Retour vers liste

Détail de la contribution


A comparison of the conditional complementizers falls and if

Abstract/Résumé: In this paper I compare the German conditional complementizer 'falls' with its English cousin 'if'. I will show that 'falls' is acceptable only in a subset of the environments in which 'if' is acceptable: It is odd in subjunctive conditionals, and in indicatives it can only be used if the speaker does not consider it extremely likely that the antecedent proposition is true in the world of evaluation. Therefore, 'falls' in contrast to 'if' is also unacceptable in premise conditionals (Haegeman 2003), i.e. in cases where the speaker takes up in the form of a conditional antecedent somebody else’s assertion without indicating that she does not believe the respective proposition to be true. I will argue that the similarities as well as the differences between 'if' and 'falls' are best captured by assuming that while both are universal quantifiers over accessible worlds where the respective antecedent proposition is true (along the lines of Gillies’ 2011 analysis of 'if'), they impose different conditions on the propositions they take as their first argument (with the consequent proposition being the second argument): In the case of 'if' it is only presupposed that the respective proposition is not true in all the worlds that are compatible with the speaker’s knowledge in the relevant context, i.e. it rules out cases like (1) where the speaker picks up a proposition she has asserted herself, but leaves open both the possibility that the speaker considers the respective proposition to be extremely likely and the possibility that he considers the respective proposition to be certainly false. In the case of 'falls', in contrast, it is presupposed (a) that there are some worlds compatible with the speaker’s knowledge where the respective antecedent proposition is true, thus ruling out 'falls' in subjunctives, and (b) that it is not true in all the best world among the ones compatible with the speaker’s knowledge, i.e. all those worlds where the speaker’s expectations about what is most likely the case, are met. (1) A: It’s later than 11 p.m. and *if/since it’s that late already, we should leave immediately. References: Gillies, Anthony. 2010. Iffiness. Semantics and Pragmatics 3, 1-42. Haegeman, Liliane. 2003. Conditional Clauses: Internal and External Syntax. Mind and Language 18, 317-339.