through language learning, known as the Classical Method. The aim was to codify a foreign language into rules of morphology and syntax to be memorized (Richards & Rodgers, 2001).
In the 19th century, the Classical Method was known as the Grammar Translation Method (Brown, 2007b), which focused on grammatical rules and usage of the students’ native language (L1), to directly translate the target language (L2). However, this method has no relation to any linguistic, psychology, or educational theory (Richards & Rodgers, 2006). In the late 19th century, increasing opportunities for communication among Europeans created a market for oral communication, thereby a reform was developing.
After experiencing the failure of foreign language learning, a teacher named Gouin, concluded the view that “language learning is primarily a matter of transforming perceptions into conceptions” (Brown, 2007b). Gouin, the reformer, proposed the Series Method by observing how children naturally learned their mother tongue. The Series Method emphases on meaning and leaning in a context that makes the meaning clear (Richards & Rodgers, 2001, p.8). This approach placed more effort on the later Language Reform. In the beginning of 20th century, the Direct Method was widely known and practiced. The Direct Method had lots of oral interaction, no translation between L1 and L2, teachers used L2 to teach with little analysis of grammatical rules (Brown, 2007b). The Direct Method was the first EMI related
21
method where classroom instruction was conducted exclusively in the target language as Richards and Rodgers (2001) summarized. After the end of World War Two, the language-teaching revolution took place, as the United States was seeking courses that facilitated the army’s oral skills where pronunciation, pattern drills, and conversation was adopted in the Army Method. In the 1950s, the Army Method turned into the Audiolingual Method, which was firmly grounded in linguistic and psychological theory. This was the time that structural linguistics turned language learning into “scientific descriptive analysis” (Brown, 2007b).
The behavioristic psychologist, B. F. Skinner (1957), proclaimed that languages are learned through reinforcement known as the “Verbal behavior” (Brown, 2007a). Brooks (1964) further explained when a teacher provides linguistic stimuli in the form of dialogue and drills, reinforces the correct responses, and corrects the students’ errors (“The Audiolingual
Method”). This Behavioral Approach is then criticized for its focus on form and external behaviors, as it was unable to explain the complex intrinsic procedures made by the language learner. The Audiolingual Method was replaced by the Chomskyan revolution in linguistics as the Cognitive Approach. As opposed to Behaviorism, the Nativist believed that language acquisition was innately determined. Chomsky (1965) claimed that we were born with a genetic capacity installed in our brain as a language acquisition device (LAD). Researchers of the Nativist further assumed that all human beings are genetically equipped with Universal Grammar (UG) that Brown (2007a) explained as follows:
22
Universal Grammar (UG) research attempts to discover what it is that all children, regardless of their environmental stimuli (the language[s] they hear around them) bring to the language acquisition process. Such studies have looked at question formation, negation, word order, discontinuity of embedded clauses, subject deletion, and other grammatical phenomena. (Brown, 2007a, p.29)
Chomsky mentioned the language performance and language competence, and therefore after the 1960s, the Communicative Approach superseded the Audiolingual Method and the new method era has arrived, where the movement was to a “competency-based instruction that focus on the outcomes of learning rather than methods of teaching” (Richards &
Rodgers, 2001, p.15).
Teaching methodology related to EMI. The teaching method after the 1960s was
considered as the Communicative Approach, including the Silent Way, the Natural Approach, and the Total Physical Response. In the 1990s, methods like Content-Based Instruction and Task-Based Language Teaching were developed, and there were also approaches which were developed in the first language, and then extended to second language settings such as Cooperative Learning, Whole Language Approach, and Multiple Intelligences (Richards &Rodgers, 2001, p.15).
When English is considered the target language in second language learning, English as a medium of instruction (EMI) can be immersed into different methodology. Except for the
23
Grammar-Translation Method, all the other methods could adopt “English” as the medium of instruction. The rationale behind EMI is about whether the mother tongue (L1) or target language (L2) should be used when teaching a foreign language. In some teaching methods, the use of L1 is restricted and the teachers are asked to use L2 only to teach a foreign language in the Direct Method, the Audiolingual Method, and the Total Physical Response (L1 only introduced in the beginning of class) while in other methods, although maximizing the use of L2 was a common goal, native language was encouraged to be used in the process, they viewed a native language in a positive way, such as the Communicative Language Teaching, Content-Based and Task-Based Approaches (Larsen-Freeman, 2000). Within the Whole Language Approach, L1 was encouraged to facilitate the acquisition of English in a bilingual context (Freeman & Freeman, 1992).
