WEQATAL vs. QATAL statistics, cleaning up assumptions

Rolf Furuli furuli at online.no
Wed Aug 23 17:23:16 EDT 2000

Dear Randall,

It is very difficult to come under your skin, but I will try a last time
regarding the assumptions behind Henry's statistics.

>clarity requires a full citation. Please excuse the length.
>>>the following needs to be cleaned up logically:
>>>> If there should be any meaning in a
>>>>comparison of the temporal reference of QATAL and WEQATAL, we first must
>>>>decide (assume) that they ARE  two different groups (Catholics and
>>>>Protestants). But then we have assumed what we should prove, namely,
>>>>QATAL  and WEQATAL *are* two different groups.
>>>Come on, clean up the act.
>>>qaTALti and veqatalTI are two different forms and do not need an
>>>in order for them to exist. They exist.
>>>People are not interested in hearing someone throw around 'debate
>>>but they would appreciate seeing that true reflection and understanding
>>>have taken place.
>>The forms exist graphically, that is correct,
>I was speaking about the phonological forms qaTALti vs. veqatalTI.
>'Graphically', here should refer to the vocalized MT, which is graphic. And
>please remember that linguistic 'forms' (as opposed to meaning) generally
>refer to the base medium of the language. In that sense the graphs, even
>the MT vocalizations, are only a partial represenation of the form. For
>example "dbr" is a graph but the linguistic forms for this graph include
>dibber and davar, among others. [The graph 'shin' even hides the fact that
>an extra phoneme existed in the language ('sin') which later joined with
>samex and not 'shin'. The Masoretes have been good enough to record these
>'sin's for us.]
>>and the accentuation is
>>different in a few forms (enough forms to show that the Masoretes
>>distinguished between them).
>Yes. Exactly. the Masoretes did distinguish them and in another post you
>implied that this was a pre-linguistic knowledge based on intuititvely
>learning the language and not on grammatical analysis.
>>Your error, in my view, is that you *presume*
>>that the stress is phonemic, but this is completely unfounded.
>We have a communication problem here. First, stress in Hebrew is phonemic,
>as everyone agrees: E.G. SHAva 'she returned versus shaVA 'she is
>returning'.  There is no 'presumption' here but a general fact. And stress
>has many conditioned euphonic changes like in pausal forms or with nasog
>But moving the stress to the end of a 1s,2s suffix conjungation verb
>qatalTI/qatalTA is not one of the euphonic changes. However, with veqatalTI
>and veqalTA stress moves to the end. It is not phonologically motivated. It
>is a different morphological class. It is good that you agreed that the
>masoretes distinguished them. They need to be included in an system.

As we all know, a difference in stress *can* be phonemic in Hebrew, but it
*need not* distinguish between different meanings. When it is phonemic, we
have the following situation: Two words are written similarly, and they are
differentiated by the position of the stress. This differentiation is
consistently made in *all* cases of the word in question. However, nobody
will claim that pausal forms are phonemic, yet they are stressed
differently than the same forms in context. The stress of the WEQATALs is
*not* consistently made in all cases, not even in the case of the same
groups of verbs (e.g. lamed he or hollow verbs). There is no uniform
meaning/reference in the WEQATALs which differ from the meaning/reference
of QATAL, but both groups overlap completely, although the number of verbs
with particular meanings is different in WEQATAL compared with QATAL Thus
the ultimate stress of some WEQATALs do not fulfill any  of the
requirements for phonemic stress.

My principal claim against Henry's conclusion that my data showed that
QATAL  and WEQATAL were two groups with different meanings, was that that
this is circular because one first have to assume that QATAL and WEQATAL
were two distinct groups. If I understand you correctly, you admit that it
is necessary to treat them as two distinct groups in order to use the
statistics as mentioned; and I appreciate this admission.

I suppose we agree that a QATAL form is a finite verb, and WAW is a
conjunction. I also suppose you agree that there is not a single example of
WEQATAL where the syntax or context would not allow us to interprete the
WEQATAL as WAW+QATAL (conjunction+verb). In other words, there is not a
single example of WEQATAL where the WAW *must* be something other than a
simple conjunction. I therefore suggest that WEQATAL is a QATAL which is
used in a syntactic position where the conjunction WAW is needed. The onus
of proof is on you to show that this simple explanation is not true,
because it is you who say that WAW+QATAL is more than a conjunction + a
verb in the perfect conjugation.

