Heritage Language Maintenance of Turkish Immigrant Families in the Netherlands

1

Heritage Language Maintenance of Turkish Immigrant

Families in the Netherlands

Zuleyha Oprea

S4134117

Heritage Language Maintenance of Turkish Immigrant Families in the Netherlands

Family language policies to foster language maintenance are often a challenge for families in diaspora, as they may collide with prevailing attitudes towards both the languages involved and the adopted language practices in the families. In this research, the extent of heritage language maintenance and development in Turkish immigrant families in the Netherlands was investigated. Using an exploratory research design, an online survey containing closed- and open-ended questions was distributed to 25 Turkish parents. Purposive snowball sampling was used to reach the participants. The questionnaire focused mostly on the language and social background of families, their children, their language practices as well as heritage language ideology in the family context. The data was analyzed according to grounded theory and coded according to themes which are related to the research question (Bergson et al., 2003; Heyde,2018). The findings show that proficiency in Turkish language affects frequency of the language use. Furthermore, the results indicate that digital tools have the biggest role in language practice and maintenance activities and Turkish language maintenance is considered to be very important in order to have a better communication in the family and to preserve ethnic identity. This study contributes to the bulk of literature on HL practices and maintenance, family language policy.

Key words: heritage language, family language, language maintenance, language practices, language attitude, Turkish as heritage language

4 Acknowledgments

My biggest thanks goes to the all families that completed the survey and introduced me to their relatives with all their kindness. Also, thanks to my supervisor since she was always so supportive and solution-oriented during the all challenges that Covid 19 brought to our lives and she was very helpful and professional in each step of this research. Thank you-Teşekkür ederim!

5 List of Abbreviations

FLP Family language policy

HL Heritage Language

HLiE Home Language in Education LPP Language Planning and Policy

6 Table of Contents

Acknowledgments ...4

List of Abbreviations ...5

1. Introduction ...8

1.1 Education policies in the Netherlands and Turkish in the education system ...9

2. Scope and limitation of the study ... 10

3. Aims and Design ... 10

4. Background ... 10

4.1 Family Language Policies ... 10

4.1.1 Domains of Language Policy and Planning ... 10

4.1.2 Family Language Planning and immigrant languages ... 11

4.2 The Factors Effecting HL Maintenance ... 13

4.2.1 HL Literacy ... 13

4.2.2 Attitudes of parents towards HL maintenance ... 14

4.2.3 HL Practices ... 15

4.2.4 Dutch language policies on minority languages ... 16

4.3 HLs in the Netherlands ... 17

4.3.1 Moroccans in the Netherlands ... 17

4.3.2 Surinamese in the Netherlands ... 17

4.3.3 Turkish as HL in the Netherlands ... 17

5. Methodology ... 18

5.1 Participants and sampling strategy ... 19

5.2 Parental survey ... 21

5.3 Procedure & Design ... 21

5.4 Data analysis ... 22

6. Results ... 23

6.1 Effects of Parental Background on HL Maintenance and Practices ... 23

6.1.1 Effects of Parental Background on Children’ HL Education ... 25

6.2 Language competence ... 27

6.2.1 Parents’ HL competences ... 27

6.2.2 Children’s HL competence... 28

6.2.3 Language choices and practices in the family ... 29

6.3 HL management activities in the family ... 31

7

6.3.2 HL use by children in the family... 32

6.3.3 Parents’ language maintenance efforts ... 33

6.4 Parents’ attitudes towards HL maintenance... 34

6.4.1 The reasons for Turkish language maintenance ... 36

7. Discussion ... 37

8. Conclusion ... 39

9. Appendices ... 41

9.1 Appendix I: Questionnaire (English translation) ... 41

... 46

... 46

9.2 Appendix II: Questionnaire (Turkish version)... 48

8 1. Introduction

Multiculturalism and multilingualism have become a crucial phenomenon in the new world with ever increasing immigration across the globe. Especially in the last decades, borders have become growingly meaningless as means of separating the countries culturally and linguistically. Eventually, this situation has resulted in the growing presence of more than one language in a single household or adopting hybrid cultural forms within one family. One of the countries that has experienced multilingualism and multiculturalism is the Netherlands. The Netherlands has received many immigrants with different characteristics since 1945 (Vasta, 2007). Some of them settled in this country from former Dutch colonies between 1945 and 1960s. Afterwards, Eastern European, Turkish and Moroccan immigrants were recruited in the 1960s and 1970s due to a lack of labor force in the Netherlands at the time (Vasta, 2007). As a result of these immigration waves, the Netherlands has a diverse population with different ethnical backgrounds, adding to the already rich autochthone language diversity. Minority languages have become majority languages in some neighborhoods or in schools, as more and more children maintain their heritage languages (HL after here). For instance, a public Dutch school –de Panda- which is situated in Utrecht have approximately 300 students in the age of four to twelve years old and majority of these students are Moroccan and Turkish (Arts & Nabha, 2001). However, this situation has turned into a controversial issue for immigrant receiving countries including the Netherlands as immigrant pupils often attain lower school achievements (Shewbridge et al., 2010). Therefore, governments have implemented assimilation or integration policies which eventually may pressure immigrants to abandon their HL especially in schools (Ağırdağ et al., 2014).

Recent migration also led the government to develop their own language policies to safeguard the position of the Dutch language in society (Rijkschroeff et al., 2005). Immigrant families have been experiencing some challenges during this process. On the one hand, they try to integrate into society both linguistically and culturally, while on the other, they wish to preserve their culture and HL. By being multilingual and multicultural, they have turned the Netherlands into a more diverse linguistically and culturally country. However, as mentioned above, being a multilingual immigrant has been associated with lower academic achievement (OECD, 2018) and communication problems may arise, which in turn may lead teachers to intervene in HL management strategies for immigrant families since mainstream teachers believe that lower achievement in schools is caused by, among other factors, using different HLs (Bezcioglu-Göktolga & Yagmur, 2018). Turkish families are one of the immigrant communities who have been affected by these policies. Their low achievement in schools is often associated with using the Turkish language at home (Agirdag et al. 2014). As result, they are occasionally advised to stop using their HL at home. With this assimilation approach, some Turkish families have started to make some changes in their HL management strategies family (Bezcioglu-Göktolga & Yagmur, 2018). In this research, HL maintenance of Turkish immigrants in the Netherlands will be studied.

