Transnational Multilingual Families in China: Multilingualism as Commodity, Conflict, and In-Betweenness

Studies on transnational multilingual families and their language planning have mostly investigated their language ideologies and practices in relation to heritage language maintenance without exploring how such families view their multilingualism and how it might affect their language planning. Most studies have also exclusively focused on the experiences of transnational multilingual families residing in Europe and North America, with those living in other regions receiving comparatively little attention. This article reports on a qualitative study involving four transnational multilingual families located in China. Data were collected via a combination of semi-structured interviews and audio/video recordings. The study investigated the extent to which the participating families’ multilingualism mediated their language planning, as well as how they related to their multilingualism in its totality. The findings revealed that the families had a complex relationship with their multilingualism and saw it as a source of opportunities but also anxiety.


Introduction
Transnational multilingual families (TMFs) have become ubiquitous in the large cities of many a country due to the forces of globalization and increased labor mobility (Lanza, 2020). These families constitute spaces where diverse languages are spoken and where members navigate multiple cultural, ethnic, linguistic, and national identities. For policymakers, educational institutions, and other societal stakeholders, understanding the dynamics within such families is essential because these dynamics help to predict future educational, social, commercial, and even political trends as more and more families become transnational and multilingual (see Medrano, 2018;Spotti & Kroon, 2017). Indeed, TMFs challenge the concept of national identity by inhabiting complex spaces where hybrid identities and language practices are enacted (Colombo et al., 2020;Kumashiro-Wilms, 2017). Referring to the role of language in this complexity, Duff (2015) notes, "language is a key means by which this identity work is done, illustrating performers' appropriation and transformation of transnational or translocal cultural flows and ideologies as well as the indigenization of such practices" (p. 60). At present, the interplay between multilingualism, identity, and language planning has been explored in only a few TMF studies, with many studies approaching multilingualism in fractional terms (Grosjean, 1989) to explore the proficiency of TMF participants in a specific language (oftentimes a heritage language) or languages and how they relate to these (e.g., Fuentes, 2020;Fukuda, 2021;Hua & Wei, 2016;Kim, 2016;Obojska, 2019). The result is that we know little about how TMFs perceive their multilingualism in its entirety, as identity (Henry, 2017), that is, more than just their proficiency in multiple languages, and how this identity affects their language planning. Second, countries in Europe and North America are overrepresented in TMF research (Lanza & Wei, 2016), which has led some writers to emphasize the need for more studies on TMFs living in countries in other regions (Smith-Christmas, 2017). Studies indicate that multilingualism is perceived differently from country to country (Edwards, 1994), and so insights gleaned from the European and North American contexts concerning TMFs may not apply to other locales. In this respect, researching TMFs in countries like China, where the number of TMFs is growing rapidly and which constituted the focus of this study, would contribute to a more holistic understanding of TMF dynamics globally. Indeed, as Gao and Zheng (2019) point out, China likely has the largest number of language learners in Asia (if not the world) and routinely sends students to countries like France, Germany, Russia, and Spain, among others, for higher education. These students return to China more multilingual and multicultural than when they left and some start families with individuals from abroad. These families possess hybrid identities that draw from multiple languages and associated experiences, yet few studies have explored how such families relate to their multilingualism and hybridity. Alongside these demographic developments, the Chinese government is also promoting the learning of diverse languages in higher education institutions throughout Mainland China in support of the "Belt and Road" initiative, with this, likewise, having a significant impact on the country's multilingual and multicultural development (Gao & Zheng, 2019;Shen & Gao, 2019). Taken together, these changes are historic and present both challenges and opportunities for the country's educators and policymakers, who must find effective ways to harness this growing linguistic and cultural diversity as part of their efforts to promote a "harmonious society" (Shen & Gao, 2019;Tsung, 2014).
Third, and somewhat linked to the fractional approach to multilingualism employed in many studies where the focus is on the participants' languages and their proficiency in these rather than on the participants themselves, several TMF studies have conflated multilingualism with exclusively immigrant identity, often emphasizing the latter over the former. In these studies, family members share the same ethnicity, first language, and nationality, and often communicate monolingually (i.e., using their first language) at home (e.g., Hirsch & Lee, 2018;Hua & Wei, 2016). As such, the participating families represent neither truly transnational nor functionally multilingual spaces nor is their multilingualism explored beyond the languages that they purportedly speak. In general, the research focus proceeds along a familiar path, involving a discussion of how mostly monocultural immigrants, having moved to a new country, adjust to their environment, learn the country's majority language, and encounter difficulties that arise as a result of their immigration. These difficulties can range from fears that younger family members will forget their native language or lose interest in their native culture to the importance of assimilating with the host population by rapidly acquiring proficiency in the majority language. In all this, what is indirectly or directly emphasized is the immigrant experience instead of the families' multilingual identity (Henry, 2017;Pavlenko, 2006), and so the conclusions that one draws when reading such research have more to do with issues of ethnicity and nationality, as these relate to being an immigrant, than the multilingual experience.
This article reports on a study involving four TMFs residing in China. Each family comprised two parents and their child, with each parent being proficient in three or more languages. The parents did not share the same ethnicity or cultural background, nor did they have any first languages in common (i.e., they did not have the same first language background). The study focused primarily on the parents' views regarding how their multilingualism mediated their language planning, as well as how they perceived their and their children's multilingualism in its totality. In other words, the children and their multilingualism were presented through their parents' voices since they did not directly participate in the study. The study relied on interview data and audio/video recordings of family interactions provided by the participating parents. The study differentiates itself from previous research on TMFs and language planning in that it moves beyond a narrow focus on specific languages and issues of ethnicity or heritage and explores the multi-faceted views that TMFs possess regarding their multilingualism, regardless of the languages they speak and their proficiency in these, and how this mediates their language planning.

