lntroduction: Investigating Form-Focused

Instruction

Rod Ellis
University ofAuckland, New Zealand

The Introduction has three main aims. First, it provides

a historical sketch of form-focused instruction research,
documenting the origins of this branch of second language
acquisition, the research questions that have been ad­
dressed, and current trends. Second, it seeks to define and
conceptualize what is meant by "form-focused instruction"
by distinguishing it from "meaning-focused instruction"
and by describing three types of form-focused instruction
in terms of whether the primary focus is on form or mean­
ing and whether the instructional attention to target forms
is intensive or extensive. Various instructional options
relating to each type are also described. Third, the Intro­
duction offers a discussion of the main research methods
that have been used to investigate form-focused instruc­
tion in terms of a broad distinction between confirmatory
and interpretative research. Methods of measuring "ac­
quisition" in form-focused instruction research are also
considered.

In this Introduction, the term "form-focused instruction"

(FFI) is used to refer to any planned or incidental instructional
activity that is intended to induce language learners to pay

Rod Ellis, Department of Applied Language Studies and Linguistics.

I am grateful to one anonymous reviewer, Alister Cumming, and Jessica
Williams for helpful comments on a draft version of this Introduction.
Correspondence concerning this article may be sent to Rod Ellis, Univer­
sity of Auckland, Prívate Bag 92019, Auckland, New Zealand. Internet:
[email protected]

1
2 Investigating Form-Focused Instruction

attention t o linguistic form. It serves, therefore, as a cover term

for a variety of other terms that figure in the current literature-
“analytic teaching” (Stern, 1990), “focus-on-form,”and “focus-on-
forms” (Long, 1991), corrective feedbacwerror correction, and “ne-
gotiation of form” (Lyster & Ranta, 1997).Thus, FFI includes both
traditional approaches t o teaching forms based on structural
syllabi and more communicative approaches, where attention t o
form arises out of activities that are primarily meaning-f0cused.l
The term “form”is intended t o include phonological,lexical, gram-
matical, and pragmalinguistic aspects of language. Different types
of FFI are considered later in this Introduction.
FFI is an area of enquiry of interest t o both researchers
and language teachers. Researchers have investigated FFI in
order t o develop and test theories of second language (L2) acqui-
sition. For example, a number of recent studies (e.g., Allen, 2000;
Cadierno, 1995; DeKeyser & Sokalski, 1996, and this volume;
Salaberry, 1997; VanPatten & Cadierno, 1993) have sought t o
test the rival theoretical claims of skill-building and input-
processing models2 of L2 acquisition by examining the relative
effectiveness of production-based and input-based grammar in-
struction. Other studies have focused on issues that are widely
discussed in handbooks for language teachers, such as error
correction (see Seedhouse, 1997a, and Truscott, 1996, for recent
reviews of this research). Here the goal has been to try to identify
what constitutes effective pedagogic practice. FFI constitutes an
area of inquiry, then, where the concerns of researchers and
teachers can be brought together. It is perhaps for this reason that
this area has attracted considerable attention over the last 30 years.
There have been a number of comprehensive surveys of FFI
research (e.g., Larsen-Freeman & Long, 1991;R. Ellis, 1994,1998;
N. Ellis, 1995; Spada, 1997; Norris & Ortega, 2000 and this
volume). This Introduction will not seek to add t o these surveys.
Rather it will provide a brief historical account of FFI and then
address two key aspects of current research-the constructs on
which the research has been based and the methods of research
that have been employed.
Ellis 3

Form-Focused Instruction: A Historical Sketch

of the Research

Early FFI Research

Early research into form-focused instruction (FFI) was

"method" oriented; that is, it consisted of global comparisons of
language teaching methods that differed in their conceptualiza­
tions of how to teach language. At the time these studies took place
(in the 1960s and 1970s), language pedagogy assumed that the
teaching oflanguage necessarily and essentially involved focusing
on form (primarily grammatical form), and the principal debate
concerned how form should best be taught. Thus, methods were
distinguished in terms of whether form was to be taught explicitly
(as in the grammar-translation method) or implicitly (as in the
audiolingual method). Large-scale research projects were under­
taken to resolve what Diller (1978) called the "language teaching
controversy." However, these studies (e.g., Scherer & Wertheimer,
1964; Smith, 1970) were largely inconclusive, failing to demon­
strate the superiority of either method (Allwright, 1988).
At around the same time as the global method studies, second
language acquisition (SLA) researchers, drawing on the findings
and methods of first language acquisition research, began to
investigate how learners acquired an L2 in naturalistic settings
(i.e., when exposed to the use of the L2 in nonpedagogic contexts).
This research was in part pedagogically motivated. That is, re­
searchers such as Hatch (1978) sought to examine how learners
acquíred language naturally, so that copies of theír successful
experiences could be incorporated into the classroom. These stud­
ies indicated that learners tended to follow a natural order of
acquisition and also manifested fairly well-defined sequences in
the acquisition of specific target structures. Furthermore, the
order and sequences appeared to a large extent universal, rela­
tively impervious to such factors as the learner's Ll or age3
(Larsen-Freeman & Long, 1991). These findings led to a question­
ing of whether FFI was necessary for acquisition. The research
4 Investigating Form-Focused Instruction

that ensued addressed the general question "Does form-focused

instruction work?" in two main ways. First, a number of studies
(reviewed by Long, 1983) compared the ultimate level of achieve­
ment and rate of learning of groups of learners who had received
instruction (which was assumed to consist ofFFI) with groups who
had not. Second, comparative and experimental studies were
carried out to investigate whether learners who had rcceived FFI
manifested the same order and sequence of acquisition as natu­
ralistic learners (e.g., Ellis, 1984a; Pica, 1983; Pienemann, 1984;
Turner, 1979). The findings of these studies appeared to be contra­
dictory. Thus, while the majority of the studies indicated that
instructcd learners generally learned more rapidly and achieved
higher levels of proficiency than noninstructed learners (suggest­
ing that FFI assisted acquisition), other studies indicated that
instructed learners followed the same order and sequence of
acquísition as noninstructed learners (suggesting that the process
of acquisition was not influenced by instruction). This apparent
paradox has had a major impact on theoretical thinking about the
relationship between FFI and acquisition, leading to claims,
clearly evident in current research, that FFI only works by pro­
moting the processes invol ved in natural language acquisition, not
by changing them.

Classroom Process Research

The demise of the comparative method studies also resulted

in another strand of research-classroom process research. This
was directed at obtaining accurate and detailed information
about how instruction was accomplished through the observation
and description of teaching-learning events. In the case of
FFI, researchers focused initially on error treatment, developing
taxonomies of the various treatment options (e.g., Allwright,
1975; Long, 1977; Chaudron, 1977). Later, researchers (e.g.,
Ellis, 1994b; VanLier, 1988) widened the scope of their enquiry
by addressing more broadly the kinds of interactions that occurred
in language classrooms. These studies resulted in general
Ellis 5

frameworks oflanguage use and ethnographic accounts of particu­

lar aspects of classroom language such as turn-taking and repair.
Toward the end of the 1980s, process-product studies began to
appear. These attempted to relate features of classroom language
use to learning outcomes. For example, using a classroom interac­
tion scheme called the Communicative Orientation to Language
Teaching (COLT), Allen, Swain, Harley, and Cummins (1990) ex­
amined the relationship between various experiential and ana­
lytic classroom activities4 and learning outcomes, measured by a
battery of tests based on a model of communicative competence.
Correlational analyses revealed that both meaning-focused and
form-focused aspects of classroom interaction were positively re­
lated to learning, leading to the conclusion that "the analytic and
the experiential focus may be complementary" (p. 62).

Does Form-Focused Instruction Work?

The late 1980s and early 1990s saw a plethora of experimen­

tal studies that all addressed the same question: "Does form­
focused instruction work?" The studies investigated whether
learners learned the specific forros they were taught. "Learned"
was typically operationalized as statistically significant gains in
the accurate production of the targeted structures. Like the earlier
research, this research was motivated by both theoretical and
pedagogic considerations. On a theoretical level, the studies
sought to test the claims advanced by Krashen (1981) and later
Schwartz (1993) that grammar can only be acquired unconsciously
from comprehensible input and that teaching grammar or correct­
ing learner errors has no effect on the learner's "acquired" system
(interlanguage). On the pedagogic level, the studies explored
whether FFI could help learners to acquire those grammatical
structures they had failed to acquire even after years of expo­
sure to comprehensible input or those structures that were
known to be difficult to acquire from studies of naturalistic
learners. Harley (1989), for example, investigated French im­
parfait and preterit. Day and Shapson (1991 and this volume)
6 Investigating Form-Focused Instruction

examined French conditional forms in hypothetical situations and

polite requests. The choice of these target features was motivated
by research showing that immersion learners often fail to acquire
these forms, even after years of content instruction. The studies
produced mixed results. Day and Shapson, for example, found that
gains in accuracy were evident only in a doze test and written
interview, not in an oral interview. Studies by White, Spada,
Lightbown, and Ranta (1991) and White (1991) indicated that
whereas instruction resulted in clear gains in WH questions,
which were sustained over time, it resulted only in temporary,
impermanent gains in the case of adverbial positioning. In general,
however, these experimental studies did show that grammatical
form was amenable t o instruction, especially if the learners were
developmentally ready to acquire the targeted structure, and also
that these effects were often durable.

Effects ofInstruction on the Order and Sequence o f Acquisition

Running parallel to research that examined whether FFI

worked were studies that addressed the related research question,
“What effect does form-focused instruction have on the order and
sequence of acquisition?” These studies were both comparative
and experimental in nature. Comparative studies such as those of
Pavesi (1986) and Ellis (1989) compared groups of noninstructed
and instructed learners, examining the sequence of acquisition of
English relative clauses and German word-order rules respec-
tively. They provided additional support for the claim that in-
structed learners followed the same order and sequences of
acquisition as naturalistic learners but that they proceeded fur-
ther and more rapidly. Experimental studies also indicated that
instruction is powerless to change the orderhequence of acquisi-
tion. However, studies by Pienemann and associates (summarized
by Pienemann, 1989) indicated that instruction directed at struc-
tures that were next in line to be acquired according to a well-
defined developmental sequence was effective in moving learners
along the sequence. In contrast, instruction directed at structures
Ellis 7

that were too developmentally advanced for the learners proved

ineffective. On the basis of these findings, Pienemann (1985)
advanced the teachability hypothesis, according to which "instruc­
tion can only promote language acquisition if the interlanguage is
clase to the point when the structure to be taught is acquired in a
natural setting'' (p. 37). Another group of experimental studies,
however, produced very different results. These "projection stud­
ies" drew on linguistic accounts of implicational universals. They
tested the hypothesis that teaching learners a marked structure
in a hierachy would enable them to acquire not only this structure
but also the implicated less marked structures, but that the
converse was not true (i.e., teaching a less marked structure would
not enable learners to acquire the more marked structures). Gass
(1982) and Eckman, Bell, and Nelson (1988), drawing on the
Accessibility Hierarchy for relative clauses, conducted studies that
supported this hypothesis. On the face of it, these studies refute
the claim that FFI does not enable learners to "beat" a natural
sequence, because learners receiving instruction in a marked
structure appear able to acquire it concurrently with unmarked
structures that are acquired first in natural settings. However,
much depends on what is meant by "acquired," a point that will
be taken up later. Also, a later study by Hamilton (1994) did not
support the projection hypothesis (i.e., learners were able to gen­
eralize when they received instruction in an unmarked relative
clause structure).

