discourse markers in oral interaction by third-year ulis mainstream english majors = dấu hiệu diễn ngôn trong giao tiếp nói của sinh viên năm thứ 3 khoa tiếng anh

VIETNAM NATIONAL UNIVERSITY
UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES
FACULTY OF POST-GRADUATE STUDIES






LÊ THỊ THU HUYỀN






DISCOURSE MARKERS IN ORAL INTERACTION BY
THIRD-YEAR ULIS MAINSTREAM ENGLISH MAJORS

(Dấu Hiệu Diễn Ngôn trong Giao Tiếp Nói của
Sinh Viên Năm Thứ 3 Khoa Tiếng Anh, ĐHNN, ĐHQGHN)

M.A. COMBINED PROGRAMME THESIS

Field: English Linguistics
Code: 60 22 15

HANOI - 2012



VIETNAM NATIONAL UNIVERSITY
UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES
FACULTY OF POST-GRADUATE STUDIES






LÊ THỊ THU HUYỀN





DISCOURSE MARKERS IN ORAL INTERACTION BY
THIRD-YEAR ULIS MAINSTREAM ENGLISH MAJORS

(Dấu Hiệu Diễn Ngôn trong Giao Tiếp Nói của
Sinh Viên Năm Thứ 3 Khoa Tiếng Anh, ĐHNN, ĐHQGHN)

M.A. COMBINED PROGRAMME THESIS

Field: English Linguistics
Code: 60 22 15
Supervisor: Assoc. Prof. Dr. Lê Hu
̀
ng Tiê
́
n









HANOI - 2012
iv


TABLE OF CONTENTS

CONTENTS
PAGE
DECLARATION

i
ACKNOWLEDGEMENTS
ii
ABSTRACT
iii
TABLE OF CONTENTS
iv
LIST OF TABLES AND FIGURES
vii
LIST OF ABBREVIATIONS
viii
PART I: INTRODUCTION
1
1. Statement of the problems and rationale behind the study
1
2. Aims and objectives of the study
3
3. Scope of the study
4
4. Contribution of the study
5
5. Organization of the study
5


PART II: DEVELOPMENT
7
CHAPTER 1: LITERATURE REVIEW
7
1.1. Discourse markers (DMs)

7
1.1.1. Terminology
7
1.1.2. Characteristics of DMs
9
1.1.3. Selection of DMs in the present study
12
1.1.4. Main functions of four selected DMs
13
1.2. Previous studies on DMs
15
1.2.1. Related studies worldwide
15
1.2.2. Related studies in Vietnam
19


CHAPTER 2: METHODOLOGY
21
2.1. Participants
21
2.1.1. Third-year ULIS mainstream English majors
21
v


2.1.2. Teachers of English Speaking Skill
25
2.2. Data collection instruments
25

2.2.1. Interaction task
25
2.2.2. Questionnaires
26
2.2.3. Interviews
27
2.3. Data collection procedure
28
2.4. Data analysis methods and procedures
30


CHAPTER 3: RESULTS AND DISCUSSION
32
3.1. Most common DMs used by third-year ULIS mainstream English majors
in spoken interaction with native speakers of English

32
3.2. Specific discourse functions of DMs in spoken discourse
34
3.2.1. Well
34
3.2.2. You know
39
3.2.3. I mean
42
3.2.4. Like
43
3.3. Justifications for the low frequency of DMs in students‟ spoken discourse
47

CHAPTER 4: SOME PEDAGOGICAL IMPLICATIONS AND
SUGGESTED DMs TEACHING RESOURCES
52
4.1. Some pedagogical implications
52
4.1.1. Implications for learners of English
52
4.1.2. Implications for teachers of English and syllabus designers
55
4.2. Suggested DMs teaching resources
57
4.2.1. Two explicit DM teaching frameworks
57
4.2.2. Some suggested activities for teaching oral DMs
60


PART III: CONCLUSION
65
1. Major findings of the study
65
2. Concluding remarks
66
vi


3. Limitations of the current study
67

4. Suggestions for further research
68


REFERENCES
70


APPENDICES
I
Appendix 1
Topic and questions for the interaction
I
Appendix 2A
Questionnaire for students
II
Appendix 2B
Questionnaire for teachers
IV
Appendix 2C
Interview schedule for teachers
VI
Appendix 3
Excerpts in the student data
VII
Appendix 4
Transcription of the interview
XI

vii


LIST OF TABLES AND FIGURES

TABLES
PAGE


Table 1: Terminology variations of DMs (adapted from Yang, 2011)
8
Table 2: Main functions of the four DMs (Adapted from Müller, 2005: 246 and
Huang, 2011)
14
Table 3: The distribution of students in different specific major classes

