Linguistic alignment in L1-L2 dialogue. Purmohammad

Linguistic alignment inL1–L2dialogue
Mehdi Purmohammad
University ofBern
The aim of this paper is to investigate alignment in L1–L2 dialogue. More
specifically, I examine to what extent alignment in L1–L2 is different from
alignment in L1–L1 dialogue. I investigate different variables that may affect
linguistic alignment in thecourse ofL1–L2dialogue.As more variables,such as
differences in language proficiency ofinterlocutors, affect the linguistic align-
ment in L1–L2 dialogue,it appears that linguistic alignment in L1–L2dialogue
is different fromlinguistic alignment in L1–L1dialogue both quantitatively
and quantitatively.Ialso discuss the mechanisms that permit lexical alignment
during dialogue. This study interprets the alignment process in terms of the
activation threshold hypothesis (Paradis 1993) and a link is made between the
activation threshold hypothesis andPickering and Garrods’(2013) accountthat
language productionand language comprehension are interwoven. Based on
Swiss multilingualism, language selection is proposed as the macro-linguistic
alignment process.
Keywords: dialogue,linguisticalignment,lexicalalignment,L2speakers,
L1–L2 dialogue,interlocutors,macro-linguistic alignment
1. Introduction
Everyday language useis mainly in the form of dialogue whichis essentially ajoint
action (Garrod and Pickering 2009). As the conversational process is primarily
not predetermined (see Weigand, 2009), it demands linguistic co-ordination be-
tweenthe twointerlocutors for communication to succeed(Clark 1996; Clark and
Schaefer 1989). Thespeakers’ ability to co-ordinatetheir linguistic behaviour dur-
ing dialogue is fundamental to reaching a common goal. During dialogue, speak-
ers use language in an interactive rather than an isolated way. It is exactly this
interactive mechanism that makes dialogue so easy (Garrod and Pickering 2004).
Consequently, one of the main aims of research on dialogue should be to explain
how thetwointerlocutors cometo co-ordinatetheir linguistic behaviour. However,
LanguageandDialogue5:2(2015),312–333.doi 10.1075/ld.5.2.07pur
issn 2210–4119/e-issn2210–4127©JohnBenjaminsPublishingCompany

313
this problem is hardto tacklebecausedialogue is highly complex and different lin-
guistic and non-linguistic factors affecting it should be taken into account simulta-
neously. Dialogue includes avast rangeof interfaceproblems. Speakers need tode-
cide when it is socially appropriateto speakand they also need to plan what to say
while listening to their partner (Garrod and Pickering 2004). As the researchers
put it, these multi-tasking and task switching procedures are indeed challenging.
Not surprisingly then, an L1–L2 dialogue, in which an L2 speaker communicates
with an L1 speaker, will involve even more interfaceproblems.
When two interlocutors converse, they coordinate their turn-taking (Sacks
et al. 1974). In conversation, when they refer to a new object, they might create
a temporary agreement upon how to name it. The addressee conceptualizes that
new object (Brennan and Clark 1996). For example, a particular “bottle opener”
may beconceptualized as a“corkscrew”or as a“bottleopener” or as “that new bot-
tle opener” (Brennan and Clark 1996). Brennan and Clark (1996, 1484) call this
process “establishing a conceptual pact”. The conversation partners’ co-ordination
is not limited to speech, though. For example, the interlocutors might need to
match their own postures (Condon and Ogston 1966). The aim of such coordina-
tion is to make sure that their conversation flows easily and that interlocutors are
favourably disposed towards each other (Richardson et al., 2007). Co-ordination
occurs both within and between language dialogues.
Common ground, as part of the background knowledge that is shared be-
tweenthe interlocutors (Clark and Marshall 1981), has been the major direction of
research in dialogue studies for manyyears. Commonground deals with“what my
interlocutors and I have in common when it comes to our cultural, linguistic, and
other backgrounds” (Mey 2008, 256). Most researchers (e.g. Clark and Wilkes-
Gibbs 1986) have proposed that if the interlocutors share a common ground, they
will experience successful communication. For example, if I want to ensure that
you understand what I mean by “Takht-eJamshid”,I shouldknow what you know,
and what you know that I know. If such coordination occurs, we may expect to
have a successful dialogue. However, this view, which echoes therelevance-theo-
ry approach, has come under fire in recent years (Gil 2011). Recent studies (e.g.,
Garrod and Pickering 2007a, 2009) have shown that “parallel lower level interac-
tion mechanisms” (Colston 2008, 151) such as the priming effect are enormously
influential in constructing a dialogue. Current psycholinguistic studies on dia-
logue (Garrod and Pickering 2007a, 2007b, 2009) have shown, for instance, that
the mechanism underlying dialogue largely depends on joint action. That is, inter-
locutors “worktogether to establish a joint understanding of what they are talking
about” (Garrod and Pickering 2004, 8). Thus, despite traditional psycholinguistic
studies (e.g., Levelt 1989) that “focus on individual acts of production or com-
prehension” (Garrod and Pickering 2009, 292), recent psycholinguistic studies on

dialogue have indicated that a deep understanding of the speakers’ linguistic be-
haviour demands “the study of interacting individuals” (292) and the process of a
joint action.
In a conversation, speakers readjust their linguistic behaviour depending on
the need of each other. These linguistic co-ordinations occur during the course of
dialogue when communicators tend to align with each other. Alignment is con-
sidered a convergent linguistic behaviour (Branigan et al. 2010) that occurs in
dialogue. Speakers may align on their choice of syntactic structures (Gries 2005;
Branigan, et al. 2000; Levelt and Kelter 1982). For example, studies have shown
that speakers tend to use passive sentences if the conversational partner has al-
ready used it (see Hartsuiker et al. 2004). In the same vein, if a speaker uses a
prepositional object structuresuchas “Thepirate handing the caketo the sailor”, it
is more likely that his or her addressee will respond using the samestructure(e.g.,
the girl giving the book to the teacher) rather than using a double object structure
such as “the girl giving the teacher the book” (Branigan et al., 2007). This phe-
nomenon is called syntactic (structural) priming (for an overview, see Branigan
2007). As Mehler et al. (2011) put it, priming is the central mechanism that was
acknowledged within theInteractive Alignment Model proposed by Pickering and
Garrod (2004) (see Schiller and de Ruiter 2004 and Krauss and Pardo 2004 for a
contrasting interpretation). Interlocutors may use the same referring expressions
in order to refer to particular objects (Pickering and Garrod 2004). This sort of
alignment is called lexical alignment (lexical entrainment). Lexical alignment is
the most salient form of linguistic alignment that occurs in dialogue (see Garrod
and Anderson 1987; Brennan and Clark 1996; Schober and Clark 1989). During
dialogue, when speakers “refer repeatedly to the same object, they tend to reuse
the same terms as they coordinate their perspectives” (Brannan and Clark 1996,
1482). An example of linguistic alignment is found in the walkie-talkie conversa-
tion between two bank robbers, one in the bank vault and another robber who
acted as a lookout on a rooftop, in Schenkein (1980). The bank robbers re-used
each other’s words,as well as the grammatical structures.
(1) “A: …Cor, the noise downstairs,you’ve got to hear it and witness it to realize
how bad it is.
B: You have got to experience exactly the same position as me, mate, to
understand howI feel”. (Schenkein 1980, 23)
Garrod and Anderson (1987) examined how speakers establish a commonseman-
tic and conceptual system during dialogue. To that aim, the researchers used a
computer game that consisted of a maze. The maze included box-like frames con-
nectedby tiny paths along whichparticipants (acting as players) would move their
position markers. Two participants collaborated in order to solve a joint maze.

