[b-hebrew] Verbs, text-segmenting and clause-types

Rolf Furuli furuli at online.no
Sat Aug 13 02:26:15 EDT 2005

Dear Hayyim,

Your appraoch is very fine. You see one claim regarding Semitic verbs, then
you see another that contradicts the first claim, and then you ask: "what is

----- Original Message ----- 
From: <Bearpecs at aol.com>
To: <b-hebrew at lists.ibiblio.org>
Sent: Friday, August 12, 2005 7:24 PM
Subject: Re: [b-hebrew] Verbs, text-segmenting and clause-types

> In a message dated 8/1/2005 3:16:06 AM Eastern Daylight Time,
> furuli at online.no writes:
> These  numbers clearly suggest that phonological reasons and/or the
> tendency
> to  cut off endings in Hebrew words are the reasons behind apocopation and
> not  that the antecedent to the apocopated WAYYIQTOLs is a short  preterit
> YAQTUL  while the antecedent to the WEYIQTOLs and the YIQTOLs  is a long
> present/future YAQTULU.
> Doesn't this ignore the comparative Semitic data?
> _The Origins and Development of the Waw-Consecutive:  Northwest  Semitic
> Evidence from Ugarit to Qumran_ by Mark S. Smith (Scholars Press,  1991),
> p.12:
> "The converted imperfect derived from independent usage of the *yaqtul
> preterite, attested in both the Amarna and Ugaritic texts... the *yagtul
> preterite
> survives in numerous Semitic languages in specifically marked
> environments."
> p. 13: "... the BH converted perfect may be traced to the future uses of
> *qatala in apodoses of BH conditional sentences as in Amarna and
> Ugaritic...."

The basic problem in many studies of Semitic verbs is conjecture,
conjecture, and
more conjecture! Or, perhaps we should substitute "conjecture" with
Professors teach their students particular viewpoints, and
when these students become professors, they teach their students the same
things, and so on.

Webster`s defines "preterit" as "past tense," and "tense"
is defined as "grammaticalized location in time". This means that if
"preterit" is used in the normal sense by Smith, it
refers to a verb form that has an intrinsic past reference, which always
(explainable exceptions can occur) is past related to the deictic center.
In English and Hebrew the participle can have
past reference (and present and future as well), but no one would say that
the participle in these languages are preterits. In Phoenician the
infinitive aboslute is the form used for past narrative, but no one would
say this form is preterit.

And here is the important point: No study of Ugaritic of Akkadian verbs has
systematically distinguished between past tense  and past reference!
Therefore, when Smith claims that the converted imperfect (WAYYIQTOL) is
derived from YAQTUL, he just repeats what is the common viewpoint, a
viewpoint which has never been demonstrated to be true.

 To illustrate what I mean by the term "hearsay," i refer to the doctoral
dissertation of T. L. Fenton (1963) "The Ugaritic Verbal System", Oxford.
In this study Fenton analyzes 561 YAQTULs as past, 70 as past continuous,
and 191 as future. If this analysis is correct, it is obvious that YAQTUL
cannot be preterit. Smith refers to three studies of Fenton, but he does not
refer to the mentioned dissertation.

In the Amarna letters the situation is much more difficult than in Ugaritic,
because the scribes wrote in Akkadian, while their Canaanite mother tongue
sometimes influenced their Akkadian choices. A. F. Rayney`s excellent
on the Amarna tablets, "Canaanite in the Amarna tablets: a linguistic
analysis of the mixed
dialect used by the scribes from Canaan" 4 vols. (1996) shows for example
that Akkadian IPRUS (viewed as preterit) is used with future reference

There is also another problem with the claim that both the Ugaritic and the
Akkadian/Amarna short forms are antecedents of the Hebrew WAYYIQTOL. In
fact, such a claim is tantamount to comparing apples and oranges, because
the shortness is
different in the three languages. In Ugaritic the short form is YAQTUL and
the long is YAQTULU, and this means that the difference is the final U. In
Akkadian the short form is IPRUS and the long is IPARRAS, and this means
that the difference is the vowel/lack of vowel between first and second root

In Hebrew there is no form similar to IPARRAS, and both YIQTOLs and
WAYYIQTOLs lack a vowel between first and second root consonant. Moreover,
the apocopation which is seen in many WAYYIQTOLs relates to the last root
consonant and not to a final U. This means that if a short preterit was the
antecedent of Hebrew WAYYIQTOL, either Ugaritic YAQTUL or Akkadian
IPRUS could have been such an antecedent. Both forms hardly could be such an
antecedent, because their shortness is different.

If we look at the use of the short forms in the cognate languages, the basic
difference we find  is not between temporal reference (tense is lacking in
these languages), but one between indicative and modality.
For example, in Ge`ez (Ethiopic) we find a system similar to the Akkadian
one. The form YINGER (subjunctive) lacks a vowel between first and second
root consonant
while YENAGGER (imperfect) has a vowel in this position. The basic
difference between
the two is that YINGER often is modal but can be indicative as well, and
YENAGGER often is indicative, but can be modal as well. Both forms can refer
past, present, and future. Ethiopic has a perfect NAGARU, which is lacking
in Akkadian.  As for Akkadian, the short form is used both with past
reference and for modality.

Arabic resembles Ugaritic as far as the indicative/jussive distinctions are
concerned. In Arabic we find the following forms: indicative YAQTULU,
subjunctive: YAQTULA, jussive: YAQTUL, and energic: YAQTULAN(NA). So, the
Ugaritic opposition YAQTUL/YAQTULU is seen in Arabic in the opposition
between indicative and jussive. In Phoenician and Biblical Aramaic
indicative has final N while jussive does not have this N. Thus, the
indicative/jussive difference in these languages is between long and short
forms as well.

By way of conclusion I would say that Smith`s claim is wrong because:

1) No study of any old Semitic language has systematically distinguished
between past tense and past reference. And, the existence of a preterit
conjugation has not been established in any of the languages.

2) Apocopation in Hebrew means that the last root consonant is lost (whether
lack of YOD  in Hiphils is apocopation or original is not known). This is
neither the case in Ugaritic nor in Akkadian. The lack of vowel between
first and second root consonant in Akkadian is found in both YIQTOLs and
WAYYIQTOLs  in Hebrew, and the distincion between Ugaritic final U/lack of U
is not found in Hebrew.

3) The basic distinction between long and short forms in the Semitic
languages is not one of tense (between preterit and present/future) but one
between indicative and modality. This is also the case in classical Hebrew.

Best regards

Rolf Furuli
University of Oslo

> Hayyim Obadyah
> , MPA
> New York, New  York 10027

More information about the b-hebrew mailing list