15
Table 4: Participants in the interaction with native speakers of English
25
Table 5: Codes and conventions used in the study. (Adapted from Müller, 2005:
281)
31
Table 6: Positions of the DM you know in the student data
40
Table 7: Sample lesson procedures in PPP and III frameworks in Jones‟s (2011: 74)
study
59


FIGURES



Figure 1: Number of occurrences of some DMs used by third-year ULIS students
34
Figure 2: Students‟ and Teachers‟ perception of the importance of DMs in oral
communication
48
Figure 3: The extent to which DMs are displayed in students‟ oral discourse, as
perceived by the students and teachers.
49
Figure 4: Possible reasons for the students‟ infrequent display of DMs in speech,
from the perspectives of the students and the teachers
50
Figure 5: The extent to which DMs have been explicitly taught by teachers
51

Figure 6: The students‟ expectation to acquire and incorporate DMs in their speech
54
Figure 7: Ways for students to acquire the use of oral DMs as suggested by the
students and teachers
55
Figure 8: Teachers' perception towards the level at which DMs should be explicitly
taught
57


viii


LIST OF ABBREVIATIONS

DM(s)
Discourse marker(s)
EFL
English as a Foreign Language
ESL
English as a Second Language
FELTE
Faculty of English Language Teacher Education
L1
first language
L2
second language
NNSs
Non-native speakers
NSs

Native speakers
ULIS – VNU
University of Languages and International Studies – Vietnam National
University




1


PART I: INTRODUCTION
1. Statement of the problem and rationale behind the study
In the world of increasing globalization, teaching and learning English language,
especially in terms of communicative oral skill has become a matter of concern to many
linguists and educators. Ur (1996: 120) states that “of all the four skills (listening,
speaking, reading, and writing), speaking seems intuitively the most important”. Ur‟s
(1996) viewpoint was supported later on by Rudder (1999: 25) who regards speaking skill
as “one of the most important goals in language teaching” since it is “now more than ever
before (…) essential for interactive survival in a global setting.”
It is not hard to find many other researchers and linguists who stress the importance
of communicative skill. In Littlewood‟s (1981: 89) words, one‟s ability to “use real and
appropriate language to communicate and interact with others” is – and should be – “the
primary goal of most foreign language learning”. Since the adoption of the communicative
language teaching approach, the prioritized focus has been put on the achievement of
functional abilities in the target language. Language learning has therefore shifted from a
grammatical perspective to a communicative perspective that emphasizes the ability to use
the linguistic system effectively and appropriately in the target language and culture.
Similarly, Campbell and Wales (1970, cited in Nguyen, 2006: 1) stress that in order to
speak competently, not only does a person need to know the grammar and pronunciation

rules of a language but s/he is also required to discern what to say, to whom, in which
circumstances, and in what manner. It is understandable why many acknowledge the
importance of knowledge in a foreign language beyond grammar and vocabulary.
One of the aspects regarding knowledge beyond grammar and vocabulary is the use
of discourse markers (hereafter DMs) in oral communication, which can help to make the
speaker sound like a native. This aspect, however, seems to be ignored though they belong
to the speaker‟s delivery, particularly his/her fluency, which is said to define the way the
listeners perceive the speaker (Croucher, 2004: 38). The lack of linguistic devices such as
DMs may account for the fact that in social interactions in English, some ESL/EFL
learners may unintentionally come across as “abrupt or brusque” (Lee, n.d.). In other
2


words, the omission of DMs might make the speakers appear impolite or the speech might
appear somehow deficient. The following two conversations given by Lee (n.d.) can
clearly exemplify the above point:
Conversation 1
Speaker A: Could I borrow your car?
Speaker B1: No, I think I need it tonight.
Speaker B2: Well, I think I need it tonight.
Conversation 2
Speaker C: Could you take me to the
airport next weekend?
Speaker D1: I’m going to be out of town next
weekend.
Speaker D2: Well uh, I'm going to be out of
town next weekend.
In responding to the requests, thanks to the presence of DMs (Well, uh…),
Speakers B2 and D2 sound more polite and less abrupt. The messages therefore reach
Speakers A and C in a less extreme way although their requests have been refused.

