Vivian Cook's Home Page  Obscure Writings Index

Review of Peter Skehan, A Cognitive Approach to Learning Language, OUP, 1998, 324 pp

For some time discussion of task-based learning has been expanding from articles to books for teachers. In Peter Skehan’s book it has reached a book-length academic justification, bringing together work from applied linguistics, psychology and language teaching to provide a basis that many people can now cite as a support for their work. It is a useful and timely book that enables one to see how such an approach can be justified.

The book overall tries to argue for a ‘processing account’ of second language learning. Because of its wide scope, the book falls into distinct sections. The first group of three chapters concern ‘psycholinguistics’. Chapter 1 is a critique of comprehensible input and strategies-based research, arguing that syntax should play a minimal role. Chapter 2 argues for a greater role for lexicalised knowledge, seen as a memory-based collection of items. Chapter 3 presents a model of ‘psycholinguistics processes’, emphasising the stages in information processing, particularly the importance of attention. Chapter Four rejects three models of L2 learning, Universal Grammar, Multidimensional Model and Analysis/Control, in favour of a ‘dual-mode’ model having rule-based and exemplar-based systems. The next group of chapters concerns language teaching. Chapter 5 presents the history of task-based teaching based on a variety of classroom-based research, relating three goals of fluency, accuracy and complexity to the dual-mode model. Chapter 6 establishes a set of five principles for task-based instruction concerning the choice of target structures and tasks, task sequencing, continuous evaluation by the students, and a need for focus on form. Chapter 7 discusses various models of testing, considering how they need to be modified by the three goals in a model that balances rules and exemplars. The next batch of three chapters covers individual variables in L2 learners. Chapter 8 defends aptitude research against its critics; Chapter 9 looks at aptitude in relationship to the age factor; Chapter 10 reviews learning style research, relating differences to task characteristics. The last two chapters come back to teaching, Chapter 11 looking at issues about syllabuses and projects in the light of learner differences and Chapter 12 reiterating key concepts such as dual-coding and the tension between learning and performance.

It is then a continually stimulating argument, with much interesting background. It is not clear that these different threads cater for the same reader. Some sections read like surveys of areas for readers who are not up-to-date with the literature, such as aptitude. The three main sections work perfectly well in themselves but the aims at the beginning suggest far wider implications than just for learner variables and task-based teaching. How for example can the model be integrated with the more traditional approaches to language teaching? How could it be used in language teaching outside the classroom? As Spolsky (1989) wisely remarks, ‘any theory of second language learning that leads to a single method must be wrong.’ A processing approach should not tie itself to a single teaching method.

The type of language teaching concerned nowhere seems to be explicitly stated. Implicitly it seems to be adult EFL teaching in a classroom, rather than mainstream secondary education or modern language teaching or autonomous learning, since no language appears to be mentioned other than English and the stress is on older learners. Again a book giving an overall theory needs to encompass a wide range of students, situations and languages. Because of this assumed background, it is never questioned why students should be learning another language in the first place. The goals that are mentioned are fluency, accuracy and complexity. But people need to be fluent, accurate or complex because they want to use a second language for buying and selling, for translating poetry, for passing an exam, for listening to operas, for travelling, for praying, for writing a novel, for organising a revolution, or any of the other myriad reasons for which people learn second languages. Task-based learning may work like a dream in the classroom but how well does it serve the students’ general goals? Will they be content if their fluency, accuracy and complexity are much improved but this does nor help them to achieve their needs through a second language? In other words language teaching seems to be goal-free and value-free; the end of it is to be able to carry out a number of skilled tasks in the classroom. Skehan (p.96) thinks that it is desirable for tasks to have real-world relevance ‘but difficult to obtain in practice’.

A major issue is the connection between this type of work in applied linguistics and the rest of the academic community. The bulk of the theories and references are either from applied linguistics classroom-based research or from educationally-oriented psychology. Psycholinguistics for example has made enormous strides in the past few years; perhaps the only reference to orthodox psycholinguistics here as opposed to applied linguistics or to psychology is to Clark and Clark (1977), hardly the most up-to- date of references. It is a shame that the author has not made more use of academic psycholinguistics, some of which complements his ideas, some of which contradicts them. For example the dual-model contrasting rules and chunks is hardly novel in psycholinguistics: Goldman-Eisler (1968) traced it back to Hughlings Jackson in the late nineteenth century; it is the mainstay of most models of word recognition. In recent years it has become associated with the work of Steven Pinker, popularised in The Language Instinct (Pinker 1994), which is indeed in the references here but not cited in this respect. Dual processing has become a hot issue, with connectionists claiming that rules are unnecessary. Skehan should perhaps be defending himself against connectionist one-system models, not against the straw-man of a linguistic syntax system.

Within SLA the coverage is equally selective. One gap is indeed vocabulary acquisition, a vast area of current research with hardly a mention here. Another is the proliferating literature in L2 cognitive processing, summarised for example in contributions to de Groot & Kroll (1997). A processing theory of L2 learning surely needs to take in the specific L2 research as well as accounts of classrooms. In particular there is work with L2 memory using the Baddeley model alluded to here, for example, Service and Craik (1993), and with the Levelt theory (de Bot, 1993), which, if not relevant, needs to be dismissed.

Needless to say, the area that is least drawn on in any applied linguistics these days is mainstream linguistics other than corpus-based analysis of English, with no reference to aspects of language other than ‘structures’ and a lexicon consisting of exemplars. The emphasis on vocabulary is hardly as anti-linguistic as Skehan seems to feel. Most views of syntax in the past twenty years have relied on a division into a set of rules and a set of lexical items, even the UG theory that is pushed aside here several times; ‘The I-language consists of a computational process and a Lexicon’ (Chomsky, 1995, p.15); indeed the current Minimalist Program is lexically-driven in that the properties of lexical items shape the sentence rather than lexical items being slotted into pre-existent structures as hinted here.

To a large extent this review has consisted of praise for what is there and regret for what is not. If this theory is to convince people who are not already committed to this approach, it has to show that it is at least compatible with what is going on in other related fields and with language teaching in general. As a first approximation it may be possible to base a language learning theory on restricted evidence but, interesting and well-put as the argument is here, we look forward to the book that will integrate the other areas of language processing and second language acquisition and show how the model applies to a range of teaching methods, situations, and students.


Chomsky, N. (1995). The Minimalist Program, MIT Press

Clark, H.H. & Clark, E.V. (1977). Psychology and Language, New York, Harcourt Brace Jovanovich

de Bot, K. (1992). ‘A bilingual production model: Levelt’s ‘speaking’ model adapted’, Applied Linguistics, 13, 1, 1-24

de Groot, A. & Kroll, J.F. (eds.) (1997). Tutorials in Bilingualism: Psycholinguistic Perspectives, Hillsdale: Lawrence Erlbaum

Goldman-Eisler, G. (1968), Psycholinguistics: Experiments in Spontaneous Speech, London: Academic Press

Pinker, S. (1994). The Language Instinct, Harmondsworth: Penguin

Service, E. & Craik, F.I.M. (1993). Differences between young and older adults in learning a foreign vocabulary. Journal of Memory and Language, 33, 59-74

Spolsky, B. (1989). Conditions for Second Language Learning, Oxford: Oxford University Press.