Philip Newton (pne) wrote,
Philip Newton

Shorthand: alphabets and abjads and abugidas, oh my!

Duployan shorthand has been proposed for encoding in Unicode—primarily, I suppose, because it was used for writing Chinook Jargon and certain Native American languages, sometimes to the exclusion of other (e.g. Latin-based) writing systems.

I think it’s nifty that at least one form of shorthand might be encoded in Unicode.

Duployan is geometric, so something similar could be done for, say, Gregg, which is structurally pretty similar (lines and portions of ellipses).

Cursive shorthand systems such as German Unified Shorthand (DEK) or Stiefografie would probably be more difficult to encode, since in those, words are not formed by a succession of shapes which are simply concatenated, but instead, consonants are encoded by shape and vowels by the position of those shapes in relation to one another.

The document itself notes (on page 2) that Duployan is, at its core, an alphabetic (consonant & vowel) stenographic (simple line & curve) writing system (cf. Pitman shorthand, a stenographic abjad). DEK and Stiefografie would both be an abugida, I suppose: consonants have an inherent vowel (the default connection width and position encodes “e”), and different vowels are encoded by modifying the consonant.

There’s no graphic “vowel killer” like the virama, though: consonants are usually simply written very closely together if there is no intervening vowel, though in practice the distinction between “e” and “no consonant” is not necessarily made. But sometimes the connection changes if there is no vowel, so perhaps a virama could be useful; it would tell the rendering engine to use the special shapes (e.g. DEK “s” turning anti-clockwise rather than clockwise). I’m not sure whether special letters would be more appropriate, though.

For advanced forms of DEK/Stiefografie, which don’t simply connect units but involve things like starting higher or lower, writing one form above another, crossing through strokes, and other things, you’d need something a bit more involved than just encoding the consonants + combining vowels. It’s possible that the Shorthand Format characters proposed in N3895 might work for this—I haven’t read through the proposal properly (nor have I analysed DEK/Stiefo in this regard). I still think it’d be nifty if a unified DEK-Stiefo-Gabelsberger-etc. Unicode encoding could be devised, for scripts which are based in the same principles.

What I find nifty, though, is that alphabet-abjad-abugida all have equivalents in shorthand systems! Now all we need is a shorthand which is a syllabary. (Or perhaps one that’s based on logograms, though I suppose CJKV grass script [草書, 草书, cǎoshū, 草書(体), そうしょ(たい), sōsho(tai), 초서, chosŏ] is something along those lines.)

Really wild would be a pictographic one, but I really doubt that “pictographic” + “shorthand” go together at all—of the picture is detailed enough to be recognisable, it ain’t quick to write.

Tags: shorthand, stiefografie
  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded