Annual Review of Applied Linguistics (2000) 20, 182–198. Printed in the USA.

Copyright © 2000 Cambridge University Press 0267-1905/00 $9.50

STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY

Elaine Tarone

INTRODUCTION

One of the most intractable issues in the field of second-language

acquisition (SLA) research has been the attempt to identify the role of social context
in influencing (or not) the process of acquisition of a second language. The central
question has been whether a theory of SLA must account only for the
psycholinguistic processes involved in acquiring an interlanguage (IL), or,
alternatively, whether social and sociolinguistic factors influence those psycho-
linguistic processes to such an extent that they too must be included in such a
theory. It seems very clear that SLA is a psycholinguistic process. But to what
extent are those psycholinguistic processes affected by social context? In 1985,
Selinker and Douglas proposed a construct of ‘discourse domains’ to show how
social and psycholinguistic processes might be included in a theory of inter-
language; Young (1999) reviews that proposal and a recent attempt to test it,
concluding that the results are still uncertain. After 15 years, this is still a lively
issue in the field of SLA. Indeed, it is becoming a source of increasing conflict
both within the field of SLA and within such areas of applied linguistics as
second/foreign language teaching and second/foreign language teacher training. In
this article, I will briefly summarize the problem, and review and summarize the
current evidence being brought to bear upon this issue in the SLA research
literature.

CRITICISM OF SLA RESEARCH FOR IGNORING SOCIAL CONTEXT

Criticisms have been advanced both by applied linguists working in other

areas and by SLA researchers themselves. Applied linguists working in other fields
have become increasingly vocal in their criticisms of the SLA research endeavor.
One of their common themes is the charge that too much SLA research focuses on
psycholinguistic processes in the abstract and does not consider the social context of
L2 learning. It is said, for example, that IL data are typically gathered from L2

182
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 183

learners in artificial settings, removed from the social contexts in which they
normally use and acquire the L2. As a consequence, it is argued that the results of
such SLA research are irrelevant to the concerns of applied linguists who must deal
with L2 learners in social context, not in the lab. For example, Firth and Wagner
(1997) state that

SLA research takes a view of the learner that is too individualistic

and mechanistic, and...fails to account in a satisfactory way for
interactional and sociolinguistic dimensions of language. As such,
it is flawed, and obviates insight into the nature of language, most
centrally the language use of second or foreign language (S/FL)
speakers (285).

The Firth and Wagner (1997) paper seems to have touched a sensitive spot
with SLA researchers, generating several rejoinders in the same and subsequent
issues of Modern Language Journal (Gass 1998, Kasper 1997, Long 1997; 1998,
Poulisse 1997) and a reply from Firth and Wagner (1998). This article will not
summarize that debate, but merely point out that it is not confined to the pages of
the Modern Language Journal. Other applied linguists working in areas related to
SLA have made a similar point, most notably Cook (in press) and Rampton (1995):

The very undifferentiated portrait of the second language learner

that emerges in SLA no doubt partly results from its tendency to
thematise the learner’s internal psychological condition. Rather
than looking at interaction as a socio-historically sensitive arena in
which language learner identity is socially negotiated, SLA
generally examines learner behaviour for evidence of the
determining influence of psycholinguistic states and processes
(Rampton 1995:293).

Rampton argues that one problem with this approach is that SLA researchers who
have ignored social context in their studies cannot then claim that their results are
generalizable to all social situations. Crucially wrong assumptions may be made in
such studies about the social conditions which normally hold for L2 learners in
society, and wrong generalizations may be made:

In the ‘root image’ underlying SLA research, it is assumed that L2

learning generates situational anxiety, that progression along the
route towards target language proficiency is (or should be) the
learner’s abiding preoccupation, and that learner status is
fundamentally stigmatised. All of these characterisations may be
true of some, or indeed many, situations, but they are certainly not
invariable... (Rampton 1995:292).
184 ELAINE TARONE

And, indeed, Rampton (1995) shows in graphic and very convincing terms that
these very characterizations are simply wrong for the adolescent L2 learners in his
ethnographic study in Great Britain.

The criticism of SLA research for failure to include social context has
become so pronounced that, at present, some influential second/foreign language
teacher trainers are even taking the position that L2 teachers do not need to know
the results of current SLA research. So, for example, when Freeman and Johnson
(1998) “reconceptualize” the knowledge base which second/foreign language
teachers must have, they do not include any knowledge of language learners and
language acquisition/learning. They explain the omission this way:

In general, due perhaps to its roots in L1 acquisition and cognitive

psychology, the field of SLA has viewed language learning from
an individualist perspective. Thus, until recently, the field has not
examined language learning from the standpoint of socially
negotiated, constructivist processes that may be at play.... From
our point of view...a social constructivist view of language
learning would seem to interface more directly with the nature of
classroom language learning.... Because the research knowledge
per se does not articulate easily and cogently into classroom
practice, much current knowledge in SLA may be of limited use
and applicability to practicing teachers (Freeman and Johnson
1998:411).

