After Dynamic Narratology - Critical Essay

Style, Summer, 2000 by Eyal Amiran

Since Barthes's S/Z, in effect, theories of narrative have had uneasy relations with textual theory. Structuralist narratology attempts to explain narrative figuration, or thematics, in terms of a perceived textual grammar. This position assumes that narrative expresses a generic logic (poetics), a genetic code. In an effort to preserve narratological discourse, late structuralists like Seymour Chatman replace the claim to deep structure in favor of something like "description," and attempt to abandon the relation between figuration and poetics. [1] Dynamic theories of narrative, some versions of which I discuss here, step into the vacancy created by the retreat of structuralist narratology. Where structuralist theories produced static, often grammatical models based on levels of the text, dynamic theories would like to explain narratives in terms of their own interiority and energy. Despite their claims to the contrary, however, the dynamic theories I discuss also see figures in the text--images produced by the text--as a code, a key that shows how the text itself works. A code here is a privileged construct in a text, like Paul de Man's "allegory," that is supposed to open up or reveal the workings of the text or of a portion of the text. But it makes little sense to believe that linguistic operation itself depends on the images shown in the text. Textual analysis appeals to a formal and logical stratum of language that is in principle independent of the specific images produced by a given text. While narratology usually speaks, reasonably, of narrative figurations as themes, as the text's thematic understanding of itself, in doing so it absorbs textual assumptions with greater or lesser degrees of consciousness, assumptions about the nature of textual operations that are rarely addressed outside the realm of these figurations. The code, the figure which represents the narrative, appears to represent the inner workings of the text, but to do so is to figure the text and so to become again its represented conte nt or theme. The code cannot reveal the workings of narrative; it can only express an idea of these workings that cannot bridge the gap to the textual operation it itself posits. While many approaches to narrative (historicist, Marxist, and others) do not confuse figuration with poetics, they do not engage the relation between narrative and textuality that defines narratology. Exploring this logical difficulty in dynamic narratology, I offer an alternative for narratology proceeding from the assumption that texts are symptomatic of their personal and cultural contexts.

I am interested in writers who see the danger of reading the textual in the figurative, of reading what the text shows as an explanation of how it works, but who end up despite themselves in that trap. Consider, for instance, how Deleuze and Guattari discuss method in Kafka: Toward a Minor Literature. They propose to study the nature of the text, not its figurations of itself, but eventually rely on figures for it: "How can we enter into Kafka's work? This work is a rhizome, a burrow. The castle has multiple entrances whose rules of usage and whose locations aren't very well known" (3). The various doors, and kinds of doors (a burrow has only one), in Kafka's work, they say, are a "trap" set up by Kafka: "the whole description of the burrow functions to trick the enemy. We will enter, then, by any point whatsoever" (3). The description of the burrow is a trap because description in the text does not explain the workings of the text, but only leads the hermeneut ("the enemy") to think that it does. It is inte resting that Deleuze and Guattari opt to "enter" the text (3), to speak of the "Kafka-machine" that produces desire (as, in their vocabulary, any text does) as having an inside and an outside (7-8). They do what they had just said they would not do.

To acknowledge this difficulty, they immediately deny that they do any such thing: "We aren't trying to interpret, to say that this means that. And we are looking least of all for a structure with formal oppositions and a fully constructed Signifier; one can always come up with binary oppositions. [...] But that's stupid as long as one doesn't see where the system is coming from and going to" (7). They make, it seems to me, a sophisticated point. Deleuze and Guattari reject the notion that one can separate structural oppositions and imaginary archetype(s): "We believe only in one or more Kafka machines that are neither structure nor phantasm. We believe only in a Kafka experimentation that is without interpretation or significance" and that merely reports information (7). Although they provide figuration, not poetics, description without hieratic or vertical appeal, in fact they argue to the opposite extreme: that you cannot have anything but interpretation. For, by their own definition of the systematic, an y language is interpretation. To think that there are words that simply exist without meaning, as objects are said to do (outside of representation), is to ignore that it is the nature of language to be a system of interpretation. Deleuze and Guattari are interested in Kafka's use of music because "sound doesn't show up here as a form of expression, but rather as an unformed material of expression" (6). Josephine the mouse singer, for instance, cannot sing, so her sound resists formalization, is unformed. That it is not music is good, because Kafka, as Deleuze and Guattari write, is interested in the cry that escapes language, "sonorous material that is always connected to its own abolition" (6). Kafka uses sound to suggest such a theme (and sound is suggested by the text), but to say that it is more than a theme, that it is a feature of the operation itself of the text, is quite another thing.