Introduction
In the past two decades, a substantial body
of empirical research in interlanguage
pragmatics has described how speech acts
performed by non-native speakers differ
from the target language norms. These
studies have focused on either the
production or comprehension of speech acts
such as requests, refusals, apologies, and
compliments. Compared to other studies of
second-language acquisition (SLA), which
have examined variation among individuals
with respect to L2 language learning for
quite some time, most ILP studies to date
have been limited to finding how L2
learners perform a particular speech act, and
there has been relatively little inquiry into
how they acquire L2 pragmatics and which
factors affect learners’ acquisition of L2
pragmatics (Bardovi-Harlig, 1999; Kasper &
Rose, 2002). Thus, ILP researchers have
argued for additional inquiry into the
variables that are known to potentially affect
learners’ pragmatic development.
Researchers have found that “high levels of
proficiency do not guarantee concomitantly
high levels of pragmatic competence”
(Bardovi-Harlig, 1999, p. 686) and that
other variables like length of stay in the
target community, quality and quantity of
input, and level of interaction should be
taken into account when assessing L2
learners’ performance. The role of length of
residence in the target community, quality
and amount of input and level of motivation
on learners’ pragmatic development are
important issues to consider (Bardovi-Harlig
1999; Kasper & Rose, 2002). Furthermore,
the inconsistency of research findings
regarding the impact that individual
differences such as length of residence and
motivation might have on learners’
pragmatic development requires more
research (Churchill & Dufon 2006; Félix-Brasdefer 2004; Kasper & Rose 2002).
A second-language context supports the
acquisition of pragmatic issues as learners
encounter more opportunities to use the
language, and are generally exposed to the
L2 more intensively (Eslami-Rasekh, 2005;
Segalowitz & Freed, 2004). However, the
assumption that that living abroad provides
an ideal context for language learning has
been questioned by other researchers
(Yager, 1998). The factors that each
individual brings to the learning context are
both crucial and complex. Learners differ in
terms of how ready they are linguistically
and cognitively to seize and benefit from the
opportunities provided for language learning
(Segalowitz & Freed, 2004; DeKeyser,
1991). This study, thus, considers how
different variables, such as motivation, the
amount of contact with English, and type of
contact, affect Korean graduate students’
English pragmatic skills. By examining the
effects of these factors on L2 pragmatic
competence, we hope to gain more
knowledge about the dynamic interactions
between learners’ pragmatic development
and individual variables, creating a better
understanding of the potential influence of
these variables in L2 pragmatic acquisition
success.
In what follows we will first present the
theoretical framework of the study.
Following that the methodology of the study
will be presented. The results of the study
are presented next, and at the end the
findings are discussed and conclusions
provided.
Theoretical framework
The role of individual variables in the
development of second language pragmatics
Researching individual differences (ID) in
language learning has a long tradition in
SLA. However, the role of ID in the
acquisition of L2 pragmatics has rarely been
addressed. Existing research in SLA has
investigated how the social, psychological,
cognitive, and personal dimensions of L2
learning impact how much and how quickly
the individual will learn an L2 (Collentine &
Freed, 2004).
The first dimension falls under the heading
of social factors. The basic theory is that the
language learners’ emotional and social
attachment to the target language culture has
a positive effect on the amount of language
learned. Cross-cultural adjustment and
acculturation have been cited as particularly
important in determining how much
language will be learned. Additionally,
attitudes (Schumann, 1986) and intended
length of residence in the target language
area are other crucial factors. A language
learner who intends to remain in the target
language area for a long time is more likely
to develop extensive contact with the target
language members, promoting L2 learning
(Schumann, 1986). L language development
differences is also attributed to
psychological factors and includes variables
such as language shock, culture shock,
culture stress, integrative or instrumental
motivation, and ego-permeability. The third
group refers to cognitive factors and
includes the learner’s language aptitude,
intelligence, and differing attention levels.
Learners may differ in where they direct or
orient their attention to the input they
receive, as well as the output they produce,
and these differences may play a crucial role
in developmental outcomes in language
learning (Skehan & Foster, 2001).
The final set of individual variables is
related to age, gender, anxiety, self-esteem,
tolerance of ambiguity, language learning
styles, and language learning strategies
(Larsen-Freeman, 2001). Research suggests
that no single variable can account for the
rate and success of language acquisition.
Nevertheless, the study of SLA within and
across various contexts of learning would
lead to a broadening of our perspective
concerning the most important variables that
affect and impede L2 acquisition.
