The Turkish Diaspora in Europe
In this article
Introduction and summary
More than 5 million people of Turkish descent live in Europe outside Turkey itself, a human connection that has bound Turkey and the wider European community together since large-scale migration began in the 1960s.1 The questions of immigration, citizenship, integration, assimilation, and social exchange sparked by this migration and the establishment of permanent Turkish diaspora communities in Europe have long been politically sensitive. Conservative and far-right parties in Europe have seized upon issues of migration and cultural diversity, often engaging in fearmongering about immigrant communities and playing upon some Europeans’ anxiety about rapid demographic change. Relations between the European Union—as well as many of its constituent member states—and Turkey have deteriorated dramatically in recent years. And since 2014, Turks abroad, in Europe and elsewhere around the world, have been able to vote in Turkish elections, leading to active campaigning by some Turkish leaders in European countries. For these and several other reasons, political and academic interest in the Turkish diaspora and its interactions with European society and politics has significantly increased in recent years.
The Turkish and Turkish-Kurdish diaspora feels at home in Europe overall, its members expressing high levels of satisfaction with their living circumstances and general contentment with host nations’ integration policies.2 Ethnic Turks and Kurds3 living in Germany, France, Austria, and the Netherlands feel their presence is generally accepted by their non-Turkish and non-Kurdish neighbors and colleagues, and they are pleased with the educational and economic opportunities the host nations offer. For most, these positives outweigh the still meaningful levels of discrimination they encounter in their daily lives.
The diaspora is largely uninterested in European politics, with few strong grievances against the authorities and little involvement in party politics in the countries in question. Diaspora communities in France and the Netherlands appear more fully integrated into those societies than do communities in Germany and Austria. Nevertheless, across the board, most in the Europe-based Turkish diaspora continue to identify themselves first and foremost as Turks rather than as full members of the societies they inhabit, and they remain more absorbed with developments and politics in Turkey than in their current countries of residence. In short, they strongly endorse, implicitly and explicitly, Turkish President Recep Tayyip Erdoğan’s maxim that Turks in Europe should “integrate but not assimilate,” even as their precise understanding of that phrase is open to interpretation.4
These findings, and many others, are revealed in a public opinion survey of Turkish diaspora communities in four European countries, conducted from November 2019 to January 2020 by the Center for American Progress, the Foundation for European Progressive Studies (FEPS), the Foundation Max van der Stoel, and the Fondation Jean-Jaurès.5
This report builds upon previous CAP public opinion research on national identity in Turkey6 and casts light on how the Turkish diaspora feels about European host countries; how Europeans of Turkish descent identify with Turkey and the Turkish community; and how these communities feel about the European Union and crucial issues of integration, migration, and politics. The goal is to help readers better understand the political dynamics within Turkish communities in Europe, where these communities get their information, how they interact with the non-Turkish community and the state, and how the diaspora interacts with Turkey’s politics.
By providing a foundation of data, the authors hope this research can contribute to the ability of policymakers, academics, and others to address linkages between developments in Turkey and the European Union and, potentially, produce ideas that can improve European policies vis-à-vis Turkey and the Turkish diaspora. The authors also hope the data will bolster research more broadly into the development of national, ethnic, and religious identification among first-, second-, and third-generation immigrants in Europe.
About the survey
The survey was conducted by the polling firm DATA4U from November 2019 to January 2020 using computer-assisted telephone interviews on landline and mobile phones in four countries: Germany (1,064 respondents), Austria (416 respondents), France (452 respondents), and the Netherlands (425 respondents). All respondents are over age 18. The survey is representative of the roughly 5 million people of Turkish origin in these four countries, of which around 4 million are over age 18.7 For the total sample, there is a margin of error of plus or minus 2 percent; for the German sample, there is a margin of error of plus or minus 2.9 percent; and for the Dutch, Austrian, and French samples, there is a margin of error of plus or minus 4.7 percent. The interviews were mainly conducted in Turkish: 84 percent were conducted exclusively in Turkish, and 90 percent were conducted at least mostly in Turkish.
Respondents were screened to select for Turkish or Kurdish migration backgrounds. The sample is balanced by gender, age, and place of residence. Overall, about 60 percent of the sample was born in Turkey. Respondents reported living in their current country for an average of 27.5 years, with slight variation between countries; this includes 38 percent of the sample who were born in a host country. The overall sample comprises 43 percent of respondents holding only Turkish citizenship, 22 percent holding dual citizenship in their host country and Turkey, and 33 percent holding only host country citizenship.8 (For more on this, see the Appendix) The average age of the respondents was 41.
The German sample is intentionally larger to reflect the outsize presence of the German-Turkish community in the overall European-Turkish diaspora, with the 1,064 respondents representing slightly more than 45 percent of the total survey. The overall sample was 48 percent female and 52 percent male, with slight variations by country. Nearly half of the sample reported full-time employment, and the average reported net household monthly income was 2,520 euros—about $3,000 U.S. dollars—some 10 percent to 15 percent below the average for the wider population in each country.9 The average household size was 3.52 people, substantially above the average household size of the overall population in each country of 2.0 to 2.2 people. The French sample was significantly more educated than the other three,10 while the Austrian and Dutch samples were somewhat wealthier.
One in five Turks living in the sampled countries say they plan to return to Turkey to live, while 72 percent want to remain in their current country of residence. The proportion of Germany-based respondents who say they plan to return or move to Turkey is slightly higher, at 24 percent, than in the other countries.
Most respondents identify primarily as a Turk—72 percent overall—and few identify primarily as a member of the host nation, though the French diaspora community is somewhat distinct in this regard, with a larger minority identifying primarily as French. There is considerable diversity, variation, and nuance in respondents’ answers to the questions about what is important to them. Yet the concepts of “Turkishness,” religion, and passing down Turkish traditions to the next generation are all very important to respondents, with all of these being given significantly more importance than identification with the host nation.
Responses on the use of language reveal a clear—and unsurprising—divide between the language used at home and that used at work: Most respondents speak the language of the host country at work but prefer Turkish at home. They are fairly split in the language they use to get their news, but Turkish is clearly favored when it comes to entertainment. Respondents rate their knowledge of Turkish highly. (Only 6 percent of the respondents self-identify primarily as Kurds, and they likewise rate their knowledge of Kurdish highly, though not quite as highly as Turks rate their Turkish.) Not surprisingly, the younger generation reports being more proficient in the language of their host country than older members of the diaspora, who are less likely to have been immersed in it from an early age.
Respondents are divided in their sources of information. Television predominates over internet news, social media, and newspapers, but the media environment overall is fractured—a fragmentation that is also visible in Turkey itself.11 Turkish-language television is the most widely consulted, while very few respondents read Turkish-language newspapers. There is a high degree of interest in news about Turkey, far more than there is in news about the country of residence. Younger respondents are somewhat less focused on news from Turkey than are older respondents, but they still register as much interest in news about Turkey as about their country of residence.
In general, respondents have positive views toward their own Turkish community in their host country, positive views of the local non-Turkish population, somewhat positive views of the Kurdish community, and only slightly positive views of non-Turkish migrants and refugees.
Members of the diaspora community say they perceive a fair amount of discrimination toward Turks in their host country, but few respondents report being personally insulted or physically attacked for their ethnicity. Many respondents feel that discrimination hurts their career prospects. Views diverge on whether the host country government treats the Turkish community equally with the majority community.
Respondents overwhelmingly say they are happy living in their current country, but a majority—albeit a somewhat smaller one—also say they would be happier in Turkey. Most respondents say their current country is more democratic than Turkey. Nevertheless, most respondents would like to see their host country be more supportive of Turkey. Better bilateral relations, they seem to feel, would mean a better situation for Turks in their current country. Somewhat contradictorily, respondents are divided on whether it is important to defend Turkish policies themselves—and, interestingly, very few say they feel pressure to do so from Turkish government representatives.
Responses are mixed on whether their current country has done a good job integrating Turkish immigrants. Although most respondents say they “feel at home” in their host country, a strong majority say that the Turkish community should be more connected to the non-Turkish community. But equally, a very strong majority of respondents say that the Turkish community should retain its separate identity. These competing desires for connection and separate community affinity seem to reinforce a reflexive adherence to the idea of assimilation without integration—a notion promoted by Turkish President Recep Tayyip Erdogan and explicitly embraced by a broad majority of respondents.
Turks who lack citizenship in their host country are more critical of the host country’s integration efforts than those who are already citizens. Indeed, those who have citizenship are more positive about virtually every aspect of life in the host country. This is perhaps to be expected, since citizenship is among the ultimate measures of full integration on an individual level.
There is strong agreement that respondents’ children receive a good education in their current country. Respondents also feel strongly that the schools in their current country are better than schools in Turkey. Few would prefer their kids to grow up and be educated in Turkey. A strong majority of respondents feel that Turks have a fair chance to attend university in their host country. Views are mixed on whether there is sufficient access to Turkish-language and Islamic education, but few would prefer that Islamic schooling replace public schooling for their children.
Perhaps because of this overall contentment with life in their current country, and despite most respondents reporting living in their current country for a long time—27.5 years, on average, in the whole sample—most say they do not participate in the politics of their current country. Related to this, most even say they do not feel politically represented in their current country; citizens feel more represented than those who only hold Turkish citizenship, but even most citizens report a lack of a sense of representation.
Overall, there is little interest in European politicians and, in general, limited engagement with European politics and political parties, visible in high nonresponse rates to questions about those topics. In general, across countries—and among the relatively limited number of respondents who express opinions on these matters—the Turkish diaspora tends to favor left-wing politicians and political parties, usually has a mixed view of the current leader of the country, has little opinion of most other national and local politicians, and harbors understandable hostility toward far-right xenophobic leaders and parties.
