Advances in Role and Reference Grammar

ATTRIBUTIVES AND IDENTIFICATIONALS^455

the subjectively-inflected class. It would therefore follow that attributive
and identificational predicates which appear in the subjective class would
typically require animate subjects, and this is the case. The fact that the
majority of attributives and identificationals appear in the objective class
can be attributed simply to the small size of the subjective class and the
large size of the default objective class.

4.3 Russian

In Dakota, subjective vs. objective pronoun selection is lexically deter­

mined and invariant, so that it is impossible to make a distinction between

volitional and non-volitional action on the basis of this selection. In Rus­

sian, however, as reported in Pesetsky (1982), it is possible to find the same

verb patterning subjectively or objectively with respect to the genitive of

negation, as shown in (64), indicating that volition or agentivity of the sub­

ject is relevant to objectivity/subjectivity.

(64) a. V bassejne nikakoj rebenok ne plavaet.

in pool no-NOM child-NOM NEG swims

"No child swims in the pool."

[DO (child) [float' (child)]]

b. V bassejne nikakogo rebenka ne plavaet.

in pool nO-GEN Child-GEN NEG floats

"No child floats in the pool."

[float' (child)]

In (64a), plavaet patterns subjectively in selecting a nominative subject

under negation and is appropriately translated as "swim", indicating voli­

tional activity, while in (64b), it patterns objectively in selecting the geni­

tive of negation and is appropriately translated as "float", indicating lack of

volitional activity. The Logical Structures given below the examples indi­

cate the distinction. In the case of verbs of motion, there is no distinction

between motion-to-goal and undirected motion verbs with respect to geni­

tive of negation (genitive of negation is not allowed at all for motion verbs),

indicating that unlike Italian, accomplishment semantics does not distin­

guish the subjective and objective classes, as shown in the examples in (65)-

(66).