Language Learning 49:4, December 1999, pp.

677–713

Exploring the Interlanguage of Interlanguage

Pragmatics: A Research Agenda for
Acquisitional Pragmatics

Kathleen Bardovi-Harlig
Indiana University

G. Kasper and R. Schmidt (1996) have argued that the

field of investigation known as interlanguage pragmatics
has been essentially modelled on cross-cultural pragmat-
ics. Taking Kasper and Schmidt’s argument one step fur-
ther, this article shows how interlanguage itself has been
ignored in research on interlanguage pragmatics. Re-
search has not established that pragmatic competence is
independent of grammatical competence. Although gram-
matical competence may not be a sufficient condition for
pragmatic development, it may be a necessary condition. I
outline a research agenda in which the study of interlan-
guage becomes more central to the study of interlanguage
pragmatics.

The goal of this article is to set out a research agenda that

makes the interlanguage of interlanguage pragmatics a central
area of investigation. It begins by assessing the state of acquisition
research in interlanguage pragmatics; then it surveys work in

Kathleen Bardovi-Harlig, Program in TESOL and Applied Linguistics.

This paper was originally presented as a plenary at Second Language Research
Forum 1998 at the University of Hawai’i, Manoa. I thank my graduate students
at Indiana University during the fall of 1998 for their comments on this paper. I
am also indebted to Gabi Kasper and the reviewers of Language Learning for their
guidance on an earlier version of this paper. Any errors or omissions are mine.
Correspondence concerning this article may be sent to Kathleen Bardovi-
Harlig, Program in TESOL and Applied Linguistics, Indiana University,
Memorial Hall 313, Bloomington, IN 47405. Internet: bardovi@indiana.edu

677

Copyright © 2000. All rights reserved.

678 Language Learning Vol. 49, No. 4

interlanguage pragmatics that either directly examines or appeals

to grammatical competence. I will also show how acquisition
studies in interlanguage pragmatics will differ from the type of
studies we have conducted so far, and I will outline potential areas
of investigation.
As we know, interlanguage pragmatics deals with the prag-
matics of language learners; but does that in itself make it an
acquisitional endeavor? As early as 1992, Kasper observed that it
did not. She observed, “the majority of [interlanguage pragmatics]
studies focus on use, without much attempt to say or even imply
anything about development” (1992, p. 204). In contrast to this
practice, the definition of interlanguage pragmatics offered by
Kasper and Dahl (1991) the year earlier included acquisition.
Interlanguage pragmatics was defined as “referring to nonnative
speakers’ comprehension and production of speech acts, and how
that L2-related knowledge is acquired” (p. 216).1 The definition of
interlanguage pragmatics was more ambitious than the contem-
porary practice, and 8 years later, it still is: The study of how
L2-related speech act knowledge is acquired is more of a desidera-
tum than a reality.
If interlanguage pragmatics research cannot be charac-
terized as acquisitional, what is an accurate characterization? It
has been essentially comparative, comparing what learners or
non-native speakers do to what native speakers do. As Kasper
(1992) observed,
The bulk of interlanguage pragmatics research derived its
research questions and methods from empirical, and par-
ticularly cross-cultural, pragmatics. Typical issues ad-
dressed in data-based studies are whether NNS differ from
NS in the 1) range and 2) contextual distribution of 3)
strategies and 4) linguistic forms used to convey 5) illocu-
tionary meaning and 6) politeness—precisely the kinds of
issues raised in comparative studies of different communi-
ties. . . . Interlanguage pragmatics has predominantly been
the sociolinguistic, and to a much lesser extent a psychol-
inguistic [or acquisitional] study of NNS’ linguistic action.
(Kasper, 1992, p. 205)

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 679

At the time that Kasper’s (1992) article “Pragmatic Transfer”

was written, only three longitudinal studies had been published:
Schmidt’s (1983) well-known report on Wes, a learner of English,
Schmidt and Frota’s (1986) study of a beginning learner of Brazil-
ian Portuguese, and Billmyer’s (1990) study of instructed learners
of English. Cross-sectional studies at that time included Scarcella
(1979), Olshtain and Blum-Kulka (1985), Blum-Kulka and
Olshtain (1986), Takahashi and Beebe (1987), Trosborg (1987), S.
Takahashi and DuFon (1989), and Omar (1991). A significant
number of longitudinal studies (regrettably rare in any area of
SLA) were published about the same time as Kasper’s article,
reflecting that other researchers also saw the need for acquisi-
tional research. These studies included Ellis’s (1992) longitudinal
study of two children’s untutored acquisition of English requests,
and Sawyer’s (1992) study of the acquisition of the sentence-final
particle ne by American learners of Japanese. Bouton (1992)
investigated the development of comprehension as related to
implicature and Bardovi-Harlig and Hartford (1993) studied the
changes in the speech acts of advanced nonnative speakers during
their advising sessions in their 1st year of graduate school. Even
after the flurry of longitudinal studies around 1992, however, the
field could not be characterized as “acquisitional.”
Nor did the balance of studies in the field change between
1992 and 1996. However, when Kasper and Schmidt’s (1996) paper
“Developmental Issues in Interlanguage Pragmatics” repeated
the observation that interlanguage pragmatics was more com-
parative than acquisitional, it was much more striking. Although
additional cross-sectional (Kerekes, 1992; Robinson, 1992; Svanes,
1992; Trosborg, 1995) and longitudinal (Siegal, 1994) studies had
been conducted, the relative handful of longitudinal, or even
cross-sectional, studies had done very little to change the overall
character of interlanguage pragmatics.2 Kasper and Schmidt’s
paper helped me articulate my growing discontent with the com-
parative stance of most studies, and I had come to understand that
not only was interlanguage pragmatics not fundamentally acqui-
sitional, but it was, in fact, fundamentally not acquisitional.

Copyright © 2000. All rights reserved.

680 Language Learning Vol. 49, No. 4

One striking detail that reveals the extent to which acquisi-

tion has been overlooked in interlanguage pragmatics is that
many articles from 1979 to 1996—with the exception of the explic-
itly acquisitional studies with cross-sectional and longitudinal
designs—identify non-native speakers as “non-native speakers”
rather than learners, and they are described only by their first
language, with no attempt to characterize variables such as pro-
ficiency levels or length of residency in the host environment. This
detail of reporting subject profiles reveals not that they were badly
done acquisitional studies, but that they were not acquisitional
studies at all.
The consequence of the dominance of comparative studies
over acquisition studies in interlanguage pragmatics has been
that the areas of investigation in interlanguage pragmatics are
quite distinct from studies in second language acquisition. This
division seems to have real-world correlates as well. When papers
are reviewed for the annual AAAL conference, for example, there
are two separate strands: one for cross-cultural and interlanguage
pragmatics together and another for SLA. Consider also that the
relatively new Handbook of Second Language Acquisition (Ritchie
& Bhatia, 1996) has no chapter on pragmatics.
Why has interlanguage pragmatics developed as it has?
Kasper (1992) provides the central reason: because cross-cultural
pragmatics has served as the model for interlanguage pragmatics
research. A second reason is that the research has concentrated
on investigating the pragmatics of advanced NNSs rather than
learners at all levels.3 Our elicitation tasks (predominately writ-
ten discourse completion tasks that involve reading scenarios)
favor advanced learners. Moreover, the availability of English-
speaking undergraduate and graduate students at universities
around the world has reinforced the tendency to use advanced
learners. In addition, the fact that studies of this population reveal
that even very advanced learners have not mastered basic prag-
matics gives them a certain shock value: To show that advanced
learners have not mastered certain areas of L2 pragmatics is a
more stunning result than to show that low-level learners differ

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 681

from the target norm. Finally, as a pedagogical needs assessment,

the demonstration that advanced learners do not exhibit tar-
getlike norms implies that instruction is warranted at all levels of
development. As a result of SLA and interlanguage pragmatics
having evolved as specialized research areas, we find a third and
later development that will continue to separate the fields of
inquiry: In many cases, the research has come to be conducted by
different practitioners.
Assuming that an acquisitional focus is desirable in interlan-
guage pragmatics research, what would an acquisitional interlan-
guage pragmatics look like? Such a research agenda is found in
Kasper and Schmidt’s (1996) lead article to the thematic issue of
Studies in Second Language Acquisition dedicated to the develop-
ment of pragmatic competence. Following Kasper and Schmidt, an
acquisitional research agenda in interlanguage pragmatics would
investigate two essential areas: changes within the L2 pragmatic
system and influences on that system. In Table 1, I summarize the
areas of investigation identified by Kasper and Schmidt (1996).
It is important to note that some of the questions in Table 1
look as unfamiliar in an interlanguage pragmatics investigation
as they look commonsensical in second language acquisition re-
search. Presumably, these questions would be addressed by the
measures that interlanguage pragmatics has used thus far, al-
though new measurements could be developed (see Table 1, Point
b). NS and NNS speech act production can differ in at least four
ways, and any or all of these comparisons could be used to evaluate
differences in an acquisitional study across time (in a longitudinal
study), or across proficiency levels or lengths of stay (in a cross-
sectional study).4 The measures include (a) choice of speech
acts—do learners and NSs perform the same act in a given
situation? and (b) use of semantic formula—there are many ways
to realize a response to a compliment; for example, do we express
gratitude (by saying thank you), downgrade the compliment (oh,
this old thing), accept responsibility (I made it myself), or recip-
rocate (you look pretty terrific, too). Differences also include (c) the
content of the propositions we encode. When turning down an

Copyright © 2000. All rights reserved.

