Re: Merging combining classes, was: New contribution N2676

From: Peter Kirk (
Date: Mon Oct 27 2003 - 09:58:27 CST

On 27/10/2003 07:28, Philippe Verdy wrote:

>From: "Peter Kirk" <>
>>So the logical order is
>><shin, sin/shin dot, dagesh, vowel, meteg>.
>>But the canonical order is
>><shin, vowel, dagesh, meteg, sin/shin dot>;
>>up to three (and in theory
>>more, at least in biblical Hebrew) other characters may appear between
>>the base letter and the dot which fundamentally modifies it.
>Ohh, I forgot the case of the dagesh consonnant modifier.
>But why would you like to encode the meteg before the vowel that it
>modifies? Couldn't it be encoded locally as well after that vowel like:
>1) The consonnant group: <shin or other base consonnant>, <sin/shin dot>,
>2) The first vowel group: <vowel>, <meteg or other accents>.
The issue with meteg is that it can occur to the left of (commonest), to
the right of and in the middle of the vowel. See sections
3.4 and 3.5. These three positions of meteg are essentially the same
character but the three positions have subtly different meanings. It is
effectively an independent issue from vowels and sin/shin dot.

>>From a Hebrew reader perspective, this logical order makes sense, as it
>consistently groups the letters in the order they are effectively modified:
>- One reads first the <shin or other base consonnant>
>- Then alters it into a sin letter with <shin or other base
>consonnant>,<sin-shin dot>
>- Then uses the alternate phonetic by adding the <dagesh>
>- Then recognizes the first "base" vowel sign
>- Then alters it according to the added accents
>You agree with me that using "combining order overrides" must be restricted,
>so that it won't be abused. The idea of using CGJ to encode them may be
>counterproductive, but one can simply avoid such abuse by creative such CCO
>control within each script, ...
>I see the general idea of CCO control characters as a general problem rather
>than something specific to each language (like Biblic Hebrew), and I see no
>reason why it could not be admitted and generalized with its own character
>property category.
I don't see any difference between your proposed generic CCO and CGJ. As
you say, the same function may be needed in several scripts, including
perhaps IPA which uses complex diacritic stacking. So why not simply use

Peter Kirk (personal) (work)

This archive was generated by hypermail 2.1.5 : Thu Jan 18 2007 - 15:54:24 CST