There is one approach very similar to EMI, known as the immersion program. The rationale behind it is the same as EMI, in that whether L1 or L2 should be used in the specific language learning context. The only difference is that EMI insists on using English-only when instructing students in L2 learning. The immersion program, on the other hand, has a more complex composition of L1 and L2.
As opposed to EMI, bilingualism is another issue in that a curriculum is taught bilingually. One thing I have to mention here is that as a teacher/researcher, the aim of the current study is for the second language development, the goal is simply to encourage my
24
students to reach a level so that they are able to communicate in English. With this goal in mind, the immersion program and bilingualism has the main purpose of subject matter
learning, that is, a dual goal on both language learning and subject knowledge. It is not saying that EMI won’t be able to acquire the subject knowledge, however, this is not the main goal in this research. The current research has the same origin of a “second language acquisition”
that is to learn a second language successfully with the positive effects of knowledge learning.
How language is learned through input and interaction. According to Krashen’s
Input Hypothesis, learners must be exposed to the target language data which they can access and comprehended with a low affective filter that makes them open to the input (Ellis, 2003, p.45), known as the comprehensible input. Krashen claimed that acquisition happensunconsciously and the contextual information is helpful for acquiring a language (as cited Ellis, 2003, p.46). Long (1983) further explained that comprehension is crucial for language acquisition, as he proposed in his Interaction Hypothesis that “modifications to the
interactional structure of conversation which take place in the process of negotiating solutions to communication problems help to make input comprehensible to the learner” (Ellis, 2003).
These modified interactions lead to a greater amount of comprehensible input, and that leads to a greater acquisition (Gass & Varonis, 1985). These two hypothesis contributed to the basic rationale of the EMI adoption as well as the Task-Based Approach which I will discuss later
25
in the chapter.
Krashen’s and Long’s Hypothesis has been criticized in that “how comprehensive,” the degree of comprehension that is necessary for acquisition, however, it was not mentioned (Ellis, 2003, p.46). Especially in EFL countries, students with low or no proficiency in English usually “switched off” when they listen to a foreign language, I personally found the situation in my classroom was that even when I’m talking in both languages, they show their boredom by saying “I don’t understand” to avoid input to be comprehended, and thus, no acquisition has ever occurred. Another criticism found in Sharwood Smith’s argument (as cited in Ellis, 2003) that there are two ways of processing input, one involving
comprehension and the other acquisition. From Færch and Kasper’s point of view, this argument can be explained as follows (as cited in Ellis, 2003): “Interactional input
modifications will only lead to acquisition if learners recognize that a ‘gap’ in understanding is the result not of the interlocutor’s failure to make herself understood but of the learner’s own lack of linguistic knowledge.” From my own experience when I was teaching the remedial class at school, I found this concept might well explain that students hardly acquire a language by practicing the sentence patterns where acquisition does not happen, and most of the time they plainly imitated what I said and kept asking the same question next time they needed to say it.
A further challenge to Krashen’s Input Hypothesis is the role of consciousness and
26
attention in language acquisition. Schmidt’s argument (as cited in Ellis, 2003) was that attention to input is a conscious process. “The allocation of attention is pivotal when learner-internal factors (including aptitude, motivation, current L2 knowledge, and processing ability) and learner-external factors (including the complexity and the distributional
characteristics of input, discourse and interactional context, instructional treatment, and task characteristics) come together” (as cited in Ellis, 2003). As this reminds me of the
complicated process of language learning, this study is therefore focused on each individual, by adopting a task that attracts the students’ attention, as well as an observation journal in qualitative research to overcome their obstacles in their learning of the foreign language.