315
The players moved their position markers until they had both “reached their re-
spective goal positions” (185). Each player had only partial information about his
partner’s location so that verbal exchanges about their positions became neces-
sary. Garrod and Anderson found that when an interlocutor used, for instance,
the words “column” and “row”,it would be more likely that his or her interlocu-
tor would respond with “column” and “row”.Similarly, when he or she used nu-
merals and/or letters, his or her interlocutor would likely respond with numerals
and/or letters. In other words, when a speaker used a word, the same word was
highly activated in the mental lexicon of the listener. This process prompted the
listener to use the same words. Thus, interlocutors “align on a characterization
of the representational domain, for instance using coordinate systems (e.g., A4,
D3) or figural descriptions (e.g., T-shape, right indicator) to refer to positions in
a maze” (Menenti et al. 2012, 1). The linguistic behaviour of the participants also
demonstrates that informativeness is one of the factors that might affect speak-
ers’ conceptualizations. Accordingly, speakers’ referring expressions should pro-
vide enough information so that it would enable the interlocutor “to pick out the
referent uniquely” (Brannan and Clark 1996, 1482). Such co-ordination results in
linguistic alignment during conversation.
As stated above, although dialogue is indeed “the most natural form of using
language” (Ni Eochaidh 2010, 1), it appears to be the most complex form, because
interlocutors’ activities need to be coordinated with precise timing (Menenti et al.
2012). L1–L2 dialogue is a form of verbal exchangein whichoneinterlocutor uses
his or her L1 while the other uses his or her L2. In L1–L2 dialogue, an L1 speaker
typically cannot speak the first language of the L2 interlocutor. L1–L2 dialogue
appears to be more complex than L1–L1dialogue, given the fact that there are
some variables (e.g., the language proficiency of L2 speakers) in L1–L2 dialogue
that are absent in dialogues between two native speakers. For example, in order to
have successfuljoint action, an L1 speaker may use a simplified form of language
or foreigner talk (Costa et al. 2008). In foreigner talk, an L1 speaker avoids using
less frequent words, slangandidioms (seeEllis, 1997for the characteristics of for-
eigner talk). For example, L1 speakers mayrepeat their L2 interlocutors’ language
production more oftenthan whenthey are speaking with an L1 speaker. Moreover,
in a dialogue condition in which one of the interlocutors is not a native speaker of
the conversation language, we might expect that he or she is not fully competent
(Costaet al. 2008). As argued above, L1–L1 dialogue is morelikely to differ from
L2–L1 dialogue, because an L1 speaker might use some strategies that he or she
does not normally use when speaking to an L1 partner, including the use of more
frequent words, avoiding rare expressions, and recasting the interlocutor’s sen-
tence (see Gass 2003). Moreover, one of the variables that affects L1–L2 dialogue
is the non-native speakers’ language proficiency. Anon-native speaker depends

more on “non-automatic routes to alignment” (Costa et al. 2008, 537) than a na-
tive speaker. Therefore, as the researchers state, L1–L2 dialogue might not be as
fluent as the dialogues between two native speakers. Wewill discuss thesepoints
in greater detail below.
In this paper, I investigate some aspects of the dialogue between an L1 speak-
er and an L2 speaker. More specifically, the differences between L1–L1 dialogue
and L1–L2 dialogue are highlighted, focusing on linguistic alignment in L1–L2
dialogue. I investigate the extent to which alignment in L1–L2 is different from
alignment in L1–L1 dialogue. I also examine the factors that cause L1 speakers to
interact differently with L2 speakers than with native speakers during dialogue.
Non-linguistic variables that may affect the degree of alignment have rarely been
discussed in the literature. These non-linguistic factors will also be discussed. In
order to investigate the mechanism that permits lexical alignment, I interpret the
processing of lexical alignment in terms of the “activation threshold hypothesis”
proposedin Paradis (1993, 1997, 2004).
2. Linguistic alignment in L1–L2dialogue
Garrod and Pickering (2006) propose that alignment is an automatic process and
linguistic alignment is primarily unconscious. Speakers “intend to produce and
interpret expressions in the same way as their partners have just done” (204).
Although lexical alignment is considered as the most obvious type of alignment
(Ni Eochaidh 2010), alignment occurs at different linguistic levels (e.g., phono-
logical and syntactic levels) (Branigan et al. 2011). Based on the interactive align-
ment model proposed by Pickering and Garrod (2004), if “communicators come
to understand relevant aspects of the world in the same way as each other” (294),
their communication will be successful. In the interactive alignment account, it
is assumed that successful dialogue requires that the two interlocutors develop
alignment at different linguistic levels (e.g., syntax, lexicon). Sometimes duringdi-
alogue, an interlocutor uses the words or the structure that his or her interlocutor
has already used. Linguistic alignment may occur as afunctionof priming mecha-
nisms at different linguistic representations. Due to the priming mechanism, some
degree of automatic alignment is predicted in any dialogue (Costa et al. 2008). In
this account, lexical priming leads toalignment at the lexical level (lexical entrain-
ment), and syntactic (structural) priming leads to alignment at the syntactic level
(Pickering and Garrod 2004). Moreover, priming at one level enhances alignment
at other levels (Pickering and Garrod 2006). For example, Branigan, Pickering,
and Cleland (2000) found more syntactic priming effect in the same-verb pairing
condition in which the prime and the target cards involved the same verb (e.g.,

317
loan-loan) than in the different-verb condition in which the prime and the target
cards involved different verbs (e.g., loan-lend).
L2 speakers may align with L1 speakers for the purpose of improving their
language proficiency. As there are several reasons to learn a language, L2 users
may have multiple communicative goals (Bongaerts 1999; Butler and Hakuta
2004; Cook 2002). One of the main goals of non-native speakers, especially those
witha lower level of proficiency, inaconversation withanL1 speaker is to improve
their language proficiency (Costaet al. 2008). For L2speakers, nativeinterlocutors
areregarded as reliable sources tolearnlanguage. Inorder tospeakthe language as
native-like as possible, they generally consider conversation with nativespeakers
as oneof themost efficient ways toimprovetheir speakingabilities (Gass 2003). L2
speakers might deliberately align with an L1 speaker at different linguistic levels
(e.g., lexicon, phonology-accent) in order to improve their language proficiency.
When they hear new expressions that native speakers used during a conversation,
they may repeat them. The repetition of new words or expressions both facilitates
language learning and alignment. By repeating new lexical items, a non-native
speaker makes them moreavailable in his or her mental lexicon, because“this pro-
cess leads tothe development of a lexicon of expressions relevant to the dialogue”
(Pickering and Garrod 2004, 175). Wemay assume that these sorts of alignment
increase the frequency of lexical items (see Section 4 for more details). For exam-
ple, in dialogue with L1 speakers or highly proficient L2 speakers, an L2 speaker
may notice that they use some adverbs such as “actually” or some adjectives such
as “salient, nasty, weird, and wee” more frequently. An L2 speaker may decide to
use them in his or her speechas well whenever possible.
An L1 speaker’s linguistic behaviour affects the degree of alignment during
dialogue. An L1 speaker might correct or re-structureL2speakers’ ungrammatical
sentences during a conversation. Gass (1977, 2003) labels this as “recasting”,and
is defined as “instances in which an interlocutor rephrases an incorrect utterance
with a corrected version, while maintaining the integrity of the original meaning”
(Gass 2003, 239). Recasting is likely especially when the interlocutor can guess
what meaning a speaker intended to convey (see example 2).
(2) A: I eated the sandwichquickly.
B: You ate the sandwich quickly.
According to Cabaroglu, Basaran and Roberts (2010), some researchers (e.g.
Lyster andRanta 1997) consider recasting as “animplicit negative feedback” (15);
however, one of the main purposes of recasting is to promote the alignment of
situation models (Costa et al. 2008). However, in such cases, native speakers use
non-automatic linguistic alignment. As example (2) shows, re-structuring does
not carry any new message. In other words, the aim of recasting is, therefore, not

