It is often claimed that relativization in (Standard) Malay/Indonesian is restricted to subject position (Sneddon 1996:286). However, a variety of authors (Chung 1976, Cole and Hermon 1998, Musgrave 2002) have argued that relativization of nonsubjects is also wellformed. Our purpose in the present paper is to examine this question with respect to Jakarta Indonesian, the colloquial variety of Indonesian spoken in the city of Jakarta. Our study is based on three sources, 1) the MPI Jakarta Indonesian Database (a large database of naturalistic data based on the speech of children and their caretakers in Jakarta), 2) a database of adult-to-adult speech in Jakarta Indonesian collected by Yassir Tjung (one of the authors of this paper and a native speaker of Jakarta Indonesian) and 3) the linguistic judgements of Yassir Tjung.

The results of our study show that for adult speakers the preponderance of relative clauses involve subject relativization. In addition, object relativization is also an option:

(1)		pasti ada sesuatu dalam diri dia yang gua nggak punya sure exist SE-unspecified in self 3Sg REL 1Sg NEG have There must be something in her that I don't have. (object)
(2)		"the dark side of Yuli" yang banyak orang nggak tau the dark side of YULI REL a.lot person NEG know "The dark side of Yuli" that (I am sure) few people know. (object)

It might be suggested that all instances of apparent nonsubject relativization are due to the promotion of apparent nonsubjects to subject by means of Passive Type 2 (Sneddon 1996). There is, however, strong reason to believe that this is not the case. Crucially, the instances of apparent object relativization found in the databases include examples that do not exhibit the characteristic restrictions of Passive Type 2 (P2): 1) In P2 the agent is restricted to pronouns or nouns used as pronoun substitutes, and 2) the agent is preceded by auxiliaries and negation. These characteristic restrictions are not found in all the apparent instances of nonsubject relativization in our data. Thus, not all instances of subject relativization could be due to the application of P2 prior to relativization. However, while adult speakers appear to allow nonsubject relativization, object relativization is blocked when the verb has the nasal prefix (active voice).

Turning from adults to children, we found that children also exhibit nonsubject relative clauses.

(3)

yang... yang... yang Mama beli. REL REL REL mommy buy The one... the one... the one that Mommy bought.

Children differ from adults in two crucial ways: children relativize on (time) adjuncts using yang and also permit relativization when the nasal prefix is present:

(4)		yang Ica beli Kentucky ya, Bu? REL Ica buy Kentucky yes TRU-mother When I bought Kentucky, right Mom? (temporal adjunct)
(5)		yang suruh Ayah ngambil. REL order father N-take the one that Daddy asked me to take.
(6)		yang Tante Erni mbawa itu. REL auntie Erni N-bring that The one that you brought.

This suggests that children have not yet learned the morphosyntactic conditions which block nonsubject relativization.

References
Chung, Sandra. 1976. An object creating rule in Bahasa Indonesia. Linguistic Inquiry 7.41-87.
Cole, Peter and Gabriella Hermon. 1998. Subject and Non-Subject Relativization in Indonesian., Second Symposium on Malay/Indonesian Linguistics, Ujung Pandang, Indonesia, July 12.
Musgrave, Simon. 2002. Non-subject Arguments in Indonesian. Unpublished Ph.D. Dissertation, Department of Linguistics and Applied Linguistics, University of Melbourne.
Sneddon, James. 1996. Indonesian: A comprehensive grammar. London: Routledge.