What Is Language Planning?
This study uses the terms family language planning and language policy interchangeably and defines language planning as a combination of language ideologies and practices that represent how languages are learned, taught, used, and viewed within families (King & Fogle, 2017;Spolsky, 2012). Family language planning need not always entail active planning; rather, it can move "along a continuum ranging from the highly planned and orchestrated, to the invisible, laissez-faire practices of most families" (Caldas, 2012, p. 352). Language ideologies, within the confines of language planning, can be described as beliefs or attitudes held by family members concerning how languages should be learned, transmitted, and valued (Soler & Zabrodskaja, 2017). For example, some parents may harbor native-speakerist tendencies (Obojska, 2019) so that they see native speakers as the only legitimate representatives of a given language. In multilingual families, subscribing to such an ideology can lead to parents implementing the one-parent-one-language approach (OPOL). OPOL generally entails the parents restricting themselves to using only their native language(s) with their children, who become socialized into seeing languages as distinct entities linked to a specific parent, with strong national and/or ethnic connotations. Language practices, meanwhile, are the observable actions performed by family members through language. In multilingual families, such practices can include OPOL (discussed above), translanguaging, and even using a hybrid family-specific language composed of various other languages, among others (Gomes, 2020;Hiratsuka & Pennycook, 2020;Van Mensel, 2018).
The specific language practices that TMFs implement at home and outside can reveal a great deal about how they conceive of multilingualism. For example, a multilingual family where the parents implement OPOL may indicate that they view multilingualism not as a unified, dynamic state where linguistic identities draw on and reinforce each other (Jessner, 2008), but as a series of disconnected monolingualisms where language ownership and identity are decided at birth and tied to ethnicity (e.g., a child born to an English couple will always have an English speaker identity and may not replace this identity with another or adopt a hybrid identity that comprises other languages). In contrast, a family where parents mix languages freely, irrespective of whether they are native speakers of these, could represent a view of multilingualism and identity as malleable, evolving, and dynamic states. Translanguaging, which is a word that traces its origins to Welsh bilingual classrooms, best captures this mixing of languages (e.g., using multiple languages within the same sentence or alternatingly from sentence to sentence) in that it helps us to conceptualize the linguistic repertoire of multilingual individuals as being a compound whole where the boundaries between specific languages blur as they "shuttle between languages" (Canagarajah, 2011, p. 401). Beyond the merely linguistic, translanguaging also represents one way in which multilingual individuals highlight their multilingual identity (Creese & Blackledge, 2010;Nguyen, 2019), with translanguaging symbolizing a hybridization of not only their ways of speaking but also their ways of thinking and being. As already mentioned, most studies on TMFs have focused on the participants' use of particular languages, including their proficiency in these, rather than on how they relate to their multilingual hybridity in its entirety.
The dearth of studies that have taken into account the multilingual identity of TMF participants (in a way that does not reduce to this merely their knowledge of three or more languages) forms part of a larger research gap concerning TMFs and their language planning, one where their language ideologies have often been explored without much attention being paid to variables like gender, emotions, and social class, as some researchers have pointed out (e.g., Gomes, 2018). Studies like those conducted by Tannenbaum (2012) and Tannenbaum and Yitzhaki (2016) are exceptions in this respect because they expand on the language planning concept to focus not only on particular languages but on how certain families use their languages as an emotional response to sociopolitical conditions or as a defense mechanism. Yet, even in these studies, like in most others (e.g., Kwon, 2020;Oriyama, 2016; for an extensive review of FLP studies involving TMFs, see Duff, 2015;Gomes, 2018;Hirsch & Lee, 2018;Lanza & Lexander, 2019), the participating families tend to be monoethnic and even functionally monolingual (at least with each other), and the findings concern the speaking of one or another language without a deeper engagement with the participants' multilingualism (see Kozminska & Hua, 2021). For instance, in the study by Tannenbaum and Yitzhaki (2016, pp. 576-577), one parent remarks, "We live in two languages. . . We don't have a clean Arabic and we don't have a clean Hebrew. It comes out mixed. . . There's no complete sentence that comes out in one language." Yet, this description of their bilingualism as in-betweenness and hybridity, of belonging neither to Arabic or Hebrew, remains mostly unexplored.

