WHAT'S REALLY HAPPENING TO SHORT-A BEFORE L IN PHILADELPHIA?

The phonemic split of short-a into low, lax /ae/ and raised, tense /aeh/ is one of the distinctive phonological features of the Philadelphia dialect. Studies over the past 25 years have argued that /ae/ is in the process of being replaced by /aeh/ in words in which it appears before /1/, such as alley and personality, via a process of lexical diffusion. However, this article argues that the foregoing is a misinterpretation of the changes affecting /ae/ before /1/. A quantitative analysis of phonetic data in the Philadelphia Neighborhood Corpus, containing sociolinguistic interviews covering 40 years in real time and 100 years in apparent time, shows that /ae/ before /1/ is better described as having merged with /aw/; the raising of /aw/ toward the phonetic vicinity of /aeh/ then creates the illusion that /ae/ is being replaced by /aeh/ in these words. This can be taken as evidence against the proposition that lexical diffusion is a common mechanism by which regular sound changes go to completion, since a regular sound change that was thought to be an example of lexical diffusion is shown not to be.