State of the art: SLA Theory and Second Language Syllabus Design

Writer(s): 
Peter Robinson, Aoyama Gakuin University

My aim in this brief overview of second language acquisition (SLA) theory and syllabus design is to compare four recent proposals for criteria for grading and sequencing the units of second language classroom activity. These are proposals for structural, lexical, skills, and task-based syllabuses. All four proposals show continuity with, and development from, similar earlier approaches. First, Ellis (1993, 1994, 1997) argues for a role for the structural syllabus, alongside a meaning-based syllabus. This grammatical approach to syllabus design has a long history in second language pedagogy (see Mackey, 1965; Richards & Rodgers, 1986) and is clearly the basis of many currently popular English courses, such as New Horizon (Asano, Shimomura & Makino, 1997), and language programs throughout Japan and elsewhere. Second, Willis (1990) describes a lexical approach to syllabus design, inspired largely by the work of the Birmingham corpus analysis project (see Sinclair, 1987, 1991; Sinclair & Renouf, 1988). This approach can be seen as a development of earlier work on lexical grading by Palmer (1917), Thorndike (1921), West (1953, 1960) and Willis (1990, p. vi). For a discussion of the history of vocabulary control see Nation (1990). Third, Johnson (1996) proposes a skills syllabus building on his own earlier proposals for communicative syllabus design and those of others (Johnson, 1982; Munby, 1978; Wilkins, 1976: Yalden, 1983) within an expanded theoretical framework. Finally, the recent proposals for task-based syllabus design of Long (Long, 1985, 1997, in press; Long & Crookes, 1992), and Skehan (1996a, 1996b, 1998), while differing in scope and details, both develop themes apparent in earlier work by Brown (Brown, Anderson, Shillcock & Yule, 1984), Carroll (1980), and Prabhu (1987) amongst others.

While these four approaches to syllabus design show evidence of theory and research-driven evolution from earlier proposals, there is also more convergence between them than the different labels might seem to imply. These are both reasons--theoretical development and an emerging consensus--for optimism about the progress that is being made in language pedagogy. Though they vary in the extent to which they draw their motivation from SLA research findings, all four proposals stress the pedagogic importance of tasks, whether these are the sole units of analysis for syllabus design, as in Long (1985, 1997, in press; Long & Crookes, 1992, 1993), or used as vehicles for delivering a sequence of other units, as in Ellis (1993, 1997), Johnson (1996), and Willis (1990). My further aim, then, is to distinguish between the uses of the term task and the role of task as an organising principle in syllabus design in these proposals and to make connections between each proposal and recent SLA research and theory.

Units and Sequence: Options n Syllabus Design

Syllabus design is based essentially on a decision about the 'units' of classroom activity, and the 'sequence' in which they are to be performed. There are options in the units to be adopted (see Long & Crookes, 1993; Long & Robinson, 1998; Nunan, 1988; White, 1988, for further discussion). Units can be based on an analysis of the language to be learned, in terms of grammatical structures, as in Ellis (1993, 1997), or of lexical items and collocations, as in Willis (1990). Units may also be based on an analysis of the components of skilled behaviour in the second language, for example the reading microskills described by Richards (1990) and Brown (1995), or the communicative skills forming part of Munby's (1978) communicative needs profiler, and Johnson's (1996) recent work. Units may also be holistic performative acts, such as serving meals on an airplane (Long, 1985, in press) or finding a journal article in a library using library technology (Robinson & Ross, 1996). They may be either generic, or based on needs analyses of specific groups of learners.

Along with choices in the units to be adopted, there are choices in the 'sequence' in which they can be presented. A syllabus can consist of a prospective and fixed decision about what to teach, and in what order, as in Long (1985, 1997, in press). In this case the syllabus will be a definition of the contents of classroom activity. A sequencing decision can also be made on-line, during classroom activity as in Breen's 'process' syllabus (Breen, 1984: Clarke, 1991; Littlejohn, 1983). In this case the initial syllabus will only guide, but not constrain the classroom activities. Finally, Candlin has proposed that a syllabus can be retrospective, in which case no syllabus will emerge until after the course of instruction. In this case the syllabus functions only as a record of what was done, imposing no controlling constraint on the classroom negotiation of content (Candlin, 1984; Clarke, 1991). None of the four proposals under review adopts retrospective sequencing, though the extent to which they differ with regard to prospective versus on-line decision making about sequencing will be discussed.