Length of residence
Length of residence is construed as one of
the ID variables that affect learners’
different developmental stages of L2
pragmatics. Many studies have used length
of stay in a target speech community as an
indicator of L2 pragmatic acquisition (Han,
2005). Researchers argue that language
learners living in a target speech community
have many opportunities to interact in the
L2, which leads to the learners’ successful
acquisition of pragmatic competence.
Blum-Kulka and Olshtain (1986) found a
relationship between length of stay in the
target speech community and the target-like
perception of directness and politeness in an
L2. Olshtain and Blum-Kulka’s study
(1985) also showed that the amount of
external modification used by L2 learners
approximated community pragmatic norms
after five to seven years of stay in the target
language environment, and that such
convergence correlated positively with
duration of stay. Takahashi and Beebe
(1987) compared Japanese EFL and ESL
learners’ production of refusals and found
that the ESL learners’ refusals were more
target-like. House (1996) found that
learners who had stayed in English-speaking
countries consistently performed better than
their peers who had not, both before and
after instruction. Röver (1996) found that
German EFL students who had spent as little
as six weeks in English-speaking countries
outperformed learners who did not in the use
of pragmatic routines. Bouton (1999)
investigated how length of residence affects
non-native speakers’ understanding of
implicature in American English. Similarly,
Churchill (2001) recorded a decrease in
direct want statements in the English request
realizations of his JFL learners over a month
in the target language context. Overall,
these studies suggest that longer residence in
the target language community yield greater
L2 pragmatic attainments.
Contrary to what these studies claim,
however, some researchers argue that length
of residence in the target country has not
been identified as a good predictor of L2
attainment and is not sufficient in the
achievement of increased proficiency in L2.
Kondo (1997) examined Japanese EFL
learners’ apology performance before and
after one year of home stay in the United
States, and compared them with L1 speakers
of Japanese and American English. In some
respects, the students’ apologies became
more target-like, but in others they did not.
In a more recent study, Rodriguez (2001)
investigated the effect of a semester
studying in a target-language community by
examining students’ request strategies. The
findings of the study showed no advantage
at all for the study-abroad students. Roever
(2001) also observed that neither learners’
comprehension of implicatures nor
performance of speech acts in English
benefited from the learners’ time abroad. It
is possible that, much like how children
acquire L1 through continuous interaction
with adults and peers, L2 learners may need
to be involved in intensive interaction with
native speakers and fully embrace the L2
culture in order to achieve native-like
pragmatic skills in the L2 (Ninio & Snow,
1996).
Although the studies mentioned above
provide evidence of the relation between
pragmatic development and learners’ length
of residence in the target language
community, one might wonder to what
extent pragmatic ability is influenced by the
intensity of learners’ exposure to the target
language, as opposed to the quantitative
measure of length of residence in the target
language community. Related to this, from
their longitudinal study of learners’
acquisition of temporality, Klein, Dietrich,
and Noyau (1995) concluded that what
matters is intensity, not length, of
interaction. Similarly, Matsumura (2003)
asserted that acquisition of pragmatic
competence is not associated with the length
of stay, because learners vary individually in
the amount of interaction in an L2 as well as
opportunities to interact in the target culture.
Thus, intensity of interaction may account
for more of the learning process than
duration of stay in the L2 speech
community.
Kasper and Rose (2002), have raised
concerns as to whether pragmatic ability is
influenced by the quality of nonnative
speakers’ exposure and social contacts or the
quantitative measure of length of residence.
These researchers consider intensity of
interaction to be the important factor rather
than the length of residence. For example,
Bella’s (2011) study on invitation refusals
by L2 learners of Greek revealed that
opportunities for interaction are much more
critical than length of residence in the target
community for the development of learners’
pragmatic competence. Bella’s (2012) study
revealed similar results in relation to request
modification strategies. These findings
suggest that the impact of length of
residence in the target community and
intensity of interaction with native speakers
on pragmatic development remains an open
question which is worth exploring further.
As suggested by Félix-Brasdefer (2004), the
results of studies dealing with the effects of
length of residence on pragmatic ability
should be viewed with caution due to the
variation research findings present regarding
both the pragmatic measure used
(comprehension, production, etc.) and the
time span proposed for pragmatic
development to take place.
