In this talk, I reconsider the design of Philippine-type syntax drawing on new evidence from Formosan infinitives, the pronominal clitic system common in Philippine-type languages, and two similar voice systems observed in western Nilotic and Caucasian. I discuss how these observations indicate Philippine-type Austronesian languages may be viewed as agreement-based and discourse configurational featuring overt A-agreement that tracks the grammatical role of topics and relativized phrases. In this view, the Philippine-type \u2018pivot-only\u2019 extraction restriction is essentially not an extraction constraint, but A-agreement morphology that indexes the grammatical role of the relativized phrase. Finally, comparative evidence from different Philippine-type languages identifies two loci of variation: (i) the presence versus absence of overt topic movement; (ii) whether or not \u03c6-features of the topic are spelled out on the verbal complex. This variation informs us about the relation between Agree and Move on the one hand, and about the relation between syntactic Agree and morphological agreement on the other hand.<\/p>\n\n\n\n