[b-hebrew] Was Schewa really silent?

Yitzhak Sapir yitzhaksapir at gmail.com
Thu Aug 30 21:39:15 EDT 2007

On 8/30/07, JAMES CHRISTIAN READ wrote:

> Hi all,
> recent discussions have reminded me of a point which I
> have never really fully resolved in my mind. The whole
> point of the Massoretic pointing system was to preserve
> the pronunciation by introducing *non ambiguous*
> representation of vowels.
> Why would the massoretes betray this goal by giving
> the point which represents Schewa a dual role? And so,
> I guess what I'm really asking is, is there any real
> evidence that Schewa was used by the Massoretes to
> close a syllable rather than to represent a vowel?

The vocalization was not intended to be "unambiguous."
In addition, a system of vocalization was generally used
to relate more differences as time went on.  Some
systems only give some vocalization for some words.
Even the Tiberian system as we have it today does not
represent the most evolved stage of this vocalization.
Some details in the pronunciation of words are not
represented even though the Massoretes recognized
the differences, and in others the Massoretes may
not have themselves realized they were pronouncing
two different allophones.  The vocalization and
cantillation system of the Massoretes is not the only
thing the Massoretes left -- we also have the
Massorah as well as various texts on grammar
and pronunciation.  Many more of these were
discovered in the Cairo Genizah.  From the Genizah,
we also have transliterations into Arabic and sometimes
even Arabic transliteration into Hebrew with vocalization.
The schewa was not originally marked even as other
vowels were.  From the point of view of the
Massoretes, the schewa syllable was always part
of the bigger syllable unit either to its left or to its

As can be determined just from this usage, schewas
at the very beginning of a word obviously are part of
the syllable to their left, just like schewas at the end
of a word are part of the syllable to their right.  Two
schewas at the middle of a word go separate ways,
the left one is part of the left syllable, and the right
is part of the syllable to its right.  The only ambiguity
is a singular schewa in the middle of a word.  This
the Massoretes pronounced silent except for certain
conditions.  For example, if the letter had a dagesh.
Also, if the previous vowel was long, the schewa
would be pronounced vocal if also it was under a pair
of identical consonants.  There are various other relatively
exceptional cases of a schewa being vocal in the middle
of a word.

In an article on the subject in Leshonenu 12, p. 61-74,
Nehemiah Allony discusses the Medieval evidence for
this pronunciation (in an article titled, "Vocal and
Silent [Schewas] In the Middle Ages").  He provides
various evidence for this as follows:

1) The Massoretes never say a schewa is vocal after
some of the seven "kings" and silent after others.
2) The division into long and short vowels is late and
dates to the 12th-13th centuries.
3) Saadia Gaon does not deal with the schewa at all
in his commentary of the Book of Creation, when dealing
with vowels.  He opens a discussion on the gutturals
with an example suggesting he does not differentiate
a vocal from silent schewa in the middle of a word -
he writes that in place of a schewa under )p(l, there is
a hataf seghol if the p is guttural.
4) From "Diqduqe Hateamim" by Aaron Ben Asher,
"These are the kings: ... and hataf and qamats and
hataf and patah and so the schewa if it comes in the
beginning of a word, but if it comes in the middle or the
end is not a king."  Also in discussing that a schewa
takes the quality of the following guttural, he writes
that this does not take place in various words - mostly
3mp and 3fs suffix verbs.
5) In the "Article on the Schewa", it is written that the
schewa in words of the verb hlk is vocal only if they
precede a word that has a conjunctive dagesh (such
as )lkh ly in Jer 5:5) and it adds, "and if there is no
dagesh, the schewa is silent, as in Neh 9:12 ylkw bh".
(There are 12 more points of proof that Allony brings).

Geoffrey Khan ("Vowel Length and Syllable Structure
in the Tiberian Tradition of Biblical Hebrew") adds the
1) Saadia Gaon writes that the schewa in the words
"she made" (&th, "she was" hyth is quiescent.
2) In the Karaite manuscript BL Or. 2554, a schewa
in the middle of a word is sometimes represented as
a sukun, as in wbrxbwt Song 3:2, k(+yh Song 1:7,
td(y Song 1:8).
He also notes that the cantillation system has
various rules (such as "a tebir is preceded by
a darga if between the two there is at least one
full vowel and one short vowel between them,
otherwise it is preceded by a merka". But this
principle considers a schewa to be silent as in
2 Sam 17:17, l) ywklw.

Yitzhak Sapir

More information about the b-hebrew mailing list