System 30 (2002) 1–14

[Link]/locate/system

Contexts of engagement: a discourse perspective

on ‘intake’ and ‘pushed output’
Rob Batstone*
School of Culture, Language and Communication, Institute of Education, University of London,
20 Bedford Way, London WC1H OAL, UK

Received 7 August 2001; received in revised form 28 September 2001; accepted 30 September 2001

Abstract
Second Language Acquisition (SLA) research has come under fire in recent years for
focussing too much on cognitive processes (such as ‘noticing’, ‘intake’ and ‘pushed output’)
and too little on the contexts of engagement within which such processes may (or may not)
occur (e.g. Firth and Wagner, 1997. Modern Language Journal 81, 286–300). These concepts
are nevertheless of considerable importance for understanding second language learning.
Rather than abandoning them in search of alternative paradigms, perhaps we need to balance
the emphasis given previously to cognitive factors by examining how these processes are ulti-
mately dependent on particular contextual conditions. Two aspects of contextual engagement
are focussed on here—‘communicative contexts’ and ‘learning contexts’. The former are seri-
ously flawed in respect to the needs of the initial language learner. Instead, learning contexts
need to be fashioned which enable learners to exploit contextual cues for intake, and take
advantage of an interpersonal context which supports rather than inhibits the risk-taking
which necessarily accompanies learners’ early attempts to deploy new forms in their own
output. # 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Attention; Communicative; Complexity; Context; Cotext; Discourse; Intake; Motivation;
Planning; Purpose

* Tel.: +44-207-612-6535.
E-mail address: [Link]@[Link] (R. Batstone).

0346-251X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0346-251X(01)00055-0
2 R. Batstone / System 30 (2002) 1–14

1. Introduction and overview

1.1. Defining context and contextual orientations

Despite the growth over recent years in scholarly concern with contexts for lan-
guage learning (e.g. Breen, 1985; Pierce, 1995; Allwright, 1997; Firth and Wagner,
1997; Tarone, 1997; Lantolf, 2000), much research into the process of second lan-
guage acquisition over the past 20 years has paid little heed to matters of context or
to the ways in which contextual conditions might intrude on the learning process,
preferring instead to develop a set of relatively abstract concepts which beg many
questions about the kind of contexts in which they may or may not occur (Faerch and
Kasper, 1986; Kasper and Blum-Kulka, 1983; Sato, 1986; Sharwood Smith, 1986).
The purpose of this article is to examine three of these concepts in particular—
intake, noticing and pushed output—in terms of the kind of context which may
most effectively facilitate them.
Context, for present purposes, has two dimensions. One is the external context,
including all those contextual components which, following Jakobson (1960) and
Hymes (1962) have come to be referred to as features of the context of the speech
event, or the context of situation (Firth, 1957). These include participants in the
speech event; the channel (whether the discourse is spoken or written); the topic;
the discourse type (e.g. whether the discourse is dialogue or monologue); the setting;
and the linguistic context1 (in the sense of language which contextualizes specific
forms new to the learner and thereby makes them more learnable).
The second dimension is a more dynamic one, having to do with how the language
learner is oriented to context. It is assumed that learners (like any discourse partici-
pant) interpret and act on context in significantly different ways depending in parti-
cular on their sense of purpose. In essence, this second dimension integrates the
external context with the ‘internal context’ of the learners’ inner orientation. We
need both dimensions if we want to explain how learners actually behave in contexts of
engagement, and many scholars define context with this in mind, by linking it directly
to matters of discourse and discourse interpretation (e.g. Cook, 1989, p. 156).
The central question which this article addresses is: what kind of context best
supports learners as they attempt to make sense of or with novel linguistic forms for
a learning purpose? Very often scholars have attempted to answer this question by
reference to communicative contexts based on notions of social acceptability. It is
argued here, though, that language learning is facilitated by a different kind of
behaviour—one in which learners go significantly beyond what is required or what is
validated on grounds of communicative acceptability. For these learners the main
effect of communicative contexts is not to facilitate attending to new language but to
constrain or even deny it. An alternative context, fashioned specifically to enhance
pedagogic accessibility (i.e. access to new language in input and in output) over
communicative acceptability, is proposed and outlined.

