AN EXPLORATION OF PEDAGOGICAL IMPLICATIONS OF USING NETWORKED-BASED COMPUTER MEDIATED COMMUNICATION IN THE COMMUNICATIVE LANGUAGE CLASSROOM

Cheol-Houn Lee

ABSTRACT

As the ever growing use of computer networking is becoming realized in the domain of communication, many attempts have also been made to explore the pedagogical implication of Computer Mediated Communication (CMC) in the communicative language classroom. Many brilliant concepts and observations about CMC applications in the language classroom have been discussed in terms of the pedagogical advantages of employing asynchronous CMC (such as e-mail) and synchronous CMC (such as MOO) in the language classroom. However, little has been discussed specifically concerning why and how text-based CMC can be used to help language learners improve spoken fluency and how CMC technology needs to be shaped and adapted to meet the goal of communicative language learning. This is because there is little understanding of the connection between how SL (Second Language) learners learn language in terms of improving communicative competence and how the characteristics of synchronous/asynchronous CMC can support such a learning process, a practical model of CMC application in the communicative language classroom is not yet feasible for language teachers and students. Since most of CMC, except for video conferencing, is text-based communication allowing little oral interaction within the networking application, there remains a lot of room to explore how the development of computer networking technology can be directed to help learners convert the experience of language structure through text-based communication to spoken fluency.

In this paper, I will attempt to explore how second language learners become communicatively fluent through text-based a CMC setting, thereby being able to discuss how CMC (both asynchronous and synchronous) should be adapted and shaped to support such language learners' cognitive processes of language learning for communicative purposes. In doing so, I will discuss what problems are encountered in the traditional language classroom to show how CMC technology can be shaped to deal with those problems and to explore what features of CMC can be applicable to the language learning setting. A discussion of text-based features of CMC will be followed by exploring some of the practical models/activities for communicative language learning, which is a web-based simulated language learning.

INTRODUCTION

Many studies of computer networking applications have been pursued to test the pedagogical advantages for language learners in terms of individualized learning and interactive features that allow various types of discourse structures. These networking based interactive features separate themselves from other all CD-based CALL (Computer Assisted Language Learning) software programs that are not technically capable enough to deal with the varying and unlimited types of creative human discourse. Thus, networking based CMC is getting extensively studied in the four language skills: listening, writing, speaking, and reading.

CMC is divided into two communicative modes; the Asynchronous mode, such as e-mail, news groups, BBS (Bulletin Boards) and threaded discussion groups and the Synchronous or real-time mode, such as IRC (Internet Relay Chat) and Multiple User Dungeons (MUDs), MOO, and video conferencing. CMC is text-based communication, but videoconferencing technologies, such as CuSeeMee, also allows visual aspects of communication on the screen. Each type of CMC has its own pedagogical advantages and disadvantages for the different types of language learning settings. For example, asynchronous CMC makes it possible to provide individualized learning. Students can have time to read, understand and respond to the written message in text-based mode, thus having a chance to monitor and edit their or other participants' text-based dialogue structures. The asynchronous mode has been widely used in the domain of collaborative writing, brainstorming, and fostering critical thinking habits of the participants. The disadvantage is that under this mode, students can not expect immediate feedback from other participants. This delayed response can cause frustration in terms of spontaneous communicative purpose. Nor does the text-based mode allow students to practice oral communication within the asynchronous mode itself.

Conversely, the synchronous mode enables learners to communicate with others in a real-time setting where each can get used to a speech-like linguistic strategy (Kern, 1995; Kwang-Kyu, 1996). Immediate responses and feedback are also available in this mode. The disadvantage is that the synchronous mode requires much cognitive demand associated with not only thinking in TL (Target Language) but also typing skills under the fast-going time constraints, thereby providing simplified linguistic or syntactically broken inputs (Kern, 1995; Kwang-Kyu, 1996), or Traduced register ( Ferrari et al., 1991; Maynor, 1994; Murray, 1988). These features of CMC provide diversified language learning activities by being embedded into different learning contexts with different goals of language learning. The features of both modes can be incorporated into a particular context to compensate for the other's disadvantage.

Although many studies indicate that CMC can offer many pedagogical advantages to language learning in terms of its interconnectivity, few have come up with pedagogical CMC models which are practically feasible in classroom activities whose goal is to improve oral fluency. In the same vein, Norikor Nagata (1998) noted that 'there remains an important empirical question about how the CALL exercises should be formatted to optimize their instructional effectiveness in promoting different sorts of competence for different types of target structures'. The major reasons for this stem from the lack of an exploration of and empirical studies on the relation between oral and text-based communication: how the feature of text-based communication differ from oral-based communication; what language learners whose goal is oral fluency can learn from the text-based mode; and what kinds of instructional CMC activities need to be designed to ensure successful transformation from the text-based mode fluency to oral communicative fluency. Most of all, few studies have been done on how text-based CMC can support the cognitive process for speaking skills and how CMC technology takes such a learning process into consideration within its development.

Many predictions have been made about the relationship between the two modes proposing that text-based communication is the intermediate path to spoken communication (Collot and Belmore,1996; Yates,1996; Ray Archee, 1993; Pennington, 1996 ; Maynor, 1994; Kiesler,1991; Poster,1990 ; Zuboff,1988). Most of the assumptions and findings about this issue help contribute to documenting the characteristics of linguistic interaction in text-based CMC and the possible relations between text-based language and spoken language, but have not led to empirical studies and discussion on the possible pedagogical activities specifically related to oral communicative competence within the CMC setting.