always to provide correct information that an interlocutor wants during dialogue,
but communicators use “recasting” as they believe that it may enhance alignment
and communication quality. Thus, recasting should not be considered as only an
indicator of interlocutors’ preferences for well-formed rather than ungrammati-
cal sentences (seeLee et al. 2009). Clearly, an L1 speaker mayfail to align with an
L2 interlocutor on some occasions. If the language production of the non-native
speaker is not grammatically correct, and the native speaker is not able to guess
what he or she means from the context (Costa et al. 2008), there is the possibility
that an L1 speaker will be unable to align with the L2 speaker. For example, if an
L2 speaker of Persian uses a wrong collocation (e.g., pas dādan, meaning to give
something back vs. pas gereftan, meaning to take or recall something), a Persian
L1 speaker may fail to align.
Speakers’ beliefs about each other, for instance, their belief about the inter-
locutors’ cultural communities (Clark 1996), and their beliefs about the interlocu-
tors’ language proficiency (Bortfeld and Brennan 1997) affect the extent to which
they align. Speakers’ prior beliefs about their interlocutors influence “theform and
communicativeness of their messages”(Fussell and Krauss 1992, 379). Branigan,
Pickering, Pearson, McLeanand Brown(2011) examined whether speakers’ align-
ment in dialogue is mediated by their beliefs about their interlocutors. They used
an interactive labeling task. The subjects were told that they were communicating
with either a human partner or a computer. Participants were then asked to select
pictures according to their partner’s descriptions and name pictures so that their
partners could select the pictures. Participants revealed a very “strong tendency to
use their partner’s choice of object name” (47). That is, when their partners used
the favoured name, they almost always used the same favoured name. More inter-
estingly, when their partner used a disfavoured name (e.g., using settee to name
a sofa), participants often used the same disfavoured name as well. The degree of
alignment varied depending on the participants’ beliefs about the capability of
their partners. Branigan and colleagues (2011) report that alignment was more
pronounced with computer partners than with human partners. More strikingly,
they found that participants’ tendency to align with the less capable computers
was stronger than with more capable computers. The researchers concluded that
the tendency to align seems to be mediated by the speakers’ beliefs about their
interlocutors.
As stated above, speakers’ beliefs about their interlocutors’ language profi-
ciency affect the extent to which they align with L2 speakers. In experiment 3 in
Bortfeld and Brennan’s (1997) study, the researchers examined to what extent na-
tive speakers adjust their speech to non-native speakers, or as the researchers put
it, they examined the extent to which the native speakers “sacrificed idiomaticity
in adjusting to non-native speakers” (134). The researchers reported that native

319
speakers significantly aligned their referring expressions to their non-native inter-
locutors. For example, in one interesting case, a native speaker used “the one with
five little tires” to describe an “office chair” to a non-native speaker, while native
speakers usually prefer to call it “the office chair with wheels” (133). Even though
the native speaker is aware of the everyday term used to describe the object, he
or she may instead choose to rely on words that are perceived to be more acces-
sible tonon-native speakers. Thus, thefindings in Bortfeld and Brennan (1997) re-
vealed that sometimes native speakers usehighly inappropriateor unfitting names
in order to achieve lexical alignment (e.g., the use of “tires” instead of “wheels”).
Although such conscious decisions to align are cognitively costly (i.e., to deac-
tivate the appropriate lexical item and simultaneously activate an inappropriate
lexical item or a string of lexical items), it is not uncommon in L1–L2 dialogue.
These results suggest that referring to objects in L1–L2 dialogue demands a more
collaborative process.
The tendency to show “affiliation with (or disaffiliation from) the cultural
communities to which interlocutors perceive each other to belong” (Branigan et
al. 2011, 55) also affects interlocutors’ alignment. For example, a native English
speaker who knows about Muslim communities and the way they call each oth-
er within such communities may use the term “brother” when addressing an L2
Muslim speaker. L1speakers’ useof terms suchas “brother”that L2 speakers nor-
mally use when they speak in their communities promotes alignment. However,
in an interactive context, “feedback from the ongoing conversation” (Fussell and
Krauss 1992, 390) may determine the degree of speakers’ reliance on their prior
beliefs. As Fussell and Krauss (1992, 390) put it, “feedback can lead to modifica-
tions of one’s prior beliefs about the addressee’s category memberships and about
the kinds of knowledge that are characteristic of a given social category”. Garrod
and Pickering (2007b) also consider feedback as one of the main sources of align-
ment. Bevales, Coates and Johnson (2000) report that a speaker sometimes al-
ters his contribution after the addressee’s feedback. Thus, it is possible that the
utterances are reformulated with respect to the communicator’s feedback. The
quality of feedbacksignificantly affects theinterlocutors’ contribution during dia-
logue (Schober and Clark 1989). This idea is, therefore, close to what Garrod and
Pickering (2007a) label “strategic alignment”. Accordingly, in order for a speaker
to establish a successful alignment especially in an L1–L2 dialogue, the amount
of feedbacka speaker receives from his interlocutor must be sufficient. Moreover,
the implication of such behaviour during dialogue is that the reliance on com-
mon ground might change as the conversation continues. For example, during a
conversation between a Muslim L2 speaker and a non-Muslim L1 speaker, one
of the interlocutors might find that his or her belief about certain aspects of the
interlocutor’s community, such as his or her religion, was not valid. This may lead

Linguistic alignment in L1–L2 dialogue 320
to a shift in common ground during conversation. Thus, common ground can be
modified. Its utilization may be increased or decreased during the course of dia-
logue (Colston 2008), especially during an L2–L1 dialogue.
In an L1–L2 dialogue, speakers may employ two opposing strategies in order
to achieve an interactive alignment. On the one hand, L1 speakers may avoid us-
ing less frequent words or more frequent synonyms. They may also describe an
object or, as Bortfeld and Brennan put it, use “descriptive referring expressions”
(128) instead of naming it. They may also name an object followed by a descrip-
tion of that object. All these strategies help non-native speakers understand L1
speakers’ intended meaning. On the other hand, in order to avoid being consid-
ered as non-proficient-non-native speakers, proficient L2 speakers may occasion-
ally use some rare words rather than using more frequent words. Such a strategy
might also affect their topic selection. An L2 speaker may also decide to mention
cultural features specific to the L1 speakers’ culture. For example, an L2 speaker
of Persian might start speaking about Iranian ceremonies of the New Year (e.g.,
chāhār šanbehsuri, the event held in the last Wednesday of the year). Such oppos-
ing strategies adopted by L1 and L2 speakers improvealignment.
Regarding the alignment of pragmatics during dialogue, this topic has hardly
been addressed inthe literature, especially in L1–L2 dialogue. As “wedon’t always
saywhat we mean, and often don’t meanwhat wesay” (Bavelas et al.1986, 6), it is
more likely that speakers reuse indirect language and align pragmatically during
conversation. Roche, Dale and Caucci(2012) investigated whether speakers come
to reuse indirect language (IL) and align pragmatically when they hear IL from a
partner. The researchers used the term IL (indirect language) to refer to any prag-
matic discourse form that encompasses broadly any statement that has both literal
and nonliteral meanings. Saying “what a quiet day” on a busy day, is an example
of IL. They proposethat pragmatics also undergoes alignment. IL maycausesome
problems for alignment in bothnative and non-native speakers’ dialogue, because
in these cases, the interlocutors might first interpret the IL literally. According
to Roche, Dale and Caucci (2012) when a speaker finds that the literal meaning
does not match the immediate context, he/she wonders whether the partner re-
ally believes that it is a quiet day or he intends to “be playfully ironic”. We may
expect that L2 speakers havemoreproblems with ILthan L1 speakers especially in
complex cases of IL, becauseanL2speaker maynot be familiar withthosetypes of
expressions (e.g. sarcasm or ironyas types of IL), whichare parts of the pragmatic
knowledge. However, in order to avoid failing to align, the addresser may pro-
vide some pragmatic cues. For example, the speaker may use “joyful intonation”
or adopt a particular facial expression during IL. However, when alignment fails
due to the addresser’s use of IL, two outcomes may occur. The speech act may be
neglected by the interlocutor “because the pragmatic goal was never realised” or,