Understandably, abruptness may especially happen most commonly in direct
interactions in which no time is given for the interlocutors to think (Östman, 1982 and
Croucher, 2004: 41). Having to think of ideas and select words to utter at the same time,
these students may encounter problems of on-line discourse production (Chaudron and
Richard, 1986). As a matter of fact, the use of DMs such as well, you know, right, okay, I
mean, etc. can help fill the pauses. In fact, these items were considered performance errors
for a long time (Moreno, 2001: 130). Luckily enough, their important roles in
conversations and speeches have been increasingly acknowledged in a growing number of
recent studies. Stenström (1994: 17) believes that a conversation is “much less lively and
less 'personal' without [DMs] signaling receipt of information, agreement and
involvement”. As stressed by Moreno (2001: 130), DMs not only help to “build
coherence”, but they also “fulfill multiple interactive functions fundamental to the speaker
– hearer relationship”. Although DMs have been viewed from a variety of perspectives and
approaches, it is generally agreed that DMs contribute to the pragmatic meaning of
utterances, thereby playing an important role in the pragmatic competence of the speaker.
3


Similarly, the significance of the use of DMs is asserted in Moreno‟s (2001: 130-131)
study that “they form part of the L2 students‟ pragmatic competence”. Never before has
the term “pragmatic competence” been mentioned so frequently in the body of research on
second language (L2) learning and foreign language teaching. Defined as “an aspect of
communicative competence [which] refers to the ability to communicate appropriately in
particular contexts of use” (Jaworski, 1998: 249, cited in Müller, 2005: 18), pragmatic
competence is acknowledged as part of what a student should learn about a language. DMs
are much related to this kind of competence since they “constitute an intrinsic part of one‟s
communicative competence” (Wei‟s, 1996: 2).
In Vietnamese language teaching and learning context, many learners of English,
even English-major ones, find it difficult to communicate with foreigners due to the fact
that they lack the strategies to employ in conversations. Meanwhile, DMs have been

relatively neglected despite their claimed function as useful communicative strategies.
Being intrigued to further study the EFL speakers‟ use of DMs in interactions, the
researcher conducts an investigation of the issue among English-major students in ULIS
who are trained to become teachers and interpreters or do business-related jobs in the
future. It is undeniable that those students, in their prospect jobs, are required to
communicate frequently with foreigners. Nonetheless, owing to some reasons, students do
not seem to pay due attention to make such interactions as authentic and native-like as that
of the native speakers.
All the aforementioned grounds have genuinely inspired the researcher to conduct a
study entitled “Discourse Markers in Oral Interaction by Third-year ULIS Mainstream English
Majors”. It is hoped that the research would contribute to the teaching and learning of DMs
in classroom context, thereby enhancing the students‟ performance in learning foreign
languages.
2. Aims and objectives of the study
First and foremost, the current interlanguage pragmatic research is conducted with
the aims to identify the most frequently-used DMs produced by the students in their
interaction with native speakers and then to some extent, to identify the functions of those
4


markers. Based on the findings, the students‟ and teachers‟ attitudes towards the use of
DMs are scrutinized, paving the way for the pedagogical implications on the teaching of
DMs in classroom. Finally, some suggested exercises and activities for practicing DMs in
English are proposed for those who want to make their conversations authentic and native-
like. It is also hoped that the paper can raise the awareness of utilizing effective DMs in
communication.
In order to achieve the above-mentioned aims, the study has been carried out to
address the four research questions as follows:
1. What are the most common DMs used by third-year ULIS mainstream English
majors in their spoken interaction with native speakers?

2. What specific functions do DMs perform in their spoken discourse?
3. What are the students‟ perceptions towards the use of DMs in speaking?
4. What are the teachers‟ perspectives towards the teaching of DMs to the students
in classroom setting?
3. Scope of the study
There are a great variety of DMs which are classified in a number of ways by
various researchers. However, the present study would like to focus on four DMs including
well, you know, I mean, and like since they are among the items that are mostly frequently
used and universally identified in speech as DMs (Lee & Hsieh, 2004: 179-180). More
justifications regarding the choice of DMs will be presented in Section 1.1.3. The
researcher aims at investigating the use of those markers by Vietnamese learners of
English to see whether those speakers can create native-like speeches.
Notably, DMs are researched in conversations between Vietnamese EFL learners
and native speakers of English. It is the researcher‟s endeavor to explore the issue in the
speech of third-year ULIS mainstream students majoring in English. Despite not being
investigated, the native speakers are present in those interactions to help obtain as much
information from the students as possible, with the intention of enhancing the latter‟s
participation in the talks. The students‟ nonverbal communications are also beyond the
5


scope of the study; therefore, their nonverbal signals would not be tracked throughout their
spoken discourse.
4. Contribution of the study
Theoretically, this study will throw light on a research area which not many
Vietnamese researchers have ever addressed – the use of DMs by EFL/ ESL learners in
interactions. Practically, as one of the initial studies in the area in the context of Vietnam,
the current research could be useful for students, teachers, educational managers as well as
researchers who are interested in the topic.
Specifically, since the study investigates ULIS students‟ deployment of DMs in

