Re: orthographic characters for glottal stop

Date: Tue Sep 07 1999 - 10:43:40 EDT

       PC>> Can
>> someone tell me, for languages that use an
>> right curly quotation mark to indicate a glottal
       stop, is the
>> Unicode character of choice U+02BC MODIFIER LETTER
>> (has category Lm)?

       JC>That's the Right Thing.

       That much seemed reasonably clear to me.

       PC> What about for those cases where the glottal is
       written using
>> the same shape as (European) digit 7? [...] I
       haven't found
>> anything in the standard that fits (and I think it's
       an option
>> to say, "change all your literature to use a true
       glottal stop
>> glyph").

       JC>I suppose you mean "I don't think it's an option"?

       Yes, of course.

       PC>> Do we need to add a character LETTER GLOTTAL 7 with
       PC>> category Lo and bidi property L?

       JC>I think that this 7 is just a glyphic variant (caused by
>fonts) of LATIN LETTER GLOTTAL STOP, just as 3 is a variant of
>LATIN LETTER YOGH and 8 is a variant of LATIN LETTER OU (in
>3.0). A suitable typewriter font might render GLOTTAL STOP
       with the
>same glyph as 7, but internally it should use the correct

       That seems reasonably obvious, and I don't know why I didn't
       think of that. However, there's still a question here: if the 7
       used for glottal stop is just a glyph variant of U+0294, then
       why wouldn't we also consider the same to be true in the case
       of the right singly curly quotation mark glyph used for glottal
       stop? I.e. why is it that, on the one hand, we encode the
       orthographic character that looks like ' and that typically
       represents the phoneme glottal stop as a separate encoding
       character using separate codepoint, U+02BC; but, on the other
       hand, we treat the orthographic character that looks like 7 and
       which typically represents the same phoneme as a glyph variant
       of the IPA symbol for the phone glottal stop. There seems to be
       some inconsistency here.


This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:51 EDT