The Critiques Of Krashens

Published Date: 02 Nov 2017

Some researchers don’t agree with Krashen about the Input Hypothesis . They mainly thinks that it is too simple and it can be tested, and it fails to define the concept of "comprehensible input"and overstates the importance of it, etc. (Zou Weicheng, 2000; Jing Zenglin, 1991). Psychologists like McLaughlin (1987) sharply criticizes Krashen’s Input Hypothesis and holds that what Krashen does in the Input Hypothesis does not provide evidence in any real sense of the term, but simply argue that certain phenomena can be viewed from the perspective of Krashen’s theory.

Some researchers also holds that Krashen does not directly tells us what is comprehensible input. Ioup (1984) points out, "it is very hard to define the levels and determine which structures constitute i+1 level and therefore people usually can’t test the hypothesis" (qtd.in McLaughlin, 1987: 39). Larsen-Freeman and Long (1991:247) also claim that the Input Hypothesis is hard to test and they hold that this hypothesis "contains vital constructs, i and i+1, which Krashen (1984) himself recognizes are can’t operate in our daily teaching , given the state of knowledge in interlanguage studies". Gass (1988) points out that the comprehended input is more important than the comprehensible input ,and is the most important (qtd. In Ellis, 1994: 278).

In the opinion of Mclaughlin (1987), the way of comprehensible input does not the only route to acquisition. Children do profit in their language development from interacting with native speakers whose language is well formed. Heath (1983) reports that black children in the working-class community in what she studied are ignored as conversation partners until they become information givers so she says, "Children learn to speak the language by taking in and imitating the sounds what they hear around them. Furthermore, the language what they heard is beyond their current level of competence. In many societies, parents and other caretakers do not use simpler language while talking to young children; simplification is viewed as an inappropriate speech behavior. These children seem to learn without simplified input and without comprehensible input." (qtd. In McLaughlin 1987: 44). White (1987) argues, "simplified input may in fact deprive the learner of information vital to acquisition, an argument anticipated by language teachers who advocate the use of authentic speech in the classroom to fill gaps in the normal edited classroom language" (qtd. in Cook 2000a: 61).

On the claim that comprehensible input is necessary for acquisition, White (1987) has argued that a considerable part of acquisition is `input-free'. And she (1987:102) proposes, "The driving force for grammar change is that input is incomprehensible, rather than comprehensible". White's idea is that failure to understand a sentence may force the learners to pay closer attention to its syntactical properties in order to obtain clues about its meaning (qtd. in Ellis, 1994: 279). But the author holds that the Krashen’s comprehensible input and White’s incomprehensible input only see input from different points of view. No matter from what point of view, the learners still progress continually from level i to level i+1 along a series of stages. As Ellis (1994: 279) sees it, these criticisms can be accommodated if the hypothesis is modified in this way, "Comprehensible input can facilitate acquisition but it (1) is not a necessary condition of acquisition, and (2) does not guarantee that acquisition will take place".

Speaking of method comparison research, Krashen claims that the comprehension-based teaching methods have advantages over traditional audiolingual methods. But McLaughlin (1987) believes that Krashen’s dismissal of grammar teaching is lacking of thinking and the role of grammar-based teaching still deserves empirical scrutiny. But the author does not think Krashen is against grammar teaching. Because he mentions, "note that the theory does not predict that comprehensible-input methods will be superior to grammar-oriented methods on all counts" (Krashen, 1985: 15). Krashen (1982) just holds that an approach which provides substantial quantities of comprehensible input will do much better than any of the older approaches. So he does not mean other methods ignore comprehensible input.

Krashen has argued that the part which the speaker’s output plays is providing a further source of comprehensible input. Swain's (1985) comprehensible output hypothesis claims that learners need the opportunity for meaningful use of their linguistic resources to achieve full grammatical competence. Comprehension of a message can take place with little syntactic analysis of input, production forces learners to pay attention to the means of expression. Other researchers such as Long (1985) stress the importance of "negotiating meaning" to ensure the language is modified to the level the speaker can manage (qtd. in Ellis, 1994: 273). Unless learners try out the language, they are unlikely to get the kind of feedback they need to analyze the structure of the language.