This position is taken by two leaders in language teacher education in the theme-
setting initial paper in a collection in TESOL Quarterly in December of 1998. It
suggests that schools of education will not be major consumers of SLA research in
the coming century unless SLA researchers can either convert teachers to a more
psycholinguistic mode of thinking or show that they do study the impact of social
context on processes of SLA.1

If we examine the criticisms of the applied linguists outlined above in light

of actual publications in SLA, those criticisms seem unwarranted. In fact, as
already seen in the case of Selinker and Douglas (1985), there are some SLA
researchers who have for some time taken the position that SLA theory and
research should explore the relationship between social context and psycholin-
guistic processes of SLA. More recently, Preston (1996) proposes an alternative
psycholinguistic model of SLA in which social factors can constrain grammatical
marking, both in competence and performance. Larsen-Freeman’s (1997) rather
different proposal that interlanguage be studied as a complex dynamic system, using
correlational research designs in natural settings, provides a very promising
framework for SLA research relating social and cognitive factors in the L2 learner.
But such models are merely proposals; where is the evidence?
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 185

The second half of this paper will review a small but growing subset of
SLA research work which demonstrates an impact of social factors on psycho-
linguistic processes of L2 acquisition. It would be premature for other applied
linguists to dismiss the entire field of SLA research out of hand without considering
the evidence produced in that field—some of which was laid out in Young (1999)
and in Bayley and Preston (1996)—arguing that social factors are related to
systematic variation in interlanguage and to SLA itself.

But why, if this work is being done, is it relatively unmentioned by

mainstream SLA researchers with a psycholinguistic orientation? As the next
section will briefly show, these two streams of SLA research have not, as yet,
affected one another. The problem is that there is minimal overlap between them.

THE SPLIT IN SLA: RESEARCH ON COGNITIVE FACTORS AND

RESEARCH ON SOCIAL FACTORS

Research on second language acquisition that takes a sociolinguistic or even

co-constructionist orientation typically focuses on L2 use in natural social contexts
at a single point in time. Note that the emphasis is on interlanguage USE in such
contexts, but not typically on the acquisition of new elements of the interlanguage.
Further, such studies typically do NOT focus on discrete grammatical,
morphological, or phonological elements in the interlanguage. Rather, they
typically focus on general pragmatic or discourse patterns. For example, Peirce
(1995) shows that Martina, while usually hesitant in speaking English with
Canadian interlocutors, became much more fluent in her use of English L2 when
she perceived a threat to her home and had to talk to her landlord. Thus, the
interlocutor and Martina’s roles (when speaking as low status immigrant or as
mother defending her family) were shown to influence Martina’s IL USE and
fluency. However, the study provided no evidence of her acquisition of any
features of IL due to the forces of either interlocutor or social role. Similarly, Platt
and Troudi (1997) trace an ESL child’s failure to learn to read in her L2 to such
social factors as the teacher’s instructional style (following from her beliefs about
the way L2s are acquired) and fellow students’ willingness to help the learner get
by without reading for herself. Platt and Troudi show that such social factors
influence an L2 learner’s discourse patterns and overall success in reading, but they
do not show how these social factors influence the development or fossilization of
specific features in the learner’s interlanguage. And so it goes. All the studies
cited in Tarone (1988) and most of those in Bayley and Preston (1996) can be
argued to suffer from the same general deficiency: They demonstrate the impact of
social factors on interlanguage USE at a single point in time and do not show that
those social factors affect the ACQUISITION of SPECIFIC linguistic features of IL
over time.

To be fair, since sociolinguistic and co-constructionist studies take place in

very complex natural settings and focus on the language input and output of just a
few individuals in those settings, it would be very difficult and time-consuming to
186 ELAINE TARONE

establish that the acquisition of new L2 forms has been influenced by identifiable
social factors. Such an enterprise would involve longitudinal studies of those
learners and the transcription of long stretches of conversation in which it would
be very difficult to identify the new forms being acquired. Thus, while SLA
researchers who take a sociolinguistic or co-constructionist orientation have a good
deal of evidence showing that L2 learners’ IL USE is variably affected by identi-
fiable features of social context, they have usually not tried to show that those
social features change the process of L2 ACQUISITION—specifically, the
acquisition of an IL linguistic system—in any clear way. They have assumed it,
and asserted it, but not often accumulated the evidence to prove it.

Research on second-language acquisition that takes a psycholinguistic, or

cognitive orientation, whether it operates within a UG framework or not (cf.
Long’s [1998] assertion that most SLA cognitive research does not belong in a UG
tradition), is clearly focused on the goal of explaining how an interlanguage
grammar gets acquired over time. As Kasper (1997) says, in the acronym SLA,
“the ‘A’ is for acquisition,” not use. Tightly controlled studies are designed to
identify particular grammatical features to be acquired and explore the impact of
various cognitive factors on the acquisition process. Such research, exemplified in
articles by deBot (this volume) and Scovel (this volume), sets out to establish some
clear causes for the acquisition (or fossilization) of very specific phonological and
grammatical features of the L2. And since, in such studies, the social context is
greatly controlled or reduced in complexity, and is usually fairly similar across
university studies, such researchers have assumed that social factors are irrelevant
for their work.

