CHAPTER 8 Organising the message: the textual metafunction - cohesion

ANSWERS

1 Identifying cohesion in text

Cohesive reference items are underlined; conjunctive Adjuncts are in italics.
Remember that reference within a sentence is not strictly speaking cohesive and is therefore
not included in the analysis – e.g. ‘that text’ in sentence 1 of the first extract refers back to
‘the article’ in the same sentence.

from the dissertation

In the analysis below, I have used invented labels to differentiate between two types of
anaphoric reference: ‘anaphoric 1’ means that the reference is to something that has been last
mentioned earlier (quite often much earlier) in the dissertation (we can think of this as ‘long-
range’ anaphora); ‘anaphoric 2’ means that the referent is in the preceding sentence or the
sentence before that (normal ‘short-range’ anaphora).

anaphoric 1 [1] The results showed that the non-native speakers’ understanding of the
headline of the article from their own country was extremely high and that
they could easily predict the content of the article, which suggests that they
already had the background knowledge on politics and that they would have
certain expectations from that text.
conjunctive Adjunct [2] On the other hand,
anaphoric 1 the British subjects had great difficulty
anaphoric 2 in guessing its content judging only from the title as they could only talk
generally about it, being unable to make a successful prediction.
anaphoric 2 [3] The non-native speaker subjects encountered more difficulties
anaphoric 1 with the article from their country on a religious scandal, maybe because this
scandal happened during their absence from home and they were not really
informed.
conjunctive Adjunct [4] On the other hand,
anaphoric 2 they had few problems with the political article because this was mainly
based on their general knowledge about politics rather than on the specific
event that was being described.
anaphoric 2 [5] When they came to read the British article, the non-native speaker
subjects faced many difficulties as many of them did not know basic
concepts such as what is ‘Labour’ and which are the most important parties
 in the United Kingdom.
esphoric [6] Of course, the subjects who had spent more years
homophoric in the United Kingdom
anaphoric 2 were able to predict the article’s content more successfully
anaphoric 1 but at the time they could not reach a total understanding of what it was
going to talk about, because their knowledge of the political situation in
Britain was still restricted.
comparative [7] What is more surprising
anaphoric 2 is that the British subjects did not have a high understanding of the title
because, like the non-native speaker subjects,
anaphoric 1 they did not mention the reason that would cause Labour to lose safe seats.
anaphoric 2 [8] Of course this is not an indication that they do not know
esphoric the way elections work in the UK,
anaphoric 2 but maybe they did not consider it was worth mentioning.

from the examiner’s comments

In this analysis, the invented label ‘anaphoric 3’ is used when the writer is referring to
something in the dissertation text that is being assessed.

anaphoric 3 [1] In the Relevant Research Background chapter, there could be a little less
weight on general factors which affect reading comprehension, and a little
more on cultural factors specifically, but you show familiarity with, and
understanding of, a good range of concepts, and there is a clear sense of
development through most of the chapter, with different authors brought in
in a way which suggests that the ideas have been assimilated.
anaphoric 3 [2] The methodology chapter is successful: you worked out a fairly complex
esphoric procedure for setting up the various stages
anaphoric 3 of the data collection and analysis, and you not only carry it through
skilfully but explain it carefully and clearly.
esphoric [3] Although the fact
anaphoric 3 that your non-native speaker subjects had such different amounts and types
of experience of British culture makes it a little harder to compare the
results, the inclusion of the English native speaker subjects is sensible and
provides some unexpected side-lights on the topic.
anaphoric 3 [4] The results are presented in a clear way,
esphoric and you use the concepts
 anaphoric 1 introduced in the Research Background reasonably systematically.
anaphoric 3 [5] The discussion of the results is thorough and sensible though it mostly
stays at a relatively simple level: you achieve some perceptive interpretation,
but in places the analysis could be pushed deeper
esphoric – for example, I would have liked some discussion of the process by which,
in a few cases,
anaphoric 3 the readers managed to work out
esphoric the meanings of initially unknown expressions by the end
anaphoric 3 of the reading.
anaphoric 3 [6] It is useful to look at three case studies to illustrate the general findings;
but the relatively low scores of the NS readers could have been exploited a
bit more
esphoric since they reinforce the importance of background knowledge rather than
linguistic proficiency.
anaphoric 3 [7] The conclusions answer your research questions explicitly and
appropriately, but, on the other hand, the outline of teaching implications is
rather thin and generalised.

