What can interactional sociolinguistics bring to the family language policy research table? The case of a Malay family in Singapore

ABSTRACT

This research applies interactional sociolinguistics within a family language policy framework to investigate how social structures and institutional discourses outside the home trickle into daily mundane activities within a Malay-English bilingual family in Singapore. Drawing upon ethnographic interviews and naturally-occurring interactions at home, the study examines the parents’ reported linguistic ideologies and the ‘practiced’ family language policy. The analysis suggests that the family embraces language mixing in the family and reports making efforts to strike a balance between Malay and English at home as a way of raising the children bilingually. The analysis of familial interactions, however, shows that the parents’ and children’s language practices vary as certain frames are evoked in the conversation. The findings suggest that a better understanding of language maintenance and shift processes through family language policy research necessitates a nuanced examination of how languages are used to evoke, negotiate and establish certain frames during daily activities at home.

KEYWORDS:

• Family language policy
• interactional sociolinguistics
• Singapore
• Malay-English bilingualism

Family language policy

Since its inception in the early 2000s, family language policy (King, Fogle, and Logan-Terry 2008; Luykx 2003) has constantly moved away from a Fishmanian understanding of home as a private domain and ‘an unexpendable bulwark’ of language maintenance in which a language could be safeguarded (Fishman 1991, 94) to a more dynamic and complex definition of family and its language policy. Family language policy (FLP) has been thus conceptualised as ‘a confluence of discourses’ (Mirvahedi and Macalister 2017), and ‘cross-scalar complexity’ which is ‘a stratified and polycentric language-ideological construction enveloping multiple resources and scripts for their deployment across scales’ (Blommaert 2019, 3), stressing the ‘porosity’ of family and its language policy to external forces (Canagarajah 2008, 171; Mensel 2018, 238). Examining multilingual families’ language ideologies, language practices, and language management (Spolsky 2004, 2009) within this framework makes it possible to understand how the private and public sphere of social life and the accompanying requirements of each – i.e. orientations towards intimacy and family-bound affection, and orientations towards trajectories of success and mobility, respectively – intersect and compete at home (Curdt-Christiansen 2013a; Haque 2019; Macalister and Mirvahedi 2017).

FLP scholars have taken different approaches to explore how the private and public sphere of social life compete in the context of home, shedding light on how family language policy interacts with social structure outside the home. Research applying language socialisation theories to family interactions, for example, have shown children’s active role in influencing family language policies, e.g. through their resistance to parental practices or bringing the dominant language home after they start school, what has come to be known as child agency in FLP scholarship (Fogle 2012; Goodwin and Kyratzis 2012; Gregory 2001; Kheirkhah and Cekaite 2018; Obied 2009; Takei and Burdelski 2018). Anthropological studies have shown that such language socialisation in the family may not be necessarily a conscious process decided by family members (e.g. see Garrett 2012; Kulick 1992), ultimately leading to language shift or language loss out of sight.

Researchers have also drawn on other theoretical frameworks and methodological tools to understand the interplay between bi/multilingual families’ language ideologies and practices at home and social structures in which they find themselves, and how such a dialectic relation influences the vitality of languages. Lane (2010), for example, applied nexus analysis (Scollon and Scollon 2004) to investigate language shift in Bugøynes, a Kven community in Northern Norway. Having illustrated how large-scale discourses such as language policies influence language choice, she argued that while shift to another language might be seen, in part, as a choice parents make, but in reality they did not always have a choice (see also Sicoli 2011, for agency in language maintenance and shift). Taking a different approach, Mirvahedi (2020) recently applied realist social theory to understand language maintenance and shift among Azerbaijanis in Iran. He demonstrated how realist social theory, giving equal weights to agency and structure, helps avoid falling into structural determinism and individual rationalism in the analysis of family language policy and its interplay with social structure.

Contributing to this line of research, in this paper, I seek to illustrate how interactional sociolinguistics could be used as yet another approach to explore how social structures trickle into everyday conversations at home, demonstrating the complexity of language maintenance and shift processes. Drawing upon interviews and sample recordings of family interactions in a Malay-English bilingual family in Singapore, I would argue that interactional sociolinguistics with its analytical toolkit can not only show what language socialisation patterns in the family are like, but also allows us to provide a deeper analysis of how such socialisation patterns are formed within particular social structures in which the family lives.

