ABSTRACT
Phonologically “strong” positions are traditionally distinguished by their special ability to resist featural neutralization phenomena that affect other positions (Trubetzkoy 1939; Steriade 1993, 1995; Beckman 1995, 1997, 1998; Casali 1996, 1997). However, there are certain phonological requirements that sometimes hold of strong positions specifically, including requirements that stressed syllables be heavy, that long vowels be high in sonority, that initial syllables have onsets, or that roots bear stress. This dissertation has presented a theory of strong position-specific requirements that accounts for those that exist while excluding those that do not. Specifically, it is proposed that certain substantive considerations determine when a markedness constraint can be relativized to a strong position, and that the substantive considerations in question are implemented as constraint filters that use extra-phonological information to screen formally possible constraints and determine which of them are to be included in Con.
