摘  要：This is an interview with Dr. Boping Yuan, Reader in Chinese Language and Linguistics at Cambridge University, UK and professor of linguistics and applied linguistics at Shanghai Jiao Tong University, China. In this interview, Dr. Boping Yuan looks back on his 30 or so years of experience directing the modern Chinese language program at Cambridge University and working in the field of second language acquisition. He describes the changes and the future of Chinese language programs in teaching and researching Chinese as a foreign language (CFL) contexts and his viewpoints on CFL teacher development. He also gives a detailed explanation on the issue of interface and feature configuration in second language acquisition, and discusses the transfer effect in third language acquisition. Furthermore, Dr. Boping Yuan offers an outlook on the future of language acquisition research which will become more interdisciplinary and will be combined with modern technologies to advance human well-being. In the end, he provides valuable advice for young researchers on theirpath to pursing second language acquisition as a profession.

    关键词：Teaching and researching Chinese as a foreign language, Chinese language program, second language acquisition, third language acquisition, language teaching

【被采访人】

袁博平，英国剑桥大学荣休教授，现任上海交通大学特聘教授，剑桥大学博士生导师，剑桥大学丘吉尔学院院士，国际期刊 Journal of Second Language Studies (John Benjamins) 主编，同时担任若干国内、国际学术刊物的编委，其中包括Second Language Research, International Review of Applied Linguistics, Journal of Chinese Linguistics, Lingua,《外语教学与研究》，《世界汉语教学》, 《现代外语》等等，兼任世界汉语教学学会常务理事，世界汉语教学学会语言学分会副会长，剑桥大学-香港中文大学双语联合实验室主任等职务。他的研究兴趣包括理论语言学框架下的二语和多语习得及双语与人类健康。他在国际主流学术刊物和中国重要学术刊物发表过八十余篇论文，其中包括：Language, Linguistics, Transactions of the Philological Society, Second Language Research, Studies in Second Language Acquisition, Bilingualism: Language and Cognition, Language Learning, International Review of Applied Linguistics, EUROSLA Yearbook, International Journal of Bilingualism, Lingua, Journal of Chinese Linguistics, Linguistic Approaches to Bilingualism, Frontiers in Psychology,《外语教学与研究》，《世界汉语教学》, 《现代外语》等等。他还编撰两部英汉-汉英词典（牛津大学出版社）。

【采访人】 

张芃，上海交通大学语言学方向硕士生。她的研究兴趣包括失语症、言语听力科学及外语教育。

张：One area of your research focuses on second language interfaces between syntax-semantics,syntax-discourse, syntax-pragmatics, etc. Why do you think the issue of interface is important for second language acquisition?

袁：In adult second language acquisition, one of the puzzles to everyone is that a human being can learn their first language easily and effortlessly, and within four or five years, a child can master his or her first language without endless drills, repetition, and efforts. Children can learn their first language very rapidly and easily. In contrast, adults spend hours and hours, years and years studying a second or foreign language. They can make progress, but very few adults can manage to reach native competence of the target language. Then people wonder how come most of adults fail to reach native competence, even though they spent significantly more hours in learning the target language. So why? There are many theories about this. For instance, the Critical Period Hypothesis argues that after the critical period people can no longer acquire a native competence of the target language.The Universal Grammar theory argues that part of the universal grammar is still available. For example, lexical words like table, cup, and computer are still available, but functional words are no longer available to adult language learners. Then Sorace and Filiacci proposed an Interface Hypothesis, dealing with the issue from a different angle. According to them, the difficulty or the impossibility of reaching native competence is not because of the critical period hypothesis, nor the availability or unavailability of the functional categories in the universal grammar. The difficulty is from the integration, coordination, and interaction between different domains in the language faculty. The language faculty in our brain is divided into different domains, including syntax, morphology, phonetics, semantics, and the lexicon. All these domains have independent locations in the mind, and these domains also interact with each other and with other cognitive domains, like pragmatics, discourse, etc. According to Sorace, the difficulty for adult learners to reach native language competence is due to the fact that these domains cannot interact properly. In the process of forming a sentence, different domains have to interact with each other. However, because of lack of automaticity, different domains cannot be integrated and coordinated with each other, which leads to problems.

