And John Hudson encodes before. Clearly there is an ambiguity which needs to be resolved.What was the consensus in the SII on how the holam vav vowel should be encoded? And what is the normal practice in Israel and for pointed modern Hebrew? Is that holam encoded after the vav or before it?
After.
If holam is encoded after vav, it is difficult to make the typographical distinction when it is required. The renderer needs to look backwards. Basically, if vav-holam occurs after a consonant with no vowel point, it is a vowel and so should be rendered as a vowel with the holam to the right. But if it occurs after a vowel it is a consonant.
But consider the seventh word in Jeremiah 52:19, which, as you would encode it, ends qof hiriq yod zaqef-qatan vav holam tav. (This hiriq yod vav holam sequence is in fact unique in the WTS Bible text.) In this case, is the yod a consonant followed by a holam-vav vowel, or is the hiriq-yod a vowel followed by a consonantal vav and a simple holam vowel? (How would this actually be pronounced in modern Hebrew, with a V sound of not?) BHS prints holam above the right side of vav, implying that the holam-vav is understood as a vowel, and has a footnote that many manuscripts and editions have a dagesh in the yod, which makes this understanding unambiguous.
If a font is to make the rendering distinction which is required by the more careful typesetters, there need to be clear definitions of the contexts, and the font needs to be able to implement these decisions on the fly. This may be a bit much to ask. Certainly the renderer cannot engage in textual criticism as the BHS editors did! John Hudson's preferred encoding is also difficult to implement, but probably not so difficult.
On the other hand, we have to recognise that the vav holam ordering for holam-vav is in common use already, in Israel and in three already issued editions of the Hebrew Bible in Unicode. So there is a good argument for trying to implement this ordering in fonts, perhaps in addition to John Hudson's preference. And then holam vav and vav holam can perhaps be defined as compatibility equivalents (tricky because other marks may intervene in canonical order) and/or folded together for searches etc.
-- Peter Kirk [EMAIL PROTECTED] http://web.onetel.net.uk/~peterkirk/
