| This thesis aims to investigate the relevance theoretic account of metaphor(henceforth the RT account of metaphor)and verifies the related theoretical hypotheses regarding the understanding mechanism of various types of metaphorical utterances via the paradigm of experimental pragmatics,exploring the causes underlying the theoretical controversy.This thesis first critically reviews metaphor research in classical rhetoric,semantics,pragmatics,cognitive linguistics,and functional linguistics,as well as their intrinsic flaws,before delving into the development of the RT account of metaphor and its internal disputes.Generally speaking,the RT account of metaphor is founded on the critique of Grice’s standardized pragmatics model.It stresses how concepts and propositional representations fulfill the goal of comprehending metaphorical utterances by constructing ad hoc concepts.In a nutshell,the RT account of metaphor has gone through three stages: loose use account,lexical pragmatics,and Carston’s dual processing mode.On this basis,this thesis highlights the inconsistency between lexical pragmatics and Carston’s dual processing mode in the description of metaphor understanding mechanism,which is embodied in the following three aspects: the division between ordinary metaphors and literary metaphors,the relationship between metaphors and the literal language,and the role of non-propositional effects in metaphor understanding.To clarify the theoretical disputes above,this thesis sets out an experimental pragmatic study from the following three aspects,namely emergent features,paraphrasability,and non-propositional effects.The reasons are as follows.First,it is generally agreed that emergent features are an intrinsic trait of metaphor,and their amounts are usually seen as a measure of the richness of the content of metaphorical utterance and also as the criterion for distinguishing between ordinary metaphors and literary metaphors.Second,paraphrasability refers to the relationship between metaphors and the literal language.It is assumed that the richer the content of the metaphor,the more difficult it is to express in literal language.Third,the non-propositional effects refer to the imagery,emotion,attitude,impression,and other pragmatic effects.They are one of the important features of literary metaphor processing under Carston’s model,and they are probably the underlying causes which lead to more emergent features and lower paraphrasability for literary metaphors.Assuming with these,this thesis puts forward the general research hypothesis: there are differences in the understanding between ordinary metaphors and literary metaphors.Specifically,compared with ordinary metaphors,literary metaphors can produce more emergent features as well as stronger and various non-propositional effects,but at the same time they have lower paraphrasability.In order to test this research hypothesis,this thesis,employing the paradigm of experimental pragmatics,attempts to answer the following three research questions:(1)How do the emergent features in metaphor understanding come into being?Whether there is some difference in the amount of emergent features between ordinary metaphor and literary metaphor? If so,what factors might be connected with the difference?(2)Whether there is some difference in paraphrasability between ordinary metaphor and literary metaphor? If so,what factors might be connected with the difference? And,what content(s)of metaphor might not be paraphrased via literal language?(3)Do different types of metaphorical utterances have an effect on non-propositional effects,and if so,how? What other factors may affect the non-propositional effects? And what is the possible relationship between non-propositional effects and the metaphor processing effort?To that aim,this study devises three cognitive psychology experiments,recruits428 undergraduates from a local university in China,and collects 80 metaphorical sentences from Chinese literary works and corpus as experimental materials.The experimental results are summarized as follows:Experiment 1 aims to see whether there is some significant difference in the amount of emergent features between ordinary metaphors and literary metaphors.The findings reveal that the amount of emergent features of literary metaphors is significantly larger than that of ordinary metaphors,illustrating that literary metaphors are richer in content and providing some preliminary evidence for the division between ordinary metaphors and literary metaphors.Experiment 2 aims to see whether there is some significant difference in the paraphrasability between ordinary and literary metaphors.The results reveal that literary metaphors have significantly lower paraphrasability than ordinary metaphors,and that some non-propositional effects such as imagery and emotion may be lost during paraphrase.The findings suggest that the metaphor is different from the literal language in pragmatic effects and provide further evidence that there are some differences between ordinary metaphors and literary metaphors.Experiment 3 delves into non-propositional effects further by examining how utterance types and contexts affect non-propositional effects(imagery,emotion,impression,thought,attitude and poeticality)and investigating the possible relationship between non-propositional effects and processing effort of metaphor understanding.The results show that utterance types have some significant effects upon imagery,emotion,impression and poeticality while contexts present some significant effects upon emotion and impression.Moreover,the understanding of literary metaphors requires listeners(readers)to exert more processing effort while retrieving more non-propositional effects as returns.These findings further consolidate the results of the previous two experiments.In summary,this study provides some preliminary empirical support for Carston’s dual processing mode,and the following findings are drawn:(1)Literary metaphor and ordinary metaphor are understood differently,as evidenced by the amounts of emergent features,the paraphrasability,and the strength of non-propositional effects.Literary metaphors,as compared with ordinary metaphors,create more emergent features and greater non-propositional effects while having lower paraphrasability.(2)One distinction between metaphor and literal language is that metaphors generate stronger non-propositional effects such as imagery and emotions.These non-propositional effects cannot be paraphrased in literal language.Simultaneously,non-propositional effects can also render metaphors(especially literary metaphors)more impressive.(3)Metaphor comprehension is influenced by a variety of pragmatic(non-propositional)factors in addition to semantic(propositional)aspects.In other words,metaphor comprehension is the result of an interaction between propositional and non-propositional representations.The main contributions of this thesis are as follows:First,this study theoretically reviews and reinterprets the RT account of metaphor,and clarifies the internal inconsistency within it via the paradigm of experimental pragmatics.Second,besides focusing on the distinction between metaphor and literal language,this study puts more stresses on metaphor itself,allowing for a more in-depth exploration of the understanding mechanism of different types of metaphors.Third,this study attempts to integrate certain notions from cognitive linguistics with relevance theory,which can help to explain the role of non-propositional effects in utterance comprehension,finetuning the RT account of metaphor and increasing the efficacy of relevance theory,making it a more effective and influential cognitive pragmatic theory. |
