Studying Families’ and Teachers’ Multilingual Practices and Ideologies in Kindergartens: A Nexus Analytic Approach

This chapter discusses the methodological opportunities of studying multilingual practices in kindergarten through a nexus analytic approach (cf. Scollon & Scollon, 2004). It is based on an ethnographic study with fieldwork in two kin-dergartens in Norway and Germany. The first part of the chapter elaborates on how the nexus analytic approach made it possible to gain insight into the kindergarten teachers’ and parents’ views on multilingualism. As a theoretical background, I draw on views on multilingualism and language ideology theory. The second part of the chapter discusses, which insights the applied nexus analytic approach may contribute with to the field of childhood studies, based on the concept of intersectionality (Alanen, 2016) and generational order (Alanen, 2009, 2016). I argue that the nexus analysis in this study contributes with several interesting perspectives. First, it provides insights into the intersectionality of multilingual children’s lives by shedding light on the complexity of intersections of linguistic practices. Second, the analysis sheds light on the relevance of various generational categories as part of these inter-sections. A question deriving from this complexity is which forms of linguistic practice may be in the best interest of multilingual children (cf. James & James, 2008). Here, my study revealed several contrasts between parents’ and teachers’ views. With reference to article 12 (UNCRC, 1989), which emphasises the importance of listening to children’s voices, I argue that this may challenge both researchers, teachers and parents to listen to young multilingual children’s voices, especially as these children are little represented in research.


Vignette 1
It is in the middle of the day at the Sunflower kindergarten when I talk to Helena about school and cooperation with parents.Helena is a kindergarten teacher and is responsible for the preschool club in the kindergarten.She tells me that there are parents who sit down with their children and train them to write the alphabet, and that she often tries to convince them that this is not necessary.Still, it does not seem as though they always listen to her. "And it is like this", she says, "school is very important, kindergarten is not"."Oh", I ask, "is it?""Yes", she says, telling me about one girl, Finja, the older sister of one of the boys in Helena's department, who attended the Sunflower kindergarten before she started school last summer.Helena recounts that Finja always joined in telling and writing stories and that she used to enjoy this activity a lot.Telling and writing stories together with the children was a common linguistic practice in this kindergarten department.Helena tells me that Finja joined in storytelling less and less and says, "and I wondered whether I had done something stupid in some way.And then I noticed at some point that it was the letters".The girl had to write the alphabet at home.She obviously had learnt it in both Russian and German before she started school.1And in the end, she did not take part in story-telling anymore, only when her friend joined.Now Finja attends first grade, and Helena refers to the fact that she asked her mother, "How is she doing at school?" "She is bored", the mother tells her, "she keeps asking when they finally are going to do something proper"."Because they all are busy learning to read and write", I say."Yes", says Helena, "that's what I told her during the whole last kindergarten year" (Field note, Sunflower kindergarten, October 2015) Vignette 2 [And I] think that she [Finja] learnt to read and write German that quickly, because she already started with Russian lessons one year before she entered school.She still takes them.
[…] And she attends lessons once a week.And there she also learnt to write and read Russian quickly.
And since she managed that, she also could read in German quite soon.And also write.[She started] when she was close to six.One year before starting school, exactly.I did not want her to start at the same time as starting school, because I thought it might get a bit complicated, both at the same time.That this would maybe demand too much of her, and then maybe something would not work out, and then she would neither get something good out of school nor out of the Russian lessons.And then I thought, okay, one year earlier.She could, she wanted to, she always asked me, because her friend attended Russian lessons already, and then she said "I want, too" and "I also want to be able to read that".I thought, okay, we will try.We tried, she is happy, we, I am also happy.
