concepts imported from twentieth century work with language. Consider the following basic, stripped-down “structuralist analysis” of English syntax. It offers a list of the elements (or “constituents”) of English sentences, along with rules English has for their combination:

S —> NP VP   [This means: any English sentence has two major constituents: a subject (NP) and a predicate (VP)—and nothing else.]

NP —> (det) (ADJ) N   [Any English subject has a noun in it, and MAY have either an article (det) or an adjective (ADJ), or both—and nothing else.]

VP —> (aux) V (NP) (ADV)*   [Any English predicate will have a verb, and MAY have auxiliary verbs, or an object (NP), or one or more adverbs (ADV), or all of those elements—and nothing else.]

To see what such an analysis can do linguistically, consider the following two sentences:

“The spurned lover drank her tea slowly.”

“The frightened child held the puppy tightly.”

These sentences are obviously quite different in meaning. But looked at more abstractly, a structural analysis shows that in structural terms these two sentences are in fact pretty much the same. Each consists of a subject and a predicate, the subject of each is a noun phrase, and the predicate is a verb phrase, and the substructure of each of the component phrases is almost exactly the same in each sentence.

Thus, when we look through the surface meanings of sentences like this in order to find deeper structural relations, we see regularities that from the surface alone we would never see. In this case, two “texts” that look very different end up under analysis revealing that they are at the same time very much the same.

This turns out to be an important insight. We can use such understandings to think about how sentences mean, or how children learn language, or how the brain can process an infinitude of different sentences, or, even, what goes wrong in such psychologically debilitating conditions as aphasia. Indeed, the success with which structuralist linguistic analysis works is a big reason people have looked to extend it to other human-generated systems—like story telling.

II. When you think about how this model of searching for underlying structures can be applied to literary texts, it isn’t too hard to see that you would be looking to discover similarly systematic underlying structural relations in the stories we tell. And to the extent you are able to accomplish that task, you will also have (the structuralist claim goes) a “scientific” basis for explaining literary behavior, just as in the language example above you can locate a “scientific” basis for explaining the syntax of a natural language. In the study of language this has been very productive—it has led to understandings of many different kinds across many, many languages. Could it do the same for literature? That in the 1960’s was certainly the hope.

Now. There really are valuable insights to be developed this way. It is not uninteresting, for example, that Sophocles’ Oedipus Tyrannus and Shakespeare’s Hamlet share many “structural” elements—including a son-hero who spends the play enacting the role of detective in the murder of his father while engaged in a deep relationship (!) with his mother. From this point of view, in fact, the two plays are surprisingly similar, and that can lead in turn to a number of conversations about the significance of that similarity. And certainly, the Willa Cather quotation I give above suggests a way of understanding all of literature as the repetition of a very few basic elements. No doubt she’d add something about how they could be combined in different orders, just as is the case with the elements of English sentences, but she is pointing towards something that to her is a deep and powerful truth about what literature is and does.

And if you were to push towards other levels of structural analysis, you would find other ways of reducing the myriad complexity of the universe of discourse to a more handle-able set of meaning units. Thus in an obvious case, you can greet someone in many, many ways, from “Hello” and “Hi” and “What’s happening?” to no more than a nod of the head or a smile and raised eyebrows. Each is distinct, yet at a certain level each is also the same—a primary move in a semi-ritualized behavior pattern—“greeting”—that goes on millions (or billions) of times a day all over the English speaking world.

Moreover, once you see that as a way of capturing a fact about English speaking societies, you might also go on to wonder whether this was not so much society-specific as universal to human cultures. So you might ask whether ALL human languages have similar greeting systems. That’s actually an empirical question—it’s something you’d have to go out and find out. You’d find a lot of words and phrases that don’t look or sound anything alike, but which are structurally very similar. “Ni hao ma?”—a standard greeting in Chinese—obviously is as different as can be on the surface from “Hi.” But the fact that Chinese, and Italian, and Tagalog and language after language have greeting terms may also be a step towards capturing something deep and powerful about human behavior and about the role of language in structuring, enabling, and regulating that behavior.

Coming back to the literary, theorists have been highly intrigued by being able to see deep regularities in literary behaviors we study. For if we can see a certain amount of systematicity in our literary behaviors, we are likely to be curious about how far that idea can be taken. That’s why people like Barthes start experimenting with new claims about the deep structure similarities to be found in literary discourse.

