Journal of Pragmatics 43 (2011) 3745–3756

Contents lists available at SciVerse ScienceDirect

Journal of Pragmatics
journal homepage: www.elsevier.com/locate/pragma

Investigating acquisition of discourse markers through a developmental

learner corpus
Brittany Polat *
Georgia State University, 34 Peachtree Street, Suite 1200, Atlanta, GA 30303, United States

A R T I C L E I N F O A B S T R A C T

Article history: Despite the importance of discourse markers for pragmatically effective communication,
Received 30 May 2011 surprisingly little research has been conducted on the acquisition of these markers by
Received in revised form 6 September 2011 immigrant second language learners. The present paper approaches this understudied
Accepted 12 September 2011
research area by using a developmental learner corpus (Belz and Vyatkina, 2005, 2008) to
Available online 11 October 2011
examine discourse marker use by one naturalistic adult language learner for one year.
Results show very different patterns of use and development among three focal discourse
Keywords:
Discourse markers markers. You know was heavily overused by the participant although its occurrences
Developmental learner corpus declined by 50% over the year; like increased from almost zero uses at the beginning of the
Second language acquisition year to over 2300 occurrences per 100,000 words by mid-year, then dropped by 50% by the
Naturalistic learner study’s end; well was not used at all as a discourse marker. Possible reasons for these
patterns are discussed, including the participant’s need for pragmatic coping devices as a
naturalistic language learner, or the more nuanced functions of well compared to like and
you know. Overall, the paper aims to show the usefulness of developmental learner
corpora as a tool for studies of pragmatic acquisition, as well as the importance of
considering naturalistic learners in our picture of second language pragmatics.
ß 2011 Elsevier B.V. All rights reserved.

1. Introduction

Within the field of second language pragmatics, the study of discourse markers has proved fruitful in understanding
differences in the ways that native speakers and language learners use conversational language. Studies have repeatedly
demonstrated that not only do discourse markers (DMs) perform essential and nuanced functions in speech, but that non-
native speakers very often do not use these pragmatic markers in the same way that native speakers do (Fung and Carter,
2007; Gilquin, 2008; House, 2009; Müller, 2005; Romero Trillo, 2002). Due to the relationally important functions of DMs,
their underuse or misuse in conversation can lead to semantic or pragmatic misunderstandings, which can in turn have
negative consequences for non-native speakers who do not use these markers in expected ways (Lam, 2010; Wierzbicka,
2003). The continued study of non-native speaker (NNS) learning and use of discourse markers, therefore, has important
implications for helping language learners to communicate effectively in a target-language community.
While a good deal of research has been conducted on DM use in native speech and even English as a foreign language (EFL)
settings, surprisingly little research has been aimed at the development of discourse marker competence in one important
group: immigrant language learners. Immigrant learners are, by definition, those who intend to make their home
permanently in a new country, with the ramification that they will likely benefit by gaining and applying pragmatic

* Tel.: +1 11 404 944 1539.

E-mail addresses: eslbrpx@langate.gsu.edu, brittanypolat@gmail.com.

0378-2166/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.pragma.2011.09.009
3746 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756

knowledge in the host discourse community. For those immigrants who wish to participate in their host country’s culture,
discourse markers can serve both as an aid in the expression or negotiation of meaning, and as a signal of competence in the
host culture’s dominant language.
To my knowledge, only one major study investigates DM use among English as a Second Language (ESL) learners
(Hellermann and Vergun, 2007), and its focus is on reporting use rather than describing development of competence. This
lack of longitudinal, development-oriented research may simply reflect trends in the study of pragmatics in general, as there
seem to be very few longitudinal studies of pragmatic development (Taguchi, 2010; Vyatkina and Belz, 2006). This curious
dearth of truly developmental studies means that while we are beginning to understand how and why speakers (both native
and non-native) ‘‘do’’ pragmatics through language use, we do not yet know how this competence might develop in second
language speakers.
The present study seeks to narrow this gap by following the discourse marker development of an immigrant second
language learner in the United States. The focal participant was recorded in informal conversation every two weeks over a
one-year period, allowing for dense, longitudinal analysis of his use of discourse markers over time. Below, I outline existing
research on NNS use of discourse markers, introduce the three focal DMs, and then explain the methodology used in the
study, a developmental learner corpus. I then introduce the study and its results, suggesting implications for pragmatic
development as well as paths for future developmental learner corpus research.

2. Discourse markers and non-native English speakers

Discourse markers are perhaps one of the most ambiguous of pragmatic phenomena, having been described and
categorized differently by various authors over the past three decades (Aijmer and Simon-Vandenbergen, 2011; Müller,
2005; Schourup, 1999). In general, they are considered the pragmatic ‘glue’ of conversation (Schiffrin, 1987), indexing the
speaker’s attitude toward the listener and the discourse (Aijmer, 2002; Fraser, 1990). Although DMs do not in themselves
contain grammatical or semantic meaning (Romero Trillo, 2002), they are multi-functional and can be crucial in expressing a
speaker’s intentions (Müller, 2005; Fung and Carter, 2007). As Wierzbicka (2003) notes of discourse markers, ‘‘their meaning
is crucial to the interaction mediated by speech. . .If learners of a language failed to master the meaning of its particles, their
communicative competence would be drastically impaired (p. 341).’’ Although they elude simple definition, for the purposes
of this study discourse markers will be considered pragmatic devices which convey no semantic meaning but which perform
expressive or interactional functions in discourse.
Because of the vital pragmatic functions that discourse markers serve, researchers in second language pragmatics have
begun to study non-native speakers’ competence in using DMs in conversation. However, these have almost always been
conducted in EFL and/or classroom settings. Romero Trillo (2002), for example, used a corpus-based approach to compare
English discourse markers used by Spanish adults and children to those used by British and American adults and children. He
found that while the native-speaking and non-native-speaking children used English DMs in similar ways, the Spanish adults
used common English discourse makers very infrequently and used uncommon discourse markers more frequently than
native speakers. This lack of English pragmatic competence on the part of the Spanish adults could lead to communicative
failures, Romero Trillo concluded, and should be remedied by appropriate and scaffolded instruction in discourse markers.
Perhaps the most in-depth exploration of differences in NS and NNS use of discourse markers is Müller (2005), in which
the English speech of German university students is compared to that of American university students. As part of a larger
corpus collection project, Müller elicited descriptions of a film from students by asking them to describe the plot to a
classmate, a scenario designed to simulate a genuine communicative context. She found significant differences in the usage
of the four focal markers, like, you know, well, and so. So was used twice as often by the native speakers as the non-native
speakers, and like and you know were used almost five times more frequently by NSs as NNSs, although interestingly, the
German students used well twice as often as the native speakers in its discourse marker function. Müller suggests that
discourse markers can only be properly taught through native speaker contact and outside of a classroom, and that much
more research is needed to understand NNS learning of them.
Fung and Carter (2007) also examined DMs in an EFL context, this time among undergraduate students in Hong Kong. By
comparing the speech of these students to that of native British English speakers using the CANCODE corpus, Fung and Carter
found that the NNS students used certain functional discourse markers (including but, ok, because) more than the NSs. At the
same time, the Hong Kong students significantly underused frequently occurring DMs from the British corpus such as really,
sort of, well, right, and actually, resulting in a wide discrepancy between NSs and NNSs in both frequency and breadth of
discourse marker use. The authors believe that, although Hong Kong English speakers should not necessarily strive to
emulate British English, they should receive more instruction on discourse markers to facilitate their real-world
interactional competence.
Several authors have taken advantage of the Louvain Corpus of Native English Conversation (LOCNEC; De Cock, 2004) and
Louvain International Database of Spoken English Interlanguage (LINDSEI; Gilquin et al., 2010) to compare British English
with European learners of English. In particular, Gilquin’s (2008) study of what she calls hesitators—pauses, smallwords, and
miscellaneous hesitation devices—provides further confirmation that EFL learners tend to use a smaller range of discourse
markers than native speakers. In this case, French EFL learners overused well at the expense of other markers with
comparable functions, something that Gilquin (following Hasselgren, 2002) calls the ‘‘lexical teddy bear’’ effect. Similarly,
Aijmer (2011) found that Swedish learners of English used well more frequently than British native speakers as a speech
 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756 3747