9 1.1 Education policies in the Netherlands and Turkish in the education system

As mentioned in the previous section, low academic achievement is generally associated with being multilingual and, as a result, research has suggested implementing multilingual pedagogies which include immigrant pupils’ HL in the school curriculum (Van Avermaet et al. 2018). Yet, this suggestion is often rejected by education systems in countries with a dominant monolingual habitus (Gogolin, 2002) and bilingual children mostly follow their education in the mainstream language which is often their second language (Sierens & Ramaut, 2018). When we look at the education system in the Netherlands in terms of minority language policies, immigrant languages are not given a place in the education system (Yagmur, 2004). In addition to growing diversity also the quality of HL provision has been assessed as inadequate, which in turn decreases the demand for education in those languages (Yagmur, 2010). On the other hand, how to deal with this language diversity in schools is an important issue as well. Since there are limited studies and projects in which HLs were implemented in the education system (Sierens & Ramaut, 2018), it is a challenge for teachers to implement a multilingual pedagogy in classes. That is why, education in the mainstream language is a still a predominant phenomenon in the Netherlands (Yagmur, 2004).

When we look at the historical development in terms of immigrant languages, most immigrant students entered the Dutch education system in the 1970s and these immigrants were predominantly from Turkey and Morocco (Yagmur, 2010). Dutch was taught as a second language and special classes were provided in which students could study their HL. In the 1990s, the demand for HL education was quite high. However, deficiency in the quality of materials and a highly structured curriculum for Turkish language education caused a decrease of demand. A study by Yagmur (2010) indicated that in 1990, the total number of Turkish students in the Netherlands was 38.294 and 82% of their children attended HL education. In 1993, the total number of Turkish students in the Netherlands was 42.619 and 77% of the children attended HL education. As it can be seen in these figures, the number of students who got access to Turkish education decreased.

Knowledge of the Turkish language is often considered as an obstacle for Turkish immigrants instead of an opportunity (Agirdag et al., 2014). This concern may lead families not to engage in a formal education of the Turkish language. However, to be able to maintain Turkish as a HL and hand it down to the next generations, formal education is needed. Although there are enough institutions providing social and religious services to Turkish communities, there are not enough quality institutions to promote and teach the Turkish language (Yagmur, 2010). Even if the Netherlands has started to improve some materials for Turkish language instruction, these projects are not enough to cater for suitable language education (Yagmur, 2010). On the other hand, the teachers who are normally recruited from Turkey in order to provide Turkish language education may themselves also struggle to integrate into the new society (Yagmur, 2010). According to Yagmur (2010), teachers returned to Turkey before feeling integrated into the Dutch society and this situation resulted in an unsuccessful attempt. As a result, many Turkish immigrants will not be provided with enough opportunities to improve their heritage language.

10 2. Scope and limitation of the study

The current study is based on an online survey (Heyde, 2018), sent with an explanation in which it was stated that the questionnaire must only be completed by a Turkish background immigrants living in the Netherlands, speaking Turkish language and having at least one child regardless of their generation. However, some participants who filled up the questionnaire did not have a child. Therefore, their answers were not included in the study. Apart from this, in the initial phase of the study, it was planned to observe Turkish families in their family context to be able to compare their answers with their practices and to get more reliable results. Furthermore, it was planned to interview children, observe them while they communicate with their parents and ask some questions regarding their Turkish language practices and attitudes. Yet, due to the Covid 19 circumstances, the interviews could not be conducted. That is why questions about children were obtained by their parents’ report and the conclusions were drawn based on their reports.

3. Aims and Design

In this study, the research question: “To what extent does language maintenance and development occur in Turkish immigrant families in the Netherlands?” will be answered. To answer the main research question five sub-questions were formulated:

a. How does the background of the parents and children influence their language practices, maintenance and attitudes in the family?

b. Does proficiency in the HL affect the frequency of use of the language in the family? c. What activities do Turkish immigrant families specifically engage in in order to

maintain their HL in the Netherlands?

d. To what extent do they use (digital) tools to maintain their HL?

e. What are the attitudes of parents towards their children’s HL use and development? This research provides an interdisciplinary perspective on HL use, practices and attitudes. Another aim of this research is to deliver a contribution to language planning and policy (LPP henceforth) in the context of Turkish immigrants in the Netherlands. Also, it provides a clear understanding about the Turkish language as a HL and its usage in the Netherlands.

This thesis consists of five parts. The first part is an introduction which provides an overview about multilingualism, Dutch policies and Turkish as a minority language in the Netherlands. The second part is a literature framework in which family language policies in general and in the Netherlands, HL and its advantages and immigrant communities in the Netherlands were reviewed. The third part gives information about the methodology of the study. The fourth part consists of the results and is followed by a discussion and conclusion.

4. Background

4.1 Family Language Policies

4.1.1 Domains of Language Policy and Planning

Language policy and planning have always been a significant issue in the world. In the 1950s and 1960s, classical language policy was adopted to eliminate language-related problems after the independency of new nations. Independency of new nations brought along new language policies in order to boost the status of these indigenous languages, develop

11

their writing system, in short; give a competitive status to these languages in light of current dominant languages (Spolsky, 2012). According to Spolsky (2004), the classical model of LPP was a “top-down” process in which impacts on other domains besides the states’ were generally neglected. However, a few scholars tried to show the influence of LPP on other domains, such as on families.

Fishman (1970) proposed a three-generation model which focused on the effect of families on immigrant language shift. According to this theory, first generation immigrants adopt a new language and start using it in their home. Second generation immigrants are thus raised as bilingual individuals and third generation immigrants mostly grow up as monolingual individuals with the dominant language. As such, it can be seen that language attrition and language loss happens in the family across the generations. For this reason, the family factor was growingly included in the domains of LPP.

Language policy and planning has four main components, namely language practices, language beliefs, language ideology and the explicit policies and plans (Spolsky, 2004; 2012). Language policies may differ according to circumstances, places or the languages involved, and they can be implicit or explicit. In implicit policies, countries tend to ignore the recognition of other languages while in explicit policies, countries provide communities with multilingual services. These policies are implemented within a speech community and their implementation depends on the power and authority. However, according to Cooper (1989), language policies gain strength with the support of powerful authorities but when they lose their power and strength, language policies are no longer available. Therefore, language practices and beliefs are the factors that keep languages alive and powerful.

In modern LPP, contrary to the classical one, many other domains are recognized in addition to the states and family. These domains “range from the supra-national organization (the European Union being one of the best studied cases through the state and regional or local governments (polities) to the army, business, work, media, education, religion” (Spolsky, 2004, p.4). Each domain has their own different stakeholders, their own language beliefs and practices. That is why, language policies can differ in accordance to the different domains. On the other hand, these domains affect each other and lead to modern, biological and applicable language policies for the communities (Spolsky, 2004). The current study zooms in on the concrete case of LPP in immigrant families.

4.1.2 Family Language Planning and immigrant languages

According to Spolsky, family language policy (FLP, henceforth) entails three factors which are ideology, practice and management (Spolsky 2004). Language ideologies of families relate to their beliefs about their HL and majority language use in the family context. Language practices entail the ways in which families act according to their language ideology and how they use the languages. Finally, language managements are the deliberate external as well as internal efforts for HL usage (Spolsky 2004; Bezcioglu-Göktolga & Yagmur, 2018). To be able to understand the concept of FLP, we should also look at its definition. According to Curdt-Christiansen, FLP is defined as “a deliberate attempt at practicing a particular language use pattern and particular literacy practices within home domains and among family members” (Curdt-Christiansen, 2009, p. 352). FLPs are based on families’ ideologies for the sake of family

12

composition itself (Curdt-Christiansen, 2009). Language ideologies in the family are the ultimate factor in shaping family language planning, practices and maintenance (King, Fogle, & Logan-Terry, 2008). Hence, it can be stated that family is the first place in which FLP and planning take place, as parents shape “their children’s attitudes towards the HL” (Bezcioglu-Göktolga & Yagmur, 2018, p.223; Li, 1999).

School is another decisive factor in FLP. Until children start school, parents are the only factor affecting children's attitudes towards language use in immigrant families. However, when they start school, they encounter dominance of other languages and parents as a unique decisive factor for FLP must share their influence with schools and peers (Baquedano-López & Kattan, 2008). In these situations, we should not think of schools as a whole and consistent structure in influencing parents for language ideologies. There might be a school language policy but teachers – on their own – can also affect parents’ language ideologies by focusing on classes’ demographic, social-political structure, etc. (Bezcioglu-Göktolga & Yagmur, 2018; Young, 2014).

There are many empirical studies about FLP. Hua & Wei (2016) studied transnational experiences on FLP. In this study, it becomes evident how language ideologies, practices and management influence FLP. Three transnational families from China living in Britain showed different attitudes towards Chinese use which in turn affected their FLP. The first family was a Korean family from China living in Britain. They preferred to give up the Chinese language and develop their Korean in Britain. Consequently, they put forward their Korean identity which was maybe suppressed in China and improved their English as well. The second family was a Chinese family living in Britain. This family preferred to maintain their HL to expand their transnational networks. Lastly, the final family was a first generation retired Chinese couple who were highly educated in China. For this family, Chinese was more important than other languages for cultural interaction. Where they are coming from was more important than where they are now for them (Hua & Wei, 2016). So, as it can be seen from this study also, language ideologies and beliefs in the family affect language maintenance activities and practices. For this reason, in this study, FLP of participants were investigated by questioning their beliefs and ideologies towards HL use in order to see the effects of their attitudes in HL maintenance.

When we look at the effects of LPP on family languages, we should refer to the longitudinal study conducted by De Houwer & Bornstein (2016). In this study, 31 bilingual interactions between mothers and their children were observed. All children were exposed to two languages from birth. Mothers mostly used the same single until their children reached 20 months. When their child reached 53 months, they started to use a second language occasionally. According to the authors, “These findings reveal a possible effect of children's overall level of language development and demonstrate the difficulty of adhering to a strict ‘one person, one language’ policy” (De Houwer & Bornstein, 2016, p. 680).

Each empirical study on FLP shows that there are many factors affecting FLP which in turn causes different types of language practices and attitudes towards HL. Luykx (2005) and Curdt-Christiansen (2014) illustrated that political and social-cultural circumstances in a country can influence FLP against HL use. The former study showed us that Quechua-speaking families in

13

Bolivia started to use Spanish within their families as a result of school pressure (Luykx,2005). The second study also illustrated that Chinese families living in Singapore shaped their FLP according to the national language policy. This situation led parents to have lower expectation for their children’s Chinese proficiency (Curdt-Christiansen, 2014). In other words, the dynamics out of the families can influence their FLP as well. On the other hand, there are also studies focuses on the dynamics within the family which affect their FLP. The attitudes towards a language use in a family play a crucial role in FLP. In this respect, King (2001) found that children prefer to speak only Spanish and abandon their mother tongue Quechua since their parents believed that speaking both languages at home will affect children’s school achievement negatively. These studies demonstrate that FLP can be shaped by external factors (political, socio-economic circumstances) and internal factors (families’ attitudes, beliefs) and also, FLP affect their HL maintenance and practices.

In the current study, providing the studies about FLP in the immigrant context enables us to see the interrelation between FLP and the dynamics out of the families and the dynamics within the families. This theoretical framework in FLP will facilitate understanding the effective factors on HL maintenance of Turkish immigrant families (Bezcioglu-Goktolga & Kutlay Yagmur, 2018). Theoretical background of these factors are provided in the following sections.

4.2 The Factors Effecting HL Maintenance 4.2.1 HL Literacy

HLs are defined as “languages other than the dominant language(s) in a given sociopolitical, geographical, psycho-linguistic and structural-linguistic context. Often, the term HL is similarly referred to as ‘foreign’ and ‘minority’, ‘immigrant’, ‘indigenous’ and ‘colonial’, ‘ethnic’ or ‘immigrant minority ‘or ‘community’, ‘immigrant’ and ‘regional’ languages” (Stavans & Ashkenazi, 2020, p:1). There are also accepted differences between heritage language and home language. According to Spolsky and Shohamy (1999), HLs are often considered as endangered languages while HLs are the languages of other immigrant groups. However, in this study, HL is accepted as a language apart from mainstream language (Stavanz & Ashkenazi, 2020). HLs include immigrant, ancestral and indigenous languages (Seals & Shah, 2017).

“HL literacy encompasses not only knowing how to write and read in the HL, but also understanding the socioculturally relevant and appropriate ways in which literacy is used in the heritage communities” (Lee, 2013, p.2). Furthermore, HL literacy is also associated to some advantages for its speakers. Firstly, improvement of HL literacy provides maintenance of that language. As some languages encounter endangerment due to decreases in the number of its speakers, HL literacy can help foster language protection. Secondly, HL literacy can enable its speakers to communicate with each other via different types of sources which in turn provides maintenance of HLs not only in the home sphere but also in texted channels (Lee, 2013). Another advantage of HL literacy is personal success in education. According to Krashen (1998), HL literacy influences its speakers positively when they learn other foreign languages. That is, apart from linguistic and social advantages, it also provides academic advantages to its speakers (Lee, 2013). Considering these advantages, the current study investigated HL literacy and HL practices of Turkish parents and their children in order to clarify whether HL literacy affect their language use in their lives or not.

14 4.2.2 Attitudes of parents towards HL maintenance

According to De Houwer (2007), parents’ language ideologies and practices have a significant effect on a child's attitude towards HL and, consequently, their bilingual education. De Houwer (1999) created a model (see figure 1) to clarify the influences of parents on their child’s bilingualism.

Figure 1: Relationship between parental beliefs/attitudes and children’s language

development (De Houwer, 1999).

