RE: Missing Phonetic Symbols (A-M)

From: Peter Constable (petercon@microsoft.com)
Date: Mon Apr 25 2005 - 10:27:11 CST

  • Next message: Hans Aberg: "Re: Germa Umlaut (was: String name and Character Name)"

    > From: unicode-bounce@unicode.org [mailto:unicode-bounce@unicode.org]
    On
    > Behalf Of David Starner

    > My library has a book called the Phonetic Symbol Guide, by Geoffrey K.
    > Pullum and William A. Ladusaw (1986) listing all the phonetic symbols
    > they could find and their meanings. This is a list of the ones I can't
    > find in Unicode.

    I went through PSG a couple of years ago. You can find the results of my
    analysis at http://scripts.sil.org/PSGSymbolsVsTUS4. Note that I used
    the 2nd edition (you're using the first edition). Also, I haven't
    updated this for additions in Unicode 4.1 (maybe I should do that).

    > This isn't to imply they should be. (Many of the
    > characters I don't mention can't be found precomposed, but that
    > shouldn't be a problem in theory.)

    In my file above, I indicate character sequences where appropriate.

     
    > Reversed Turned Script A - Mentioned by Kurath (1939), for an
    > unrounded low back vowel. "Not in general use", "Extremely rare".

    In the 2nd edn, they renamed this Inverted Script A (a more sensible
    naming convention).

     
    > Barred B (with a line across the bowl): used by Pike and Smalley for a
    > voiced bilabial fricative. "still sometimes found: see e.g. the
    > transcriptions in Danesi (1982)." Also seen in Meillet and Cohen
    > (1952).

    This is a glyph variant of U+0180.

    > Barred D: voiced interdental frivative; Pike (1947), Smalley (1963),
    > Danesi (1982).

    This is a glyph variant of U+0111.

     
    > D with Upper-left Hook (hook attached to top of bowl): Used by Daniel
    > Jones to indicate a voiced retroflex stop, esp. Sinhalese. Replaced by
    > IPA usage in 1927. See Jones and Laver (1973).

    In the 2nd edn, they renamed this Front-Hook D.

     
    > Reversed E (lowercase, not 01DD): Suggested in JIPA 5 for a half-close
    > central unrounded vowel. Used by Abercrombie (1967) and Catford
    > (1977), Kurath (1939) and Trager (1964).

    This is U+0258.

     
    > Barred G (through loop): Pike and Smally for voiced velar fricative.
    >
    > Crossed G (through descender): Jespersen (1949) and Meillet and Cohen
    > (1952) for voiced velar fricative. (=01E5?)

    Both of these are glyph variants of U+01E5. (Note that they have never
    been used contrastively.)

     
    > Hooktop H with rightward tail: Prokosh (1939) for a voiceless fortis
    > velar fricative. "it could simply be a form of italic hooktop _h_ with
    > an exaggerated right leg."

    In the 2nd edn, they renamed this Right-Tail Hooktop H.

     
    > Heng: Yuen-Ren Chao (1934), used half-seriously for h and eng in
    > English phonemics. "Included here [...] because calling it _heng_
    > allowed us to devise a name for the otherwise unnameable <hooktop
    > heng>)."

    I've heard that this was used in one orthography for Tat, but I haven't
    seen clear evidence.

     
    > Hooktop Barred Dotless J: Palatal implosive, used by Ladefoged (1971)
    > and Maddieson (1984)

    This is U+0284.

     
    Re the glyphs that were at one time proposed by one author or another
    but not otherwise used, I concluded that these do not need to be
    encoded. These would include:

    Reversed Turned Script A (Inverted Script A)
    A-O Ligature
    Inverted Small Capital A
    Small Capital A-O Ligature
    Slashed D
    D with Upper-left Hook (Front-Hook D)
    Small Capital Delta
    Front- and Back-tailed Gamma
    Hooktop H with rightward tail (Right-Tail Hooktop H)
    Hooktop J
    Barred J

    There are others that I might consider candidates for encoding if I
    found them in use (though after doing some searching I didn't find
    reason to propose them): Slashed B, Barred C, Small Capital F, Heng,
    Crossed K .

     
    > I'm sure there's a few above that are actually in the charts
    > somewhere, but hopefully not too many.

    Why do you hope they're not in Unicode?
     

    Peter Constable



    This archive was generated by hypermail 2.1.5 : Mon Apr 25 2005 - 10:27:55 CST