Re; Silent or vocal shewa?

Henry Churchyard churchh at
Wed Oct 25 10:53:54 EDT 2000

>>>> From: "M & E Anstey" <anstey at>

>>>> It is usually stated that a shewa after an accented syllable is
>>>> silent, but what of the case of irregularly stressed 3fs qal
>>>> perfects, such as found in Gen 19:38, 30:21; Joe 4:13; Mal 1:9,
>>>> etc which have the accent on the first syllable? Is this shewa
>>>> then silent or vocal, ie yaldah or yaledah, mal'ah or male'ah,
>>>> haytah or hayetah, etc?

>>> From: "Trevor & Julie Peterson" <spedrson at>

>>> One principle that may apply is stated in Jouon-Muraoka (and
>>> elsewhere I'm sure): "In the state of Hebrew as recorded by the

>>> Naqdanim, the primary stress occurs only on the ultima . . . or on
>>> the penultima" (15b).  They don't give any exceptions to this
>>> principle (although it could be an oversight), which may suggest
>>> that the vocal shewa readings are impossible.

>> From: Henry Churchyard [mailto:churchh at]

>> as discussed by William Chomsky in his 1971 article "The
>> Pronunciation of the _Shewa_", in _Jewish Quarterly Review_ v. 62,
>> pp. 88-94, sh at wa after a long vowel + consonant is not automatically
>> vocal, so that _yaaldhaa_ and _haaythaa_ would have silent _sh at wa_
>> regardless of whether the "stress retraction" process of _n at siga_
>> has applied or not.  However, there are indisputable cases of
>> antepenultimate stress, and rather clear cases of vocal _sh at wa_
>> after the main-stressed syllable, in the Biblical text.  For a list
>> of antepenultimate and quasi-antepenultimate stressed cases, see
>>  For a comprehensive
>> list of forms affected by _n at siga_ a.k.a. the Hebrew Rhythm Rule,
>> see  For a
>> discussion of the environments of vocal vs. silent _sh at wa_, see my
>> dissertation, downloadable from

> From: "Peter Kirk" <Peter_Kirk at>

> Thank you, Henry, for the interesting list of antepenultimate
> stresses.  I note that in every one of these cases except for the
> five in section 1d the penultimate syllable is a vocal shewa or
> compound shewa.  There is one more apparent exception, Exodus 7:10
> WAY.A(A&W., but here BHS follows L in writing pathah in the text but
> notes that many MSS and editions have a compound shewa here (as in
> Jos.10:23 etc.).  So if we modify Matthew's and Jouon-Muraoka's rule
> so that vocal shewas and compound shewas are not counted as
> syllables, then there are only five exceptions to the rule plus one
> textually doubtful case.  That is about as close to an absolute rule
> as anything I have seen in the area of Hebrew pointing etc.

Two principles were brought up, neither of which is valid in its
simplest and most categorical formulation, but both of which have some
validity if a higher level of phonological sophistication is taken into
account.  So "shewa after the main-stressed vowel of a word is silent"
is not a universally valid principle, but it is true provided that
words with retracted stress position and words with a directional -aah
suffix are left out of consideration (this directional -aah suffix can
act as a clitic affix, attached after stress-placement and segholate
epenthesis have already applied to an otherwise unsuffixed noun -- if
you're operating within a theory that allows sequential phonological
derivations).  (Also see the caveats in the ANTEPNLT.TXT file about the
interpretation of silent vs. vocal _sh at wa_ in a few forms in
Hosea 5:15, Psalms 32:2, etc. which might be further minor exceptions
to a "shewa after main-stress is silent" generalization.)

The other proposed generalization, "stress is always word-final or
penultimate", is not universally true as stated either -- but as you
pointed out, it is basically true if _sh at wa_ vowels of all types are
ignored.  I would argue that what's relevant here is the distinction
between unreduced (or full) vowels and prosodically reduced vowels,
where the term "reduced" not only covers vocal _sh at wa_ and compound
_sh at wa_ (i.e. vowels -- but also (as I discuss in section
1.3 of my dissertation) covers non-sh at wa vowels which originate as
epenthetic vowels after a guttural consonant, and which are then
promoted to full vowels by deletion of the vowel in the following
syllable (e.g.  na`modaa -> na`amodaa -> na`amdhaa in Isaiah 50:8
etc.; see end).  This last type of vowels are actually also reduced
vowels according to certain phonological criteria (despite not being
surface _sh at wa_ vowels in the Tiberian orthography).

I agree that the unreduced pointing of the middle vowel of the form
in Exodus 20:7 is bogus (in the tabulation of _n at siga_ cases in the
file NSIGRHYR.TXT I silently counted this among the 68 cases of
_n at siga_ in "which the rhythm rule moves stress onto an unreduced
open antepenultimate syllable (CV or CVV), skipping over an
intervening light open syllable with a reduced vowel --- where the
antepenultimate syllable is followed by a consonant +
vowel --- with a guttural consonant before the").

 ;  d.Rhythm rule (_n at siigaa_), where penultimate syllable is closed:

 Exo15. 8           nE6`Er.muw  ma3yim
 Deu33.28           ya6`ar.puw  TA1l;
 Isa40.18          t:a6`ar.kuw  lo1w;
 Isa50. 8           na6`am.dAh  y:A2Had
 Job12.15         w.ya"hap.kuw  'A"rEc;

Henry Churchyard   churchh at

More information about the b-hebrew mailing list