<wayyiqtol> again

Rolf Furuli furuli at online.no
Wed Dec 29 02:48:16 EST 1999

Dear Peter,

>Dear Rolf,
>I am glad to see you active on this list again. Thank you for sharing
>with us the results of your surveys of QATAL and WEQATAL.
>It is clear that the categories into which you classified QATAL and
>WEQATAL are not the categories which distinguish these forms from
>other verb forms (not absolutely in the sense you wish, probably not
>even prototypically in the sense recently discussed). If they had been
>those categories, I would have expected to find for each verb form one
>or a small number of categories widely represented and others almost
>empty. So I accept that we can conclude that neither QATAL nor WEQATAL
>is a tense pure and simple. However, we cannot go on from there to
>suggest alternative distinguishing categories without first defining
>those categories carefully and making an analysis according to those
>categories. The problem with your suggestion that YIQTOL and WAYYIQTOL
>are imperfective, QATAL and WEQATAL are perfective, is that, having
>admitted that you do not follow well-known definitions of perfective
>and imperfective like Comrie's, you have not clearly defined how you
>define these aspects and can distinguish between them in a Hebrew

Regarding the definition of aspect, I recommend L:J: Brinton, 1988, "The
Development of English Aspectual systems", Cambridge: Cambridge University
Press. On p 1 Brinton wrote about aspects: "There seems to be more
uncertainty of the very definition of this grammatical category than any
other. There is no consensus about the object of study; widely diverse
phenomena are subsumed under the label of 'aspect'." In the following we
find a long description of different approaches. I respect Comrie highly,
but his definition of aspect has been criticised by several writers. He is
aware that many writers strictly differentiate between "aspect" and
"Aktionsart" but he himself has not made this distinction. However, there
is a strong tendency today among linguists (including Brinton) scrupulously
to distinguish between the two.

There is, however, a common denominator between Comrie, Brinton and most
persons who work with aspect, which can be expressed in three points:
(1) Aspect is a subjective viewpoint; i.e. it shows how the reporter
chooses to report an event.
(2) Aspect somehow makes visible, either a part (the imperfective aspect)
or the whole (the perfective aspect) of event time. ("Event time" is
non-deictic and is the time it takes to complete an event, say, from the
beginning to the end of drinking a glass of water.)
(3) The beginning and/or the end of an event is important for
distinguishing between the imperfective and the perfective aspect.
(Comrie would agree with all three points.)

What I have done with Hebrew is NOT to take *one* particular definition of
aspect in a deductive-nomologic way (using it just as a law of nature is
used in the natural sciences), but to use it in a hypothetic-deductive way
(almost as a hypothesis which is tested against the data). So I started
with the three points above (and the way they are expressed by Mari Broman
Olsen) to find out how they could be applied to Hebrew verbs. This lead to
a modification of Broman Olsen's definition, but the basic aspectual
properties in which most persons agree (including Comrie) are included in
my definition. This approach is simply to use the hermeneutic circle: to
interpret the data in the light of the hypothesis (the original definition
/three points/ of aspect) and to modify the hypothesis in the light of the
data. I have earlier described this in detail, and I suppose it can be
found in the archives. For interested ones I will recommend my own book
about the role of theology and bias in Bible translation. I has an Appendix
of 37 pages with a description of Greek and Hebrew verbal systems (see

>I am grateful to Galia for her clear definition of Reichenbach etc's
>three times. I think your deictic point (C) corresponds to Galia's
>speech time (S). It seems that when you decide whether a particular
>QATAL is past, present or future you are considering the relationship
>between E (event time) and S (speech time) - for I think your RT is
>not Reichenbach's R (reference time) but his E.
>Now I wish that you had been able to make a slightly different
>analysis of QATAL and WEQATAL, to determine the relationship between E
>and R, as defined by Galia. For my thinking (which could be confirmed
>or disproved by the type of analysis which you have done but with
>different categories) is that the Hebrew verb system is built around
>distinguishing relationships between E and R, and that S is of minor

Your view about E,R. and S in Hebrew studies is definitely wrong! The
important point for the study of tense is to find the C (deictic point)
which is contextually determined. ST (Galia's S, speech time) appears to be
the default or unmarked value for C, and the relationship between C and RT
is what really counts.

My system is in principle the same as Reichenbach's, Comrie's and
Galia's,that is, it accounts for the relationship between C (deictic
point), RT (reference time), and E (event time). However, Broman Olsen has
modified the scheme somewhat, to the effect that all tenses are relative
tenses (C is pragmatically determined) and Reichebach's "relative tenses"
are basically aspects. This means that the relationship between E (event
time) and RT (reference time) is *intersection* /RT intersects E either at
the nucleous (E is not completed)or at the coda (E is completed)/. Based on
this scheme, tense is the relationship between C and RT: Past tense- RT>C
(RT comes before C), present tense RT=C (RT coincides with C), and future
tense C>RT (RT comes after C). This means that tense is semantic and

Galia uses the scheme of Reichenbach and Comrie, her E is identical with my
E, her R is identical with my RT, but her S is not identical with my C. The
deictic point (C) is the point of time on which an expression is anchored.
Broman Olsen's modification is that she distinguishes between speech time
as (C) and other time points for (C) i.e. C is pragmatically determined,
while Comrie allways takes speech time as (C) in absolute tenses. Two
examples illustrating Broman Olsens scheme:

(1) I send an affidavit for your signature.
(2) I sent an affidavit for your signature.

