International Journal of Multilingualism

ISSN: 1479-0718 (Print) 1747-7530 (Online) Journal homepage: http://www.tandfonline.com/loi/rmjm20

Ethnic identity and language choice across online

forums

Jessica Rae Birnie-Smith

To cite this article: Jessica Rae Birnie-Smith (2016) Ethnic identity and language choice
across online forums, International Journal of Multilingualism, 13:2, 165-183, DOI:
10.1080/14790718.2015.1078806

To link to this article: http://dx.doi.org/10.1080/14790718.2015.1078806

Published online: 10 Sep 2015.

Submit your article to this journal

Article views: 227

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rmjm20

Download by: [The University Of Melbourne Libraries] Date: 26 July 2016, At: 04:51
 INTERNATIONAL JOURNAL OF MULTILINGUALISM, 2016
VOL. 13, NO. 2, 165–183
http://dx.doi.org/10.1080/14790718.2015.1078806

Ethnic identity and language choice across online forums

Jessica Rae Birnie-Smith
Department of Linguistics, Monash University, 28 Canterbury St, Hughesdale, Melbourne, VIC 3166, Australia

ABSTRACT ARTICLE HISTORY

This paper examines the language choice and ethnic identity Received 18 September 2014
Revised 13 June 2015
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

construction online of four young Chinese Indonesians from West

Kalimantan, Indonesia. The study draws on a combination of Accepted 24 July 2015
Social Identity model of De-individuation Effects (SIDE) theory and
KEYWORDS
audience design theory to formulate a hypothesis about Chinese diaspora; audience
participants’ linguistic behaviour within different social design; SIDE theory; Internet;
circumstances online. Broadly, the hypothesis stated that bilingualism
participants would adjust their language choice and self-
presentation to suit different online circumstances. Online social
variables such as levels of anonymity, audiences, group identity
and personal identity were predicted to impact on participants’
linguistic behaviour. The hypothesis was evaluated through
analysis of online interviews with participants and observations of
participants’ interactions on two social networking sites, Kaskus ®
and Facebook®. The paper shows that participants draw on their
multilingual linguistic repertoires to construct their ethnic identity
in different ways according to the social variables they encounter
on each site. The results are generally consistent with the
hypothesis of the study; however SIDE theory and audience
design could not definitively explain the code-switching practices
of participants in interactions with others of the same ethnicity on
Facebook®.

Introduction
The peculiarity of the Internet has been of interest to researchers studying multilingualism
and identity for the past few decades (Childs & Mallinson, 2006; Kendall, 1998; Kolko, Naka-
mura, & Rodman, 2000; Nakamura, 2002; Turkle, 1995). Inescapable facets of one’s offline
identity such as race can be lost in the space between offline and online worlds (Christo-
pherson, 2007; Kolko et al., 2000). However, research has shown that ethnic identity online
is usually grounded in some physical reality (Christopherson, 2007; Kolko et al., 2000; Naka-
mura, 2002), but can vary in presentation according to contextual shifts (Androutsopoulos,
2006). Previous studies of ethnic identity online (Androutsopoulos, 2006; Childs & Mallin-
son, 2006; Dorleijn & Nortier, 2009; Hinrichs, 2006; Kumar Sahoo & de Kruijf, 2014; Miller &
Slater, 2000), have shown that members of ethnic migrant or diaspora groups can select
languages from their multilingual linguistic repertoires to construct various identities
online.

CONTACT Jessica Rae Birnie-Smith jessica.birniesmith@monash.edu

© 2015 Taylor & Francis
 166 J. R. BIRNIE-SMITH

Growing research on the ‘multilingual internet’ (Danet & Herring, 2007; Wright, 2004)
has thus far largely neglected migrant and diaspora communities (Childs & Mallinson,
2006; Kumar Sahoo, & de Kruijf, 2014; Miller & Slater, 2000). Migrant and diaspora commu-
nities present important opportunities to researchers of multilingualism online because
the Internet is one medium through which members of these communities can ‘negotiate
their dual identity’ (Mitra, 1997, p. 64) through use of their multilingual repertoire.
The ethnic Chinese (Tionghoa) community of Indonesia possess a complex sociolinguis-
tic situation which impacts on patterns of language choice and identity online. Tionghoa
identity consists of Chinese ancestry and Indonesian nationality (Handoko, 2009; Oetomo,
1987). The linguistic repertoire of Tionghoa can consist of both Chinese and Indonesian
languages. Although the ethnic Chinese community has been stigmatised in the past
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

(Handoko, 2009; Heidhues, 1996; Oetomo, 1987; Purdey, 2006), Chinese language is
now highly valued and is considered an important aspect of Tionghoa identity. Therefore,
online Tionghoa users may draw on their Chinese and Indonesian languages to negotiate a
dual identity.
Prior research on different diaspora and migrant communities has often been focussed
on one online site for research (Childs & Mallinson, 2006; Kumar Sahoo & de Kruijf, 2014;
Miller & Slater, 2000), and has not yet thoroughly investigated how different online con-
texts can impact on language choice and construction of identity.
In this paper, I will respond to the following research questions:
(1) How does participants’ language use and presentation of identity differ across Face-
book ® and Kaskus ®?
(2) How well can Social Identity model of De-individuation Effects (SIDE) theory and audi-
ence design predict language choice and identity construction across the two sites?
The two sites have very different characteristics that may have different effects on the
way users interact and construct identities online. Facebook ® is an extremely popular
global social network which invites users to find and connect with friends from around
the world (Alexa, 2013). Kaskus ® is the largest Indonesian forum on the Internet. The site
consists of numerous forums and threads where users can discuss a wide range of
topics that range from pop culture to political issues. As a result, participants draw on
various languages including Bahasa Indonesia, Teochew Chinese and English to construct
identities differently across different contexts on Facebook ® and Kaskus ® forums. Partici-
pants’ language choice and identity construction varies across the two sites due to vari-
ation in levels of anonymity and group salience. In addition, participants’ language
choice and self-presentation is also impacted by their perceived audiences.
SIDE theory will be used alongside audience design theory to analyse the impact of
anonymity, groups and audiences online on participants’ language choice and identity
on Facebook ® and Kaskus ®. The following section includes a discussion of SIDE theory
and audience design theory and how they are applied to this study.