Multilingualism as Identity
In this study, multilingualism is defined as an individual's knowledge and use of three or more languages (Kemp, 2009). Yet, it is also more than just the sum of languages that one knows and uses. Pavlenko (2006), in her study involving 1,039 bi-and multilinguals, found that 65% of the participants felt that their bilingualism or multilingualism had led to them acquiring multiple personalities and viewpoints. One of the participants stated that "speaking another language causes me to assume certain cultural perspectives that also entail certain behaviors" (Pavlenko, 2006, pp. 11-12) while another noted that "speaking a different language means being a different person, belonging to a different community, character type, emotional type." The participants also had strong opinions about their multilingualism in its totality (regardless of what languages, how well, or how many of these they spoke), with many seeing it as a beneficial state that gave them choices they would not have had as monolinguals. They enjoyed the hybridity that being bi-or multilingual entailed and the fact that it made them think and feel in a multiplicity of ways. Some participants, however, reported feeling incomplete or split into multiple parts as a result of being bi-or multilingual. One of the participants noted that "sometimes I feel like being two different persons or just a person with two incomplete languages" and that the "worse moments are when I feel like not having a language identity" (Pavlenko, 2006, p. 25). Henry (2017) similarly found that the students he interviewed had distinct ideas about their multilingualism, including the presence of deeper feelings and a "personal value attached to being/ becoming multilingual," and awareness of "the social and intersubjective benefits of a multilingual repertoire," regardless of what languages comprised their repertoire. As one student noted: Being able to speak lots of languages, it is a feeling. That you are not like stuck in your little box. You are open to things and go out and discover, so to say. It is a feeling, how should I put it, of expanding your awareness in some way (p. 558).
In other words, multilingualism was as much an identity for the participants as it was about how many languages they used (or planned to use) seeing as it encompassed how they saw themselves in relation to others and the world around them (Romaine, 2011). Building on this link between languages and identity, Kroskrity (2001, p. 106) notes that "identity is defined as a linguistic construction of membership in one or more social groups or categories," with "language and communication" representing "critical aspects of the production of the wide variety of identities expressed at many levels of social organization," including in families. Mills (2006), drawing on social theories like language socialization, observes that an "individual's verbal repertoire works toward defining the self" and "acts as an index of identities" so that "it is language that constructs identity rather than various social identities being summoned up by language." Yet, as already mentioned, such an approach to multilingualism has often not been applied by researchers in TMF studies, with most either focusing on the participants' ethnic and immigrant identities instead of their multilingual identities or exploring their use of, proficiency in, or emotional attachment to a particular language or languages without delving more deeply into the identity-related dimensions of their multilingualism.

The Interplay Between Multilingualism and Language Planning
In TMFs, multilingualism might mediate language planning in various ways by giving family members access to certain identities and associated practices while simultaneously limiting their access to others (Burck, 2005). For example, children in TMFs could experience difficulties in finding complete acceptance among monoglossic or monoethnic peers, with this influencing their language practices and even their views regarding multilingualism. This is because their hybridity would be seen by some as in-betweenness, representing a lack of purity or loyalty to any one country or group, similar to how multilingualism was viewed in the past (Ter-Minasova, 2005). In countries where monolingualism and monoculturalism are widespread, TMFs can seek to hide their multilingual identity and use only the country's majority language so as not to face social ostracism and discrimination. Referring to this dynamic, Colombo et al. (2020, p. 78) observe that "monolinguals tend to conceptualize plurilinguals as "the other," which explains why plurilinguals often seem reluctant to reveal their plurilingual identity, as it may be used as grounds for othering." In such instances, language planning in TMF homes can acquire a monolingual character in that the parents, despite being multilingual, do not desire the same for their children due to fears that they might be "othered." Multilingualism could also mediate language planning in TMFs via ideologies that accord more power to some languages (and associated identities) than others. In these instances, languages, and therefore multilingualism, come to be viewed as commodities and treated as "an objective skill, acquired and possessed, that affords status, recognition legitimacy, and ultimately material remuneration, to those who possess it" (Block, 2017, p. 126). Due to this commodification, some multilingual profiles (based on a particular combination of languages) may be perceived as affording their speakers a higher status than do others (Gogonas & Kirsch, 2018; see also Tannenbaum & Yitzhaki, 2016).
For families, the commodification of multilingualism leads to language planning that prioritizes the learning of high-status languages and discourages the learning of lowstatus ones. These power dynamics are visible in countries like the United Arab Emirates, where parents have been found to promote the learning of English and other foreign languages (e.g., French) while according little or no importance to Arabic, despite its status as the country's sole national language (Tang & Calafato, 2021). Lastly, the interplay between multilingualism and language planning can lead to instances of conflict, especially in TMFs, due to the presence of multiple languages and, therefore, perspectives, discourses, and behavior. As Michalski (2021, p. 3) notes, "the capacity for syntactical language assumes a key role regarding conflict. . . language facilitates communication, shapes definitions of situations, reinforces group boundaries, and establishes cultural identities." Conflicts may affect language planning dynamics in TMFs in various ways, for example, by leading to family members harboring negative attitudes concerning the benefits of being part of a multilingual, multicultural family. This does not mean that conflicts do not occur in monocultural/monoethnic/monolingual families, yet studies indicate that multiethnic, multicultural marriages are often susceptible to higher instances of conflict (Hohmann-Marriott & Amato, 2008;Smith et al., 2012). At the same time, these studies focus mostly on static concepts like ethnicity rather than languages as a marker of identity, meaning that the multilingualism of the participants remains largely unexplored, including what role it might play in fueling or mitigating conflicts.

Research Questions
In light of the limited number of studies on TMFs residing in countries outside Europe and North America, as well as significant research gaps when it comes to our understanding of how their multilingualism (including their multilingual identity) influences their language planning and how they relate to their multilingualism in its totality, this study explored the following research questions (RQs):