Respondents are divided on whether they are proud to live in an EU country, divided on whether the European Union serves their economic interests, and divided on whether Turkey should become an EU member. But despite this ambivalence, respondents are generally more satisfied than not with the European Union.
When it comes to a football (soccer) match between Turkey and respondents’ current country—sometimes seen popularly as a measure of spontaneous identification—76 percent say they would support Turkey, 5 percent say they would support their current country, and 11 percent say they would support both. But when their current country plays a third country that is not Turkey, 79 percent say they would support their current country, while just 3.5 percent say they would support the other country.
The survey reveals mixed views on Turkish politics, including on President Erdoğan personally and on whether he cares about the welfare of Turks in Europe. But Erdoğan is more popular than any other Turkish political figure polled, including main opposition leader Kemal Kılıçdaroğlu, nationalist leader Devlet Bahçeli, and jailed Kurdish political leader Selahattin Demirtaş. Not surprisingly, respondents have very negative views of PKK leader Abdullah Öcalan, though almost half of the 6 percent who self-identify primarily as Kurds have a positive view of him.
Among the roughly 66 percent of the sample holding Turkish citizenship, a clear majority (roughly 56 percent) say they voted in the 2018 Turkish elections. Their self-reported preferences roughly mirror those in Turkey, although it is striking that the ultra-nationalist right wing seems to have minimal appeal in the diaspora: Among those who say they voted, the Justice and Development Party (AKP) took 51 percent, the Republican People’s Party (CHP) took 30 percent, the Peoples’ Democratic Party (HDP) took 10 percent, and other parties collectively took but 9 percent. Their self-reported votes in the 2018 Turkish presidential election followed a similar pattern, and responses were roughly in line with the reported votes for the respective parties in parliament.
Detailed findings and country analyses
Identity, language, and citizenship
The survey captured differences in the composition of the diaspora communities in the four countries, including substantial variations in citizenship dynamics and subtle differences in language. But a clear finding across all four communities is the enduring power of Turkish identity and language.
Austrian respondents are more likely to report having been born in Turkey—at 75 percent—than those from the other countries surveyed (ranging from 50 percent to 64 percent). The German respondents, on the other hand, are more likely to have been born in Germany—at 45 percent—than respondents in the other countries; Austria was on the low end with 23 percent Austrian born, and France and the Netherlands were in the mid-30 percent range. The German respondents have also spent more time in Germany.
There are also differences in the citizenship composition of the samples (Turkish vs. host country vs. dual); fewer than half the German respondents have German citizenship, the only country for which this was true. The Dutch respondents are the most likely to have citizenship, with 79 percent holding Dutch citizenship and 52 percent also holding Turkish citizenship. For more on the dynamics behind these citizenship figures and the migration history that defines them, please see the Appendix.
On the question of identity, respondents rated the importance of different aspects of their identity on a scale from 1 to 10, with 1 being the least important and 10 being the most important. Traditional sources of identity—including ethnic background, religion, and cultural traditions—are quite important to nearly all respondents, though relatively less so to those in the French sample. In general, across questions, younger respondents place less emphasis on the various components of Turkish identity surveyed, while less educated respondents and those with less knowledge of the local language place more emphasis. Respondents uniformly place a very high importance on their Turkish or Kurdish identity, with an overall weight of 8.70.
Religion is also deemed very important, given an overall weight of 7.84, ranging from a high of 8.59 in Austria to a low of just 6.20 in France. As with ethnic identity, older respondents place more importance on their religion than do younger respondents. Respondents say it is very important to maintain Turkish/Kurdish traditions and pass them on to their children, giving this an overall weight of 8.57, ranging from 8.80 in Germany to 7.88 in France. Again, the older generation cares more about passing on Turkish traditions than do the younger respondents.
Respondents place less emphasis on their host country identity—German, Austrian, French, or Dutch—with an overall weight of 5.92, with the French and Dutch respondents placing more emphasis, at 6.54 and 6.41, respectively. Here, the age breakdown is reversed, with younger respondents placing noticeably more emphasis on their European identity than their older counterparts, though still less emphasis than they place on their Turkish/Kurdish identity. Again, these results reveal a more integrated French diaspora community; French respondents place relatively more importance on their French identity, relatively less importance on being a Turk or Kurd, relatively less importance on their religion, and relatively less importance on passing on their traditions to their children, though these factors are still important to them.
In terms of language, again, most respondents speak the language of the host country at work but prefer Turkish at home. This is not surprising, as most of the younger generation of the diaspora are fully comfortable in the language of their host country—used in school and in their careers—but will often return in the evening to multigenerational households in which the older generation is more comfortable using Turkish, having come to the host country as adults. The French respondents also appear more linguistically integrated: A higher percentage report using French in their daily life, including for reading and entertainment, than respondents in other countries report using their host country languages.
Media, news, and information
Respondents rely on a variety of news sources to get information about politics, business, and social issues but place a heavy emphasis on Turkish-language sources in all four countries. The Austrian diaspora community relies the most on Turkish-language sources—across platforms—while the French community is the least reliant on Turkish sources. Older respondents tend to rely more on Turkish-language sources, while younger respondents are more likely to consult news sources in the language of their current country. Less-educated respondents are far more likely to rely on Turkish-language sources, as are those who only hold Turkish citizenship and those with lower incomes.
These trends are particularly stark regarding television. Turkish-language television is the most influential source across the survey, with an average weight of 6.79 out of 10; French respondents place slightly less importance on Turkish television sources, with a weight of 5.50. Television news in the language of the host country is given significantly less importance, with an overall average weight of 4.58; Austrian respondents ascribe the least weight to host country news sources, with a weight of just 3.29. The importance of Turkish-language television may be generationally tied to the medium itself: Television is generally preferred by older respondents, who are also less likely to be fluent in the language of the host country.
Regarding online news sources, the gap between Turkish sources and sources from the host country narrows significantly, perhaps because younger respondents use those sources more frequently and are also more likely to be fluent in the language of the host country. Respondents say Turkish-language internet news portals play an important role, with an overall average weight of 5.73 across the surveyed countries, compared with 5.51 for internet news portals in the host country’s language. There are some differences between the countries on this question, however, with Austrian respondents more reliant on Turkish internet sources, German respondents less reliant on internet sources overall, and both French and Dutch respondents more reliant on French and Dutch internet news sources than Turkish-language sources. These results should not be overinterpreted, given how small the differences are, but they again point to slightly more integrated French and Dutch diaspora communities. As indicated, younger respondents rely more heavily on online sources than their older counterparts, and younger respondents tend to favor host country internet sources.
In terms of reliance on social media sources for information about politics, business, and social issues, respondents again rely more on Turkish-language sources in all four countries, but the differences are small, except in Austria. Turkish-language social media sources are given an average weight of 5.74 across the survey, compared with a weight of 5.32 for social media in the language of the host country. The split is widest in Austria, where respondents are far more likely to consult Turkish social media sources than Austrian-German sources. Across all four countries, younger respondents rely more heavily on social media than do their older counterparts, and host country social media sources are the primary news source for those ages 18 to 29, though a significant minority of the younger demographic relies primarily on Turkish-language social media sources. This finding indicates that even for younger respondents—who are more integrated into the language, society, and information ecosystem of the host countries—the Turkish language remains a powerful point of distinction from the wider society.
As in most markets around the world, the survey shows much less use of newspapers and magazines overall. But perhaps due to reduced access to Turkish newspapers, this is one area where a host country medium proves slightly more important than the counterpart Turkish-language medium: Across the survey, newspapers in the language of the host country have an average weight of 3.96, while Turkish-language newspapers have an average weight of 3.56. Germany is the exception, with Turkish-language newspapers narrowly edging German newspapers. French respondents favor French newspapers, with a weight of 4.96, whereas the Austrian sample relies less on Austrian newspapers, with a weight of 3.54. Again, these data are somewhat indicative of a slightly more integrated French diaspora community and a somewhat less integrated Austrian diaspora community, but the differences are not large. The primary takeaway is that the Turkish diaspora is, unsurprisingly, affected by worldwide shifts away from newspapers to other news sources.
The survey asked respondents how closely they follow the news from different areas—specifically, Turkey, the Kurdish regions, the host country, and other European countries and the wider world. It found uniformly high interest in news about Turkey, with an average of 7.87 across the countries surveyed and little country-to-country variation. Young, more educated, and higher-income respondents are relatively less interested in news from Turkey but still register a high absolute level of interest. There is less interest in news from the Kurdish region, with an average weight of 5.44 and little variation—again, no doubt, reflecting the limited number of Kurds among the respondents. There is reasonably high interest in news about the host country, with an average of 6.58 across the survey; French respondents are the most interested in host country news, with a weight of 7.37, compared with just 6.20 in Austria and 6.34 in Germany. Respondents are least interested in news about other European countries and the wider world, with an average weight of just 5.35; Turks in France show the most interest with 6.02, while Germany has the least at 4.82. These findings point, again, to a more educated and integrated French diaspora community.