In making the employment of CMC available in the language classroom, many advanced studies on computer technology have been made to support the learner's complex learning process for language learning. Those attempts include, for example, developing speech technology such as Automatic Speech Recognition (ASR) (Farzad Ehsani ,1998), artificial intelligence(AI) and natural language processing(NLP). As one of attempts to emphasize the importance of boosting technology to an appropriate level enough to serve as a perfect human-like learning counterpart controlled by the student, Farzad Ehsani(1998) mentions that Reducators and second language acquisition (SLA) researchers are now demanding intelligent, user-adaptive CALL systems that offer not only sophisticated diagnostic tools, but also effective feedback mechanisms capable of focusing the learner on areas that need remedial practice'. What he noted here belongs to the current paradigm of CALL courseware development and research. However, many questions from the pedagogical approach also arise as to whether or not only the issue of 'state of-the-art' technology should be paramount as a paradigm to compensate for the problems of adopting text-based CMC in oral communicative activities. David Crookall, D. Wells Coleman, Edward B. Versluis (1990) pointed out that the overemphasis of the technology in CALL results in a 'dangerous neglect of the learner and the learning process.' The development of appropriate technology needs to be based on such an understanding of how language learners perceive the various types of TL (Target Language) as their input and how technology supports such a learning process. With these problems in mind, as one alternative way to seek a rationale for shaping CMC technology use in the language classroom, it would be valuable to reconsider, within the CALL context, such issues as the features language learners receive as input and which are used for output, and a discussion of CMC technology within the context of the realization of how and where traditional language methodology falls short of training language learners for the communicative competence.

What are the constraints of Communicative Language Teaching?

The major problems found in the conventional language teaching methodology is that the methodology tends to provide students with discrete parts of language skills, such as grammar and vocabulary, thereby failing to prepare students for dealing with active conversation outside the classroom. This realization concerning traditional methods has resulted in the communicative language teaching (CLT) methodology being hailed currently as the most favorable one among many language teachers. CLT is to teach students how to speak competently, not only in terms of knowing the grammatical rules of a language, but also knowing what to say to whom, in what circumstances, and how to say it (Hymes 1972, 1974). Learning languages through scenarios and Role-Play (1), and information gap activities (2) are good examples of such applications. Swain (1984) introduced the concept of four communicative competence, consisting of grammatical, socio-linguistic, discourse, and strategic competence, which means that not only do learners have to know TL structures but also know the other factors to be communicatively fluent. The assumption of CLT about language learning is that if students can be actively pushed to evoke TL in such communicative activities as role-play and information gap, they would stand high chances of receiving abundant linguistic inputs as well as practicing verbal output through interaction, which are said to be essential factors for SLA (Second Language Acquisition)(Krashen,1987-comprehensible input; Swain, 1985-comprehensible output). However, many pedagogical concerns still arise here in terms of how to motivate learners to evoke linguistic structures that the students might know, how to present those linguistic inputs that they might not know, and how to document what they pick up during their communicative interaction. Is it just a matter of telling them directly of vocabularies or syntactic language structures or just expecting them to evoke TL by putting them in communicative activities where they are pushed to say something?

Under fast-going communication, students tend to get discrete parts of language structure. For example, as seen in the following real speech sample between NSUs (Native Speakers) and a NNSUs (Non Native Speakers), the learner is hardly expected to be able to cognitively notice the general linguistic systems underlying TL systems except for the targeted item - the vocabulary meaning of 'rectangle' through this communicative interaction. The actual conversation is so fast that they are not given opportunity to notice the entire syntactic structure of what is being spoken from the verbal interaction.

NS: it's a rectangular bench

NNS: rectangular?

NS: yeah it's in the shape of a rectangle with um you know a rectangle has two long sides and two short sides

NNS: rectangle?

NS: re-rectangle it's it's like a square except you flatten it out NNS: square except

NS: uh a rectangle is a square

NNS: uhuh

NS: except a square has four equal sides NNS: yes

NS: a rectangle has two sides that are much longer and two sides that are much shorter

NNS: OK

(discussed in Pica, 1994, p. 512)

In this linguistic exchange, the interaction provided opportunities for learners to comprehend message meaning and target items(vocabulary) but not to notice the entire linguistic structures of the dialogue they got involved in as well as any kinds of communicative strategies that NSs are cognitively, psychologically, resorting to in getting through his message. For example, how does NS change his utterances, what grammatical features does NS use (except, in the shape of, etc.)?

The notice on the structural forms can be done by verbal feedback and recast, which are said to provide potential intake for SLA as a key element in the process of acquisition (e.g., recast, clarification requests, confirm, etc.) (Gass, 1991; Schmidt, 1990). However, this verbal feedback can likely only touch the segmented languages (lexical, tense,etc.) All syntactic structures can not be noticed by the students in a face-to-face setting. Thus, the students tend not to have much to reinforce except the noticed items (mostly segmented linguistic system such as tense, subject/verb agreement) from this interaction. Nor can teachers, at the same time, draw the students' attention to every possible language structure in the verbal interaction. It is assumed that these segmented items are not necessarily linked to the syntactic mode of processing that is important for oral output unless they are connected to other supporting activities.

