Henry Churchyard churchh at
Wed Sep 27 15:35:17 EDT 2000

> Date: Tue, 26 Sep 2000 08:42:22 -0500
> From: "Ray Clendenen" <rclende at>
> Subject: shtey

> I have started reading Horsnell's very helpful _A Review and
> Reference Grammar for Biblical Hebrew_.  I ran across a note on
> p. 33 that says the word for 'two', shtey/shtayim, begins with a
> silent sheva followed by a dagesh lene (as I have transliterated
> it).  He argues that since the dagesh is a dagesh lene, the sheva
> must be silent.  My question is, How do we know it is a dagesh lene?

That's an interesting question.  The problem arises from the
interaction of three phonological/orthographic patterns in Tiberian

1) A _dagesh_ in a _begadkephat_ letter which is located between
between two pronounced vowels in a word always means that the
_begadkephat_ letter is pronounced as a doubled consonant (i.e. the
_dagesh_ is "dagesh forte" or "dagesh h.azaq").

2) A Hebrew syllable never begins with two consonants.  Instead, every
word can be divided into syllables such that each syllable always
starts with exactly one consonant (except possibly the "vocalized" u-
form of the prefixed conjunction w at -), then contains a vowel, and then
can optionally end with one consonant (or two consonants if at the end
of a word).

3) A pronounced or "vocal" _sh at wa_ never occurs in a closed syllable
(i.e. a syllable that ends in a consonant), but always occurs directly
before the consonant that begins the next syllable.

Together, all this means that if you take the orthographic sequence
_shin-sh at wa-taw-dagesh_ which begins the forms _shtayim/shtey_ etc.
($:T.AYIM etc. in the electronic BHS "betacode" transliteration), then
one of the three principles above has to bend -- since a pronunciation
of [sh at ttayim] with vocal _sh at wa_ and doubled consonant would violate
the constraint against a _sh at wa_ vowel in a closed syllable (given
that double consonants always straddle two syllables); while a
pronunciation as [sh at tayim] with vocal _sh at wa_ and single consonant
would violate the principle that a _begadkephat_ with dagesh between
between two pronounced vowels is pronounced double (a non-doubled
_begadkephat_ between two pronounced vowels would be liable to
spirantization, in which case it wouldn't be written with dagesh at
all); and finally, a pronunciation of [shtayim] with "silent" _sh at wa_
would violate the third principle (that no Hebrew syllable should
begin with more than one consonant).

In fact, there seems to be evidence that it's the third principle
which is violated, and that the words _shtayim_, _shtey_ etc. are the
only forms in the language which contain a syllable that begins with a
consonant cluster (instead of with a single consonant).

One piece of evidence for this comes from the distribution of the
accent _y at thibh_.  This orthographic symbol is only supposed to occur
on words whose initial syllables are main-stressed.  Therefore this
accent generally does not occur on words which begin with
consonant+_sh at wa_+consonant, since _sh at wa_ vowels are never
main-stressed (so that words beginning with a syllable containing
"vocal" _sh at wa_ cannot be stressed on the first syllable).  However,
in many manuscripts, including manuscript L / B19a which is the basis
for BHS, _y at thibh_ does in fact occur on four occurrences of the word
_shtayim_ in Leviticus 23:17, Ezekiel 1:11, and twice in Ezekiel
41:24, showing that in this particular form the _sh at wa_ was actually
silent.  (In fact these four cases are the only occurrences of
_y at thibh_, out of 1170 occurrences in the electronic BHS text, located
on words beginning with a consonant followed by an orthographic
_sh at wa_ grapheme.)  See _Introduction to the Tiberian Masorah_ by
Israel Yeivin (translated by E. J. Revell, 1980) page 276, or section of my dissertation, for further discussion.

There have also been two brief articles in scholarly journals devoted
to the topic of the pronunciation of _shtayim_.  One is "Hebrew
_shtayim_ in the light of Syriac and Turcic" by Meir M. Bravmann, in
the Proceedings of the American Academy for Jewish Research v. 21
(1952), and the other is "Initial Consonant Clusters in Hebrew and
Aramaic" by Robert D. Hoberman in the _Journal of Near Eastern
Studies_ v. 48 (1989).  Both articles supported the same conclusion
(silent _sh at wa_) from other evidence.  The Bravmann article was on the
indirect evidence of typological cross-linguistic comparisons.
Unfortunately, I can't find my photocopy of the Hoberman article right
now (if I made one).  (I think Blau also discusses it somewhere, but I
can only find a brief footnote on p. 76 of his 1976 grammar.)

Henry Churchyard   churchh at

More information about the b-hebrew mailing list