Modeling Vowel Quantity Scales in Q Theory

Abstract

A growing body of research suggests that vowels vary in degree of strength. These strength differences are borne out in the degree to which these segments undergo or trigger phonological processes such as stress assignment or harmony. Traditionally, this variability has been accounted for through binary differences in phonological representations, such as presence or absence of a segment in the underlying representation, presence or absence of a phonological feature, and moraicity or non-moraicity of the relevant segment. While distinctions in underlying status and moraic structure are an effective tool for capturing some of the observed differences in vowel strength, they do not capture all attested differences. In this paper, we offer evidence supporting a four-point strength scale to which faithfulness and markedness constraints can refer. This model allows for strength differences among underlying and inserted vowels, and within monomoraic and bimoraic vowels as well, subject to scalar implications. We argue that Q-Theoretic representations offer the necessary representational tool to capture the full range of vowel strength.