A problem with an analysis in terms of RST

[Holger Schauer]

>>>>"GR" == Gisela Redeker schrieb am Wed, 20 Oct 1999 20:32:43 +0200:

 GR> Considering the genre (news article), I'd guess 1-4 is the
 GR> evaluative (or commentary) end of the last (or only) "take" of
 GR> the article, followed by what Van Dijk might call an
 GR> "expectation" segment (probably the final paragraph of the
 GR> article).

Some more information on the excerpt: the article [1] is a commentary
on the matter of compensation. It is entitled "The german disgrace in
forced labourers compensation". Units (1)-(4) actually constituate the
third sentence of the article, where sentence 1 is about "which offer
causes shame for who is a matter of cultural shame" (Welches Angebot
wem die Schamröte ins Gesicht treibt, unterliegt bekanntlich der
jeweiligen Schamkultur.) Sorry for the crude translation, I am not a
native speaker of English. Sentence 2 is "In this regard, Wolfgang
Gibowski, the speaker of the industry fond for compensation of forced
labourers in the third reich, obviously seems to <idom: have a skin
made of leather = be very unsensitive." What I perhaps should have
indicated more clearly is that there is a paragraph break between my
units (1-4) and the units (5-7). (5-7) is a paragraph for itself,
continued with another paragraph which comments on the negotiations.
So, (5-7) is not further elaborated.

 GR> If that is true, we should not expect a strong rhetorical
 GR> relation between these two paragraphs. As I have argued elsewhere
 GR> (Redeker 1996/forthcoming), RST lacks special 'machinery' to
 GR> handle genre-specific structures (where sequential relations
 GR> between text segments are defined in terms of the structural
 GR> elements expected, and not in terms of content or argumentation),

I would be very interested in these "elsewheres" :-)

 GR> and thus has to go with the nearest semantic or pragmatic
 GR> relation-- in this case I'd say "SEQUENCE"--with the
 GR> understanding that I assume 1-4 is the EVALUATION satellite
 GR> within a larger segment, to which 5-7 can then form the 'next' in
 GR> the sequence.

I don't think (1-4) evaluates the previous context. I do think,
however, that both (1-4) and (5-7) evaluate the "negotiations" which
is obviously the main topic of the text. U.Wassner send a comment to
me in private e-mail, in which he makes (after having seen the whole
article) a similar point: he basically states that there is only a
(text-) structural (hierachical) relation between (1-4) and (5-7) and
the only "relatedness" depends on the main topic of the text.
Similarily, he claims that coherence is mainly a matter of paragraphs.

I think that we basically all agree: what I had in mind when proposing 
the "Topic Shift" relation is essentially this: there is not an
informational rich relation between (1-4) and (5-7), the main topic
of (1-4) (or sentence 1-3, if you like), namely the "shame" one should 
feel, shifts to a different topic in (5-7). It is just the main topic
of the text that is still elaborated on.

 GR> This analysis would seem particularly appropriate if the earlier
 GR> global strategy was also sequential (chronological report). If
 GR> this is not the case, other possibilities might offer themselves,
 GR> such as INTERPRETATION (with 5-7 as the satellite introducing the
 GR> wider/new perspective) or a simple JOINT if the article is built
 GR> up like a series of (not chronologically ordered)
 GR> observations+commentary.

The latter option is in my opinion more appropriate.

 GR> The lattter is the option Holger
 GR> discusses in his original mail:

 >> Okay, now here's the problem: decide to which unit (5-6) relates,
 >> and with which relation. I tend to say (1-4) and "Joint" (or in
 >> D. Marcus terms "Topic Shift"), although one could argue that in
 >> (5) the "result" from (2) is ELABORAT-ed ("finally"). This however
 >> would imply that (5-7) are sub-ordinate to "feeling shame" (1) and
 >> I am not sure about this reading (i.e. I am not sure whether the
 >> author wanted to convey it, my own feelings are not an issue
 >> here).

What I still don't understand though is why is this latter option
ruled out ? I cannot gain further insight by introspection, so I am 
interested in the introspection of others. :-)

[guideline, to which I agree, snipped]

 >> Apart from this, I would say, 1 is related to 2-4 via a
 >> CIRCUMSTANCE relation, so the conditional in (5-6) would also be a
 >> part of this CIRCUMSTANCE which does not look too convining to me.

