语言学termpaper：Discussion on the research hypothesis of second(2)

2.2.1 Sources of output hypothesis

Swain (1985) examined Canadian students who have studied French for seven years. She was through a survey of immersion teaching to find that the students’ French proficiency was not ideal and although they achieved a high level of listening and reading, they were unable to reach the native speakers' level in both oral and written terms. From her research, it is understood that comprehensive input is important in language acquisition, but it is not the only important, and it is not sufficient for learners to fully develop their L2 proficiency (Swain, 1985). Output also has its importance, in understanding input information, learners usually only need to know the meaning of the word without the need for syntax analysis. However, learners need to have a clear understanding of the rules of the language when carrying out the output of the language, so the output can force learners to pay more attention to the form of new language information. Output can effectively help learners to correct their mistakes by accepting the correct feedback information while testing their hypotheses about a target language (Swain, 1985; Romney, Romney and Menzies, 1995).

2.2.2 Contents of output hypothesis

Swain (1985) believed that output hypothesis has four functions for language acquisition, namely: noticing, hypothesis testing function, metalinguistic function and fluency function.

Noticing means that when learners express themselves, learners are more aware of their own language problems, and they will concentrate more on future language problems and solve relevant language problems so as to promote language acquisition (Swain, 1985).

Hypothesis testing function means that output provides learners with the opportunity to try to express themselves in various ways and to test whether these underlying hypotheses about the target language are correct (Swain, 1985).

Metalinguistic function refers to that L2 learners notice their problems in the language system through output, then making conscious analysis of the language form, and then outputting the corrected output to improve the accuracy of the language. The output allows learners to participate more in syntactic cognitive processing than in semantic cognitive processing, which is only involved in understanding (Swain, 1985; Branden, 1997).

Fluency function refers to that language output can improve the fluency of expression. The more frequently learners use language, the more possibly they gain the fluency of the language. Language output provides learners with the opportunity to use their language resources for meaningful drills, which allows learners to strengthen stored language knowledge, develop language processing automation and enhance the fluentness of presentation (Swain, 1985; Romney, Romney and Menzies, 1995).

2.2.3 Contributions and weakness of output hypothesis

The contribution of output theory first lies in not only recognizing the importance of teachers, but also affirming students' important position in L2 learning. Students do not passively accept teachers’ knowledge but take the initiative to practice (Branden, 1997). Second, it deepens students’ and teachers' understanding of L2 learning that students not only have to learn semantic knowledge, but also master the knowledge of the entire meta-language (Romney, Romney and Menzies, 1995). Finally, it provides a way for L2 learners to improve the level of L2 use, triggering a major shift in L2 education and learning (Romney, Romney and Menzies, 1995; Branden, 1997).

Of course, there are some shortcomings in the output hypothesis, which is mainly reflected in the following three aspects: First, output is only a way to improve L2 learning performance, not all students need to make use of a lot of output to raise the level, and research has shown that some students do not like to learn L2 through output to improve their grades, they can still get good grades without output. Second, without sufficient relevant knowledge and positive motivation for learning, a large number of inflexible output does not improve L2 learning performance. Finally, output hypothesis assumes that the output hypothesis helps students to find themselves having problems with linguistic knowledge during communication, but for most L2 learners, there is a lack of an adequate L2 communication environment, it's not easy for them to get high-quality feedback in communication to correct their mistakes in their own language.

2.3 Interaction hypothesis

2.3.1 Sources of interaction hypothesis

As what Krashen thought that, Long (1981) also believed that comprehensible input facilitates linguistic acquisition, and language environment facilitate the acquisition of comprehensible input. However, Long also held that language learning is communicatively driven, emphasizing the importance of interactive adjustment. Through an in-depth study on foreign accent, Long found that the study on foreign accent confuses two related but distinct linguistic phenomena: input and interaction, the former refers to linguistic forms such as words, sentences and articles, and the latter refers to the communicative function exercised by these language forms; interactive adjustment can better explain the role of input adjustment in SLA, which is the unity of language input and output (Long, 1983; Pica, 1987).