SIDE theory and audience design

Prior research on online communities has demonstrated that individuals’ ethnic identity
construction is influenced by several social variables in their online interactions (Kolko
et al., 2000; Miller & Slater, 2000; Nakamura, 2002). The level of anonymity made available
to them, the language preferences of the forums, the audiences of their interactions as
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 167

well as their own personal identities are all factors which affect participants’ identity con-
struction online (Christopherson, 2007; Haines & Mann, 2011; Kim, 2009; Klein, Spears, &
Reicher, 2007; Lea & Spears, 1991). These variables are linked to key differences in the
online forums within which individuals interact.
SIDE theory accounts for how behaviour within groups is mediated by anonymity and
self-categorisation (Christopherson, 2007; Postmes, Spears, Sakhel, & De Groot, 2001).
Drawing on SIDE theory, Lea and Spears (1991, p. 288) suggest that when all members
of the same group are anonymous to one another, differences between members are
invisible. And so, group salience will increase and members will identify more strongly
with the group. The impact of the group is generally strong in CMC environments as anon-
ymity is high, and so CMC participants are likely to enforce group norms (Christopherson,
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

2007, p. 3048). By contrast, if members of a group are made identifiable to one another
(e.g. through real names and photographs), then differences between individual
members will be visible (Lea & Spears, 1991, p. 288). As a result, personal identity will
become more salient than social (group) identity online and members will identify less
strongly with the group. Therefore, according to SIDE theory varying levels of anonymity
and personal identification impact on how users behave online.
In the current study, the SIDE model is extended to predict linguistic behaviour on Face-
book ® and Kaskus ® forums. Prior research using SIDE theory has shown that language
choice can be impacted by contextual factors including personal identity, anonymity
and group salience (Christopherson, 2007; Klein et al., 2007; Lea & Spears, 1991;
Postmes et al., 2001). Lea and Spears (1991) state the behaviour of the individual is
mediated by their awareness of the group norms, which are heightened in anonymous
conditions. Bell’s (1991, 2002) audience design theory goes a step further by suggesting
that speakers’ language choices are primarily motivated by their perceived audiences.
The purpose of audience design theory is to explain how speakers adapt their linguistic
practices and self-presentation to resemble their perceived audiences (Bell, 2002, p. 148).
According to audience design, there are different kinds of audiences. First there are
addressees who are listeners who are known, ratified and addressed by the speaker.
There are also auditors, who are not directly addressed but are known and ratified. Over-
hearers are non-ratified listeners whom the speaker is aware of. Finally, eavesdroppers are
non-ratified listeners whom the speaker is not aware of (Bell, 1991). Bell (1991, p. 13)
explains that the degree to which audience members have an impact on speakers’ style
shifting is dependent on the degree to which speakers are aware of their various audi-
ences and ratify them.
Bell (2002, p. 153) stated that the salience of the audience has a significant impact on
language choice and style. Using SIDE theory, it is possible to account for the impact of
situational factors such as anonymity in different forums, on the salience of the audience.
According to SIDE theory, anonymity increases the salience of groups (Lea, Spears, & De
Groot, 2001; Postmes, Spears, & Lea, 2000; Postmes et al., 2001). Postmes et al. (2001,
p. 1244) explained that group identities and group norms often formed prior to the occur-
rence of interactions. If groups exist prior to interactions, self-categorisations as either in-
group or out-group members occur before individuals interact with the group (Postmes
et al., 2001). Individuals can establish themselves as in-group or out-group members by
converging towards or diverging from the linguistic preferences of the groups with
whom they interact (i.e. their audiences) (Bell, 2002, p. 147). In anonymous circumstances,
 168 J. R. BIRNIE-SMITH

the influence of the audiences on language choice will increase as the audiences are more
visible. And so, individuals continually redefine themselves in relation to their audiences.
The current study focuses on the impact of participants’ audiences, including their
addressees, auditors and overhearers on their language choice and identity construction.
Addressees are considered the individual users with whom participants interact on Face-
book ® and Kaskus ®. Auditors are members of the groups they interact with (their friends on
Facebook ® or thread participants on Kaskus ®), whom they are not directly addressing, but
nonetheless recognise. The overhearers on both these sites would be all those who are
able to view participants’ status updates on Facebook ®, or their forum posts on Kaskus ®.
All of the audiences participants encounter may have an impact on their language
choice to varying degrees. The extent to which audiences impact on participants’ linguistic
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

practices in anonymous and identifiable conditions online will be discussed with reference
to interview responses and their interactions on Facebook ® and Kaskus ®.
Using the combination of SIDE theory and audience design theory, I formulated a
hypothesis on the four Tionghoa users’ language choice and identity construction
across Kaskus ® and Facebook ® forums. On Kaskus ®, anonymity is higher than it is on Face-
book ® as users select pseudonyms and avatars to represent themselves. Interactions on
Kaskus ® occur between users who are unknown to one another. Therefore, on Kaskus ® par-
ticipants are predicted to identify more strongly with the groups with whom they interact,
than their own personal identities. Additionally, the rules of each forum and thread are
outlined in introductory posts, therefore the prescribed norms of each group (including
language choice) would be highly salient. Participants are predicted to follow the norms
of the groups with whom they interact and choose whichever language is prescribed to
communicate.
On Facebook ® users are generally visually identifiable to one another through photo-
graphs of themselves in display pictures and the use of their real names as usernames.
The social factors of interactions on Facebook ® thus suggest that participants would
likely adhere more strongly to their personal identity than any group identity. On Face-
book ®, participants will use a wider variety of languages including Indonesian, English
and Teochew, or even some combination of languages to demonstrate their personal
identity. However, participants are nonetheless aware of the groups with whom they inter-
act on Facebook ®, as their friends are ratified in their friend lists. Participants are predicted
to design discourse for their audience of majority Indonesian speakers by writing status
updates using Indonesian language.
In sum, the four Tionghoa participants were predicted to express their ethnic identity
through use of their multilingual repertoire in different ways across the two sites. The fol-
lowing sections will examine the validity of this hypothesis.

Participants of the study

Participants for this study were recruited on the Kaskus ® forum Bisa Bahasa Tiociu Kumpul
Sini ‘Teochew Speakers Meet Here’. The criteria for selection were that participants be Tion-
ghoa aged between 18 and 30, and have originated from Pontianak, West Kalimantan.
Young users were sought because they form part of the ‘Internet Generation’ (Herring,
2008). Four Kaskus ® users (Kaskusers), two female and two male, were chosen to participate
in this study to enable an in-depth analysis of their online identity practices. The four
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 169

individuals were chosen because they were all quite different from one another in terms of
their experiences, language proficiencies, attitudes, ideologies and physical locales. It was
thought that these differences may impact on each participant’s language use and self-
presentation online, and would therefore be effective in assessing the generalisability of
predictions made using SIDE theory and audience design theory.
Drawing on online ethnographic methodology of prior online studies (Androutsopou-
los, 2008; Childs & Mallinson, 2006; Miller & Slater, 2000), participants underwent obser-
vations and interviews online. The ethnography investigated the impact of participants’
personal identities, group identities, their interlocutors and the online setting on their lin-
guistic styles.
Participants’ daily interactions on Facebook ® and Kaskus ® were monitored over a period
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

