Among the recent innovations in the field of second langauge teaching, task-based language teaching is probably the most promising and productive one, the one which has drawn much attention from both second langauge teaching profession and second langauge researchers. Three distinctive variations of it have emerged over the last ten years. They all can be put under the general category of task-based language teaching because they all consider the use of content-based tasks as the most important feature of their annovations. However, they do differ from each other in some significant ways, particularly in terms of syllabus design. The first of them is characterised by having no predetermined syllabus and represented by Prabhu (1987), Breen (1984), Candlin (1984), and Foley (1991). Advocates of this approach argue that the content of the language classroom should not be predetermined. Among the reasons they give, the following two are the most important. First, our current knowledge of second language learning process doesn't allow us to "clearly identify learning "items"" and "prespecify the optimum sequence for their presentation in order to optimize learning by the audience, whatever its size." (Candlin, 1984 p. 40), and second, even if we predetermine what is to be taught, we can't determine what is actually learned by our learners. The result is the "continuing disparity and conflict between intention and reality, between theory and realization." (Candlin, 1984 p. 32). To them a syllabus should be open and negotiable and they should be "retrospective records rather than prospective plans" (Candlin, 1984 p. 35).

In the second approach, as represented by researchers such as Long and Crookes (1992) and Nunan (1989, 1991), on the other hand, the syllabus does have predetermined content. However, different from tranditional syllabuses which use grammer, lexis, notion or function as units of analysis, it is organized around the tasks which are derived from real-life tasks for pedagogical purposes and sequenced accrording to the complexity of the tasks (Long & Crookes, 1992). The third variation, as represented by Willis (1990, 1993), Sheen (1994), Widdowson (1984), and Brumfit (1984), makes a clear distinction between syllabus and methodology. They adopts a traditional structural (grammatical or lexical) syllabus, but the method is task-based. They argue that it is the methodology that can be communicative or task-based, but "There is no such thing as a communicative syllabus" (Widdowson, 1984).

Advocates of task-based language teaching claim that such a teaching approach is "compatible with current SLA theory." (Long & Crookes, 1992 p. 43). It has also been claimed that many studies have produced evidence which is in support of the effectiveness of this approach, specifically the use of tasks, in facilitating SLA. However, compared to the discussion of the practical aspects involved in this teaching approach, such as the designing of a task based syllabus and task derivation and sequencing, little has been done to explain the general question of how the use of tasks can better promote SLA than other teaching approaches, or, in the cases of the impirical studies, the results which are interpretated to support such claims, in terms of current SLA theory. In this paper I hope to answer three questions: what is a task? What does a task do as a learning activity? According to the current theory of SLA, broadly defined, how does the use of tasks facilicate SLA.

A. What is a task?

This is an important question to answer for two reasons. First, there is much inconsistency and confusion about what a task is at present. Second, a proper definition is necessary and important to the discussion of how the use of tasks should be able to promote SLA. The following are some definitions of the term "task" one can find in the literature:

"...a piece of work undertaken for oneself or for others, freely or for some rward. Thus, examples of tasks include painting a fence, dressing a child, filling out a form, buying a pair of shoes, making an airline reservation, borrowing a blirary book, taking a driving test, typing a letter, weighing a patient, sorting letters, taking a hotel reservation.... and helping someone across a roda. In other words, by 'task' is meant the hundred and one things people do in everyday life, at work, at play, and in between." (Long, 1985 p. 89)

"An activity or action which is carried out as the result of processing or understanding langauge (ie. as a respone). For example, drawing a map while listening to a tape, listening to an instruction and performing a command, mat be refereed to as tasks. Task may or may not involve the production of a langauge. A task usually requires the teachrr to specify what will be regarded as successful completion of the task. The use of a variety of different kinds of tasks in language teaching is said to make langauge teaching more communciative ...since it proficeds a purpose for a classroom activity which goes beyond the practice of a language for its own sake. (Richards, Platt and Weber, 1986 p. 289)

