LINGUIST List 12.2288

Mon Sep 17 2001

Review: Ellis, Instructed 2nd Lang Acquisition

Editor for this issue: Terence Langendoen <>

What follows is another discussion note contributed to our Book Discussion Forum. We expect these discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for discussion." (This means that the publisher has sent us a review copy.) Then contact Simin Karimi at or Terry Langendoen at


  1. Steve Bird, review of Rod Ellis - Instructed Second Language Acquisition

Message 1: review of Rod Ellis - Instructed Second Language Acquisition

Date: Mon, 17 Sep 2001 01:26:33 -0700 (PDT)
From: Steve Bird <>
Subject: review of Rod Ellis - Instructed Second Language Acquisition

Ellis, Rod (1990) Instructed Second Language Acquisition, 7th reprinting.
Blackwell Publishers, paperback, ix+230 pp., ISBN: 0-631-16202-X, $32.95,
Applied Language Studies.

Stephen A. Bird, University of Cambridge

If one can use the phrase 'classic text' at this point in the history of
applied linguistics literature, Rod Ellis's Instructed Second Language
Acquisition would aptly fit the description. It is one of the most readable
and popular (this is its 7th printing) introductory books in applied
linguistics; it provides an excellent historical overview of theory and
practice in the field while remaining very accessible for newcomers; and, as
with all classic texts, it is very dated -- the book has not been revised
since its publication in 1990.

The book asks a very difficult question: "How does second language learning
take place in a classroom?"(p.1). Chapter 1 explores some of the ways and
means of answering the question. Ellis concisely surveys and critiques the
major approaches that have been used in applied linguistics research.
Although many of the references are almost 30 years old, the basic
distinctions between research paradigms still hold true -- for example,
descriptive, ethnographic approaches to documenting classroom processes
versus 'scientific' hypothesis testing procedures. Ellis makes it clear he
sees a place for both sorts of research: "Given the complexity of the task
facing the researcher it would seem wise to employ as many and varied
strategies of research as are available' (p.5). However, descriptive,
process-oriented classroom observation at some point needs a framework
through which to observe classroom behaviour if it is going to yield useful
pedagogical and theoretical commentary. Ellis argues that a clear and
explicit theory of classroom learning needs to be formulated, " that
statements about how learners learn and how teachers ought to teach can be
subjected to critical scrutiny" (p.1). To anticipate the second to last
chapter of the book, Ellis has a theory of his own to put forward.

While useful and interesting, Chapter 1 is, like most of the book, marred by
its dated research citations: on page 13 the reader is introduced to "recent
surveys of research..." into formal instruction - a core topic of the book -
but the citations are from 1983, 1985 and 1988. No doubt many current
readers will be put off by such a dated review. If the text is aimed at
newcomers to applied linguistics, it is advisable to revise the book to
include recent citations, if only to bolster confidence that the discussion
is still relevant (which, in this reviewer's view, it is).

What might a theory of instructed second language acquisition look like?
Ellis devotes Chapter 2 to a cursory inspection of early behaviourist
learning theory applied to language teaching methods in the 1950s. The
chapter is invaluable to new students of applied linguistics because it is
in this era that students of human psychology came to realise that the
dominant learning theory of the day could not explain language learning in a
satisfactory way -- it could not adequately explain, for example, the
infinitely creative use of language simply as habit formation based on
experience and reinforcement -- and therefore was perhaps not the best theory
to employ when thinking about how to teach and learn a foreign language.
Ellis covers all of the main topics in a clear and readable way,
demonstrating the links between behaviourist psychology and teaching methods
like audiolingualism. The chapter encourages the reader to think about just
how complex the process of acquiring a language is, and how difficult it is
to articulate an explanatorily adequate theory of classroom learning.

Chapter 3 continues the historical survey, introducing the reader to
influential mentalistic theories of learning that emerged in the late 1960s
and which generated myriad models of the mind and its capacity to acquire
language. The reader is introduced to Chomsky's views on child language
acquisition and the effects of those views on theories of adult second
language learning. Krashen's Monitor Model and Cognitive Anti-method are
both described, theories based on Chomsky's essential idea that language
grows in the mind on the basis of simple exposure rather than explicit
teaching. Ellis lays out the relationship between these theories and
methods, sketching the links between Chomsky's work and Krashen's proposals
for how best to facilitate second language acquisition. The chapter offers
the reader a chance to reflect on the way in which, either implicitly or
explicitly, the assumptions we make about language learning drive the way we
approach classroom teaching methods. One glaring assumption made in the
early 70s was that first and second language acquisition are identical
cognitive processes. Moreover, the application of L1 theories to classroom
second language learning methods required enormous assumptions about the
transfer of cognitive learning mechanisms across contexts. Many researchers
now agree that these assumptions are untenable because of obvious key
differences between the two learning environments, and Ellis ends Chapter 3
by arguing that only by entering and studying the classroom context can one
really understand classroom language learning.