Long (1981) divided 48 adult native speakers and 16 adult non-native speakers of different linguistic and cultural backgrounds into 32 pairs, of which the 16 pairs of native speakers were in Group A and the other 16 pairs of native speakers and non-native speakers were in Group B. Each pair accomplished the same 6 tasks in the same order. Each pair exchanged 25 minutes and it was recorded. Long compared 5 input variables and 11 interactive variables of Group A and B. As a result, it was found that in foreign languages, the interaction adjustment was more frequent than the input adjustment. This shows that in foreign languages, interactive adjustments are more frequent than input adjustments. In foreign languages, two-way information exchange tasks provide more interactive adjustment than one-way information exchange tasks. Long (1983) reanalyzed the previous experiment and found that 15 specific interaction adjustment measures commonly used by native speakers from 6 tasks: remising control over topics, tolerating the freedom to switch topics, confirming whether self understands, tolerating ambiguities, etc.  

2.3.2 Contents of interaction hypothesis

Based on this, Long (1981) put forward the interactive hypothesis theory: when the communication comprehension is difficult, the two sides of conversation must make linguistic adjustments based on the feedback of whether the two sides understand, that is to say, meaningful negotiation makes the input comprehensible to promote learning. Meaning consultation refers to the mutual adjustment between learners to achieve communication, including input and output. For the purpose of negotiation, a variety of feedback methods can be used to prompt or correct the grammatical errors in discourse. These means have important meanings to learner's second language learning. In his later interactive hypothesis, Long (1983) adjusted earlier arguments to point out that meaningful negotiation, especially interactive adjustments made by native speaker or speaker with high verbal skills is conducive to language acquisition. Since then, Pica (1987) made further research on interaction hypothesis. It is found that learners constantly adjust language output during interaction, which helps to understand language and in turn promotes language output. Language input and language output is precisely connected in the process of interaction, and ultimately achieving the internalization of language knowledge.

2.3.3 Contributions and weakness of interaction hypothesis

The significance of interactive hypothesis lies in that it expounds the importance of communication in second language learning. It not only shows the positive influence of communication on students' second language input and output, but also affirms the impact of second language input and output of high quality on successful communication, which is very useful for second language learning and education (Pica, 1987). It is relative to input hypothesis that communicative hypothesis does not neglect a students' own importance to second language learning, including their own motivation, ability to learn, motivation to learn and so on (Pica, Young and Doughty, 1987).

There are two main shortcomings in the interaction hypothesis. First was that those non-native speakers in his experiment lived in a native language environment. In other words, they themselves have a lot of exercise opportunities to get in touch with and familiar with their own language. However, the vast majority of second language learners in the world does not have such good conditions to guarantee high-quality communication to promote second language learning. Second, the birth of interactive hypothesis is mainly aimed at the ability of oral English, whether it can still play the same effect on the ability of other aspects of English still requires further in-depth observation.

2.4 Discussion 

Input hypothesis, output hypothesis and interaction hypothesis reveal the law of L2 learning and education from different perspectives, and they provide scientific theoretical guidance for the L2 learning and education methods. The three hypotheses play an important role in promoting the overall development of L2 acquisition, but it is noteworthy that the three hypotheses currently have the following problems. First of all, each hypothesis corresponds to different educational methods and strategies (Pica, Young and Doughty, 1987), but different hypotheses conflict logically, such as input hypothesis and output hypothesis, for teachers and education authorities, to apply these theories to specific teaching practices is a question that is worth of considering. Second, the birth and beginning of the study of each hypothesis is aimed at specific groups, but on a world-wide basis, different L2 learners face totally different teaching environment, personal English proficiency, motivation and purpose of learning (Pica, 1987), whether these hypotheses can be applied to each L2 learning group needs more substantive research to support. Finally, in the L2 teaching and learning process, affected by a variety of factors, what kind of roles the three hypotheses will play in the process, if they all work or there cooperate to play the roles, which hypothesis takes effect faster, these problems need to be solved in order to carry out more target teaching and learning strategies, but there is no corresponding empirical study to answer the questions.