web hosting short URLs photo sharing

Shavian SYvWn SYvWn a new alphabet for E n g l i s h   FNliS   FNliS

Shavian is a non-Roman font developed to display phonetic English.  The rationale for Shavian is based on the fact that English is so inconsistent that the best phonemic transcriptions would mispell 60% of the words.  They would match the dictionary pronunciation guides but not the dictionary spelling.
The spellings for Shavian and English, however, do not have to be as unintuitive and odd as SYvWn and FNliS.  They could be `Sovian and `EGliS which is still odd but readable.  The transcriptions are odd because of the digraphs commonly used in English where an added letter is used as a marker for a different sound.  When the systematic script employs digraphs as in the case of RES, the transcriptions is more familiar Shawvian and English.

Shaw wrote:

Mark Twain wrote:
 

I have developed a notation called Unigraf that would provide a more readable ascii keyboard version of for the Shavian font.  In order to adopt the Unigraf keyboard locations, the some of the Shavian phonograms would need to be remapped.  Only about 16 charcter shapes would need to be relocated because most of the keyboard locations are intuitive and.  To make this work, one needs to remap the keyboard location of the Shavian shapes.

Pick up the ShawScript or Lionspaw font. The ones used here.
There are other Shaw Alphabet fonts:
 
 

I have developed a notation called Unigraf that would provide a readable ascii version of the Shavian font if it the location of the Shavian phonograms were altered somewhat.  To make this work, one needs to remap the keyboard location of the Shavian shapes.

Unigraf is based on TO not IPA.  This is to make it easy for TO readers.
Keyboard positions.  9 assignments look arbitrary.
 

Vowel Phonemes - with IPA identifiers  Arbitrary  key correspondences in red.
checkt |  free  |  difthongs  | R-combinations

a /ae/A@ q /aa/y     I /ai/F Y  qr R  Ir (are, ire)
e        R /3:/D     A /ei/E  x  er (air)
i        E /i:/I     Q /oi/q    ir (ear)CD looks lyk an ear ia W
q /o/o   o /o:/YB    O /ou/O    or (P) (aura)
c /u/UJ  C /u:/MW    U /ju/VU    Ur Va(your) [turned or lazy U's used for hook and hoop]
u /^/u   @ //a      V /au/Q    Vr (our)
K /ts/c  j /dzh/J    y /j/ j    zh /z:/ Z

ModifiedKeyboard Shavian - some odd assignments (no dash)
kEbPd SYvWn - sum Yd asFnments
kEbPd SYvWn - sum od asFnments.  Use W for hoop, J for hook.

A        a y      F       ar R   ire
A       y      F      ar R   yC--
e        D        E        x er X air
e       Darray?arE E        x  X
i        I        q        ir  C  ia W
i        I        q        ir  C  W
o        Y        O        or  P
o       Y        O        or  P
U        M kMl    V        ur The /u/ should be J short U
U        M kMl    V        Mr
u        a        Q        aur     7 lazy ^
u        a       Q AM
dh th    HT       ng N N G   KL

There are a few problems with Read's orthography.  a'r is listed as array rather than /3:/ There is no need for a ligature for array.  other  aHa,  her ha?  It is more complicated than it needs to be.  A fact
corrected by Read in his QuickScript.  Where did the key positions come from.  90% are logical,  the
remainder are arbitrary.  Perhaps you can complete the grapheme phoneme correspondence table that you use.

A         a              F    F                             dh/H    th/
e         D,  D        E                 x (air)
i           I               Q
o        _Y_            O         P
?       _M_           _V_
u          x             ___

I would have to make unigrafs of some of the r combinations to connect to the Shavian ligatures.

X@ pQnt iz Xat Xis nOtASN iz rEd@bl or at lEst mor rEd@bl Tan

/kristafD /kalumbas, an /itAlian-bPn nAvigEtD hM sEld in H sxvis v /spEn, iz komanli diskrFbd
Az H diskuvDD v H /nV /wxld - /amerika. YlHO kalumbas woz in sxc v a westwDd rMt t /EZa bF sI,
H diskuvDiz hI did mEk wx mP impPtant n vAljUabal HAn H rMt hI fEld t fFnd.

/kristafD /kalumbas, an /itAlian-bPn nAvigEtD hM sEld in H sxvis v /spEn, iz komanli diskrFbd
Az H diskuvDD v H /nV /wxld - /amerika. YlHO kalumbas woz in sxc v a westwDd rMt t /EZa bF sI,
H diskuvDiz hI did mEk wx mP impPtant n vAljUabal HAn H rMt hI fEld t fFnd.

kristafD kalumbas, An itAlian bPn nAvigEtD hM sEld in H sxvis v /spEn,

/krist@fR k@lumbus, an /itali@n-born navigAtR hC sAld in X sRvis v /spAn, iz kqm@nli diskrYbd
az X diskuvRR v X /nU /wRld - /@merik@.  olxo  k@lumbus woz in sRK v @ westwRd rVt t /AZ@ bY sE,  X diskuvREz hE did mAk w mor important n valyU@b@l Xan X rVt hE fAld t fYnd.

kristafD kalumbas, An itAlian bPn navigEtD hM

I would have to make unigrafs of some of the r combinations to connect to the Shavian ligatures.