Participants
Four TMFs participated in the study, with all four residing in different cities in China on a mostly full-term basis, especially due to coronavirus-related restrictions. Table 1 provides sociobiographical information regarding the participants, including the parents' home provinces or cities, the languages they reported using in their daily life, and their level of education. The participants' names have been altered to protect their anonymity.
In each family, as can be seen in Table 1, both parents were fully multilingual in that they used at least three languages and none of them shared a common first language with their partner (i.e., they were not of the same ethnicity or nationality, unlike the participating families in many other TMF studies). It is also worth noting that each of the parents, regardless of their nationality and first language, reported using English, something that constitutes a hallmark of multilingualism in the modern era. Indeed, English has witnessed an "unparalleled spread" as an international language so that, in many countries, regardless of how geographically, culturally, historically, or politically removed these may be from the English native-speaking world, one frequently encounters "multilingualism with English" (Aronin & Singleton, 2008, p. 3). Sampling was convenience-based, with TMF recruitment taking place through acquaintances. The only criterion for participating in the study was that the parents should be from different nationalities. This was done to ensure that the families were truly transnational and multilingual, unlike in many other studies on TMFs where, as already mentioned, participating family members share the same ethnicity and nationality, and are not functionally multilingual (at least not at home or with each other). During the recruitment phase, the families were emailed an information sheet explaining the study's objectives, data collection procedures, the families' role in the study, their rights as participants should they choose to participate, and our contact details. They were assured that their participation in the study would be anonymous, that they could withdraw from the study at any point, and that all data collected would be encrypted, securely stored, and ultimately anonymized once the study concluded (all recordings would be deleted). The families were asked to sign a consent form to indicate that they had agreed to participate in the study.
At the time of the project, Ahmed and Nuwa had been married for 9 years and resided in Ningbo in Zhejiang province. Ahmed worked for a trading company and often traveled to Guangzhou on business for long periods. Originally from Egypt, he had been living in China for 15 years when the family volunteered for the study. Nuwa stayed home to look after their 7-year-old daughter, Mayleen, who had become a child model. All three spoke a mix of Arabic, Chinese, and English at home, whereas they only spoke Chinese with each other outside. Li and Vadya had been married for 8 years and had been living in China for 3 years when they volunteered for the study. Before living in China, they had spent a decade in Russia where they completed their graduate and post-graduate studies. Their return to China had been precipitated by financial necessity and they had decided to open their own business in the country, which they saw as offering them the best chance to secure their future financially. Both were fluent in English and Russian, and Vadya had advanced proficiency in Chinese. At home, they mostly spoke Chinese and Russian, and some English. Nastya and Wei had met at a university in China where Nastya was completing her undergraduate studies. When we interviewed them, they had been married for 8 years and Nastya had been living in China for close to 10 years. Nastya and Wei spoke English with each other when they first became a couple, though Wei started taking Russian lessons soon after they met and they gradually started communicating in a mix of Chinese, English, and Russian at home.
Metin and Yue had met and married in China. Metin, who was originally from Turkey, had founded a company in Hangzhou, having lived in China for close to 12 years when the study commenced. Yue stayed busy managing affairs at home and looking after their son, Han, although she sometimes helped Metin with administrative work at his company. The couple had initially only communicated in English because Yue did not know Turkish and Metin's Chinese was still somewhat poor at the time. Yue had subsequently

Data Collection
The study collected data from each TMF via semi-structured online interviews and audio/video recordings of their family interactions. The interviews were conducted with the parents over Skype in English and Mandarin Chinese since all the participants spoke these two languages. Some of the children, on occasion, made an appearance to talk to their parents about something but they did not actively participate in the interviews, which lasted around 40 to 45 minutes. The online interviews served as the primary source of data for the study while the audio/video recordings were used as a secondary data source to complement the interviews. This study takes the view that individuals in TMFs are socialized into new dispositions and behaviors through the acquisition of multiple languages (Duff, 2015), which leads to the formation of unique and dynamic multilingual identities (Henry, 2017;Pavlenko, 2006) that extend beyond their original ethnic or national identities, as well as their proficiency in the languages that constitute their multilingual repertoires. These identities can then interact with their language planning in various ways, something which we sought to explore as part of the study's research aims. During the interviews, the parents were asked about how they met, the languages they used with each other and their children, their language learning experiences, how they felt as a multilingual family, the significance that this held for them, and how they perceived their multilingualism, including their identity as multilinguals, in relation to themselves, their relatives, and society, among other questions. As for the recordings, these were provided by the parents following the interviews. It is important to note that the recordings were not made especially for the project; rather, they consisted of video and audio material that the families had recorded on their smartphones over the years to capture their family life (e.g., mundane interactions throughout the day like banter while watching a TV show, chatting when doing chores around the house, etc.). As such, they served as a memory device, a store of content mediatized through the act of being recorded (Kapur, 2018), as well as an integral part of overall family functioning (Coyne et al., 2014).
In the present, especially with the proliferation of social media networks and the significant advances in video and audio capture technologies contained within smartphones, it is normal for individuals and families to record various aspects of their daily life, from the routine to the special, in the form of audio/video clips. Our decision to seek out these recordings was based on the belief that they represented authentic, ready-made digital traces of family life (Legewie & Fasang, 2021) as opposed to if the participants had been asked to make custom recordings for the study. The latter would have inevitably led to staged, artificial interactions, to some extent, given the participants' awareness that their recordings were being created for the researchers. In eliciting the recordings from the participating parents, we informed them that these would provide us with additional insights into their language planning, alongside the interviews, and help us better understand their multilingualism, specifically, how they, as a multilingual family, actually interacted with each other on a daily basis using their linguistic repertoires (i.e., observed versus reported language practices). We also requested that they avoid selecting recordings that contained sensitive topics, for example, where they or their children might discuss personal information related to someone's health or refer to third parties not involved in the study. The parents sent in their selected recordings via encrypted file transfer, with these varying in length, though most were no more than 2 minutes long.

