Testing the implicit prosody hypothesis: Data from English

Abstract

One of the recent theories in sentence processing hypothesized that the prosodic phrasing implicit to a certain syntactic structure determines how speakers interpret the structure. According to this implicit prosody hypothesis (IPH; Fodor, 1998, 2002), in a structure of relative clause (RC) with a complex head noun, speakers/languages tend to prefer low attachment (i.e., an RC is modified by the adjacent head noun) if the prosodic break adjacent to RC is smaller than that between the head nouns. However, speakers/languages tend to prefer high attachment (i.e., an RC is modified by the nonadjacent head noun) if the prosodic break adjacent to RC is greater than that between the head nouns. Jun and Koike (2003) and Jun and Kim (2004) have shown that Japanese and Korean speakers who prefer high attachment produced a bigger prosodic break adjacent to RC than between the head nouns, confirming the IPH. The current study examines the prosodic phrasing of English sentences to investigate whether speakers of American English, known to prefer low attachment, would produce a larger prosodic break between the two head nouns than that between the RC and the adjacent head noun. Data from 30 speakers will be discussed. [Work supported by UCLA Faculty Grants.]