The Role of the Learner in Approaches to Syllabus Design

Another useful distinction in conceptualising options in syllabus design was made initially by Wilkins (1976) and refers to the learner's role in assimilating the content provided during group instruction and applying it individually to real world language performance and interlanguage development (also see Long & Crookes, 1992; Nunan, 1988, White, 1988: White & Robinson, 1995). Synthetic syllabuses involve a focus on specific elements of the language system, often serially and in a linear sequence, such as grammatical structures, or language functions. The easiest, most learnable, most frequent, or most communicatively important (sequencing decisions can be based on each of these ultimately non-complementary criteria, and on others) are presented before their harder, later learned, less frequent, and more communicatively redundant counterparts. These syllabuses assume the learner will be able to put together, or synthesize in real world performance, the parts of the language system they have been exposed to separately.

In contrast, analytic syllabuses do not divide up the language to be presented in classrooms, but involve holistic use of language to perform communicative activities. The learner's role in these syllabuses is to analyse or attend to aspects of language use and structure as the communicative activities require them to, in line with: a) their developing interlanguage systems; b) preferred learning style and aptitude profile; and c) to the extent that they are motivated to develop to an accuracy level which may not be required by the communicative demands of the task. For these reasons analytic approaches to syllabus design have been argued to be more sensitive to SLA processes and learner variables than their synthetic counterparts (Long & Crookes, 1993; Long & Robinson, 1998; Nunan, 1988; White, 1988; White & Robinson, 1995).The extent to which the four proposals for syllabus design under review imply these roles for the language learner is also discussed below.

SLA Theory and Approaches to Syllabus Design

How does SLA theory inform recent proposals for structural, lexical, skills and task-based syllabus design? The rationales for each proposal are described below.

1. Ellis' structural syllabus

Ellis (1993, 1994, 1997) draws extensively on SLA research and theory to motivate his arguments for a role for a structural syllabus. Ellis' argument rests on two distinctions: between explicit conscious knowledge, and implicit tacit knowledge (see deGraaff, 1997; Robinson, 1993, 1994, 1996b, 1997a; Schmidt, 1995); and between declarative knowledge of facts, and procedural knowledge of how to do things (see Anderson, 1983, 1992; DeKeyser, 1996, 1997, 1998). He argues that explicit, declarative knowledge of L2 grammar can influence the development of implicit declarative knowledge, and that, through communicative activity, implicit declarative knowledge can be proceduralised and used in spontaneous skilled performance. This is a 'weak interface' model, which allows explicit knowledge, under some conditions, to influence the development of tacit representations or competence. The main condition is that the learner must be developmentally ready to incorporate the explicit grammar instruction into their interlanguage. Ellis cites research by himself (1989), Pienemann (1989), and others showing that learners pass through stages of development in the acquisition of, amongst other things, word order rules, question forms, and negation. Unless grammatical instruction is timed to the learner's point of development it will not influence the developing implicit knowledge base. Since stages of development are learner internal and hidden from the teacher, timing is difficult to manage. However, Ellis argues explicit grammatical knowledge serves a number of other functions: it can be used to monitor production; it can help learners notice features in the input; and it can help learners compare their own production with a target model, and in some cases notice the gap between them. Knowing about grammar, Ellis argues, is therefore useful. Tasks promote consciousness-raising, and noticing of target grammar rules. Tasks are therefore pedagogic devices for teaching units of grammar (examples are described in Ellis & Noboyushi, 1993; Fotos & Ellis, 1991), and are used to implement a prospective synthetic structural syllabus.

2. Willis' lexical syllabus

Drawing on a different type of empirical evidence--large scale corpora of spoken and written language use--Willis also argues for a synthetic syllabus, where word and collocation are the units of analysis. Willis nowhere draws on SLA research to the extent Ellis does to motivate his proposal, but does conclude that SLA research findings show "input does not equal intakeモ and that "the assumption that language can be broken down into a series of patterns which can then be presented to learners and assimilated by them in a predictable sequence" is wrong (Willis, 1990, p. iii). Arguing against "a methodology which presents learners with a series of patterns" in a presentation, practice, production sequence Willis proposes taking "meaningful exposure as a starting point" (Willis, 1990, p. iv). Exposure should be organised in three ways: a) language is graded in difficulty; b) language exemplifying the commonest patterns is selected; and c) the language syllabus is itemised to highlight important features. Exposure is thus tightly controlled. Rather than linguistically grading the content of the syllabus Willis argues for lexically grading it using corpora of language use to identify word frequency at the 700 word, the 1,500 word, and the 2,500 word levels. Words in the corpora are itemised as collocations exemplifying each word's typical patterns of use. In effect, though, lexical grading leads to linguistic grading, since as Willis notes, by identifying the commonest words, "inevitably it focuses on the commonest patterns too...the lexical syllabus not only subsumes a structural syllabus, it also indicates how the structures which make up the syllabus should be identified" (1990, p. vi). In the lexical syllabus these three corpora are the bases of exposure at three levels of learner development. Willis claims that exposure is not sequenced or controlled within these levels, and the lexical syllabus "does not dictate what will be learned and in what order," rather "it offers the learner experience of a tiny but balanced corpus from which it is possible to make generalisations about the language as a whole" (Willis, 1990, p. vii). In other words, the learner corpus which forms the basis of exposure at each level is carefully itemised, but these items are not presented individually and serially.

