UvA-DARE (Digital Academic Repository) Are traditional minority languages a bench marking for the rights of migrant languages in the European Union?

In this paper a normative position will be defended. We will argue that minimal territorial minority language rights formulated in terms of the personality principle referring to traditional minority languages granted in the framework of the European Union (EU) are a benchmark for non-territorial linguistic rights. Although territorial minority languages should be granted collective rights this is in large parts of Europe not the case. Especially in the Central and Eastern European Member States language rights granted to territorial languages are assigned on the basis of personal language rights. Our argumentation will be elaborated on the basis of a comparative approach discussing the status of a traditional territorial language in Romania, more in particular Hungarian spoken in the Szeklerland area with the one of migrant languages in the Netherlands, more in particular Turkish. In accordance with the language hierarchy implying that territorial languages have a higher status than non-territorial languages both in the EUs and Member States’ language regimes nonterritorial linguistic rights will be realized as personal rights in the first place. Hence, the use of non-territorial minority languages is conditioned much as the use of territorial minority languages in the national Member States. So, the best possible scenario for mobile minority languages is to be recognized as a personal right and receive full support from the states where they are spoken. It is true that learning the host language would make inclusion of migrant language speakers into the host society smoother and securing a better position on the labour market. This should however be done without striving for full assimilation of the speakers of migrant languages for this would violate the linguistic rights of migrants to speak and cultivate one’s own heritage language, violate the EUs linguistic diversity policy, and is against the advantages provided by linguistic capital in the sense of BOURDIEU (1991).


INTRODUCTION
In this paper, we will compare the legal status of traditional minority languages and migrant languages in the European context and its practical implications for the trade-off between mobility and inclusion. 1n fact, below it will be demonstrated that territorial minority languages as traditional minority languages are usually referred to enjoy restricted linguistic rights in the European Member States, where they are spoken.It has been argued that traditional minority languages in Central and Eastern European Member States have less rights compared to official languages and that their status and position is best described by language hierarchies, asymmetries, subordination, and threshold restrictions.This against the background of international and EU treaties, such as article 22 of the European Charter of Fundamental Rights (ECFR) stating that the Union respects cultural, religious, and linguistic diversity. 2lthough linguistic inequality in these cases is an unwanted state of affairs violating the ECFR, and other international treaties and obligations of Member States, it is realistic to suppose that these traditional minority language cases will function as a benchmark for the assignment of linguistic rights to languages of migrants to which we will refer in the remainder of this paper as non-territorial, mobile minority languages.Although among the traditional minority cases of linguistic inequality there are instances of successful practices because peaceful co-existence is guaranteed.This, however is realized at the expense of the traditional minority languages in a language regime that assigns these languages restricted rights only.The language rights of non-territorial minorities in the EU are not only restricted by a language policy favouring full support for the official majority language at the expense of traditional territorial minority languages, but they are further restricted by the mobility-and-inclusion trade-off, which is detriment to the mobile minority languages.
Governments tend to intervene in the language regime on behalf of the official host language of the state at the expense of mobile languages.Therefore, it is safe to hypothesise that the quality of the linguistic rights of non-territorial minorities in the EU or in individual Member States due to a lack of a Union wide supranational language policy for mobile language minorities will be under the assigned level of the linguistic rights which territorial minority languages in the EU or individual Member States enjoy.
Our discussion will take the linguistic rights within the new Member States of Central and Eastern Europe as a starting point, for these states are clearly defined as national states with one official language spoken by the majority population.All other languages spoken in these states are considered minority languages.These traditional minority languages are not integrated into the country's language policy on the basis of the territoriality principle but on the basis of the personality principle, even if the minority language speakers are in the majority in a specific sub-national jurisdiction, such as in the case of the Hungarian speaking Szeklerland in Romania which we will discuss.The application of the personality principle granting a lower ranking official status to the traditional minority language results into bi-multilingual asymmetries, subordinations, and so forth, in reference to the official, majority language.