This model shows us an interactive process in which parental beliefs and attitudes affect the development of children while the development of children can, in turn, affect parents’ beliefs and attitudes (King et al., 2008). Three types of parental attitudes influence HL practices and maintenance: “First, parents often have clear ideas about which languages should be used for what purposes with their children. Second, parents’ attitudes concerning particular types of interactions, such as mixing or use of slang, impact their own child-directed speech. Third, parents’ attitudes towards language learning and bilingualism also come into play in influencing their interactional strategies” (King et al., 2008, p. 912). There is a much research on how parental attitudes affect children’ attitudes towards HL and their bilingualism. In her study with 1899 parents in Belgium, De Houwer (2007) showed how parents’ linguistic contribution affected the bilingualism of children. In each family, at least one of the parents spoke a second language other than the majority language. HL use in the family was assessed for both parents and their children who ranged from 6 to 10 years. The findings show that children who were raised with two languages dominantly speak the majority language. Differences in parental input were also investigated in this study and the results show that if both parents use the minority language and one of them also speaks the majority language, children have a better achievement in the minority language. Whereas, if both parents speak the majority language, children are less likely speak the minority language. In other words, this study indicates us that raising children as bilingual individuals needs parental language input since parents’ contribution will affect their children’s achievement in minority language. Another study by King and Fogle (2006) illustrated that experiences of Spanish-English bilingual parents influenced their linguistic competence in a positive way. Against this backdrop, the current study is going to illustrate how parents’ attitudes are towards HL education of their children and what types of HL practices are done in the family and finally how these situations affect children’ HL use and attitudes.

15

When we look at the Turkish parents, there are some studies focusing on their attitudes towards HL in the family. Bezcioglu-Goktolga & Yagmur (2018) studied HL policy of second-generation Turkish families in the Netherlands. In this study, observations and interviews were conducted with 20 Turkish families in the Netherlands. Most of the participants mentioned preserving their Turkish identity by speaking the Turkish language and teaching it to their child as well. On the other hand, almost all participants wanted their children to learn both Turkish and Dutch equally. The reasons behind this were related to educational success and welfare in the Netherlands. Consequently, according to this study, Turkish parents generally teach their HL to their children and support them to improve their HL literacy. Another study about language maintenance patterns of Turkish immigrant communities in Australia, and western Europe was conducted by Yagmur in 2004. This research revealed that the Turkish language has a value in child rearing and communication with family and friends. In this respect, this study will demonstrate the Turkish parents’ attitudes towards HL use and the underlying reasons of these attitudes in order to have a broad perspective with regard to how Turkish families in the Netherlands maintain and develop their HL.

4.2.3 HL Practices

According to Pillali et al. (2014, p. 81) “Language practice refers to the extent to which a language is used with family members and it includes rules for the appropriacy for each named language while language management refers to choices made in relation to language use at home”. HL literacy mostly depends on parents’ practices at home since there are not so many schools or government initiatives supporting bilingual education or HL schooling. Without home practices, it is almost impossible to develop HL literacy (Hinton, 1999). Most of the studies in this domain generally focused on attitudes of parents toward bilingual education and on the influence of the language of school on HL. Therefore, “how HL is used, taught, and practiced in immigrant homes remains largely unknown" (Li, 2006). Li’s study (2006) regarding HL practices at home shows us that focusing on mainstream language (English in this case) and abandoning HL activities at home caused children to lose HLs within one-year of public-school experiences in English. This case proves that home practices and public-schools affect HL development. Additionally, positive attitudes of parents and teachers can trigger development of HLs. As a result, HLs should not be restricted to home; they should be used at schools as well. Also, parents should communicate with their ethnic groups as much as possible to be able to maintain HLs and they should use any type of appropriate materials to develop HL literacy of their children (Li, 2006).

To increase HLs literacy of immigrant children, the schools must be involved in the process. However, it is quite hard to expect teachers to speak all minority languages that are spoken in the class, digital tools, more specifically computer-based learning, is the most convenient alternative to teach immigrant pupils their HL. Apart from the teaching HL, the studies showed that computer-based learning environments (CBLEs) influence students’ cognitive abilities positively (Van Laere et al., 2017; Nir-Gal & Klein, 2004). There are several projects using CBLEs and “aiming to foster pupils’ knowledge acquisition through the inclusion of support in their

home language” (Van Laere et al., 2017, p.99). e-3M

(http://www.3mproject.nl/producten.html), EDINA (https://edinaplatform.eu/), Teach-D (http://www.teach-d.eu/) projects are some the projects using CBLEs (Burdese & de Jong,

16

2020). In this regard, the current study will explore to what extent Turkish immigrants and their child use (digital) tools to maintain their HL (Oprea, 2020).

4.2.4 Dutch language policies on minority languages

The Netherlands is one of the most immigrant receiving countries in Europe. 24% of the population of the Netherlands is formed by minority ethnic groups (Central Bureau for Statistics [CBS], 2019) and this percentage is significantly higher for school aged children. Ever increasing migration movements to Europe have caused governments to change their national language policies. Language policies towards minority groups in Western countries including the Netherlands have been implemented in order to foster integration strategies for immigrants. However, these policies have been perceived as being assimilative by minority groups. When we look at previous Dutch strategies, we indeed see different approaches towards minority languages. The Netherlands first adopted a social country policy which is also known as ‘civic ideology’ (Yağmur & Van de Vijver, 2012), in which there was little interference in the languages of minority groups. However, after 1994, Dutch language policy has turned into an ‘ethnist ideology” which mostly excludes minority groups from society (Yagmur, 2009). Before this change, immigrants’ languages were taught in schools to support Dutch language and improve their HL (Bezcioglu-Goktolga & Yagmur, 2018). “This practice began in 1974 in primary schools under the title ‘Education in One’s Own Language and Culture’. In 2004, these courses were abolished, with the idea being that they contradicted the policy of integration of minority children” (Bezcioglu-Goktolga & Yagmur, 2018, p.47). However, this pluralistic approach has turned into an ethnist approach in the Netherlands (Driessen & Merry, 2011). Promoting mainstream language linked to economic and social opportunities made this language more valuable among immigrant groups (Rijkschroeff et al., 2005). Additionally, these policies caused some minority groups to be marginalized which in turn prevented them from reaching a higher status within the Dutch society (Vasta, 2007; Yagmur, 2009).

Apart from implicit policies towards immigrants in the Netherlands, there are also some explicit processes to support their language policies. When analyzing common beliefs about immigrant groups, the media shows up in first place in terms of influencing people's attitudes towards immigrants. Immigrant groups are often associated with societal problems such as criminality, unemployment, illiteracy etc. (Brands et al., 1998; Yagmur, 2009). These attitudes may lead immigrant groups to feel ashamed of their ethnic identity which in turn causes them not to use their HL in the public sphere. Even if it does not seem an intentional implementation, it may eventually induce immigrants to give up their language and behave accordingly to the integration policy (Extra and Verhoeven, 1993). As a result of recent Dutch immigration policies, the current study will analyze Turkish parents’ attitudes towards HL use and its development in the family. The reasons behind these attitudes will also be investigated in a small scale to be able to see whether Dutch policies affect Turkish immigrants’ attitudes or not.