Data Analysis
Using MAXQDA, we grouped the recordings and interviews based on family and then fully transcribed them following the recommendations laid out in Kuckartz and Rädiker (2019), although we excluded silences, pauses, background noises (i.e., external interruptions), and affirmative utterances (e.g., aha, mhm, etc.). Unlike the Skype interviews, where the participating parents sat mostly immobile in front of their computer screens, and the audio recordings, where non-verbal actions and location could only be surmised, the video recordings featured physical movement quite prominently at times, providing additional context to the families' interactions, similar to found footage. When transcribing the videotaped recordings, we transcribed the participants' nonverbal actions within square brackets (see Kuckartz & Rädiker, 2019). As already mentioned, the audio and video recordings provided by the participants contained authentic instances of family interactions during everyday activities (e.g., watching TV, eating dinner, etc.), with these frequently being brief and containing no instances of long dialogue or extended discourse. Once transcribed, the data underwent thematic analysis (Braun & Clarke, 2006). Table 2 provides an example of the coding process using an extract from one of the interviews.
During the coding procedure, the interview transcripts were read multiple times, with each reading leading to the creation of codes that were refined during subsequent readings. The coding process was inductive and the interview transcripts were read and reread until saturation was reached in that no more codes could be generated. The codes were then checked with the transcripts of the recordings and additional codes were generated and existing ones refined where needed. In general, since the recordings contained instances of everyday family interactions, the changes that were made to the codes during this stage of analysis concerned the families' language practices, that is, how they used the languages, and not how they perceived their multilingualism. This process continued until the codes could be refined no further, resulting in a final set of codes that were checked with the interview transcripts to ensure that the changes made during the readings of the transcribed recordings reflected the data from the interview transcripts. Next, the codes were studied and collated into themes, which were then checked with the entire data set and finalized.

Findings
The following sections contain a discussion of the findings organized based on the study's research questions. As such, the first section concerns how the participating families' multilingualism mediated their language planning (RQ1) while the second one covers the participants' views regarding their multilingualism (RQ2).

RQ1: Multilingualism and Language Planning
All four couples reported engaging in translanguaging at home and described the mixing of languages as a normal aspect of their family life. The parents reported using languages freely and did not ask their children to use a particular language (or languages) at home. They also avoided adopting a maximalist approach to language use in that they did not feel that their varying levels of proficiency in the languages they knew hindered their use of multiple languages with each other. The audio/video recordings appeared to confirm this, with those provided by Li and Vadya, for example, containing several instances where they can be heard speaking in a mix of Chinese and Russian with each other, switching between the two languages within sentences. The two also used English at home, although they reported feeling freer in Chinese and Russian. Li said that they could more accurately convey their thoughts and emotions when mixing languages. When asked what languages they spoke with Bolin, the couple reported speaking in Chinese, English, and Russian in front of him. Vadya felt that, being a multilingual family, it was important to show Bolin that using languages interchangeably was natural and important, and that this "switching or crossing between languages" benefitted their son's development. The parents' views regarding translanguaging being an important aspect of their identity as multilinguals also contained an element of pride in that it was viewed as a unique skill to which not everyone had access and one that not all multilingual individuals knew how to use well. For instance, Nastya and Wei compared each other's ability to engage in translanguaging and measured a multilingual individual's translanguaging competence based on how deftly they could switch between languages when strong emotions (e.g., anger, sadness, joy, etc.) were involved. Wei specifically praised Nastya's translanguaging, noting that she could "switch between Chinese and Russian on a moment's notice." He said that he could not do this as skillfully as his wife, adding that Nastya had a gift when it came to learning languages. Like Li and Vadya, Nastya and Wei also used multiple languages at home, mostly a combination of Chinese and Russian with each other, and Chinese, English, and Russian with their daughter, Alma. The recordings confirmed that the family used multiple languages during interactions, although, unlike Li and Vadya, they did not switch between languages within sentences: Asked why they no longer used English as frequently as Chinese and Russian when communicating with each other, Nastya said that it had happened "organically" over time and that she still regularly used English at her workplace, an international school in Hangzhou. Wei said that he found it difficult to express strong emotions when using English and so he would often switch to speaking Chinese or Russian with Alma and Nastya. He observed that this habit had intensified in recent years so that he no longer spoke English as frequently. Questioned why they spoke English with Alma, the couple said that they wanted their daughter to become multilingual, much like they had, though the two differed in how multilingual they wanted her to be. Nastya said she would be satisfied to see Alma acquire proficiency in Chinese, English, and Russian, whereas Wei said that he wanted Alma to learn more than just Chinese, English, and Russian: Let the child learn as much as possible. The three languages should be considered few. After growing up, she can learn a few languages abroad, for example, Spanish, Japanese, Korean, etc. She will naturally learn foreign languages in the future. (Wei) Meanwhile, for three couples (excluding Metin and Yue), their multilingualism also influenced their language planning in that they reported strategically and deliberately using specific languages within the family, both separately and in combination, to conceal or reveal information to those outside the family, for instance, when out in public. Referring to one iteration of this, Li and Vadya said that, as a family, their language practices had evolved over the years from using only Russian in Russia and Chinese in China outside the home so as not to attract attention to themselves to now consciously using English and Chinese in Russia and English and Russian in China. Vadya said that one of the advantages of being a multilingual family was that they could choose which languages to use in front of others and thereby decide how others understood and perceived them. He said that using different language combinations altered their appearance in the eyes of their interlocutors, something that they could use to their advantage to influence people's emotions and behavior. He said that using languages in this way also allowed them to talk more freely about anything in public without fear of anyone listening in and understanding what was being said.

