On Dec 19, 2007 8:24 AM, Yahya <yahya@...> wrote:
> I have much more experience, however, of actual *speakers* than
> readers. Those who went to school (mostly or entirely) in Australia
> usually make the distinctions I've described (Baal/bawl/boll
> are pronounced in three different ways, and bawl is pronounced
> exactly as ball).
I make that distinction, too, and I think that's common for south-east
British English in general.
> Now, this could be where I'm going wrong. Forgive me if I seem
> obtuse, but isn't the point of writing in Shavian that it is an
> unambiguous phonetic transcription of a given speaker's speech?
I don't think so -- it doesn't have enough characters for that.
If you really wanted a phone_t_ic transcription, for example, you'd
need three separate letters for voiced unaspirated alveolar stop,
voiceless unaspirated alveolar stop, and and voiceless aspirated
alveolar stop, for the sounds of _d_un, s_t_un, and _t_un. At least
for some speakers.
I was under the impression that the point of writing in Shavian is
that it gives a _phonemic_ transcription -- that is, it makes only the
distinctions necessary which make a difference in meaning. (For
example, voiceless unaspirated alveolar stop and voiceless aspirated
alveolar stop don't contrast in meaning, since the sound depends on
the environment and they're both allophones of the phoneme /t/.)
Though it even falls short on the phonemic line for those who make
more phonemic distinctions than the dialect the alphabet was based on
(for example, those who distinguish between "horse" and "hoarse", or
"witch" and "which", or "weight" and "wait").
As for the "of a given speaker's speech", consider these paragraphs
from the introduction to _Androcles_:
In personal and intimate writing the forty-eight
(40 + 8) characters of the Shaw alphabet may faithfully
portray the pronunciation of the individual; but, as Shaw
pointed out, too eccentric a dialect may hamper, and
even destroy, effective communication. He considered
that, though there was no need to standardize writing if
not intended for publication, there was every need for
conformity in print; standard spellings being particularly
desirable when that print is itnended for circulation
throughout the English-speaking world.
In his Will, Shaw specified just such a standardization
for this play. He laid down for it a 'pronunciation to
resemble that recorded of His Majesty our late King
George V and sometimes described as Northern English'. [...]
But by all means _write_ as you think fit, and leave
experts to standardize printers' spelling.
Which I've interpreted to mean, roughly, that everyone can write in a
manner which they think best represents the way they speak -- someone
who pronounced "cot" and "caught" the same may use only one symbol,
likewise one who has the same vowel sound in "buck" and "book". While
those, on the other hand, who make the distinction in speech can do so
in writing as wel.
If spelling is to be standardised, however, then a good
standardisation would make a moderately large number of distinctions,
even if those distinctions are not made in this or that writer's
speech, and would use "the pronunciation [...] sometimes described as
Northern English". If a writer wishes to use this standardised
spelling, then, they could leave it to "experts" to edit their
spelling, or use a dictionary of that standard pronunciation for
phonemes about which they are unsure.
I think it was Hugh Birkenhead who pointed out that the AHD4,
available online, uses a pronunciation key that makes roughly the same
phonemic distinctions that "Northern English" does, and so could be a
suitable dictionary for a writer who wishes to use this standardised
spelling.
> Only the two pairs of these I've indicated are pronounced alike.
> Some few speakers will diphthongise bawl, Beale and bowl; most
> don't. For those that don't, we require Shavian to supply 13
> distinct phonetic transcriptions. Any fewer would mean that the tool
> is too blunt to use for its stated purpose.
And even 13 are not enough for those people who distinguish "bale" and
"bail"; as <http://en.wikipedia.org/wiki/Phonological_history_of_English_diphthongs#Pane-pain_merger>
says, those two sounds are the same for nearly all English speakers
but not quite all. But I think that the Shaw alphabet was not designed
to cater for _all_ possible distinctions used anywhere in Anglophonia.
> This doesn't mean that Philip can't pronounce three of these words
> (or even all 15 of them!) the same. The correct phonetic
> transcription for Philip's English probably uses fewer vowels than
> for mine.
Agreed, providing you mean "phonemic" here.
> I'm happy for Philip to use one letter to record his
> speech where I would need to use three different letters to record
> mine.
*nod*
Cheers,
--
Philip Newton <philip.newton@...>