17 4.3 HLs in the Netherlands

4.3.1 Moroccans in the Netherlands

With the end of World War II, economic changes and need for low-skilled workers in the Netherlands made it an immigration country. This need for the workers drove the government to recruit people from different countries and Morocco was one of these countries. These workers generally worked in factories and in jobs that require physical strength. Moroccans were settled in the cities like Amsterdam, Rotterdam and Utrecht (Beets et al. 2008; Butter, 2000). They came to the Netherlands without their families since they were thought to be temporary immigrants. However, the oil crisis in 1973 affected economic circumstances and this situation led the government to stop recruiting from other countries. During this crisis, most of the southern immigrants returned to their home countries due to economic development in their home country. However, most Moroccans did not return to their countries, on the contrary, migration from Morocco increased since their family members started to migrate to the Netherlands as well. Currently, Moroccans are the third largest non-western immigrant groups in the Netherlands after Surinamese and Turks. In 1980, immigration from Morocco reached the highest level (Beets et al. 2008).

4.3.2 Surinamese in the Netherlands

Unlike the other communities, Surinamese migrated to the Netherlands in the colonial period (1967-1975) since Surinam was among Dutch colonial possessions. While some Surinamese migrated to the United States, the majority of them migrated to "the former mother country,” where, already Dutch citizens, they would immediately enjoy a secure legal status” (Niekerk, 2004, p.161). Even if migration of Surinamese started before other communities, their immigration level reached the highest point after World War II as well. The economic situation and unstable security circumstances in Surinam were the main reasons for the immigration to the Netherlands. There are two main different Surinamese communities in the Netherlands and they diverge in terms of ethnicity. Afro-Surinamese and Indo-Surinamese are two main Indo-Surinamese groups in the Netherlands and around 350.000 Surinamese live in the Netherlands (Oostindie, 2011; Niekerk, 2004). Amsterdam and The Hague host the majority of Surinamese community.

4.3.3 Turkish as HL in the Netherlands

Immigration of the Turkish community to the Netherlands started in the 1960s during the economic boom after World War II. The first generation of Turkish immigrants came from rural areas in Turkey (Backus, 2013). Favorable economic opportunities in the Netherlands caused more Turkish people to migrate (Bezcioglu-Goktolda & Yagmur, 2018). Today, approximately 400.000 immigrants with a Turkish background live in the Netherlands (CBS, 2020). After Germany and France, the Netherlands hosts the third largest Turkish immigrant group in Europe. Turkish immigrants generally reside in urban areas like Rotterdam, Amsterdam, Utrecht, The Hague and they generally tend to live in the same areas (Bezcioglu-Goktolda & Yagmur, 2018).

The average age of the Turkish community is younger than the autochthone population in the country. 80% of Turkish immigrants are below 40 years while this rate is just 50% in Dutch society. Also, the birth rate of Turkish immigrants (2.2) is on average higher than that of the

18

Dutch society (1.6) (Yagmur, 2009; Central Bureau for Statistics [CBS] 2019). According to the Central Bureau for Statistics, the Turkish population will double by 2050 as a result of this demographic situation. Also, marriage is another factor which causes the increase of the Turkish population. Usually, Turkish immigrants marry a person from the same ethnic background (Yagmur, 2010). That is why, a majority of Turkish immigrants find their partners in Turkey and move them to the Netherlands for family reunification. This also increases the Turkish population in the Netherlands.

Since Turkish immigrants generally live near to each other or keep close connections to each other by means of existing organizations, they normally have a large social network. Also, they are quite active in economic areas, mostly by owning their own businesses. These circumstances make the Turkish language available in almost every part of their lives. Turkish TV and radio channels, magazines and newspapers are easily accessible. As a result, the Turkish language has a dominant role for Turkish immigrants (van Heelsum & Tillie, 1999; Yagmur 2009). For the first-generation immigrants who were temporary workers in the country, the Dutch language was not a 'must' since they were planning to return to their home country. They worked with other Turkish immigrants and lived in predominantly Turkish-speaking communities. Hence, their Dutch skills often did not improve (Smets and Kreuk 2008; Crul and Doomernik 2003). They could not give a highly literate Turkish language education to their children since they were mostly low educated individuals. As a consequence, their children often had a low level of proficiency in both languages (Leseman & van den Boom, 1999).

To sum up, to be able to understand the language dynamics, parents’ beliefs and efforts in terms of HL use within the family, the studies about FLP of immigrant families should be firstly focused on. FLPs are related to families’ attitudes, practices and efforts about their HL and majority language use in the family context (Spolsky, 2004). There are different factors affecting HL maintenance of immigrant families. As it is mentioned previously, HL literacy, HL practices, attitudes towards HL and countries’ policies on HL are the main factors which shape families’ HL maintenance. On the one side, the studies (Luykx, 2005; Curdt-Christiansen, 2014; King, 2001) showed us that language practices, attitudes and governments’ policies against HL use in immigrant families resulted in abandoning HL. On the other side, the studies (Hua & Wei, 2016; De Houwer & Bornstein, 2016; Bezcioglu-Goktolga & Kutlay Yagmur, 2018; De Houwer, 2007; King and Fogle, 2006) revealed how parents’ attitudes towards HL, HL practices in the family and supports of LP for HL affected HL maintenance positively in the family. In this respect, the current study about HL maintenance of Turkish immigrants in the Netherlands will enable us to understand how the families maintain Turkish language in the family and how HL literacy, HL activities and attitudes towards HL affect their HL maintenance.

5. Methodology

An exploratory research design (Stebbins, 2001) was used in this study.In this design, both qualitative and quantitative data can be collected by the researchers. Although an exploratory design is prevalently thought as a qualitative method, quantitative methods to gather descriptive statistics can also be used in this design (Stebbins, 2001). In the current research, a survey was used in order to collect both quantitative and qualitative data. Concretely, the

19

research was conducted by means of an online survey to investigate FLP en heritage language maintenance of Turkish families in the Netherlands. Also, their HL practices, language ideologies and management strategies were investigated in the survey (Bezcioglu-Goktolga & Yagmur, 2018). The online survey was developed by focusing on the language and social background questionnaire (LSBQ) (Li et al., 2006) to be able to gather some detailed information about participants’ socio-demographic background as well as their language competences (Anderson et al., 2018). Data was collected from 25 Turkish participants who have at least one child and live in the Netherlands.