RQ2: Perceptions of Multilingual Identity
The themes concerning the participants' perceptions of their multilingualism are illustrated in Figure 1. Conflict represents the tension and disagreements that the participants linked to being multilingual while Valuable Asset covers the participants' view that their multilingualism provided them with several benefits, both tangible and intangible. Expensive Commodity concerns the participating parents' beliefs that their multilingualism was a profitable commercial product but also financially draining in that it required significant investment. Valuable Asset and Expensive Commodity are discussed together because of a slight overlap between the two. Meanwhile, Hybridity represents how the participants' viewed their multilingualism as changing their ways of thinking, doing, and being. Finally, In-betweenness relates to the participants' perception that their multilingualism had made it difficult for them to find acceptance in, or identify with, any one group. As with Valuable Asset and Expensive Commodity, Hybridity and In-betweenness are discussed together.
Conflict. All four couples referred to their multilingualism as a source of conflict and tension. According to them, multilingualism involved two related yet ultimately separate aspects: culture and language, and they felt that proficiency in multiple languages did not make one proficient in multiple cultures. Some of the parents felt that being multilingual, therefore, meant not only using several languages but also incorporating the cultures, ways of thinking, and behavior associated with speakers of these languages into one's self. For instance, Li and Vadya emphasized that multilingualism sometimes created uncertainty for them because it was not always easy to understand its various aspects, which then led to conflict in the family: Researcher: As a multilingual couple, how were things when you first met? Li: In the beginning, it was like two actors. After all, they were from two countries, with different cultural backgrounds, different ideological systems, and different language structures, which led to conflicts in behavior and ideas. Vadya noted that being part of a multilingual family meant being exposed to the very different ways of seeing things that were represented in the family. Reflecting on their years together, Wei talked about how he felt that becoming multilingual had contributed to more arguments between Nastya and himself. He thought that the languages they had learned had affected their thinking differently, adding that they used to understand each other more easily when they only spoke English to one another. He said that he had realized over time that a multilingual family was more than just speaking several languages; it was about being tolerant of each other and learning about and respecting each other's worldview. The couples' multilingualism also led to interesting and somewhat unexpected dynamics during periods of conflict. In two families (Ahmed and Nuwa, and Nastya and Wei), the couples reported going completely silent when they quarreled in that they stopped talking to each other during such instances despite being able to communicate in multiple languages. Nuwa used the term "Cold War" to describe these periods, revealing that their first years of being a family had been "emotional torture." She said that while they spoke multiple languages in the family, they could not get each other to fully understand what the other meant.
Hybridity and in-betweenness. Two couples (Li and Vadya, and Metin and Yue) described how their transnational multilingualism had hybridized their ways of thinking and being, although with very different results. For Li and Vadya, for example, their multilingualism had led them to integrate each other's "ideological systems" into their respective worldviews. Li felt that Vadya's use of Chinese had made him "think like a Chinese person." Vadya similarly felt that Li had become more Russian as a result of learning Russian and that she understood "Russian culture and the Russian ideological system." He referred to this fusing of multiple "ideological systems, cultures, and languages" in each of them as "a perfect combination." In contrast, hybridization in Metin and Yue had caused them to gravitate toward each other's culture and language while leaving behind parts of their own to some extent. Yue, for instance, noted that Metin had become more Chinese in this thinking and had even forgotten bits of English and Turkish. She said that he preferred to speak Chinese with her and their 5-year-old son, Han, and revealed that he would often reply to her in Chinese even when she spoke English or Turkish with him. She added that he sometimes struggled to recall certain words in Turkish and asked her for help when doing so. On the other hand, Yue was strongly drawn to Turkish and Turkey. She studied the language diligently, had enrolled in online courses for Turkish, and wanted to move to Turkey, something which she had discussed with Metin. She said that Metin had agreed to think about it, although he had been noncommittal and preferred that they stayed in China. Asked why she took online courses when she had Metin to help her, she pointed out that he seemed to have lost interest in Turkish and did not make an effort to help her learn it: He doesn't want to teach me. He started to speak Chinese to me after teaching me a few sentences. I am impatient. I feel that I still need to learn more of the language and am a little sad that I haven't. (Yue) She said that she had found Turkey pleasant when she visited the country and that her in-laws were very welcoming and supportive of her desire to learn Turkish. She said that Metin's family, especially his brother, were proficient in English, in addition to Turkish, and she communicated with them in a mix of the two languages. Discussing her desire to move to Turkey, she felt that the country was also a good option for Han in terms of higher education and that it would be easier for him to get into a good university in Turkey as opposed to China, where competition was more intense. She said that being part of a family that was multilingual and transnational meant that Han would have more opportunities for education and employment in diverse countries since he had been exposed to multiple languages and cultures from birth; however, she also worried that Han would feel that he did not belong anywhere: I have always worried that my son does not have a sense of belonging. I told him that you can say that you are Chinese and it doesn't matter because his nationality is Chinese. But when he went to Turkey, the children played with him and asked, "Where are you from?" He said he was a Turk, and in China, he says he is Chinese. He looks more Chinese and does not have any distinctive foreign features. Some adults think he is of mixed-race when they see him but when playing with children, they can't tell. (Yue) Valuable asset and expensive commodity. All four couples explicitly referred to multilingualism as a valuable asset that helped them with employment, mobility, and business opportunities, and felt that it boosted cognitive flexibility and intelligence. One couple (Ahmed and Nuwa) also described multilingualism as an expensive albeit highly desirable commodity among parents in China that required a lot of financial investment. For instance, both Ahmed and Nuwa expressed a strong desire for their daughter, Mayleen, to learn several languages beyond just Arabic, Chinese, and English, although they noted that learning languages was an expensive endeavor in China. Ahmed said that developing their daughter's multilingualism "required funds" that they did not currently have, explaining that they were saving money to buy a house for Mayleen. Nuwa thought Mayleen's multilingual proficiency was lower than that of other children her age from the families they knew in Zhejiang. She said that the children from these families were learning several languages with tutors and that language institutes had become a very profitable business in China due to the high demand for foreign language education. She said that while they were dissatisfied with Mayleen's situation, they would focus on investing in her language learning after they bought the house: We must first solve the housing problem and then consider studying languages. Later, she will make up for it but, in the early stage, the house is our main concern. (Nuwa) As Nuwa saw it, "it doesn't matter if you and your husband don't own a house, but your child must have a house." At the same time, she noted that multilingual and multicultural individuals were in strong demand among companies for a variety of positions, including for marketing and promotional purposes, especially children. Nuwa said that Mayleen attracted a lot of interest due to her background and knowledge of multiple languages, revealing that she received many offers to work as a child model, which had helped the family immensely with their finances. She said that being multilingual was also crucial for Ahmed given his involvement with a trading company and the ensuing need to interact with people from around the world. She said that he needed to further develop his multilingual proficiency, which would bring him more customers. Meanwhile, Nastya and Wei felt that in addition to the financial benefits, being multilingual also allowed one to move to a new place and settle in without any difficulties. He noted that multilingual children had an easier time adapting to different countries and making friends and that he had even witnessed this with children in China who knew several Chinese dialects. This is the advantage of multilingual children, for example, even children from different cities in China can speak several dialects. (Wei) Lastly, multilingualism was associated with greater mental flexibility and intelligence in the minds of the participating families. For instance, Li and Vadya stressed the importance for Bolin to grow up to become multilingual, much as they had. They felt that it would ensure that he was more intelligent and had a more multi-faceted worldview: Because just like a person exercising, if there are more languages, the child's way of thinking will develop in many ways and it will be more flexible, which means that his mind will be more intelligent. (Vadya)