Theory-Driven FFI Reserach

The early 1990s were characterized by developments in L2

acquisition theory and by a concomitant change in the questions
FFI researchers were interested in. SLA began to draw extensively
on theories of information processing and skill learning drawn
from cognitive psychology. Schmidt (1990, 1994, 1995a) advanced
what has become known as the "noticing hypothesis." This
claims that for acquisition to take place, learners must consciously
notice forms (and the meanings these forms realize) in the input.
8 Investigating Form-Focused Instruction

Noticing, however, is not seen as guaranteeing acquisition. It is

only "the necessary and sufficient condition for the conversion of
input to intake for learning" (Schmidt, 1994, p. 17). That is,
noticing enables learners to process forms in short-term rnemory
but does not guarantee they will be incorporated into their devel­
oping interlanguage. The noticing hypothesis contradicts
Krashen's (1981) claim that the process of acquisition is uncon­
scious. It is compatible with the claim that FFI can aid acquisition
by drawing learners' attention to forrns in the input that otherwise
they rnight not notice and thus fail to intake. However, Schmidt's
hypothesis, while widely accepted by SLA researchers, remains
controversia! (see Tomlin & Villa, 1994, and Truscott, 1998, for a
different view of the role of attention in language acquisition).
VanPatten (1990, 1996) has also drawn on information
proccssing theory to suggest that learners, especially beginner
learners, experience difficulty in concurrently attending to
meaning and to form and thus often prioritize one at the expense
of the other. He argues that learners will only be able to attend to
form when the input is easy to understand and that when learners
are primarily concerned with processing meaning, they will rely
on "default strategies" that prevent them from attending to forms
in the target language that do not conform to these strategies. 5
Other SLA researchers (e.g., DeKeyser, 1998; Johnson, 1996) have
drawn on skill building theory (Anderson, 1993) to suggest that
FFI can enable learners to proceduralize declarative knowledge
through practice, especially if this is accompanied by negative
feedback on learners' attempts to produce a target structure under
"real operating conditions" (i.e., when trying to communicate). The
importance of negative evidence for L2 acquisition has also been
emphasized by sorne researchers who base their research on a
theory ofUniversal Grammar (e.g., White, 1989).
This theorizing about the role of consciousness, attention,
and negative evidence in the process of L2 acquisition has led to
several new questions in FFI research in the 1990s: "Do sorne
types of form-focused instruction work better than others?"; "In
what ways can input (positive evidence) be enhanced to promote
Ellis 9

noticing?“; and “What kinds of feedback (negative evidence) pro-

mote acquisition?“ There has been a plethora of experimental
studies, both classroom-based and laboratory, investigating the
effects of different instructional approaches on learning. VanPatten,
in a series of studies (VanPatten & Cadierno, 1993; VanPatten &
Oikennon, 1996; VanPatten & Sanz, 19951,compared the relative
effects of input processing and production-based instruction on
comprehension and production, finding that the former resulted
in larger gains in comprehension and equivalent gains in produc-
tion. However, this finding has subsequently been challenged by
other studies (including DeKeyser & Sokalski, 1996 and this
volume; Allen, 2000). Other experimental studies have examined
the difference between explicit and implicit instructional ap-
proaches (e.g., DeKeyser 1994, 1995; de Graaff, 1997; Robinson,
1996), generally finding in favor of explicit learning (see the
conclusion reached by Norris and Ortega [ZOO0 and this volume1
in a meta-analysis of FFI experimental studies). Yet other stud-
ies (e.g., Fotos & Ellis, 1991; Fotos, 1994) have explored which
type of explicit instruction-traditional, teacher-centered, or
consciousness-raising (CR) tasks-works best, finding no differ-
ence. Other studies have examined the effects of “enhanced
input” (Sharwood Smith, 1993) on “noticing”and on acquisition.
Alanen (1995)and Jourdenais, Ota, Stauffer, Boyson, and Doughty
(1995) provide evidence t o suggest that highlighting forms in the
input increases the likelihood of their being noticed and sub-
sequently used. However, Trahey and White (1993) found that
positive evidence in the form of an “input flood” failed to enable
learners to discover the ungrammaticality of placing an adverbial
between the verb and the direct object in English. A number of
studies have explored the effects of negative feedback on acquisi-
tion. Tomosello and Herron (1988, 1989) found that inducing
learners t o make errors and then correcting them worked better
than traditional grammar instruction involving production prac-
tice. Carroll, Swain, and Roberge (1992) and Carroll and Swain
(1993) examined the differencebetween explicit and implicit feed-
back, finding the former more effective. More recently, however, a
10 Investigating Form-Focused Instruction

particular type of implicit feedback, “recasts”(i.e., reformulations

of deviant learner utterances), has been examined (Doughty &
Varela, 1998; Mackey & Philp, 1998; Long, Inagaki, & Ortega,
1998),with results suggesting that these can have a positive effect
on acquisition. Still other recent experimental studies (e.g., Robin-
son, 1996; Rosa & ONeill, 1999) have attempted t o explore the
effect of different instructional treatments on learners’ level of
awareness of form in the input as they perform some task and t o
relate these levels t o acquisitional gains. This research makes use
of learner self-reports, concurrent or retrospective, in the attempt
t o examine what “noticing”has taken place.
There is another strand of theorizing that has had a major
impact on current form-focusedresearch. Long (1988,1991) draws
on research showing that instruction appears to facilitate learning
but only if it supports the natural processes of acquisition. He has
argued that attention to form will work most effectively for acqui-
sition if it occurs in the context of meaning-focused communication
rather than in instruction that is specifically directed at linguistic
forms. According to Long‘s revised Interaction Hypothesis (Long,
1996), attention to form in meaning-focused communication oc-
curs when learners have the opportunity t o negotiate for meaning
following a breakdown in understanding. Such negotiation serves
to highlight linguistic forms that are problematic to them. It helps
them to “notice the gap” (Schmidt & Frota, 1986) between the
input and their own interlanguage and gives them opportunities
for “pushed output” (i.e., t o improve linguistic accuracy by refor-
mulating utterances that were initially misunderstood).A number
of experimental-type studies have examined the effects of mean-
ing negotiation on acquisition (e.g., Nobuyoshi & Ellis, 1993;
Mackey, 1999) indicating that opportunities to participate in
meaning negotiation centered (without the learners’ knowledge)
on specific grammatical forms results in acquisitional gains.
Ellis 11

Descriptive Studies of FFI and Teacher Cognitions About FFI

The 199Os, then, have seen a rapid growth in experimental-

type research directed at testing a variety of hypotheses drawn
from SLA theory. This research has been very much theory driven
and, as such, has to some extent sacrificed pedagogic relevance.
Some researchers have (quite legitimately) been more concerned
with addressing theoretical claims in carefully controlled labora-
tory studies than in investigating what constitutes effective in-
structional practices in real classrooms. As a result, something of
a gap has grown between theory-driven and pedagogically moti-
vated research. In the view of Borg (1999a), the results of the
theory-driven research have been “largely inconclusive” (p. 201,
with no consensus having been reached on how best to teach
grammar. Borg also argues that an essential element in FFI-the
teacher-has been ignored entirely. This is a justifiable criticism
insofar as researchers wish t o claim pedagogic relevancy. However,
a number or recent studies have focused more closely on the
teacher and on teaching behaviors. Drawing on teacher-cognition
research, Borg’s own research (Borg, 1998,1999131has used quali-
tative research methods (e.g., observation and interviews) t o gain
an understanding of teachers’beliefs about grammar teaching and
of their actual practices. There have also been a number of detailed
descriptive studies (in the tradition of process research mentioned
above) of how teachers handle form in actual classrooms. Lyster
and Ranta (1997) examined teachers’ corrective feedback in im-
mersion classrooms (where the primary focus is on content and
meaning rather than on form). Lyster (1998a, 1998b)subsequently
carried out fine-grained analyses of teachers’ corrective feedback
in relation to error types and uptake (i.e., the students’ response
to feedback). Ellis, Basturkmen and Loewen (in press) have car-
ried out a similar descriptive study looking at two teachers’
handling of form in communicative ESL lessons.
12 Investigating Form-Focused Instruction

Some Retrospective Comments

Looking back over some 30 years of research into FFI, what

conclusions can be reached? One pleasing observation is that the
range of languages under study has broadened considerably.
Whereas early research was directed at the effects of FFI on
learning English and, t o a lesser extent, German, the later re-
search has examined a number of different languages, including
Chinese, French, Italian, Japanese, and Spanish. The studies
included in this volume address English, French, and Spanish.
Regarding the actual research findings, however, there are few
certainties. Researchers have employed very different methods
and, as Norris and Ortega (2000) point out, there have been few
replications. A large number of variables have an impact on
whether FFI is effective-the learners’ developmental stage, the
structure being taught, the instructional context, and the instruc-
tional materials. It is, therefore, not entirely surprising that dif-
ferent findings emerge from different studies. Two findings are
pervasive, however: (1)FFI, especially of the more explicit kind, is
effective in promoting language learning, and (2)FFI does not
alter the natural processes of acquisition.6 The key question is how
these two findings can best be reconciled. The next section will
examine the conceptual framework that informs current FFI
research.

Conceptualizing Form-Focused Instruction

The last 30 years of FFI research have seen progressive

changes in the way in which the phenomenon has been investi-
gated. Initially, as already noted, FFI was conceptualized in rela-
tion t o method, a little later as a type of exposure distinct from
natural exposure, a little later still as a set of classroom processes,
and, increasingly, as a set of psycholinguistically motivated peda-
gogic options. This section will explore two of these conceptualiza-
tions-pedagogic options and classroom processes. The first is
“external” in the sense that the options have been derived from
Ellis 13

theory, in particular theories of L2 acquisition, and this has in-

formed the construction of form-based syllabi. The latter is “inter-
nal” in the sense that the constructs have been derived from
observing and describing classroom discourse; it affords an ac-
count of FFI as teaching.