5.1 Participants and sampling strategy

The target population of this study was immigrant Turkish families with at least one child living in the Netherlands. To be able to get sufficient participants and “to narrow the range of variation and focus on similarities”, purposive snowball sampling was used (Palinkas et al. 2015, p.534). This technique can be implemented in two contexts: “The first context involves surveying members of a rare population. The second involves studying mutual relationships among population members. In both cases, respondents are expected to know about the identity of other members of the same population group” (Lavrakas, 2008, p.2). In this research, the second context was used to raise the number of participants. Participants were selected according to the following criteria: families had to have a Turkish language background, to be living in the Netherlands regardless of their immigrant generation status, to have at least one child and at least one of parents had to speak the Turkish language (Bezcioglu-Goktolga & Yagmur, 2018). Participants were reached by the author's personal contacts, posts on three different Facebook pages (Turkish communities in the Netherlands) and by visiting predominant Turkish neighborhoods in Dordrecht since the researcher is living in this city. The Turkish language was used when contacting the participants.

Contacting, recruiting and informing participants took place in April 2020. Because of the Covid 19 crisis and social distancing restrictions, data collection was carried out using online channels. Interviewing participants was one part of the research design in order to observe language practices of the participants in the family context, yet; it was not conducted due to the Covid 19 circumstances. The participants were from 13 different cities in the Netherlands and the data was obtained from 6 male participants and 19 female participants who ranged in age from 26 to 60 years. 9 participants were born in the Netherlands and the rest was born in Turkey and moved to the Netherlands at different ages. In addition, there were 38 children in total and they ranged in age from 1 to 34 years old. Some parents who were born in Turkey moved to the Netherlands prior to entering school while others moved during school age or after school age. As a consequence, the sample is composed of both first generation and second-generation immigrant participants. Those who were born in the Netherlands and have at least one child are regarded as second generation immigrants and the ones who were born in Turkey and then moved to the Netherlands with their family or alone are regarded as first generation immigrants (Yagmur, 2018). An overview of the sample can be seen in Table 1.

20 Table 1: Gender, age, generation, education, occupation, number of children and ages of children of the participants.

Parents Age Gender Children and

age Country of birth Life in Holland Mother tongue Second Language

Parent 1 30 Male 1 The Netherlands 30 Dutch & Turkish English & Dutch

Parent 2 32 Male 5 The Netherlands 32 Dutch & Turkish English & Dutch

Parent 3 34 Male 1.5 The Netherlands 34 Dutch & Turkish English & German

Parent 4 45 Male 9 & 4 The Netherlands 45 Dutch & Turkish English

Parent 5 26 Female 3 The Netherlands 26 Dutch German & English

and Turkish

Parent 6 30 Female 5 The Netherlands 30 Dutch & Turkish English

Parent 7 35 Female 6 The Netherlands 35 Turkish English & German

Parent 8 37 Female 19 & 17 & 11 & 3 The Netherlands 37 Turkish Dutch

Parent 9 38 Female 8-13 The Netherlands 38 Dutch German & English

and Turkish

Parent 10 49 Female 19 & 30 The Netherlands 49 Turkish Dutch

Parent 11 42 Male 8 Turkey 14 Turkish English & Dutch

Parent 12 30 Male 2 Turkey 3 Turkish Dutch

Parent 13 26 Female 4 Turkey 10 Turkish Dutch & English

Parent 14 40 Female 23 Turkey 21 Turkish Dutch

Parent 15 60 Female 25 & 26 Turkey 28 Turkish English & Turkish

Parent 16 28 Female 2 Turkey 3 Turkish English & Dutch

Parent 17 47 Female 15 & 18 Turkey 30 Turkish Dutch & English

Parent 18 33 Female 3 Turkey 4 Turkish Dutch

Parent 19 50 Female 29&19 Turkey 40 Turkish Dutch

Parent 20 57 Female 25 & 32 Turkey 40 Turkish Dutch

Parent 21 60 Female 39 & 34 & 24 Turkey 40 Turkish English & Dutch

Parent 22 49 Female 34 & 19 Turkey 43 Turkish Dutch

Parent 23 47 Female 17 & 20 Turkey 44 Turkish English & Dutch

Parent 24 54 Female 22 & 30 Turkey 47 Turkish Dutch

21 5.2 Parental survey

The data was collected by developing and applying a questionnaire which focused mostly on the language and social background of families, their children, their language practices as well as HL ideology in the family context (see appendix I). The questions to assess the socio-demographic information and language competences were adopted from an existing questionnaire (Li et al., 2006). The survey consisted of 26 questions, divided into three sections: in the first section of the questionnaire, socio-demographic information such as age, gender, education, occupation was obtained; in the second part of the questionnaire, social background and language competences of parents and their children were collected. Questions about the number of languages that they and their children can speak, proficiency in reading, listening, speaking and writing in Turkish both for themselves and their children (with self-rated assessment) were asked to the participants in this section. Linguistic competence of parents’ children was evaluated by parents as well. Questionnaire also included questions about Turkish language activities at home and in their daily life. Expectations regarding their children’s development in Turkish were questioned also (Bezcioglu-Goktolga & Yagmur, 2018). Finally, in the third section, practices for language maintenance and HL ideology in the family context were assessed. Both the questions concerning the demographic information, social background and language competences and those regarding language practices and attitudes towards HL were developed on the basis of the same questionnaire (Li et al., 2006).

The questionnaire used in this study was created with multiple-choice questions, closed-ended questions, open-closed-ended questions, partially open-closed-ended questions and rating scales. An overview of the questionnaire can be seen in Table 2.

Table 2: An overview of the questionnaire. (Please find the complete questionnaire in

Appendix I)

Sections Aims Items Format

Section 1 Demographic information and social background

9 Open-ended questions

3 Multiple choices questions

1 Partially open-ended question

Section 2 Language competence and

Language Practices

1 Open-ended questions

1 Multiple choices questions

1 Partially open-ended question

2 Rating scale

2 Close-ended questions

Section 3 Attitudes towards HL 3 Multiple choices questions

3 Partially open-ended question

5.3 Procedure & Design

Data was collected through an online survey and a web-based survey tool – Qualtrics – was used to send the questionnaire to the participants. 25 Turkish parents who were contacted via social media, personal contacts and by visiting Turkish predominant neighborhoods in Dordrecht were asked to fill out the questionnaire. While conducting the

22

research, anonymity of the participants and confidentiality of their responses were ensured (Bordens & Abbott, 1988).

Also, a printed questionnaire was given to a few participants who were reached by visiting Turkish predominant neighborhoods in Dordrecht and their responses were later saved to the Qualtrics Survey. Other participants accessed the survey directly through provided links on social media posts. Before filling out the questionnaire, it was stated that this questionnaire should be filled out only if one meets the following criteria: (i)Living in the Netherlands, (ii)having a Turkish ethnic background, (iii)being able to speak Turkish, (iv)having at least a child. Data of those who do not meet the requirement or the surveys which were partially completed was not included in the research.