When it comes to communicative competence, however, the learners should not only have the opportunity to comprehend a variety of functions in the target language but should also be able to verbally use what they mentally comprehend through interaction in TL. In other words, when they were asked to re-explain 'rectangular' to others in TL after this interaction, they have to be able to restate what was linguistically experienced to others regardless of the degree of simplified or elaborated expression from the some sort of intuition built upon through the interaction. In this vein, this type of communicative verbal interaction may fall into the same categories of discrete language teaching methods. In other words, CLT may also fall short of providing such a learning mechanism that students can get chances to notice and analyze the various types of syntactic language structures that can be creatively manipulated by NSUs linguistic intuition although it allows the students to be immersed in spontaneous oral inputs. Thus, the interaction issue in CLT needs to be reconsidered in CMC context for its new practical application supporting an enhancement in the learners' awareness of how to manipulate language structures according to their intent.

New paradigm of providing interaction in CMC:

Interaction is said to be the essential factor for SLA. Through interaction, students can negotiate meaning, thus making input comprehensible and practice verbal output. When it comes to the role of output(Swain 1985) in second language acquisition, many studies(Noriko Nagata,1998; Swain 1985, 1995; Lapkin,1995) showed that the role of output is also important for SLA as well as comprehensible input (Krashen, 1980,1987;Swain, 1985; Lapkin, 1995; DeKeyser and Sokalski,1996; Gass & Madden, 1985). From this perspective of interaction, CMC can provide expanded opportunity for the participants to take abundant linguistic input and to practice what they know and learn(oral output) by interacting with many different types of other participants on the networking systems.

Terri Cononelos and Mauriaio Oliva (1994) note that 'the learner will benefit by communicating with the greatest possible number of competent speakers' We believe network use facilitates language development by allowing communication with a wide range of native speakers on topics of student's choices. However, it is noted that the interaction is directed to support how the participants can notice how NSs use different linguistic structures in different linguistic situation at the discourse level. I assume that the importance of the language-specific interaction at the discourse level doesn't necessarily lie in the number of the participants. Nor is the frequency of the interaction with many participants although those conditions are said to be helpful for SLA. Rather, it is in how those quantity and frequency are directed for the language learners to notice and retain entire linguistic interaction that they experienced so that they can maximize their chances of digesting any linguistic input on the syntactic level, not on the discrete level. Any communicative interaction needs to be maintained to support such a learning process so that students can raise structural awareness of what is being spoken during interaction and can, linguistically, test out such an awareness again in the other connected linguistic situations. In this regard, the expanded opportunity for interaction during CMC can be marginal unless it is designed to provide the chance for students to explore and notice the creative aspect of language structures through interaction.

All language structures have categorical features such as lexical and syntactical ones within their systems which are collaboratively used by NSUs linguistic intuition(language universal). In the case of children's language learning, it is said that they master 2nd languages easily based on language universals(Gass,1979; Keen and Comrie, 1977 ;Chomsky,1972 ;Dwyer,1997), but many second language studies show that although adult 2nd language learners have difficulty mastering 2nd languages for many reasons such as language transfer and language universal(Gass,1983), adult language learners can have an advantage in learning something in comparison with a child's learning in terms of matured cognitive development. Thus, SL (Second Language) learners can be pushed to manipulate those features creatively as long as they, at the same time, can be provided opportunity to notice such a linguistic mechanism. Through noticing and analyzing various syntactic processes of language structure that vary among people and different situations, learners can have a chance to 'lessen the difference between L1 and L2 in terms of semantical and logical differences of language systems as well as language structures such as ir/regular, infrequent, or a semantically opaque form (Gass, 1994-language transfer)', thereby being able to raise awareness of such noticed forms and experiment with those in their verbal practice. In other words, learners need to be exposed to a variety of spontaneous authentic utterances as a linguistic input which should not be dissected by any discrete part of notion or topic or grammatical sequence during interaction and at the same time, are given a chance to notice how the different syntactic features of language structures can be combined to produce different meaning semantically, logically. Besides noticing the language structure at the discourse level, they also need to be able to analyze such a noticed structural knowledge for reinforcement so it they can be used in later, different linguistic situations. With those controls over the learning process, learners can stand a better chance to build up their linguistic intuition that helps manipulate the knowledge of TL in a way they want, although this linguistic intuition might be different from native speakers.

chapter.

I have so far reviewed the constraints in traditional language classroom in terms of the traditional method's lack of leading the students' communicative interaction into students' syntactic analysis of what they said and heard. Then, the question is that how text-based CMC compensates for the constraints encountered in classroom and shapes the learners' language learning experience in a way to support their oral production.

The general implications of employing text-based Communication in language earning classroom:

CMC is an interactive text-based medium, which means that it can make the texturalization of the linguistic intuition of native speakers available on the screen. This allows learners to notice the entire structure of their communication during text-based interaction in both synchronous and asynchronous modes. In other words, CMC enables what can not be easily targeted and noticed during face-to-face verbal interaction to be texturalized on the screen so that students can have better chances to analyze not only discrete parts but syntactic, the semantic variation of language structures including a native speaker's linguistic strategy-how they change language structures according to their intent, how they start and end and even how they correct their verbal mistakes, etc. From this perspective, it carries significant meaning to the CMC setting, that the recording system of chat log in CMC needs to be technically adapted so as to ensure that the entire syntactic structures of dialogue can be noticeable and recordable for analysis and reinforcement. In the traditional classroom, it is not possible for students to have such a chance to notice the entire syntactic structures of dialogue when they make verbal expression. Gass commented on this by saying that Rone would like to know if learners on their own, without any intervention, are in fact analytical about language as they put utterances together into a text. That is 'Do learners perform some form of rapid-fire mental analysis, using whatever resources they may have at hand, as they produce language?'