682 Language Learning Vol. 49, No. 4

Table 1

Basic Questions About SLA With Respect to Interlanguage Pragmatics

a. Are there universals of pragmatics and do they play a role in interlan-
guage pragmatics?
b. How can approximation to target language norms be measured?
c. Does L1 influence L2 pragmatics? (Transfer)
d. Is the development of L2 pragmatics similar to learning a first language?
e. Do children enjoy an advantage over adults in learning a second lan-
guage?
f. Is there a natural route of development as evidenced by difficulty, accu-
racy, or acquisition orders or discrete stages of development?
g. Does type of input make a difference? (foreign language vs. second language)
h. Does instruction make a difference?
i. Do motivation and attitudes influence level of acquisition?
j. Does personality play a role?
k. Does a learner’s gender play a role?
l. Does (must) perception or comprehension precede production in acquisi-
tion?
m. Does chunk learning (formulaic speech) play a role in acquisition?
n. What mechanisms drive development from stage to stage?
Note. The questions in Table 1 are derived from the section headings in
Kasper and Schmidt (1996).

invitation, we could be vague (I have something to do) or specific

(I have to go to my cousin’s wedding). And finally, (d) differences in
linguistic form: Is the speech act realized with downgraders or
aggravators? (Could you do me a favor versus Just do me a favor).
Measuring change in interlanguage pragmatics systems either
cross-sectionally or longitudinally would result in more acquisi-
tionally oriented interlanguage pragmatics studies, at once link-
ing interlanguage pragmatics research more directly to SLA
research and maintaining its connection to cross-cultural prag-
matics. This is a necessary stage in the maturing of the field of
interlanguage pragmatics research, and an increasing number of
studies in this area will be undertaken in the coming years.
Although this stage of research has not yet been fully ex-
plored, I propose that we go still further in studying the acquisition
of L2 pragmatic competence by investigating the relation of the

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 683

development of the grammatical and pragmatic systems. The first

step is to examine how grammatical competence had been acknowl-
edged in interlanguage pragmatics research; I will begin with a
brief review of how proficiency in general has been handled. (For
more thorough coverage of the issue of proficiency, see Bardovi-
Harlig, 1999a; Kasper & Schmidt, 1996).

Previous Studies on Acquisition in Pragmatics

A consequence of the comparative focus of interlanguage

pragmatics is that there have not been enough longitudinal stud-
ies to allow comparison across learners, contexts, or languages.
However, there have been sufficient cross-sectional studies to
begin to compare effects of levels of proficiency and on length of
stay of pragmatic development.
Low- and high-proficiency learners were found to differ in
the order and frequency of semantic formulas they used. Lower-
proficiency learners were also more direct in their refusals than
higher-level ESL learners (T. Takahashi & Beebe, 1987). The use
of external modifiers may also increase with linguistic proficiency,
as does the number of words used (Blum-Kulka & Olshtain, 1986).
The use of lexical downgraders (downtoners, understaters, hedges,
subjectivizers, intensifiers, commitment upgraders, and cajolers)
also improves with proficiency (Trosborg, 1987).
In a rare study that included low-level learners, Scarcella
(1979) found that when making requests, the low-level learners
invariably relied on imperatives, whereas higher-level learners
showed sensitivity to status, restricting the use of imperatives to
equal familiars and subordinates. Koike (1996) also found a pro-
ficiency effect in the recognition of the intent of speech acts in a
study of the perception of Spanish suggestions by English-speak-
ing learners of Spanish.
Proficiency may also influence transfer. Advanced learners
were found to be better than intermediate learners at identifying
contexts in which L1 apology strategies could be used successfully
(Maeshiba, Yoshinaga, Kasper, & Ross, 1996). Japanese learners

Copyright © 2000. All rights reserved.

684 Language Learning Vol. 49, No. 4

of English as a second language showed a greater tendency to

soften the directness of their refusals than did lower-level Japa-
nese ESL learners, and they also showed a greater level of formal-
ity, both of which T. Takahashi and Beebe (1987) attributed to
transfer from Japanese refusals.
Other areas are apparently less sensitive to level of profi-
ciency. Kasper and Schmidt’s review of the literature found that
proficiency may have little effect on the range of realization
strategies that learners use (Kasper & Schmidt, 1996), and S.
Takahashi (1996) found little proficiency effect on learners’ per-
ception of the transferability of L1 request strategies as a whole.
Both low- and high-intermediate proficiency learners equally re-
lied on their L1 request conventions in L2 request realization (S.
Takahashi, 1996, p. 210). However, as Kasper (personal communi-
cation, March 1999) pointed out, the absence of a proficiency effect
may be due to the fact that real beginners were not included in
the studies. The child L2 learners observed by Ellis (1992) did not
produce a full range of request strategies at the end of the 15–21-
month observation period. The reduced range may be in part
attributable to the classroom context in which the requests were
collected; nevertheless, there is a possibility that request strate-
gies are constrained by proficiency.
Length of stay is also a factor in pragmatic development.
Blum-Kulka and Olshtain (1985) reported an increase in accep-
tance of direct request strategies by NNS speakers of Hebrew as
their length of stay increased. In fact, Blum-Kulka and Olshtain
(1986) showed that proficiency is not as good a determiner of
pragmatic development as length of stay with respect to length of
utterance. The longer requests of the NNSs of Hebrew were
attributed to their greater use of external modification (i.e., ele-
ments added to the head act, such as reasons and justifications for
the requests). Such external modification increased across profi-
ciency levels—higher-proficiency learners used more external
modification moving away from the native-speaker
norm—whereas external modification decreased the longer an
NNS had lived in Israel. This led Blum-Kulka and Olshtain to

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 685

conclude that “in terms of pragmatic competence, length of stay is

a much more interesting measure than level of linguistic profi-
ciency” (1986, p. 174). Blum-Kulka and Olshtain’s work points to
the necessity of considering proficiency and length of stay inde-
pendently in a context where foreign-language learners may ar-
rive in the host environment with a high level of proficiency, but
no experience in the target culture. In the same setting there may
be learners with lower proficiency who have been in the target
environment longer (see Blum-Kulka & Olshtain, 1986, fn. 9).
In a longitudinal study of the advising session talk of NNSs
with high grammatical proficiency who had just arrived at an
American university, Bardovi-Harlig and Hartford (1993) found
that NNSs showed an increase in the use of speech acts favored
by NSs in the academic context, and they showed a decrease in
speech acts not used by NSs as length of stay increased. In the
same period, however, NNSs did not conform to NS use of mitiga-
tors and nonuse of aggravators (Bardovi-Harlig & Hartford, 1993).
Bardovi-Harlig and Dörnyei’s (1998) study of awareness of devia-
tions from NS norms showed that tutored ESL learners in an
intensive program were more sensitive than EFL learners to
pragmatic infelicities. Within the ESL group, learners at a high
level of proficiency showed greater pragmatic awareness than
learners at lower proficiency. Bouton (1992, 1994) also found that
ESL learners enrolled at an American university without specific
training in pragmatics became increasingly targetlike in their
interpretation of implicature as length of stay increased.
Even shorter lengths of stay might help learners become
more targetlike, particularly with respect to highly salient conver-
sational functions such as greetings. Omar (1991, 1992) found that
American learners of Swahili who had been to Tanzania showed
much more targetlike use of multiple turns in lengthy Swahili
greetings (Omar, 1991, 1992). Similarly, Hoffman-Hicks (1999)
observed that American university students of French also ad-
justed their greetings to be more targetlike during a semester in
France, but theirs became shorter and less frequent.

Copyright © 2000. All rights reserved.

686 Language Learning Vol. 49, No. 4

As we have seen, some acquisition studies compare learners

using a broad concept of proficiency. Our next step is to ask
whether interlanguage pragmatics studies have taken specific
note of the grammatical development of their learner-participants.
In the next section I consider pragmalinguistic competence, the
linguistic competence that allows speakers to carry out the speech
acts that their sociopragmatic competence tells them are desir-
able. Because this discussion focusses on what grammatical
knowledge is necessary for pragmalinguistic competence, the fol-
lowing section divides pragmalinguistic competence into its prag-
matic and grammatical parts. The term grammatical as I am using
it here encompasses anything having to do with a linguistic
grammar, including phonology, morphology, syntax, semantics, or
the lexicon of the developing language.

