The Ban Against Illeism and Indexical Shift in Farsi
Amir Anvari
October 2019
 

Illeism, the act of referring to oneself, and by extension the addressee, using a third person noun phrase, is usually infelicitous. One salient explanation of this "Ban Against Illeism" is that third person noun phrases compete with indexical (i.e., first or second person) pronouns and the latter are preferred if they convey the same meaning, which happens in particular when third person noun phrases refer to the actual speaker / addressee. This explanation predicts an interaction between (in-)felicity of illeism and "indexical shift" (Deal 2017 for a recent survey). Specifically, we predict that in an environment in which indexical shift is obligatory, using a third person noun phrase to refer to the actual speaker / addressee is acceptable. Building on data from Farsi, I will use the Shift Together constraint on indexical shift (Anand & Nevins 2004, Anand 2006) to construct environments in which indexical shift is obligatory, and I will provide evidence that this prediction is indeed correct. I will then argue that, at least in Farsi, another generalization holds which is the mirror-image of the first one: in an environment in which indexical shift is obligatory, using a third person noun phrase to refer to the reported speaker / addrressee is not acceptable. To capture this second generalization, a modification is required to make sure that the principle is "blind" to the de re / de se distinction. By modifying the principle to check for covaluation of referential noun phrases (as opposed to contextual equivalence of sentences), I will argue that Sharvit's (2010) notion of "type-II covaluation" can be used to solved this problem. The data set explored here constrains the space of possible analyses of indexical shift: I will argue that while the operator-based approach of indexical shift (Anand & Nevins 2004, Anand 2006) captures the relevant data without further ado, the same cannot be said for the most straightforward implementations of the binding-based approach (Schlenker 1999, and subsequent work). Finally, I will provide evidence that the two generalizations mentioned above apply, not only to indexical pronouns and third person noun phrases, but also indexical locative and temporal adverbials and their non-indexical counterparts.
Format: [ pdf ]
Reference: lingbuzz/004823
(please use that when you cite this article)
Published in:
keywords: illeism, indexical shift, farsi, persian, semantics
previous versions: v2 [October 2019]
v1 [October 2019]
Downloaded:97 times

 

[ edit this article | back to article list ]