of two months. All of participants’ posts in Kaskus ® forums as well as all their status updates
and comments on their own personal Facebook ® walls over the two months were collected
and analysed. The analysis procedure involved performing discourse analysis on interac-
tional data and comparing the analysis with the predictions made about participants’
language choice across the two sites to evaluate the hypothesis. Further information on
the analysis can be found in the discussion of extracts of participants’ interactions on
both sites which exemplify the overall findings of the study.
Additionally, three semi-structured interviews were conducted periodically with each of
the four participants via email over the two months. In the first interview, participants were
asked about their basic information including their work, living situation, linguistic reper-
toire and their ethnicity, to establish sociolinguistic variables that may impact on their lin-
guistic practices online. The second interview required participants to describe their online
social networking activities including the linguistic features they used on different forums
and who they interacted with online. Finally, participants were asked to discuss their
language ideologies and perceptions of their ethnicity and their identity overall. The
aim of the final semi-structured interviews was to gain an in-depth understanding of par-
ticipants’ attitudes towards their ethnic identity and language ideologies that may have
influenced their online behaviour. For further information on the interview questions,
see the appendix.
Interviews were also supplemented by ‘ongoing conversations’ (Oetomo, 1987) which
emerged as participants were invited to email me outside of our scheduled interviews if
there was anything they wished to discuss further. Participants and I therefore built on
the discussion stimulated by the three interviews in a number of unstructured conversa-
tions via email, which formed part of several ongoing conversations about ethnicity, iden-
tity and language online.
In this study, interviews were conducted in Indonesian because that was the preference
of the participants and myself as it is our shared language code. All quotes from partici-
pants’ interviews presented in this paper were translated from Indonesian to English by
myself.
The participants are outlined in Table 1 and discussed below.
The first participant to self-select for this study was Hadi.1 Hadi is 26 years old, and lives
with his wife (also Chinese Indonesian) in Sydney, Australia, where he works as a graphic
designer. Hadi studied Graphic Design for two years at a Sydney college then studied
Digital Media for another two years at a Sydney university. He moved to Australia eight
years ago from Pontianak where he lived with his parents and his older brother and
 170 J. R. BIRNIE-SMITH

Table 1. Participant information.

Gender First
Name M/F Age Current location Occupation language(s) Other languages
Dewi F 23 Yogyakarta, Student (Business Teochew, Indonesian, Hakka, Mandarin,
Indonesia Management) Cantonese English, Dutch, Russian
Hadi M 26 Sydney, Graphic Designer Teochew Indonesian, Mandarin, English
Australia
Yanti F 28 Pontianak, Administrative Teochew Indonesian
Indonesia Assistant
Agus M 21 Semarang, Student (Civil Teochew Indonesian, Mandarin, Malay,
Indonesia Engineering) Javanese

sister. He also returns to Pontianak once or twice each year. Hadi’s first language is
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Teochew, but he also speaks Indonesian, English and some Mandarin that he learnt at
school.
The second participant of this study is Yanti. She is 28 years old and lives with her
mother in Pontianak, where she has resided all her life. Yanti works in administration at
a diesel engine supply warehouse in Pontianak. Yanti’s first language is also Teochew
and she also speaks Indonesian. She did not have the opportunity to learn Mandarin at
school because Mandarin language classes were only introduced to private catholic
schools in Pontianak just after she graduated high school.
The third participant is 21-year-old Agus. Agus is currently living in Semarang, in Java,
where he attends university. He moved to Semarang from Pontianak one year ago to
pursue studies in civil engineering. In Pontianak, he lived with his parents but in Semarang
he is living with friends. Like the other two participants living away from Pontianak, he is
able to return to his city of origin twice a year during the university holidays. His first
language is Teochew and he also speaks Indonesian, Malay and Javanese. He has also
learnt some Mandarin at university.
The final participant, Dewi, is 23 years old and is currently living with her younger sib-
lings in Yogyakarta, Java where she studies Business Management. Before she moved to
Yogyakarta almost six years ago, she lived in Pontianak where she was born and raised.
Since moving to Yogyakarta she has travelled back to Pontianak once or twice each
year, during the university holidays, to spend time with friends and family. Dewi is of
mixed Chinese ancestry as her mother is Cantonese and her father is Teochew (tio ciu
nang ‘Teochew person’). Her first languages are Cantonese Teochew and Indonesian.
She also speaks Hakka, English and some Russian. She also learnt some Mandarin language
in primary school in Pontianak.
These four participants are all Tionghoa from Pontianak who share an ethnic language,
Teochew, however they differ in that they all live in different cities and have different life
experiences that may result in significant differences in their perspectives and attitudes
towards their ethnic identity, as well as their linguistic practices and self-presentation
online.

Language and identity online

Online, participants’ language use and identity construction was influenced by the contex-
tual factors they encountered on Kaskus ® and Facebook ®. This section will discuss the
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 171

impact of situational specific variables including anonymity, group norms, individual iden-
tity and audiences, on language choice and identity online. Additionally, this section will
evaluate the validity of the hypothesis on participants’ language use and identity in
relation to SIDE theory and audience design theory.

Language choice and identity construction on Kaskus ®

As previously discussed, on Kaskus ®, participants were presented with a high level of
anonymity as pseudonyms and avatars veil the identity of individual users. Interactions
on Kaskus ® occur within forums, the creators of each forum thread or thread starters
(TSs) often outline the rules and regulations of the thread in an introductory post.
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

These rules can govern the topic and content of posts and, in some cases, the
language that posts are to be written in. On most threads where no language is pre-
scribed, Indonesian appears to be the language of choice for most Kaskus ® users.
Although many Indonesian-medium Kaskus ® threads do not specify a particular
group identity, participation on these threads forms part of a larger Kaskus ® user (Kas-
kuser) identity. The existence of Kaskuser identity is evidenced by the use of Kaskus ®
Indonesian language style and technical terms that are viewed as typical of Kaskus ®
language.
By contrast, the creator of the thread Bisa Bahasa Tiociu Kumpul Sini ‘Teochew speakers
meet here’ created a more specific Kaskuser group identity – that of Teochew speaking
Kaskusers. On this thread, Teochew is the expected code choice as the TS states that
users’ posts should contain as much Teochew language as possible.
The rules governing topic and language increase the visibility of the group and the
group norms in the larger forum as well as within particular threads. High salience of a
group and group norms were predicted to increase association with the group (Lea
et al., 2001; Postmes et al., 2000, 2001). When individual users choose to be part of the
group, they establish in-group status by designing discourse to follow the group norms,
which further highlights the group identity (Christopherson, 2007). All four participants
of the study followed the norms of the groups they interacted with by using Indonesian
language and Kaskus ® style features in Indonesian-medium forums and by using
Teochew in the Teochew-medium forum.
In interviews, participants demonstrated a high awareness of their audience. Partici-
pants claimed that their use of Indonesian on Kaskus ® was practical because ‘it’s the
national language, everyone can understand it’ (Agus). As Kaskus ® is an Indonesian
forum, the website is composed in Indonesian. Therefore the participants on the
forums consisted mostly of Indonesian speakers. Kaskusers have no control over who
views their posts on forums because forums are open to the public. Therefore an indi-
vidual’s perceived audiences can consist of their addressees (members of each thread
or forum for whom they design their posts), auditors (wider Kaskuser community of
Indonesian speakers) and overhearers (other Internet users who are able to view
their posts). Thus in choosing Indonesian to write posts on Kaskus ® threads, participants
were designing discourse for a wide range of Kaskusers who made up their audience
(cf. Bell, 2002).
 172 J. R. BIRNIE-SMITH