[…] And I also think that [when] you have this language as your mother tongue, why not be able to write and read it?That is an advantage, I think.(Excerpt from interview with Finja's mother Susanne, February 2016) The two texts presented above are excerpts from the data collected for my PhD study (Pesch, 2017).Both excerpts are centred on Finja, who was a first grader at the time I collected the data but had attended the Sunflower kindergarten, where I carried out part of my fieldwork.I chose these excerpts to begin this chapter because they point to several contrasts between the kindergarten teacher's view and that of Finja's mother regarding language practices, multilingual language development and formal or non-formal language education.In my dissertation, I discuss these contrasts in relation to multilingualism, including how views on multilingualism create discursive conditions for linguistic practice with multilingual children and influence cooperation between kindergarten teachers and parents.The first aim of this chapter is to elaborate on how the nexus analytic approach developed by Scollon and Scollon (2004) helped me gain insight into the teachers' and parents' language ideologies and views on multilingualism.The second aim is to discuss in which way these insights may contribute to the field of childhood studies, drawing upon the concepts of intersectionality (Alanen, 2016), generation and generational order (Alanen, 2009(Alanen, , 2016;;Honig, 2009), transformation (James, 2009) and the best interests of the child (James & James, 2008).The question of the best interests of the child is connected to the UN Convention on the Rights of the Child (in the following: UNCRC) (United Nations, 1989), and has in Norway and the other Nordic countries often been discussed related to juridical issues, as violence and assault, divorce, adoption and taking children into care (Adolphsen et al., 2019).The topic of this chapter does not involve juridical considerations, and hence the question of the best interests of the child is treated slightly different.The discussion is based on article 12 of the UNCRC (United Nations, 1989) and evolves around the importance of listening to young multilingual children's perspectives on multilingualism.

Theoretical and Methodological Background
The methodological opportunities and challenges of studying multilingual practices in kindergarten with a nexus analytic approach are starting points for this chapter.For my PhD, I conducted ethnographic fieldwork in two kindergartens, Sunflower in Northern Germany and Globeflower in Northern Norway, both of which were public and located in medium-sized cities.The data are organised into two cases, and the study employed both a case study (see Yin, 2014) and ethnography (see Gulløv & Højlund, 2010;O'Reilly, 2012) as methodological approaches.2The data include interviews with the teachers and parents of multilingual children, pictures of the kindergartens' semiotic landscapes, field notes about teachers' linguistic practices and relevant policy documents.For this chapter, I draw upon the data I gathered from Sunflower.
Although this chapter is based on the data as a whole, the field notes and interviews are of particular interest.I first introduce the theoretical framework and then present nexus analysis as an analytical approach.My PhD study draws upon theory from the fields of early childhood education and sociolinguistics.It adopts a socio-epistemological view of kindergarten, proposed by Ødegaard and Krüger (2012), in which kindergarten is understood as a social and cultural arena where people (i.e.children, teachers and parents) with various agendas, aims, views and desires meet.One important aspect is that these 'actors' -as Ødegaard and Krüger (2012, p. 28) refer to them -have different roles in the kindergarten context.The authors emphasise the relevance of talking to these actors to gain insight into their implicit views and understandings as well as observing their practices to understand what they actually do (Ødegaard & Krüger, 2012, p. 28).A related debate in the field of language ideology about whether language ideology can or should be studied through observation of linguistic practices in addition to other methods (Kroskrity, 2004;Woolard, 1998) forms part of the theoretical background for my study.In my study, I gained insight into the language ideologies of parents through discursive reading of interviews about their linguistic practice with multilingual children and into the ideologies of kindergarten teachers through discursive reading of observations and interviews.I found that the different views on multilingualism between teachers and parents were connected to their different language ideologies as well as different choices, aims and agendas (cf.Ødegaard & Krüger, 2012), which framed the conditions in which multilingual children developed their language practices.Here, I focus on the differences between teachers' and parents' views.
Another important theoretical aspect of the socio-epistemological framework is that kindergarten is understood as a social, cultural, historical and political field with different practices that create discursive conditions for learning, formation and development (Ødegaard & Krüger, 2012, p. 20).The present study focuses on the discursive conditions for linguistic practice with multilingual children, a main theoretical part of which are views on multilingualism and norms of linguistic behaviour (Garcia & Li Wei, 2014;Jørgensen, 2008).Garcia andLi Wei (2014) andJørgensen (2008) refer to different ways of understanding multilingualism and how these connect to different views on multilinguals' linguistic practices.Both authors discuss traditional views of multilingualism as a form of double monolingualism or an additive view of multilingualism and recent views of multilingualism as integrated linguistic repertoires with features from several languages (Garcia & Wei, 2014;Jørgensen, 2008).In addition, both emphasise and argue for a distinction between multilingualism and polylingualism (Jørgensen, 2008, p. 169) or translanguaging (Garcia & Li Wei, 2014, p. 13), which are complex linguistic practices in which individuals draw upon various linguistic features in a communication context and the question of which language these features belong to becomes immaterial.3The different views of multilingualism are illustrated below.Translanguaging (Garcia & Li Wei, 2014, p. 14) Behind these distinctions lies the ideological question of what counts as language (Woolard, 1998, p. 16) and whether languages are relatively solid systems or more fluid and dynamic constructions.In my data, language ideologies are expressed through teachers' and parents' views on multilingualism as well as through the choice to use particular languages and the values attached to them (see Jaffe, 2009) or a focus on language separation on the one hand and Pesch translanguaging practices on the other (see Garcia & Li Wei, 2014;Riley, 2011).In this chapter, I use the term multilingual to refer to children who grow up speaking more than one language because one or both of their parents' mother tongue is different from the country's majority language and used with the child.The term mother tongue is complex and has various definitions (Sollid, 2014;Øzerk, 2016).I choose to use it here because it best covers the dynamic nature and changeability of the participants' multilingualism.