The Relation of Pragmatic and Grammatical Competence

What do we know about the relation between grammatical

competence and pragmatic competence? Working with highly
grammatically proficient learners and non-native speakers has
shown that high levels of grammatical competence do not guaran-
tee concomitant high levels of pragmatic competence. As early as
1985 Olshtain and Blum-Kulka observed, “It has been shown
repeatedly in the literature that second language learners fail to
achieve native communicative competence even at a rather ad-
vanced stage of learning (or acquisition in the natural setting)” (p.
321). In 1990, Bardovi-Harlig and Hartford made a similar obser-
vation regarding graduate students at an American university.
From these and other studies that investigate advanced NNSs we
learn that interlanguage grammatical competence is not a suffi-
cient condition for interlanguage pragmatic competence—but is it
a necessary condition? Asked another way, is pragmatic compe-
tence built on a platform of grammatical competence? Very few
studies make the link between pragmatics and the interlanguage
system. How have researchers talked about the relation of gram-
matical competence and pragmatics?

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 687

When we review the literature from this perspective we find

that, with very few exceptions, mention of grammatical compe-
tence is very brief and appears only in the discussion section as a
possible interpretation of results. The real exception to this claim
is a 1987 article by Blum-Kulka and Levenston entitled “Lexical-
Grammatical Pragmatic Indicators.” The article presents an over-
view of lexical simplification, morphological constraints, and
syntactic development as they affect the realization of speech acts.
This study takes an error analysis perspective showing what can
go wrong in the grammatical and lexical realization of speech acts
in L2 English and L2 Hebrew, but it is nonetheless a step in the
direction that I am advocating. Similarly, in their article “ ‘I Very
Appreciate,’ ” whose very title suggests that a degree of grammati-
cal knowledge is necessary for pragmatic realization, Eisenstein
and Bodman (1986) observed “extensive syntactic and lexical
problems” in the written responses of advanced learners. They
listed problems such as intensifiers, tense, word order, “mis-
used/mangled idioms,” and word choice (p. 175).
Effects of development in the lexicon were noted by Blum-
Kulka and Levenston (1987) where learners of Hebrew use the
equivalent of have in requests such as “Can I have your notes?”
rather than the more specific terms such as borrow or lend which
according to Blum-Kulka and Levenston are less scary to the
addressee. Cohen and Olshtain (1993) also identified development
in the lexicon as the culprit behind the awkward request “I want
to drive with you” in a context where “Could you give me a lift?”
would be appropriate. In a retrospective task the learner reported
not knowing how to use the word lift, which appeared in the
role-play scenario. T. Takahashi and Beebe (1987) also noted that
when making refusals, higher-level learners used, and in fact
overused, a greater range of intensifiers, really, awfully, terribly,
truly, deeply, and extremely. Such use seems to be dependent on
lexical acquistion.
In interpreting a similar observation made by Trosborg
(1987) that higher-proficiency learners used more downgraders
(downtoners, understaters, hedges, subjectivizers, intensifiers,

Copyright © 2000. All rights reserved.

688 Language Learning Vol. 49, No. 4

commitment upgraders, and cajolers) than lower-proficiency

learners, Maeshiba et al. (1996) observed that “it is difficult to say
whether this pattern truly reflects a development of pragmalin-
guistic competence or merely an extension of the learners’ lexical
repertoire” (p. 160).
More general linguistic competence is discussed as well. In
her discussion of apology strategies, Trosborg (1987, p. 165) noted,
“The low number of explanations used by learners is a likely
outcome of insufficient linguistic knowledge. In order to provide a
convincing explanation or give an adequate account, you need the
relevant linguistic means. Likewise, a high degree of proficiency is
demanded to query the preconditions on which an accusation is
built” (emphasis added). Koike (1996) observed that lower-level
learners of Spanish as a foreign language had difficulty identifying
the illocutionary force of suggestions and particular difficulty with
negative interrogative suggestions such as ?No has pensado en leer
este libro? and ?No deberías leer este libro? (illocutionary-force
equivalents of the non-negative suggestions in English “Have you
thought about reading this book?” and “Shouldn’t you read this
book?” respectively). She concluded, “this process implies a knowl-
edge of the target language speech acts at both the grammati-
cal/lexical level as well as the pragmatic level of use” (p. 275).
T. Takahashi and Beebe (1987) found that their lower-
proficiency group used direct expressions such as “I can’t” when
making refusals, and they attributed this to a developmental
stage, stating, “the point is that the higher frequency of direct
expressions among lower proficiency learners is not a function of
NL transfer, but rather most probably a developmental stage
where simpler, and also more direct, expressions are being used”
(p. 150).
Olshtain and Cohen (1989) reported that “it often happens
that nonnative speakers are aware of the sociolinguistic need to
apologize, yet because their linguistic competence is limited, they
use erroneous language forms and produce speech acts that sound
deviant or even create communication failure” (p. 62, emphasis
added).

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 689

Just as transfer studies have taken the lead in investigations

of acquisition in interlanguage pragmatics, they also have also
paid the most attention to the role of proficiency. Probably the best
known study to investigate the effect of the level of proficiency on
transfer was T. Takahashi & Beebe’s (1987) study. They hypothe-
sized that
the peak of the [pragmatic transfer] curve would be ex-
pected to fall at higher proficiency levels than the peak of
the curve for phonological or morphological transfer. Simi-
larly for the developmental curve of any one individual, we
would expect phonological and morpho-syntactic transfer
to rise to their peaks earlier than pragmatic transfer since
pragmatic transfer requires more fluency to surface. (p. 153)
This echoed earlier observations by Blum-Kulka (1982) that
Canadian learners of Hebrew did not transfer request strategies
to Hebrew as much as they might have been expected to because
they lacked the complex L2 knowledge necessary to implement
indirectness, and Cohen and Olshtain (1981) who found “that L2
learners sometimes avoided using a semantic formula which had
high frequency in their L1 and which would have allowed for
positive transfer into L2, because they were lacking linguistic
competence in L2” (Olshtain & Cohen, 1989, pp. 62–63). Wildner-
Bassett (1994) observed that a learner ”wants to carry [a] routine
over into her use of German, to the extent that her proficiency
allows” (p. 11). And Maeshiba and colleagues (1996) observed that
in the case of positive transfer, advanced learners used the same
strategies as native speakers in both the L1 and the L2, and they
concluded that this “makes sense in light of the assumption that
advanced learners are likely to be more acculturated, and have the
linguistic facility to transfer pragmatic strategies from the native
language” (p. 169, emphasis added).
House (1996) described the dilemma of interpretation when
she attempted to tease out why some EFL learners were superior
to others at the end of a semester-long instructional study. Learn-
ers who had had longer stays in an English-speaking environment
were better than other learners both before and after instruction,

Copyright © 2000. All rights reserved.

690 Language Learning Vol. 49, No. 4

and instruction did not close the gap between them. Were these
same learners superior grammatically at the outset of their host
experience, or was the communicative experience alone sufficient
to increase their pragmatic competence? She observed that
It was not possible to answer the question of whether those
students who were more competent linguistically had been
superior before they entered the English-speaking envi-
ronment—such that they had a more solid basis for the
development of their pragmatic competence in the first
place—or whether their pragmatic competence strongly
benefited from the inimitably rich variety of communica-
tive experience real everyday life affords. (p. 245, emphasis
added)

Investigating the Relation of Grammatical

and Pragmatic Development

House (1996) accurately summed up the state of the field

when she wrote, “whether the level of linguistic competence is a
necessary precondition for efficient learning of pragmatic phe-
nomena such as the ones discussed here cannot be further elabo-
rated here” (p. 245). However, as we consider the future of
interlanguage pragmatics, I would say that we have sufficient
resources to take up the challenge and address the issue now. But
where should we begin? How do we move beyond very broad
concepts of proficiency or competence to examine particular and
well-defined interlanguage development as it relates to pragmatics?
One excellent place to start is House and Kasper’s (1981) list
of mitigators and downgraders. Consider the presupposition of
linguistic competence that lies behind the list in Example 1.
1. Downgraders (House & Kasper, 1981)
a. Politeness markers (please)
b. Play-downs (past tense, progressive, modals, negation, in-
terrogative)
c. Consultative devices (would you mind)
d. Hedges (kind of, sort of, somehow)

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 691

e. Understaters (a little bit, a second)

f. Downtoners (perhaps, possibly)
g. “Minus” committers (I think, I guess, I suppose)
h. Forewarnings (anticipatory devices such as, you’re a nice
guy, Jim, but or far be it from me to belittle your efforts, but)
i. Hesitators (deliberate malformulations used to indicate
reluctance to perform the ensuing speech act such as stutter-
ing or repetition)
j. Scopestaters (I’m afraid that, I’m not happy that)
k. Agent avoiders (passive and impersonal constructions)
l. Embedding (Blum-Kulka & Levenston, 1987)
In play-downs a speaker draws on knowledge of modals, tense, and
aspect, and on syntactic knowledge of negation and question
formation. With hedges and understaters a speaker must have
enough syntax to properly position them in the sentence. With
consultative devices and scopestaters a learner needs knowledge
of the complements that particular formulas take, and with agent
avoiders, the learner needs to know formation and use of passive.
Even minus committers put a strain on a learner’s suprasegmen-
tals, and embedding as a means of marking politeness has obvious
syntactic requirements.

