The sociopragmatics of writing

Peter Grundy, Northumbria University

A broad distinction originally owed to Thomas and given wide currency by Leech in Principles of Pragmatics (1983) is that between sociopragmatics (broadly, when to use a particular pragmatic strategy) and pragmalinguistics (broadly, what form to use to achieve an intended pragmatic effect). Although I borrowed the term from Thomas/Leech, in my presentation I tried to extend the definition of sociopragmatics to include the default first language pragmatic interpretations of particular expressions which find their way into second language writing. I called this phenomenon the 'sociopragmatics of writing' because these default interpretations, or ideologies, are a true reflection of the culture of each group of users. More importantly, these commonsensical pragmatic interpretations of ways of saying things are largely invisible to first language users. As Verschueren puts it:

Language use, just like other forms of social behaviour, is interpreted by the actors involved. In the realm of social life in general, more or less coherent patterns of meaning which are felt to be so commonsensical that they are no longer questioned, thus feeding into taken-for-granted interpretations of activities and events, are usually called ideologies. (Verschueren, 2000:450).

Because of the invisibility to themselves of 'taken-for-granted interpretations', second language writers have special difficulties recognizing that they are creating problems for their readers. These problems are sometimes problems of intelligibility and sometimes the cause of that sense of 'otherness' that we have when reading the writing of members of other cultural groups. We feel that there is something 'wrong', but just what it is that's 'wrong' is relatively intangible.

At this point, I need to admit that my talk was given from the perspective of the reader who expects a target language norm in second language writing, and that this may be unattainable and inappropriate. I also admit that this perspective flies in the fact of much contemporary work (e.g. Jenkins, 2000) on the need of first and second language users of a global language to find a mutual intelligibility that may be unlike the norms established in first language use. And I suppose that the problems first language users of English have understanding the writing of second language users is as nothing when compared to the problems second language users have understanding the writing of first language users. However, my audience included largely those with the responsibility of helping second language writers to achieve 'native'-like second language writing in a very special context, an aim that is probably more devoutly wished by their learners than by themselves. Therefore I fully admit the political incorrectness of what follows and welcome approaches that set out from a more even-handed perspective.

So, where to start this journey into the perilous waters of the pragmatics of second language writing? Let's begin with Levinson's important distinction of three levels of meaning (1995, 2000). Working within a neo-Gricean paradigm, Levinson suggests that we have sometimes focused on the distinction between entailment (that meaning associated with linguistic forms whatever the context they occur in) and implicature (inferred meaning), assuming that a distinction only between context-free entailment and context-determined implicature is sufficient. However, as Levinson reminds us, Grice distinguishes generalized implicatures (i.e. context-free inferences) and particularized implicatures (context-determined inferences). Most work in pragmatics has focused on the latter, for the obvious reason that particularized implicatures are at the very heart of the kind of pragmatics which aims to study the relationship of context and language. However, these token-inferences, as Levinson calls particularized implicatures, should not blind us a further level of meaning, that provided by what he calls type-inferences, the characteristic inferences associated with the ways of putting things that are typical amongst any group of language users, and, as we noted above, that typically go unnoticed by them. Thus a more or less universal type-inference is to understand that when we say some, we intend those we speak to to infer 'not all'. (This is clearly an inference and not an entailment because it's easy to say some, in fact all without giving rise to a contradiction.) The point is that these type-inferences go far beyond relatively simple case such as some implies not all. For example, illocutionary force may also be involved. As Levinson states

[U] a level of systematic pragmatic inference based...on general expectations about how language is normally used. These expectations give rise to presumptions, default inferences, about both content and force. (Levinson, 2000:22).

In my talk, I illustrated this distinction with a simple case taken from a Chinese / English parallel text. The text cited was a public information leaflet about AIDS published in Hong Kong with a Chinese text on one side and an English text on the other. At one point, an imaginary character who is infected with the HIV virus says in English

(1) I don't drink or smoke

In my judgement, the type-inferences associated with this expression certainly include the notion that it is alcohol that the speaker does not drink (rather than that the speaker never drinks anything) and that the speaker does not smoke cigarettes (rather than that the speaker isn't on fire periodically). In addition, I also infer that the speaker regards drinking and smoking as same-category activities because of the conjunction reduction - i.e. if the speaker had said I don't drink and I don't smoke, I would not have drawn this inference.