alternatively, morecues related to thepragmatic intent may be provided by thead-
dresser that attempt to further prevent the pragmatic goal from failing during or
after IL (Roche et al., 2012). For example, the speaker may say “oh, I was joking”.
In doing so, the two interlocutors return to what I would call their “established
literal alignment”.
3. Obstacles to the alignmentprocess inL1–L2 dialogue
Whereas some strategies facilitate linguistic alignment during L1–L2 dialogue,
some linguistic behaviour may hinder it. For example, naming an object or an
event differently might be problematic for lexical alignment during the course of
dialogue, especially in a dialogue between an L1 speaker and an L2 speaker. If the
twointerlocutors in L1–L2 dialogue usedifferent terms to refer to the sameobject
or to address the same event, this situation might not enable strong alignment.
While such a situation might also arise in a dialogue between two L1 speakers,
using different referring terms occurs more frequently in an L1–L2 conversation
than in an L1–L1 conversation (Costa et al. 2008). An L1 or L2 speaker might
prefer to use a specific name for an object as the speaker may believe that the use
of this particular term increases communicative success (Branigan et al. 2011).
Moreover, a speaker might believe that using a particular name is more appropri-
ate when speaking to a particular interlocutor. Using a specific label also implies
an underlying conceptualization of the event or object (Brannan and Clark 1996).
Theinterlocutor, however, mayor maynot agree with the sameconceptualization.
Thespeaker’s referring expression is, thus, provisional until it is accepted or modi-
fied by his or her partner (Brennan, 1996). Since past histories may underlie using
some words, naming things or events sometimes has serious social and/or politi-
cal consequences (Jackson2005).This problem mayexist in dialogue betweentwo
speakers whoprefer to usedifferent terms for the same thing or event. As Jackson
(2005) states, to “call an act of political violence terrorist is not merely todescribe
it but to judge it. …. Two names for the same act have very contrasting meanings
and would likely elicit very different responses” (22) from the two interlocutors.
For example, calling international forces in Iraq “peacekeepers” or calling them
“occupying forces”might trigger vastly different reactions from an Iraqi interlocu-
tor or even a Middle Eastern speaker. Another example is calling “Persian Gulf”
“Gulf ”,whichmay yield different reactions from an Iranian interlocutor.
Thereare, however, somecaseswhereaspeaker might chooseadivergent label
or name for an object or event, even though he is awarethat that particular label or
name is not appropriate from the perspective of the interlocutor. In other words,
the interlocutors do not express any intent to align on the same labels or names. A

MehdiPurmohammad
speaker might also be aware that choosing a particular term dramatically decreas-
es communicative success.Toput it differently, aspeaker might use a disfavoured
name for an object or an event regardless of whether it might disrupt alignment.
It is reasonable to assume that this would occur more frequently when a speaker
talks about asensitive topic suchas religion, politics, war, gender andsoon. Under
the circumstances, conveying a message might be much more important for some
speakers than establishing linguistic alignment. Difficulties arise as soon as dia-
logue is viewed as pure exchange of information (Stalnaker 1974). “The difficul-
ties come with contexts in which other interests besides communication are being
served bythe conversation”(201). Inthesesituations, it appears that the automatic
linguistic alignment seems to vary in different dialogues (Costa et al. 2008).
One of the more interesting cases in conversations between a non-native
speaker and a native interlocutor, which has not been specifically addressed in
literature, is when a native speaker chooses not to promote alignment (e.g., word
choice, phonology) with his or her non-native interlocutor. One of the reasons
for this might be that native speakers believe that they should not be concerned
about whether their non-native interlocutor understands what they are saying. A
speaker might assume that it is the interlocutor’s responsibility to improve his or
her language proficiency. Thus,therearesomesituations in whichspeakers do not
appear to be concerned with the interlocutors’ level of language proficiency.This
might be rooted in the economic status of L1 speakers’ countries. Interlocutors’
political views or general economic status mayplay pivotal roles in adopting such
non-aligned linguistic strategies. For example, when the rate of unemployment
increases in a society, nativespeakers of the country’slanguagemight assumethat
one of the reasons for unemployment is that foreigners (for our purposes here,
nonnative speakers) who are residing in their country have robbed them of their
job opportunities. This might causenative speakers tonot express anytendency to
align withnon-native speakers, which, inturn, causes them to continue using their
own non-simplified/non-aligned language. Some degree of deviance from com-
mon groundis, therefore, expectedfor various reasons. It is obvious that too much
deviance from the common ground can stifle talk (Colston 2008). An imbalance
caused by the degree of language proficiency, and the lack of tendency to align
with the interlocutor can harm interaction. The interlocutor in such situations
may feel lost or even insulted. For example, a foreign student at the University of
Edinburgh or Glasgow who is not familiar with the Scottish accent may feel in-
sulted or at least uncomfortable if during a meeting a professor continues using a
strong Scottish accent, ignoring the student’s difficulty with it. Using non-aligned
language in L1–L2 dialogue might be prevalent in some societies. Thus, it is sug-
gested that questions related to theways in whichsociocultural and economic fac-

