Biblical Hebrew Forum

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

On 9/8/07, James Christian Read wrote:

> Aren't the quotes from much later than the Massoretes who invented the
> system? If so, is it not possible that they are commenting on a later phase
> of the evolution of pronunciation? If vocal schewas are a result of laziness
> of speech in unstressed syllables with short vowels would not the next
> logical step in such an evolution to be to not pronounce a vowel at all in
> places where they are not necessary to produce a pronounceable word i.e in
> between a consonant which closes a syllable and a different consonant which
> starts a syllable? Double consonants would survive this change because of
> the difficulty of pronouncing double consonants outweighing the advantages
> of dropping a vowel.
>
> I also asked if there was any transliterational evidence that predates the
> Massoretes from a language like Greek whose vowel system I well understand
> that supports the theory of silent schewas.

Vocal schewas are not a result of "laziness of speech."  They include
historically
short vowels that were preserved in the language.  Historically, Hebrew
differentiated vowel length phonemically.  As late as the Hexaplaric evidence,
vowel length was still differentiated phonemically.  However, in the centuries
that followed, much of these differences were transferred to vowel quality, 
and
vowel length became predictable based on the syllable structure of the word.
There are still some rare examples where vowel length still served as a 
function
phonemically -- such as the word dɔmi "silence" vs. dɔ:mi "my blood."  While
vowel length was pretty much predictable, the Massoretes still understood and
recognized short and long vowels.  I am not sure whether this has to do with
the fact that they also spoke a language that did differentiate between the 
two
(Arabic) or not, but it doesn't really matter.  Also, I personally
think there are
hints in the Massoretic cantillation system that shows it has its
origins during
a time where vowel length was still differentiated.  In any case, the absence
of a vowel between two consonants ("a silent schewa") or a short open vowel
("a vocal schewa" or hataf) could be the result of any number of reasons --
including an originally short vowel or original absence of vowel or a 
shortened
or dropped vowel or an inserted vowel where there previously was none.
However, the double consonant rule is a special case of a dropped vowel -- in
most cases, a doubled consonant would be written once in the consonantal
script.  A doubled consonant indicates that while the spelling at one point in
the past recognized a vowel present between the two letters, the vowel was
later dropped when preceded by a short vowel.  Thus, only when preceded by
a long vowel we see a short vowel that was maintained.  Just because the
double consonant rule indicates that a schewa is to be pronounced vocal
in that specific instance and this indicates that for a doubled consonant, 
there
was a short vowel that was lost in other comparable instances does not mean
all vocal schewas have their history in the loss of a vowel.

The Tiberian system (as well as other systems) developed over a period of
centuries.  However, the consonantal and vocalization system that we have
in the texts before us dates from the 10th century.  The sources I mentioned
are written by those same Massoretes who lived in the 10th-11th centuries,
although the Leshonenu article goes on to later centuries.  One of the
pieces of evidence is written by the same person who vocalized the Aleppo
codex!  Thus, the sources mentioned are of the same people whose
vocalization system we now use.

The Arabic transliteration also dates from this time and, even though Arabic
does not normally indicate short vowels, in this case the Arabic 
transliteration
indicated the absence of a vowel explicitly in those words quoted.

That a hataf indicates a short vowel can be learned from a comparison of
forms in Hebrew:
yispor "will count" (3ms) - yisparu: "will count" (3mp)
ya(amod "will stand" (3ms) - ya(amdu: "will stand" (3mp)
In both cases, there was an absence of vowel between the first and second
root consonants.  As part of the "slide" process of the gutturals, the 
gutturals
were vocalized with a short vowel.  In the 3ms case, the guttural is vocalized
with a hataf patax because it is an open syllable, but in the 3mp case it is
vocalized with a regular patax because it is a closed syllable.  But this is 
the
same linguistic process and in both cases it is the same length!  The only
difference is that one is an open syllable and one is a closed syllable.  The
vowel between the second and third root consonants of the 3mp case fell
in the guttural case because the syllable structure during Massoretic times
did not allow two consecutive short vowels, and the guttural added a short
vowel to the word after it "slid".

Going on, we also see from the vocalization of the second root consonant
that it was not preceded by a vowel already in early times.  Otherwise, it
would be vocalized as an [f] because of the bgdkpt rule.  In fact, we know
from comparative Semitic data that this form never had a vowel in this
position.  The above verb form is also a common structure for names so
you can check this easily in the Septuagint.  Some names that come to
mind and show the absence of a vowel in this position include Jacob,
Jephthah, and Ismael, all of which show an absence of a vowel.  The
Mesha inscription also shows an absence of a vowel in the verb krty
(line 25) from the verbal root krt -- compare the next word hmkrtt -- but
with the same form as mlkty (line 2-3), except because there is no vowel
between the verb's root and the -ty, the t doubles and is written as one
consonant.  A few weeks ago, I also provided Karl with evidence from
Akkadian sources that various words such as the place name Ashdod
had no vowel between two consonants.

Yitzhak Sapir