SLLT, Vol. 1, 2001

James McLaughlin

This project evaluates the English language-learning program implemented at Andong National University in Korea. The program's workbooks were designed in house according to task-based principles. Testing and grading also showed a strong orientation towards task-based principles.

The task-based approach falls under the umbrella of Communicative Language Teaching (CLT). However, it differs from other CLT approaches in some important ways.

First, I will review the literature on CLT and task-based approaches. I will then briefly assess to what extent the A.N.U. program followed principles for a CLT program as described by Canale and Swain (1980) and Breen and Candlin (1980). Finally, through a content analysis of the workbooks, I will do a more thorough evaluation of the program against Long and Crooks' (1992) and Skehan's (1996) recommendations for a task-based approach.


"Approaches that were popular prior to Communicative Language Teaching such as Situational Language Teaching and The Audiolingual Method tended to place emphasis on the elemental components of language" (Richards & Rodgers, 1986, pp. 35, 48-50). They assumed language to be a collection of parts. Language proficiency, then, requires mastery of the parts - phonology, morphology, lexicon and syntax. Communicative Language Teaching (CLT) in contrast places emphasis on the functions and purposes of language in use.

Linguistic theory is also divided into two contrasting schools of thought paralleling the ones above. One contends that linguistics' primary focus should be language elements, while the other insists that it is equally important to study language in use. Chomsky, a proponent of the elemental approach, distinguishes competence and performance. He claims that competence is the internalized linguistic system of an ideal native speaker of a given language. "Performance concerns linguistically irrelevant psychological factors involved in the perception and production of speech including perceptual strategies, memory limitations, and emotional factors" (Canale & Swain, 1980, p. 3).

Other linguists, however, do not accept that language can be studied in a way that is abstracted from situational use. Halliday developed the alternative school of Functional Linguistics. Feez (1995) describes the key ideas of this perspective:

Every time we use language we make meaning by representing our experience about the world (experiential meanings) and meanings, which are to someone else by establishing a relationship with them (interpersonal meanings) and we make meaning which are relevant and meaningful in the context (textual meaning). We also make meanings, which are logical... From the point of view of systematic functional linguists, when language users require language to do something, it is never in a vacuum. Language is a social semiotic (p. 27).

It may be inferred from this perspective that language use must be understood in terms of context. Since language does not occur in a social vacuum it is pointless to study it as if it did.

Systematic Functional Linguistics considers that the very nature of language is determined by the uses people have for it. In other words the kinds of systems that exist in language and the relationships between them and so the kinds of choices languages users make, have evolved and continue to evolve, because of what language is required to do  (Feez, 1995, p. 27).

Along this line, Campbell and Wales (1970) state, "by far the most important linguistic ability is that of being able to produce or understand utterances which are not so much grammatical but, more important, appropriate to the context in which they are made." They proposed the notion of communicative competence. This includes contextual or sociolinguistic competence, as well as, grammatical competence. Hymes, as well, asserts, "there are rules of use without which the rules of grammar would be useless" (cited in Canale & Swain, 1980, p. 4).

Richards and Sukwiwat (1982) use the term conversational competence in a similar way to the previous definition of communicative competence:

Grammatical competence describes a speaker's knowledge of the underlying system of morphology, and syntax, which are required to construct grammatical sentences in a language. The sentence is the unit of description for grammatical competence. Conversational competence, however, is defined not with reference to the sentence, but to the utterance. This refers to the speaker's knowledge of how speech acts are used in social situations (p. 113).

Canale's and Swain's (cited in Yalden, 1987) theory of communicative competence includes three main competencies: grammatical competence (knowledge of the rules of grammar), socio-linguistic competence (knowledge of the rules of language use), and strategic competence (verbal and non-verbal communication strategies that compensate for breakdowns in communication due to performance variables or to insufficient competence). Canale (cited in Yalden 1987) extends his model to include discourse competence (mastery of how to combine grammatical forms and meanings to achieve unity of a spoken or written text).

Canale's (1983) model of the elements of communicative competency

Canale (cited in Yalden, 1987) writes the following about the nature of linguistic communication:

  1. is a form of interaction, and is therefore normally acquired through interaction;
  2. involves a high degree of unpredictability and creativity in form and message;
  3. takes place in discourse and socio-cultural contexts which provide constraints on appropriate language use and also clues as to the correct interpretations of utterances;
  4. is carried out under limiting psychological and other conditions such as memory constraints, fatigue and other distractions
  5. always has a purpose (for example, to establish social relations, to persuade, or to promise);
  6. involves authentic, as opposed to textbook-contrived language; and
  7. is judged as successful or not on the basis of actual outcomes (pp. 3-4).

Accordingly, knowledge of the grammatical system is necessary, but not sufficient, for language proficiency. The ability to comprehend the intended meaning of something that is heard or read in context and having the ability to use language to do things in context are equally important. CLT assumes this model of language.


I will now review research and theories supporting the case for task-based approaches. Stern (1983) states, "What emerges from the interlanguage research is a conviction that the learner's degree of proficiency can legitimately be conceived as a system created by the learner for himself" (p. 155). Interlanguage is that system of L2 understanding that the learner creates.