It is difficult to characterise the EU as a political entity.It can be argued that it is a hybrid structure compromising institutional aspects of the traditional European system of nation states and institutional aspects of federalism, such as the European Commission and other institutions of the EU. 3 In order to understand the politics of multilingualism in the Member States of Central and Eastern Europe, it is necessary to investigate political and institutional aspects of both systems, i.e. the nation state system and the more recent supranational features of the EU.Seemingly, there is a conflict in this system.Most of the traditional nineteenth century nation states are characterised by the concept of "one state-one nation-one language," whereas the supranational aspects of the EU imply a different relation between the different levels of governance.In the European framework, there is no one-to-one mapping between citizenship-institutions and language.Here comes in the politics of multilingualism, for the language regimes and the use of individual languages is depending on power constellations.4Although a complete multilingual regime is guaranteed in the EU by language regulation 1/1958 this language regime has not "percolated down" to the Member States.By and large, a restricted set of languages in the Member States are official and might function as a lingua franca for the citizens that live in the territories where those languages are used.This traditional language situation, characterised by a so-called upgrading of national languages onto the level of the EU, does not hold factually in the Member States of the EU.European language policy is further guided by an educational resolution that was adopted at the Barcelona European Council in 2002, the so-called '1 + 2 formula'. 5The Barcelona formula states that all citizens should be taught to master at least two foreign languages in addition to their mother tongue.A recent resolution on multilingualism adopted by the European Parliament on 24 March 2009 states that that the Union "[r]eiterates its political priority of the acquisition of language skills through the learning of other EU languages, one which should be the language of a neighbouring country and other international (...) 'lingua franca' (...)" (compare article 36).
The conflict consists in the fact that in the traditional model of nation states the matching between the official language of the nation state and the mother tongue of its citizens is taken for granted.The language policy schemes of the EU leave more space for the detachment of mother tongue and official languages, as compared to the traditional nation state scheme.This detachment has been recognised by the Council of Europe, an organisation monitoring the issue of human and minority rights in the Member States of Europe and which is closely involved in the conditioned accession of candidate Member States to the EU. 6 Two conventions of the Council of Europe guarantee the linguistic rights of traditional minorities in the states that are a partner to these legal treaties, i.e. the Framework Convention for the Protection of National Minorities (FCPNM) signed on 1 February 1995 in Strasbourg, and the European Charter for Regional and Minority Languages (ECRML) adopted on 5 November 1992 also in Strasbourg. 7Both charters have been included by the Union as one of the four accession criteria -the so-called Copenhagen criteria -for EU membership.Especially the new EU-candidate countries from Central and Eastern Europe having traditional ethno-linguistic minorities on their territory had to fulfil the obligations of these charters before they could enter the Union.As a result of these supranational interventions concerning member state accession, a supranational linguistic space in the EU has been developing. 8The conflict can be resolved in a flexible framework of multi-level governance (MLG) which is able to absorb the characteristics of the national Westphalian system and the federal aspects of the EU system. 9he idea is that policy is generated as an outcome of the interplay between different levels of governance, distinguishing different tiers that might operate autonomously but also in an interactive fashion, including the supranational Brussels, the national, regional, and local tiers.
Below the working of the European system in the case of linguistic diversity will be discussed and compared.Hence, if linguistic rights of mobile minorities in the EU or in the Member States will be granted, it is to be expected that this will take place in reference to the existing linguistic rights for traditional minorities.A relevant case is presented by the Hungarian speaking minority in Romania's Transylvania.Although the conditions for an application of the territoriality principle are fulfilled, the linguistic rights situation of traditional minority languages in Romania is regulated in accordance with the personality principle. 10This gives us a point of reference of what the limits of the linguistic rights for migrant languages in the European space are.In most European Member States, non-territorial minority languages or the languages introduced by migrants or internally mobile European citizens have hardly or no legal position at all.Realistically speaking migrant languages will not enjoy in the near future a higher level of linguistic rights than traditional 6 See the following references : GÁL 2000;SKOVGAARD 2007;SCHWELLNUSS 2005. 7 See SKOVGAARD 2007;MARÁCZ 2016. 31-32. 8 Compare BRUBAKER 1996;CSERGÕ -GOLDGEIER 2013;MARÁCZ 2014aMARÁCZ , 2015aMARÁCZ , 2015bMARÁCZ , 2016. 9 . 9 See the papers in SCHOLTEN -PENNINX 2016. 10The territoriality versus personality principle was first discussed in McRae (1975).The 'territoriality versus personality debate' is one of the most important debates concerning language policy and language rights, roughly speaking whether rights should be attached to the person, i.e. the so-called 'personality principle', or whether they should be connected to territory, i.e. the so-called 'territoriality principle'.The debate plays a central role in language policy and language planning; it is a key issue in understanding ethno-linguistic struggles and the demarcation of the rights of linguistic minorities.See also LAPONCE (1987) and DEMBINSKA et al. ( 2014) for further discussion.See for an extensive discussion of these principles in connection with the Hungarian language minority in Romania CSATA -MARÁCZ (2016).
minority languages that have a legal position in the Member States.In order to make a realistic estimation of the linguistic rights to be granted to mobile minority languages in the EU or individual Member States these rights will not surpass the linguistic rights of traditional minority languages in national states, such as Romania.Hence, the linguistic rights assigned in those cases will be in terms of the personality principle.