2.1
Nexus Analysis Nexus analysis is a type of discourse analysis that uses human action as a starting point (Scollon & Scollon, 2004, p. 64).It draws upon theories from different linguistic and anthropological fields as well as critical discourse analysis (Hult, 2017;Lane, 2014;Scollon & Scollon, 2004).Since sociocultural theory is an important theoretical background for nexus analysis, action is always regarded as social and mediated (Lane, 2014, p. 2;Scollon & Scollon, 2004, p. 12).However, action is not connected to a particular group, which distinguishes the concept of a nexus of practice from a community of practice (Lane, 2014, p. 6).This also becomes evident in the connection to ethnography; Scollon and Scollon (2004) point out that nexus analysis adopts ethnography not only as a research approach but also as a theoretical position: A nexus analysis is a form of ethnography that takes social action as the theoretical center of study, not any a priori social group, class, tribe, or culture.In this it departs to a considerable extent from traditional ethnography in anthropology or sociology.(Scollon & Scollon, 2004, p. 13, italics in original document) This distinction -that nexus analysis studies social action and not a group of people -is important, not least as an ethical consideration when presenting the findings, and relates to Ødegaard and Krüger's (2012) view of teachers, children and parents as actors.Discourses may seem personal when they are revealed in interviews or actions, but they are always connected to one's role and aim within the kindergarten context.Teachers and parents are not studied as people, but as actors with linguistic practices in relation to the discursive frames they experience.In this context, it is important to identify participants' motives for action, not objectively but in relation to relevant discourses ( Scollon & Scollon, 2004, p. 11).

2.2
Central Terms and Concepts Nexus analysis depends upon three terms action, practice and the nexus of practice (Scollon & Scollon, 2004, p. 62f.).5Both Scollon and Scollon (2004) and Lane (2014) point to time and repetition as important differences between action and practice.Similarly, I understand a practice as an action that has been established over time.Moreover, a practice is a nexus of trajectories of participants, places and cycles of discourses.These trajectories and discourses intersect and enable action, and an action or practice may alter these discourses or trajectories (cf.Scollon & Scollon, 2004, pp. 28, 159).Cycles of discourse are related to three key factors that intersect in social action: participants' historical bodies, the interaction order and the discourses in place (Hult, 2017, p. 94;Lane, 2014, pp. 7-8;Scollon & Scollon, 2004, p. 19-20).
Hult (2017) argues that, due to its integration of principles from different research traditions, nexus analysis makes it possible to focus on three complementary scales: (inter)personal, community and societal scales.The historical body is about a personal scale and involves beliefs that are related to an action and based on one's experiences through education and socialisation as well as the beliefs of earlier generations passed on through an individual's language socialisation (Hult, 2017, p. 94).It also includes the possibility for individuals to influence society.Interaction order refers to the typical patterns of interaction between participants that occur during an action or practice at a particular location and time (Scollon & Scollon, 2004), and it takes place mainly at the interpersonal scale (Hult, 2017, p. 95).Important aspects of the interaction order include individuals' social positions, their expectations for each other and the possibility of developing certain kinds of interactions during encounters between individuals (Hult, 2017).Interaction orders often relate to norms and expectations that have developed over time, and to understand why an interaction order works as it does, it is important to map its sociohistorical evolution (Hult, 2017, p. 96).Thus, one could argue that the interaction order also involves the community and societal scales.
Discourses in place are connected to particular places.Even though they become relevant for an action at a particular moment in time, they also cycle on wider community and societal scales (Hult, 2017, p. 97).Some discourses are more foregrounded and thus more relevant for a particular action or practice.One main aim of nexus analysis is to find the foregrounded discourses within the studied practice, and one challenge is that some discourses are so implicit that they may be difficult to find (Lane, 2014, p. 8;Scollon & Scollon, 2004, p. 14).Hult (2017, p. 93) suggests searching for joint values, attitudes, stances and ideologies to which certain actions relate.