"...a piece of work or an activity, usually with a specified objective, undertaken as part of an educational course, or at work." (Crookes, 1986)

"...any sturctured langauge learning endeavour which has particular objective, appropriate content, a specified working procedure, and a range of outcomes for those who undertake the task. "Task' is therefore assumed to refer to a range of workplans which have the orverall purpose of facilitating languate lerna8ing--from the smple and brief exercise type, to more compelx and lengthy activities such a sgroup problem-solving or simulations and decision making." (Breen, 1987 p. 23)

"An activity which required learners to arrive at an outcome from given information through some process of thought, and which allowed teachers to control and regulate that process, was regarded as a `task'. (Prabhu, 1987, p. 24)

"...a piece of classroom work which involves learners in comprehending, manipulating, producing or interacting in the targe tlanguage while their attention is principlaly focused on meaning rather than form. (Nunan, 1989 p. 10)

"...an activity which involves the use of language but in which the focus is on the outcome of the activity rather than on the language used to achieve that outcome. (Willis, 1990, p. 127)

"...some kind of activity designed to engage the learner in using the language communicatively or reflectively in order to arrive at an outcome other than that of learning a specified feature of the L2. (Ellis, 1994, p. 595)

These definitions differ in specificity. Long's definition which covers virtually everything people do, is a definition of the term in its general sense, with no particular reference to language teaching. The rest of the definitions all define the task in the pedagogical context. A difference here is that Breen's definition includes exercises while the rest of them do not. Based on the the way most researchers define and use the term and the actual examples people give in their discussion of task-based langauge teaching, and following Hichey's (1993) approach to defining formulas by means of a preference rule system originally used by Jackendoff (1983), some defining features of tasks are outlined below.
1. A task is an activity carried out for the ultimate purpose of language learning. Since the need for a clear definition of the term "task" rises because it takes on a particular meaning when being used in language teaching context, it is only reasonable to exclude activities carried out for purposes other than langauge teaching and learning.
2. A task is a meaning-oriented activity whereby the exchange of information is essentail for its successful completion. This rules out those learning activities in which the learners' attention is focused on linguisitc forms, such as imitation or pattern drills.
3. A task is an activity which leads to a specific outcome. The outcome can be a decision, a solution to a problem, an agreement reached among participants. This rules out activities which, though involving exchange of information, do not result in a specific outcome such as telling a story, questions and answers as independent tasks though they can be part of a task.
Based on these defining features, a task can be defined as a meaning-focused language learning activity which leads to a specific outcome at its completion. These defining features are the necessary conditions an activity has to meet to be considered as a task. Besides these conditions, a task typically employs language as the primary media for communication, though paralinguistic or any other means may be also involved. It should also have a definite beginning and ending point to make it a complete and independent activity. A task also typically has a set of precedures usually predetermined by the instructor for the learners to follow for its completion.

B. What does a task do as a learning activity?

A task thus defined plays an important role in providing comprehensible input and promoting communictive interaction among the learners using the target langauge. This can be seen in a number of studies which have produced evidence suggesting there is a relationship between variation in task types and and variation in the quanity and quality of negotiated interaction. (e.g. Pica et al., 1989; Varonis & Gass, 1985).

The amount of comprehensible input and interaction, in turn, may promote acquisition. This is the major claim of the input hypothesis of Krashen (1985) and the interaction hypothesis of Swain (1985). It has also been supported by several lines of evidence, such as relationships between input frequency and output accuracy (Larsen-Freeman, 1976a, 1976b; Lightbown, 1980; for a summary of the studies, see Ellis, 1994) and input frequency and the use of formulaic expression which leads to the creative construciton of rules (Wong Fillmore, 1979; Ellis, 1984; also see my answer to Dr. Saville-Troike's question), positive effect of communicative teaching on the acquisition of grammatical structures (Hammond, 1988; Terrell, Gomez, & Mariscal, 1980), and acquisiton of L2 in naturalistic settings (Wong Fillmore, 1979) (but see Spada & Lightbown, 1989; Ellis, 1992; Snow & Hoefnagel-Hohle, 1982; Krashen and Scarcella, 1984 for conflicting evidence).

