“Alexandru Ioan Cuza” University of Iaşi (Romania)
Communication, Knowledge and Discourse. Remarks on Teun Van Dijk’s Sociocognitive Interpretation of Context
Abstract: My paper focuses on Teun Van Dijk’s approach of the concept of context. The novelty of this approach consists in the argumentative effort to overcome the difficulties brought by the behaviourist fallacy present in the theories regarding context.
Linguistics, conversational analysis, rhetoric and communication studies underline – each in a very specific way – the importance of context when it comes to coding, decoding, interpretation and understanding of a discursive element. In the same time, these disciplines have not done enough to clarify the role of the mental entities in creating and maintaining the discursive context. Without the use and the proper comprehension of the mentalist vocabulary (intention, action, goal, awareness, meaning coordination), the explanations that we offer to common communication situations seem inappropriate.
Keywords: context, mental models, discourse, communication, sociocognitive approach
1. In need of a theory
The importance of context in explaining and interpreting everyday situations of communication is not a matter of proof anymore, but of refined clarification. In Van Dijk’s (2007, 283) terms, “context obviously does matter. The question thus is not whether context should be a part of a sound theory of discourse and conversation, but rather how such contextual influence should be accounted for, and how context should be analyzed”. A part of the problem is the fact that the concept of context is not defined in a good, convenient manner; as Teun Van Dijk puts it (2007, 284), “in the social sciences, there are thousands of books that bear the notion of ‘context’ in their titles, but there is not a single monograph that theoretically deals with the notion of context, although there are some collections of articles [...]. This means that the notion is mostly used in a rather intuitive, pre-theoretical sense, namely as some kind of social,
cultural or political environment for an event or action – and more often as a condition than as a consequence of such an event”.
The traditional linguistic view of context was focused mainly on the verbal aspects; moreover, terms like “understanding”, “interpretation” or
“intention” were used freely, without any deep cognitive investigation. In other connected fields of inquiry (study of literature, discourse analysis, communication studies or conversation analysis), Van Dijk claims, the notion of context was used rather as an equivalent for “social situation”.
So, up until now, scholars have seldom mixed linguistic and social cues in their use of the concept of context. Sometimes, further distinctions brought more light on the problem, as it is the case with the sociological categories local/global and micro/macro. These four terms have been successfully applied to context; but we are still short of a “theory of context and its relations to talk and text” (Van Dijk 2007, 288).
This means that we do not have yet answers to questions such as:
how much context do we need in order to communicate correctly? How much do we have to know about context so that we might be able to rise up to the communicative expectations? What are the contextual parts that really have an influence on the conversational interactions? Last, but not least, is there a direct conditional relationship between context and the communicative performance?
Moreover, a theory of context should incorporate a consistent part of the knowledge previously acquired (see, for instance, Van Dijk 1977 for the problem of semantic and pragmatic dimensions of discourse, and Sperber and Wilson, 1990, also Sperber and Wilson, 1993, and Sperber, 1995, for the problem of relevance in communication).
2. Context and mental models
Van Dijk’s development of a sociocognitive theory of context includes a wide range of works, which started probably as early as 1989 (Van Dijk 1989) and continued for about two decades (see Van Dijk 1997, 2001, 2003, 2005, 2006, 2007). I shall focus primarily on Van Dijk’s “Discourse, context and cognition” (2006) and “Comments on context and conversation” (2007).
First of all, the sociocognitive approach is, in Van Dijk’s view, based on the following assumption: “There is no direct causal or conditional relationship between social characteristics (gender, class, age, roles, group membership, etc.) of participants and the way they talk or write. Rather, it is the way participants as speakers (writers) and recipients subjectively
understand, interpret, construct or represent these social characteristics of social situations that influences their production or understanding of their talk or text”. This way, the sociocognitive paradigm reaches two goals in the same time: it gives the mentalist vocabulary the right place in the explanation and also helps us to understand why we have difficulties when we try to talk about context without referring to the (social) constructs that become active during a communicative situation. I personally think that we should add the concept of communicative investment to the above list. Surely, there is an important part of what we call context that is objectively present (time, place, physical conditions, environment etc.) – of course, it may be argued that many times we can decide even on these aspects of context (we choose to talk outside, for instance). But, on the other hand, only by interpretation and construction (investment) a certain context is created for communication. This is seen also in the distinction between speaker’s meaning and sentence meaning.
A mental model is defined as a mental representation of an object in episodic memory (Van Dijk 2007, 290). When people participate in a conversation or discourse (Van Dijk 2006), they create dynamic mental models that are built to fit both the conversation itself and the social situation. Creating such a mental model helps people understand what is happening, take turns in speaking and listening or opt for actions.
