1.

INTRODUCTION
It has long been assumed both by teachers and students of English as a Second Language (ESL) or English as a Foreign Language (EFL) that one can learn a second language (L2) effectively by using it productively as soon as possible. The more active the students are in oral practice, the faster, it is believed, they will learn L2. Consequently, an ESL/EFL teacher often demands that students talk right away and expends great effort to make them repeat after his model from the first day of the class. Observations and studies of children's second-language acquisition (see Krashen 1985) have revealed that in the initial phase of the language acquisition process, there is typically a 'silent period' during which children acquiring a new language in natural settings are silent and concentrate on comprehension. And they may respond, if necessary, only in a non-verbal way or by making use of a set of memorized phrases. This phenomenon is also observed when we see how children acquire their mother tongue. A baby spends many months listening to the people around it long before it ever says a word. Since comprehension always comes before production in a natural process of language acquisition, a number of ESL/EFL methodologists and teachers are beginning to pay attention to this phenomenon. They hypothesize that the existence of a 'silent period' in ESL/EFL classrooms should be of great benefit in facilitating the acquisition of L2, in listening comprehension and also in other skills such as speaking, writing and reading. A number of experiments were conducted to test a 'silent period' hypothesis and results reported seem to constitute arguments in favor of a 'silent period' in initial stages of L2 learning even in the formal environment. There are also several researchers who have developed teaching strategies based on a 'silent period' hypothesis. The purpose of this paper is to search through the literature concerning such experiments and researches and to consider the possibility of introducing such strategies in ESL/EFL classrooms.

2. EXPERIMENTS AND RESEARCHES

2.1 The Total Physical Response In Asher's Total Physical Response (see Asher 1965, 1966, 1967, 1969, 1972, 1982 and also see Kunihira and Asher 1965), students listen to commands in a target language and then immediately respond along with the instructor with an appropriate physical action. When those commands become familiar, the teacher remains seated and only the students will continue to respond to the teacher's commands. From time to time, some novel utterances (unfamiliar utterances created by recombining familiar elements) will be given. The students do not have to respond verbally until they are ready. Asher (1982) claims that most of the grammatical features in a language can be nested in the imperative and that almost any grammatical constituent can be taught through the skillful use of the imperative. Asher's experiments show that experimental subjects are superior to the control subjects not only in the listening skill but also in both pronunciation and control of grammar. Asher added (1972:138) "Perhaps even more striking than the high level of listening skill is the transfer-of-learning to reading...." According to his report, after 32

hours of training in German, the adults who ranged in age from 17 to 60 and met twice a week for two hours per evening outperformed the control subjects not only in listening skills but also reading. The control subjects were finishing their advanced college course in German which took more than 80 hours. The experiments also show that the experimental subject has attained a long-term retention of the learned items. One of the explanations Asher (1982) gives is that this procedure may help facilitate long-term recall for the same reason that the practice of any manual skill such as ice skating, bicycling or swimming produces long-term recall. 2.2 Postovsky's Delayed Oral Response Approach In this method (1974, 1977), Postovsky makes use of writing as a meaningful, nonverbal response mode in teaching Russian. "The student hears a foreign utterance and sees its transcription simultaneously and then writes a foreign word in response to an auditory stimulus." (Postovsky 1974:231) Written responses were used as a substitute for oral practice. In this experiment, too, students develop better overall language proficiency when oral practice is delayed in the initial phase of instruction. 2.3 The Comprehension Method Winitz and Reeds (1973) make use of a machine called TAPAC (Totally Automated Psychological Assessment Console), which provides students with auditory input and about two seconds later four projected pictures. Students, then choose one picture which matches the auditory input by touching the panel. The machine responds to each answer either by saying 'correct' in a light and pleasant tone or by unpleasant buzzer sounds along with an 'incorrect' sign on the panel. In this method, speaking is discouraged until a high degree of comprehension is achieved. The program is only made up of basic structures until a considerable number of vocabulary items (about 3000) are learned. The students are never given the rules of the language, but they problem-solve to arrive at the rules. Winitz (1981) conducted an experiment to compare his comprehension method with three other orally active training procedures; such as imitation and two types of paraphrasing, elaboration with contrasts present, and elaboration without contrasts present. The results of the experiment confirm the previous studies. He says that 'production' practice is not essential to improvement in production, as the comprehension group achieved the same degree of improvement without any practice in production. 2.4 The Natural Approach Terrell (1977) suggests in his Natural Approach (NA) that in order to enable students to communicate meaningful contents from the first day of class, they should be permitted to respond in Ll. Although the target language is stressed, when necessary, teachers can use the students' native language. In this way students can rapidly expand their listening comprehension abilities to a wide variety of topics. In his updated research (1982) and Krashen and Terrell (1983), he expands his original proposals and suggests specific teaching techniques. He divides the language acquisition process into three stages; comprehension (reproduction), early speech