Amount of interaction
Seliger’s 1977 study of the role of
interaction patterns of ESL students
provides empirical support that target
language use is essential in second-
language acquisition. Seliger claimed that
the more learners seek out opportunities to
use the target language and interact
intensively with native speakers, the more
competent they become. Stern (1983) also
believed that committed language learners
“seek communicative contact with target
language community members and become
actively involved as participants in authentic
language use” (p. 411). Pica (1996) and
Ellis (1994) also offered evidence to validate
the positive correlations between interaction
in the target language and success in
language learning. Learners acquire
comprehensible input through target
language interactions that provide input on
how to successfully use the language, enact
speech acts, and carry out redressive action
(LoCastro, 2003). Marriot (1995) study
examined the acquisition of sociolinguistic
competence by Australian secondary
students who participated in exchange
programs in Japan. She observed how
learners benefit more from “self- and other-correction” procedures in interactive
situations in a Japanese homestay context.
Cooperative interactants who surrounded the
learners contributed significantly to the
development of these learners’ L2 pragmatic
awareness.
Likewise, Edmondson and House (1991)
suggested that exposure to proper pragmatic
input in the target language does have a
beneficial effect on the development of
pragmatic competence. Kasper (1998) noted
that “sustained contact with the target
language and culture may be required to
attain native pragmatic knowledge and skill”
(p. 200). Resonating this, Wray (1999)
proposed that interactions with native
speakers helps language learners obtain the
pragmatic rules of use in the target language.
Additionally, in a study on Japanese ESL
learners’ perception of appropriateness in
advice situations, Matsumura (2003) found
that the amount of exposure to the target
language was a significant factor predicting
learners’ pragmatic ability.
Motivation
Although there are a number of studies in
SLA that suggest motivation is one of the
variables that provide the primary impetus to
initiate L2 learning, and the driving force to
sustain the long-term learning process, there
is a relative dearth of data that specifically
focus on a possible link between motivation
and L2 learners’ pragmatic competence.
Additionally, depending on the domain of
language to be examined, motivation has
been found to have more or less effect. Au
(1998) pointed out that a number of studies
have revealed zero or even negative
relationships between motivation and L2
proficiency (Clement, Gardner, & Smythe,
1980; Kasper & Schmidt, 1996). The
importance of motivation in interlanguage
pragmatics was raised as one of twelve basic
questions by Kasper and Schmidt (1996).
Niezgoda and Rover (2001) showed that
environment may not be the only factor
influencing the development of pragmatic
competence and affective variables may also
play an important role in learners’ L2
pragmatic acquisition. Schmidt (1993)
observed that “those who are concerned with
establishing relationships with target
language speakers are more likely to pay
close attention to the pragmatic aspects of
input and to struggle to understand than
those who are not so motivated”(p. 36).
The first systematic studies to examine the
effects of motivation on L2 pragmatics were
by Takahashi, 2001 and 2005. Takahashi
(2001) speculated that motivation could be
one of the most influential individual
variables influencing differences in learners’
noticing of target request forms. The study
shows that highly motivated learners
willingly adopt target standards for
pragmatic action, whereas less-motivated
learners are more likely to resist accepting
target norms. Takahashi argued that
learners’ personal values may influence how
much effort they expend on understanding
L2 pragmatics and sociolinguistic practices
and how much of a positive affect they have
toward a target-language community.
Evidence from research studies indicates
that availability of input through
interlocutors or models is a necessary
condition for development of pragmatic
competence. However, learner-internal
factors may control the conversion of input
to intake and consequently hinder or boost
the development of pragmatic knowledge.
Accordingly, the present study examines the
role of motivation in interlanguage
pragmatics.
Considering the importance of length of
residence, amount of interaction and
motivation in second language acquisition, it
is worthwhile to examine whether these
three variables play a role in the pragmatic
competence of ESL learners. The study
focuses on the performance of compliments
and compliment responses by Korean ESL
learners.
Compliments
Compliments are one of the frequently used
speech acts in everyday encounters, yet they
are intricate and could be challenging for L2
learners. They are studied in different
languages and compared across languages
and cultures (e.g., Golato, 2005; Lorenzo-Dus, 2001; Maíz-Arévalo, 2012, Manes,
1983; Wolfson & Manes, 1981). One of the
earliest studies is Wolfson and Manes
(1981) empirical and descriptive work on
compliments in American English. Wolfson
and Manes (1981) argued that compliments
in American English are highly patterned,
with a very restricted set of syntax and
lexicon. Wolfson and Manes (1981) also
found that the most frequent topics of
compliments fall into two major categories:
those having to do with
appearance/possessions, and those
addressing ability/performance. Under the
category of appearance/possessions,
compliments tend to be on clothing and
other personal features such as hairstyles
and on possessions such as cars and
household items.
