Re: IPA and Unicode

From: Richard Cook (
Date: Thu Sep 17 1998 - 12:38:00 EDT

Michael Everson wrote:
> RC:
> >> "small-capital omega", for a rounded mid back vowel;
> Not sure I've seen this.

This is the char. used by, e.g. Bloch and Trager (1942, 22) for a
"mean-mid back rounded" vowel, written in IPA as lowered "o" or raised
"backward c".

In the STEDT Font Reference this is:
"round mid-back vowel"; at decimal 152

> >> "Long-Leg Turned Iota", (see Pullum + Ladusaw, 1996:89);
> Isn't this a dotless j (not a UCS character)?

This is a so-called (by Sinologists) "apical vowel", the more front
unrounded one. It's shape is basically that of a "dotless i", the bottom
of which descends slightly below the baseline, the head of which curves
up to the left and then down. The form given in P+L is slightly
different from versions I've seen printed in China, which are seriffed,
and actually sit on the baseline.

In the STEDT Font Reference this is:
"syllabic alveolar fricative, plain"; at decimal 189

> >> "Right-Tail Turned Iota", (see Pullum + Ladusaw, 1996:90);

This is another "apical vowel", the more back unrounded one. It is
basically a small backward "esh".

In the STEDT Font Reference this is:
"syllabic retroflex fricative, plain"; at decimal 180

In addition to these two unrounded apicals, there are also symbols for
two rounded apicals:

In the STEDT Font Reference this is:
"syllabic alveolar fricative, rounded"; decimal 243

The second rounded one, "syllabic retroflex fricative, rounded" is not
in our font, but has a similar shape to the alveolar, with the addition
of a bottom hook.

Sinologists use all four of these "apical" symbols as extensions to IPA.
I do not recall seeing the "syllabic retroflex fricative, rounded"
symbol in Tibeto-Burman transcriptions, which is perhaps why it was
omitted from STEDT font.

I should perhaps emphasize that the omission of a character from STEDT
font does not necessarily imply that a particular symbol should not be
supprted. As I mentioned in a private reply off-list, part of the
motivation for standardizing one version of our
character set is that characters have been jammed into every possible Macintosh
codepoint, which inhibits porting the font to other platforms. In this
jamming process, bargains have been made with the devil, symbols
occasionally (though I think rarely) sacrificed to the greater good.

I should like to work with people interested in this problem, to
assemble a list of characters currently or formerly used by Asian
linguists for phonetic transcription, to determine exactly which are
present in the standard and which ought perhaps gain consideration for inclusion.

Martin Heijdra wrote:
> In fact, Richard's characters are exactly the ones I once talked about,
> characters which are used as extensions (NOT obsolete versions) to IPA, which
> is deemed by many Tibeto-Burmese linguists as incomplete.
> At that moment, I did collect material of proof of widespread use, official
> guidelines, Chinese encyclopedias, their use in Chinese typesetting programs,
> and started checking them against Unicode, in order to officially report them;
> Ken and Joan send me information on how to do that.
> What bogged me down, and what made my job unfinished, is that once I looked
> into the Chinese typesetting programs, I did not find only the four or five
> symbols widely used, but more a collection of some 100 IPA and non-IPA
> phonetical symbols accessible. I started to map those to Unicode, but found
> this difficult in many cases, partly intrinsically, partly because the
> printing in my samples seemed to be multiinterpretable.
> I think I 'm writing this in the hope that Michael Everson will say, hey, send
> what you found to me, and I'll take it from there...
> Martin Heijdra

It seems in fact that Martin Heijdra <> may have
done some of this work already, and this being the case, his collection
should be cross-referenced against the STEDT font and these against
currently accepted IPA extensions.

Martin- How may I have a look at your collection? Where you have found
questionable forms, perhaps you could simply note these as problematic.
Please don't let a few questionable forms stop you from reporting your
results. Have you or could you perhaps put these together as a GIF?

Thanks to all who responded to my initial query. You've given much to
think about.


This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:41 EDT