Questions for Rolf (LONG)

Rolf Furuli furuli at
Sun Apr 22 06:36:04 EDT 2001

Greg Doudna wrote:

>Your posts are very interesting. For the benefit of those on
>the list who may have tuned in late and missed earlier discussions
>(a way of asking for myself!), could you just state very simply
>what you see as the difference between yqtl and qtl (prefix and
>suffix) conjugations in your view? What are these two conjugations
>marking, why are they used, what is their difference in meaning,
>etc., or are you saying it is all totally free variation?
>You cite many examples from BH showing inconsistencies
>with conventional assumptions. However I have done a lot of
>work on the Qumran pesharim and related texts, and curiously,
>it looks to me (in the pesharim; not in 1QH) like the verb forms
>function very much like the naive ways which you attempt
>to refute, i.e. there is a consistent pattern of wicked figures named
>with sobriquets: their crimes are named as qtl/perfects (past),
>whereas their punishments (future) are named with yqtl (imperfects).
>The consistency of this pattern argues against any notion of
>free variation as the explanation.
>(I hasten to say this is not quite totally consistent; but the
>rare exceptions are reasonably explained as scribal errors.)
>And the waw-consecutives seem to 'reverse' the tense just as
>in the beginner's Hebrew-learning conception that most on this
>list would talk the beginner out of.
>Three possibilities: (a) is QH a reduction or development such
>that the 'naive' tense rules, which in fact don't work for BH
>(as per your and other data) DO work here, because they
>have BECOME that way? (b) maybe I'm reading the
>pesharim wrongly?, or (c) maybe some underlying single
>explanatory system would account for both QH and BH
>verb uses by a single set of rules.  (Maybe you have such
>a set of rules to propose?)
>In any case, thanks: basically could you just give your
>one-paragraph (or more) basic beginners' intro to your

Dear Greg,

I appreciate your questions, both because I know you are an expert on the
dead Sea Scrolls, and because your questions are important.


In my view there are several weaknesses with modern Hebrew studies:
(1) Conclusions regarding the meaning and use of verbs are delivered from
one generation to the next without ever being cheched (e.g. prophetic
(2) No attempt has been made for a systematic study of the *whole* corpus
of Classical Hebrew, so even those consclusions which are based on fresh
studies are based on a part of the corpus.
(3) Technical terms are often used without clear definitions (e.g. no
differentiation between "tense" and "time" or between Aktionsart and aspect)
(4) No attempt has been made to differentiate between semantics and
pragmatics (I am not aware of a single study which systematically has
sought to do that).


The combination between tense and aspect in English leads to twelve
different groups of verbs, simple past, past progresseve, pluperfect, past
perfect progressive etc. Suppose now that the tense elements were lost and
we only had the imperfective aspect "writing", the perfective aspect (has)
"written", and the infinitive "to write" to express our thoughts,how would
we do that? One could do as in modern Hebrew,namely, to grammaticalize the
forms (QATAL= past tense, YIQTOL=future tense, and participle=present
tense), but here the aspectual meanings are lost. Another way would be to
develop a system for the *use* of the forms which were at the disposition
of the speakers in accordance with their basic meanings. This can be a
process leading toward grammaticalization-a particular form gets fewer and
fewer uses until it has just one use (say past tense) which now has become
its uncancellable semantic meaning.

In my view, Classical Hebrew has four (or five) fundamental forms at its
disposition (in QAL): YIQTOL (=the imperfective aspect), QATAL (=the
perfective aspect), QOTEL (the participle, which is not the imperfective
aspect), and QETOL/QATOL (the infinitive(s)). In accordance with the
peculiar meaning of each form the linguistic convention which we find in
classical Hebrew has developed. A good place to start to realize this, is
to study the use of the infinitive and the participle. The infinitive
constructus and the participle are two different forms with different
meanings, but they have much in common as well. A study will reveal
particular patterns; sometimes the use/meaning of the two do overlap, but
for the most part the pattern is kept (e.g. preposition+infinitive+suffix
as a temporal subordinate clause with the suffix as subject).

