Coherence and Cohesion in English Discourse

1 Introduction
Even though English performs the role of a lingua franca of international 
academic communication, many studies on written discourse used in academic 
settings show cross-cultural variation (e.g. Clyne 1987, Ventola and Mauranen 1991, 
Mauranen 1993, Čmejrková and Daneš 1997, Duszak 1997, Chamonikolasová 
2005, Stašková 2005, Mur-Dueňas 2008, Wagner 2011, Dontcheva-Navratilova 
2012, Povolná 2012). This variation in the global lingua franca of academia, which 
concerns all text characteristics including form and content, results mainly from 
the influence of L1 writing habits and culture- and language-specific conventions 
which authors working in different fields of research transfer from their mother 
tongue to the academic texts they write in English.

---

30
Anglo-American academic texts in general tend to be more dialogic and 
interactive, thus providing more space for negotiation of meaning between 
the author(s) and the prospective reader(s) (Clyne 1987) and bearing a greater 
resemblance to non-academic texts. The fact that these texts are considered more 
reader-oriented stems from an overall linear organization of discourse through 
explicit signposting which includes text organizers such as DMs – notably 
conjuncts, which are at the core of the present chapter. These characteristics are 
in contrast to rather monologic, less interactive texts, which sometimes include 
numerous digressions and provide readers with knowledge and theory rather than 
giving them space for negotiation of meaning. Such writer- or text-oriented texts 
are usually connected with Teutonic intellectual traditions attributed to academic 
texts written in some Central European languages such as Czech, Slovak, Polish 
and German (cf. Galtung 1985). These academic backgrounds and intellectual 
traditions prefer a more impersonal style of writing with fewer reader-friendly 
devices such as text organizers and fewer explicit clues concerning content, which 
can provide reader guidance and discourse predictability. Instead, a considerable 
amount of intellectual effort and an ability to process rather demanding texts 
filled with knowledge and theory are required of the reader(s).
Since the overwhelming majority of writers and readers of academic 
texts written in English are not native speakers of English, the question arises 
whether it is justified to impose the linguistic standards and style conventions 
typical of the Anglo-American discourse community on international academic 
communication and whether the qualities such as clarity and effectiveness in 
communication should be considered from the perspective of native speakers 
of English, i.e. “the native speaking minority” to use Mauranen et al.’s words 
(2010), or from those who come from communities that speak other languages.
Discourse communities share certain discourse patterns and expectations and 
utilize and hence possess one or more genres in the communicative furtherance 
of their aims (Swales 1990: 26). In the case of the international academic 
community it is the genre of research articles that mostly serves as a tool for 
transmitting scholarly matters. When experts from language backgrounds other 
than the Anglo-American want to be recognized internationally in their respective 
research fields within their native speech discourse communities and in particular 
within the international academic community, they have to produce their research 
articles in English and undergo what is sometimes called “a process of secondary 
socialization” (Duszak 1997), i.e. the process of developing academic credentials 
within their non-native environment. This concerns both novice writers and 
experts from academic traditions other than the Anglo-American (Dontcheva-
Navratilova 2012, Povolná 2012).
Scholarly texts vary in degree of interactivity, understood here in agreement 
with Duszak (1997: 19) as a “form of realization of interpersonal meanings 

---

31
in discourse”; this is particularly important once we agree that academic 
communication is dialogic in nature. While academic cultures and intellectual 
traditions subscribing to a more impersonal style of academic texts (e.g. some 
discourse communities in Central Europe) give preference to less interactive 
and thus less dialogic texts, Anglo-American academic traditions favour more 
interactive and more dialogic texts with reader-friendly devices such as text 
organizers, clear relations between discourse segments, clear division of text 
into sections and subsections, chapter and section headings, and explicit clues 
on content. Therefore, the negotiation of preferred degrees of interactivity and 
dialogicality in academic texts is one of the main issues to be considered in 
international academic communication.
Conceived as explicit signals of semantic relations between segments of 
discourse (Fraser 1999) and thus clearly contributing to both cohesion and 
coherence, DMs (Schiffrin 1987, Fraser 1999, Biber et al. 1999) – including 
what are called conjuncts (Quirk et al. 1985) in this chapter – are expected to be 
relatively frequent in academic discourse, since convincing argumentation and 
clear presentation to the reader of the author’s standpoints are of great importance. 
Consequently, a more specific aim of the present study is to discover which 
semantic relations, such as apposition, result, contrast, and concession, tend to be 
expressed overtly by conjuncts, because these are applied intentionally by writers 
as guiding signals to help the prospective reader(s) arrive at an interpretation 
which is coherent with the author’s communicative intentions, and, moreover, 
which of the semantic relations overtly expressed by conjuncts contribute most 
of all to the negotiation of meaning between discourse participants (i.e. the 
author and the readers), thus enhancing the higher degree of interactivity and 
dialogicality in academic texts.