物质环境、符号还是认知模型？——谈韩礼德、马丁、范迪克的语境观

A Reading Report About ‘Context’ (Oct. 15-28, 2020)

刘立华.(2019)马丁对语类研究的贡献. 外语学刊(01),12-17. doi:10.16263/j.cnki.23-1071/h.2019.01.003.

Halliday, M. A. K. & Matthiessen, C. (2004). An Introduction to Functional Grammar (Third Edition). London: Hodder Arnold.

Halliday, M. A. K. & Hasan, R. (1989). Language, context, and text: aspects of language in a social semiotic perspective. Oxford: Oxford University Press. (pp. 5-14)

Halliday, M. A. K. (1978). Language as a Social Semiotic: The social interpretation of language and meaning. London: Edward Arnold

Martin, J. R. (1998). ‘Modeling Context: The Crooked Path of Progress in Contextual Linguistics’. In: Ghadessy, M. (Ed.), Text and Context in Functional Linguistics. Amsterdam: John Benjamins Publication.

Cloran, C. (1998). ‘Context, material situation and text’. In: Ghadessy, M. (Ed.), Text and Context in Functional Linguistics. Amsterdam: John Benjamins Publication.

• Introduction

I was wondering why Martin proposed ‘genre’ as a complement to Halliday’s register theory, while Halliday left out genre in his theory. Liu Lihua’s paper offers a brief introduction on the development of Martin’s theory about genre, which gave me some hint on the divergence of Halliday’s and Martin’s views on the concept of ‘context’. Based on this information I made a superficial inquiry on ‘context’ by going through some available literature.

 

• Halliday’s and Martin’s definition of ‘context’

To define the concept of ‘context’, Halliday & Hassan (1989) started with the literal meaning of ‘con-text’, namely ‘with the text’, which can be both verbal and non-verbal. In other words, context is a situation, or environment where the text unfolds. This notion of ‘context’ mentioned by Halliday is inherited from Malinowski’s context of situation, i.e. the immediate environment of the text; and the other kind of context introduced by Malinowski is the context of culture, i.e. the total cultural background.

Worth noticing, in Halliday’s earlier book, Language as a Social Semiotic: The social interpretation of language and meaning (1978), the semiotic aspect of context is emphasized; ‘a context of speech is itself a semiotic construct’ (ibid., p. 2). This view of context leads to a contrast of context and the ‘what is going on’, i.e. context as an abstract representation of the environment can be remote from the actual physical environment (cf. Halliday 1978, p. 109, Cloran 1998). But in IFG (2004), context seems to be the equivalent to ‘experience and interpersonal relationships’ (ibid., pp. 24-25); though it may be natural to argue ‘experience’ and ‘interpersonal relationships’ have a semiotic aspect as well as a physical/material one, context as a semiotic construct or an information system is not clearly stated. Another case that may give rise to doubts on the consistency of Halliday’s perspective on ‘context’ is in Language, context, and text: aspects of language in a social semiotic perspective (1989), where ‘goings-on’ and ‘the total environment’ are immediately used to define context.

Anyway, Martin (1998) clearly reemphasized the semiotic aspect of context: ‘following Halliday (1978), we were as far as possible trying to model context as a semiotic system rather than something material or mental’.

• Halliday’s and Martin’s description of ‘context’

To give a description of ‘context (of situation)’, Halliday outlined a framework of three headings—the field of discourse, the mode of discourse and the tenor of discourse, while Martin, following Gregory, proposed a framework of four heading with an extra variable of ‘functional tenor’. Debates mainly pertain to the notion of ‘functional tenor’. Several alternative terms have been used to refer to ‘functional tenor’—'text structure’, ‘schematic structure’, and ‘genre’; and they are all related to the global text organization (Martin 1998). Specifically, Martin made two major modifications to Halliday’s framework. First, He introduced ‘genre’ because he found that ‘context manifested itself by skewing probabilities in linguistic systems’; thus, a Second, he replaced the term ‘context of situation’ with ‘register’ to get rid of the materialist readings the term context invites (ibid.). Here we can see the ‘more extreme perspective on the relationship between context and text’ held by Martin (cf. Liu 2019).