2.3.2 The Advantages of Krashen’s Input Hypothesis

It’s very obvious , which is the deficiency of Krashen’s theory. In the theory, his hypothesis does not have a strong research foundation and it cannot be regarded as a good theory without overall research support. After reading the criticisms, one might hold Krashen’s Input Hypothesis is not useful at all . But it still has some advantages in guiding foreign language teaching. In fact, theories can be divided into two categories. One comes from researches and facts. During the study of the facts, the theorists find the law and gradually perfect it by examining and modifying it for many times. The other is to present the hypothesis first and then the theorists try to find the supports and arguments in the researches and studies. And the hypothesis is proved or modified until being accepted as scientific. Krashen’s theory belongs to the second one. He puts forward his hypothesis first, and then tries to use his theory to interpret the phenomena in SLA. So his theory is criticized for lacking of research arguments. But the novel hypothesis can enlighten people to study further. The criticisms and discussions about Krashen’s theory will stimulate others to perfect the theory or develop better ones. And if a research can inspire a controversy, it surely will facilitate related studies and the improvement of the theory.

The Hypothesis of Input has received many attacks ,but it also attracts many people dealing with L2 learning, particularly teachers. Ellis (1990) talks of "the lucidity, simplicity, and explanatory power of Krashen’s theory" and Lightbown (1984) praises its combination of a "linguistic theory, social psychological theory, psychological learning theory, discourse analysis and sociolinguistic theory" (qtd. in Cook, 2000a: 66). Cook (2000a: 66) sees that older methods ignore comprehensible input are incorrect and she states, "Comprehensible input has been the core of many teaching methods, in spirit if not in name". Therefore the Input Hypothesis enjoys great popularity at its emergence and many researchers and language teachers accept it in no time. Brown (1994: 89) sees that although Krashen’s theories are oversimplified, "we owe a debt of gratitude to Krashen for his bold, if brash insights. They have spurred many a research to look very carefully at what we do know, what the research evidence is, and then in the process of refutation to propose plausible alternatives". Ellis (1994) holds the same opinion and sees that the Input Hypothesis has motivated a substantial amount of researches. Nunan (2004:76) states, "to this day they (Krashen’s hypotheses) remain popular, widely cited and infuential, particularly in North America". In fact, Krashen’s Input Hypothesis still has its strengths which are significant for the study of SLA and which have great implications for language teaching. We can see them from the following three aspects:

(1) Comprehensible input is of great significance to teaching. Krashen cites caretaker speech as evidence to support this idea and he (1985: 5) states, "...caretaker speech, while `simplified' in several ways, is intended for communication...The fact that caretaker speech tends to be limited to the `here and now' aids comprehension, supplying the extra-linguistic context that helps the child in decoding the message". As the child develops linguistically, caretaker speech tends to get more complex. The simplified speech will be helpful when it provides the acquirer with i+1 in a context that makes the message comprehensible. Nevertheless, children and adults have many differences in acquiring a second language. The children are generally superior in second language attainment in the long run, but adults acquire at a faster rate. In addition, older children acquire faster than young children do. Krashen’s (1985)explanation is that older acquirers progress quicker in early stages because they obtain more comprehensible input, while younger ones do better in the long run because of their affective filters. The older acquirers obtain more comprehensible input because of their greater experience and knowledge of the world, their strategy of falling back on first language syntactic rules and their superior skills in conversational management.

(2) Input is primary and output is secondary

Many researchers proved that when there is enough input, production ability can emerge without any output practice. It has often been noted that children acquiring a second language in a natural, informal linguistic environment may say very little for several months following their first exposure to the second language. The child is building up competence in the second language via listening and by understanding the language around him. Speaking ability emerges on its own after enough competence has been developed by listening and understanding. In China, many self-taught language learners often develop their language competence by reading a lot ; they can reach a higher level of writing without training in language class.

On the other hand, no evidence proves that language competence can be developed by production practice without further input. Suppose a group of six-year-old children would be sent to a desolate and uninhabited island to live for ten years and they are allowed to talk to each other. It is hard to imagine that they can develop to the same language competence as the adults without any input from the adults. Language competence can be developed by input even without production practice, but production practice itself cannot. So we can draw the conclusion that input is primary and output is secondary. This is what Krashen’s input hypothesis tries to tell us. Thus in second language class, more emphasis should be put on listening and reading. Only when there is enough comprehensible input, can output emerges and helps to develop language competence.