One piece of evidence in support of the integration issue in adult L2 acquisition is problems caused by the third person singular -s and the past tense -ed for English learners like Chinese-speaking learners of English. In the past, people claim that Chinese-speaking learners of English can never acquire the third person singular marking and the past tense marking -ed because of the critical period. After the critical period, adult Chinese-speaking learners can never acquire the agreement marking, nor tense marking, because Chinese does not have agreement marking and tense marking. Nevertheless, according to a Mapping Hypothesis, Chinese people do have the knowledge of the third person singular marking and past tense marking. They know the past and they know the present. The problem is due to the fact that they are unable to map the agreement and the tense onto the verb with the required morpheme -ed or -s. So the problem is resulted from the mapping problem between morphology and syntax. If Chinese speaking learners of English are asked to fill the blanks, the accuracy is very high. But if Chinese people are asked to have simultaneous communication or natural communication in English, they will miss the third person singular -s and the past tense -ed very frequently. So, that means if they have sufficient time, they can connect morphology with syntax. However, if they are under time pressure, they will not be able to connect the two domains. L2 learners have processing difficulties in integrating and coordinating different types of information pertaining to different domains. This is a breakthrough in linguistic theory as well as in explaining difficulties for learners like Chinese-speaking or Japanese-speaking learners of English, where their first language does not have tense marking or agreement marking. This theory makes a significant contribution to the understanding of L2 acquisition, and that is why it is important.

张：Could you share with us some of the most interesting findings from your previous research regarding interfaces in SLA?

袁：One of my studies investigated English speakers’ second language acquisition of Chinese wh-topicalization at the syntax-discourse interface. The wh-word has to move to the sentence initial position in wh-questions in English, whereas wh-words normally stay in situ in Chinese. However, some of the Chinese sentences do allow the wh-word to move to the sentence initial position. For example, we can say (1), but we can’t say (2). However, in a certain situation, we can say (3) and we can also say (4). This shows that wh-words in this particular sentence can either stay in situ or can move to the sentence initial position. Then what is the difference between these four sentences? When people say “Shenme cai ni mei chi”, the phrase shenme cai “what dish(es)” is linked to a list of dishes known to the speaker and the listener. For example, we went out for dinner yesterday. After coming back, everyone got food poisoned except for you, then I asked you zuotian wanshang shenme cai ni mei chi “what dish(es) did you not eat yesterday”. In this situation, both the speaker and the listener know that there was a set of dishes on the table, so shenme cai “what dishes” is linked to the previous discourse. The above examples show that only the discourse-linked wh-word can appear in the sentence initial position, while non-discourselinked wh-words cannot be moved to the sentence initial position. In contrast, the wh-word is fronted to the sentence initial position in English wh-questions. For English-speaking learners of Chinese, making distinctions between discourse-linked wh-words and non-discourse-linked wh-words is a big challenge. My finding shows that English-speaking learners have difficulty in making decisions about which wh-word can be fronted and which wh-word cannot.

(1) “ 你 吃 什么?”

You eat what

What do you eat?

(2) * “ 什么你吃?”

What you eat

What do you eat?

(3) “ 什么 菜 你 没 吃?”

What dish(es) you didn’t eat

What dish(es) did you not eat?

(4) “ 你 没 吃 什么 菜?”

You didn’t eat what dish(es)

What dish(es) did you not eat?

张：The acquisition of new and additional grammatical features has been a hot topic for discussion. In your view, how are new and additional features acquired in adult second language acquisition?