The Development of the Tense-Mood-Aspect System

Related to Pragmatics

Modals have been demonstrated to play an important role in

mitigation. In House and Kasper’s (1981) inventory of downgrad-
ers alone, modals were listed as a group under play-downs and
individually under consultative devices (see Example 1). Other
studies that highlight the pragmatic role played by modals include
House and Kasper (1987), Blum-Kulka (1989), Faerch and Kasper
(1989), and House (1989). Understanding the acquisition of mo-
dals and modal expressions in interlanguage thus becomes rele-
vant to understanding how learners use modals as downgraders.
A paper by Dittmar and Terborg (1991), entitled “Modality and

Copyright © 2000. All rights reserved.

692 Language Learning Vol. 49, No. 4

Second Language Learning,” offers a model for research that

integrates interlanguage research with interlanguage pragmat-
ics. Although modality is recognized as both a semantic and
pragmatic category (p. 348), Dittmar and Terborg’s study did not
set out to examine the use of modals related to the development
of pragmatic competence per se. Nevertheless, their research
yields important lessons for students of interlanguage pragmatics.
The paper uses a concept-oriented inquiry to investigate the
expression of modality in German, and thus, in keeping with the
European studies of this type, investigates a range of linguistic
devices used to express modality in interlanguage. These include
modal verbs, modal adverbs, intonation, and, in interlanguage, the
request marker bitte “please.” The concept-oriented approach has
been used very successfully in the study of the acquisition of
temporal expression (e.g., Dietrich, Klein, & Noyau, 1995) and
often results in an interesting mix of linguistic devices being
studied together, whereas the form-oriented approach in SLA
would focus on a specific form such as modal verbs.
In this study Dittmar and Terborg (1991) focussed on one
learner drawn from a longitudinal study of 16 native Polish-speak-
ing informants learning German. Their informant was observed
from a very early stage of interlanguage development and partici-
pated for the full 2.5 years of the study, during which time 18
transcripts were collected. Throughout the study, learners per-
formed narrative tasks, instruction-giving tasks, and problem-
solving tasks. The instruction-giving tasks are particularly
important to request formation. In one of the instruction-giving
tasks, for example, the ashtray task (adopted from the ESF project;
see Becker & Carroll, 1997), a researcher silently acted out a scene
for the learner in which a guest in a cafe steals an ashtray.
Following the initial demonstration, the learner gave instructions
to a second researcher so that he could act out the same scene
(p. 353).
In interlanguage, modal expressions include bitte “please,”
which is the first linguistic device used to mark modality. Dittmar

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 693

and Terborg (1991) identified the order of acquisition of linguistic

devices used to express modality as shown in Example 2.
2. Order of acquisition of means of modal expression (Dittmar
& Terborg, 1991, p. 358)
i. bitte “please”
ii. müssen “must”
können “can”
iii. denken “think”
iv. möchten “would like”
wahrscheinlich “probably”
v. sicher “sure/certain”
vi. wollen “want”
vii. vielleicht “maybe”
As the first expression of modality, bitte “please” marks requests
and necessity in early stages of interlanguage because other
means of request or necessity are unavailable. According to
Dittmar and Terborg, the function of bitte as a politeness marker
comes into the interlanguage later: “Not until so-called ‘true’
deontic means are available is bitte ‘please’ used primarily to mark
politeness” (p. 359). An inventory of expressions shows that learn-
ers use “bitte ‘please’ in the sense of a request for action . . .
especially frequently when müssen ‘must’ is not to be found or is
only found to a small degree” (p. 359).
Table 2 shows the interaction of emergent modal expressions
in the interlanguage of an individual learner (Dittmar & Terborg,
1991). Bitte emerges earlier than muss and weiss. Note that when
bitte is the only means of modal expression (in Interview 3), it is
used robustly with 36 tokens. Muss and weiss emerge tentatively
(with 4 and 5 tokens, respectively) at Interview 5. The use of muss
peaks at Interview 9 (37 tokens) with a corresponding decline in
the use of bitte (only 2 tokens). A second increase in the use of bitte
(22 tokens) is seen before it stabilizes at a low rate of use in
Interviews 12–15, as muss (with 29 to 48 tokens) assumes the
modal functions that bitte carried in earlier stages. Thus, the
longitudinal evidence shows that even a very simple lexical item

Copyright © 2000. All rights reserved.

694 Language Learning Vol. 49, No. 4

Table 2

Order of Emergence of Modal Expressions for an Individual Learner

Interview no. Modal expressions

bitte muss weiss
3 36 0 0
5 4 4 5
7 9 4 0
9 2 37 5
11 22 5 3
12 1 29 11
13 4 30 4
15 0 48 5
Note. The frequency counts in Table 2 are based on the instruction-giving
task reported in Dittmar and Terborg (1991).

like please takes its place in the acquisitional sequence. According

to data from the more advanced learners, bitte surfaces again later
as a politeness marker.
Continuing with modals, consider the case of will as used by
NNS graduate students in advising talk (Bardovi-Harlig & Hart-
ford, 1993) as in Example 3b. Whereas NS graduate students
always mitigated their suggestions as in 3a, “I was thinking of
taking syntax,” in addressing an advisor about courses they
wanted to take, NNSs often did not, resulting in turns such as 3b,
“I will take syntax.” The use of will seems to have the opposite
effect of a mitigator, operating instead as an aggravator indicating
a strong commitment by the student to his suggestion for a course.
3. Graduate students addressing a faculty advisor
Advisor: OK, let’s talk about next semester.
a. NS: I was thinking of taking syntax.
b. NNS: I will take syntax.
After a presentation of the 1993 paper, a conference partici-
pant suggested that the NNS in Example 3 simply did not know
his modals. He asserted that given better grammar training, the
learner’s use of will would take care of itself. Anxious to get our

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 695

pragmatic message across I replied that this was not simply a case
of grammar learning. I wanted to head off the claim that if learners
knew or were taught grammar better, they would get pragmatics
for free.
After thinking about this exchange for several years, I’d like
to go back to this example and make the opposite claim from the
EFL teacher. I’d like to suggest that the NNS in Example 3 shows
an excellent understanding of the use of will, specifically an
understanding of the core, referential meaning of will as a marker
of the future. “I will” has a transparent form-meaning association
with the future, whereas “I was going to” does not. Although the
reader is likely to find the native-speaker’s suggestion more ap-
propriate, the learner is likely to find was going to to be an odd
equivalent to will in the pair of sentences “I will take syntax next
semester” and “I was going to take syntax next semester” when
the propositional content is held constant. Acquisitionally, we
cannot expect pragmatic extension of tense-mood-aspect forms
until the core deictic meanings have been acquired.
In fact, Andersen and Shirai (1996) captured this claim in
their treatment of the acquisition of past morphology in a proto-
type semantics model. Employing a prototype model, Andersen
and Shirai hypothesized the acquisitional sequence in Example 4
for simple past beginning with the core meaning of deictic past,
expanding to a marker of habitual or iterative past, and finally,
extending to a pragmatic function as a softener (or play-down
following House & Kasper, 1981).
4. The spread of the simple past
Deictic past > habitual or iterative past > counterfactual or
pragmatic softener (play-down) (Andersen & Shirai, 1996, p. 557)
Andersen and Shirai also hypothesized an acquisitional sequence
for the progressive. In the progressive sequence a pragmatic device
was not included, but because it is used as one, I have provisionally
inserted it in the last position in Example 5, analogous to the
sequence for the spread of the simple past.

Copyright © 2000. All rights reserved.