The following extracts are taken from data collected over two months on participants’
daily interactions on Kaskus and Facebook. The original text is presented on the left and a
translation into English is provided on the right. In the original text, Indonesian is rep-
resented in italics, Teochew is italicised and bolded and English is in plain text. The one
word written in Javanese is underlined and italicised in Extract 6.
The first example below illustrates Hadi’s use of Indonesian to communicate with other
Kaskusers within Indonesian-medium forums. In this example, Hadi writes an opening
message on a thread he starts about the banning of Lady Gaga in Indonesia. Hadi is
designing discourse for his addressees and conforming to the norms of the Kaskus ® com-
munity by using Indonesian.
Extract 1: Hadi’s post on Kaskus ® forum
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Hadi
1 sekarang gue mengerti, kenapa gaga [Now I understand why Gaga is banned
2 dilarang di indo..inilah in Indonesia..this is
3 yang terjadi. anak muda Indonesia yang what happens when young Indonesians
4 TERLALU GAGA are TOO GAGA
5 GAGA … jadinya GAGAL dah.. GAGA … they become CRAZY..
6 di TV ONE nih gan.. this is on TV ONE..
7 ane screenshot cepet2 tadi ^^ I’ll screenshot it really soon ^^]

Hadi used Indonesian to open the discussion on this thread. Hadi’s language choice
can be analysed in relation to two predictions made using SIDE theory and audience
design. First, as predicted, Hadi designed discourse for his addressees. Bell (1984)
explains that speakers converge towards the language preferences of their addressees.
Hadi responded to his addressees by designing discourse for them in Indonesian, the
language that he perceived them to prefer. Secondly, Hadi conformed to the norms of
the wider group of Kaskusers (auditors) and reinforced the norms in the smaller group.
Lea and Spears (1991) stated that in anonymous online conditions, individuals tend to
adhere to and enforce the group norms. As the TS, Hadi subtly outlined the norms of
the thread by stating the topic of the thread in the preferred language on Kaskus ®
(Indonesian).
Hadi used a Kaskus ® style to write his post, which conforms to the norms of the wider
Kaskuser group and establishes Hadi as an in-group member. The pronoun ane ‘I’ is a
feature of Kaskus ® style. Ane is originally an Arabic word that has made its way into
slang used on Kaskus ®. Dewi explains this pronoun is used because ‘“ane” sounds informal
(way more informal than “aku”) and more slang … “ane” is the hallmark of Kaskus kids,
Kaskus kids always use “ane” rather than “aku”’.
Hadi and other participants’ use of Indonesian and Kaskus ® style to post on Indonesian-
medium Kaskus ® forums confirms the hypothesis that participants would follow the group
norms, associate strongly with the group and design discourse using Indonesian language
within the wider Indonesian-medium Kaskus ® forum.
On the Teochew language thread (Bisa Bahasa Tiociu Kumpul Sini), participants all
selected Teochew to communicate with others. Participants expressed an awareness of
their addressees that motivated their use of Teochew on the thread. Yanti stated that
‘there are only Tionghoa people [on the thread] because there we only use Teochew’.
Dewi said that she similarly assumes that all users on the thread are Tionghoa ‘because
of the eloquence of their written Teochew, very few pribumi Indonesians can speak
Teochew’. Hadi also explained that he uses the Teochew thread ‘to communicate
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 173

among fellow tiociu nang “Teochew people”’. Referring to the people on the thread as
tiociu nang ‘Teochew people’ suggests that Hadi presumes his audience on the thread
to be members of the Tionghoa ethnic group, and more specifically tiociu nang
‘Teochew people’.
In the following example, Agus and Yanti converse on the Teochew thread in Teochew.
Yanti and Agus followed the rules of the group by using Teochew and designing discourse
for their audience of Tionghoa Teochew speakers.
Extract 2: Yanti and Agus interacting on Kaskus ® Teochew forum
Yanti
1 oh le ken jawa toh … .. [Oh you’re in Java yeah …
2 wa kin khuntien a I’m in Pontianak]
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Agus
3 si wa , tho jaw a wa..ai cho wong [Yeah I’m in Java..and I’ve become
4 jowo ow Javanese.
5 oh khuntien nang hee? Oh a Pontianak person yeah?
6 wa kai mak khai thi heeng a. My mum is a Hulu
7 hahah.thak che apa cho kang ow? hahah are you in school or working?
8 oh he wa , mek kai mia le? Oh yeah me, what now?
9 wa Agus I’m Agus
10 wa thoi siooo yammie , yamkwetiau , I want to eat yammie, yamkuetiau and
11 phue kii mue congee]

By using Teochew in this instance, Yanti and Agus are conforming to the rules of the
thread to use ‘as much Teochew as possible’. Additionally, Bell (2002) stated that speakers
with a high awareness of their audiences are likely to use the languages that they perceive
their audiences to prefer. On the Teochew thread, the participants’ awareness of their
audience is high, as discussed above. Therefore, Yanti and Agus’ use of Teochew here
can be considered a response to their addressees. In this instance, participants are more
focussed on conforming to the norms of their addressees than their auditors (wider
Kaskus ® community). The norms of the auditors include using Indonesian language or Kas-
kuser style whereas the norms of the addressees involve the use of Teochew language. Bell
(1984, p. 160) analysed this phenomenon as the result of the graded influence of different
audiences that directly correlates with the proximity (physical or abstract) of the audience
to the speaker. And so, addressees have the closest proximity to the speaker, and therefore
have the greatest influence on the speaker’s language choice, followed by the auditors
and finally overhearers.
Additionally, the addressee group identity and norms are more salient than those of
the wider auditor group because the use of Teochew is a specified prerequisite of the
thread and the thread was created for ‘Teochew speakers’. The norms of the Kaskus ®
community in general are not ostensibly prescribed on the wider forum. Researchers
using SIDE theory have suggested that in circumstances where the norms of the sub-
group are more salient than those of the larger group, individual members are more
likely to conform to the norms of the smaller, more salient group (Christopherson,
2007; Postmes et al., 2000).
In sum, as on the Indonesian forums, participants linguistic practices followed the
norms of their addressees in the Teochew language forum, as predicted. They designed
discourse for their audience and continually reinforced the group identity (Tionghoa)
rather than promoting their individual identities.
 174 J. R. BIRNIE-SMITH