Another main aim related to the three different scales is to connect discourses at the (inter)personal and local levels to discourses at different macro levels.This is possible due to a twofold understanding of discourse in nexus analysis.As Scollon and Scollon (2004, p. 2) point out, in the simplest sense, discourse can be understood as 'the use of language in social interaction' .In addition, they make use of Gee's (1999) understanding of discourse as a connection between linguistic and non-linguistic elements, including emotions, values, symbols and artefacts (Scollon & Scollon, 2004, p. 4).This understanding implies that we attach values to material and non-material elements and view some as more valuable than others.In that respect, discourse involves power.While the first understanding of discourse is connected to action at the micro level, the second understanding includes the development of discourses over time and within society, groups or institutions.Still, it is important to note that the distinction between the micro, meso and macro levels might only be applicable to analysis, and in reality, discourses are interconnected and not necessarily found on only one level.

Reading the Excerpts from a Nexus Analytic Perspective
The excerpts presented in the introduction to this chapter are two different descriptions of Finja's participation in and enjoyment of non-formal activities at a kindergarten and formal Russian lessons that she attended in the afternoon.Helena and Susanne almost seem to be describing two different children.Looking at these descriptions from a nexus analytic perspective might not give insight into Finja's actual experience, but by viewing the choices made by Helena and Susanne as mediated actions, one can ask questions about the discourses connected to them.Helena's description of Finja's decreasing participation in story-telling highlights to a discourse in my data from Sunflower regarding the relevance of exposing children to literature, encouraging them to create stories and helping them to understand how to construct a story to support their language development.Another discourse in place circulating through this practice is connected to the child-centred approach at Sunflower and its opposition to the common pedagogical approach of learning through memorisation.For Helena, the joy children feel while telling stories is important, as evidenced by her criticism of Susanne and other parents who train their children to memorise and write the alphabet before they enter school.In the excerpt, these discourses seem to be situated mostly in Helena's historical body, which is connected to her professionalism and experience as a kindergarten teacher.But, as mentioned above, they also are connected to discourses in place at Sunflower.Regarding interaction orders, the excerpt also highlights the expectations that, as an expert, Helena's advice should be heeded.In the last sentence of the excerpt, for example, after Susanne mentions that Finja is bored at school, she points to her recommendations the previous year.Many of Helena's colleagues also believe in this interaction order (i.e.kindergarten teachers are the experts, not parents).Thus, in many respects, it is a general interaction order that exists at Sunflower regarding linguistic practice and preparation for school.It also directly references the kindergarten as a context of interaction (Ødegaard & Krüger, 2012), in which studies on cooperation have identified a field of tension between professional roles and equal partnerships between kindergarten teachers and parents (Alasuutari, 2010;Einarsdottir & Jonsdottir, 2018;Kultti & Samuelsson, 2016).6 The excerpt of the interview with Susanne confirms Helena's statement that parents do not always take her advice or the advice of teachers in general.Susanne might not share the expectation that she is less of an expert, partially due to the different discourses affecting her choices regarding Finja's language development.For example, at the end of the interview excerpt, she emphasises the advantage of being able to read and write in one's mother tongue.In addition, she considers it to be too complicated for children to start reading and writing in two different languages simultaneously.Thus, she chooses to let Finja start learning Russian before she enters school, related to a discourse regarding the importance of formal schooling.At the beginning of the excerpt, she supports her choice, saying that her daughter learnt to read and write in German so quickly because the groundwork was laid during her Russian lessons.In line with Garcia and Li Wei (2014), this refers to a view of multilingualism as dual and languages as interdependent (see Cummins, 2000).From a language ideological perspective, she emphasises the importance of separating languages (Riley, 2011), but with the underlying idea that children benefit from certain competences in all their languages, even though they are acquired in one language first.Both Susanne's emphasis on formal schooling and view on multilingualism contrast the kindergarten's practice and discourses.