1
I use the term ‘linguistic context’ where others prefer the term ‘co-text’ (e.g. Brown and Yule, 1983)
or ‘previous discourse co-ordinate’ (Lewis, 1972).
 R. Batstone / System 30 (2002) 1–14 3

1.2. Contextual requirements for initial learning

This article focuses on the early stages in the learning of new linguistic forms.
Initial learning of this kind involves being able to make sense of new linguistic forms,
and in particular learning about how particular lexical and grammatical forms serve
to encode particular meanings. This ‘form/function mapping’ is a vital but relatively
under-researched aspect of a wider process whereby learners begin to internalise
information about language by converting some of their input (the language sur-
rounding them) into ‘intake’. In other words, they begin to make sense of new
language in ways which help them progressively to develop and organize their
developing interlanguage system (Van Patten, 1990, p. 227). It is generally agreed
that new intake of this kind is likely to require conscious attentional processing; in
other words, the new linguistic form will need to be consciously ‘noticed’ if it is to
become part of intake (Schmidt, 1990).
Equally important is the need for learners to engage with language in their own
output which is similarly developmental, so that by readily calling on a rich lin-
guistic repertoire they can progressively ‘automatize’ their knowledge (Schmidt,
1992). As with new intake, the learners’ early efforts to output new forms are likely
to require conscious attention, since the ease with which competent users call on
language in their output is something which is only gradually accomplished
(Schmidt, ibid.). Because of this, and also because such output often requires the
selection of a more complex and challenging form over a simpler paraphrase, it is
sometimes known as ‘pushed output’ (Swain, 1985).
These early encounters with new language are difficult to achieve without con-
textual support. It is argued that context plays a critical role in helping learners
to achieve new intake by drawing inferences about how form and function work
together. This involves providing contextual sources of meaning which serve to
reinforce the meaning of new forms. For example, a learner of Italian might be able
to infer that ‘la carta!’, uttered in a restaurant by an expectant patron, means ‘the
menu’ by observing such a form in use with appropriate contextual backup (waiter
disappearing and returning with menu in hand, etc). Or instead of the setting, the
source of such information might be in the linguistic context, as with the sentence ‘I

Table 1
Contextual dimensions and requirements for learning

Contextual aspect Requirement for initial learning

Ideational context
How context can ‘contextualize’ the meaning of Context insofar as it can provide situational or
specific linguistic forms in learners’ input by cotextual signals which potentially help the
reinforcing their meaning learner infer the meaning of new language as intake

Interpersonal context
How learners’ awareness of others in the context Learners need contextual support to counter
potentially affects their ability to focus on new the potentially inhibiting conditions associated
linguistic forms for learning with output pushing and consequent risk-taking
4 R. Batstone / System 30 (2002) 1–14

went to the cinema yesterday’ where the word ‘yesterday’ acts as a cue to the past
time semantic in the verb (‘went’). I refer to this aspect of context as ‘ideational’
because it deals with the contextual signalling of content, and hence with ‘‘the
expression of experience’’ (Halliday, 1973, p. 99).
Secondly, deploying new forms in one’s output can be difficult for socio-
psychological reasons. Pushing one’s output means taking the risk of making
errors, thereby causing embarrassment in the eyes of others. This aspect of context is
therefore referred to as the ‘interpersonal’ aspect. Table 1 summarised these points.

2. Two kinds of context

2.1. Communicative contexts and their constraints on learning

Over the past 20 years the kind of discourse behaviour and discourse contexts
most often proposed in applied linguistics as a basis for language pedagogy has been
communicative (e.g. Widdowson, 1983, 1998). This is taken here to refer to a view of
how competent users behave in discourse—one where language is used primarily as
a communicative means to an end—as a linguistic resource exploited to act on and
change context in some way (Widdowson, 1983, p. 25). It is clear that learning a
language involves learning to be communicatively adept. What is less clear is that
communicative language and contexts of communicative language use are the best
equipped to meet initial learning needs. It is argued here that there are two senses in
which communicative contexts serve these particular learning needs very poorly.
Firstly, communicative behaviour in its ideational context. As was suggested in
Section 1.2, in order to help learners map functions onto new forms in their input,
contexts need to be contrived which provide learners with necessary conceptual cues.
The difficulty here is that context is not just an external configuration but also a
matter of internal orientation, and the communicative orientation typically means
exploiting contextual cues so as to avoid the need to engage any further with new
language, rendering redundant the very target language forms which ought to be the
object of the exercise.
Secondly, communicative behaviour in its interpersonal context. In commu-
nicative contexts, as is well known, there are norms of appropriacy which limit how
much is deemed acceptable to say, while it is a fundamental tenet of commu-
nicatively appropriate language use that one strives at all times to remain coherent.
Each of these norms serves as a contextual constraint on the learner who might need
to risk temporary incoherence, and who might benefit from being able to hold the
floor indefinitely, whilst trying to utilise more complex forms in their output (e.g.
Aston, 1986; Foster, 1998).
In essence, the problem is that in communicative discourse there are boundaries to
acceptability which delimit the learner’s capacity to exploit context for learning,
both in its ideational and in its interpersonal functions. Ideationally it limits what is
acceptable and relevant to pay attention to; interpersonally it limits what and how
much it is acceptable to say.
 R. Batstone / System 30 (2002) 1–14 5