Community relations and discrimination
Respondents are generally positive toward the wider population of their host country—i.e., not Turks, Kurds, or refugees—with an average favorability weight of 7.13 out of 10. Respondents in the more integrated communities in France and the Netherlands are even more positive, with weights of 7.49 and 7.51, respectively, whereas Turkish respondents in Germany and Austria rate the host population at 6.99 and 6.71, respectively. There are no meaningful differences along generational lines. The mostly Turkish respondents have slightly less favorable views of Kurds in the European countries surveyed, with an average weight of 6.04, and are also ambivalent toward refugees in the country of residence, with an average weight of 5.35.12
This generally positive view of relations between the diaspora community and the wider national community is complicated somewhat by responses to a series of specific questions about discrimination, which paint a more nuanced picture. The survey asked how much a series of statements applied to the respondent, with 1 being “not at all” and 10 being “very much.” Respondents generally agree that there is discrimination against Turks and Kurds in their country of residence, with a weight of 6.17. There is variation among countries on this score, with respondents in Germany (6.75) and Austria (6.78) reporting higher levels of discrimination than respondents in France (4.71) and the Netherlands (5.71). Younger respondents are more likely to say there is discrimination, and men are more likely to do so than women.
At least at first glance, it is surprising that younger respondents, by most measures more integrated into host country societies than their elders, would perceive greater discrimination. As among the first to arrive in the host country, older residents have likely experienced equal or greater levels of discrimination. One might speculate that it is precisely because they feel more at home in the country that younger generations have higher expectations of equal treatment; better understand subtle forms of discrimination, thanks to their greater linguistic and cultural awareness; and feel more comfortable speaking out. In contrast, those from earlier generations—who, the survey shows, feel less at home in their adopted country—may have few expectations from the host country other than the right to earn a living; less comprehension of the cultural environment; and, for those raised under more authoritarian governance in Turkey, perhaps a lingering sense that criticizing the host country, even to a pollster, could lead to negative consequences.
This trend holds when respondents are asked if they have ever been personally insulted or physically attacked by xenophobes; Germany reports the highest incidence (3.41), followed by Austria (2.78), France (1.69), and the Netherlands (1.48). Men and younger respondents are again more likely to report such attacks. The trend holds again when respondents are asked if they have been discriminated against because of their Turkish name or origin, with Germany reporting the highest incidence (4.48) and the others following in the same order as above, with an overall average of 3.89. In sum, respondents tend to see ample evidence of discrimination against the diaspora as a group, particularly in Germany and Austria, but do not tend frequently to experience the worst forms of discrimination personally.
To get at these issues another way, respondents were asked how they feel they are perceived by non-Turkish locals in their host country. Overall, they estimate that 46 percent of locals view Turks/Kurds as their equals, whereas 43 percent harbor racist feelings toward them. By country, the respective numbers are 45 percent and 46 percent, respectively, in Germany, 39 percent and 50 percent in Austria, 48 percent and 34 percent in France, and 54 percent and 37 percent in the Netherlands.
As elsewhere, these findings are consistent with the general picture of a French and Dutch diaspora that is more fully integrated into the life of the host country and happier with their circumstances and with the Turkish-origin community’s relationship with the wider population than are their counterparts in Germany and Austria. But they also point to the ambivalence with which many Turkish immigrant communities see the wider population—significant minorities in France and the Netherlands, a plurality in Germany, and an outright majority in Austria feel that non-Turkish locals harbor racist feelings toward them.
Moreover, members of the Turkish/Kurdish diaspora feels that discrimination and racism do affect their career prospects, with an average agreement of 3.61 across the sample; again, the highest levels were in Germany (4.05), and the lowest levels were in the Netherlands (2.91). And again, younger respondents and men are more likely to say discrimination affects their career. The diaspora community is divided on whether the government in their country of residence treats all citizens fairly, with an average weight of 5.39 agreeing with that notion but with significant numbers of responses grouped at either end of the spectrum. The responses are fairly consistent across three countries; Austria is notably lower with a weight of 4.71.
Perhaps the best news in the survey is that respondents strongly agree that their German, French, Dutch, and Austrian neighbors and colleagues accept their presence, with an average weight of 8.83 across the survey and consistently high levels of agreement across all four countries and all demographic categories. This response is the closest to a note of unanimity in the entire survey and an impressive indication of a feeling of belonging or acceptance.
Integration and immigration
The survey posed a range of more specific questions about integration, immigration, and relations between the diaspora communities and the host communities in their countries of residence. The results present a nuanced picture of a diaspora that, as indicated, feels at home in Europe, despite occasional discrimination, but also values its separate identity. Respondents explicitly say they feel at home in their current country of residence, with a weight of 7.58, ranging from a low of 7.15 in France to a high of 8.10 in the Netherlands. France’s relatively low score on this question is a surprise, given that the French sample appears more integrated than the others, as already indicated. There are not large differences by age. Women tend to feel more at home than men, which lines up with the above finding that men, more than women, report higher levels of discrimination. Respondents are divided on whether their country of residence has done a good job integrating the Turkish community, with an average agreement of 5.31, ranging from a low of 4.61 in Austria to a high of 5.69 in the Netherlands—again, with women more positive than men.
Two basic pillars of the “integrate but don’t assimilate” attitude are visible in this section of the survey. On the one hand, respondents strongly agree that the Turkish community should be more connected to the wider, non-Turkish community, with a weight of 7.83, ranging from a low of 7.54 in Germany to a high of 8.27 in Austria. Interestingly, older respondents are more likely to support deeper connections with the non-Turkish community. On the other hand, respondents also strongly feel that the Turkish community should keep its separate identity, with an overall weight of 8.46, ranging from a low of 7.77 in France to a high of 8.90 in Austria.
Respondents are divided about further immigration to their countries of residence; overall, they feel their countries should accept fewer immigrants, with a weight of 5.68, ranging from a low of 4.68 in France to a high of 6.01 in Germany. France is the only country where respondents favor more immigration, albeit narrowly.
Significantly, across all questions on integration, the largest demographic schism is between those with citizenship in their country of residence and those without. It is perhaps to be expected that those lacking but desiring citizenship would be more critical of the pace of progress toward integration.
Views on education
The survey also queried views on education and schooling, given how important schools are for integration, assimilation, and interaction between immigrant communities and the wider population. Respondents strongly feel that they or their children receive (or received) a good education in schools in their current country of residence, with an overall weight of 8.05—ranging from a low of 7.84 in Germany to a high of 8.31 in the Netherlands. Respondents also strongly agree that the schools in their country of residence are better than those in Turkey, with an overall weight of 7.50 and little variation among countries. Respondents do not prefer that their children grow up and be educated in Turkey; that idea rates an average of only 4.68 overall, from a high of 5.16 in Austria—the only country where respondents agree overall, if marginally—to a low of 4.04 in France. Older respondents are significantly more likely to say they prefer (or perhaps, would have preferred) that their children grow up and be educated in Turkey; indeed, those above age 50—and the Austrian sample overall—are the only subsets that rate this idea positively.
Respondents are generally pleased with the opportunities for higher education in Europe; they strongly agree that ethnic Turks have a fair opportunity to attend university in their country of residence, with an overall weight of 7.91, ranging from a low of 7.68 in Germany to a high of 8.22 in the Netherlands. Respondents are broadly satisfied with the level of access to Turkish-language and Islamic education in their countries of residence, with a weight of 6.34. Few prefer that their children go to an Islamic school full time—a notion given an overall weight of 4.05, ranging from 4.68 in Germany and 4.83 in Austria, to 3.57 in the Netherlands and just 2.47 in France. Older respondents are more likely to prefer an Islamic education, with roughly half of those above age 50 expressing that desire (or perhaps regret, on the part of those with children beyond school age) for their children. Overall, across these various questions about education, the younger, more educated, and wealthier respondents—as well as the Dutch sample across the board—are generally the most positive elements in an overall positive result regarding educational opportunities in Europe.
Sports and national affinity
To gauge broad affinity for national symbols and emotional ties to Turkey among the diaspora, the survey asked a few questions about sports. Asked who they support when Turkey plays their current country of residence in a sports competition—for example, an international football (soccer) match—overall, 76 percent say they support Turkey, 11 percent say they support both, and 5 percent say they support their host country. France is the only (slight) outlier here, with just 65 percent of respondents saying they support Turkey, 17 percent saying they support both countries, and 9 percent saying they support France. Perhaps this greater support for France is explained by the team’s 2018 FIFA World Cup victory—or simply by the community’s higher degree of integration in France. Just 2 percent of respondents in Austria say they support Austria, with 81 percent saying they support Turkey. Nevertheless, respondents reject the proposition that an ethnic Turkish athlete from Europe should feel obligated to play for Turkish national teams (which eligibility rules allow); agreement with that notion rated only 4.02 on a 1–10 scale, ranging from a high of 4.40 in Germany to a low of 3.27 in France.13
To further query relative attitudes toward their current country of residence, the survey also asked who they would support in a sports competition against a third country—for example, Germany against Spain, or Austria against England. Here, the country of residence earns the most support, with an average of 79 percent saying they would support their country of residence against a third country, 5 percent saying they would support both, and just 4 percent saying they would support the third country. A low of 75 percent in France and a high of 88 percent in the Netherlands say they would support the host country.
Quality of life in Europe
Members of the diaspora generally show high levels of satisfaction with life in Europe and their respective countries. The survey asked respondents to weight how strongly they agreed with different statements, with 1 signifying strong disagreement and 10 signifying strong agreement. Respondents say they are happy living in their country of residence, with an average weight of 7.68; Dutch respondents are the happiest, with a score of 8.01, but there is little variation across countries. Respondents give a decidedly lower rating to the idea that they would be happier living in Turkey, albeit with a not insignificant weight of 5.57; Austrian respondents are the most likely to say they’d be happier living in Turkey (6.01), while French respondents are the least likely (4.65). Older respondents are much more likely to say they would be happier living in Turkey. Meanwhile, younger, wealthier, and more educated respondents are happier with life in Europe. These answers are polarized, with many either strongly agreeing or strongly disagreeing. Most respondents feel their country of residence is more democratic than Turkey, with a weight of 7.07, though that weight fell to 6.72 in Austria.
