Word susceptibility to phoneme changes

Gordon, Matthew J. GordonMJ at MISSOURI.EDU
Fri Nov 16 22:28:44 UTC 2012

You have an odd collection of pronunciation features here. I see a few that might qualify as diagnostic of regional accents in a broad sense (e.g. r-lessness, /ai/ monophthongization, cot/caught merger). But, you also have medial /t/ flapping (butter > budder), which is used by (AFAIK) all American dialects and several allegro speech characteristics (vowel reduction in function words like 'to'; him > im) and stylistic variables (final cluster reduction (and > an); ing > in') that would be regularly used by most Americans in everyday speech.

The article you cited offers a description of the Northern Cities Shift. Perhaps this is why you included "kids > keds", which is something that can be heard with the Shift, but why didn't you include the other elements in this shift?

If you want to perform this analysis more completely - and it's not clear to me what the benefits of doing so would be - you might consult some of the research that's been done on American dialects. A good place to start would be Wolfram & Schilling's book, American English, which contains an appendix of diagnostic pronunciation (and grammatical) features.

Matt Gordon
From: American Dialect Society [ADS-L at LISTSERV.UGA.EDU] on behalf of Tom Zurinskas [truespel at HOTMAIL.COM]
Sent: Friday, November 16, 2012 3:41 PM
Subject: Word susceptibility to phoneme changes

At least half the words on a typical text page in US English are susceptible to regional phoneme swaps or drops (accent or other shifts).   See http://screenr.com/BVH7   A link to copy the text is at http://justpaste.it/englishswapsdrops .  No one region would have all these affects.

The data show the vulnerability of words to various phoneme swaps or drops according to word frequency across US accents as related to word frequency (Collins Cobuild database).  This does not include word mergers like "gonna" (dropped i, g and t) "wanna" (dropped t) or "gimme" (dropped v) etc.  See http://tinyurl.com/divertingdialects  for a map of regionally shifting US dialects.  There are more phoneme affects listed there.

The data in the table aren't representative per item, because if a word were flagged for one affect it was not counted for a second.  So if "mother" were counted for a "th" to "d" swap it wasn't counted for an "ending r" drop.  To get an actual count by item another analysis need be done.

Tom Zurinskas, Conn 20 yrs, Tenn 3, NJ 33, now Fl 9.
See how English spelling links to sounds at http://justpaste.it/ayk

The American Dialect Society - http://www.americandialect.org

The American Dialect Society - http://www.americandialect.org

More information about the Ads-l mailing list