Biblical Hebrew Forum

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

On 10/10/06, Peter Kirk wrote:
> On 09/10/2006 08:12, Yitzhak Sapir wrote:
> > ...
> > I would like to suggest a different theory.  In the article on
> > Afroasiatic in the
> > Cambridge Encyclopedia of the World's Ancient Languages, Huehnergard
> > suggests that in West Semitic, the perfective adjective of the Proto Semitic
> > replaced the inherited Semitic form yaqtVl perfective form (contrasted
> > with the inherited Semitic form yaqattVl form).  Following the same lines
> > of argument as suggested by Goerwitz for noun vs verbal pausal forms, we
> > might suggest the following development:
> >
> > 1) the perfective adjective replaced the perfective verb in sentences of the
> > form: *bayta wathibna ("we lived in a house") from the previous: bayta
> > yawthVb.  this did not occur in sentences of the form: wayawthVb bayta
> > ("and he lived in a house") because adjectives were not used as often at
> > the beginning of the word.
> > 2) this created a situation in which yawthVb and yawaththVb at the middle
> > of the sentence were confused or reanalyzed as prefix-verbs contrasted
> > with "suffix verbs" which developed from the perfective adjective.
> > 3) by analogy, the waw in such cases was analyzed as a waw conversive
> > and so applied to the adjective
> >
> >
> If I understand this correctly, and to restate it in terms more like
> Hebraists are used to, you are suggesting that the QATAL form in Hebrew
> derives from some kind of participle, with subject suffixes, with a
> perfective meaning.

This is suggested by the Afroasiatic article.

> Apart from this, the distinction between perfective
> and imperfective was more one or word order, with WAYYIQTOL X as
> perfective and WE-X YIQTOL as imperfective.

No.  The imperfect had a geminate in the second consonant in PS,
according, again, to the Afroasiatic article.
yaqtVl = perfective, yaqattal = imperfect
The article also posits the possibility of several additional forms based
on the perfective, including one ending in *na (yaqtVlna?), which, it says
replaced the inherited geminate imperfect form in Central Semitic.  It is
thus perhaps better to speak of YIQTOL vs YIQTOLUN, and WAYIQTOL
vs WAYIQTOLUN.  I suggest that the change whereby the participle
replaced the perfective only or even just usually took place in middle
or end of sentence surroundings.  Thus, the changes were:
WAYIQTOL X -> WAYIQTOL X
WA-X YIQTOL -> WA-X QATAL
WAYIQTOLUN X -> WAYIQTOLUN X
WA-X YIQTOLUN -> WA-X YIQTOLUN
where X is a noun.
In this way, the change came to be perceived more as YIQTOL(UN)
vs QATAL than as YIQTOL vs YIQTOLUN.  YIQTOL and WAYIQTOL
were perceived to be equivalent, except that for YIQTOL, WA acted
as the "waw conversive", and for YIQTOLUN as a "waw conjunctive".
Later, WEQATAL X came in as analogous to the WAYIQTOL.  All
this also occurred in other situations representing the start of a
sentence, such as ")AZ YIQTOL".

Other forms are posited by the Afroasiatic article, an imperative
equivalent to the perfective without prefixes, a l(a)yaqtVl jussive,
and perhaps other forms such as yaqtula or an "energic" yaqtulan(na),
but these are difficult to reconstruct because they only appear in one
or two branches.  For what I called the "YIQTOLUN", it posits some
form that was the origin of the -u or -ni (probably *-na) ending on
Akkadian verbs in subordinate clauses, and which replaced the
inherited yaqattal for the imperfect.  But perhaps the Biblical
YIQTOLUN is really derived from that "yaqtulan(na)" form, for
example.

> This kind of development process makes reasonable
> sense.
>
> So, this kind of process may have happened. But if it did, it must have
> happened way back in proto-Semitic history, as the distinction between
> prefix and suffix forms is found at least in all West Semitic languages
> (from memory).

The article does say that QATAL replaced YAQTAL in West
Semitic.  The reanalysis came later, and was probably triggered by
additional factors.

> So what is the particular relevance to Hebrew? I suppose
> you could argue that this was the situation in proto-West Semitic
> languages, but WAYYIQTOL had been lost in all but Hebrew
> before the time of the earliest inscriptions. Is that in fact what
> you would suggest?

Sort of.  But first, it's not just Hebrew: The Mesha inscription and the
Balaam inscription both show a "wyqtl" for past or perfective:
(mry mlk y$r)l wy(nw )t m)b ymn rbn - Omri ruled Israel and oppressed
Moab a long time (Mesha)
spr [bl]([m brb(]r )$ xzh )lhn h) wy)tw )lwh )lhn blylh - The book of Balaam
son of Beor, a seer of the gods, and the gods came to him at night (Balaam)
Also, I looked up Shlomo Izreel's Canaano-Akkadian book, and it writes
that the prefix conjugation was appended suffixed morphs which marked its
tense-mood-aspect application, with imperfective suffixed with u or na,
perfective with 0 (zero morph), and modal with a.  He gives the following
examples for the perfective prefix conjugation, which bear past or modal
implication only:
"and my mare was shot, and I stood behind him, and I rode with Yashdata"
(EA 245:8-12, Megiddo), "u tussaxmi", "u izzizmi", "u irkabmi", past
"may the king, my lord, ask Yanhamu, his servant what is being done in
his land" (EA 271: 23-7, Gezer) "yi$a:l", modal
"lest the Apiru destroy us" (EA 299:24-6, Gezer), "la: tugammeru:nu", modal
The only example for past tense that he gives is used in a "waw conversive"
sort of context.  Also, his derivation of the suffixes with 0 = perfective 
does
seem to connect with the basic perfective form in the Afroasiatic article.

So it's not "all but Hebrew", but even Hebrew lost it by the Mishnaic period.

Yitzhak Sapir