2 Discussing cohesion in text

Both texts use a restricted range of cohesive signals, with anaphoric reference by far the most frequent
device, especially through the use of the marker the. As I pointed out in Chapter 8, ellipsis is strongly
associated with spoken discourse, so it is unsurprising that it does not occur in these written texts. The
absence of conjunctive Adjuncts in the examiner’s comments is perhaps more surprising at first sight,
and I will comment on that below.
As I have indicated with my invented labels, the anaphora can be seen as falling into three
groups. The first is ‘normal’ anaphora in which the referent is in an immediately preceding sentence.
The second, longer-range anaphora, relates to cases where the referent is something that can be taken
to be ‘on the table’ more or less throughout the thesis, although it may not have been mentioned
recently in the text. The third is where the referent is in the other text. This latter type could be
regarded as exophoric, in that the other text is part of the situation in which the examiner is writing his
comments; but, since the reference is to a previous language event, it seems better to code it as a
special case of anaphora. In some ways, it is similar to one speaker referring back to something
mentioned by the previous speaker.
It is noticeable that the first two types are found only in the extract from the dissertation. This
reflects the fact that the dissertation is, as it were, self-contained: it is expected that all the information
needed by the reader to make sense of the text will be in the text. The longer-range anaphora occurs
because of the constant focus on the same topic throughout the dissertation: the writer assumes that
the reader will have in mind the basic design of the research project (with particular texts,
experimental subjects, etc.) and will understand that mention of entities such as ‘the headlines’ are to
be interpreted as referring to those which were introduced further back in the dissertation. The writer
does still have to decide at what point it is more reader-friendly to use a full nominal group with
definite reference, such as ‘the British subjects’, and when she can use a pronoun such as ‘they’. This
is highlighted particularly with references to the two different groups of experimental subjects or the
various types of texts used in the experiment. The general pattern is that when one of the groups of
subjects or text types is brought back into the text, a full nominal group with the is used; and then a
pronoun is used until the next switch. What constitutes a ‘switch’ may vary. Obviously, if the writer
changes from discussing, for example, the non-native speakers to the native speakers, that will form a
switch point (to simply continue using ‘they’ would be confusing). But note that, whereas in sentence
4, ‘they’ is used to refer back to the non-native speaker subjects mentioned in sentence 3, ‘the non-
native speaker subjects’ is used again in the main clause in sentence 5 even though the same group is
being discussed. This seems to be partly because there has been a relatively long stretch of text since
the full nominal group was last used, but perhaps mainly because there is a switch from discussing the
article from their own country to discussing the British articles. The fact that the writer chooses to
start a new paragraph is also implicated in the choice of a nominal group: these two choices work
together.
The dissertation extract includes two conjunctive Adjuncts. As we saw in the exercises for
Chapter 7, the intra-sentence conjunction focuses especially on causal relations; the conjunctive
Adjuncts, on the other hand, contribute to setting up the contrasts between the different groups of
experimental subjects (the Adjunct in sentence 2) or the different text types (the Adjunct in sentence
4). They therefore mark off larger chunks of the text, within which the causal relations operate. It is
worth noting that these larger shifts are not all marked by conjunctive adjuncts: in sentence 5, a
subordinate clause functioning as marked Theme performs this role, and in sentence 7, a thematic
equative with a comparative form marking contrast. This underlines the fact that conjunctive and
thematic choices can both, separately or in combination, guide the reader towards the organisation of
carefully planned discourse like this.
In the examiner’s comments, virtually all the cohesive reference – the only type of cohesive
marker which occurs – is of the third type. This construes the comments as keyed to the dissertation
text. This is precisely their main role: they are not intended to be self-contained in the same way as
the dissertation. They are primarily addressed to the writer of the dissertation, who is naturally
expected to be intimately familiar with that other text and to be in a position to identify the intended
referents. At the same time, the examiner is aware that the comments will probably be read by other
people, particularly other examiners; and, even if they do not read the dissertation, they can be
expected to recognize the standard sections of that kind of text (chapters on relevant research
background, methodology, findings, etc.) and the kinds of information that is likely to be found in
those sections (data collection, analysis, etc.). In that sense, there is an element of homophora in play.
I mentioned above that there are no conjunctive signals. The marking of stages in this text is
performed just by the Themes, which establish the section of the dissertation that is being commented
on (e.g. ‘In the Relevant Research Background chapter’, ‘The methodology chapter’). The balance
between praise and criticism that is shown by the analysis of clause complexing is realized within
each of the stages: the writer opts for a stage-by-stage organisation. The most likely alternative that he
could have chosen was a point-driven organisation: all the praise followed by all the criticism (or vice
versa). In that case, it is probable that there would have been a conjunctive Adjunct such as However
at the switch point in the text. The stage-by-stage organisation allows him, amongst other advantages,
to project himself as having read the whole dissertation thoroughly. The text is essentially a list of
good and bad points, with each stage largely separate from the others. There is therefore no reason to
mark any conjunctive relations between the stages.