Language choice and identity construction on Facebook ®

On Facebook ®, participants encounter vastly different circumstances to those within
Kaskus ® forums. Anonymity on Facebook ® is near non-existent. Interactions on Facebook ®
typically occur between ratified friends on individual users’ Facebook ® walls. The process
of becoming ‘friends’ on Facebook ® involves performing a search of the individual’s user-
name (often their actual name), before they can be ‘added’ as a friend (Facebook ®, 2012).
Users are identifiable in profile pictures which often include photographs of themselves.
On Facebook ®, there are no rules governing discussion topics or language choice in inter-
actions on Facebook ® walls.
Unlike on Kaskus ®, there is no ubiquitous Facebook ® identity which users can claim
membership to by using a particular language style, nor are there as clearly defined
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

sub-groups as that of the Teochew forum on Kaskus ®. Therefore, as stated in the hypoth-
esis, it was predicted that participants would focus more on expressing their personal
identity through language rather than conforming to perceived group norms. Lea and
Spears (1991) explained that low levels of anonymity and low group salience but high indi-
vidual salience result in increased adherence to personal identity. However, interactions
on Facebook ® are often initiated by users posting status updates on their personal walls.
And so, participants were predicted to design discourse in status updates for their audi-
ence of ratified friends.
Interviews showed that participants perceived the vast majority of their audience to
consist of Indonesian speakers. Agus rationalised his use of Indonesian as ‘it’s the national
language, everyone can understand it’. Yanti elaborated that ‘not everyone can under-
stand English or Teochew’.
The reasons, ‘everyone can understand it’ and ‘not everyone can understand English
or Teochew’, suggest that participants’ perceive Indonesian as the language that the
majority of their Facebook ® friends can understand, if not prefer to use. Participants’
responses showed that they had a high awareness of their addressees, auditors and
overhearers on Facebook ®. Therefore, when they used Indonesian to write status
updates, the participants designed discourse for their Indonesian-speaking audiences
(cf. Bell, 1991, 2002). Alternatively, participants’ interview responses may also indicate
the existence of unwritten norms of Indonesian Facebook ® users that include employing
Indonesian language. Participants may highlight their group identity as Indonesian
users by using Indonesian language or emphasise their personal identity by deviating
from the norms. The results of the analysis that follows demonstrate both practices
as all the participants wrote status updates in Indonesian to design discourse for
their audience and contribute to the norms of Indonesian users. However, all four par-
ticipants’ regularly deviated from the norm by writing in other languages, and so
appear to have highlighted their personal identities or, perhaps, an alternative group
identity.
In the following example, Hadi designs discourse for his Indonesian-speaking
Facebook ® friends in a status update. Users A and B then respond to the status
update by posting comments in the same language and creating a joke at Hadi’s
expense.
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 175

Extract 3: Hadi’s status update in Indonesian

Hadi
1 Saya mirip orang [I resemble a human]
User A
2 Utan [An orang-utan]
User B
3 Berarti situ, hehehehe … nyemot donk [That’s it, hehehehe … monkey dude]

Hadi’s use of Indonesian in this example could be motivated by a number of contextual

factors. Firstly, Hadi is designing discourse for his addressees, his Facebook ® friends, almost
all of whom are Indonesian speakers. Bell (1991, 2002) explained that speakers converge
towards the language preferences of their addressees. Secondly, Hadi’s use of Indonesian
in this example can be analysed as him conforming to perceived norms of Indonesian
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Facebook ® users and highlighting his own membership to this group.

All participants wrote status updates in Indonesian on Facebook ®, however Hadi, Dewi
and Agus regularly deviated from the Indonesian norms by writing status updates in
English. This finding is especially significant because all three stated that the majority of
their Facebook ® friends were non-English speaking. Dewi wrote approximately three quar-
ters of her status updates over two months in English. In the following example, Dewi uses
English to write a status update on Facebook ® about the Shmuel HaNavi bus bombing in
Israel.
Extract 4. Dewi’s status update in English
1 Dewi Wondering why the media did not report about the Shmuel
2 HaNavi bus bombing

The majority of Dewi’s Facebook ® friends are Indonesian (or Tionghoa) and speak little or
no English. Therefore she is not designing discourse for her perceived audience. Dewi’s
code choice in Ex. 4 may thus be indicative of her desire to distance herself from Tionghoa
and non-Chinese Indonesian communities. Dewi criticised the Tionghoa and Indonesian
communities in interviews, and therefore may have preferred to distance herself from
both identities. She said that although she considered herself to be Tionghoa, she
prefers to call herself Cantonese to distance herself from some groups of Tionghoa. She
states that she does not particularly like the Tionghoa of Pontianak:
Dewi Over there [in Pontianak], whilst they [Tionghoa] don’t have much money, they like to act
like they have a lot of money, usually they don’t want to lose anything (in a material sense) and
they like pampering, feeling exclusive and lots of other things that make me not like Pontianak
Chinese. I don’t hate Pontianak Chinese, it’s just most that I have met are like that, but not all
are. There are some that aren’t like that.

Dewi is also critical of the Indonesian community, and therefore has equivalent distaste
for her classification as an Indonesian. She says that she does not value her Indonesian
identity because she is so disappointed with the Indonesian government and ‘some
narrow-minded people who like to associate matters with religion when they are actually
purely political’. She clarifies that she is not un-nationalistic but her sense of nationalism
has worn away over time because ‘this country is heading for destruction but no one
cares’.
Dewi further stated that she actively tried to use English on Facebook ® to practice her
language skills so they do not become ‘rusty’. She also stated that ‘there are words in
 176 J. R. BIRNIE-SMITH

English that if transferred into Indonesian lose some meaning’. Postmes et al. (2000,
p. 1246) stated that differences between individuals and the groups they belong to are
more visible in identifiable conditions than in anonymous conditions. Therefore Dewi’s
use of English in the above example highlights the differences she perceives between
herself and other Indonesian and Tionghoa users, and increases the salience of her per-
sonal identity over the group identities.
The trends demonstrated in the two examples above conform to predictions that par-
ticipants would design discourse for their audience of Indonesian speakers and adhere to
their personal identity over the group identity.
Data on interactions between fellow Tionghoa users on Facebook ® did not necessarily
conform to the predictions of the study. Analysis of participants’ language choice on Face-
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

book® revealed that in interactions with Tionghoa users, participants consistently code-
switched between Teochew and Indonesian. It is possible, given the regularity of the
code-switching, that speaking both Teochew and Indonesian forms part of the communi-
cative norms associated with Tionghoa on Facebook ®.
In interviews, participants all expressed the importance of speaking Teochew with other
Tionghoa people in both offline and online social contexts. Agus stated that he considered
it essential for young Tionghoa to continue to maintain Chinese languages which he
viewed as ‘the characteristic, the identity, the custom and the culture of Tionghoa citizens’.
Likewise, Hadi and Dewi claimed that speaking Teochew was an integral part of ‘being
Tionghoa’. Yanti also agreed that Teochew was important; however, when discussing
her use of Teochew with other Tionghoa on Facebook ®, she stated ‘we [Tionghoa] just auto-
matically talk like that’. All of the participants’ comments expressed the perceived impor-
tance of using Teochew to speak with fellow Tionghoa. However Yanti’s comment further
suggests that Teochew language use is a norm of Tionghoa users on Facebook ®. Partici-
pants’ interview responses may explain the motivations behind the use of Teochew on
Facebook ®, however they do not account for the code-switching practice. The question
arises therefore as to why individuals alternate between Teochew and Indonesian on Face-
book ® but not on Kaskus ®.
On Kaskus ® participants used only Teochew, which may be due to the regulations of the
Teochew language forum. On Facebook ®, language choice is not prescribed so participants
had more freedom to select the language(s) that they perceived as appropriate for each
individual communicative setting. In this sense, participants may use Teochew with Tion-
ghoa users to respond to the expectations of their addressees. By code-switching Teochew
and Indonesian, participants may be highlighting their personal identities as a mixture of
Chinese ancestry and Indonesian nationality. Or, from a different perspective, the code-
switching practices may be an attempt to create a different group identity, that is not
purely Indonesian or Chinese, and whose norms involve code-switching between
Teochew and Indonesian. The code-switching practice was not mentioned by any of
the participants when they were asked to describe their language use when communicat-
ing with Tionghoa friends on Facebook ®. When I asked participants about their tendency to
code-switch they were reluctant to label it as code-switching. They considered it to be
merely them speaking Teochew on Facebook ®. This may suggest that the code-switching
is not a marked expression of personal identity, but may form part of the norms for com-
municating with Tionghoa on Facebook ®.
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 177

