Speaker Series: Craige Roberts (Ohio State)

Apr 19, 2018 at - | Fisher-Bennett, room 201


Talks will take place on Thursdays from 3:30pm - 5pm, followed by a reception in the Linguistics Department.


The talks themselves are open to members of the greater university community.


Character assassination: de se semantics for indexicals
Craige Roberts, OSU


The classic account of indexicality due to Kaplan (1977) has in recent years been challenged by an anaphoric approach (Büring 2005, Heim 2008a,b, Heim & Kratzer 1998, Jacobson to appear, Sauerland 2003, 2008b, Sauerland, Anderssen & Yatsushiro 2005, Schlenker 1999, 2003a,b, von Stechow 2003; and for demonstratives, Roberts 2002, Elbourne 2008), the central motivation being that anaphora resolution in the global context of utterance suffices to account for the properties of indexicals pointed out by Kaplan. Thus, Kaplanian direct reference is not necessary. I investigate a deeper problem for Kaplan’s Character of indexicals, one which shows that direct reference is not sufficient to satisfy all the desiderata for a theory of indexicality. Nor does simple anaphoric binding to participants in the global context of utterance suffice to address this problem.
Indexicals are one subtype of a much broader class of perspectival expressions, all of which presuppose as anchor the center of a relevant doxastic perspective. In indexicals, I will argue, this is evidenced by the fact that their use always gives rise to a de se interpretation, something initially observed in 3rd person pronominals (Casteneda 1966,1967,1968; Morgan 1970; Lewis 1979b; Richard 1983; Perry 1993; Maier 2009; Ninan 2010, etc.). We see this de se character not only in fixed indexicals like English I, but in shiftable indexicals like English now and Amharic –ññ (Schlenker 1999,2003). It is especially evident in reported speech, and has been observed for all 1st and 2nd person shifted indexicals in which this has been investigated, across languages.
I sketch a formal semantics and pragmatics for the de se, based on Stalnaker (2008), and use them to develop a presuppositional, perspectival meaning for English I and shiftable Amharic –ññ. All of the elements required for the proposed account are independently motivated to account for other types of perspectival expressions (e.g., see Barlew 2016 on deictic motion verbs in English and Bulu). And the approach preserves the virtues of the Kaplanian direct-reference account while overcoming its known short-comings, permitting us (a) to explain the de se character of indexical reports (the focus of this talk), but ultimately also (Roberts, in progress) (b) to permit an improved account of demonstratives, (c) to account for bound indexicals and demonstratives, (d) to provide a uniform account of so-called “fake” indexicals (Partee 1989, Heim 2008, Kratzer 1998b,2009, von Stechow 2003; Rullmann 2004; Cable 2005; Maier 2009,2009b; Jacobson to appear, Sudo 2012:144ff) without faking them, and (e) to extend to natural accounts of shifted indexicality in the large number of languages where 1st and 2nd person indexicals may be anchored to 3rd person agents of certain embedding attitudes. No monsters required