Mutual Prediction Between Context And Text

The relationship between context of situation and text enjoys a long history, during which many researches on either of the definitions or the relationship between them have been done by many scholars at home and abroad. Researches on both the definitions can be found everywhere. As for researches on the relationship between them, studies on the influence of context upon text are considerable while those on how text affects context are poor according to the survey done in this thesis.According to Halliday, the fundamental relationship between the two is their mutual prediction, so one can not only predict context from text, but predict text from context in a contrary direction. The present thesis hold the view that in the course of discourse understanding the functions of both text and context on each other are of equal importance. So this thesis covers not only the contributions of context which have been done in predicting text in discourse understanding but how the prediction of context of situation from text works in the process of analysis.Discourse analysis must be based on certain data. The data of this thesis is selected from an old but special literature form—play. Due to the fact that the previous studies on play mainly focused on its basic elements from the perspective of literature, while only a minority of linguistic theories like the register variation in pragmatics was found in the analysis of play, this can be said to be a great pity.Based on the understandings above, the current author employs Hallidy's Theory of Mutual Prediction between Context and Text to analyze the dramatic language in Arthur Miller's play script Death of a Salesman. There are altogether 12 selected texts which are key events related to Willy's death, among which 5 are analyzed in the direction of predicting text from context and the rest 7 in the direction of predicting context from text. In the process of analysis, qualitative analysis, quantitative analysis, data statistics, and probability statistics are adopted mainly in forms of tables and figures.The analysis results show that both predicting text from context and predicting context from text contribute a great deal to discourse understanding. And this point is proved further in Chapter 4 where the expected results has been got that according to the predicted information by virtue of Halliday's Mutual Theory, the selected texts are closely connected to Willy's'death', which is the criterion the current author follows when choosing the data. This not only proves the practical application of this theory in play, but reminds us of that this theory will provide an entirely new perspective for discourse analysis of any genre in the future.