It is also suggested that certain recent ideas for supporting intake and pushed
output in language pedagogy fail to take adequate account of these contextual
constraints. Two areas of contemporary research are examined. The first relates to
intake, and to the belief that learners can effectively be forced to attend to new
language by removing any possible contextual distractions (e.g. Cadierno, 1995;
Van Patten, 1996). This proposal may be effective in other respects but as regards
the initial need to make form/function mappings it suffers from a logical difficulty:
learners cannot be expected to make sense of new language if the contextual cues
they need to process it have been removed. The second set of proposals relates to
output, and to task-based provisions for planned or repeated task production (e.g.
Bygate, 1996; Foster and Skehan, 1996). These proposals have earned consider-
able empirical support, but it is argued that their effectiveness is nonetheless con-
tingent on the kind of interpersonal context and interpersonal support in which
they occur—a factor not generally considered by the researchers who advocate
them.

2.2. The fundamental orientation in a learning context: accessibility

The communicative context rests on what is primarily a sociolinguistic conception

of the world, with language accessed in reference to criteria of situational need and
interpersonal acceptability. As already suggested, this imposes certain constraints on
learners which they need to override. What is needed is a context which validates
such behaviour by placing less emphasis on acceptability in sociolinguistic terms and
more emphasis on accessibility in psycholinguistic terms, in the sense that it facil-
itates access to new language in both input and pushed output. Such a ‘learn-
ing context’ contrasts with the communicative context both in terms of the external
context and in terms of the learners’ orientation to context.
For example, in the case of input, external contextual cues are provided which
need to be perceived as relevant (rather than redundant) by learners because they are
acting explicitly on a learning purpose. In the case of output, the potential for
interlanguage pushing implicit in the task and other external features of context can
more easily be converted into actuality, for instance because the learner is function-
ing as part of a wider learning culture, shared by other learners, which validates and
encourages what in other circumstances might be prohibitively face threatening. The
key differences between the two are summarised in Table 2.

3. Context and its ideational function: making sense of new input

3.1. The paradox of communicative input

On encountering the following sentence the learner might use the phrase ‘three
days ago’ (part of the linguistic context) to infer the meaning of completed past time
and hence pass over its grammatical encoding in the past tense (which would con-
sequently remain unnoticed):
6 R. Batstone / System 30 (2002) 1–14

Table 2
Key differences between communicative and learning contexts

Contextual aspect Communicative context—emphasis Learning context—emphasis on

on acceptability, sociolinguistic in accessibility, psycholinguistic in
orientation orientation

Ideational Problems of learner orientation to Situational and cotextual cues

context: learners attend to language deliberately and systematically
only to the degree necessary, so that provided, and learners are
cotextual cues are used instead of oriented to see them as
(rather than in alliance with) new pedagogically salient and
language in the input necessary to support intake
Problems of external context: if such
cues are removed initial intake is
effectively prevented from occurring

Interpersonal Interpersonally appropriate behaviour Tasks provide various external

generally means avoiding incoherence contextual features which
and not saying more than necessary— potentially resource output
conditions which are inherently in pushing (planing, task repetition,
tension with the need for output etc) but are cotingent upon
pushing and general interlanguage learners sharing an interpersonal
risk-taking orientation to context which
supports and validates such
behaviour

1. I telephoned her three days ago

Van Patten (1996, p. 21) argues that many of the semantic notions encoded
grammatically in a language tend also to get encoded lexically. As well as past time,
Van Patten cites the third person -s morphology (also signalled lexically by the
choice of noun as subject) and plurality (often marked by lexical items such as
‘many’ or ‘two’). According to Van Patten, learners process such lexical markers in
preference to their grammatical counterparts, and he cites a number of studies where
learners provided with both lexical and grammatical cues are observed consistently
to prefer the former over the later (Musumeci, 1989; Cadierno et al., 1991).
The solution proposed by Van Patten and by a number of other scholars is to set
up conditions where such contextual cues are removed, their reasoning being that if
the learner is deprived of the contextual means for by-passing the target form, then
she will have no processing option other than to engage with it. Thus, instead of
encountering the sentence in 1 above, the learner is given a sentence such as the
following, and is asked to say whether past or present time is being referred to:

2. I telephoned her

This is exactly what is proposed in a study undertaken by Cadierno (1995), and a

similar methodology is used with respect to other aspects of the language system
 R. Batstone / System 30 (2002) 1–14 7

(Van Patten and Cadierno, 1993). Van Patten and colleagues argue (quite rightly)
that learners who become dependent on cotextual/lexical cues will never develop
processing strategies necessary to attend adequately to grammar, and one can cer-
tainly make a strong case that the kind of careful cotextual depletion used in exam-
ple 2 strongly orients learners who are already at least partially acquainted with the
target forms to continue to attend to them for further processing.
However, there are serious difficulties here if such a methodology is applied in the
case of learners who are struggling to make initial links between form and function
in the input. Indeed, if we were to look for a deficit learning context—one which
systematically eliminated opportunities for learning—then we could do no better
than engineer a scenario such as this. In a very real sense what such contrivances do
is remove any vestige of context which might be of use to the learner.

3.2. Making sense of input in a learning context; the external context

In a communicative context, as was noted earlier, the learner’s primary purpose is

to deploy the linguistic code ‘‘as a resource for the creation of meaning’’ (Widdow-
son, 1983, p. 25). The direction of resourcing in a communicative context is the
reverse from what is required for intake. Instead of being positioned as the supplier
in a system based on contextual demand, what is needed is a context which supports
the learner, so that it is not the language which is the resource, but the context.
This brings us back to the conditions which Van Patten and Cadierno find so
problematic, for these very contextual cues may initially be necessary sources of
information about language. Such cues might be located in the linguistic context, so
that the adverbial phrase in example 1 functions to orient the learner to the meaning
of past time so that she can then use this knowledge to infer the connection with the
past tense grammar—a connection necessary if new learning is to occur, and one
which requires an attentional shift from lexis (or situational context) to grammar, as
is indicated in sentence three:

1. I telephoned her three days ago

Situational context is another potentially rich source of conceptual/linguistic

information, particularly in the case of stereotypic situations like restaurant rou-
tines or service encounters. Assuming that the learners already have the requisite
schematic familiarity with such events upon which to draw, they can again make
connections between recognised convention, context and new linguistic forms,
thereby inferring the meaning (for instance) of various lexical items and forms of
request.

3.3. Orienting to input in a learning context

But according to Van Patten (1996), it is a ‘given’ that learners will process
lexical cues before and in preference to grammatical forms, adopting a least-effort
strategy in order to focus on meaning with the minimum necessary effort (p. 21). If
8 R. Batstone / System 30 (2002) 1–14

this were universally true, then there would be little point in engineering the kind
of contextual guidance argued for in Section 3.2. The solution to this apparent
paradox cannot be simply to remove such cues from the external context, but
rather to look to the other aspect of context for a solution—the learners’ con-
textual orientation.
Learners surely need to let go of their default least-effort orientation, instead
adopting a learning purpose through which they understand the avowedly pedago-
gic purpose of (for instance) classroom presentations of new language—a process
which is supported and guided so that they can more easily make the critical pro-
cessing shift from context to targeted linguistic item (e.g. Samuda, 2001). Through
such a process, context is seen as a construct deliberately designed to illuminate new
language, and learners are oriented to it (and within it) as learners rather than as
least-effort users.

3.4. Contexts for intake in language pedagogy

It might well be objected that this is no more than already happens quite commonly
in language pedagogy, where it is customary to focus very explicitly on new lan-
guage through presentation-stage work which routinely involves using contexts to
accompany new items of language (restaurants and other service encounters being
particular favourites). Sometimes, though, such presentations are flawed because
although context and new language co-occur, they are not made to converge in the
eyes of the learner.
Very often pedagogic contexts for intake are constructed with inadequate thought
to how they might be made pedagogically salient. For example, one long established
and very popular textbook (Doff et al., 1983, p. 37) presents various possible uses of
the present progressive form, including describing events as they occur. A text is
presented for each meaning and the leaner is presented with the following decep-
tively straightforward rubric: ‘all the passages are talking about ‘now’. In what ways
are they different?’ In principle the task is well conceived, setting up written contexts
designed to prompt learners to draw contextual inferences about tense and time. As
it stands, though, the exercise works more as a test than a task. For instance, the
passage for describing events at the time of speaking begins ‘‘ . . .and now the Royal
Coach is turning into Parliament Square.’’ The effectiveness of such a cue clearly
depends on the learner having the requisite background schematic knowledge, and
might anyway require some careful ‘scaffolding’ from the teacher for the right con-
ceptual connections to be made (Donato, 1994).

