Title of Work
The Ohio State University, Columbus, OH (hosted as a virtual conference)
According to Bauer & Benedict (1997: 419-420), descriptions of Cantonese published before the 1940s describe a contrast between the om/op and am/ap rime groups (ex: gom2, ‘thus, so’, 噉vs. gam2, ‘embroidered’, 錦). They say that “one presumes that these earlier scholars were making a distinction that actually existed in the Cantonese language of their time; nonetheless, it was one which eventually disappeared from standard Cantonese” (1997: 420). This presentation addresses two questions about this purported contrast by analyzing sociolinguistic interviews (spontaneous speech samples) from speakers born between 1922 and 1998: (1)Is there acoustic evidence of this contrast among Cantonese speakers born before 1940? (2)Does this distinction persist among any speakers born after 1940?Acoustic evidence comes from Lobanov normalized (Thomas & Kendall 2007) midpoint F1 and F2 measurements of 32 speakers (including 24 from Toronto and 8 from Hong Kong), recorded as part of the Heritage Language Variation and Change in Toronto Project (Nagy 2011). Words identifiable in Eitel’s (1877) dictionary as containing either om/op or am/ap were analyzed. This amounted to a grand total of 1321 usable vowel tokens (510 tokens of om/op and 811 tokens of am/ap). Pillai Scores (Nycz & Hall-Lew 2015) were calculated for each individual speaker. Speakers with a score of 0.300 and below are considered merged while speakers above 0.300 are considered not merged in vowel production (following a rough guideline set in Hall-Lew 2009).Results show speakers born before 1940 (n=4) with scores ranging from 0.444 to 0.783. Thus, the oldest speakers examined all produce a distinction. 15 of the remaining 28 speakers born after 1940 have scores above 0.300 and thus also produce a distinction. In fact, the speaker with the highest Pillai Score (0.855) was born in 1969. Nevertheless, results from a Pearson Correlation Test show a significant inverse correlation (r = -0.493, p < 0.01, n = 32) between Year of Birth and Pillai Score. Thus, there is still a gradual trend towards merger of the two rime groups.To conclude, this talk presents acoustic evidence of the am/ap ~ om/op merger in Cantonese spontaneous speech. The merger appears to have developed some time after 1940 and has since become a sound change still in progress (affecting both Hong Kong and Toronto Cantonese).
Tse, Holman, "The om/op ~ am/ap merger in Cantonese: Acoustic evidence of a not quite completed sound change" (2020). English Faculty Scholarship. 50.