Evidentials and Questions in Cheyenne
Recent research on evidentials has led to formally explicit semantic analyses of various languages, including Cuzco Quecha [2, 3], St’át’imcets [5], Kalaallisut [1], and Japanese [6]. Interaction with questions and other diagnostics distinguish illocutionary evidentials from propositional evidential...
| Other Authors: | |
|---|---|
| Format: | Text |
| Language: | English |
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.405.5442 http://conf.ling.cornell.edu/sem/Evidentials-and-Questions-in-Cheyenne_abstract.pdf |
| Summary: | Recent research on evidentials has led to formally explicit semantic analyses of various languages, including Cuzco Quecha [2, 3], St’át’imcets [5], Kalaallisut [1], and Japanese [6]. Interaction with questions and other diagnostics distinguish illocutionary evidentials from propositional evidentials. Roughly speaking, illocutionary evidentials (found e.g. in Cuzco Quechua [2, 3] and Kalaallisut [1]) are interpreted like English parentheticals, whereas propositional evidentials (found e.g. in St’át’imcets [5] and German [3]) are more like modals. My fieldwork on Cheyenne (15 weeks over three field trips) reveals a more complex semantic typology. Cheyenne evidentials behave like illocutionary evidentials in declaratives, but not in questions, where they can shift the illocutionary force. Building on [4], I propose a semantic analysis of Cheyenne evidentials in declaratives and questions that satisfies three desiderata: (i) it is fully compositional (like [4]); (ii) it distinguishes the evidential and the propositional contributions without positing a separate level of illocutionary meaning (contra [2]), and (iii) it accounts for the intuition that both contributions affect the truth conditions (contra [2]). Cheyenne Data: Crosslinguistically, illocutionary reportatives do not imply that the proposition in their scope is true, false, or even possible ([2], [3], [1], a.o.). In contrast, propositional reportatives imply that it is at least an open possibility [5]. Cheyenne exhibits the illocutionary pattern, shown in (1). However, as shown in (2), illocutionary reportatives do imply that the speaker has heard the proposition in their scope (both diagnostic examples are based on [2]). (1) É-háéána-s˙estse 3-hungry-rpt.3sg naa+oha but é-sáa-háéána-he-∅ |
|---|