discourse analysis and circular reasoning

Rolf Furuli furuli at online.no
Mon Apr 15 06:40:16 EDT 2002

Dear Peter,

There is no reason why you should be a linguistic back-packer even 
though discrourse analysis is just descriptive. The simple way is to 
study individual forms one by one, and then draw your conclusion. 
This can reveal whether a particular form is a tense, an aspect, or 
has other characteristics. Please look at (1) and (2) below (I use 
participles to avoid the problem of conjugation number).

(1) 1 Kings 1.22 While she was still speaking ( (t®rR;bådVm 
hΊn®dwøo) with the king, Natan the prophet arrived (aD;b).

(2) 1 Kings 6:33 While she was still speaking (rE;bådVm ŠwŠn®dwøo 
with them, look! the messenger came down  (dérOy) to him.

In both examples the action of the second verb (in (1) the second 
verb is a QATAL or a participle, in (2) it is  a participle) 
intersects the first verb. From this we draw the "normative" 
conclusion that a participle *can*  be intersected by another verb in 
the middle, even by another participle; no disocurse analysis is 
necessary for that conclusion. If we find hundreds of examples of of 
participles where not the end but the middle is focussed upon, we can 
conclude that this is an important characteristic of the form. If we 
go further and analyse *all* the occurences of the participle (we 
need not even put much stress on whether the form is foreground or 
background) and compare the results, we can find a generalization 
that covers all the results of our analyses.

You can see my analysis of the participles below. First vertical 
column: the passive participles, second column: the active 
participles, third column: both participles, fourth column: 
percentage of all participles.

PAST	 	 339	1406	1745	20,2
PRE-PAST	   25	    69	    94	  1,2
PERFECT	 299	  143	  372	  4,3
PRESENT	 221	1772	1993	23,1
FUTURE	  	  97	  573	  670	  7,8
MODAL	  	  74	   66	  138	  1,6
GNOMIC	  	  42	 569	  610	  7,0
IMPERATIVE	    3	    5	     8
SUBSTANT	 304	2073	2377	27,5
ADJECTIVE	 329	  283	  612	  7,1

TOTAL	1733	6959	8 692

I have done exactly the same with the finite forms. I would like to 
mention that  the results of a simple discourse analysis of the 1.020 
YIQTOLs with past reference, is that 889 of them have some word 
element preceding them, and this element prevents them from from 
getting a prefixed wa(y). If the word order was reversed, the verbs 
would have preceded the mentioned word elements, and they would have 
been written as WAYYIQTOLs. These YIQTOLs are of the same roots as 
verbs realized as WAYYIQTOLS, and the context suggests that they have 
the same aspect as their WAYYIQTOL counterparts. It is high time that 
the fairy tales that the YIQTOLs with pase reference represent 
"durative past" and have a different aspect compared with the 
WAYYIQTOLs be abandoned!

So instead of being a linguistic back-packer, I suggest that you 
study the smallest units of language, because this can establish * 
verb meaning*; discourse analysis can only establish stylistics and 
function, not meaning.



Rolf Furuli

University of Oslo

>But Rolf, if Niccacci's method is "descriptive" and so wanting, what is
>different about any other method including yours? As far as I can see,
>by your argument there is no method by which we can find out how many
>conjugations there are or anything about the meaning of Hebrew. So
>should we all give up and go home? Or how can we find a method which is
>not circular (and which is not based, as yours is, on subjective
>judgments of which verbs are past, present or future) which can actually
>tell us something about the semantics of Hebrew?
>Peter Kirk
>>  -----Original Message-----
>>  From: Rolf Furuli [mailto:furuli at online.no]
>>  Sent: 15 April 2002 08:10
>>  To: Biblical Hebrew
>>  Subject: RE: discourse analysis and circular reasoning
>>  Dear Peter,
>>  I have never said that Niccacci's or anybody else's method is
>>  "suspect". Niccacci has done a great job by describing discourse
>  > patterns in detail. My principal point, however, has been that
>>  discourse analysis is *descriptive*, and  the method can never be
>>  used to show how many conjugations there are in Hebrew, let alone
>>  their meaning. That  this is possible is implied by Niccacci's
>>  "Syntax..", and in this area I criticise him. My second point has
>>  been that the method entails a measure of circularity, and this is
>>  admitted by the linguist on whose work Niccacci builds, namely,
>>  Harald Weinrich. In his work "Tense and Time" Archivum Linguisticum 1
>>  (new Series), p 41, he admits that his method is "unassailable",i.e.
>>  it cannot be tested by other means.
>>  Therefore, if you put it five conjugations, the output is five
>>  conjugations. My advise is to use discourse analysis in the ares
>>  where it belongs, namely, as describing patterns. And do not pretend
>>  that we by the help of this method can learn anything about the
>>  number of conjugations or the meaning of each conjugation. This is
>>  reserved for a study of the smallest parts of language
>  >
>  >
>  >
>  >
>  > Regards
>  >
>  > Rolf
>  >
>  >
>  > Rolf Furuli
>  >
>  > University of Oslo
>  >
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ibiblio.org/pipermail/b-hebrew/attachments/20020415/dcb17fb0/attachment.html 

More information about the b-hebrew mailing list