Shawalphabet YahooGroup Archive Browser

1. I had an idea today to save some time. When a page has several untransliterated words it's a nuisance to have to keep going and creating the relevant pages. So now the first twelve unknown words have little boxes at the bottom of the page where you can fill them in in bulk, if you like. The old way still works, too, of course. This saves quite a lot of time.

2, We have a new user, Baisong, who has contributed a lot of new documents. But he has also been modifying some existing words in the lexicon, many of which appear in _Androcles_: "marriage", "sorrow", "religion", "beauty" and even "with". I don't agree with many of his choices of vowels, but I thought I'd ask here for someone else to check his work out before I had a word with him about it.

http://shavian.marnanel.org/wiki?limit=50&title=Special%3AContributions&contribs=user&target=Baisong&namespace=0&year=&month=-1

3. On a similar subject, do you think we should have a policy that _any_ word from Androcles shouldn't be changed? We don't have much in the way of standard policies at all at present, but that should probably change.

4. What other documents would you like to see on the wiki? They have to be cc-by or public domain. Project Gutenberg is a good source.

Thomas

2009/4/11 Thomas Thurman <tthurman@...>:
> 2, We have a new user, Baisong, who has contributed a lot of new documents.  But he has also been modifying some existing words in the lexicon, many of which appear in _Androcles_: "marriage", "sorrow", "religion", "beauty" and even "with".  I don't agree with many of his choices of vowels, but I thought I'd ask here for someone else to check his work out before I had a word with him about it.
>
> http://shavian.marnanel.org/wiki?limit=50&title=Special%3AContributions&contribs=user&target=Baisong&namespace=0&year=&month=-1

I don't agree with many of his choices, either, and I think a word
with him is in order.

This ties in with your next point:

> 3. On a similar subject, do you think we should have a policy that _any_ word from Androcles shouldn't be changed?  We don't have much in the way of standard policies at all at present, but that should probably change.

Yes, I support such a policy. And I would suggest this policy, too:
for words not in _Androcles_, try to spell it the way _Androcles_
would (likely) have spelled it, by extrapolating from existing words
with similar sound combinations.

This means things such as "ado" in the unstressed -es and -ed endings
and "if" in most -y endings, including -ly. And also means
"ash"+"roar" instead of "air", for example.

I don't agree with some of the spelling conventions in _Androcles_ (in
the sense that they don't represent my personal pronunciation as well
as they could in all cases), but it's as close to a standard as we
have.

For that matter, I think that even _Androcles_ spelling is not what it
could have been; I got the impression that whoever fixed the spelling
would have liked to do away with apostrophes ("To help unpractised
readers, many apostrophes omitted in the orthodox version are restored
in this Shavian text; but any negative verb (dont, wouldnt etc) is
transcribed withotu an apostrophe"), but would spell (for example) 's
with an apostrophe since that's what _Androcles_ did, even if that
convention is for "unpractised readers" and was destined to be omitted
"eventually".

Another thing I'm a bit disappointed with is that "used" in "used to"
is spelled with "so"+"tot" but that "have" in "have to" is spelled
with "vow" rather than "fee"; I presume that this is due to note (1)
on spelling: "It is desirable that a given word should appear always
in a given spelling and not vary from time to time", though I'd argue
that "have" as the present perfect auxiliary and "have" as "possess"
on the one hand, and "have" in "have to" on the other hand, are
different words synchronically and should accordingly be spelled
differently just as much as "used" (ju:zd) and "used to" (ju:st) are.

But here again, I follow _Androcles_ and think that all contributors
to the wiki should also (or at least should try to).

If they prefer a different pronunciation, they can by all means make
their own wordlist. (Basing it on CMUdict would save a lot of work,
for example.)


So, in short: I think we should adopt the _Androcles_ spelling as a
standard (for words contained in it) and as a guideline (for words not
contained in it), and that contributors (such as Baisong) who
contribute words with divergent spellings should be advised to
contribute spellings more in conformity with _Androcles_, or
contribute elsewhere instead.

Cheers,
Philip
--
Philip Newton <philip.newton@...>

The lexicon of http://shavian.marnanel.org now comprises ten thousand words and is still growing. That's only about 7% of CMUdict, but it's still an impressive figure for a system that's only been around a fortnight. Thank you to everyone involved.

I have an idea that when it reaches a certain size-- I'm not sure what-- I'd like to take the lexicon, format it into a PDF (containing credits to all contributors), and run it through lulu.com to make a print-on-demand printed lexicon. Maybe it's even large enough for a first edition already. What do you think?

Thomas

re: further news from the wiki

http://shavian.marnanel.org/index.php/Main_Page
http://shavian.marnanel.org/transliterate

Remarkable.

If these attractive sites do not encourage more people to take an
interest in Shavian, its use and propagation, I don't know what will.

dshep

re: further news from the wiki

 Philip (11.04.2009) wrote:

> So, in short: I think we should adopt the _Androcles_ spelling
> as a standard (for words contained in it) and as a guideline
> (for words not contained in it), ...

Something to consider is the request of Shaw that his alphabet
be based upon Northern English. One of the characteristics of
the speech of northern England is the usual absence of the
trap/path split, the fronted, open /a/, almost a continental /a/
(close to
the cardinal vowel French 'la', and similar to Italian
'pasta') used in both categories of these words. While this is not
the same as the American ae, being more open, it is closer to
the American example than it is to the fully retracted /ah/ of RP
(cardinal vowel French 'pas').

The bi-alphabetic version of Androcles however is written in RP.
Examples: path, page 20, spelt (in the usual Shavian formatting)
/pyT/; danced, page 34, spelt /dynst/; clasp, page 135, spelt /klysp/.

