A Cross-Linguistic Study of the Non-at-issueness of Exhaustive Inferences

Several constructions have been noted to associate with an exhaustive inference, notably the English it-cleft, the French c’est-cleft, the preverbal focus in Hungarian and the German es-cleft. This inference has long been recognized to differ from exhaustiveness associated with exclusives like English only. While previous literature has attempted to capture this difference by debating whether the exhaustiveness of clefts is semantic or a pragmatic phenomenon, recent studies such as (Velleman et al. 2012, Proceedings of Semantics and Linguistics Theory (SALT) 22, pages 441–460) supplement the debate by proposing that the notion of at-issueness is the culprit of those differences. In light of this notion, this paper reconsiders the results from previous experimental data on Hungarian and German (Onea and Beaver 2011, Proceedings of Semantics and Linguistic Theory (SALT) 19, pages 342–359; Xue and Onea 2011, Proceedings of the ESSLLI 2011 Workshop on Projective Meaning, Ljubljana, Slovenia) and presents new data on English and French, showing that the “Yes, but” test used in these four languages to diagnose the source of the exhaustive inference (semantics vs. pragmatics), in fact diagnoses its status (at-issue vs. non-at-issue). We conclude that the exhaustiveness associated with clefts and cleft-like constructions is not at-issue, or in other words, exhaustiveness it is not the main point of the utterance.