Re: triple diacritic (sch with ligature tie in a German dialect writing document)

From: Richard Wordingham (
Date: Tue Jun 13 2006 - 16:30:45 CDT

  • Next message: Cristian Secară: "Re: Yahoo groups support for Unicode"

    Karl Pentzlin wrote on Tuesday, June 13, 2006 at 12:34 PM

    > Am Dienstag, 13. Juni 2006 um 09:33 schrieb Michael Everson:

    >ME> At 01:48 +0200 2006-06-13, Karl Pentzlin wrote:
    >>> ... I would propose a single "triple diacritic"

    > ME> Is it productive, or just used with sch in German?

    > At this time, I only know of sch.

    We've now seen the Greek examples. In general, I would expect it to have
    the same behaviour as *COMBINING TRIPLE INVERTED BREVE ABOVE. I've found an
    example of that used to illustrate French liaison (in Bodmer's 'Loom of
    Language'). Delimiting the letters associated by the 'inverted breve' with
    '(...)', we have examples such as:

          o(n e)(n a) pour so(n a)rgent.
          le(s e)nfants

    This is clearly productive usage. I was disappointed when I dug out this
    example I had partly remembered - I had though that the slurs (as I think of
    them) were below as in the 'sch' case.

    I has been going to suggest that instead of encoding <s, COMBINING TRIPLE
    BREVE BELOW, c, h>, we should consider interpreting <s, U+035C COMBINING
    DOUBLE BREVE BELOW, c, U+035C, h> as specifying the desired form. It seems
    that it is already spoken for - unless someone can successfully argue that
    they are glyph variants. The attraction for *encoding* is that it would
    have enabled arbitrarily long groups of letters to be grouped together
    without using bracketing schemes analogous to <U+1D177 MUSICAL SYMBOL BEGIN
    SLUR ... U+1D178 MUSICAL SYMBOL END SLUR>. I'm not certain that this pair
    can't be used - does it matter whether the slur is above or below?


    This archive was generated by hypermail 2.1.5 : Tue Jun 13 2006 - 17:46:32 CDT