1. 研究目的与意义（文献综述包含参考文献）

1. IntroductionMetaphorical competence (MC), according to Gardner and Winner (1978), is the capacity to paraphrase a metaphor, to explain the rationale for the metaphors effectiveness, to produce a metaphor appropriate to a given context and to evaluate the appropriateness of several competing metaphoric expressions. Danesi (1986) was the first to introduce the notion of MC into the domain of second language (L2) learning. Foreign language learners are bound to be confronted with all kinds of metaphors in the target language and comprehending them is necessarily important. A lack of MC will result in such a situation where learners can express L2 in a correct and acceptable way, but not as typically as native speakers (Danesi 1995). Consequently, developing MC and mastering metaphorical ways of the target language should necessarily become an important goal of L2 learning.Idioms are a common phenomenon and the term idiom has a wide range of meanings, and generally refers to those fixed phrases with specific forms. The meaning of its implication cannot often be inferred from the meaning of a single word. It is one of the most active and sensitive parts of English, and it is also an effective tool for communication. As we all know, nowadays, English idioms have been widely used in many fields. Although L2 learners have more and more opportunities to watch English movies, TV series, magazines, etc., some people still find it difficult to understand these English materials without knowledge of relevant idioms. English idioms are usually not taught in middle school and university English classes, and students need to understand and master them by themselves. Many students have problems in understanding idiom variants due to a lack of cultural background knowledge, low language proficiency and cognitive ability, etc. This thesis aims to explore the relationship between the level of MC and the understanding of idiom variants. This thesis is divided into five chapters. The first chapter is the introduction, which presents the research background, research questions and the overall structure. The second chapter reviews previous studies on idiom variants, MC and elaborates the research gap. The third chapter introduces the research methodology. Chapter Four analyzes the data and provides a detailed discussion of the result. Chapter Five is the concluding section, which summarizes the main findings, points out the limitations and puts forward suggestions for further research.2. Literature Review2.1 The definition of idiomOne of the most difficult problems in idiom research has been the question of how to define idioms. For a long time, the study of idioms has run into two main difficulties, i. e. the terminology involved and the delimitation of the concept resulting in a proliferation of definitions by various researchers working in the field (Meier 1975). According to Moon (1998), idiom is an ambiguous term, used in conflicting ways. In lay or general use, idiom has two main meanings. Firstly, idiom is a particular manner of expressing something in language, music, art, and so on, which characterizes a person or group. Secondly, an idiom is a particular lexical collocation or phrasal lexeme, peculiar to a language. Huo Ying and Song Yan (2020) believe that English idioms are rationally motivated and can be divided into rational and irrational idioms. Rational idiom builds its meaning on the various parts that make up the idiom, while irrational idiom cannot be inferred and understood simply from the words constituent parts of the idiom. 2.2 Studies on English idiom variants2.2.1 Researchs by foreign scholars The actual use of idioms seems to deviate from its basic characteristics. In other words, idioms can have different variations. In the 1960s, Chomskys transformation-generative grammar theory ruled out the possibility of idiom variants and he recognized words as the greatest fixed units in the language. However, Fraser (1970) recognized the variant potential of idioms and argued that idioms make one or a series of components rather than a fixed whole. He put forward the frozen level table of idioms, which are: 6 levels of unlimited, 5 levels of reconstruction, 4 levels of extraction, 3 levels of permutation, 2 levels of insertion, 1 level of addition, and 0 level of complete freezing. Frasers research is a bold attempt to classify idiom variants. However, his hierarchy is subjective in a sense, because there is no clear dividing line between these types. Following Fraser, Nunberg (1994) also tried to classify the variants of idioms. He believes that they can be divided into four categories: attributive insertion, objective preposition, component omission and noun antecedent. His classification explains several changes in detailed examples. But he did not notice that these four types are just the tip of the iceberg of idiom variants.Fernando (2000) classified idiom variants from the perspective of traditional grammar. In her depiction, there are four variants: replacement, addition, arrangement, and deletion. She also divided the above-mentioned variants into two categories: one is the normal variant that is part of the language system, and the other is the variant that exhibits innovation and novelty that breaks the rules, but she did not elaborate further. Classification of normal variants and innovative variants. Fernandos grammatical classification is not comprehensive, because she ignores some other forms, such as copying, reversal, etc. Her research mainly focuses on the perspective of grammar, and grammar is not comprehensive in terms of idiom variations.Burger et.al (1982), Moon (1998) and Langlotz (2006) agree on another