Introduction
The importance of developing pragmatic
competence has long been acknowledged by
researchers in the field of second language
acquisition (Bardovi-Harlig, 1999; Kasper &
Rose, 1999; Kasper & Schmidt, 1996). In
fact, successful communication requires the
mastery of social usage as well as linguistic
forms (Glaser, 2009).
Among all communicative activities,
argumentative discourse is one that
permeates all aspects of life, and performing
well in this discourse domain is an important
pragmatic skill for every person, either in
the first (L1) or second language (L2)
(Dippold, 2011). To negotiate ones' own
ideas successfully and perform well in
argumentation, a person should acquire this
important pragmatic skill. Therefore,
understanding of how arguing is
accomplished would contribute to the
understanding of the negotiation of social
structures and vice-versa.
Most of the research on argumentative
discourse focuses on the expression of
disagreement (Hayashi, 1996; Holtgraves,
1997; Locher, 2004; Rees-Miller, 2000).
This is because conversation analytic
approaches to the study of argument see
disagreement as the ‘marked answer’
(Dippold, 2011). Disagreement is defined as
“the expression of a view that differs from
that expressed by another speaker”
(Sifianou, 2012, p. 1). This speech act is
generally considered dispreferred
(Pomerantz, 1984) because it threatens the
speaker’s face (Brown & Levinson, 1987).
However, researchers studying argument
and conflict talk in interaction have
illustrated that, in arguments, ordinary
preference structures are sometimes
removed or even reversed; that is,
disagreements may take preferred forms,
while agreements are produced as
dispreferred (Kotthoff, 1993). Disagreement
has also been observed to enhance
sociability and relationship in some contexts
(Angouri & Tseliga, 2010; Georgakopoulou,
2001; Kakavá, 2002). In recent research on
disagreement, it is believed that
disagreement is an "everyday phenomenon"
that is needed in decision making and
problem solving interactions (Angouri &
Locher, 2012, p.1). It is suggested that
context plays an important role in the
acceptance of disagreement as a preferred
speech act.
Due to such conflicting views on the nature
of disagreement as preferred or dispreferred,
and in order to investigate possible effects of
context on the expression of disagreement,
the present study investigates disagreement
strategies used in arguments in the context
of English as foreign language.
Background
Disagreement: A multidirectional and
multifunctional act
Brown and Levinson’s (1987) politeness
theory has been applied to various studies of
speech act realization and conversational
interaction (Garcia, 1989). Brown and
Levinson (1987) considered disagreements
as acts which threaten the addressee's
negative face when “a speaker is imposing
her/his will on the hearer' (Sifianou, 2012, p.
65).
Due to this negative-face threatening aspect
of disagreements, Brown and Levinson
(1987) proposed two positive politeness
strategies to avoid this threat: 1. 'Seek
agreement' (e.g., by engaging in safe topics),
and 2. 'Avoid disagreement' (e.g., by using
token agreement, hedging, and white lies)
(p. 112–113). They suggested that more
direct strategies of disagreement are
preferred to less direct strategies in three
situations: 1. when there is less social
distance between the speaker and addressee,
2. when the speaker has greater power than
the addressee, and 3. when the severity of
disagreement is less.
The notion of preference can best be
explained by conversation analytic work of
Pomerantz's (1984), according to which
participation in speech act involves making
assessments, “with an assessment a speaker
claims knowledge of that which he or she is
assessing” (p. 57). Initial assessments are
followed by second assessments which are
“subsequent assessments that refer to the
same referents as in the prior assessments”
(Pomerantz, 1984, p.62). Pomerantz viewed
disagreements as dispreferred second
assessments; therefore, turns and sequences
in talk should be structured so as to soften
the disagreement. As a result, disagreements
are expressed with delayed components and
in a way that they are not positioned early
within turns. To redress the threats to the
addressee’s positive face the speaker may
use partial agreement, colloquial language,
and first person plural. The use of
interrogatives, hedges, and impersonal forms
can soften the threat to the addressee's
negative face (Brown & Levinson, 1987).
All in all, the above theories view
disagreement as “a form of conflict . . .
taxing communication events” (Waldron &
Applegate, cited in Locher, 2004, p. 94),
dispreferred second (Pomerantz, 1984;
Sacks, 1987), which 'is largely destructive
for social solidarity' (Heritage, 1984, p. 268)
and should, therefore, be avoided in the
interest of interlocutors’ face (Brown &
Levinson, 1987; Leech, 1983). However,
recent studies have led to new and somehow
contradictory findings regarding the
expression of disagreement.