So is there, then, a lexical syllabus, apart from the superordinate distinction between level 1, 2, and 3 corpora? Willis describes the development of the COBUILD Course (an exemplar of the lexical syllabus) as a process of first intuitively deciding on interesting topics, then developing tasks and choosing texts to complement them, and then highlighting lexical items within, e.g., the first 700 word level, as they occurred in the texts. This series of highlighted items is the syllabus, but sequenced according to no criteria that are discussed, apart from teacher intuition (see D. Willis, 1990, pp. 74-90). The methodology accompanying the syllabus (described in D. Willis, 1990; and in detail by J. Willis, 1996a, 1996b) involves a pre-task introduction to a topic, and exposure to texts; a task cycle where a task is planned, drafted and rehearsed; and a final language focus where learners consciously focus on forms used during the task. Course planning and content, hence the syllabus, is thus largely determined by the choices of texts and tasks--topics about which the lexical syllabus says nothing. This is, then, a language-focussed synthetic syllabus, but with some control given to the learner about which forms to attend to and focus on, since the itemised corpora at each level function as a guide, rather than as a prospective plan, allowing more on-line negotiation of content than Ellis allows. Surprisingly, given Willis' invocation of SLA research findings to support his approach, no account is taken of research into learnability and learning processes (a literature Ellis draws on) in selecting the collocations presented in corpora at each level of exposure, though these inevitably contain word order combinations, as well as tense and aspectual distinctions which are developmentally scheduled (for the SLA of tense/aspect see Anderson & Shirai, 1996; for the SLA of English collocations see Gitsaki, 1996).

3. Johnson's skill syllabus

Drawing on the work of Anderson (1983, 1992) and the declarative, procedural distinction referred to by Ellis (1997), Johnson argues that SLA and general skill learning draw on the same general cognitive mechanisms. Traditionally, skill acquisition has been viewed as a speed-up in the use of initially attention-demanding declarative knowledge. With practice, attentional demands diminish and declarative knowledge is proceduralised. Johnson argues that many aspects of second language learning can be viewed as the reverse process, from initially fast, unattended and unanalysed use, drawing on procedural knowledge alone, to declarative knowledge. This occurs when formulaic language is used fluently at first, without any knowledge of its internal structure. As this becomes attended to and analysed, declarative knowledge emerges. Declarative knowledge is valuable because it allows greater generalizability of language use, and is not context dependent, in contrast to procedural knowledge. Johnson concludes that his proposals support a skills syllabus, similar to, but going beyond the earlier attempts of Munby (1978) and Wilkins (1976) to specify the units of communicative syllabus design.

In essence, Johnson proposes a four-tier model of syllabus design. Occupying the first tier are what Johnson, following Munby, calls language specific skills, such as メidentifying the present perfect,モ or correctly contrasting /i/ and /i:/: "In our attempts to break language behaviour down into subskills, the general areas of phonetics/phonology and syntax would, then, follow traditional lines and would not pose any new difficulties for syllabus designers" (1996, p.164). But the old difficulties are surely difficulties enough. Are separate subskills to be identified for each phonetic contrast, for example? And how does 'learning difficulty' affect decisions about selecting and sequencing subskills? Another tier would contain semantic categories, such as notions and functions, "but only those about which pedagogically accessible generalizations can be made" (Johnson, 1996, p.165), that is, notions and functions which can be generalized to many contexts. An example given is inviting versus being polite. Johnson claims inviting need not be taught, and so need not be part of the syllabus, since it is largely phrasal and situation specific, whereas in being polite "useful generalizations...can be made about such things as 'being circumspect and indirect in approach'" (Johnson, 1996, pp. 165-166). A third tier would involve skills often referred to in 'process' approaches to teaching writing skills, such as generating new ideas, drafting essays, structuring and evaluating them. It seems then that skill is being used as a term to cover three different types of unit: language item, semantic category, and writing strategy. This is because Johnson is concerned with the transition from knowledge state--procedural to declarative and vice versa--that learning all these units has in common. The fourth and final tier of Johnson's skills syllabus concerns processing demands; the level of complexity of the classroom task should also be specified and enter into sequencing decisions. In summary, Johnson also favors a synthetic syllabus, prospectively organized, based on subskills at a number of levels: linguistic, semantic and pragmatic, and strategic. The daunting job for the syllabus designer is to inventory the subskills at each of these levels (as Munby 1978 attempted to do) then sequence them, and weave them together in a principled way.