management device (to stall or plan), but less frequently as an attitudinal device. Aijmer compares this to House’s (2009)
finding from a different learner corpus that German EFL learners appropriated and overused you know not for the
interpersonal reasons that native speakers use this DM, but rather in a ‘‘self-serving’’ planning or monitoring function.
In addition to these foreign-language settings, Hellermann and Vergun (2007) specifically investigated discourse marker
use in an ESL immigrant context. Examining DMs in the speech of low-proficiency English learners enrolled in community
college ESL classes, they found that the students in general used very few discourse markers, but that there seemed to be a
correlation between language proficiency and frequency of DM use. Based on their analysis of the students’ classroom
environments, Hellermann and Vergun also believe that the learners did not have the opportunity to learn pragmatic
markers from their language classrooms. Although they did not compare the data to native speaker frequencies, this finding
supported their conclusion that the learners who used discourse markers were those who were more acculturated to the
target culture and spent the most time interacting with native speakers.
In sum, almost all of the research conducted so far on non-native speakers and DMs shows that, despite the critical
importance of conversational discourse marker competence, English language learners fall very short of target-like use.
Based on existing evidence, it appears that learners seldom acquire discourse marker competence in classrooms, but the
results of Hellermann and Vergun’s (2007) study suggest that interacting with native speakers may enhance NNS use of these
pragmatic markers. Because so little research has focused on the process of developing discourse marker competence,
however, we do not know how learners might acquire these skills. Additional research is needed to investigate the
acquisition of discourse markers, which might allow us to better prepare learners to acquire this complex but essential
element of communicative competence.

3. Focal discourse markers

Clearly, the literature presents a great diversity of words and multi-word units that could be considered discourse
markers, and any study attempting to investigate them is necessarily forced to select a few focal markers. The present study
takes its cue from previous studies of DM acquisition by second language learners, focusing on three markers that appear
throughout the literature: you know, like, and well. These were the three DMs examined by Hellermann and Vergun (2007),
the study which closest resembles the present study in terms of setting (United States) and learner type (immigrants). In
addition, these are three of the four markers considered in Müller (2005); they feature prominently in Fung and Carter
(2007); and you know and well are both considered in Romero Trillo (2002). Because of the native speaker and non-native
speaker data available for these discourse markers, as well as the extensive literature on their characteristics, these three
markers will be investigated in the present study.

3.1. You know

Müller (2005) calls you know ‘‘one of the most versatile and notoriously difficult to describe’’ discourse markers (p. 147),
one that has been credited with up to 30 different discourse functions in the literature. As one of the more interactional
discourse markers, you know frequently refers to shared knowledge or understanding between speaker and listener. Its
interactive functions, according to Müller’s (2005) classification, include ‘‘imagine the scene,’’ ‘‘see the implication,’’
acknowledge that the speaker is right, and appeal for understanding (p. 157). House (2009), however, follows Redeker (2006)
in highlighting the non-interactional functions of you know, which may provide a coherence function in discourse. In the
present study, only occurrences of you know that are not syntactically necessary to an utterance are considered discourse
markers (i.e., utterances such as you know how. . . and do you know if. . . were excluded).

3.2. Like

Contrary to its portrayal in popular culture, like in its discourse marker form seems to be used more often by native
speakers as a functional rather than as a purely stylistic device (Fuller, 2003). Its functions have been classified as
approximating, exemplifying, and hedging (Jucker and Smith, 1998), as well focusing and searching for the next word
(Müller, 2005). Although like also occurs frequently as a quotative, the present study does not consider this usage, which may
be a syntactically essential part of an utterance (Hellermann and Vergun, 2007). Only instances of like which are not
grammatically necessary to the meaning of a sentence are considered in this study, which excludes its appearances as a verb,
preposition, or adverb. As Biber et al. (1999) note, however, distinguishing between like in its purely discourse marker sense
and its use as a stance adverbial is often not possible. For this reason, the present study includes both types in like frequency
counts.

3.3. Well

Well assumes some of the same functions as like and you know, such as stalling for time or rephrasing and correcting, but it
also plays roles such as contradicting, evaluating, and indirectly replying to previous turns in the discourse (Müller, 2005). As
such, well can mitigate face-threatening disagreement or confrontation, and it can also act as a framing device to introduce a
new topic or new information in the conversation. These diverse and nuanced functions make it considerably more
3748 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756

specialized than like and you know. Also in contrast to like and you know, which can occur at the beginning, middle, and end
(for you know) of many types of utterances (House, 2009; Schiffrin, 1987), well usually occurs turn-initially and only in
certain contexts (Aijmer, 2011), such as indirect responses. In this study, any non-adverb occurrence of well will be
considered a discourse marker.