3.5. Intake in SLA: Noticing the Gap vs Noticing the Link

Amongst applied linguists, too, this is a contextual function little exploited. The
basic principle was clearly stated long ago by Krashen (1982, p. 21) who argued
that since new input must by definition be incomprehensible, the learner will need
to circumvent the problem by going outside the target forms and taking her bear-
ings from the surrounding context. But the idea has been little taken up, in part
 R. Batstone / System 30 (2002) 1–14 9

perhaps because input studies in SLA have tended not to discriminate explicitly
between intake of functional knowledge and intake of formal knowledge. The lat-
ter has received much greater attention that the former, often being discussed in
terms of the learners’ need to ‘notice the gap’ between their existing hypothesis
about how a certain form is structured, and its actual encoding in target language
(Schmidt and Frota, 1986). But despite its utility and notwithstanding its huge
influence in SLA, the ‘gap’ metaphor has left little room for context. Implicitly, at
least, it tends to highlight significant points of difference between contrasting
forms, rather than significant points of linkage between contexts and forms which
occur within them.

4. Context and its interpersonal function: output

4.1. Claims made for tasks designed to ‘push’ output

The past 15 years or so have seen the emergence of a new kind of learning activity
in SLA research. Its primary exemplars are task repetition (e.g. Hulstijn and Hul-
stijn, 1984; Ellis, 1987; Bygate, 1996), and pre-task planning (e.g. Crookes, 1989,
Williams, 1992; Foster and Skehan, 1996; Ortega, 1999). These studies continue to
attract interest because they are said to lead to learner output which is more com-
plex, and/or more accurate, and/or more fluent than is often the case with the
unplanned/first-time-around task. In general, the claim is that such tasks enable
the learners to collect their thoughts and to reach for language which is richer or more
accurate on the basis of having had the time either to plan to do a task twice over.
Given that on the whole the results of such research seem to show a clear positive
effect on learner language, it is easy to fall into the kind of shorthand (particularly
common in cognitive accounts) whereby features of task design are discussed as if
they can lead directly to positive effects on learner performance. The simplest effect
of allowing learners planning time, says Skehan (1998), is to free-up on-line proces-
sing resources. He goes on:

The more that is planned. . ., the less computational work needs to be done
during the task. Other things being equal, the result will be that more attention
is available as a general-purpose tool to achieve a variety of goals: greater flu-
ency, complexity, or accuracy (pp. 73–74).

The level of generalisation in such accounts, combined with the emphasis given to
task design features, leaves one with the impression that task themselves have the
capacity to ‘push’ learners output to new levels. . .other things being equal.

4.2. Face sensitivity and the contextual cost of output pushing

But other things are not always equal. If we de-emphasise the task perspective
momentarily and look instead at such tasks in terms of the potential contextual
10 R. Batstone / System 30 (2002) 1–14

constraints on learner production, then a different picture emerges. One thing which
is particularly striking about output ‘pushing’ is that it incurs threats to ‘face’
(Goffman, 1967) which might well deter many learners from taking full advantage of
the opportunity offered in theory by pre-task planning. By placing such faith in task
design per se, we run the risk of focussing too much on aspects of the external con-
text and not enough on the learners’ internal orientation within it.
In part, pushing ones’ output in an attempt to access novel language which is
complex is face-threatening, because it risks incoherence—something which in
everyday communicative discourse we do all we can to avoid, both in our own dis-
course and in interpreting the discourse of others (Cook, 1989, pp. 28–29). In one
study on the effects of repeating a problem-solving task, Irving (1999) interviewed
the learners after the repeated task and discovered an unanticipated and very tangi-
ble awareness of this kind of risk, and a consequent reluctance to take too many
linguistic chances. She quotes one of the participants as follows:

The words are not coming and you think. . .I don’t know it’s different you just
had to talk and you say stupid things and you make stupid mistakes which you
realize. . .and it was eas—yeh this time was like ‘arghh! stupid mistake!’ and you
think ‘oh no no no!’. (Irving, 1999, p. 106)

