Subject: Re: [b-hebrew] wayyomer as imperfective aspect
Date: Fri, 26 Jun 2009 17:45:28 +0200
Dear list-members,
I have some comments to the challenge below. I discuss WAYYIQTOL as
a form rather than particular verbs. And I give no examples to avoid
fruitless discussions like the discussions regarding "plod". The
important thing in BH is to see the big picture and the reasons why
different scholars paint this picture differently. Hairsplitting
discussions of minute details that some list-members are so fond of,
tend to cloud the picture rather than illuminating it. I do intend to
continue this discussion, but I hope that some list-members may have
some benefit from it. Examples showing that the WAYYIQTOLS and
YIQTOLs of the verbs of speech )MR and DBR have the same meaning are
found in examples 22- 34 on pp. 208-217 in my dissertation.
I'd like to repeat my question since you may have missed it from my
earlier post. Can you demonstrate for us how wayyomer in BH is
imperfective in aspect?
TWO OR FOUR CONJUGATIONS - MORPHOLOGY AND HISTORY
1. In Ugaritic, Phoenician, and Aramaic we find only two different
conjugations, the prefix forms and the suffix forms (I leave Akkadian
alone at this stage.)
2. In BH we also find the prefix forms and suffix forms. But a little
more than half of the prefix forms and between a third and a fourth
of the suffix forms have a prefixed WAW. So the question is whether
these prefixed WAWs signal two extra conjugations, or whether they
only are prefixed conjunctions,
3. On the basis of the DSS we can give no answer because of lack of
vowels (patah patrticularly) and gemination, Origen in his Hexapla
makes no distinction between the so-called WEYIQTOL and WAYYIQTOL.
The first time we can see a difference is in the Masoretic text.
4. The WAY- element of WAYYIQTOL need not be a semantic element,
because the geminiation and different stress can be explained on the
basis of the Masoretic phonetic rules of stress and vowel shift, as
well as the resitation of the text in the synagogue. The same is true
with the WE-element if WEQATAL and the particular stress shift in
some persons. So there is nothing in the use for verb forms, as seen
in the manuscripts before the Masoretes, or in the morphology of the
forms (including vowels and accents) that demand that WAYYIQTOL is a
conjugation different from YIQTOL, or that WEQATAL is different from
QATAL.
LINGUISTIC APPROACHES THAT CAN ILLUMINATE THE ISSUE
5. Different approaches have been used: Diachronic approaches trying
to follow the evolution of verb forms in the Semitic languages were
popular in the first part of the 20th century. We have also seen
approaches based on the assumption that BH is a tense language or an
aspect language. And studies of grammaticalization processes have
also been performed. The problems I see in all these studies are all
the assumptions and axioms that are used, and there are few controls
because of lack of distinction between semantic and pragmatic factors,
6. A much better approach, in my view, is to study the language
without any preconceived ideas regarding the BH verb forms. To
achieve that, one needs a set of fundamental parameters that can be
applied to any living or dead language, and by the help of which one
can eliminate some supposed properties and argue in favor of others.
As I have mentioned several times, I have used the parameters deictic
center, event time, and reference time. By the help of these
parameters one can demonstrate that a language has or do not have
tenses; one can demonstrate whether a language has aspects, and if
so, the properties of these aspects. And one can find out (if a
diachronic study is possible) whether there is an unfinished
grammaticalization process in the verbal system.
7. In order to use as few assumptions as possible and get results
that can be tested by others, it is necessary to distinguish between
semantic and pragmatic factors. For example, in order to find whether
BH has tenses, it is necessary to accept that there is an
unchangeable semantic relationship between the deictic center (C)
and reference time (R). When the tense is past, R always comes
before C, and when the tense is future, R always comes after C. This
semantic relationship is the universal understanding of tense, even
when scholars dot not use C and R. This is not to say that when R.
comes before C, it is evidence of past *tense,* because this can also
happen in purely aspectual languages. Then we only speak of "past
reference". In addition to the semantic difference between R and C
when we speak of tense, we must also see uniformity of verb forms and
temporal references. If BH had tenses, we would expect that the same
verb form (conjugation) had only one temporal reference, except in
cases that can be linguistically explained. This is not the case of
BH, and the conclusion must be that BH does not have tenses. (As a
basis for the conclusion, an unfinished grammaticalization process
has been excluded on the basis of the evidence.)
THE WAYYIQTOL FORM
8. As already mentioned, there is no morphological difference between
YIQTOL and WAYYIQTOL(if the WAY- element is a conjunction). So, a
semantic difference between the two cannot be concluded on the basis
of morphology. And please note that here we have a good example of
how other scholars use the distinction semantic/pragmatic. From the
Middle Ages it has been claimed that the distinction between YIQTOL
and WAYYIQTOL is not only pragmatic (use/function), but it is
semantic. It has been believed that the WAYYIQTOL somehow reverses
the meaning of YIQTOL. Some have said that YIQTOL is futue tense,
and WAYYIQTOL is the reverse, namely, past tense. Others have said
that YIQTOL is the imperfective aspect and WAYYIQTOL the perfective
one. This is nothing but a *semantic* distinction!
