Alternating ditransitives in English : a corpus-based study

• Main Author: Ozon, G. A.
• Published: University College London (University of London) 2009
• Subjects: 428
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.564709

This thesis is a large-scale investigation of ditransitive constructions and their alternants in English. Typically both constructions involve three participants: participant A transfers an element B to participant C. A speaker can linguistically encode this type of situation in one of two ways: by using either a double object construction or a prepositional paraphrase. This study examines this syntactic choice in the British component of the International Corpus of English (ICE-GB), a fully tagged and parsed corpus incorporating both spoken and written English. After a general introduction, chapter 2 reviews the different grammatical treatments of the constructions. Chapter 3 discusses whether indirect objects have to be considered necessary complements or optional adjuncts of the verb. I then examine the tension between rigid classification and authentic (corpus) data in order to demonstrate that the distinction between complements and adjuncts evidences gradient categorisation effects. This study has both a linguistic and a methodological angle. The overall design and methodology employed in this study are discussed in chapter 4. The thesis considers a number of variables that help predict the occurrence of each pattern. The evaluation of the variables, the determination of their significance, and the measurement of their contribution to the model involve reliance on statistical methods (but not statistical software packages). Chapters 5, 6, and 7 review pragmatic factors claimed to influence a speaker’s choice of construction, among them the information status and the syntactic ‘heaviness’ of the constituents involved. The explanatory power and coverage of these factors are experimentally tested independently against the corpus data, in order to highlight several features which only emerge after examining authentic sources. Chapter 8 posits a novel method of bringing these factors together; the resulting model predicts the dative alternation with almost 80% accuracy in ICE-GB. Conclusions are offered in chapter 9.