classification. Langlotz (2006) argues in her book Idiom Creativity that regardless of the potential language motivation and communicative effects of a given idiom variant, we can describe and classify the form and semantic changes of the idiom variant. This group of scholars analyzed the changes in form and semantics, including formal morphological-syntactic changes, formal syntax and lexical changes, and semantic lexicalization, discourse changes, and meaning adaptation. However, the use of language itself is a process of complexity---one changed idiom may go through the above-mentioned techniques instead of one technique. Some can be transformed both formally and semantically. Idiom variants are a latecomer in the study of English idioms. There are few updates in their classification. The basic classifications of Fernando (2000) and Langlotz (2006) are still used by scholars.2.2.2 Researchs by home scholarsHua Xianfa (1998) is one of the pioneer scholars in China to classify idioms. Hua (1998) divides idiom variants into two categories: legal variants (or normal variants) and illegal variants (temporary or innovative variants). He mainly studies the illegal variants of idioms and proposes four illegal variants: vocabulary variants, phonetic or spelling variants, grammatical variants and semantic variants. Huas classification is multiple branches of linguistics which involves lexicology, phonology, grammar and semantics. This is a new point of view, but there are also flaws. First, in his classification, there is a boundary with limited nodes between the temporary variant and the normal variant. Second, the term illegal does not apply to innovative variants, because once the variant succeeds, the variant becomes legal in a pragmatic sense, and the term innovative becomes one-off. Third, the same variant may be legal for some idioms but illegal for other idioms. Therefore, the term legal or illegal cannot distinguish idiom variants from each other. For the above reasons, this classification of idiom variants is not perfect. Luo (2006) also agreed to divide idiom variants into legal and illegal ones. In his book, legal variants include phonetic variants, component substitutions, number of noun variants, verb ending variants, and other minor variants. Compared with the classification of Hua (1998) and Ferernando (2000), this is a more detailed classification. However, this classification of legal and temporary variants has the same disadvantages as Huas. Shi Luoxiang (2013) also analyzed the temporary changes of idioms under the premise of the classification of Hua (1998), and did not make a breakthrough in the classification of idiom changes.In short, domestic scholars have reached a consensus on the classification of idiom conversion. They all agreed that there are two types of idiom conversion: legal conversion (normal conversion) and illegal conversion (temporary conversion). But this classification has its own shortcomings.2.3 The definition of metaphoric competence and relevant studies2.3.1 The definition of metaphoric competenceContemporary cognitive linguistics theories regard metaphor as the basic way of human cognition, and it also determines the way of language and cognition expression. Therefore, training L2 learners to understand, interpret, and create metaphors can help them form object languages, mode of thinking and improve their language skills. However, the current academic research on MC is less than 40 years, and scholars have not yet reached a consensus on the definition of MC.MC can be defined in both a narrow and a broad sense. The narrow meaning of MC mainly includes the acceptance of metaphors and the productivity of metaphors. Gardener Winner (1978) once proposed that the competence of metaphor should include some interpretation of metaphors and their rationale, and the ability to evaluate and produce appropriate metaphors under specific circumstances. Pollio Smith (1980) mentioned that Metaphorical competence mainly includes the originality of metaphor output, fluency, and the ability to discover metaphorical meaning. Since then, researchers Danesi and Low have conducted a more in-depth exploration of the definition of MC. Among them, Danesi (1992) pointed out that Metaphoric competence is one of the characteristics of native speakers different from foreign language learners, and it is also one of the basic requirements for native speakers to achieve conceptual fluency. If foreign language learners have mastered metaphor reasoning ability, they can also reach the fluency level of native speakers. Later, Littlemore (2006) added the ability to extract metaphorical meaning speed. According to Iijima (2) The ability to obtain metaphorical meaning from semantic contradictions; (3) The ability to use traditional metaphors reasonably; (4) The ability to use typical hedges with metaphorical properties; (5) Interpretation metaphor at multiple levels; (6) The social sensitivity of metaphor use; (7) The consciousness of metaphor interaction. The broad definition of MC has the characteristics of social interaction and includes various related factors. However, contrary to its narrow definition, it lacks a clear and operable research framework. In contrast, its narrow definition can highlight the cognitive nature of metaphor. 