Chapter 4 describes classroom process research, a research approach
initially designed to collect behavioural data for study, much as the
naturalist goes out and collects butterflies before formulating a theory of
evolution. The aim of early researchers in this paradigm was simply to
collect enough data to begin to see patterns of behaviour that might lead to
explanatory theories, hence better teaching methods. Here the reader comes
up against a fundamental problem with classroom processes and behaviour:
unlike butterfly collecting, where the examples are fairly clearly
delineated on the basis of a broad and simple definition of the species, it
is exceedingly difficult to know beforehand which samples of 'learning
behaviour' ought to be collected, and how many samples are adequate for
useful generalisations to emerge. Nevertheless, some interesting patterns
have emerged through process research. One good example is given on page 80
(Table 4) where the reader is introduced to teacher communication
strategies, a set of fairly clearly delineated teacher behaviours observed
in classrooms. The trouble remains, however, that in order to make sense and
use of these classifications one really needs some a priori learning theory.
For example, if one has a theory that explicit error correction is
ineffectual (as Krashen for example argues), one can observe a teacher's
behaviour, watch for time spent explicitly correcting errors, and thereby be
able to say just how much time was wasted during that classroom hour. Alas,
things are not so simple. As Ellis points out, we should not underestimate
the amount of variability there is in classrooms (p.90) and in individual
students. Few teachers would accept that no student, regardless of
motivation and learning context, would be better off not being told when
they have made errors. Ellis discusses Allwright's (1984) "interaction
hypothesis" which states that "the process of interaction is the learning
process", but the reader is left with the feeling that this is not a
hypothesis in any useful sense. As Ellis says, we all glean that
"interaction somehow results in L2 learning" (p.91) but neither the type of
interaction nor the learning mechanism are explained, rendering such
statements unhelpful as explanatory or predictive theories. For teachers,
this means one may be vaguely aware that interaction leads to learning but
not know how or why, nor which specific types of interaction are going to
work with individual students in individual classrooms on individual days.

Can we do better than this? Ellis thinks so. Chapter 5 explores classroom
language learning as an interactive, communicative process, Chapter 6 from
the point of view of formal instruction. The chapters are to be regarded as
complementary: Ellis favours neither viewpoint and embraces both as valuable
frameworks for thinking about what goes on in classrooms. Viewed as an
interactive process, classroom language learning is about the relationship
between input (i.e., language that learners see and hear), output (i.e.,
language learners produce) and the kinds of "negotiation" between input and
output that lead to meaningful communication, hence learning. Viewed as
formal instruction, the focus of investigation is on the consequences of
direct intervention into communicative interaction that teachers use to
facilitate learning. The question in Chapter 6, then, is what kind and how
much intervention ought to be employed to encourage learning. Chapter 6
summarises important studies in the literature, but again is marred by the
dated citations -- the most recent study cited in Table 6.1 (pp. 134-135) is
1986. Nevertheless, the conclusions Ellis comes to in both chapters remain
true today: interactive, communicative learning classrooms that mimic
authentic communication generally facilitate learning (Chapter 5), but so
too does a formal learning environment where teachers draw attention to
specific forms.

Chapter 7 is devoted to Ellis's theory of instructed second language
acquisition. Ellis attempts to reconcile a paradox of language learning:
learning appears to be driven by cognitive mechanisms that determine what is
learned and when, but at the same time this acquisition can be bolstered by
teachers who intervene at appropriate moments to facilitate both implicit
and explicit knowledge of the target language. Ellis attempts to draw
together an enormous number of relevant issues - theories of cognition and
learning found in psycholinguistics, factors related to individual learning
styles, as well as syntactic theories derived from universal grammar. The
theory reads like a catch-all theory and a fairly common sense approach to
second language acquisition: an acceptance that learners are in charge of
what they acquire, but that teachers can facilitate learning by focussing
attention on relevant, appropriate forms. At least in respect of the claim
that teacher mediation in the classroom contributes to better overall
learning, research in the 1990s has shown Ellis to be largely correct:
Norris and Ortega (2000) for example present a meta analysis of a decade of
research into the effects of explicit instruction. They concluded that,
overwhelmingly, explicit instruction yields greater learning than implicit
techniques. Ellis was on the mark when he placed explicit instruction and
declarative knowledge in the centre of effective learning.

As with most applied linguistics theories, however, readers looking for
specific advice about how to improve learning in their own classroom will be
disappointed: any such theory must remain vague because, as Ellis pointed
out in 1990 and Norris and Ortega point out in 2000, classrooms are highly
variable environments. If there has been one obvious development in the last
decade, it is that number of potential factors that have been identified in
classroom learning. Norris and Ortega point to aptitude, age , learning
style, structural complexity of linguistic forms, degree of noticing, timing
of instruction, duration, and so on - a daunting list of fuzzy variables.
Ellis argued that classrooms were the place to find out about classroom
learning behaviour, and a decade later what has been found is that the
number of factors is huge, and the number of specific predictions that can
be made are few. It may be that, for all its vagueness, Ellis's theory went
as far as theory building can go in a complex domain like a classroom. This
perhaps explains why the book has not been revised ten years after its first
publication, leaving the reader with a question: are we at the end of useful
theoretical discussion in L2 classroom research?

Norris, J.M., and L. Ortega (2000) Effectiveness of L2 instruction: A
research synthesis and quantitative meta-analysis. Language Learning 50:3,

Dr Stephen A. Bird works in the private sector, consulting on educational
and commercial software development
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue