Language Practices within the Mixed Spanish-/Italian-/French-and Estonian-Speaking Families in Tallinn

: This phenomenological study examined six mixed families living in Tallinn who are composed by French-/Italian-/Spanish-Estonian native speakers, who have at least one child who is being raised simultaneously with the combination of French-/Italian-/Spanish-Estonian and who all appeared to follow the one parent one language strategy as family language policy. The semi-structured interviews were conducted with parents. The theoretical aspect features family language policies and strategies, identity and its types, globalisation forces, bilingualism, and multiculturalism. The research aimed at highlighting the reasons behind parents’ ideological decision, more speciﬁcally, on how these bilingual families manage and adapt their language policies. The study shows how families control their chosen strategies. Research revealed in which languages children prefer to speak if they have been raised in multilingual environment. The results demonstrated that parents prefer to use a one parent–one language approach and they are led by their intuition and desire to speak in their own mother tongue with their children. It was found that bilingual reading to children during their ﬁrst years contributes to their ability to speak in both parents’ mother tongues. Data showed that bilingual children living in Tallinn prefer to speak Estonian while having competency in both languages. This study revealed that parents were content about their children being bilingual. Author Contributions: Conceptualization, R.S. and A.Z.; methodology, R.S.; software, R.S.; valida-tion, R.S. and A.Z.; formal analysis, R.S.; investigation, R.S.; resources, R.S. and A.Z.; data curation, R.S.; writing—original draft preparation, R.S. and A.Z.; writing—review and editing, R.S. and A.Z.; visualization, R.S.; A.Z.


Introduction
Globalisation has enabled ethnic diversity in Estonia, a country that is known for its innovative technology and entrepreneurship and called the "next/new/better Silicon Valley" [1][2][3]. It is certainly not the only reason, but definitely a huge advantage for businesses to attract innovative employees from all over the world to come and work here. People from other countries come to Estonia to either work, study, or begin a new life. In the course of their work and/or study experience, they may meet someone and start a new life and begin a family in Estonia. This research focused on communication within multilingual families, and how they are supporting their children's language development process within multilingual families. Each of these families are prone to make many decisions before the child is even born. Bi-and multilingual families need to establish linguistic preferences bearing in mind the "spaces" inside or outside the family boundaries such as grandparents and/or school [4] (p. 1). Even when the boundaries and strategies have been set ideally, they tend to differ in practice. A very common phenomenon is that both parents speak their own native languages with their children and a common home language can be a third language, for example a language of globalisation, such as English, and the parents use the foreign language, e.g., English, at home between them and sometimes with their children [5].
This article explored the phenomenon of Family Language Policy (hereafter FLP) within bi-and multilingual families in Tallinn. FLP is defined as "explicit and overt planning in relation to language use within the home among family members and provides an integrated overview of research on how languages are managed, learned, and negotiated within families" [6] (p. 907). When it comes to migrant families and their FLP, one must bear in mind how many generations a family has lived in a new linguistic environment. This topic has been widely researched by Joshua A. Fishman and one of his language assimilation theories is a three-generation theory which describes that when two or more languages are in an uneven situation in the family, then, generation-by-generation, a dominant language will most probably prevail by the third migrant generation [7] (p. 478). Another theory by Fishman, which needs to be taken into consideration when speaking about FLP, is the theory of socio-cultural contact and change. There exist many external forces which can influence language maintenance and/or shift it [8] (p. 50); those external forces definitely have an impact for FLP, whether short-or long-term (for example, historical, political, economic, or even military predominance of the societal language).
Although multilingual families in Tallinn are not rare, the object itself has not been researched enough, see ref. [5]. This study focused on a wide group of subjects by including Spanish, Italian, and French speaking bi-and multilingual families living in Tallinn and their language practices. The purpose of this research was to identify whether there is a pattern on how those families apply FLP in Tallinn and their motivation.
Bernard Spolsky [9] said that the basic principle of the FLP is best understood by connecting it to three factors: ideology, management, and practice. If Spolsky's three components of language policy are adapted to the family, then ideology can be understood as the family's beliefs and perceptions about language and language use, management as the steps taken by parents to guide language use in the family, and practice as the way language is actually used in the family. Regarding language ideology, he means the beliefs towards the language, what kind of language is appropriate, and what language has the prestige, etc. Under the language management, he defines the effort for changing language use by intervening, planning, or managing what language or languages are to be learned (status, corpus, acquisition planning). Language practices are defined as conscious choices that are made by language speakers. The aim of this work was to find answers to those three research questions, inspired by Spolsky's FLP factors: Therefore, in the first two research questions, the focus is on the parental strategies and decisions, whereas the third research question aims at investigating children's language preference and their motivations. In order to answer these questions, a qualitative study was conducted by collecting data from semi-structured interviews with six families in Tallinn. Research data were analysed with coding methodology (see Section 3) and the results were divided largely into three sections to answer the three questions on family language ideology, management, and practice (see Section 4).
This article is organised as follows. The introduction provides an insight to what the research is all about by providing a brief background of the issue being investigated. Section 2 presents relevant theoretical literature and previous scholarly research important for this article. Terms, such as language policy and family language policy, globalisation, identity, and bi/multilingualism are described in more detail. Earlier research conducted in Tallinn and which is relevant to this study is observed. Section 3 describes methodological background, providing an overview of the participants and data collection alongside methods used to collect and analyse the data. Section 4 is devoted to the results of this research, answering all three research questions. Section 5 constitutes a discussion of the research. Section 6 presents conclusions, where the limitations are highlighted and future research topics are proposed.