In the course of this paper, we will elaborate on two case studies representative for the linguistic rights situation of a traditional minority language and a mobile minority language, i.e.Romania's minority languages, especially Hungarian and migrant languages in the Netherlands, i.e.Turkish respectively.Comparing the two cases will also give insight into what policy measures can contribute to the optimal equilibrium between mobility and inclusion.

TRADITIONAL MINORITY LANGUAGES IN ROMANIA
There are several territorial minority languages spoken in Romania, including Hungarian, German, Roma, and others.In principle, these minority languages have a similar legal position and have been granted the same type of linguistic rights.Although the conditions for a territorial language regime granting traditional minority languages equal linguistic rights to the majority language are fulfilled, i.e. in sub-national territorial jurisdictions the minority language is in majority only the personality principle is operative in these cases and territoriality is used as a negative restriction due to a threshold of twenty percent of minority language speakers in a certain territorial jurisdiction.Hence, even if there is a majority of minority language speakers in a certain jurisdiction, linguistic rights are only granted at an individual, personal basis and not as a collective right.Let us consider the case of traditional, territorial minority languages in Romania in more detail.
The Hungarian minority in Romania counted by the latest 2011 census 1,227,623 persons who make up 6.5 percent of the total population of Romania.11However, the Hungarian speaking minority is predominately concentrated in the north-western part of the country, i.e. the Transylvanian region, stretching from the Hungarian-Romanian country border to Szeklerland at the feet of the Eastern Carpathians mountains.12This "stroke" is a traditional multi-ethnic region and the ethnic Hungarians are not present in it in equal concentrations.The percentages of the ethnic distribution of ethnic Hungarians and Romanians in Transylvania clearly differ from the national percentages.
In the whole area of Transylvania, ethnic Hungarians make up around 19 percent of the total population, while ethnic Romanians make up around 75 percent.The remaining six percent are other ethno-linguistic groups, like Germans, and Roma.Transylvanian Hungarians inhabit three spatially connected sub-regions displaying a different and heterogeneous geo-ethnic distribution.
The first sub-region 'Partium' is located in the Hungarian-Romanian border area in northwest Romania.In this region, a substantial percentage of ethnic Hungarians constitute a majority in a number of municipalities and districts, except the greater towns like Oradea (Hun.Nagyvárad) and Satu Mare (Hun.Szatmárnémeti).The second sub-region, the area landward is central Transylvania with the major city of Cluj-Napoca (Hun.Kolozsvár).In this region, the ethnic Hungarians are often smaller minorities and live frequently in mixed Hungarian-Romanian-Roma communities, but in some municipalities and districts they can have a relative or absolute majority. 13he third sub-region, which is matching the historical area of Szeklerland is of approximately 13,000 km 2 and consists of the three counties Harghita (Hun.Hargita), Covasna (Hun.Kovászna), and Mures (Hun.Maros). 14 After the collapse of communism, the Romanian language regime anchored in the 1991 Constitution is clearly path dependent on the historical state tradition displaying two "diagnostics". 16The first diagnostic is the language hierarchy.This hierarchy is caused by the official status of the Romanian language in the Romanian Constitution providing the Romanian language a hegemonic position outranking all other minority languages spoken on the territory of Romania, including Hungarian.
