Re: [hebrew] Re: Hebrew composition model, with cantillation marks

From: Philippe Verdy (
Date: Wed Nov 05 2003 - 16:15:54 EST

  • Next message: Philippe Verdy: "Re: UTF-16 inside UTF-8"

    From: "Jony Rosenne" <>

    > Philippe Verdy wrote:
    > > In fact, even spelling errors are often made on purpose, and
    > > wanted by their authors. What seems a spelling error at one
    > > time often happens to become a normal spelling for these
    > > words (look at the many abreviated forms of English written
    > > by American natives, such as "color" instead of "colour": I
    > > would not like that "color" be counted as 6 characters simply
    > > because a "u" is implied in the spirit of a British native,
    > > when the text was in fact written by an American).
    > >
    > > We are in a similar position with the so-called "missing" Yod
    > > letter in
    > > "Yerusalaim": the modern Hebrew reader attempts to imply
    > > something that was not intended in the original text by its
    > > author, and there's no reason to try adding a "missing"
    > > character in this text.
    > The original text of the Ketiv is without the Yod, the original text of
    > Qere is with the Yod, and it isn't marked in the margin because the reader
    > is supposed to know it - know, not guess or assume. It isn't implied. It
    > isn't unintended.

    If it's not in the written text, it is not implied by the writer.
    It may be implied by the reader, but we are not discussing about the implied
    meaning for the writer, which must not be encoded, but about the possible
    interpretation by the reader (which should not be encoded as well as this is
    not in the text!)

    So if the writer uses the original Ketiv, we must trust him and not imply
    the Qere that may have been used by other authors. If the writer uses
    the newer Qere, we must trust him too that it is not the Ketiv and the
    author wanted to insert its own interpretation of the Ketiv.

    I don't see any reason why a transcription of an original text to Unicode
    should change the intent of the author of using either form. Both versions
    of the same text may exist in various books from distinct authors, and they
    don't have to be unified with such sort of annotation. In fact waht you want
    is to allow a modern reader to read the ancient text with the rule of his
    newer language. If this is so, then the publisher will insert the necessary
    Qere marks (the actual mark, not an ambiguous ersatz) and will warn its
    reader that this annotation was not in the original text. In that case,
    there's no need of any new code to represent a missing letter.

    There are better ways to annotate a text without modifying it: it is very
    common in almost all publications that quote someone, and the margin
    space, calls of footnotes, or interlinear annotations can be inserted in
    the final document, but they come out of the normal flow of the original
    text, and they are made clearly distinct from it because they were not
    created by the original author. This is even a requirement for correct
    quotation of someone else, in some international ISO agreement about
    quotations and protection of intellectual rights in publications.

    This archive was generated by hypermail 2.1.5 : Wed Nov 05 2003 - 18:27:51 EST