Complimenting can be treated as a social
strategy employed to start or maintain
solidarity in mundane interactions between
colleagues, neighbors, or close friends.
Holmes (1988) essentially agreed with this
view by treating compliments as “positively
affective speech acts directed to the
addressee that serve to increase or
consolidate the solidarity between the
speaker and addressee” (p. 486). According
to Herbert (1989), compliments establish
solidarity with the listener by praising some
feature relevant to that listener, of which the
listener approves. Compliments serve many
other social functions as well. Under certain
conditions, compliments replace speech acts
such as apologies, thanking, and greetings.
Compliments can also be used to soften the
effects of criticism or other face-threatening
acts such as requests (Billmyer, 1990). As
Wolfson (1983) suggested, compliments
may even be used as sarcasm (e.g., “You
play a good game of tennis — for a
woman”) (pp. 86-93).
Compliment responses
Compliments trigger a number of response
options for the addressee (Holmes, 1995;
Maíz-Arévalo, 2012; Pomerantz, 1978).
One early study focusing specifically on
compliment responses is Pomerantz’s (1978)
descriptive analysis of compliment
responses in American English. Based on
her data, Pomerantz posited that
agreement/acceptance and
disagreement/rejection were the
predominant compliment response type in
American English.
Gracefully accepting compliments without
seeming to praise oneself can result in a
dilemma for the recipient of the compliment
(Herbert, 1986). Manes (1983) also
recognized the dilemma posed to receivers
of compliments and offered a set of
strategies which enable speakers to both
accept but not necessarily agree with the
compliment.
As Herbert (1990) pointed out, “thank you”
is considered the most appropriate response
to a compliment in the United States. While
this response is appropriate in most
situations, researchers have stated that “an
unadorned ‘thanks’ may unintentionally
limit or even end an interaction between
status equals, and deflecting compliments
may serve to extend the interaction between
interlocutors, which may lead to
interlanguage development” (Billmyer,
Jakar, & Lee, 1989, p. 17). Wolfson (1989)
agreed stating that a native speaker of
English may strategically use compliments
to open and to lengthen the conversation.
Using a simple “thanks” then may
inadvertently result in the opposite outcome
by limiting opportunities to extend the
interaction. As a result, interaction
opportunities for the nonnative speakers
may be hindered (Wolfson, 1989). Being
able to compliment others and to respond to
compliments effectively will enhance
interaction possibilities for the learners and
therefore, should promote their pragmatic
development.
Purpose of the study
This study examined the Korean ESL
learners’ level of approximation to native
speakers’ use of giving compliments and
responding to compliments, and the effect of
the three research variables (motivation to
learn English, the amount of interaction in
English, and length of residence in the
target-language area) on the pragmatic
competence level attained. The following
research questions were addressed:
1) How do differences in the Korean
ESL learners’ degree of motivation
correlate with their achievement of
pragmatic competence?
2) How does the amount of interaction
in English contribute to the
differences in the Korean ESL
learners’ pragmatic competence?
3) How do differences in the Korean
ESL learners’ length of residence
contribute to the differences in the
Korean ESL learners’ pragmatic
competence?
Methodology
Participants
The participants of the study were 50
Korean graduate students majoring in
different academic fields at Texas A&M
University in the United States. The length
of time the participants had spent in the
United States ranged from two years to eight
years. The participants were recruited from
various Korean communities (e.g., Korean
students’ association, Korean churches).
Instrumentation
The data for the present study were collected
using three types of elicitation instruments: a
written background questionnaire, a
discourse completion test, and the mini-Attitude/Motivation Test Battery.
Background Information Questionnaire
The researchers used the background
questionnaire to identify the amount of
interaction in English the participants
experienced in their daily encounters and
their length of residence in the United
States. The questionnaire elicited
information on the total amount of time
participants used English during a typical
week, both inside and outside the classroom
(e.g., the time spent speaking English,
watching television or listening to the radio,
reading books in English, and writing
email), and the number of years spent in the
United States.
Discourse Completion Test
Data for examining pragmatic competence
of Korean ESL learners in the speech acts of
compliment and compliment responses were
collected via a written DCT. Social
variables of power and distance were
considered in designing the DCT situations
and only complimenting scenarios assumed
by the researchers to be experienced by the
participants in their daily living in the L2
community were used for the study.
