Among languages that employ specialized particles for answering polar questions, two major groups are known: Positive-Negative (P-N) and Agreement-Disagreement (A-D). English and Italian are examples of the former group, Japanese an example of the latter. It has occasionally been remarked in the literature that at least in some P-N languages a narrow focus in a negative polar question affects the choice of responding particles: namely, the answering pattern shifts to A-D. In the present contribution, this claim is thoroughly tested experimentally with two forced choice experiments that investigate the choice of responding particles in answers to Italian negative polar questions both with and without narrow foci. Our results bring clear support to the claim of the exceptional answering pattern to negative polar questions with narrow foci. Two candidate accounts are presented, in accord with the two main current lines of research on responding particles: an ellipsis account and an anaphoric account. In both cases, we will argue that a crucial role is played by the interaction of the narrow focus with sentential negation in the questions.

Previous research has proposed that languages diverge with respect to how their speakers confirm and contradict negative questions. Taking into account the classification between truth-based and polarity-based languages, this paper is mainly concerned with the expression of REJECT (a semantic operation that signals a contradiction move with respect to the common ground, along Krifka's lines) in two languages belonging to two typologically distinct answering systems, namely Catalan (polarity-based) and Russian (a mixed system using polarity-based, truth-based, and echoic strategies). This investigation has two goals. First, to assess empirically the relevance of prosodic and gestural patterns in the interpretation of confirming and rejecting responses to negative polar questions. Second, to test the claim that in fact speakers resort to strikingly similar universal strategies at the time of expressing rejecting answers to discourse accessible negative assertions and negative polar questions, namely the use of linguistic units that encode REJECT in combination with ASSERT. The results of our investigation support the existence of a universal answering system for rejecting negative polar questions that integrates lexical and syntactic strategies with prosodic and gestural patterns, and instantiate the REJECT and ASSERT operators. We will also discuss the implications these results have for the truth-based vs. polarity-based taxonomy.