Theoretical Background
Section 2 introduces the relevant concepts for this research: language policy and FLP, the one parent one language strategy, and parental discourse strategies.
To understand the evolution of FLP, it is necessary to start with the language policy (hereafter LP). LP is something that is legislated by government, also referred to as a macro level of language policy. FLP is a rather emotive issue and is referred to as a micro level of language policy. Language policy is a large and regulated structure to meet national priorities [10]. Language planning at the national level began in the immediate post-colonial period; language policy came to deal with the aftermath of the issues that colonisation caused for its colonised territory languages. Spolsky explains that the top-down model on what language or variety of language is assigned to a specific purpose, space, or group of people in society was used to actualise the LP, but in reality it became bottom-up. That happened because "each domain within sociolinguistic ecology has its own variety of language policy, and each influences and is influenced by all the other domains" [4] Estonia's current official sociolinguistic situation is monolingual, but reality shows that it is multilingual [11]. According to the last census of population data [12], there are less than 70% of Estonians, roughly 25% Russians, and 5% of other ethnic groups living in Estonia. Although Estonia has a monolingual sociolinguistic environment, Estonia is tolerant towards foreign languages. For example, according to national curricula, Estonian students can learn as their first (A) and second (B) foreign language English, Russian, German, or French. For their second foreign language, students can choose any of the foreign languages that their school is offering. By the time of graduation, pupils should have reached at least B1 level in two foreign languages [13].

Family Language Policy
Family and its language are one of the most important domains when it comes to acquiring a language as a mother tongue. Fishman argues that the family acts as a united front against external pressures, and the language of the family is the inevitable ground for this [14] (p. 94).
Acquiring a language in early childhood is already a complex process itself; it becomes even more multiplex when there is more than one language to acquire, for example, when parents speak different mother tongues or when it is typical for a multilingual region such as Belgium, Switzerland, or India. Here rises the importance of FLP, which, as the name suggests, is the established language policy used within family. In earlier years of researching FLP, it was defined as "explicit and overt planning use of languages in relation to language use within the home and among family members" [6] (p. 907). In 2017, the definition was revised, and the two most important adjectives acquired the opposite meaning by defining FLP as "in fact implicit, covert, unarticulated, fluid, and negotiated moment by moment" [15] (p. 322). This shows that if one plans the usage of languages at ideological level, then management and practice might differ. This shows how complex FLP is and why any research conducted in this area is and will be a tremendous help to further research.
FLP becomes more and more important and distinct when discussing it at different levels; first is the macro level (LP level), then there is the meso level (includes groups, organisations, communities, political parties), and the micro level, which is individual level and includes families and relationships [16] (pp. 52-53). The micro level is considered very important as it has a crucial role on children's sociolinguistic environment [17] (p. 172). Yates and Terraschke claim that the meso level has a significant role for a minority language within society. Communities and their activity are important sources for multilingual families in cultural contexts and for preservation of their heritage languages [18] (p. 107). On the state level, LP level gains importance to the prestige of the minority language in the host country (for example, teaching it in school) and language prestige around the world [19] (p. 225).
Bernard Spolsky connects many aspects with FLP: political, social, demographical, religious, cultural, psychological, and bureaucratic, etc. He emphasises that FLP exists even when it has not been expressed clearly to the outside world [9] (p. 6) Spolsky also says that there are many factors which influence FLP. These include, for example, family structure, assimilation, parental education, language action, and cultural identity. Considering family structure, he means that in case there are older siblings in family, then they will most certainly have an effect on younger children's languages [20] (p. 433). Zhu adds that when older siblings penetrate mainstream society, then, willingly or not, they affect younger children's fluency and motivation with relation to their heritage language [21] (p. 175). There is research conducted on how older children in family contexts influence younger children's language, and more and more evidence surfaced suggesting that older siblings influence younger siblings toward using the mainstream language (see ref. [22], and references therein). However, there have been opposite outcomes in the heritage and mainstream languages when parents insist that older siblings shall speak only in a heritage language with other family members [17] (pp. 173-174). In case there are grandparents who insist to speak in their heritage language in the family structure, then the possibility to preserve heritage language is more likely as well [23,24].
Assimilation and cultural identity are closely linked. Regarding assimilation, it shows that when parents introduce children earlier to the mainstream language and culture in their sociolinguistic environment, the more prone they are to assimilate. Then, the culture starts to play a role-parents who have a strong connection to, awareness of, and pride for their heritage language could even slow down children's assimilation to mainstream language [25].
Parental education also plays a role in FLP. Although there are controversial studies in some means, for example, one study indicated that parents that are more educated are more likely to preserve their heritage language while another study suggested that parents with lower education find it harder to assimilate to a mainstream language, which means that their heritage language is better preserved for their children [26] (pp. 407-408). These two studies seem to have come to similar results, despite the different socioeconomic or educational status of the parents; this suggests different motives for preserving the heritage language. Spolsky adds that the success of FLP, i.e., the parents being able to effectively transmit the heritage language, is influenced by parents' language awareness [4]. Parental language awareness is an important part of language policy that practically influences children's language acquisition [27].
Under the language action, it is meant that parents literally need to take action to transmit a language, especially when children hit their teens, as they are becoming more independent, including their language choices [28] (p. 119). Intervening, such as enrolling children in heritage language classes, might be emotionally intensive and energy consuming but necessary if parents want to preserve their heritage languages as the minority languages [29]. This confirms the complex FLP model and illustrates how difficult it might be for parents to raise bi-or multilingual children.
Doyle states that all family members usually valorise bilingualism and family language. He also says that all family members are co-constructing the FLP. Doyle's research shows that younger children sometimes showed resistance to communicate in a non-societal language. The latter happened more when one of the parents could not understand the non-societal language (refs. [30,31] had similar results). He also brings out the fact that Estonian is a rather small language on the global scale, meaning that it provides fertile ground for larger languages to penetrate the sociolinguistic landscape thereof [32] (p. 9).
Tannenbaum discusses that the reason why bilingual families have a FLP and follow one of the possible FLPs described, for example, in ref. [33], is that parents want to keep the contact between children and heritage and to protect the unity of the family [34]. Baker says that the attitudes of people using language have the greatest effect on its maintenance or language death in society [35] (p. 2), so the parents parenting in their mother tongue is vital for keeping their heritage alive inside the (extended) family. Not only the heritage on a macro level is at stake but also the micro level; in family relationships, a common language helps to create better bonds between family members and the absence of the heritage language in some of the family's members could have a negative impact on the family wellbeing [36].

