Question about IPA: Is [a] a front vowel?

Lazar   Fri Aug 25, 2006 4:22 am GMT
As Travis pointed out (I assume he meant to type [a]), use of [a] in "hot" or "pot" is indicative of the Northern Cities Vowel Shift, or NCVS-influenced dialects, so that's why Chicago popped into my mind.
Travis   Fri Aug 25, 2006 6:39 am GMT
Yeah, I did mean to write [a].
Kelly K   Sat Aug 26, 2006 3:26 am GMT
Boberg (2000) sums up the result rather succinctly: the word stack is pronounced in Windsor, Ontario (indeed, in the rest of Canada) with the same vowel of the word stock as pronounced in Detroit, just across the river.
Zeph Agayo   Thu Jul 19, 2007 4:51 am GMT
Does someone know a site that has an illustration of Vietor's triangle? If someone does, please let me know. Thank you very much!
Guest   Thu Jul 19, 2007 5:09 am GMT
Does someone know a site that has an illustration of Vietor's triangle? If someone does, please let me know. Thank you very much!

P.S.
Please reply ASAP!
Kess   Thu Jul 19, 2007 8:33 am GMT
Trawicks   Fri Jul 20, 2007 2:29 pm GMT
[a] can be EITHER a front or central vowel. I believe it is actually, technically speaking, the only IPA symbol that represents two vowel positions. The reason is that the centralized [a] isn't distinguished in any language from any sound near it. It tends to be allophonic with either [A] (as in many GenAm speakers) or the fronted [a] (as with Spanish).
ivy lombendencio   Sun Dec 02, 2007 7:42 am GMT
i have this homework.. to draw the vietor triangle of 14 vowels but the page does only have the fivi fundamental vowel of sounds.. can you give me what im needing? thanks!!
Milton   Sun Dec 02, 2007 5:52 pm GMT
Dutch distinguishes between back and front a, and so does English (in Standard US English pop is /pAp/ (and not /pap/) but pipe is /paIP/ (and not /pAip/)...In many accents of WestCoast lost is /lAst/ but last is /last/.
{ lowering is much more frequent than /A/ rounding...
Travis   Sun Dec 02, 2007 7:42 pm GMT
Not all English dialects have separate front and back unrounded low vowel phonemes (/{/ or /a/ and /A/ or /A:/). Notably, Scottish English has a single phoneme /a/ for both. SImilarly, NCVS dialects shift historical /{/ towards being a front-to-central or open-to-close diphthong or a middle monophthong. While classical NCVS dialects unround /Q/ to /A/ resulting in two unrounded low vowel phonemes, there are those which do not, leaving a contrast of /a/ versus /Q/; for instance, a contrast between two unrounded low vowel phonemes only exists after /w/, where there has been a three-way phonemic split into /wa/, /wA/, and /wQ/ (even though many may in turn merge /wA/ and /wQ/).
Milton   Sun Dec 02, 2007 8:50 pm GMT
The people from Nevada will object to the nev/A/da pronunciation
but will not object to the nev/a/da pronunciation (/{/ shifted to /a/)...

in Reno, ''Nevada'' is many times pronounced with /a/ (Cali influence)
but never with /A/ (Southern and BackEast pronunciation of this word)


I don't know about ''Chicago'' /a, A, Q/ in Chicago...
Travis   Mon Dec 03, 2007 3:50 pm GMT
About the word "Chicago", the vowel in it seems to vary. In these parts of the Midwest, the vowel seems to be historical /Q/, which may actually be [Q] or [A] in various dialects here (as some unround historical /Q/ and some don't). If it had historical /A/ such would actually be [a] or even [{] around the southeastern coast of Lake Michigan, which I practically never actually hear.