Practically, they construct two types of mental models, namely a semantic one (concerned with the meaning of the conversation) and a pragmatic one (related to the relevant issues of the conversation).
Some of van Dijk’s claims need further examination. For instance,
“The relativity of context”, Van Dijk says, is linked to the known fact that something may be relevant to me (at a given moment in time, say), but not relevant for someone else. This thesis goes hand in hand with “The subjectivity of the context”, according to which “contexts are not objectively out there, but the result of personal, subjective understandings of the communicative situation. That is, contexts are personal constructs”
(Van Dijk 2007, 291). The use of the term “construct” is also interesting and worth exploring, but I will not enter its details now. Although Van Dijk’s observations seem intuitive enough, we have to point out that we have plenty of communicative situations in which the relative elements are blended with “absolute” ones, while subjective traits are mingled with objective traits. Let's think about an oral exam, for instance. Let us presume we have 5 students and 2 teachers in a classroom talking about different matters of the subject “Public Debates”. These seven individuals – within normal conditions – have reasonably predictable behaviours: each
student will have to make a brief presentation of her or his theme, teachers comment and ask questions, questions are answered (or not), the student gets a grade and so on and so forth. Of course, such a “normal”
situation does not exclude students singing or protesting instead of talking, or teachers behaving erratically. Why can we predict what happens? If things were totally relative, our degree of uncertainty would have been higher. But such examples of institutional facts show how people are drawn towards compatible behaviours. The explanation, in my view, is found within the sociocognitive approach; to rephrase van Dijk’s words, people invest in certain elements and situations, creating constructs that finally lead to inter-subjective communication, the closest thing we have for objective elements in communication. Thus, the student's grade (A, say) exists (both as a physical entity and an institutional entity) because a bunch of people choose to believe in the concept of university studies (and its subsequent issues) and provide intentional content accordingly. This intentional coordination finally makes that A “objective”. Van Dijk also agrees, saying that
“Mental models of contexts are subjective, but not arbitrary. After experiencing and participating in many thousands of unique communicative situations, language users tend to generalize and normalize such situations, so that also their mental models of such situations are generalized to shared, social representations of such situations. Such social representations will abstract from ad hoc, personal and other specific aspects of communicative situations, and hence reduce the subjectivity of each context model. It is in this fundamental way that (this aspect of) the social order is reproduced, how the rules of conversations and other interactions are being acquired, and how context models may be coordinated by different participants” (Van Dijk 2007, 293).
The thesis of “partiality of context” is easily acceptable, the notion of flexibility being one of its most interesting traits. People are free to ignore big parts of the social situation and also a big part of the linguistic features. They select only what is important for them, they make a communicative investment that fits the relevance stance: they only include the aspects that seem salient enough. The cognitive dimension of the context models provides “the missing link between social situation and discourse”, and thus we avoid two traps: on the one hand, we avoid the danger of determinism (there is no direct influence of the social situation on discourse); on the other hand, we get rid of the idea of an exclusive account of context.
3. Knowledge, discourse and mental models
In Van Dijk’s terms, “to exclude mental models from our theories and analyses of conversation because they are not ‘observable’, and to exclusively admit only what is ‘demonstrably displayed’ in talk, is in my opinion based on a behaviourist (or maybe better an interactionist) fallacy” Van Dijk 2007, 299). This fallacy is not only dangerous, but also unfruitful. The context models are evidently related to a vast area of shared social representations (attitudes, values, gender culture, power issues, ideological issues, community issues etc.). The communicative events in our lives are influenced by such representations, but they also produce changes in our representations; many times, changes emerge after a long time of conflict between diverse representations. The amount of knowledge that an individual has on a specific subject makes him evaluate and interpret that subject. His or her inferences construct the basis of the strategic approach that finally becomes language in action.
We must note that people make up discourses in all the possible situations, from zero or little information to the situation of total cognitive control. The scarcity of information made very few people abstain from talking, by the way. Also, both general and particular elements of knowledge are present in the construction and performance of discourse.
For Van Dijk (2007, 293), “(‘pragmatic’) context models control interaction, speech acts, style, rhetoric, and all other aspects of discourse that make discourse appropriate in the current situation, these ‘semantic’
models and social representations control the meaningfulness (both in production and understanding) of the debate” (for the problem of mindfulness in communication, see Burgoon, Berger and Waldron 2000).