Teacher Talk works as input. This section is based on Wong-Fillmore’s discussion in
the article “When does teacher talk work as input” (Wong-Fillmore, 1985), where she would like to search for answers of “How greatly teachers can influence language learning in their classes by the way they use language in instructional events, and by the opportunities they make available to students during these events to practice the new language”(Wong-Fillmore, 1895), which gave me insight about teacher talk in more detail, and thus, supporting the usage of EMI in the current research.
The issue mentioned in Wong-Fillmore’s study was that the students with limited English proficiency (LEP) were not learning well enough to achieve the required academic level. Among these majority users of English, they do not have the chance to talk to peers in
27
English outside the classroom, as the LEP students’ only English Resources were from the teachers in the classroom, she claimed that bilingual instruction did not work well since the students understood they wouldn’t need to use English in the classroom, therefore, the teacher’s language played an essential role. There are two aspects of the teacher's language use: first, it is the means by which they impart to their students the information and skills they are supposed to be learning in school, and second, it also serves as the linguistic input on which these students can base their acquisition of English.
Wong-Fillmore (1985) offered some suggestions to help the LEP students in these particular circumstances. She observed the differences between classes, and gave a
recommendation that two sets of characteristics determined the success of language learning.
The first set related to the way the lessons are structured or area organized for instruction.
Structural characteristics of lessons that work for language learnings are:
formal lessons with clear boundaries (non-linguistic features like movement of students’
seats, teacher's voice volume, gestures, teacher's talk signals the class begin);
routines were well established and lesson scripts (the same lesson format) has high regularity in the lesson routine;
the use of instructions and signals, and
the recapitulation phase of the lesson.
Wong-Fillmore further mentioned, turn-allocation in lessons, in that everyone has a chance to
28
be called is considered important.
The second set is the way language is used in the lesson. The statement of how language used in lessons will affect language learning were profoundly discussed by researchers Long (1983), Pica, Young and Doughty (1987), and Ellis (1994). As Long (1990) proposed that the interactionally modified input worked well in second language acquisition. As in his
Interaction Hypothesis, “communication breakdowns encourage L2 learners to negotiate solutions to these problem and in the progress internalized the new linguistic knowledge” (as cited in Yousofi & Bahramlou, 2014). Instead of pre-modified input (where teacher talk was modified by reducing the complexity and increasing the quantity and redundancy), the interactionally modifications were made by both teachers and students in negotiation of meanings, and thus, facilitated language acquisition
Pica, Young and Doughty (1987) did a research on comparing 16 non-native speakers’
comprehension by using pre-modified input and interactionally modified input. The results showed that comprehension was best assisted with repeated and rephrased content
information rather than the simplification of input. Furthermore, the research found that the native speaker’s and non-native speaker’s interactional modifications in the form of a comprehension and confirmation check, and a clarification request, served as a mechanism for the native speaker modification of input played a critical role in comprehension.
Ellis (1994) mentioned repetition, confirmations, reformulations, comprehension checks,
29
recasts, confirmation checks, and clarification request as instances of modification and restructuring of interaction (as cited in Yousofi & Bahramlou, 2014). Ellis, Tanaka and Yamazaki (1994) conducted two studies to investigate the effects of modified interaction on comprehension and vocabulary acquisition. They found that input received through
interaction would lead to a higher level of L2 comprehension (as cited in Yousofi &
Bahramlou, 2014). Later, researchers Maleki and Pazhakh (2012) further compared pre-modified input, interactionally pre-modified input, non-pre-modified input, and pre-modified output.
They also concluded that interactionally modified input outperformed the others in the comprehension of new words. These are confirmations of the successful language usage in second language learning.
In addition to Krashen and Long, the argument that comprehensibility is crucial in determining whether the language spoken to learners works as input, Swain (1985) contends that “comprehensible output” was also needed. Swain (1985) concluded that students taught in the French Immersion courses had not performed as equals in the level of the native speakers in their grammatical communicative competence, however, there was no significant difference in discourse and in sociolinguistic communicative competence. The reason was due to the lack of comprehension output, as he defined the comprehension output as follows:
Comprehensible output: output that extends the linguistic repertoire of the learner as he or she attempts to create precisely and appropriately the meaning desired.