Respondents generally agree that relations between their country of residence and Turkey affect the way Turks living in that country are treated, with an average weight of 7.35. Germany-based respondents feel the impact of relations with Turkey most strongly, with a weight of 7.91. In that regard, respondents broadly agree that their country of residence should be more supportive of Turkey, with an average weight of 7.64. Austrian respondents agree most strongly with this proposition, with a weight of 8.51, while French respondents are less emphatic, with a weight of 6.60. Older respondents are more likely to say their current country should be more supportive of Turkey, but all age groups feel this way.
Respondents overall have a strong basis for comparing life in Europe with life in Turkey. Most report having spent an extended period of time in Turkey; 59 percent say they have spent an extended period of time in Turkey, ranging as high as 72 percent among Austrian respondents. Asked to say how long they have spent in Turkey, many respondents report spending more than 10 years there, with some spending more than 20 years. Still, about one-third of respondents in Germany and the Netherlands report having never spent an extended period in Turkey. These figures reflect generational divides—the first generation came to Europe as adults, and the second and subsequent generations were largely born in Europe, with some exceptions. These divides are obscured by the average time spent in Turkey, which clusters around 20 years across the survey—from 19.3 years among Turks in the Netherlands to 20.5 years among Turks in Germany and Austria.
The survey also asked for open-ended descriptions of the advantages of living in their current country of residence. Broken down into broad categories, these responses reflect a widely held belief that the European countries surveyed offer greater prosperity, freedom, and stability than Turkey. Overall, 19 percent say the opportunity to work in their current profession and the earnings potential thereof is the most important advantage—the leading answer. Issues related to freedom, democracy, and human rights are the most important advantage to 16 percent of respondents. Social security protections and workers’ rights are the top advantage for 13 percent of respondents. The higher standard of living and superior living conditions are the biggest advantage to another 13 percent of respondents. The advantages of having an organized society, greater tolerance, and legal certainty are most important to 10 percent of respondents. Access to good schools, vocational training, and the overall quality of the educational system is the biggest advantage to 9 percent of respondents. Only 20 percent of the answers fell outside these broad categories. The four countries are broadly similar in their responses on this question, though responses in France tend to place more emphasis on freedom and less emphasis on social security and living conditions.
The survey also queried, in an open-ended manner, respondents’ views on the disadvantages of life in Europe. These answers are more heterogenous, but the most frequently cited disadvantage is the loss of one’s own culture and the distance from family ties, cited by 19 percent of respondents. The effects of racism and discrimination is the next-most-cited disadvantage, named by 17 percent of respondents overall. Many complain about a social coldness or lack of warmth in their current countries of residence, and associated loneliness—a complaint voiced by 8 percent of respondents. Finally, living amid a foreign culture is listed as a disadvantage by another 8 percent of respondents.
There are significant differences among the countries surveyed on this question. For example, just 12 percent of Turks in Germany list the loss of family ties and culture as the biggest disadvantage, perhaps reflecting more family-based migratory patterns or, perhaps relatedly, a better-defined Turkish community within Germany. But this loss of family and culture is a top concern in France, with 29 percent voicing this lament. Racism and discrimination, on the other hand, is the top complaint of just 9 percent in France, while it is the main disadvantage to 29 percent of Austrian respondents and 17 percent of German respondents.
Asked for open-ended descriptions of what values characterize society in their countries of residence, some familiar tropes emerge, but these concepts still generally reflect an overall positive view of society. Twenty-three percent cite order, discipline, and bureaucracy as the defining characteristics of their current country, but this figure ranges from 32 percent in Germany to just 12 percent in France. Some 15 percent of respondents say sincerity, honesty, and respect are defining values, reaching 26 percent in the Netherlands. One in 10 say that hard work and a focus on one’s profession are defining values. And 9 percent—including 28 percent of Turks in France—say that a social, democratic, and freedom-loving spirit is the most distinctive characteristic.
Political engagement and European politics
The survey shows minimal political engagement in and widespread ambivalence toward European politics on the part of the diaspora. An optimistic interpretation might be that this reflects the generally high levels of satisfaction with life in Europe, obviating the urgency of political involvement. Of course, other explanations are possible, including the possibility that political parties in Europe fail to target the diaspora in their communications and outreach. Certainly, those lacking citizenship in the host country could be excused for limited interest in its partisan politics. In general, when Europe-based Turks do express an opinion about European politics, they lean toward social democratic parties and the Greens, with deep skepticism toward conservative parties and near-total rejection of populist, anti-immigrant parties.
This ambivalence toward government and politics among diaspora Turks is visible in high nonresponse rates to political questions, expressed low affinity for political parties, and an apparent widespread perception of distance and disrespect emanating from established political parties.
Few respondents feel politically represented in their country of residence, rating their overall average sense of representation at just 4.09. Those who only hold Turkish citizenship consistently reported feeling less politically represented, across all countries; overall, those who hold Turkish citizenship give a weight of 3.28 to this question, those with citizenship only of the host country give 4.59, and dual citizens give 4.73. Austrian and French respondents rate their sense of being represented somewhat lower than those in Germany and the Netherlands, while Dutch respondents feel somewhat more represented, perhaps reflecting the presence of a major political party founded by Turkish-Dutch politicians. The level of engagement does not vary significantly by generation.
When respondents are asked which political party most respects them and their Turkish/Kurdish compatriots, many say none do. In fact, in each of the four countries surveyed, the sum of “no answer” and the more emphatic “none” formed roughly half of the responses to this question about respect.14
In Germany, for example, a plurality of 26 percent of respondents say the Social Democratic Party (SPD) most respects their community; 14 percent say “no party” respects their community; 11 percent say the Greens; 7 percent say the Christian Democratic Union (CDU), along with its Bavarian sister party the Christian Social Union (CSU); and 4 percent say The Left (or Die Linke, a left-wing party popular in the former East Germany). But a full 35 percent give no answer—itself probably reflecting a negative judgment on all the parties.
The overall pattern is roughly similar in Austria, though the Social Democratic Party of Austria (SPÖ) has made significant inroads, with 41 percent saying it most respects the Turkish community; 15 percent saying no party; 9 percent saying the Greens; and just 1 percent saying the far-right populist Freedom Party (FPÖ). Thirty percent give no answer.
The Netherlands yields slightly different results, due to the presence of Denk, a staunchly pro-immigrant breakaway group of the Labour Party founded by two Turkish-Dutch politicians. Thirty percent of respondents say Denk most respects their community; 10 percent say the social democratic Labour Party; 9 percent say “no party”; 5 percent say the Socialist Party; and 4 percent say the Greens. Thirty-six percent give no answer. Here, again, the dominance of center-left and left-wing parties among politically conscious elements of the Turkish/Kurdish diaspora community is nearly complete.
Finally, France offers a fragmented picture but one that is again dominated by the political left. Twenty-four percent of respondents say the Parti Socialiste (PS) most respects their community; 18 percent say “no party” respects them; 9 percent say President Macron’s centrist La République En Marche! (LREM); 6 percent say La France Insoumise (the left-wing party of Jean-Luc Mélenchon); and 5 percent say the Greens. Thirty-three percent give no answer. According to polling partner DATA4U, LREM won 46 percent of the Turkish vote in the last elections but has lost the confidence of the majority of these voters and stands to lose significant ground in the next election.15
The same ranking of parties, albeit at lower levels of support, applies when respondents are asked with which party they “identify.” The SPD wins a plurality in Germany, with 21 percent; the SPÖ in Austria (47 percent); Denk in the Netherlands (37 percent); and the PS in France (20 percent). Again, the sum of “no answer” and the more emphatic “none” varies from 35 percent to 50 percent in each country. In fact, “none” is the leading answer in Germany (28 percent) and France (43 percent); in the Netherlands (22 percent), it is second only to Denk; and in Austria, it receives 20 percent.
The pattern of friendliness to the left and center left generally held when respondents rated various European politicians on a 1–10 favorability scale (see Figure 16). In Austria, Federal President Alexander Van der Bellen, a former Green Party leader, led the way with a 6.36 rating—the highest of any politician surveyed across the four countries. In the Netherlands, Tunahan Kuzu, one of the Turkish-Dutch founders of Denk, was the most popular (5.75). In France, that accolade went to Parti Socialiste leader Olivier Faure (4.95).
The one exception to the pattern is Germany, where conservative CDU leader Chancellor Angela Merkel received by far the highest favorability rating (5.32). This possibly reflects the inclusiveness she has projected by welcoming close to 1 million Middle Eastern refugees,16 as well as her government’s de facto role as Europe’s primary liaison with Turkish President Recep Tayyip Erdoğan. She outpaces her closest pursuer, Federal Minister of Finance and SPD Chancellor candidate Olaf Scholz, who scores 3.65, despite the overall social democratic slant of the Turkish community in Germany. Interestingly, the country’s most prominent politician of Turkish descent, Green Party veteran Cem Özdemir—a frequent critic of Turkey’s human rights policies17—scores only 3.42 among Germany-based Turks.
