Coherence and Cohesion in English Discourse

collaboratively in the development of the text. I use the concept of the 
‘reader-in-the-text’ (Thompson and Thetela 1995) to explore a central 
form of interactional resource: the inclusion in the text of a voice that 
is intended to be attributable to the reader. I identify a particular set 
of discourse contexts in which this happens – where the writer brings 
in the reader’s view in order to contradict it – and outline
 the lexico-
grammatical features which signal the other voice in those contexts; and 
I place
 these in a broader perspective on written text as a stage-managed 
form of dialogue. The impetus for the study comes from working with 
novice writers; and I discuss a number of examples where written drafts 
were improved by exploiting the interactional resources described, and 
argue for the value of raising students’ awareness of these resources. 
(AL/4 Abstract)
In co-authored texts the plural author-reference pronoun can be used 
similarly to the single author-reference pronoun. However, while the referent of 
singular author-reference pronouns is definite, the plural form 
we
is inherently 
ambiguous as it may be used both exclusively and inclusively. This is evidenced 
by an extract taken from the conclusions of a co-authored article (2) in which the 
authors use author-reference pronouns to stress their position of active research 
subjects committing themselves to a subjective interpretation of a phenomenon 
and to a discussion of strengths, limitations and future research perspectives. 
Apart from the inclusive personal pronoun 
we
used in the last sentence of the 
extract (
we now have a baseline
), the authors use the exclusive 
we
to refer 
anadeictically to themselves, thus assuming responsibility for the limitations of 
their interpretation of the phenomenon under investigation. In the last sentence, 
however, the referent of 
we
is extended to comprise the writer and the reader 
as members of a disciplinary community who are assumed to share the opinion 
expressed by the author, thus enhancing the persuasive force of the discourse. 
Although the mental representations of the writer and the reader are available in 
the minds of the participants, this inclusive use of 
we
is primarily contextually 
retrievable and allows for multiple interpretations.

---

18
(2)
There is of course a danger that we are, for the purposes of discussion and 
clarification of the genre, exaggerating the problem of comprehension. 
Indeed, we believe
 
that most authors are capable of understanding the 
intent of the editorial letters and the guidance that they provide about 
reading the reviews and revising the manuscript. We must affirm again 
that because our study
 
is of the letters of one single editor we cannot
 
make 
any generalizations concerning the editorial letter genre. Nevertheless 
we believe
that 
our study
 
may go some way towards demystifying the 
editorial process and contribute to our understanding of the nature and 
function of the genre in question. At least we now have a baseline
 against 
which further research can be measured, the next step, of course, being a 
study of a corpus of letters from a range of journals. 
(AL/5 Conclusions)
The function of the possessive determiner 
our
in (2) also deserves attention 
as it is dependent on the interpretation of the personal pronoun 
we
. Thus while 
the occurrences of 
our study
present the authors as the agents of the research 
process,
 
our understanding
is ambiguous and may be interpreted as referring to 
the authors, the authors and the readers or the whole disciplinary community. 
What is important from the point of view of cohesive relations, however, is that 
possessive determiners contribute to the availability of referents in discourse 
processing. As Table 1-2 indicates, possessive determiners are not very frequent 
in the corpus under investigation or in academic discourse in general (Biber et al. 
1999). The most frequent use of possessive determiners in academic texts is with 
deictic nouns (nouns that point to the text or textual segments, e.g. 
study
, 
paper
, 
article
) and shell nouns (abstract nouns that enclose or anticipate the meaning 
of the preceding or succeeding discourse, e.g. 
problem
, 
fact
, 
understanding
), 
although some lexical items (e.g. 
research
, 
study
) may function both as deictic 
and 
shell nouns (cf. Hunston and Francis 1999, Aktas and Cortes 2008, Gray and 
Cortes 2011). There are two cohesive relations established in such cases – one is 
based on the reference of the possessive determiner, and the other is established 
by the deictic or shell noun which points to a part/section of the text. It should be 
mentioned that while similarly to author-reference pronouns deictic nouns have 
the potential to create global and local cohesive chains, the scope of shell nouns 
is typically restricted to local cohesion.