The second diagnostic is asymmetric bilingualism.Successive Romanian Constitutions guarantee the language use of the other languages of Romania, including the minority languages, only on the basis of the personality and not on the basis of the territoriality principle. 17Territoriality is relevant for the use of the Hungarian language and other minority languages, but only as a negative condition restricting the personality principle by a demographic factor, i.e. the twenty per cent threshold rule per territorial-administrative unit. 18This implies that Romanian mother tongue speakers do not need to learn or to speak Hungarian, even if they live in administrativeterritorial units with a Hungarian majority, like the counties of historic Szeklerland.However, all Hungarian minority speakers have to learn and to speak Romanian.The latter but not the former is a requirement in the official primary and secondary school curriculum.As a consequence, the Hungarian language use is restricted in scope and although legally not banned from Romanian educational institutions where the language of instruction is Romanian, Hungarian is not offered to their pupils.This leads then to asymmetric bi-and multilingualism.Bi-an multilingual ethnic Hungarians always speak the official language of the country, i.e.Romanian and their own mother tongue Hungarian, whereas ethnic Romanians only speak the official language of the country. 19owever, research makes clear that asymmetric bilingualism is disadvantageous for the weaker language, in this case the Hungarian minority language as opposed to the Romanian state language. 2015 According to CSATA -MARÁCZ 2016, out of the total population of Szeklerland, the ethnic affiliation of 38,096 persons is unknown.These persons were added to the results obtained in the original census survey using a very controversial methodology.So, if we distribute this population according to the ethnic ratios of the original census data, the number of Hungarians would increase with 20,665 persons to 629,698 and their proportion to 58.7 percent.Using the same method of estimation, the number of Hungarians in Transylvania would increase by 56,487 persons to 127,3153 and their proportion would change to 18.8 percent. 16See MARÁCZ 2017 for a detailed discussion of Romanian path dependency.Although Romania's accession to the EU in 2007 is a clear critical juncture in the country's state tradition affecting many features of the political, socio-economic and institutional spheres both "diagnostics" of the traditional language regime can clearly be observed in the societal context. 21he critical juncture of 2007 has not brought a fundamental change in the language policy of the country.There is no equality of the languages used on the territory of Romania, or in its Transylvanian parts where most of the minority languages are being spoken, nor is there symmetric bilingualism in Transylvanian administrative-territorial units with a Hungarian majority, such as in the traditional counties in historic Szeklerland. 22From the point of language policy the accession of Romania to the EU can hardly be called a critical juncture.
It is true that the liberal regime in Central and Eastern European space of which Romania is a part too has some positive side effects on the language use of the minority languages that are due to the liberal democratic and commercial market society empowering the use of minority languages in the public spheres.European transnational actors and structures have positively affected the language situation of minority languages after the accession of Romania to the EU as well. 23fter the collapse of communism and the expansion of the EU eastwards, global and transnational structures have led to the introduction of European human rights norms and standards in the field of minority rights and minority language rights in Central and Eastern European countries.Even more robust policies in support of indigenous minority rights and languages have been adopted by the Council of Europe.All the Member States of the EU are members of the Council of Europe and its arrangements in these domains have been part of the set of conditions to enter the EU. 24onditionality as a part of the EUs accession criteria has guaranteed the implementation of the Council of Europe's minority rights arrangements. 25Of the conditionality requirements two legal treaties of the Council of Europe are relevant in this case, namely the Framework Convention for the Protection of National Minorities (FCPNM) and the European Charter for Regional or Minority Languages (ECRML) signed on February 1, 1995 and November 5, 1992 in Strasbourg respectively. 26he Framework Convention supports the positive discrimination of national minorities on the basis of human rights and general freedom rights, although the explanatory report text is ambiguous on the status of minority groups. 27It emphasises the individual belonging of persons to national minorities and does not recognise the collective status of national minorities. 28The ECRML, "Language Charta" has been motivated by similar considerations.Languages are seen as part 21 See for a discussion of the notion of critical juncture the introduction of SONNTAG -CARDINAL 2015.
22 Compare MARÁCZ 2017. 23Compare MARÁCZ 2014c, 2015c. 24See GÁL 2000, SCHIMMELFENNIG -SEDELMEIER 2005, GRABBE 2006, SASSE 2005, 2008, MARÁCZ 2015b. 25 The Council of Europe has its own sanctioning mechanism through the legally binding judgements within the framework of the European Convention of Human Rights (ECHR) and its Court.Minority rights are included in the ECHR, such as the prohibition of discrimination on the base of language in article 14. Romania is part of this treaty since 1994.