While the evidence seem to favor the effectiveness of tasks in promoting L2 acquisition by means of comprehensible input and interaction, little has been done, in the context of task-based language teaching, to explain such effectiveness in terms of current SLA theory. Poeple touch upon theories and ideas such as the input hypothesis, the frequency hypothesis, the interaction hypothesis (Fotos & Ellis, 1991), experiential learning (Nunan, 1991), but they seldom go into details in explaining the mechanism whereby comprehensible input and interaction can actually bring about acquisition (but Loschky & Bley-Vroman 1993 is an exception), partly because these theories themselves are not specific enough about the mechanism or the process to allow such explanation . Thus, explicit theoretical explanations are scarce. However, several models or theories of langauge acquisition in general or second language acquisition in particular can be brought to bear on this issue. I will focus on three of them, the teachability hypothesis, the competition model and asymmetrical connetion model of bilignual lexicon.

C. Theorectical accounts of the effect of tasks on L2 acquisiton

1. Tasks and the competition model

The Competition Model (Bates and MacWhinney, 1982, 1989; MacWhinney, 1987) was first proposed mainly to account for sentence processing in first languages. However, it has been applied to the study of second language learning (McWhinney, 1987, 1992) and proved to be a very productive model for examining issues of second language acquisition, particularly sentence processing in second languages (Bates and MacWhinney, 1981; Miao, 1981; Wulfeck et al., 1986; McDonald, 1987; Gass, 1987; Harrington, 1987; Kilborn and Ito, 1987; Kilborn and Cooreman, 1987; Kilborn, 1989; Sasaki, 1991; McDonald and Heileman, 1991; Liu, Bates and Li, 1992).

Two features of the model seem to be of particular importance in discussing tasks and SLA. One is form-function mapping as a mechanism of langauge learning and the other is the proportional nature of the learning process. The proponents of the Competition Model have described this model as following a functional approach to the study of language. The language is considered first and essentially as a means of communication, rather than a system of rules. The most important idea of the Competition Model is form-to-function mapping, that is, any formal feature of a language such as phonology or word order exit to express certain communicative intention or function and any communicative function is realized by certain linguistic feature. The learning of a language, according to this model, is a process whereby such mappings are formed. It involves the recognition of a particular linguistic form as a cue to a particular function and the building-up of the appropriate cue strength. Following such a view of langauge learning, a favorable learning condition would be one under which the learner is able to see how linguistic forms are used to fulfil particular communicative functions and able to try out these forms for communication so that the initial association between the form and its function can be established and then strengthened. That means the learning activities the learners engage themselves in should be content-oriented and should allow the learners to use the langauge to fulfil a purpose. A task, as defined above, where the use of the langauge has a specific objective and the exchange of information and negotiation of meaning is essential for its completion, is exactly what such a model of language learning calls for.

The competition model is also a connectionist model because of its proportional nature of the learning process. It is recognized that form-function mapping does not have to be in a one-to-one fashion, i.e., a linguistic feature can express more than one communicative function and a function may be realized by more than one linguistic form. When more than one form are associated with a function, these forms are said to be in a competing relationship. The result of the competition will depend on the "cue strength", i.e., the relative strengths between each of these forms and the function. The linguistic form, or "cue", with greater cue strength will be more likely to be relied on in carrying out or inferring about the underlying function. Cue strength is claimed to be "a function of both task frequency and cue validity" (MacWhinney, 1987) and "directly proportional to cue validity" (MacWhinney, 1992). Cue validity refers to the extent a linguistic form reliably and consistently tell us the function it expresses. It is determined by "overall availability" (i.e., the extent to which a cue is actually used or "available" to express that function), "cue deductibility" (i.e. the extent to which If a cue can be detected by the learner as related to certain function), and "conflict reliability" which refers to the extent to which a cue can win over the other competing cue when they suggest the opposite. The concepts of availability and reliability can be expressed numerically, the former being "the ratio of the cases in which the cue is available over the total number of cases in a given task demain" and the latter "a ratio of the cases in which a cue leads to the correct conclusion, over the number of cases in which it is available" (MacWhinney, 1987).