The presence of discursive schemata has been already discussed in the scientific literature of the rhetorical domain (Sălăvăstru 1996, Sălăvăstru 2006), but the sociocognitive approach shows how these schemata are related to the mental and conduct schemata. The use of schemata creates different routines, and the routines, at their turn, lead to the formation of discursive categories and genres. The use of schemata and routines does not entail the fact that we are in danger of „losing” the particular aspect of a communicative event. On the contrary, every mental model is unique, and the process of understanding and of problem solving involves
“socially shared minds” (Van Dijk), mental interactivity, coincidence of the passing theories (Davidson 1996, also see Grădinaru 2011) and context management (Van Dijk 2007, 296):
“At each point in talk a speaker needs to 'calculate' what the recipients already know, and take this meta-knowledge into account in the production of the next discourse fragment. In other words, talk is not only socially interactive, but also mentally so. Moreover, much of the kinds of knowledge involved here are not at all personal or subjective, but socially shared in a group, institution, nation or culture, and hence should be part of the social theory of interaction as well: they are inherent part of the social order and its reproduction through discourse”.
4. Conclusions
According to the sociocognitive perspective, context models are strategic tools that we use in our daily conversations. In order to function properly, they require a minimum of knowledge from the participants;
they can be incomplete or misguided (thus possibly leading to conflict).
This theory of mental models is consistent with the concept of relevance, and is also providing a cognitive basis for communicative events and interactions: “all relevant properties of interaction, such as mutual awareness of participants, mutual knowledge, coordination, and so on, are unthinkable without such a cognitive dimension” (Van Dijk 2007, 298).
Even though these mental models of context are not directly observable, both our thinking process and our interactions with other people strongly argue in their favour. Creating a bridge between social structures and communicative events, the context models prove necessary. Sadly enough, “models are always much more detailed than the discourses based on them” (Van Dijk 2006, 170), and this reminds us never to forget to repeat our reading, in the perpetual search for hidden beauty.
References
BURGOON, Judee K., BERGER, Charles R. and WALDRON, Vincent R..
2000. “Mindfulness and Interpersonal Communication”. Journal of Social Issues 56 (1): 105-127.
DAVIDSON, Donald. 1996. “A Nice Derangement of Epitaphs”. In The Philosophy of Language, third edition, edited by A. P. Martinich, 465- 475. New York: Oxford University Press.
GRĂDINARU, Ioan-Alexandru. 2011. “Donald Davidson's Approach of Malapropisms – Understanding Communication from the Standpoint of the Concept of Passing Theories”. Argumentum – Journal of the Seminar of Discursive Logic, Argumentation Theory and Rhetoric 9 (2):
109-120.
SĂLĂVĂSTRU, Constantin. 1996. Raţionalitate şi discurs. Perspective logico- semiotice asupra retoricii. Bucureşti: Editura Didactică şi Pedagogică.
SĂLĂVĂSTRU, Constantin. 2006. Mic tratat de oratorie. Iaşi: Editura Universităţii “Al. I. Cuza” Iaşi.
SPERBER, Dan. 1995. “How Do We Communicate?”. In How things are: A Science Toolkit for the Mind, edited by J. Brockman and K. Matson, 191-199. New York: Morrow.
SPERBER, Dan and WILSON, Deirdre. 1990. “Rhetoric and Relevance”. In The Ends of Rhetoric: History, Theory, Practice, edited by D. Wellbery and J. Bender, 140-155. Stanford: Stanford University Press.
SPERBER, Dan and WILSON, Deirdre. 1993. “Linguistic Form and Relevance”. Lingua 90 (1-2): 1-25.
VAN DIJK, Teun. 1977. Text and Context. Explorations in the Semantics and Pragmatics of Discourse. London and New York: Longman.
VAN DIJK, Teun. 1989. “Social cognition and discourse”. In Handbook of Social Psychology and Language, edited by H. Giles and R.P. Robinson, 163-183. Chichester: Wiley.
VAN DIJK, Teun. 1997. “Cognitive Context Models and Discourse”. In Language, Structure, Discourse and the Access to Consciousness, edited by M. Stamenow, 189-226. Amsterdam: Benjamins.
VAN DIJK, Teun. 2001. “Discourse, Ideology and Context”. Folia linguistica XXX/1-2: 11-40.
VAN DIJK, Teun. 2003. “The Discourse-Knowledge Interface”. In Critical Discourse Analysis. Theory and Interdisciplinarity, edited by Gilbert Weiss and Ruth Wodak, 85-109. Houndsmills: Palgrave-MacMillan.
VAN DIJK, Teun. 2005. “Contextual Knowledge Management in Discourse Production. A CDA Perspective”. In A New Agenda in (Critical) Discourse Analysis, edited by Ruth Wodak and Paul Chilton, 71-100.
Amsterdam: Benjamins.
VAN DIJK, Teun. 2006. “Discourse, context and cognition”. Discourse Studies 8(1): 159-177.
VAN DIJK, Teun. 2007. “Comments on Context and Conversation”. In Discourse and Contemporary Social Change, edited by Norman Fairclough, Giuseppina Cortese and Patrizia Ardizzone, 281-316. Bern:
Peter Lang.