Thus, two strands of SLA research exist in parallel, but rarely focus on the
same data or the same questions. The issue of the relationship between social
factors and psycholinguistic processes of acquisition is, as Eckman (1994) points
out, an empirical question, best resolved by the presentation of the right sort of data
and not by argument alone. But neither strand of SLA research has consis-tently
and systematically set out to gather the sort of data which might show whether
social factors affect cognitive processes of acquisition in specific ways and thereby
enable both strands to see how their work is related.

Long (1998) states the position of cognitively-oriented SLA researchers

succinctly, and makes two assertions which can be empirically tested:

Remove a learner from the social setting, and the L2 grammar

does not change or disappear. Change the social setting
altogether, e.g., from street to classroom, or from a foreign to a
second language environment, and, as far as we know, the way the
learner acquires does not change much either, as suggested, e.g.,
by comparisons of error types, developmental sequences,
processing constraints, and other aspects of the acquisition process
in and out of classrooms... (Long 1998:93).
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 187

These two assertions seem to provide useful ways to structure the

remainder of the present article. There is evidence in the recent SLA research
literature which may address both of the above assertions.

RESEARCH EVIDENCE AT THE INTERFACE OF SOCIAL CONTEXT AND

INTERLANGUAGE GRAMMAR

1. Remove the L2 learner from the social setting: Does the IL grammar change?

Let us assume that we are not talking about instantaneous changes—instan-

taneous acquisition of entirely new rules, or a switch to a completely different
learner grammar—as the learner moves from the classroom to the hallway.
Positing such switches from one grammar to another would certainly violate
Occam’s Razor. But we do know from research on English for Specific Purposes
(Bhatia 1993, Swales 1990) that different social situations often require different
registers of English, some of them with rules for grammatical usage that differ
markedly from those for general English. For example, the rules for tense usage in
academic research papers are distinctly different from those for general English.
(See Swales 1990 for a review.) Active and passive aspect are patterned quite
differently in astrophysics papers than in publications for researchers in other fields
(Tarone, et al., 1981 [1985; 1998]). We could assume that the IL grammar stays
the same but is in fact variable, sensitive to social setting at any given point in time
(a position consistent with studies in Bayley and Preston 1996, Preston 1996,
Tarone 1988, and others). In this view, the entire IL grammar at any given time
consists of a range of styles. As the learner moves from one social setting to
another, the IL grammar does not change wholesale—the learner just accesses one
or another style of a grammar that stays the same.

Perhaps a more empirically interesting way to paraphrase the question

above is as follows: If two L2 learners acquire English in two different social
settings, will those learners internalize two different IL grammars? The answer to
this question, based on the research evidence, is surely yes. As just seen, ESP
research shows that the TL rules to be learned vary in specific ways from one
social situation to another. The classroom may expose L2 learners to only one
rather formal register (Cohen 1997, Cohen and Tarone 1997, Tarone and Swain
1995). Tarone and Swain (1995) show that the immersion classroom provides
input only in an academic register of L2, while an L2 adolescent vernacular register
is only available in other social settings. Thus, it is well-established that, for any
given target language, the L2 learner receives different input on the grammatical
and lexical features to be acquired in different social situations.

But there are also other ways in which social context can influence the TL
input given to learners. The degree to which native speakers adjust their language
for learners may differ in different social contexts, and the amount of overall
modeling and collaborative assistance given to learners may also differ in different
social situations. For example, social context affects the degree to which
188 ELAINE TARONE

interlocutors make linguistic and conversational adjustments for learners. It was

initially claimed (Long 1980; 1983) in fairly general terms that all native speakers
adjust the input they give to L2 learners. Based on his study of 16 elementary level
non-native speakers of English performing six tasks with native speakers they had
never met before in a lab at UCLA, Long identified a set of linguistic and
conversational adjustments in the input given to the learners. Long’s study and his
review of the literature suggested to him that foreigner talk could be found in
virtually all groups of people addressing speech to learners:

(1) in children as well as adults, (2) in upper-middle, middle and

working-class adults, (3) individuals with or without prior FT
experience, and 4) second language teachers, content-teachers and
non-teachers.
...linguistic and conversational adjustments...appear to be
immune to differences among groups/types of speakers (Long
1983:184).

