End of ADS-L Digest - 10 Jan 1994 to 11 Jan 1994 ************************************************ There is one message totalling 23 lines in this issue. Topics of the day: 1. consonantal /r/ ---------------------------------------------------------------------- Date: Thu, 13 Jan 1994 09:44:10 -0500 From: Ellen Johnson Subject: consonantal /r/ Back in June, Bill Kretzschmar and I gave a paper at the Assoc. for Literary and Ling. Computing using an analysis of postvocalic /r/ from the LAMSAS records as a demonstration. We distinguished /r/s in 2 ways: retroflex vs. non-retroflex (non- being those places, e.g. Deep South "lacking" postvocalic /r/) and vocalic vs. consonantal. The latter distinguishes between the "hooked-schwa" and the true consonant. Consonantal /r/ is found along the Eastern Seaboard in 2 areas in Pennsylvania. Our computer analysis validated the description of distribution in PEAS. The two areas have very different /r/s, however. The PA German-influenced area of SE PA is a "weak velar fricative", while the "apical" /r/ under discussion here is found in an area centered around Pittsburgh and extending to the west out of the LAMSAS region. Ellen Johnson ellenj[AT SYMBOL GOES HERE]atlas.uga.edu