Re: USAGE: Finnish and English vowels (was: Adapting non-Latin scripts)

> > >Umm... So does that mean a word like /kAt/ you'd consider to rhyme
> > >with a word like /pA:t/?
> >
> > No. The first has one /A/; the second has two. Nor would /pAAt/ rhyme
>with
> > /pAkAt/, since this has two syllabes; the other has only one.
>
>Yeah, I realised after I wrote that that things like stress will make
>that different; not even "thing" and "anything" rhyme in English
>because of stress.
Bizarre. They certainly rhyme to my ears. :) Stress is not crucial in
Finnish poetry, tho poets do prefer to rhyme (primarily or secondarily)
stressed syllabes with other stressed syllabes.


>(I think I've asked the question before, but I
>cannot recall the answer: In sung Australian English, you normally
>(certainly to the extent you can) put long vowels on long notes and
>short ones on short notes. I've heard in Finnish you just completely
>ignore length 110%. Is that actually the case?)
Checking the nearest songbook, I indeed don't notice any particular
correlation between the lengths of notes and syllabes.


> > trade-in of phonemic length for phonemic syllabicity. Here's a neat
>minimal
> > triplet:
>
>A "trade-off". (Trade-ins completely replace one thing with another.)
Uh? So how is a trade-off any different then?
Anyway, yes, I indeed mean to completely get rid of phoneMic lenght. (While
on the phoneTic level, length of course exists.) Maybe it's because of our
orthography, but long phonemes *are* seen fully equivalent to two short
phonemes in a row. You could make a stand for _suprasegmental_ length, but
even then, nobody thinks of long phonemes as unitary, as I see the /:/
symbol suggesting.


> > > > A syllabe-final glide interpretation works too, but then at least
> > > > an additional schwa must be positioned.
> > >
> > >(I also thought there's a set of diphthongs
> > >/ie/, /y2/, /uo/, which I can't see how you'd interpret them as
> > >having syllable-final glides.)
> >
> > That's where that additional schwa comes in: /i@ y@ u@/. Long vowels
>will be
> > /ij yv\ uv\ e@ 2@ &@ o@ a@/. It's something of a cheat, but it works,
>and
> > makes the system a nice regular 4x8 block.
>
>Youch! (Are the long vowels ever realised with /@/ offsets?)
No. ;) Some dialects however insert epenthetic schwas in various consonant
clusters, and this schwa always assimilates to the previous vowel, so
there's some sort of a precedent for this usage of /@/ ... altho I confess
never having seen anyone actually using the /@/ symbol for it.


> > -I should stress that this system is my own invention and would probably
>be
> > scowled on by professional Finnish linguistics.
>
>Yes; it strikes me as over-complicating the system; vowel length does
>the trick well-enough...
Except with /ie y2 uo/. Well, unless those are analyzed as /e: 2: o:/ (from
which they diachronically stem) and the modern [e: 2: o:] as /E: 9: O:/...

I have overengineering tricks for consonants up my sleeve, too; how would
analyzing [hAmp:u] as /hAn?pu/ sound? :D


>Confusion?! You were analysing "bird" as /"b@d/? That strikes me as
>more than a bit funny, considering /@/ a long vowel [;)]. Fact of the
>matter is, /@/ *does* occur only in unstressed syllables. You could
>variously (depending on context) unify it with /e/, /i/ and /A/ if you
>want to eliminate it, which has the advantage of being somewhat
>intuitive.
No, /b@rd/. I'd have problems with the difference (of rhotic 'lects) between
"India" and "indier" otherwise.

I still think that the unrhotic vowel traditionally transcribed as /3:/
sounds closer to /@/ than any of /I E V/ do; and schwa behaves like a long
(unchecked) vowel anyway, since it can end a word.


>Coming from a Finnish perspective (but biased by my native
>dialect), the Finnish vowel /2:/ seems the best choice for the vowel
>in "bird", even though it means we've got a vowel that's only ever
>long---but that accords with the nature of the language. Last but not
>least, there's the possibility of considering [@] to be /2/. Front
>rounded vowels and central vowels are somewhat similar phonetically,
>and seeing as the vowel is only used in unstressed syllables it'll
>lose parts of its quality anyway...
>
>(Then again, I suppose you were trying to be a bit outlandish---it
>just didn't seem all that much, considering again my native dialect :)
I could have tossed /2 2:/ in too - that's indeed what less educated Finns
tend to use - but since I was going for rhotic American English, I
restrained.


>[And if you were going for non-rhotic, then you forgot an equivalent
>of /e:/, /I@/ and /U@/.]
Those "look like"  /e@ i@ u@/. Or /e2 i2 u2/ if you want to use "rälli
traivör inglis", as ridiculously bad English pronounciation is known around
here.


> > Yeah. That works. Of course, without a central regulating body, it'll
>never
> > actually break into usage... I suspect English orthography won't get any
> > major revamps until it's already broken down into a full-grown language
> > family a few centuries from now. I hope to get at least minor fixes in
> > before that, however.
>
>You reckon we respell then? You are an optimist aren't you!
>
>--
>Tristan.
Heh, I fully expect at least some dialects to keep the traditional spelling
even after their <i>'s are pronounced (varyingly) /a e i a: e: i: ai ei je
ju/ and their <qu>'s /pf g Z/; but certainly some of the countries resulting
from the 2nd American Civil War of the 2300s are going to revamp, too. :)

John Vertical

Reply