Evidentials are syntax-sensitive: The view from Bangla

Abstract

This paper studies an evidential in Bangla which changes its evidential flavor based on its syntactic position. Forging novel connections with the literature on finiteness, indexical shift, and complementizer agreement, this paper demonstrates how evidentials can be sensitive to the presence of syntactic heads that represent the point-of-view of an utterance. Three main claims are made: (i) evidentials always take finite clauses which are perspective-sensitive, (ii) this perspective-sensitivity is syntactic, i.e. it is the result of control by speech-act heads, (iii) coindexation or contraindexation among these perspectival heads can have very important effects on word order in evidential constructions. This paper thus offers a comprehensive syntactic profile of evidential particles, which have generally been investigated with regard to their semantic-pragmatic contribution, arguing that the structural configurations these elements appear in have undeniably crucial effects on the interpretative component.

The Morphosyntax of Magahi Addressee Agreement

“…I locate it in a specifier of FinP. This is in accord with Bhadra 2018's proposal that there are null speaker and addressee coordinates in every finite clause. Secondly, I propose that the functional head that bears ɛ and agrees with the Hr DP is relatively low in the clause.…”

“…However, I propose that this locus of AA is relatively low in the clause in Magahi: in the FinP domain (in the cartographic structure of Rizzi 1997). This is in accord with Bhadra 2018: in her analysis of evidentials in Bangla, Bhadra proposes that there are null “speaker” and “addressee” coordinates in every finite clause. Furthermore, I claim that the functional head that agrees with the Hr DP is also relatively low in the clause: it is Fin, the head of FinP.…”

“…In other words, it has an entire sentence in its scope. We could, in principle, find a phenomenon where the DPs are contraindexed, that is, where they cannot have the same referent (see the discussion of Bhadra 2018's analysis of evidentiality below). In this article, though, I limit myself to the default configuration and leave the contraindexation possibility for future work (see footnote 18, which sheds some light on the complexity related to this issue in light of Magahi AA).…”

“…Independent evidence in support of two instances of the speaker and addressee comes from Bhadra 2018. Bhadra proposes that, in addition to the null DP coordinates corresponding to the speaker and addressee that occur in the SAP domain, there are also such coordinates in every finite clause.…”

On the syntax of addressee in imperatives: insights from allocutivity

“…However, this predicts obligatory matching between the features of embedded and matrix allocutivity, which is not borne out. In contrast, recent studies (Bhadra 2018;Alok 2020) have shown that there are null DP coordinates corresponding to 'speaker' and 'addressee' in every finite clause. If the above-mentioned representations of the addressee can be shown to host their own [status] feature, the facts in (i) can be explained.…”