696 Language Learning Vol. 49, No. 4

5. The spread of the progressive (modified)

Process > iterative > habitual or future > stative progressive
> pragmatic softener (play-down) (Andersen & Shirai, 1996, p. 557)
Both of these sequences remain to be tested in their later stages,
and both point to good places to start an investigation.
Grammatical competence may also limit the value of the
pragmatic input to the learner. A learner who has not achieved
control over prototypical uses of tense-mood-aspect morphology
may not be ready to extend the use of those forms to politeness
markers in incoming data. Consider the case of an intermedi-
ate ESL student who was engaged in a pedagogical task. The
students had been asked to rate videotaped scenarios for
appropriateness in pragmatics and grammar (see Bardovi-
Harlig & Dörnyei, 1998). After students compared their re-
sponses, a student questioned the use of could in one of the “good”
scenarios (i.e., a scenario with no grammatical or pragmatic
problems). In the scenario, a student responds to his teacher’s
invitation “Peter, we need to talk about the class party soon” with
“Yeah, if tomorrow is good for you, I could come any time you say.”
My ESL student identified the use of could as problematic,
explaining that could was used for past, but that the utterance
was oriented to the future as reported in Example 6.
6. ESL learner, IEP Level 4, summer 1997
“Dr. Bardovi, there is something strange about this sentence:
If tomorrow is good for you, I could come any time you say. He
is talking about tomorrow, but he uses could, which is past.”
It appears that the learner has made a primary association of
could with past which at this stage disallows its use in nonpast
contexts. Her form-meaning association will eventually expand
beyond one-form one-meaning (the one-to-one principle; see An-
dersen, 1984), possibly by encountering more positive evidence of
this type. There is clearly quite a bit of work that can be done in
the area of modals, grammatical competence, and interlanguage

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 697

pragmatics. However, the acquisition of modals is not the only area

to be investigated.

The Development of Prosody Related to Pragmatics

The most obvious and most often-identified facet of pronun-

ciation that is singled out in discussions of pragmatics is the
suprasegmental level, namely sentence stress and intonation.
How does a learner’s interlanguage phonology affect his or her
delivery or perceived illocutionary force?
A minus committer such as I thínk quickly becomes an aggra-
vator with the opposite effect if the stress pattern changes to Í think,
in which a speaker contrasts her opinion with that of her interlocutor
rather than softening her utterance, as in Example 7.
7. NNS, Chinese (Taiwan)
A: What is it you were thinking about taking next fall?
S: Um, there are two required courses. So Í think I need to take
for this semester . . . (Bardovi-Harlig & Hartford, 1996)
Kasper (1984) and Wildner-Bassett (1984) provide two examples of
the very few studies in interlanguage pragmatics to include intona-
tion. Other work outside of interlanguage pragmatics such as the
collection of conversational analyses in Atkinson and Heritage (1984)
could provide a useful model for interlanguage pragmatics studies.
As part of their investigation of modality, Dittmar and Ter-
borg (1991) also analyzed intonation in German interlanguage, in
addition to modal development. Dittmar and Terborg found that
the earliest requests had very little linguistic structure, as shown
in Example 8.
8. Time 1 (Dittmar & Terborg, 1991)
a. komm komm komm komm ja come come come come yes
b. platz kleine platz seat little seat
c. zeitung zeitung tasche newspaper newspaper bag
In addition, these syntactically simple requests exhibited an into-
nation pattern that Dittmar and Terborg described as “dramatic”

Copyright © 2000. All rights reserved.

698 Language Learning Vol. 49, No. 4

and “strange” (p. 365). In the requests in Example 8, each word is

its own intonational phrase. They observed that at Time 1 the
informant is very uncertain of his production and the illocutionary
force of a request can hardly be inferred from the way that it is
delivered with respect to intonation. The context in which the
examples in 8 were produced makes it possible for the utterances
to be understood as requests; otherwise without context they
would be more likely to be understood as questions.
A striking difference is found between the utterances in
Example 8 from Time 1 and those in Example 9, 10 months later.
By Time 2, the requests in 9 show terminal intonation. Overlook-
ing certain issues of syntax and semantics, Dittmar and Terborg
(1991) observed that, with respect to intonation, the requests in 9
are reasonably good realizations of requests.
9. Time 2 (Dittmar & Terborg, 1991)
a. sie bitte setzen you please set
b. bitte diese tasche please this bag place small chair
platz kleine stuhl
c. bitte zietung raus please newspaper out
The study of the relation of interlanguage intonation and
speech acts is in its infancy, but Dittmar and Terborg (1991, p. 367)
identified three hypotheses for future study:
1. The illocutionary force of an utterance is primarily
expressed by intonation in the initial stages of SLA.
2. The basic function of the progradient, nonterminal, pattern
[Time 1] is the interactive signalling of the learner that s/he
is uncertain about the meaning and performative aspects of
the utterance and prefers that the conversational partner
qualify the utterance with the right illocutionary modality and
take over the conversational work.
3. It may be that speech acts in elementary learner varieties
are not very distinct from each other on the basis of formal
criteria (syntax, intonation, morphology, etc.).

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 699

The Development of Morphology Related to Pragmatics

In addition to the examples of the English tense-aspect forms

discussed earlier that are instances of morphological acquisition,
it is important to consider examples from languages with richer
inflectional systems than English. Consider the range of forms
used in two judgment studies of requests in Spanish (Rodríguez,
1997; Walters, 1979). Request forms may include, but are not
limited to (e.g., Blum-Kulka, 1989), items in Example 10. The
sample of Spanish request forms represents a range of grammati-
cal knowledge including verbal morphology representing the pre-
sent (10a, b, and c), conditional (10d and e), future (10f), formal
imperative (10g), and formal past subjunctive (10h). The second-
person forms used in the judgment tasks were all formal (usted)
forms (10a, b, d, g).5
10. Spanish request forms
Expression Grammatical English equivalent
description
a. Me puede dar 2p, pres, formal “Can you give me”
b. Tiene 2p, pres, formal “Do you have”
c. Quiero 1p, pres “I want”
d. Podría 2p, conditional, “Could you,
formal Would you”
e. Querría 1p, conditional “I would like”
f. Tomaré 1p, future “I will have”
g Déme 2p, formal “Give me”
imperative
h. Quisiera 2p, formal past “I would like”
subjunctive
Judgment tasks that ask learners to draw on such a broad base
of grammatical knowledge assume that learners can distin-
guish among the forms. Moreover, by including only formal
verbal morphology in second person, they may further assume
that learners can make the distinction between informal and
formal verbal morphology corresponding to the use of tu and usted.

Copyright © 2000. All rights reserved.

700 Language Learning Vol. 49, No. 4

This assumption must be examined for classroom foreign language

learners who are engaged in a status-unequal encounter (Kasper,
1997b), an environment in which the learners are addressed by their
teachers informally (Rodríguez, personal communication, 1997) and
thus only the second-person tu forms are modelled. To use judgment
tasks successfully, it seems that learner familiarity with morphology
must somehow be determined independently of the judgment task
before the learner politeness rankings can be assessed.
Finally, consider how grammatically sophisticated the re-
quest alternatives in Example 10 compare to the German produc-
tion data for interlanguage requests (Dittmar & Terborg, 1991).
By analogy to the German production data, we might expect
production by beginning learners of Spanish to include por favor
“please” (by analogy to the use of bitte “please” in early German
interlanguage) and isolated NPs (as in b, platz kleine platz “seat
little seat”), followed by use of informal morphological forms by
more advanced learners (such as the classroom use of digame “tell
me” used by students to surprised teachers, Rodríguez, personal
communication, 1997), and finally emergent use of the morphology
in Example 10.

The Development of Grammatical Complexity Related to Pragmatics

The grammatical competence required for syntactic devices

is perhaps more straightforward than in the other areas. In a
study of requests, S. Takahashi (1996) found that learners favor
monoclausal request formulas, whereas NSs prefer biclausal re-
quest formulas, as summarized in Example 11.
11. Use of embedding (S. Takahashi, 1996, 1998)
Monoclausal requests (preferred by NNSs)
a. Would you (please) VP
b. Could you (please) VP
Biclausal requests (preferred by NSs)
c. Would it be possible for you to VP?
d. I was wondering if you could VP.

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 701

A typical interlanguage pragmatics study would compare native-

speaker production in 11c and d to learner production in 11a and b.
Adding a focus on grammatical competence would also ask why
learners do not use the biclausal requests in 11c and d: Are the
learners simply unfamiliar with the pragmatic value of the em-
bedded realizations or do they additionally show only emergent
use of such constructions throughout the grammar that precludes
their pragmatic use? I will take this case up again later.