By way of contrast, the Chinese text, translated literally, reads

(2) Do not smoke, do not drink alcohol.

Not only do I have to recover the subject I, but I also infer that smoking and drinking (notice the order) aren't same-category activities and that if someone said in Cantonese I don't drink, they wouldn't be understood to be referring to alcohol.

Imagine a Cantonese speaker translating from their mother tongue and saying in English 'I don't drink alcohol' - what inference do I draw? That the speaker is rather self-satisfied, slightly smug, rather prissy? But the poor speaker isn't to know that I don't drink gives rise to the type-inference that the speaker doesn't drink alcohol and that I don't drink alcohol gives rise to the type-inference that the speaker is self-satisfied. And of course, this is merely the tip of an iceberg.

In the second part of my talk, I listed some of the differences between Chinese and Anglo cultures as they have emerged in the very considerable social psychology literature, and tried to frame some first, naïve hypotheses about their possible effects on communication. Thus, for example, taking Hofstede's (1980, 1991) characterizations of Anglo cultures are small Power / Distance, high Individualism cultures and Chinese cultures as large Power / Distance, low Individualism cultures, and bearing in mind Brown and Levinson's (1978/1987) theory of politeness, according to which similar linguistic resources are available to speakers of all languages whilst their functional employment is triggered by culture specific computations of imposition and the hierarchical power and social distance differentials between speaker(s) and addressee(s), I hypothesised that: Small PD, high IDV cultures favour positive politeness (language structures used to encode mutual esteem and equality of status); and that large PD, low IDV cultures favour negative politeness (language structures used to encode respect for ascribed status). I then ran through a number of other distinctions between the two cultures (viz, Confucian Work Dynamism as identified by the Chinese Culture Connection and Yum's account of the influence of Confucianism) before listing the kinds of effect that I would expect to emerge as default type-inferences from Chinese and English texts and the kinds of pragmatic choices that I would expect writers to make in order to reflect the underlying value systems of the two cultures.

This then set the stage for an examination of two parallel Chinese:English texts, one a letter from a Hong Kong bank to its customers and the other a letter from the President of a University to members of staff. Despite apparently setting out to convey the same meanings, these texts did indeed exhibit striking differences with respect to the kind of default, type-inferences that a reader might be expected to draw from the structures they used to convey their messages.

This examination of parallel texts caused me to think a little bit about the implications for EAP methodology, which seemingly needs to react to the following:

So in my talk I raised a topic that isn't frequently discussed - how EAP teachers are to deal with the sociopragmatic properties of our students' writing. As you can see, I don't have any real answers.


Brown Penny & Stephen C. Levinson 1978 Universals in language usage: politeness phenomena. In Goody, Esther N. (ed.) Questions and Politeness. Cambridge: Cambridge University Press, 56-311. Reprinted with new introduction and revised bibliography as Politeness: Some Universals in Language Usage. 1987. Cambridge: Cambridge University Press.
Chinese Culture Connection (Michael Bond) 1987 Chinese values and the search for culture-free dimensions of culture. Journal of Cross-Cultural Psychology 18: 143-64.
Hofstede, Geert 1980 Culture's Consequences. Beverley Hills: Sage.
Hofstede, Geert 1991 Cultures and Organizations. London: McGraw-Hill.
Levinson, Stephen C 1995 Three levels of meaning. In Palmer, Frank Grammar and Meaning - Essays in Honour of Sir John Lyons. Cambridge: Cambridge University Press, 90-119.
Levinson, Stephen C 2000 Presumptive Meanings MIT Press.
Verschueren, Jef 2000 Notes on the role of metapragmatic awareness in language use. Pragmatics 10/4, 439-56.
Yum, June Ock 1988 The Impact of Confucianism on interpersonal relationships and communication patterns in East Asia. Communication Monographs 55 374-88.