In message <9906080807.AA16059@unicode.org> you recently said:
> 1.) The are linguistically/phonologically two different holams:
> - Lo (letter and following an /o/ or an /ô/)
> - oL (/o/ or /ô/ before the letter)
> 2.) In some MSS and prints these different holams are
> indistinguishable when before or after a waw. On the much wider
> alef the difference is always visible. Both holams have the same
> shape but are placed differently.
I know next to nothing about Hebrew, but after pouring over an English book
on Hebrew grammer, could I ask your opinion of the following "rule" for a
holam appearing on the right side of an alef? (Assuming the user wants
A holam will move from the character before an alef onto the right side of
the alef if either
the alef has no vowel points, or
the alef has a sheva below it
[I've never seen an example of the second case, but include it for
My reasoning behind this is as follows:
A holam will move onto an alef if the alef ends the syllable (or word).
It does not move if the alef starts a syllable.
Since Hebrew syllables must start with a consonant, an alef bearing a vowel
point must be the start of a syllable. (Syllables are CV or CVC.) A holam
will *not* move on to it.
Silent alefs come at the end of a syllable and have no point attached. A
holam will move on to it.
An attached sheva can either represent the vowel schwa or indicate that the
consonant is the end of a syllable. However a long vowel before a gutteral
changes a vowel sheva on the gutteral to hataf patah. Alef is a gutteral and
holam represents a long vowel. So an alef will never bear a vowel sheva in
these circumstances and a sheva on an alef can only represent an end of
syllable, and a holam will move on to it.
The rule for shin is much simpler. (By shin which I mean the basic U+05E9.)
If the character before a shin has a holam, the holam will move onto the
right side of the shin. If the shin also has a shin dot (U+05C1) the holam
interacts with this typographically often vanishing completely.
If a shin has both a sin dot and a holam (an ordinary holam on the left, not
one that has moved from a preceding character) then the dot and holam again
interact with the holam usually vanishing.
My references don't explicitly mention holam jumping onto waw, but
inspection would suggest that an unpointed waw attracts a preceding holam.
Can you confirm this?
My personal inclination would be to make jumping holams a standard
behaviour, since the result still seems understandable by people not aware
of the fully correct convention. I would allow a zero width non-joiner to be
put before the alef etc to prevent it, if the user had strong views about
the representation. Is there an official standards body position on this
As for other languages using the Hebrew script, they seem to be unpointed or
to use Arabic vowel marks (therefore no holem to jump). The only exception I
can find is the rare Yiddish with redundant pointing. Here most alefs
represent a vowel and have a point on them, so no risk of jumping. Alef
without a point (shtumer alef) only seems to occur at the start of a word,
but is retained in composite words so there is a small risk of a prefix
ending in o coming before an unpointed alef. According to "The World's
Writing Systems", ed Bright and Daniels page 737, redundantly pointed Yiddish
is today only used by ultra-Orthodox groups, so they might welcome a jumping
-- Tim Partridge. Any opinions expressed are mine only and not those of my employer
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:46 EDT