Discussion
This study sought to explore how four TMFs residing in China perceived their multilingualism and how it mediated their language planning. Regarding the families' language planning, the findings indicated that multilingualism mediated this in three primary ways. First, their multilingualism led the parents to implement translanguaging with each other and their children, with this translanguaging being both intersentential and intrasentential. Moreover, all the families considered translanguaging a natural part of being multilingual and reported using the languages in their repertoire freely during interactions (except for perhaps Metin). They also mixed their languages without assigning them to particular life domains for use or expressing sentimental attachment to a particular language, nor did they speak only their first languages due to a belief that they were at their most authentic when doing so, as the participants in several studies on multilingual families and their language planning have been found to do (e.g., Hua & Wei, 2016;Soler & Zabrodskaja, 2017). In fact, most participants felt that they were at their most accurate when expressing their thoughts and emotions when mixing languages. Second, concerning language ideologies, the participants emphasized the naturalness of switching between languages and the need to see languages as interchangeable, which they felt was both a sign of being multilingual and also good for one's development. Several writers (e.g., García, 2018) have described translanguaging as a natural aspect of being multilingual and how it is used in complex meaning-making practices. Yet, it has often not been explored in studies on TMFs (for a review, see Gomes, 2018), where, as already mentioned, the participants were frequently monoethnic and functionally monolingual.
It is also worth mentioning that, in this study, the families' language planning did not involve them limiting themselves to speaking a specific language with each other or their children, for example, in the form of the one-parentone-language (Soler & Zabrodskaja, 2017) or only-theheritage-language-at-home approaches (Hua & Wei, 2016) that many studies have found families subscribing to. Worries about proficiency (e.g., Yue and her desire to improve her Turkish), likewise, did not hinder the parents' use of multiple languages when interacting with each other, in contrast to the findings from other studies (e.g., Kozminska & Hua, 2021). The free use of languages and their emphasis on seeing languages as interchangeable may have stemmed from the participants seeing a reduced distance between the languages in their repertoires as a result of being multilingual (Benet-Martínez & Haritatos, 2005), similar to some of the US-European couples in the study by Piller (2002). Alternatively, they may have integrated all the languages in their repertoires (Canagarajah, 2011) in such a way so that these no longer constituted different languages but rather parts of one unified language. There is evidence of this occurring in the families, for instance, when Li and Vadya discuss how they feel freer when expressing themselves in a mix of languages. Similarly, Nastya and Wei refer to translanguaging as a particular skill, with Wei praising Nastya's ability to switch between languages regardless of the emotions involved. One reason for these findings might be that the four couple families in this study were fully multilingual, where each member spoke at least three languages, and the family was multiethnic and multicultural, unlike, as already mentioned, in other studies on multilingual families and their language planning.
Third, the participants' multilingualism affected how they used their languages outside the home, providing them with several options to use it strategically to influence others, including denying or giving them access to information. Such practices indicated that their multilingualism had changed the way they behaved in society through language, from evincing conformity to using languages deliberately in specific combinations as a mark of their identity as multilinguals and even to ensure confidentiality and privacy in public spaces. Li and Vadya's deliberate use of Chinese and English in Russia and English and Russian in China, for example, may have been their way of indexing aspects of their multilingual identity (Duff, 2015) and how they chose to relate to their interlocutors and the general public in the two countries. The participants' use of their languages in this way was also at odds with the language practices implemented by the participating families in other studies (e.g., Cui & Zheng, 2021;Fuentes, 2020), where the participants generally opted to blend in, sometimes due to fears of drawing attention to their immigrant identity and being "othered" (Colombo et al., 2020). As for how the participating couples perceived their multilingualism, the data revealed five themes that were consistently referenced during the interviews (for a visual representation of these, see Figure 1). The participants perceived multilingualism as a valuable asset in that they felt that it specifically boosted executive function and cognition (mental flexibility and intelligence) and led to them possessing a more open personality (i.e., a broader mindset). They also felt that it enhanced their mobility, allowing individuals to move to new places and adapt to their new surroundings quickly. Multilingualism was also financially rewarding in that it could enhance one's employment prospects and attract more customers to one's business.
Several studies have touched on the cognitive and psychological benefits of being multilingual (Festman, 2021;Kroll & Dussias, 2017), which the findings support. As for the economic benefits of multilingualism, this has been covered in a limited number of studies (Curdt-Christiansen, 2016; Gogonas & Kirsch, 2018), where the participants mostly linked the learning of specific languages (e.g., English) to better job opportunities or discussed the potential benefits of being multilingual without experiencing these benefits for themselves. One of the families in this study (i.e., Ahmed and Nuwa) also touched on how the acquisition of multilingualism had become an expensive undertaking and profitable industry in China, where an increasing number of parents were sending their children to language institutes to learn several languages. These findings provide meso-and macro-level insights into language learning trends in China and the growing importance that parents in the country have attached to being multilingual. They can also be viewed as confirming, at least to some extent, that the language learning initiatives implemented by the Chinese government to promote multilingualism in the country (Gao & Zheng, 2019) have positively affected families and their language planning. At the same time, the commodification of multilingualism has several implications for society. On the one hand, it signals the increasingly high value attached to multilingualism while, on the other, it suggests that multilingualism may not be economically feasible for everyone. For families with insufficient funds, their limited ability to invest in their children's multilingualism, as was the case with Ahmed and Nuwa, could affect their access to opportunities later in life.
A situation where multilingualism is only accessible to a select number of individuals based on their wealth may ultimately exacerbate power imbalances and inequality in society. Studies on elite multilingualism and the effects of neoliberal policies on language education certainly allude to the commodification of multilingualism (De Costa, 2019; Sharma & Phyak, 2017), although this commodification has seldom been explored through the eyes of families and their language planning (see Bae, 2013), including how they view such commodification and the financial and psychological toll it takes on them. Given the Chinese government's promotion of the learning of multiple languages, especially languages other than English, as part of their Silk Road initiative (Gao & Zheng, 2019;Shen & Gao, 2019), and the expenses associated with formally acquiring foreign languages in the country, as reported in this study, it is hoped that future research will investigate the effects of the commodification of multilingualism on families and their language planning in greater detail, ideally through quantitative means that involve surveying a large number of participants. The families also perceived their multilingualism as a source of tension and conflict. Conflict in multiethnic marriages is not a new phenomenon and has been reported in several sociological studies (e.g., Johnson & Warren, 1994); however, in this study, conflict was not linked to ethnicity or race; rather it was linked to multilingualism and how it led to divergent mindsets. As Wei observed, he and Nastya understood each other better (i.e., fewer conflicts) when they only communicated in English.
At the same time, the conflicts that the participants reported having also signaled the beginnings of hybridity, that is, a melding of ways of thinking and being as the families' multilingual identities solidified over the years. This is evident in the interviews where the participants mentioned becoming more attuned to each other's ways of thinking and mastering each other's ideological systems. For instance, Vadya alluded to the meshing of Chinese and Russian cultural and ideological systems when he talked about how well he and Li understood each other. This meshing was an indication that what were previously mono-systems involving a single language, mindset, and culture had evolved to encompass multiple cultures, ideologies, and languages. The findings support the observation that multilingualism can lead to changes in the way individuals think and see themselves (Colombo et al., 2020;Medrano, 2018). In this study, hybridity was dynamic and individual-specific in that some integrated their various identities in equal measure, whereas others reduced their attachment to some identities in favor of others. These dynamics can be seen in Li noting that Vadya had started to think like a Chinese person, as well as in Metin's gravitation toward Chinese at the expense of English and Turkish. Finally, multilingualism was also perceived as in-betweenness. This was poignantly communicated by Yue, who feared that her son did not have a sense of belonging to either China or Turkey. As Burck (2005) notes, multilingualism can provide individuals with access to some identities while closing off access to others. In Han's case, while multilingualism could provide him with more opportunities in terms of career prospects, education, and mobility, it could also lead to him feeling "rootless" (Tannenbaum & Tseng, 2015, p. 289) and no longer finding complete acceptance among the cultural and ethnic groups of his parents.