9. Apart from the difference in WAY- and stress, one basic reason
scholars have for viewing WAYYIQTOL as different from YIQTOL, is
that WAYYIQTOL occurs extensively in narratives with past reference
and the YIQTOL occurs often in future contexts. Some would see a
tense difference, and others would see an aspectual difference. Those
who see an aspectual difference, cannot fathom that WAYYIQTOL can be
imperfective, because it so often has past reference.
10. One problem with many modern studies of BH is that our modern
experience with languages are projected into BH. For example, the
narrative form in aspectual languages is the perfective aspect and
not the imperfective one. A look at Ugaritic may show the futility of
projecting this into Semitic languages. A few semesters ago, in class
we read the whole sagas of Danil and Aqhat, and Kirta, and
particularly in connection with Kirta some students were stunned
because of the use of verbs. First we have a long account of what
Kirta should do and what would happen, where the verbs (mostly prefix
forms, but also some suffix forms) have future reference. Then we
have exactly the same account with past reference -it had happened-
almost completely similar with account in the future and with the
same verbs, but now with past reference. Obviously, the forms have
the same meaning in both cases, yet they have completely different
temporal references. If YIQTOL and WAYYIQTOL both have the same
meaning, the Ugaritic examples would be an interesting counterpart to
the Hebrew use: the YIQTOLs are often used with future reference, but
the same forms with prefixed conjunctions (WAYYIQTOL) are also used
with past reference. The difference being the extensive use of the
conjunction in Hebrew narrative. However, Kirta is not a narrative
account, although it has some narrative traits. In any case, the
Ugaritic examples show that the argument that the WAYYIQTOL must be
different from YIQTOL because of the different uses is futile. In
Phoenician the narrative form is the infinitive absolute, with an
extensive use of prefixed WAWs, and in the Aramaic of Daniel, the
prefix form (somttimes the same root) is used both with future and
past reference. So the arguments in favor of WAYYIQTOL being
different from YIQTOL on the basis of morphology and use, are weak
indeed, if we compare them with other old Semitic languages. And, to
the best of my knowledge, it is unprecedented in any language that a
particular verb form (here a prefix form) by the addition of a
conjunction (or another element that no one has been able to
identify) should signal the reverse meaning of the form without the
conjunction/element!
11. By the help of event time and reference time the BH aspect can
be described or defined. And the BH aspects are both similar and
different compared with the English aspects. An analysis of those
WAYYIQTOLs where the relationship between E and R is visible, shows
that WAYYIQTOL has imperfetive traits just as the YIQTOLs: Some of
the characteristics are:
a) The form can have past, pre-past, present, present completed, and
future reference. It can also be modal.
b) In many instances the reference time of two or more forms
coincide, so they are not consecutive.
c) In many instances it can be shown that the WAY- element is not
more than an ordinary conjunction.
d) The form expresses conative events (attempted but not carried through)
e) The form expresses ingressive events (beginning and first part of
the action being made visible).
f) The form can be resultative (the resultative state being made visible).
g) The form can be intersected by another verb ("When Moses was
reading the law (WAYYIQTOL), Joshua entered the tent.")
h) One WAYYIQTOL can be modified by another WAYYIQTOL.
12. As Kevin Riley wrote, the last word regarding BH verbs has not
been said. I do not claim that my conclusions are the truth and
nothing but the truth. But both my approach and my conclusions are
very different from other studies. Because of this they deserve to be
carefully scrutinized. And the most important step in that direction
is to get a full understanding of the meaning and use of the basic
parameters. This is particularly important in order to understand the
differences between English and BH aspects that I outline. It is not
fair to evaluate my approach in the light of different other
approaches, such as discourse theories, prototype theories and other
theories. Such theories could be applied in other studies, and these
studies be evaluated in their own right. So my approach should also
be evaluated in its own right, whether I have been faithful to my own
principles, and whether my results follow because of my data and the
application of them. When considering the WAYYIQTOL form, which is
almost universally believed to be different from the YIQTOL, it is
important to try to put away one's biases (we all have such), and in
a balanced way consider the material. Two important questions can be
asked, 1) What are the reasons why WAYYIQTOL *must* be different
from YIQTOL? Are these reasons conclusive?, and 2) Does WAYYIQTOL
really have imperfective characteristics? And if so, why should we
not accept that it has the same imperfective aspect as YIQTOL.
Addition: When I began to study Hebrew, and even up to the present,
the Ugaritic use of verbs has been used to to bolster the distinction
between WAYYIQTOL and YIQTOL. It has been argued that WAYYIQTOL came
from a short Ugaritic preterit (YAQTUL) in contrast with the longer
(YAQTULU). This is to turn the evidence upside down, as Kirta and
Danil and Aqhat shows. An interesting article discussing the Ugaritic
verb occurs in "Biblical Hebrew in its Northwest Semitic setting :
typological and historical perspectives / edited by Steven E.
Fassberg and Avi Hurvitz.