We have been reading some papers from the book Language and Gender, edited by Jennifer Coates.

The book is a collection of sociolinguistic research work from the seventies and eighties. It is fascinating in a number of ways – first, that it gives objective evidence about the way various social groups use language, and, second, that, like most advances in science, it depends on a breakthrough in technology. The research workers use the sensitive and discreet taperecorders available in the seventies to record long stretches of actual conversation, which they then transcribe and analyse. This work throws new and often unexpected light on how conversation actually works, often contradicting the myths which surround such discourse.

Jennifer Coates’ own contribution, Gossip Revisited: Language in All-Female Groups, is a very interesting example of this kind of work. She recorded some hours of informal conversation between five middle-aged academic women, who had been friends for twelve years.  She calls such conversation “gossip” and says that “the aim of such talk is to create and maintain good social relationships.”

One unexpected finding is that this undirected informal chat follows a quite formal pattern in which the women co-operate to explore different topics. In the furtherance of this aim they interject supportive murmurs (such as “yeah” and “mhm”), which do not interrupt the speaker but shows they are listening. The middle section of each topic consists of a passage in which, instead of clear turn-taking, a number of people speak at once to develop the theme. This “simultaneous speech” is not competitive but, again, supportive.  She also produces evidence that what she calls “epistemic modality” is frequently used – that is, phrases such as “I think”, “perhaps”, “sort of”, “probably” – in order not to assert too strongly the truth of the propositions they are making, which might distort the co-operative exploration of the topic under discussion.

The mind boggles at the sheer amount of work that goes into recording, transcribing and analyzing such conversations. It is clearly worthwhile in that it opens up whole areas of oral language for linguistic description.

In my last blog I gave a summary of Tomasello’s triadic theory of human cognition (different from the dyadic communication of the great apes).

Tomasello’s  main theoretical aim is to contest Chomsky’s hypothesis of the Language Acquisition Device,  namely that a Universal Grammar is programmed innately into the human brain. Tomasello argues that, on the contrary, the learning of language can be accounted for more satisfactorily in terms of usage (nurture rather than nature).

There is already a large body of empirical research to support this hypothesis.  Tomasello’s Construction Grammar, as it is called, has the advantage over Chomsky’s Universal Grammar in that, rather than narrowly isolating grammar from all the semantic information which any real utterance has to incorporate, it examines whole utterances at any linguistic level. It also encompasses figurative and abstract language, which Universal Grammar cannot handle.

What I find curious is that the only linguistic schools considered by Tomasello are those derived from Chomsky’s model. There are a number of alternative schools including Systemic Functional Linguistics associated with the work of M.A.K. Halliday* and his successors in the Sydney School. SFL is based on a triadic functional analysis of human language: Continue reading →