Both the American and northern English versions can easily be
represented by the phoneme allophone (Shavian) /A/, while RP
demands Shavian /y/.

Shaw did not care for RP; he frequently railed against what he
called 'Oxford' English. I suspect he avoided the term RP because
the term 'Received Pronunciation' had been coined by a friend of
his, Daniel Jones, author of the "English Pronouncing Dictionary',
who stated clearly in the early
editions of his book that this
pronunciation was based upon that of the great "public' schools
of southern England, i.e. Eton et al, something which could not,
I doubt, have appealed to an old Fabian like Shaw, who wanted
the speech his alphabet was to represent to be readily accessible
to as many as possible. In modern times RP has been generally
adopted by many and perhaps most university students in England,
but still (according to J. C. Wells) is spoken by only about ten per
cent of the British population.

So, which is to be canonical, 'Androcles'--as printed--or Shaw's

wishes for the alphabet he sponsored?

Moreover, the use of /a/ in all trap/path words would eliminate the
greatest source of contention between the American and English
(and perhaps Australian as well) versions of our 'common' language.
In exchange, Americans could accept the short-o spelling of all
the /lot/pot/shot/ group of words (as printed in Androcles). This is
a reasonable compromise. 
 
dshep

2009/4/15 dshep <dshepx@...>:
>
> Both the American and northern English versions can easily be
> represented by the phoneme allophone (Shavian) /A/, while RP
> demands Shavian /y/.

Agreed.

> So, which is to be canonical, 'Androcles'--as printed--or Shaw's
> wishes for the alphabet he sponsored?

A good question indeed.

I could go either way. I'd prefer the trap-bath split since I use it
myself, but not using the split would, as you say, be more
representative of Northern English.

> Moreover, the use of /a/ in all trap/path words would eliminate the
> greatest source of contention between the American and English
> (and perhaps Australian as well) versions of our 'common' language.
> In exchange, Americans could accept the short-o spelling of all
> the /lot/pot/shot/ group of words (as printed in Androcles). This is
> a reasonable compromise.

And not merging father/bother is something I do feel more strongly
about than merging or not merging trap/path.

So if the community consensus is for merging (actually, not splitting)
trap/path, I'll modify my spelling of newly-entered words on the wiki
accordingly.

Cheers,
--
Philip Newton <philip.newton@...>

H desemvirF

...

0 JenerAlitIz 5 sFens

1 filosofi 6 teknoloJi

2 reliJion 7 Rt

3 sosFeti 8 literacur

4 lANgwaJ 9 histori

...

H wOrld'z noliJ iz dubliN everi 8 jIrz.

--jMnFted teknoloJIz....0

...

konsistensi, YlHO helpful in speliN n mAT iz not a sinonim fOr vircM.

--t.n.....1

...

prEr iz spIkiN t god. meditESon iz liseniN t god. trust trAnkwiliti

--r.t.....2

...

mF moHer jMsd t sE "tYk iz cIp," but Hen SI didnt hAv muc eksperiens in
hFriN lYjerz.

--r.kIler....3

...

H mOr spiricMal H eksperiens, H les RtikjMlat its ekspreSon.

--m.m....4

...

sFens iz An eksersFz v H hjMmAn brEn t grAsp H prinsipelz bF wic H
jMnivers wOrks n t rFt Hem dQn, if posibel, in krisp, presFs,
mATemAtikal termz.

--mAgnus pFk.....5

...

estAbliSd teknoloJi tendz t persist in H fEs v nM teknoloJi.

--gerit blQ....6

...

mAni v H grEt ilustrEtOrz v H pAst hAv ben HOz hM hAv kompFld, Over H
jIrz, TQzAndz v skecez v Ordinari TiNz arQnd Hem.

--k.g....7

...

literacur TrFvz on tAbMz, Just Az Yl Rt TrFvz on teknikal difikultIz.

--AnToni burJes....8

...

nekst t H rFter v rIl estEt Adz, H YtObFogrAfer iz H mOst suspekt v prOz
Rtists.

--donal henahAn....9

...

(c) desimus 2009....082PRL.... the consociation

.

A rough draft of a printed copy of the lexicon from the Shavian wiki is at /wiki/lexicon.pdf in the files section. I still want to go through it and take out all the obscure Biblical names and possibly some of the inflected forms, and proofread it all. I've caught a couple of errors this way alreay.

(I was also considering putting the Latin-alphabet words which occur in _Androcles_ in italics.)

Thomas

2009/4/16 Thomas Thurman <tthurman@...>:
> (I was also considering putting the Latin-alphabet words which occur in _Androcles_ in italics.)

I agree that marking those words somehow is good. (I probably would
have gone with a symbol - say, an asterisk - rather than italicising.)

Cheers,
--
Philip Newton <philip.newton@...>

The wiki's lexicon now stands at a very creditable 14,000 words. We now have the whole of _Alice's Adventures in Wonderland_ transliterated, which could also make quite a workable book.

We are currently considering some thorny policy questions about how words should be spelt both on the wiki and in any works produced from it, and I'd like feedback from anyone interested. (The numbers are from the wiki's policy document at <http://shavian.marnanel.org/wiki/Shavian:Policy>.)

3. We currently have a rule that all spellings in _Androcles_ are canonical, and set precedents. Should this rule be kept?

7, 8. Should syllabic -n or -r have "ado" before them? For example, should "bottle" be "bib on tot loll" or "bib on tot ado loll"?

10, 11. Should we retain apostrophes where they're used in the Latin alphabet? _Androcles_ appears to use them for possession ("beast's") but not for elision ("don't"). Two editors have suggested removing them for possession too.

There are other questions, but these seem the most pressing. Making choices about any of these seems a large step, and I'd like to hear feedback from the larger community on the subject before we do anything. What are your thoughts?

Thomas