(photo thanks to Alex; full size)
Interested in the Shavian alphabet? Come join us at the Shavian Wiki
where we're translating documents and building a lexicon as we go.
[ About | Fonts | Transliterator | Desktop | Keyboard | Firefox ]
If you can't see the letters here, install one of the Unicode fonts from this list and come back. I rather like "Androcles".
George Bernard Shaw was frustrated by the inconsistencies of the spelling system of the English language. When he died, he left a large sum towards the design of a new alphabet. His will stipulated that it had to have a one-to-one letter to sound mapping, and it had to be distinct from the Latin alphabet so that it didn't look like mispelled words and confuse people.
However, various people contested the will, on the grounds that you can't leave money to an idea. In the end, they only took away most of the money and gave it to libraries and so on; a little bit was left which was used to hold a competition to design the alphabet and print one book in it. I have a copy of this book and I think it was done rather beautifully:
(photo thanks to Alex; full size)
The winning alphabet was designed by Mr Kingsley Read. Kingsley Read had the idea of using a set of letters which rose above the line ("tall" letters), and then also turning them a half turn to make them sink below the line ("deep" letters):
饜憪=P 饜憫=T 饜憭=K 饜憮=F 饜憯=TH 饜憰=S 饜憱=SH 饜憲=CH 饜憳=Y 饜憴=NG Tall letters
饜憵=B 饜憶=D 饜憸=G 饜憹=V 饜憺=DH 饜憻=Z 饜憼=ZH 饜憽=J 饜憿=W 饜懀=H Deep letters
In general, tall letters represent unvoiced sounds (like 饜憫 "t" for tall) and deep letters represent voiced sounds (like 饜憶 "d" for deep). This makes it a lot easier to remember which is which. The only exceptions to this are the last two pairs shown, which don't represent sounds which have voiced or unvoiced counterparts in English: 饜憳 and 饜憿 (y and w) and 饜憴 and 饜懀 (ng and h). It has often been speculated that this was an error.
You might be happy to realise that 饜憲 is a 饜憫 with a 饜憱, which is in fact what the letter sounds like.
There are also "short" letters which have no tail or ascender, like most lowercase letters in the Latin alphabet. They include all the vowels, and four more consonants which don't have voiced/unvoiced pairs: 饜懁=L, 饜懏=R, 饜懃=M, 饜懐=N. Note that these pair off in an easy-to-remember way, and look a bit like vowels, since these letters and all vowels fall into the class of sonorants. The vowels are:
饜懄 饜懓 饜懅 饜懕 饜懆 饜懖 饜懇 饜懗 饜應 饜懘 饜懌 饜懙 饜懍 饜懚 饜懎 饜懛
You can find a key on Wikipedia-- the link is at the end of this post; I won't bother giving a key here because the Latin alphabet is really bad at representing the vowel sounds of English. [TODO: Now this page is getting popular, I should probably write something more here.] (It is worth noting, though, that 饜懄 and 饜懘 are very similar to one of the sounds made by the respective Latin letters of similar shape.) I would have to say "饜懄 is the vowel sound in 'if'", and so on. In fact, this is how the original chart was drawn up; the letters don't really have names in theory, but in practice people name them after the example words in the original chart. This means that, for example, 饜懀 gets called "ha-ha". Yes, honestly.
There are also a few ligatures. They represent a few pairs of letters that often appear together-- for example, 饜懣 is 饜憳 plus 饜懙. You don't really have to bother with them.
In addition, certain very short words are represented with a single letter: 饜懐=AND, 饜憹=OF, 饜憺=THE, 饜憫=TO.
There are no capitals or lowercase, but you place a mid-line dot before proper nouns: "饜憺 路饜憱饜懕饜憹饜懢饜懐 饜懆饜懁饜憮饜懇饜憵饜懅饜憫". (One way to look at this is that a dot represents a capital in the Latin alphabet but you don't use them at the start of sentences.)
People often say that Shavian is a problematic alphabet because it reverses letters and this is bad for dyslexics. I don't know whether people who say this know any dyslexics, but they never seem to level the same argument at the Latin alphabet. If you confuse 饜憫 and 饜憶, you will read "top" as "dop", which sounds a bit similar; if on the other hand in the Latin alphabet you confuse "b" and "q" the situation will be a great deal more hopeless.