Corder (1967) suggests that "hypothesis testing is the primary way first and second language learners develop language knowledge. Possible language constructs are first considered and tested. Correction or reinforcement determines if a given construct is maintained or modified" (p. 167).

Ellis (1985) expands on this principle. He states, "when learners first internalize new linguistic items, they do not know precisely what functions they realize in the target language" (p. 85). The Diffusion Model (Gatbonton cited in Ellis, 1985) posits two broad phases of the development of an interlanguage rule. In the acquisition phase, the learner uses a given form in his/her interlanguage in every situation or context then introduces a second form. To begin with, the new form is used together with the first form in all environments - in free variation. In the replacement phase, the learner restricts one of the forms to a specific environment, while continuing to use both for all other environments. The learner, gradually, restricts both forms to their respective environments. The progressive reorganization of form-function relations occurs with the eventual elimination of redundant forms. Language use is, of course, essential for this process to occur.

The Natural Order Hypothesis suggests language elements are acquired following a tract determined by the innate characteristics of the learner. Most of the evidence to date comes from first language studies which include Brown (1973) who found that morphemes were acquired in a similar sequence longitudinally. Jill and Peter de Villis (cited in Littlewood, 1984) in a cross sectional study found a corresponding sequence. Krashen (1982) cites "other studies lending support to the hypothesis" (p. 13).

In a study of language learning strategies, O'Malley and Chamot (1990) report more effective students monitor their comprehension and production for meaningfulness rather than concentrating only on individual components. They, as well, use their prior knowledge while working on a task.

Krashen distinguishes language acquisition from language learning. "Language acquisition is the unconscious incorporation of the target language system through its use in communication" (Richards & Rodgers, 1986, p. 72). Language learning results from formal instruction in which students learn the rules and structures of the language. Language learning, according to Krashen, does not lead to language acquisition. The Input Hypothesis describes the process of acquisition. Language is acquired when the learner is exposed to L2 language he/she understands, but which is a little more advanced than his/her current level of competence - Comprehensible Input. "The process is described by the formula i + 1 "(Krashen, 1982, p. 21).

Krashen's Monitor Hypothesis states a cognitive device called a monitor (the product of learned language) acts in conjunction with acquired language to produce second language speech. The monitor modifies and corrects L2 speech (Krashen 1982). Although not scientifically verified, Krashen's ideas have been influential.


I will now discuss general attributes of CLT. Richards and Rodgers (1986) claim "CLT is an approach (rather than a method) that aims to make communicative competence the goal of language teaching and develop procedures for the teaching of language skills that acknowledge the interdependence of language and communication" (p. 66).

Yalden (cited in Richards & Rodgers, 1986) lists major communicative syllabus types of CLT:

Type Reference
1. structures plus functions Wilkins (1976)
2. functional spiral around a central core    Brumfit (1980)
3. structural, functional, instrumental Allen (1980)
4. functional Jupp and Hodlin (1975)
5. notional Wilkins (1976)
6. interactional Widdowson (1979)
7. task-based Prabhu (1983)
8. learner generated Candlin (1976) Henner-Stanchina and Riley (1978)

Finocchiaro and Brumfit (cited in Richards & Rodgers, 1986) list distinctive features of CLT:

Breen and Candlin (1980) describe teacher, student and material roles in the CLT classroom. These are summarized in Figure 2.

Canale and Swain (1980) present five principles that must guide the communicative approach for a general language program. 

  1. As communicative competence is composed of grammatical competence, sociolinguistic competence, and strategic competence the primary goal of a communicative approach must be to facilitate the integration of these types of knowledge for the learner.
  2. A communicative approach must be based on and respond to the learner's communication needs. It should take into account the situations they are likely to use the L2 and the level of competence expected by native speakers in those situations. The needs must be specified with respect to the three competencies: the levels of grammatical accuracy required for communication; needs relating to setting, topic, communicative functions (sociolinguistic); the compensatory strategies used when breakdowns occur in one of the other competencies.
  3. The language learner must have the opportunities to take part in meaningful communicative interaction with highly competent speakers of the language in realistic second language situations. Exposure to realistic communication situations is crucial if communicative competence is to lead to communicative confidence.
  4. Particularly in the early stages of second language learning, optimal use must be made of those aspects of communicative competence that the learner already has. The grammatical code should be practiced in environments which are less arbitrary and more familiar. Social functions such as greetings, requests, compliments etc. may share the same appropriateness aspects as in the L1. These common features can be taken advantage in practicing the uncommon elements of code.
  5. Classes conducted in the L1 should teach aspects of the L2, such as grammatical categories, communicative functions, appropriateness conditions, and rules of discourse. A social studies program may provide the sociocultural knowledge of the L2 group that will be necessary for drawing inferences about the social meanings or the values of utterances. They suggest a curriculum wide approach to the development of communicative competence may facilitate continued study of the L2 (p. 27).

Canale and Swain (1980) while acknowledging a lack of empirical evidence (at the time) on its effectiveness, argue for a functionally organized approach. They suggest that this approach should contain, "a level of grammatical organization that is adequate for effective second language teaching and learning" (p. 32). They propose a functionally organized approach is beneficial at all stages of second language learning, if only, because of the face validity of its materials and syllabus. In their view, "a functionally based communicative approach in particular, one in which the units are organized and labeled with reference to communicative functions - is more likely to have positive consequences for learner motivation than is a grammatically based communicative approach with reference to grammatical forms" (p. 33).