4. The task-based syllabus

In many discussions of tasks, and examples of what claim to be task-based materials, tasks are used to force attention to, or to practice a particular structure, function or subskill. Skehan (in press) refers to these as 'structure-trapping' tasks. These include the tasks advocated by Ellis (1997), and Loschky and Bley-Vroman (1993), where the use of tasks to direct attention to grammatical form is theoretically motivated and an explicit part of the rationale for their use, as well as those in commercially available task-based courses, such as Richards, Gorden and Harper (1995), and Nunan (1996). In these latter cases, what were typically called exercises or activities in older coursebooks are now called tasks, but there is no difference between them. The organising principle of these coursebooks, apparent from the syllabus descriptions at the front, are grammatical structures, listening microskills, functions, topics, and often more. In contrast to structure-trapping tasks, and in contrast to coursebooks using task as a synonym for language exercise, Skehan and Long view tasks as purely meaningful activities. Tasks do not implement a covert grammatical or lexical syllabus, tasks alone are the units of syllabus design.

Long (1997; Long & Crookes, 1993; Long & Robinson, 1998) and Skehan (1996b, 1998) are in broad agreement about the SLA motivation for analytic syllabuses, and task-based syllabuses in particular, citing research showing: a) little resemblance between acquisitional sequences and instructional sequences based on linguistic forms (e.g., Ellis, 1989; Lightbown, 1983); b) evidence that learning is non-linear and cumulative, rather than linear and additive as synthetic language syllabuses imply (see Selinker & Lakshmanan, 1992 on backsliding; see Kellerman, 1985, on U-shaped behaviour); and c) research showing the influence of learnability on the order in which items can be learned (e.g., Mackey, 1995; Pienemann, 1989). Even if a structural syllabus could be sequenced based on what is known of learnability and language development it would be impossible to accurately time and target instruction at the stage learners are ready to progress to, since there is variation in rate of acquisition, meaning groups of learners do not progress in lockstep, homogeneously through acquisition sequences (see Long, 1988, 1997; Long & Crookes, 1992; Long & Robinson, 1998; Robinson, 1994; Rutherford, 1988; Skehan, 1996a, 1996b, 1998). Additionally, as Long (1997) points out, linguistic grading, as required by many synthetic structural approaches, at least in the early stages, results in classroom language and texts which are artificial, and functionally and linguistically impoverished, prohibiting exposure to language learners may be ready to learn. Given their broad agreement over the motivation for choice of task-based syllabuses, there are some differences of scope and focus in their proposals.

Long (1985, 1997; Long & Crookes, 1992) describes a number of steps to be taken in implementing task-based language teaching. First conduct a needs analysis to identify the target, real world tasks learners need to perform in the second language, then classify the target tasks into types or superordinate categories such as 'making/changing reservations.' From the target tasks derive pedagogic tasks: "Adjusted to such factors as learners' age and proficiency level, these are a series of initially simple, progressively more complex approximations to the target task" (Long, 1997, p.10). These tasks are then sequenced to form a syllabus, and the program is implemented with appropriate methodology and pedagogy. One methodological principle Long advocates is 'focus on form.' That is, where individuals or groups of learners are heard repeatedly producing non-target-like forms, teacher intervention to provide corrective feedback is recommended. This can take several forms, such as implicit negative feedback, or recasts of learner forms, brief written illustration of the correct form, brief rule explanations, input enhancement of forms in aural and written texts used on task, and a variety of other techniques. For research on input enhancement see Jourdenais, Leeman, Arteagoitia, Fridman, & Doughty (1995); and White, Spada, Lightbown & Ranta (1991). For research on corrective feedback see Carroll & Swain (1993); Lightbown & Spada (1990); and Mackey (1998). For research into the use of pedagogic rules see DeKeyser (1995); and Robinson (1996a). For summaries see Doughty & Williams (1998); and Long & Robinson (1998).

While Long places great importance on the opportunities to focus on form in the context of meaningful interaction that task work provides, in line with his 'interactionist' theory of L2 development (Long, 1996; see also Gass, 1997), Skehan takes a more cognitive, information processing approach to task-based instruction. These are not oppositional perspectives, of course, since there is a substantial amount of cognitivist research into task design and performance within Long's framework. However, Skehan has steadily pursued a research agenda aimed at identifying the effects of factors such as planning time (Crookes, 1989; Foster & Skehan, 1996) on the complexity, accuracy and fluency of learner production, as well as the influence of learner variables such as aptitude on language processing (Skehan, 1996a, 1996b, 1998, in press; Skehan & Foster, 1997, 1998). Like Long, Skehan rejects linguistic grading as a criterion for task and syllabus design, defining a task as an activity in which, "Meaning is primary; There is a goal which needs to be worked on; the activity is outcome-evaluated; There is a real world relationship" (Skehan, in press). Skehan concludes that this definition rules out "an activity that focuses on language itself" such as a transformation drill, or the consciousness-raising tasks described by Ellis (1997), and many of the tasks in Nunan (1996).