tors affect linguistic alignment in L1–L2 dialogue remain largely unansweredand,
consequently, require further study.
Costa, Pickering and Sorace (2008) propose that there is a correlation be-
tween the degree of an L2 speaker’s language proficiency and alignment with an
L1 speaker. When the speaker is highly proficient, automatic alignment occurs
normally between L1 and L2 speakers as would be the case with L1–L1 dialogue.
If, however, a non-native speaker is not fluent enough to provide the automaticity
needed for successful alignment, this “may disrupt the timing of the L1 speak-
er’s responses” (Costa et al. 2008, 549). Moreover, language proficiency of an L2
speaker is a crucial factor that determines to what extent he or she is engaged
in conversation with an L1 speaker. For example, L2 speakers might feel intimi-
dated if on someoccasions their native speaker classmates askthem to repeat what
they have already said. This might “preclude them from generating productive
discussions”(Kawanani and Kawanani 2009, 2), meaning that lack of proficiency
decreases interaction during dialogue, which, in turn, further decreases the pos-
sibility of alignment.
The degree of similarity or difference betweenthe L1 and the L2 has a crucial
role in alignment. When the representations in the twolanguages are similar, they
may be transferred from language A to B or vice versa. Thus, similarity across the
two languages may increase priming (Costa, et al. 2008). For example, if a bilin-
gual speaker hears a construction such as a passive structure that has the same
construction in the other language, wemay expect cross-linguistic priming at syn-
tactic level to occur (see Hartsuiker et al. 2004 for cross-linguistic syntactic prim-
ing). As many studies show, cross-linguistic syntactic priming occurs only when
the twolanguages of bilinguals havethe same structures (seeBernolet et al. 2007).
Thus, wemaypredict that in general, a French-English speaker has a better chance
of creating morealignments witha native English speaker thana Japanese-English
speaker, since there are simply many more cognate words in French and English
than in Japanese and English. This means that “an L2 speaker’s profile is strongly
influenced by his L1 knowledge” (Costaet al. 2008, 541).
Native speakers of a language typically share morecommonknowledge about
the cultural aspects of their communities compared to interlocutors in L1–L2 dia-
logue. Torephrasethis, L2 speakers maysharemuchless information that is part of
theL1 speakers’ socio-culturalbackground. This typeof knowledgeis called“com-
munity membership” by Clark and Marshall (1981). In circumstances in which
L1 speakers find that their addressees do not know much about the socio-cultural
aspects of their lives, they may be required to frequently establish meanings for
such unknown facts during dialogue. Once the information has been grounded, it
would be possible for interlocutors to “make use of these shared perspectives on
subsequent trials of task”(Horton 2008, 205). Thus, somebackground knowledge,

MehdiPurmohammad
that is shared among L1 speakers, only becomes accessible to an L2 speaker dur-
ing the course of the dialogue. These occurrences make the linguistic alignment
process challenging for both parties. Under the circumstances, the L1 speaker
may need to spend considerable time explaining such things as “what it is”, “what
people aredoing”, and“how things are”totheL2speaker. Theaim of suchexplana-
tions is to establish more common ground between the two interlocutors. I would
call this a “backward process tore-establish commonground”.AnL1 speaker may
need to producelonger descriptions, thus more hedges areexpected during L1–L2
dialogue thanL1–L1 conversation.
The lack of sociocultural information may make an L2 speaker appear as out
of touchwiththe reality of the L1 speaker during the courseof dialogue. L2 speak-
ers may not be completely engaged in the various aspects of the culture of those
whospeakthat language natively. (seeCouncil of Europe 2001, for moredetails on
thesociocultural knowledge). Theseareas might include aninadequate knowledge
of traditional foods, social events, famous musicians and music works, actors and
actresses, history, popular TV shows, and so on. In such cases, interaction and
alignment would be suboptimal, because alignment typically occurs whenspeak-
ers believe that they understand each other. In other words, alignment is achieved
to the extent that speakers “share much relevant background knowledge” (Garrod
and Pickering 2009, 294).
4. Mechanisms involved in linguisticalignment
As discussed above, linguistic alignment occurs in dialogue. The question that
arises here is what mechanisms permit lexical alignment? As mentioned above, the
priming mechanism leads to alignment. As a function of the priming mechanism,
some degree of automatic linguistic alignment is, thus, predicted in any dialogue
(Pickering and Garrod 2004). Lexical alignment may also be interpreted in terms
of the “activation threshold hypothesis” proposed in Paradis (1993, 1997, 2004).
According to this hypothesis, when there are two or more competitors for selec-
tion, thelexical item with thelower activation threshold level is selected. Inabilin-
gual setting, there are at least two competitors for any word selection (Kroll et al.
2008), one competitor is from the base language and the other is its equivalent in
the other language. The hypothesis asserts that the item that has the lower activa-
tion threshold level is selectedregardless of whichlanguageit belongs to. However,
the base language continues to govern the exchange and to define communica-
tion patterns. According to the hypothesis, there are two main factors that lower
the activation threshold level of a lexical item: the word’s frequency of use, and
its recency of use (Paradis 2004). This idea, therefore, shares important features

with the very well-established effects (primacy and recency) in human memory.
According totherecency effect,“recallwouldlikely be greatest, all else held equal,
for items most recently grounded in conversation” (Colston 2008, 154). Thus, it is
possible that recency influences the overall structureof dialogue (seeBrennan and
Clark 1996 for recency effect). If a word has already been used, or if it has been
used more frequently, it is more likely to be selected instead of its equivalent. This
is consistent with the account of lexical alignment that each term selection is af-
fectedby the availability of the terms. Under this interpretation, therecencyof past
references affects the speakers’ choice of words (Brannan and Clark 1996). Thus,
when a speaker uses a word during the course of dialogue, its activation threshold
level is lowered both in the memory of the speaker and the interlocutor. In other
words, aspeaker’s production of awordenhances theavailability of that item in the
memory of the interlocutor as well (see Jucks et al. 2008 for extensive discussion).
Thus, abridge is made betweenthe mental lexicon of thespeaker andhis addressee
during dialogue. Theactivation of lexical items does not always occur as afunction
of “autonomous processes within the interlocutor, but can just as well be due to
prior contributions by her partner” (Pickering and Garrod 2006, 201).
The account that a speaker’s production of a word lowers the activation
threshold level of that word in the mental lexicon of his or her comprehender
is in line with Pickering and Garrods’ (2013) proposal that language produc-
tion and language comprehension are interwoven. The researchers postulate that
language production and comprehension are not quite distinct from each other.
Accordingly, interlocutors’ tendency tobecomealigned with each other during di-
alogue indicates the close link between language production and comprehension.
Pickering and Garrod (2013) postulate that “the tight coupling between interlocu-
tors in dialogue is the evidence for effects of comprehension processes on acts of
production and vice versa”(9). I submit that the activation threshold level account
could well be integrated with the alignment account in dialogue. Together, they
might ultimately be captured in one single integrated account that includes both
production and comprehension processes.
As argued above, the two interlocutors co-ordinate their linguistic behaviour
in different ways. Co-ordination during dialogue is made possible through low
level mechanisms of linguistic alignment that function to increase the similarity
of mental representations across interlocutors (Horton 2008). According to the
interactive alignment account (Pickering and Garrod 2004) the automatic prim-
ing mechanism enhances the likelihood that a particular linguistic behaviour pro-
duced by a speaker during dialogue will be produced more frequently by both the
samespeaker again andtheinterlocutor as well. Thus, themental representations of
interlocutors are aligned through the automatic priming processes (Horton 2008).
The aim of this coordination is to make sure that their conversation flows easily