Interactional sociolinguistics

Built on the work of Erving Goffman (1964, 1979, 1981, 1983, 1986) in Sociology and John Gumperz (Gumperz 1982, 1999) in anthropology and linguistics, interactional sociolinguistics (IS) has evolved into an approach to discourse analysis that ‘attempts to “bridge the gap” between “top-down” theoretical approaches which privilege “macro-societal conditions” in accounting for communicative practices, and those, such as CA which provide a “bottom-up” social constructivist account’ (Stubbe et al. 2003, 358). IS achieves this goal by shifting away from accounting for grammar-like rules of language use to questions such as ‘(1) how and by what signalling devices language functions to evoke the contextual presuppositions that affect interpretation, and (2) what presuppositions are at work in particular talk exchanges’ (Gumperz 2015, 313). Taking language as situated in particular circumstances of social situations both reflecting and adding meaning to those social contexts, these questions are addressed by focusing on what Gumperz calls contextualisation cues. That is, those ‘constellations of surface features of message form … by which speakers signal and listeners interpret what the activity is, how semantic content is to be understood and how each sentence relates to what precedes or follows’ (Gumperz 1982, 131). These contextualisation cues that ‘signal the speech activity in which participants perceive themselves to be engaged’ are viewed as ‘a type of frame’ (Tannen 1993, 4), the analysis of which can help us empirically investigate how frames are invoked, re-negotiated, and established (Canagarajah 2020).

The notion of frame (Bateson 1972; Goffman 1986), as Tannen (1993, 6) puts it, captures ‘what people think they are doing when they talk to each other.’ In other words, as people engage in conversations with others, their use of linguistic, paralinguistic, and non-verbal cues evoke certain frames that render ‘what would otherwise be a meaningless aspect of the scene into something that is meaningful’ (Goffman 1986, 21), and guides the doings of the participants. Goffman (1981) notes that as interlocutors create frames, they also construct footings, i.e. ‘alignments we take up to ourselves and other participants within interaction’ (Goffman 1981, 128). What is of paramount importance to note here is that conversations do not often happen within only one frame; rather, individuals laminate both frames and footings within a single stretch of interaction (Goffman 1981). This means that interlocutors often use a variety of linguistic features to switch quickly between various footings and frames, embed frames and footings within one another, or frames can be in conflict and accidentally ‘leak’ into one another (see Gordon 2009), thereby altering their ‘interpretative edge’ (Karrebæk 2011, 2912). This takes place when, for example, a child while doing homework signals completing his/her homework, drifting off to begin playing with a sibling. What happens in such a situation is that the participant transforms the already meaningful material in accordance with a schema of interpretations, what Goffman calls ‘keying’ (Goffman 1986, 45).

This conceptualisation of framing and family interactions as layered in a multilingual context such as Singapore is important as it allows us to investigate the ‘dovetailing of public and private discourse,’ i.e. intertextuality (Tannen 2006, 598) in family language policy scholarship. While this is essential to an analytic understanding of how shifts in the frame take place (Buchbinder 2008), it enables us to move away from Fishmanian questions of who speaks what languages to whom to examining what frames, through what semiotic repertoires, and by whom are evoked, (re)negotiated, and established in daily mundane family interactions. Addressing these questions will not only capture the dynamic, fluid, and moment-by-moment nature of FLP, but it also allows us to go beyond the here-and-now of family interactions, and find explanations in societal and institutional structures outside the home. This ultimately helps us better understand the language maintenance and shift processes.