袁：Whether a new feature or an additional feature can be acquired by second language learners or third language learners depends very much on the availability of positive evidence in the input. This involves the quantity of the positive evidence in the input, as well as the quality of the possible evidence in the input. By quantity, it means how frequent new features will be in the input, which can tell the learner that this is a new feature or an additional feature. By quality, it refers to whether this evidence in the input is obvious and salient and can tell the learner the feature is required for the target language. However, the positive evidence is not always robust and clear enough to send a signal to the learner. For example, when English-speaking learners learn Chinese, they can tell the plural meaning of the plural marking -men, but they cannot tell that -men also indicates definiteness from the phrases xueshengmen “the students” or laoshimen “the teachers”. This [definite] feature is hidden behind the plural marking, so this feature is not salient. If we do the analysis, the plural marking is very obvious. And this plural marking-men can only attach to human nouns, which is probably relatively obvious because they can't say chabeimen “the cups”, jisuanjimen “the computers”, or qichemen “the cars”. They can probably detect the [animate] feature, but it’s very difficult for learners to detect the feature of definiteness. How new features or additional features are acquired very much depends on the type of features, the availability of the features, and also the quantity and quality of the features available in the input. It's very complicated, but this is one of the focuses in my research.

张：According to the Feature Reassembly Hypothesis (FRH), learners need to map and reconfigure the linguistic features encoded in L1 onto L2. What is your view of feature reassembly and feature reconfiguration? And how does feature reassembly and feature reconfiguration work in second language acquisition?

袁：I like this Feature Reassembly Hypothesis. There are many theories arguing that learner’s difficulties are related to the unavailability of certain features in their first language. For example, the Failed Functional Features Hypothesis (FFFH) proposed by Hawkins and Chan (1997) argues that if a particular feature is not available in learners’ L1, this feature can no longer be acquired by the second language learners anymore. Coming back to the morpheme “-men”, English language has plural markings and also has thevfeature of definiteness. According to the FFFH, English-speaking learners of Chinese should not have problems of acquiring the feature of definiteness because this [definite] feature is available in English. However, according to the FRH, the difficulty is not whether the feature is available in learners’ L1 or not. Instead, the difficulty is related to the learners’ ability to reassemble these features. In English, the [definite] feature is mostly expressed in the article system, such as “the”. In Chinese, we don't have an article system. And the Chinese morpheme -men has this feature. So English-speaking learners learning Chinese have to reassemble the [definite] feature from the definite article “the” onto -men. According to FRH, this process leads to difficulty because it requires the learner to reconfigure this feature from one item in their L1 to another item in their L2. And this shows that the problem is not whether the feature is available in the L1 or not. The problem is whether learners are able to reassemble a particular feature from one item in their L1 onto another in the target language.

张：As you have mentioned, the availability of positive evidence in the input will significantly influence the acquisition of an additional feature. Your empirical studies have also identified other variables influencing successful second language acquisition, such as computational complexity and cross-linguistic influence. What implications do these findings have for teaching Chinese as a foreign language?

袁：Second language acquisition (SLA) research can inform the teaching profession by identifying what is difficult and what is easy, and it can also inform the teaching profession of what is transferred and what is not transferred. For the easy part, teachers do not have to spend precious time in the classroom and can save the time for those aspects that SLA research has found difficult for learners to acquire.Thus, findings in SLA can enable teachers to allocate the precious time in the classroom more sensibly and more economically. Moreover, SLA research findings can also inform professionals working in related areas, such as textbook writers, syllabus designers, language testing people, etc. Those findings an inform them of what needs more attention in their teaching, in their design of syllabus, and in their writing of textbooks and what areas do not require lots of attention in their writing. In that way, the teaching can be informed by the language acquisition research. And this is what is happening in English language teaching. In this sense, second language acquisition is a kind of basic research, just like other research fields. For example, in the field of life science, people try to find what genes lead to the lung cancer and the changes in what genes result in the breast cancer or the lung cancer. So people target the particular genes. Once they have found potential genes which may lead to cancers, their findings can help the clinicians to pay more attention to these genes. And that can also be used for pharmaceutical companies to invent targeted medicines. They form an academic chain or technique chain. In language research, we can also work to find what language features lead to difficulties and what languages features will not lead to difficulties. And that can inform language teachers, textbook writers, and syllabus designers. In this way, we can have a systematic chain of research. That will make our profession healthy, scientific, and systematic.