Summary: Units and sequencing in recent approaches to syllabus design

Each of the approaches to syllabus design I have described chooses different units of analysis. How are these sequenced? A brief summary of this complex issue is given here.

1. The structural syllabus.

Ellis acknowledges that the issue of how to sequence units of grammatical instruction is problematic, and suggests using traditional criteria, such as the intuitively judged relative difficulty, and the relative frequency of grammar items. In addition he suggests marked features should receive explicit instruction, since 'unmarked features may be learned by most learners naturally, and therefore do not require explicit attention'(1993, p.106), a suggestion which begs at least two questions-which definition of markedness is to be adopted, and is it true that unmarked features are learned naturally, without being explicitly attended? If Ellis means learners can learn them without paying focal attention to them in the input Schmidt would answer no to the last question, since he argues all learning requires focal attention accompanied by awareness of the form of input (Robinson, 1995b; Schmidt, 1990, 1995). Finally Ellis suggests learner's errors should be targeted as the forms for instruction, suggesting a need for on-line modification of the syllabus, as these errors occur. Taken together, these are weak, and potentially non-complementary sequencing criteria.

2. The lexical syllabus

As described above, the basis for sequencing items in the lexical syllabus is frequency, and coverage. Those lexical items occurring most frequently are presented first, in their most common sentence patterns. However, this applies only to the establishment of the corpora at the 700 word, 1,500 word, and 2,500 word level. Within each level it is not clear on what criteria items are chosen for inclusion in texts, or why tasks making use of the texts are sequenced in the way they are.

3. The skill syllabus

The skill syllabus is the least explicit of the four proposals about sequencing criteria. Is one level, e.g., language subskills, to be developed and sequenced before others, such as pragmatic and strategic subskills? Like earlier proposals for notional/functional syllabuses, Johnson seems to agree that some notions and functions are more core than others, and should be taught first, but as with those earlier proposals he offers no psycholinguistic rationale or SLA research evidence for what they are, and what sequence they should be taught in. As Paulston noted (1981) the notional/functional approach of Wilkins is メatheoretical as regards learning theory,モ a point which Johnson concedes (see Johnson, 1996, p. 174), and has attempted to address. Nonetheless, problems remain.

4. The task-based syllabus

Research into the criteria determining task sequencing has been increasing in recent years, and findings have emerged. One line of research has been to identify cognitive dimensions of the difficulty of tasks, and to assess the effects of tasks performed at easy and complex ends of each dimension on measures of learner language (Robinson, 1995a, 1997b, 1998a, 1998b; Robinson, Ting & Urwin, 1995). A general finding is that easier tasks tend to result in more fluent speech, since cognitive and processing demands are low. More complex tasks force learners to attend to the language used on task, resulting in less fluent but more complex and accurate production. Thus tasks can foster fluency and accuracy. Incorporating these dimensions into task design and implementation provides a way of slowly increasing the difficulty, and authenticity of the task being practised. Some dimensions of the cognitive complexity that have been proposed and researched include: a) planning time (tasks with planning time are easier than tasks without planning time); b) single versus dual task (tasks making only one demand, such as describing a route marked on a map to another person, are easier than tasks with two demands, such as thinking up the route and describing it at the same time); c) prior knowledge (tasks in a domain the learner has prior knowledge of are easier than tasks in a domain the learner has no prior knowledge of); d) number of elements (tasks involve few elements are easier than tasks involving many elements). Using these dimensions, an example of the staged increase in the complexity of a task (giving directions to another person using maps) is given in Figure 1. For other research into sequencing tasks see Brown, Anderson, Shillcock & Yule (1984); and Skehan (1996a, 1998). For taxonomies of features intuitively judged to influence task difficulty see Brindley (1987); Nunan (1989); Prabhu (1987); and Long (1985).

