Re: No Vowels?

Jörg Rhiemeier wrote:

[Plan B]
>>It most certainly has 16 consonant phonemes. Whether it has 8 or 16
>>vowel phonemes is a matter of interpretation. If phonemic status is
>>given to /r/, then we have 8 vowel phonemes plus 8 combos of /r/ plus a
>>vowel. The rule is then that syllables must be of the form: C(r)V(C)
>
>
> I did not remember that the "16 vowels" included 8 /r/+vowel combos.
> I'd say that /r/ is a phoneme, so there are 17 consonants, one of them
> being /r/ that patterns fundamentally different than the other 16, and
> 8 vowels.
Yep - I think in summary if describing Plan B _phonologically_ one would
say that it has:
- sixteen consonants that function at syllable boundaries, i.e. /p t k b
d g f T s S v D z Z m n/
- eight vowels, namely /i: I ej E aj A ow u:/ *
- a 'glide', namely /r/ (alveolar approximant) which can occur only
between a consonant and vowel.

*it will be seen that the 'vowels' contain diphthongs as well as simple
vowels, so one could derive different _phonemic_ analysis, e.g. the
first six above could be analyzed as /ij/, /i/, /ej/, /e/, /aj/, /a/ -
that is as three simple vowels plus three diphthongs with /j/. But that
would then leave /ow/ and /uw/ a bit anomalous.

>>
>>>this is probably a better analysis than saying
>>>it had 16 phonemes each with a consonantal and a vocalic allophone.
>>
>>That IMHO is a simply a ridiculous analysis!
>
>
> Concurred.  It is very far-fetched, and if someone was to analyse
> a natlang that way, he should ask himself what he is doing.
Most certainly!

 >>Yep - and IMO in view of Plan B's odd use of each quartet to represent
 >>either a consonant or a vowel, it would have been better to use the
 >>plain ol' hexadecimal digits, i.e. have the "alphabet": 0 1 2 3 4 5 6 7
 >>8 9 A B C D E F.
 >
 > Yes.  That would not really look like a pronounceable *language*,
 > but Prothero's sequences of consonants of which every other is
 > to be pronounced as a (seemingly) unrelated vowel doesn't, either.

Quite so. His use of traditionally consonant symbols is IMO unhelpful in
that sometimes they are pronounced with sounds associated with those
symbols (tho pronouncing _h_ as /T/ and _l_ as /D/ is a bit strange),
but at other times they are pronounced as vowels or diphthongs. It is
just plain crazy IMO to pronounce _c_ as /ej/ and _d_ as /I/.

At least if they were the hex digits it is a tad more obvious that the
symbols map to quartets of bits and that they will unusual rules of
pronunciation.

 > (Nor does the all-consonant orthography of X-1, I have to admit.)

At least the consonants keep values which are consonantal and similar to
their traditional pronunciation in most languages. It is just that built
into the system is a method of determining the unwritten vowels.

Yes, I know that using only consonant symbols in the orthography might
lead to the naive assumption that the language is vowel-less. But that
really is a naive assumption. Well might just as well claim that any
language written with an abjad, e.g. Arabic or Hebrew, is vowel-less!

[X-1 and my 'experimental loglang']
[snip]
>>>I have to concede that that project quickly lost momentum soon after
>>>I started it.  Well, I am more interested in naturalistic diachronic
>>>artlangs.  Don't expect X-1 to go beyond that outline anytime soon.
>>
>>i know the feeling - the same has happened to my 'experimental Loglan'   :)
>
>
> It is a well-known (to me) mistake of mine to start all sorts of
> projects even though I have enough to do already, most of which
> soon fall by the wayside.  And X-1 is one of those projects.
Oh - I do still intend to complete my 'experimental loglang' sometime -
if only to discover what its name is    :)

[snip]
>>>and that the orthography was phonemic; but it is probably more
>>>reasonable to analyse it as having 7 consonant phonemes /p t k s m n l/
>>>and 4 vowel phonemes /E i O u/.
>>
>>Basically I agree, tho I think I would include the 'zero consonant' as a
>>phoneme, i.e. 8 consonant phonemes. This simplifies the rule that all
>>syllables must be of the type CV
>
>
> That's a legitimate analysis, though I'd avoid it - I am not all that
> comfortable with zero phonemes.
The so-called 'zero consonant' of my experimental Conlang can be [j]
before front vowels or [w] before back vowels, i.e. it does have
phonetically realized allophones, so I think it is legitimate to refer
to it as a phoneme. However, in my latest revision I have given it the
phonemic symbol /ɰ/ and it may be:
- a velar approximant in any position;
- a palatal approximant before front vowels;
- a labiovelar approximant before back vowels;
- silent.

[Phonemic & phonetic vowels]
[snip]
>>One does need to be very careful how we define vowel & consonant. These
>>terms have somewhat different meanings depending upon whether we are
>>talking in terms of phonetics or of phonology (Pike suggested using the
>>terms 'vocoid' and 'contoid' when speaking phonetically, reserving
>>'vowel' and 'consonant' for strictly phonological meanings).
>
>
> Yes.  Phonetics is not phonology, and phonology isn't phonetics.
Quite so.

[snip]
> And orthography is neither phonetics nor phonology.  But many people
> fail to realize.
Indeed - but not, one would hope, on this list.

>
>>Didn't we discuss this on the Conlang list in February this year?
There was a "Why consonants" thread.

--
Ray
==================================
ray@carolandray.plus.com
http://www.carolandray.plus.com
==================================
Nid rhy hen neb i ddysgu.
There's none too old to learn.
[WELSH PROVERB]