In the following examples, participants switch to Teochew when conversing with other
Tionghoa on Facebook ®. The following examples will show that participants use Teochew
in different ways for a variety of possible reasons. In the first example, Dewi responds to
her friend’s request to play a Chinese online game called Ant Wars by code-switching
between Teochew and Indonesian. Her code-switching can be interpreted as a response
to her addressee, an indication of her personal identity or a norm of a Tionghoa group
identity.
Extract 5: Conversation on Dewi’s Facebook ® wall
User A [Come join me, play Ant Wars
1
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Ayo ikuti saya main ke Perjuangan Semut.

2 Game tembak yang manis, santai A shooting game that’s sweet, relaxing
3 dan rangsang, Anda bias and exciting, you can
4 berjuang bersama teman dengan fight alongside friends with
5 berbagai macam senjata dan alat! different kinds of weapons and tools!
6 Cepat ikut berjuang! Quickly come fight!]
Dewi
7 ternyata aing he. haha. [In fact it’s Aing, yeah. Haha
8 wa sio kaw diang. Who am I thinking of]

User A initiates the conversation by posting an automated message on Dewi’s wall. The
post is in Indonesian because the User A has adjusted their Facebook ® settings so that
Indonesian is the default language for all sent and received Facebook ® notifications (Face-
book @, 2012). The Chinese Mandarin text in the picture reads: Insect Special Attack Forces.
Dewi response begins in Indonesian, with the adverbial ternyata ‘in fact’, and then
switches to Teochew for the remainder of the phrase.
The code-switching can be interpreted in several ways in relation to audience design
and SIDE theory. Firstly, Teochew may be a response to Dewi’s addressee. As previously
explained, Facebook ® users who are ratified friends tend to be known to each other in
the offline world. The utterance ‘in fact it’s Aing’ further suggests that Dewi knows who
the user is and may have a pre-existing relationship with him or her. Therefore she may
use Teochew to communicate here because she knows that Teochew is preferred by
User A (cf. Bell, 1984, 1991). However, Dewi established in interviews that the majority
of her Facebook ® friends were non-Chinese Indonesians, and so, whilst she is designing dis-
course for her addressee, she is not ratifying her auditors or overhearers. As stated pre-
viously, addressees have a greater influence on language choice than auditors or
overhearers (Bell, 1984, p. 160). And so, Dewi’s addressee, User A, has evidently more
impact on Dewi’s language than her other Facebook ® friends (auditors) or potential onloo-
kers (overhearers).
Additionally, Dewi’s code-switching can be interpreted on a personal identity or group
identity level. As previously stated, applications of SIDE theory have demonstrated that in
visible conditions (as is the case in the above example), individuals are more likely to
adhere to the norms of their personal identity than those of the group identity (Christo-
pherson, 2007; Postmes et al., 2000, 2001; Reicher, Spears, & Postmes, 1995). Hence, on
 178 J. R. BIRNIE-SMITH

a personal level, Dewi’s code-switching may symbolise her personal identity, and highlight
differences between herself and both Chinese and Indonesian groups. Dewi’s interview
responses also give credence to this interpretation because, as previously mentioned,
she did not like being associated with either Indonesian or Tionghoa groups. Therefore
she would be expected to attempt to distance her personal identity from both groups.
However, Tionghoa identity is defined as being both Chinese and Indonesian (Handoko,
2009; Oetomo, 1987, 1988), therefore, code-switching between Indonesian and Teochew
may be representative of the norms of the larger Tionghoa group identity. The code-
switching occurred in every instance of participants’ Teochew language use recorded
on Facebook ®. As previously stated, participants claimed that they always used Teochew
when communicating with Tionghoa interlocutors. Myers-Scotton (1993, p. 122) explained
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

that this kind of regular code-switching is used by speakers who are not content with the
identity associated with one language or the other, on its own. Instead, these speakers
code-switch when they want to express two or more positively evaluated identities simul-
taneously (Myers-Scotton, 1993, p. 122). Therefore, code-switching may be the norm of the
group of Tionghoa on Facebook ®.
In the next example, Agus converses with other Tionghoa on Facebook ® in a mixture of
English, Teochew and Indonesian. This interaction, like the previous one, exemplifies the
difficulty in trying to ascertain whether code-switching practices follow the norms of
the individual or the group identity. However, Agus’ use of English as well as Teochew
and Indonesian in this example, suggest that he may be simultaneously contributing to
both the norms of the Tionghoa group identity as well underlining his personal identity
as ‘cool’ and ‘modern’.
Extract 6: Agus’s status update and conversation with friends on Facebook ®
Agus
1 Don’t beg anyone to stay in your life.
2 If they want to be a part
3 of your life they will stay
User A
4 eh … bujor e? [Eh … true yeah?]
Agus
5 KMM llai lai lai le :p [Come come come on KMM, you :P
6 @ayu : messageeee kan no hapeee gpllllll @ayu: message me, got no cellphone]
User A
8 le pake BB meh to? [You’re using BB too?
9 w minta pin le ya..hihih I’ll ask for your pin yeah..hihih]
User B
11 woi mei2 minta pin kau [I haven’t asked for your pin]
Agus
13 cie bb am die..add am kmm. [I’m on BB now … add me to KMM]