4. Learner corpora

To investigate the important but under-studied phenomenon of discourse marker development, this paper takes a
developmental learner corpus case study approach. While learner corpora have been used in applied linguistics since at least
the 1990s (Nesselhauf, 2004), there have as yet been very few applications of developmental learner corpora. However, as
Granger (2004) points out in her review of learner corpus research, ‘‘A detailed longitudinal study of one single learner is of
great value if the focus is on individual interlanguage development’’ (p. 125). Indeed, Ortega and Byrnes (2008) draw
attention to the great need for more longitudinal studies of language learner development, and long-term, developmental
studies are especially under-represented in second language pragmatics (Taguchi, 2010; Vyatkina and Belz, 2006) and
corpus linguistics (Belz and Vyatkina, 2008).
Although several researchers (e.g., Hilton, 2008; Housen, 2002) have used cross-sectional corpus data to investigate
second language development, the only pragmatics-related longitudinal corpus project to my knowledge is Belz and
Vyatkina’s (2005, 2008; Vyatkina and Belz, 2006) research of ‘‘telecollaborative’’ interaction. Their research investigated
development of German modal particles as native English speakers interacted electronically with native German speakers
for a two-month period. While the students also received some instruction on modal particles, results indicated that the
learners mainly produced patterns similar to those they had encountered in interactions with native speakers. The
researchers cite the important role of their bilingual developmental learner corpus in allowing systematic study of this
acquisitional process and in providing students with important linguistic input (Belz and Vyatkina, 2005, 2008; Vyatkina and
Belz, 2006).
Belz and Vyatkina (2008) distinguish between a longitudinal learner corpus, which they consider to measure data at
distant points in time, such as the beginning and end of a semester, and a developmental learner corpus, which measures
production at higher intensity and more frequent intervals. They suggest that ‘‘such dense examinations. . .offer insight into
SLA at the level of individual micro-changes in performance and awareness over time,’’ citing their developmental corpus as
a useful tool in ‘‘facilitating developmental analyses of individual pathways of L2 use’’ and identifying ‘‘the idiosyncratic
nature of learners’ development toward NS norms’’ (Belz and Vyatkina, 2008:34). This type of microgenetic, longitudinal
data collection method seems very well-suited to corpus-based investigations of language acquisition, particularly for
pragmatic markers, which have been shown to be challenging for learners to acquire.
In many ways, however, the present study differs from existing learner corpora, developmental or otherwise. By far the
majority of learner corpora are written rather than spoken (Nesselhauf, 2004), including the well-known ICLE corpus
(Granger, 2003; Granger et al., 2009) and its subfamilies (e.g., Hatzitheodorou and Mattheoudakis, 2011). Several spoken
learner corpora have also been collected, such as the University of Savoie’s PAROLE corpus (Hilton, 2008; Osborne, 2011), but
these are generally set in academic contexts and use university student participants. Perhaps the closest step toward a non-
academic learner corpus is the Louvain International Database of Spoken English Interlanguage, known as LINDSEI (De Cock,
2011; Gilquin et al., 2010). This corpus contains transcripts from interviews in which third- and fourth-year university
students from eleven different language backgrounds were asked to informally discuss (in English) topics such as hobbies
and travel. Even here, however, the participants were recruited from the university, interviews were structured, and the
sessions concluded with a picture interpretation activity. No learner corpus that I know of has collected speech samples in a
naturalistic context with a non-university student participant, and none have examined discourse marker development over
a twelve-month period.
The study reported on in this paper, then, extends existing corpus studies in several ways. First, it uses data collected in a
naturalistic setting with an untutored learner, a data source that has all too often been ignored by applied linguists. This
research therefore represents a significant departure from the academic and written contexts that dominate corpus-based
studies (Granger, 2004) and recent research on second language pragmatics. Second, it is based on a longitudinal case study
conducted over one year, which is much longer than Belz and Vyatkina’s developmental corpus (2005, 2008). In addition to
being longitudinal, the data is also densely microgenetic, as the participant was recorded every two weeks for 30 min. The
resulting learner corpus of over 100,000 words therefore provides extensive documentation of the learner’s language
development over one year, allowing for a truly developmental picture of this participant’s learning processes. This research
thus represents the first time, to my knowledge, that a corpus approach has been used to investigate the acquisition of a
second language pragmatic feature over an extended length of time.

4.1. Native speaker corpus comparison

Though I fully agree with the viewpoint that second language learners should not be seen merely as deficient native
speakers (Bley-Vroman, 1983; Hunston, 2002), the present study nevertheless employs a comparison between the learner–
participant’s developing use of discourse markers and those used by native speakers. Such a comparison is advantageous in
allowing us to see ways in which the non-native speaker’s use of the language approaches that of native speakers and those
 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756 3749

in which it does not. Granger (2004) points out that native comparison corpora are perhaps most appropriate when used
with highly proficient speakers, and the fact that the participant in the current study has learned naturalistically—i.e.,
directly from native speakers, rather than in a classroom—makes this comparison all the more relevant and interesting. It
therefore does not seem misguided to look at how this particular speaker uses language similarly or differently from native
speakers.
An additional reason for looking at native speaker comparisons is the nature of corpora themselves. The availability of
many types of reference corpora in English allows researchers to choose data that reflect the type and register of their own
data, leading to a more accurate comparison of syntactic, stylistic, and lexical choices (Granger, 2009). In the present study,
the participant’s informal spoken data are compared to the informal spoken data of American native speakers, as described
below. This is similar to research by Romero Trillo (2002), Müller (2005), Aijmer (2011), Fung and Carter (2007), and others in
which the researchers compare non-native speaker corpus data to one or more native speaker corpora.
In short, it is not suggested that the NNS should aim to speak in exactly the same way as the native speakers in reference
corpora; but the corpus provides a useful way of studying the NNS’s pragmatic development by reflecting the extent to which
he has picked up the language spoken by those around him. By using available corpus tools, which provide authentic
information about conversational usage, we can build our understanding of the second language acquisition process.