The audio and video (with caption) can be good examples of helping the participant to structurally notice what they interact with others. However, it differs from CMC in that there is no interaction with the contents, the participants, and even the teachers. All they have to do in the audio/video setting is to dictate the dialogue structures by repetitively playing the tapes backward and forward to analyze them. This may be more boring, less challenging, less motivating than interactive text-based CMC. In addition, the scope of linguistic input that the learners are going to explore is likely confined only by what was being spoken in the classroom setting.

Another feature of CMC in allowing students to notice dialogue structures is that the synchronous/asynchronous modes of CMC can be easily combined to maximize students' awareness of different types of language structures in various ways within one consistent learning context. For example, after the real-time interaction which can be saved for self-analysis, learners can post any questions on the asynchronous BBS arising from their analysis in terms of grammar, vocabulary, or idioms that they missed during the real-time chat or discussion.

Third, CMC can be technically, easily combined with another networking environment (internet), which can enrich such a language learning experience of various authentic types of language structures with more complexity and flexibility.

Fourth, the pattern of dialogue structures that the language learners will analyze during CMC interaction is the one collected not from other speech sources but from what they struggled to produce, which means that analyzing TL can be bi-directionally related to their actual structural experience. It helps the learners be able to not only analyze their weakness of perceiving the difference between what they know and how they produce oral outputs but also verbally experiment with the analyzed knowledge in another linguistic situation.

Fifth, Under text-based CMC setting, students can test their on-going learning process in a safer mode than in a face-to-face mode. They are free to take risks or a laid-back attitude, which is not recommendable, but sometimes necessary as a part of linguistic strategy, without feeling embarrassed not to "talk".

These features of text-based communication have a great implication for language learning in terms of giving students an opportunity to internalize all linguistic inputs through the written discourse, which could not be all retained in a face-to-face classroom setting. However, in order for the text based system to be available for the communicative language learning, the text-based CMC system needs to be more discussed in terms of how it contributes to the cognitive process involved in speaking skills.

Real time text-based CMC and its relation to oral communication.

The present studies and observations of a real-time genre of electronic discourse shows that text-based communication in CMC has a unique characteristic in between the writing and speaking mode. It is said that the characteristics of 'text-based mode' CMC are not inherently and solely driven by the cognitive process related to writing skills alone, but by Tintermediate one between speaking and writing skills' (Collot and Belmore,1996; Yates,1996). Ray Archee (1993) indicated that 'text-based format of a communication mediated communication is a hybrid of speech and writing, displaying features of both.'

Pennington (1996) and Maynor (1994) noted that the communication through these CMC modes promote conversation-like language and represent a convergence of both oral and written modalities. Kiesler (1991) mentioned that, "though people 'talk' using text, it is not the equivalent of a fast letter. Nor is it a transcribed face-to-face discussion." Poster(1990) said that CMC substitutes writing for spoken conversations and extends the domain of writing to cover areas of communication that previously were limited to face-to-face interactions, the mail, and the telephone.

As one of few empirical studies to analyze the structural characteristics of real time CMC, Kwang-kyu ko (1996) compared written text samples of synchronous CMC (daedalus software) with spoken discourse data collected from all face-to face conversation in terms of the frequency of 28 linguistic features claimed in previous research (Biber, 1983). Kwang-kyu characterized prototypical attributes of speaking and writing ( pronoun-it, word length, amplifiers, private verbs, that deletion, etc.). According to his findings, text-based synchronous CMC is not only simply more similar to the pattern of spoken discourse than written but also shows its unique linguistic patterns on its own, partly shaped from written-like constraints(typing), partly from spoken like constraints(cognitive demands imposed by the real-time communication situation). The similar observation was also made by Kern (1995) about this relationship. These findings that real-time text based communication is shown to pattern the features of spoken language may provide practical rationale in employing text-based CMC as the effective intermedium for the mastery of speaking skills. Under normal face-to-face verbal interaction, there are so many cognitive demands required simultaneously for adult 2nd language learners to notice and retain structural aspects of dialogue in terms of not only fast going constrains, but also social, physical, emotional factors. Thus, if students can have an intermediate learning setting between classroom and individualized learning which requires less cognitive demand for processing all those inputs but still requires similar cognitive demand associated with producing speech-like sentences, they can better transfer what they notice and analyze from intermediate learning activities into the development of spoken fluency. In this way, prior to getting into face-to-face conversation, students can get used to using the speech-like discourse pattern of TL under a similar cognitive learning demand required for speaking, but in a safer position which affords the opportunity of noticing dialogue structures.

The disadvantage of using text-based CMC for the preliminary activities of oral communication is that text-based CMC alone doesn't provide any oral interaction within the CMC technology as follow-up activities to facilitate such a transformation of cognitive process involved in text-based communication into oral-based one. Thus, the listening part is not considered as a contributory factor to SLA. James Fernandes, Gail Ellis, and Barbara Sinclair(1983) said that 'listening skills can expand learners' awareness of the language--the discovery that English is a stress-timed language, that stressed words convey meaning, and that these are therefore the words to concentrate on when listening for gist.'

Second, text-based CMC lacks meta-linquistic factors such as gesture and facial impression which are said to serve as facilitative factors for SLA.