The predilection of the Turkish diaspora—at least that portion of it that expresses any sort of political identity in the European context—for the European left stands in seeming contrast to a rightward tilt in many of the diaspora’s views on Turkish politics. Pending further study, this is presumably explained by the minority-friendly policies of left-wing political parties in Europe. The German Social Democratic Party, for example, pushed through a 1999–2000 reform to Germany’s citizenship law explicitly aimed at better-integrating Turkish immigrants—then, as now, seen as a large and somewhat distinct immigrant group worthy of special consideration—likely securing some affinity from the diaspora community.18 The French socialists have also been generally more responsive to immigrant demands than other French parties.19 Certainly, the harshly anti-immigrant—and often, Islamophobic—rhetoric and policies of the European far right make the diaspora’s hostility toward those parties unsurprising.
Views on the European Union
The survey reveals generally positive views of the European Union. Asked to rate how much they agree with the statement, “I am proud to live in a country that is part of the EU,” respondents give an average weight of 5.83, with France slightly higher at 6.27. Asked if the European Union serves their economic interests, respondents give an average weight of 6.07, with France again slightly higher at 6.49. Asked if they are basically satisfied with the European Union, respondents agree with an average weight of 6.57, ranging from a low of 6.22 in Austria to a high of 7.05 in France. Asked if Turkey should become an EU member, respondents are roughly divided, with an average weight of 5.81 agreeing that it should, ranging from a low of 4.88 in Austria to a high of 6.40 in France. Overall, the French sample tends to have slightly more favorable views of the European Union across these questions, though the differences are not large and the demographic variations are minimal.
Attitudes on Turkish politics
The survey also examined the Turkish and Kurdish diaspora’s attitudes toward and involvement with Turkish politics, and how Turkish politics affects attitudes toward integration in their country of residence. Respondents feel that President Erdoğan cares about the welfare of Turks in their current country with a weight of 5.32, ranging from the Austrian response at 5.87 to the Dutch response at 4.65; younger respondents are more skeptical of that notion. Asked about Erdoğan’s past statement that Turks in Europe should “integrate but not assimilate,” respondents strongly agree, with an overall weight of 8.63 and little variation among the four countries, though with slightly stronger support in Austria and slightly weaker support in France. Again, the younger cohort is relatively less supportive of this statement, though overall they still strongly agree with Erdoğan’s view.
Asked to weight their favorability toward various Turkish political figures, President Erdoğan emerges as the top choice overall, though with significant variations among countries and a high degree of polarization. Overall, respondents give Erdoğan a favorability score of 5.46, ranging from a high of 6.61 in Austria to a low of 4.28 in France—perhaps reflecting the fact that the French sample is more Kurdish and more educated, two demographic factors that tend to negatively correlate with support for the Turkish president. President Erdoğan is also viewed more favorably by older respondents, mirroring the trend in the Turkish domestic electorate.
Ekrem İmamoğlu, the mayor of Istanbul, is the second-most-popular Turkish political figure, with a weight of 5.04. No other Turkish political leader earns a positive favorability weighting overall. Perhaps surprisingly, given his mobilization of young supporters in Turkey itself, İmamoğlu is rated more favorably by older respondents than younger respondents, which could reflect older respondents’ higher baseline interest in Turkish politics—many of them, after all, grew up in Turkey—and, therefore, greater name recognition for a relative political newcomer. Indeed, İmamoğlu is more popular than Erdoğan in France and the Netherlands. Devlet Bahçeli is the next most favorably viewed figure, with a favorability score of 4.27—clearly below the 5.0 “neutral” mark and in-line with his favorability ratings domestically in Turkey—managing a narrowly positive weighting of 5.26 in Austria. He is significantly more unpopular in both France and the Netherlands. Bahçeli is also more favored by older respondents. Kemal Kılıçdaroğlu has a positive weighting of just 3.80 overall and is not viewed positively in any country. Likewise, Selahattin Demirtaş has a positive weighting of just 3.02, though he has a significantly more favorable profile of 4.63 in France. PKK leader Abdullah Öcalan is predictably unpopular across the board, given the history of terrorism and conflict between the PKK and the Turkish state, with a weight of just 1.61, though almost half of the 6 percent overall who self-identify as Kurds have a positive view of him.
Overall, then, the diaspora communities reflect many of the political and generational divides of Turkey itself. The French and Dutch diaspora communities register relatively more liberal views and sympathies, while the Austrian and German diaspora communities tend toward relatively more conservative views—based on averages that obscure an underlying polarization of opinion in all four countries.
Asked if they voted in the 2018 Turkish elections, 56 percent of the Turkish citizens—single and dual, both eligible to vote—in the sample say they did, while 39 percent say they did not; 5 percent gave no answer. (Again, Turkish citizens and dual citizens together constituted 65 percent of the total sample.) This self-reported participation rate varied by only a statistically insignificant 2 percent among the four countries. Older respondents reported voting at a higher rate than others. This once again could reflect their greater interest in Turkish politics.
Of those who say they voted, the AKP took 51 percent of the votes, the CHP took 30 percent, the HDP took 10 percent, and other parties took 9 percent. The AKP did significantly better in Austria, with 64 percent, while the CHP and the HDP both did significantly better in France and the Netherlands. Indeed, the CHP won 39 percent in France, compared with the AKP’s 37 percent and the HDP’s 21 percent.
Responses are broadly similar regarding support for Turkish presidential candidates in the 2018 presidential election, with Recep Tayyip Erdoğan earning 55 percent support from declared voters, the CHP’s Muharrem İnce earning 31 percent, the HDP’s Selahattin Demirtaş earning 10 percent, and others earning just 4 percent. Erdoğan did better in Germany and Austria, with 63 percent and 69 percent, respectively, than in the Netherlands, where he received 54 percent. In France, İnce won 39 percent support to Erdoğan’s 38 percent support, according to the poll.
Again, these breakdowns point to a generally more liberal French and Dutch diaspora and a generally more conservative Austrian and German diaspora. They also largely mirror the political composition of the domestic Turkish electorate, with one important exception. The ultra-nationalist Nationalist Movement Party (MHP), with 11 percent, and its perhaps slightly less nationalist breakaway, the İyi Party (10 percent), together won more than 20 percent of the electorate in Turkey, whereas they score less than half that in the survey. This is noteworthy in regard to the attention that the extremist, MHP-associated Grey Wolves have garnered in Europe.20
Whatever their citizenship or views on Turkish politics, most Europe-based Turks feel it is important to defend Turkish policies. (see Figure 18) Asked to rate how important it is for the Turkish community in their country of residence to defend Turkey’s policies, respondents give an average weight of 6.37, with a slightly lower level in France. Perhaps relevant to this, respondents feel that the Turkish community in their country of residence is strong and unified, with an average weight of 5.92. Younger voters care less than their elders about supporting Turkish policies, though they still think it more important than not.
Asked if they feel pressure from their community to support Turkish government policies, respondents give an average weight of just 2.04, though young respondents are slightly more likely to say they feel pressure. Very few report feeling pressure from representatives of the Turkish government itself to support its policies, with an average of just 1.78 saying that this is the case. That figure is slightly higher in France, but there is little other variation by country or demographic category.
The survey presents a decidedly mixed picture for those who wish to see Turkish immigrants fully integrated into European societies. It reflects a generally positive—though far from perfect—record of cross-cultural accommodation and understanding. European Turks deeply appreciate their “adopted” countries—not only for the material advantages and professional and educational opportunities but also for less tangible values, such as freedom, organization, and the rule of law. Younger Turks, those educated in Europe, increasingly feel more at home in local national languages than in Turkish. The diaspora communities in France and the Netherlands appear more fully integrated into those societies than do communities in Germany and Austria, perhaps reflecting the differing migratory patterns to the German-speaking countries. An overall average of the diaspora respondents say that 46 percent of their host country’s population view Turks as equals, slightly more than the 43 percent who they say hold racist feelings toward Turks. This is a mixed picture, certainly, evidence perhaps of halting progress on the part of European societies that did not initially envision Turkish immigrants staying permanently and that had themselves been largely ethnically homogenous until the late 1960s.
Yet the Turkish community’s primary sense of identity remains overwhelmingly Turkish. On a 1–10 scale, the community rated the importance of its Turkish identity at 8.70 and the importance of its religion at 7.84. These numbers exceed substantially the importance that European Turks place on identification with their country of residence. In short, most think of themselves first as Turkish or Muslim, rather than as Austrian, Dutch, French, or German. Moreover, a significant majority primarily use Turkish, rather the local language, in their home for reading, entertainment, and most news purposes. They are far more engaged with political developments in Turkey than with those in the nations of their residence. Turkish-language television is their main source of news. Even the younger generation, linguistically so much more comfortable than their forebears in the host country language, largely shares the older generation’s conception of identity and its focus on news about Turkey. Clearly, if European political leaders want to speak to the Turkish diaspora, they need to go to where they are and conduct sustained outreach across Turkish-language outlets, particularly television.
Perhaps reflecting its generally positive view of life in Europe and its continuing interest in and connection to Turkey, the diaspora is largely ambivalent about, if not alienated from, European politics. On the one hand, Europe-based Turks display few strong grievances against the governments of the countries in question, but they also do not display much interest in the politics of those countries. Most European Turks do not have strong political party preferences—indeed, most say they do not identify with political parties in Europe—besides notable antipathy toward the anti-immigrant right-wing parties and leaders. Those that do express preferences tend to favor the social democratic parties and leaders of the left. But most simply do not feel politically represented in Europe.
For Europeans concerned about Ankara’s influence on the European Turkish population, the message is also mixed. On the one hand, the diaspora community overall feels it is important to support Turkish policies, though the survey did not ask about specific policies. On the other hand, the Turks in this sample are dismissive of the idea that Turkish state and government representatives put pressure on them to support Turkish policies. Moreover, lack of interest in the politics of their host countries may diminish the diaspora’s influence on European countries’ foreign policies.