5. Study design

5.1. Participant

The case study participant is a native speaker of Turkish who immigrated to the United States at age 25. This participant,
Alex (a pseudonym), interacts daily with Americans, both with his American wife and in his job at a large mainstream
supermarket. In several of our conversations, Alex indicated a positive orientation toward American culture and a
willingness to adapt to the target community, although he remains very proud of his own linguistic and cultural heritage.
While he does not spend time with any Turkish speakers in the United States, Alex estimates that about half of his time
outside of work is devoted to Turkish language activities (online forums, computer games, talking with his family in Turkey),
and that the other half of his time is spent in English language activities (conversing with his wife and friends, watching
American movies and television shows).
Despite five years of formal, grammar-oriented English instruction in Turkey, Alex describes his English proficiency as
very low at the time of his arrival in the U.S. (This may not be unusual within the Turkish educational system; see Zok, 2010.)
Because most of Alex’s English acquisition occurred in an untutored environment after his arrival in the United States, he can
be considered an untutored learner (Klein and Dimroth, 2009). According to Klein and Dimroth (2009), untutored learners
have access to the target language through everyday communication, rather than in a classroom setting; they are not
exposed to any kind of ‘‘systematic external control’’ (p. 508) such as tests or grades; and they experience rapid
communicative pressure, which forces them to make constant use of their limited linguistic systems. This description of
untutored learning maps quite well onto Alex’s situation, although he perhaps has an advantage over many immigrant
learners in that his spouse is a native English speaker.
At the time this study began, his English proficiency was, impressionistically, quite high, despite the fact that his
conversation displayed the low grammatical accuracy characteristic of naturalistic learners (Klein and Perdue, 1993). Alex’s
English-language job requirements, which included managing 25 employees and high volumes of perishable food inventory
every week, along with the fact that he received three promotions in three years on the job, confirm that his functional
English ability is advanced. The study follows Alex from 2.5 years to 3.5 years after his immigration, beginning when he was
promoted to a new management position in his job.

5.2. Data collection and analysis

I recorded conversations with Alex every two weeks for one year, from January through December 2010. Because I have
known Alex for six years and have spoken with him as a friend on many occasions, the data intentionally represent more
informal conversation rather than formalized research interviews. To maintain as naturalistic a setting as possible, the
conversations were held in Alex’s home and I encouraged him to discuss whatever topics interested him. Topics we discussed
over the year include sports teams, world and Turkish politics, his cultural and linguistic adaptation to the United States,
childhood memories, and other subjects of interest to him. My role was to make Alex feel at ease and keep the conversation
going, so I encouraged him to speak as much as possible during the 30-min conversations by providing backchanneling cues,
asking follow-up questions, and providing new topics of discussion as needed.
The 24 recorded conversations resulted in a learner corpus of 104,041 words, and a concordancer, AntConc (Anthony,
2011), was used to find frequencies for the three focal markers. Because you know, like, and well could be used in ways other
than as discourse markers, each instance of these terms was analyzed manually according to the criteria listed in section 3
above, and non-DM instances were thrown out. At the same time, I felt it was important to ascertain whether Alex was using
the markers in a target-like way. I therefore evaluated each occurrence of the focal DMs for pragmatic appropriateness, based
on my judgment of whether it would have been pragmatically possible for a native speaker to use the discourse marker in a
similar context. Well did not occur in the data as a discourse marker, so it was not analyzed for appropriateness.
3750 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756

A second rater also analyzed 50% of the data for contextual appropriateness. Perhaps because of the functional flexibility
of like and you know, we were in 100% agreement that there were no pragmatically inappropriate uses, with one possible
exception. The bundle you know for example appeared 42 times in the data, probably as a result of Alex’s frequent overall use
of for example, which occurred 165 times throughout the year. An example of this type of usage is seen in the following
excerpt from April 18, in which he discusses mortgages:

I think the mortgage is for certain kind situation. You know for example, if you don’t have enough money to buy, and but
you have to have a place to live, and mortgage give you chance to giving same money but you own. (Alex, April 18)

Although this collocation did not appear in the Santa Barbara Corpus of Spoken American English (Du Bois et al., 2000,
2003; Du Bois and Englebretson, 2004, 2005), used as a reference corpus in this study, a similar collocation (you know for
instance) did occur. For this reason, and because you know for example is not necessarily pragmatically inappropriate, it was
retained in the data.

5.3. Reference corpus

The native speaker corpus used in this comparison was the Santa Barbara Corpus of Spoken American English (Du Bois
et al., 2000, 2003; Du Bois and Englebretson, 2004, 2005). Parts 1–4 of the Santa Barbara Corpus (SBCSAE) contain 60
recordings with a total of approximately 249,000 words. While the corpus features data from speakers of all ages, summaries
of the files indicate that speakers in at least 14 conversations are in the same age range as Alex (20s through early 30s,
including university students; Du Bois et al., 2000, 2003; Du Bois and Englebretson, 2004, 2005). The discourse setting and
participants in this corpus, therefore, are similar enough to the developmental learner corpus in this study to be a valid
reference corpus.
For analysis, data from the SBCSAE were downloaded and analyzed in the same way as the developmental learner corpus.
Each occurrence of you know, like, and well was examined and all non-DM occurrences were removed from the data.
Quotative like was also removed, as this function is generally not considered a discourse marker (Müller, 2005). The
frequencies for each discourse marker are shown in Table 1.

6. Results and analysis

Alex’s use of the focal discourse markers is striking for its extremes: he uses two of the markers (you know and like) far
more frequently than native speakers and does not use the third marker (well) at all (see Table 2). The patterns of use over the
one-year period are also very different for each of the three markers, as shown in Fig. 1. Alex’s usage of you know starts out
very high, at 4122 occurrences per 100,000 words, then peaks in April and gradually declines for the rest of the year; by
December, he uses you know only half as often as he did in January. Like tells quite a different story. Although Alex only uses
like once or twice in each conversation at the beginning of the year, his usage begins increasing exponentially in April so that
by July, like occurs as frequently as you know. In the same way as you know, like then declines for the rest of the year, finishing

Table 1
Native speaker usage of focal discourse markers.

You know Like Well

SBCSAE 580 586 559

Note: All frequencies are normalized per 100,000 words.

Table 2
Alex’s raw and normalized frequency counts by month.

