University of Manchester
The file is open to annotations via Hypothes.is. Comments are always welcome!
In this talk I discuss the nature of evidence for the stratal organization of morphophonological grammar. Much — of course by no means all — of the currently available evidence in favour a stratal approach to morphology-phonology interactions comes from languages such as English, where the stratification of the lexicon has a strong historical rationale, or the Romance languages, where one can often easily observe morphological constituency at the level below the word. In this talk I discuss three cases that support the division between stem-level and word-level morphophonology as assumed in Stratal Phonology, from languages where the morphological rationale for stratification is less obvious. I consider evidence from Welsh svarabhakti, Irish vowel allophony, and the [e]~[o] alternation in Russian, and show that in all cases we observe not only cyclic misapplication effects but also properties more generally associated with the ‘lexical syndrome’.