Structuralist poetics has often been criticized for its claim that literature, like any other domain of external reality, can be approached with objective methods. Extreme positions have denied poetics-structuralist or otherwise-any relevance in the understanding of literature, dismissing it as yet another variety of logocentric and paternalistic discourse. In a more moderate vein, some critics, notably Paul de Man, have argued against the over-confidence of formalist poetics to handle all areas of literature with equal effectiveness; they have insisted that the distinction should be maintained between poetics and hermeneutics, each with their own areas of application. Since division of labor requires cooperation, de Man was prepared to allow, perhaps even to recommend, the cross-fertilization of the two fields. But division of labor also implies that each domain's tasks be undertaken with the best available techniques. And while the more radical stands against rational poetics cannot, by virtue of their very extremism, be properly answered, some of the doubts expressed regarding the exaggerated scientism of earlier structuralist poetics are worth consideration. If a lasting cooperation should indeed be established between the study of structure and that of meaning, poetics would have to clean up its own house by carefully scrutinizing the principles and methods it has used. Among the more pressing problems poetics should tackle, I would include the formalisms it employs and what expectations may be reasonably entertained as to their relevance. Focusing on the symptomatic case of Greimasian narrative semiotics -a school of thought that, in spite of its remarkable conceptual con-
Read full abstract