production, and speech emergence. He claims that the classroom should be devoted to activities which foster acquisition and that providing comprehensible input is of the utmost importance throughout the process. At the comprehension stage, students are not required to speak. They respond to the teacher physically as in a TPR strategy or answer the teacher's questions making use of pictures, objects around them or the other students' names. At the early speech production stage, the students still answer in single words. Teachers speak as caretakers do in natural settings and utilize context, gestures and objects available around the students to make the input comprehensible. Terrell says that if students can successfully pass the first two stages, speech ability will emerge. 2.5 Swaffar and Woodruff's Experiment Swaffar and Woodruff (1978) started a new program of teaching German to students at the University of Texas by emphasizing comprehension. In this case, during the first 4 weeks, Asher's learning strategy of physical response to oral commands was used, and then from the fifth week reading for global meaning was emphasized. Students were encouraged to problem-solve in order to guess at meanings. After two semesters, most students assessed themselves positively. They showed confidence both in reading German to grasp the main ideas and in understanding spoken German. The Modern Language Association listening and reading comprehension tests were administered at this point and the results show the median score for University of Texas students was well above the national norm. Swaffar and Woodruff added that the rate of students dropping the German course between the first and second semesters decreased dramatically when compared with the years before this program had started.

3. IMPLICATIONS
What do the results of all these experiments imply to ESL teachers? Can they introduce a delayed oral practice in their classroom? Yes; but as Gary (1975:95) mentioned, "this approach cannot be used haphazardly. It must be a carefully sequenced program requiring active listening, where the students are both required to demonstrate their comprehension and are given corrective feedback about their response." I would like to consider the characteristics of this approach along with some of the crucial conditions which make this approach effective and successful. 3.1 Comprehensible Input and Modes of Non-verbal Response One of the important characteristics of these strategies is that students are exposed to ample comprehensible input during a 'silent period'. Terrell (1982) says acquisition does not take place by listening to speech that is not understood by the student. Krashen (1985) claims that the optimal level of the input should contain i+1, structures slightly beyond the acquirer's current state of competence. Krashen also introduces Cross's research (1977) and argues that not only the level of the input but also the amount of comprehensible input plays an important role in child language acquisition. According to Cross, children rapidly acquiring language abilities usually heard a great deal of comprehensible input. Another characteristic of these strategies is that they are designed not only to avoid oral practice in initial phases but also to make students demonstrate their