Contemporary Singapore: language management and social structures

Singapore’s independence in 1965 introduced unprecedented language management to the city–state. Faced with no fewer than 33 languages in a new nation state/city of only 721 square kilometres, language appeared to be a sensitive and key issue to the new government and their ambition to build a cohesive modern nation (Cavallaro and Ng 2014; Kuo 1985). In order to deal with the ‘language problems’ of the new postcolonial nation (Neustupný 1974), the government found the solution in ‘oligolingualism,’ i.e. reducing the sociolinguistic complexity by promoting certain languages to the status of national and official through conscious policy and planning (Blommaert 2019, 1). Malay and Tamil were then selected as the mother tongues for the entire Malay and Indian communities, whose heritage languages could be, amongst others, Bugis, Boyanese, Sinhala, Punjabi, or Urdu (Cavallaro and Ng 2014). Mandarin, with almost no native speakers (0.1%), was also designated as the only ‘mother tongue’ for the Chinese community whose heritage languages were different Chinese varieties such as Hakka, Hainanese, MinNan (Hokkien), Teochew, Yue (Cantonese) and others (Kuo 1985). While the three mother tongues, Malay, Tamil, and Mandarin, were assigned based only on the generic ethnicity and race of the parents rather than the first language acquired from infancy, the other non-official languages were disparaged as ‘stagnant dialects’, and were viewed as varieties with low cultural and no economic value, which if spoken to the child, will ‘ruin his future’ (Lee Kuan Yew 1980). The Speak Mandarin Campaign was also launched in 1979 – which is celebrated annually – seeking to encourage all Singaporean Chinese to use Mandarin at home, stressing the economic and instrumental value of Mandarin in trade with China as a powerful economy (Curdt-Christiansen 2014). The corollary of these policies has been a drastic increase in using Mandarin as the dominant home language, from 13.1% in 1980 – 34.9 in 2015, and a concomitant decrease in the use of the Chinese vernaculars from 76.2% in 1980 –12.2% in 2015 (Singsat 2015).

With an emphasis on four official languages – English, Mandarin, Malay and Tamil, a bilingual educational policy was then instituted in 1966, whereby English – perceived as an ethnically neutral language – was used as the medium of instruction, and the three ethnic languages were taught as subjects at schools (Low and Pakir 2018). Though it was possible to attend a number of Chinese-Malay-Tamil-medium schools at the time of independence, the families, perhaps under the influence of state-level policies, stopped sending their children to these schools, leading to their shut-down due to a low student enrolment (Lee 2012). This bilingual policy was intended to contribute to the country’s economic development and achieve a competitive edge in international markets by training what later came to be known as an English-knowing bilingual labour force (Chua 2010; Pakir 2003, 2004); the type of ‘wordforce’ that Heller (2010) considers to be necessary for nations to perform successfully in this globalised new economy. At the same time, the policy was to support the maintenance of the three main ethnic languages as ‘cultural ballast’ in Singapore (Lim 2015, 262; Wee 2003).

Over the course of 55 years, however, the state’s language policies and campaigns such as the Speak Good English Movement have increasingly institutionalised English as the mainstream language in the society. As a result, Singaporeans across all the ethnic groups – Chinese (76%), Malays (13%) and Indians (8%), and a small minority of others (3%) – have increasingly come to choose English as a home language and de facto ‘mother tongue’ for young children (Tan 2014). According to the recent Census of Singapore, 36.9 per cent of the population claimed English as their dominant home language, a dramatic increase from 1.8% in 1957 (Department of Statistics 2011, 2015; Lim 2015). The families’ preference to speak English at home has resulted in a decrease in the children’s desire to learn their mother tongues (Chua 2010), and consequently in their decreasing proficiency in the mother tongues (Lee 2012). Because of the younger generation’s low proficiency in the mother tongues, the Ministry of Education introduced a ‘B’ syllabus at the secondary level in 2004 and a Modular Approach at the primary level in 2007. The aim was to ‘solve’ the problem of increasing ‘language learning difficulties’ by reducing the content of the curriculum (Curdt-Christiansen 2014).

Although Singaporean Malays are often depicted as exceptionally successful guardians of their language (Chong and Seilhamer 2014), Cavallaro and Serwe (2010) suggests a different scenario. The survey administered to 233 participants from 12 to 72 years of age revealed that English is increasingly used by the younger generation (18–24 years of age) as well as those with a higher socioeconomic and educational background in domains of language use once considered a safe space for Malay. Viewing language shift as the children’s gradual loss of interest, preference and proficiency in their mother tongue (Harrison 2007), this paper seeks to contribute to this body of research by documenting and explaining how language shift in a Singaporean Malay-English bilingual family at home is ‘talked into being’ (Gafaranga 2010, 241). Drawing upon interactional sociolinguistics in the analysis, I seek to argue that the language shift that is realised in everyday family interactions can be traced to home-external social structures and discourses.