Agus’ initial status update is a quote written in English, however User A comments on
his status in mixed Javanese (underlined) and Teochew. User A’s response then leads Agus
to code-switch between Teochew, English and Indonesian (lines 5 & 6). Users A and B then
converge towards Agus’ language use, in lines 9 and 11 respectively, by code-switching
between Indonesian and Teochew (cf. Bell, 2002).
Agus, like all the participants of the study, stated in interviews that he knew all of his
Facebook ® friends and claimed that the vast majority spoke limited English at most. There-
fore, Agus did not design discourse for his audience in the above instance. The use of
English in this case can be considered strategic on Indonesian Facebook ®. In interviews,
Agus presented a positive attitude towards the use of English on Facebook ®, ‘it seems
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 179

nicer to read and to look at if it’s in English’. This quote suggests that perhaps English on
Facebook ® is not considered inappropriate but symbolises his personal identity as ‘cool’
rather than ‘rigid’. Other participants expressed similar views on the use of English on Face-
book ® as a means of establishing an identity that was ‘modern’ and ‘international’ (Hadi), or
simply to maintain linguistic skills (Dewi). These perspectives suggest that English
language use on Facebook is not necessarily a response to audiences, but rather a
desire to fulfil personal goals and enact a personal identity. Establishing personal identity
rather than group identity confirms the predictions made using SIDE theory that partici-
pants would focus on augmenting their personal identities on Facebook ®.
Agus’ code-switching between Teochew and Indonesian does not definitively support
the original hypothesis of the study. His use of Teochew and Indonesian exhibits his per-
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

sonal differences from other members of each of the associated groups, thus establishing
his personal identity. On the other hand, as noted above, Agus’ and Dewi’s code-mixing
may also contribute to a social identity that is separate from the purely Chinese and
purely Indonesian groups (cf. Myers-Scotton, 1993). Agus’ and his Facebook ® friends’
code-switching between Indonesian and Teochew appears to be conventional as all the
users converge towards the code-switching practice (cf. Bell, 1984).
Therefore, on Facebook ®, as predicted, participants appear to emphasise their personal
identities more than group identities by concurrently using multiple languages from their
repertoire to create unique identities. However, the regular code-switching practices
amongst Tionghoa Facebook ® users may suggest that participants contributed to and
reinforced a group identity that is both Chinese and Indonesian (i.e. Tionghoa). This
element of the study was not predicted in the hypothesis.

Conclusion
The combination of SIDE theory and audience design in the hypothesis created predictions
that were proven accurate to a certain extent. It was predicted that due to the high levels of
anonymity and group salience on Kaskus ®, participants would associate strongly with group
identities and conform to group norms. Participants were hypothesised to establish their in-
group membership to both Indonesian and Tionghoa groups by following the rules and
regulations of the forums and designing discourse for their audiences. These predictions
held true on Kaskus ®. However, in the Teochew language thread, participants designed dis-
course for their addressees but did not accommodate their auditors or their overhearers.
This occurrence could be explained through audience design theory and SIDE theory.
Bell (1984, p. 60) explicated that addressees have a greater impact on language choice
and style than auditors and overhearers because they have closer social proximity to the
speaker. Additionally, applications of SIDE theory (Lea et al., 2001; Postmes et al., 2000;
Reicher et al., 1995) have shown that the relative salience of group norms affects the
degree to which individuals adhere to them. The Teochew thread had more salient
group norms than the wider Kaskus ® forum because language choice was prescribed
and enforced by the TS. Therefore although this finding was not initially predicted, the
combination of audience design and SIDE theory can nonetheless account for its
occurrence.
It was originally predicted that on Facebook ®, participants would adhere more strongly
to their personal norms than they did on Kaskus ® due to the low levels of anonymity and
 180 J. R. BIRNIE-SMITH

high salience of personal identity. Participants were also predicted to design discourse for
their audience of Indonesian speakers (addressees, auditors and overhearers) in status
updates. The predictions appeared accurate because participants conformed to the
group norms by designing discourse for their majority Indonesian-speaking audience.
Although, Dewi and Agus did not design discourse for their audience and instead
highlighted their personal identities. Dewi and Agus both posted status updates in
English on Facebook ®, despite the majority of their Facebook ® friends having limited
English proficiency. Their use of English was therefore considered an expression of their
personal identity. Dewi used English to distance herself from Indonesian and
Tionghoa groups. Agus used English to construct a ‘cool’ personal identity. This finding
challenges the assumption that participants would design discourse for their audiences,
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

however it conforms to the prediction that on Facebook ®, participants would be more con-
cerned with constructing their own personal identities than following the norms of their
groups.
On the contrary, participants’ use of code-switching on Facebook ® did not reflect the
hypothesis. In interactions with fellow Tionghoa users, participants regularly code-
switched between Teochew and Indonesian. The possible reasons for this code-switch-
ing are various. Participants may code-switch between Teochew and Indonesian to
respond to their addressees (Bell, 2002). Code-switching may also be a means
through which they express their personal identities as both Chinese and Indonesian.
Alternatively, code-switching may be considered a norm of Tionghoa users that
reinforces a Tionghoa group identity rather than personal identity. If the latter interpret-
ation is correct, then the predictions about participants conforming to their own per-
sonal norms rather than perceived group norms were contradicted on Facebook ®.
Further research is needed to determine whether or not the code-switching practice
is a result of group norms or personal identities of participants. The code-switching
practice could be evaluated by gathering more data on the code-switching practices
of a larger sample size on Facebook ®. The larger data set could elucidate the frequency
of code-switching amongst Tionghoa users, as well as the particular conditions which
elicit code-switching.
The combination of SIDE theory and audience design theory appears to be a useful one
in the examination of the impact of different online contexts on language choice and iden-
tity construction. However, it is difficult to definitively assess the applicability of the SIDE
theory framework and audience design theory to the peculiarity of code-switching prac-
tices among Tionghoa users on Facebook ®. The current case study thus serves as a stepping
stone towards wider generalisable findings on SIDE theory and audience design in exam-
inations of a variety of diaspora and migrant groups’ language and identity on the Internet.

Acknowledgements
I am greatly appreciative of the journal’s reviewers and to Dr Simon Musgrave and Dr Howard Manns
for their invaluable advice and comments on earlier drafts of this paper.

Disclosure statement
No potential conflict of interest was reported by the authors.
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 181

Note
1. Pseudonyms are used to protect all participants’ privacy.