comprehension of materials in a variety of non-verbal ways. The difference between these strategies and those of the traditional audio-lingual or other attempts to teach listening and speaking of a foreign language almost simultaneously is additional nonverbal responses as well as 'lack' of oral practice. As we have already seen, in the case of Asher's technique, students respond physically. Postovsky makes use of writing as a non-verbal response. In the Comprehension Method, students choose correct answers by touching the panel of the machine, TAPAC. In the Natural Approach, students can respond physically or by uttering simple words such as peers' names or "Yes"/'No", or even in their first language. Swaffar and Woodruff, in their program, also use Asher's TPR for the first few weeks. These two points above explain why the traditional way exposing students to authentic English does not work. It is often proposed by some ESL/EFL teachers that listening comprehension is so important that it is useful to let students watch TV programs produced in English speaking countries, or, if students are young, to let them listen to nursery rhymes recorded by a native speaker of English. Then, it is assumed, the more the students are exposed to this kind of authentic English, the more chance they have to master listening comprehension ability. These activities may be useful to motivate students in learning English, but are not enough by themselves. Firstly, if the input is far above the students' comprehension, it will be very painful, frustrating and of little use for the students to keep listening to it. Secondly, if the students are satisfied just with watching TV programs or listening to records, there is no active response on the part of the students. The language acquiring process is always an interaction of stimuli-responses on both sides. Children acquiring a language in natural settings are not passive at all. Even in a 'silent period', there is no such thing as one-way communication. Even a new-born baby can respond to its parents or caretakers by stopping crying or changing the tones of its cries. In addition to this, it is often reported that children who have been taken care of using TV programs show a delay in language acquisition. Though TV programs are made up of audio-visual stimuli, they do not demand viewers to respond to them. They just keep sending stimuli independently of the viewers. In a natural setting, caretakers and teachers try to simplify their speech using many contexts and referents which make the verbal input comprehensible for children. In this kind of environment, children always respond to verbal stimuli in some way. Through such interactions children rapidly internalize a language. The most important point for ESL/EFL teachers seems to make the input comprehensible for their students and also to choose the optimum mode of non-verbal response according to the age and learning styles of students. It may vary widely depending on the class size, length of class time, students' interests and objectives, and so on. If teachers, after taking all these things into consideration, can develop suitable teaching methods and materials, delayed oral practice can be efficiently realized in the classroom situation. 3.2 Lowering Anxiety Another characteristic of this approach is that allowing students to be silent in class serves to present a learning situation with less anxiety. To put it in another way, in a classroom with this kind of approach, the students receive comprehensible input in a low anxiety environment. Asher (1966:81) mentioned that to force speaking from the beginning of training may be somewhat analogous to electroshock experiments with

rats. If students are compelled to utter alien sounds from the start of training, the forced noise-making may function as a stressful stimulus similar to electroshock for rats and it may tend to inhibit understanding. Terrell (1977:333) points out that the standard problem of embarrassment is reduced considerably by allowing the students to respond in their first language. This idea is well explained by the 'filter' theory discussed precisely in Dulay et al. (1982). According to this theory, the filter is a part of the internal processing system that subconsciously screens incoming language based on motives, need, attitudes, or emotions. Incoming language data can be filtered out depending on the students' affective conditions. Research has shown that the less anxious the learner, the better language acquisition proceeds. Similarly, relaxed and comfortable students apparently can learn more in shorter periods of time. This may be especially true for the learners who are past puberty. Most of them are self-conscious and they are anxious and nervous about making errors in front of their peers and very sensitive to sounding strange. Therefore we can say students in an L2 classroom, especially at high school or adult levels, feel much more comfortable and acquire the target language more rapidly if they are not compelled to respond until they feel ready. 3.3 Authentic Model The next characteristic of this approach is that teachers can give students only an authentic model of the target language in class. ESL/EFL teachers always try hard to give their students an authentic model of English. When teachers are non-native speakers of English, they usually make use of tape-recorders. But ironically enough, the more active the oral activities become, the less the students are exposed to a target language. As Postovsky mentioned (1974:231), even if ESL teachers are native speakers of English and can present students with an authentic model, in the audiolingual class where each student is vocally active, students hear themselves more than they hear the teacher. The auditory input which they are processing, then, is not the authentic language, but the classroom dialect rich with all the distortions. This is not the case in a delayed oral practice approach. Students only hear their teachers and/or taped voices. 3.4 Problem-solving The other characteristic of this approach is to encourage students to problem-solve and guess at the meaning of unfamiliar elements in foreign utterances on the basis of context and other cues in the given linguistic environment. Teachers should not present grammar as a set, but rather should help students construct their own grammar through problem-solving. Children acquiring a language in natural settings learn grammar in this way. Swaffar and Woodruff (1978:29) mentioned that in their training "preparations with the aid of a dictionary were actively discouraged, and students were not penalized for incorrect guesses." Once the students are familiar with the way of problem-solving, then they will not have any trouble confronting unknown utterances. According to Asher's experiments

