Relations that are seldom or never signaled

[Bill Mann]

Dear RSTlist subscribers:

One of the aspects of RST that has received relatively little commentary or
explanation concerns unsignaled relations in text.  It has been found, over and
over, in hundreds of cases now, that the number of explicit signals of
relations in a text is usually very much lower than the number of links in an
RST analysis of that text.

One of the sources of this is an important set of relations that almost never
are signaled.  I have put together a list of these based on my impressions,
without trying to do any counting at the moment.

Using the list of relations that appears on the tool page on the web site, here
are 12 of those 28 that seem to be very seldom indicated explicitly:

Background
Elaboration
Enablement
Evaluation
Evidence
Interpretation
Justify
List
Motivation
Preparation
Restatement
Solutionhood


I wonder whether this agrees with the impressions of other folks who have used
RST.  Please let us know about your experience.

I wonder also whether there are languages that signal some of these relations
frequently or predominantly.

There are issues about how people recognize these relations.  They include some
of the very consequential relationships in text, such as Background,
Elaboration, Evidence, and Solutionhood.

We will have a chance to discuss each of these questions.

Best wishes,

Bill Mann


The web site is at http://www.sil.org/linguistics/RST