Recently the ECHR has been interpreted also as a collective obligation to protect minority rights.Compare VIZI 2012 for detailed discussion.
of a common cultural heritage and the protection of languages is deemed necessary to counterbalance assimilatory state policy and uniformisation by modern civilisation. 29Note that Romania, just like all the other Central and Eastern European states with Hungarian ethno-linguistic minorities, has also ratified these international instruments, as is shown in Tables 1 and 2: Therefore, national and ethno-linguistic minorities receive recognition and protection from these supranational arrangements, So, the position of the Hungarian minority language is governed by the personality principle, especially in the domains of education and public administration, it enjoys "external," supranational protection from the CoE's treaties and from the transnational spaces which have been created in Europe. 30Due to the transnational spaces information and communication flows via Internet, social media and so on the use of the Hungarian language both in the private and public sphere is heavily intensified.Before we compare the linguistic situation of traditional, territorial linguistic minorities with the one of mobile linguistic minorities let us first discuss in the following section the linguistic rights of mobile minority languages in the EU.The section will focus on migrant languages in the Netherlands, especially Turkish.

MIGRANT LANGUAGES IN THE NETHERLANDS
In this section, the practical angle referring to mobile minority languages will be discussed.Would it actually be beneficial for mobile minorities and the host society, if non-territorial linguistic minorities are granted more rights and facilities in order to construct an optimal equilibrium between mobility and inclusion, or would such a step be counter-productive?This question will be answered by drawing upon historical and contemporary examples from the Netherlands, a country that has received a significant number of immigrants since the 1960s, and has experimented with different cultural and linguistic integration policy frameworks.Furthermore, the experiences from Turkish migrants with Dutch language policy will be discussed.This information is based on a dataset compiled of in-depth interviews collected between 2015 and 2017. 31ince the 1960s, the Netherlands has been, similarly to other Western-European countries, a main destination, for both European and non-European migrants.Due to the rapid growth of the post-war economy and the increasing disinterest of the Dutch population in accepting work at the lower end of the job market, the Dutch government actively recruited temporary foreign workers between the 1950s and in the mid-1970s.In the early periods, most workers originated from southern-European countries such as Italy and Spain, but in the 1960s and 1970s the majority came from states such as Morocco and Turkey.Among government officials the question rose what the appropriate policy response to this newfound ethno-linguistic diversity in the Netherlands should be.One of the first documents that sought to formulate the beginnings of an answer is the "Nota Buitenlandse werknemers" (Note on Foreign workers), drafted jointly in 1970 by four Dutch ministries, including Social Affairs & Health Ministry, Ministry of Justice, Ministry of Economic Affairs, and Minstry for Culture, Recreation and Societal Affairs.The document emphasises the economic nature of the foreign migration flow and discusses potential problems regarding recruitment. 32Dutch policy at the time emphasised the temporary nature of these migration flows, assuming that most foreign workers would return to their countries of origin once their labour contracts had ended.When it came to the adaptation of migrants to Dutch culture and society, the notion of "mutual adaptability" was of key importance.Both migrants and the native Dutch population needed to get accustomed to the new situation.The concept remained vague however, and was rarely backed up by actual policy measures.In practice, the guest worker system revolved around circular migration, with migrants coming to and leaving the country according to the time periods of their labour contracts.This situation changed in the late 1970s.The Dutch authorities acknowledged that many migrants planned to stay in the Netherlands indefinitely, and even made use of their right to family reunification to bring their wives and children to the Netherlands.This new reality called for concrete policy measures.One of the first comprehensive policy briefings is the so-called "Nota Minderhedenbeleid" (Minorities policy document), which continued the philosophy of "mutual adaptation," but also attempted to make it concrete with a set of policy interventions.One of the main interventions announced in the Minorities policy document was the so-called "Onderwijs in Eigen Taal en Cultuur" (Education in mother tongue and culture) policy, also abbreviated as OETC.The main argument to implement this policy was practical: it was assumed by the Dutch authorities, on the basis of socio-linguistic research, that migrant children would be better capable of learning Dutch if they mastered their mother tongue first.This view is still shared by many linguistic and psychological studies.Learning one language that the child is already somewhat familiar with at a proper level makes the transition towards learning a second language, in this case Dutch, much easier.Thus, the Dutch authorities organised classes in the main migrant languages, such as Turkish and Arabic at primary schools, that were specifically designed for the migrant children.The Dutch government primarily perceived itself as a facilitator rather than the main organiser of mother tongue education.Initiatives of immigrant communities to self-organise mother tongue education were financially supported by the authorities, as long as these initiatives did not hinder Dutch language acquisition.This system of shared responsibilities, but with a focus on the role of the migrant communities, produced rather mixed results.In the early 1990s, the main scientific advisory bureau of the government, the "Wetenschappelijke Raad voor het Regeringsbeleid" (WRR) (The Netherlands Scientific Council for Government Policy) critically analysed Dutch integration policies and also discussed OETC extensively.The WRR criticised the unproven effectiveness of the mother tongue education programme and the immigrants' children poor mastery of the Dutch language.Mother tongue education was according to the WRR poorly organised, mainly because of a lack of good study material and, most importantly, a severe lack of qualified teachers.Teachers were often recruited from the migrants' country of origin and usually lacked a connection with the Dutch society and sometimes also did not have a teaching degree.This combination of a poor evaluation of the mother tongue education system, and an increased focus on Dutch language acquisition, inspired governments to gradually deconstruct the mother tongue education facilities, by first reducing its teaching hours and eventually by abolishing it all together in 2004.