their interactions with foreigners, its findings will help these students as well as other
English-major students improve their performance in oral communication. Meanwhile,
teachers, after going through this research, will be more aware of their students‟
communicative performance, and thus will be able to help them gradually develop their
pragmatic competence. As for educational administrators, the paper would provide them
with a close and comprehensive view into the current situation, which may then reveal
some pedagogical suggestions. Finally, researchers who share the same interest will find
helpful information from this research to conduct further studies into this relatively new
and so far ill-explored issue in the EFL context of Vietnam.
In general, students, teachers, educational administrators and researchers are those
who are likely to benefit from the study. The present paper is hoped to contribute to a
small but growing amount of research into the speaking performance of Vietnamese
speakers of English.
5. Organization of the study
The thesis consists of three main parts:
PART I: INTRODUCTION
The rationale, aims, scope of the study as well as its methodology are presented in this
initial part.
6


PART II: DEVELOPMENT
The main part of the study is divided into three chapters.
Chapter 1 (Literature Review) lays the theoretical foundations for the whole study,
including the definitions of key terms as well as a concise review of related studies on
DMs undertaken by researchers in the world and in Vietnam.
Chapter 2 (Methodology) elaborates on the research methods, participants,
instruments, data collection procedure and data analysis methods.
Chapter 3 (Results and Discussion) presents the results of the research and gives
interpretation and analyses of major patterns found in the data. Besides, the discussion

referring back to the literature in the research areas is also included to show the similarities
and differences in the findings.
Chapter 4 (Some pedagogical implications and suggested DMs teaching resources)
puts forwards some implications for EFL learners and teachers of English. It also proposes
some recommended activities and two explicit DMs teaching frameworks for teachers‟
references.
PART III: CONCLUSION
This part summarizes the major findings of what has been carried out in the
research. Subsequently, the limitations of the research are also pointed out before some
suggestions for further studies are made. Following this chapter are the references and
appendices.
7


PART II: DEVELOPMENT
CHAPTER 1: LITERATURE REVIEW
This chapter, as its name suggests, elucidates the literature related to the study,
thereby helping to lay concrete foundations for the development of the succeeding parts of
the paper. Initially, the term “Discourse markers” and its characteristics will be defined.
Following the characterization of the key term is a comprehensive review of related studies
to reveal the research gap which will be bridged to a certain extent by the present study.
1.1. Discourse markers
DMs have been the focus of many studies, becoming a matter of concern to
researchers from the 70s onwards. The general consensus in the literature is that DMs are
difficult to fit into traditional grammatical categories. However, little agreement has been
reached on their terminology, definition and classification. This section begins with the
justification for the use of the term “discourse marker” (DMs), followed by their widely-
accepted characteristics. It then moves towards the selection of DMs in the current study.
1.1.1. Terminology
Research has shown the fundamental roles of DMs in interactions (Carter and

McCarthy, 2006) due to their high frequency of occurrence in spoken discourse. During
the last few decades, numerous studies have dealt with DMs under a number of names
including “discourse signaling devices” (Polanyi and Scha, 1983, cited in Yang, 2011),
“discourse particles” (Schourup, 1985), “pragmatic particles” (Östman, 1995), “discourse
connectives” (Blakemore 1987, cited in Huang, 2011: 22), “discourse operators” (Redeker,
1991), “discourse markers” (Schiffrin, 1987; Stenström, 1994; Jucker & Ziv, 1998; Fraser,
1990, 1999; Trillo, 2002; Müller, 2004; Carter and McCarthy, 2006), inter alia. The last
term (i.e. discourse markers) is probably the most frequently-used term and it is regarded
as a broad covering term (Lewis, 2006; Jucker and Ziv, 1998). Besides, concerning DMs‟
functions, this term is also used under such labels as: “discourse connectors, turn-takers,
8


confirmation seekers, intimacy signals, topic-switchers, hesitation markers, boundary
markers, fillers, prompters, repair markers, attitude markers, and hedging devices” (Jucker
and Ziv, 1998: 1-2). A generalization of those linguistic labels of DMs and many others is
summarized in the following table (Table 1).
Label
Example
 backchannels/ backchannel cues
 Verschuren, 1999
 continuatives
 Trillo, 1997
 cue words
 Horne et al, 2011
 discourse markers
 Schiffrin, 1987; Jucker & Ziv, 1998; Fraser,
1990, 1999; Trillo, 2002; Müller, 2004;
Carter and McCarthy, 2006
 discourse signaling devices