 

 

 


Versions Of Map Task
 

 


Simple
     

 


Complex
Dimensions of complexity

 


1

 


2

 


3

 


4

 


5
planning time (before speaking)

 


+

 


-

 


-

 


-

 


-
single task (route marked)

 


+

 


+

 


-

 


-

 


-
prior knowledge (of familiar area)

 


+

 


+

 


+

 


-

 


-
few elements (a small area)

 


+

 


+

 


+

 


+

 


-
  (simplified data/map) (authentic data/map)
(From Robinson, 1998b)

 

 


Figure 1: Five map tasks at increasing levels of complexity

 

Conclusion

Clearly, decisions about the units and sequence of classroom activity must accommodate what is known of learning processes, since these are what they are trying to facilitate. Of the proposals for syllabus design reviewed here, SLA research has had the strongest influence on task-based approaches. The structural, lexical and skills syllabuses all show signs of theoretical, and research-driven development from earlier proposals, but in a number of cases SLA research findings pose problems or raise unanswered questions for them. Further development and evaluation of these proposals will involve a research agenda in which SLA research plays a large role. I have been most optimistic about task-based syllabus design, since it appears most in line with what we know of SLA processes, and since it offers the prospect of meeting the goals of each of the synthetic syllabuses reviewed (i.e., the development of L2 structural, lexical and skill ability) in the context of practice on tasks with real world relevance and application. There are signs of convergence too, between the proposals reviewed, evident in the common interest in the use of tasks to implement each syllabus, especially the lexical syllabus (see Willis, 1996a, 1996b). For this reason information about task complexity, critical for sequencing tasks in the task-based syllabus, will also be of interest to syllabus designers adopting other units of analysis and is an area where further SLA research will be of great value to pedagogy.

 

References

Andersen, R., & Shirai, Y. (1996). The primacy of aspect in first and second language acquisition: The pidgin-creole continuum. In W. Ritchie & T.K. Bhatia (Eds.), Handbook of second language acquisition, (pp. 527-570). New York: Academic Press.

Anderson, JR (1983). The architecture of cognition. Hillsdale, NJ: Lawrence Erlbaum.

Anderson, J.R. (1992). Automaticity and the ACT* theory. American Journal of Psychology, 105, 165-180.

Asano, H., Shimomura, Y., & Makino, T. (1997). New Horizon 1. Tokyo: Tokyo Shoseki.

Breen, M. (1984). Process syllabuses for the language classroom. In C.J. Brumfit (Ed.), General English Syllabus Design, (pp. 47-60). Oxford: Pergamon.

Brindley, G. (1987). Factors affecting task difficulty. In D. Nunan (Ed.), Guidelines for the development of curriculum resources, (pp. 45-56). Adelaide: National Curriculum Resource Center.

Brown, G., Anderson, A., Shillcock, R., & Yule, G. (1984). Teaching talk: Strategies for production and assessment. Cambridge: Cambridge University Press.

Brown, J.D. (1995). Elements of language curriculum. Boston, MA: Heinle and Heinle.

Candlin, C. (1984). Syllabus design as a critical process. In C.J. Brumfit (Ed.), General English Syllabus Design, (pp. 29-46). Oxford: Pergamon.

Carroll, B. (1980). Testing communicative performance. Oxford: Pergamon.

Carroll, S., & Swain, M. (1993).Explicit and implicit negative feedback: An empirical study of the learning of linguistic generalizations. Studies in Second Language Acquisition, 16, 205-230.

Clarke, D. (1991). The negotiated syllabus: What is it and is it likely to work? Applied Linguistics, 12, 13-28.

Crookes, G. (1989). Planning and interlanguage variation. Studies in Second Language Acquisition , 11,183-199.

De Graaff, R. (1997). The eXperanto experiment: Effects of explicit instruction on second language acquisition. Studies in Second Language Acquisition, 19, 249-268.

DeKeyser, R. (1995). Learning second language grammar rules: An experiment with a miniature linguistic system. Studies in Second Language Acquisition, 17, 379- 410.

DeKeyser, R. (1996). Exploring automatization processes. TESOL Quarterly, 30, 349- 357.

DeKeyser, R. (1997). Beyond explicit rule learning: Automatizing second language morphosyntax. Studies in Second Language Acquisition, 19, 167-189.

DeKeyser, R. (1998). Automaticity. In P. Robinson (Ed.), Cognition and second language instruction. Under submission to Cambridge University Press.

Doughty, C., & Williams, J. (1998). Pedagogical choices in focus on form. In C. Doughty & J. Williams (Eds.), Focus on form in classroom SLA., (pp. 197-262).New York: Cambridge University Press.

Ellis, R. (1989). Are classroom and naturalistic acquisition the same? A study of the classroom acquisition of German word order rules. Studies in Second Language Acquisition, 11, :305-328.

Ellis, R. (1993). The structural syllabus and second language acquisition. TESOL Quarterly, 27, 91-113.

Ellis, R. (1994).A theory of instructed second language acquisition. In N. Ellis (Ed.), Implicit and explicit learning of language, (pp. 79-115). London: Academic Press.

Ellis, R. (1997). SLA research and language teaching. Oxford: Oxford University Press.

