The capacity of narrative to submit to the desires of this or that mind without giving up secret potential may be crudely represented as a dialogue between story and interpretation. This dialogue begins when the author puts pen to paper and it continues through every reading that is not merely submissive. In this sense we can see without too much difficulty that all narrative, in the writing and the reading, has something in common with the continuous modification of text that (...) takes place in a psychoanalytical process or in the distortions induced in historical narrative by metahistorical considerations. All that I leave to Roy Schafer1 and Hayden White. My immediate purpose is to make acceptable a simple proposition: we may like to think, for our purposes, of narrative as the product of two intertwined processes, the presentation of a fable and its progressive interpretation . The first process tends toward clarity and propriety , the second toward secrecy, toward distortions which cover secrets. The proposition is not altogether alien to the now classic fabula/sujet distinction. A test for connexity is that one can accurately infer the fable . The sujet is what became of the fable when interpretation distorted its pristine, sequential propriety . · 1. Not forever, I hope; his essay and its "refined common sense" have powerful implications for a more general narrative theory. Frank Kermode is King Edward VII Professor of English at Cambridge University. The author of The Sense of an Ending: Studies in the Theory of Fiction, Continuities, and Shakespeare, Spenser, Donne: Renaissance Essays, his works also include The Classic and The Genesis of Secrecy. His previous contributions to Critical Inquiry are "Novels: Recognition and Deception" , "A Reply to Denis Donoghue" , and "A Reply to Joseph Frank". (shrink)