Regarding the finite verbs one of my fundamental assumptions is that a
difference in form suggests a difference in meaning. But how many finite
forms are there? In *all* instances of WAYYIQTOL,WEYIQTOL, and WEQATAL the
prefix WA/WE functions syntactically as a conjunction (particularly "and"),
and there is no uncancellable meaning connected with either of the forms
(although 95 % of the WAYYIQTOLs have past reference). I therefore conclude
that there are only two different forms of finite verbs YIQTOL and QATAL,
that the prefixes of some of them are syntactic (conjunctions) and not
semantic, and that the pattern that we can see is the result of linguistic


To elucidate the Hebrew aspects I take your questions as a point of departure.
It seems to me that the DSS use the verbs in the same way as we find in the
Tanach with minor differences (e.g. Hiphil is used as Qal etc). So I think
you read the texts correctly, and what you see is the conscious pattern of
the author. As suggested above, this is based on linguistic convention, but
what about the *meaning* of the fundamental forms used.

I have recently completed a study of all the YIQTOLs with prefixed WAW in
the 70 % of the DSS which are found on the Gramcord CD. Of the 607 examples
which the CD analyses as WAYYIQTOLs I found 342 which fulfilled the
criteria, 1) the verb was seen on the fragment (not reconstructed), and 2)
there was a context by the help of which temporal reference and modality
could be ascertained. The results of my analysis of temporal reference and
modality is as follows:

	PAST	200	58,5%
	PRESENT	67	19,6%
	PERFECT 56	16,3%
	FUTURE	12	3,5%
	GNOMIC	2	0,6
	MODAL 	3	0,9

My analysis of of the forms which the CD identified as WEYIQTOL is as follows:

	PAST	36	14,5%
	PRESENT	53	21,3%
	PERFECT 11	4,4%
	FUTURE	59	23,7%
	GNOMIC	1	0,4%
	COMMAND	32	12,8%
	MODAL 	38	15,2%
	FINAL	11	4,4%
	OTHER	8	3,2%

The biggest difference between the WEYIQTOL data from the DSS and from the
Tanach is that only 33 % are modal at Qumran while 64% are modal in the
The numbers do not prove that there is a free variation.


I would like to make a brief presentation of my approach. I operate with
three kinds of time:
1) TENSE  This is deictic time because it is seen in relation to a vantage
point which is called "the deictic point" (C); events with past tense
happens before (C), events with future tense happens after (C), and events
with present reference coincides with (C)

2) EVENT TIME This is the time the event takes from beginning to end. It is
non-deictic because it is not seen in relation to a deictic point.

3) REFERENCE TIME. This can be compared to a "pointing finger" which our
eyes follow to a particular part of event time. Reference time makes a
particular part of event time visible by referring to it; it intersects
event time as it were. It is non-deictic because it is not seen in relation
to a particular deictic point.

The deictic time represents tense, and the interplay of event time and
reference time represents aspect. The imperfective aspect in English is
expressed by the participle and the perfective aspect is expressed by
perfect. The whole English verbal system can be seen as a combination of
tense and aspect. The aspects are "objective", because we allways can draw
the same conclusion regarding their meaning. When the imperfective aspect
is used, reference time intersects event time at the nucleus, and we know
that the event was in progression at reference time. When the perfective
aspect is used, reference time intersects event time at the coda, and we
know that the event was terminated at reference time.


No hebrew verb form has a uniform time reference, so tense is absent, As to
Hebrew aspects,we cannot take for granted that they have the same
properties as their English counterparts, but their properties should be
tested. It is possible to get a clearer understanding of the nature of
aspect by testing the interplay of reference time and event time by the
help of three parameters, "the quality of focus","the angle of focus", and
"the breadth of focus". In order to get some idea of these parameters, I
include a small section below from a chapter that will be included in an