(3) The affective filter is an important factor that second language teachers should consider in teaching, because learners are different in level of affective filter. Those with a high or strong affective filter will accept less input because the input will not reach that part of the brain responsible for language acquisition. Those with a lower or weaker affective filter will be more open to the input. Sometimes the reason of why learners of about the same level, same age in the same English class will reach different levels of language competence lies in different levels of affective filter. So a successful language class should be one that provides situations encouraging a lower affective filter, a successful language teacher should be someone who can conduct such a class.

Input Hypothesis still has its values in SLA learning and teaching. Using it properly, we can achieve high learning efficiency.

2.4 Conclusion on the Importance of Input Hypothesis

Although the Input Hypothesis has received many critiques, it is still useful to guide our foreign language teaching and its values should not be ignored. A number of researches see comprehensible input as a major causative factor in second language acquisition. Strong proposals have been put forth about the role of input in SLA. For instance, Krashen (1982: 35) calls comprehensible input in the presence of a low affective filter "the only causal variable in SLA". Ellis (1994: 26) thinks, "it is self-evident that L2 acquisition can only take place when the learner has access to input in the L2". White (1989: 37) also believes that "L2 learners are faced with the problem of making sense of input data, of coming up with a system which will account for that data, and which will allow them to understand and produce structures of the L2". They concur that input is essential to SLA, which means that learners must be exposed to samples of language that are used to communicate information. Therefore, according to the above analyses, the conclusion can be drawn that the role of input is essential.

In China, some scholars like Hu Wenzhong (1984), Zhu Zhizhong (1988), Zhang Jianzhong (1990), Jing Zenglin (1991), Hu Zhuanglin (1994), etc. have shown their opinions about Krashen’s theory in foreign language teaching journals. Many researchers in foreign language teaching studies in China agree with Krashen on basic assumptions, such as the need to turn from grammar-based to communicatively oriented language instruction and to increasing the amount of foreign language input (Hu Wenzhong （1984）, Zhang Jianzhong （1990）, Wu Dinge （1990）, Cheng Enhong（ 1988）, Ding Guocheng（1993）, Zhu Zhizhong（1988）, Zhuang Enping and Zhuang Enzhong（1990). And the role of affective factors in language learning is also studied. Wang Dawei (1997) puts input theory into practice by conducting some experiments. Inspired by Wang's empirical success and his students' learning improvement, Wu Peng (2001), adapting Krashen’s input theory to the Chinese learning context, conducts a teaching experiment. His experimental class in this program outperforms the control class in comprehensive English ability greatly. Then the input theory begins to be combined flexibly with other theories on foreign/second language teaching. The typical combinations are those of input and output (You Qida, 1997)&(Deng Yongzhong, 2000), input and emotional barriers (Wu Dinge, 2000), input and constructivism ( Pengpeng, 2001), etc.

Though a certain number of studies about input have been undertaken in China, most of them are about the role of input in English learning, few studies are about how well students make use of the input available. This dissertation tries to investigate and analyze the situation of input in EFL classroom teaching in China. On the basis of Krashen' s Input Hypothesis, the dissertation gives out some suggestions on what to input and how to input in our current English class.

2.5 Output Hypothesis

Krashen highly emphasizes the importance of input and depreciates the role of output, while it is not quite agreed by some other linguists who view learners’ output as contributing to second language development. They hold that output is very important and indispensable in language acquisition. Merrill Swai’s Output Hypothesis is one of the most influential theories of this school. Swain is not quite satisfied with Kashen's Hypothesis and argued that comprehensible output is important for language acquisition. After many researches she generalized her ideas and Put forward the Output Hypothesis.

2.5.1 Swain's comprehensible Output Hypothesis

Steven Krashen emphasizes the contribution of comprehensible input to the success of second language acquisition, but denies the importance of comprehensible output in second language acquisition. Swain(1985) put forward the output hypothesis, as a complement to the comprehensible input hypothesis, in which she points out that for successful second language acquisition, comprehensible input may not be sufficient, opportunities comprehensible output are foreign language learners to produce necessary.