The One Parent One Language Strategy and Parental Discourse Strategies
The most popular language approach in multilingual families-the one parent one language strategy (hereafter OPOL)-has been a widely researched approach in FLP. OPOL is a strategic decision that is made due to the birth of a child in bilingual family [37] (p. 696). As the name suggests, OPOL means that parents have different mother tongues but both parents have some sort of proficiency in each other's languages. However, in refs. [33,38], the authors of which being the first linguists to use this terminology, no information concerning the competence of each parent in the language of the partner is presented. Döpke adds that one of the parent's mother tongues in the OPOL approach is the dominant language in their sociolinguistic environment [39].
Unfortunately, there is no consistency in the OPOL approach regarding management, daily communication, and children's language proficiency. Research shows that, for some children, OPOL has been very effective and they turn out to be balanced, bilingual individuals, while some children may show an unbalanced behaviour with respect to the mainstream language. They may further show signs of delay in some grammatical aspects with respect to monolingual children [6] (p. 913).
The attitude of parents towards the strategy influences greatly the success of OPOL. For example, Piller says that many parents believe that OPOL will surely lead their children to balanced bilingualism, while other parents believe that OPOL is an investment in their children's future because it increases a child's educational advantage by investing in desirable language skills. She adds that, even if considering those beliefs and attitudes, many families still fail to implement OPOL as parents rarely strictly adhere to the approach. Only a few families can stick to OPOL [40]. Lanza says that an OPOL approach is the most difficult to maintain, moreover when taking into consideration that the parents will speak with each other in one or two languages, presenting to the child that at least one of the parents has competencies in both languages [41]. Families might have difficulties implementing OPOL, as other (input) factors have to be taken into consideration when exposure to a heritage and a majority language is taking place [42].
More often, parents mix languages and bring in the expressions or/and terms from each other's mother tongues [40] (pp. 67-70). This idea means, for children's language acquisition, the emergence of translanguaging [43]. Studies have shown that the families that have the greatest success are those that strictly adhere to the OPOL, that is, where the OPOL is chosen, and where families do not share a common family language (ref. [44] and references therein).
For implementing the approach in best way possible, Lanza outlined six different parental discourse strategies and they are minimal grasp, expressed guess, adult repetition, move on, adult code-switch, and borrowing [41]. Lanza's strategies are parental strategies that are used in order to "bring the child in a monolingual mode" when code-switching takes place. Döpke described translation and modelling strategies [39]. Those strategies help to carry out approach that parents have chosen, i.e., these discourse strategies help the parents to return to OPOL (one of several FLPs), in which they behave monolingually. The following is a brief overview of these strategies with constructed samples to better understand each strategy, which is helpful in better understanding the parent's methods for implementing certain strategies in the current study.
Minimal grasp reflects the strategy where parents make believe that they did not understand what the child said, meaning that the child has used a language that is undesired by the parent [41]  Move-on strategy means that a parent ignores the child's choice of language and responds in the parent's preferable language. This results in bilingual conversation and shows that the parent agrees with the child's chosen language [41] (p. 262): Child: Ma tahan süüa (I want to eat) Parent: Allora, ti preparo qualcosa (Ok, I will cook you something) Adult code-switch refers to a strategy where parent adapts to the child's language. This shows that the parent is made to shift from their preferable language to the child's preferable language [41] (p. 262): Child: Ma tahan süüa (I want to eat) Parent: Okei, ma teen sulle midagi (Ok, I will cook you something) Borrowing reflects to a strategy where the parent includes word(s) from the language that a child has used, showing that the parent understands and partially accepts the child's chosen language [41] (p. 262): Child: Ma tahan süüa (I want to eat) Parent: Allora, ti preparo süüa (Ok, I will cook you something) Döpke proposed two more strategies, which were not listed by Lanza, but for this study are important to mention.
Translation request is when a parent requests the child to speak their preferred language, meaning that the parent does not accept other languages spoken to him/her by their child. This is more typical with children with greater language proficiency [39]. Modelling is a strategy where there are new words for the child and the child does not know how to express them or says them incorrectly [39].