Thus, the task of learning a language is not only to learn the form-to-function mappings but also their associated cue strength. Since the information about the validity of various cues comes exclusively from the language input a learner receives in the course of language acquisition, it naturally follows that the amount of input a learner receives will be a determining factor in this learning process. A child needs sufficient input to establsih adult-like cue strength. In the case of second language learning, the amount of input is even more important becuase of the fact that the cue validity of the learner's two langauges might be different and he or she has to overcome the L1 influence, which has been documented in several studies of bilingual sentence processing (e.g. Wulfeck et al, 1986; McDonald, 1987; Gass, 1987; Harrington, 1987; Kilborn, 1989; Liu, Bates and Li, 1992) to establish the target language patterns. To conclude, both comprehensible input and interaction play very inportant roles in langauge learning in this model. They work tegother to allow the learner to establish the connection between forms and function and the appropriate strengths of these connections. With the constructs such as form-funtion mapping, competition, cue strength and cue validity, and given its proportional nature of the learning mechanism, the competition model seem to be able to explain how comprehensible input and interaction may facilitate SLA in more specific and coherent terms.

2. Tasks and bilingual representations

In the limited space left, I'd like to consider the effect of tasks, comprehensible output to be specific, from a representational perspective. Examining SLA from the representational perspective has not been a tradition of the field, at least not the mainstream approach. My discussion is very tentative.

It has been generally accepted that memory can be representated at two levels, conceptual and lexical. The L1 lexicon has two important features. First, at the conceptual level, concepts are closely interconnected, such that thinking of any concept would automatically activate other semantically related concepts. Second there is a strong connection between the conceptual and lexical levels such that thinking of any concept would lead to the automatic activation of the corresponding lexical item(s). These features are essentail for efficient spontanious communication.

When one begins to learn a second langauge, it is very likely that the L2 lexical items are directly connected to their corresponding L1 words, rather than the related concepts. This has two consequences in terms of the organization of the L2 lexicon. There is less and weaker interconnections between L2 lexical items and a weaker connection between the conceptual level and the L2 lexical level. Observation of L2 performance and experimental evidence seem to suggest that this is the case.

My own intuition and observation suggest that there is much greater disparity between the active and passive vocabularies of L2 users than L1 users. Many L2 users are able to recognize huge number of words in reading; However when it comes to speaking or writing, they use a much smaller vocabulary repetitively. This can be explained by assuming that direct connection between the lexical items and their conceptual representations only exist in the case of some but not other words. Words become active vocabulary when such connection exist. Otherwise, they are passive. Another related phenomenon is that L2 learners usually give fewer responses and take longer time than L1 users in word association tasks (Lambert, 1956; Magiste, 1979; Kruse, Pankhurst, & Sharwood Smith, 1987). This could be taken as indicating weaker interconnections among L2 lexicl items. More direct evidence comes from more recent studies of bilingual lexicon. In some experiments done in the Psycholinguistics Lab here at UA, it was found that, under the masked condition under which the subjects were not aware of the presense of the prime words, thus were not able to adopt any strategies, lexical decision reaction time is faster when the L2 target words are preceded by their L1 translation primes than by unrelated L1 words, but not when the L1 words are the targets preceded by L2 prime words (Jiang, 1995; Gollan, personal communication, 1995). This asymmetrical priming effects (from L1 to L2, but not from L2 to L1) can be explained by the asymmetrical connection model of bilingual memory of Kroll (Kroll & Steward, 1994). An important claim of this model is that the links between words and concepts are weaker for L2 than for L1. Their subjects' performnce in translation and picture naming tasks indicate that bilinguals take the semantic route when translating from L1 to L2, but use the lexcial route when doing translation in the opposite direction, thus supporting the claim.