Comprehensible output hypothesis is based on Swain's immersion study on problems of Canadians who learning French in Toronto in the 1980's. Swain observed that learners get too few opportunities to use the target language (French). They spent too much time on listening to the teacher, but too little on reading, writing and speaking. What's more, there was an exceedingly high tolerance of errors in class, and teachers often failed to correct even the most glaring pronunciation and grammar errors of advanced pupils. Therefore, the immersion programs, although content-based and communicative, failed to produce successful second language users with target-like accuracy.

Swain formulated the comprehensible output hypothesis based on the observations. She claims that "comprehensible input", although invaluable to the language acquisition process, is not enough for language learners to fully develop their second language proficiency. It is should be pointed out that the comprehensible output hypothesis by no means negates the importance of comprehensible input hypothesis. She acknowledges the role of comprehensible input in second language acquisition, but argues that comprehensible output is also a necessary part that aids second language acquisition in many ways which cannot be replaced by input. The intention of the hypothesis is to complement and reinforce, rather than replace. She claims that both the two hypothesis are necessary for successful language acquisition. Swain emphasized specifically, that if language learners are to be both fluent and accurate in the target language, they need both obtaining comprehensible input and producing comprehensible output, and opportunities for meaningful product of their linguistic materials to achieve full grammatical competence are also needed. Comprehensible output hypothesis claims that production will aid acquisition only when the learner is pushed. To guarantee language acquisition, only providing learners with opportunities of speaking and communicating is not enough. According to Swain, in producing the target language, learners will become aware of a linguistic problem on occasion, either externally or internally attributable. It can push learners to modify their output. That is to say, activities of producing the target language may prompt second language learners to consciously recognize some of their own linguistic problems, bringing their attention to something that they need to discover about their target language. That means learners notice a problem and then they conduct an analysis leading to modified output, when learners cannot work out a solution, they may turn to input, with more focused attention, searching for relevant input datum, and latter, successful language acquisition may be received.

After the proposition of her comprehensible output hypothesis, Swain(1995) furthers her studies and has elaborated three functions of comprehensible output in second language acquisition which involves conscious cognitive process. The first function is the "noticing/triggering"function or what might be referred to as its consciousness-raising role.

Swain claims that when learners use their target language to achieve certain communicative goals, they will find that there are some linguistic problems that may hinder their output producing, even their communication. Thus, a linguistic gap comes into being. The second one is the "hypothesis-testing" function, which means that learners have formulated hypotheses about the target language and try to test them. The third one is the "meta-linguistic" function or what might be referred to as its reflective role, which is also referred to as "conscious reflection" by which it means the learners can use the target language to talk about or think about it. The three functions proposed by Swain can facilitate second language acquisition. She claims that output activity provides chances for the active employment of the learner's limited cognitive resources. Language production may not only result in consolidation of existing knowledge, but may also contribute to restructuring by pushing learners to test hypotheses about the second language in international and collaborative contexts. In the western second language acquisition literature, there are lots of researches conducted on language interaction and language production.

These researches provide ample evidence for the fact that learners do notice problems as they produce the target language, and they do try to do something to solve them. Several experimental studies have been done to test the functions of output. Swain and Lapkin(1995) investigated the noticing function of output. By studing language learners’think-aloud protocols; they looked at their composing topic-given essays. The study results supported the argument that learners would notice a linguistic problem on occasion through the think-aloud protocols process.

According to Swain and Lapkin, 40% of the language-related episodes focused on grammatical forms. In doing so, it helps learners to move from semantic to grammatical processing. Based on this research, Swain and Lapkin proved that output not only led to noticing of the gaps in interlanguage knowledge but also facilitated language acquisition by triggering various internal processes conductive to second language acquisition.

2.5.2 The Roles of Comprehensible Output

Krashen emphasizes that language can be acquired simply by comprehensible input and minimized the role of output in the acquisition (i.e., output is just a means for receiving more comprehensible input.) Krashen summarized his position as follows: "Comprehensible input is responsible for progress in language acquisition. Output is possible as a result of acquired competence. When performers speak, they encourage input (people speak to him).That is conversation."(Krashen, 1983:61)

However, the results of the studies conducted by Swain in Canadian (Swain, 1984, qtd inSwain&Lapkin, 1995:372; Swain, 1985) present us with strong counter evidence against the"acquisition-without-output" statement. The students in the program were educated in their second language-French from the very beginning of school, i.e. kindergarten; however, in the speaking and writing tests, these students can be easily identified as non-native speakers and writers by the end of the elementary school. Swain's interest in output arouses from this phenomenon, which has led her to the conclusion that besides comprehension, production has a significant role to play in one's second language learning. Swain(1985) advances the output hypothesis, proposing that through producing language, either spoken or written, language acquisition/learning may occur (Swain, 1985, 1993, 1995).