Form-Focused Versus Meaning-Focused Instruction

FFI contrasts with meaning-focused instruction (MFI). The

former describes instruction where there is some attempt t o draw
learners’ attention to linguistic form-Stern’s (1990) “analytic
strategy.” The latter refers t o instruction that requires learners to
attend only t o the content of what they want t o communicate-
Stern’s “experiential strategy” (see note 4). Widdowson (1998)has
criticized this distinction, arguing that so-called form-focused
instruction has always required learners to attend to meaning as
well as form (e.g.,the semantic meanings realized by different verb
forms), whereas meaning-focused activities still require learners
t o process forms in order t o decode and encode messages. For
Widdowson, the key difference lies in the kind of meaning learners
must attend to-whether it is semantic meaning, as in the case of
language exercises, or pragmatic meaning, as in the case of com-
municative tasks. Widdowson’s point is well taken, but arguably,
it is nothing new. SLA researchers have always used the term
“form” t o refer not just to form (e.g., -ed in the regular past tense
in English) but also t o the semantic meaning(s) a form realizes in
use (e.g., completed action in the past). Ellis (2000) has argued
that the essential difference between form-focused and meaning-
focused instruction lies in how language is viewed (as an object as
opposed to a tool) and the role the learner is invited to play
(student as opposed to user). In this respect, it should be noted
that attention t o lexical forms and the meanings they realize,
where words are treated as objects to be learned, constitutes
form-focused i n s t r ~ c t i o nAs
. ~ noted at the beginning of this Intro-
duction, “form”involves more than grammar.
14 Investigating Form-Focused Instruction

Of course, many interactions that occur inside the classroom

will be neither entirely form-focused nor meaning-focused but a
combination of both, although achieving a dual focus is not easy
(Seedhouse, 1997b). In effect, then, it is possible t o distinguish
types of FFI according t o whether attention to form is primary or
secondary, as when it is integrated into MFI.

Three Types of Form-Focused Instruction

There have been a number of recent attempts t o develop

taxonomies of pedagogic options in FFI (e.g., Ellis, 1997; Doughty
& Williams, 199813).These have centered on a binary distinction
between what Long (1988, 1991), somewhat confusingly, calls
focus-on-form and focus-on-forms and what Ellis (1997), perhaps
no more helpfully, labels focused-communication and feature-
focused instruction. Long’s terms have become widely used (see,
for example, Doughty & Williams, 1998a) and so will be the ones
used here. It should be noted that FFI can be distinguished by
means of other binary distinctions of potential importance to
L2 acquisition (e.g., proactive versus incidental FFI and intensive
versus extensive FFI). However, given the importance attached to
the focus-on-fordforms distinction, this will inform the discussion
that follows. I will argue below, however, that it may be more
helpful t o conceptualize FFI as involving three rather than two
broad types.
Focus-on-forms is evident in the traditional approach t o
grammar teaching, based on a synthetic syllabus. The underlying
assumption is that language learning is a process of accumulating
distinct entities. In such an approach, learners are required t o
treat language primarily as an “object”to be studied and practiced
bit by bit and to function as students rather than as users of the
language. In contrast, focus-on-form overtly draws students’ at-
tention t o linguistic elements as they arise incidentally in lessons
whose overriding focus is on meaning or communication (Long,
1991, pp. 45-46). Such an approach, according to Long and Robin-
son (19981,is t o be distinguished not only from focus-on-formsbut
Ellis 15

also from focus-on-meaning (as in Krashen and Terrell’s [19831

Natural Approach), where there is no attempt to attend to linguis-
tic form at all.
In fact, the focus-on-form construct has been stretched t o
cover a type of FFI that it was initially intended to exclude. Long‘s
definition above identifies two essential characteristics of focus-
on-form: (1)Attention to form occurs in lessons where the overrid-
ing focus is meaning or communication, and (2) attention to form
arises incidentally in response t o communicative need. Re-
searchers such as Doughty and Varela (1998)’Long et al. (19981,
and Williams and Evans (1998) have latched onto the first of these
characteristics t o design experimental studies in which a pre-
selected form has been taught using various communicative de-
vices (e.g., providing feedback by means of recasts). They have
ignored the second defining characteristic of focus-on-form,
namely, that it should be incidental. Indeed, Long himself now
appears to overlook this. Long and Robinson (1998) give three
examples of focus-on-form.The first involves “seeding”a text with
ergative verbs (e.g., increased, deteriorated). The second involves
the teacher taking time out from a communicative activity to
briefly draw attention t o a linguistic problem the students are
experiencing. The third consists of using the recasts in the context
of task-based conversation, as has been found t o occur in conver-
sations children experience during first language acquisition.
While the second and third of these satisfy Long’s earlier definition
of focus-on-form as incidental, the first does not, because “seeding”
a text necessarily requires preselecting a specific form for treat-
ment. This reconceptualization of focus-on-form is clearly evident
in the definition provided by Doughty and Williams (1998~). This
mentions three definitional features: (1) the need for learner
engagement with meaning to precede attention t o the code; (2) the
importance of analyzing learners’ linguistic needs t o identify the
forms that require treatment; and (3) the need for the treatment
to be brief and unobtrusive. While (1)and (3) are compatible with
Long‘s initial definition, (2) is not, for again, it assumes a planned
rather than incidental approach to form and, thereby, constitutes
16 Investigating Form-Focused Instruction

a synthetic approach t o teaching, albeit a remedial one.8 The

motivation for this shift in the way in which focus-on-form is
conceptualized is probably the desire of researchers t o conduct
experimental studies. Such studies are not possible unless the
incidental aspect of focus-on-form is abandoned, for they require
the selection of a specific linguistic feature for treatment.
It might be argued, of course, that this shift in definition is
unimportant-on the grounds, perhaps, that what is essential in
focus-on-forminstruction is that attention t o form arises naturally
out of communicative activity and that whether such attention is
planned o r incidental is irrelevant. However, planned and inciden-
tal focus-on-form instruction are likely t o differ in one major
respect. In the case of planned focus-on-form,the instruction will
be intensiue, in the sense that learners will have the opportunity
to attend t o a single, preselected form many times. In the case of
incidental focus-on-form,the instruction will be extensive, because
a range of linguistic forms (grammatical, lexical, phonological,
pragmatic) are likely to arise as candidates for attention (see
studies by Williams [1999], Oliver [20001; Ellis, Basturkmen, &
Loewen [in press]. This difference is important both theoretically
and pedagogically, because it raises the question as t o whether
language learning benefits most from focusing on a few problem-
atic linguistic forms intensively or from a scatter-gun approach,
where multitudinous problematic forms are treated randomly and
cursorily and where the treatment may or may not be repeated.
Perhaps, then, FFI needs t o be conceptualized in terms of
three rather than two types, according t o (a) where the primary
focus of attention is t o be placed and (b) how attention t o form is
distributed in the instruction. Thus, as shown in Table 1,focus-on-
forms (Type 1)is characterized by a primary focus on form and
intensive treatment of preselected forms. Planned focus-on-form
(Type 2) differs from focus-on-forms with respect to where the
primary focus of attention lies (on meaning rather than form), but
like Type 1 involves intensive attention to preselected forms.
Incidental focus-on-form(Type 3) also involves primary attention
t o meaning but differs from both focus-on-forms and planned
.Ellis 17

focus-on-form in distributing attention to a wide range of forms

that have not been preselected. It should be noted that whereas
Types 1 and 2 can be investigated experimentally, Type 3 can only
be examined by means of interpretative studies.
There is a problern of perspective in all this. The typology of
types of instruction is based on the participants' orientation to the
instruction they are experiencing (i.e., whether they give prece­
dence to form or meaning). In part, this rnay be evident in the
discourse they construct, but ideally it requires entering the rninds
of the participants, as it were, by collecting retrospective self­
report data. A general weakness ofFFI research, especially experi­
mental research, is that such data have rarely been collected.

Type 1: Focus-on-Forms

Focus-on-forrns irnplies that the teacher and students are

aware that the primary purpose of the activity is to learn a
preselected form and that learners are required to focus their
attention on sorne specific form intensively in order to learn it.
There are various options for achieving this. The options consid­
ered below have been selected because they are of psycholinguistic
interest and because they have figured widely in studies of in­
structed L2 acquisition. There are, of course, a number of other
options widely recognized in language pedagogy (e.g., different
types of controlled practice).
Explicit vs. implicit focus-on-forms. Explicit focus-on-forms is
instruction that involves "sorne sort of rule being thought about

Table 1

Types of Form-Focused Instruction

Type ofFFI Primary Focus Distribution
l. Focus-on-forms Form Intensive
2. Planned focus-on-form Meaning Intensive
3. Incidental focus-on-form Meaning Extensive
18 Investigating Form-Focused Instruction

during the learning process" (DeKeyser, 1995). The rule can be

addressed deductively or inductively. A deductive presentation
occurs when the rule is presented to the learners. An inductive
treatment involves learners attempting to arrive at a rule them­
selves by analyzing data containing exemplars of the feature in
question. Another way of referring to these two types of explicit
instruction is as "didactic" and "discovery."Robinson (1996) opera­
tionalized this distinction in treatments that asked learners to
(1) read through written accounts of rules and (2) identify the
rules illustrated by a set of sentences.
According to DeKeyser (1995), implicit learning involves
learners memorizing instances or inferring rules without aware­
ness, or both. Thus implicit learning contrasts with explicit learn­
ing with regard to the absence of awareness of what is being
learned. In both cases, learners may construct rules to represent
the form they are studying, but whereas explicit instruction is
directed at helping learners make the rules explicit, implicit
instruction is geared to helping learners acquire implicit rules.
Implicit instruction is typically operationalized by researchers in
terms of instructions for learners to memorize a set of sentences
that model a specific feature (DeKeyser, 1995; Robinson, 1996).
The idea here is that learners' attention will be focused on form,
as this is essential for memorization, but that they will not become
aware of what specific feature has been targeted. It should be
noted that, in this respect, the experimental treatments differ
somewhat from pedagogic practices that are often labeled "im­
plicit" (Stern, 1983). These latter involve drilling students in the
production of the target form and correcting errors. Such instruc­
tion is likely to result in learners becoming aware of the target
structure and, perhaps, in their attempting to construct a con­
scious rule.
Structured input vs. production practice. The second distinc­
tion draws directly on the computational model of L2 acquisition
that informs mainstream SLA. This model posits three types of
processes: (1) intake (i.e., noticing forms in the input and storing
them in short-term memory), (2) acquisition (i.e., the incorporation
Ellis 19

of new forms in long-term memory and the restructuring of the

interlanguage system), and (3) language production (i.e., the use
of stored forms in speech or writing). Instruction can be directed
at (1)or (3); instruction directed at (2) is not viable, because the
complex processes of accommodation and restructuring that are
involved in interlanguage development are not amenable t o envi-
ronmental control. Traditional focus-on-forms instruction is di-
rected at (3). That is, opportunities are created for learners to
practice producing the target structure. However, an alternative
instructional approach involves presenting learners with struc-
tured input (i.e., input specially designed t o provide plentiful
examples of the target structure) and asking them to perform
some task designed to make them notice the form in question and
t o process its meaning. In structured input, “the learner is pushed
to attend to particular feature of language while listening or
reading” (VanPatten, 1996,p. 6).
Structured input should be viewed as a focus-on-forms op-
tion, because it is designed t o enable learners to give primary
attention to form rather than meaningg and because it is designed
to focus learners’ attention repeatedly on a specific, preselected
linguistic feature. The structured-input option, in fact, is simply
another way of teaching a structural syllabus. Thus, structured
input differs from an input flood, which also exposes learners to
input rich in some specific linguistic feature but which requires
them to process this input primarily for meaning. An input flood,
therefore, constitutes an example of incidental instruction of Type
2 FFI, planned focus-on-form,and is discussed below.
Traditional production practice involves both mechanical
and contextualized activities. VanPatten and Cadierno ( 1993),for
example, operationalized this instructional option by means of
oral and written transformation and substitution drills and more
open-ended communicative practice. Structured input entails the
provision of oral or written data containing exemplars of the target
feature together with some task requiring learners to interpret
(but not produce) the input. VanPatten and Cadierno operational-
ized this through tasks that required learners to select drawings
20 Investigating Form-Focused Instruction

that best represented what they heard or read or t o respond to

sentences by checking “agree” o r “disagree.”
Functional language practice. This involves the development
of instructional materials that provide learners with the opportu-
nity to practice producing the target structure in some kind of
situational context. Such kinds of activities were previously re-
ferred to as “situationaUcontextua1 language exercises” but are
now more commonly referred to as “functional.” Good examples of
such materials can be found in the work of Day and Shapson (1991
and this volume), Harley (1989), Nunan (1998), and Widdowson
(1986). It is important to note that such materials lie within the
realm of focus-on-formsinstruction, because despite the apparent
concern for meaning, the primary focus remains on form rather
than meaning, and learners are aware that the purpose is t o
master accurate use through repeated use of the target feature.