Ellis, R., & Noboyushi, J. (1993). Focused communication tasks. ELT Journal, 47, 203-210.

Foster, P., & Skehan, P. (1996). The influence of planning time and task type on second language performance. Studies in Second Language Acquisition, 18, 158-184.

Fotos, S., & Ellis, R. (1991).Communicating about grammar: A task-based approach. TESOL Quarterly, 25, 87-112.

Gass, S. (1997). Input, interaction and the language learner. Mahwah, NJ: Lawrence Erlbaum.

Gitsaki, C. (1996). The second language acquisition of English collocations by Greek learners of ESL. Unpublished Ph.D. dissertation, University of Queensland.

Johnson, K. (1982). Communicative syllabus design and methodology. Oxford: Pergamon.

Johnson, K. (1996). Language teaching and skill learning. Oxford: Basil Blackwell.

Kellerman, E. (1985). Input and second language acquisition theory. In S. Gass & C. Madden (Eds.), Input in second language acquisition, (pp. 345-353). Rowley, MA: Newbury House.

Leeman, J., Arteagoitia, I., Fridman, B., & Doughty, C. (1995). Integrating attention to form with meaning: Focus on form in content based Spanish instruction. In R. Schmidt (Ed.), Attention and awareness in foreign language learning. (pp. 217- 258). University of Hawai'i at Manoa, Second Language Teaching and Curriculum Center, Technical Report # 9.

Lightbown, P. (1983). Exploring relationships between developmental and instructional sequences. In H.G. Seliger and M.H. Long (Eds.), Classroom oriented research in second language acquisition, (pp. 217-243). Rowley, MA: Newbury House.

Lightbown, P., & Spada, N. (1990). Focus-on-form and corrective feedback in communicative language teaching: Effects on second language learning. Studies In Second Language Acquisition, 12, 429-448.

Littlejohn, A. (1983). Increasing learner involvement in course management. TESOL Quarterly, 17, 389-403.

Long, M.H. (1985). A role for instruction in second language acquisition. In K. Hyltenstam & M. Pienemann (Eds.), Modeling and assessing second language acquisition, (pp. 77-99). Clevedon, Avon: Multilingual Matters.

Long, M. (1988). Instructed interlanguage development. In L. Beebe (Ed.), Issues in second language acquisition: Multiple perspectives, (pp.115-141). Rowley, MA: Newbury House.

Long, M.H. (1996). The role of the linguistic environment in second language acquisition. In W. Ritchie & T.K. Bhatia (Eds.), Handbook of second language acquisition, (pp. 413-468). New York: Academic Press.

Long, M.H. (1997). Focus on form in task-based language learning. Paper presented at the McGraw-Hill Teleconference on Approaches to grammar instruction in communicative language teaching, California State University, Long Beach, October 29, 1997.

Long, M.H. (in press). Task-based language teaching. Oxford: Blackwell.

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

Long, M.H. & Crookes, G. (1993). Units of analysis in syllabus design: The case for task. In G. Crookes & S. Gass (Eds.), Tasks in a pedagogical context, (pp. 9-54). Clevedon, Avon: Multilingual Matters.

Long, M.H., & Robinson, P. (1998). Focus on form: Theory, research, and practice. In C. Doughty & J. Williams (Eds.), Focus on form in classroom SLA., (pp. 15-41). New York: Cambridge University Press.

Mackey, W. (1965). Language teaching analysis. London: Longman.

Mackey, A. (1995). Stepping up the pace: Input, interaction and interlanguage developmentムAn empirical study of questions in ESL. Unpublished Ph.D. dissertation, University of Sydney, Department of Linguistics.

Mackey, A. (1998). Interactional modifications and the development of questions in English as a second language. Manuscript under submission .

Munby, J. (1978). Communicative syllabus design. Cambridge: Cambridge University Press.

Nation, P. (1990). Teaching and learning vocabulary. New York: Newbury House.

Nunan, D. (1988). Syllabus design. Oxford: Oxford University Press.

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

Nunan, D. (1996). Atlas. Boston, MA: Heinle and Heinle.

Palmer, H. (1917). The scientific study and teaching of languages. London: Harrap.

Paulston, C.B. (1981). Notional syllabuses revisited: Some comments. Applied Linguistics, 2, 93-95.

Pienemann, M. (1989). Is language teachable? Applied Linguistics ,10, 52-79.

Prabhu, N.S. (1987). Second language pedagogy. Oxford: Oxford University Press.

Richards, J. (1990). The language teaching matrix. Cambridge: Cambridge University Press.

Richards, J., Gordon, D., & Harper, A. (1995). Listen for it: A task-based listening course. Oxford: Oxford University Press.

Richards. J. & Rodgers, T. (1986). Approaches and methods in language teaching. Cambridge: Cambridge University Press.