More recently, however, Gass (1997) draws a different conclusion in

reviewing the research on simplification of input to L2 learners. She cites several
studies which show that some individuals in some social situations do not modify
their speech, even for people experiencing obvious language difficulty. She notes
two studies which document such “counter-accommodating behavior”: Arthur, et
al. (1980) and Varonis and Gass (1985). Data from the latter study are discussed
in Gass (1997:64–66), particularly a long telephone conversation between a native
speaker at a TV repair shop and an L2 learner who believes he is calling a TV sales
shop. In that conversation, which is fraught with misunderstandings, the native
speaker does not make linguistic accommodations for the learner: She speaks
rapidly, uses contractions and idiomatic language, and shows none of the
conversational adjustments which are supposedly constant across social situation
and interlocutor. Bondevik (1996) discusses similar findings in a more controlled
study in a Minnesota electronics store. Each of four salesmen from Minnesota
failed on three different occasions to accommodate to different L2 learner listeners
by making linguistic and conversational adjustments; one even complexified his
syntax after learners indicated noncomprehension. In subsequent interviews with
the salesmen, one said that he felt it would have been insulting to simplify his
speech in that context. Yet the listeners in Long (1980; 1983) did simplify their
speech. Thus, the research evidence indicates that interlocutors in differing social
situations provide L2 learners with differing amounts of adjusted TL input; some
do not accommodate to learners by simplifying linguistic input at all. If simplified
input is important to the SLA process, then we must conclude that the process of
SLA is affected differently in these different social contexts.

But the input provided to learners in different social contexts differs in one
more way: The overall modeling and collaborative assistance given to learners may
also vary from one context to another. In conversation in certain formal
institutional contexts, learners may not have direct access to the L2 input they need.
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 189

Bardovi-Harlig and Hartford (1996) studied the sort of input which is available to
L2 learners in the institutional setting of the university academic advising session.
They described this session as “an unequal status encounter that by nature is a
private speech event and cannot be observed by other learners.” Thus, the L2
learner entered this encounter with no previous input or preparation on the sort of
language that would be needed. Although the advisors attempted to guide the
students in choosing courses for the coming term by teaching them that student
suggestions are expected, and that certain types of content are appropriate for those
suggestions, they did not teach the students appropriate linguistic forms for the
invited speech acts, and they provided corrective feedback on meaning, not form.
When the higher-status, more powerful advisors tried to make suggestions for the
students, those suggestions could not serve as direct models of appropriate language
use for the students. Examples were “I’m going to have you take...,” “we’ll ask
you to take...,” and “I would suggest....” Additionally, learners whose mastery of
tense-mood-aspect morphology did not enable them to mitigate as required would
have had especial difficulty with this type of situation. In a similar study of
interactions in institutional settings, Tarone and Kuehn (in press) taped L2 learners
going through intake interviews in a welfare office. They showed that this too, as a
private unequal status encounter, provided little or no useful input for L2 learners
to use in order to know what applicants should say, or how they should say it. As
a consequence, L2 learners had demonstrable difficulty understanding and
responding to directives and suggestions being made by the welfare worker. Their
attempts to communicate in this situation were not well supported.

In contrast with the restricted available input in these two institutional

situations, there was very collaborative input and co-construction provided to L2
learners in social situations documented in two other recent studies. Parks and
Maguire (1999) reported on the way in which francophone nurses, newly hired in
an English-medium hospital in Montreal, acquired skill in writing nursing notes in
English L2. In a highly collaborative process, the francophone learners observed
more experienced nurses as they interacted with patients about to be discharged,
interacted themselves under observation, solicited input from supervisors and co-
workers, got and used advice given, and revised and produced notes in a highly
collaborative process. Parks and Maguire concluded:

...when nurses had questions about language they did not have to
rely solely on their internalized language resources but were able
to have access to resources in the form of relevant documentation
or more knowledgeable others, who confirmed or disconfirmed
hunches or supplied genre-specific language. Indeed, all acts of
language use, whether oral or written, are inherently social and
context specific (1999:166).

The data provided in Parks and Maguire (1999) actually could allow us to move
beyond language USE, since it shows how, over time, input provided by others
gradually becomes incorporated into each new revision of nursing notes. Though
190 ELAINE TARONE

the authors did not point this out, the data in the paper even show changes in a
grammatical construction produced by the learner, from an initial erroneous “was
been given” to a correct “was given.” Co-construction is a central feature of
another study: Swain and Lapkin (1998) recorded and analyzed the way in which
two French immersion students worked together in a dyad to co-construct a story.
In this study, the authors focused on ACQUISITION of L2 linguistic items. The
learners worked together to help each other identify lexical and grammatical forms
which they later used in their report; forms were proposed and then revised to
become more target-like over the course of the learners’ interactions. In this study,
collaboration in this social context permitted the learners to focus on form and
internalize specific new features of an IL grammar.

In summary, we have seen that L2 learners in different social contexts

receive input which contains different grammar rules (cf. the ESP research), or
which contains more or fewer linguistic/conversational adjustments, or which
provides more or less explicit modeling and co-construction to support the
acquisition process. Thus, because the TL input provided in different social
situations is different, the IL grammars which can be acquired in those contexts
must also be different.

