Phonemic distinctions and ESL students

Mi5 Mick   Sunday, September 19, 2004, 03:07 GMT
Just one small comment: One vowel sound utterance can be made to sound like any other. Likewise for consonants.
Mxsmanic   Sunday, September 19, 2004, 10:14 GMT
No, vowels are distinct even in isolation. Making "one vowel sound like another" is simply equivalent to pronouncing a different vowel.
Mi5 Mick   Sunday, September 19, 2004, 12:58 GMT
I was exaggerating by looking to closely at the earlier "Mary;Mary;Mary" wording, where each instance of "Mary" was made, through context, to sound like a different word, ie. merry, Mary, marry. At least this is how it is perceived by those who make the dinstinction (me for one) between each word, audially.
........................................   Sunday, September 19, 2004, 13:07 GMT
...........................................................................................
Tom   Monday, September 20, 2004, 14:07 GMT
I do teach the distinction between the three vowels, because they are all different in British English and I feel it is important to have a knowledge of British English even if you want to speak American English.

The original poster should remember that pronunciation is seldom taken seriously in ESL classes. 99.999% of English teachers probably don't even know the mary/marry/merry distinction is an issue. They certainly neglect much more serious points.
Mxsmanic   Monday, September 20, 2004, 19:30 GMT
I'm not sure what the utility of teaching the 3M distinction is if a majority of English speakers don't make the distinction, particularly since it is not phonemic (minimal pairs involving Mary vs. merry are extremely rare). Even without training, they'll recognize the words in all three forms; and unless they want to speak with a _specific_ pronunciation without an accent, there's no point in learning the distinctions that I can see.
Jim   Tuesday, September 21, 2004, 07:08 GMT
Mxsmanic,

How do you say that it's not phonemic? What's your definition of a "phonemic distinction"?
Tom   Wednesday, September 22, 2004, 01:27 GMT
Mxsmanic uses his own definition of a phonemic distinction.

Normally, we say that the difference between sounds A and B is phonemic if there are two different words X and Y such that the only difference between them is that X contains A and Y contains B in the same place.

http://www.antimoon.com/terms/phonemic_transcription.htm

For example, saying "d" instead of "t" in the word bet changes the meaning (the word becomes bed), therefore we say that "t" and "d" are two separate phonemes.

The "w" in "wet" and the "w" in "twice" are two different sounds, but there are no words where the difference changes the meaning, so we say that the difference is not phonemic.
Tom   Wednesday, September 22, 2004, 01:34 GMT
Mxsmanic,

Even if you're learning American pronunciation, it's useful to know what to expect when listening to British speakers. Besides, learners have to know [@], [e..(r)] and [e] anyway, so it's no big deal teaching them to pronounce ['m@ri(:)], ['me..ri(:)] and ['meri(:)].

Pronouncing "marry" as ['me..ri(:)] is silly anyway. I think you'll agree it's more logical to pronounce it ['m@ri(:)].
Mi5 Mick   Wednesday, September 22, 2004, 02:03 GMT
Mxs' consideration for "phonemic distinction" is in a complete phrase not in individual words. That's his own definition. Like he says, in 3M, it doesn't matter that each M word is pronounced the same way. However, you could find other phrases where similar word pairing are significant.
Sanja   Wednesday, September 22, 2004, 16:58 GMT
Actually, the "w" in "wet" and the "w" in "twice" sound the same to me and I pronounce them exactly the same. What's the difference?
Tom   Thursday, September 23, 2004, 00:50 GMT
The "w" in "twice" is voiceless, as described in http://www.antimoon.com/terms/phonemic_transcription.htm
Steve K   Thursday, September 23, 2004, 05:20 GMT
Any distinction between "w" in twice and wet is only in important to the teacher and the learner who is more interested in theory than in practice. Just listen to them and get close. My Cantonese speaking student says "trelve" for "twelve". I haven't the heart to mention voiced and voiceless. If most of my students could get Mary, merry ,and marry anywhere close to anyone of the three they would be doing fine.Get real!
Mxsmanic   Thursday, September 23, 2004, 19:21 GMT
A phonemic distinction is one that changes meaning.

If two distinct sounds exist such that exchanging one for the other in an otherwise unchanged utterance changes the meaning of that utterance, then the sounds form a "minimal pair," and they are _phonemes_. The distinction made by exchanging the sounds is a phonemic distinction.

Some minimal pairs are far more common than others, which means that some phonemes are much more important to learn than others. Additionally, the set of phonemes that exists for complete utterances in connected speech is different from the set that exists for isolated words. Minimal pairs that exist between isolated words may disappear when those words are part of complete utterances (because the context of the complete utterances removes any ambiguity or renders one member of the pair impossible in correct speech).

The distinctions between the 3Ms are phonemic in a few dialects of English and nonexistent in others. Even in the most significant cases, they form virtually no minimal pairs of utility. Therefore they can be ignored, unless one wishes to learn a _specific_ pronunciation in which they are clearly distinguished (even then, one is simply suppressing any accent, not improving comprehension).

The distinction between cot and caught is similar (although it is far more common and still made by most English speakers), as is the distinction between voiced and voiceless 'w'.
Jim   Friday, September 24, 2004, 01:08 GMT
Consider this sentence.

"The murder of Caesar was an historical event."

Compare it to the following.

"The murder of Caesar was a historical event."

They mean exactly the same thing. Is /n/ not a phoneme?

How about this one?

"The secretary went to Asia."

It could be pronounced

/TH.. sekr..tri(:) went tu: eiZ../

or

/THi: sekr..teri(:) went t.. eiS../

There'd be no change in meaning. Are /S/ and /Z/ the same phoneme? Is there no phonemic distinction between /u:/, /../ and /i:/? Is /e/ some kind of non-phoneme?

"Some minimal pairs are far more common than others ..."

Let's see how differently this can be pronounced ...

/s^m min..m.l pe..z a: fa: mo: kom.n TH@n ^TH..z/
/s^m min..m.l perz a:r fa:r mo:r ka:m.n TH@n ^TH..rz/
/som minimow pe..rz a:r fa:r mOur komon ven overz/
/s^mu minim^lu pe^zu a:ru fa:ru mo^ru ka:ma:n zen ^za:zu/

Have I changed any meaning? It might have got harder and harder to understand what was said but the clause never took on any different meaning.

Okay, how about "I thought a lot." verses "I fought a lot."? Quite different, right? Exchanging the /th/ for a /f/ makes a world of difference to the meaning. How about "I thought a lot about you, my love. I'm always thinking about you." verses "I fought a lot about you, my love. I'm always finking about you." Have we changed the meaning?

For that matter, how about "I thought a lot." and "I'm always thinking." verses "I sought a lot." and "I'm always sinking."? Quite different, again. Compare it to "I thought a lot about you, my love. I'm always thinking about you." verses "I sought a lot about you, my love. I'm always sinking about you." Have we changed the meaning?

In both cases we can tell by context that by "fought"/"sought" the speaker really means "thought" and by "finking"/"sinking" the speaker means "thinking". Take the context away and we're left with ambiguity. Are you telling us, Mxsmanic, that whether the difference between /f/ or /s/ and /f/ is phonemic depends on context? Speak to a Cockney first before you start talking mispronunciation.

How many minimal pairs can you find for /S/ verses /Z/? I can only think of two and they're a little obscure. How about /z/ verses /Z/ or /th/ verses /Z/ or even /ai/ verse /N/? Stumped? According to your "definition", Mxsmanic, it seems that if you can't find a minimal pair then the two sounds are not phonemically different.

Consider, on the other hand, the following question.

"Why did Adam eat the apple?"

We can easily change the meaning by emphasising different words.

"WHY did Adam eat the apple?"
"Why DID Adam eat the apple?"
"Why did ADAM eat the apple?"
"Why did Adam EAT the apple?"
"Why did Adam eat the APPLE?"

Each of those questions have a different meaning but aren't we using the same phonemes? How about these?

"The dog, which I despise, eats meat."
"The dog which I despise eats meat."

They mean completely different things (dependent verses independent clauses) but there is no phonemic difference between them.

How about giving us a definition that doesn't rely on the "meanings" of "utterances"? Minimal pairs are a good way to illustrate phonemic distinctions but they're not so great a way of defining them.