Angouri and Locher (2012), in their review
of the literature on disagreement, provided
new insights for future research on
disagreement. They proposed that
disagreement is an everyday speech act
which is expected in certain interactional
practices such as problem solving and
decision making. According to them, it is
erroneous to consider disagreements as
primarily negative; various aspects of
context, culture and social norms and
practices determine the nature of
disagreement as a preferred or dispreferred
speech act.
Sifianou (2012) describes disagreement as
“a situated activity, interactionally managed
by interlocutors” (p.4), which is a
multidirectional (i.e. disagreements can
affect either or both positive and negative
face of the interlocutors) and multifunctional
(i.e. disagreements can serve various
functions such as establishing hostility or
solidarity) speech act. She believes that the
conceptualization of disagreements
primarily as face-threatening acts which
should be avoided in favor of agreements is
only a part of the story. She argues against
such views as follows:
they ignore the possibility that
even agreements may be face
threatening if, for instance, they
are interpreted as insincere,
manipulative or ingratiating.
Moreover, agreements may also
be self-face threatening acts if
one feels impeded in voicing
one's own views openly and
freely. (Sifianou, 2012, p. 6)
On the other hand, she emphasizes on the
face-enhancing function that disagreements
may play in various situations. For example,
the speaker may display interest in his/her
interlocutor's argument through
disagreement (i.e. through involvement in
interaction rather than indifference by just
agreeing or being silent) or help in
investigating different perspectives in a
discussion to find a solution which is helpful
to the addressee (Georgakopoulou, cited in
Sifianou, 2012). Disagreements can also be
face enhancing in the case of self-belittling
statements or even compliments. In these
cases, the speakers protect their face through
disagreement as agreeing with self-praise is
face-threatening (Pomerantz, 1984).
Disagreement may also enhance one’s face
when speakers disagree to present
themselves as skillful contesters
(Hernandez-Flores, 2008; see, also, Angouri
& Tseliga, 2010; Corsaro & Maynard, 1996;
Kakavá, 2002; Locher, 2004; Tannen,
1984).
Therefore, recent studies on disagreement
can help us to view this speech act not
simply as a threat to our interlocutors' face,
but as a multidimensional act which may
foster solidarity among people in their
interactions.
Disagreement and context of interaction
Viewing disagreement as a
multidimensional and multifunctional act
has led researchers to investigate the
expression of this speech act in relation to
the context of interaction. They have tried to
explore the possible effects of contextual
variables such as age, gender, power,
solidarity, personal traits, and the degree of
formality of the interaction on the
expression of disagreement.
Sifianou (2012) believes that context is not
static and simple and disagreement is not by
itself impolite or self-threatening; rather, it
is the context of interaction which makes
disagreement face threatening or face
enhancing. She views interlocutors' personal
traits and relational histories as influential in
predisposing them to particular strategies:
Some individuals are more
argumentative than others,
and some may be aversive to
any kind of
opposition…some people
object to certain kinds of
FTAs [face-threatening acts]
more than others. It is highly
likely that such personal
traits will influence both
interlocutors’ linguistic
behavior. (Sifianou, 2012, p.
5)
Parvaresh and Eslamirasekh (2009)
investigated the effects of contextual
variables of solidarity and deference on the
ways in which young women in Iran argue
in their first language (i.e. Persian). They
have concluded that the non-western culture
of Iran causes the interlocutors to seek
deference rather than solidarity while
disagreeing with their close male friends.
However, they observed that their
participants used ‘conflictives’ in cases
where their interlocutor was of the same
gender.
In another study, Mehregan, Eslamirasekh,
Dabaghi, and Jafari Seresht (2013) explored
the effect of gender and the degree of
formality of situation on the expression of
the speech act of disagreement in Persian.
They observed that the degree of formality
of the situation causes their participants to
disagree conservatively.
In another study on the effect of context on
the production of disagreement, Netz (in
press) supported the claim that disagreement
is not inherently face-threatening and needs
to be contextualized. The author studied
disagreements in the gifted classes and
found that in this context disagreement was
unmarked and less mitigated and did not
undermine solidarity among interlocutors.