Turks of all age groups remain deeply engaged with political developments in Turkey—closely connected through Turkish-language news sources and social media—and say the state of relations between the host nation and Turkey significantly affects the way they are treated by the host population. For the many European Turks who consider it important to support Turkish policies, this could mean they want host countries to accommodate Turkish policies more than vice versa. President Erdoğan, a source of frustration to European leaders in recent years, is the most popular Turkish political figure among the European Turks surveyed. Erdoğan is also highly polarizing in the diaspora community, much as he is in Turkey itself. Erdoğan and his party were also favored by a slight majority among Europe-based Turks who voted in the 2018 Turkish elections. The Turkish community firmly agrees with Erdoğan’s exhortation that they should “integrate but not assimilate” in European societies. The interaction between this conviction—visible across a number of responses to questions in the survey—and the desire of many Europeans to more fully assimilate Turkish minorities will undoubtedly continue to be a source of tension.
Europeans of nearly all stripes—at least, those that favor assimilation, multiculturalism, or some hybrid form of integration for the Turkish diaspora—can find at least some support for their inclinations in the results of this survey. The data reinforce a paradox underlying this debate. Across the survey’s questions on integration, views diverged between those with citizenship in their country of residence and those without; those lacking citizenship were more critical of the host country’s integration policies and most other aspects of life in the host country.21 This reflects a simple reality: People feel more integrated and happier in their place of residence when they are citizens. In the sometimes-heated European political debate, some hesitate to extend full citizenship to diaspora communities, saying they are not fully integrated; yet as the survey shows, citizenship itself is a primary tool of full integration.
The Turkish community seems to be increasingly satisfied with—and at home in—Europe. Yet that same community also remains steadfast in its allegiance to its Turkish identity, which supersedes by far any self-definitions as Austrian, Dutch, French, or German, or as European. It remains to be seen what tendency ultimately will come to define this evolving community: integration, assimilation, multiculturalism, or some other concept entirely.
About the authors
Max Hoffman is the associate director of National Security and International Policy at the Center for American Progress, focusing on Turkey and the Kurdish regions.
Alan Makovsky is a senior fellow for National Security and International Policy at the Center and is a longtime analyst of Middle Eastern and Turkish affairs.
Michael Werz is a senior fellow at the Center, focusing on emerging democracies and migration.
This study was made possible by the expertise and hard work of the polling firm DATA4U, to whom the authors are grateful. The authors are equally grateful to the Foundation for European Progressive Studies (FEPS), the Foundation Max van der Stoel, the Fondation Jean-Jaurès, and Stiftung Mercator for their support for this research. This study does not represent the collective views of FEPS or the cooperating organizations. The responsibility of FEPS and the other cooperating organizations is limited to approving this publication as worthy of consideration for the global progressive movement. The Center for American Progress is an independent and nonpartisan organization and retains final editorial decision-making authority on the contents of its reports. This study was published with the financial support of the European Parliament. The views expressed in this report are solely those of the authors and do not necessarily reflect the views of the European Parliament.
Appendix: Citizenship laws and migration history in brief
Citizenship facilitates integration. Previous studies in Germany and the Netherlands, based on online and offline surveys, show that the degree of the Turkish diaspora’s integration into European society increases with length of stay in Europe and exposure to European mainstream media, and it decreases with the retention of Turkish citizenship.22 A 2018 representative study by the German Federal Office for Migration and Refugees (BAMF) showed that significant numbers of German citizens of Turkish background identified with the German state, while those who remained Turkish citizens identified more strongly with Turkey. Acquiring the same rights as the native population and thereby becoming part of the European community were key drivers of integration.23 These findings help contextualize the results of the study conducted by the Center for American Progress and the Foundation for European Progressive Studies, which similarly show the ways in which citizenship facilitates integration.
The Turkish diaspora in Germany has its roots in the 1961 guest worker program, meant to temporarily bring Turkish workers to Germany to address the labor shortage caused by the economic recovery after World War II. The agreement was part of a broad effort by the German government to secure cheap labor from Southeast Europe, the Balkans, and North Africa. The expectation was that this labor influx would be temporary, and very little effort was made to integrate the workers or encourage them to learn German. In fact, however, few guest workers returned to Turkey, and many opted to bring their families from Turkey and settle permanently in Germany.24 By 2002, one-third of Germany’s Turkish population had arrived as guest workers, roughly half had immigrated under family reunification visas, and 17 percent were their descendants born in Germany.25 In 2016, the nearly 3 million Turks in Germany constituted the country’s largest ethnic minority, but only approximately 246,000 of them held German citizenship.26
This lack of full integration has its roots in German citizenship law, which until 2000 did not provide for birthright citizenship—and even then only with restrictions—and still makes dual citizenship quite difficult. From 1913 to 2000, German citizenship law generally required a person to have at least one German parent. Until January 1, 2000, “children born in Germany to non-Germans—no matter how long the parents’ stay had been—had no right to German citizenship.”27 This meant that many Germany-born children of guest workers were raised in Germany but did not have citizenship. In almost all cases, German law also required those wishing to naturalize to forfeit any other citizenship, which many Turkish immigrants were disinclined to do.28
The 1999–2000 reform of the citizenship law was explicitly aimed at better integrating Turkish immigrants—seen as distinct from the general population and other immigrant communities—and was driven by the Social Democratic Party, likely securing some affinity from the Turkish diaspora. The new law gave those born in Germany citizenship, provided that at the time of their birth one parent had lived in Germany for at least eight years. The law also made provisions to allow for the retention of prior citizenship when naturalizing. But the law also required a child who acquired multiple citizenships to choose by age 23 whether to keep German citizenship and give up any others, or vice versa. While some naturalization requirements later were eased, such as the necessary time of residence, new requirements were introduced, such as a loyalty oath and a German-language test. Since 1981, the Turkish authorities had allowed Turks abroad to briefly relinquish Turkish citizenship—allowing them to naturalize elsewhere, such as Germany—then reclaim Turkish citizenship and quietly maintain dual citizenship. But the 1999–2000 German reforms removed that possibility, closing off a narrow path to dual citizenship.29 Given Germany’s federal structure, however, these national-level reforms did not always have the envisioned policy outcomes in the states or in administrative courts, with some areas, such as Baden-Württemberg and Bavaria, tending to be more restrictive in both naturalization decisions and in the toleration of dual citizenship.30 Taken collectively, the changes, in fact, contributed to a decline in naturalization, and Turkish naturalizations in Germany “peaked in 1999 at over 100,000, then declined to 19,695 in 2015, and just 16,290 in 2016.”31
Another reform, in 2014, ended the requirement for German-born children to choose between citizenships by age 23, allowing for permanent dual citizenship provided the person has lived eight years in Germany, attended school there for six years, or graduated from a German school or training program. For those born elsewhere, however, such as the first generation of Turkish immigrants to Germany, dual citizenship is still not allowed except in very narrow cases; full naturalization of older-generation Turks is still therefore relatively rare.32
Austrian citizenship rules remain largely rooted in a 1949 law dominated by the principle of ius sanguinis—whereby a person’s citizenship is defined primarily by their parents’ citizenship—and have little toleration of dual citizenship.33 Austrian law sees naturalization as “only the last step of a ‘successful integration process’.”34 The Migrant Integration Policy Index (MIPEX) describes the path to citizenship in Austria as “one of the longest and most burdensome, discretionary and expensive … Austria’s policies are the most restrictive in Western Europe.”35
Austria has not had an attempt at liberalization comparable to Germany’s 1999–2000 changes, and the principle of descent still applies. Children born to married parents acquire Austrian citizenship if one of the parents is Austrian. Until 2013, children born out of wedlock only acquired citizenship if their mothers were Austrian, but in the case of a foreign mother, the law required an Austrian father to acknowledge a child born out of wedlock in order to validate Austrian citizenship. A 2013 court ruling eased some of these restrictions, but not in all cases, and the law still requires Austrian fathers of children born out of wedlock to foreign mothers to recognize the child before birth or within eight weeks of birth in order to provide citizenship, though even without this recognition naturalization is eased for such children.36 Naturalization is difficult, requiring 10 years’ legal residence, financial resources, proven language skills, and passage of tests on Austrian civics, history, and culture, among other restrictions. The strict requirements and financial costs have resulted in less naturalization of financially disadvantaged immigrants.37
As in Germany, Austria began to allow guest workers into the country in the 1960s to address labor shortages. Naturalizations increased along with this influx, though they remained low in absolute terms, but they slowed significantly in the 1980s, only to grow again in the 1990s. Turkish nationals accounted for only 17 percent of all naturalizations in 1985, increasing to 31 percent in 2004.38 This showed the substantial interest in naturalization among the Turkish community in Austria but was also due in part to 1995 changes allowing Turkish immigrants to keep most citizenship rights in Turkey while avoiding certain Turkish military obligations, making dual citizenship particularly attractive.39
The implementation of Austrian naturalization and citizenship law is overseen by local authorities, however, and some localities, such as Vienna, have adopted more permissive naturalization criteria. This localism—as well as the growth in Turkish and Balkan immigration—sparked a backlash, and laws passed in 1998 and 2005 introduced stricter, uniform national requirements, including language and civics tests.40 These years also brought evidence of de facto administrative hurdles to naturalization, which reduced the numbers.41 As a result, total naturalization numbers, as well as the naturalization rate, have plummeted since the 2005 legislation, even as the number of foreign-born Austrian residents has continued to increase. Indeed, from 2003 to 2011, naturalizations dropped by 85 percent to their lowest point since 1973.42 Total naturalizations have ticked back up in recent years, though the naturalization rate remains quite low at 0.7 percent.43 According to official Austrian statistics from 2019, some 282,800 people of Turkish descent live in Austria, of which 160,300 are first-generation immigrants and 122,500 are of the second generation and born in Austria.44