• Conclusion (very subjective and premature)

The notion of genre is necessary for Martin’s framework since Martin sees context as strictly semiotic. To him the gap between context (with material readings) and language is impossible to be negotiated. In contrast, Halliday’s view on context remains vague to me, though he also defined context in semiotic terms in his early work.

 

(Update) About the inconsistency in Halliday's theory of context; Van Dijk's view on context as mental model

Van Dijk, T. A. (2008). Discourse and Context: A Sociocognitive Approach. New York: Cambridge University Press. (pp. 1-55)

Van Dijk, T. A. (2012). 'Critical Context Studies'. In: Tian Hailong & Zhao Fan (Eds.), Critical Discourse Analysis: Essential Readings. Tianjin: Nankai University Press. (pp. 263-295).

Van Dijk, T. A. (2014). Discourse and Knowledge: A Sociocognitive Approach. New York: Cambridge University Press.

 

Van Dijk's telling criticism on Halliday's theory of context justifies my doubts. Though Halliday is a 'staunch anti-mentalist' (i.e. one who rejects a psychological interpretation of 'context'), mental terminology such as 'planning' is used in his theory of context (cf. 'Context and Language'. In: Van Dijk 2008, pp. 28-55).

I am more into van Dijk's view on context as mental model, which is clearly defined and practical in discourse analysis. I am going to explore more of it.

 

• Van Dijk’s definition of context

According to van Dijk, a theory of context 'must avoid social positivism, realism and determinism at the same time' in that 'if context were objective social conditions or constraints, all people in the same social situation would speak in the same way' (Van Dijk 2008, p. x). In other words, a theory of context 'needs to account for a more complex way to deal with 'covariation', conditioning or causation between social situation and discourse structure' (Van Dijk 2012, p. 267). Martin also saw the covariation between social situation and discourse structure, as he put it, 'context manifested itself by skewing probabilities in linguistic systems' (Martin 1998, p. 29). While Martin introduced ‘genre’, Van Dijk proposed to view contexts as mental models.

 

Mental models are updated in working memory and stored in episodic memory, part of long-term memory. A mental model is parameterized by schematic categories including Setting, Participants, etc. Contexts or context models are instances of the mental models that dynamically represent the communicative situations. In addition, context models have a special knowledge (here, knowledge is defined as a kind of belief) device, or K-device which defines the Common Ground (i.e. the shared knowledge) of the participants (cf. Van Dijk 2008, 2014).

 

• Van Dijk’s description of context (or mental modes)

In consideration of the limitations of short-term memory (i.e. working), van Dijk claimed that must be relatively simple, including the following schematic categories:

• Setting: Time/Period, Space/Place/Environment;
• Participants (self, others)
  • communicative roles (participation structure);
  • social roles types, membership or identities;
  • relations between participants (e.g. power, friendship);
  • shared and social knowledge and beliefs;
  • intentions and goals;
• Communicative and other Actions/Events.

(see Van Dijk 2008, p.76. In a later version of this list, ‘Goals’ is listed separately, see van Dijk 2014, p.50)

• Conclusion

Linguists look into ‘context’ from different perspectives, including at least the following three:

• A material/physical perspective: Context is the physical setting of discourse. Though a division between ‘linguistic context (i.e. co-text)’ and ‘non-linguistic context (i.e. situation/environment)’ may be made by scholars who hold this perspective, their inquiry into the ‘non-linguistic’ aspect of context seems to fundamentally related to the physical world.

Representatives: Malinowski(?), Firth(?)

• A semiotic perspective: Context is the semiotic construct representing the environment of discourse.

Representatives: Halliday (1978), Martin

• A cognitive perspective Context is the subjective participants construct (i.e. the mental model or the instance of mental model) of communicative situation.

Representatives: Van Dijk