Georgakopoulou (2001, p. 1881) argues that
future research on disagreement should be
“context-sensitive”. In his study of Greek
conversations between young people, he
found that disagreements were implied and
indirectly constructed through a) turn-initial
markers, b) stories used as analogies for the
debated issues, and c) questions. However,
he argued that this indirectness in the
expression of disagreement was neither an
indication of sociability nor was due to
increased politeness. Instead, he
demonstrated that disagreements in his data
were shaped by contextual factors such as
the participants' relationship, their shared
background information, type of activity and
the norms of argumentation.
Other studies have considered contextual
factors like power, severity of disagreement,
ethnicity, personal traits, relational histories
(Rees-Miller, 2000; Sifianou, 2012), and
professional training (Edstrom, 2004)
important in the expression of disagreement
in interaction. Garcés-Conejos Blitvich
(2009, p. 282) argues for the importance of
interlocutor's identity in the interpretation of
what is said.
The above findings call for further context-sensitive research on disagreement in natural
settings. Therefore, the present study tries to
investigate the expression of disagreement
in arguments through conversation analysis
techniques in various contexts of interaction.
Theoretical framework
In this study the argument transcripts were
analyzed according to Rees-Miller’s (2000)
taxonomy. Rees Miller categorized the
expression of disagreement as softened
disagreement, aggravated or strengthened
disagreement and disagreement which is
neither softened nor aggravated (see Table
1). She justified the use of this taxonomy in
contrast to the existing taxonomies by
Brown and Levinson (1987) and Blum-Kulka, Shoshana, House, and Kasper (1989)
and argued that description of the content of
interaction by using terms like 'head act and
adjunct' (Blum-Kulka et al., 1989)
presupposes a discrete speech act which is
accomplished in one adjacency pair and
does not apply to the natural data of
disagreement.
On the other hand, she argues against using
ambiguous terms such as 'direct and
indirect' (Brown & Levinson, 1987) as a
particular disagreement turn may appear
direct or indirect to different observers. In
her taxonomy, the category of softened
disagreement is divided into positive and
negative politeness. In positive politeness,
the speaker uses some softeners and
linguistic markers that increase solidarity
with the addressee, such as positive
comment and inclusive first person pronoun
(we, us). In negative politeness, the speaker
avoids imposing on the addresses’ autonomy
and uses softeners such as questions or verbs
of uncertainty (Rees-Miller, 2000).
On the other hand, some turns of
disagreement are considered as neither
softened nor strengthened by explicit
linguistic markers. These turns were
recognized as disagreement because they
contradicted their previous utterance by
using a negative, the words ‘yes’ or ‘no’, or
repeating a previous speaker's utterance with
altered words or intonation ('verbal
shadowing') (Rees Miller, 2000, p. 1094).
The other type of disagreement is
aggravated disagreement in which the
disagreement is strengthened by the use of
rhetorical questions, intensifiers, the
personal accusatory you, or judgmental
vocabulary.
Method
The study of speech acts has gained much
attention in analyzing pragmatic competence
of L2 users. However, the method of
analysis has often been discourse
completion tests (DCT) which fail to
provide natural discourse data (Bardovi-Harlig & Hartford, 1993; Bardovi-Harlig &
Salsbury, 2004; Johnston, Kasper, & Ross
1998; Kasper & Dahl, 1991; Rose, 1992;
Rose & Ono, 1995). These written tests are
of limited use in the analysis of talk in
interaction (i.e. discourse) since written
production differs from the actual
conversation in that it allows planning time
and it is non-interactional (Bardovi-Harlig &
Salsbury, 2004). As the study of talk in
actual interaction provides a deeper insight
into what people do with talk, in this study
natural arguments were used instead of
written tests.
Data Collection
All of the participants were informed of the
research purpose before the discussions and
they were asked to express their arguments
on different aspects of the mentioned issues.
The researchers participated in all the
discussions as participant observers. The
recordings were then transcribed by the
researchers. The analysis of the arguments
was focused on the disagreement turns
following Rees-Miller’s (2000) taxonomy of
linguistic markers of disagreement.
Participants
The participants in this study were 26
Iranian English-as-a-foreign-language (EFL)
learners within the age range of 18-50 (see
Table 2). As the context of interaction is an
important factor which influences the ways
disagreement is expressed, the researchers
collected the data in three different settings
to see the relationship between the
expression of disagreement and contextual
factors. The three settings were chosen by
following these criteria: 1. the settings
should have differing degree of formality (to
investigate disagreement in formal versus
informal interactions), 2. There should be
both male and female participants to see the
effect of gender, 3. The participants should
have different relationships (e.g. family,
friends, classmates) to see the effect of
interlocutors' relationship on the ways they
disagree.