30
Comprehensible output, it was argued, is a necessary mechanism of acquisition independent of the role of comprehensible input. Its role is, at minimum, to provide opportunities for contextualized meaning use, to test out hypotheses about the target language, and to move the learner from a purely semantic analysis of the language to a syntactic analysis of it. (Swain, 1985, p.252)
Swain (1985) claimed the reason that the immersion students did not demonstrate native-speaker productive competence was because their comprehensible output was limited in two ways; first, the students were simply not given adequate opportunities to use the target language in the classroom-context. Second, they were not being “pushed” in their output.
There is little social or cognitive pressure to produce language more precisely, especially in later grades as they are understood by teachers and peers. Students in the French Immersion program managed to convey meaning adequately in their early age, however, the second language used is not as precise or appropriate as the native speaker. Although grammatical competence can be explained by comprehensible input, the comprehensible output on the other hand ought to achieve a native-like utterance and be affected by the discourse and sociolinguistic competence.
As a teacher/researcher, I considered comprehensible output as an important element, which explained my own learning experience, in that the language used was Chinese outside the classroom. As long as my teachers and friends can understand the meaning, they will not
31
give negative feedback of my speaking, and thus, communication is adequate, however, but not as precise as the native speaker. The current research conducted the Task-Based Approach as the teaching method, which I think will overcome the problem that has been discussed.
During the post-task stage, which is designed to disclose the language problems or mistakes that have been observed by the teacher.
Back to the classroom scenario of Wong-Fillmore’s study in 1985, where the students ranged from full proficiency in English to none at all. That is a very similar situation faced in Taiwan, a diversity of proficiency level students, students in English proficiency distribution is as bimodal distribution. In this kind of learning environment, language learning occurs in the following conditions: first, when students try to understand what their teachers and classmates are saying; second, when teachers through their efforts to communicate with learners by providing them with enough extralinguistic cues to allow them to comprehend what is being said; and third, when the situation allows the learners to make astute guesses at the meaning of the language being used in the lesson. Teacher talk once again plays an important role as language input. Wong-Fillmore (1985) concluded that characteristics of teacher talk that works as input are as follows:
1. Clear separation of languages: bilingual instructions is not suggested, different language should be used by a different person and time. The negative effects of using LEP students’ native language are: Translation short-circuits the process of
32
language learning from both sides of communication. Teachers do not modify their language, and English is not used as an input in the situation of translation. On the other hand, the learners ignore the language usage only to ensure the meaning, they no longer need to pay attention to the second language as a result. “Children tend to tune out when the language they do not know is being spoken” (Legarreta, 1979;
Wong-Fillmore 1982).
2. Emphasis on communication and comprehension: the use of pictures, demonstration, gestures, enactment, to communicate some of the information to the students. As in an immersion programs, adjustments are made for the goal of communication by putting the new information on what was being presented in the context of work that the students had already completed. The conveyance of meaning is thus successful.
3. Grammaticality and appropriateness of the language used in lessons where language used are appropriate to the activity, teachers are suggested to use the techniques such as simpler structures, avoidance of complex structures, repeated use of the same sentence patterns and/or routines, repetitiveness, and the use of paraphrases for variation, but Foreigner-talk forms were never used in the class.
4. Tailoring of elicitation questions to allow for different levels of participation from students: elicit one-word answers, ask more open-ended questions. teachers should tailor questions to fit the levels of proficiency of individual students, expand
one-33
word answers to a full sentence. Teachers made modification by tailoring questions according to the students' proficiency level, thus lessening the anxiety that language learners are likely to feel when more is expected of them than they can give.
5. Richness of language use: teacher talk in successful classes tends to focus on language itself. Although it might neither be the content of the lesson planned, nor the stripped-down language features, the teachers tend to take every opportunity to impart a feel of the language to their students.
6. Playfulness of the curriculum design: the objectives throughout the lesson is “to help students to develop a greater control of the forms, functions and uses of the new language” (Wong-Fillmore, 1985, p.43). Teachers show their creativities in the curriculum design. Furthermore, teachers found in a successful classroom were considered as effective communicators, a more effective factor of determination.
(Wong-Fillmore, 1985, p.33-43)
From the above findings, acknowledging the theoretical background that
From the above findings, acknowledging the theoretical background that