Role of definiteness, specificity & animacy in the prominence of object NPs in Hindi

Overt case marking on a direct object is associated with its prominence, animacy
and definiteness being dimensions along which prominence can be assessed (Aissen
2003). Enç (1991) proposes that the syntactic distribution of case marking in Turkish can
be stated in terms of specificity. She suggests the widely known definiteness effect is
caused not by definiteness but by specificity. The sentences (1) and (2) below exhibit the
so called definiteness effect. According to Enç, sentence (2) is ungrammatical as the NP
“most Pharaohs” is not specific. Her proposal suggests that specificity could be
considered as a dimension to assess prominence on the object.

An examination of Hindi suggests that animacy and specificity1 function in Hindi
for object case marking. Definiteness does not play any role in it despite its apparent
contribution, e.g. examples (3) and (4) below seem to show that definiteness affects case
marking in Hindi. It seems here that since pronouns are higher in definiteness2 scale, they
are case marked and the NP lower on the scale is not case marked. But, also notice that
the pronoun “us ko” is [+human], so it is high on animacy scale, the NP “seb” is [-
animate], so it is low on the same scale as well as on the specificity scale being [-
specific]. Examples (5) and (6) below show definiteness does not have any role in case
marking. Notice in both the sentences, specificity and animacy are kept constant, in (5)
the object is [+def], in (6) it is [-def].

We notice, specificity forces accusative case marking on the direct object
(example (6) below), although objects can have nominative case and still have specific
reading (example (7) below). Similarly animacy scale does not put an obligatory
requirement on the object to have accusative case (example (8) below), but generates a
preference for it (example (9) below). In this paper, I propose an account for variation
exhibited by Hindi object case marking in terms of animacy and specificity. I propose
that specificity and animacy interact to contribute to the prominence of an object NP3. If
the object is high in prominence, it deviates more from its prototypical configuration
(Aissen 2003) and hence is incompatible with VP internal position, it moves out of VP to
a higher node in the tree and gets overtly case marked. This account can explain not only
the languages with overt object marking (Hindi) but also with verb- object agreement
(Bantu language Ruwund).

1. An NP is specific if the speaker has some specific reference in mind.
2. Definiteness scale: Pronoun > Proper Noun > Definite NP > Indefinite Specific NP > Indefinite Non- specific
NP
3. Also I show that animacy adds more to the prominence of a direct object NP in Hindi than specificity does (at
least feature [+human] has more weight than specificity).

Data:
(1) There are exactly two Pharaohs buried here. (Forbes)
(2) * There are most Pharaohs buried here. (Forbes)
(3) raam us ko dekh-ta hai
Ram-ø her ACC see-hab pres ‘Ram sees her (pronoun).’
(4) raam seb khaa-ta hai
Ram-ø apple-ø eat- hab pres ‘Ram eats an apple (indefinite non- specific NP).’
(5) pichle dinoon raam ek DawkTar se milaa thaa, aajkal raam us DawkTar ko
DhuunDh rahaa hai
‘Ram had met a doctor some day back, he is searching for that doctor these days.’
(6) raam ek DawkTar ko DhuunDh rahaa hai
‘Ram is searching for a doctor.’
(7) raam chomski ka sabse nayaa pepar DhuunDh rahaa hai
‘Ram is searching for the Chomsky’s recent most paper.’
(8) raam bacce dekh rahaa hai
‘Ram is watching children.’
(9) raam baccoon ko dekh rahaa hai
‘Ram is watching the children.’

References:

1. Aissen, J. (2003) “Differential Object Marking: Iconicity vs. Economy,” Natural
Language and Linguistic Theory 21, 435- 483.

2. Enç, M. (1991) “The Semantics of Specificity,” Linguistics Inquiry 22, 1- 25.

3. Forbes, G. “Depiction Verbs and the Definiteness Effect”
http://www.tulane.edu/~forbes/pdf_files/D_Vbs&DefEffect.pdf