MehdiPurmohammad
and intelligibly and that interlocutors are favorably disposed toward each other
(Richardson et al. 2007). The speakers’ formulation of utterances in dialogue has
at least twomain components:first, thespeakers’ prior beliefs about the interlocu-
tor, his or her culture, norms, and background; second, the immediate feedback
that is received from the interlocutor. Feedback from the interlocutor might affect
a speaker’s word and topic selection, and so on. Thus, communication is achieved
jointly through the “alignment of information states” (Pickering and Garrod 2006,
209) andthe dynamic emergence of feedback during the courseof dialogue.
5. Language selectionandthe linguisticalignmentinmultilingual
communities
I consider language selection to be the “macro-linguistic alignment” process in
multilingual communities. Language choice alignment is a necessary factor that
makes successful communication possible in multilingual communities such as
Switzerland in which more than two languages are spoken (four national lan-
guages — German, French, Italian, and Romansh). (see Werlen 2007, for Swiss
multilingualism). Moreover, as English is taught as a foreign language at school,
many people have some degree of proficiency. As is the case in Switzerland, it is
not unusual tofind people whospeakdifferent languages or different dialects (e.g.,
Bernese dialect in Switzerland) in multilingual communities. Naturally, the first
step to establish linguistic alignment is likely to be the selection of an appropri-
ate language or even an appropriate dialect. In fact, in a multilingual community,
where people have the option to select one of the languages or dialects for com-
munication, language or dialect choice alignment is the most obvious example of
linguistic alignment. Selecting an appropriate language in multilingual communi-
ties enhances alignments at other levels of language production (e.g., wordchoice,
accent). Extending the notion of linguistic alignment to “language choice align-
ment” might first be interpreted as a deviation from the main accounts of linguis-
tic alignment. However, considering multilingualism and the various ways people
with different languages communicate with each other in Switzerland may point
to the fact that in some multilingual communities “language choice alignment” is
considered to be both a sort of linguistic alignment and the most important step
towardestablishing different types of linguistic alignments.
As stated above, it is common to find people who can speak more than one
language in Switzerland. Themain question, therefore, is what mechanism governs
the choiceof language to usefor communication. Werlen (2007) first reviewed the
two models proposed by Kolde (1981) and then introduced two additional mod-
els. According to Kolde (1981, cited in Werlen 2007), in the “Swiss model”, the

secondspeaker answers in his or her ownfirst language. It means that in the Swiss
model everyone speaks their own language. But if a speaker follows the second
model (Biel/Bienne model), known also as the monolingual model, the second
speaker will answer in the same language as the first speaker (the interlocutor).
Thus, the main difference between the “Swiss model” and the Biel/Bienne model
is that, while in the second model the addressee’s language choice is accepted, in
the first model the addressee’s language is not accepted (Werlen 2007). In a third
model (default model) proposed by Werlen (2007), everybody chooses “the de-
fault language of theterritory”(142). However, applyingalingua francamodelwill
lead speakers to use a language that is “neither participant’s language”(142). Lüdi
(2007) argues that further communication models may exist. For example, the
monolingual model might be integrated with the lingua franca model. As Swiss
multilingualism shows, speakers’ language choice can have a great effect on lin-
guistic alignment during dialogue. Each of the language choice models permits
different levels of linguistic alignment during dialogue. For example, while vari-
ous types of linguistic alignment might occur inthe Biel/Bienne and lingua franca
models, no or not much alignment is expected in the Swiss model. For example,
in the Swiss model, lexical alignment may only be possible through using mixed
speech(e.g., codeswitching).
6. Conclusion
Dialogue is themost natural form of language use(Ni Eochaidh 2010). However, it
is also the most complex form of language use, becauseprecise timing is required
to properly coordinate interlocutors’ activities (Menenti et al. 2012). Dialogue is
also easy (Garrod and Pickering 2004). As Garrod and Pickering put forward,
part of the explanation why dialogue is easy is that dialogue is a joint activity and
that interlocutors work jointly to construct a mutual understanding of what they
are speaking about. Both the interlocutors and the dialogue are highly dynamic.
Interlocutors perform a joint action during dialogue (Garrod and Pickering 2009).
Thus, dialogue is not merely a transferred messagefrom one speaker to another in
a neutral space, but speakers rather negotiate meaning by“interweaving their con-
tributions” (Pickering and Garrod 2013,1). This is, however, true for both L1–L1
dialogue andL1–L2 dialogue.
It follows that, as L1–L2dialogue may beaffected byother resources andstates
that are not available or present in dialogue between two native speakers, there
will naturally be some differences between L1–L1 dialogue and L1–L2 dialogue.
Thesedifferences influencelinguistic alignment in L1–L2 dialogue. Thedegrees of
alignment in L1–L2 depend on some other variables that may be absent in L1–L1

MehdiPurmohammad
dialogue. Therefore, we might expect more different patterns of alignment in L1–
L2 dialogue than in dialogue betweentwonative speakers. What makes L1–L2 di-
alogue more complicated than L1–L1 dialogue is that in the former, communica-
tion partners are not only differentiated by linguistic factors. They may have very
different beliefs, identities, cultures and backgrounds. This situation may exist in
L1–L1 dialogue, but as stated above, more differences of thesetypes are expected
in L1–L2 conversation. Thus, oneof the main steps in establishing alignment is to
know about theinterlocutors’ beliefs, identities, cultures, andbackground, because
the extent to which speakers align is affected by their beliefs about the identity of
their interlocutors (Branigan et al. 2011). There is evidence (Giles et al. 1991) that
wheninterlocutors show someaffiliation withthe cultureof the other interlocutor,
they are morequickly engaged to build successfulalignment.
Given that dialogue betweena nativespeaker anda non-nativespeaker is likely
todiffer from L1–L1dialogue, morealignment inL1–L2dialogue needs tobemade
bythe twointerlocutors at different linguistic levels (e.g., syntax,pragmatic andlex-
ical) in order for communicatorstomaintainsuccessfuldialogue. This suggests that
alignment in L1–L1 dialogue might also be different from L1–L2 dialoguequanti-
tatively. Such alignment of differing linguistic representations may occur through
priming mechanisms (Costa et al. 2008). However, a speaker’s affiliation with the
culture of the other interlocutor (Branigan et al. 2011), feedback from the inter-
locutor, theL2speaker’sgoaltolearn thelanguage, recasting, andthedegree of lan-
guageproficiency ofL2speaker willall impact alignment inL1–L2dialogue as well.
Finally, linguistic alignment may also be interpreted in terms of the activation
threshold hypothesis (Paradis 2004). According to this interpretation, language
production of speakers may lower the activation threshold levels of those items or
structures in the memory of their interlocutors. Consequently, a bridge is made
betweenthe mental lexicon of the speaker and theaddresseeduring dialogue. This
account is in line with Pickering and Garrods’ (2013) account that language pro-
duction and language comprehension are interwoven.
References
Bavelas, Janet B., Alex Black, Charles R. Lemery, and Jennifer Mullett. 1986. I Show How You
Feel: MotorMimicry asaCommunicative Act.”JournalofPersonality and SocialPsychology
50 (2): 322–329. DOI: 10.1037/0022-3514.50.2.322
Bavelas, Janet B., LindaCoates, and Trudy Johnson. 2000. “Listeners as co-narrators.”Journalof
Personality and Social Psychology 79: 941–952.
Bernolet, Sarah, Robert J. Hartsuiker, and Martin J. Pickering. 2007. “Shared Syntactic
Representations in Bilinguals: Evidence for theRole of Word-Order Repetition.” Journalof

