Re: Discourse Spans (hierarchy)

Date: Tue Mar 09 1999 - 15:04:59 EST


In a message dated 3/8/99 10:59:40 PM Mountain Standard Time, writes:

> I have no problems with the statements you make here. I grouped
> informational and teleological together because they are both semantic
> and not "formal" structures. It is precisely because I have had nothing
> but trouble doing analysis starting out with "formal" patterns that in
> desperation to make something actually work I have decided to give
> semantic patterns a try. Semantic patterns, once detected are often
> reinforced by formal patterns. But using formal patterns to find the
> semantic patterns just got me nowhere at all. One of the reasons is
> there are too many kinds of patterns to look for and they are often used
> in really complex ways that make them difficult to spot.

In fact, my position is that semantic patterns and formal patterns are not two
different strands running through the text, but that meaning is realised in
the intersentential structure, grammar and lexis, and that there is no divorce
between form and content. So an approach to structure that results in a
serious tension between the two should be questioned. Now, I'm not saying that
starting with a content/topical/thematic analysis will necessarily end up in a
tension with form. But when it does, you've got it wrong (not you personally,
of course).

But I quite agree that there are many complex patterns that are used in many
complex ways. It's not like chapter headings, subheadings and indentation.
However, we are dealing with some of the same kind of criterion when we
attempt to analyze oral speech patterns and organization. We generally can
follow speech and even in unplanned informal speech we know a number of
complicated ways to organize material, to signal a shift, or to emhasize a
main point.

> > The grammatical 'spans' are extremely useful in detecting how the author
> > breaks up (or shifts) and presents the material--it can function like
> > paragraph indentation. And what is good about the grammatical features is
> that
> > they are formal indicators
> To this I would say it depends on which author and which genre. I am not
> quite convinced by the actual examples I have seen that grammatical
> 'spans' are a universal method used for structuring a text. In actual
> exegetical practice, what do you look for? You don't know in advance
> what sort of pattern to look for so you end up looking for some sort of
> pattern. Looking for some sort of pattern can be a frustratingly vague
> process.

Yes, there can be quite a distinction between genres.

 Well, like I said a while ago when we were talking about sources, I don't
think that Young's discussion is the last or best word in discourse analysis,
but I appreciate the attempt. Jeffrey Reed's description in <<A Discourse
Analysis of Philippians>> is quite good. However, I know you've read it, and
I would imagine that you are frustrated by the emphasis on theory, and the
lack of pragmatic methodology that you can really apply.

Instead of looking at it as just span, think of complementary patterns of
repetition and variation. And don't expect to see the same patterns repeat
through the entire discourse. Different sections are often characterized by
different patterns. To use an example, I think that the use of rhetorical
questions, and the variation of indicative and modals (imperatives,
prohibitions, etc) are used quite often to 'chunk' material, even in
narrative. Every once in awhile, you hit on some signal that pervades an
entire discourse, but I think it's the exception.

> > .
> > Once I saw how some analysts overrode grammatical 'spans' on the basis of
> > their intuitive notion of topic, I realised how important it was to
> respect
> > them, and even use them as a starting point of anaysis (I'll give one
> > example--it is quite common to place a break between Hebrews 1:3 and
> > 1:4--which breaks up a periodic sentence that the author unified
> > structurally).
> >
> How "intuitive" are topic or information or purpose? Once you can read
> the language reasonably well, so you can sit down and go through large
> sections of a text over and over, then the semantic structures based on
> the categories like topic, information and purpose are not just wild
> guesses, if that is what you mean by "intuitive". The semantic
> structures are intuitive in the sense that your mind is able to process
> the text in ways that defy your analysis to nail down. But that does not
> mean that the results of this process are not real or of no value. This
> is getting into the whole epistemological side of this question which I
> alluded to in my private post. This is not an epistemology list so I
> will drop it.

Not at all, but my personal goal is to try to determine the intended message
of a given discourse and how the original readers would have understood and
interpreted it. And if my mind interprets a text in some way that defies my
analysis to nail down in terms of form, grammar or lexis (and I should throw
in context to cover my bases), I look at either my methodology and/or my
conclusions with some suspicion. If you are going to make a case, what counts
for evidence?

But hey, if you resolve your falling out with Chomsky, you can just say that
your interpretation resides in the deep structure! (joke)

Now I'm going to attempt to retire to lurker status, even though I know I've
raised more questions (and eyebrows) than I've answered!

Cindy Westfall
PhD Student, Roehampton

B-Greek home page:
You are currently subscribed to b-greek as: []
To unsubscribe, forward this message to
To subscribe, send a message to

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:40:19 EDT