The past decade has seen the Netherlands partially abandon its historic commitment to open immigration. Citizenship requirements remain less stringent than in Germany or Austria, however, and in some ways are less restrictive than in France. To naturalize, a person must be eligible for permanent residence, have lived in the Netherlands for five years, and pass Dutch-language and civics exams. Children born or raised in the country are entitled to become citizens at age 18. Still, many applicants are discouraged in practice, and the anti-immigration pressure of right-wing political parties in coalition governments has led to cuts to support for linguistic and civic integration. Finally, the personal financial cost of naturalization is quite high.45 While the Netherlands only allows dual citizenship as an exception, its immigrant population has generally lived in the country for a long time, and three-quarters of Dutch immigrants born outside the European Union now have Dutch citizenship, according to 2011–2012 estimates—one of the highest naturalization rates in the developed world.46
As with most of Europe, the Netherlands only slowly realized that migration beginning in the 1960s was likely to be permanent. The country adopted liberal policies in the 1980s designed to ease naturalization, and it embraced multiculturalism to a large extent, resulting in an increase in naturalization, primarily from former Dutch colonies. The Dutch Nationality Act of 1984 essentially established naturalization as a right rather than a privilege, subject to residency requirements, a clean legal record, and the renunciation of other citizenships. Nevertheless, while the number of immigrants from Turkey and Morocco grew in this period, their naturalization rates remained low, suggesting administrative hurdles or perhaps a lack of interest.47
The requirement for a naturalizing citizen to renounce other citizenships was abolished in 1991 but reinstated in 1997, shaping recent naturalization patterns. In the period when the requirement was suspended, the number of naturalizations increased to an average of 50,000 per year, compared with 19,000 people per year in the 1980s. This period had a considerable impact on Turkish naturalization in the Netherlands, as Turkey by that time permitted Turks abroad to relinquish citizenship (and then subsequently easily reacquire it). In 1992, the Turkish naturalization rate rose to 20 percent of eligible immigrants and subsequently dropped to 5 percent in the period from 1999 to 2001.48
The 1997 reinstatement of the renunciation requirement was part of a broader shift away from integrative policies, though one accompanied by efforts to normalize the status of second- and third-generation noncitizen residents who desired naturalization. Beginning in the late 1990s and accelerating with several high-profile cultural controversies—such as the murder of Dutch film director Theo van Gogh49—and growing right-wing influence in the early 2000s, however, the Netherlands further slowed its integrative and multicultural approach. Instead of viewing citizenship as an important tool of integration, emphasis shifted to viewing citizenship as the culmination of the integration process. A new Dutch Nationality Act became law in 2003, requiring passage of a new naturalization test to prove sufficient knowledge of Dutch society and language. While “proof of integration” had always been a condition for naturalization, previous Dutch governments had “also made it clear that ‘integration’ was not the same as ‘assimilation’.”50 The 2003 law began to change this emphasis, formalizing testing requirements, though these changes at least offered some transparency and removed testers’ ability to de facto discriminate against certain applicants in their administration of the test. But this formalization also increased the difficulty—and the significant financial costs—of the tests, resulting in a precipitous drop in naturalization applications.51 Still, counteracting this restrictive tendency has been a slow effort to normalize the situation of longer-term residents in some ways, including by providing avenues to citizenship, as with the 2010 extension of an option for people born before 1985 of Dutch mothers and non-Dutch fathers to become citizens. Previously, these people had been denied citizenship due to the Netherlands’ gendered ius sanguinis rules.52
French nationality law has been a mixture of ius soli—or birthright citizenship, where anyone born in the country receives citizenship—and ius sanguinis. But as with the other World War II combatants, postwar France recognized the need for immigrants and adopted an even more integrative approach, though the de facto administration of migration policies has sometimes differed from the de jure policies.
Today, French nationality is secured if a single parent is French, regardless of the child’s place of birth, or if the child is born in France and has one parent also born in France. A person born in France to non-French parents automatically becomes a citizen at age 18 if they still reside in France and do not refuse citizenship.53 Naturalization barriers are formally low, requiring just five years of residence, but the government has historically been biased against certain classes of migrants and has selectively slowed down their processing.54 France also has an employment requirement, and its language testing is among the most difficult in Europe.55 Citizenship also can be obtained through marriage to a French citizen, but the process takes several years. France is tolerant of dual citizenship and does not require a naturalized citizen to renounce previous citizenship.
Modern France has generally adopted a progressive integrative approach, believing that full integration requires citizenship.56 This tradition has come under some pressure periodically in modern French history. In the 1970s and 1980s, high unemployment and the perception that large numbers of ethnically and religiously “different” immigrants were unassimilated led to a backlash against progressive citizenship laws. Likewise, in the past 30 years, Europe’s broader culture wars over religion and ethnic difference led to laws in 1993, 1998, and 2003 that placed new restrictions on the citizenship eligibility of foreign-born spouses and family and that required tests on citizenship and the French language for naturalization.57
Separate from citizenship law itself, France has long placed heavy emphasis on civics courses, highlighting the importance of le vivre ensemble, or social/community cohesion—not just secularity, or laïcité. In recent years, the government has instructed public school teachers to make clear that cultural diversity, including religious differences, is part of the country’s overall, unified culture.58
- Estimates vary, but the Turkish government estimates that around 5.5 million Turkish people live in Western Europe. Republic of Turkey Ministry of Foreign Affairs, “Turkish Citizens Living Abroad,” available at http://www.mfa.gov.tr/the-expatriate-turkish-citizens.en.mfa (last accessed September 2020).
- “Host country,” “current country,” and “country of residence” are generally used interchangeably throughout this report, which is not meant to imply a value judgment on the relative membership or belonging of the Turkish diaspora, nor their intentions or desire to stay in their current country or return to Turkey. These terms are used primarily for utility and variety.
- Only 6 percent of the respondents primarily self-identify as Kurdish, so the survey overwhelmingly reflects the views of ethnic Turks in the countries surveyed. Nevertheless, the linkages—primarily, shared Turkish citizenship and national background—between the communities are deep, justifying the decision to survey them together. This report frequently refers to the Turkish diaspora as useful shorthand, but unless specifically noted, the results reflect the opinions of both self-described Turks and Kurds. As a point of further clarification, the Kurds referred to in the report are, in all cases, Turkey-origin Kurds.
- See, for example, Özlem Gezer and Anna Reimann, “Erdogan Urges Turks Not to Assimilate: ‘You Are Part of Germany, But Also Part of Our Great Turkey’,” Der Spiegel, February 28, 2011, available at https://www.spiegel.de/international/europe/erdogan-urges-turks-not-to-assimilate-you-are-part-of-germany-but-also-part-of-our-great-turkey-a-748070.html.
- The full results of this survey are on file with the authors and are available upon specific request for academic, noncommercial purposes.
- John Halpin and others, “Is Turkey Experiencing a New Nationalism? An Examination of Public Attitudes on Turkish Self-Perception” (Washington: Center for American Progress, 2018), available at https://americanprogress.org/issues/security/reports/2018/02/11/445620/turkey-experiencing-new-nationalism/; Max Hoffman, Michael Werz, and John Halpin, “Turkey’s ‘New Nationalism’ Amid Shifting Politics: Further Analysis of Polling Results” (Washington: Center for American Progress, 2018), available at https://americanprogress.org/issues/security/reports/2018/02/11/446164/turkeys-new-nationalism-amid-shifting-politics/.
- Personal communication from DATA4U on its methodology, received March 2020. DATA4U provided the following information as well: Contact databases for the respective countries were set up from public telephone directories. Using name analysis (onomastics), a targeted search for Turkish first and last names yielded an anonymized contact database of around 350,000 telephone connections (landline and mobile) with Turkish subscribers identified. Numbers were selected at random and presented to the interviewers to dial. ZIP codes were used to achieve proper regional distribution, and the sample was controlled for representative criteria such as age, gender, place of residence, education, and other key demographic categories. The questionnaire contained around 120 open and closed questions and was available to the native-speaking interviewers in Turkish, English, German, French, and Dutch. Interviewees’ consent to data protection and the temporary storage of their anonymized answers was obtained, and their Turkish migration background confirmed. The average duration of interviews was 50 minutes. Demographic information on the populations of Turkish origin in Germany, Austria, France, and the Netherlands was taken from the respective official statistical offices and supplemented by previous DATA4U studies. There are only rough estimates about the population of Turkish origin in France, as the collection of official ethnic information is prohibited. Upon request, the Turkish Embassy in Berlin informed us that around 700,000 Turkish citizens are registered with the Turkish Embassy or consulates in France. Supplemented by previous DATA4U studies on the Turkish diaspora in France, this study estimates that there are currently around 800,000 people of Turkish origin in France. Of these, around 624,000 should be in the 18+ age segment. The populations of Turkish origin for the other countries are as follows:
- Germany: 3.5 million, thereof 2.73 million over age 18 (sources: Federal Statistical Office (DESTATIS) and calculations of DATA4U)
- Austria: 0.37 million, thereof 0.29 million over age 18 (sources: STATcube – statistical database from STATISTIK AUSTRIA and own calculations by DATA4U)
- Netherlands: 0.48 million, thereof 0.38 million over age 18 (sources: Centraal Bureau voor de Statistiek and calculations of DATA4U)
- The remaining 2 percent only hold the citizenship of third countries (i.e., neither Turkey nor any of the countries surveyed).
