In the last decades, linguistics has ‘jumped the border’ of the sentence and moved towards larger units of description such as text and discourse. In this seminar we will cover some principal linguistic approaches to text, addressing frameworks such as cohesion, rhetorical structure theory and introductory segmented discourse representation theory. Class work will consist of overviews of the theoretical approaches (supported by readings) followed by group-based analysis and discussion of example texts. Final credit for the module can be obtained by carrying out a more detailed analysis of a collection of short texts (possibly in groups) and motivating the decisions made. Considerations of corpus linguistic approaches to discourse structure and organisation will also be addressed and some particular tools for supporting such analyses introduced. Successful participation in the course should enable the analysis and critical discussion of texts in general, as well as raising awareness of current open topics and issues in linguistic discourse research.

In this course participants will be introduced to how grammatical analysis can be embedded into social critical analyses of texts, moving from detailed grammatical analysis to methods for evaluating both whether the analysis has been performed accurately and whether one has actually found differences between texts are not using basic methods of statistical comparison and corpus analysis. The main kind of grammatical analysis that will be introduced and practised is transitivity. Transitivity is the part of grammar that encodes the speaker or writer's view of reality--literally the 'who did what' part of grammar. We will explore a wide range of English texts in order to practice recognising the basic types of transitivity patterns in English. The main emphasis will be on doing, so that all successful course participants will become proficient in analysing texts according to their transitivity and testing the reliability of their results. In short: Particular attention will be paid to evaluating the reliability of analyses and to demonstrating whether or not bodies of data actually exhibit statistically significant differences.