5.4 Data analysis

Analysis of the data was a continuous process parallel with data collection (Charmaz, 2003; Zhang & Slaughter-Defoe, 2009). The data was analysed according to the grounded theory and coded according to themes which are related to the research question (Bergson et al., 2003; Heyde, 2018). In other words, first of all, the research questions were formulated to collect the data and then data was collected andanalyzed (Bergson et al., 2003). Two steps were followed during the analysis process namely an initial coding and a focused coding. In the first step, all possible ideas related to the research topic were generated and then each response was coded for the entire data. In the second step, initial codes were categorized, conceptualized and synthesized (Sbaraini et al., 2011; Charmaz, 2003). Since the questionnaire was conducted in Turkish, responses were translated in English.

The answers were classified according to the research questions. In order to answer the first research question on the influence of the background of parents and children on their language practices, maintenance and attitudes, the items 1 to 13 of the questionnaire were analyzed. These items consisted of information about demographic and language background of the participants and their children.

In order to answer the second question on influence of HL proficiency on the frequency of use of the language in the family, the questions (14 and 19) were asked. Parents rated their and their children’ HL skills. This data and the data from language practices were analyzed together to see whether HL literacy affects the frequency of HL use.

In order to answer the third research question on the activities that Turkish immigrant families specifically engage in order to maintain their HL in the Netherlands, the items related to language practices (15, 16, 17, 18, 20) were analyzed. These items consisted of the information about what types of activities Turkish parents and their children do to maintain the HL. In order to answer the fourth research question on the importance of using (digital) tools to maintain their HL, digital tool options were added to the items in 15 and 20. With these items, participants were asked to determine the frequency of their HL use in the provided options including digital tools. With these options, the importance of digital tools in HL maintenance were analyzed by comparing to other options.

In order to answer the last research question on the attitudes of parents towards their children’s HL use and development, the items from 21 to 26 were analyzed. These items

23

provided us information about parents’ beliefs and attitudes towards HL use and the underlying reasons of these attitudes.

6. Results

6.1 Effects of Parental Background on HL Maintenance and Practices

In the first part of the questionnaire, demographic information of participants was gathered. This section was included in order to answer the research question “how does the background of the parents and children influence their language practices, maintenance and attitudes in the family?”. This question will be addressed by analyzing participants’ socio-demographic features, such as ages, country of birth, years in the Netherlands, HL and second languages to be able to get an overview about influence of background on language practices. As it can be seen in the table 1, the majority of the participants were female (n=19) and their age ranged from 26 to 60 years. Slightly over half of the participants (n=15) are first generation immigrants and were born in Turkey and migrated to the Netherlands in their early to middle ages. When we look at the languages they claim to speak, we can see that there is only one participant who stated that she is monolingual. The rest of the participants (n=25) consider themselves to be multilingual parents. In each family, at least two different languages are spoken namely Turkish and Dutch. English and German are the other two languages that are spoken by some of the participants as second/foreign languages.

All Turkish parents who were born in Turkey and moved to the Netherlands identified their HL as the Turkish language, regardless of when they moved to the Netherlands. This is also valid for Parent 22 who is 49 years old, migrated to the Netherlands at the age of six and started school in the Netherlands, thus never getting any formal education in Turkey. Even if she migrated to the Netherlands at a very early age, Dutch was assessed as a second language by her. Similarly, Parent 23 (age 47) migrated to the Netherlands when she was three years old, but she also reported Dutch as not being her HL. Six first generation Turkish parents can speak two additional languages other than their HL Turkish. These two languages are Dutch and English. Nine parents stated that they can speak only Dutch as a second language. There was only one participant (Parent 11) who was not able to speak any other language other than Turkish. Since he moved to the Netherlands just three years before, it is understandable that he did not state that he can speak Dutch at the moment.

Regarding the second-generation Turkish parents, we can see that five parents classified themselves as bilingual (Dutch and Turkish as HL), two parents stated their first language as being Dutch and three parents claimed their first language to be Turkish. The majority of the second-generation Turkish parents (n=8) can speak English as a second language. German is the second most spoken language among the participants as a second language (n=4). Overall, Turkish is stated as mother tongue of 72% of the participants and it is the second language of 8% of the participants. 20% of the participants reported themselves as bilinguals (Please see Figure 2). Since 40% of the participants are second generation immigrants, the proportion of bilinguals were expected bigger.

24 Figure 2: Difference between Turkish as HL and Turkish as second language by Turkish

parents living in the Netherlands.

The Turkish language for both first- and second-generation parents was obviously acquired in the family. The two parents (Parent 8 and Parent 10) who stated that Dutch is their first language claimed that they have learned Turkish through formal education and communication with people. Due to the small sample size, we are not able to conduct further analyses in order to check what ways are generally used to learn Turkish as a second language. However, in this research with the current participants, it can be concluded that HL is acquired in the family.

Figure 3: Differences between language preferences of first- and second-generation parents.

To sum up, generation statuses and language backgrounds of the parents were discussed above. In the light of these data, we can say that these factors affect language maintenance of the participants in their family. This result can be obtained from Figure 3. Differences in generations naturally gives rise to diversity in language preferences in the family as well. Although the majority of parents in both generations use Turkish and Dutch at the same time, only Turkish language use in the family predominate in first generation parents. On the other hand, using both languages in the family predominate saliently in second generation parents

72% 8%

20%

Language Background of Parents

Turkish as mother tongue Turkish as second language Bilingual /Turkish-Dutch) 47% 0% 53%

Language Preferences

of First Generation

Parents

Language Preferences

of Second Generation

Parents

25

(see Figure 3). In other words, first generation parents use Turkish more than second generation immigrant which in turn give them more opportunities to develop their HL in the family.

6.1.1 Effects of Parental Background on Children’ HL Education

HL knowledge for the parents is quite predictable as first generation parents were born in Turkey and some of them got their formal education there. For the second-generation children, the use of Turkish as a HL is quite common since they were raised in a family in which Turkish was dominant and their parents were not able to speak Dutch in the beginning of the immigration. However, when considering the third-generation migrant children, knowledge and use of HL is less easy to predict than for their parents. Since priorities and preferences change from family to family in terms of language choice and use, the results for HL knowledge of children show significant differences.

As it can be seen in Table 3, that eight parents provided a formal education for their children’s HL acquisition while the rest did not enable any formal education for HL. The remarkable difference stands out between first- and second-generation parents in terms of providing HL education for their children. There are only eight parents in total who provided formal HL education for their children and seven of these parents are from first generation immigrants. In other words, almost 50% of first-generation parents were in favor of teaching HL formally. However, this situation is quite different for second generation parents since just one out of ten parents provided HL education for their children. We are going to look at the parents’ attitudes towards HL education in detail in another part, yet, we can hypothesize that first-generation parents are more attached to their HL and they still speak Turkish language dominantly in the family. These factors lead them to place more importance to HL education for their children. On the other hand, third generation children generally do not get any formal HL education since their parents were born and raised in the Netherlands with dominance of Dutch.

