[b-hebrew] BH verbal system

Rolf Furuli furuli at online.no
Fri Jan 29 11:29:37 EST 2010


Dear Joseph,

I have  Hebrew grammar of Schneider  (In German) 
in my library, and I have a short discussion of 
it in my doctoral dissertation. Both Schneider 
and Alviero Niccacci base their works on Harald 
Weinrich (1974) "Tempus: Besprochende und 
Erzählte Welt." Weinrich admits that his analysis 
of the verbal systems of  languages on the basis 
of discourse analysis is "unassailable," that is, 
the conclusions cannot be tested. So why should 
we believe them?

In addition to this very grave weakness, I would 
say that Schneider's work to some extent is 
tautological and to some extent is circular. 
Further, it uses grammatical terms in a 
sub-standard way-if we judge his work from the 
point of view of elementary linguistics!

One examples of a tautological conclusion is as 
follows: "The primary tense of the narrative is 
WAYYIQTOL." What is the reason for the 
conclusion? Because the WAYYIQTOL form so often 
is used in the narrative texts. This is similar 
to saying: "John is the father of Charles, and 
Charles is the son of John." Hebrew writers used 
such statements as stylistic devices, but they 
should not occur in modern grammars, because they 
explain nothing.

It is a fundamental linguistic truth that verbs 
used in narratives have past reference. Thus, the 
past reference need not be a semantic part of the 
*verb form*, the past reference may be pragmatic; 
its basis being the context. In Phoenician, 
infinitive absolute is used as the narrative 
verb, but it is not a past tense. In In the 
Ugaritic saga of Kirta (Keret)  prefix forms 
(YAQTUL) are used with past reference as 
narrative verbs. Then, exactly the same account 
with exactly the same prefix forms (a few 
exceptions, though) are used with future 
reference. Therefore, the USE of WAYYIQTOL in 
Hebrew narratives does not tell us anything about 
whether it is a tense (that past tense is a 
semantic part of the very form). To the contrary, 
I have a list of about one thousand WAYYIQTOLs 
with non-past reference. And they show that 
WAYYIQTOL  is not a tense at all. The term 
"tense" is defined as "a grammaticalized location 
in time." Schneider does not distinguish between 
"past tense" (semantic past) and past reference 
(pragmatic past), and this is linguistically 
speaking sub-standard.

The tautological nature of an analysis of the 
Hebrew verbal system based on discourse 
analysis, that is, on the difference between 
narrative and discourse, is less problematic that 
the circularity. The verb forms YIQTOL, 
WAYYIQTOL,  and QATAL  have been assigned 
different functions in narrative and discourse. 
Now, it is often easy to identify narrative 
texts, but in many contexts it is not easy on the 
basis of the context to distinguish between 
narrative and discourse, between foreground and 
background information etc. But this is often 
done on the basis of the theory. This means that 
when a verb form is assigned to represent a 
certain kind of communication, the clause or 
context where it occurs is viewed as representing 
such a communication (for example foreground or 
background) because the verb form occurs there. 
So the nature of the clause(s) is defined on the 
basis of the verb that occurs, and the occurrence 
of the verb form there confirms the correctness 
of the linguistic theory. This is circular 
reasoning!  We find this in Schneider, Niccacci, 
and in numerous discussions regarding discourse 
analysis that have occurred on this list through 
the years.

In the natural sciences the smallest parts of 
particular objects are studied. On this basis, 
balanced conclusions can be drawn. A basic 
principle of the Philosophy of Science is that 
the certainty of a scientific conclusion is 
inversely proportional with the number of 
components that must be accounted for at the same 
time. And please keep in mind that Schenider and 
others who use discourse analysis, do not study 
single words, nor single sentences (although this 
can also be done). But they study chunks of text 
above the sentence level. So they must deal with 
scores of factors at the same time. No certain 
conclusions can be drawn on the basis of so many 
factors. Therefore, a preconceived theoretical 
system is made beforehand, and the text is 
studied in the light of this system. Thus, the 
conclusions cannot be tested.

