Organised by the Centre de recherches sur les arts et le langage (EHESS-CNRS) in collaboration with the Narratology Research Group (University of Hamburg) and the Département de Littérature Comparée (Université de Paris III)
Institut Goethe, Paris- 29-30 November 2002
It is generally acknowledged that a narrative is the narration of events and that narrative is thus divided into two distinct levels: the level of narration and the level of narrated events. Any contamination of one level by the other would thus seem to run counter to the very nature of narrative.
In actual narratives, however, things are more complex. One of the narrative devices that contradicts the thesis of such a distinction of essence between level of narration and level of narrated events is the embedded or metadiegetic narrative. As a result of embedding, the metadiegetic narrative enters into a relation with the narrative containing it that, functionally, is either explanatory or thematic (an extreme form being the mise en abyme) or, independently of the content of the embedding narrative, that represents a contamination between the two levels, the act of narrating itself providing a link between two domains that are normally distinct from one another. This latter effect, known as metalepsis (Genette), is a kind of forced marriage between levels where the narrator or the narratee strangely enters the domain of the characters, or vice versa. Usually, metalepses are marked and thus perceptible. However, when a metalepsis dispenses with the diegetic relay, it undermines the distinction between the two narrative levels and, more generally, destabilises the operability of representation as such. In the theatre, as can be seen in some of Pirandello's and Handke's plays, for example, violations of the 'representational pact' spill over the stage where the dramatic action takes place, disrupting the very framework of representation. But however transgressive metalepsis might be, it still remains among the means of narrative embedding, and must thus be counted among the possibilities open to narrative.
 Embedding is a phenomenon that does in fact exist elsewhere than in narratives. In various forms, embedding can be found in logic, mathematics, linguistics, semiotics, cognitive psychology, ontology, etc. Research carried out in these fields has not failed to influence the study of metalepsis. While metalepsis in narrative calls for a more general reflection on narrative levels, the status of boundaries, the criteria for segmenting, etc., research in other fields may well prove fruitful for reflection on metalepsis. Hence, the work of Hofstadter (and others) makes it possible to compare metalepsis with the paradoxes of the 'Strange Loop' or the 'Tangled Hierarchy'. Such paradoxes have their equivalents in art, as for example in Escher's Print Gallery (paradox to be distinguished from the recursive character of the mise en abyme found in Drawing Hands, also by Escher) or in Magritte's Ceci n'est pas une pipe.
 Metalepsis must also be examined (re-examined) in light of the various approaches to narrative currently practiced in narrative analysis. What is the place of metalepsis in models of narrative communication that distinguish between concrete author/reader, implied author/reader, etc. What is the perimeter of metalepsis in a narratology which, on the basis of story-telling in everyday life and research in cognitive psychology, incorporates the existence of various levels of experience of the world and the idea of stratification of consciousness? How does narrative semantics, based on the logic of possible worlds with 'accessibility' relations between textual strata, look at metalepsis? Does the computer hypertext, with its innumerable possibilities for linking textual segments, open up new perspectives on metalepsis?
Recent developments in narratology give evidence not only of integrating new epistemological models, but also of broadening its field of analysis to other forms of artistic expression. Metalepsis, an enigmatic and disturbing device that calls into question the means of representation, is one of the crucial junctions for any understanding of narrative and, more generally, of representation. There is thus every reason to study metalepsis in light of current theoretical developments through a transdisciplinary approach. It is our conviction that what may appear at first view as a mere technique in fact opens up paths for a renewed reflection on narrative in its relations with other cultural phenomena.
For further information:
John Pier: j.pier@wanadoo.fr