2. Change the social setting altogether: Will the way the learner acquires L2
change much?

This second question focuses more specifically on certain types of change

in acquisition processes that might arise due to different social settings. These
processing changes might include error types, developmental sequences, processing
constraints, and other aspects of the acquisition process. There is, in fact, research
evidence that error detection, developmental sequences, and negotiation of meaning
may all be sensitive to social context.

Kormos (1999), in an extensive review of processes of monitoring and self-

repair in the use of an interlanguage, shows that while Levelt’s perceptual loop
theory of monitoring can account for monitoring in IL speech, it needs to be
supplemented with recent research on consciousness, attention, and noticing if it is
to account for mechanisms of error detection in using an IL. The studies she
reviews suggest that error detection in L2 depends on several factors, some of them
psycholinguistic, like the availability of attention. However, error detection also
depends on social context factors such as the “accuracy demand of the situation”
and “various listener-based discourse constraints” (1999:324). Kormos concludes
that diverse results on the rate of well-formed repairs in studies with different
research designs (cf. Levelt 1983, van Hest 1996) suggest that the well-formedness
of corrections is both speaker-based and listener-based. “...[S]peakers will strive
to produce well-formed repairs not only because their original speech plan needs to
be encoded again, but also because they want to aid their inter-locutors”
(1999:330). We must assume that if the learner is correcting errors differentially in
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 191

different social contexts to accommodate to different interlocutors, then the errors

which remain uncorrected in those social contexts will be different.

Developmental sequences of SLA can also be influenced by factors of

social context. Here we have only a small amount of intriguing research evidence,
and a puzzling lack of attempt to obtain any more such evidence. Liu (1991), and
Tarone and Liu (1995; see discussion in Young 1999) show that a 6-year-old
Chinese boy acquiring English L2 in Australia over a two-year period acquired
Pienemann and Johnston’s (1987) six stages of interrogatives in a different order
from that predicted. Stage 4 and 5 questions were acquired before Stage 3
questions. Each new stage of interrogative formation except for one was produced
in conversation in a single social context: in play sessions with the researcher at the
learner’s home. Each new stage of question then gradually spread to other social
contexts, appearing next in conversations with other peers in desk work at school,
and last in conversations with the teacher in school. Liu (1991) argues that there
was something about the social context of play sessions at home with the researcher
which caused later-stage questions to emerge before earlier-stage questions in this
learner’s interlanguage. Aspects of input (Young 1999), and aspects of play (cf.
Tarone 1999) have both been cited as possible characteristics of this social context
that altered the sequence of acquisition of questions for this learner. To date, there
have been no other reports of research studies which examine the impact of
different social contexts on the sequence of acquisition of any phonological,
morphological, or grammatical feature of interlanguage. This absence of studies
designed to obtain relevant data on possible interactions between social context and
sequence of acquisition is hard to account for. Certainly it represents a gap in the
literature that needs to be addressed.

In contrast to the above, there have been many studies focusing on

interaction and negotiation as a central process in second-language acquisition.
Negotiation is important because it leads to the psycholinguistic event of noticing
that a linguistic form needs to be acquired:

The input-interaction view must take the position that noticing is

crucial. In negotiation the learner is focusing on linguistic form,
and that focus, or specific attention paid to linguistic form, is the
first step toward grammar change (Gass 1997:101).

Several recent studies of negotiation show that the negotiation process can
be highly sensitive to social context. Gass (1997), in her substantive review of
input, interaction, and second language learner, states:

It would be too simplistic to assume that these integral parts of negotiation

sequences occur without influence from the context in which they appear.
To the contrary, many factors affect the structure of conversation (Gass
1997:117).
192 ELAINE TARONE

She then reviews studies which show the following features of social context to be
important factors affecting negotiation:

1. Task type (where a task requiring the exchange of information results

in more negotiation; a task which is student-student involves more
negotiation than one which is teacher-fronted),
2. Background knowledge and status differences (where great status
differences inhibit negotiation, and gaps in background knowledge
provide more opportunities for negotiation),
3. Familiarity (where conversation partners who know each other
negotiate more), and
4. Gender (where same gender partners negotiate more).

Disturbingly, Foster (1998) has found that the social context of the ESL
classroom may be one which promotes little negotiation of meaning. Foster
observed 21 intermediate level ESL students as they worked in dyads vs. small
groups, doing tasks with optional or required information exchange. She found that
many learners in small groups did not speak at all, and many more in dyads and
groups did no negotiated interaction. Very few modified their utterances on the
basis of interaction. Lyster (1998) found that French immersion teachers also did
not respond to learner errors in ways that allowed for much negotiation to occur.
In particular, teachers did not promote negotiation that drew learners’ attention to
errors of form. Thus, in different social contexts, the degree of negotiation of
meaning (or, in theory, of focus on form) may vary tremendously.