"The quality of focus

When we speak about the "quality" of the focus, we are concerned with how
the nature of the event is made visible. This relates to the "distance"
between the eyes of the observer and the area where reference time
intersects event time. We can allustrate this with a camera. A closeup
picture of a thing reveals the details of a small area of the object quite
clearly, but a picture taken from some distance, while showing a greater
part of or the whole object, reveals no details.
If we look at (7) and (8) we can learn something about the quality of
focus.  Example (7a) is perfective, and what is made visible is that
sometime in the past Abigail knocked at a door; whether she knocked one or
several times is not visible. Only the end of the event is made visible but
no other details. Example (7b) is imperfective, and the only possible
interpretation is that Abigail knocked several times, so we learn that the
knocking event was in progress at reference time. When progression is made
visible, details are focused upon. Thus the quality of the imperfective
aspect is a closeup view of the area where reference time intersects event
time while the perfective aspect is a view from "some distance" without
details made visible.
(7a) Abigail had knocked at the door.
(7b) Abigail was knocking at the door.
(8a) Elisabeth was reaching the top.
(8b) Elisabeth was reaching the tops.
(8c) The people were reaching the top.
(8d) Elisabeth has reached the top.
(8e) The people have reached the top.
To further illustrate the quality difference between the two aspects, let
us concentrate on the imperfective one, and the fact that when this aspect
is used in English, the reader or listener expects to see action in
progress *somewhere*, that is, one expects to see details. This is evident
in (8). In (8a) both subject and object are singular and definite.
Therefore any progressivness cannot be accounted for by a plurality of
persons or things. The only possible interpretation of (8a) is that
Elisabeth was "on the point of" reaching the top. The linguistic reason for
this interpretation is that the imperfective aspect is used, and that
progressive action is visible *somewhere* is therefore expected. To "reach
the top" is not an event but rather a point, and therefore it is void of
progression. The use of the imperfective aspect with momentary actions such
as "knocking", implies an iterative interpretation - several knocks occur.
But the nature of the "reach the top"-event excludes such an
interpretation. Thus, to make progressive action visible in (8a) the only
place to see it is just before the top is reached.
In example (8b) the object is plural, and therefore the event is iterative,
top after top is reached. In (8c) the subject is plural, and
progressiveness is made visible by the notion that one after the other
reached the top. In (8c) and (8d) the perfective aspect is used, and
regardless of whether subject or object is singular or plural, the only
thing that is made visible, is that the top was reached. We therefore see a
fundamental difference between the aspects as to the "distance" from which
an event  is described. The imperfective aspect is a closeup view of
progressive action while the perfective aspect is a viewpoint from some
"distance" where the termination of the event is made visible but not its
details.(The term "distance2 is conceptual rather than factual.)

I have applied the three parameters to the Hebrew and English aspects with
the following result:

		   The quality of foc.	The angle of foc. The breadth of foc.

Particularly the fact that the angle of focus (the position of the area
where reference time intersects event time) is different in both aspects
makes the Hebrew aspects very different from the English ones; we cannot on
the basis of the use of Hebrew aspect know whether the event which is
described was in progression or was terminated at reference time, as we can
in English. To account for the relationship of reference time and event
time in relation to the three parameters, I define the Hebrew aspects this

"The imperfective aspect is a closeup view of a small section of the event
where the progressive action is made visible. The perfective aspect is a
view, as if from some distance, of a great part, or of the whole of the
event, where the progressive action is not made visible."


There is not a totally free variation in the use of the two aspects, but
there are many restrictive factors. But even without the other factors,
there are restrictions on  the use of YIQTOL  versus QATAL because of the
nature of the aspects themselves. To continue along the line of the
"reach-the-top"-example above and the quality of focus: when YIQTOL,
WEYIQTOL, and WAYYIQTOL are used we expect to see progressive
action/continuing state. This is the reason why conative situations allways
are expressed by one of these forms and never by QATAL and WEQATAL, as in
(9) and (10)

(9) Genesis 37:21 NIV When Reuben heard this, he tried to rescue him
(WAYYIQTOL) from their hands.
(10) Exodus 8:18 NIV But when the magicians tried to produce (WAYYIQTOL)
gnats  by their secret arts, they could not.

Other restrictions for the use of the aspects are procedural
characteristics (which are functions of durativity, dynamicity, and
telicity), the number and definiteness of subject and object, the nature of
particles and adverbials, syntactical traits, and last but not least,
linguistic convention (e.g. modal verbs (YIQTOL) and direct speech (QATAL)
only are sentence initial).

In my model, the two aspects, the two infinitives, and the one participle
were the forms at the disposition of the Hebrews in ancient times. On the
basis of these a linguistic convention (pattern) had to be developed. Forms
signalling tense are absent, and as a general tool for past reference QATAL
was chosen (7.446 (53.5%) QATALs in the Tanach have past reference. For
present and future reference YIQTOL was chosen (5.451 (38%) YIQTOLs have
future refence and 3.376 (23,6%) have present reference.) To express
modality, short or extra long forms are used,the negation )AL, and sentence
initial position. (At Qumran )AL and LO are used interchangeably). We may
perhaps call this the basic pattern. There is some variation of the basic
pattern due to suntactical and contextual factors,for instance, 394 (2,8%)
QATALs are modal and 762 (5,3) YIQTOLs have past reference.

