Re: þe getisbyrg adres

On Wed, 2004-08-04 at 22:07, Alan Beale wrote:
> On Wed, 4 Aug 2004 17:58:32 +1000, Tristan Mc Leay wrote:
> > On Wed, 2004-08-04 at 17:21, Philip Newton wrote:
> >> Where did you get the words THOUGHT, LOT, FATHER, TRAP from? Is
> >> there
> >> a sort of standardised set of words to illustrate certain vowels?
> >>
...
> >>
> >> Or did you make up the example words yourself?
> >
> > AFAICT, they're just made up.
> >
>
> See http://www.ic.arizona.edu/~anth383/lexicalsets.html
Hm, interesting. I don't understand why it calls TRAP a broad a and PALM
a flat one. Totally backwards to what I'm used to (which makes perfect
sense because the PALM vowel is longer than the TRAP one :)---and I'd
never seen it before Mark J. Reed's post minutes before yours.

Mostly well-chosen I suppose, except I can't work out why 'idea',
'Korea', 'museum', 'real', 'ideal' are members of NEAR (American didn't
split them out, RP merged them in---but even if American *had* split
them out, it mostly seems that splittings get their own set for the
dialect of the split).

(Splits/Changes IMD:
  - STRUT into STRUT and PULSE; the latter are rounded to LOT (before
    l+alveolar in general)
  - BATH isn't a single group (I use the START vowel in staff, path,
    calf, half, TRAP in chaff and plastic, and DANCE in dance and
    prance).
  - CLOTH into CLOTH and GONE (containing only gone), phonetically a
    long form of LOT, /O:/
  - THOUGHT: all->GOOSE; bald, halt, alter, fault -> GOAT
  - oral -> CLOTH, aural -> THOUGHT (which feels unnatural, it's quite
    clearly a change made simply to provide a distinction).
  - CURE is more complicated. I think it probably represents a sound
    change in progress, except that it's been in progress some time and
    hasn't worked out how to sort itself out. Looking at the exampleset
    they have:
    - Words that never had a yod (boor, poor) or lost the yod in ancient
      history (sure), except 'tour' have the THOUGHT-FORCE-NORTH vowel.
      This suggests to me that at the time of this soundchange, tour was
      either not current in AuE (or it's ancestors) or perceived as
      foreign. This is dialectal and everyone's happy to say it about
      AuE, even relatively conservative sources.
    - Tour, as well as words that lost their yod earlier on (we'll say
      lj > l for simplicity's sake, but stress has something to do with
      it,* and rj, sj, zj, Tj > r, s, z, T might all have been early
      enough), retain a diphthong [u@)].** This may be dialectal or it
      could be less general, I'm not sure. (Note that /u/ and hence [u]
      in broad notation in AuE represents the FOOT vowel, not GOOSE.)
    - All words that retain the yod, as well as words that caused
      palatisation and lost the yod (i.e. tj, dj, sj, zj > tS, dZ, S, Z)
      retain a sequence [i\u\)w@]. This may be dialectal or it could be
      less general, I'm not sure. (Note that [i\u\)] or something as
      ugly is the realisation of the GOOSE vowel.)
    - In either case, [@] is replaced with [:] before an r, so that fury
      and lurid have monophthongs. I'm happy to say that lurid's
      monophthong is [u:]. In the case of fury, I think I'm more likely
      to say [i\:] or even [r\=:]; but if [u:] turned up, I probably
      wouldn't register it as different.

    So basically, not having a yod caused the first sound change, then
    having a yod caused the second. The ones that remained as-is lost
    their yod in between... One of the all-to-few cases of consonants
    before vowels causing/affecting the sound change in modern English.

*  Stress is also important in determining whether /sj/ > /s/ (early) or
   [S] (late).
** In all cases, be it in a diphthong or standalone, [@] at the end of
   a phrase(?---some level of speech relevant to intonation) is realised
   as [a].



--
Tristan  <kesuari@...>

Replies