Social distance was kept constant in all
situations (only acquaintances), since
research has indicated that the great majority
of compliments occur between interlocutors
who are friends or acquaintances, rather than
strangers (e.g., Manes, 1983; Wolfson,
1981, 1989).
The Mini-Attitude/Motivation Test Battery
The mini-attitude/motivation test battery
(mini-AMTB) was used to measure the
participants’ degree of motivation to learn
English. Developed by Gardner (1985), the
AMTB is the most frequently used
assessment tool to measure students’
attitudes and motivation to learn another
language, and to assess various individual
difference variables based on the socio-educational model. The mini-AMTB is
made up of 11 items that fall into five
dimensions of motivational constructs:
integrativeness (items 1-3), attitudes toward
learning (items 4 and 5), motivation (items
6-8), instrumental orientation (item 9), and
language anxiety (items 10 and 11). The
mini-AMTB uses a seven-point interval
scale anchored at the end points, with the
mid-point as neutral. The mini-AMTB has
recently been used in many studies of L2
motivation (e.g., Baker & Macintyre, 2000),
because it reduces administration time while
measuring the basic constructs of the
original AMTB.
Data collection procedures
A pilot study was conducted prior to the
main study to determine the practical
feasibility of the inquiry and to ensure
clarity of the questionnaire and the discourse
completion test. In the main study,
participants first signed a consent form
confirming their willingness to participate.
The researchers provided the participants
with detailed instructions about the tasks in
their L1. The instruments were
administered individually and each
participant was asked to complete the
written open DCT first and then the
background information and motivation
questionnaire.
Data analyses
A statistical analysis of the data was carried
out using version 14.0 of the Statistical
Package for the Social Sciences. Tests for
normality of variables, multicollinearity
among variables, and interrater reliability
were taken to prevent against validity issues
and to improve the reliability of the
quantitative analyses. Descriptive statistics
were used and the means for level of
pragmatic competence, amount of
interaction in English, length of residence in
the target environment and motivation were
converted to standardized scores (z scores)
for each participant. The standardized data
were then analyzed by performing a Pearson
product-moment correlation and multiple
regression (α = .05).
Pearson’s r correlation coefficients were
carried out to examine if there is a
statistically significant correlation among
three independent variables (motivation for
learning English, the amount of interaction
in English, and length of residence in the L2
community) and Korean ESL learners’ L2
pragmatic competence.
Following bivariate (correlational)
relationship analysis, multiple regression
analysis was performed to determine the
joint effects of all independent variables on
the dependent variable. A multiple
regression analysis was conducted to
determine if the findings in correlation
coefficient analysis are upheld by the
multiple regression analysis.
Results
This study aimed to account for the different
levels of pragmatic development among
fifty graduate-level Korean ESL learners
and whether the learners’ pragmatic ability
was influenced by motivation levels for
learning English, the amount of interaction
in English, and length of their residence in
the target-language community.
First, univariate descriptive statistics were
conducted to obtain mean, standard
deviation, skewness and kurtosis of the raw
data for each observed variable. Table 2
displays a summary of univariate descriptive
statistics for the three observed variables.
The mean level in the DCT representing the
L2 pragmatic competence of the Korean
ESL learners when performing
complimenting behavior indicated that the
Korean ESL subjects attained a relatively
high level of English pragmatic competence.
A mean of 1.85 suggests that the Korean
ESL learners’ DCT rating is close to the
“acceptable” category, which means that
their dialogues contained small errors with
respect to pragmatic norms. The total
amount of time the participants spent
interacting in English each week had a mean
of 32.04. The seven-point scale to
determine subjects’ level of motivation for
learning English had a mean of 4.67, which
implies that the participants had a favorable
attitude toward learning English.
Measures of skewness and kurtosis were
examined to ensure that the data of
individual variables represented a normal
distribution. As seen in Table 2, the
skewness and kurtosis values of the three
variables all lie between ± 1.0, which means
that all three variables fall within the
“excellent” range as acceptable variables for
further analyses (George and Mallery,
2001).
The Pearson Product Moment Correlation
was used to examine the degree of
consistency in the two independent raters
who scored the participants’ DCT scores.
There was a correlation coefficient of .61
using Eisenstein and Bodman’s (1993)
nativeness rating scale for assessing the
participants’ pragmatic competence.
Meanwhile, multicollinearity was found to
pose no threat to the reliability of the
subsequent regression analyses.
