Prosody as syntactic evidence

Abstract

A subset of Mayan languages feature “prosodic allomorphy,” a phenomenon involving morphological alternations at certain prosodic boundaries. In previous work, Henderson (2012) proposes that prosodic allomorphs in K’iche’ provide evidence for non-isomorphisms in the correspondence between syntax and prosody. In this paper, I argue against this view by building on a related extraposition analysis in Aissen 1992. I contribute novel data from prosodic allomorphy from two Mayan languages, Chuj and K’iche’, and show that upon further inspection, there is strong evidence for a syntactic analysis different from the one assumed in Henderson 2012. The new syntax leads to several predictions that are borne out, and crucially, does not force us to posit mismatches, allowing for a one-to-one correspondence between syntax and prosody. By taking apparent instances of mismatches as suggestive that the syntactic analysis must be revisited, the proposal aligns with work such as Steedman (1991), Wagner (2005, 2010), and Hirsch and Wagner (2015). Finally, I discuss how the proposal could be restated within phase theoretic approaches to the interface between syntax and phonology, concluding that Mayan prosodic allomorphy poses an interesting challenge for such accounts.