Provided that the affidavit was enclosed, (1) expresses the sender's
persepctive and C (the deictic point) is identical with ST (speech time).
Example (2) takes the reader's viewpoint and C is not ST but the time the
letter is received. It is important to take this into account in Hebrew
studies in connection with a possible epistolatory QATAL. My C is therefore
in most cases identical with Galia's S, but not allways.The points above
indicate that I have taken all the important sides of Reichenbach's system
into account in my analysis of QATAL and WEQATAL, but because Broman
Olsen's system is simpler, for most practical purposes, all I have to note
is the relationship between RT and C for each QATAL/WEQATAL.

I would also suggest that the waw conjunction, when
>prefixed directly to the verb only, rather than the verb form is what
>signals the building of a new R time, i.e. it functions rather like
>the conjunction "then" in English. Thus both WAYYIQTOL and WEQATAL
>would build a new R time, and that is the essential difference between
>these two and the forms QATAL, YIQTOL.

Here we are in agreement, in fact, I have repeatedly stressed that the
we/wayy-element solely are conjunctions which has a syntactic and no
semantic role. Let me make some additional comments. Please compare the
following three passages from the NRSV:

(3) Deut. 11:26  See, I am setting (NOTEN) before you today a blessing and
a curse:
(4) Deut. 30:15  See, I have set (NATATI) before you today life and
prosperity, death and adversity.
(5) Jer. 21:8 .. See, I am setting (HINNENI NOTEN) before you the way of
life and the way of death.

In (3) and (5) a participle functions as a finite verb and in (4) we find a
QATAL in the same position. Do the participle and the QATAL have the same
semantic meaning? Nobody would claim that. But they have exactly the same
function - and there need not even be a difference in stress between (4)
and the other two, even though the NRSV translates them differently. Why
can two different verb forms be used similarly? Because the Aktionsart of
the verb, the nature of the verbal arguments (whether subject/object are
singular/plural, countable/uncountable, and definite/indefinite),
linguistic convention, and other factors are much more important than
*aspect* to convey meaning - and because the participle and QATAL have so
much in common that they can function similarly.

/As an illustration I will point out that regarding the Sumerian verb
neither the conjugation suffixes (roughly comparable til Hebrew binyanim)
nor the two basic conjuugations "Hamtu" and "Maru" (roughly comparable to
the Hebrew conjugations) are understood. And just the same seemingly good
translations can be made./

The participle has much in common both with the QATAL  and with the YIQTOL,
and it can have exactly the same function as both. However, there are also
differences. In some cases an author will use either the participle or the
QATAL  to stress a certain point, in other cases the choice seems to be a
matter of style or just incidental, and the meaning is the same. (A
participle of MWT could mean "on the point of dying" while a QATAL could
mean "was dead", but both a QATAL and a participle of YD( could mean "know"
without any difference.)

Just as the participle has something in common with both aspects, so YIQTOL
and QATAL has something in common. This indicates, in my view that both
aspects in many situations can be used without any difference in meaning
(e.g. the QATALs  and YIQTOLs in Psalm 2:1,2)- the *similarities* of the
aspects are here taken account of. In combination with verbs with
particular Aktionsart and particular syntactic factors - one of the aspects
is choosen to stress a certain point (e.g. conative situations are allways
expressed by the imperfective aspect) - the *differences* of the aspects
are here taken account of.

These points may give the following model for the use of YIQTOL,WAYYIQTOL,
QATAL, AND WEQATAL, neither of which are grammaticalized tenses:

In a world without tense, there may be a system based on linguistic
convention. Because of the nature of the perfective aspect, it is chosen to
depict events with past meaning, and because of the nature of the
imperfective aspect, it is chosen to depict situations with present and
future meaning. However, because the aspects only have a limited role in
conveying objective meaning, compared with Aktionsart and the other
mentioned factors, while one aspect basically was chosen for a particular
time reference, the Hebrew felt it was natural to use the other aspect to
portray a consecution of events inside this time sphere. Thus we got past
situations starting with QATALs and continuing with WAYYIQTOLs, and future
situations starting with YIQTOLs and continuing with WEQATALs. The waws in
both cases simply were conjunctions used to build new R-times as you say.
The human mind likes system, so no wonder that a QATAL, for instance in a
chain of WAYYIQTOLs could be used for pluperfect, for background
information etc. In narrative the rules of the linguistic convention are
strict,  in other material they are not so strict, and the "true" nature of
the verbs are easier seen. So I see Alviero's and Bryan's discourse systems
and Galia's R-time system as excellent results of long study describing the
linguistic convention of Classical Hebrew, but not as systems describing
semantic meaning.

My viewpoints are quite radical compared with the mainstream. Three good
questions for those who seriously want to test the viewpoints:

(1) It is taken for granted that Hebrew has four conjugations and not two.
Have you ever tried to figure out how greatly this aksiom influence your
interpretation of texts, and how important it is for your theory? Have you
compared this with the view that Hebrew just has two conjugations?
(2) Do you know any principles for differentiating between "semantic
meaning" and "conversational pragmatic implicature"? Have you ever tried to
use them systematically?
(3) Do you systematically use the difference between "past meaning" and
"past tense" in your studies?


Rolf Furuli
University of Oslo


Rolf Furuli
University of Oslo

More information about the b-hebrew mailing list