So, while rejecting a system based only on grammatical organization, they argue for an organized and labeled system organized primarily on functions. Grammar, also, should receive a degree of organizational attention.  

Howatt (cited in Richards & Rodgers, 1986) distinguishes strong and weak versions of CLT. The weak version, which he describes as becoming more or less standard practice, provides learners opportunities to use their second language  for communicative purposes and attempts to integrate communicative activities into a wider program of language teaching. The strong version advances the claim that language is acquired through communication and does occur not by activating an existing but an inert knowledge of language, but by stimulating the development of the language learning system itself.

Teacher, Student and Material Roles in CLT (Breen & Candlin, 1980)

Teacher Roles

  Main roles:
  (1)Communication process facilitator

  • between classroom participants
  • between classroom participants and the activities and texts

  (2)Independent participant in the learning-teaching group

  Implied secondary roles of the teacher:

  • organizer of resources
  • resource for the learners
  • guide for class procedures and activities (making clear to students what they need to do to accomplish tasks)
  • giver of feedback
  • seeker of feedback
  • researcher, learner (monitoring the process of learning-teaching)
  • exploiter of learning opportunities
  • shaper of future directions of learning

Student Roles

  Main roles:
  (1)Active participant in a learning community
  (2)Participant in matacommunicative activities

  Implied secondary roles:

  • assists the teacher
  • assists other learners
  • provides feedback to the teacher (a source of information to the teacher)
  • receives feedback from the teacher
  • receives feedback from other learners
  • acts to direct his own learning (employs strategies)

Material roles

  • specifies student tasks
  • provides samples of the target language
  • promotes internalization of L2 characteristics
  • encourages student independence
  • encourages student motivation
  • promotes L2 language skills

It would seem in their influential paper Canale and Swain (1980) advocate what was referred to (by Howatt) as the weak version of CLT. They, in addition, seem to strongly suggest an additive approach where learners incorporate predetermined language structures organized primarily on a functional basis and secondarily on a grammatical basis. This as we will see, is an approach rejected by task-based advocates. Task-based syllabuses as we will also see, fall under Howatt's strong description of CLT.  


Long and Crooks (1992) discuss and advocate task-based syllabuses. About these syllabuses they contend:

They are distinguished from most earlier syllabus types by the fact that their rationale derives from what is known about human learning in general and/or second language learning in particular rather than, as is the case with lexical, structural, notional, functional, and relational syllabuses primarily from an analysis of language and language use. In addition, while differing from one another in important ways, all three reject linguistic elements (such as word, structure, notion or function) as the unit of analysis and opt instead for some conception of task (p. 27).

Long and Crooks (1992) state "syllabus types can be divided into two classes, synthetic and analytic" (p. 28). These represent two points on a continuum rather than as a strict dichotomy. Synthetic syllabuses are characterized by their tendency to segment the target language into discrete linguistic items for presentation one at a time (Wilkins cited in Long & Crooks, 1992).

Different parts of the language are taught separately and step by step so that acquisition is a process of gradual accumulation of parts until the whole structure of language has been built up… At any one time the learner is being exposed to a deliberately limited sample of language…The learner's task is to re-synthesize the language that has been broken down into a large number of small pieces with the aim of making his learning task easier (Wilkins cited in Long & Crooks, 1992, p. 2).

The synthetic syllabus, therefore, relies on the learners' assumed ability to learn a language in parts (e.g., structure and functions) independently of one another. They must integrate, or synthesize, the pieces of learned knowledge for when they are used for communicative purposes. Syllabus designers who choose a linguistic element - word, structure, notion, or function as the organizational unit simultaneously commit to a synthetic syllabus. Therefore, lexical, structural, notional, and functional syllabuses are synthetic.  Crook and Long (1992) insist, that even though developers may describe synthetic approaches as communicative they are, in fact, artificial. They conform to sets of linguistic specifications defining levels of proficiency, and so do not reflect how people speak, write or learn language.

Analytic syllabuses in contrast do not control for structure or lexis in the above manner. Again, they quote Wilkins:

Prior analysis of the total language system into sets of discrete pieces of language that is a necessary precondition for the adoption of a synthetic approach are largely superfluous… Analytic approaches… are organized in terms of the purposes for which people are learning a language and the kinds of language performance that are necessary to meet those purposes… Since we are inviting the learner, directly or indirectly, to recognize the linguistic components of the language behavior he is acquiring, we are in effect basing our approach on the learner's analytic capabilities (Wilkins cited in Crook & Long, 1992, pp. 13-14)

They argue the use of task-based syllabuses, following an analytic model is supported by SLA research. Research summarized to support the case for task-based include, findings that learners usually do not move from zero to targetlike use of new items in a single step, there can be lengthy periods of nontargetlike use; that learners move through a developmental sequence (the Natural Order Hypothesis); that progress is not always unidirectional – second language acquisition may involve temporary deterioration in learner performance; and that learners may alternate between correct word use and non-nativelike forms for extended periods.