References
Alexa The Web Information Company. (2013). Facebook® traffic stats. Retrieved from http://www.
alexa.com/topsites
Androutsopoulos, J. (2006). Multilingualism, diaspora, and the Internet: Codes and identities on
German-based diaspora websites. Journal of Sociolinguistics, 10(4), 520–547. doi:10.1111/j.1467-
9841.2006.00291.x
Androutsopoulos, J. (2008). Potentials and limitations of discourse-centred online ethnography.
Language@Internet, 5, Article 8.
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Bell, A. (1984). Language style as audience design. Language in Society, 13(2), 145–204.
Bell, A. (1991). The language of news media. Cambridge: Blackwell.
Bell, A. (2002). Back in style: Reworking audience design. In P. R. Eckert & John R. Rickford (Eds.), Style
and sociolinguistic variation (pp. 139–169). Cambridge: Cambridge University Press.
Childs, B., & Mallinson, C. (2006). The significance of lexical items in the construction of ethnolinguis-
tic identity: A case study of adolescent spoken and online language. American Speech, 81(1), 3–30.
Christopherson, K. (2007). The positive and negative implications of anonymity in Internet social
interactions: “On the Internet, nobody knows you’re a dog”. Computers in Human Behavior, 23
(6), 3038–3056. doi:10.1016/j.chb.2006.09.001
Danet, B., & Herring, S. (2007). The multilingual internet: Language, culture, and communication online.
New York, NY: Oxford University Press.
Dorleijn, M., & Nortier, J. (2009). Code-switching and the Internet. In B. Bullock & A. Toribio (Eds.), The
Cambridge handbook of linguistic code-switching (pp. 127–141). Leiden: Cambridge University
Press.
Facebook@. (2012). Information for Parents and Educators. Retrieved from http://www.Facebook.com/
help/parents
Haines, R., & Mann, J. (2011). A new perspective on de-individuation via computer-mediated com-
munication. European Journal of Information, 20(2), 156–167. doi:10.1057/ejis.2010.70
Handoko, F. (2009). Education, language use and shifting identities among ethnic Chinese
Indonesians. Chinese Diaspora Studies, 3, 183–192.
Heidhues, M. (1996). Identity and the minority: Ethnic Chinese on the Indonesian periphery. Indonesia
Circle, 24(70), 181–192. doi:10.1080/03062849608729877
Herring, S. (2008). Questioning the generational divide: Technological exoticism and adult construc-
tion of online youth identity. In D. Buckingham (Ed.), Youth, Identity and digital media (pp. 71–94).
Cambridge, MA: MIT Press.
Hinrichs, L. (2006). Codeswitching on the web: English and Jamaican Creole in E-mail communication.
Amsterdam: John Benjamins Publishing.
Kendall, L. (1998). Meaning and identity in “cyberspace”: The performance of gender, class, and race
online. Symbolic Interaction, 21(2), 129–153.
Kim, J. (2009). “I want to be different from others in cyberspace” The role of visual similarity in virtual
group identity. Computers in Human Behavior, 25(1), 88–95. doi:10.1016/j.chb.2008.06.008
Klein, O., Spears, R., & Reicher, S. (2007). Social identity performance: Extending the Strategic Side of
SIDE. Personality and Social Psychology Review, 11(1), 28–45. doi:10.1177/1088868306294588
Kolko, B., Nakamura, L., & Rodman, G. (2000). Race in cyberspace. London: Routledge.
Kumar Sahoo, A., & de Kruijf, J. (2014). Indian transnationalism online: New perspectives on diaspora.
Farnham: Ashgate.
Lea, M., & Spears, R. (1991). Computer-mediated communication, de-individuation and group
decision-making. International Journal of Man-Machine Studies, 34(2), 283–301.
Lea, M., Spears, R., & De Groot, D. (2001). Knowing me, knowing you: Anonymity effects on social
identity processes within groups. Personality & Social Psychology Bulletin, 27(5), 526–537.
Miller, D., & Slater, D. (2000). The Internet: An ethnographic approach. Oxford: Berg.
 182 J. R. BIRNIE-SMITH

Mitra, A. (1997). Virtual commonality: Looking for India on the Internet. In S. Jones (Ed.), Virtual culture:
Identity and communication in cybersociety (pp. 55–79). London: Sage.
Myers-Scotton, C. (1993). Common and uncommon ground: Social and structural factors in codes-
witching. Language in Society, 22(4), 475–503.
Nakamura, L. (2002). Cybertypes: Race, ethnicity, and identity on the Internet. New York: Routledge.
Oetomo, D. (1987). The Chinese of pasuruan: Their language and identity. Canberra: Australian National
University.
Oetomo, D. (1988). Multilingualism and Chinese identities in Indonesia. In J. Cushman & G. Wang
(Eds.), Changing identities of the Southeast Asian Chinese since World War II: The ethical challenge
of biotechnology (pp. 95–106). Hong Kong: Hong Kong University Press.
Postmes, T., Spears, R., & Lea, M. (2000). The formation of group norms in computer-mediated com-
munication. Human Communication Research, 26(3), 341–371.
Postmes, T., Spears, R., Sakhel, K., & De Groot, D. (2001). Social influence in computer-mediated com-
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

munication: The effects of anonymity on group behavior. Personality & Social Psychology Bulletin,
27(10), 1243–1254.
Purdey, J. (2006). Anti-Chinese violence and transitions in Indonesia. June 1998–October 1999.
Honolulu: University of Hawai’i Press.
Reicher, S. D., Spears, R., & Postmes, T. (1995). A social identity model of deindividuation phenomena.
European Review of Social Psychology, 6(1), 161–198. doi:10.1080/14792779443000049
Turkle, S. (1995). Life on the screen: Identity in the age of the Internet. New York, NY: Simon & Schuster.
Wright, S. (2004). Multilingualism on the internet – Thematic introduction. International Journal of
Multicultural Societies, 6(1), 5–13.

Appendix: Semi-structured interview questions

The following outlines the interview schedules for the three interviews conducted with each of the
four participants. The questions were used as a basic guide for the conversations, however, partici-
pants were able to direct the conversation towards whichever topic they wished to discuss. The first
interview was aimed at gaining basic information about participants; the second was focused on dis-
covering their language attitudes and perceptions of their own linguistic behaviour online. Partici-
pants were asked follow up questions in Interview 2 about all the languages they mentioned in
Interview 1. The final interview delved into participants’ perceptions of their own identities and
orientation towards their Tionghoa ethnicity and Indonesian nationality. The semi-structured inter-
views were also supplemented by ongoing conversations (Oetomo, 1987) with participants that
were unstructured and permitted participants to highlight what they thought was important for
me to know and understand.

Interview 1:
How old are you?
How long have you lived in Pontianak?
Who do you live with?
What languages do you speak?
What languages do you use online?
Are you working or studying? Where at? Doing what?
What is your ethnicity?
What is Tionghoa?
How can you tell if someone is Tionghoa?
Where do you often encounter other Tionghoa in your daily life?
What’s the difference between Tionghoa Indonesian and other Indonesians?
Interview 2:
When do you use Bahasa Indonesia? Why?
 INTERNATIONAL JOURNAL OF MULTILINGUALISM 183

When do you use Teochew? Why?

When do you use English? Why?
How often do you use Kaskus®? What do you use it for?
What language do you usually use on Kaskus®? Why?
Who do you usually interact with on Kaskus®?
How often do you use Facebook®? What do you use it for?
What language do you usually use on Facebook®? Why?
Who do you usually interact with on Facebook®?
How many (approximately) of your friends on Facebook are Tionghoa? How many are non-
Chinese Indonesians?
How many (approximately) of your friends offline are Tionghoa? How many are non-Chinese
Indonesians?
Downloaded by [The University Of Melbourne Libraries] at 04:51 26 July 2016

Interview 3:
In your opinion, is your ability to speak Teochew important or not? Why?
Is your ability to speak Bahasa Indonesia important to you or not? Why?
How often do you use Bahasa Indonesia in real life?
How often do you use Teochew in real life?
How often do you use English in real life?
How often do you use Bahasa Indonesia online?
How often do you use Teochew online?
How often do you use English online?
Do you often practice Chinese culture in your everyday life?
Is your identity as Tionghoa important to you? Why/why not?
Is your identity as Indonesian important to you? Why/why not?
Do you prefer to identify yourself as Tionghoa or Indonesian? Or something else?