This very brief overview of the Dutch situation makes it clear that simply granting educational rights to non-territorial languages is not sufficient to guarantee a vivid linguistic landscape in practice.The Dutch history regarding mother tongue education presents several important considerations for future policy-makers who might wish to experiment with new facilities for immigrant languages.Firstly, a government should not merely be a facilitator, but an active organiser of mother tongue education, to ensure the consistency and quality of the courses offered.Secondly, it is of paramount importance whether a government treats non-territorial language rights as accessory to integration, or as an intrinsic value.The Dutch government pursued a policy of what could be called "pragmatic multiculturalism": mobile minority cultures were nurtured and facilitated, but mainly for a pragmatic reason (i.e.serving as a bridge to Dutch language acquisition).It is therefore unsurprising that after the WRR concluded there was no empirical proof that mother tongue education benefitted Dutch language acquisition in any way, that OETC (later renamed to OALT, 'Onderwijs in Allochtone Levende Talen' [Education in allochtonous living languages] was slowly deconstructed and eventually abolished.There was another reason why the Dutch government abolished the mother tongue education for immigrant children.In-depth studies on the language proficiency of migrants' children, who had participated in home language education in elementary school, radically changed the policy perspective.It turned out that especially Turkish and Moroccan youngsters in their last year of elementary education at the age of twelve faced serious deficient language skills in Dutch compared to their Dutch classmates.Due to the delay in language development, migrant children had to qualify for lower types of secondary education, resulting into a much worse position on the labour market.In 2004 the Dutch government concluded that mother tongue education was a detriment to solving this problem, since immigrant children also faced deficient language skills in Dutch.The Netherlands decided to abolish the state-sponsored heritage language education of migrant children in elementary school.From then on, all educational efforts were concentrated on teaching migrants and their children only Dutch at school.The switch from a multicultural to an assimilatory language policy was thus motivated first and foremost on economic grounds.It was argued that improving Dutch language proficiency among migrants and their children was needed for strengthening their position on the labour market.But also among the migrant communities the opinions with reference to the education of immigrant children in their heritage language, even if it was offered in a restricted way, was also different, as interviews with members of the Dutch Turkish community underline. Migrant communities are naturally one of the most important actors in the non-territorial language right debate.The question whether they actually desire the rights that are (or are not) granted upon them, and how they evaluate the policy measures that are targeted at them is a valid consideration when designing a policy framework.In practice, immigrants and their descendants seem to have a rather mixed view on both the principle of migrant rights and its practical application in the Netherlands.The interviews are part of ongoing field work and analyses attitudes of Turkish (and Polish) migrants in France, Sweden, and the Netherlands.This section will discuss preliminary results of the study conducted in the Netherlands.The interviews were held between 2015 and 2017.Eleven Dutch Turks participated in the research.They lived in Amsterdam or Veendam (near Groningen).The sample is well partitioned in terms of education level, social class, migration generation (first, second and third), age (but all older than 18), and gender.Participants have been selected using a snowball method.It is however complicated, if not impossible, to generalise the results of these interview sessions to the whole of the Turkish community in the Netherlands.The sample size is not sufficient for such a purpose.The value of this research is however (at least) twofold.Firstly, it is one of the few studies about language policy towards immigrants that places the immigrants' perspective at the forefront.Secondly, the in-depth interview method allows for new perspectives to emerge that might have been unknown beforehand to the researcher, hence providing perspectives for new hypotheses and future policy-oriented studies.
The interviews covered two very significant components of language policy towards migrants: citizenship courses and/or training in the host language (here Dutch in the Netherlands) and mother tongue education.Lastly, the interviews covered the connection between the respondents'(perceived) language skills and their opportunities to be mobile within the EU, hereby also explicitly including the role of their mother tongue.