o p e 2: Planned Focus-on-Form

Studies of planned focus-on-form have also drawn exten-

sively on the computational model of L2 acquisition. The options
t o be considered here relate to (1)input and (2) production.
Enriched input. Instruction involving a planned focus-on-
form frequently makes use of enriched input. Like structured
input, enriched input consists of input that has been specially
contrived or modified t o present learners with plentiful exemplars
of the target structure. In the case of enriched input, however,
learners are invited t o focus primarily on meaning. Thus the tasks
that accompany the enriched input are all communicative in
nature, requiring learners t o respond t o the content of the input.
In effect, then, enriched input is designed t o cater to incidental
rather than intentional language acquisition (see Schmidt, 1994,
for a discussion of this key distinction). The aim of enriched input
is to induce noticing of the target form in the context of meaning-
focused activity.
Various options exist for enriching input. Input flood consists
of input that has been enriched by including plentiful exemplars
Ellis 21

of the target feature without any device to draw attention to the

feature. For example, Trahey and White (1993)and Trahey (1996)
developed materials consisting of stories, games, and exercises
with the aim of simply exposing learners to adverbs. The rationale
for such an option is that acquisition occurs as a result of frequent
exposure to a target feature (see N. Ellis, 1996)Jnputenhancement
(Sharwood Smith, 1993) involves some attempt t o highlight the
target feature, thus drawing learners’ attention to it. For example,
Leeman et al. (1995)enhanced input by highlighting instances of
Spanish preterit and imperfect verb forms in the written texts
used in content-based instruction. Enriched input can function
entirely by itself (e.g., learners can simply be asked to listen to or
read texts that have been enriched) or can be accompanied by some
kind of meaning-focused activity that incidentally assists learners
t o focus their attention on the target feature (e.g., comprehension
questions that can only be answered correctly if the learners
process the target feature).
Focused communicative tusks. These are tasks that are de-
signed t o elicit production of a specific target feature in the context
of performing a communicative task. Such tasks have all the
characteristics of communicative tasks. That is, meaning is pri-
mary, there is a goal that needs t o be worked toward, the activity
is outcome-evaluated, and there is real-world relationship (Ske-
han, 1998). However, in contrast to communicative tasks in gen-
eral, focused communicative tasks are intended t o result in
learners’ employing some feature that has been specifically
targeted. Such tasks differ from functional materials, because
they require the primary focus t o be on meaning rather than on
form. Thus, in focused communicative tasks, any acquisition of
the target feature that takes place is incidental, whereas with
functional materials, it is intentional. However, the difference
between functional and focused communicative tasks is a slen-
der one and is not acknowledged by some researchers or
teachers. It is dependent ultimately on the learners’ perspec-
tive (i.e., whether learners treat the instruction as requiring
22 Investigating Form-Focused Instruction

them t o view language as an object or tool and to function as

students or as users).
As Loschky and Bley-Vroman (1993) have noted, focused
communicative tasks can be designed so that the production of the
target feature is useful, natural, or essential t o the performance
of the task. Ideally, it needs t o be essential, but this is difficult to
achieve, because learners cannot use features they have not yet
acquired and also are adept at avoiding using those features they
find difficult t o process (Schachter, 1974). Samuda (2000) provides
an example of a focused communicative task and also illustrates
the problem of avoidance. However, as Samuda’s study shows, this
problem can be overcome methodologically; the teacher in her
study introduced a brief explicit focus in the context of the learn-
ers’ performing the task. In effect, this amounted to supplement-
ing Type 2 FFI (planned focus-on-form) with Type 3 FFI
(incidental focus-on-form).

Type 3: Incidental Focus-on-Form

Whereas the previous options reflect constructs that have

been derived from theories of language acquisition or language
pedagogy, the options that will be considered now have been
derived from studies of classroom processes-that is, they reflect
what actually transpires during FFI. The options relate to two
kinds of incidental focus-on-form:pre-emptive and reactive. Both
kinds of focus-on-formcan arise either because there is a problem
of communication (i.e., the interactants have not understood each
other) or because there is a problem of form (i.e., the interactants
have understood each other but nevertheless wish t o focus on some
form that has arisen in the course of communicative activity).
Pre-emptiue focus-on-form.l o In pre-emptive focus-on-form,
the teacher or a learner takes time out from a communicative
activity t o initiate attention t o a form that is perceived to be
problematic even though no production error in the use of the
form or difficulty with message comprehension has arisen. Such
time-outs involve the teacher and learner briefly switching from
Ellis 23

viewing language as a tool and functioning as a user to viewing

language as an object and functioning as a student. Below is an
example:
T: What’s an alibi?
(4.0)
T: S has an alibi.
(3.0)
T: Another name for girlfriend?
(laughter)
(4.5)
T: An alibi is a reason you have for not being at the bank
robbery (.) Okay (.>Not being at the bank robbery.

Here the teacher and students are about to perform a communi-

cative activity that involves the students concocting alibis for a
crime. The teacher pre-empts by briefly checking that the students
understand the meaning of the word “alibi” and supplying a
definition. She then proceeds with the communicative activity.
Pre-emptive focus-on-formhas been little studied t o date, and
thus the extent to which it occurs during FFI is uncertain. Wil-
liams (1999) looked at the ways in which learners initiate atten-
tion to form in learner-learner interactions, reporting that this
occurred most frequently when learners requested assistance
from the teacher. Ellis, Basturkmen, and Loewen (in press)
examined teacher- and learner-initiated attention t o form in
communicative ESL lessons, finding that this occurred as fre-
quently as reactive focus-on-form. However, little is known
about the kinds of options teachers and learners select from
during pre-emptive focus-on-form, and even less about whether it
facilitates acquisition.
Reactive focus-on-form. Reactive focus-on-form has received
much more attention. It consists of the negative feedback teachers
provide in response t o learners’ actual or perceived errors. This
feedback occurs in all types of FFI and involves the same set of
options, although there may be differences in frequency of choice
of specific options according t o FFI type. The options entail various
24 Investigating Form-Focused Instruction

ways of providing implicit and explicit negative feedback.ll This

distinction is perhaps best viewed as a continuum with options
being more or less implicit/explicit. The distinction is consid-
ered important because it potentially affects noticing and has
been shown t o influence whether learners notice corrected
forms and uptake them (see Lyster, 1998a; Oliver, 2000; Ellis et
al., in press).
Implicit negative feedback. As Seedhouse (1997a) has noted,
teachers display a general preference for implicit negative feed-
back in what he calls “form and accuracy contexts” (p. 552), that
is, in Type 1FFI. He comments, “When learners supply a linguis-
tically incorrect response in reply t o a teacher initiation . . . the
teacher tends t o avoid direct, explicit, overt negative evaluation”
(p. 554). The same appears t o be true in meaning-focused contexts
(i.e., Type 3 FFI). Lyster and Ranta (1997) show that teachers in
immersion classrooms rely extensively on recasts (i.e., reformula-
tions of all or part of the learner’s deviant utterance). Some
researchers (e.g.,Oliver, 2000) distinguish recasts and negotiation
of meaning, with recasts defined as reformulations that negotiate
form and negotiation of meaning as involving confirmation checks
used t o clarify understanding. Such a distinction is an uncomfort-
able one, because it necessitates interpreting the intention of the
teacher, which is not easily achieved by simply inspecting a tran-
script of what was said. As noted earlier, a number of studies have
examined recasts (e.g., Long et al., 1998; Mackey & Philp, 1998),
suggesting that they assist acquisition. Other implicit options are
available t o teachers-e.g., requests for clarification and repeti-
tion@ (often with the learner’s error highlighted by intonation).
Lyster’s 1998a study of negative feedback in immersion class-
rooms found that uptake is more likely t o occur after requests for
clarification than after recasts, thus supporting the findings of
earlier laboratory studies (e.g., Pica, Holliday, Lewis, & Morgen-
thaler, 1989). Experimental type studies of clarification requests
involving Type 2 FFI (Ellis & Takashima, 1999; Noboyoshi &
Ellis, 1993) have also provided evidence of long-term effects on
acquisition.
Ellis 25

Explicit negative feedback. Explicit negative feedback is dis-

preferred in all types of FFI. That it should occur infrequently in
Type 2 and Type 3 FFI is not surprising, because explicit negative
feedback is clearly more obtrusive than implicit. However, its
relative scarcity in Type 1 FFE is more puzzling, especially be-
cause teachers these days are advised to view errors positively, as
the means by which learners can test hypotheses about the lan-
guage. Most likely, it reflects a sociolinguistic need on the part of
teachers t o protect the face of their students.
Lyster and Ranta (1997)identify a number of explicit options.
“Explicit correction” occurs when a teacher clearly indicates that
the learner has said something wrong and provides the correct
form. “Metalinguistic feedback consists of “comments, informa-
tion, or questions related to the well-formedness of the student’s
utterance” (p. 47). “Elicitation” constitutes an attempt t o directly
elicit the correct form from students. Samuda (in press), in the
study referred t o earlier, found that explicit feedback involving
metalinguistic comments and elicitations was necessary t o
prompt students into using the feature targeted in her focused
communication activity.