Here, by output, Swain refers to comprehensible output which means the need for a learner to be" pushed toward the delivery of a message that is not only conveyed, but that is conveyed precisely, coherently, and appropriately"(Swain,1985:249)Furthermore, as to in what way output works in SLA, Swain(1995)puts forward three functions of output:

1.The"noticing/triggering"function or what might be referred to as its consciousness-raising role

2. The "Hypothesis-testing"function 11

3 .The "meta-linguistic" function or what might be referred to as its reflective role.

2.5.3 Other Studies on output Hypothesis

Several qualitative studies have been conducted to support Swain' s Output Hypothesis since its proposition in 1985. Pica, Holliday, Levis，and Morgenthaler(1989)，for example，found that in response to requests for clarification or confirmation，learners tended to modify their output. According to the Output Hypothesis，such modifications contribute to the process of second language acquisition. Nobuyoshi and Ellis(1993) conducted an exploratory study involving six learners(three experimental and three under control)in a two-way information gap task, seeking to study whether clarification requests resulted in subsequent accuracy of learners，output and whether pushed out resulted in improved performance over time. Experimental subjects experienced "focused meaning negotiation" and received a clarification requests from the researcher each time they produced an utterance that contained a past tense error while the control subjects did not. As a result of the requests for clarification two of the three experimental subjects showed improvement in the accuracy of their use of the past tense and maintained this improvement one week later, while no improvement was shown by any of the control subject. This study found evidence which suggests that pushing learners to produce output leads to better acquisition and it may have a long-term effect, but not necessarily for all learners. It is possible that when learners produce output they are able to develop greater control over the features they have demonstrated in Nobuyoshi and Ellis’study. It is not clear whether pushed output can result in the acquisition of new linguistic features.

Swain and Lapkin(1995)，conducted a study involving 18 students from an eighth grade early French immersion class who were trained to think aloud while writing an article for a newspaper and found that, as they were encouraged to produce their second language, learners did recognize problems during production and notice gaps in their linguistic knowledge. They also found that recognition of problems often triggers cognitive processes that have been implicated in second language learning: processes that either generate linguistic knowledge that is new for the learners or consolidate their existing knowledge. Based on these findings, Swain and Lapkin argue that output not only leads to noticing of the gaps in second language knowledge, which plays a consciousness-raising function, but also facilitates language acquisition by triggering various internal cognitive processes conducive to second language acquisition.

Izumi(1999，2000) attempted to investigate the question of whether the awareness of Problems during Production can prompt learners to notice relevant features with more focused attention if input is subsequently provided to them. Learners were provided opportunities to receive relevant input to see whether they would notice and learn the targeted features in the input. In his study, Izumi compared two groups of ESL students regarding their learning of a grammatical structure in English. The experimental group produced written output: essay-writing tasks and text reconstruction tasks, and then was presented with relevant input in the form of short reading passages: The comparison group was exposed to the same input for the Purpose of comprehension only. The studies have three major findings as follows: First, the experimental group and the comparison group demonstrated increased noticing of the target form，as measured by participant's underlining as they read the input passages. Second, the significant rate of uptake of experimental group demonstrated a the target from in their production immediately following exposure to the input. Third, the experimental group made significantly larger gains on the production tests after the treatment than did the comparison group. These results lend some support to the noticing function of output in second language acquisition. Gass(1985) discussed the possible ways in which output may serve important language learning purposes:(a)testing hypothesis about the structures and feedback for meaning of the target languages,(b)receiving crucial the confirmation of these hypotheses,(c)developing automatically in learners’ interlingua system, and (d)forcing a shift from more meaning-based process of the L2to a syntactic mode.