Third, real-time CMC tends to be syntactically fragmented under timing constraints and to have a relatively low degree of information focus and elaborateness, which means ironically that students can lack the experience of learning elaborated forms of linguistic expression in cases of delicate dialogue tasks, although the chances of noticing entire structures of discourse is expanded by CMC's texturalization. This kind of disadvantage shows clearly how CMC technology needs to be developed so as to compensate for those constraints. For example, video conferencing would be the best exemplary technology to compensate for those constraints of CMC.

Two models for the technology application of CMC (web-based and videoconferencing) in Communicative language learning classroom:

There are two possible models for combining the text-based CMC application with spoken communication. One is to make oral communication available along with the text-based mode. Video conferencing such as CuSeeMe is the best example for this. CuSeeMe video conferencing allows not only textual discourse, but also oral communication via microphone along with a small visual screen. When learners talk to other participants through video and audio, what they communicate can be, at the same time, available to the next small screen as a text format which can be technically saved. The students don't have to pay attention to both modes at once since the text chat log can be saved so it can be referred to later. Or they can quickly refer to some sentences that they didn't understand while communicating with each other. In this case, the participants can actually speak with no burden on cognitive demand associated with typing as in text-based CMC, but still have an opportunity to notice and retain the entire dialogue structures while seeing each other on the video screen. The participants can reflect structurally what they communicate with others at their individualized setting and then, they can get back to the next conference secession with the same or consistent topic and test out the acquired awareness of the previous structural experience in the verbal mode. They can go back and forth from text-based to oral-based communication in negotiating meaning or to debate some hot issues periodically according to the pedagogical design of the dialogue task. In this way, if students are under a pedagogically well-designed context along with other supplementary activities such as grammar and vocabulary, the chances would be that they can keep track of how the linguistic structures of TL vary according to different situations and different types of personalities of the participants, thereby raising the awareness of creative aspects of TL structure and having communicative confidence in manipulating TL.

We do not know which features of language structures or how often learners might pick these up from such an individualized analysis in order to use them in speaking. They might use the same vocabulary or idioms used or similar syntactic features that frequently took place in the previous communication, or they might paraphrase the same dialogue sentences with some modification when they speak at a different time with different partners about a similar or the same topic. If these transmissions/transformations take place consistently, it would offer a significant pedagogical implication indicating that students decide, process the linguistic input, develop their own learning processes, and test out their spoken awareness in such a video conferencing setting. This kind of mode rules out an attempt to identify 'input' whose nature is not known and lets students be in charge of identifying, developing input and testing their output based on reflection. Although the detailed and empirical analysis on the more elaborate design of dialogue task using CMC is necessary to test this assumption, it is assumed that the participants can build up their own linguistic intuition about TL which might not be the same as a native speaker's as learners go through trail and error in this kind of CMC learning setting.

Computerized Simulation on the networking Internet (WWW):

The other possible model is to embed text-based CMC(synchronous/asynchronous mode) within a more real-life web-based context which is 'a simulated linguistic context on the internet' and to combine those real-life linguistic experiences on the net with verbal interaction in the classroom/video conference where they can connect their experience of structural exploration to making verbal reports or discussion.

The internet's hypertext system, which can be very complex, makes learning contexts as complex as real-linguistic ones. Sue Barness (1994) said that "these applications enable students to interact with blocks of text by picking and choosing topics of interest as they navigate through hypertext documents. This text-based simulation transforms the student into active language structure explorer who blazes trails through excited simulated space." In other words, the internet can provide a learning setting where students can navigate various types of linguistic choices within the hypertext domain. This kind of well-structured language learning context can be easily utilized as a simulated format in the internet's hypertext system so that the students' interaction can be easily led to the language specific one.

In the realm of business language, these kinds of simulated linguistic tasks have been actively used for learning the appropriateness of language use in different business situations. For example, students can be given multiple linguistic choices to select what would be the best to say hello to the boss or to someone in the particular linguistic setting. After clicking one option, the participants are led to the next linguistic situation according to the result of their choice. All possible dialogue structures for the given linguistic situation are complicatedly simulated as multiple options.

The values of the simulation methodology in communicative language learning are that "simulation can provide students with a whole range of conversational models between a variety of speakers (e.g., doctor/patient, boss/employee, lawyer/client) in a variety of speech situations. In addition, it can expose students to a variety of speech acts not normally found in teacher-centered classes, such as apologies, excuses, promises, threats, and compliments. (Robin, David Crookabll, 1990)S.

Marion Geddes, Gill Sturtridge, Rebecca L. Oxford, and Hana Raz (1990) also noted that 'language learners play a key role in shaping their own learning in accordance with the axiom that the mind is an instrument to be developed, not a receptacle to be filled...The process of learning how to learn pushes students to take responsibility for learning, identify their preferred learning style. This process also helps learners to develop the skills of planning, arranging, and evaluating their own learning, to notice the specific strategies or behaviors(such as imagery, repetition, and naturalistic practice) they use in language learning to expand their strategy repertoire if necessary.'

When it comes to the role of input in the simulated activities, Rebecca L. Oxford and David Crookall (1988) said that 'the input, which can be understood because of the meaningful and engaging context of the activity, is at the discourse level, above the level of single words or isolated sentences; it is connected, contextualized speech embodying all sorts of language functions. Krashen (1982) has demonstrated the importance of language amounts of such authentic inputs. This allows attention to be given to language form as well as to content.'

The concept of computerized simulation has also been explored among many language teachers who are interested in employing such a CD based simulated game in the language classroom such as ICONS, LONDON ADVENTURE, OREGON TRAIL(3).