Instructional Options in Research and Pedagogy

Describing the various types of FFI and the options available

within each is of potential value to both researchers and teachers.
For both groups, it affords a basis for making systematic selections
of what t o investigate or to teach. However, the different require-
ments of research and pedagogy point t o very different ap-
proaches. Researchers are encouraged to isolate specific options
(variables) in order to test theory That this has not always hap-
pened is seen as cause for criticism. Norris and Ortega (ZOOO), for
example, argue that the inconclusive findings of much of the FFI
research to date is the result of a failure to investigate particular
subtypes of instruction (i.e., to examine the effect of discrete
instructional options). They also lament the fact that researchers
have inconsistently operationalized instructional approaches. In
26 Investigating Form-Focused Instruction

contrast, teachers can be expected to respond in a very different

way. They are likely to construct lessons that make use of a number
of different options, both because this may enhance the effective-
ness of the instruction and because they need t o provide variety
in a single lesson. Furthermore, they are likely to vary their
instructional approach from lesson to lesson.
Finally, it is necessary to reiterate the point made at the
beginning of this account of the various options: conceptualizing
FFI in terms of types and options is not unproblematic. The three
types of FFI rest on the distinction between focus on form and
focus on meaning. The question arises as t o how this focus is to be
determined. Whose perspective should be considered? Is the focus
to be determined in terms of the researcher’s or teacher’s intention
or in terms of particular learners’ response to instruction? It
certainly does not follow, for example, that instruction intended t o
focus learners’ primary attention on meaning will always achieve
this. Classroom learners may or may not respond in the way
intended. Ideally, therefore, researchers (and, perhaps, teachers
too) need to investigate learners’ responses t o different treatments
to ensure that these are in line with expectations.

Research Methods in FFI Research

FFI research, like classroom research in general, reflects

two broad traditions: confirmatory and interpretative research
(Anderson & Burns, 1989). The former tradition is evident in
correlational and experimental studies that involve manipulation
of the learning context and quantitative analyses. The latter
tradition is evident in descriptive and ethnographic studies of
contextualized practice in real classrooms and, more recently, in
studies of teachers’ cognitions about FFI; it emphasizes qualita-
tive analyses. While Tarone (1994) is probably right in asserting
that “researchers typically agree, in theory, that both qualitative
and quantitative methodologies are essential,” there is also evi-
dence of tension and opposition between advocates of these two
approaches. This is reflected in the tendency of researchers t o try
Ellis 27

to make a case for their chosen approach. Hulstijn (1997), for

example, argues the case for laboratory-based research, noting the
difficulty of keeping variables constant in natural learning envi-
ronments. Borg (1998),on the other hand, presents the case for an
exploratory-interpretative approach to FFI in order “to under-
stand the inner perspectives on the meanings of the actions being
studied (p. 11).This tension is also reflected in the obvious bias
toward confirmatory research in the journals that publish FFI
research (i.e., Language Learning, Modern Language Journal,
Studies i n Second Language Acquisition, and TESOL Quarterly).
Lazaraton (2000) reports the results of a study of data-based
articles in these journals over a 7-year period. She found that 88%
of the articles were quantitative, 10% were qualitative, and 2%
were partially qualitative. Researchers with a preference for work-
ing in the confirmatory tradition have a vested interest in protect-
ing their privileged position in the key journals; interpretative
researchers will have an understandable desire to challenge it.

Confirmatory Studies of FFI

Two main types of confirmatory research are evident in FFI

research: comparative and experimental studies. As noted in the
historical sketch of FFI research, there has been a gradual shift
from comparative t o experimental research. In fact, there have
been no comparative studies published in recent years. Compara-
tive studies (e.g., Ellis, 1989; Pavesi, 1986; Pica, 1983) compare
groups of naturalistic and instructed learners t o investigate
whether there are differences in the order or sequence of acquisi-
tion of grammatical features. There is an inherent problem with
such an approach. An assumption is made that might not be
justified-namely, that the instructed learners have indeed re-
ceived FFI (as opposed to some other kind of instruction) and that
the naturalistic learners have not engaged in any FFI (e.g.,
through consulting a grammar reference book or receiving explicit
negative feedback). Indeed, classifying learners as instructed or
naturalistic on the basis of the setting in which they are learning
28 Investigating Form-Focused Instruction

is necessarily a very crude procedure. Also, of course, when com-

parative studies do find evidence of differences between two such
groups of learners, it is not possible t o establish what aspects
of FFI are responsible, because no data on the FFI itself are
available.
Experimental studies can be carried out in laboratory-type
settings, in which case they may involve either a real language
(e.g., Robinson, 1996) o r an artificial language (e.g., de Graaff,
1997; DeKeyser, 1997), or in real classrooms (e.g., Doughty &
Varela, 1998), where for ethical reasons they invariably involve a
real language. The advantage of laboratory studies is that they
allow for rigorous control of extraneous variables and also for
careful replication; their disadvantage is that they may be lacking
in ecological validity (especially if such studies wish to propose
pedagogical applications). The obverse is true of experimental
studies conducted in real classrooms; they have strong ecological
validity, but control of variables and replication are problematic.
Experimental research assumes a “scientific”paradigm in which
theoretically derived hypotheses relating t o specific cause-and-
effect relationships are tested in a manner that will allow for
generalizable findings. To this end, it needs to conform t o a set of
rigorous design and reporting requirements. In so doing, meta-
analyses of the kind carried out by Norris and Ortega (2000) are
made possible. However, as Norris and Ortega point out, the
experimental FFI studies to date have frequently failed t o meet
these requirements. For example, they have tended t o use complex
designs t o investigate multiple variables and the interactions
among them rather than simple designs involving two or three
variables; they do not always incorporate pre- and post-tests and
a control group; they do not have designs that are easily replicable;
they often fail t o provide adequate information about the inde-
pendent, dependent, and moderator variables; they do not report
results in a manner that permits further interpretation and accu-
mulation of findings; or they suffer from some combination of these
failures. Such failures make it difficult to compare results across
studies and arrive a t definite conclusions. Doubtless, much can
 29

be done t o address these failures. Nevertheless, experimental

researchers will have t o continue t o grapple with the inherent
complexity of FFI and are likely t o resolve this difficultyin varying
ways, some opting for the kind of “simple”study Norris and Ortega
recommend but others preferring to try to reflect the natural
complexity of FFI in their designs. Calls for researchers t o conform
to a set of guidelines frequently go unheeded when the phenome-
non under investigation is a complex one, in part because of an
understandable desire on the part of researchers to achieve eco-
logical validity
There are no examples of comparative studies in this book.
Section 2 includes three experimental studies. Two of these (Day
& Shapson, 1991, and DeKeyser & Sokalski, 1996) took place in a
real classroom setting, while Leow’s study took place in a language
laboratory. Leow’s study, although pseudo-experimental in design,
employed a think-aloud task, a method of data collection found in
interpretative research (see below). All three studies were in-
cluded in Norris and Ortega’s meta-analysis of experimental stud-
ies and thus met the rigorous reporting requirements of these
authors.

Interpretative Research

Interpretative FFI research can be subdivided into two main

types-descriptive and introspective. Descriptive research can be
further classified according to whether the focus of the description
is the language produced by instructed learners or the classroom
discourse of FFI. Introspective studies of FFI are few and far
between. They entail attempts to investigate the perspective of the
participants in FFI (i.e., the teacher and the learners). Both
descriptive and introspective FFI research typically seeks an emic
perspective on FFI through qualitative analysis and in this con-
trasts with confirmatory FFI research with its etic perspective and
quantitative procedures.
Descriptive studies of learner language ideally need t o be
longitudinal t o show how instructed learners’ interlanguage
30 Investigating Form-Focused Instruction

develops over time. However, there have been surprisingly few

longitudinal studies of classroom learners. One reason, perhaps,
is the problem of collecting adequate data from individual learners
within a classroom setting. As has been widely noted (see
Chaudron, 1988, for a review of relevant studies), teacher talk
dominates in the classroom, so individual learners have limited
opportunities to speak. And when they do speak, their speech
typically consists of utterances shorter in length than a clause
(Allen et al., 1990). Furthermore, the kind of speech learners
typically produce in the classroom is highly controlled and, there-
fore, not ideally suited to an investigation of interlanguage devel-
opment, a point that is discussed more fully below. For this reason,
researchers make extensive use of what Corder (1981) has called
“clinical elicitation” (i.e., devices for eliciting relatively “natural”
language). For example, Ellis (1985) asked learners t o give oral
descriptions of What’s Wrong cards, and Bardovi-Harlig (1997 and
this volume) used learners’ written compositions and “guided
conversational interviews.” Ellis (1984b, 1992) is the only longitu-
dinal study of instructed learner language based solely on the
language that learners produce in the classroom itself. The
strength of descriptive studies of learner language lies in the
potential for plotting the effect of FFI on interlanguage develop-
ment. However, it is difficult for such studies to determine what it
is in the instruction that promotes (or impedes) acquisition, unless
they incorporate a detailed study of the instructional discourse.
With the exception of Lightbown (1983), no such comprehensive
FFI study has yet been undertaken.
Descriptive studies of FFI discourse have been plentiful.
These have examined instruction involving all three types of FFI
described in the previous section. Descriptive studies of Type 3
FFI have been particularly in evidence recently. These studies
involve the recording and transcribing of samples of instructional
discourse and the construction of data-driven taxonomies of dis-
course moves, instructional options, teaching strategies, etc. Such
studies undoubtedly contribute greatly to what Van Lier (1988)
sees as the main aim of interpretative research, an understanding
Ellis 31

of how instruction is accomplished in context. They reinforce what

critics of experimental studies have always argued, namely, that
FFI cannot be viewed as a general phenomenon but must be seen
as highly contextualized and variable. However, because each
study advances its own descriptive taxonomy, it is difficult to
compare results and thus t o make any generalizations. Descrip-
tive studies of Type 3 FFI suffer from another drawback. Whereas
it is possible to examine what effect focus-on-form has on learner
uptake (see Lyster, 199813 and this volume), it is very difficult to
investigate its effect on acquisition. Because, by definition, the
features attended to in this type of FFI are incidental, it is not
possible t o pre-test learners. Also, post-tests will need t o be
tailor-made t o target the specific features that each learner has
addressed during a 1ess0n.l~Such a testing procedure is prob-
lematic and time-consuming, although as Swain (1995) has
shown, it is not impossible.
Introspective studies seek to examine what beliefs the class-
room participants have about FFI and what their views and
interpretations of specific FFI events are. They make use of the
kinds of think-aloud and retrospective data collection methods
described by Faerch and Kasper (1987). These methods have been
used on both learners (as in Leow, 1997 and this volume) and
teachers (Borg, 1998). The strengths and disadvantages of these
methods are well known. They provide important information
about participants’ mental worlds and processing, which is en-
tirely lacking in both experimental and descriptive studies of
learner language and FFI. However, the validity and reliability of
introspective data is ~ontestab1e.l~
Despite Lazaraton’s (2000)finding that experimental studies
greatly outnumber interpretative studies in the leading journals
(includingLanguage Learning),the number of descriptive studies
in the present volume matches the number of experimental stud-
ies.15 Bardovi-Harlig’s 1997 study is a longitudinal study of
learner language, focusing on the acquisition of a specific gram-
matical feature (English present perfect tense). The studies by
Lyster (1998b) and Williams (1999) are descriptive studies of FFI;
32 Investigating Form-Focused Instruction

where Lyster considers teacher-student interactions, Williams

examines interactions among learners.