ExperimentalPsychology:Learning,Memory,andCognition33:931–949.
DOI: 10.1037/0278-7393.33.5.931
Bongaerts, Theo. 1999. “Ultimate Attainment in L2 Pronunciation: TheCaseof Very Advanced
LateL2Learners.”InSecondLanguageAcquisitionandtheCriticalPeriodHypothesis,ed.by
David Birdsong, 133–159. Mahwah, New:LawrenceErlbaum.
Bortfeld, Heather, and Susan E. Brennan. 1997. “Use and Acquisition of Idiomatic Expressions
in Referring by Nativeand Non-nativeSpeakers.”Discourse Processes 23:119–147.
DOI: 10.1080/01638537709544986
Branigan, Holly P.2007. “SyntacticPriming”. Language and Linguistics Compass 1 (1-2):1–16.
DOI: 10.1111/j.1749-818X.2006.00001.x
Branigan, Holly P., Martin J. Pickering, Jamie Pearson, Janet F.McLean, and Ash Brown. 2011.
“The Role of Beliefs in Lexical Alignment: Evidence from Dialogs with Humans and
Computers.”Cognition121:41–57.DOI:10.1016/j.cognition.2011.05.011
Branigan, Holly P., Martin J. Pickering, Jamie Pearson, and Janet F.McLean. 2010. “Linguistic
Alignment between Peopleand Computers.” Journalof Pragmatics 42:2355–2368.
DOI: 10.1016/j.pragma.2009.12.012
Branigan, Holly P., Martin J. Pickering, Janet F. McLean, and Alexandra A. Cleland. 2007.
“Syntactic Alignment and Participant Rolein Dialogue.”Cognition104 (2):163–197.
DOI: 10.1016/j.cognition.2006.05.006
Branigan, Holly P.,Martin J. Pickering, and AlexandraA. Cleland. 2000. “SyntacticCoordination
in Dialogue.” Cognition 75: B13–B25. DOI: 10.1016/S0010-0277(99)00081-5
Brennan, Susan E., and Herbert H. Clark. 1996. “Conceptual Pacts and Lexical Choice in
Conversation.” Journalof ExperimentalPsychology: Learning, Memory, and Cognition22:
1482–1493. DOI: 10.1037/0278-7393.22.6.1482
Butler, Yuko G., and Kenji Hakuta. 2004. “Bilingualism and Second Language Acquisition.”
In The Handbook of Bilingualism, ed. by Tej K. Bhatia and William C. Ritchie, 114–144.
Massachusetts: Blackwell Publishing.
Cook, Vivian. 2002. “Background to theL2 User.”In Portraitsofthe L2 User, ed. by Vivian Cook,
1–28. Clevendon: Multilingual Matters.
Cabaroglu, Nese, Suleyman Basaran, and Jon Roberts. 2010. “A Comparison between the
Occurrence of Pause, Repetitions and Recasts under Conditions of Face-to-face- and
Computermediated Communication: A Preliminary Study.”The Turkish Online Journal of
Educational Technology. April. 9 (2).
Clark, Herbert H. 1996. Using Language. Cambridge, England: Cambridge University Press.
DOI: 10.1017/CBO9780511620539
Clark, Herbert H., and Edward F.Schaefer. 1989. “Contributingto Discourse.”Cognitive Science
13: 259–294. DOI: 10.1207/s15516709cog1302_7
Clark, Herbert H., and Deanna Wilkes-Gibbs. 1986. “Referring as a Collaborative Process.”
Cognition 22: 1–39. DOI: 10.1016/0010-0277(86)90010-7
Clark, Herbert H., and CatherineR. Marshall. 1981. “DefiniteReferenceandMutualKnowledge.”
In Elements of discourse understanding,ed. by Aravind Joshi, BonnieWebber, and Ivan Sag,
10–63. Cambridge, England: Cambridge University Press.
Colston, Herbert L. 2008. “ANew Look at Common Ground: Memory, Egocentrism, and Joint
Action.”In Intention, CommonGround, and the Egocentric Speaker-Hearer, ed. by Istvan
Kecskes and Jacob Mey, 151–188. New York:Mouton deGruyter.

Condon, William S., and William D. Ogston. 1966. “Sound Film Analysis of Normal and
Pathological Behavior Patterns.” Journalof Nervous and MentalDisease 143:338–347.
DOI: 10.1097/00005053-196610000-00005
Costa, Albert., Martin J. Pickering, and AntonellaSorace. 2008. “Alignment in Second Language
Dialogue.”LanguageandCognitiveProcesses23(4):528–556.
DOI: 10.1080/01690960801920545
Council of Europe. (2001). CommonEuropeanFramework ofReference forLanguages: Learning,
teaching,assessment.Cambridge:CambridgeUniversity Press.
Ellis, Rod. 1997.SecondLanguageAcquisition.OxfordUniversity Press.
Fussell, Susan R., and Robert M. Krauss. 1992. “Coordination of Knowledgein Communication:
Effects of Speakers’ Assumptions about What Others Know.”Journal of Personality and
Social Psychology 62: 378–391. DOI: 10.1037/0022-3514.62.3.378
Garrod, Simon C., and Anthony Anderson. 1987. “SayingWhat YouMean in Dialogue:A Study
in Conceptualand Semantic Co-ordination.”Cognition27 (2):181–218.
DOI: 10.1016/0010-0277(87)90018-7
Garrod, Simon, and Martin J. Pickering. 2009. “Joint Action, Interactive Alignment, and
Dialogue.” Topics in Cognitive Science 1:292–304. DOI: 10.1111/j.1756-8765.2009.01020.x
Garrod, Simon, and Martin J. Pickering. 2007a. “Alignment in Dialogue.”In Oxford Handbookof
psycholinguistics,ed.by GarethGaskell,443–451.Oxford:OxfordUniversity Press.
Garrod, Simon, and Martin J. Pickering. 2007b. “Automaticity in Language Production in
Monologue and Dialogue.” In Automaticity and Control in Language Processing, ed. by
AntjeS. Meyer, LindaR. Wheeldon, and AndreaKrott, 1–21. Hove:Psychology Press.
Garrod, Simon, and Martin J. Pickering. 2004. “Why Is Conversation So Easy?” Trends in
Cognitive Sciences 8: 8–11. DOI: 10.1016/j.tics.2003.10.016
Gass, Susan M. 2003. “Inputand Interaction.”In The HandbookofSecond Language Acquisition,
ed. by Catherine J. Doughty and Michael H. Long, 224–255. Massachusetts: Blackwell
Publishers. DOI: 10.1002/9780470756492.ch9
Gass, Susan M. 1997. Input, Interaction, and the Second Language Learner. Mahwah, NJ:
LawrenceErlbaum.
Gil, José Maria. 2011. “Relevance Theory and Unintended Transmission of Information.”
InterculturalPragmatics8(1):1–40.DOI:10.1515/IPRG.2011.001
Giles, Howard, Nikolas Coupland, and Justine Coupland. 1991. “Accommodation Theory:
Communication, Context,and Consequence.”InContextsofAccommodation:Developments
in Applied Sociolinguistics, ed. by Howard Giles, JustineCoupland, and Nikolas Coupland,
1–68.Cambridge: Cambridge University Press.DOI:10.1017/CBO9780511663673.001
Gries,Stefan Th.2005.“SyntacticPriming:aCorpus-basedApproach.”JournalofPsycholinguistic
Research34:365–399.DOI:10.1007/s10936-005-6139-3
Hartsuiker, Robert J., Martin J. Pickering, and Eline Veltkamp. 2004. “Is Syntax Separate
or Shared between Languages? Cross-linguistic Syntactic Priming in Spanish/English
Bilinguals.” Psychological Science 15: 409–414. DOI: 10.1111/j.0956-7976.2004.00693.x
Horton, William S. 2008. “A Memory-based Approach to Common Ground and Audience
Design.”In Intention, CommonGround, and the Egocentric Speaker-Hearer, ed. by Istvan
Kecskes and Jacob Mey, 189–222. New York:Mouton deGruyter.
Jackson, Richard. 2005. “Writing the War on Terrorism: Language, Politics and Counter-
terrorism(New Approaches to Conflict Analysis).” Manchester University Press.

