Biblical Hebrew Forum

["Vadim Cherny" <VadimCherny AT mail.ru>]

> >Gemination is, of course, simply post-tonic (Calanni - stress shift
> > and the resulting gemination).
>
> I don't think so.  I refer you again to the table:
>
>  http://www.exc.com/JoelHoffman/Excerpts/ITB-p95.pdf
>
> The Masoretes record RIVKAH and ZILPAH, but the LXX gives us REBEKKA
> and ZILFA.  Either the LXX transliteration is inconsistent or the
> Masoretic rendering is inaccurate.

Joel, you are not right. There are simple phonetical differences between
Rivkah and Zilpah, which account for the LXX difference.

Kaf in Ribkah is plosive. Pey in Zilpah could be aspirated.
Bet in Ribkah is plosive. Lamed in Zilpah is not.

First, let us deal with the difference in the first vowel. Both names have
hirek in Hebrew. In Greek, hirek in open syllable became e (R/e/bekka),
while hirek in closed syllable remain i (Zil.pha).

Option 1: schwa was silent, stress on the final.
Syllabification, then, runs: Ri.bkAh, Zi.lpAh.
Try saying Ri.bkah with really plosive kaf. You are losing bet: Ri.'kAh. If,
however, you try to pronounce bet and preserve kaf as plosive, you need
epenthesis: Rib(e)kAh. Plosive bet works in the same direction: it also
forces epenthesis. Which influence was stronger, kaf's or bet's, we don't
know. Greeks heard that as RebekAh, changed to RebEkah to conform to their
stress pattern, and geminated the post-tonic to RebEkka.
Zi.lpah also has plosive pey, and epenthesis *Zil(e)pAh was one option to
prevent lamed from falling out, Zi.'pAh. A
simpler solution, however, was to aspirate pey, ZilphAh. This is how the
Greeks heard it.

Option 2: schwa was silent, stress on penultimate.
Epenthesis is driven by plosive bet more than by kaf: RIbkah - RIb'kah - 
RIb(e)kah -
Greek RebEkka.
No need for epenthesis in ZIlpa because lamed is not plosive. In unaccented
syllable, pey lost its "plosive force" and acquired considerable aspiration.
(Compare to /C/ain versus /H/ettura.) Thus, Zil.pha.

Option 3: schwa was vocal.
Not very plausible option. Hebrew had a long history as spoken language by
the time of LXX. From other languages (even French) we know that schwa
becomes silent very fast, in a few centuries.
Your other examples from LXX suggest that what is now silent schwa was also
silent then. Few examples to the contrary (Heshbon - Esebon, Admah - Adama,
Ribkah - Rebekkah, Calneh - Halanni) are explainable by epenthesis to break
hard-to-pronounce consonantal clusters. The nature of that unpleasantness is
clear: impossibility to elongate the first consonant in a cluster (which
would semi-break the cluster, Milcah - Mil:cah ~ Mil(e)cah) or when the
sounds originate in incompatible areas (ln in Calneh - Halanni).
Since I always defend an idea that the Masoretes could not mean two
different things by one mark, and so intended schwa to sound the same, how
did vocal and silent schwa appeared? Since the language originally had only
CV (later CVC final) syllables, every schwa marks a lost vowel, and
theoretically sounds like apostrophe. Basically, silent schwa occurs
immediately before the accent, while vocal schwa is pro-pretonic. This is
easily understandble.
Accented syllable draws the adjacent consonants (ni.zcAr) and jams the weak
apostrophe between them (niz'car - ni.zcAr). Pretonic schwa becomes silent.
When the accent is farther away from the apostrophe (k'turAh), the
consonantal cluster (k't) is not getting drawn into the accented syllable.
Consequently, the interconsonantal apostrophe is not jammed. Moreover, the 
apostrophe comes
handy in otherwise hard-to-pronounce clusters far from the accent.
Pro-pretonic schwa thus remains vocal.
Modern Hebrew lost pro-pretonic (vocal) schwa. One possibility is that
European speakers tolerate consonantal clusters, and consonants, at any
rate, became considerably aspirated. Another possibility is that vowels
shortened (long vowels lost), and modern pro-pretonic position corresponds
to ancient pretonic in the air flow.
Now, why the Masoretes insisted that all schwas are vocal (they all
are marked similarly)? Because the Masoretes sung! They had extra-long
vowels. What was pretonic schwa in regular speech, was equal to pro-pretonic
to them (niz'ca:r ~ niz.caar). This phenomenon is very common in singing;
next time you go to the Metropolitan Opera, listen to the singers: those
less than brilliant introduce epenthetic sounds (breathing) in consonantal
clusters - or blur those clusters (the last case is not common at the
Metropolitan level, but you can listen to it on Broadway shows). Whether or 
not the Masoretes actually knew that pretonic schwa should be theoretically 
vocal even in speech, we don't know.

(A note for Peter: Joel's examples show that as early as the LXX, begedkefet
were aspirated. In a discussion about schwa some time ago you preferred the
view, if I recall correctly, that the aspiration is a late phenomenon.)

Vadim Cherny