Lexical Acquisition and Formulaic Learning in Pragmatics

Finally, in the realm of the lexicon, Cohen and Olshtain (1993)

provide a retrospective method for uncovering the influence of
lexical development on the production of semantic formulas as
previously discussed. Related to lexical development is formulaic
learning. Some of the syntactic arguments are simplified if empiri-
cal work reveals that some of these issues are better described by
an appeal to formulaic speech. Note that Kasper and Schmidt
(1996) identified the role of formulaic learning as one of the areas
in need of investigation (see Table 1, Point m).
Scarcella (1979) interpreted expressions such as I would like
as formulaic devices, not representative of a speaker’s L2 compe-
tence, especially in the case of low-level learners (p. 283). Scarcella
concluded that L2 learners seem to use politeness features before
they acquire rules that govern their distribution. Schmidt (1983)
also provided examples of formula learning by Wes, who appropri-
ately used the formula “Shall we go?” as a suggestion, but did not
extend it to any other contexts at an early stage of development.
At the same stage, he used expressions such as “Sitting?” to mean
“Shall we sit down?” or “Let’s sit down.” By the end of the obser-
vation period of Wes’s interlanguage development, gross errors in
the performance of directives had largely been eliminated and the
use of progressives like “sitting” had disappeared. Schmidt sum-
marized Wes’s progress as follows:

Copyright © 2000. All rights reserved.

702 Language Learning Vol. 49, No. 4

[Because] Wes is highly motivated to engage in inter-

action and communication and in general has developed
considerable control of the formulaic language that acts as
social grease in interaction, we might expect that he would
show more development over time in the area of sociolin-
guistic competence compared with his very limited devel-
opment in grammatical competence. This is, in general,
the case. (p. 154)

In this case, and perhaps in the case of other learners like Wes,
pragmatic development is not dependent on grammatical devel-
opment. Although other studies that I have cited suggest that
there is a relation between grammatical and pragmatic develop-
ment, both outcomes are consistent with the line of investigation
I have argued we should pursue. It is important to bear in mind
that identifying the relationship—whatever that turns out to
be—is necessary to advance our understanding of interlanguage
pragmatics. It is likely that, as in other areas of SLA research,
there will be some individual variation, and there will also be
varying degrees of involvement of grammatical competence de-
pending on the linguistic complexity of the speech act realization.

Grammar in Interlanguage Pragmatics Instructional Studies

One of the ramifications of factoring in the role of grammati-

cal competence that immediately comes to mind is for instruc-
tional studies in interlanguage pragmatics (see Table 1, Point h).
Next to transfer studies, I think that influence-of-instruction
studies are going to form the most significant body of acquisitional
interlanguage pragmatics studies. Examples of such work are
House (1996), LoCastro (1997), and S. Takahashi (1998). Other
pedagogical evidence is summarized in Kasper (1997a), and a new
collection edited by Rose and Kasper (1999) on instructional
experiments that focus on the teaching of L2 pragmatics will
appear shortly.
Adding a relevant grammar-oriented probe at the outset of a
pragmatics instructional study is not unlike the syntax screening

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 703

tests used in SLA syntax studies. For example, White (1989)

described a test to determine whether learners have mastered
WH-movement that was administered to determine which learn-
ers could provide valid judgments of subjacency violations. In the
case of pragmatics, we would not necessarily be screening partici-
pants, just collecting additional data to aid our interpretation.
What kind of information would tests of grammatical devel-
opment provide when added to the obligatory pretest-posttest-
retention test design of instructional-effect studies? I will
construct a hypothetical study modelled on S. Takahashi’s (1998)
study of instructional effects on the production of monoclausal and
biclausal requests to illustrate some potential benefits of such an
approach. In this line of investigation, an appropriate grammati-
cal pretest would probe the learners’ ability and willingness to
produce multiclausal sentences.
If learners who show no embedding in any context do not
respond to pragmatic instruction on the pragmatic use of embed-
ding, we might interpret this as evidence that a certain level of
syntactic competence must precede this particular aspect of prag-
matic competence (Table 3). If learners who show no embedding
in any context before instruction show embedding in requests and
other contexts after instruction, we might interpret this as evi-
dence that syntax and pragmatics can be learned at the same time.
On the other hand, if learners who show no embedding in any
context before instruction show embedding in requests only, in
response to instruction (but in no other contexts), we might inter-
pret this as evidence of formulaic learning. Finally, if learners who
show use of biclausal constructions in nonpragmatic uses before
instruction show use of biclausal constructions as a mitigator after
instruction, we might interpret this as a case where syntactic
development is a prerequisite to pragmatic development. We
might further interpret this as a case of the teachability hypothe-
sis (Pienemann, 1989, 1998) in one of two ways: that a learner’s
having completed a prior acquisitional stage has increased the
potential influence of instruction on interlanguage development, or
that the instruction has focussed on the next stage of acquisition.

Copyright © 2000. All rights reserved.

704 Language Learning Vol. 49, No. 4

Table 3

A Grammatical Interpretation for an Instructional Experiment

in L2 Pragmatics

Before instruction After instruction Possible interpretation

No embedding No embedding Syntactic competence

in any context precedes pragmatic
competence
No embedding Embedding in requests Syntax and pragmatics
in any context and other contexts simultaneously
No embedding Embedding in requests Formulaic learning
in any context only
Use of embedding, Embedding in Syntactic development
but not generalized requests also necessary to pragmatic use
to pragmatic use

Expanding Elicitation Tasks to Accommodate Acquisition Studies

Finally, whether we broaden our focus to include interlan-

guage studies as I have argued for here, or broaden our interlan-
guage pragmatics studies to include acquisition as part of the
research agenda following Kasper and Schmidt (1996), it means
including learners at all levels, especially at the lower levels.
Including lower-level learners will require some modifications to
standard elicitation practices in order to make them more acces-
sible. Visually oriented tasks such as presenting scenarios on video
(Leary, 1994) or the use of printed cartoons (Rose, in press) will
become more important because lower-level learners can interpret
them more easily than the common written presentation (Bardovi-
Harlig, 1999b). The more obvious and easier solution of delivering
scenarios in the first language is much less desirable because that
could increase both first language and culture influence.
Oral data will also become increasingly important as we seek
to understand the role of chunk learning. What may appear to be
a smooth delivery of a formula in a written response may show

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 705

difficulty of retrieval in authentic oral production as shown in the

conversations between students enrolled in an intensive English
program and the program advisor in Examples 12 and 13
(Bardovi-Harlig & Hartford, 1997).
12. Level 3 (low-intermediate), male, L1 Brazilian Portuguese
S1: Maybe, maybe I, uh, maybe I can (..) I can (.) jump one
level, do you understand?
13. Level 7 (advanced), male, L1 Korean
S2: So, I just, I’m just asking you, um, do you, could you, please
let me, could you please um, give me some information, if, if
possible?
In Examples 12 and 13 the learners struggle with modal
expression. The learner in Example 12 seems to build his request one
word at a time, whereas the learner in 13 tries out a variety of request
formulas, “I’m just asking,” “do you,” “could you,” and “please let me,”
before settling on “could you please give me some information.” Such
important production information is obscured in written tasks.
Lastly, we note that in Dittmar and Terborg’s (1991) study,
the instruction-giving task had a single addressee, simplifying the
sociolinguistic variable of the addressee. This may or may not
prove to be a useful design simplification for working with the
lowest-level learners. It is worth considering such a simplification
when working with beginning learners. We must also keep in mind
that oversimplification with respect to addressees may run the
risk of missing the stages at which learners are able to linguisti-
cally differentiate between addressees. The ultimate goal will be
to increase our acquisitional savvy without sacrificing our prag-
matic orientation. Arriving at that balance will undoubtedly take
some experimentation.
As we incorporate different levels of learners into our studies,
we should be prepared for additional challenges in analysis as
well. Lower-level-learner speech acts will be harder to identify,
especially in conversation (as opposed to more controlled elicita-
tion tasks), as shown in Example 14, collected as part of the corpus

Copyright © 2000. All rights reserved.

706 Language Learning Vol. 49, No. 4

for Bardovi-Harlig and Hartford (1997). Potential ambiguity of

speech acts was also noted by Dittmar and Terborg (1991).
14. Level 3 (low-intermediate), male, L1 Brazilian Portuguese
A: If you start in March, then you can stay in the IEP until (.)
December.
S1: December?
A: Or
S1: I don’t think so, December, because if I have (.) um
A: Level 3
S1: Level 3?
A: Next session Level 4
In this example we see what might be categorized as a negative
response by S1 (in italics); it is probably a disagreement, but it is
also possibly a rejection of a suggestion. It is unlikely that we
would be able to make fine-grained distinctions between similar
speech acts in the speech acts of lower-level learners as Murphy
and Neu (1996) did when they distinguished between complaints
and criticisms in a role-play by advanced non-native graduate
students. This anticipated difficulty in speech act identification is
consistent with Dittmar and Terborg’s (1991) observation that
speech acts in elementary learner-language are not very distinct
on the basis of formal criteria.