DCT rating results
Results of the DCT rating showed that 71
(35.5 percent) of the dialogues achieved a
rating of 1 (native-like). A large number, 92
(46 percent), obtained a rating of 2
(acceptable) and contained small errors that
did not affect understanding or
appropriateness. Thirty-one dialogues (15.5
percent) received a rating of 3 (problematic)
which meant that they contained errors that
might cause misunderstandings. There were
6 dialogues (3 percent) that were rated as 4
(not acceptable) meaning that they were
difficult to comprehend and/or there were
instances of a violation of a social norm.
The analysis of factors that contribute to
success in achieving L2 pragmatics were
performed using DCT scores as the criterion
measure of learners’ pragmatic skills.
Correlation coefficient analysis was
performed to investigate the relationship
among the four variables of interest.
Research Question One
The first research question examined to what
extent learners’ pragmatic competence is
related to their degree of motivation. To
examine this relationship, a Pearson
product-moment correlation analysis with
alpha set at .05 was performed. The analysis
indicated a significant and strong
relationship between DCT scores and the
level of motivation (r = -.305, p = .031).
Next, we examined which subcomponents of
motivation are correlated with pragmatic
competence. Descriptive statistics (table 3)
show that the measures of skewness and
kurtosis of the five motivation subscales
were within acceptable levels and consistent
with a relatively normal distribution. Thus
we followed with the correlation analysis.
As shown in table 4, motivation subscale
showed the highest correlation (r = -.287, p
= .043) with participants’ DCT scores,
followed by language anxiety with the
second highest correlation (r = -.245, p
=.086). Both constructs show moderate,
statistically significant correlations. The
positive relationship between anxiety and
pragmatic competence was rather
unexpected because previous studies in
general have found a negative relationship
between anxiety and L2 achievement
(Gardner, Day, & MacIntyre, 1992).
However, some researchers have found
experimental evidence that anxiety could be
beneficial in language learning (Brown,
Robson, & Rosenkjar, 2001).
The other subscales on motivation survey
(integrative and instrumental orientation,
attitude toward learning situation) and the
pragmatic competence did not reveal any
statistically significant relationships.
Research Question Two
The second research question examined
whether learners’ pragmatic ability is related
to amount of interaction in the target
language. A Pearson product-moment
correlation matrix was used to examine the
relationship between the amount of
interaction in English and the students’ level
of L2 pragmatic ability.
Contrary to what was expected, the
correlation coefficient for amount of
interaction was not statistically significant (r
= -.194, p = .177). Research has shown that
the type of interaction, rather than the
amount of interaction, is instrumental in
developing pragmatic ability (Parr, 1988;
Freed, 1990; Ward & Rana-Deuba, 2000).
Thus, a Pearson product-moment correlation
matrix between separate types of interaction
in the target language and the learners’
pragmatic competence was performed.
Descriptive statistics in Table 5 show mean,
standard deviation, skewness and kurtosis of
the raw data for the four types of interaction.
A review of the summary statistics showed
an abnormal distribution for two of the
subcomponents of amount of interaction
variable (speaking and writing). Thus, a
data transformation on the variables
(speaking and writing) which did not show
normal distribution was executed.
Table 6 presents the correlation between the
participants’ DCT performance and the three
interaction subfactors.
The correlation coefficients between the
different types of interaction and the
pragmatic competence were small and a
salient relationship was identified only
between time spent reading books and the
DCT scores (r= -.315, p = .026).
Research Question Three
The third research question examined to
what extent achievement of L2 pragmatic
competence is related to the length of
residence in the second language
community. The correlation analysis
showed that the relationship between the
two variables was in the desired direction;
that is, longer length of residence was more
likely to lead to better outcomes in L2
pragmatics. However, the correlation
coefficient (r= -.141, p = .329) was not
significant.
Regression analysis
The third phase of our analysis consisted of
multivariate statistical analyses. When
examined individually, the regression model
of the effect of motivation on pragmatic
competence was significant and about 10
percent of the variation in the dependent
variable (R2
=.093) was accounted for by
motivation variable. However, the model
including either amount of interaction or
length of residence as the independent
variable failed to demonstrate the powerful
relationship between these variables and the
dependent variable.
