English vowel allophones... omfg

<b>Forum for the discussion of Applied Linguistics </b>

Moderators: Dimitris, maneki neko2, Lorikeet, Enrico Palazzo, superpeach, cecil2, Mr. Kalgukshi2

Post Reply
creztor
Posts: 3
Joined: Tue Dec 29, 2009 3:42 am

English vowel allophones... omfg

Post by creztor » Fri Mar 16, 2012 4:14 pm

OK, hoping someone can shed some light on the predicament I am in. I am basically "scoring" vowel pronunciation and have been told that in addition to "right" and "wrong" I need to score as "partially" correct. A partially correct answer should be a vowel allophone. Now, I can only assume that a vowel allophone fall into one of the six possible rules listed at the bottom of this page:

http://gaozhiyan.wordpress.com/2010/10/ ... ladefoged/

Basically, if someone said "bat" but the "a" was a little too long and sounded like "bad", am I correct in assuming that this would be an allophone of the short "a" in bat?

fluffyhamster
Posts: 3031
Joined: Tue Oct 26, 2004 6:57 pm
Location: UK > China > Japan > UK again

Post by fluffyhamster » Sat Mar 17, 2012 2:19 pm

Well, to mark items as wrong or borderline you'd need IMHO to demonstrate how not using a standard pronunciation would genuinely affect intelligibilty, impede contextualized communication. (Versus, as I suspect, testing single words, e.g. should that be and/or did the student say: But? Butt? Baht? Bart? Bert? Bet? etc).

creztor
Posts: 3
Joined: Tue Dec 29, 2009 3:42 am

Post by creztor » Sun Mar 18, 2012 2:46 am

Thanks for the post, fluffy. You raise a good point, but intelligibility isn't taken into account. Basically, I would like to know if the link above that points out the six rules is what vowel allophones are.

fluffyhamster
Posts: 3031
Joined: Tue Oct 26, 2004 6:57 pm
Location: UK > China > Japan > UK again

op

Post by fluffyhamster » Sun Mar 18, 2012 2:49 pm

Those may indeed be the rules for vowel allophones, but as your omfg implies, they hardly seem that helpful for your purposes, do they! ("Hmm, it's something to do with length and quality, as the actress said to the bishop").

I think like I implied before (regardless of whether intelligibility is taken into account) that you'd be better taking your set of test items and establishing how many have reasonably frequent minimal pairs (=sound substitutions that result in actual, certainly "dictionary word", differences in meaning (which is basically the opposite of allophony, which as we know is where differences [though this time only very slight ones] in phoneme sound~realization have no real impact on meaning)).

The easiest and most thorough way of doing this would probably be to construct a table (a 'phonemic inventory'?) with the e.g. RP consonantal phoneme(s) on or down the left hand side, and all the RP vowel phonemes arranged across the top (and with vowel-initial words you simply go from top to left rather than from left to top). You enter some test item(s) and then, with the help if need be of a learner dictionary, you fill in the table with at least some of the more likely vowel-based minimal pairs.

This would seem a more practical way of dealing with the issue of what exactly might NOT be (the) vowel allophones than "working" in the abstract from very generalized descriptions or supposed rules; that is, if your students~test takers clearly utter any of the minimal pairs on your table(s), you will then be in much more of a position to "categorically" state that they did not utter any acceptable or even halfway-acceptable vowel allophone (though like I implied in my first post with all those possible [mis]pronunciations of 'but', there could be a lot more leeway in what seems acceptable when the items are so decontextualized...which might call into question the value of composing a lot of~on these tables in the first place! "Bert you can't be serious!" - "That depends on if my name is Bert/Burt. Fortunately it isn't, Ernie, nor are many other people's, so we'll all be more than likely to interpret what you said as simply 'But...', despite your pronunciation being a tiny bit off according to the standard").

creztor
Posts: 3
Joined: Tue Dec 29, 2009 3:42 am

Post by creztor » Tue Mar 20, 2012 7:24 am

fluffyhamster, many thanks for the reply. You are absolutely correct that the rules for an allophone aren't very helpful, and the only reason I am marking something as partially correct based on an allophone is because it was "suggested" to me by someone higher up on the food chain than myself.

If I understand your post correctly, and I probably don't, are you suggesting that one would first see if what the student uttered matched any other "word" beside the target (eg, the target word was "bad", but what they uttered matches "bed"), and if it did then they clearly didn't utter an allophone but instead another word with a completely different meaning? If so, that does help immensely with making it more mechanical about how I can determine something as correct and incorrect before delving into the dark abyss of allophones. I'll look at how I can use something like that.

fluffyhamster
Posts: 3031
Joined: Tue Oct 26, 2004 6:57 pm
Location: UK > China > Japan > UK again

Post by fluffyhamster » Tue Mar 20, 2012 12:49 pm

Yes, you read my post correctly, Crez. :wink: It just seems a lot easier to say what isn't an allophone (i.e. to say what is rather one item in a minimal pair). But like I say, maybe even the "wrong" item from any set of minimal pairs will work, be acceptable, in certain contexts. (That is probably a bit too relativistic though, even in this "communicative" age LOL).

Post Reply