Month Total words You know Like Well

Raw Normed Raw Normed Raw Normed

January 7642 315 4122 1 13 0 0

February 8172 295 3610 2 25 0 0
March 8326 329 3952 0 0 0 0
April 8244 353 4282 50 607 0 0
May 8172 302 3696 120 1468 0 0
June 8093 177 2187 174 2150 0 0
July 8261 201 2433 191 2312 0 0
August 8913 162 1818 135 1515 0 0
September 9025 172 1906 117 1296 0 0
October 8267 178 2153 100 1210 0 0
November 9516 234 2459 110 1156 0 0
December 8518 172 2019 86 1010 0 0

Note: Monthly figures represent two biweekly recordings combined. All frequencies are normalized per 100,000 words.
[(Fig._1)TD$IG] B. Polat / Journal of Pragmatics 43 (2011) 3745–3756 3751

4500
4000
3500

Normed Frequency
3000
2500
2000 well
1500
you know
1000
like
500
0

Jan.
Feb.
March
April
May
June
July
August
Sep.
Oct.
Nov.
Dec.
Month

Fig. 1. Alex’s usage of focal discourse markers over one year.

in December at half the frequency of its high point in July. In contrast to the very high frequencies and volatility of like and
you know, well stays at zero occurrences all year. Each of these discourse markers will be discussed in more detail below.

6.1. You know

Compared to the SBCSAE norm of 580 occurrences per 100,000 words, Alex’s usage of you know is rather extraordinary.
His high point of 4283 uses per 100,000 words is over seven times more frequent than the SBCSAE, and even his lowest
frequency of 1818 is over three times more frequent than the native speaker norm. These results are especially interesting
given that previous studies found EFL students significantly underusing you know. Müller (2005) reported that the German
students in her study used this marker five times less frequently than the native speaker comparison group, and in Fung and
Carter (2007), you know occurred in only .16% of the student data, compared to .38% of the native speaker data. Romero Trillo
(2002) found that although you know was heavily used in the native speaker corpora, the Spanish-speaking adults rarely used
this DM, resulting in the biggest gap between native and non-native frequencies of all six discourse markers in the study.
While Hellermann and Vergun (2007) did not compare their L2 data to a native speaker corpus, their results show that their
participants used you know only 25 times in 4966 turns, which seems to be far less than Alex and native speakers (Fig. 2).
Specific features of this discourse marker, coupled with Alex’s experiences learning language naturalistically within the
target culture, may contribute to this interesting pattern. In Müller’s (2005) study, an appeal to the listener’s understanding
through you know is related to a lexical or content search, and it was often used by the German students when they could not
find the right word and so relied on the hearer’s own understanding to fill in the gaps. It seems likely that this could be an
effective device for a non-native speaker to use with a native speaker interlocutor, who can probably guess the non-native
speaker’s meaning when appealed to. Since Alex frequently conversed with native speakers at home and at work, he may
sometimes use you know as a strategic prod for native speakers’ more complete linguistic knowledge.
Just as Müller notes that you know can be an appeal or a ‘‘gap-filler,’’ House’s (2009) analysis of German students’
discourse suggests that it acts as ‘‘a stock phrase mainly used to help speakers process and plan their output, and link spans of
discourse’’ (p. 187). Examples from Müller’s and House’s data of both native and non-native speakers using you know in this
way show that the speaker may be searching for content or lexical information, and it is often accompanied by pauses or
hesitation phenomena. This function especially makes sense in the context of a second language learner, who may need
additional time to find the right word or grammatical structure but does not want to give up the floor while searching for the
appropriate phrase. Many examples from the present data show that Alex indeed seems to be using you know this way, of
which only two are included here.

You know, if somebody see the you know the old eh, you know, some part of machine everybody said oh such a bad things.
(April 3)

That was the biggest one because, we had bigger than that for the, you know, stronger than that. (October 4)

In much the same way, Alex’s use of you know seems to frequently mask repair and false starts (Erman, 2001; Holmes,
1986; Müller, 2005), as in the following example:

Ok right now is, for me, you know, you know I supposed to be one side because my father soldier, but I’m not one of the
side. (April 3)

Given that these functions of you know are associated with hesitation and disfluencies, it is interesting to consider what
Alex’s general production of disfluency markers might tell us about his use of discourse markers. I therefore compared the
use of major disfluency markers uh/eh, um, er in Alex’s overall data and the reference corpus. The combined data for all these
disfluencies together, shown in Fig. 3, reveals that Alex uses disfluencies roughly five and a half times less than the native
[(Fig._2)TD$IG]
3752 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756

6000

5000

Normed Frequency
4000

3000

2000

1000

Data Source

Fig. 2. Comparison of native speaker and Alex’s usage of you know.

[(Fig._2)TD$IG]
1200

1000
Normed Frequency

800

600

400

200

0
SBCSAE Alex
Data Source

Fig. 3. Disfluency marker usage by native speakers and by Alex.

speakers. This is quite unexpected, but it does suggest one possible cause of his excessive use of this discourse marker: Alex
may be using you know as a catch-all gap-filler, replacing the disfluency markers and some other discourse markers that are
used by native speakers.
By using you know instead of unfilled pauses or less interactive hesitation markers such as um or er, Alex is working through
the disfluencies of a non-native speaker but still holding his position in the conversation. His use of you know allows extra time
for a lexical or content search while at the same time actively participating in the conversation and appealing to the listener’s
understanding of his meaning. This may be quite an effective strategy for a learner who has developed his knowledge of English
in a naturalistic setting such as a fast-paced workplace, where we can guess that he has had to communicate quickly while still
learning the language (Klein and Dimroth, 2009). You know seems to occur so frequently because it may be a coping device that
Alex has adopted to deal with the linguistic and interactional demands of real-life contexts.

6.2. Like

Alex’s use of like as a discourse marker stands out for its huge fluctuation in frequency over the year. Barely present at all
during the first months of conversation, like quickly catches up with you know and then from this high point halves in
frequency by the end of the study period. It also seems that once Alex started using like, he employed it in ways very similar
to those used by native speakers. Müller (2005) reports four different discourse marker functions of like in her corpus—
approximating quantity, introducing an example, indicating lexical focus, and marking a lexical or content search—and it
appears that Alex is using each of these four types. The first 5 min of one transcript from July 17 produced the following
examples of these four functions:

Number approximation: Ok what I was doing for editing part, you know sometimes, we were taking like two hours long.

Focusing attention: So many different things. And, like you have to make sure its match, together, and everybody
understand what you are trying to say.