It is important to note that, during the interview, Susanne voices some concern about her daughter's competence in German when she started school, which seems to be an important aspect of her choice regarding formal schooling.She also states that, in hindsight, her concerns probably were exaggerated, adding an interesting aspect to the expectations regarding roles in the interaction order.To a certain degree, Susanne now confirms Helena's role as an expert.Concerning motives (Scollon & Scollon, 2004), both Helena and Susanne refer to their motive to do what is in Finja's best interest (cf.James & James, 2008).Since this is related to their discourses on multilingual children's language development and education, their choices -and evaluations of these choices -are quite different.

Cycles of Discourses at Different Levels
My data contain several references to connections between discourses at the inter-or intrapersonal level and the macro level.One such connection is revealed in the excerpts with regard to the question of formal or non-formal education.The focus on non-formal linguistic practices with a child-centred approach at Sunflower is not only connected to discourses at this kindergarten but also general discourses on children's participation and language development in national policy documents (Schleswig-Holstein Ministerium für Soziales, Gesundheit, Familie und Gleichstellung, 2012).In the interviews with parents, school emerged as an important discursive condition for the choices regarding language and linguistic practices they made on behalf of their children.Concerns regarding the children's language competence in German was a main factor in their choices; to compensate for the kindergarten's non-formal approach, the parents engaged in various formal linguistic practices with their children.This is also evident in the following excerpt from an interview with Manuel, who describes his son, Niko's, linguistic practice: Yes, sometimes it is like that, well, that you catch them when you fetch him or so, you know?That you catch them, if they now speak the same language, Turkish or so.Then they babble Turkish.Where I tell them, "Guys, you have to speak German".So it will be a bit easier in school later on.But it is, well, they are young -in here, out there and then they still do what they want.(Interview excerpt with Manuel, Sunflower kindergarten, February 2016) In this excerpt, Manuel focuses on his son's German language development, referencing school as a reason for the relevance of German.It is also interesting that he chooses the word 'babble' when referring to his son's choice to use Turkish, implying criticism of this linguistic practice.During the interview, Manuel expressed that German should be the primary language at the kindergarten, while Turkish should be spoken at home, in contrast to Sunflower's view of the kindergarten as a multilingual space.To some degree, this view was shared by all the parents I interviewed, although Manuel made the strictest distinction.Unlike Susanne, Manuel chose to enrol Niko in formal German lessons roughly a year before Niko started school.Manuel refers to the same discourse as Susanne -that learning two languages at the same time would be too much -but he regards German as the more important language.In terms of language ideology theory, this points to a view of languages as separate entities (Riley, 2011) and having different values (Jaffe, 2009); German is seen as the main language that is important for participating in society, while Turkish is seen as belonging to the family.The discourse about the importance of school was also referenced by the third family I interviewed.The father, Thorben, describes how he often sits down with his children in the evening to learn English, including names of colours and numbers, using an iPad.Thorben notes that the children are able to learn new languages quickly and that he wants to give them an advantage when they start learning English in school: And I try to lead them a bit closer to English, so they maybe through this, maybe get a little help.I think, in third grade, they start with English already.My son, that he at least knows some words by then.Maybe also the numbers, so up to, what do I know, ten or twenty, and so.(Interview excerpt with Thorben, Sunflower kindergarten, February 2016) Thorben is focused on his children's English language development, not German, but the discourse to which he refers is similar to that of the two other parents.School is an important discursive condition affecting many of the language choices the parents make for their children, and it is connected to their motive (Scollon & Scollon, 2004) to do what is best for their children.
Another important discursive condition that emerged in the interviews is migration, and it intersects with school in many of the parents' practices.While these discourses primarily relate to the parents' inter-or intrapersonallevel historical bodies (cf. Hult, 2017), they also relate to society-level discourses regarding equality in the education system, as Oberhuemer (2015) points out.Programs such as Sprach-Kitas (Bundesministerium für Familie, Senioren, Frauen und Jugend, 2017), for which kindergartens can apply to get extra funding to support multilingual children's language development, are part of this discourse.By making these connections between micro-and macro-level discourses, I do not mean that they connect in only one way (i.e. from the society level to the micro level or vice versa).Rather, they meet and intersect in the families' choices.Still, the macro-level discourses are important.As Lane (2010) points out in her study on language shift from Kven to Norwegian, parents do what they think is best for their children, but their choices are influenced by societal discourses and attitudes towards minority languages.Although my study was carried out in a quite different context, some of the same patterns of connections between micro-and macro-level discourses emerged.One of the interesting aspects of my study is that what the parents and kindergarten teachers regard as best for the children (James & James, 2008) is fundamentally different.