I have mentioned it in several posts on this 
list, and I mention it again. There is one basic 
linguistic principle that is almost completely 
ignored in studies of Semitic verbs. This is the 
very opposite principle of discourse analysis. 
First semester students of linguistics learn it, 
and if they do not use it in their works, they 
will not get their exams-yet many Semitic 
scholars ignore it. The principle is that in any 
study of the verbal system we must distinguish 
between "semantic meaning" (meaning that is an 
uncancellable part of the verb form) and 
"conversational pragmatic implicature" (meaning 
that is caused by the context and is 
cancellable).  On the basis of this distinction, 
a very simple test can be made to find out 
whether tense is a part of the Hebrew verbal 
system: If a verb form is a tense, it will have 
the same temporal reference in any context, save 
special cases that can be explained, such as 
hypothetical conditional clauses.  I have made 
this test of all verbs of the Tanakh, and I have 
found that no verb form has a uniform temporal 
reference. Therefore, "tense" (grammaticalized 
location in time) is non-existent in Classical 
Hebrew. Schneider has not made such a test, but 
he has just assumed that tense is existent.

The ancient Semitic languages are typically 
aspect languages. This is true with Hebrew and 
Aramaic. However, modern Hebrew and modern 
Aramaic are tense languages, and at least in the 
case of Aramaic and Syriac (which is a part of 
the Aramaic family) there has been a gradual 
grammaticalization process from aspect to tense. 
But aspects die hard!  It is assumed in grammars 
and textbooks that Syriac (witnessed from the 1st 
century C.E.) is a tense language. But clearly, 
the grammaticalization was not completed at the 
time of the Peshitta translation. In my Syriac 
class today, we studied the last 15 verses of 
Matthew, chapter 24. Syriac QATAL is normally 
used with past reference, but in Matthew 24:33 
the QATAL form of the verb XZ) (to see) is used 
with future reference. Similar uses of QATAL are 
found elsewhere as well, and this shows that 
Syriac QATAL is not a tense.

I do not say that students of Hebrew students 
should not use discourse analysis. A small or 
great measure of discourse analysis is necessary 
in any linguistic study. But we should keep in 
mind what discourse analysis can do and cannot 
do. In my view, to understand the nature of the 
verbal system of Classical Hebrew it is 
completely useless.



Best regards,

Rolf Furuli
University of Oslo














>Dear List Members
>
>I want to invite your evaluation of the BH 
>verbal system as proposed by Wolfgang Schneider 
>and summarized below by John Sailhamer in his 
>class lecture notes.
>
>"In the Hebrew verb system, tense is used to 
>signal the orientation of the speaker to the 
>listener...In Hebrew the choice of tense depends 
>on whether the action is part of a conversation 
>or part of a narration....In Hebrew there are 
>two sets of tense systems: a primary tense and a 
>secondary tense. The primary tense is used to 
>express the main action of a passage in the 
>Hebrew Bible. The primary tense has only two 
>forms.
>      The secondary tense has only one form. It 
>is used to express actions which lie in the 
>background of those actions expressed in the 
>primary tenses. In the Hebrew Bible, these two 
>tense systems have merged to form one complete 
>system of tenses."
>
>He then identifies the primary tense for 
>narrative as the wayyiqtol and the primary tense 
>for discourse as the yiqtol. The qatal is 
>secondary in both narrative and discourse. I can 
>provide more details if anyone is interested, or 
>even thinks this is a valid approach to the 
>verb. It certainly is radically different from 
>time and aspect models. It seeks to ask not 
>"What type of action does a verb represent?", 
>but rather What type of communication does the 
>verb represent?".
>
>Your thoughts?
>Joseph Justiss			  		 
>_________________________________________________________________
>Hotmail: Trusted email with Microsoft's powerful SPAM protection.
>http://clk.atdmt.com/GBL/go/196390706/direct/01/
>_______________________________________________
>b-hebrew mailing list
>b-hebrew at lists.ibiblio.org
>http://lists.ibiblio.org/mailman/listinfo/b-hebrew




More information about the b-hebrew mailing list