An Empirical Study Of The Cognitive And Communicative Principles Of Relevance

The thesis reports an empirical study of the cognitive and communicative principles of relevance the author has performed. Through the tests of relational problems, it attempts to explore whether human cognition is relevance-oriented; whether optimal relevance is expected and fulfilled by interlocutors in communication.The thesis starts with the definition of Relevance: a property of inputs to cognitive processes. Relevance theorists propose that human beings tend to achieve the greatest possible effect at the lowest possible cost, that is to say, they are inclined to obtain a Maximal Relevance. In an ostensive communication, however, the hearer can only expect an Optimal Relevance due to the speaker's abilities and preferences in producing the utterance while still believing this utterance is intended by the speaker as relevant enough as claimed by the Presumption of Optimal Relevance. Therefore, the Cognitive Principle of Relevance is defined as that"Human cognition tends to be geared to the maximisation of relevance"; whereas the Communicative Principle goes as follows:"Every act of ostensive communication communicates a presumption of its own optimal relevance".Then the thesis turns to a critical review of the tests on the principles of relevance. By pointing out the factors which should have been taken into account in those tests, the author highlights the necessity of similar tests in different social backgrounds with varied testing questions and a more comprehensive as well as concise design. This consideration leads to the following improvements in the present study: a Chinese version of three-term relational problems (For example, one set of the premises goes on like this:"老马比老曹大;老马比老郑大". Participants are requested to supply their conclusions.); two newly-added types of questions and two differentiated types of instructions (the one allowing participants to write"nothing follows"as a valid answer and the one without such hint), thus integrating the tests on the two principles. This new design eliminates the influence of the instruction from the experimenter's side and the uncontrollable factors from the participants'side.Two hundred and forty students (all were freshmen then in National University of Defence Technology) in the first test and eighty students (all from Guangdong University of Foreign Studies) in the second study were randomly chosen to complete the test papers in Chinese. No limitation of answering time was required. Through the classification of answers (determinate answers including the single-subject determinate ones, single-subject indeterminate ones plus double-subject determinate ones; indeterminate answers including double-subject ones and"nothing follows"answers; and others) and calculations of the original data (Bayesian statistics was used in some of the analyses.), and in view of the hypotheses derived from the principles, we achieved the following results:a. According to the cognitive principle and as proved by the test results, more participants process the relational problems in the most productive way, so that to both determinate and indeterminate problems there are more determinate answers depicting relations between either two or three of the terms mentioned.b. Still in view of the cognitive principle and illustrated by the results, people would like to derive more single-subject than double-subject conclusions from indeterminate problems as single-subject answers require less mental effort to be remembered and retrieved than double-subject ones.c. In order to reduce future effort and increase cognitive gains, in the previous test performed in France, more participants gave single-subject than double-subject responses to different-subject premises, whereas in the Chinese test, due to the different syntactical grammatical systems (topic-prominent sentence pattern in Chinese language versus subject-prominent in French), thinking modes (conservational and introspective versus open and extroversive) and logics in thinking (holistic and linear versus analytic and spatial), Chinese participants do not pay equal attention to subjects when constructing a sentence, which reflects in relatively less single-subject conclusions to different-subject indeterminate problems. The divergence here, as proposed by the thesis, indicates the direction and degree of relevance adopted by individuals from distinctive contexts so it cannot simply be taken as an objection to the principle. And this in turn throws light on the importance of alteration rather than adoption of former research hypotheses and designs in empirical studies.d. Resulting from the communicative principle and as shown by the test data, participants in Group B without the permission to write'nothing follows'supply more determinate answers to both determinate and indeterminate problems than those in Group A do since their expected relevance aroused by the survey instructions vary, so that their answers satisfy different degrees of relevance. The result, nevertheless, proves the practicability of the author's modification of relational problems as a test for the communicative principle.The findings of the present and previous tests imply, as pointed out by the thesis, that Relevance Theory as well as other pragmatic principles and theories, can be tested by instruments borrowed from experimental psychology and psycholinguistics, through hypotheses derived from the theoretical framework. The altered or invented designs, conversely, promote psychological and psycholinguistic studies. Another implication of the present findings is pedagogical: no matter in communication or language teaching and learning, the different direction and degree of individual's search for relevance should be considered.The empirical study of the principles of relevance presented in this thesis has great significance because it has initiated a bold try in Experimental Pragmatics, a rather underdeveloped research realm in China. The limitations in the study are expected to lead to future explorations in this'wonderland'.