Methodology

This study is part of a larger project on Malay and Tamil speaking families’ linguistic ideologies and practices in Singapore. In order to investigate the micro-level dynamics of language use at home and its impact on language maintenance/shift, a qualitative inquiry was chosen to explore things as ‘they are undertaken, experienced and narrated by people’ (Heller, Pietikäinen, and Pujolar 2018, 9). Five families (three Malay, and two Tamil families) from different walks of life were recruited through ‘friends of friends’ approach (Milroy 1980, 47), and according to convenience and criterion sampling (Cohen, Manion, and Morrison 2011, 229–230). That is, those parents who expressed interest and availability could be recruited; however, only those who met the main criterion of the research, i.e. having at least one child between 2 and 12, were admitted as the participants. The data comprising ethnographic observations, six hours of interviews with parents, and approximately 150 hours of audio-recordings of naturally-occurring family interactions were collected in five families over nine months (September 2017–May 2018). The parents were asked to self-administer the recording at their own comfort. They were trained by the researcher how to use the recorder and asked to ideally place it out of the children’s sight to reduce the Observer’s Paradox impact on the authenticity of the collected data. The transcriptions in fact proved that natural family interactions in different settings, e.g. putting the children into bed, doing homework, children’s play, to name a few, had been recorded. Nonetheless, in some recordings, parents had inadvertently placed the recorder next to a fan, rendering those recordings useless.

The interviews are used in this paper to both provide demographic information about the family and explore the parents’ language ideologies. As language managers in the family, parents’ language ideologies and their agency have proved significant in language maintenance (Mirvahedi 2020); however, parental ideologies and agency do not always bring about the expected, anticipated or claimed outcome (Mirvahedi and Jafari 2018; Yu 2010). Ideologies and practices reported by the parents are then explored along with samples of real-life language socialisation at home to illuminate what the families did with languages they knew (Spolsky 2004, 2009). Out of the collected corpus of familial interactions, in this paper only those key moments were selected for analysis (Wei 2011) that best describe the dynamics and complexity of language use at home. In other words, the conversations that were only in Malay or English were not included here due to the space limitation. Although each of those sheds light on interesting aspects of language use in the family, I avoided the ones reflecting total language maintenance or language shift, for example, when the mother lectures her child on her pronunciation of English, when she and her daughter practice reading the Holy Quran, or when grandparents attempt to engage the children in Malay conversations.

Findings and discussion

Family profile

The family presented here lives in an HDB flat, and they perceive themselves a middle-class Malay family. The mother, 34 years old, works at a school library and the father, 42 years old, is a technical officer. Both parents consider themselves bilingual; yet, the father evaluated his English a little weaker than that of his wife. Their two daughters are 4 and 7 years old. At the time of the data collection, the older daughter was attending Primary 1 at a Madrasah school, and her younger sister was attending a three-hour programme at a Madrasah kindergarten. It is worth mentioning that those Malay families who would like to instill more Islamic values in the children prefer to send their children to Madrasah schools where emphasis is also put on the Arabic language by offering courses such as Arabic Language, Islamic Theology, Islamic Jurisprudence and so on. Yet, Madrasah schools equally use English as the medium of instruction for all subjects except for the mother tongue so that they do not lag behind national schools in preparing students for the knowledge-based economy of Singapore (see Buang 2010; Tan and Kasmuri 2007). The family also had employed a domestic helper from Indonesia who spoke Bahasa Indonesia, the variety different from Singaporean Malay (Bahasa Melayu) spoken by Singaporean Malays. According to the parents, the helper was not fluent in English at all upon her arrival in Singapore, but her English has improved considerably since then because of the children’s interactions with her in English at home.