 Polanyi & Scha, 1983
 discourse connectives
 Blackmore, 1987, 1992
 discourse operators
 Redeker, 1991
 discourse particles
 Goldberg, 1980; Schourup, 1985
 fillers
 Brown & Yule, 1983
 gambits
 Keller, 1979
 linguistic markers
 Redeker, 1991
 pragmatic expressions
 Erman, 1992
 pragmatic devices
 van Dijk, 1979
 pragmatic operators
 Ariel, 1994
 pragmatic particles
 Östman, 1995
 utterance particles
 Luke, 1990
Table 1: Terminology variations of DMs (adapted from Yang, 2011)
The aforesaid terms are coined based on different theories and approaches, which
are not discussed here due to space limit. In this paper, the researcher adopts the term
“discourse markers” which was seen by Schourup (1999: 228) as “the most popular of a
host of competing terms used with partially overlapping reference”. This term is,
moreover, deemed as “more popular and theoretically neutral” (Huang, 2011).
Despite their different labels in previous studies, DMs are quite numerous and very

easy to be spot out in spoken discourse. Typical DMs are linguistic items or expressions
which belong to distinct word classes “as varied as conjunctions (e.g. and, but, or),
interjections (oh), adverbs (now, then), and lexicalized phrases (y’know, I mean)”
(Schiffrin, Tannen, and Hamilton 2001: 57). Schiffrin (1987) is said to be among the first
scholars to instigate the characterization of DMs as her works have been cited in many
9


linguistic research and articles. DMs are defined by Schiffrin (1987: 31) as “sequentially
dependent elements which bracket units of talk”, which means nonobligatory utterance-
initial items that function in relation to ongoing talk and text (Schiffrin, 1987: 31). As a
matter of fact, DMs often help to make interactions coherent by bringing together the
different aspects of discourse in a meaningful way. Fraser (1999) describes DMs as:
a class of lexical expressions drawn primarily from the syntactic classes of conjunctions,
adverbs, and prepositional phrases. With certain exceptions, they signal a relationship
between the interpretations of the segment they introduce, S2, and the prior segment, S1.
They have a core meaning, which is procedural, not conceptual, and their more specific
interpretation is "negotiated" by the context, both linguistic and conceptual.
(Fraser, 1999: 931)
Fraser‟s (1999) specification is once again supported in Levinson (1983)‟s idea that
DMs refer to the expressions which indicate the relationship between an utterance and the
prior discourse. In the same vein, Redeker (1991: 1168) defines DMs as words or phrases
that are uttered with the key function of drawing to the listener‟s attention a particular kind
of linkage of the upcoming utterance with the immediate discourse context.
1.1.2. Characteristics of DMs
In general, there is no clear consensus about the definition of DMs and the question
as to which words and phrases should be treated as DMs remains confusing. Despite this,
the following characteristics are generally agreed upon by most researchers as the basic
criteria. Among these, common features such as connectivity, optionality, non-truth
conditionality, flexibility of position and multigrammaticality are discussed below.

Connectivity
The clearest attribute of DMs, as can be seen in many a definition, is their use to
connect the messages in ongoing interactions. Connectivity refers to both the “local” and
“global” link. At the local level, Fraser (1990: 383) defines DMs as signals of “a sequential
relationship between the current basic message and the previous discourse”, which means
the link between two adjacent textual units, as in the following example:
10


(1.) I was searching for the key in the dark with just a torch, and suddenly I heard a loud
scream…
As can be easily identified, the and is a DM used to link two neighboring parts of a
narrative.
Apart from the “local” relation, Lenk (1998, cited in Huang, 2011: 26) adds that
DMs‟ coherence-building function exists at a more global level within the discourse. In his
viewpoint, DMs signal the relationships with other segments of the discourse such as
earlier topics or the contexts or situations. This global level means that the relations
between two textual units do not need to be adjacent (Fraser, 1999). The instance below
best summarizes this point:
(2) [On seeing someone carrying lots of parcels]
So you‟ve spent all your money.
(Blakemore 1987: 85, 106, cited in Huang, 2011: 27)
In Example (2) above, the speaker uses so to connect his or her utterance to the context.
Other DMs such as right, OK and now seem to suggest a different kind of
relationship of connectivity. These three DMs might be used to open a topic, for instance
when a teacher begins a lecture with OK or now. They mark the disconnection from the
previous utterance (e.g. students‟ chats) and the context (e.g. a break between lectures).
Optionality
Another characteristic which is generally accepted is that DMs are optional, both
syntactically and semantically. In other words, the addition or removal of a DM does not