Searching for appropriate word/Introducing Example: Yeah but you know like you can show one video, and you can look
them. Like you had a message line, like one hour program right.
[(Fig._4)TD$IG] B. Polat / Journal of Pragmatics 43 (2011) 3745–3756 3753

2500

2000

Normed Frequency
1500

1000

500

Data Source

Fig. 4. Comparison of native speaker and Alex’s usage of like.

Furthermore, although Alex’s use of like appears to be twice as frequent as the native speakers in the SBSCAE by the end of
the study period, other studies of American English native speakers have shown the use of like to be much higher. For
instance, Fuller (2003) found that the number of like occurrences in interviews with graduate students (in the same age range
as Alex, mid-twenties through early thirties) ranged from 1.3 to 18 times per thousand words. The upper frequency
corresponds 1800 times per 100,000 words, which is right in line with Alex’s usage. This suggests that near the end of the
study Alex may have in fact achieved a very native-like frequency for this discourse marker (Fig. 4).

6.3. Well

Well as a discourse marker is completely absent from Alex’s speech from the beginning through the end of the data
collection period. Although Alex uses well frequently as an adverb, he does not appear to have recognized it as a discourse
marker. This corresponds with Romero Trillo’s (2002) finding that Spanish-speaking adults rarely used well, especially when
compared with the high frequencies in the native speaker comparison corpus, and to Fung and Carter’s (2007) results
showing a sizeable difference between native speaker and non-native speaker use of well.
This finding contrasts, however, with data from the German and French learners in Müller (2005), Gilquin (2008), and
Aijmer (2011). All of these studies showed EFL learners overusing well compared to native speakers, although Aijmer found
that this only applied to the speech management functions of well and not to its interpersonal functions. These studies
suggest, then, that the learners in these EFL contexts have latched on to well as a pragmatic catch-all, possibly similar to the
way that Alex uses you know. This points toward the conclusion that some learners simply adopt a favorite pragmatic marker
and use it rather indiscriminately in discourse to manage their speech performance (e.g. House, 2009). It seems that if
classroom learners are introduced to well, they may appropriate it for their own pragmatic purposes without fully
understanding its use in native speech (Aijmer, 2011).
On the other hand, Alex, in his naturalistic learning context, did not pick up on this discourse marker at all, even though he
acquired like and you know. We cannot attribute this discrepancy to lack of exposure, since the high frequency with which
well occurs in native speaker data implies that Alex was exposed to this marker often in interactions with native speakers.
(Interestingly, the conversation transcripts show that, when conversing with Alex, I used well quite frequently to indicate
polite disagreement and other functions shown in the literature.) The difference may therefore lie in the discourse functions
of well, which are possibly less salient in authentic communication because of their more hedged or indirect nature. Its more
relationally nuanced functions could have prevented Alex from detecting it as a discourse marker in the linguistic input
around him, or if he did notice well, its attitudinal or face-threatening functions may have led him to avoid trying to use it. It
therefore does not seem unreasonable to surmise that the more specific and delicate uses of well have prevented Alex from
picking up this discourse marker as he has the other two.

7. Participant feedback

In order to ascertain just how much Alex was aware of his use of discourse markers, I held a final, brief conversation with
him specifically to elicit his perspective on this topic. At first, he was hesitant to discuss anything about his speech, declaring
that he was not interested and did not pay attention to how he used these words. This in itself was somewhat revealing, as it
suggests that Alex was not intentionally trying to improve or change his use of discourse markers or any other linguistic
feature. However, we did discuss the findings, and when I brought up the fact that he used like very frequently, he was quite
surprised. He did not realize he used it so often, and he had no explanation as to why he had suddenly started using it,
increased his use of it, or decreased his use over the year. Alex was also completely unaware that well existed as a discourse
marker; when I mentioned that he had not used it at all, he protested that he says well very often (which of course he did, as
3754 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756

an adverb). Apparently he was not aware of its discourse marker sense at all, since my explanations and examples seemed to
be the first time he understood well in any function other than adverb.
Likewise, when I first told Alex that he used you know up to 200 times per 30-min conversation, he refused to believe it
until I showed him the transcripts and word counts. In response to my question about why he used this expression so
frequently, he said, ‘‘I wanted to make sure you understood me. I don’t trust my English, I am saying right or wrong, and I
want to get approval from you.’’ This is a fascinating explanation specific to non-native speakers, one that is not mentioned in
the literature on second language learners and you know. Of course, just because Alex believes this is why he used you know
does not necessarily mean this is the correct or only explanation; after all, he had not even been aware that he was using it so
much. But it does indicate that when non-native speakers are involved, and particularly untutored speakers, discourse
markers in their developing language systems may take on a different meaning or function than that associated with native
speech. This also reveals a potential gap between what learners appear to be doing and what they might actually be thinking
or intending to express. Future research, then, would likely benefit from mixed methods approaches that combine the broad
perspective of corpora with a qualitative study of the learner’s own perspective.