The parents pointed to the ‘lingua franca’ status of English in Singapore and the world making it the most important and necessary language for them; however, they simultaneously expressed mixed feelings about the equal significance of other languages:

Father: ‘All languages is sort of equally important. You have to know all the basics of all languages, somehow we have to learn like those Chinese words, from friends everything, so language wise I think it’s not that varies to us because most of the time we communicate in English. So English is a must, because if we would talk in our own language, it would be a problem for them to understand us, so we tend to end up in talking in English. Actually, I don’t think it’s only in Singapore, it is in the world.’

Believing that they need to know the basics of languages, including the Malay language, the parents have adopted a one-parent-one-language policy (OPOL) at home. As both parents mentioned in different instances during the interviews, they ‘try to balance languages at home, if one talk in English, the other will have to speak in Malay.’ Rather than the standard OPOL policy where each parent sticks to only one language, the parents here use both languages fluidly. However, as it was expressed in the interviews and illustrated in the interaction where the children were involved, their presumably balanced input in the two languages were far from ‘balanced’ (see also Mirvahedi and Cavallaro 2019). The outcome of such a fluid OPOL was the children’s predominant interest and proficiency in English.

Mother: ‘Our girls are not into Malay that much, they are more to English.’

Notwithstanding their children’s predominant use of English at home, the parents in this family believed in a strong link between the Malay language and Malay ethnic identity that will arguably contribute to its maintenance:

Father: ‘I think Malay won’t fade away, it’s in the blood, it won’t go away that easily, somehow we have to master, know at least the basics, at least they must understand the language, we cannot just chuck aside our second language.’

While the father’s remark shows a strong attachment to their ethnic language and identity, his sense of satisfaction with the children knowing only the basics of the language suggests the parents’ ambivalent, if not conflicting, ideologies. While they claim making efforts to strike a balance between the two languages at home because Malay is equally important to them, they seem to be happy with speaking Malay in ‘bits and piece.’ In other words, Malay maintenance for this family seems to be defined merely as their children’s ability to comprehend the language and respond sporadically in bits and pieces rather than becoming fluent speakers of Malay.

In short, a glance at the family’s language ideologies suggests that they place English at the top position within what Liddicoat (2013) calls ‘hierarchies of prestige.’ Nonetheless, the parents express their stronger attachment to the Malay language and identity, and their attempt in raising their children bilingually by striking a balance between the two languages at home. Despite the parents’ pro-Malay ideologies, what is clear from the corpus of 30 hours of familial interactions collected in the household is that parents’ ad hoc language mixing practices have not resulted in their children’s preference and/or adequate language skills to predominantly use Malay at home. The samples presented from the family below illuminate (1) what the dynamics of language use in the home is like, and (2) how such language socialisation patterns are informed by social structures and institutional discourses outside the home.

Interactions at home

Excerpt one: Malay teacher and a mother

This extract from the family’s interactions at home offers interesting insights into what the ‘real’ language policy is at home despite the parents’ reported ‘balanced’ use of Malay and English. Here, Amirah (the older daughter) is doing her Malay homework with her mother (M).

The seven-year-old daughter initiates the conversation by asking a question in which she uses the English word number and the Malay word satu. Notwithstanding the fact that the conversation is taking place at home, a frame of Malay class in school and the roles of a teacher and student are evoked as the mother, rather in a scolding manner, corrects her and asks her child to learn. While this suggests the mother’s insistence in her daughter’s learning Malay, it also implies that the child’s default word for ‘number’ is its English rather than Malay equivalent, making the mother emphasise it in a scolding voice. The child aligns with the mother’s shift to Malay-only language use, and in lines 3–13 she and her mother continue checking the answers. In lines 14–15, the daughter once again switches to using both Malay and English in her utterance. Although in lines 15 and 16, the recording was not clear, what is obvious is the use of Malay language. Note in line 18, however, that English becomes dominant as the frame shifts from doing homework in a Malay class to talking about tomorrow; the mother, rather than a Malay teacher above, switches to English instructing the child not to forget, either the lesson or the notebook, and put it in her younger sister’s (adik’s)¹ bag. The conversation between the daughter and mother continues in English up to line 24. The contrast between using English in these lines and the Malay language in the previous lines suggest that the Malay language might have been used merely because of doing Malay homework rather than being the usual medium of interaction at home. In Agha’s words, it may be further argued that by using Malay, the mother performs a figure through speech (Agha 2005, 39), which projects an image of a Malay teacher; however, by choosing English, this figure changes from a Malay teacher to the mother at home who is worried about her child’s school the next day.