A private language institute in Tehran
provided the first setting for the collection of
the data. Fifteen female students who had
enrolled in FCE (First Certificate in English)
and IELTS exam-preparation courses
participated in this setting. The students had
been put in these exam-preparation classes
through a placement test. Their proficiency
levels were B2 and C1 on the Common
European Framework of Reference (CEFR).
They had learned English through
communicative methods in private language
schools. A total of 70 minutes of argument
were recorded on the topic of male and
female roles in society and the educational
system.
In the second setting that was home
environment, there were 5 participants.
Three of them were female and the other
two were male. Two of the participants were
close friends, and the remaining three were
family members. Two of these participants
(family members) had lived in English
speaking countries for 6 years. Arguments
took place in the participants' personal
dwellings. A total of 70 minutes of argument
on the topic of mixed or single-sex schools
and the advantages and disadvantages of
education abroad was recorded in this
setting.
The third setting was a Graduate University
in Tehran. Six MA students of TEFL
(Teaching English as a Foreign Language)
took part in this environment. Of the six
participants three were male and three were
female. They had a 40-minute discussion on
the topic of male and female roles in society
and their cultural background.
Results and discussion
Disagreement turns were identified
according to the definition of this speech act
as an utterance which is ‘Not P’ (i.e.
proposition) in response to a proposition
which is P (Sornig, cited in Rees-Miller,
2000, p. 1088).
The linguistic markers which identify the
type of disagreements were located and the
frequency of each type in the three settings
was found. The results are shown in Table 3.
It has been argued that as learners develop
their pragmatic competence they use more
native-like disagreement strategies; that is
they move toward using more mitigation
strategies and avoiding more direct forms of
disagreement to save both aspects of face
(Bardovi-Harlig & Salsbury, 2004; Behnam
& Niroomand, 2011; Dippold, 2011). In
setting 1, in which participants are preparing
themselves for proficiency exams, there is
little difference in the frequency of different
types of disagreement. However, in
comparison with the other two settings,
there is a higher number of ‘neither softened
nor aggravated disagreements’ (32 %).
A closer look at the data reveals that the
participants in setting 1 have used more
contradictory statements than disagreements
which are neither softened nor aggravated.
This may account for the fact that their
interlanguage competence is developing and
they have not yet mastered strategies to
mitigate disagreement to save the
addressee’s face or their own face. Example
1 shows one instance of direct disagreement:
Example 1
1 S: I think e(hh) men and women equal to
each other in most of the situations because
2 e(hh)I think that e(hh) they have e(hh) they
have some capacity (.) they have some
3 abilities that they are same to each other
and they can e(hh) somehow they can
4 complete to each other …….
5 M: Yes (.) but (.) do you think in our society
men and women are equal ↑? they are
6 not (.)because we- when we are child
(.)when we want to do something they say
7 that let your father or brother do it he is
stronger ↑(.) he can do it better, and
8 we believe that we are weak in these things
and when we grow up we believe
9 that OK↑ we cannot do these things I have
my father he can do it better than me
10 S: So if we (.) OK maybe it’s wrong e(hh)
– I don’t know- it’s a wrong opinion if
11 we accept these things OK in the future all
girls should think like this and after
12 sometimes after some years nothing
change. So we should start from ourselves
13 - we should start- [we
14 M: [They won’t they won’t
let us to start
15 …………
16 S: OK we should change [these things
17 M: [we can’t change
(.) Our father says let me talk with
18 this man or the teacher we can’t say
anything
In this example the speakers expand their
disagreement over a number of turns which
is a sign of higher proficiency level
(Dippold, 2011); however, the use of
contradictory statements as direct
disagreement may be due to their lower
pragmatic competence.
Georgakopoulou (2001) found the type of
activity as an influential factor in shaping
disagreements as face-threatening or face-enhancing. Dippold (2008, p.147)
distinguishes between the
‘argument/discussion frame’ and ‘language
task frame’. In the language task frame
learners try to display their accuracy and
fluency and do not care much about face
requirements. As it was observed, learners
used direct and 'yes, but' disagreement
strategies more than complex and indirect
strategies. It is assumed that in this setting
the participants may have considered the
argument as a language task. However, in
the other settings, which included natural
arguments between friends and family
members and graduate students, the
interlocutors argued in the discussion frame.
