Theoretical Implications of Directionally Asymmetric Transparency

Transparent segments have been a well known challenge for accounts of patterns long distance agreement, such as vowel and consonant harmony. Two standard ways to account transparency are autosegmental feature spreading with underspecification (e.g. Kiparsky 1981; Steriade 1987) Agreement by Correspondence (ABC; Walker 2000; & Rose 2004; Hansson 2001). Both, however, fail derive the multiple instances encountered in Tsilhqút'í­n (Cook 1993; 2013). Here, non-retracted dorsals act both transparent opaque process retraction, depending on which side trigger, retracted sibilant, they located. On other hand, sibilant harmony, sibilants forced agree retraction. I propose superset approach that combines ABC: All because outside correspondence relation. At first step derivation, not specified allowing them be regressive later step, negatively hence able block progressive