Concluding Remarks

The expansion of interlanguage pragmatics to include acqui-

sition is not in conflict with continuing the practice of interlan-
guage pragmatics research as we know it. What I am advocating
is a broadening of the field of inquiry. The goals of the research
agenda that I have outlined here can be summarized as follows:
expanding learner populations to include beginning-level learners
and modifying elicitation procedures appropriately; implementing
cross-sectional studies in which acquisition can be studied across
levels of proficiency; instituting longitudinal studies when possi-
ble; and integrating the investigation of the development of inter-

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 707

language grammar with investigations of emergent pragmatic

competence.
In closing, I believe that in order for interlanguage pragmat-
ics to mature as an area of investigation in second language
acquisition, we must begin to ask acquisitional questions of the
type outlined by Kasper and Schmidt (1996), addressing the
question “How does L2 pragmatic competence develop?” To also
understand how L2 pragmatic competence is related to linguistic
competence, we must begin to explore the interlanguage of inter-
language pragmatics.
Revision version accepted 13 March 1999

Notes
1
Kasper and Dahl (1991) employed a narrow definition of interlanguage
pragmatics as speech act research for the specific purposes of the review
article. Kasper’s definition of pragmatics is much broader (e.g., Kasper,
1997b).
2
To give a perspective on the field at the time the article by Kasper and
Schmidt (1996) was written, I have included only work circulated before 1996.
Cross-sectional studies from 1996 include Houck and Gass (1996), Koike
(1996), Maeshiba, Yoshinaga, Kasper, and Ross (1996), and S. Takahashi
(1996), and more recently Hassall (1997), Hill (1997), Bardovi-Harlig and
Dörnyei (1998), and Rose (in press). Longitudinal studies include Cohen
(1997) and Kanagy and Igarashi (1997).
3
Six of the longitudinal studies cited earlier investigated beginning-level
learners and constitute an exception to this claim (Cohen, 1997; Ellis, 1992;
Kanagy & Igarashi, 1997; Sawyer, 1992; Schmidt, 1983; Schmidt & Frota,
1986). These studies are a small fraction of the acquisitional studies (cross-
sectional and longitudinal together) and an even smaller portion of interlan-
guage pragmatics research as a whole.
4
In this article I focus on speech act research in interlanguage pragmatics
where the investigation of the interface of grammar and pragmatics is likely
to be the most revealing. For a broader definition of pragmatics see Kasper
(1997b).
5
This is not the case with native-speaker production, as Blum-Kulka’s (1989)
crosslinguistic investigation of requests shows. Native speakers of Ar-
gentinian Spanish used both formal and informal morphology in their
requests in response to the variety of scenarios presented by the discourse
completion task. Using formal morphology exclusively in the judgment tasks
is both a simplification of the number of choices presented to the learners

Copyright © 2000. All rights reserved.

708 Language Learning Vol. 49, No. 4

and a reflection of the shopping scenario used by Walters and Rodríguez for
the judgment tasks.

References

Andersen, R. W. (1984). The one to one principle of interlanguage construction.

Language Learning, 34, 77–95.
Andersen, R. W., & Shirai, Y. (1996). The primacy of aspect in first and second
language acquisition: The pidgin-creole connection. In W. C. Ritchie &
T. K. Bhatia (Eds.), Handbook of second language acquisition (pp.
527–570). San Diego, CA: Academic Press.
Atkinson, J. M., & Heritage, J. (Eds.). (1984). Structures of social interaction:
Studies in conversational analysis. Cambridge: Cambridge University
Press.
Bardovi-Harlig, K. (1999a). Evaluating the empirical evidence: Grounds for
instruction in pragmatics? In K. Rose & G. Kasper (Eds.), Pragmatics in
language teaching. Manuscript submitted for publication.
Bardovi-Harlig, K. (1999b). Researching method. In L. F. Bouton (Ed.), Prag-
matics and language learning (Vol. 10, pp. 237–264). Urbana-Champaign:
University of Illinois, Division of English as an International Language.
Bardovi-Harlig, K., & Dörnyei, Z. (1998). Do language learners recognize
pragmatic violations? Pragmatic vs. grammatical awareness in instructed
L2 learning. TESOL Quarterly, 32, 233–259.
Bardovi-Harlig, K., & Hartford, B. S. (1990). Congruence in native and
nonnative conversations: Status balance in the academic advising session.
Language Learning, 40, 467–501.
Bardovi-Harlig, K., & Hartford, B. S. (1993). Learning the rules of academic
talk: A longitudinal study of pragmatic development. Studies in Second
Language Acquisition, 15, 279–304.
Bardovi-Harlig, K., & Hartford, B. S. (1996). Input in an institutional setting.
Studies in Second Language Acquisition, 17, 171–188.
Bardovi-Harlig, K., & Hartford, B. S. (1997, April). Academic advising sessions
with students in an intensive English program. Paper presented at the
Eleventh International Conference on Pragmatics and Language Learn-
ing, Urbana, IL.
Becker, A., & Carroll, M. (1997). The acquisition of spatial relations in a second
language. Amsterdam: John Benjamins.
Billmyer, K. (1990). “I really like your lifestyle”: ESL learners learning how
to compliment. Penn Working Papers in Educational Linguistics, 6, 31–48.

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 709

Blum-Kulka, S. (1982). Learning to say what you mean in a second language:

A study of speech act performance of learners of Hebrew as a second
language. Applied Linguistics, 3,29–59.
Blum-Kulka, S. (1989). Playing it safe: The role of conventionality in indirect
requests. In S. Blum-Kulka, J. House, & G. Kasper (Eds.), Cross-cultural
pragmatics (pp. 37–70). Norwood, NJ: Ablex.
Blum-Kulka, S., & Levenston, W. A. (1987). Lexico-grammatical pragmatic
indicators. Studies in Second Language Acquisition, 9, 155–170.
Blum-Kulka, S., & Olshtain, E. (1986). Too many words: Length of utterance
and pragmatic failure. Studies in Second Language Acquisition, 8,
165–180.
Bouton, L. F. (1992). The interpretation of implicature in English by NNS:
Does it come automatically—without being explicitly taught? In L. F.
Bouton & Y. Kachru (Eds.), Pragmatics and language learning (Vol. 3, pp.
53–65). Urbana-Champaign: University of Illinois, Division of English as
an International Language.
Bouton, L. F. (1994). Can NNS skill in interpreting implicatures in American
English be improved through explicit instruction? A pilot study. In L. F.
Bouton & Y. Kachru (Eds.), Pragmatics and language learning (Vol. 5, pp.
88–109). Urbana-Champaign: University of Illinois, Division of English as
an International Language.
Cohen, A. D. (1997). Developing pragmatic ability: Insights from accelerated
study of Japanese. In H. M. Cook, K. Hijirida, & M. Tahara (Eds.), New
trends and issues in teaching Japanese language and culture (Technical
Report #15; pp. 133–159). Honolulu: University of Hawai’i, Second Lan-
guage Teaching and Curriculum Center.
Cohen, A. D., & Olshtain, E. (1981). Developing a measure of sociolinguistic
competence: The case of apology. Language Learning, 31, 112–134.
Cohen, A. D., & Olshtain, E. (1993). The production of speech acts by EFL
learners. TESOL Quarterly, 27, 33–56.
Dietrich, R., Klein, W., & Noyau, C. (1995). The acquisition of temporality in
a second language. Amsterdam: John Benjamins.
Dittmar, N., & Terborg, H. (1991). Modality and second language learning. In
C. A. Ferguson & T. Huebner (Eds.), Crosscurrents in second language
acquisition and linguistic theories (pp. 347–384). Amsterdam: John
Benjamins.
Eisenstein, M., & Bodman, J. W. (1986). “I very appreciate”: Expressions of
gratitude by native and non-native speakers of American English. Applied
Linguistics, 7, 167–185.
Ellis, R. (1992). Learning to communicate in the classroom: A study of two
language learners’ requests. Studies in Second Language Acquisition, 14,
1–23.

Copyright © 2000. All rights reserved.