While maintaining SLA research rejects a focus on forms or in other words the use of synthetic syllabuses, in language teaching, they allow for a focus on form (singular). This, refers to the use of pedagogic tasks and other methodological options to draw students' attention to aspects of the target language code.

In terms of program design, tasks are described as, the unit of analysis to provide an, integrated, internally coherent approach. While disclaiming that learners can acquire language a task at a time, just as they cannot learn language a structure or function at a time, they maintain tasks provide a vehicle for presenting appropriate target language samples to learners. What results is that learners perceive new form-function relationships. Through their use, form-function relationships not destabilized by negative input are strengthened, their accessibility is increased, and they become incorporated into more complex associations within long term memory. This then, adds to the complexity of the grammar that constitutes SL development. This is essentially the same description of interlanguage development provided earlier in this project.

Long and Crooks (1992) maintain syllabus designs must involve identification of student language needs in terms of the real world tasks they are preparing to undertake, for example, buying a train ticket, renting an apartment, reading a technical manual, solving a math problem, reporting a chemistry experiment, or taking lecture notes (This is much the same as what Canale and Swain (1980) recommend). Once the student needs are identified, they should be classified into target task types. For example, in a course for training flight attendants, food and beverages may be a heading under which the tasks serving breakfast, lunch, dinner and snacks and refreshments could be classified. Pedagogic tasks derived from the real world task types can then be sequenced to form the task-based syllabus. Increasingly complex approximations of the target tasks would be developed for the students and teachers to work from. Assessment in task-based syllabuses is usually criterion-referenced. Student performance on tasks is set against criterions established by  experts in the field.

Sheen (1994) reacting to what he sees as an excessive pendulum swing away other approaches and towards the task-based syllabuses raises several red flags:

  1. Research, according to Sheen (1994), indicates that approaches involving deductive teaching methods are more effective in bringing about language acquisition than inductive methods.
  2. He criticizes Long and Crook (1992) for declaring that research shows task-based syllabuses reflect the processes used in language acquisition and then failing to cite adequate research data. He contends a consensus on how language acquisition occurs does not yet exist.
  3. Long and Crooks (1992), according to Sheen (1994), do not specify the nature of tasks intended to draw student attention to form.
  4. He states task-based approaches and French Immersion programs share a dependency on Comprehensible Input (Krashen) as the mechanism to induce language acquisition. Students in French Immersion programs tend to show high usage rates of non-target structures. Doubt, therefore, is cast on the effectiveness of task-based syllabuses (as they are also supposedly CI based) to bring about accurate second language use.
  5. The suggestion by Long and Crooks (1992) that language-learning programs must consider the specific conditions in which learners will use the target language is unrealistic. Most students in a given class have various reasons for learning the L2 and one cannot say definitely under what conditions they will use the L2 in the future. Task-based approaches, therefore, limit themselves to Survival English and ESP roles.

Skehan (1996) elaborates on specific pitfalls associated with the task-based approach. He then recommends tactics to avoid the dangers of task-based syllabuses while preserving the advantages.

Skehan (1996) explains that processing language only for the purpose of extracting meaning does not necessarily lead to form sensitivity. It as well, will not guarantee the pressures for interlanguage development assumed by the Input Hypothesis. As communicative pressure becomes the primary concern of the learner, he/she employs strategies to handle or compensate for the immediate communication pressures. The compensation strategies that are developed, however, may cause disengagement of other processes that would be used to stretch interlanguage and lead to change. Compensation strategies that succeed in getting the point across successfully a number of times (without necessarily being correct or appropriate) may become habitual. Lexicalisation is an example. As a result of communicative pressure during language learning, language tends to be stored in word combinations. This enhances accessibility when time pressure is high and efficiency is important. However, the build up of understanding of how the words are used in various combinations within the L2 structure is retarded. Proceduralization of improvised strategies can over time become a stumbling block to language growth (Sinclair cited in Skehan, 1996).

Skehan, unlike Sheen, does not propose a step back from task-based syllabuses. He contends the conventional Presentation, Practice and Production (PPP) approach is problematic. He describes the approach as tacitly assuming language development happens in the presentation phase, and is translated onto accuracy and fluency in the succeeding stages (Rivers cited in Skehan, 1996). This represents a synthetic approach as described by Long and Crooks (1992). According to Skehan, as well, it ignores the language acquisition research on learner systematicity (Ellis cited in Skehan, 1996). It assumes that whatever the teacher decides is worth presenting is appropriate for incorporation into the learner's current interlanguage system. It, finally, assumes language rules have a linear sequence - that covered rules need not be repeated. He contends the PPP approach is now widely discredited (Stern cited in Skehan, 1996).

A task-based approach, in contrast, may achieve the goal of restructuring, if interaction opportunities can extend interlanguage development, and engage acquisitional processes. To do so, tasks must be designed to allow enough attentional capacity in learners to attend to form by reducing the communicative pressure in tasks to a manageable load (Patten cited in Skehan, 1996).

