Jony Rosenne wrote on 07/02/2003 05:55:02 AM: > I would like to summarize my understanding:
I agree with you on most points, but would quibble on the first, as I find it overgeneralizes and is not explicit enough. > 1. The sequence Lamed Patah Hiriq is invalid for Hebrew. It is invalid in > Hebrew to have two vowels for one letter. It may or may not be a valid > Unicode sequence, but there are many examples of valid Unicode sequences > that are invalid. We need to state more carefully *what* is invalid. The facts are that spellings such as lamed patah hiriq *are* attested in literature and encoded representations are needed for them. These spellings are invalid as written representations of Hebrew that are consistent with Hebrew phonology; but their use in literature is not assumed to be consistent with Hebrew phonology; they are used *in spite of the fact* that they are inconsistent with Hebrew phonology. It is not normal for Hebrew spelling, but the literature to be encoded includes abnormal spellings, and they have as much need to be represented as the normal spellings. It appears to me that you are trying to establish invalidity of such sequences as a basis to argue that encoded representations should involve some character between the two vowels. I consider this reasoning flawed, however: the encoded representation is a representation of the *text*, not the phonology, and the text most certainly does include sequences such as lamed patah hiriq. It may be that we end up deciding to adopt an encoded representation for this that involves a character between the two vowels, but that is a technical-design choice, and not something that we are compelled to do because of the nature of the Hebrew language and normal conventions of Hebrew spelling. - Peter --------------------------------------------------------------------------- Peter Constable Non-Roman Script Initiative, SIL International 7500 W. Camp Wisdom Rd., Dallas, TX 75236, USA Tel: +1 972 708 7485