Hybrid Research

The distinction between the experimental and interpretative

research paradigms is, of course, not as clear cut as the above
discussion might suggest. Many FFI studies are examples of
hybrid research. Grotjahn (1987) identifies a number of “mixed
forms.” Leow’s (1997) study is an example of what Grotjahn calls
“experimental-qualitative-statistical”research (i.e., the design is
experimental, a qualitative method of data collection is used, and
the data are analyzed statistically). Williams’ (1999) study is an
example of “exploratory-qualitative-statistical~7research (i.e.,
the design involves an exploratory study of a real classroom, the
data consisted of samples of classroom interaction, and statistical
methods of analyzing the data were employed). Such hybrid
research is becoming increasingly common. It is motivated by
researchers’ felt need t o examine the social and mental processes
that contribute t o learning outcomes, a need that is best met
through a combination of experimental and interpretative
methods.

Measuring Acquisition i n FFI Research

Irrespective of whether the research is confirmatory or

interpretative, there is a need t o operationalize acquisition, the
dependent variable of all FFI enquiry. This section on the meth-
odology of FFI research, therefore, will conclude with a discus-
sion of this important matter.
In fact, the term “acquisition”can mean very different things.
First, it can mean onset (i.e., the initial introduction of new forms
into a learner’s interlanguage). This is the definition preferred by
Pienemann, Johnston and Brindley (1988). They suggest that a
feature can be considered acquired if a learner uses it twice in her
creative (nonformulaic)speech. The advantage of such a definition
Ellis 33

is that it acknowledges that learners’ interlanguages incorporate

nontarget language forms and rules and thus avoids what Bley-
Vroman (1983) has referred to as the “comparative fallacy.” The
problem with such a definition is that it is not always easy to
decide whether a particular utterance is creative or formulaic.
Also, for many researchers, onset is not an adequate definition
of acquisition precisely because it fails t o take into account
targetlike accuracy. Thus, FFI studies have typically opted for
“target-language accuracy” as the definition of acquisition. In
other words, they have proceeded on the assumption that the
more accurately a learner uses a feature, the more it has been
acquired. Such a definition accords more closely, perhaps, with
teachers’ own ideas of what acquisition means, but it is also
problematic. First, it falls foul of the comparative fallacy (ie.,
assumes that acquisition can only be measured in terms of target-
language norms). Second, it ignores the fact that interlanguage
development is U-shaped and that, in some cases, a learner may
use a specific form more accurately at an earlier stage of acquisi-
tion than a later. Cross-sectional FFI studies have little alterna-
tive but t o adopt accuracy of use as the definitional criterion.
Another advantage of longitudinal studies, therefore, is that they
can adopt a more organic view of acquisition.
Acquisition has been measured in terms of grammaticality
judgments, comprehension, and production. The use of grammati-
cality judgments is widespread in SLA research, including FFI.
However, controversy exists regarding both the validity and the
reliability of such a measure (Ellis, 1990;Gass, 1994).Comprehen-
sion measures involve designing sets of sentences that can only be
correctly comprehended if the learners process the target struc-
ture successfully. For example, VanPatten and Cadierno (1993)
asked learners t o listen t o sentences and match them with pictures
t o demonstrate their ability to comprehend the syntactic function
of clitic pronouns in Spanish. Production measures come in all
shapes and sizes, from highly mechanical tests (e.g., fill-in-the-
gap) t o the clinical elicitation of spontaneous language use by
means of communicative tasks.
34 Inuestiguting Form-Focused Instruction

Three points can be made about these various measures. The

first is that different measures can produce different results. For
example, the ability t o judge a sentence as grammatical or un-
grammatical does not necessarily correlate with the ability to
produce the targeted feature correctly. This is another reason
why comparisons of results across different studies are highly
problematic. Second, whereas mainstream SLA has viewed spon-
taneous oral language use as the best source of data for investi-
gating interlanguage development, FFI research has tended t o
rely on measures of controlled language production t o obtain
data on the specific target structure under study. The problem
here is that it is one thing t o demonstrate that FFI has an effect
on controlled production and entirely another t o demonstrate
that it affects spontaneous production. For example, VanPatten
and Sanz (1995) tested the effects of input-processing instruc-
tion with a variety of measures-an interpretation (i.e., compre-
hension) test, a sentence-completion test, a structured interview,
and a storytelling task. Furthermore, measures of production in
both oral and written modes were obtained. They found an effect
for the instruction on all measures except the oral storytelling
task. In other words, although they were able t o demonstrate that
input processing improved accuracy in controlled language use,
they failed to show that it had made any impact on spontaneous
production. Third, surprisingly-given the importance that jour-
nals such as Language Learning attach to rigorous design-FFI
researchers rarely bother t o demonstrate that their tests have
validity and reliability. VanPatten and Sanz, for example, do not
report reliability measures for any of the instruments they used.
In short, the problem of how to measure acquisition in FFI studies
remains not only unsolved but also largely ignored. A number of
researchers now opt for multiple measures (see, for example,
White, Spada, Lightbown, & Ranta, 1991;J. White, 1998),which,
when the results obtained concur, allow conclusions t o be made
with greater conviction. However, until FFI studies, as a matter of
routine, include some measure of learners’ ability t o process a
structure under real operating conditions (as in spontaneous
Ellis 35

speech), doubts will remain about the nature of the reported

instructional effects. This remains the single greatest problem
facing FFI researchers today.

Conclusion

Today, FFI constitutes a rich vein of enquiry. Whereas the

earlier research was concerned with whether FFI “worked,” cur-
rent research is directed at examining how it is accomplished in
different settings and what effect different types of FFI have on
L2 acquisition. This shift in direction has been informed by and
has also contributed t o a progressively refined conceptualization
of FFI options. These options have tended t o be based on develop-
ments in L2 acquisition theory rather than language pedagogy.
Likewise, the research examining these options has been increas-
ingly motivated by theoretical rather than pedagogical considera-
tions. As a result, there is now a clearer divide between what might
be called theoretically and pedagogically driven research than
there was a decade or so ago. This divide is reinforced by the
preponderance of experimental studies of FFI, which by necessity
seek to control variables rigorously, in ways that teachers may not
recognize as pedagogically valid. This research also suffers from
various design and reporting weaknesses that make it difficult to
make generalizations. One way forward is obviously to improve
the quality of these studies, but there is also a need for descriptive
and introspective studies that adopt a more emic and holistic
perspective on FFI and that view instructed L2 acquisition as an
organic process. Such studies may help t o narrow the divide
between theoretical and pedagogical concerns. They are also nec-
essary if Type 3 FFI, that is, incidental focus-on-form, is t o be
investigated. It is a healthy sign that descriptive studies of FFI do
finally appear t o be on the increase, as reflected in the number of
such studies published in Language Learning in the last year or
two and included in this book.
36 Investigating Form-Focused Instruction

Notes
lThe term “form-focusedinstruction” is not always used with such a general
meaning. Spada (1997),for example, defines FFI as consisting of “pedagogical
events which occur within meaning-based approaches to instruction but in
which a focus on language is provided in either spontaneous or predetermined
ways”(p.73).Spada’s definition excludes traditional instruction involving the
presentation and practice of discrete forms, while the definition followed in
this Introduction includes them.
2According t o skill-building models, learners acquire forms by proce-
duralizing explicit knowledge through production practice. In contrast, the
input-processing model claims that learners acquire forms by consciously
attending to them and the meanings they encode in the input.
3WhileL1 and age were not found to affect the order of acquisition,the method
of collecting data was. Specifically, the “universal” order was found in data
that reflected spontaneous language use but not in data that reflected
controlled language use (e.g., data from discrete-item tests).
4Analyticactivities are those that are based on “some kind of analysis of the
language, with an emphasis on grammar” (Stern, 1990, p. 94). Experiential
activities are those that focus on “some substantive theme or topic” (p. 101)
and that “create conditions for real language use” (p. 102).
5An example of a default strategy is the “first-noun strategy,” according to
which learners automatically assume that the first noun in an input string
is the agent, which in Spanish sentences such as “La sigue el sefior” leads t o
misinterpretation (see VanPatten, 1996, pp. 62ff).
%pads and Lightbown (1999) produced results that do challenge the claim
that form-focusedinstruction does not affect the sequence of acquisition. They
found no evidence to support the teachability hypothesis; i.e., in their study,
instruction in English questions was effective irrespective of the learners’
stage of acquisition.
7The idea that such attention to word meanings constitutes form-focused
rather than meaning-focused instruction is supported by Jessica Williams
(personal communication, July 28,2000). She gives this example: “If a learner
is reading a text for comprehension and some planned intervention draws
her attention briefly away from text comprehension and toward a word
(maybe textual enhancement or a gloss), then we have lexical FFI.”
8A synthetic syllabus consists of a list of linguistic features t o be taught
one-by-one. There are different ways of determining the linguistic content of
such a syllabus. One way is by consulting a pedagogic grammar. Another way,
which is the way advocated by Doughty and Williams (1998c),is to identify
the errors learners make as a basis for a remedial synthetic syllabus.
sVanPatten (1996) proposed that “the input strings must encode some mean-
ing that the learner is required t o attend to and respond t o in some way” (p.
67). However, what VanPatten had in mind is that learners give primary
attention not to meaning but to the meaning(s) realized by some specific
El 1is 37

target feature. Instruction involving structured input is often accompanied

by some form of explicit instruction, the two together comprising what
VanPatten (1996) calls “input-processing instruction.” Obviously the pres-
ence of explicit instruction means that learners will be aware of the target
structure and that the purpose of the instruction is focus-on-forms.However,
even if there is no explicit instruction (as in VanPatten & Oikennon, 1996)
the nature of the structured-input activities is likely to make it clear to
learners that the primary goal is to attend to form.
locare needs to be taken not to confuse preemptive and proactive FFI. The
latter term refers to Type 1and Type 2 FFI-that is, it involves preplanned
form-focused activities. Pre-emptive FFI arises incidentally and thus is Type
3 FFI. Of course, the kinds of discourse moves that occur in pre-emptive FFI
(e.g.,a teacher or learner query about a linguistic form) can also occur in other
types of FFI.
‘INegative feedback has also been explored in the context of FFI involving
focus-on-forms.A number of studies (e.g., Carroll, Swain, & Roberge, 1992;
Carroll & Swain, 1993) have examined the relative effects of explicit and
implicit negative feedback on acquisition following this kind of instruction.
121tis interesting to note that Spada and Lightbown (1993) classify repeti-
tions as a type of explicit negative feedback. This reflects the earlier comment
that the distinction between implicit and explicit negative feedback is to be
seen as a continuum. Repetitions are clearly more explicit than recasts but
also are clearly more implicit than explicit correction.
‘3Focus-on-formepisodes are typically not directed at the whole class but a t
individual learners, as when an individual student asks a question about the
meaning of a word or a teacher responds to an error made by an individual
student. However, other students have the opportunity to benefit from such
episodes by listening in.
14For example, think-aloud tasks require learners t o engage in dual
processing-performing some learning task and commenting on their think-
ing processes. Such dual processing is problematic to L2 learners, especially
less proficient ones. One cannot be sure, therefore, that the processes learners
report accurately reflect what they are thinking.
15Thisbalance reflects the overall balance of experimental and interpretative
FFI studies in Language Learning over the last 10 years. It would seem,
then, that Lazaraton’s 2000 finding that experimental studies greatly
outnumber interpretative does not apply to Language Learning (although
she included this journal in her analysis). There are two possible explana-
tions. First, Lazaraton looked at all empirical studies, not just those investi-
gating FFI. Second,it is not always easy to classify studies as “experimental”
or “interpretative,”because many involve mixed forms (see section on hybrid
research),
38 Investigating Form-Focused Instruction