It is surely of some significance that this learner opted to talk in a long, un-
interrupted monologue the second time around, despite having an interlocutor with
whom she could have interacted. Monologues are commonly used in this kind of
research (e.g. Ellis, 1987, 1994, 1995; Crookes, 1989; Crookes and Gass, 1993;
Dougherty and Williams, 1998; Williams, 1992; Bygate, 1996; Ortega, 1999). Yet
despite their potential advantages in terms of allowing greater attentional space,
monologues also put the learner in a very exposed and isolated position in terms of
the interpersonal context—the presence of fellow students and a researcher can itself
trigger considerable self-consciousness, while the learner does not have the comfort
or the communicative camouflage afforded by more interactive encounters where
one can hide behind shifts in topic or turn-taking.
In fact this particular learner was able nonetheless to push her language to a
considerable extent. One wonders, though, how often individual learners simply fail
to push their language to any significant degree on account of contextual constraints
which are obscured in the drive towards averaging out overall statistical results—a
process in which the positive performance of some participants may have a dis-
proportionate effect on the overall statistics (Foster, 1998). In one recent study on
the effects of prior planning, Gilanlioglu (2001) quotes one learner whose reluctance
to take linguistic chances was such that he utilised the planning phase as an oppor-
tunity simply to close down on any foreseeable error. In a subsequent interview
Gilanlioglu quotes this learner as follows:

I have tried not to use any words that are difficult for me because I can’t be sure if
I use them correctly or not. . .so I used. . .I tried to use simple words and all the
words that I definitely know their meaning and how can I use them in sentences
 R. Batstone / System 30 (2002) 1–14 11

It is a moot point whether this learners’ performance might be noted simply as an

example of a positive effect for the planning task in terms of increased accuracy,
thereby entirely disregarding a potentially significant insight—that the presence of
an interlocutor (notwithstanding one disbarred from interrupting the learner) was
enough to trigger a level of self-consciousness sufficient to determine this learner’s
entire strategy of caution and risk-avoidance.
Another aspect to such interpersonal sensitivity is the learners’ wariness about
intruding too liberally onto the territory of the interlocutor (Widdowson, 1984, p.
85). Attempting to be more linguistically complex might well involve saying more
and holding the floor for longer than is normally deemed communicatively neces-
sary, thereby incurring threats to face. Ortega (1999) investigated the effects of giv-
ing her learners 10 minutes to plan a story re-telling task told to a silent interlocutor
(just as in Gilanlioglu’s study). The interlocutor’s task was to sequence a series of
jumbled pictures in accordance with the learner’s narrative. Ortega argues that the
presence of the interlocutor served to ‘‘build in a communicative need’’ (1999, p.
123). Yet this very communicative orientation might also have an inhibiting effect,
by providing a level of interpersonal awareness which competes unprofitably with
the learners’ potential orientation to learning through greater risk taking. In some
cases it appears that this is what happened, and Ortega notes how some of her
learners remained inhibited through being excessively alert to the needs of the
interlocutor. She quotes one learner describing her experience in the following way:
‘‘when I hear a mistake I know it’s a mistake, but I couldn’t stop and fix it, because
then that messes up the person I’m telling. . .’’ (1999, p. 128).

4.3. Being ‘between contexts’

Arguably these are learners ‘between contexts’. They are in a learning context to
the extent that they are trying to use task repetition and pre-task planning as con-
structs to help them stretch their interlanguage resources. But at the same time there
are certain default behaviours associated with communicative/social contexts which
have to do with face and deference to the other which run very deep and can be
difficult to let go of, even when one is engaged in more straightforwardly commu-
nicative activity (Dascal, 1977).
How are such face sensitivities to be contextually supported? One suspects that
such risk-taking behaviour might take time to develop, and that it might be
achieved most effectively through teachers working, discussing, explaining and
negotiating with learners about why they do such tasks, helping to create a sense of
a ‘learning culture’ (Riley, 1997; Clemente, 1999) in which learners become pro-
gressively more able and more prepared to create a different set of contextual roles
in the classroom. Such roles require a detachment from communicative norms in
relation to language, in terms of engaging with complex or redundant forms as
form rather than merely as a pragmatically expedient resource. Interpersonally,
learners need to support each other’s explorations into language from this more
detached perspective; recognising that accessibility can itself be the main objective
in classroom discourse.
12 R. Batstone / System 30 (2002) 1–14

5. Concluding remarks

Undoubtedly the learning context has social and psychological dynamics which
need further investigating (utilising the kind of learner protocols employed to such
useful effect by Irving, Gilanloiglu and Ortega) so as to provide a stronger empirical
base for some of the admittedly speculative arguments made here. Nevertheless it is
surely arguable that such research is warranted, and is in tune with the many calls
from contemporary researchers to examine the social contexts of language learning
as thoroughly as its cognitive aspect has already been probed.

