Re: Arabic Joining Classes

From: Gregg Reynolds (
Date: Mon Jun 06 2005 - 11:57:48 CDT

  • Next message: John H. Jenkins: "Re: ku/ten syntax for kGB? in Unihan.txt"

    Andreas Prilop wrote:
    >>U+0649 (ALEF MAKSURA) is used in the Arabic Qur'an in the middle of words.
    > No, not U+0649 but U+0670.
    Those are distinct semantically and visually. In a Unicode
    representation of Quranic text, both will occur midstring.

    >>But it would be better to call it DOTLESS YEH. The
    >>_letterform_ DOTLESS YEH occurs in all four forms in written Arabic.
    > I do not believe this. In Arabic, initial and medial glyphs always have
    > two dots. (The letter with hamza has no dots.)

    Suit yourself. Notice I said "in written Arabic", not "in Unicode".
    Unicode is a very bad guide indeed for understanding written Arabic. The
    various HAMZA codepoints in Unicode involving waw, ya, alef, etc. have
    no place in the conceptual model of written Arabic. In particular,
    hamza on ya is precisely that, hamza with a ya "seat" regardless of
    where it occurs in a word. The same hamza in the same word may take
    different seats depending on word form.

    > So far, I haven't seen evidence/proof that *initial* and *medial*
    > forms of letter ya *without dots* (and without hamza) exit in Arabic.

    Keep looking. Both occur in the Quran.

    This archive was generated by hypermail 2.1.5 : Mon Jun 06 2005 - 11:58:29 CDT