However, simulation on the internet has a greater advantage compared with CD based or classroom -simulation activities in that internet based simulation not only allows text-based interaction among many human participants which can elicit a large quantity of spontaneous linguistic input, but also allows those text-based CMC to be visible, saved. In addition, the participants' linguistic interaction can be easily, effectively led to language specific interaction in terms of the quality of the given context design where the students can simplify or elaborate their oral output. Students can also have more to say and discuss what they encountered from the complex context and report what they got in a classroom conference.

The combination of CMC and web-based simulated activities using the hypertext system take full advantage of employing all required language learning skills within text/oral-based communication and well-structured linguistic context. This combination can prevent students' linguistic interaction from falling into the syntactic simplification or broken one. Thus, this combination can maximize students' getting access to the various syntactic manipulations of TL within a well organized real-life context by expanding students' structural experiences into their full potential.

The other advantage of using CMC on the net is that the internet's multi-media features combined with networking, such as graphics, sounds, and animation can also be effectively used to support more real-life simulated language learning on the internet, becoming more challenging, fun, and motivational.

When it comes to the teacher's role and intervention, the teacher plays a crucial role in this process. A teacher can easily mediate students' interaction by participating in the activities as another 'chief-partner' or by arranging students into the appropriate level of simulated activities. Besides the well-structured linguistic context, this kind of teacher's mediation can also serve to prevent the use of CMC from falling into non academic discourse in terms of the formal modality of linguistic communication.

Teachers can easily update web-based simulated contexts by changing/adding parts of linguistic structures as long as they know some basic networking technology, such as web-page editors, FTP (PC), Fetch (MAC). Using such a basic networking technology, teachers can reflect the number of students, different needs, and levels on the design of the context, which is another big advantage in comparison with CD-based simulation game software.

The disadvantage of this simulation, however , is that the design of simulated activities has problematic variables to be considered in terms of the level and number of learners, the design of every possible linguistic context involved, the balance between linguistic and social aspects, and situations teachers alone can not deal with.

Net-based CD game:

There is a good example to sense what some of these activities may be like, which is web-based CD gaming such as Diabola and Starcraft. These are highly interactive internet-based CD game programs that are favored world-wide by many game players. Although this commercial CD is solely designed for playing games thus mostly associated with the action involved, an insightful idea about how to design a web-based learning context can be induced from the observation of these kinds of programs.

This net-based CD game combines game playing with synchronous communication on the internet so that the participants on the net can cooperate in beating enemies by moving each character, sharing information and collected items by real-time text-based communication. Each participant can create their own domain by choosing a single player mode or can join other domains created and maintained by other players. All participants can post their game domain or join other's domain on the net based bulletin board designed and provided by the game's production company. Some advanced players who already have a lot of resources, such as gold and powerful weapons, can share some of those with other naive players who request donations or advice. Although the content of the real-time text-based dialogue in this game setting is limited by the nature of game play (they have to keep navigating, searching items and defeating enemies, thus, most of language interaction is simplified and broken language), this kind of interactive setting may shed light on how to incorporate the communicative mode into well-structured context(simulated activities) with some modification tailored to the language learning purpose.

For example, the net-based simulation activities for language learning can be designed to invoke more language specific interaction through brainstorming, negotiated meaning, and cooperative learning than the behaviorally centered navigation such as clicking a mouse to move around characters in game world. Each participant can share information with others in solving the encountered linguistic problems or guide the beginners in scoring or rewards through real time text-based communication. In this way, participants can also have many options to choose how to say and discuss their choices with others in both a/synchronous CMC.

The chat log (not an option in the above games) can also be saved so that students can analyze what they communicated and in the classroom conference, they can report what they experienced during simulated activities. Students can be given such a special task as finding or reporting unknown vocabularies, idioms, syntactic, or semantic features of language structures while moving around the simulated world. Teachers can raise questions, give feedback to those reports, and guide the class discussion about a complicated situation that the teacher thinks the participants suffer in getting through in the simulated activities. In addition, it would be really fun and challenging, and motivating according to the real-life quality of how such a simulation can be competitively, interestingly designed containing well-structured linguistic contexts.

Internet-based simulation doesn't necessarily need animation and cinematic features like these CDUs. It can be graphics-oriented or text-based situations as long as every linguistic situation is linked to each other so as to evoke many unexpected but possible linguistic situations under a hypertext system.

The teacher's role in a future language classroom:

The future ESL/EFL class might not require the physical attendance of students when the technology is at everybody's fingertips. However, it does not mean that such a formal class is not necessary. Rather, its major function might shift from teaching discrete parts of language to managing, facilitating linguistic interaction based on pedagogical perspectives. Keith Miller, Susann LuperFoy, Esther Kim(1996) mention that 'tools for electronic communication in use today offer limited dialogue mediation or no mediation at all...As collaborative groups expand to include more participants and multiple modes of interaction (e.g., speech, keyboard text), the task of dialogue management becomes increasingly unwieldy' This issue could be an important one for language teachers since synchronous CMC, which is under time constraints and the cognitive demand of typing and thinking, still have the tendency to take on the characteristics of 'reduced register', which might serve as poor linguistic input for language learners without pedagogical intervention.