At this place, we will discuss the results pertaining to the topics pointed out above, and if needed in connection with the respondents' migration history.The Dutch Turks that were interviewed were mostly planning to stay in the Netherlands for the mid-long term (at least 5-10 years).This fact alone incentivised all the interviewees to at least gain a basic understanding of the Dutch language, and also make sure their children master the language, either by speaking it actively at home or learning it at school.Especially for the respondents in the lower social classes, Dutch is considered by far the most important language for them to speak, as they see it as a prerequisite for economic and social mobility in the Netherlands.Respondents thus supported the idea of the government facilitating Dutch language acquisition courses.However, Dutch Turks reported a lack of proper organisation of the language acquisition classes, both in the past (from the 1970-80s onward) and in the present.An often cited issue is the lack of an emphasis on oral communication and a too strong focus on teaching the basics of Dutch grammar.Respondents often felt that simply being active in Dutch society by working and socialising with native Dutch speakers would have a significantly more positive effect on their oral communication skills than spending the evenings in a classroom.Views on Dutch language training were quite coherent both within and across the migrant groups.It can easily be summarised as follows: it works very well in principle, but partly fails in practice.
They think the courses would be much more effective if the focus would lie more on oral communication skills.It is beyond the scope of this paper to analyse whether this claim holds true from an educational point of view, but it is at least a perspective to keep in mind when potentially reforming host society language acquisition policies.
Secondly, perspectives on mother tongue education, and multilingualism in general, seems to be a class-issue.Low SES parents are sometimes discouraged from raising their children bilingually or give them a bilingual school education due to the fact they perceive it might hamper their chances of learning Dutch, and thus decrease their socio-economic opportunities in the Netherlands.This perspective is not supported at all by socio-linguistic research.If a large chunk of low SES immigrants persist in this misguided view, the linguistic gap between poorer and richer immigrants may grow, even if immigrant languages are granted extensive rights.The government could counteract this development with good information campaigns and by for example instructing teachers to stimulate bilingual education with parents.Field work conducted in Sweden and France suggests teachers at primary and secondary school play a crucial advisory role for parents.
Thirdly, some immigrants think and see that mother tongue education policies, and multicultural policies in general, can foster segregation.Again, this could be an undesirable side effect of facilitating mother tongue education for migrants.

CONCLUDING REMARKS AND OUTLOOK
The main question that was put forward in this paper is whether the languages of non-territorial minorities could benefit from the same facilities currently granted to traditional, territorial minority languages in the EU.There are different angles from which to approach this topic.In order to answer the central question of this paper, namely 'are traditional minority languages a bench marking for the rights of migrant languages in the European Union?' we have to make a comparative analysis of the two case studies presented here, i.e. the status and position of the Hungarian minority language in Romania and the ones of the Turkish language in the Netherlands.The analysis unambiguously demonstrates that the traditional minority language is higher on the language hierarchy in its own national context than a mobile non-territorial language in its Member State context.
The strongest option for minority language protection, i.e. the territorial option is not even available in the case of traditional minority languages in Central and Eastern European Member States, like Romania. 33However, traditional minority languages like Hungarian in Romania are assigned linguistic rights on the basis of the personality principle., that is in terms of individual "personal" rights, and not in terms of collective or territorial rights, even though in some cases the conditions to introduce such a language regime would be fulfilled leading to more respect for linguistic diversity, the balancing of rights and a more equal position of the traditional minority languages, such as Hungarian in Szeklerland.For now, the benchmark in which the linguistic rights of mobile languages in EU Member States can be drafted is under the denominator of "individual rights." In both cases, the dominant position of official state languages is not challenged.Traditional territorial minority languages enjoy however recognition.The use of these languages is regulated on the level of the Constitution or otherwise in the legal system.The history of Turkish in the Netherlands demonstrates that the recognition of migrant languages in the context of Member States is fuzzy.Although Turkish was recognized as a language in the Dutch educational system without real legal anchoring of migrant languages it was slowly less facilitated in the educational system.However, mobile minority languages just as territorial minority languages within a national context could be included into the language regime on a personal basis.This would however not guarantee equality of the languages involved resulting in asymmetric bilingualism which is at the disadvantage of the minority language.Majority language speakers, i.e.Romanian in Romania and Dutch in the Netherlands do not speak the minority languages but vice versa is normally the case.Setting asymmetric bi-and multilingualism aside recognition of migrant languages on the personality basis would satisfy the concept of "linguistic diversity" celebrated in the European Union, although in a somewhat limited form.
