Arnold Zwicky's Blog

In yesterday’s NYT, a piece by Patrick Healy, “For 2016 Run, Scott Walker Washes ‘Wiscahnsin’ Out of His Mouth”, beginning:

Columbia, S.C. — Out on the presidential campaign trail, Gov. Scott Walker has left “Wiscahnsin” back home in Wisconsin. He now wants to strengthen the economy, not the “ecahnahmy.” And while he once had the “ahnor” of meeting fellow Republicans, he told one group here this week that he simply enjoyed “talkin’ with y’all.”

The classic Upper Midwest accent — nasal and full of flat a’s — is one of several Walker trademarks to have fallen away this month after an intense period of strategizing and coaching designed to help Mr. Walker capitalize on his popularity in early polls and show that he is not some provincial politician out of his depth.

Although Healy leads with pronunciation matters, they are not the focus of the piece, which is about how Walker is being coached in general on ways to make himself attractive to a wide range of voters.

Now on the main dialect feature in question, the Upper Midwest “flat a”.

The vowel at issue is the vowel of HOT (and, with the nasal stop n also involved, WISCONSIN. ECONOMY, and HONOR) in American English. Start with a chart of some relevant vowel sounds:

(The vowels in the two regions in the upper right are rounded; all the rest are not.)

The HOT vowel is the phoneme /a/, with different phonetic realizations in different dialects: for most American speakers, the realization is back [ɑ] (often called “broad a” by non-specialists), but for many speakers in the Upper Midwest (in Chicago as well as Wisconsin), the realization is central [a] (often called “flat a” by non-specialists).

The sociolinguistic point is that central [a] pegs you as coming from a relatively small geographical area, while back [ɑ] is much less specific; the coaching is then to make Walker sound less “provincial”. I’m not sure how effective this strategy is; certainly, many other politicians have achieved career success with notably regional linguistic features.

Oh yes, nasality. The fact is that vowels in general, especially low vowels, are significantly nasalized before nasal stops, especially n. The accented vowel of WISCONSIN is notably nasalized for everyone, regardless of its vowel quality; in fact, for a fair number of speakers, a lot of the time no n is actually produced, and nasalization of the vowel alone signals the involvement of nasality.

Nevertheless, [ɑ] speakers tend to hear central [a] before n as strongly “nasal” (while not noticing the nasality of their own [ɑ] before n). In general, people tend to detect the nasality of vowels before n when the vowel quality is not their own; for instance, raised variants of the phoneme /æ/ are routintely judged to be strongly “nasal” by speakers with lower variants of this phoneme (even though users of these lower variants have very substantial nasalization for their vowel).

I’m not denying the strength of these impressions that some vowels are strongly “nasal”, but I am suggesting that such impressions are sometimes not phonetically accurate, and instead are a kind of perceptual illusion.

This entry was posted on March 22, 2015 at 4:54 am and is filed under Dialects, Language and politics, Pronunciation. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.