(1972:138), the experimental students showed high competence with novel utterances, which they had never heard before.

4. CONCLUSION
The results of these several experiments fully support the 'silent period' hypothesis; i.e. at the initial phase of language acquisition, it is advisable to avoid oral practice and concentrate on comprehension. There is no evidence that the students who started their L2 learning in this strategy were outperformed by students in the traditional orally active classes. As Gary (1975:93) mentioned, "If it is true that students who weren't required to speak in initial stages would do at least as well as students who were required to speak, then teachers would no longer have to feel compelled to spend a lot of time on oral drill and have more flexibility for individualization of instruction." As already seen above, this approach is effective to lower anxiety and therefore enhance learning. A great deal of transfer-of-learning to other skills has been observed in the achievement of the experimental students. Gary (1975) found in her experiment that the experimental students' rate of learning appeared to be superior to that of the control students. Additionally, Postovsky (1982) mentions that this approach helps to prevent interference. Interference occurs when beginning students are tasked with the production of a foreign sentence, because in this case they are actually asked to retrieve something they have not yet stored in their memory. Therefore they will make an attempt to process the foreign sentence through the only channels available to them: those of their native language. Then why not start speaking later? As we have already seen in the researches presented above, the students should be exposed to a great amount of input in the target language which is comprehensible and authentic. They should not be demanded to respond in the target language but should respond in an optimum mode of response until they are ready to speak. This will certainly enhance their acquisition and give good results after the transition to oral or written production. If we assume some similarity between the language learning process in natural settings and that in formal settings and try to apply some knowledge of the former to the latter, we cannot chop out a certain phase in question alone from the natural environment and transfer it to formal settings. Even if a 'silent period' does exist in a language acquisition process in natural settings, it might be of little use to introduce a 'silent period' alone in our classroom. We have to transfer a lot of elements concomitant with natural settings. There are some more reservations to be considered. (1) In natural settings children are exposed to a target language almost all day long. If a 'silent period' continues for three months in the case of a child who attends an English speaking school for eight hours a day, that period is equivalent to four years of learning in a formal class which meets only three hours a week. Can teachers wait such a long time for students to produce any sounds? (2) Children acquiring language in natural settings seem to be strongly motivated. Children are interested in every interaction in the environment. According to the degree of their understanding, they may or may not eat something good, join in an interesting activity, miss their turn, etc. It can easily

be imagined how difficult it is to always realize such an urgent need or desire in a classroom. If language teachers use only the target language in class, they can serve as a comprehensible model and give the students a chance to know that understanding a language is more than processing quizzes. But native-like proficiency in the target language is required on the part of the teachers. Not all of the teachers satisfy this requirement. (3) The natural environment is rich in concrete referents which help children figure out the meaning of strange sounds in the target language. Children can grasp the meaning to some extent without knowing the meaning of all the words. It is almost impossible for teachers to bring all the referents into a classroom. (4) In natural settings there are a lot of simple expressions which are used frequently and repeatedly, which students in a formal classroom rarely have a chance to use. Instead they are usually required to proceed step by step to more complicated expressions. So the program should include a great deal of repetitions without making students tired of them. If we, as ESL/EFL teachers, pay full attention to these concomitant factors and succeed in realizing a true language environment in the classroom, our 'silent period' hypothesis will give us fruitful results.