Tridha Chatterjee and Marlyse Baptista (University of Michigan)

This study examines the hocche copula in Bengali within equational sentences, as illustrated in example (1).

1) Tar baba ho-cch-e ei bisshobiddaloy-er oddhapok
3sg.GEN father be-PRS.PROG-3P this university-GEN professor
His/her father is a professor at this university. (Thompson 2010)

The occurrence of hocche in that position is unusual given that Bengali is an SOV language, as shown in (2)

2) ami ├Žkhon bari ?a-cch-i
1SG now home go-PRS.PROG-1P
I am going home now.

Second, this particular position seems to result from a relatively recent development in the language, unattested in 18th century Bengali texts we consulted, in which equational sentences lacked a copula (as shown in 3 below).

3) tini diner rokkhok
2 SG.HON poor keeper
He/She is the keeper of the poor. (1860, Bengali play)

Third, a similar construction is attested in only a minority of south Asian languages such as Assamese and Oriya in addition to Bengali (Ferguson 1967, Dasgupta and Ghosh 2006, Nath 2009).
This talk sheds more light on the hocche copula of Bengali by comparing them with the copulas of other Indian languages such as Hindi and Marathi (Western Indo-Aryan), Telugu and Malayalam (Dravidian) and Assamese and Oriya (Eastern Indo-Aryan). Through these comparisons I show why it is a unique feature among south Asian languages and why its position of occurrence is outstanding given that south Asian languages have SOV word order and the copulas across all of these languages except in a handful occur sentence finally. Crucially, I examine equational sentences in Modern Bengali that lack a copula (2) and those with an overt copula (1) in relation with each other to show to what they have equivalent semantics. I argue that NP NP and NP be NP are two competing constructions that occur in free variation but that potentially originate from different sources.