Recognition of the traditional minority language in the Member States' legal system also triggers recognition at the supranational level in terms of European minority and minority language treaties, as discussed above.Migrant languages completely lack recognition at the European level.It is reasonable to suppose that these languages could be integrated under the umbrella of the Language Charta of the Council of Europe.What makes a pan-European regulation and implementation complicated is what the interviews with the Dutch Turks unambiguously demonstrate, although due to the limited scope of the research presented here it might be too early to draw definite conclusions.In any case, the speakers of migrants languages themselves, like Dutch Turks do not have a uncontested position on the protection of their own linguistic heritage.Note, that this is very different from the case of the traditional minorities which have their own interest organizations that lobby for more linguistic rights. 34Hence, there are clear constraining circumstances that do not support the practical implementation of migrant languages in the educational system of a host country.
There is an unjustified but persistent belief among policy makers and a part of scientific researchers that mother tongue education of the heritage language does not benefit language acquisition of the host language and spoils changes of migrant language speakers on the EU labour market.Interviews show that even some parents of migrant children from low SES are in principle not in favour of extending language rights for their own heritage language because they think that mother tongue education will lead to unsuccessful integration and negative host language acquisition results.
Hence, a way out of this situation more in line with respect for the principle of linguistic diversity in Europe that has been the official stance of many different organs of the EU in a number of its official documents (see a collection below) and BOURDIEU (1991) idea of linguistic capital.BOURDIEU (1991) argues that the more languages you speak the more your financial-economic value will be at the "language market" as an individual citizens of a country.By providing the facilities for reciprocity it would imply access to hidden, untapped resources of linguistic capital in the sense of Bourdieu. 35 The official and educational authorities will have to counterbalance misguided views that plurilingualism narrows the chances for integration and a better position on the labour market.An integration policy without full linguistic assimilation should also be granted to children from migrant and intra mobile European citizen families by introducing language learning facilities for mobile minority languages in educational curricula and open these courses also for children speaking the host language.In fact, to teach these languages in e.g. the elementary school context as it was done in the Netherlands until 2004 and is still done in Sweden.This implies that there is no segregation in the educational context but reciprocity would guarantee that the teaching of these mobile minority languages is available in elementary school education where these languages or other languages of wider communication could be offered for both children of newcomer families and children speaking the host language.
The question is whether social cohesion which is normally safeguarded in national states by speaking and using the official language of the state will be undermined by introducing reciprocity in language education.PUTNAM (2007) argues that a high degree of ethnic and cultural diversity is associated with a decrease in social trust.In other words: in diverse neighbourhoods all citizens, be they part of a majority or minority group, tend to become more individualistic.Social capital for the population as a whole thus decreases, which logically leads to lower levels of social cohesion.To counteract this trend, Putnam pleads in favour of reinvigorating institutions that can positively influence social capital, such as community centres and sport associations.He also argues strongly in favour of solid, state sponsored courses in the host language, so all citizens have at least one common language they can use to communicate.Putnam's solutions are steering in the right direction.We also deem good quality host language courses of the utmost importance.Nonetheless we would like to propose an addition to his solution.It is our claim that if reciprocity is introduced in the language educational system and autochthonous children learn to speak mobile minority languages as well social cohesion does not need to be undermined in a context of symmetry.The recognition of the symbolic value of mobile minority languages could contribute to a quick inclusion and integration of the children from these groups into society.
The introduction of mobile minority languages in the educational curricula should further be conditioned by linguistic subsidiarity which is also conditioning the language use of territorial minority languages.Granting linguistic rights to mobile minority languages should only apply in cases, when the mobile minority languages are spoken in sufficient numbers and in a territorial concentration, such as urban spaces.In sum, in order to guarantee the linguistic rights of territorial and non-territorial minority languages state recognition, sponsoring and intervention are highly relevant; reciprocity is considered as a contribution to social inclusion and does not have to be viewed as a threat for social cohesion; and linguistic subsidiarity and multi-level governance are interpreted as relevant for securing linguistic diversity.Hence, state intervention, reciprocity, and linguistic subsidiarity provide background facilities that make possible a language policy that strives for the optimal equilibrium between mobility and social inclusion.