Volume 8, Issue s2 - Non-canonical questions from a comparative perspective; Editors: Andreas Trotzke & Anna Czypionka

In this brief introduction, we provide an overview of this special collection on a variety of linguistic means to convey non-canonical interpretations of the speech act of questioning.

This paper concerns wh-interrogatives with French comment ‘how’ in their reason interpretation. Some of the peculiarities that make them non-canonical are the modified semantic type of the domain of the wh-item, which is closer to propositions than to ‘manners’, and a high base generation position. These questions are used by a speaker to put on hold a (possibly implicit) invitation from the interlocutors to admit the prejacent into the common ground, and tackle preconditions to admitting it. We aim to gain insight about the characterisation of reason- comment by testing potential constraints on its occurrences, using data about lack of subject-auxiliary inversion from corpora research. The paper reports on a survey that reveals that these questions facilitate the setting up a jousting discursive procedure, in monological texts. They are not among the characteristics of informal French.

This paper presents an account of the peculiar properties of Negative Interrogatives (NI). In uttering Don’t you speak Italian? , the speakers is biased towards the underlying positive proposition, expects a positive answer, and seeks a confirmation of that expectation from the hearer. What’s more, NI involves uncertainty with respect to p , a novel observation that we bring to the fore by comparison with epistemic modals. Using a framework by which speech-acts are derived by the two operators Speaker and Other representing sources of information, the high negative is assigned to the Other projection, the positive proposition being left under the responsibility of the Speaker. Thus, the NI is motivated by evidence contradicting the speaker’s belief, which is expressed by the negation that is attributed to another source and that therefore has full referential force. Because the questions is addressed by the Speaker in spite of evidence that ¬ p ${\neg}p$ , NI displays uncertainty as to p and gives rise to a confirmation request. With direct mapping from syntax to semantics and pragmatics, the parsimonious account thus explains the morphosyntactic and interpretative properties of Negative Interrogatives.

The French wh -word comment ’how’ may be used with a reason interpretation, that is in a question that inquires about how the situation described by the proposition that follows comment (the pseudo-prejacent ) might occur or have occurred, which goes against the expectations of the speaker. In this paper, we report on a rating study whose goal was to understand the extent to which reason- comment questions – compared to manner- comment questions – are interpreted as having some questioning force or, conversely, as being rhetorical. We also test whether reason- comment questions express a certain degree of surprise and confirm our hypothesis that with reason- comment questions the speaker tries to recover from an expectation failure, while results concerning the questioning force show that reason- comment questions can be true questions (though questioning is not their sole function), and invalidate analyses that systematically treat them as rhetorical questions.

In this paper, we investigate a type of non-canonical question that is not information-seeking, but rather expresses the speaker’s surprise and/or disapproval with respect to the event denoted by the propositional part of the question. Our work is empirically based on claims that have been put forward in the syntactic literature, and our study aims at empirically assessing those theoretical claims about the illocutionary status of surprise-disapproval questions by exploring their pragmatics.

This paper discusses the nature of non-standard questions in English, German, and Japanese. I will first introduce some basic properties of non-standard questions and then show that non-standard questions can be classified into the speaker-oriented type and the addressee-oriented type. It is suggested that the addressee-oriented non-standard questions can be further classified into the counter-expectation type and the aggressive question type.

In this paper we introduce the notion of Timing of Belief (ToB) as a relevant factor of variation in common ground (CG) management in sentence-peripheral particles across different languages. CG management traces the epistemic development of mutual beliefs between speaker and addressee. Evidence for the relevance of ToB comes from a small-scale acceptability study which tested the relevance of ToB for particles in English, German, and Spanish. While these languages all possess grammaticalized structures to encode different types of knowledge asymmetries between speaker and addressee, they vary with respect to the sensitivity or encoding of ToB. The evident relevance of ToB for CG management suggests that models which focus on the dynamic character of CG development require further expansion. We hope that the fine-graded differences in CG management reported here serve to inspire an engagement with the notion of ToB and the variation we find across languages and dialects.

Besides information-seeking questions, there are all kinds of non-standard questions, whose function is not to request information from the addressee, but rather to convey information about the speaker’s epistemic or emotional state. These include rhetorical questions, surprise questions, incredulity questions, etc. This paper focuses on rhetorical questions (RQs), a sub-type of non-canonical questions, and presents the first production experiment which has investigated them in French to date. We compared string-identical rhetorical and information seeking questions (ISQs) in French. Experimental design and preliminary results are presented and discussed. In the final part of the paper, we discuss possible correlations between these findings and the semantic and pragmatic properties that distinguish RQs from ISQs.

This paper offers a pragmatic account of two wh -questions in French used non-canonically - c’est quoi, ce N and qu’est-ce que constructions. It is claimed that in this non-canonical use, both constructions express surprise. As these constructions may be syntactically ambiguous between an information-seeking reading and a surprise reading, it is expected that the two readings differ in terms of prosody. This hypothesis is supported by the results of a production experiment. Experimental evidence shows significant prosodic differences between syntactically similar information-seeking questions and surprise questions. Surprise questions exhibit an increase in lengthening, slower speech rate and less frequent rising final contours.