Taken together, the above observations and evidence seem to lead to one conclusion, i.e., L2 memory representations are different from those of L1. The difference is well captured in Tulving's distinction between episodic and semantic memeory systems (Tulving, 1972, 1983). One of the differences with regard to their operations between these two systems is that "Access to, or actualization of, information in the episodic system tends to be deliberate and usually requires conscious effort, whereas in the semantic system it tends to be automatic." (Tulving, 1983, p. 46). L2 lexicon of a non-balanced bilingual, or at least part of it, is more like an episodic system in its lack of close association between lexical items and the deliberate nature of information retrieval.

Viewed in this context, learning a second language can be considered a process whereby initally stored episodic memory becomes a semantic system. This means at least three things. First the learner should be able to develop a set of concepts specific to L2. Second these concepts should be integrated into the conceptual system of the learner. Three, direct links between the conceptual representations, whether they are new or originally acquired through L1, and the L2 lexical items should be established so that activation of any semantic information would automatically activate the appropriate L2 words. Such links can be best established through message-oriented productive use of the language, where a top-down information flow is involved, beginning from the communicative intention and concepts, through the selection of lemmas, and ending in the actualization of the output. This is exactly what tasks encourage the learners to do. A task, as defined above, differs from traditional form-oriented learning activities in that it involves the learners in meaning negotiation and encourage message-oriented output. In a form-centered activity whose primary concern is accuracy or fluency in using lingusitic form, rather than exchange of information, though speech production is involve, it doesn't start from the conceptual level, thus not able to facilitate the establishment of the links between concepts and words. To summarize, one aspect of non-nativeness is the lack of direct links from the concepts to the L2 words, which limits the learners' active vocabulary and hinders automaticity in speech production. The use of tasks help learners establish such links by involving them in message-oriented productive use of the langauge.

In addition to the competition model and the representa- tional model of bilingual lexicon, the use of tasks can find support in some other SLA thoeries. In the terms of the teachability hypothesis of Pienemann (1985, 1989), engaging the learners in meaning-oriented activities may help render more structures in a "ready" state to be affected by teaching, thus speeding up the learner's progress through developmental stages. The use of tasks can also be supported by any theory which incorporates affective factors in its framework (e.g., Krashen, 1981; Gardner, 1985) because it can be argued that meaning-oriented activities with specific objective are usually more intersting to do and the learners might be more motivated. However, a task-based language teaching may have its limitations. One potential problem is that the extensive use of pair or small group work may result in much exposure to ungrammatical input (Fotos & Ellis, 1991). It also remains an open question as to how to monitor and measure the learners' acquisition in a task-based approach.

References:

Bates, E., & MacWhinney, B. (1981). Second language acquisition from a functionalist perspective: Pragmatic, semantic and perceptual strategies. In H. Winitz (Ed.), Annals of the New York Academy of Sciences conference on native and foreign language acquisition. New York: New York Academy of Sciences. pp. 190-214.

Bates, E., & MacWhinney, B. (1989). Functionalism and the competition model. In B. MacWhinney & E. Bates (Eds.), pp. 3-73.

Bates, E., & MacWhinney, B. (Eds.) The crosslinguistic study of sentence processing. New York: Cambridge University press.

Breen, M.P. (1984). Process syllabuses for the language classroom. In C.J. Brumfit (Ed.), General English syllabus design. (ELT Documents No. 118, pp. 47-60). London: Pergamon Press & The British Council.

Breen, M.P. (1987). Learner contributions to task design. In C.N. Candlin & D. Murphy (Eds.), Lancaster Practical Papers in English Language Education: Vol. 7. Language learning tasks (pp. 23-46). Englewood Cliffs, NJ: Prentice Hall.