But to what extent do these differences in negotiation lead to differences in

noticing, or in a focus on linguistic form? We are beginning to see research on the
impact of negotiation on noticing of particular linguistic features. Mackey, et al.
(in press) found that L2 learners who are given implicit negative feedback while in
the process of negotiating for meaning do not always notice that feedback.
Feedback on phonology and lexis is noticed readily, but feedback on morpho-syntax
is not, nor is it incorporated into subsequent learner production. How does social
context affect this process of noticing? Does one notice more linguistic form when
one’s interlocutor is a teacher, as opposed to when the interlocutor is a student?
Research on style shifting suggests that learners certainly adjust their speech
towards the TL norm more in the presence of a teacher (e.g., Blanco-Iglesias,
Broner and Tarone 1995, Liu 1991), but studies on differential noticing in response
to such features of social context as interlocutor do not seem to exist.

As Gass (1997) points out, negotiation for meaning is only one possible
means of getting L2 learners to focus on form. Another means proposed more
recently by Cook (1997; in press), Tarone (1999), and Tarone and Broner (1999) is
language play, which occurs in some social contexts more than others. Play with
L2 forms at all linguistic levels may be another contextually-influenced way of
drawing L2 learners’ attention to language forms that need to be acquired. Work
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 193

on language play, the social contexts in which it occurs, and its impact on second-
language acquisition is just beginning.

To conclude this section, there does appear to be evidence in the literature

to suggest that if we change the social setting altogether, the way the learner
acquires [L2] does seem to change, at least with regard to error correction,
developmental sequences, and negotiation of meaning.

CONCLUSION

There are SLA researchers who are exploring the ways in which social
context may affect the acquisition of specific L2 forms. It thus appears premature
for other applied linguists, or for SLA researchers themselves, to assert that the
field has no interest in examining SLA in its social context. But there is also a
good deal of research study to be done, particularly in documenting the impact of
social factors on the psycholinguistic processes of acquisition of specific inter-
language morphosyntactic, lexical, and phonological forms. And what might an
SLA theory look like that incorporated both social and psycholinguistic factors?
Two theoretical proposals, those of Preston (1996) and Larsen-Freeman (1997),
appear especially promising in their capacity to bring the two strands of research,
the psycholinguistic and the sociolinguistic, together in a single framework. Work
in this area appears important, and quite promising.

NOTES

1. For one response to Freeman and Johnson (1998), see Yates and Muchisky
(1999).

ANNOTATED BIBLIOGRAPHY

Bondevik, S.-G. 1996. Foreigner talk revisited: When does it really occur and
why? Tromso, Norway: University of Tromso. MA thesis.

Exploring the factors which trigger foreigner talk, the researcher conducted
a study in an American electronics store. Over a period of several weeks,
three L2 learners and a native speaker customer entered the store at
different times, and each one talked to each of five native-speaker
salesmen. All of the “customers” indicated serious noncomprehension at
similar points in the conversation, and the salesmen’s subsequent speech
was analyzed to see if they had made linguistic or conversational
194 ELAINE TARONE

adjustments to help their customers understand. Results showed that four

of the five Minnesota salesmen did not use foreigner talk with the learners,
not even with one who looked foreign and had a pronounced accent, while
the one salesman who had lived in California did. One Minnesotan even
complexified his speech after noncomprehension had been indicated. In
interviews, he said he felt it would have been insulting to simplify his
speech to his customers.

Larsen-Freeman, D. 1997. Chaos/complexity science and second language

acquisition. Applied Linguistics. 18.141–165.

The author outlines a number of ways in which interlanguages exhibit the

features of complex nonlinear systems: IL is characterized by dynamic
processes; it is complex, with many interacting factors that determine its
developing trajectory; it does not develop in a linear fashion; and it is an
open, self-organizing system with continued input. The author uses this
perspective to examine five issues in SLA: mechanisms of acquisition,
definitions of learning, the instability and stability of interlanguage,
differential success, and the effect of instruction. Adoption of this
perspective would discourage theory construction through the aggregation
of simple univariate cause-effect links, underscore the importance of
details, and encourage a focus on the whole process of SLA rather than just
a part.

Long, M. H. 1998. SLA: Breaking the siege. University of Hawai’i Working

Papers in ESL. 17.79–129.

The author responds to three published criticisms of SLA research—that

too many SLA researchers (a) focus overly narrowly on learner-internal,
cognitive processes; (b) use an outdated “realist” perspective, ignoring the
insights of post-modernist scholars; and (c) claim implications for second-
language teaching from their research when they do not exist. The author
takes on one paper claimed to represent each critical position (papers
written by Firth & Wagner, Lantolf, and Nunan respectively). He
concludes, rather contentiously, that none of these positions has any
validity at all. While this reviewer is sympathetic with some of Long’s
positions, his rhetoric in this paper appears too extreme. Unfortunately,
Long concludes by suggesting that research journals in the field which
publish such attacks on SLA research are thereby not maintaining high
enough standards of scholarship.
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 195

Preston, D. 1996. Variationist perspectives on second language acquisition. In R.