Skehan (1996) distinguishes three areas of learner goals: accuracy, fluency and complexity. Accuracy concerns the learner's capacity to perform to the level of interlanguage complexity he/she has already attained. Fluency is concerned with the learner's capacity to mobilize an interlanguage system to communicate meaning in real time. Finally, complexity, with its attendant process, restructuring, refers to the elaboration of the underlying interlanguage system. The complexity goal promotes risk taking; encourages incorporation of new forms; and directs the learner's attention to assembling the interlanguage system (Foster & Skehan cited in Skehan, 1996).

Skehan (1996) explains the importance of accuracy, fluency and complexity to communication and describes the problems that may result when there is a deficiency in any one. He, also, explains how too much concern with either accuracy or fluency can lead to a learner not taking the steps or risks necessary to expand his/her interlanguage. A preoccupation with accuracy leads to a reluctance to take risks and attentional resources get diverted away from language expansion. A preoccupation with fluency leads to a lower priority being attached to accuracy, and the use of new forms (Bygate cited in Skenhan, 1996). Skehan concludes, to avoid the pitfalls that result from over or under emphasis on accuracy, fluency and complexity, a balance must be achieved in attentional allocation. 

Figure 3 (Skehan, 1996) presents the factors and conditions necessary for an effective task-based syllabus. Code complexity concerns the traditional areas of syntactic and lexical difficulty and range. This is is not described any greater detail. Communicative stress concerns a group of factors impacting upon the pressure of communication. Time pressure concerns how quickly the task has to be done, and if there is any urgency in the manor in which it is done (Bygate cited in Skenhan, 1996). Modality concerns the macro skill contrasts speaking/writing and listening/reading. It is assumed that speaking leads to more pressure than writing, and listening more than reading (Ellis cited in Skehan, 1996). Scale refers to a range of factors associated with tasks, including the number of participants, the number of relationships involved and so on (Brown, Anderson, Shilcock & Yule cited in skehan, 1996). Stakes depend on how important the task is, and to do it correctly. If it is important not to make mistakes while doing the task, than the stakes are high (Willis cited in Skehan, 1996). Finally, control refers to the extent to which the participants within a task can exert an influence on the task and on how it is done. Whether, for example, participants can negotiate task goals or ask clarification questions to reduce the task difficulty (Pica et al. cited in Skehan, 1996). Cognitive complexity concerns the content of what is said. It distinguishes processing and familiarity. Processing concerns the amount of on-line computation required while doing a task, and highlights the extent to which the learner has to actively think through task content. Familiarity, in contrast, involves the extent to which the task draws on ready-made or pre-packaged solutions. It implies the accessing of relevant aspects of schematic knowledge.

I will now discuss the Goals of task design presented in Figure 3. Pre-emptive or pre-task activities increase the chance that restructuring will occur in the language system; new elements will be incorporated or some re-arrangement of existing elements will take place during the task. This is accomplished by making salient language form relevant to task performance (Foster & Skehan cited in Skehan, 1996). Teachers can be explicit immediately before the task is done as to whether they want accuracy to be stressed, or whether they want specific structures to be used.

During the task the choice of the task itself is the main factor effecting performance during the task. The goal is to choose tasks of appropriate difficulty. Tasks should not be so difficult that excessive mental processing is required to communicate meaning. If they are, a reliance on communication strategies can reduce the pedagogic value of the exercise. Conversely, excessively easy tasks will cause students to be bored and will not lead to extension of the interlanguage system (Swain cited in Skehan, 1996). In terms of cognitive complexity visual supports such as diagrams will make tasks easier by easing the amount of material a learner has to keep in mind while performing the task itself. The surprise elements not matching learner expectations of task requirements add to difficulty. Communicative stress can be adjusted by manipulating the stress factors - time, modality, scale, stakes and control.

During tasks the teacher withdraws and allows the acquisition processes to operate. However, this brings about the danger that communication goals will dominate to the extent that lexicalised communication strategies will override the behaviors required for language development. Post-task activities can redirect learner attention back to the task, reminding learners that fluency is not the only goal during task completion, and that restructuring and accuracy also have importance (Willis & Willis; Tarone cited in Skehan, 1996).

Factors and elements pertinent in developing a task-based syllabus (Skehan, 1996)

Goal Code complexity Communicative stress Cognitive complexity

- establish target
- reduce cognitive load



- conventional
- parallel tasks
- rehearsal of elements





Solve simmilar tasks

- cognitively
- linguistically

- activate


Mediating accuracy
and fluency


Accuracy focus

Conformity pressure


Task choice



Support Available

Surprise Elements

Post 1

Increases accuracy

Encourages restructuring

excessive synthesis

Post 2

Cycle of synthesis
and analysis


Public performance
- teacher
- group
- camera

Degree of analysis testing

The task sequence
-parallel tasks

Task families



The analysis of the Andong program has two main components. Firstly, the results of a teacher questionnaire are assessed regarding the extent generally accepted CLT principles were followed by instructors and students in the program. Secondly, the program workbooks are assessed against principles for a task-based curriculum.   


Before beginning my assessment I will briefly describe conditions in the Language Center at Andong National University in Korea, where the program to be assessed was implemented.

The Language Center (L.C.) building was equipped with state of the art facilities including multi-media technology.

The teaching staff in the program were all native English speakers. Although all had university degrees plus minimal TEFL credentials such as an RSA certificates only a small minority had degrees focusing in TESOL or applied linguistics.