710 Language Learning Vol. 49, No. 4

Faerch, C., & Kasper, G. (1989). Internal and external modification in inter-
language request realization. In S. Blum-Kulka, J. House, & G. Kasper
(Eds.), Cross-cultural pragmatics (pp. 221–247). Norwood, NJ: Ablex.
Hassall, T. J. (1997). Requests by Australian learners of Indonesian. Unpub-
lished doctoral dissertation, Australian National University, Canberra.
Hill, T. (1997). The development of pragmatic competence in an EFL context.
Unpublished doctoral dissertation, Temple University Japan, Tokyo.
Hoffman-Hicks, S. (1999). A longitudinal study of L2 French pragmatic
competence: Study abroad vs. FL classroom learners. Unpublished doctoral
dissertation, Indiana University, Bloomington.
Houck, N., & Gass, S. M. (1996). Non-native refusals: A methodological study.
In S. M. Gass & J. Neu (Eds.), Speech acts across cultures: Challenge to
communication in a second language (pp. 45–64). Berlin: de Gruyter.
House, J. (1989). Politeness in English and German: The function of “please”
and “bitte.” In S. Blum-Kulka, J. House, & G. Kasper (Eds.), Cross-cultural
pragmatics (pp. 96–119). Norwood, NJ: Ablex.
House, J. (1996). Developing pragmatic fluency in English as a foreign
language: Routines and metapragmatic awareness. Studies in Second
Language Acquisition, 17, 225–252.
House, J., & Kasper, G. (1981). Politeness markers in English and German.
In F. Coulmas (Ed.), Conversational routine: Explorations in standardized
communication situations and prepatterned speech (pp. 157–185). The
Hague: Mouton.
House, J., & Kasper, G. (1987). Interlanguage pragmatics: Requesting in a
foreign language. In W. Lörscher & R. Schulze (Eds.), Perspectives on
language in performance: Festschrift for Werner Hüllen (pp. 1250–1288).
Tübingen, Germany: Narr.
Kanagy, R., & Igarashi, K. (1997). Acquisition of pragmatic competence in a
Japanese immersion kindergarten. In L. F. Bouton (Ed.), Pragmatics and
language learning (Vol. 8, pp. 243–265). Urbana-Champaign: University
of Illinois, Division of English as an International Language.
Kasper, G. (1984). Pragmatic comprehension in learner-native speaker dis-
course. Language Learning, 34, 1–20.
Kasper, G. (1992). Pragmatic transfer. Second Language Research, 8,
203–231.
Kasper, G. (1997a). Can pragmatic competence be taught? (Network #6;
HTML document). Honolulu: University of Hawai’i, Second Language
Teaching & Curriculum Center. Available: http://www.lll.hawaii
.edu/nflrc/NetWorks/NW6/
Kasper, G. (1997b). The role of pragmatics in language teacher education. In
K. Bardovi-Harlig & B. S. Hartford (Eds.), Beyond methods: Components
of language teacher education (pp. 113–136). New York: McGraw Hill.

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 711

Kasper, G., & Dahl, M. (1991). Research methods in interlanguage pragmat-

ics. Studies in Second Language Acquisition, 13, 215–247.
Kasper, G., & Schmidt, R. (1996). Developmental issues in interlanguage
pragmatics. Studies in Second Language Acquisition, 18, 149–169.
Kerekes, J. (1992). Development in nonnative speakers’ use and perceptions of
assertiveness and supportiveness in a mixed-sex conversation (Occasional
paper no. 21). Honolulu: University of Hawai’i at Manoa, Department of
English as a Second Language.
Koike, D. A. (1996). Transfer of pragmatic competence and suggestions in
Spanish. In S. M. Gass & J. Neu (Eds.), Speech acts across cultures:
Challenge to communication in a second language (pp. 257–281). Berlin:
de Gruyter.
Leary, A. (1994). Conversational replicas on computer for intermediate-level
Russian: A theoretical model of a design paradigm. Unpublished master’s
thesis, University of Iowa, Iowa City.
LoCastro, V. (1997). Pedagogical intervention and pragmatic competence
development. Applied Language Learning, 8, 75–109.
Maeshiba, N., Yoshinaga, N., Kasper, G., & Ross, S. (1996). Transfer and
proficiency in interlanguage apologizing. In S. M. Gass & J. Neu (Eds.),
Speech acts across cultures: Challenge to communication in a second
language (pp. 155–187). Berlin: de Gruyter.
Murphy, B., & Neu, J. (1996). My grade’s too low: The speech act set of
complaining. In S. M. Gass & J. Neu (Eds.), Speech acts across cultures:
Challenge to communication in a second language (pp. 191–216). Berlin:
de Gruyter.
Olshtain, E., & Blum-Kulka, S. (1985). Degree of approximation: Nonnative
reactions to native speech act behavior. In S. M. Gass & C. Madden (Eds.),
Input in second language acquisition (pp. 303–325). Rowley, MA: Newbury
House.
Olshtain, E., & Cohen, A. D. (1989). Speech act behavior across languages.
In H. Dechert & M. Raupach (Eds.), Transfer in language production
(pp. 53–67). Norwood, NJ: Ablex.
Omar, A. S. (1991). How learners greet in Kiswahili. In L. Bouton & Y. Kachru
(Eds.), Pragmatics and language learning (Vol. 2, pp. 59–73). Urbana-
Champaign: University of Illinois, Division of English as an International
Language.
Omar, A. S. (1992). Opening and closing conversations in Kiswahili: A study
of the performance of native speakers and learners. Unpublished doctoral
dissertation, Indiana University, Bloomington.
Pienemann, M. (1989). Is language teachable? Psycholinguistic experiments
and hypotheses. Applied Linguistics, 10, 52–79.

Copyright © 2000. All rights reserved.

712 Language Learning Vol. 49, No. 4

Pienemann, M. (1998). Language processing and second language develop-

ment: Processability theory. Amsterdam: John Benjamins.
Ritchie, W. C., & Bhatia, T. K. (Eds.). (1996). Handbook of second language
acquisition. San Diego, CA: Academic Press.
Robinson, M. A. (1992). Introspective methodology in interlanguage pragmat-
ics research. In G. Kasper (Ed.), Pragmatics of Japanese as a native and
target language (pp. 27–82). Honolulu: University of Hawaii Press.
Rodríguez, S. (1997). Appropriate requests and the card sorting judgment task.
Unpublished manuscript, Indiana University, Bloomington.
Rose, K. R. (in press). An exploratory cross-sectional study of interlanguage
pragmatic development. Studies in Second Language Acquisition.
Rose, K. R., & Kasper, G. (Eds.). (1999). Pragmatics in language teaching.
Manuscript submitted for publication.
Sawyer, M. (1992). The development of pragmatics in Japanese as a second
language: The sentence-final particle ne. In G. Kasper (Ed.), Pragmatics
of Japanese as a native and foreign language (Tech. rep. no. 3, pp. 83–125).
Honolulu: University of Hawai’i at Manoa, Second Language Teaching and
Curriculum Center.
Scarcella, R. (1979). On speaking politely in a second language. In C. A. Yorio,
K. Perkins, & J. Schachter (Eds.), On TESOL ’79 (pp. 275–287). Washing-
ton, DC: TESOL.
Schmidt, R. (1983). Interaction, acculturation, and the acquisition of commu-
nicative competence: A case study of an adult. In E. Judd & N. Wolfson
(Eds.), Sociolinguistics and language acquisition (pp. 137–174). Rowley,
MA: Newbury House.
Schmidt, R., & Frota, S. N. (1986). Developing basic conversational ability in
a second language: A case study of a learner of Portuguese. In R. Day (Ed.),
Talking to learn (pp. 237–326). Rowley, MA: Newbury House.
Siegal, M. (1994). Looking East: Learning Japanese as a second language in
Japan and the interaction of race, gender and social context. Unpublished
doctoral dissertation, University of California, Berkeley.
Svanes, B. (1992). En undersoekelse av realisasjonsmoenstret for spraakhan-
dlingen “aa be noen om aa gjoere noe” [An investigation of the realization
pattern of linguistic action “to ask someone to do something”]. Maal og
Minne, 1-2, 89–107.
Takahashi, S. (1996). Pragmatic transferability. Studies in Second Language
Acquisition, 18, 189–223.
Takahashi, S. (1998, March). Instructional effects on development of IL request
realization competence. Paper presented to the Third Pacific Second Lan-
guage Research Forum, Tokyo, Japan.
Takahashi, S., & DuFon, M. A. (1989). Cross-linguistic influence in indirect-
ness: The case of English directives performed by native Japanese speakers.

Copyright © 2000. All rights reserved.

 Bardovi-Harlig 713

Unpublished manuscript, University of Hawai’i at Manoa, Honolulu.

(ERIC Document Reproduction Service No. ED 370 439)
Takahashi, T., & Beebe, L. (1987). The development of pragmatic competence
by Japanese learners of English. JALT Journal, 8, 131–155.
Trosborg, A. (1987). Apology strategies in natives/non-natives. Journal of
Pragmatics, 11, 147–167.
Trosborg, A. (1995). Interlanguage pragmatics: Requests, complaints and
apologies. Berlin: Mouton de Gruyter.
Walters, J. (1979). The perception of politeness in English and Spanish. In
C. A. Yorio, K. Perkins, & J. Schachter (Eds.), On TESOL ‘79 (pp. 288–296).
Washington, DC: TESOL.
White, L. (1989). Universal grammar and second language acquisition. Am-
sterdam: John Benjamins.
Wildner-Bassett, M. (1984). Improving pragmatics of learners’ interlanguage.
Tübingen, Germany: Narr.
Wildner-Bassett, M. E. (1994). Intercultural pragmatics and proficiency:
“Polite” noises for cultural appropriateness. IRAL, 22, 3–17.

Copyright © 2000. All rights reserved.