Canadian Shift

From Infogalactic: the planetary knowledge core
Jump to: navigation, search

The Canadian Shift is a chain shift of vowel sounds found primarily in Canadian English, but also possibly in some other dialects (for example, younger Pacific Northwest English).[1] It was first described by Clarke, Elms and Youssef in 1995,[2] based on impressionistic analysis. The shift is structurally identical to the movement of front vowels in the California Shift of California English; whether this a coincidence or not is not yet clear.[1][3][4]

The shift involves the lowering of the tongue in the front lax vowels /æ/ (the short-a of trap), /ɛ/ (the short-e of dress), and /ɪ/ (the short-i of kit).

It is triggered by the cot–caught merger: /ɒ/ (as in cot) and /ɔ/ (as in caught) merge as [ɒ], a low back rounded vowel.[5] As each space opens up, the next vowel along moves into it. Thus, the short a /æ/ retracts from a near-low front position to a low central position, with a quality similar to the vowel heard in Northern England [a]. The retraction of /æ/ was independently observed in Vancouver[6] and is more advanced for Ontarians and women than for people from the Prairies or Atlantic Canada and men.[7]

However, scholars disagree on the behaviour of /ɛ/ and /ɪ/:

The Canadian Shift according to Clarke (red), Boberg (blue), and both (purple)
  • According to Clarke et al. (1995), who impressionistically studied the speech of a few young Ontarians, /ɛ/ and /ɪ/ tend to lower in the direction of [æ] and [ɛ], respectively: hence, bet and bit tend to sound, respectively, like bat and bet as pronounced by a speaker without the shift.
  • Labov et al. (2006),[8] through acoustic analysis of 33 subjects from all over the country, noted a backward and downward movement of /ɛ/ in apparent time in all of Canada except the Atlantic Provinces. No movement of /ɪ/ was detected.
  • Boberg (2005)[9] considers the primary movement of /ɛ/ and /ɪ/ to be retraction, at least in Montreal. He studied a diverse range of English-speaking Montrealers, and found that younger speakers had a significantly retracted /ɛ/ and /ɪ/ compared with older speakers, but did not find that the vowels were significantly lower. A small group of young people from Ontario were also studied, and there too retraction was most evident. Under this scenario, a similar group of vowels (short front) are retracting in a parallel manner, with /ɛ/ and /ʌ/ approaching each other. Therefore, with Boberg’s results, bet approaches but remains different from but, and bit sounds different, but remains distinct.
  • Hagiwara (2006),[10] through acoustic analysis, noted that /ɛ/ and /ɪ/ do not seem to be lowered in Winnipeg, although the lowering and retraction of /æ/ has caused a redistribution of backness values for the front lax vowels.
  • Sadlier-Brown and Tamminga (2008)[11] studied a few speakers from Vancouver and Halifax, Nova Scotia, and found the shift to be active in Halifax as well, although not as advanced as in Vancouver. For these speakers, the movement of /ɛ/ and /ɪ/ in apparent time was diagonal, and Halifax had /æ/ diagonal movement too; in Vancouver, however, the retraction of /æ/ was not accompanied by lowering.

Due to the Canadian Shift, the short-a and the short-o are shifted in opposite directions to that of the Northern Cities Shift, found across the border in the Inland Northern U.S., which is causing these two dialects to diverge: the Canadian short-a is very similar in quality to the Inland Northern short-o; for example, the production [maːp] would be recognized as map in Canada, but mop in the Inland North.

Related phenomena in the U.S.

Most U.S. speakers with the cotcaught merger appear not to undergo the Canadian shift in part because typically in the U.S. the merged vowel is less rounded and/or less back and slightly lower than the Canadian vowel, and therefore less room would be left for the retraction of /æ/.

In the Western U.S., one in four Western speakers in the Atlas of North American English exhibit the Canadian Shift, as defined quantitatively by Labov et al. based on the formant values for /æ/, /ɒ/, and /ɛ/.

The California Shift in progress in California English contains the features of the Canadian Shift; the two phenomena may however be different, since in California the retraction of /æ/ has occurred even though Californian /ɒ/ is more centralized and less rounded than Canadian /ɒ/.[12]

In Pittsburgh, where cot and caught are merged to a back rounded vowel, the mouth vowel /aʊ/ is traditionally a monophthong [äː] that fills the low central space, thus preventing /æ/ from retracting. Among younger speakers, however, /æ/ begins to move backward as the monophthongization of /aʊ/ declines.[13]

Durian (2008),[14] via instrumental analysis, found evidence of the Canadian Shift in the vowel systems of men born circa 1965 and later in Columbus, Ohio, a city located within the U.S. Midland. /ʌ/ is undergoing fronting without lowering, while still remaining distinct from the space occupied by /ɛ/. At the same time, historical /ɑː/ (the vowel in "father") is merged with the /ɒ/ class. This allows a "free space" for the retraction of /æ/, a possibility also suggested for Western U.S. dialects being potentially capable of showing the Canadian Shift by Boberg (2005). In Columbus, the Canadian Shift closely resembles the version found by Boberg (2005) in Montreal, where /ɒ/ and /ɔː/ are either merged or "close," /æ/ shows retraction of the nuclei (as well as "rising diphthong" behavior—i.e., ingliding with a lower nucleus than the glide), and /ɛ/ as well as /ɪ/ show retraction of the nucleus. (However, the retraction of /ɪ/ was not found among all speakers, and is more mild among the speakers that do show it than the retraction of /ɛ/ among those speakers. Also, the outcome of low back merger-like behavior is more like the California Shift outcome noted above than the rounded variant found in Canada.)

See also

Notes

  1. 1.0 1.1 Conn, Jeff (2002). "An investigation into the western dialect of Portland Oregon." Paper presented at NWAV31. San Diego, CA.
  2. Lua error in package.lua at line 80: module 'strict' not found.
  3. Hagiwara (2006)
  4. Lua error in package.lua at line 80: module 'strict' not found.
  5. Labov, p. 128.
  6. Esling, John H. and Henry J. Warkentyne (1993). "Retracting of /æ/ in Vancouver English."
  7. Charles Boberg, "Sounding Canadian from Coast to Coast: Regional accents in Canadian English."
  8. Lua error in package.lua at line 80: module 'strict' not found.
  9. Lua error in package.lua at line 80: module 'strict' not found.
  10. Robert Hagiwara. "Vowel production in Winnipeg."
  11. Sadlier-Brown, Emily and Meredith Tamminga. 2008. "The Canadian Shift: Coast to Coast." Canadian Linguistic Association (CLA 2008), Vancouver, BC. May 2008.
  12. Hagiwara (2006)
  13. Corrine McCarthy. "Language Change in Pittsburgh: The Decline of /aw/ Monophthongization and the Canadian Shift." Poster presented at NWAVE 33, University of Michigan, October 1, 2004. [1]
  14. Lua error in package.lua at line 80: module 'strict' not found.