affect the syntactic value of the sentence and a DM does not create meaning (Fraser, 1990)
and will not alter the conceptual meaning of the utterance. Similarly, Rouchota (1998, cited
in Jucker and Ziv, 1998) concludes that DMs are optional and can be omitted and such
omission does not affect the truth values of the propositions in which they occur. This also
coincides with Akande‟s (2008: 81) idea that DMs has a “peripheral” relationship with
their surrounding sentence or clause since they can be removed “without causing any
11


damage to the structure of the sentence or clause”. This can be clearly seen by the fact that
when the DM is left out, the grammaticality of a sentence remains unchanged.
Consider the following two examples:
(3) I like this subject better.
(4) you know, I like this subject better.
In traditional grammar, words classed as adjective and adverbs can be syntactically
optional. In Example (3), better can be deleted and the clause is still syntactically correct
yet it loses some semantic meaning. In Example (4), you know can be omitted without
affecting the utterance‟s syntactical or semantic values.
In short, though being optional, DMs are by no means considered useless,
irrelevant or redundant items. In fact, they do facilitate the process of interpretation in
interactions (Fraser, 1990; Carter and McCarthy, 2006; Fung and Carter, 2007). Serving
certain functions in spoken discourse, DMs make the utterances becomes more explicit.
Flexibility of position
Depending on their functions, DMs may appear at the initial, medial or final
position of an utterance. The following set of examples illustrates the various positions of
DMs:
(5) … Well, I believe that a better solution is needed here… like… very strict punishments
for those who break the laws, like doing community service or paying very high fines.
(6) A: So you are allergic to animals?
B: Yes, well, but it doesn‟t mean that I don‟t like cats, you know.

Multigrammaticality
DMs come from different grammatical classes and they cannot be grouped under
any single grammatical category. They can be adverbs (now, then, therefore), verbs (look,
12


say, see, listen), conjunctions (and, but, also, nevertheless), interjections (oh, well)
sequencing conjuncts (first, next, finally), or non-finite clauses (to be frank, to be honest, I
mean, you see, you know). According to Hansen (1998, cited in Fung, 2003: 58), they are
intermediate between grammatical and lexical items.
Prosodical independence
DMs have to be prosodically independent (Fung and Carter, 2007: 413). Schiffrin
(1987: 328) suggests that a DM “has to have a range of prosodic contours, e.g. tonic stress
and followed by a pause, phonological reduction”. Therefore, the prosodic clues that go
with DMs include pauses, phonological reductions and separate tone units which are
distinguished from other linguistic items in the discourse units bearing the same
manifestation.
All things considered, among those listed characteristics, it should be noted that
any attribute alone is only a necessary, not a sufficient condition for the verification of DM
status. Instead, a combination of criteria needs to be taken into consideration to identify an
item as a DM.
1.1.3. Selection of DMs in the present study
As Jucker (1993: 436) noted, no research has been able to come up with an
exhaustive list of DMs either in English or in any other languages. However, it is not
difficult to identify some prominent DMs in spoken discourse which are acknowledged as
DMs by various analysts. For this research, four DMs to be investigated include well, you
know, I mean, and like. The selection of DMs is grounded on theoretical basis. To be more
specific, these items, also known as fluency devices, are often used by interlocutors to
maintain the floor while thinking of what to say next in the stream of speech (Ngo Huu

Hoang, 2002: 74, cited in Nguyen Thi Hong Nga, 2006: 29). For that reason, these DMs
are regarded as effective interactive strategies which should be employed by speakers to
attain successful communication goals.
The four items well, you know, I mean, and like as DMs were selected as they
appear frequently in native corpora. Their status as DMs in oral discourse is disclosed by
13


Schiffrin‟s (1987) preliminary study on DMs. These DMs, according to Trillo (2002), are
reported to have the highest frequency of occurrences in native discourses in the London-
Lund Corpus, except for affirmative or negative elements, i.e., yeas, yeah, no and the
hesitation marker um. In addition to that, Fox Tree and Schrock (1999, cited in Lee and
Hsieh, 2004: 180) assert that the employment of well and I mean is among the most salient
features of spontaneous talk. According to Stenström (1994), I mean, well and you know
are “peculiar to spoken discourse” and they are used frequently in native-speakers‟ spoken
discourse to start, carry on, and terminate the conversation, or organize or hold the turn,
mark boundaries in the discourse.
Regarding like, this item has become a common DM among NSs and received
much attention from a number of researchers such as Jucker and Ziv (1998), Fuller
(2003b), Müller (2005) and Huang (2011). Like as a DM is claimed to be frequently
attributed to young people (Andersen, 2001 cited in Müller, 2005: 202).
1.1.4. Main functions of four selected DMs
While you know, I mean and well have a number of individual functions which are
analyzed as working at the textual and at the interactional level, like only functions at the
textual level (Müller, 2005: 242; Huang, 2011). Textual functions organize the content of
what is said or mark (parts of) utterances as specific types of utterances, while interactional
functions address the hearer directly or organize the sequence of turns between the
participants.
Following is the summary of the selected DMs together with their main functions.
Reviews of the literature on each marker under investigation are to be briefly provided,