Bayley and D. Preston (eds.) Second language acquisition and linguistic
variation. Amsterdam: J. Benjamins. 1–46.

The author states that sociolinguistics as a discipline has much to contribute

to the field of SLA, but that several factors have prevented a healthy
relationship between the two fields: 1) variationists have not advanced
plausible psycholinguistic models of SLA; 2) SLA researchers do not
understand the aims of sociolinguistics as a field, or its tools and methods;
and 3) many SLA researchers tend to rely overmuch on a UG framework
which is too narrow to account for SLA. A variable psycholinguistic
model is proposed to account for SLA data, one based on the analogy of
weighted coins being flipped, the “coins” being selection devices in
competence and performance which are sensitive to linguistic and stylistic
factors.

Tarone, E. and M. Broner. 1999. Is it fun? Language play in a fifth grade Spanish
immersion classroom. Unpublished manuscript, University of Minnesota
[under review].

The authors discuss two kinds of language play which appear in the
discourse of fifth grade Spanish immersion students: private speech used
for purposes of rehearsal of target forms and ludic speech used for
purposes of self-amusement and creativity. Examples of both kinds of
language play are identified in the natural classroom L2 discourse of three
learners, observed and taped over a period of five months. Five criteria
are used to distinguish effectively the two kinds of play in most cases. The
paper concludes with a brief discussion of the differing possible roles of
each kind of language play in the process of SLA.

UNANNOTATED BIBLIOGRAPHY

Arthur, B., R. Weiner, M. Culver, Y. J. Lee and D. Thomas. 1980. The register
of impersonal discourse to foreigners: Verbal adjustments to foreign
accents. In D. Larsen-Freeman (ed.) Discourse analysis in second
language research. Rowley, MA: Newbury House. 111–124.
Bardovi-Harlig, K. and B. Hartford. 1996. Input in an institutional setting. Studies
in Second Language Acquisition. 18.171–188.
Bayley, R. and D. Preston (eds.) 1996. Second language acquisition and linguistic
variation. Amsterdam: J. Benjamins.
Bhatia, V. K. 1993. Analysing genre: Language use in professional settings.
London: Longman.
196 ELAINE TARONE

Blanco-Iglesias, S., M. Broner and E. Tarone. 1995. Observations of language use

in Spanish immersion classroom interactions. In L. Eubanks, L. Selinker
and M. Sharwood Smith (eds.) Festschrift for William Rutherford.
Amsterdam: J. Benjamins. 239–251.
Cohen, A. 1997. Developing pragmatic ability: Insights from the accelerated study
of Japanese. In H. M. Cook, K. Hijirida and M. Tahara (eds.) New trends
and issues in teaching Japanese language and culture. Honolulu:
University of Hawaii, Second Language Teaching and Curriculum Center.
137–163.
_________ and E. Tarone. 1997. Language learning in an accelerated college
program: A case study of a learner in search of the vernacular. Paper
presented at the Second Language Research Forum (SLRF). East Lansing,
Michigan, October 1997.
Cook, G. 1997. Language play, language learning. ELT Journal. 51.224–231.
________ In press. Language play, language learning. Oxford: Oxford University
Press.
Eckman, F. R. 1994. The competence-performance issue in second-language
acquisition theory: A debate. In E. Tarone, S. Gass and A. Cohen (eds.)
Research methodology in second-language acquisition. Hillsdale, NJ: L.
Erlbaum. 3–15.
Firth, A. and J. Wagner. 1997. On discourse, communication, and (some)
fundamental concepts in SLA research. Modern Language Journal.
81.285–300.
_____________________ 1998. SLA property: No trespassing! Modern Language
Journal. 82.91–94.
Foster, P. 1998. A classroom perspective on the negotiation of meaning. Applied
Linguistics. 19.1–23.
Freeman, D. and K. E. Johnson. 1998. Reconceptualizing the knowledge-base of
language teacher education. TESOL Quarterly. 32.397–417.
Gass, S. 1997. Input, interaction and the second language learner. Mahwah, NJ:
L. Erlbaum.
_______ 1998. Apples and oranges; Or, why apples are not oranges and don’t need
to be. A response to Firth and Wagner. Modern Language Journal.
82.82–90.
Kasper, G. 1997. ‘A’ stands for acquisition: A response to Firth and Wagner.
Modern Language Journal. 81.307–312.
Kormos, J. 1999. Monitoring and self-repair in a second language. Language
Learning. 49.303–342.
Lantolf, J. P. and G. Appel (eds.) 1994. Vygotskyian approaches to second
language research. Norwood, NJ: Ablex.
Levelt, W. J. M. 1983. Monitoring and self repair in speech. Cognition.
33.41–103.
Liu, G.-q. 1991. Interaction and second language acquisition: A case study of a
Chinese child’s acquisition of English as a Second Language. Melbourne,
Australia: La Trobe University. Ph.D. diss.
 STILL WRESTLING WITH ‘CONTEXT’ IN INTERLANGUAGE THEORY 197