What is interesting here is that the mother does not align with her daughter’s use of Malay in line 25, and she, rather than answering in Malay, uses English to respond, asking her daughter to keep the activity book in the library. Furthermore, the child’s choice of Malay and the mother’s use of English, or vice versa, in interaction with each other, which is also repeated in the following excerpts, suggests that the two parties understand each other’s languages but they each use a different language in communication, a case which illustrates what Saville-Troike (1987) calls ‘dual-lingual communication.’ The corollary of such language socialisation practices is said to be that the two languages will not develop ‘harmoniously’ in children (De Houwer 2015), increasingly leading to the weakening of the heritage language.

In lines 28–29, the mother uses three Malay words in her English utterance to emphasise that the daughter is supposed to belajar (learn), and she does not need to fold the book, as it delam koyak (it tears). In line 30, having finished the homework, the child asks her mother for permission to play, which serves as a keying device to change the frame of the activity and talk from doing homework and getting prepared for school the next day to play (Goffman 1981, 1986). What ensues is that with change of the frame from doing Malay homework to playing, the participants change too. The conversation in the play takes place between the two sisters, in which no Malay is used, suggesting how the children at home can actively exert their agency to influence the language ecology of the home in favour of a particular language (Fogle 2012; Gafaranga 2010; Goodwin and Kyratzis 2012; Gregory 2001; Kheirkhah 2016; Kheirkhah and Cekaite 2018; Obied 2009; Paugh 2005; Rindstedt and Aronsson 2002; Takei and Burdelski 2018; Williams and Gregory 2001). The following excerpt casts more light on how children’s play can be a complex stretch of talk in which the world outside the home is reconstructed.

Excerpt two: Playing games

This extract of the interaction in the family shows, in particular, the role the older daughter plays in ‘talking language shift into being’ (Gafaranga 2010). The children – Alifah (the younger daughter) and Amirah (the older daughter) – with the help of their parents – father (F) and mother (M) – are making some shapes/toys out of cardboard.

While Alifah, the younger child, keeps singing ‘boom boom cak’, Amirah, the older daughter, speaks English to herself asking a question about her cake, and then signalling she is going to make a donut instead, because she is sure somebody will buy her donut (lines 5–11). The mother’s participation in the play in lines 11–12 takes place through a question, a guess at the answer, and an instruction for the child, all of which are carried out in English. Amirah replying back to her mother in English expresses her intention of decorating the things she is making. In line 17 she uses the address term Abah (dad) in talking to her father, and asks him if he is done with a piece of equipment he is using. Then she asks if her father likes the cat, and then continues talking to herself in English. While Alifah is still singing to herself in the background, Amirah asks her father (line 20) a question using the English question word where and its Malay equivalent. Alifah inserts a Malay sentence here (line 21) with only one English word (cake). Aligning with the medium of interaction, i.e. Malay, the father asks his child which one of the things made by the father she wants. Using the English ‘I’ and the Malay words nak ini (want this), Alifah answers her father. This is the only sequence where Malay is actively used in interaction, which reflects how the parents’ fluid OPOL as well as their children’s speaking Malay in bits and pieces are realised in practice. This could be also arguably due to the fact that this part of the imaginative play takes place within the frame of baking a cake with parents at home, the requirement of which is using Malay, even though in bits and pieces.

What happens in the next line (line 26) is of great importance. While the previous interaction between the father and the younger daughter was in Malay with minimum use of English, Alifah, addressing her older sister, uses English to ask for her help. Then the remaining interactions between the two sisters (lines 26–50) continue in English. While the children’s choice of English to interact amongst themselves demonstrates their role to influence the language ecology of the home in favour of English, it can be also traced to the frames in which their imaginative place is situated.

