< previous page page_214 next page >

Page 214
It is easy to be wise after the event: the second topic is discovered only at a second reading. Those whom the gods wish to destroy they first drive mad. Drame displays its discoursive strategies to madden (or rather blind) its naive reader as far as the strategy of the fabula is concerned.
8.3. The Strategy of Narrative Structure
8.3.1. Inferential Walks and Ghost Chapters
Taking chapter 2 as a reduced model of Drame, one finds in it a clear example which shows the difference between actualizing discoursive structures and actualizing narrative structures.
Raoul pursues Marguerite /la main levée/. The reader, resorting to the frame «conjugal quarrel», realizes that Raoul is raising his hand to beat his wife. 5 But, in performing this semantic disclosure, the reader is in fact* accomplishing a double inferential movement:
(i) he realizes that Raoul wants to beat Marquerite;
(ii) he expects Raoul actually to beat Marguerite.
The discursive inference (i) is correct. The narrative inference (ii) is false, since the further course of the story will disprove it.
At the level of discursive structure the reader is invited to fill up various empty phrastic spaces (texts are lazy machineries that ask someone to do part of their job). At the level of narrative structures, the reader is supposed to make forecasts concerning the future course of the fabula.
To do this the reader is supposed to resort to various intertextual frames among which to take his inferential walks. Every text, even though not specifically narrative, is in some way making the addressee expect (and foresee) the fulfillment of every unaccomplished sentence: /John will not arrive because . . ./ makes one hazard forecasts about the missing information. Obviously, these expectations are more evidently requested in a narrative text. They are anticipations of the global course of events represented by the fabula when it reaches its final state. Frequently, a fictional text not only tolerates but anxiously awaits these inferential walks in order not to be obliged to tell too much.
Frequently, given a series of causally and linearly connected events a . . . . . . . e, a text tells the reader about the event a and, after a while, about the event e, taking for granted that the reader has already anticipated the dependent events b,c,d (of which e is the consequence, according to many intertextual frames). Thus the text implicitly validates a 'ghost chapter', tentatively written by the reader. In other words, the author is sure that the reader has already written by himself a chapter

 
< previous page page_214 next page >