In 1997, Susan M. Gass combined and further developed Klashen’s input Hypothesis and Swain’s Output Hypothesis in her study, thus deepened the understanding of the theory and also applied it to wider fields. According to Gass，there are five stages that account for the conversion of input to output :apperceived input, comprehended input, intake, integration, and output. Once the prior knowledge get apperceived, then it goes through the stages one by one till giving output after the integration stage，and gets tested and finishes one cycle. It is a dynamic learning system with each stage as an integral.

2.5.3 Researches on Output Hypothesis in China

Since Swain put forward the Output Hypothesis, some Chinese researchers and teachers are interested in it. They constantly publish articles to introduce this theory to China. The earlier studies are mostly introduction of this theory and the discussion of the roles of input and output and its inspiration to our foreign language teaching, such as ZhaoPing(2000), Lu Renshun(2002) and ZhaoPei(2004). ZhaoPing(2000)discusses the two roles of output(noticing and hypothesis- testing) in the process of writing. Lu Renshun(2002)introduces the researches done on the Output Hypothesis and the functions of output and put forward several implications of researches for China's English teaching. ZhaoPei(2004) thinks that undue emphasis on either input or output is not proper and only through the close combination of the two can improve learners' foreign language learning.

There are also some experimental studies on the influence of production-based instruction on learners, production abilities. WangYing(2003) addresses the functions of output entraining learners’ language intuitions. The study by FengJiyuan and HuangJiao(2004) closely followed the experimental procedure of the studies done by Izumi(2000), making only few modifications. The findings of this study consistent with those of Izumi et a1.(2000) proved the effectiveness of output practice in helping learners acquire linguistic forms. The research done by Wang Chuming(2000)shows that composition-writing can improve learners’ English production ability. WuFei(2005) testifies the positive effect of comprehensible output on the efficiency of language learning, which conforms to Wang Chuming’s proposal. Zhou Dandan(2006) conducted a research on the effects of different input and output frequencies. Li Hong(2002} put forward her doubt on the noticing function of the output hypothesis. In her opinion, both language meaning and form compete for learners’ limited cognitive resources，so the premise for output to play a role in language acquisition is that learners have enough cognitive resources. She thought that output hypothecs is neglected this problem.

2.6 Longs Interaction Hypothesis

According to Krashen, SLA depends completely on comprehensible input. However, evidence shows that "comprehensible input alone is insufficient, particularly with adults and if native-like proficiency is the goal" (Long, 1996). Based on the study of Krashen's. Input Hypothesis, Long proposed the Interaction Hypothesis in 1983.There are two main claims in the Interaction Hypothesis. One is the negotiation of meaning and the other is the interactive modification.

2.6.1 The Negotiation of Meaning

The "Interaction Hypothesis" not only highlights the importance of comprehensible input in language acquisition, but also emphasizes the importance of the negotiation of meaning in it. Long argues that input that has not been comprehended may become comprehensible through the process of negotiation. Strategies used by the speakers to avoid or overcome difficulties in interaction such as repetition, reformulation, confirmation check, comprehension check, clarification request and so on are called negotiation of meaning (Gass, 1997). Ellis defines negotiation of meaning as "the interactive work that takes place between speakers when some misunderstanding occurs". (Rod Ellis, 1997:141) Negotiation of meaning refers to the conversational exchanges that take place when interlocutors tryto avoid communicative impasse exchanges. It is a process of preventing and modifying the gap of understanding, which brings interactive comprehensible input. And at the same time, it offers opportunities for output. The negotiation of meaning has the following two results. First, it keeps the interaction going on and gets things done. Second, more importantly, it makes the learners not only pay attention to the language form of their own but that of their encounters'. By making a comparison between the language forms, learners will notice the gap between their interlanguage and the target language or the more target-like language. The following example may well illustrate the reasons why the negotiation of meaning can make the comprehensible input more effective.

Example 1:

Hiroko: A man is uh drinking c-coffee or tea with uh the saucer of the uh coffee set uh in

his knee.

Izumi: In his knee?

Hiroko: Uh on his knee.

Izumi: Yeah.

Hiroko: On his knee.

Izumi: So sorry. On his knee. (S.Gass &E.Varonis, 1994:283-302)

From the interaction between Hiroko and Izumi, we can see that learners often receive negative feedback. That is, their interlocutors suggest that they have not understood and, in the course of so doing, may present the correct target language forms. Consequently, learners can receive input relevant to those aspects of grammar that they have not yet fully mastered. Finally, their communication in second language will be assisted by interaction.

