Tense and Connective Constraints on the Expression of Causality

Tense and Connective Constraints on the Expression of Causality Pascal Alnsili TALANA, Universit6 Paris 7 2, pl. Jussieu, case 7003 F-75251 PARIS Cedex 05, France Pascal. hms ili@linguist, jussieu, fr and Corinne Rossari Unlverslt6 de GenSve, Facult6 des Lettres, dpt de Linguistique 2, rue de Candolle, CH-1211 GENI~VE 4, Switzerland Corinne. Rossari~lettres. mlige, ch Abstract Starting from descriptions of French connectives (in particular "done"--therefore), on the one hand, and aspectual properties of French tenses passd simple and imparfait on the other hand, we study in this paper how the two interact with respect to the expression of causality. It turns out that their interaction is not flee. Some combinations are not acceptable, and we pro- pose an explanation for them. These results ap- ply straightforwardly to natural language gen- eration: given as input two events related by a cause relation, we can choose among vari- ous ways of presentation (the parameters being (i) the order, (ii)the connective, (iii)the tense) so that we are sure to express a cause relation, without generating either an incorrect discourse or an ambiguous one. 1 Introduction The work reported in this paper aims at deter- mining which constraints hold on the interac- tion between the expression of causality (with or without connective) and aspectual properties of the eventualities and of the tenses used to ex- press them. As a matter of fact, it turns out that, at least in French, the choice of one of the two tenses passd simple (ps) or imparfait (IMP) is not neutral with respect to the expression of causality, in particular realised by means of the connective "done" (therefore). It has been observed that even when con- cerned only with temporal localisation, it is not enough to characterize tenses if one do not take into account the effects of discourse relations be- tween eventualitiesl: (la-b)(Molendijk, 1996); it has also been observed that the use of the 1 We use the term eventuality to refer to either events, states or processes, as is traditional since (Bach, 1981). connective "donc" is itself subject to various ac- ceptability constraints (1c-d) (Jayez, 1998). (1) a. Paul attrapa une contravention. @ I1 roulait avec plaisir Paul got fined, lie was driving with pleasure 2 b. Paul attrapa une contravention. I1 roulait trop vite Paul got fined, lie was driving too fast c. La branche cassa. # Paul tombait donc dans le vide The branch broke. Paul was therefore falling down d. Sa premi6re demande rut refus6e. I1 en r6digeait donc une autre His first application was refused. tie was therefore writing another one Our objective in this paper is twofold: we want to study systematically the interaction between the various parameters we have men- tionned, in order to provide a general expla- nation for the acceptabilities that have been observed, and we also want these explanations be formulated in terms of "conditions of use", so that our results are exploitable for text gen- eration. As a matter of fact, the choice of an appropriate form …

Content truncated to 500 words