a)   c-command is not crucial for accounting for the anaphor-antecedent relationship in Hindi, and       b)   the notion of subject being the antecedent of anaphors is not adequate to account for the anaphor-antecedent relationship. The proposal to treat NP as the governing category of a pronoun and S and some instances of NP as governing category of an anaphor (Huang; 1983, Chomsky; 1986a) is not way persons. The proposal has surely some merit, at least it accounts for more facts than the theory proposed in LGB. For instance, LGB has no way of accounting for the occurrence of the pronoun in each of the following: viii)   Ram lost his way       ix)   The artists expected their pictures to be on sale. In (Viii), "his" is a bound pronoun, bound by "Ram" alone; in other words, it cannot have an antecedent outside the sentence. In contrast, the pronoun in (ix) can pick up an antecedent outside the sentence but can also have "the artists" as its antecedent, which, in fact, is the preferred interpretation. As mentioned above, LGB indeed has no way of accounting for these facts. However, once NP is regarded as the governing category, the pronoun in (viii) can have "Ram" as its antecedent and in (ix), it can have "the artists" as its antecedent. Semantic considerations involving "lost" and "way" can restrain the pronoun from choosing an antecedent outside S and pragmatic considerations would yield the preferred interpretation of the pronoun in (ix). But then, if such considerations have to be invoked in order to obtain the desired results, the proposal to treat NP as a governing category loses its appeal to a certain extent.   Consider the following NPs: x)   the senators' criticism of them            i                         j          *i                         i       xi)   the senator's criticism of themselves            i                         i One would not like to treat "of-them/themselves" as the governing category in the above. One would like to have the entire NP as the governing category of the pronoun and the anaphor in both these sentences. But why is "of-insertion which, however, does not lead to node-creation (Chomsky, 19811b:51: Chomsky: 1986a:198). Now, if the governing category of the pronoun has to be an NP there is no reason why "of-them" cannot be the governing category of "them". If "of" is not inserted but is postulated in the D-structure representation then, of course, it cannot be a semantically empty element and must contribute in the same way to theta-role assignment as any other preposition. But then, that would mean treating the verb "criticize" and the verbal noun "criticism" differently for purposes of theta-role assignment; the verb and its verbal noun, then, cannot have the same argument structure -- consequences one would hardly regard welcome. The proposal to regard the NP as the governing category of the pronoun aims to ensure that the pronoun in the specifier position of the NP remains unbound because there a pronoun cannot obviously have a binder within that NP. But, then although it accounts for facts in (viii) and (ix), it makes the theory immune to falsification. It does not seem unreasonable, then, to try and formulate a different theory of binding for Hindi which can satisfactorily account for the relevant facts in this language.