| A major trend in the study of Grammatical Metaphor occurred in1985whenHalliday first proposed the notion “Grammatical Metaphor (GM)” in his book AnIntroduction of Functional Grammar. Unlike the previous theories, Halliday proposedthat Grammatical Metaphor is the grammatical variant of meaning, though oftenaccompanied with the lexical variants. Since then, great numbers of papers and workshave been published involving the nature, categories, realizing forms, workingmechanisms. In1976, when Halliday and Hasan first proposed the notion “cohesion”in their work Cohesion in English, GM has not been regarded as a means of discoursecohesion. In the late1990s, Fan Wenfang (1999) noticed the discourse coherencefunction played by nominalization, which is the most powerful means of creating GM(Halliday,1994:352), and did qualitative analysis in its realizing forms.However, in spite of the numerous qualitative research concerning the applicationand cohesion function of nominalization (N), the quantitative research concerningthe evaluation of the discourse coherence of academic thesis on the basis ofnominalization (N) is still rare. Therefore，this thesis is aimed to use the quantitativeand qualitative research method, two self-built corpuses(one for English abstracts ofexcellent doctoral dissertation and the other for English abstracts of average doctoraldissertation), to explore the co-relation of nominalization and the degree of coherencein meaning.Based on the two self-built corpuses, this thesis needs to testify whether the twohypotheses are real and answer three questions. The two hypotheses are the drive ofimplementing the research: firstly, the English abstracts of excellent doctoraldissertation are judged more coherent than the abstracts of the average doctoraldissertation; secondly, the degree of coherence is positively correlated with thefrequency of nominalization. Specifically, the thesis attempts to answer threequestions in the analysis of academic discourse where nominalization has an extensiveutilization.First, to what extent are the abstracts coherent in meaning? Second, how aredegree of coherence of the abstracts and the use of nominalization co-related? Third,how are degree of coherence of the abstracts and forms of realization co-related? Statistics and Data in the fifth chapter all prove the validity and reliability of thetwo hypotheses. To make the findings explicit, subjects from the excellent group(corpus A) perform better than the subjects from the average group (corpus B). Apartfrom that, all the four functions show a positive co-relation between the degree ofcoherence in meaning and the frequency of the use of nominalization. So do thevariants of nominalization.Thanks to the previous qualitative analysis, some rewarding efforts were made inthe following aspects. First of all, two new corpuses with certain features were builtby the author. Secondly, it is the bold use of the quantitative method that marks.Thirdly, the thesis created a brand-new evaluation mechanism based on“three-dimension” evaluation method, which can provide new thought and possibilityfor the later quantitative analysis of coherence.The thesis is composed of six chapters. First, the introduction; Second, theliterature review of abstracts, concept and theory of cohesion and coherence; third, theGM theory under the theoretical framework of Systematic Functional Grammar;fourth, research design involving the whole procedure; fifth, specific correlationsexpressed by math functions after data analysis; sixth, conclusion and implications. |
PDF Full Text Request