Methodological Background
This qualitative research dealt with the examination of the social and personal experiences of the study participants, its description, and interpretation. The families were found via Facebook by writing a post on the Estonian expats page with the following criteria: a family should live in Tallinn, at least one child should be growing in the family, and one of the home languages should be Estonian and other language either Spanish, Italian, or French.
Due to the COVID-19 pandemic, the semi-structured interviews were conducted via Zoom. The interviews lasted an average of forty-six minutes and were recorded on the researcher's computer. Six families met the requirements: two of the families were Estonian-Italian, two Estonian-French, and two Estonian-Spanish speaking families. As the representative of the family, one of the parents was interviewed. It was not important if the interviewee was a mother or a father, as each parent knew all the answers for all the questions asked in the course of the interview.
Interviews took place on three consecutive days from 28 February to 2 March 2022. Both parties signed the consent form. Participants' anonymity and confidentiality was assured. The language of interviews was chosen in order to obtain the maximum freedom of expression and understanding of both parties and, as a result, three interviews were conducted in English and three in Estonian. The mother tongues of all recipients were familiar to an interviewer-Estonian, Italian, Spanish, and French-hence they was ready to support them if they needed to express themselves somehow in those languages.
The aim of each interview was to collect data to answer research questions that were formulated as follows: 1.
On what basis do bilingual families implement FLP on children in Tallinn? 2.
How do the bilingual families carry out or control their chosen FLP? 3.
What languages do children in bilingual Tallinn families prefer to use and why?
These questions represent three different sections: ideology, management, and practice of FLP. An interview frame was divided largely into six parts, comprising: introduction, three main blocks of questions, and the ending part. The cultural background data block focused on the family's linguistic proficiency and cultural history. The colloquial arrangement block gathered data regarding the family's everyday life arrangements in their sociolinguistic environment. The last main part focused on children's agency where questions regarding children's multilingual experiences in different sociolinguistic environments, relationships with grandparents, reading habits, the schools/kindergarten they are going to, etc. were asked (i.e., "In what language do children communicate with their grandparents?", "What books and in what language do your children read?", etc.). The aim for those questions was to gain data in order to support family language management and practice given by the parents in earlier blocks.
After conducting the interviews, transcripts for all of them were created manually, meaning each word and sound was written down, and each pause was marked. Phenomenological approach to data analysis was used. It "focuses on the conscious experiences of how a person relates to the world, and with such focus, it moves away from traditional empirical research and toward inclusion of the subject group members' perspective, as they are best positioned as the expert on their own life and life experiences" [45] (p. 580). The aim of this study was to understand ideological, practical, and managemental issues of FLP within families: participants had to share their personal views and experiences about raising their children bilingually. Coding was performed manually: ideological, practical, and managemental topics had different colours for reducing, relating, and conceptualising the data; those three categories had smaller sections within, which were separately marked to differentiate data.