Figure 4: Differences between first- and second-generation parents in terms of providing

formal education of HL for their children. 7 1 8 9 0 1 2 3 4 5 6 7 8 9 10

First generation parents Second generation parents

Providing Formal Education to Children

26 Table 3: Summary of parental choices regarding HL education.

Parents Number of Children Age of Child 1 Age of Child 2 Age of Child 3 Age of Child 4 Education for HL Acquisition Parent 1 1 1 No Parent 2 1 3 No Parent 3 2 9 4 No Parent 4 1 5 No Parent 5 2 13 8 No Parent 6 4 19 17 11 3 No Parent 7 1 6 No Parent 8 1 1.5 No Parent 9 1 5 No Parent 10 2 30 19 Yes Parent 11 2 26 25 Yes Parent 12 2 34 19 No Parent 13 2 29 19 No Parent 14 1 8 No Parent 15 1 4 No Parent 16 2 30 22 No Parent 17 1 2 Yes Parent 18 1 23 Yes Parent 19 1 2 Yes Parent 20 2 18 15 Yes Parent 21 2 32 25 Yes Parent 22 2 20 17 Yes Parent 23 1 3 No Parent 24 1 5 No Parent 25 3 39 34 24 No

27 6.2 Language competence

Language competence of parents and children mutually affect each other; there are also some external factors that eventually might determine HL use in the family. As it was previously mentioned, language practices in the family have an important influence on children’s competence for HL. Equally, children’s competence and proficiency on HL might have an effect on frequency of HL use in the family which in turn can affect HL competences for both parents and children in the long term (De Houwer, 2007). The reason why language competences were collected was to answer the sub-question: “Does proficiency in the HL affect the frequency of use of the language in the family?”. In the following sections, we are going to look at the parents’ and children’s HL competences and their language choices in order to answer whether proficiency in the HL affect the frequency of use of the language in the family or not. The question to rate their HL competence was formed by restricted items which is also called closed-ended items. “A restricted item with ordered alternatives lists these alternatives in a logical order” (Bordens & Abbott, 1988, p.262).

6.2.1 Parents’ HL competences

Parents were asked to rate their Turkish competence for reading, writing, listening and speaking with seven restricted items. All of the parents stated knowing the Turkish language, but their proficiencies differed for each skill. First of all, we can see that the majority of both first- and second-generation parents reported that they are mostly native/native-like in each skill. Writing the skill rated as least native/native-like with 15 rates followed by reading, speaking and listening respectively. Listening is the most native/native-like rated skill with over 75% of the participants (n=19).

Figure 5: Parents’ HL self-reported competence in reading, writing, listening and speaking. 0% 20% 40% 60% 80% 100%

Reading Writing Listening Speaking

Native/Native like 5 15 19 17 Very Good 3 4 1 1 Good 3 2 3 5 Enough 3 4 1 2 Avarage A little 1 Poor Very poor

28

71% (n=16) of parents’ reading ability is rated as native/native-like while 13% (n=3) of the parents stated their reading ability as very good, 13% (n=3) of them as good and finally 13% (n=3) of them as enough competence. Regarding writing skills, 60% (n=15) of parents reported their Turkish writing skill as native/native-like, 15% of them (n=4) as very good, 10% (n=2) of them as good and 15% (n=4) of them as enough. Listening competence is higher for the parents with the rates 76% (n=19) as native/native-like, 4% (n=1) as very good, 12% (n=3), 4% (n=1) as enough and lastly, 4% (n=1) as a little. Finally, speaking is the second most native/native-like rated skill among parents with 68% (n=17) claiming to be native/native-like, 4% (n=1) as very good, 20% (n=5) as good and 8% as enough competence. When we look at the difference between first- and second-generation participants, we can say that first generation immigrants stated their skills more as native-like than second generation immigrants. Over 70% of first generation immigrants reported their skills as native-like in each category whereas only 50% of second generation immigrants reported their skills as native-like in each skill.

These results were actually expected in the beginning of the research, since formal education of HL for immigrants is not a prevalent phenomenon (Yagmur, 2009). Without a formal education, it cannot be expected to be proficient in a language even if it is stated as HL. It is a well-known fact that minority languages are generally spoken languages for that community. Although there is a quite large Turkish community in the Netherlands, the Turkish language is often spoken in the domestic domain or in the community/ neighborhood (Yagmur, 2009). As such, it is reasonable to assume that listening and speaking competences are higher than reading and writing since they can be more easily improved without formal education.

6.2.2 Children’s HL competence

Children’s HL competence was rated by their parents with the same ranking question. The majority of the parents reported that their children are proficient in the Turkish language (see figure 6). However, when we look at figure 6, there are more variations in the reported HL language skills and there is hardly a pattern for children in these four different skills. We should again take into consideration that these evaluations were provided by children’s families. That is, the data does not depend on an objective measurement or on the children’s own self-rating. Apart from parents’ thoughts about their children's HL competence, any other assessment for them was not implemented (Gkaintartzi et al., 2016).

29 Figure 6: Children’s HL competence in reading, writing, listening and speaking.

Similarly, to the parental data, results indicate that native/native-like skills are rated higher for listening and speaking skills. 32% (n=8) of the children on listening skills and 28% of the children on speaking skills were reported as native/native-like by their parents. When we look at the reading and writing competences, these results show the opposite. 28% of the children on reading skills and 28% of the children on writing skills were reported as having very poor competences. In other words, their oral receptive and speaking skills are rated better than their reading comprehension and writing ability (Gkaintartzi et al., 2016). A small proportion of children are reported as not having listening and speaking competence. When we consider the fact that only eight parents provided a formal HL education for their children, it is an expected result that their HL competences in reading and writing are not as good as that of their parents. However, there is another significant issue that must be taken into consideration and that is that there are some parents having more than one or two children. To be able to make the questionnaire clear and short as much as possible, HL competence evaluation for children was asked just one time regardless of the number of children. As a result, although the language skills of siblings may differ from each other, their parents reported their skills either by focusing on one child or by considering an average answer for this question. That is why, it should be mentioned that the data on children’s’ language skills does not represent accurate results but rather a broad indication.

6.2.3 Language choices and practices in the family

In the second part of the questionnaire, language use in the family was assessed. Two main questions were asked to be able to obtain data regarding linguistic interaction between parents and children: Which language/languages do they speak with their children? and Which language/languages do their children speak to them? Some parents - especially the second generation – stated that they are able to speak different languages other than Turkish

0% 20% 40% 60% 80% 100%

Reading Writing Listening Speaking

Native/Native like 5 5 8 7 Very Good 4 3 4 4 Good 2 2 6 7 Enough 3 4 2 2 Avarage A little 2 2 1 2 Poor 2 2 2 1 Very poor 7 7 2 2