X@ pQnt iz Xat Xis nOtASN iz rEd@bl or at lEst mor rEd@bl Tan

/krist@fR k@lumb@s, an /itali@n-born navigAtR hC sAld in x sRvis v /spAn, iz kqm@nli diskrYbd
az x diskuvRR v x /nU /wRld - /@merik@.  olXO k@lumbus woz in sRK v @ westwRd rVt t /AZ@ bY sE,  x diskuvREz hE did mAk w mor important n valyU@b@l xan x rVt hE fAld t fYnd.

Regards,

Bob Richmond           Needs the chart letrmatrix Sent but no comment.

I was interested to hear your proposal. I also find that the characters are
assigned to unusual keys; I have to try and cope with the strange
arrangement every time I write something in a Shavian font (and that is
quite often nowadays). If a reform of the scheme were to take place, now
would be the best time - before too much Shavian literature has been
written.

I would like to point a couple of things out, if I may:

1. You have used the '@' character for the shwa vowel in your scheme.
While this does help make Shavian text viewed in Roman clearer to read,
typing it is not so straightforward; the location of the '@' character on
keyboards often varies between just left of Enter to in between '!' and '£'
on the number row. Where is it on your keyboard? My point is that since the
shwa is the most often used character in Shavian, shouldn't it be in a more
'reachable', and, indeed, consistent place?

K&n olsO Uz "&" for the vowel in 'at'.

2. The current scheme is designed for typists.
One thing that was considered by the original designer was that people
learning how to type in Shavian would, most probably, already be fluent in
conventional English typing. The resultant character mapping was designed so
as to place as many letters in 'conventional' places as possible, which
means that short and/or common words written in Shavian often read very
similar when viewed in Latin (e.g. but, on, it, sin, men, if, fit, etc.).
This makes fast Shavian typing not so much a nightmare. It's true that your
scheme has similar traits; I'm only trying to say what I believe the
designer was originally aiming for. In fact, he probably did not consider
that anyone would have to be able to read Shavian text in a Roman font; it
could be seen as defeating the point of a different alphabet...

Unfortunately, I don't know an awful lot about fonts myself. I have never
successfully designed and used one; I leave it all to people who have the
know-how. :) If you want to refer your proposal to the original designer of
the Shavian character mapping, Ross DeMeyere, E-Mail him at:
ross@demeyere.com. He should be able to offer more feedback than I ever
could!

Many thanks for the E-Mail; hope to hear from you again!

Hugh Birkenhead

P.S. I'd also be interested to find out how you discovered Shavian... was it
through Androcles and the Lion?

----- Original May 11, 1999 Message -----
 

 Hugh,

 My one gripe with the existing Shavian fonts is that insufficient effort
went in to developing the underlying phonascii and key positions.
 

This has to be changed @ V for   I am going Vp to the  tV nU
can leave u with up.  J V U for u u: and ju.
Where available @ would be replaced with the ae ligature.
Where available x would be replaced by eth
Should V be for up or for hoop, hVp.  w=oo?
x cW gOz mV, the c@t sez mU.
 
 

LIONSPAW lFnzpY

SHAWSCRIPT lFnzpY

LIONSPAW lFnzpY

 Vowel Phonemes - with IPA identifiers
 checkt | free  | difthongs | R-combinations

 a         q /aa/       I  /ai/    qr   Ir   (are, ire)
 e         R /3:/      A /ei/     er  (air)
 i          E /i:/       Q /oi/     ir  (ear)
 q /o/    o /o:/       O /ou/    or (P) (aura)
 c /u/    C /u:/      U /ju/      Ur (your)      [turned or lazy U's used
for hook and hoop]
 u /^/    @ shwa   V[au]     Vr   (our)

 Perhaps you can complete the grapheme phoneme correspondence table that
you use.

 A         a              F                                 dh/H    th/
 e         D, x          E
 i           I               Q
 o        _Y_            O         P
 ?       _M_           _V_
 u          x             ___

 I would have to make unigrafs of some of the r combinations to connect to
the Shavian ligatures.

 X@ pQnt iz Xat Xis nOtASN iz rEd@bl or at lEst mor rEd@bl Tan

 /kristafD /kalumbas, an /itAlian-bPn nAvigEtD hM sEld in H sxvis v /spEn,
iz komanli diskrFbd
 Az H diskuvDD v H /nV /wxld - /amerika. YlHO kalumbas woz in sxc v a
westwDd rMt t /EZa bF sI,
 H diskuvDiz hI did mEk wx mP impPtant n vAljUabal HAn H rMt hI fEld t
fFnd.

 /krist@fR k@lumbus, an /itali@n-born navigAtR hC sAld in X sRvis v /spAn,
iz kqm@nli diskrYbd
 az X diskuvRR v X /nU /wRld - /@merik@.  ----- k@lumbus woz in sRK v @
westwRd rVt t /AZ@ bY sE,  X diskuvREz hE did mAk w mor important n
valyU@b@l Xan X rVt hE fAld t fYnd.__

 Of course, one can adapt to any crazy code.  The objective is to make it
as intuitive as possible.
 With a correspondence table, I believe that I can learn the underlying
ascii code for the Shavian
 font.
 
 

 Regards,

 Steve