Study Participants
The aim of this research was to study language practices among Italian, Spanish, and French speaking multilingual families living in Tallinn, and to find if there was a pattern on how those families apply OPOL and what the motivations are for that. For this reason, one of the criteria was that one parent's mother tongue must be Estonian and the other parent's is either Italian, Spanish, or French, which are all minoritised languages in the territory of origin. The interviewer has a bachelor's degree in Roman languages and cultures. Thus, during the interviews, when someone could not express their thoughts in English or Estonian as they would like, they could use their native language to do so. Four mothers and two fathers had Estonian as their mother tongue. There was a criterion for the couple to have a child growing in their family. What follows are brief summaries of the families interviewed (see also Table 1). Family 1 (F1) is an Estonian-Italian family. The mother was 38 years old, and the father was 50 years old at the time of interviewing. The mother was born and raised in Estonia and the father in Italy. The mother had vocational education and the father had unfinished higher education. The mother was a stay-at-home mom. They met five years ago in Italy and have one child together, who was three years old at the time of interviewing. The child has a dual citizenship. They are currently living in Tallinn, but the father owns a restaurant in Spain, and they spend a lot of time there as well. The father speaks only Italian, so the mother learned to speak Italian and their home language is Italian. The mother speaks Estonian with the child, and the child prefers speaking Estonian as well as they have mostly spent the last years in Estonia due to COVID-19. The child attends an Estonian kindergarten.

Family 2
Family 2 (F2) is an Estonian-Mexican family. The place of residence of the F2 was first Mexico where their child had spent almost all his life, before Estonia. The mother and the father both were 32 years old at the time of interviewing. The father was born and raised in Mexico and his mother tongue is Spanish. The mother was born and raised in Estonia and her mother tongue is Estonian. The father has a master's degree and works in Tallinn as an office manager; the mother has a doctoral degree and works as a scientist from home. They met 10 years ago in the Czech Republic and started speaking with each other in English, but today the mother speaks Spanish at B1 level and the father speaks Estonian at C1 level. They have one child together, who was two and a half years old at the time of interviewing. The child has a dual citizenship. Parents speak with each other English and both parents speak in their mother tongue with their child. The child prefers to speak Estonian. The child attends an Estonian kindergarten.

Family 3
Family 3 (F3) is an Estonian-Venezuelan family. The mother was 35 and the father was 32 years old at the time of interviewing. The father was born in Estonia and his mother tongue is Estonian. The mother was born and raised in Venezuela and her mother tongue is Spanish. The mother has a bachelor's degree and works in IT but she is currently on maternity leave. The father has a master's degree and works as a lawyer. They have been together for four years and have one 10-month-old child and one on the way. The child only has Estonian citizenship; they would have liked to have him obtain Venezuelan citizenship as well, but at the time of the interview, it was stuck in the bureaucracy. Parents speak in English with each other and use their mother tongues with their child. The child began to speak his first words in Spanish, but understands some words in Estonian as well.

Family 4
Family 4 (F4) is an Estonian-French family. The mother was 37 years old, and the father was 34 years old at the time of interviewing. The father was born in Estonia and the mother was born and raised in France. The mother has a master's degree and the father has a vocational degree. The mother works as a digital seller but is currently on maternity leave and the father is currently studying. They have been together for 10 years and speak English with each other. They have two children; older is 3.5 years old and younger is 6 months old. The children have dual citizenships. Both parents speak their mother tongue with their children. The older child speaks French and Estonian equally and does not have a preference of language between those two.

Family 5
Family 5 (F5) is an Estonian-French family. The mother was 39 years old, and the father was 35 years old at the time of interviewing. The mother was born in Estonia and father was born and raised in France. Both parents hold a master's degree. The father works in IT and the mother is self-employed. They have been together for 13 years and speak English with each other. They have two children: one of them is 9 years old and the other is 5 years old. The children have dual citizenships, and both parents speak their mother tongue with their children. The children prefer to speak in Estonian. The older child can speak French well, but answers in Estonian to her father.

Family 6
Family 6 (F6) is an Estonian-Italian family who lived in Italy before moving to Estonia. The mother was 42 years old, and the father was 47 years old at the time of interviewing. The mother was born and raised in Estonia and the father in Italy. The mother has a master's degree and the father has an unfinished bachelor's degree. The mother has not been working since the children were born and the father works in Italy as a human resource manager. They have been living together 10 years and speak Italian with each other. F6 have two children: one of them is 8 years old and other is 4 years old. The children have dual citizenships and speak both languages fluently. As F1, they have chosen the heritage language as their home language, which has "boosted" the language proficiency in this language. This is also what other studies have found, for example, ref. [31].

Findings
First, results on family language ideology are presented. Next, results on family language management are given, and lastly, the results on family language practice are described.