Brumfit, C. (1984). Function and structure of a state school syllabus for learners of second or foreign langauges with heterogeneous needs. In Brumfit, C. (ed.).

Brumfit, C. (ed.). General English Syllabus design. Oxford: Pergamon, 1984.

Candlin, C.N. (1984). Syllabus design as a critical process. In Brumfit, C. (ed.).

Crookes, G. (1986). Task classification: A cross-disciplinary review (Tech. Rep. No. 4) Honolulu: University of Havaii at Manoa, Social Science Research Institute, Center for Second Language Classroom Research.

Ellis, R. (1984). Formulaic speech in early classroom second language development. In Handscomber et al (eds.). On TESOL '83: The question of control. Washington D.C.: TESOL.

Ellis, R. (1992). Learning to commujciate in the classroom. Studies in Second Lanaguge Acqusition, 14, 1-23.

Ellis, R. (1994). The study of second langauge acquisition. Oxford: Oxford University Press.

Foley, J. (1991) A psycholignuistic framework for task-based approaches to language teaching. Applied Linguistics 12 (1): 62-75.

Fotos, S., & Ellis, R. (1991) Communicating about grammer: A task-based approach. TESOL Quarterly 25(4): 605-628.

Gardner, R. (1985). Social Psychology and second language learning: The role of attitude and motivation. London: Edward Arnold.

Gass, S. (1987). The resolution of conflicts among competing systems: a bidirectional perspective. Applied Psycholinguistics, 8, 329-350.

Harringtion, M. (1987). Processing transfer: language-specific processing strategies as a source of interlanguage variation. Applied Psycholinguistics, 8, 351-377.

Jiang, N. (1995). The organization of the bilingual lexicon: evidence from masked priming. Paper presented at the annual conference of AAAL, Long Beach, CA., March 1995.

Kilborn, K. (1989). Sentence processing in a second language: the timing of transfer. Language and Speech, 32, 1-23.

Kilborn, K., & Cooremann, A. (1987). Sentence interpretation strategies in adult Dutch-English bilinguals. Applied Psycholinguistics, 8, 415-431.

Kilborn, K., & Ito, T. (1989). Sentence processing strategies in adult bilinguals. In E. Bates & B. MacWhinney (Eds.) pp. 257-291.

Krashen, S. (1981). Second language acquisition and second langauge learning. Oxford: Pergamon.

Krashen, S. (1985). The input hypothesis: issues and implications. London: Longman.

Krashen, S., & Scarcella, S. (1978). On routines and patterns in language acquisition and performance. Language Learning, 28, 283-300.

Kroll, J.K., & Stewart, E. (1994). Category interference in translation and picture naming: evidence for asymmetric connections between bilingual memory representations. Journal of Memory and Language, 33, 149-174.

Kruse, H., Pankhurst, J., & Sharwood Smith, M. (1987). a multiple word association probe in second language acquisition research. Studies in Second Language Acquisition, 9, 141- 154.

Lambert, W.E. (1956). Developmental aspects of second lanauge acquisition. Journal of Social Psychology, 43, 83-102.

Larsen-Freeman, D. (1976a). Teacher speech as input to the ESK learner. Working Papers iN TESL (Unviersity of California), 10, 45-49.

Larsen-Freeman, D. (1976b). An explanation for the morpheme acquisition order of second lanauge learners. Langauge Learning, 26, 125-134.

Lightbown, P. (1980). The acquisition and use of questions by French L2 learners. In S. Felix (Ed.) Second Language Developemnt: Trends and Issues. Tubingen: Gunter Narr.

Long, M.H. (1985). A role for instruction of second language acquisition: Task-based language teaching. In K.Hyltenstam & M.Peinemann (Eds.), Modelling and assessing second lanauge acqusiition (pp. 77-99). Clevedon, England: Multilingual Matters.

Long, Micheal, H., and Graham Crookes. (1992) Three approaches to task-based syllabus design. TESOL Quarterly 26: 27-56.