The interlocutors’ relational history is also
considered influential in the amount and
types of disagreements expressed in
arguments (Georgakopoulou, 2001;
Sifianou, 2012).
In the first setting, participants were
classmates and they knew each other well.
The outstanding point was that although
there was not any significant difference
among the three types of disagreements
expressed by the participant, softened
disagreement was the most employed type
(40%). Interestingly enough, positive
politeness strategies were employed more
that negative ones (see Table 2). The
participants sought solidarity through using
positive strategies in the expression of
disagreement. On the other hand, the
percentage of expressing softened
disagreements in setting 3 (39.5%) was very
much similar to that of setting 1 (40%). This
can be due to the same relational histories
among the participants of these two settings.
In setting 3 all the participants were
classmates and knew each other for two
years.
In setting 2, in which family members and
friends were present, the lowest total
number of disagreements (15) occurred.
This implies that in this setting disagreement
is considered as a more face-threatening act.
Most of these disagreements (73.5 %) were
softened and more politeness markers were
used to maintain social harmony (Rees-Miller, 2000). Unlike setting 1 in which
positive politeness strategies were employed
more than the negative ones, in setting 2 the
use of negative politeness strategies was
significant (72%). The influence of
participant relationships on the expression of
disagreement is clear here. By using these
negative politeness strategies the speakers
try to avoid direct disagreement. Members
of the family who participated in arguments
were a couple and the wife's father-in-law.
The relationship among these participants in
the Iranian culture may have contributed to
the use of more negative politeness
strategies by the son and her wife to avoid
imposing their ideas on their interlocutor
(father in law).
In the following example the son of the
family starts disagreeing indirectly with
what his father has said by asking a question
and using phrases like 'I don’t know' which
are negative politeness strategies.
Example 2
1 Son: well (.) just before you go to the next
topic speaking about critical thinking
2 and other things … (.) ok for example
something which is not available
3 here in our country but is more valued in
Europe or in I don’t know
4 English system countries is critical thinking ok
but the question I have is
5 that so (.) what is the benefit of going abroad
and doing this critical
6 thinking↓ is it just having a short experience of
being in an
7 environment in which critical thinking is
fostered and you know > what
8 I wanna say is that ok you go there and stay
there for 4 or 5years you do
9 your phd< and when you are back here I mean
at the end of the day
10 you wanna come back to your own country
for example………
11 they still don't have (.) I don’t know (.) that
they don’t have the chance
12 for preparing an environment for critical
thinking one question I have
13 from dad is that do you really believe that the
critical thinking is
14 something that comes with system or is it
dependent on professors or
15 individuals↑
16 Dad: (it’s the system) let me give you an
example…
Among the politeness strategies used in this
context, downtoners such as ‘maybe’ and
verbs of uncertainty such as ‘seems’ and
‘may’ and the preface ‘I think’ were used
more than the others. In the following
example two friends are discussing the issue
of mixed schools. In the disagreement turn,
speaker A uses ‘you know’, ‘I think’ and
‘may’ to soften the disagreement and avoid
imposing her personal view on the
addressee.
Example 3
1S: Because two genders should have some
experience living together growing
2 up and there is no problem (.) it seems there is
no problem if they grow up
3 together (.) but in high school level it seems it
distracts them and=
4 A: =you know (.) I think that it's just more
things about this separation and
5 mixing sexes I think that if we want to
segregate schools or universities or to
6 mix them we should do something basic …
The significant finding in setting 3 was that
the total number of disagreements in this
setting was higher than the other two,
though they took place in a shorter time
period (40 minutes). This may be due to the
fact that graduate students know how to
delve into academic topics. Despite Brown
and Levinson’s conceptualization that all
disagreement acts are face threatening acts
(FTAs), Sifianou (2012) argues that the
context determines if disagreements are
polite or impolite acts. In some contexts
disagreements threaten the addressee’s
positive face to claim solidarity with the
speaker, but in some other contexts such as
political debates or social science
discussions, disagreements present the
speaker as a skillful contester, so it is face-enhancing.