Family Language Ideology
The first research question concerned language ideology and was structured as follows: on what basis do multilingual families implement FLP in Tallinn? To answer this question, the next aspects were investigated more closely: (a) what kind of communication strategy, i.e., FLP, they have chosen and (b) why they have chosen this strategy. The interviewer asked them to describe the parenting discourse strategies parents use in simple terms.
All six families made a conscious choice before their first child was born that they are going to be a multilingual family, supporting Spolsky's argument [4] (p. 1). Five out of six parents' ideological decisions were practicing OPOL. Those families did research before the birth of their children and decided that OPOL would be best in their linguistic environment. F3 mentioned, more than once, success stories within her own social network of raising a bilingual child with the OPOL approach. Another decision that parents have to make consciously or not is the choice of the language they speak to each other. F3, F4, and F5 spoke in English before their children were born and continued to do so afterwards. In two families (F1, F6), mothers were Estonians and fathers were Italians. Both Estonian/Italian parents in this research opted to speak with each other in Italian, unlike other families, who opted for English. F1M did not speak Italian before she met her husband, but she learned to speak it for the relationship, though she continues to regard English as the language with the greatest advantages. F6M mother spoke Italian already before she met her husband and they continued to speak in Italian. According to F1F and F6F wives, the Italian fathers' English was not good and their Estonian was non-existent at the time of interviewing. One family, F2, decided to switch to their own language after the child was born. Although the father is Mexican, and his mother tongue is Spanish, he speaks fluent Estonian.
Two families (F1, F2) mentioned the importance of culture; moreover, a language being part of their culture and that continuing the language helps to preserve their heritage, supporting Yates's and Terraschke's arguments [18]. To conclude, their ideology is that the child must learn both parents' mother tongues; however, F1, F2, and F3 had more confined and firm goals for the languages their children will speak in the future (parents noted the need to learn English, etc.). For example, F1 intentionally showed the child cartoons only in English so that the child would learn a new language. F4, F5, and F6 had no specific expectations, but they were quite sure that the children would speak at least two languages. All the parents had been guided mostly by their intuition, the willingness to speak in their mother tongue with their children, and to share jokes and feel openly within home environment. Parents said that the choice of languages was discussed, but it came very naturally. Many parents brought up the usefulness of the languages, how they are going to acquire them, and how they will help them in the future. For example, the difficulty of Estonian and the advantage of learning it as a native language were mentioned several times.

Family Language Management
The second question concerning family language management was set as follows: how do the bilingual families carry out or control chosen FLP? Families' reading habits were studied, attention was paid to languages they were watching TV programs in, how they chose schools and kindergartens, and general conscious language guidance within the family.
All the participants' children who were at least 2 years old were currently studying in an Estonian school or kindergarten. F6's eldest child tried an Italian school a couple of years ago, but it did not work well for them, so they wanted to come and study in Estonia. F5, which is a French-Estonian family, had an opportunity to give their eldest child an education in a school known for a high level of French teaching, but all the other subjects, beside foreign languages, are taught in Estonian. The child already had an advantage in school, as she has known French from birth. The eldest child in F6 tried an Italian school while they were living in Italy and the child did not like it because, according to her mother, the school was quite strict, and atmosphere was chilly; the mother thought that it was not an environment where she wants her children to study. For the previous experience in Italy, she was specifically searching for a smaller, different, yet still Estonian school to support her children's needs for a more personal approach and help them to adapt better.
Audio-visual media and printed media in acquiring languages plays an important role. For young children as those in this study (6 months to 9 years old), parents read for them, bought them books, or at least told them what books to read. Only F1 family, who's child was 3 at the time of the interview, did not read to their child, but the parents of the mother (F1M) read Estonian books to their grandson. F2, F3, F4, and F5 parents found reading to their children in both parent mother tongues important. F6M said that now her child reads only in Estonian, because he has to read at school already and is not motivated to read more than required. F2, F3, and F4 children's ages are between 6 months and 3.5 years and they are going through the most intense language-learning phase [46] and, consciously or not, most parents of this research that had children less than 4 years old found it very important to read in both languages. Children who were mentioned by parents of F5 and F6 both go to school, meaning that they had obligatory literature already and parents did not want to overwhelm them with extra-curricular literature activities unless they opted to read light literature themselves. Parents of F5 decided to share the literature they read, and both parents read in their mother tongue daily. Regarding television and screen time, all the parents let them have it. In families in which parents speak English between themselves, regarding cartoons enters a third language-English. F5 and F6 focused more on one parent's mother tongue. For F5, it appears to be more of a decision of convenience, as Estonian channels show cartoons in the mornings around the time children are waking up and preparing to go to school. F6F works in Italy and therefore spends a lot of time away from home, and for this reason, F6M lets their children to watch cartoons more in Italian.
Although F3 parents have agreed to use the OPOL strategy, both parents would like to introduce their own mother tongue through their spouse as well. F3 parents are using a translation strategy diligently with their child. As the parents explained, it is their first child and the child had recently started to say their first words, so understandably both of them prefer to hear their own mother tongue. F4 also opted for the translation strategy, not only with the child, but with each other as well. A representative of a French-Estonian family in which spouses use English among themselves and try to stick to OPOL as much as possible, F4M also said in her interview that for her it is "frustrating that I don't understand everything, but this is me personally, who just likes to understand everything". F6M knows "many women" from her own social circle which were not the examples she would have like to follow, and moreover were the opposite of her goal, so for this reason she has been very strict with using only her mother tongue with her children. The move-on strategy was used by F5F; the father uses the move-on strategy because he thinks that if he were to use the translation strategy, his daughter would not answer to him at all.
To summarise, all the participants took into consideration the Estonian societal majority language and the fact that one of the parents speaks the majority language as a mother tongue. It concluded by placing their children into Estonian kindergartens and schools, not international schools, which could have been one of the options as well. To add, there are no schools in Estonia where all the subjects would be taught in Italian, French, or Spanish. In terms of guiding children's reading, five families considered it important, and three of the families rated it very highly. Each parent read in their mother tongue to their children. Concerning screen time, in many participant families, with cartoons and movies, English as a third language was used. Some families (F2, F3, F4, and F6) had English language cartoons additionally to cartoons in the parents' mother tongues. F5 showed a short amount of cartoons daily and in Estonian. When it came to parental discourse strategies, participants mainly used the translation strategy, meaning that the parent did not accept other languages spoken to him/her by child; this is more typical with children with greater language proficiency, as the participants' children here have.