The only thing you have presented which can be called an argument against
this, is the stress difference. But a shown above, this argument is weak

>>Henry has
>>correctly argued that the stress distinction between QATAL  and WEQATAL
>>cannot be diachronically traced back to some older forms.
>And here is your error. You have added a 'diachronic' argument to the
>synchronic status and interpretation of veqatalTI. You are arguing or
>implying that because veqatalTI does not have a proto-Hebrew, or
>proto-Semitic source you may remove it from the synchronic layer that it is
>in. BH as recorded by the Masoretes has a veqatal morphological category.
>Most people I know assume that the stress shift is an innovation in
>Biblical Hebrew within the Northwest Semitic languages. But the pre-history
>of its pedigree or source is irrelevant for its existence. Would you mind
>retracting your statement "you presume that the stress is phonemic, but
>this is completely unfounded",?
>>My suggestion is
>>that in the chanting in the Synagogue the 'default' understanding ( to use
>>your words) was that QATALs with prefixed WAW for the most part was used
>>with future reference, and therefore these forms were marked with ultimate
>I don't understand 'therefore', here. Actually, the forms are quite often
>used to mark habitual in past contexts, and forms like Hif`il plurals
>retained their stress on the 'i' vowel and not on the suffix. So once again
>the Masoretes surprise everyone by NOT systematically regularizing and
>changing something but apparently leaving things the way they received them
>as best they could. [I am not arguing that MT dialect retains ancient
>phonetics. Only that they did not regularize the grammar. Fortunately for
>everyone, they remained grammatically naive and didn't fudge with the
>data.] by the way, most Israelis read right over this accentual
>distinction. while it shows the categoriy status to the grammarian the
>system is able to function without it.

By "completely unfounded" I mean that there is not a single piece of
evidence which *must* be interpreted in the way you say, that is, there is
no single piece of evidence which cannot be interpreted in a different way
than you do.

I for one view the Masoretes as extremely faithful copyists. Most likely,
the ultimate stress on some WEQATALs was something they heard in the
synagogue. However, the biblical text was not written for the purpose of
chanting it in the synagogue. Therefore, at the point when the synagogue
use started, or some time later, the tones and accents connected with the
chanting were instituted. True, the chant often follows the supposed stress
patterns, but not allways. What we primarily are efter, is not synagogue
use, but the original stress and meaning. Regardless of whether Jeremiah
the prophet actually said what he is reported to have said, the author
presented it this way. So the question is: With what kind of intonation was
the expressions uttered or read originally (I use the book of Jeremiah)?
There are two problems with your view that this question reveals. First, we
find a great spread of meaning in the WEQATALs with ultimate stress: past
meaning, future meaning, imperative,volitional, apodosis, final clause etc.
(in other books we have present, perfect, prothasis, gnomic etc). What is
the common denominator of all these that would justify putting them in one
group with one kind of special stress? Second, there is a fact that the
intonation is different in different kinds of clauses; e.g. we expect an
imperative to have another intonation than a final clause. An ultimate
stress, therefore, is not to be expected in all kinds of clauses in the
original expression or writing of the words. That Jeremiah should have used
an artificial ultimate stress contrary to his natural intonation, is not
substantiated with a single piece of evidence, and it simply is not likely.

Chanting, however, is different. We see this in the pausal forms. I suppose
you don't claim that the long vowels and the position of stress in pause
has a distinctive meaning compared with the same forms in context. Nobody
knows how Hebrew was pronounced in the days of the second temple, let alone
the first temple. A little before the Masoretes started their work with
pointing, Jerome (whom you recently mentioned) pointed out that there were
different pronunciation and stress traditions. Wrote he (Epist.73 ad
Evangelum) "It does not matter whether it is called Salem or Salim, because
the vowel letters (=matres lectionis) are used by the Jews very rarely in
medial positions and same words are realized with different (vowel) sounds
and accents in accordance with the will of readers and regional
distinctions." Evidence of this is also seen in the difference in pointing
in the Palestinian, Babylonian, and Tiberian systems of Masoretic pointing.
this is not only a difference in the graphic signs, but in the pointing of
WAYYIQTOLs versus WEYIQTOLs as well.

A suggestion, therefore, is that a combination phonological rules regarding
stress, syllables and vowel length and the rhytm in reading narrative
accounts can account for the penultimate stress of some WAYYIQTOLs and the
opposite in WAW+QATALs of 1st and 2nd person, singular (when they are used
contexts which are different from the normal past contexts). This view does
not presume that the Masoretes invented something new, although they *may*
have done that by their choice of points, even without having this
intention. But I do not think that the pausal forms with the stress on the
end of sentences, reflect the way Jeremiah and others pronounced their
sentences, and this is the point we have to  address if we want to know the
*meaning* of the verbal forms. The consequence is that at some point after
people started to chant the text in the synagogue, the pausal forms and
possibly other accentuations were invented. This might have happened before
the Masoretes started their work.

I therefore do not see a single piece of evidence against the view that the
ultimate stress of some WEQATALs, is of the same nature as pausal stress
and is non-distinctive, although it may have evolved on the basis of the
use of QATAL + waw in contexts which were different from the normal past
use of QATAL.