compared and contrasted with the student data in the analysis chapter.
Level / DMs
Textual Level
Interactional Level
well
 (as a delay device) searching for the
right phrase
 rephrasing/correcting

 acting as a face-threat mitigator
 marking continuation in speech
14


you know

 marking reported speech
 prefacing important information
 signaling a search for lexical words or
content information
 indicating clarifications and
explanations
 signaling exemplifications
 marking approximations.
 marking shared or general
knowledge
 appealing for acceptance
 appealing for patience,
understanding and sympathy
 softening the force of utterance

 acting as a question.
I mean

 marking a self-repair
 elaborating, clarifying, modifying or
expanding what has been said.
 orientating the speaker’s talk
 acting as a mitigator
like

 searching for the appropriate
expression
 marking an approximate number or
quantity
 introducing an example



Table 2: Main functions of the four DMs (Adapted from Müller, 2005: 246 and Huang, 2011)
It can be seen from the table above that interactional functions are particular to
each discourse marker; in contrast, textual functions can be similar across different
markers.
Those markers were asserted by Müller (2005: 26) to hold “a range of different
characteristics which might potentially be relevant for the usage by EFL speakers”.
However, their frequency of occurrence in a person‟s speech should be appropriate so that
his/her speech is not seen as irritating (Müller, 2005).
All in all, the present study aims at scrutinizing the use of DMs including well, you
know, I mean, and like as produced by third-year ULIS students. The next part is dedicated
to a brief overview of third-year ULIS mainstream English majors – the informants in the
present study.

15


1.2. Previous studies on DMs
Research on DMs has steadily proliferated and diversified since the 1980s. Most of
the studies provide analyses and descriptions of DM use in different languages. In
Schiffrin‟s (2001) conclusion, DMs have been examined in an array of genres and
interactive contexts, and in a variety of different language contact situations. Within the
scope of the current study, only those related to DM use in English spoken discourse will
be reviewed.
1.2.1. Related studies worldwide
Studies on DMs and L2 learners’ listening comprehension
Although the present paper does not study DMs to check L2 learners‟ listening
comprehension, the researcher still spares this part to reflect the value of DMs in spoken
interaction, which lies in the comprehensibility of speech production.
Jung (2003) conducts a study on the role of discourse signaling cues in L2 listening
among 80 Korean learners of English as a Foreign Language. This study was based on
Chaudron and Richards (1986)‟s classification of cues into macromarkers (cues used to
signal the relationship among main segments or to mark the major transition points in
discourse; e.g. what I’m going to talk about today…) and micromarkers (cues indicate
intersentential relations or to function as pause fillers; e.g. and, so and well). His aim was
to discover whether and how the combination of macromarkers and micromarkers affect
L2 listening comprehension. The study has demonstrated that the group listening to a
lecture with cues recalls significantly more high- and low-level information when
performing summary tasks than the group listening to a lecture without such cues.
With a similar ultimate aim, Eslami and Eslami-Rasekh (2007) investigates the
effect of the use of DMs on academic listening comprehension of Iranian university
students in an English as a Foreign Language setting. Two groups of students listened to
two different versions of a lecture. The two versions were dissimilar with respect to the
quantity and type of DMs used. Listening comprehension tests and their mean scores were

16


then compared. The findings clearly indicate that subjects comprehended the lecture better
when DMs were included than when they were omitted.
The value of DMs in aiding listening comprehension has also confirmed by Tyler
and Bro (1992):
in any communicative situation, participants bring a set of expectations
concerning how discourse-structuring cues signal relationships among the expressed
ideas When [these] cues are missing or are used in unexpected ways [listeners]
find a meaningful interpretation difficult to construct, and therefore judge the
discourse as incoherent.
(Tyler and Bro, 1992: 74-75, cited in Wei, 1996: 34).
It should be hence interpreted that the absence or misuse of DMs fails to provide
appropriate direction to the hearers and may prevent them from building a coherent
interpretation of the discourse (Wei, 1996: 35).
Studies on NNSs’ use of DMs in oral communication
As DMs received much attention due to their significance in spoken discourse,
a small but growing number of studies have been undertaken on the employment of DMs
by L2 learners. Some of those studies consider the uses of the DMs by NNSs while some
on the comparative DM usage by NSs and NNs. This section will first deal with some most
widely-cited and influential studies on the use of DMs by NNSs in oral discourse.
Müller‟s (2005) research, based on a 350,000-word corpus of spoken English by
American NSs and German NNSs, provides a detailed analysis of the frequencies and
functions of the four DMs so, well, you know and like. Her study was well-designed for
collecting comparable data from silent movie narratives and discussions. Rather than
adopting an existing framework, Müller (2005) manually identified the functions of the
four DMs and classified them at two levels, the textual and the interactional level. The
functions at the textual level are not used to address the listeners but focus on
lexical expressions (e.g. a speaker‟s search for words, restarting and repairing), the