In setting 3 the frequency of aggravated
disagreements is higher than the other two
settings and rhetorical question was the most
often used linguistic marker by the
interlocutors. The use of linguistic markers
which strengthened the disagreement in this
setting can be attributed to the severity of
disagreements. According to Rees-Miller
(2000, p. 1098), severe disagreements
“threaten the personal or professional
identity, worth, beliefs, or values of the
interlocutors. The more personally
threatened the interlocutors feel, the more
severe the disagreement.” In this setting the
topic of argument was people’s ‘cultural
background’ and as participants were of
both male and female genders,
disagreements were expressed severely. The
following example shows a rhetorical
question to aggravate the disagreement.
Example 4
1 M: Ali says that cultural backgrounds
somehow fueling this trend of thought =
2 A: =a (hh) cultural backgrounds say that that
women are weak!? and women cannot
3 be for example in this position ?
In some contexts a severe disagreement that
threatens the speaker’s beliefs and identity
may attract more aggravated disagreement.
In these cases one’s own face is more
important than the addressee’s face. This
was again the case in the third setting in
which graduate students are talking about
people’s ‘cultural background’. The topic
threatens some participants’ beliefs and
makes them use strong disagreements such
as using the personal accusatory ‘you’ and a
verb like ‘must’ in the following example.
Example 5
1 A: yes we are just drawing circle around the
wrong I don’t know e(hh) – negative
2 points and if you do not look at the context and
around that yes cultural background
3 would be the worst thing in our all life so we
should omit it!
5 M: if you say that cultural background is a
complete thing that for example leads you to
6 perfection so you mustn’t ignore the negative
things (.) you know (.)
7 if there is a negative thing
In this example speaker A tries to defend her
opinion, so uses ‘we’ to soften her
disagreement. However, M uses ‘you’ to
force A to consider negative aspects of one’s
cultural background too.
Use of disagreement generally has been
considered a dispreferred second pair part
that is likely to be delayed and elaborated to
enhance politeness. However, in many
situations like the academic setting,
disagreement has been viewed as a preferred
act (Tannen, 2002). In this setting, as
suggested by data, disagreement is a means
of “sociability rather than disaffiliation”
(Sifianou, 2012, p. 11) and the number of
disagreement turns (38) is much higher than
the other two settings.
Another point worthy to be mentioned here
is that participants in setting 3 used softened
disagreements (39.5 %) to the same degree
as aggravated types of disagreement (39.5
%). This may be due to the influence of their
relationship on the expression of
disagreement. As participants in this setting
were classmates and friends, in some parts
they tried to soften their disagreement to
avoid threatening their interlocutors' face.
Furthermore, similar to setting 2, the
proportion of negative politeness strategies
employed in this setting is more than that of
positive politeness strategies. Also, in
setting 2 the proportion of softened
disagreements was much higher than that of
aggravated disagreements. This finding
again reminds us of the argument over face
threatening nature of disagreement and the
belief that as learners develop their
pragmatic competence to the level of native
speakers, they move toward using more
mitigation strategies and avoiding more
direct forms of disagreement to save their
interlocutors' face (Bardovi-Harlig &
Salsbury, 2004; Behnam & Niroomand,
2011; Dippold, 2011). Participants in setting
2 learnt English in an English environment
and were more native-like; similarly, in
setting 3 participants were were highly
proficient in the pragmatic sense.
Conclusion
The findings of the current study suggest
that disagreement cannot be studied without
consideration of context, and that the
linguistic markers cannot simply categorize
disagreement turns into polite/impolite or
preferred-dispreferred acts.
This paper had the advantage of analyzing
longer discourse in natural settings;
however, it was not without limitations.
According to Sifianou, (2012) “our daily
encounters are not finished products but
processes related to previous and future
ones” (p. 8), and that being preferred or
disprefferd acts for disagreements may also
depend on previous encounters of
interlocutors. Sifianou further argued that
controlling the effect of all these factors may
not be possible. Not being an exception, in
our study, some intervening contextual
variables unknown to the researchers might
have influenced our participants' use of
linguistic markers.
Future studies are therefore suggested with a
larger sample size merging various methods
of data collection such as discourse
completion tests and conversation analysis
to gain richer data. As the literature on
disagreement has shown, L1 culture and
social norms (Angouri & Locher, 2012) may
influence the way people disagree. This
study, therefore, suggests future comparative
research to investigate firstly how the speech
act of disagreement is expressed in different
languages and secondly to what extent one's
practice of disagreement in L1 can have
effects on L2.