>>in contrast with WAYYIQTOLs with penultimate stress.
>Again, a little problem. Most BH vayyiqtols have final stress. It is only
>some of the weak verbs that show the accent shift.
>>This need not
>>be more than a *functional* view, an observational view, not a semantic
>to be honest, this statement doesn't communicate. 'Observational' is
>linguistic term? Most linguists would call differing morphology for
>different interpretations to be 'semantic'. If by 'semantic' you mean
>'uncancellable meaning', then by that definition it is not semantic and
>will become unverifiable. [Have you read Jerry Katz  "Common Sense and
>Semantics"? He is an old time generative-schoolist who is not very
>appreciative of semantic theories that contravene what language users
>intuitively know. And if semanticists can contravene living languages, just
>think what they could do with the dead.]
>>To use Catholics and protestants again. The people exist, regardless of
>>their religious faith. But to make a study of the difference in income, it
>>is not enough to study people in Catholic neighborhoods and compare them
>>with people in Protestant neigborhoods. What is necessary, is to make a
>>*semantic* distinction and decide that these particular people are
>>Catholics and these are protestants. Similarly, it is not enough that
>>QATALs and WEQATALs *exist*, but a semantic decision is necessary. You
>>decide that the stress difference is phonemic, and that they are two
>>different semantic groups. In other words, you must in a scientific way
>>demonstrate that my claim that they are one and the same group, is wrong.
>>That is my challenge, and this is not tactics.
>I think that what has been said above already answers and reverses your
>statements here. A semantic distinction is not necessary for defining the
>groups veqatalTI and qaTALti. They are different formal categories and as
>different formal categories may be assumed to have a different meaning. And
>it doesn't matter whether that meaning is an uncancellable semantic meaning
>or some kind of pragmatic-semantic meaning. The difference is there. I
>think you've already admitted that the Masoretes, in fact, did see that
>they had a different meaning, though I don't accept that the Masoretic
>meaning was specifically 'future'.

The list-members may be bewildered because of the many different
expressions. Let me try a last, simple illustration. It is possible to
group the participles in two ways: 1) because of temporal reference, and 2)
because of prefixed waw/no prefix. If we use classification 1) we get four
groups: past reference, present reference, perfect reference, and future
reference. We can say that each group has a different function, or if we
like, we can say that each group has a different meaning. But there is no
difference in the "semantic meaning", because in all the four cases, we are
dealing with the same small semantic unit - the participle. To apply this
to our discussion, it is not meaningful to use the 4 different time
references of the participle to establish four different semantic groups.

 It is exactly the same if we group the participles in a group without
prefix and another group with prefixed waw. Here we even have a graphic
difference. But it is completely meaningless to compare all the functions
of the group of WAW+participle with the group of participle without WAW,
and then calculate the odds for the two groups having different meanings.
Both are participles and therefore they have the same meaning! For such
calculations to have meaning, we must first decide that WAW+participle *is*
a different fundamental unit compared with participle without WAW. If this
is established, the calculations are meaningful. The only way to circumwent
the circularity of assuming what is going to be proved (WEQATAL  and QATAL
are two semantic groups, and on this basis we can show by their different
use/references that they are two different semantic groups), is to
demonstrate that the functions of The WAW+participle group and the
participle-without-WAW group are mutually exclusive. This means that the
functions are so diverse that it is impossible that the same single group
can have them all.

Applied to Henry's statistics, either it must be assumed (before we start
the calculations) that QATAL and WEQATAL *are* groups with different
semantic meaning, or it must be assumed that Hebrew is a tense-language,
and the same forms therefore cannot have past and future reference. I claim
that Hebrew is no tense-language, and that QATAL and WEQATAL is one and the
same form. You have not demonstrated the contrary in any of the cases.

>which leaves us with four-plus morphological categories needing
>definitions: qatal, yitqtol, vayyiqtol, veqatal.
>Now, how are you going to prove that yiqtol has the same 'uncancellable'
>meaning as vayyiqtol? You cannot just assume them to be the same. And
>proving that the abstract linguistic parameters of tense and aspect are
>'not uncancellable' with yiqtol and vayyiqtol becomes irrelevant since that
>only says what they are not, and that negative result is only according to
>a particular constricted definition. I further wonder if such definitional
>constrictions aren't crossing over that common sense line that Jerry Katz

My approach start from the very opposite point of yours. I start with the
graphic difference that we see in unpointed texts - prefix-forms and
suffix-forms.(That some prefix-forms are shortened does not at the outset
indicate a third group). I cannot imagine that any linguist will say that
this is not a logical starting point.

After noting that Origen did not distinguish more than two conjugations, my
next step is to look at the Masoretic text where we find  the WAYYIQTOLs
which *seem to* be different from the YIQTOLs. But in contrast with you,
who seem to take for granted that the two *are* semantically different, and
demand that I prove that they are not, I ask: What is this WA(Y)-element
before all these YIQTOLs? Looking at the attempts to explain this element,
I find an enormeous amount of speculation, but *not a single* study that
has been able to substantiate a view that this WA(Y) is anything but a
simple conjunction. Therefore I need not prove that it is not
WAW+conjunction, but those who deny this, should prove their case.

So to my next question: If WAYYIQTOL simply is a YIQTOL+WAW, which by
linguistic convention has come to be the verb form that carries the
narrative forward, can I account for the fact that 95 % of the WAYYIQTOLs
have past reference while only 5 % of the YIQTOLs and 4,5 % of the
WEYIQTOLs have this reference? And further, can I account for the fact that
the short form is chosen in 95 % of the WAYYIQTOLs where there is a short
form, in view of the fact that the short form is connected with modality
both in Hebrew and in the other Semitic languages?
This is the best approach in my view because it is not dependent upon so
many assumptions as your approach is.



Rolf Furuli

University of Oslo

More information about the b-hebrew mailing list