The extract above shows how children play a crucial role in ‘social and linguistic reproduction and change, through their active linguistic and discursive production of their social worlds’ (Howard 2017, 156). As the later lines illustrate, the two sisters engage in an act of buying and selling, something that happens outside the home. This is in line with previous studies (Mirvahedi and Cavallaro 2019; Smith-Christmas 2020) that show children’s imaginative play that places them in certain frames related to the outside world can be a site of language shift. These frames are often evoked and established by older siblings who have more experience of the outside world. For example, a study by Rindstedt and Aronsson (2002) on language shift from Quichua-Spanish bilingualism to Spanish monolingualism in San Antonio, Ecuador documented the older children’s consistent use of Spanish in addressing their younger siblings. Older siblings led language shift to Spanish through exposing the toddlers whom they took care of for the most part of the day. In a similar study, Bridges and Hoff (2014), found the older siblings’ influence on US-born toddlers who were being raised in bilingual homes. As the older siblings exposed the toddlers to English more than other members of the family, the toddlers with older siblings became more advanced in English in the following years.

Excerpt three: Putting the children to bed

This excerpt from the everyday routine of putting the children to bed takes place between the mother (M), the two sisters – Amirah and Alifah – and the father who does not actively contribute to the on-going conversation. In this extract, some Arabic is used, but never as a medium of interaction, rather as a liturgical language.

Through her question in line 1, the mother reminds the children of a religious routine that is supposed to be done before sleep. Mixing Malay and English, the older daughter says that she wants to read some parts of the holy book, Quran. These are known as three Quls (written as Kul in Malay), as the chapters begin with an Arabic word Qul meaning ‘say’. The mother aligns with her daughter’s use of Malay verb baca (read), and she asks the children in line 3 to read. In line 4, the older daughter asks her father to join them in reciting the three Quls. The older daughter here again uses Malay indicating her competence in the language. Yet, the mother’s immediate choice of English in line 5 triggers a shift in the language of the conversation. After reciting the three Quls, the mother says ‘good night’, ‘I love you’, ‘sweet dreams’ and ‘sleep now’ to the girls (lines 7–20), to which the girls respond back in English – very similar to practices of English native speakers. What is interesting here is the insertion of Arabic titles such as umi that is commonly used in some Malay families, and formulaic expressions such as inshaAllah, Assalamualaikum and Waalaikumsalaam. These words obviously suggest the influence of practicing Islam among Malays indexing the families’ religious identity. Overall, this extract also illustrates how the main medium of communication is English when it comes to the parent–child interaction with minimal use of Malay. Of paramount importance in this particular excerpt is also the language through which emotions are expressed. Given that emotions often remain to be expressed in the mother tongue in the context of language shift (Pavlenko 2004, 2005), the use of English here can be also taken as further evidence of ongoing language shift in this family. While the parents could have made the most of the moments when Malay was used to reinforce Malay use at home, they, in particular the mother, opted for English to interact with the children.

Concluding remarks

The study departed from the premise that family and its language practices are not immune from home-external forces and influences, thus moving away from a Fishmanian understanding of the role of family in language maintenance and shift processes. To unravel the complexity of the interplay between language ideologies and practices of families and social structures outside the home, home is hence conceptualised as a stratified and polycentric language-ideological construction that draws upon multiple resources and scripts for their deployment across scales (Blommaert 2019, 3). It was argued that examining family language policy within this framework entails analysing how social and institutional influences trickle into, and inform, every day moment-by-moment interactions in the home, and how this helps us understand the interplay between agency and structure in FLP scholarship. Contributing to this line of research, interactional sociolinguistics was then introduced as one viable approach that can not only illustrate what language socialisation in a family is like, but also document its complexity in relation to home-external forces and influences. Extending and expanding the scholarship on language maintenance and shift in Singapore, which has been mainly studied through interviews and surveys to date (exceptions are Curdt-Christiansen 2013b; Hu and Ren 2017; Ren and Hu 2013, which are on Chinese speaking families), the study then applied interactional sociolinguistics to three snapshots of everyday linguistic practices in a Malay family, providing a complex picture of what the ‘real’ (Spolsky 2009, 4) ‘practiced’ (Bonacina-Pugh 2012) family language policy is in the family.

