Epistemic and evidential bias in Catalan/Sardinian yes-no questions

This investigation is concerned with Catalan and Sardinian yes-no questions expressing an epistemic or evidential bias on the speaker’s side. The distinction between epistemic and evidential bias was first introduced by Sudo (to appear) who following Büring & Gunlogson’s (2000) proposal about contextual evidence considers that two different biases have to be recognized in yes-no questions: the epistemic bias which refers to “the speaker’s belief/expectation or what he/she takes to be a norm” and the evidential bias which has to do with “evidence available in the current context of conversation”. Typologically, Catalan and Sardinian constitute a good test case to investigate the epistemic and evidential bias in yes-no questions since different studies have highlighted a set of linguistic markers that seem to be related to evidential/epistemic properties in questions, such as the following: (a) yes-no questions can be headed by particles such as o or que (Catalan) and a or nachi (na(rat) chi, ‘he/she says that’) (Sardinian), as in (1)-(3); (b) Sardinian yes-no questions can be marked through fronted constituents as in (5) and (c) different intonational patterns can be found in yes-no questions (¡H+L* L%, H+L* L% both for Catalan and Sardinian and L+H* L% and H*+L L% only for Catalan and Sardinian respectively). (1) O te ne n’anaves? ‘Are you leaving?’ (Catalan) (2) A benis a jocare chin mecus? ‘Are you coming to play with me?’ (Sardinian) (3) Nachi benis a mandigare? ‘Are you coming to eat (I suppose so)?’ (Sardinian) (4) Mortu in s’ispidale est? ‘Did he die in the hospital?’ (Sardinian) The goal of this research is therefore twofold: fist, we seek to determine what kind of bias is encoded by the different linguistic markers and second, to investigate the interplay between lexicosyntactic and intonational markers in both languages. The pilot corpus analyzed was obtained by means of the Discourse Completion Test methodology (Nurani 2009) using a prompted response questionnaire. Following Sudo’s (to appear) proposal, we created a set of situations which contained different combinations of evidential and epistemic conditions leading to an affirmative, negative or free response (‘It is you birthday and a friend of yours gives you a present. You had talked to her about a specific book many times and since the packet has the shape of a book, you ask her whether it is “that book”. ’ [Positive epistemic bias, positive evidential bias.] Target question: ‘Is it that book?’). The results of the Sardinian pilot database clearly confirm that the use of different linguistic markers in yes-no questions is related to the epistemic/evidential bias as well as its polarity. Particle a is thus related to unspecified epistemic/evidential bias while fronting of the constituents conveys positive epistemic/evidential bias. Keywords: epistemic bias, evidential bias, Catalan, Sardinian, intonation. Selected references Büring, Daniel/Gunlogson, Christine (2000): Aren’t positive and negative polar questions the same? <http://hdl.handle.net/1802/1432> (14 October 2011). Sudo, Yasutada (to appear): “Biased polar questions in English and Japanese”, en: Gutzman, Daniel/Gärtner, Hans-Martin (eds.): Expressives and Other Kinds of Non-truth-conditional meaning. Oxford: Oxford University Press.