Jucks, Regina, Becker Maria Bettina, and Rainer Bromme. 2008. “Lexical Entrainment in
Written Discourse:Is Expert’s Word UseAdapted to theAddressee?” Discourse Processes
45: 497–518. DOI: 10.1080/01638530802356547
Kawanani, Sachiyo, and Kawanani Kazuya. 2009. “Evaluation of World Englishes among
JapaneseJunior and Senior High schoolStudents.” Second Language Studies 27 (2):1–69.
Kolde, Gottfried. 1981. “Sprachkontakte in gemischtsprachigen Städten. Vergleichende
Untersuchungen über Voraussetzungen und Formen sprachlicher Interaktion verschie-
densprachiger Jugendlicher in den Schweizer Städten Biel/Bienne und Fribourg/ Friburg
i.Ue.” Zeitschrift für Dialektologie und Linguistik, Beihefte H.37. [Language contacts in
mixed-language towns. Comparative studies on conditions and voice interaction forms dif-
ferent language of young people in the Swiss cities of Biel / Bienne and Fribourg / Friburg
i.Ue. Journalof dialectology and linguistics, supplements H.37]
Krauss, Robert M., and Jennifer S. Pardo. 2004. “Is Alignment Always theResult of Priming?”
Behavioural and Brain Science 27: 203–204. DOI: 10.1017/S0140525X0436005X
Kroll, Judith F., Susan C. Bobb, Maya Misra, and Taomei Gue. 2008. “Language Selection in
Bilingual Speech:Evidence for Inhibitory Process.”ActaPsychologica 128:416–430.
DOI: 10.1016/j.actpsy.2008.02.001
Lee, Danielle H., and Peter Brusilovsky. 2009. “Reinforcing Recommendation Using Implicit
Negative Feedback.”In User Modeling, Adaptation, and Personalization, Lecture Notes in
Computer Science, ed. by Geert Jan Houben, Gord McCalla, Fabio Pianesi, and Massimo
Zancanaro, 422–427. Vol. 5535. Springer. DOI:10.1007/978-3-642-02247-0_47
Levelt, William J. M. 1989. Speaking fromIntentiontoArticulation. MassachusettsInstituteof
Technology.
Levelt, William J. M., and Stephanie Kelter. 1982. “Surface Form and Memory in Question
Answering.” Cognitive Psychology 14: 78–106. DOI: 10.1016/0010-0285(82)90005-6
Lüdi, Georges. 2007. “The Swiss Model of Plurilingual Communication.” In Receptive
Multilingualism.LinguisticAnalyses,LanguagePoliciesandDidacticConcepts,ed.byJanD.
Ten Thijeand Ludger Zeevaert, 159–178. Amsterdam: John Benjamins.
DOI:10.1075/hsm.6.10lud
Lyster, Roy, and Leila Ranta. 1997. “Corrective Feedback and Learner Uptake: Negotiation of
Formin Communicative Classrooms.”Studies inSecond Language Acquisition20:37–66.
Mehler, Alexander, Andy Lücking, and Peter Menke. 2011. “Modelling Lexical Alignment in
SpontaneousDirectionDialogueDatabyMeansofaLexiconNetworkModel.”InProceedings
of12thInternationalConferenceonIntelligent TextProcessingandComputationalLinguistics
(CICLing).
Menenti, Laura, Martin J. Pickering, and Simon Garrod. 2012. “Toward a Neural Basis of
Interactive Alignment in Conversation.”Frontiers inhumanneuroscience 6 (185):1–9.
DOI: 10.3389/fnhum.2012.00185
Mey, Jacob L. 2008. “Impeach or Exorcise?” or, What’s in the(Common) Ground.”In Intention,
CommonGround, and the Egocentric Speaker-Hearer, ed. by Istvan Kecskes and Jacob Mey,
254–275. Berlin: Mouton deGruyter New York. DOI:10.1515/9783110211474
Ni Eochaidh, Ciara. 2010. The Role of Conceptual and Word Form Representations in Lexical
Alignment: Evidence fromBilingualDialogue. (Masters’thesis). Edinburgh:TheUniversity
of Edinburgh.
Paradis, Michel. 2004. A Neurolinguistic Theory of Bilingualism. Amsterdam:John Benjamins.
DOI: 10.1075/sibil.18

Paradis, Michel. 1997. “The Cognitive Neuropsychology of Bilingualism.” In Tutorials in
Bilingualism: Psycholinguistic Perspectives, ed. by Annette M. B. De Groot and Judith F.
Kroll,331–354.Mahwah,NJ:LEA.
Paradis, Michel. 1993. “Linguistic, Psycholinguistic, and Neurolinguistic Aspects of Interference
in Bilingual Speakers: The Activation Threshold Hypotheses.” International Journal of
Psycholinguistics: 133–145.
Pickering, Martin J., and Simon Garrod. 2013. “AnIntegrated Theory of Language Production
and Comprehension.” Behavioral and BrainSciences 36 (4):1–64.
DOI: 10.1017/S0140525X11002160
Pickering, Martin J., and Simon Garrod. 2006. “Alignment as the Basis for Successful
Communication.” Research onLanguage and Computation4:203–228.
DOI: 10.1007/s11168-006-9004-0
Pickering, Martin J., and Simon Garrod. 2004. “TowardaMechanisticPsychology of Dialogue.”
BehavioralandBrainSciences27:169–225.
Richardson, Daniel C., Rick Dale, and NatashaZ. Kirkham. 2007. “The Art of Conversation Is
Coordination Common Ground and the Coupling of Eye Movements during Dialogue.”
Psychological Science 18 (5): 403–413. DOI: 10.1111/j.1467-9280.2007.01914.x
Roche, Jennifer M., Rick Dale, and Gina M. Caucci. 2012. “Doubling up on Double Meanings:
Pragmatic Alignment.”LanguageandCognitiveProcesses27(1):1–24.
DOI: 10.1080/01690965.2010.509929
Sacks, Harvey, EmanuelA. Schegloff, and Gail Jefferson. 1974. “ASimplest Systematic for the
Organization of Turn-taking in Conversation.” Language 50:696–735.
DOI: 10.1353/lan.1974.0010
Schenkein, Jim. (1980). “ATaxonomy for RepeatingAction Sequences in NaturalConversation.”
In Language productionVol. 1. Speech and talk. ed. by Brian Butterworth, 21–47. London:
Academic Press.
Schiller, Niels O., and Jan Peter de Ruiter. 2004. “Some Notes on Priming, Alignment, and
Selfmonitoring.” Behavioral and BrainSciences 27:208–209.
DOI: 10.1017/S0140525X0441005X
Schober, Michael F., and Herbert H. Clark. 1989. “Understanding by Addressees and
Overhearers.” Cognitive Psychology 21:211–232. DOI:10.1016/0010-0285(89)90008-X
Stalnaker, Robert A. 1974. “Pragmatic Presuppositions.” In Semantics and Philosophy, ed. by
Milton K. Munitz and Peter K. Ungerer, 197–213. New York:New YorkUniversity Press.
Weigand, Edda. 2009. Language as Dialogue: From Roles to Principles of Probability, ed. by
Sebastian Feller. Amsterdam:John Benjamins. DOI:10.1075/ds.5
Werlen, Iwar. 2007. “Receptive Multilingualism in Switzerland and theCaseof Biel/Bienne.”In
ReceptiveMultilingualism.Linguisticanalyses,languagepoliciesanddidacticconcepts,ed.by
Jan D. Ten Thijeand Ludger Zeevaert, 137–158. Amsterdam: John Benjamins.
DOI: 10.1075/hsm.6.09w er

Linguistic alignment in L1-L2 dialogue. Purmohammad

More Related Content

What's hot (20)

Similar to Linguistic alignment in L1-L2 dialogue. Purmohammad (20)

Linguistic alignment in L1-L2 dialogue. Purmohammad