Then we have the YIQTOLs with prefixed WA/WE and the QATALs with prefixed
WE. When we realize that tense is absent in Hebrew, these forms cannot be
said to be "reversed tenses", rather they are prefix-forms and suffix-forms
with prefixed WA/WE which to a great or to a lesser degree are used with
the same temporal reference as the "opposite" suffix-forms and prefix-forms
without prefixed WE. (14.091 (95%) WAYYIQTOLs and 53 (4,4%) WEYIQTOLs have
past reference, and 4.100 (67,4%)  WEQATALs have future reference, and 240
(3,9%) have present reference. Also interesting: 2.505 (18%) of the QATALs
have present reference.) How can the pattern of these forms be explained?

WEYIQTOL. Of this form 783 (64.3%) are modal while 1258 (20,7)% WEQATALs
are modal. The kind of modality of the two forms are generally different,
so WEYIQTOL is clearly a YIQTOL which primarily is used when two or more
particular modal ideas are expressed, and therefore it has a prefixed WE

WAYYIQTOL. This form basically has a discourse function - it is the
narrative form. Again, if tense is excluded from our thinking, and we
accept that the aspects which are expressed by YIQTOL and QATAL are
subjective, i.e. we cannot conclude on the basis of the aspect whether the
event was terminated or not at reference time, it is not necessarily
strange to choose this form for narrative past. Given my model, there are
just two options for the narrative form, either to choose the form which
generally are used with past reference - QATAL, or to choose YIQTOL. The
choice of YIQTOL would mark the narrative as a particular genre, and there
can be argued for finer aspectual nuances as a reason for the choice as
well. And remember, the other form QATAL do neither represent completed nor
complete action, even though complete action can often *pragmatically* be
associated with QATAL. Therefore, the notion of the chain of completed
actions in the narrative must be gathered from the context and not from the
form used. Remember also that the participle in Hebrew is not equivalent
with the imperfective aspect. A narrative expressed by WAYYIQTOLs,
therefore, does not tell the reader that "they were coming, they were
fighting, and they were killing". WAYYIQTOL (or other YIQTOLs with past
meaning) more seldom express a notion which best ca be expressed by
durative past. What is done when WAYYIQTOL is used in narrative, is that a
small area of event time is made visible, and progressive action is seen -
something which to some extent can be compared with historical present
(which earlier has been suggested by others on the list). The conjunction
WAW is typical for narrative and in all cases of WAYYIQTOL is a conjunction
syntactically required or at at least is syntactically compatible.

WEQATAL. This form is a QATAL with a prefixed conjunction, and as in the
case of WAYYIQTOL, the conjunction syntactically required or at at least is
syntactically compatible. The 1258 (20,7%) examples with modal meaning
occur generally where another modal thought is expressed, so they are bound
together by WAW. Many of the 4.100 (64,7%) of the forms with simple future
can be used on the basis of aspectual nuances, opposite of those connected
with WAYYIQTOL. The 357 (5,9%) examples with past reference occur in
contexts where events which are bound together occur.

I would like to add that there are several YIQTOLs which occur in contexts
where we would expect a WAYYIQTOL, and the reason why they are not
WAYYIQTOLs is that there is an element between the WAW and the YIQTOL.
Similarly there are several QATALs in contexts where we would expect
WEQATALs, and the reason why they are not WEQATALs is that there is an
element between the WAW  and the QATAL.

I see absolutely nothing strange in the pattern of Hebrew verbs, at Qumran
or in the Bible, given that there are two aspects as described above. We
should also remember another important factor where I yet have no complete
statistics, namely, the difference between stative and fientive verbs.
Because any part of a state is similar to any other part or to the state as
a whole, the intersection of the state by reference time between its
beginning and end will make visible exactly the same when either aspect is
used. So the choice of aspect in such situations is not necessarily
important. This can also account for the use of a few or many verbs.

My conclusion is that while the imperfective aspect *can* be used with all
kinds of events, and the perfective aspect *can* be used with most kinds of
events, there are so many restrictions in Classical Hebrew that we are very
far from a free use of the aspects. The pattern we see has been remarkably
stable through several hundred years, and we see the same general pattern
in all the the books of the Tanach and at Qumran, although we see some
minor changes in the last books and at Qumran.



Rolf Furuli

University of Oslo

More information about the b-hebrew mailing list