Another objection to the use of Shavian is that English speakers from different parts of the world speak differently. But Shavian is not a phonetic but a phonemic alphabet: the letters represent the units which have meaning, not the actual sounds with which they're realised. (Consider that people from London consistently say "oo" for an L at the end of a word and people from elsewhere say "l", but nobody thinks that's a problem with English spelling. Or that the "c" in "cane" and "crane" is pronounced differently by almost all speakers. They are the same phoneme, the unit of meaning, not the same phone, the actual sound.) However, there are occasionally splits and mergers which give a speaker difficulty in, for example, distinguishing the sounds in "cot" and "caught", and this is a problem. It is also a problem for the Latin alphabet, of course.
At one point people used to represent Shavian using the Latin alphabet with special fonts (the "Latin mapping"). Now there is space in Unicode for Shavian. However, it uses a particularly unusual part of Unicode which requires the use of a technology called "surrogate pairs" to represent it. Some software does not yet support surrogate pairs, which is a bug in that software; other software works fine. Twitter allows it in tweets but not in searches. Firefox supports them but treats them as two characters which display as one. I hope these will be fixed soon.
There is a quasi-standard keyboard layout for Shavian called Shavorak. You can make at least Windows and Linux machines use it. I don't know whether there's anything to make the Mac type with Shavorak, but I wouldn't be surprised.
Shavian is designed to work around the shortcomings of English spelling with the Latin alphabet, yet a few people have tried to use it for other languages. 臏an 努esli Starling has created an Esperanto mode, and its application to Welsh is an interesting thought-- I have some notes on this which I won't bore you with-- though both of these are ultimately rather pointless because the Latin alphabet is almost perfectly regular for both languages.
If this has interested you, here are some more places to read:
饜懃饜懖 饜懐饜懕饜懃 饜懄饜憻 路饜懃饜應饜懏饜懐饜懗饜懐饜懅饜懁. 饜懖 饜懀饜懇饜憹 饜憯饜懏饜懓 饜憭饜懇饜憫饜憰. 饜懖 饜憵饜懄饜懁饜懓饜憹 饜懄饜懐 饜憺 饜憪饜懍饜懟 饜憹 饜憪饜懓饜憪饜懗饜懁 饜憿饜懟饜憭饜懄饜憴 饜憫饜懗饜憸饜懅饜憺饜懟 饜憫 饜懀饜懅饜懁饜憪 饜憿饜懗饜懐 饜懗饜懐饜懗饜憺饜懟. 饜懖 饜憵饜懄饜懁饜懓饜憹 饜懄饜懐 饜懄饜憭饜憿饜應饜懁饜懗饜憫饜懓, 饜憭饜懗饜懐饜憰饜懅饜懐饜憰饜懗饜憰 饜懐 饜懏饜懄饜憰饜憪饜懅饜憭饜憫. 饜憺 饜憮饜懏饜懓 饜憰饜懛饜憮饜憫饜憿饜懅饜懏 饜懃饜懙饜憹饜懃饜懗饜懐饜憫 饜懄饜憻 饜憹饜懅饜懏饜懓 饜懄饜懃饜憪饜懛饜懏饜憫饜懗饜懐饜憫 饜憫 饜懃饜懓. 饜懖 饜憯饜懄饜憴饜憭 饜懅饜憹饜懏饜懓饜憿饜懗饜懐 饜憱饜懌饜憶 饜憫饜懕饜憭 饜憺 饜憫饜懖饜懃 饜憫 饜懄饜懐饜憮饜懛饜懏饜懃 饜憺饜懅饜懃饜憰饜懅饜懁饜憹饜憻 饜懗饜憵饜懍饜憫 饜憺 饜懄饜憱饜懙饜憻, 饜懐 饜懁饜懌饜憭 饜懄饜懐饜憫饜懙 饜憿饜懗饜憫 饜憺饜懕 饜憭饜懇饜懐 饜憶饜懙 饜憫 饜懀饜懅饜懁饜憪. 饜懖 饜懏饜懗饜懐 路饜懙饜憵饜懗饜懐饜憫饜懙 饜懇饜憫 饜懀饜懘饜懃 饜懐 饜憿饜懟饜憭, 饜憵饜懘饜憯 饜憳饜懙饜憻饜懄饜憴 路饜懐饜懘饜懃 饜懄饜懐 路饜憿饜懅饜懁饜憲. 饜懃饜懖 饜憮饜懕饜憹饜懟饜懄饜憫 饜憭饜懗饜懁饜懟 饜懄饜憻 饜懛饜懏饜懗饜懐饜憽. 饜懖 饜懁饜懗饜憹 饜憲饜懛饜憭饜懁饜懗饜憫.
