ABSTRACT
The study of literary language by linguists has always promised two types of insight. The first is insight into the individuality of a writer’s style; indeed the very name ‘stylistics’ reflects that concern. Milic (1967) on Swift, and Ohmann (1962, 1964) on Shaw and Faulkner, represent early successes in this area. The second is as a tool for the interpretation of individual works; here successes have perhaps been more sporadic and have been more dependent on the individual than on the development of useful techniques of analysis, with literary critics having at times justification for their oft-voiced suspicion of linguists’ interference in their discipline. One reason for this has been that systems of analysis centring on the sentence have only been revealing about eccentric works. Those many works whose individuality does not rest in any syntactic deviance have for the most part not been illuminated by sentential analysis. The burgeoning research in discourse analysis over the past decade offers, however, the possibility of approaching these works in a new and more productive way. If every discourse has a unique organization, then it follows that one way of identifying the individuality of a work is by analysing its organization as discourse. In Hoey and Winter (1981), based on an earlier unpublished paper by Winter, we attempted to show how a new interpretation of Julius Caesar grew naturally out of the detailed analyses of the speeches of Brutus and Mark Antony at Caesar’s funeral. In this essay I seek to demonstrate that the discourse analysis of a poem by John Donne results in a greater number of readings for the poem and a synthesis of those readings; at the same time it serves to test and challenge the descriptive system used.