2.6.2 The Interactive Modification

According to Long, in order to have a successful communication, both of the interactive parties should "negotiate" meaning in conversation. Negotiation that involves the restructuring and modifying interaction may occur when the interactive parties have to cooperate to get comprehensibility by repeating a message word by word, adjusting its syntax, changing its words, or modifying its form and meaning and many other ways until what he or she said is comprehensible. This process of communication was described by Long (1983) as Interactive Modification. The interactive modification assists learners in avoiding non-comprehension in communication and making conversation going on. According to Long (1996), interactive modification was most probably to facilitate input comprehension and language acquisition. He summarized the interactive modification to the following three kinds:â‘ Comprehension check. It means that the speaker asks whether the interlocutor understands what he or she said or not. He or she may ask questions such as "Are you clear?" "Do you follow me?" "I was born in Nagasaki. Do you know Nagasaki?" etc.â‘¡Confirmation check. It refers to the fact that the speaker checks whether he or she has understood the real meaning of the interlocutor or not. He or she may ask questions such as "Is this what you mean?" "When can you visit me? Visit?" etc.â‘¢Clarification request. It means that the speaker asks the interlocutor to provide more information or help him or her understand what the interlocutor said before. He or she may ask questions such as "What?" "Huh?" "Research? I don't know the meaning." "Sorry what did you say?" etc.

2.7 Limitations of previous studies and starting point of the current study

From the above discussion we can see that the findings of these studies have shown that activities of producing the target language may, under certain circumstances, promote the process of second language acquisition. Although the previous studies of output theory have been insightful, there still leaves some room for the present study.

Firstly, the existing researches on output are mostly descriptive in nature, focusing primarily on describing the output of the subjects rather than acquisition or whether and how output can be a source of competence in second language acquisition (Shehadeh, 2002). These studies do not relate directly the output of the subjects to the acquisition of the linguistic forms. Two notable exceptions are the study done by Izumi, et al.(1999, 2000). They on the other hand, examined whether the awareness of problems during language production would prompt second language learners to find out subsequent input with more focused attention and lead to the noticing and learning of a specific grammatical form. However, their target forms to be taught are always past hypothetical conditional in English.

Secondly, foreign experimental researches on the functions of output fail to take into consideration the environmental differences between second language and foreign language teaching. Their experimental participants are always the English language learners in English as a second language, so the findings in western countries may not be very helpful for the English classroom teaching in China, for in China English is taught as a foreign language.

Thirdly, at home, Zhao's (2000) study on Swain's comprehensible output hypothesis in the English writing class was conducted empirically. It was primarily to test and verify that output did work its functions in subjects' improvement of college English writing. Study done by Feng and Huang (2004) was designed to measure the effectiveness of output practice in helping learners acquire linguistic forms. And the target forms are also past hypothetical conditional in English. Moreover, the study by them closely followed most experimental procedures of the studies done by Izumi, etal.(1999), making only few modification.

After taking a retrospection of the achievements reported in this field the author has not found anyone who has ever tried to do any experimental research on the functions of output in improving learners’overall foreign language proficiency. Therefore, the author attempts to perform an experimental study to further explore the utility of output in promoting foreign language acquisition by working out a productive-skills developing model of English as a foreign language teaching with the hope of yielding some useful and empirical evidence to Swain's output hypothesis and making some contribution to enhance output theory in second language acquisition as well as shedding some light on English as a second language teaching and learning in China.

2.8 Summary

In this part, the theoretical foundations of the current study, the Input Hypothesis, the Output Hypothesis and the Interaction Hypothesis are clearly illustrated. Through the illustration of the three hypotheses, it is apparent that both the Output Hypothesis and the Interaction Hypothesis are put forward in an attempt to modify the Input Hypothesis as there are some shortcomings in this theory. So both of them are the results of the development of the Input Hypothesis, not its substitutions. The relationship between them is very complex and close. The combination of the three hypotheses as a whole can provide comprehensive and solid theoretical foundation for the study of classroom interaction. If any hypothesis of the three is not included, the study will not be intact.