References

Alanen, R. (1995). Input enhancement and rule presentation in second

language acquisition. In R. Schmidt (Ed.), Attention and awareness in
foreign language learning (pp. 259-302 ). Honolulu: University of Hawai’i
Press.
Allen, L. (2000). Form-meaning connections and the French causative: An
experiment in input processing. Studies i n Second Language Acquisition,
22,6944.
Allen, P., Swain, M., Harley, B., & Cummins, J. (1990).Aspects of classroom
treatment: Toward a more comprehensive view of second language educa-
tion. In B. Harley, l? Allen, J. Cummins, & M. Swain (Eds.),The development
of second language proficiency (pp. 57-81). Cambridge: Cambridge Uni-
versity Press.
Allwright, B. (1988). Observation i n the language classroom. London:
Longman.
Allwright, R. (1975).Problems in the study of language teachers’ treatment
of error. In M. Burt & H. Dulay (Eds.j, On TESOL ’75: New directions in
second-language learning, teaching and bilingual education (pp. 96-109).
Washington, DC: TESOL.
Anderson, R. (1993). Rules of the mind. Hillsdale, NJ: Erlbaum.
Anderson, R., & Burns, R. (1989). Research i n classrooms: The study of
teachers, teaching and instruction. Oxford: Oxford University Press.
Bardovi-Harlig, K. (1997).Another piece of the puzzle: The emergence of the
present perfect. Language Learning, 47,375-422.
Bley-Vroman, R. (1983). The comparative fallacy in interlanguage studies:
The case of systematicity. Language Learning, 33, 1-17.
Borg, S. (1998). Teachers’ pedagogical systems and grammar teaching: A
qualitative study. TESOL Quarterly, 32,9-38.
Borg, S. (1999a). Studying teacher cognition in second language grammar
teaching. System, 27, 19-31.
Borg, S. (1999b).The use of grammatical terminology in the second language
classroom.Applied Linguistics, 20, 95-126.
Cadierno, T. (1995). Formal instruction from a processing perspective: An
investigation into the Spanish past tense. Modern Language Journal, 79,
179-193.
Carroll, S., & Swain, M. (1993). Explicit and implicit negative feedback: An
empirical study of the learning of linguistic generalizations. Studies in
Second Language Acquisition, 15,357-366.
Ellis 39

Carroll, S., Swain, M., & Roberge, Y. (1992). The role of feedback in adult
second language acquisition: Error correction and morphological generali-
zations. Applied Pyscholinguistics, 13, 173-198.
Chaudron, C. (1977). A descriptive model of discourse in the corrective
treatment of learners’ errors. Language Learning, 2 7 , 2 9 4 6 .
Chaudron, C. (1988). Second language classrooms. Cambridge: Cambridge
University Press.
Corder, F! (1981). Error analysis and interlanguage. Oxford: Oxford Univer-
sity Press.
Day, E., & Shapson, S. (1991). Integrating formal and functional approaches
to language teaching in French immersion: An experimental study. Lan-
guage Learning, 41,25-58.
de Graaff, R. (1997). The experanto experiment: Effects of explicit instruction
on second language acquisition. Studies in Second Language Acquisition,
19,249-276.
DeKeyser, R. (1994). Implicit and explicit learning of L2 grammar: A pilot
study. TESOL Quarterly, 28,188-194.
DeKeyser, R. (1995). Learning second language grammar rules: An experi-
ment with a miniature linguistic system. Studies in Second Language
Acquisition, 17,379410.
DeKeyser, R. (1997). Beyond explicit rule learning: Automatizing second
language morphosyntax. Studies in Second Language Acquisition, 19,
195-221.
DeKeyser, R. (1998).Beyond focus on form: Cognitive perspectives on learning
and practicing second language grammar. In C. Doughty & J. Williams
(Eds.),Focus on form in classroom second language acquisition (pp. 42-63).
Cambridge: Cambridge University Press.
DeKeyser, R., & Sokalski, K. (1996). The differential role of comprehension
and production practice. Language Learning, 46,613-642.
Diller, K. (1978). The language teaching controuersy. Rowley, MA: Newbury
House.
Doughty, C., & Varela, E. (1998).Communicativefocus on form. In C. Doughty
& J. Williams (Eds.), Focus on form in classroom second language acquisi-
tion (pp. 116138).Cambridge: Cambridge University Press.
Doughty, C., & Williams, J. (Eds.). (1998a). Focus on form in classroom second
language acquisition. Cambridge: Cambridge University Press.
Doughty, C., &Williams, J. (1998b). Issues and terminology. In C. Doughty &
J.Williams (Eds.),Focus on form in classroom second language acquisition
(pp. 1-11).Cambridge: Cambridge University Press.
40 Investigating Form-Focused Instruction

Doughty, C., & Williams, J. (1998~). Pedagogical choices in focus on form. In

C. Doughty & J. Williams (Eds.), Focus on form in classroom second
language acquisition (pp. 197-262). Cambridge: Cambridge University
Press.
Eckman, F., Bell, L., & Nelson, D. (1988). On the generalization of relative
clause instruction in the acquisition of English as a second language.
Applied Linguistics, 9, 1-20
Ellis, N. (1995). Consciousness in second language acquisition: A review of
field studies and laboratory experiments. Language Awareness, 4,
121-146.
Ellis, N. (1996). Sequencing in SLA: Phonological memory, chunking, and
points of order. Studies in Second Language Acquisition, 18, 91-126.
Ellis, R. (1984a). Can syntax be taught? A study of the effects of formal
instruction on the acquisition of WH questions by children. Applied
Linguistics, 5,138-155.
Ellis, R. (198413).Classroom second language development. Oxford: Pergamon.
Ellis, R. (1985). Teacher-pupil interaction in second language development.
In S. Gass & C. Madden (Eds.), Input in second Eanguage acquisition (pp.
69-85). Rowley, MA: Newbury House.
Ellis, R. (1989).Are classroom and naturalistic acquisition the same? A study
of the classroom acquisition of German word order rules. Studies i n Second
Language Acquisition, 11,305-328.
Ellis, R. (1990). Grammaticality judgments and learner variability. In H.
Burmesiter and P. Rounds (Eds.), Variability i n second language acquisi-
tion (pp. 25-60). Oregon: Oregon University Press.
Ellis, R. (1992).Learning to communicate in the classroom. Studies in Second
Language Acquisition, 14, 1-23.
Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford
University Press.
Ellis, R. (1997). S L A research and language teaching. Oxford: Oxford Univer-
sity Press.
Ellis, R. (1998).Teaching and research: Options in grammar teaching. TESOL
Quarterly, 32, 39-60.
Ellis, R. (2000, April). Focusing on form: Towards a research agenda. Paper
presented at annual Regional English Language Centre Conference,
Singapore.
Ellis, R., Basturkmen, H., & Loewen, S. (in press). Learner uptake in commu-
nicative ESL lessons. Language Learning.
Ellis, R., & Takashima, H. (1999). Output enhancement and the acquisition
of the past tense. In R. Ellis, Learning a second language through interac-
tion (pp. 173-1881, Amsterdam: John Benjamins.
Ellis 41

Faerch, C., & Kasper, G. (Eds.). (1987). Introspection in second language

research. Clevedon, Avon: Multilingual Matters.
Fotos, S., and Ellis, R. (1991). Communicating about grammar: A task-based
approach. TESOL Quarterly, 25,605-628.
Fotos, S. (1994). Integrating grammar instruction and communicative lan-
guage use through grammar consciousness-raising tasks. TESOL Quar-
terly, 28,323-351.
Gass, S. (1982). From theory to practice. In M. Hines & W. Rutherford (Eds.),
On TESOL '81 (pp. 129-139). Washington DC: TESOL.
Gass, S. (1994). The reliability of second-language grammaticalityjudgments.
In E. Tarone, S. Gass, & A. Cohen (Eds.),Research methodology in second
language acquisition (pp. 303-322). Hillsdale, NJ: Lawrence Erlbaum.
Grotjahn, R. (1987). On the methodological basis of introspective methods. In
C. Faerch & G. Kasper (Eds.), Introspection in second language research
(pp. 54-81). Clevedon, Avon: Multilingual Matters.
Hamilton, R. (1994). Is implicational generalization unidirectional and maxi-
mal? Evidence from relativization instruction in a second language. Lan-
guage Learning, 44,613-642.
Harley, B. (1989). Functional grammar in French immersion: A classroom
experiment. Applied Linguistics, 19,331-359.
Hatch, E. (Ed.). (1978). Second language acquisition. Rowley, MA: Newbury
House.
Hulstijn, J. (1997). Second language acquisition research in the laboratory:
Possibilities and limitations. Studies i n Second Language Acquisition, 19,
131-143.
Johnson, K. (1996).Language teaching and skill learning. Oxford: Blackwell.
Jourdenais, R., Ota, M., Stauffer, S., Boyson, B., & Doughty, C. (1995). Does
textual enhancement promote noticing? A think-aloud protocol analysis.
In R. Schmidt (Ed.), Attention and awareness in foreign language learning
(pp. 183-216). Honolulu: University of Hawai'i Press.
Krashen, S. (1981).Second language acquisition and second language learn-
ing. Oxford: Pergamon.
Krashen, S., & Terrell, T (1983).The natural approach: Language acquisition
in the classroom. Oxford: Pergamon.
Larsen-Freeman, D., & Long, M. (1991).An Introduction to second language
acquisition research. London: Longman.
Lazaraton, A. (2000). Current trends in research methodology and statistics
in applied linguistics. TESOL Quarterly, 34, 175-181.
Leeman, J., Arteagoitia, I., Fridman, D., & Doughty, C. (1995). Integrating
attention to form in content-based Spanish instruction. In R. Schmidt
(Ed.),Attention and awareness i n foreign language learning (pp. 217-258).
Honolulu: University of Hawai'i Press.
42 Investigating Form-Focused Instruction

