Minimal pairs (was Re: Vowel for 'dad' vs 'mad')

Question

[nq:1]No, just more than one way to isolate phonemes. The set of phonemesfor isolated words is larger than it is ... than 100%, since there are many pairs that have no meaning (changing the first vowel, forexample, produces no minimal pairs).[/nq]
OK, you begin to make yourself clear. I've heard phonemes described as being defined by the existence of minimal pairs, but I long ago discarded this definition as useless. Whilst in your example changing the first vowel cannot produce another word, it nonetheless produces something which is not an English word at all, which will itself lead to misunderstanding. Moreover, is it not the case that by your analysis the schwa is not a phoneme?
This inspires a more general question (and thus a change of subject line). Can anyone think of a satisfactory minimal pair for /T/ (as in "thin") and /D/ (as in "this")? The best I've been able to come up with are "loth" and "loathe" which I find unsatisfactory because the first I would deem archaic, and also because the dictionaries claim that they are variants of the same word.

Mark Barratt

Usenet · Accepted Answer

--

Usenet · Answer

[nq:1]This inspires a more general question (and thus a change of subject line). Can anyone think of a satisfactory minimal ... the first I would deem archaic, and also because the dictionaries claim that they are variants of the same word.[/nq]
I don't consider 'loath' (as I would spell it) and 'loathe' to be archaic, just a bit formal/literary, so I think those work.

Here's a good one: teeth/teethe.

Usenet · Answer

[nq:1]I don't consider 'loath' (as I would spell it) and 'loathe' to be archaic, just a bit formal/literary, so I think those work. Here's a good one: teeth/teethe.[/nq]
Sooth/soothe (but I suppose 'sooth' might be considered archaic too).

Katy Jennison
spamtrap: remove the first two letters after the @

Usenet · Answer

[nq:1]OK, you begin to make yourself clear. I've heard phonemes described as being defined by the existence of minimal pairs, ... another word, it nonetheless produces something which is not an English word at all, which will itself lead to misunderstanding.[/nq]
Not necessarily; most listeners can guess the correct word.

The longer the utterance (in terms of number of segments), the smaller the set of segment strings of that length that make actual sense, and so the fewer the number of sounds or groups of sounds that are truly phonemic in that they form minimal pairs.
This is why people have a very high tolerance for mispronunciations of individual words in connected speech, but not when any of these words are spoken in isolation.
[nq:1]Moreover, is it not the case that by your analysis the schwa is not a phoneme?[/nq]
Absolutely. However, that particular phoneme has a great many allophones. It's like a placeholder.

Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

Usenet · Answer

[nq:1]This inspires a more general question (and thus a change of subject line). Can anyone think of a satisfactory minimal ... the first I would deem archaic, and also because the dictionaries claim that they are variants of the same word.[/nq]
Breath/breathe.

Lea V. Usin

Usenet · Answer

[nq:2]This inspires a more general question (and thus a change ... dictionaries claim that they are variants of the same word.[/nq]
[nq:1]Breath/breathe.[/nq]
You have an odd accent then...are you saying you use the same vowel for both?

Usenet · Answer

[nq:1]Breath/breathe.[/nq]
This isn't a minimal pair: /brET/ vs. /briD/. Two phonemes are different.

Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

Usenet · Answer

[nq:1]This inspires a more general question (and thus a change of subject line). Can anyone think of a satisfactory minimal ... the first I would deem archaic, and also because the dictionaries claim that they are variants of the same word.[/nq]
"Thy" and "thigh".
David

Usenet · Answer

[nq:2]This inspires a more general question (and thus a change ... dictionaries claim that they are variants of the same word.[/nq]
[nq:1]"Thy" and "thigh".[/nq]
"Either" and "ether".
Those are the canonical two pairs that you hear about in Intro Ling.

"Loth" and "loathe", like "teeth" and "teethe" and some others, are morphologically related, thus not a true phonemic minimal pair. There's a minor derivational morpheme that consists of voicing a final fricative when deriving a verb from a noun. Often a vowel gets changed, too, like "bath" and "bathe", "breath" and "breathe". A similar process occurs in the irregular plurals of "hoof", "path", "house", etc.

But in general, the contrast between /T/ and /D/ has the lowest functional load of any consonant pair in English. Not only are there almost no minimal pairs, but the ones that occur all (both?) have defects: "thy" is archaic, and "either" is pronounced /'ayD@r/ by a significant proportion of speakers.

In fact, the distinction, like the other voiceless/voiced fricative distinctions, is novel and was allophonic in many Middle English dialects. It used to be that there was a phoneme /T/ that was pronounced (D) in certain contexts, likewise a phoneme /f/ with a (v) allophone, /s/ with a (z), and /h/ with an (x). The (x) (spelled "gh", by the way), disappeared, but the others stayed around and eventually became phonemes in their own write.
Except for /T/~/D/, which are still written identically. This low functional load is the reason we can get away with that. People complain all the time about English vowel spellings, but nobody ever complains about "th" representing two phonemes; indeed, the fact that it does always comes as a surprise when the topic comes up in Intro Ling.

Usually we can tell by word class whether to pronounce "th" as /T/ or /D/. If it's a word in a closed class (pronoun, preposition, conjunction, determiner), the odds are high it's /D/. If it's in an open class (verb, noun, adjective), it's /T/. Unless there's some morphology involved, like "breathe" or "teethe". Indeed, with some words, they're in free variation, even in a single individual's speech. I never know whether I'm gonna say /wIT/ or /wID/ when I go to say "with".
-John Lawler http://www.umich.edu/~jlawler U Michigan Linguistics Dept "Our language can be seen as an ancient city: a maze of little streets and squares, of old and new houses, and of houses with additions from various periods; and this surrounded by a multitude of new boroughs." Ludwig Wittgenstein, 'Philosophische Untersuchungen'

Usenet · Answer

[nq:2]Breath/breathe.[/nq]
[nq:1]You have an odd accent then...are you saying you use the same vowel for both?[/nq]
OK, I guess that explains why no one had used such a common word as an obvious example. Psigh, I just got carried away with the /T/, /D/ change and ignored the rest. You should see some of the examples that I rejected just in time.
Cheers, Lea

Lea V. Usin

Minimal pairs (was Re: Vowel for 'dad' vs 'mad')

Top answer

Get the Weekly English Kit 📬