8. Conclusion

We have seen that the untutored learner in this study showed very striking patterns of discourse marker use and
development over the year-long study period, even though the study began two and a half years after his arrival in the United
States. Alex’s extreme use of you know declined by 50% over the year, and his use of like went from almost zero to over 2000
instances per 100,000 words, then back down to 1000. In contrast, he did not use well as a discourse marker at all, even
though he used it regularly as an adverb. A discussion with Alex about these discourse markers reveals that he was not aware
of his use of these markers and had probably made no intentional effort to modify his use of them.
These results have interesting implications for pragmatic development in target-language settings, particularly where
the learner receives significant exposure to the language and has a positive orientation toward the target language and
culture. The study suggests that, when left completely to their own devices, learners may pick up on some discourse markers
more readily than others, resulting in uneven distribution of these important pragmatic devices. The fact that Alex seemed to
use a few all-purpose markers with great frequency, while perhaps ignoring markers such as well with more specific or
nuanced functions, indicates that certain DMs may in fact be more difficult to use appropriately than others. While further
studies on this aspect of discourse marker acquisition are certainly needed, this finding may have implications for classroom
instruction. If some discourse markers are more difficult to pick up than others, it may be most beneficial for teachers to focus
on these in the classroom, especially in situations where learners will be interacting primarily with native English speakers.
A further implication relates to the great pragmatic need of naturalistic learners to communicate effectively—and often
beyond their linguistic means (Klein and Dimroth, 2009)—with native speakers. The participant in this study works in an
English-language workplace and manages employees in English, also dealing extensively with customers, supervisors, and
food safety regulations in English. His practical communicative needs are therefore fairly demanding, and in this context it is
hardly surprising that he has developed certain strategies for dealing with lexical or content search issues while holding his
place in interactions. If he uses a limited repertoire of discourse markers such as you know and like to cover all his pragmatic
needs, this raises questions about the extent to which his discourse marker use will continue to become more native-like. It is
possible that Alex, and perhaps other untutored learners like him, may never acquire the depth and breadth of discourse
markers that native speakers command. On the other hand, we have seen that within a year he went from barely using like to
using it in contextually appropriate ways, which means that the same pattern could potentially occur in the future with well.
The study has also shown the usefulness of a developmental learner corpus as a tool in second language studies. Although the
present corpus differs from previous learner corpora in that it follows just one learner, the microgenetic and longitudinal nature
of the case study revealed very interesting developmental patterns that could not have been identified in a larger but less
detailed corpus. This suggests a promising future for developmental learner corpora in areas of second language acquisition,
including the study of pragmatic features such as discourse markers. As more and better developmental corpora are compiled,
we may be able to see where learners tend to have the most difficulty, thus allowing instructors and curriculum designers to
more effectively address problematic areas for second language learners. Such corpora may also contribute to our
understanding of developmental sequences, both furthering our knowledge of the language acquisition process and suggesting
fruitful areas for future research. In short, despite its time-consuming and labor-intensive collection process, the use of a dense
developmental corpus seems to be a very promising research approach, especially if paired with more qualitative analyses.
The present study is by no means without its limitations, some of which may be addressed by future research. First, while
the case study approach permits in-depth exploration, it cannot necessarily be directly extended to other learners, which
means that much more research on other learners is needed to create an extensive, reliable pool of data. Following learners of
different language backgrounds, educational experience, and proficiency levels would certainly provide more insight into
general patterns of discourse marker development. Second, even though one year is a much longer study period than any
previous developmental corpus approaches, it is still not a long period in the language acquisition process. Longitudinal
studies on the scale of many years would be needed to trace the full developmental arc of pragmatic features. It will also be
important to complement corpus analysis with more detailed functional and qualitative analyses, which space constraints
prevented in the present study. Although the learner corpus featured here provides a sense of what kinds of developmental
trajectories are possible in DM acquisition, much research remains to be done in this important pragmatic area.
 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756 3755

In sum, this study has presented a longitudinal picture of one learner’s development of three focal discourse markers, you
know, like, and well. In doing so, it has attempted to demonstrate the usefulness of developmental learner corpora and has
suggested future mixed-methods research with developmental corpora and qualitative approaches. Results suggested that
acquisition of discourse markers may depend on individual features of each marker, such as interactional salience, as well as
each learner’s particular pragmatic needs. Overall, I hope to have convincingly shown that research with new methodologies
and under-studied learner groups can contribute important insights to our understanding and teaching of second language
pragmatics.

Acknowledgements

I would like to thank Dr. Eric Friginal for providing valuable suggestions and insight during the early stages of this project,
and Luciana Junqueira and Lenora Bryant for their helpful feedback and encouragement.

