Biblical Hebrew Forum

[Peter Kirk <peter AT qaya.org>]

On 10/03/2007 13:04, Rolf Furuli wrote:
> Dear Peter,
>
> The question under discussion was whether we can distinguish between 
> WAYYIQTOLs and WEYIQTOLs in unpointed texts. The answer is No! I have only a 
> part of my data base where I am at present, but I quick look  at what I 
> have, revealed that 90.5 % of 3.p.s.m. WEYIQTOLs of lamed he verbs were 
> apocopated. ...

OK. How many of these WEYIQTOLs are jussive? What is the percentage of 
those which are not jussive? Even 90.5% is statistically significantly 
different from 98.7%. Actually I would dispute these figures based on 
the ones I find below (which combine masculine and feminine) which 
suggest more like 50% of forms which can be apocopated (63 apocopated, 
57 not apocopated).

> ... While jussive forms may be apocopated, that is not always the 
> case. I analyse 5,117 YIQTOLs as modal, yet I found only 298 apocopated 
> forms of these. ...

Out of how many which could be apocopated? On what basis did you analyse 
some YIQTOLs as "modal"? ("You should not be too quick to give your 
interpretation when you see verb statistics.") How does your basis 
compare with the traditional understanding of "jussive", or with the 
"jussive" markings in the Westminster Hebrew database? It would be 
possible and interesting to analyse this database to find the 
distributions of YIQTOL, WEYIQTOL etc forms of lamed he verbs which are 
and are not apocopated. See below where I do this.

> ... Many occur after the negativeparticle )L. Of 598 exmples of 
> YIQTOLs preceded by )L, I found 123 apocopated forms and 30 long forms of 
> those verbs that could be apocopated. ...

This is the expected pattern in what are clearly jussive forms; it seems 
that rather more than 10% of jussives are not apocopated when they can 
be, a much higher proportion than for WAYYIQTOLs.

> ... You should not be too quick to give 
> your interpretation when you see verb statistics. While quantity comparisons 
> are important, quality assessments are even more important. To do that you 
> need to work with the text from which the statitics is taken, using 
> discourse analysis, looking for word order, modality, stress patterns, 
> suffixes, and peculiarities of each root. Without this your interpretation 
> cannot bebalanced.
>
>   
I agree that these factors may also be important. However, if I see a 
clear correlation between Masoretic pointing as WAYYIQTOL and 
apocopation, even if the match is not precise, that is evidence that 
there is a distinction here which goes back before the Masoretes.

So, let's look at the evidence:

From a morphological tool based on the Westminster Hebrew database, I 
count a total of 1491 WAYYIQTOLs of all lamed-he verbs which are marked 
as apocopated (tags "v?w3?s?a"), and 127 which are not marked as 
apocopated (tags "v?w3?s"), excluding forms with suffixes. But in fact 
only 34 of the 127 end with a he and so are truly not apocopated. So I 
conclude that there are in fact 1584 apocopated and 34 non-apocopated 
3rd person singular WAYYIQTOLs of lamed-he verbs, that is 2.1% 
non-apocopated - actually a higher figure than you gave.

For YIQTOLs of lamed-he verbs, 3rd person singular only with no 
suffixes, the figures are as follows:

Not apocopated, not jussive meaning: 933 (tags "v?i3?s"), of which 51 
are preceded by the conjunction (tags "v?i3?s Pc")
Not apocopated, jussive meaning: 20 (tags "v?i3?s{1}Jm"), of which 6 are 
preceded by the conjunction (tags "v?i3?s{1}Jm Pc")
Apocopated, not jussive meaning: 20 (tags "v?i3?s?a{1}Jf"), of which 8 
are preceded by the conjunction (tags "v?i3?s?a{1}Jf Pc")
Apocopated, jussive meaning: 150 (tags "v?i3?s?a{1}Jt"), of which 55 are 
preceded by the conjunction (tags "v?i3?s?a{1}Jt Pc")

Of all these non-jussive forms (as interpreted by the Westminster team), 
20 out of 973 are apocopated, 2.1%. Of the WEYIQTOLs in this group, 8 
out of 59 are apocopated, 13.6% (1 Samuel 10:5; Job 36:15; Isaiah 63:3; 
Lamentations 3:50; Ezekiel 14:7; Daniel 11:4; Joel 2:20; Micah 7:10). 
Although this shows a clear tendency for apocopation to be more common 
after the conjunction, this is still only a small minority of the cases. 
Let me summarise the results:

          apoc  non-apoc
WEYIQTOL      8     51
WAYYIQTOL  1584     34

Thus there is a clear and strongly statistically significant correlation 
between WAYYIQTOL forms (as pointed by the Masoretes) and apocopation of 
lamed-he verbs, if we do not include forms with jussive meaning. This is 
clear proof that the Masoretic pointing of WAYYIQTOL was not a Masoretic 
invention but goes back to the consonantal text.

--
Peter Kirk
E-mail:  peter AT qaya.org
Blog:    http://www.qaya.org/blog/
Website: http://www.qaya.org/