Loschky,L., & Bley-Vroman,R. (1990). Creating structure-based communication tasks for second language development. University of Hawai'i Working Papers in ESL, 9(1):161-212.

MacWhinney, B. (1987). The competition Model. In B. MacWhinney (Ed.), Mechanisms of Language Acquisition. Hillsdale, NJ: Erlbaum.

MacWhinney, B. (1987). Applying the competition model to bilingualism. Applied Psycholinguistics, 8, 315-327.

MacWhinney, B. (1992). Transfer and competition in second language learning. In R.J.Harris (Ed.), Cognitive Processing in Bilinguals. Elservier Science Publishers B.V. pp. 371-390.

Magiste, E. (1979). The competing linguistic systems of the multilingual: a developmental study of decoding and encoding processes. Journal of Verbal Learning and Verbal Behavior, 18, 79-89.

McDonald, J. (1987). Sentence interpretation in bilingual speakers of English and Dutch. Applied Psycholinguistics, 379-413.

McDonald, J., & Heilenman, K. (1991) Determinants of cue strength in adult first and second language speakers of French. Applied Psycholinguistics, 12, 313-348.

Nunan,D. (1989). Designing tasks for the communicative classroom. Cambridge: Cambridge University Press.

Nunan, D. (1991). communicative tasks and the langauge curriculum. TESOL Quarterly 25(2): 279-295.

Pica, T., Hollday, L., Lewis, N., & Morgenthaler, L. (1989). Comprehensible output as an outcome of lingusitics demands on the learner. Studies in Second Language Acquisition, 11, 63-90.

Pienemann, M. (1985). Learnability and syllabus construction. In Hyltenstam, K., & Pienemann, M. (Eds.), Modelling and assessing second language acquisition. Clevedon, Avon: Multilingual Matters.

Pienemann, M. (1989). Is lanauge teachable: Psycholinguistic expreiments and hypotheses. Applied Linguitics, 10, 52-79.

Prabhu, N.S. (1987). Second Language Pedagogy. Oxford: Oxford University Press.

Richards, J., Platt, J., & Weber, H. (1986). Longman dictionary of applied linguistics. London: Longman.

Sasaki, Y. (1991). English and Japanese interlangauge comprehension strategies: an analysis based on the competition model. Applied Psycholinguistics, 12, 47-73.

Sheen, R. (1994). A critical analysis of the advocacy of the task-based syllabus. TESOL Quarterly 28(1): 127-151.

Snow, C., & M. Hoefnagel-Hohle. (1982). School age second lanauge learners' acess to simplified linguistic input. Language Learning, 32, 411-430.

Spada, N., & Lightbown, P. (1989). Intensive ESL programmes in Quebec primary schools. TESL Canada, 7, 11-32.

Swain, M. (1985). Communicatve competence: some roles of comprehesible input and comprehensible output in its developemtn. In S. Gass & C. G. Madden (Eds.), Input in second language acquisition. Rowley, Mass.: Newbury House.

Tulving, E. (1972). Episodic and semantic memory. In e. Tulving & W. Donaldson (Eds.), Organization of memory. New York: Academic Press.

Tulving, E. (1983). Elements of episodic memory. Oxford: Clarendon Press.

Varonis, E. and Gass, S. (1985). Non-native/non-native conversations: A model for negotiation of meaning. Applied Linguistics, 6, 71-90.

Widdowson, H.G. (1984). Educational and pedagogic factors in syllabus design. In Brumfit, C. (ed.).

Willis, D. (1990). The lexical syllabus. London: Collins.

Willis, D. (1993). Comments on Michael H. Long and Graham Crookes's "Three Approaches to Task-Based Syllabus Design." TESOL Quarterly 27(4): 726-733.

Wong Fillmore, L. (1979). Individual differences in second language acquisition. In C.J.Fillmore, D.Kempler, & W. S-Y. Wang, (Eds.), Individual differences in language ability and langauge behavior. New York, NY: Academic Press.