In the case of Usenet which is asynchronous CMC, for example, where the enormous number of readers can participate in choosing various topics and discussing the chosen issues, many practical questions remain in terms of dialogue management. What if students can not actively interact with others on the Netnews in case their counterpart who is a native speaker, loses interest talking to the nonnative speaker students due to a language barrier or to the counterpart's lack of information about the chosen topics, thereby resulting in frustration or even the termination of dialogue tasks? In terms of providing a more language learner friendly setting, as well as academic linguistic feedback, it hardly expects native or non native participants to assume the responsibility of a language teacher in providing an appropriate level of linguistic input. Thus, the level arrangement as well as authentic linguistic inputs needs to be considered as important factors. According to KumiKo's tele-collaboration writing project (1995), she also observed that only when all the participants obtain enough communicative competence in the same working language will synchronous CMC be utilized effectively, emphasizing the level issue in the CMC setting. There are so many unexpected variances for language learners to be able to use this interactive CMC which is open for anyone all over the world for the purpose of language learning. In order to solve the dilemma of expanding the interaction through cooperative language learning but still ensuring the appropriate guidance of the language -specific interaction or the authentic linguistic input for language learners, there is a need for language learners to have an interactive domain (web-based class setting) tailored to language learning purposes where native teachers should get involved to monitor and manage the participants' interaction.

Under this model, the size and number of physical classrooms will be on the decrease whereas the number of students on the increase globally in the net based classroom. This administrative down-cost for the construction of buildings and the increased number of enrolled students over the world will also make it possible to increase the number of English mentors in terms of quality dialogue management (feedback, guide, comments for individual linguistic development) on a small group basis.

Conclusion:

Although no single instruction method can ensure the mastery of spoken skills of TL, there should be more underlying pedagogical approaches reflecting the exploration of the learning process of language learners in shaping the learning environment and developing learning materials. This understanding needs to be reflected within the CALL context. With this issue in mind, I have discussed the possibility of CMC as the best intermediate medium to lead students mastery of communicative skills by exploring the various supporting features of CMC as well as text-based features for the improvement of communicative competence.

Most of all, the features of interactive text-based communication can provide language learners with not only the advantage of their being able to notice the variety of syntactic manipulation of language structures on the screen while they are actually participating in the asynchronous and synchronous written communication but also with the advantage of their being able to reflect upon these structural experiences with the saved data later. By having a chance to notice the dialogue structure they were involved with, students can not only relate their communicative experience into their building up the basic linguistic mechanism of Target Language to be used for oral fluency that NSs use, but also reinforce this knowledge. Gass (1994) addressed this point, saying that 'attention to form and discourse function may contribute to the internal consistency of learner systems'. In addition, the synchronous text/oral-based CMC can especially provide a speech-like linguistic strategy to students so that they can experiment with those speech-like linguistic patterns under similar time-constraints but in a safer environment than face to-face social situation, thereby being better prepared to improve spoken based communicative competency.

The concept of CMC-embedded simulation activity is also studied in terms of how a well-structured context can lead students' communicative interaction into a syntactic analysis of what they experience during/after their text based communication.

The text-based features of CMC can be effectively combined with the internet's hypertext system to guide linguistic interaction under simulated linguistic contexts. Otherwise, students' oral/text-based interaction would fall into the categorization of simple chat-like discussion, which serves as impoverished linguistic input. The CMC within the well-structured linguistic context/task gives more power for students to deal with many other issues besides language structures, such as semantic, socio-linquistic, pragmatic as well as vocabulary and grammar, according to the design of simulated tasks on the net. During this integrated context where the language learners can switch back and forth from CMC to the simulated activities as they explore language structurally, they can also get access to 'syntactic-based language and meaning /pragmatic-based comprehension strategy language'.(Gass, 1987; Harrington, 1987; Kilborn & Ito, 1989; Sasaki, 1991). Under this setting, students are in charge of developing their communicative awareness of TL out of this abundant spontaneous linguistic input at the discourse level. It is not feasible to categorize or sequence those linguistic inputs into discrete parts since the natural process of students' minds regarding such information can not be easily identified. Instead of dissecting the language structures by categorizing some features into discrete parts, to use spontaneous speech input as live learning materials, as in CMC, would help enhance the students' chances to build up native-like communicative fluency. In other words, language structures at the discourse level collected from spontaneous oral interaction means much to language learners in that spontaneous oral interaction may reveal various aspects of creative syntactic features of language tied with the NSUs linguistic mechanism for spoken language including all four factors of communicative competence(Hyme,1972), not necessarily being tied as a fixed rule, not being limited by fixed discourse patterns that can be seen in text-books or in CD-based software programs.

The combination of CMC and web-based simulated contexts should help extend the students' actual verbal or written experiences of language interaction into the stage of analysis and reinforcement at their individualized learning setting so that the learners can actually relate what they learn to what they practice in a connected creative way tailored to their own intent. Although this design of web-based learning materials along with CMC might be demanding for some teachers, cooperatively working with the specialist of CALL courseware or of commercial CD games can make it possible to come up with the finest program for a real-life language learning setting, ensuring the reinforcement of all the learners' structural experiences.

Finally, I briefly pictured the possible transformation of the language learning classroom of the future in that the increasing number using CMC and its global urge will precipitate the transformation of the local-centered ESL/EFL learning setting into more multi-featured and world-bound educational settings and emphasis on the authentic dialogue management maintained by teachers of native speakers.

There still remains much room for empirical researches concerning the effect of the use of text-based CMC on spoken fluency in the off-screen mode and instructional design of CMC-embedded simulation on the net in terms of its format and content. Thus, I hope that this exploration of why and how CMC should be considered in the domain of language learning would contribute to the development of the future research method regarding the CMC issue in the language classroom.