Multiple regression modeling was then used
to analyze the overall contribution of each
independent variable with the influence of
other independent variables controlled for,
evaluating the contribution of total
independent variables to the total explained
variation in the dependent variable. The aim
was to examine two questions: Was it
possible that students’ L2 pragmatic
achievement was best predicted as a
combination of all three predictor variables
of motivation, amount of interaction, and
length of residence? Or did a single
predictor variable yield greater
predictability? To answer these questions, a
series of multiple regressions were
performed by first entering two predictor
variables (amount of interaction and length
of residence) after controlling for the
strongest predictor identified based on the
correlation analyses (motivation).
Inspection of the squared multiple
correlations (R2) suggests that overall, 7.5
percent of the variance related to
participants’ L2 pragmatic competence was
explained by two variables (amount of
interaction and length of residence). Based
on Cohen (1988), this effect size is
considered to be small and not significant (F
(2, 47) = 1,899, p = 0.161).
Next, in order to explore the presence of
possible relationships between predictors
and outcomes, all three independent
variables (motivation, amount of interaction,
and length of residence) were added to the
model, and changes in the values and
direction of parameter estimates as well as
changes in the significance and the size of
the R
2
were recorded. When motivation
predictor was added to the model, the value
of R2 did change substantially (from R2=.075 to R
2=.154). Inspection of the squared
multiple correlations (R2) suggests that
moderate and statistically significant
relationships were found among these
predictors, F (3, 46) = 2,802, p = 0.050. The
three independent variables explained about
16 percent of the variance.
To explain the degree to which the
independent variables (motivation, amount
of interaction, and length of residence)
affect the L2 pragmatic achievement of the
learners, the weight of their respective
standardized regression coefficient, or beta
(β), was calculated for each predictor
variable. The predictor variable of
motivation yielded a beta of -.286 and a t
value of -2.083 resulting in a significant
relationship (p = .043) while the predictor
variable of the amount of interaction and
length of residence yielded a beta of -.197/-.206 and a t of -1.395/-1.477 resulting in a
non-significant relationship (p = .170/.147),
respectively.
Findings from multivariate regression
analysis are consistent with those obtained
through examination of simple correlations,
and suggest that among all predictors
considered in the present study motivation
was the main predictor of the criterion
variable (pragmatic competence).
Discussion and conclusions
This study was undertaken in an attempt to
account for Korean ESL learners’ pragmatic
competence, as functions of their motivation
levels for learning English, amount of
interaction in English, and length of
residence in the target-language community.
Pearson’s r correlation coefficients were
calculated to assess which variable was the
better predictor of participants’ pragmatic
competence. The correlation between
pragmatic competence and motivation was
moderately significant; contrary to what was
expected, however, the correlation
coefficient for the amount of interaction and
length of residence was relatively low.
The results support other research findings
that indicate motivation as an important
factor in second-language pragmatic
acquisition (e.g., Cook, 2001; Niezgoda
&Rover, 2001; Schmidt, 1993; Takahashi,
2001, 2005). With respect to this finding,
data was further examined to check the
extent to which the subcomponents of
motivation were related to the participants’
L2 pragmatic competence. The results show
that learners’ pragmatic competence is
associated with some motivational factors
but not with all motivation subscales. In
particular, the learners’ motivational
intensity was found to be closely related to
their pragmatic competence. The one
exception to this finding was a positive
relationship between language anxiety and
the participants’ pragmatic performance.
This finding is incongruent with other
research in which language anxiety has been
shown to correlate negatively with language
achievement (Gardner & Macintyre, 1993;
Horwitz, 2001). Our results are not
unexpected given that some studies indicate
what would typically be labeled as
detrimental anxiety could be sometimes
facilitative for language learning (Brown,
Robson, & Rosenkjar, 2001).
Our study indicates that the relationship
between amount of interaction and
pragmatics competence was weak and non-significant. The findings of the present study
are inconsistent with the findings of some
previous studies, which found a statistically
significant relationship between interaction
and students’ pragmatic abilities (Bacon,
2002; Hashimoto, 1993; Lapkin, Hart, &
Swain, 1995). Our findings, however, agree
with some other studies indicating that
informal contact does not necessarily result
in pragmatic development (e.g., Bouton,
1994; Masumura, 2003). The fact that
increased opportunities to interact in the L2
did not necessarily result in L2 pragmatic
achievement is an important issue for future
research. Possible explanations might be
that the amount of interaction itself was
insufficient and thus failed to contribute to
increases of learners’ pragmatic knowledge.