References

Aijmer, Karin, 2002. English Discourse Particles: Evidence from a Corpus. John Benjamins, Philadelphia.
Aijmer, Karin, 2011. Well I’m not sure I think. . .The use of well by non-native speakers. International Journal of Corpus Linguistics 16 (2), 231–254.
Aijmer, Karin, Simon-Vandenbergen, A.M., 2011. Pragmatic markers. In: Zienkowski, J., Östman, J., Verschueren, J. (Eds.), Discursive Pragmatics. John
Benjamins, Philadelphia, pp. 223–247.
Anthony, Laurence, 2011. AntConc (Version 3.2.2) [Computer Software] Waseda University, Tokyo, Japan Available from http://www.antlab.sci.waseda.ac.
jp/
Belz, Julie, Vyatkina, Nina, 2005. Learner corpus analysis and the development of L2 pragmatic competence in networked intercultural language study: the
case of German modal particles. Canadian Modern Language Review 62 (1), 17–48.
Belz, Julie, Vyatkina, Nina, 2008. The pedagogical mediation of a developmental learner corpus for classroom-based language instruction. Language
Learning & Technology 12 (3), 35–52.
Biber, Douglas, Johansson, Stig, Leech, Geoffrey, Conrad, Susan, Finegan, Edward, 1999. Longman Grammar of Spoken and Written English. Pearson
Education Limited, Harlow, England.
Bley-Vroman, Robert, 1983. The comparative fallacy in interlanguage studies: the case of systematicity. Language Learning 33, 1–17.
De Cock, Sylvie, 2004. Preferred sequences of words in NS and NNS speech. Belgian Journal of English Language and Literatures 2, 225–246.
De Cock, Sylvie, 2011. Preferred patterns of use of positive and negative evaluative adjectives in native and learner speech: an ELT perspective. In:
Frankenberg-Garcie, A., Flowerdew, L., Aston, G. (Eds.), New Trends in Corpora and Language Learning. Continuum, New York, pp. 198–212.
Du Bois, John, Englebretson, Robert, 2004. Santa Barbara Corpus of Spoken American English, Part 3. Linguistic Data Consortium, Philadelphia.
Du Bois, John, Englebretson, Robert, 2005. Santa Barbara Corpus of Spoken American English, Part 4. Linguistic Data Consortium, Philadelphia.
Du Bois, John, Chafe, Wallace, Meyer, Charles, Thompson, Sandra, 2000. Santa Barbara Corpus of Spoken American English Part 1. Linguistic Data
Consortium, Philadelphia.
Du Bois, John, Chafe, Wallace, Meyer, Charles, Thompson, Sandra, Martey, Nii, 2003. Santa Barbara Corpus of Spoken American English, Part 2. Linguistic
Data Consortium, Philadelphia.
Erman, Britt, 2001. Pragmatic markers revisited with a focus on you know in adult and adolescent talk. Journal of Pragmatics 33, 1337–1359.
Fraser, Bruce, 1990. An approach to discourse markers. Journal of Pragmatics 14, 383–395.
Fuller, Janet, 2003. Use of the discourse marker ‘like’ in interviews. Journal of Sociolinguistics 7 (3), 365–377.
Fung, Loretta, Carter, Ronald, 2007. Discourse markers and spoken English: native and learner use in pedagogic settings. Applied Linguistics 28 (3), 410–439.
Gilquin, Gaëtanelle, 2008. Hesitation markers among EFL learners: pragmatic deficiency or difference? In: Romero-Trillo, J. (Ed.), Corpus and Pragmatics. A
Mutualistic Entente. Mouton de Gruyter, Berlin, New York, pp. 119–149.
Gilquin, Gaëtanelle, De Cock, Sylvie, Granger, Sylviane, 2010. The Louvain International Database of Spoken English Interlanguage. Handbook and CD-ROM.
Presses universitaires de Louvain, Louvain-la-Neuve.
Granger, Sylviane, 2003. The International Corpus of Learner English: a new resource for foreign language learning and teaching and second language
acquisition research. TESOL Quarterly 37 (3), 538–545.
Granger, Sylviane, 2004. Computer learner corpus research: current status and future prospects. Language and Computers 23, 123–145.
Granger, Sylviane, 2009. The contribution of learner corpora to second language acquisition and foreign language teaching: a critical evaluation. In: Aijmer,
K. (Ed.), Corpora and Language Teaching. John Benjamins, Philadelphia, pp. 13–31.
Granger, Sylviane, Dagnaux, Estelle, Meunier, Fanny, Paquot, Magali, 2009. International Corpus of Learner English (Version 2). Presses universitaires de
Louvain, Louvain, Belgium.
Hasselgren, Angela, 2002. Learner corpora and language testing: smallwords as markers of learner fluency. In: Granger, S., Hung, J., Petch-Tyson, S.
(Eds.), Computer Learner Corpora, Second Language Acquisition and Foreign Language Teaching. John Benjamins, Philadelphia, pp. 143–173.
Hatzitheodorou, Anna-Maria, Mattheoudakis, Marina, 2011. The impact of culture on the use of stance exponents as persuasive devices: The case of GRICLE
and English native speaker corpora. In: Frankenberg-Garcia, A., Flowerdew, L., Aston, G. (Eds.), New Trends in Corpora and Language Learning.
Continuum, London, pp. 229–246.
Hellermann, John, Vergun, Andrea, 2007. Language which is not taught: the discourse marker use of beginning adult learners of English. Journal of
Pragmatics 39, 157–179.
Hilton, Heather, 2008. The link between vocabulary knowledge and spoken L2 fluency. Language Learning Journal 36 (2), 153–166.
Holmes, Janet, 1986. Functions of you know in women’s and mens’ speech. Language in Society 15 (1), 1–22.
House, Juliane, 2009. Subjectivity in English as Lingua Franca discourse: the case of you know. Intercultural Pragmatics 6 (2), 171–193.
Housen, Alex, 2002. A corpus-based study of the L2 acquisition of the English verb system. In: Granger, S., Hung, J., Petch-Tyson, S. (Eds.), Computer Learner
Corpora, Second Language Acquisition and Foreign Language Teaching. John Benjamins, Philadelphia, pp. 77–117.
Hunston, Susan, 2002. Corpora in Applied Linguistics. Cambridge University Press, Cambridge.
Jucker, Andreas, Smith, Sara, 1998. And people just you know like ‘wow’: discourse markers as negotiating strategies. In: Jucker, A., Ziv, Y. (Eds.), Discourse
Markers: Descriptions and Theory. John Benjamins, Philadelphia, pp. 171–202.
Klein, Wolfgang, Dimroth, Christine, 2009. Untutored second language acquisition. In: Ritchie, W., Bhatia, T. (Eds.), The New Handbook of Second Language
Acquisition. Emerald Group, Bingley, UK, pp. 503–521.
Klein, Wolfgang, Perdue, Clive, 1993. Utterance structure. In: Perdue, C. (Ed.), Adult Language Acquisition: The Results. European Science Foundation, New
York, pp. 1–40.
Lam, Phoenix, 2010. Discourse particles in corpus data and textbooks: the case of well. Applied Linguistics 31 (2), 260–281.
Müller, Simone, 2005. Discourse Markers in Native and Non-native English Discourse. John Benjamins, Philadelphia.
Nesselhauf, Nadja, 2004. Learner corpora and their potential for language teaching. In: Sinclair, J. (Ed.), How to Use Corpora in Language Teaching. John
Benjamins, Philadelphia, pp. 125–153.
3756 B. Polat / Journal of Pragmatics 43 (2011) 3745–3756

Ortega, Lourdes, Byrnes, Heidi, 2008. The Longitudinal Study of Advanced L2 Capacities. Routledge, New York.
Osborne, John, 2011. Oral learner corpora and the assessment of fluency in the Common European Framework. In: Frankenberg-Garcie, A., Flowerdew,
L., Aston, G. (Eds.), New Trends in Corpora and Language Learning. Continuum, New York, pp. 181–197.
Redeker, Gisela, 2006. Discourse makers as attentional cues at discourse transitions. In: Fischer, K. (Ed.), Approaches to Discourse Particles. Elsevier,
Amsterdam, pp. 339–358.
Romero Trillo, Jesus, 2002. The pragmatic fossilization of discourse markers in non-native speakers of English. Journal of Pragmatics 34, 769–784.
Schiffrin, Deborah, 1987. Discourse Markers. Cambridge University Press, Cambridge.
Schourup, Lawrence, 1999. Discourse markers. Lingua 107 (3–4), 227–265.
Taguchi, Naoko, 2010. Longitudinal studies in interlanguage pragmatics. In: Trosborg, A. (Ed.), Pragmatics Across Languages and Cultures. De Gruyter
Mouton, New York, pp. 333–362.
Vyatkina, Nina, Belz, Julie, 2006. A learner-driven corpus intervention for the development of L2 pragmatic competence. In: Bardovi-Harlig, K., Felix-
Brasdefer, J., Omer, A. (Eds.), Pragmatics and Language Learning. University of Hawaii Press, Honolulu, HI, pp. 315–358.
Wierzbicka, Anna, 2003. Cross-cultural Pragmatics: The Semantics of Human Interaction. Walter de Gruyter, Berlin.
Zok, David, 2010. Turkey’s language revolution and the status of English today. The English Languages: History, Diaspora, and Culture 1 Retrieved from
http://jps.library.utoronto.ca/index.php/elhdc/issue/current/showToc.

Brittany Polat is pursuing her PhD in Applied Linguistics at Georgia State University. Her primary research interests are individual differences in second language
acquisition and second language pragmatics.