On 23/11/2011 20:24, Michael Everson wrote:
> The more I try to like it, the more I think that
> Roman-letter syllabaries are just a bad idea. Just use
> Tengwar. Then you can get nice and featural.

Tengwar is certainly attractive.  Building one's own
featural syllabary is even more fun.

I agree entirely that a featural syllabary is a much better
way of going about things if one is starting from scratch.

Why I use a Roman-letter syllabary for Brx is because of the
origins of the language.  It began life as trying to do
better what Reginald Dutton tried to do with his Speedwords,
i.e. to create a language:
1. which, when written alphabetically, can serve as an
alphabetic shorthand;
2. which can, if desired, serve as an international
auxiliary language.

As many on this list know, I don't take artificial IALs too
seriously; but, in theory, aim (2) is retained.

Dutton's first attempt at international alphabetic shorthand
was his 'International Symbolic Script' (ISS) of 1935. This
was not intended only to be read, not to be spoken.

His friend, the Rev. F.W.G. Foat, observed that the language
would have a better chance of catching on if it could be
spoken.  Dutton did this by changing the language and
adopting fairly traditional values for letter sounds but, to
keep word forms brief, he created quite complicated rules of
pronunciation with quite a few exceptions to the rules!

It seemed to me long ago that it would have better if he had
kept ISS and pronounced each letter as a syllable, i.e. a
Roman-letter syllabary.  That is where Brx has come from.

As Fuishiki Okamamoto and his Babm, I quote:
"... Roman letters are used in a quite different way from
existing European languages, namely the pronunciation of the
letters is specially determined for every letter
respectively, never changing its utterance without a
phonetic sign.  The respective utterance is the same as the
name of letters in Babm carefully chosen among most clear
voices, avoiding any similar sound, so that there is no
troublesome need to make some pronunciation by combining a
consonant with a vowel."

I must admit the second sentence is curiously worded; I
think I see what he means.

As for "Yemls, I leave that Jeff Jones to explain himself.

Jacques Guy, in his satirical 'Plan C', observed that Jeff
Prothero could have achieved his aim of "a pronunciation
scheme which makes all sequences of letters equally
pronounceable" by using "a strict CV(V)."  I show three
possible ways in which this could have been done:
http://www.carolandray.plus.com/Exp/Appendix1.html
http://www.carolandray.plus.com/Exp/March2006.html

There are, of course, other ways a syllabary of 16 CV
syllables could have been devised. Indeed, I point out that
Tengwar could have been used: "a stream of characters
(whether alphabetic or of binary digits or, indeed, any
other characters such as Tengwar" - see
http://www.carolandray.plus.com/Exp/index.html

Indeed, my suggestions for mapping the 16 bit-patters of
'Plan B' to a CV syllabary use the bits _featurally_ and
clearly a featural syllabary, whether base on Tengwar or
some other system, would be more appropriate.

But to return to Brx: the reason for adopting a Roman-letter
syllabary is historical.  It does present problems if one is
going to:
- produce a systematic syllabary;
- not depart too far from traditional uses of the letters.

It seems to me that compromises or fudges cannot be avoided;
the aim is to keep these to a, hopefully, acceptable
minimum.  It's a sort of intellectual exercise; I guess
that's what attracts me.  Thus I find it interesting to
discover how other people, who have attempted the same
thing, resolve these problems.

But, as I wrote above, I entirely agree that a featural
syllabary is a much better way of going about things if one
is starting from scratch

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