|
Clause reduction and agreement in Makassar
Dan Finer & Hasan Basri SUNY Stony Brook & Tadulako University
Languages of the southwestern peninsula of Sulawesi (Indonesia), exhibit a construction potentially analyzable as a species of ECM. Examples the alternation are shown in (1)-(4). Given that the absolutive clitic is attached to the main verb and not attached to the lower verb, the straightforward interpretation of this data is that the ECM’d argument corresponds to the absolutive argument in these examples (the definite direct object in (1b), (3b) and the subject in (2b) and (4b)). The examples in (5) and (6c), however, show the absolutive clitic of the main clause corresponding to the ergative argument of the lower clause, and here, the ergative agreement prefix remains on the lower verb. Note also the alternation in (6b-c). Either argument, therefore, is available to the ECM process, and this in turn suggests that neither argument is closer than the other to the position of the agreeing absolutive clitic attached to the main verb. Further, questions of multiple case checking arise, if it is assumed that the presence of the ergative and/or absolutive morphology correlates with checked case of the respective sort. I argue in this talk that (i) the relation between the position of the absolutive clitic in the main clause and the ECM’d argument of the lower clause is subject to the locality constraints that typify A-movement; (ii) the absolutive clitic is derived via phi-feature movement from the position of the corresponding argument to a position adjoined to vP (it receives second-position placement in PF (cf. Finer 2001)), (iii) the case features of both the ergative and absolutive arguments are checked in a multiple vP-Spec configuration (cf. Massam’s (2001) proposal for Niuean), thus accounting for the equidistance phenomenon with respect to feature movement mentioned in (iii) (cf. (6b-c)); (iv) these languages share with Niuean the property discussed in Bejar and Massam 1999 that accounts for multiple case checking (case features do not follow under movement, and only the top of the A-chain is spelled out).
Makassarese (1) a. ku-asse -i a kanayya na-cinik-ko i-Baso? 1E-know-3A that 3E-see-2A class-Baso? 'I know that Baso? saw you.'
b. ku-issek-ko na-cini? i-Baso? 1E-know-2A 3E-see class-Baso? 'I know Baso? saw you.'
(2) a. ku-asse -i a kanayya am-muno-ko bawi b. ku-asse -ko am-muno bawi 1E-know-3A that itr-kill-2A pig 1E-know-2A itr-kill pig 'I know that you killed a pig.' 'I know you killed a pig.' Bugis (3) a. u-isse?-i makkadae na-ita-ko i-Baso? 1E-know-3A that 3E-see-2A class-Baso? 'I know that Baso? saw you.'
b. u-issek-ko na-ita i-Baso? 1E-know-2A 3E-see class-Baso? 'I know Baso? saw you.' Selayarese (4) a. la-isse?-i lako attamburu-mo-i i-Basse? 3E-know-3A that itr-play.drum-prt-3A class-Basse? ‘They know that Basse? is playing the drum’ <idiom: ‘They know that Basse? is pregnant’>
b. la-isse?-i attamburu-mo-i i-Basse? 3E-know-3A itr-play.drum-prt-3A class-Basse? ‘They know Basse? is playing the drum’ <idiomatic meaning is preserved> Makassarese (5) ku-asse -ko nu-buno-i bawi-a (6) a. na-asse -i a kanayya nu-bau-a? 1E-know-2A 2E-kill-3A pig-the 3E-know-3A that 2E-kiss-1A ‘I know that you killed the pig’ ‘He knows that you kissed me’
(6) b. na-asse -a? nu-bau c. na-asse -ko nu-bau-a? 3E-know-1A 2E-kiss 3E-know-2A 2E-kiss-1A 'He knows you kissed me.' 'He knows you kissed me.'
References: Bejar, S and D. Massam, 1999 Multiple Case Checking. Syntax 2.2:65-79. Finer, D. 2001 Phases and Movement in Selayarese, AFLA 8, MIT. Massam, D. 2001. Fully Internal Case: Surface Ergativity can be Profound, AFLA 8, MIT.
|