Leow, R. (1997). Attention, awareness and foreign language behavior. Lan-

guage Learning, 47,467-505.
Lightbown, P. (1983). Exploring relationships between developmental and
instructional sequences in L2 acquisition. In H. Seliger & M. Long (Eds.),
Classroom-oriented research in second language acquisition (pp. 2 17-243).
Rowley, MA: Newbury House.
Long, M. (1977). Teacher feedback on learner error: Mapping cognitions. In
H. Brown, C. Yorio, & R. Crymes (Eds.), O n TESOL '77 (pp. 278-293).
Washington, DC: TESOL.
Long, M. (1983). Does second language instruction make a difference? A
review of the Research. TESOL Quarterly 17, 359-382.
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. (1991). Focus on form: A design feature in language teaching
methodology. In K. de Bot, R. Ginsberg, & C. Kramsch (Eds.), Foreign
language research in cross-cultural perspective (pp. 39-52). Amsterdam:
John Benjamins.
Long, M. (1996). The role of the linguistic environment in second language
acquisition. In W. Ritchie & T. Bhatia (Eds.),Handbookof second language
acquisition (pp. 413-468). San Diego: Academic Press.
Long, M., Inagaki, S., & Ortega, L. (1998). The role of implicit negative
feedback in SLA: Models and recasts in Japanese and Spanish. The
Modern Language Journal, 82,357-371.
Long, M., & Robinson, P. (1998). Focus on form: Theory, research and practice.
In C. Doughty & J. Williams (Eds.), Focus on form in classroom second
language acquisition (pp. 1541). Cambridge: Cambridge University
Press.
Loschky, L., & Bley-Vroman, R. (1993). Grammar and task-based methodol-
ogy In G. Crookes & S. Gass (Eds.),Tasks and language learning: Integrat-
ing theory and practice (pp. 123-167). Clevedon, Avon: Multilingual
Matters.
Lyster, R. (1998a).Recasts, repetition, and ambiguity in L2 classroom dis-
course. Studies in Second Language Acquisition, 20,51-81.
Lyster, R. (1998b). Negotiation of form, recasts, and explicit correction in
relation to error types and learner repair in immersion classrooms. Lan-
guage Learning, 48,183-218.
Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake:
Negotiation of form in communicative classrooms. Studies in Second
Language Acquisition, 19,37-66.
Ellis 43

Mackey, A. (1999).Input, interaction and second language development: An

empirical study of question formation in ESL. Studies in Second Language
Acquisition, 21,557-587.
Mackey, A., & Philp, J. (1998). Conversational interaction and second lan-
guage development: Recasts, responses and red herrings? The Modern
Language Journal, 82,338-356.
Nobuyoshi, J., & Ellis, R. (1993). Focused communication tasks and second
language acquisition. ELT Journal, 47,203-218.
Nonis, J., & Ortega, L. (2000). Effectiveness of L2 instruction: A research
synthesis and quantitative meta-analysis. Language Learning, 50,
417-528.
Nunan, D. (1998).Teaching grammar in context. ELT Journal, 52, 101-109.
Oliver, R. (2000). Age differences in negotiation and feedback in classroom
and painvork. Language Learning, 50,119-151.
Pavesi, M. (1986).Markedness, discoursal modes and relative clause forma-
tion in a formal and informal context. Studies in Second Language Acqui-
sition, 8,38-55.
Pica, T. (1983). Adult acquisition of English as a second language under
different conditions of exposure. Language Learning, 33,465497.
Pica, T., Holliday, L., Lewis, N., & Morgenthaler, L. (1989). Comprehensible
output as an outcome of linguistic demands on the learner. Studies in
Second Language Acquisition, 11, 63-90.
Pienemann, M. (1984). Psychological constraints on the teachability of lan-
guages. Studies in Second Language Acquisition, 6, 186-214.
Pienemann, M. (1985).Learnability and syllabus construction. In K. Hylten-
stam & M. Pienemann (Eds.), Modelling and assessing second language
acquisition (pp. 23-112). Clevedon, Avon: Multilingual Matters.
Pienemann, M. (1989).Is language teachable? Psycholinguistic experiments
and hypotheses. Applied Linguistics, 10,52-79.
Pienemann, M., Johnston, M., & Brindley, G. (1988). An acquisition-based
procedure for second language assessment. Studies in Second Language
Acquisition, 10,217-243.
Robinson, P. (1996). Learning simple and complex rules under implicit,
incidental rule-search conditions, and instructed conditions. Studies in
Second Language Acquisition, 18,27-67.
Rosa, E., & O’Neill,D. (1999).Explicitness, intake, and the issue of awareness:
Another piece to the puzzle. Studies in Second Language Acquisition, 21,
5 11-553.
S a l a b e q , M. (1997).The role of input and output practice in second language
acquisition. Canadian Modern Language Review, 53,422-451.
44 Investigating Form-Focused Instruction

Samuda,V. (in press). Getting relationship between form and meaning during
task performance: The role of the teacher. In M. Bygate, P. Skehan, &
M. Swain (Eds.), Task-based learning: Language teaching, learning and
assessment. Harlow, Essex, England Pearson.
Sanz, C., and VanPatten, B. (1998). On input processing, processing instruc-
tion, and the nature of replication tasks: A response to Salaberry. Cana-
dian Modern Language Review, 54, 263-273.
Schachter, J. (1974). An error in error analysis. Language Learning, 27,
205-214.
Scherer,A,, & Wertheimer,M. (1964).Apsycholinguisticexperiment in foreign
language teaching. New York: McGraw Hill.
Schmidt, R. (1990). The role of consciousness in second language learning.
Applied Linguistics, 11, 129-158.
Schmidt, R. (1994).Deconstructing consciousness in search of useful defini-
tions for applied linguistics. AlLA Review, 11, 11-26.
Schmidt, R. (1995a).Consciousness and foreign language learning: A tutorial
on the role of attention and awareness in learning. In R. Schmidt (Ed.),
Attention and awareness i n foreign language learning (pp. 1-63). Hono-
lulu: University of Hawai'i Press.
Schmidt, R. (Ed.) (199513). Attention and awareness in foreign language
learning. Honolulu: University of Hawai'i Press.
Schmidt, R., & Frota, S. (1986). Developing basic conversational ability in a
second language: A case study of an adult learner of Portuguese. In R. Day
(Ed.),Talking to learn: Conversation i n a second language (pp. 237-326).
Rowley, MA: Newbury House.
Schwartz, B. (1993). On explicit and negative data effecting and affecting
competence and linguistic behavior. Studies in Second Language Acquisi-
tion, 15,147-163.
Seedhouse, P. (1997a). The case of the missing 'no': The relationship between
pedagogy and interaction. Language Learning, 47,547-583.
Seedhouse, P. (1997b). Combining meaning and form. English Language
Teaching Journal, 51,336-344.
Sharwood Smith, M. (1993). Input enhancement in instructed SLA. Studies
in Second Language Acquisition, 15,165-179.
Skehan, P. (1998).Task-based instruction.Annua1Review ofApplied Linguis-
tics, 18,268-286.
Smith, P. (1970).A comparison of the audiolingual and cognitive approaches
to foreign language instruction: The Pennsylvania Foreign Language
Project. Philadelphia: Center for Curriculum Development.
Spada, N. (1997).Form-focused instruction and second language acquisition:
A review of classroom and laboratory research. Language Teaching, 30,
73-87.
Ellis 45

Spada, N., & Lightbown, P. (1993). Instruction and the development of

questions in L2 classrooms. Studies i n Second Language Acquisition, 15,
205-224.
Spada, N., & Lightbown, P. (1999). First language influence and developmen-
tal readiness in second language acquisition. The Modern Language Jour-
nal, 83, 1-21.
Stern, H. (1983). Fundamental concepts of language teaching. Oxford: Oxford
University Press.
Stern, H. (1990). Analysis and experience as variables in second language
pedagogy. In B. Harley, P. Allen, J. Cummins, & M. Swain (Eds.), The
development of second language proficiency (pp. 93-109). Cambridge:
Cambridge University Press.
Swain, M. (1995). Three functions of output in second language learning. In
G. Cook & B. Seidhofer (Eds.), For H. G. Widdowson:PrincipZes andpractice
in the study of language (pp. 125-144). Oxford: Oxford University Press.
Tarone, E. (1994). Analysis of learner’s language. TESOL Quarterly, 28,
676-678.
Tomlin, R., & Villa, V. (1994). Attention in cognitive science and second
language acquisition. Studies i n Second Language Acquisition, 16,
183-203.
Tomosello, M., & Herron, C. (1988). Down the garden path: Inducing and
correcting overgeneralization errors. Applied Psycholinguistics, 9,
237-246.
Tomosello, M., & Herron, C. (1989). Feedback from language transfer errors:
The garden path technique. Studies in Second Language Acquisition, 11,
385-395.
Trahey, M. (1996). Positive evidence in second language acquisition: Some
long term effects. Second Language Research, 12,111-139.
Trahey, M., &White, L. (1993). Positive evidence and preemption in the second
language classroom. Studies in Second LanguageAcquisition, 15,181-204.
Truscott, J. (1996).The case against grammar correction in L2 writing classes.
Language Learning, 46,327-369.
Truscott, J. (1998). Noticing in second language acquisition: A critical review.
Second Language Research, 14,103-135.
Turner, D. (1979). The effect of instruction on second language learning and
second language acquisition. In R. Andersen (Ed.), The acquisition and use
of Spanish and English as first and second languages (pp. 107-116).
Washington, DC: TESOL.
Van Lier, L. (1988). The classroom and the language learner. London:
Longman.
46 Investigating Form-Focused Instruction

VanPatten, B. (1990). Attending to content and form in the input: An expen-

ment in consciousness. Studies in Second Language Acquisition, 12,
287-301.
VanPatten, B. (1996). Input processing and grammar instruction in second
language acquisition. Norwood, NJ: Ablex.
VanPatten, B., & Cadierno, T. (1993). Explicit instruction and input process-
ing. Studies in Second Language Acquisition, 15,225-243.
VanPatten, B., & Oikennon, S. (1996). Explanation versus structured input
in processing instruction. Studies in Second Language Acquisition, 18,
495-5 10.
VanPatten, B., & Sanz, C. (1995).From input to output: Processing instruction
and communicative tasks. In F. Eckman et al. (Eds.), Second language
acquisition theory and pedagogy (pp. 169-185). Mahwah, NJ: Lawrence
Erlbaum.
White, J. (1998). Getting the learners’ attention: A typographical input
enhancement study. In C. Doughty & J. Williams (Eds.), Focus on form i n
classroom second language acquisition (pp. 85-113). Cambridge: Cam-
bridge University Press.
White, L. (1989). Universal Grammar and second language acquisition.
Amsterdam: John Benjamins.
White, L. (1991).The verb-movement parameter i n second language acquisi-
tion: Some effects of positive and negative evidence in the classroom.
Second Language Research, 7,133-161.
White, L., Spada, N., Lightbown, P., & Ranta, L. (1991). Input enhancement
and L2 question formation. Applied Linguistics, 12,416-432.
Widdowson, H. (1986).Design principles for a communicative grammar. In C.
Brumfit (Ed.), The practice of communicative teaching (ELT Document No.
124). Oxford: The British Council.
Widdowson, H. (1998).Context, community and authentic language. TESOL
Quarterly, 32,705-716.
Williams, J. (1999). Learner-generated attention to form. Language Learning,
49,583-625.
Williams, J., & Evans, J. (1998).What kind of focus and on which forms? In
C. Doughty & J. Williams (Eds.), Focus on form in classroom second
Language acquisition (pp. 139-155). Cambridge: Cambridge University
Press.