The students came from the general university population – most were not English majors. They tended to originate from the surrounding region and were generally from lower middle class households. The majority were false beginners; having studied English structure in public school they had had few chances to use English realistically.

According to their author each of the workbooks implicitly seeks to address a different learning characteristic. Year one is intended to address confidence. It is organized along survival English themes. It employs information gap activities such as surveys and guided role-plays. Year two works towards motivation. It is topic oriented and includes guided conversation exercises plus the same sorts of activities as in year one. Year three works towards independence. It is project oriented with students being expected to complete a number of longer-term tasks, independently or in groups, in which they have to make use of their acquired English knowledge.

Students were graded for class participation, homework and project completion, and a final oral test. Midterm tests were at the teacher's discretion. A listening project for first year students was implemented where students received points for completing listing activities independently in the Language Lab.


I will now assess the program against Canale and Swain's (1980) recommendations.

As recommended, the activities in the workbooks tend to integrate grammatical, sociolinguistic and communicative competencies. In some activities students get information from one another using structures provided in the book (e.g., Do you ____ ?). They then record, summarize and report their findings to the class. Such activities clearly integrate grammatical and communicative skills. Less explicitly, the students utilize sociolinguistic skills as they report their findings to the class. Role-play activities such as making a request from a teacher or mailing a package, integrate socio-linguistic skills more overtly. 

Students in the program were likely to stay in Korea and use English for work or travel. As concerns the recommendation that the program prepare the students for the specific situations they would encounter, the workbooks attempt to address this. They provide practice in everyday situations like going to a party, buying a train ticket, mailing a letter and so on. There are few activities, however, that provide practice in realistic professional or work related situations.

Canale and Swain's third criteria for a CLT program is that it provide opportunities for taking part in meaningful communicative interaction with highly competent speakers of the language in realistic situations. Class sizes, in the program, did not exceed twenty-five students. Confident students would have had opportunities to interact with instructors for contrived and real purposes (a few instructors deliberately gave vague instructions to encourage students to come forth and seek clarification). Instructors were required to keep regular office hours. In addition, most instructors engaged in extracurricular activities with students, such as magazine clubs.

As recommended, the early stages of the program, especially Book One, employed activities making use of existing student knowledge of L2 structure for communicative tasks. As well, group and partner discussion questions were on topics generally familiar to Korean university students.

Canale and Swain's final recommendation is that a curriculum wide approach be taken to prepare students to operate in the target language environment. On this final point, the A.N.U. program was somewhat lacking. The students were sent to the Language Center for just two 50-minute sessions a week. Except for scheduling, there was no coordination between the L.C. and other university departments.

Surveys were conducted based on Breen & Candlin's (1980) description of teacher, student and material roles in the CLT classroom. The results are summarized in Tables 1, 2 and 3.

The teachers' survey (Table 1) yielded very positive outcomes for the first three sets of questions dealing with teacher, student and material roles in class (the fourth set of questions is discussed later). In regards to student roles, six of seven items yielded average results of 4.00 or higher on a scale of one to five. For teacher roles, ten of eleven items yielded average results higher than 4.00. Finally for material roles, all the averages were over 4.00. 

Since the survey scale summarizes perceptions and not observations we do not truly know the extent to which Breen and Candlin's conditions were truly met. The results do, however, clearly show a strong positive bias on the part of the responding instructors towards the CLT approach. Most of the instructors seemed to adhere to the principles of CLT even if it cannot be shown they implemented these principles to the extents indicated.

It should also be noted that the survey was conducted during a period of upheaval at the L.C. The Deputy Director (and author of the books) was preparing to leave on rather unhappy terms. Many instructors were preparing to leave with him. The highly positive teacher survey outcomes may, also, be a manifestation of political cohesiveness.

The student surveys produced more modest and widely varying results. Interestingly, the only item in the students' survey scoring higher than the equivalent item in the teacher's survey was the one on the extent they saw themselves helping the teacher. It was the highest scoring item on the student surveys but the lowest on the teacher surveys. In other categories the teachers' scores were higher to those of the students in a range of 0.71 to 1.08.

The student scores indicate a more modest and more widely varying degree of success in the application of CLT principles to the classroom. The significantly higher teacher yields are probably attributable to the teachers coming from a smaller, more cohesive group with a strong positive regard for CLT principles. 

The fourth section of Table 1 assesses the extent the instructors supported the analytic to the synthetic approach, as defined by Long and Crooks (1992).

The results for the first five questions tend towards the center. The first three items scored 3.17 indicating a slight preference for something more synthetic. The next two items seem to indicate a slight preference towards a more analytic approach. The final item, which is written so that it reverses the direction of synthetic or analytic preference, indicates a stronger orientation towards a synthetic approach.

It would seem that the consensus, among the instructors, towards the general features of CLT did not extend towards the analytic orientation. More specifically some instructors seemed to want a workbook that was more structurally or functionally organized.