Decisions about Children's Future Made in the Present Based on the Past
Nexus analysis studies action situated in a moment in time and space (Lane, 2010, p. 68) with the underlying idea that action refers to past experiences and future expectations.My data revealed many such trajectories, where both parents and kindergarten teachers referred to past experiences on the one hand and goals for the children's future on the other.In the excerpts, school is a clear reference to the children's future, and choices related to the future are made in the present.However, the relevance of the past became visible through the kindergarten teachers' references to the development of their multilingual practice and through the parents' views on migration as part of their historical bodies.As Hult (2017) states, historical bodies contain the beliefs of earlier generations, which became salient when the parents referred to their own or their parents' experiences of migration.Some parents referred to their own childhoods and experiences with the German school system as factors affecting the choices they made for their children.Some viewed migration as part of their personal history that distinguished them from non-migrant Germans, while others included their own migration in the German society in general.
As with school, these discourses on migration connect to different levels and intersect in various ways in the parents' choices for their children.In my opinion, it is the insight in the intersection of discourses affecting choices regarding children's future that contribute to the field of childhood studies, as I will discuss in the last sections of this chapter.
As mentioned in the introduction, two important concepts from the field of childhood research are intersectionality and generation or generational order (Alanen, 2009(Alanen, , 2016)).According to Alanen (2016, p. 158), intersectionality in research has been used as an additive approach to individual identity as well as a non-additive approach to differences between individuals.In relation to the view that children's lives are intersectionally structured, she criticises the fact that intersectional thinking "appears to be a […] thought experiment" in childhood studies (Alanen, 2016, p. 159).Referring to Qvortrup (2008), she argues for generation or (inter)generationality as an important category for confronting the challenges of intersectionality, as childhood can only be understood as interdependent with a counter-category, such as adulthood or a "differently constructed generational category" Alanen (2016, p. 159).Honig (2009, p. 46) argues that children become children -and adults become adults -through institutionalised practices of differentiation.7In this regard, the positions of children and adults in relation to the constructed concept of childhood are important (Honig, 2009).This relates to generational order (Alanen, 2009, p. 161), according to which childhood exists in relation to other social categories and involves the idea of a system of social order.It connects children to social circumstances in which they participate in social life.Childhood extends beyond the differences between generations as age categories; Honig (2009, p. 48) argues that childhood is a social position that is influenced by various factors, such as age, gender, social-cultural environment and ethnicity.
Both Alanen (2016) and Honig (2009) describe childhood as a fluid category that is interrelated to other categories.I also understand both of these descriptions as intersectional views on childhood.My point is not to analyse these intersections for single children, focus on diversity within the category or add different sections to create a full picture of multilingual childhood.Rather, I think that these concepts, similar to nexus analysis, draw attention to the complex intersection of linguistic practices and the choices experienced by the children in my data.As James (2009, p. 42) points out, "children live their lives in and between any numbers of social institutions", including families, educational institutions and society.These institutions, and discourses at different levels (Hult, 2017), contribute to the complexity of children's lives.
Generation and generational order are important not necessarily in relation to age, but in relation to the roles of parents, children and kindergarten teachers in the kindergarten context (Ødegaard & Krüger, 2012).James (2009, pp. 42-43) argues that people (in this case, teachers and parents) occupy specific social positions and may transform the social structure, thus shaping the conditions for children's agency.Bergroth and Palviainen (2017) make a similar point in their discussion of the interplay of educational and language policies in bilingual kindergartens and bilingual children's agency.Their analysis shows how practice structures in the studied kindergartens, the teacher's pedagogical linguistic solutions and official language policies and educational policies shape conditions for bilingual child agency.In nexus analysis, this process of transforming social structures can be connected to the individual's historical body, which has the potential to influence other people's life experiences through mediated action (Hult, 2017).In interrelationship with their parents, children belong to a certain social category, and in interrelationship with their kindergarten teachers, they belong to a different category.Honig's (2009) reference to institutionalised practices of differentiation applies to the institutions of both kindergartens and families, but as they intersect with macro-level discourses, society also plays a role.
The multilingual children in my data have different positions in their family and in their kindergarten.Nexus analysis of their parents' and teachers' language choices reveals the complex intersections of linguistic practices that frame their childhoods.In the interview excerpt presented in the introduction, figure 12.1 Traditional bilingualism: Two autonomous linguistic systems(Garcia & Li Wei, 2014, p. 14)