However, in the rest of the US, the vowel seems to be historical /A/, which is actually [A] or in some dialects [Q]. However, I am not sure if the /A/ in the rest of North American English is a spelling pronunciation or is due to the reanalysis of [A] as being historical /A/ rather than historical /Q/, or partly both. As the NCVS is not that old, dating back to about the end of WW2, if this pronunciation is older than that it is probably a spelling pronunciation, whereas if it is newer than it then it likely is at least partly a borrowing with reanalysis.
Travis   Mon Dec 03, 2007 5:01 pm GMT
There are some words with seemingly anomalous /Q/ in them here. Outside of the placename "Chicago", the word that sticks out to me the most with respect to such is the placename "Waukesha", as it actually ends in /Q/ rather than the orthographically-implied /@/ in most dialects here (even though apparently the name is pronunced differently in Waukesha itself). There are some other cases like it as well, such as "Oconomowoc" and "Manitowoc", which has /wQ/ in its last syllable rather than the orthographically implied /wA/ (as most, but not all, less-frequent cases of historical or orthographic /wQ/ became present /wA/ here). However, variability due to the change from pre-father-bother merger /wA:/, /wQ/, and /wO:/ to present /wa/, /wA/, and /wQ/ here, though, is actually rather complex and ill-defined here, and I am not really sure as to whether any firm rules can be established as to whether a word with pre-father-bother merger /wQ/ would have present /wA/ or /wQ/ here.

On that note, that would either indicate that the lot-cloth split actually occurred at a rather late date (considering that Chicago was incorporated as a city in 1837), or, more likely, that the lot-cloth split has actually broadened, with words moving from the LOT lexical set to the CLOTH lexical set, since when it originally took place. Such movement of words from the LOT set to the CLOTH set would probably have taken place way of analogy or more random variability resulting from not being sure whether certain words have historical /A/ or historical /Q/.
Travis   Mon Dec 03, 2007 5:43 pm GMT
>>I've read a theory that the diphthonging of /{/ in the NCVS may have been started by the mixed settlement of the Erie Canal area, with New York City {-tensing and non-phonemic pre-nasal {-tensing mixing to produce a globally raised /{/.<<

That still doesn't answer the question of why the distribution of /A/ > /a/ extends further west than the global raising of /{/ and why /A/ > /a/ seems to actually be stronger than the raising of /{/ here in southeastern Wisconsin. Those two things make me think that the two were originally separate shifts that have coalesced over time, with dialects that already had /A/ > /a/ gaining the raising of /{/ and at the same time dialects that already had the raising of /{/ gaining /A/ > /a/ as a separate pull-shift.

The other thing that makes me think this is that younger speakers here tend to have higher /{/ (with less stressed /{/ often becoming [E_o] or [E]) than older speakers *but* do not have further fronting of historical /A/ in suit. This makes me doubt whether a true pull-shifting of historical /A/ is present here, because if such were present, then such younger speakers should also have a fronter historical /A/, when many such speakers seem to often have [a_"] sporadically and [A] is actually normally preserved here as an allophone of present /a/ in certain positions. Similarly, few words with historical /wA/ actually shifted to [wa], resulting in a phonemic split between /wa/ and /wA/, with /wA/ being far more common than /wa/.

That would also explain the extreme case of /A/ > /{/ in some Chicago-area dialects. If these dialects *already* had /A/ > /a/ before the raising of /{/, but also participated in the pull-shifting of /A/ with dialects to the east of Chicago, it would explain the extreme final position of historical /A/ in many words.
Travis   Mon Dec 03, 2007 6:54 pm GMT
>>Hmm, I've never heard that before; very interesting. Though that raises the question of why LOT started fronting in the first place.<<

Yeah, the classical explanation of /A/ > /a/ in NCVS dialects is that it is a pull-shift, but it just does not seem to work out around here, where it seems as if it really should be a pull-shift (but that would not explain the diphthongization of more stressed /{/) or that the /A/ > /a/ shift and /{/ raising/diphthongization are unrelated here.

As for why a separate /A/ > /a/ *before* the NCVS would have occurred, my guess is that it is either due to substratum influence or is just a sporadic shift. If it were due to substratum influence, it would have to be from German, as Norwegian and Swedish actually have /A(:)/ (and in the case of Norwegian, /{/). However, that would not be that strange, as even the areas of the Upper Midwest that are classically thought of "Scandinavian" actually have more German settlement than Scandinavian settlement overall (even though that could potentially be in part due to internal immigration as well).