2.3.2 Relevant studies of metaphoric competenceAs a high-level language competence, MC is one of the signs that L2 learners can master and use a language authentically. However, in the actual use of English, students rarely use metaphors, especially some metaphorical expressions. In the process of English reading, students often encounter such a phenomenon: most of the words in the English reading materials are recognized, but it is difficult for students to understand the meaning expressed by these sentences or chapters. That is to say, students cannot identify some metaphorical words and sentences that deviate from the original meaning of the dictionary in the process of reading. In addition, in the process of English writing, students seldom use some metaphorical language to express their views and make the article rich in language. The expression of sentence pattern is diversified, and there are many straightforward expressions. Even if some metaphorical language is used, it is not expressed in English idiomatically, and most of them are transferred from Chinese. All these phenomena indicate that students MC are deficient in the process of English learning.With the further development of cognitive linguistics, people have begun to realize the importance of cultivating MC in language learning. The cultivation of MC is conducive to L2 learners acquisition of basic linguistic knowledge of the target language, strengthening the understanding of the conceptual system of the target language and its cultural background, as well as the mastery of the thinking mode of the target language and the improvement of their cognitive ability. It is of great significance to the improvement of L2 learners comprehensive language use ability.Reilly and Marsden (2020) discovers that in order to study MC, researchers use natural data, for example, silent oral production (Pitzl 2016), written assignments (Kathpalia Nacey 2013) and heuristic methods. Compared with heuristic methods, naturalistic data are believed to provide better insight into the true, spontaneous language in use, but their description of the unknown of knowledge is less clear. For example, if the learner does not produce a specific metaphor or a certain type of metaphor, then the researcher will not be able to know whether the L2 learner does not know it or just has no chance or is unwilling to use it. Of course, it is often difficult to make a clear distinction between natural data and derived data, because it can be seen that the two overlap, such as citing spontaneous works in a role-playing job interview or citing pre-prepared content in a written article.2.4 The research gapWith the continuous development of metaphor theory, the definition of metaphor is also expanding, from the original rhetoric category to any non-literal category. However, the unclear definition of metaphor has led to many different opinions in specific research, which has caused the problem of unclear classification. This thorny problem exists in the current research, which directly limits the development of metaphor standardization research. Although the current metaphor research literature can be described as numerous, the most basic problem of the definition of metaphor has not yet been resolved. Whats more, the content of testing MC in the existing research is not comprehensive. Existing researches mostly test MC for the external manifestations, such as comprehension ability and output ability, while ignoring the study of its inherent distinguishing characteristics, such as the relationship between different conceptual domains, mapping capabilities between different domains, semantic association capabilities between different domains, etc. Therefore, it is difficult to guarantee the validity of the study.Then, the above literature review on idiom variants is mainly divided into two parts: meaning-retained variation and structure-retained variation. There are no examples of idiom variation that experience changes in both basic structure and meaning. However, it must be pointed out that the structure-retained here is a relative concept which refers to the reservation of its basic form. And categorization from the language users perspective is subjective to some extent. As to the motivations of idiom variants, there may be other extra-linguistic factors that the scholar fails to mention. Linguistic signs, include idiom variants, are complexities that interact widely with the world around them. Other extra-linguistic factors that contribute to idiom variants are endless in amount and will be sorted out in pace with our cognition of this special linguistic phenomenon.All in all, many researches on MC and many meaningful conclusions have been drawn, but there are still deficiencies. The most important shortcomings are the definition of MC and the testing of MC. Moreover, there are countless studies on idiom variants, but they are not comprehensive enough. In summary, although the respective researches have in-depth discussions, previous studies have not explored the relationship between the two. And this paper will explore the relationship between the two through two questionnaires.

2. 研究的基本内容、问题解决措施及方案

This study uses two questionnaires to collect data. The first was a MC level survey, which consisted of two types of questions. The first type asks the subjects to judge whether the sentences have a metaphorical meaning; the second type is concerned about how to interpret the sentences that they previously judged to be metaphorical. The second questionnaire asks the subjects to explain the idiom variants. Rate the level of difficulty in explaining each idiom variant in the following grading questions. SPSS 16.0 will be used for statistical analysis of the average, median, t-test and other values in the questionnaire.The questions to be addressed in this thesis are as follows:1) What is the MC level of English major students?2) How does the MC level influence the understanding of idiom variants? What type of idiom variant is more significantly influenced on understanding?3) How to improve the MC level of English major students?