张：In recent years, you have also paid attention to third language acquisition. How did you become interested in this area?

袁：This is purely by coincidence. The world has become globalized. And with the cross-border mobility of people, not only do people speak two languages, more and more people can speak three or four languages. This has become a kind of eventuality, that is, people naturally learn a second, third, and forth language. For language acquisition research, we should not only look at learners’ first language as a potential source of transfer. We also have to look at what other languages learners have, in order to get a reliable and valid assessment of the learner. And in China, now universities and the Ministry of Education also encourage teachers to teach students not only a second language but also a third language. In what way we can identify possible influences from learners’ L1 or L2 in their L3 acquisition is a challenging question in our research. We cannot have a blind eye and pretend that participants in our research only have their mother tongue in their mind, and we should not pretend that the target language is learners’ second language. In Europe, many students have learned two languages before they come to learn Chinese. So we have to bear this in mind.

张：The source of transfer has been a focus in third language acquisition research. How do you identify the transfer effect? And what did your projects reveal about the transfer effect?

袁：As third language learners have already acquired two languages in their mind, transfer definitely occurs. But we have to find what is the source of the transfer. Does the transfer come from the first language or second language or both? Does the transfer take place in a wholesale manner? That is, they transfer the whole L1 grammar to L3 or transfer the whole grammar of L2 into L3. Or do they transfer the grammatical features piece by piece in a non-holistic manner? And whether the transfer only comes from the L1 which is the strongest and most stable language, or in L3 the transfer comes from L2 because both L2 and L3 are acquired by learners as adults, and they may be stored in the same location in their brain. The knowledge of L2 and L3 is most likely to be conscious knowledge, whereas L1 is most likely to be subconscious knowledge or implicit knowledge. Another influencing factor of language transfer is the typological distance between languages. Some people argue that transfer only occurs between languages that are typologically close and transfer is less likely to occur between languages that are typologically distant. For example, in our project, Cantonese is typologically close to Mandarin, whereas English is typologically distant from Mandarin. And in L3 Mandarin, transfer is more likely to take place from Cantonese into Mandarin than from English into Mandarin. So we also have to look at the typological distance when we investigate the transfer effect. These are the factors we have to bear in mind when investigating transfer in L3 acquisition.

Our results are very preliminary. Initial investigations have found that the typological distance plays an important role in language transfer. And the transfer does not seem to occur in a wholesale manner. It depends on individual linguistic phenomenon. L3 research is a relatively young area and is waiting to be explored. We want to find a model with which we can explain L3 phenomenon. We need to do more research and collect data to prove or disprove what we hypothesize. We're still working on it.

张：How do you see the future of language acquisition research?

袁：I think that the field of language acquisition has a very bright future. This field has become interdisciplinary. People will try to do research more precisely or more accurately by incorporating modern technology into language acquisition research. And particularly, people want to find the connection between language acquisition with human well-beings. For example, researchers are investigating the linguistic features of people with aphasia and autism. So language acquisition is not just about learning the language. People want to use the language learning to look at how the human brain operates and whether the language acquisition or language research can detect problems that people with autism or people with aphasia have. Again, the interdisciplinary approach to language research in combination with human well-being is the future and the right direction. This is very challenging, but this is a new field. The field of arts and humanities should become interdisciplinary. Researchers should not just follow the old fashion. Instead, they should open their arms and embrace new development and new advancement in different fields.

本文来源：《国际汉语教学学报》2021年第1期

感谢《国际汉语教学学报》授权发布！

关注“语言学心得”公众号

后台回复“袁博平”可获取

袁博平先生论文发表集锦文档