------------------------------------------------

(1) See the following web site about "role-play/simulation".

http://www.aitech.ac.jp/~iteslj/Techniques/Tompkins-RolePlaying.html

(2) See the examples of 'information gap activities' at the following address;

http://wwwsv1.uaizu.ac.jp/~luther/Professional_Stuff/Seattle_Presentatin.htm

(3) OREGON TRAIL - in which the player adopts the role of 'the head of a family of five with $700, a wagon, and a dream of reaching Oregon City in five or six months' (from simulation, gaming, and language learning (1990).

REFERENCES

1. Carol Chapelle (1997). Still in search of research paradigms? Language Learning and Technology. Vol. 1, No. 1, PP. 19-43. Call in the year 2000.

2. David Crookall (1990). Simulation, Gaming, and Language Learning p29. Oxford.

3. C. J. Brumfit and K. Johnson (1983). Communicative approach to language. Oxford University Press.

4. Christopher/Martin/Peter (1985). Computers in English Language Teaching. The British council by Pergamon Press.

5.Clark, H., & Schaefer, E. (1987). Collaborating on contributions to conversations. Language and Cognitive Processes, 19-41

6. Dwyer (1996). The consequences of Language; Manuscript, 1993 - one of reading collections in the course of 'Language and Culture' in Fall semester in1998.

7. Deridre C. Donvan (1995). Computer Mediated Communication and The Basic Speech course. Interpersonal Computing and Technology: An Electronic Journal for the 21st Century. Vol 3, Num 3, pp.32-53. http://www.emoderators.com/ipct-j/1995/n3/donovan.txt

8. J. Michel Metz (1994). Computer-Mediated Communication; Literature review of a new context. Interpersonal Computing and Technology: An Electronic Journal for the 21st Century. Vol 2, Num 2, pp 31-49. http://www.emoderators.com/ipct-j/1993/n2/greller.txt

9. Jackendoff (1993). Patterns in the mind. Chapter 2. Basic Books

10. Kumiko Aoki (1995). Synchronous Multi-user Textual communication in International tele-collaboration. Communication Institute for Online Scholarship, Inc. http://www.cios.org/getfile/Aoki_V5N495

11. Kiesler, S. R. (1985). Talking, teaching, and learning in network groups. In A. R. Kaye (Ed.), Collaborative learning through computer conferencing (pp147-165) Oxford, England: Pergamon Press.

12. Kwang-Kyu Ko (1996). Structural Characteristics of Computer-Mediated Language: A comparative analysis of interchange. Discourse.Communication Institute for Online Scholarship, Inc. Vol. 6, No. 3, 1996 http://www.cios.org/getfile\Ko_V6N396

13. Keith Miller, Susann LuperFoy, Esther Kim, David Duff(1996). Dialogue Management for Computer-Mediated Spoken Bilingual Dialogue Communication. Institute for Online Scholarship, Inc. EJC/REC Vol. 6, No. 3. http://www.cios.org/getfile\Miller_V6N396

14.Leonore M. Greller and Sue Barnes(1993). Groupware and Interpersonal text: The computer as a medium of communication. Interpersonal Computing and Technology: An Electronic Journal for the 21st Century. Vol 1, Num 2
http://www.emoderators.com/ipct-j/1993/n2/greller.txt

15. Martha C. Pennington (1996). The Power of the Computer in Language Education p33. Houston: Athelstan.

16. Maynor, N. (1994) The language of electronic mail: Written speech? In G. Little & M. Montgomery, Centennial usage studies. American Dialect Society, 78, 48-54

17.Maurizio Oliva and Terri Condonelos (1994). Teaching languages with NETNEWS. Interpersonal Computing and Technology: An Electronic Journal for the 21st Century. Vol 2, Num 1, pp. 40-49. http://www.emoderators.com/ipct-j/1994/n1/cononelo.txt

18. Mannulea Gonzalex-Bueno (1998). The effects of Electronic mail on Spanish L2 Discourse. Language Learning &Technology Vol.1,No.2, Jan. pp. 50-65 http://polyglot.cal.msu.edu/llt/vollnum2/article3/default.html

19. Noriko Nagata (1998). Input VS. Output Practice in educational software for second language acquisition. Language Learning and Technology . Vol.1. No.2, Jan. 1998, pp.23-40. http://polyglot.cal.msu.edu/llt/vollnum2/article/default.html

20. R. William Maule (1993). The network classroom. Interpersonal Computing and Technology: An Electronic Journal for the 21st Century. Vol 1, Num 1.
http://www.emoderators.com/ipct-j/1993/n1/maule.txt

21. Ricoeur (1978). Word, Structure, Event. Chapter 8. Boston, Beacon Press.

22. Robert J. Di Pietro (1987). Strategic interaction. Cambridge University Press

23. Ray Archee (1993).Using Computer Mediated Communication in an educational context: Educational outcomes and pedagogical lessons of computer conferencing.. Vol. 3, No. 2. Communication Institute for Online Scholarship, Inc. http://www.cios.org/getfile\archee_V3n293

24. Steven Pinker(1994). The Language Instinct Chapter 2-chatterbox. HarperPerennial.

25.Sue Barnes (1994). Hypertext Literacy. Interpersonal Computing and Technology: An Electronic Journal for the 21st century. Vol 2, Num 4, pp. 24 36. http://www.emoderators.com/ipct-j/1994/n4/barnes.txt

26. Slobin (1979). Language Development in the Child, language and cognition. P 144-185. Scott Foresman and Company.

27. Susan Gass (1983). A review of interlanguage syntax: Language transfer and language universals.