References

Allwright, R.L., 1997. Contextual factors in classroom language learning: an overview. In: Malmkjaer,
K., Williams, J. (Eds.), Context in Language Learning and Language Understanding. Cambridge Uni-
versity Press, Cambridge, pp. 115–134.
Aston, G., 1986. Troubleshooting in interaction with learners: the more the merrier? Applied Linguistics
7, 128–143.
Breen, M.P., 1985. The social context for language learning—a neglected situation? Studies in Second
Language Acquisition 7, 135–158.
Brown, G., Yule, G., 1983. Discourse Analysis. Cambridge University Press, Cambridge.
Bygate, M., 1996. Effects of task repetition: appraising the developing language of learners. In: Willis, D.,
Willis, J. (Eds.), Challenge and Change in Language Teaching. Heinemann, London, pp. 136–183.
Cadierno, T., 1995. Formal instruction from a processing perspective: an investigation into the Spanish
past tense. Modern Language Journal 79, 179–193.
Cadierno, T., Glass, W.R., Lee, J.F., Van Patten, B., 1991. Processing Tense in Second Language Input:
Lexical Cues Versus Grammatical Cues. University of Illinois, Urbana-Champagne.
Clemente, A., 1999. SLA research on self-directed learning: theoretical and practical issues. PhD thesis,
Institute of Education, University of London.
Cook, G., 1989. Discourse. Oxford University Press, Oxford.
Crookes, G., 1989. Planning and interlanguage variation. Studies in Second Language Acquisition 11,
367–383.
Crookes, G., Gass, S., 1993. Tasks and Language Learning: Integrating Theory and Practice. Multilingual
Matters. Multilingual Matters, Clevedon, UK.
Dascal, M., 1977. Conversational relevance. Journal of Pragmatics 4, 309–328.
Doff, A., Jones, C., Mitchell, K., 1983. Meanings into Words: Intermediate (Student’s Book). Cambridge
University Press, Cambridge.
Donato, R., 1994. Collective scaffolding in second language learning. In: Lantolf, Appel (Ed.), Vygotskian
Approaches to Second Language Research. Ablex Publishing Corporation, Norwood, NJ, pp. 33–56.
Doughty, C., Williams, J., 1998. Focus on Form in Classroom Second Language Acquisition. Cambridge
University Press, Cambridge.
Ellis, R., 1987. Interlanguage variability in narrative discourses: style shifting in the use of the past tense.
Studies in Second Language Acquisition 9 (1), 1–19.
Ellis, R., 1994. The Study of Second Language Acquisition. Oxford University Press, Oxford.
Ellis, R., 1995. Interpretation tasks for grammar teaching. TESOL Quarterly 29 (1), 87–103.
Faerch, C., Kasper, G., 1986. The role of comprehension in second language learning. Applied Linguistics
7 (3), 257–274.
Firth, J.R., 1957. Papers in Linguistics, 1934–51. Oxford University Press, Oxford.
Firth, A., Wagner, J., 1997. On discourse, communication, and (some) fundamental concepts in SLA
research. Modern Language Journal 81, 286–300.
Foster, P., 1998. A classroom perspective on the negotiation of meaning. Applied Linguistics 19 (1), 1–23.
 R. Batstone / System 30 (2002) 1–14 13

