Biblical Hebrew Forum

["Yitzhak Sapir" <yitzhaksapir AT gmail.com>]

Dear David,

It is very surprising to me that you refer me to the full quote in
Geoffrey Khan's article.  If
you do have access to his article, then just look up p. 54-55, where
he discusses the
schewa.  As for the ga(ya, if you read the extended context you'll see
that he is making the
same point that I am making.  This is a list, and I can't bring the
entire article, so I provided
the quote, without the extended discussion by Khan, but I definitely
did not misrepresent
Khan's views.  Khan himself is very adept and very resourceful, and
has access to a wide
variety of evidence, from various types of Genizah documents to
reading commentaries
in their Arabic and Hebrew originals, as well as knowledge of a large
body of scholarship.
After reading this and other follow-up articles (this article I had to
re-read some three
times before I felt all my questions were settled), I am convinced by
Khan's arguments and
I recommend you do likewise.

For direct evidence about the schewa, I suggest as a good place to
start Nehemiah Allony's
"Na(im veNaxim Biymei Habenayim" ("Mobiles and Quiescents in the
Middle Ages") published
in Leshonenu 12 (1942-3), p. 61-74, and referenced by Khan in his article:

He quotes from Diqduqe Hateamim:
p. 20, Sec 19: "These are the 'kings', [qamats], [patah], [sere],
[seghol], [holam], [shurek],
[hirik], [hataf and qamats], [hataf and patah], and so too the schewa
if it comes at
the beginning of a word will be counted as a 'king' but if it comes at
the middle or its
end, shall not."
p. 13, "And so too other letters that are not with dagesh and are next
to a schewa, shall
not be vocalized by the tongue, such as pnxs bn )l(zr, lqxw, $m(w,
yc)w, qr(w, zr(w, .."
A different version of the text: "If there is a ga(ya at the start of
the word, and the second
letter is a guttural )hx( and it is with one of the kings, shall be
vocalized as it is vocalized
as in w(wmsym (l hxmwrym, but if it is in the middle of the word shall
not be vocalized as
in $m(w, qr)w..."  He points out that although the examples are
defined by the guttural,
this describes the more extreme case and clearly if next to a guttural
the schewa is not
vocalized, then it is not vocalized next to a regular letter.
He quotes from the "Article on the Schewa", but notes that it is
parallel to section 50 in
Diqduqe Hateamim: "The schewa in the word )lkh, nlkh 'I shall go',
etc. is mobile only if
the next word has a dagesh such as )lkh ly Jer 5:5, nlkh $m 1 Sam 9:6.
 And if the
dagesh is absent so too is the schewa as in ylkw bh."  He adds that
from what is also
noted regarding the schewa which is mobile on the first of two similar
letters, whereby
examples are given such as hxwqqym, bwzzym, hwllym, which would not be
interesting
if the schewa after a long vowel was always mobile.  He also notes a
bit later regarding
a similar quote about the root hlk from Mahberet Hatijan, which states
that in the word
)klyh the kaf has a mobile schewa if the lamed has seghol except in
Qoh 5:10 where it
is quiescent.  From this, he notes, we can understand that if the
lamed is not with a
seghol, as in )wklym, )wkly, etc. the schewa must be quiescent. Elsewhere he
quotes the Tijan as saying "any sole schewa in the middle of a word is always
quiescent, as in zmry, m$(y, dbry."

Geoffrey Khan which I quoted before adds (p. 54-55), the grammar book
of Saadya which
does not include the schewa in the middle of a word after a long vowel
as a mobile schewa,
a work of Saadya (published in Leshonenu 44, p. 124) where the schewa
in words such
as hyth, (&th, is said to be quiescent.  Khan also refers to the
accent system.  According
to a rule of the accent system, tebir is preceded by darga if at least
a schewa and a full
vowel separate them, otherwise (if less than this separates them) a
merkha is used.  Thus
we have l) tklh Lev 19:9 but wyqm qyn Gen 4:8.  Following this rule,
the schewa in the kaf
of l) ywklw 2 Sam 17:17 is quiescent even though it is preceded by a long 
vowel.
Additionally, Qaraite manuscripts from the Genizah which sometimes
vocalize a transliterated
Biblical text using Arabic vowels, use a sukun for a schewa in these cases.

As for the issue about the pausal form of w$m(, I mean no general rule
about pausal forms.
We would like to make up laws and rules, and language is bound by
certain rules.  However,
each word has a unique history, and its development is specific to it.
 In the case of w$m(,
the ga(ya/non-ga(ya form differentiate pausal forms from non-pausal
form but this does not
mean that every pausal form will have a ga(ya.  However, it shows that
the ga(ya is not an
abnormal case, but rather a very normal and very regular occurrence,
where in this particular
word we can tell the rule that differentiates them.  Other words might
have different reflexes
of the ga(ya, so we can't generalize.  I don't have a specific verse
and chapter, I just looked
it up in the concordance which notes the different vocalizations based
on pausal and non-
pausal forms.

Incidentally, the specific case of vayomru is taken from Khan's
article, p. 54.

Yitzhak Sapir