 GR> This is VERY difficult to analyze without further context. It
 GR> sounds a bit like the article has just reported on someone being
 GR> upset and ashamed about the result. In any case, 1-4 could not be
 GR> the first units of the text, and 1-4 as a whole might be a
 GR> RESTATEMENT or, if the details in 2-4 have not been stated
 GR> before, an INTERPRETATION of the preceding cluster. Given some
 GR> such global analysis, I'd analyze 1 as an EVALUATION satellite at
 GR> 2-4, which in turn consists of a CONTRAST relation (3-4) that
 GR> forms an ELABORATION satelite to 2.

This is very interesting, as my analysis is different. First, I do
think that (1-4) could be the first sentence of a text - but of course 
only, if the broader context is available. I think that in this
article the author just made the assumption that the context is indeed 
clear to his readers. I just checked the issue, there is no other
article on the topic in this issue, but I guess that there are more
articles on the topic in previous issues.

Second, I would say that (1-4) are connected to the previous sentence
by a very weak relation as well, in this case I would argue for
"TopicDrift" in Marcus terms, as both sentences are concerned with
"shame" but the behaviour of the speaker (see above) is not further
commented on in (1-4).

Third, I do agree that in some respect (1) seems to be an
INTERPRETATION or perhaps a COMMENT on (2-4). However, I would like to
explain why I choose CIRCUMSTANCE: I did this because of the cue
phrase "in face of" (german: "angesichts"). I think (2-4) constitute
the context on which to interpret the suggestive, rhetorical question
in (1). Of course, an INTERPRETATION or EVALUATION without the context
what is interpreted or evaluated doesn't make to much sense,
either. But, is (1) really an evaluation of the situation in (2-4)?
The speaker does not say: "It's a shame that (2-4)" but instead he
describes a feeling he has and which he more or less claims anybody
should have. I think that this makes a difference. In so far I
believe that EVALUATION would be an over-interpretation.

Again, I would like to know why you have come to the conclusion that
(1) is an EVALUATION/INTERPRETATION on (2-4). 

Fourth, I have a similar question with your analysis of CONTRAST
between (3) and (4). I have analysed this in terms of the weaker
"TEMPORAL-AFTER", and again mainly based on the linguistic cues,
i.e. "begin" (3), and "end soon" (4). What I have translated as "only" 
in (4) might be understood as a cue-phrase for contrast in English, so 
looking again at the analysis I see that I miss this. However, the
german construction has the conjunction "um" at that place and this is 
not necessarily a cue-phrase for contrast, e.g.,

"Er stand frueh auf, um sich dann wieder hinzulegen."
"He got up early, just to go to bed again."

Thinking it over, I would say that (4) might stand in a CONCESSION
relation to (3), i.e. after starting with a nice gesture (the
installation of the industrial fond [2], one wouldn't expect that the
negotioations would go on like (4). But is the focus of the author on
such a reading ? I am not sure.

The problem that I see with all analyses is that one looses some
information, e.g. when I just analyse in terms of temporality I loose
the contrast/concession and vice versa.

 GR> As for 5-7, I'd analyze 5 as a CIRCUMSTANCE satelite
 GR> left-attached to 6, with 7 right-attached to the combination 5-6
 GR> as an INTERPRETATION satelite (introducing a new perspective).

I find the CIRCUMSTANCE plausible but would say that 7 gives an
EXPLANATION-ARGUMENTATIVE (Marcu) or BACKGROUND (Original-RST) to 5-6, 
i.e. 7 says why "it is sink or swim".

 GR> But maybe Holger or some other list readers will find these
 GR> useful to mull over and criticize...

Yes, thank you and also Luuk Lagerwerf and U.Wassner. I find this
discussion very useful.

I hope that there will be some more feedback.

Holger


Footnotes: 
[1]  Available on-line via http://www.taz.de/tpl/1999/10/07.fr/
     tbox?Ueber=&Tname=a0003&idx=0&re=a1&qu=TAZ&mon=

[2]  This is a relative clause to (3), which I did not include in the
excerpt as I thought I would not be relevant.

-- 
---          http://www.coling.uni-freiburg.de/~schauer/            ---
"Kann mir jemand kurz das Anliegen der newsgroup erklären?"
"Wahrheitssuche, eingeschränkt auf das Thema Fahrrad."
                  -- C. Karl und Bernd Sluka in de.rec.fahrrad