While the interviews with the family showed their favourable views about bilingualism, ethnic language maintenance and the link associated between the Malay language and the Malay ethnic identity, the analysis of daily interactions suggested a far more complex scenario. For the parents, Malay formed a core value of their ethnic identity and culture that would never ‘fade away’. However, in order to maintain Malay, they reported using both languages at home in a fluid and dynamic type of one-parent, one-language style. That is, whenever one used Malay, the other needed to strike a balance by using English, and vice versa. In their effort to maintain Malay, they also sought help from home-external forces (Schwartz, Moin, and Leikin 2011) by sending their children to Madrasah schools where the Malay and Islamic values are stressed more than other public schools. As the three excerpts from the family show, however, Malay, English, and Arabic use are not distributed equally when it comes to evoke, negotiate and establish frames of talk by different family members in the home. While Arabic is used as a liturgical language when a religious frame is established, Malay and English are used dynamically in various frames. This means that while certain frames such as doing Malay homework contribute to language maintenance, others such as children’s imaginative play contribute to language shift. Given that more frames are evoked and established through the predominant use of English by children, the parents’ wishful thinking that their one-parent, one-language family language policy would result in the maintenance of Malay seems to be far from the reality. As De Houwer (2007) also found through her questionnaire data from 1,899 families, the often-cited ‘one-parent, one-language’ strategy to develop bilingualism in children provides neither a necessary nor sufficient input condition. Similarly, for this family, the parents’ proclaimed dynamic one-parent, one-language policy does not seem to have brought about an adequate level of proficiency in Malay among the children, ultimately having led to the predominant use of English at home – at least as long as the children are involved in an interaction.

Through parents’ OPOL policy coupled with mother tongue classes, the children growing in this family, however, seemed to receive just enough input to develop, at least, what Nakamura (2017) calls ‘hidden bilingualism.’ That is, the children may be able to, at least comprehend, or even use Malay, even though in bits and pieces, in private contexts where no English-speaking interlocutors are present, e.g. when grandparents and other relatives with little knowledge of English are around. In this light, the parents’ language policy could be considered ‘successful’ as they have achieved their goals in exposing their children to just enough Malay so that they can understand and speak Malay in bits and pieces. However, viewing this as the maintenance of Malay is problematic because the children’s current predominant proficiency and interest in English and their tendency to use it at home makes it very unlikely to imagine that they will use Malay to their children in the future. Moreover, a paramount question with far-reaching implications for language maintenance that also arises here is what will happen when those individuals with little knowledge of English are gone? In other words, in the long run, in what situations will these children as well as their own children in the future be motivated to use Malay?

Overall, this study shows that of the two important components of Singapore’s language management over the years, i.e. English as the medium of instruction and mother tongue education, English medium policy coupled with many other social pressures associated with it has produced a bigger ‘discursive ripple effect’ (Hult 2010, 19). This ripple, which has now reached Malay families, brings with it messages with respect to the significance of English as the key to academic success and more well-paid job opportunities, and as the national and international lingua franca. Given the societal dominance of English (Pennycook 2017), on the one hand, and the children’s role in influencing the language ecology of the family (Fogle 2012; Fogle and King 2013), on the other, what ensues in daily language socialisation activities is a shift to English. In Gafaranga’s (2010, 266) words, ‘The macro-sociological order OCCASIONS the conversational order’ (emphasis in original). As shown above, this impact of the sociological order on familial interactions is subtle which contributes to language shift taking place out of sight.

This means that, first, the private sphere of family in Singapore can no longer be taken as ‘an unexpendable bulwark’ of language maintenance’ (Fishman 1991, 94). Second, English-knowing bilingualism, as envisaged in Singapore’s language management, can and should not be taken for granted as a by-product of mother tongue education unless natural socialisation through and into mother tongues happens at home, talking language maintenance into being. To achieve such a goal, the parents need to be aware of how societal dominant languages trickle into their and their children’s language practices.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Seyed Hadi Mirvahedi

Seyed Hadi Mirvahedi is a postdoctoral fellow at Center for Multilingualism in Society across the Lifespan (Multiling) at University of Oslo.

Notes

1 The Malay words Kakak (older sister), Adik (younger sister), and Abah (dad), which connote affection, love, and endearment, are often used by both parents and children in Malay households (see also Said-Sirhan 2014).