Lapkin, Hart, and Swain(1995) suggested
that many factors affect how informal
contact relates to acquisition, including the
type and quality of informal contact, and
individual differences, such as students’
second-language level, language experience,
learning style, attitude toward the host
culture, awareness of cultural differences in
language use, and willingness to
accommodate to pragmatic norms in a L2
and motivation. Related to this suggestion,
Siegal (1994) and LoCastro (1998) focused
on learners’ pragmatic development in
relation to their subjectivity and agency.
McKay and Wong (1996) argued that we
should study L2 learners’ subjectivity when
we examine their L2 use and development.
As DuFon (1999) asserted, little is known
about how individual learners take
advantage of opportunities to interact, and
what factors influence their willingness and
ability to do so.
Additionally, Schmidt (1993) argued:
Simple exposure to appropriate input is
unlikely to be sufficient for acquisition of L2
pragmatic knowledge because the specific
linguistic realizations are sometimes opaque
to learners and the relevant contextual
factors to be noticed may be defined
differently or may not be salient enough for
the learner (p.36).
Also, Kasper (1998) proposed that while
authentic L2 input is essential for pragmatic
learning, it does not secure successful
pragmatic development.
Another major finding of the study was the
lack of correlation between the learners’
pragmatic competence with their length of
residence in L2 community. This is contrary
to the findings of studies which revealed the
positive effects of length of residence on
pragmatic competence (Churchill, 2001;
House, 1996; Kuriseak, 2006). While these
studies claim that a lengthy residence in the
target-language area would tend to promote
second-language learning, many questions
remain about the validity of that assumption.
Regarding this, the result in the present
study supports findings from other studies
that show length of residence may have a
negligible effect on the eventual attainment
of pragmatic skills in English (Bouton,
1994; Kondo, 1997; Roever, 2001;
Rodriguez, 2001).
One might expect that students living for an
extended time in the target-language
community take advantage of the many
opportunities to interact in the L2 and, in
turn, would have shown greater achievement
in the target language. However, additional
variables that influence language learning
success have been incorporated need to be
taken into consideration. For example, it is
possible that people with greater interest in
long-term stay (e.g., U.S. permanent
residents and naturalized citizens) would
display a greater willingness to relinquish
aspects of their native culture and
acculturation into the host country.
International students, on the other hand,
might be more committed to maintaining
their cultural heritage, and therefore show
lower acculturation to life in the United
States. Clearly, there is a need for further
investigation into the relationship between
pragmatic competence and the experiences
that students have during their stay in the
target community, which are greatly affected
by the myriad of factors that are experienced
differently by each learner.
Additionally, as noted by Ward and Rana-Deuba (2000), we do not know whether it is
the quality or quantity of informal
interaction that is of primary importance in
language learning. Thus, it may not be the
amount but rather the type of interaction that
most affects the level of participants’
pragmatic ability. Our results demonstrate
that the time subjects reported reading
books, magazines, or English-language
newspapers was a significant predictor of
the criterion measure. These results agree
with Freed’s (1990) study that show
interactive contact with native speakers did
not predict changes for students at the high
intermediate and advanced levels. Freed
concluded that perhaps the amount of
interaction in L2 has less influence on
advanced students’ L2 pragmatic
achievement. There is a need for future
studies to explore the possibility that
different types of activities interact in
different ways with the process of language
learning at different stages in the acquisition
process. Learners vary in terms of how
linguistically and cognitively ready they are
to seize opportunities and to benefit from
them once they do. This study documents
examples of these complex interactions. It
remains for future studies to identify
additional variables that influence learners’
pragmatic acquisition. Such interactions
may help explain the enormous individual
variation one sees in learning outcomes and
underscore the importance of studying such
variables together rather than in isolation.
The study has limitations due to the data
collection methods (Eslami & Mirzaei, in-press). We used self-reported data to
measure the amount and type of contact and
pragmatic performance of the learners.
Future studies should also use qualitative
information such as daily diary, interviews,
and collect more detailed information about
type and context of interactions.
Furthermore, we only focused on the speech
act of complimenting and compliment
response. Additional research is needed to
further examine the effect of motivation,
interaction, and length of residence using
different pragmatic measures, in other
speech acts, and with different groups of
learners.
In conclusion, our findings show that simple
exposure to language is unlikely to be
sufficient for acquisition of L2 pragmatic
knowledge because the specific linguistic
realizations are sometimes not salient
enough for the learner. For L2 pragmatics
to develop, input should be noticed and
some explicit techniques such as input
enhancement and form focused instruction
that would make the learners attend to the
targeted linguistic features are necessary.
