Copular agreement in Hindi-Urdu

Abstract

In this paper, we present an analysis of Copular Agreement in Hindi-Urdu. We examine assumed identity copula structures (e.g., For today, I am him.), where we show that the broad characterization of the Hindi-Urdu agreement generalization – ‘agree with the highest unmarked DP’ – is insufficient: a structurally lower unmarked nominal has a demonstrable impact on the availability of agreement with a higher unmarked nominal. This interference arises as a function of the person specifications of the various unmarked nominals in the structure. Previous approaches account for such effects either in terms of the licensing needs of the nominals involved in identity copulas (Coon et al. 2017; Keine et al. 2019), or in terms of the properties of the person agreement probe (Coon & Keine 2020). We argue that in order to account for the interference effects observed in Hindi-Urdu, a combination of both perspectives is required. Given the impact of realization on the grammaticality of identity copula sentences – verbal morphology in certain tenses and the presence/absence of copulas modulate grammaticality – a feature gluttony analysis (Coon & Keine 2020) is required. Under this approach, for certain combinations of person features, the finite agreement probe is in an agreement relationship with both the nominals in the structure. Ungrammaticality arises when there is no morphological exponent that can realize the features associated with those two agreement relationships. Further, we establish that both the DPs in identity copula structures are licensed by finite T, and thus an analysis entirely without a licensing component is untenable. We also show that the issue of licensing in copular constructions (and elsewhere) in Hindi-Urdu can be handled through the adoption of a Kalin (2018) style analysis.