Family Language Practice
The third research question concerning family language practices was posed as follows: what languages do children in bilingual Tallinn families prefer to use and why? In F1, at the time the family was living in Italy, although the mother had spoken only in Estonian and spent more time with the child during his infancy period, the father's mother tongue and home language, between spouses, had been Italian and the child picked up the Italian first. After moving to Estonia, the child went to an Estonian kindergarten and switched to Estonian and at the time of interviewing preferred speaking in Estonian, although the family language and OPOL strategies were kept the same. In F2, one child started to speak in Spanish because, additionally to the F2F, had grandparents in Mexico who spoke in Spanish to him. In these early ages, FLP and input play an important role. The mother spoke with the child in Estonian, but he still started to speak in Spanish at first. When they moved to Estonia, he preferred to speak in Estonian. Even now when F2F speaks to him in Spanish, the child often responds in Estonian, because he knows that the father can understand him. The father has competences in the community language, Estonian in our case, where the child is growing up. It has already been observed that competences in the languages spoken at home (particularly when this language is the heritage language) promote language competence [31]. In F3, the child was only 10 months old by the time of interviewing and he spoke one Estonian word and two Spanish words-mama and papa-so it is too early to say which language F3's child prefers. F3's parents had been together for 4 years and lived this whole time in Estonia, so F3M, who is originally from Venezuela, has started to speak some Estonian with her husband. They may encounter the same problem as the F2 family when the children begin to favor the majority language while the parents speak English among themselves.
In F4, where parents exclusively use the OPOL approach, the older child (the younger was 6 months old at the time of interviewing) started to speak both languages at the same time. The mother does not speak Estonian at all; the father attended French courses and tries to speak French at home. However, as the mother says, Estonian surrounds her every day. Interestingly, the child started to speak both languages equally; the mother says that the child did not prefer one language to another. F4M claims that the child's diction is better than her peers are and thinks it is like that because she must make a greater effort to make herself understood when she speaks. Although the mother claimed that F4F and F4M speak in English with each other, it appears that French is penetrating this communication increasingly, while the child speaks in and listens to Estonian during the day and French later at home; they might have found a very good balance. F4M adds that the child has learned some English simultaneously when as she has spoken with her spouse in this language. This statement also refers to equal preference of Estonian and French languages, while adding a notion of understanding of a third language, which is English. The child has not only acquired one more language by listening to another family member, but the mother says that she also understands Estonian now as she has listened to her child and husband speak with each other in Estonian.
The children in F5 started to speak very differently: the first child started to speak at 1 year old and the second at 4 years old. Parents believe that the second child simply lacked individual attention and maybe courage to speak with his peers, which he has now overcome. In F5, there have been few changes in language use over the ten years of life together between spouses. F5F kept using English, but added some phrases in French, which he found useful or hard to translate, so sometimes he uses it with his children and wife. F5M uses either English or Estonian. The children use mostly Estonian, but eldest child (9 years old) understands French as well. The children are not encouraged to speak in French, as the father uses the move-on strategy with them, meaning that he accepts when children are answering in Estonian to him. The eldest child can speak French and understands it, but the younger child speaks only Estonian and understands some English as the parents speak in English at home. F5F described a situation when they spent Christmas in France with the father's relatives the and child spoke in Estonian with everyone. This shows that even when the child did not speak French like other children, he still managed to play with them in Estonian, which may not be motivating him to speak in French as well.
The children in F6 first started speaking Estonian, even though they were living in Italy at the time, and their mother was the only one who spoke Estonian to them. Thus, quantity of input and exposure in the early years is very important for language acquisition (see also ref. [47]). F6M says that the reason for their starting to speak Estonian is because mothers are "cuddling and snuggling" more with babies. Only one child occasionally brings in words from his Italian vocabulary to express himself better.