- It is difficult to make exact comparisons on income, given the differing demographic profile of the diaspora community compared with the nonimmigrant community in the host country and the variations among the countries surveyed. Previous studies have found that “in every decile average income among natives is significantly higher than that among immigrants. Thirdly, quantile regression results show us that immigrants from Turkey mostly have lower incomes than natives with the same background factors.” See, for example, Asena Caner and Peder J. Pedersen, “Income Distribution among Turkish Immigrants in Germany and Denmark” (Copenhagen: The Rockwool Foundation, 2019), available at https://www.rockwoolfonden.dk/app/uploads/2019/01/Study-Paper-135_Income-distribution-among-Turkish-immigrants-in-Germany-and-Denmark-1.pdf. Taking Organization for Economic Cooperation and Development (OECD) data and converting from U.S. dollars to euros according to OECD exchange rates of the applicable (most recent) years yields the following median net household incomes for the overall population of the surveyed countries:
- Germany (2018): $40,699, or 34,472 euros, which amounts to 2,873 euros per month
- France (estimated 2018): $35,385, or 29,971 euros, which amounts to 2,498 euros per month
- Austria (2018): $38,333, or 32,468 euros, which amounts to 2,706 euros per month
- Netherlands (estimated 2019): $37,810, or 33,764 euros, which amounts to 2,814 euros per month
Weighting those median net household incomes by each country’s share of the survey sample (Germany: 45.2 percent of sample; France: 19.2 percent; Austria: 17.7 percent; and Netherlands: 18 percent) yields a weighted average net household income for the whole survey of 2,761 euros per month for the 2018–2019 period. This figure undoubtedly grew further in the intervening time before the CAP/FEPS survey was conducted in December 2019 and January 2020, yielding the approximate difference of 10 percent to 15 percent between diaspora and nondiaspora median net household incomes. For OECD household disposable income data, see OECD data, “Household disposable income,” available at https://data.oecd.org/hha/household-disposable-income.htm (last accessed August 2020). For OECD exchange rates, see OECD data, “Exchange rates,” available at https://data.oecd.org/conversion/exchange-rates.htm (last accessed August 2020).
- DATA4U pollsters, who administered the poll, speculate that there could be several reasons for this. First, this could be attributable to the character of Turkish migration to France, which was less dominated by migrant workers from rural regions of Turkey, where educational opportunities were more limited. The Turkish-French population was relatively more heavily composed of liberals, intellectuals, left-wing politicians, and Kurds seeking protection or asylum. Second, the phrasing of the question may have resulted in some misunderstanding or miscoding of results, as the middle school qualification in France is referred to as “collége” and may therefore have been confused in part with the English “college” (university degree).
- Andrew O’Donohue, Max Hoffman, and Alan Makovsky, “Turkey’s Changing Media Landscape” (Washington: Center for American Progress, 2020), available at https://americanprogress.org/issues/security/reports/2020/06/10/485976/turkeys-changing-media-landscape/.
- The German respondents were outright negative toward refugees in Germany, with an average weight of 4.98. Dutch respondents were the most positive toward refugees in their country of residence, with an average weight of 6.01.
- Above 5 would indicate agreement, while below 5 would indicate disagreement.
- Combining those who responded “no answer” or “none do” to this question about which party respects Turkish people yields the following percentages, by country: Germany, 49 percent; Austria, 45 percent; Netherlands, 44 percent; and France, 51 percent.
- Personal communication with polling firm DATA4U based on its past private polling, September 2019. Communication is on file with the authors.
- Peter Hille, “’We can do this!’ — Merkel’s famous words five years on after refugee influx to Germany,” Deutsche Welle, August 31, 2020, available at https://www.dw.com/en/merkel-germany-refugees/a-54769229.
- See, for example, Kay-Alexander Scholz, “Cem Özdemir: ‘Erdogan wants to establish Turkey in Germany’,” Deutsche Welle, July 19, 2017, available at https://www.dw.com/en/cem-%C3%B6zdemir-erdogan-wants-to-establish-turkey-in-germany/a-39761625. Or see Gezal Acer, “No place for Turkish ‘climate of fear’ in Germany says Green’s Cem Özdemir,” Deutsche Welle, December 23, 2017, available at https://www.dw.com/en/no-place-for-turkish-climate-of-fear-in-germany-says-greens-cem-%C3%B6zdemir/a-41918067.
- Susan Willis McFadden, “German citizenship law and the Turkish diaspora,” German Law Journal 20 (1) (2019): 72–88.
- Christophe Bertossi, “Country Report: France” (Florence, Italy: EUDO Citizenship Observatory Robert Schuman Centre for Advanced Studies and Edinburgh, United Kingdom: Edinburgh University Law School, 2010), available at https://cadmus.eui.eu/bitstream/handle/1814/19613/France.pdf?sequence=1&isAllowed=y.
- Elizabeth Schumacher, “Germany seeks to ban Turkish ‘Gray Wolves’ far-right symbols,” Deutsche Welle, October 9, 2018, available at https://www.dw.com/en/germany-seeks-to-ban-turkish-gray-wolves-far-right-symbols/a-45809792.
- Although the question was not asked on the survey, the authors believe that those Turks living in Europe without citizenship of their host country would largely like to have it—perhaps, in many cases, provided they could also maintain Turkish citizenship.
- Roya Imani Giglou, Leen d’Haenens, and Baldwin Van Gorp, “Determinants of degree of integration of Turkish diaspora in Belgium, the Netherlands, and Germany,” International Communication Gazette 81 (3) (2019): 259–282.
- Susanne Schührer, “People of Turkish origin in Germany: Findings from the representative study ‘Selected Migrant Groups in Germany 2015’” (Nuremberg: German Federal Office for Migration and Refugees (BAMF), 2018), pp. 44, 46, available (in German) at https://www.bamf.de/SharedDocs/Anlagen/DE/Forschung/WorkingPapers/wp81-tuerkeistaemmige-in-deutschland.html?nn=283640.
- Willis McFadden, “German citizenship law and the Turkish diaspora.”
- Claus Hofhansel, “Citizenship in Austria, Germany, and Switzerland: Courts, Legislatures, and Administrators,” International Migration Review 42 (1) (2008): 163–192.
- Willis McFadden, “German citizenship law and the Turkish diaspora.”
- Hofhansel, “Citizenship in Austria, Germany, and Switzerland: Courts, Legislatures, and Administrators.”
- Joachim Stern and Gerd Valchars, “Country Report: Austria” (Florence, Italy: EUDO Citizenship Observatory Robert Schuman Centre for Advanced Studies and Edinburgh, United Kingdom: Edinburgh University Law School, 2013), available at https://cadmus.eui.eu/bitstream/handle/1814/60232/RSCAS_EUDO_CIT_2013_28.pdf?sequence=1&isAllowed=y.
- Thomas Huddleston and others, “Migrant Integration Policy Index 2015: Austria,” Barcelona Centre for International Affairs and Migration Policy Group (last accessed August 2020).
- Stern and Valchars, “EUDO Citizenship Observatory, Country Report: Austria.”
- Dilek Cinar and Harald Waldrauch, “Austria,” in Acquisition and Loss of Nationality, Volume 2: Country Analyses: Policies and Trends in 15 European Countries (Amsterdam: Amsterdam University Press, 2010), pp. 19–62, available at https://www.jstor.org/stable/j.ctt46n00f.5?refreqid=excelsior%3A943b2e79741e300b71270db04c9b57b7&seq=1#metadata_info_tab_contents.
- Stern and Valchars, “EUDO Citizenship Observatory, Country Report: Austria.”
- Statistics Austria, “Naturalisations: Latest annual results,” available at http://www.statistik.at/web_en/statistics/PeopleSociety/population/naturalisation/index.html (last accessed September 2020).
- Statistik Austria, “Bevölkerung mit Migrationshintergrund im Überblick,” available at http://www.statistik.at/web_de/statistiken/menschen_und_gesellschaft/bevoelkerung/bevoelkerungsstruktur/bevoelkerung_nach_migrationshintergrund/033240.html (last accessed August 2020).
- Thomas Huddleston and others, “Migrant Integration Policy Index 2015: Netherlands,” Barcelona Centre for International Affairs and Migration Policy Group (last accessed August 2020).
- Ricky van Oers, Betty de Hart, and Kees Groenendijk, “Country Report: The Netherlands” (Florence, Italy: EUDO Citizenship Observatory Robert Schuman Centre for Advanced Studies and Edinburgh, United Kingdom: Edinburgh University Law School, 2013), available at https://cadmus.eui.eu/bitstream/handle/1814/19629/RSCAS_EUDO_CIT_2013_1.pdf?sequence=3&isAllowed=y.
- See, for example, Marlise Simons, “Dutch Filmmaker, an Islam Critic, Is Killed,” The New York Times, November 3, 2004, available at https://www.nytimes.com/2004/11/03/world/europe/dutch-filmmaker-an-islam-critic-is-killed.html.
- Bertossi, “Country Report: France.”
- Thomas Huddleston and others, “Migrant Integration Policy Index 2015: France,” Barcelona Centre for International Affairs and Migration Policy Group (last accessed August 2020).
- Bertossi, “Country Report: France.”
- Isabelle Saint-Martin, “Teaching about religions and education in citizenship in France,” Education, Citizenship and Social Justice 8 (2) (2013): 151–164.
The positions of American Progress, and our policy experts, are independent, and the findings and conclusions presented are those of American Progress alone. A full list of supporters is available here. American Progress would like to acknowledge the many generous supporters who make our work possible.