structure of propositional contents (e.g. explanations and exemplifications) and the
17


distinction between the speaker‟s voice and reported speech. The functions at the
interactional level work for the relationship between speakers and listeners by marking
a speech act, response, opinion, evaluation, appeal to the listener, etc. So, you know and
well have been found to serve a number of functions at both the textual and interactional
levels. Like only functions at the textual level. It has been remarked that German students
use the DMs so, you know and like less frequently than American students do and the DM
well with similar frequency; some functions are used only by American students and some
only by German students.
Unlike Müller‟s compilation of two comparable corpora, Wang and Zhu‟s (2005)
research investigates fifteen types of DMs in the Spoken English Corpus of Chinese
Learners (SECCL) and the spoken component of the British National Corpus (BNC). The
former consists of Chinese-speaking NNSs‟ monologues and dialogues, while the latter
includes NSs‟ informal conversations. Three main differences between Chinese NNSs and
NSs in the use of DMs were highlighted. First, the NNSs and NSs use different types of
DMs. Second, the NNSs underuse DMs in terms of frequency and type. Finally, the NNSs
overuse some of the additive and emphatic DMs, such as and, but and very and fillers with
semantic meaning, such as I think. Huang (2011) notes that using the terms under- and
over-representation of a particular DM would be more appropriate than the terms underuse
and overuse in this case.
Another of the few studies carried out on the use by NNSs of DMs is Fung and
Carter‟s (2007). The two researchers compared the use of DMs by NSs and Hong Kong
NNSs in pedagogic settings. Based on a 460,055-word sub-corpus of the Cambridge and
Nottingham Corpus of Discourse in English (CANCODE) and a 14,157-word learner
corpus of interactive classroom discourse, they examined a wide range of DMs in a
framework of interpersonal, referential, structural and cognitive categories. Native
speakers are found to use discourse markers for a wider variety of pragmatic functions.

More specifically, the Hong Kong NNSs extensively use referential and structural DMs
(and, but, because, I think) but have limited use of others that are frequently used by the
NSs (and, right, yeah, well, so, now, sort of, you know, actually, see, say and cos).
18


Huang‟s (2011) paper explores the use of DMs in the speech of Chinese NNSs of
English and NSs using corpus methodologies. It reports that the DMs for analysis, like, oh,
well, you know, I mean, you see, I think and now, occur more frequently in the dialogic
genres than in the monologic genres extracted from the Chinese NNS corpus (SECCL) and
two NS corpora (MICASE and ICE-GB). This supports his hypothesis that the more
interactive the genre or activity type is, the more DMs occur. In most cases, the
markers appear less frequently in NNSs‟ speech than the native one.
While the above studies compared DM use by NSs and NNSs through two
respective corpora, some others approach the same issue using the NNSs only.
Hays‟s (1992) work with the classroom oral discourse by some Japanese learners of
English studied seven DMs including both ideational and interactional markers (cited in
Müller, 2005). Only three students were found to be able to use well, while the great
majority of students were able to use and, but, and so ideationally. Hays (1992) claims that
ideational DMs are acquired earlier (cited in Müller, 2005: 67) because they are overtly
taught while markers on the interactional plane seem to be neglected. This study makes us
aware of a possible difference in the acquisition of the two different types of DMs by
nonnative learners; more competent learners are expected to be better able to use
interactional DMs.
Moreno (2001) also investigates NNSs‟ use of DMs based on the data taken from
the direct interaction between NNSs and NSs. In his study, Moreno (2001) analyzed some
DMs such as well, you know, right, okay, I mean, etc. which appeared in fifteen
conversations in English between NNSs and NSs. The NNSs were undergraduate students
in their third, fourth and fifth year of English Language and Literature at the University of
Seville. Her analyses indicate that those items are hardly used as DMs in the students‟

discourse, resulting in distinctly non-native discourse, which can negatively affect the
students‟ images.
Likewise, Chen (n.d.) compared the use of DMs in interactions between natives and
non-natives when social roles changed. Specifically, six English as a Foreign Language
(EFL) college freshmen and six English as a Native Language speakers (ENSs) took part