Robinson, P. (1993). Problems of knowledge and the implicit/explicit distinction in SLA theory construction. University of Hawaii Working Papers in ESL, 12, 90-138.

Robinson, P. (1994). Implicit knowledge, second language acquisition and syllabus construction. TESOL Quarterly, 28, 160-166.

Robinson, P. (1995a). Task complexity and second language narrative discourse. Language Learning, 45, 99-140.

Robinson, P. (1995b). Attention, memory and the 'noticing' hypothesis. Language Learning, 45, 283-331.

Robinson, P. (1996a). Learning simple and complex second language rules under implicit, incidental, rule-search and instructed conditions. Studies in Second Language Acquisition, 18, 27-67.

Robinson, P. (1996b). Consciousness, rules and instructed second language acquisition. New York: Peter Lang.

Robinson, P. (1997a). Individual differences and the fundamental similarity of implicit and explicit adult second language learning. Language Learning, 47, 45-99.

Robinson, P. (1997b). Giving pushed output a push: The role of task complexity. EUROSLA Newletter, 3, 2, 36-40.

Robinson, P. (1998a). Task complexity, cognition, and second language syllabus design. In P. Robinson (Ed.), Cognition and second language instruction. Under submission to Cambridge University Press.

Robinson, P. (1998b). Task complexity and second language syllabus design. Under submission to Cambridge University Press.

Robinson, P. & Ross, S. (1996). The development of task-based testing in EAP programs. Applied Linguistics, 17, 455-476

Robinson, P., Ting, S., & Urwin, J. (1995). Investigating second language task complexity. RELC Journal, 25. 62-79.

Rutherford, W. (1988). Second language grammar: Learning and teaching. NewYork: Longman.

Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129-158.

Schmidt, R. (1995). Consciousness and foreign language learning: A tutorial on the role of attention and awareness in learning. In R. Schmidt (Ed.), Attention and awareness in foreign language learning. (pp. 1-63). University of Hawai'i at Manoa, Second Language Teaching and Curriculum Center, Technical Report # 9.

Selinker, L., & Lakshamanan, U. (1992).Language transfer and fossilization. In S. Gass & L. Selinker (Eds.), Language transfer in language learning, (pp. 196-215). Amsterdam/Philadelphia: John Benjamins.

Sinclair, J. (Ed.). (1987). Looking up: An account of the COBUILD project in lexical computing. London: Collins.

Sinclair, J. (1991). Corpus, concordance, collocation. Oxford: Oxford University Press.

Sinclair, J., & Renouf, A. (1988). A lexical syllabus for language learning. In R. Carter & M. McCarthy (Eds.), Vocabulary and language teaching, (pp. 140-160). London: Longman.

Skehan, P. (1996a).Second language acquisition research and task-based instruction. In D. Willis & J. Willis (Eds.), Challenge and change in language teaching, (pp. 17-30). Oxford: Hienemann.

Skehan, P. (1996b). A framework for task-based approaches to instruction. Applied Linguistics, 17, 34-59.

Skehan, P. (1998). A cognitive approach to language learning. Oxford: Oxford University Press.

Skehan, P. (in press). Task-based instruction. In W. Grabe (Ed.), Annual Review of Applied Linguistics. New York: Cambridge University Press.

Skehan, P., & Foster, P. (1997). Task type and task processing conditions as influences on foreign language performance. Language Teaching Research, 1, 185-212.

Skehan, P., & Foster, P. (1998). Cognition and tasks. In P. Robinson (Ed.), Cognition and second language instruction. Under submission to Cambridge University Press.

Thorndike, E.L. (1921). The teacher's wordbook. New York: teachers College, Columbia University.

West, M, (1953). A general service list of English words. London: Longman.

West, M. (1960). Vocabulary selection and the minimum adequate vocabulary. ELT Journal, 8, 121-126.

White, L., Spada, N., Lightbown, P.M., & Ranta, L. (1991). Input enhancement and L2 question formation. Applied Linguistics, 12, 416-432.

White, R. (1988). The ELT curriculum: Design, management, innovation. Oxford: Blackwell.

White, R., & Robinson, P. (1995). Recent approaches to syllabus design. RELC Guidelines, 17, 93-101. (Video recording, RELC library, Singapore).

Wilkins, D. (1976). Notional syllabuses. Oxford: Oxford University Press.

Willis, D. (1990). The lexical syllabus: A new approach to language teaching. London: Collins.

Willis, J. (1996a). A flexible framework for task-based learning. In D. Willis & J. Willis (Eds.), Challenge and change in language teaching, (pp. 163-76). Oxford: Hienemann.

Willis, J. (1996b). A framework for task-based learning. Oxford: Longman.