Now, that said, to this point structuralist methods have not carried the day as a hermeneutic. I think one reason they have not is that in practice no structuralist critic has ever produced anything nearly subtle enough to account for the complex entities that texts turn out to be. It may be interesting to think about the ways texts (like sentences) can be analyzed into constituent structures, but that way of thinking just hasn’t said enough about the sorts of things that interest us about how texts mean—and a major reason for this is that in excluding the actual surfaces of texts (the language choices “authors” actually make, for example, along with what we readers can and cannot do in response to those choices) structuralism also excludes most of what we care most about in literary texts. You can only get texts to be this simple if you ignore reference and use. True, as Barthes declares, literary texts are not “transitive” in the normal sense of the word. But just because they don’t work exactly like things we say in ordinary human conversations doesn’t mean that they “do” nothing. In fact they do indeed do things—lots of things, including educate, frighten, amuse, urge, create sympathy or anger.

Radway’s explorations of middlebrow literature offer an example of the limits of the structuralist model. For while sharing Barthes’ anti-establishment bias, and sharing as well a shift of interest from text and/or author to reader, Radway’s way of seeing literary discourse is not finally a place Barthes himself would be happy. For in the end she is interested in texts in use—how different readers use different texts for purposes of their own—based on experience, age, class, and so on, and in how these texts affect and structure their naïve readers. And that for her quite naturally leads to other questions—such as what people use texts for, and what significance those uses have for them. Thus by the end of her piece she’s explained her ambivalence about middlebrow writing by noting that while on one hand she is still carried away by the rush of reading excitement such books produce in her, on the other hand her analytic bent reveals to her some disquieting things about what she thinks these books ALSO do, all under cover of being nothing more than exciting and powerful reading experiences.

Moreover, while a certain conception of authorial intention can indeed be (just as Barthes argues) a misleading way of limiting and even suppressing interpretation, maybe the right response to that isn’t abolishing the notion of “intention” altogether as a guide to interpretation (thus removing the author entirely from consideration of a text’s interest—what Barthes calls the “death” of the author). Maybe we should respond to the downside of the concept of “author” by rethinking that concept in order to demystify it. Thus if we imagine authorial “intent” as something (like all of our intentions) that is limited in scope, capable of being misunderstood or even badly expressed by the writer him-/herself, for example, and if we understand that what texts do for readers can be related to an author’s intention in a very different ways, then we might suggest it isn’t “intention” that limits reading. Rather it is the fetishization of intention such that the act of interpretation becomes defined exclusively or even just primarily as the recovery and account of “what the author meant.” So maybe we shouldn’t just throw authors and their “intentions” into the dustbin of history.

Does that mean we can just ignore Barthes after all? I don't think so! For Barthes is right to argue that to the extent authorial intention excludes understanding much of what Hamlet (for example) actually does for its readers, it is both limiting and falsifying. At the same time, however, what theory does next, and what this course will be doing next as well, is follow this conversation through its next few turns. As we shall see in our remaining readings, one can explain a lot of post-structural and post-colonial discourse as a set of revisions of the notion that the author is dead. So don’t worry. Authors as a subject of literary interest are in fact far from dead. Indeed, they may actually be more alive than ever! It’s just that in most interpretive frames they are much more “normalized” than traditional canonical criticism had allowed them to become.

III. To close this discussion of structuralism, I’ll suggest that while it is certainly the case that there are structures in literature, the structuralist perspective hasn’t taken over because at a word-by-word level these structures occur and shift and reoccur in bewildering complexity. But even if structuralism hasn’t finally been all that productive in terms of reading practices, it still has had a major effect on the tradition of “literary” interpretation. The ways it has made a difference include:

Its attack on the “elitism” of the literary establishment resonated strongly, especially with younger scholars, and encouraged other forms of rebellion against conventional thinking.

Its “scientific” dimension helped to promote the introduction of a whole range of new texts (like romances) into the arena of the “study-able.” Barthes, for example, wrote not just on literary topics but on the systems of style as well.

It helped to breathe the oxygen of interdisciplinarity into literary studies, which in time led to cultural studies (Radway is an example) and, more recently, to postcolonial studies.

In more abstract forms, the attractions of structuralist thought seem very much at the root of much post-structuralist theorizing in Derrida and Foucault. Neither would identify themselves as "structuralists," but they are very obviously concerned with structural problems in how and what we know, and how best to identify and respond to them.

As we move to Rabinowitz, Said and Spivak, then, we will be thinking of what they argue in terms of the way they extend this ferment in the world of textual studies by exploring even newer realms of textual complexity. There are indeed, as Hamlet tells Horatio, “more things in heaven and earth than are dreamt of in your philosophy”—at least so long as Horatio’s philosophical dreaming hasn’t been much focused on postcolonialism.