Teacher Questionnaire on L.C. Curriculum

For each question the respondent circled a number on a five-point scale: 5 4  3  2  1

5 indicates strong agreement
1 indicates strong disagreement

Twelve questionnaires were received out of a total of nineteen teachers

Learners develop language skills gradually through realistic use 4.55 (n=11)
I see the student (in my class) as one who
- assists the teacher 3.33
- assists other learners 4.50
- provides feedback to the teacher  4.00
- receives feedback from the teacher 4.42
- receives feedback from other learners 4.33
- acts to direct his own learning 4.17
- is an active participant in a learning community 4.58
As a teacher, I
- organizes resources 4.67
- am a resource for learners 4.83
- guide to class activities 4.67
- give feedback to students 4.50
- seek feedback from students 4.25
- research 3.83
- exploit learning opportunities 4.42
- shape future directions of learning 4.25
- participate in class activities 4.50
- facilitate communication between classroom participants 4.58
- facilitate interaction between students and learning material 4.83
The teaching material I used this term
- specified student tasks 4.58
- provided samples of English 4.42
- promoted internalization of English language characteristics 4.17
- encouraged student independence 4.50
- encouraged student motivation 4.33
- promoted English language skills 4.50
- I think the student workbooks need to
- be more systematic in their coverage of grammar 3.17
- be more systematic in their coverage of vocabulary 3.17
- be more concerned with correct usage 3.17
- ensure students master a skill before continuing 2.92
- to have the teacher play a more central role in the classroom 2.42
- be less structured 1.92

Student Questionnaire on L.C. Curriculum

Approximately 400 questionnaires were given to teachers who were asked to distribute them randomly to students during the week of final tests. 159 questionnaires were received.

For each number the respondent circled a number on a five-point scale: 5  4  3  2  1

5 indicates strong agreement
1 indicates strong disagreement

In English class I... Ave. Y1 Y2 Y3 u.k.
- helped the teacher (n = 156) 3.85 3.85 3.70 4.10 4.18
- helped other students (n = 159) 3.72 3.70 3.45 4.10 4.24
- gave other students my opinions about learning English well (n = 159) 3.25 3.32 3.33 3.10 3.71
- gave the teacher my opinions about the class (n = 159) 3.13 3.22 3.00 3.14 3.29
- received the teacher's opinions about my English (n = 157) 3.47 3.44 3.42 3.50 3.71
- got help from other students (n = 161) 3.55 3.55 3.53 3.77 3.76
- tried to help myself learn English (n = 159) 3.64 3.86 3.63 3.81 3.59
- was part of an English learning team (n = 159) 3.59 2.62 3.58 4.00 3.76
- think free is better than strict (n = 159) 3.59 3.89 3.83 4.00 3.89

The Y1 column contains the results from those students indicating they had used the first year textbook. Y2 and Y3 indicate the other textbooks. The u.k. column is the results from respondents who did not clearly indicate which textbook they used. Y1 and Y2 together comprise about two-thirds of the responses. The fifth question had two non-responses resulting in an n of 157. The n of 161 in the sixth question is likely the result of tabulation error.

Student - teacher Comparison of Questionnaire Results

The results below compare student and teacher aggregate responses to the questions meant to assess the extent to which student classroom conduct follows the CLT model. The wording in commas represents the form students received in their questionnaires.

  Teachers Students
help other students 4.50 3.72
help the teacher 3.33 3.85
feedback to the teacher (gave opinions to the teacher) 4.00 3.13
feedback from the teacher (received the teacher's opinion) 4.42 3.47
feedback from other learners (gave opinions about learning English) 4.33 3.25
directs own learning (helped myself to learn English) 4.17 3.46
active participant in a learning community (part of an English learning team) 4.58 3.59

The reason for different wording in the two versions of the questionnaire was an attempt to keep the language simple enough in the student version to be understood by the students. Although, such differences make comparisons invalid in the social sciences, language and cultural difference between the groups would also have had a confounding effect even if wordings were the same. The table is to illustrate the possible differences in the two groups perceptions about student learning behavior in broad terms.


The first workbook, Tell Me More is partitioned into twelve chapters with the following themes: greetings and introductions, class information and learning goals, family members, household items and my room, likes and dislikes, routines and hobbies, descriptions, food and restaurants, shops and clothes, directions; instructions, story telling, and phone conversations. It contains a section of extra activities at the end.  

The second workbook, Now You're Taking has twelve chapters with the themes: welcome, study skills, the newspaper, planning a trip, travel in Korea, making reservations, the hotel, sports and pastimes, at the doctor's, the post office and the bank, entertainment, and cultural identities and differences. It, too, contains a section of extra activities.

The third workbook, The Way Ahead is broken into eight project sections. It's content is beyond the scope of this analysis.

Earlier chapters seem to have a personal orientation while later chapters seem more community oriented.

Tasks tend to be thematically related to the chapters in which they are found. Some activities are self-contained while others build on previously covered material. Some tasks require students to refer back to previous work.

I will quickly reiterate Skehan's (1996) recommendations for designing tasks. Pre-task activities should improve the chances for restructuring and incorporation of new elements into the language system. They may, also, serve to reduce the cognitive load by familiarizing students with task requirements. Activities should be the appropriate level of difficulty so that students will be neither bored nor have to rely on improvised communication strategies. They allow students to use language elements to achieve a goal. Post activities are intended to redirect student attention back to task elements, reminding them of the importance of restructuring and accuracy.  

