Re: YAEPT: apparently bizarre 'A's (was Re: YEAPT: f/T (was Re: Other Vulgar Lat

As to what began this subthread, I have absolutely no idea what the
realisation of coda /r/ was at any of the times T0, T1, T2, T3 or T4.
I don't even know what times they were, except that if my hypothesis
is correct they happen in that order. I saw no reason to over-specify
the consonant, particularly because it's phonetic character wasn't
relevant. /r/ is perfectly suitable for any rhotic in a language with
only one and it's easier to read than /4/ or /r\/.

(The pronunciations I gave as T4 weren't even my own; I wouldn't
hypothesise an underlying /r/ IMD, and there's definitely no phonetic
realisation of it in the ones with word-final consonants! I make them
/ha:T h3:d 3:T sta: v3:s/ [ha_":T h8_+:d 8_+:T sta: v8_+:s], assuming
[_+] means "fronted".)

On 23/02/06, Mark J. Reed <markjreed@...> wrote:
> On 2/22/06, R A Brown <ray@...> wrote:
> > BTW you seem to have interchanged [ ] and / / throughout. The square
> > brackets [] denote _phonetic_ values; these are not language dependent.
> > the slashes denote phonemes of a given language.
>
> As your explanation and mine have opposite specificities, perhaps I
> should clarify further.
>
> A symbol between [..] refers to a specific sound, exactly as
> pronounced; you can be confident that wherever you see the same symbol
> between [...], exactly the same sound is meant, regardless of the
> language being discussed.
Actually, you can't be. Particularly with vowels, which vary over a
continuum, it would be completely impractical to have exactly the same
sound meant. The major difference is that when you're using slashes,
you're making a theoretical statement ("every different symbol refers
to a different phoneme, and every different phoneme has a different
symbol") whereas when you're using square brackets, you're making an
observation ("every different symbol refers to a different sound"). It
comes close to the same thing, but in many languages with only one low
vowel, [a] can be used to denote the low central vowel, rather than
using [ä] all the time.

Also, there's issues of focus. (YAEPD follows, IMD implied.) When
discussing the loss of aspiration when a voiceless stop follows /s/,
there's no need to point out that the vowel of "steep" is
substantially shorter than the vowel of "team" (or indeed the
diphthongal character of the vowel); we can say they're [sti:p] and
[t_hi:m] respectively. But if the issue happens to be the clipping, we
can say they're [sti:\p] and [ti:m].

...

> different dialects.  But similar considerations apply.  Theoretically,
> there is some underlying Original English and our various dialects are
> just surface realizations of this Ur-language. :)   So I might still
> write these words phonemically as /spOt/ and /pOt/, even though in my
> particular 'lect they are [spat] and [p_hat].
Well, there's no *historical* Original English. Any such "Ur-language"
will have a lot of distinctions that were *never* made in the history
of certain dialects, or come short of its goal. e.g. in
northern-English dialects & their ancestors, there has never been a
difference between "can't" and "cant" or "ran" and "Rann" or "put" and
"putt", but I distinguish them all (the vowels are /a: &: & &: U a/,
resp.).

--
Tristan.

Reply