Foster, P., Skehan, P., 1996. The influence of planning and task type on second language performance.
Studies in Second Language Acquisition 18, 299–324.
Gilanlioglu, I., 2001. The effects of task features on lexis and grammar in L2 oral performance. PhD thesis
(draft), Institute of Education, University of London.
Goffman, E., 1967. Interaction Ritual: Essays on Face to Face Behaviour. Garden City, New York.
Halliday, M.A.K., 1973. Explorations in the Function of Language. Edward Arnold, London.
Hulstijn, J.H., Hulstijn, W., 1984. Grammatical errors as a function of processing constraints and explicit
knowledge. Language Learning 34, 23–43.
Hymes, D., 1962. The ethnograpy of speaking. In: Fishman, J.A. (Ed.), Readings in the Sociology of
Language, 1968, pp. 13–53. Mouton, The Hague.
Irving, R., 1999. An investigation into social and psychological effects in oral interactive tasks. MA thesis,
University of London, Institute of Education.
Jakobson, R., 1960. Concluding statement: linguistics and poetics. In: Sebiok, T. (Ed.), Style in Language,
pp. 353–357. MIT Press, Cambridge, Massachusetts.
Kasper, G., Blum-Kulka, S., 1993. Interlanguage Pragmatics. Oxford University Press, Oxford.
Krashen, S.D., 1982. Principles and Practice in Second Language Acquisition. Pergamon, Oxford.
Lantolf, J.P. (Ed.), 2000. Sociocultural Theory and Second Language Learning. Oxford University Press,
Oxford.
Lewis, D., 1972. General semantics. In: Davidson, D., Harmon, G.H. (Eds.), Semantics of Natural Lan-
guage. Harmon, Dordrecht.
Musumeci, D., 1989. The ability of second language learners to assign tense at the sentence level: a cross-
linguistic study. Unpublished doctoral thesis, The University of Illinois, Urbana-Champagne.
Ortega, L., 1999. Planning and focus on form in L2 performance. Studies in Second Language Acquisition
21, 109–148.
Pierce, B.N., 1995. Social identity, investment and language learning. TESOL Quarterly 29, 9–31.
Riley, P., 1997. The guru and the conjurer: aspects of counselling for self-access. In: Benson, R., Voller, P.
(Eds.), Autonomy and Independence in Language Learning, pp. 114–130. Longman, England.
Samuda, V., 2001. Guiding relationships between form and meaning during task performance: the role of
the teacher. In: Bygate, M., Skehan, P., Swain, M. (Eds.), Researching Pedagogic Tasks: Second Lan-
guage Learning, Teaching and Assessment, pp. 119–140. Addison-Wesley Longman, London.
Sato, C., 1986. Conversation and interlanguage development: re-thinking the connection. In: Day, R.
(Ed.), Talking to Learn: Conversations in Second Language Acquisition. Newbury House, Rowley,
MA, pp. 23–45.
Schmidt, R.W., 1992. Psychological mechanisms underlying second language fluency. Studies in Second
Language Acquisition 14 (4), 357–385.
Schmidt, R.W., 1990. The role of consciousness in second language learning. Applied Linguistics 11 (2),
129–158.
Schmidt, R., Frota, S., 1986. Developing basic conversational ability in a second language: a case study of
an adult learner. In: Day, R. (Ed.), Talking to Learn: Conversations in Second Language Acquisition.
Newbury House, Rowley, MA, pp. 237–322.
Sharwood Smith, M., 1986. Comprehension versus acquisition: two ways of processing input. Applied
Linguistics 7 (3), 239–256.
Skehan, P., 1998. A Cognitive Approach to Language Learning. Oxford University Press, Oxford.
Swain, M., 1985. Communicative competence: some roles of comprehensible input and comprehensible
output in its development. In: Gass, S., Madden, C. (Eds.), Input in Second Language Acquisition.
Newbury House, Rowley, MA, pp. 235–253.
Tarone, E., 1997. Analysing IL in natural settings: a sociolinguistic perspective on second language
acquisition. Culture and Cognition 30 (1), 137–149.
Van Patten, B., 1990. Attending to form and content in the input: an experiment in consciousness. Studies
in Second Language Acquisition 12, 287–301.
Van Patten, B., 1996. Input Processing and Grammar Instruction in Second Language Acquisition. Ablex
Publishing Corporation, Norwood, New Jersey.
14 R. Batstone / System 30 (2002) 1–14

Van Patten, B., Cadiemo, T., 1993. Explicit instruction and input processing. Studies in Second Language
Acquisition 15, 225–243.
Widdowson, H.G., 1983. Learning Purpose and Language Use. Oxford University Press, Oxford.
Widdowson, H.G., 1984. Explorations in Applied Linguistics 2. Oxford University Press, Oxford.
Widdowson, H.G., 1998. The conditions of contextual meaning. In: Malmkjaer, K., Williams, J. (Eds.),
Context in Language Learning and Language Understanding. Cambridge University Press, pp. 6–23.
Williams, J., 1992. Planning, discourse marking, and the comprehensibility of international teaching
assistants. TESOL Quarterly 26, 693–711.