Discussion
The research participants took into consideration the societal majority language, Estonian. Many families emphasised that children should know the language of the country they are living in and when one of the parents speaks the language as well, then it is a good opportunity to develop the skills of Estonian. Only one family had a child who was not yet placed in any educational institution, and this was so because the child was 10 months old, and the mother was staying at home with her. All other families placed their children into Estonian day care, kindergartens, or schools. Yates and Terraschke noted that communities and their activities are an important resource for multilingual families in a cultural context and for preserving their heritage language [18] (p. 107). In this research, children are placed in Estonian schools, which contribute to preserving one parent's mother tongue that is actually the societal language. Yates and Terraschke believe the meso level that, for example, focuses on the sociocultural institutions where learning may take place (schools), will be the most useful for the preservation of minority languages. One French-Estonian family of the research (F5) had the opportunity to place their eldest child into an Estonian school with a French incline. Palviainen and Boyd said that the prestige of the minority language in a host country, which French, in Estonia, has, is important when acquiring it in school [19] (p. 225). Tran said that regarding assimilation, when parents introduce children earlier to a mainstream language and culture in their sociolinguistic environment, they are more prone to assimilate. Then the culture starts to play a role-parents who have strong connections to, awareness of, and pride for their heritage language could even slow down children's assimilation to a mainstream language [25].
On a practical note, families brought out the importance of knowing many languages and that being able to communicate with more people confers a greater advantage in the job market. All participants appreciated the emotional connection, the opportunity to speak their native language, and to continue their family heritage, which is in line with ref. [39]. Tables 2 and 3 summarise language knowledge and use inside families F1-F6.  Five families valued reading, but three families (F2, F3, and F5) out of those five stressed the importance of reading very highly. Interestingly, one family (F1) who did not read to their child had a child of 3 years old (by the time of interviewing) and the child started to speak rather late. The F1 child's grandparents read to him in Estonian. Luo and Wiseman said that in case grandparents insist to speak in their heritage language in the family structure, then the possibility to preserve the heritage language is more likely [23].
Cantone found similar results for four extended families she investigated [24]. This means that grandparenting and its effectiveness should not be underestimated. Cummins stressed the importance of the parental education in FLP by saying that parents that are more educated are more likely to preserve their heritage language [4] (pp. 407-408) while Spolsky added that the success of FLP is influenced by parents' language awareness [4]. In this study, F1M had a vocational degree and F1F an unfinished higher degree, while F2F had a master's degree and F2M a PhD. After reviewing the observations from this study, we can conclude that the first statement is not relevant, but parents' language awareness is an important part of FLP that practically influences children's language acquisition.
In conclusion, regarding family language practice, parents who described that their children started to either speak early or spoke both mother tongues by the age of two also valued reading a lot. According to this study, regular reading to children by both parents with different mother tongues could contribute to earlier bilingualism (cf. [48]).

Conclusions
In a semi-structured interview via zoom, three main aspects of raising children bilingually were investigated: language ideology, language management, and language practice. A tendency towards sticking to OPOL was observed as much as possible, while some families opted for a family language (either the heritage or the majority language) and a language that is spoken exclusively between the parents. When it comes to the discourse strategies followed by the parents in order for them to remain monolingual in the parent-tochild interactions, they mostly opted for the translation strategy. Finally, reading practices were not equally carried out in all families. Some families used reading activities most frequently than others.
This study contributes to the body of literature regarding FLP among bilingual families in Estonia. It helps communities understand the needs and nature of bilingual families of different ethnic backgrounds to ensure they integrate better and smoother into society. These results also enrich research on globalisation and multiculturalism, since the abovementioned ethnicities have never been studied in an Estonian setting.
This study has certain limitations: it is based on the views of only six participants, which makes measurement more difficult. In addition, participants may provide responses that the researcher is waiting for them to give or present themselves in a way that makes them look better [49]. Still, the study is important for the expats living in Tallinn, as there is a large community of expats in Estonia. Researching FLP patterns among multilingual families who live in Tallinn will influence current and future expatriates. This article contributes to the FLP field in overall as it is still an understudied sociolinguistic concept and any new research helps to explore it more. The topic is important for locals as well, because the result might contribute to creating a friendlier and more welcoming environment for expatriates living in Estonia.
Finally, yet importantly, the most important limitation of this study is that its sample is small and unequal. For future research, it will be important to collect a larger sample, grouping the participants, taking into account some more variables that are important in a study of this type, including the language of the interviewee, homogeneity in the age of the children, which parent spends more time with the children, and the level of education of the parents, etc. In other words, more factors and variables that can condition (bilingual) language transmission and that have been outlined in the theoretical framework should be taken into account. Furthermore, future research could compare the results obtained in the interviews in terms of the OPOL strategies used with the children's level of heritage language.
Funding: This research received no external funding.

Institutional Review Board Statement:
The study did not require ethical approval because of national laws and non-existence of such practices as IRB.
Informed Consent Statement: The written informed consent was obtained from all respondents before participating in this study.

Data Availability Statement:
The data presented in this study are available on reasonable request from the corresponding authors.