The assessments in the Analysis Section describe student behaviors for pre, during and post activities and analyze them according to complexity, stress, and cognitive goals. The word none is written where no pre or post tasks can be inferred. 

Every fifth lesson in the books, for a total of forty-three lessons, are described and assessed in the Analysis Section. Sixteen of the lessons are assessed as having a form focus. In doing them, students attend especially to form. In most cases (but not in every case) students, as well, engage in authentic L2 communication. Another nineteen seem to mediate between accuracy and fluency goals. These tasks require students to attend to form by use of multiple repetitions of a form in a communicative exercise (most often a survey). Four of the activities have discussion questions that relate to a topic. These are categorized as fluency oriented. An additional four activities are for the purpose of expanding or reviewing vocabulary. These activities are closely linked to chapter themes and support other tasks. Twenty-nine lessons have stated pre-tasks or are such that an experienced instructor can easily infer a suitable pre-task. Thirty-six have stated or inferable post-tasks. A few tasks have wider pedagogic purposes, for example, the one that requires a student to collect his classmates' email addresses.

Total 44 100%
Accuracy focus 16 36%
Mediated accuracy and fluency focus 19 43%
Fluency focus 4 9%
Vocabulary expansion focus 4 9%
Stated or inferable pre-task 29 66%
Stated or inferable post-task 36 82%

Tasks with wider pedagogic purposes (encouraging email use) and those that support chapter themes may superficially fall short of Skehan's recommendations. The above results, therefore, may under represent the extent his recommendations were followed in principle.

Generally, the tasks appear to be designed to bring about minimal communicative and cognitive stress in the students. Lexical and structural cues are usually present to assist students in task completion. Page layouts suggest what to do.

The term-end tests used during the period the workbooks were in use supported Skehan's principles. They required students to perform parallel tasks to those in the books. The tests usually covered a range of skills covered over the term. The students were told what they would be expected to do in advance. They were, thus, encouraged to re-analyze the form and functions presented to them.

In line with Long and Crooks recommendation for task-based testing, the tests were designed by experienced instructors and implemented program wide. 


The Andong program succeeded in the following ways:

Possible weaknesses are:

In summary, the Andong program was a bold and innovative task-based course. It generally met criteria discussed in the literature for such a program and was well suited the students using it. Certain refinements and additions would benefit future applications. 


Content analysis of workbooks using Skehan's (1996) criteria for task-based lessons

Tell Me More


Breen, M. P., & Candlin, C. N. (1980). The essentials of a communicative curriculum in language teaching. Applied Linguistics, 1(2), 89-110.

Campbell, R., & Wales, R. (1970). The study of language acquisition. In J. Lyons (Ed.), New horizons in linguistics. Harmondsworth: Penguin Books.

Canale, M., & Swain, M. (1980). Theoretical basis of communicative approaches to second language teaching and testing. Applied Linguistics, 1(1), 1-47.

Canale, M. (1983). From communicative competence to communicative language pedagogy. In J. C. Richards & R. W. Schmidt (Eds.), Language and communication (pp. 2-27). New York: Longman.

Corder, S. P. (1967). The significance of learners' errors. International Review of Applied Linguistics, 5(4), 162-170.

Ellis, R. (1985). Understanding second language acquisition. Oxford: OUP.

Feez, S. (1995). Systematic functional linguistics and its applications in Australian language education: A short history. Interchange, 27, 5-10.

Finocchiaro, M., & Brumfit, C. (1983). The functional-notional approach: from theory to practice. New York: OUP.

Halliday, M. A. K. (1979). Modes of meaning and modes of expression: types of grammatical structure, and their determination by different semantic functions. In D. J. Allerton, E. Carney & D. Holcroft (Eds.), Function and contexts in linguistic analysis (pp. 57-79). Cambridge: CUP.

Halliday, M. A. K. (1994). Functional grammar (2nd ed.). London: Edward Arnold.

Howatt, A. (1984). A history of English language teaching. Oxford: OUP.

Krashen, S. D. (1982). Second language acquisition theory. In Principles and practice in second language acquisition (pp. 9-32). London: Pergamon Press.

Littlewood, W. (1984). Foreign and second language learning (pp. 4-16). Cambridge: CUP.

Long, M. H., & Crooks, G. (1992). Three approaches to task-based syllabus design. TESOL Quarterly, 26(1), 27-56.

O'Malley, M. J., & Chamot, A. U. (1990). Learning strategies in second language acquisition. Cambridge: CUP.

Richards, J. C., & Rodgers, T. S. (1986). Approaches and methods in language teaching. New York: CUP.

Richards, J. C., & Sukwiwat, M. (1982). Language transfer and conversational competence. Applied linguistics, 4(2), 113-125.

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

Skehan, P. (1996). A framework for the implementation of task-based instruction. Applied Linguistics, 17(1), 38-62.

Stern, H. H. (1983). Models of second language learning and the concept of proficiency. In Fundamental concepts of language teaching (pp. 137-359). Oxford: OUP.

Yalden, J. (1983). The communicative syllabus: evolution, design, and implementation. Oxford: Pergamon Press.  

Yalden, J. (1987). Principles of course design for language teaching. New York: CUP.

| Back |