Long, M. H. 1980. Input, interaction and second language acquisition. Los

Angeles: University of California at Los Angeles. Ph.D. diss.
___________ 1983. Linguistic and conversational adjustments to non-native
speakers. Studies in Second Language Acquisition. 5.177–193.
___________ 1997. Construct validity in SLA research: A response to Firth and
Wagner. Modern Language Journal. 81.318–323.
Lyster, R. 1998. Recasts, repetition, and ambiguity in L2 classroom discourse.
Studies in Second Language Acquisition. 20.51–81.
Mackey, A., S. Gass and K. McDonough. In press. How do learners perceive
implicit negative feedback? Studies in Second Language Acquisition.
Parks, S. and M. Maguire. 1999. Coping with on-the-job writing in ESL: A
constructivist-semiotic perspective. Language Learning. 49.143–175.
Peck, S. 1978. Child-child discourse in second language acquisition. In E. Hatch
(ed.) Second language acquisition: A book of readings. Rowley, MA:
Newbury House. 383–400.
Peirce, B. N. 1995. Social identity, investment, and language learning. TESOL
Quarterly. 29.9–32.
Pienemann, M. and M. Johnston. 1987. Factors influencing the development of
language proficiency. In D. Nunan (ed.) Applying second language
acquisition research. Adelaide, Australia: National Curriculum Resource
Centre. 45–141.
Platt, E. and S. Troudi. 1997. Mary and her teachers: A Grebo-speaking child’s
place in the mainstream classroom. Modern Language Journal. 81.28–49.
Poulisse, N. 1997. Some words in defense of the psycholinguistic approach.
Modern Language Journal. 81.324–328.
Rampton, B. 1995. Crossing: Language and ethnicity among adolescents. London:
Longman.
Rounds, P. and R. Kanagy. 1998. Acquiring linguistic cues to identify agent:
Evidence from children learning Japanese as a second language. Studies in
Second Language Acquisition. 20.509–542.
Selinker, L. and D. Douglas. 1985. Wrestling with ‘context’ in interlanguage
theory. Applied Linguistics. 6.190–204.
Swain, M. 1995. Three functions of output in second language learning. In G.
Cook and B. Seidlhofer (eds.) Principle & practice in applied linguistics:
Studies in honour of H. G. Widdowson. Oxford: Oxford University Press.
125–144.
________ and S. Lapkin. 1998. Interaction and second language learning: Two
adolescent French immersion students working together. Modern Language
Journal. 82.320–337.
Swales, J. M. 1990. Genre analysis: English in academic and research settings.
Cambridge: Cambridge University Press.
Tarone, E. 1988. Variation in interlanguage. London: Edward Arnold.
_________ 1998. A sociolinguistic perspective on an SLA theory of mind. Studia
Anglica Posnaniensa. 23.431–444.
_________ 1999. Getting serious about language play: Language play,
interlanguage variation and second-language acquisition. Paper presented at
198 ELAINE TARONE

the annual Second Language Research Forum. Minneapolis, MN,

September 1999.
Tarone, E., S. Dwyer, S. Gillette and V. Icke. 1981. The use of the passive in two
astrophysics papers. ESP Journal. 1.123–140. [Reprinted in J. Swales
(ed.) 1985. Episodes in ESP. Oxford: Pergamon. 191–205.] [Updated and
reprinted, 1998. English for Specific Purposes. 17.113–132.]
_________ and K. Kuehn. In press. Negotiating the social services oral intake
interview: Communicative needs of non-native speakers of English. TESOL
Quarterly.
_________ and G.-q. Liu. 1995. Situational context, variation, and second
language acquisition theory. In G. Cook and B. Seidlhofer (eds.) Principle
and practice in applied linguistics: Studies in honour of H. G. Widdowson.
Oxford: Oxford University Press. 107–124.
_________ and M. Swain. 1995. A sociolinguistic perspective on second-language
use in immersion classrooms. Modern Language Journal. 79.166–178.
van Hest, E. 1996. Self-repair in L1 and L2 production. Tilburg, The Netherlands:
Tilburg University Press.
Varonis, E. and S. Gass. 1985. Miscommunication in native/non-native
conversation. Language in Society. 14.327–343.
Yates, R. and D. Muchisky. 1999. On the status of disciplinary knowledge in
language teacher education. Paper presented at the International Language
Teacher Conference, University of Minnesota. Minneapolis, MN, May
1999.
Young, R. 1999. Sociolinguistic approaches to SLA. In W. Grabe, et al. (eds.)
Annual Review of Applied Linguistics, 19. Survey of applied linguistics.
New York: Cambridge University Press. 105–132.