Conclusion
Researching TMFs and their language planning contributes to our understanding of how individuals in today's globalized world are being socialized into complex identities via the acquisition of multiple languages, how they perceive multilingualism, and how it informs their language ideologies and practices. The findings from this study indicated that the couple families perceived their multilingualism in diverse ways, highlighting the complexities of their relationship with it and the fact that it was a source of many benefits but also conflict. In addition, the participants' responses helped to shed light on how a growing number of parents are investing in their children's multilingualism in China, with significant implications for multilingual and multicultural development in the country, something which researchers have only recently begun to explore. In any event, more research, both qualitative and quantitative, is needed on the language practices and experiences of TMFs residing in countries outside of Europe and North America so that we may obtain a more comprehensive understanding of the many ways in which transnationalism and multilingualism affect family life and, by extension, society globally. At the same time, studies should more explicitly cover how families' views of their multilingualism, as both knowledge of multiple languages and identity, affect their language planning while avoiding using multilingual as a foil to explore heritage language maintenance, immigrant identity, or the learning of a specific language, as has been done in much of the research on TMF language planning until now. Secondly, studies may obtain deeper insights if they distinguished between bilingual and multilingual families, as well as monoethnic and multiethnic families, when investigating how multilingualism affects language planning. At present, the majority of studies have mostly investigated bilingual families with the same national background. Lastly, in this study, the participating families' perceptions of multilingualism brought up themes of commodification, financial investment, hybridity, and in-betweenness, which we found to be especially interesting because these provided a nuanced view of multilingualism as a dynamic force that was both inclusionary and exclusionary. It is hoped that these aspects of multilingualism will be explored more deeply in future language planning studies involving TMFs.

Author Note
The submitted manuscript has not been published previously in any form.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.