Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The Netherlands was one of the first countries in Europe to formulate a coordinated national policy on migrant integration, in the early 1980s. In time this policy would become well known internationally as the Dutch multicultural model, because of its emphasis on the institutionalisation of cultural diversity and cultural emancipation of ethnic minority groups (see also Chap. 3). However, the history of Dutch migrant integration policy has been very dynamic. In the 2000s, the Netherlands was one of the first European countries to make a sharp ‘assimilationist turn’ (Joppke and Morawska 2003). More recently, Dutch policymakers have been withdrawing from this policy area, delegating responsibility for integration almost entirely to migrants themselves.

In the Netherlands migrant integration policy became politicised around the year 2000. Since then, political anti-migrant movements have gradually become stronger. Public perceptions of Muslims in particular have deteriorated significantly (SCP 2009: 29–30). Influential events in this regard were 9/11 and the killings of politician Pim Fortuyn by an animal rights activist (2002) and film director Theo van Gogh by a fundamentalist Muslim (2004). Both were outspoken opponents of immigration, in particular from Muslim countries. Today, migrant integration issues are still high on the political and public agenda (Table 12.1).

Table 12.1 Policy frames in Dutch migrant integration policy since the 1970s (Adapted from Scholten 2011)
Full size table

At the same time, the extensive ethnic monitoring structure constructed in the 1980s and 1990s still exists (Guiraudon et al. 2005). Dutch policies still aim at so-called evenredigheid, i.e. proportional participation for comparable groups with or without a migrant background, though less explicitly than before (Dagevos 2012). Each year, either CBS (Statistics Netherlands) or the SCP (the Social and Cultural Planning Office, a governmental research organisation that provides data on social and cultural developments) still publishes an integration report on the progress made so far and on remaining challenges in various societal domains.

The broader field of migrant integration research (i.e. the field of knowledge production) has changed significantly over the years. From a relatively small and coherent field in the 1980s, with a strong knowledge consensus (the ‘minorities paradigm’), migrant integration research has evolved into a large and fragmented field, which is characterised by various ‘schools of thought’. There has been a growing manifestation of knowledge conflicts amongst migrant integration scholars, often working from very different frames of migrant integration. Moreover, Dutch migrant integration research has internationalised rapidly, especially over the last decade or so. There has been a rapid proliferation of comparative studies.

Since the late of the 1980s, the positivist belief in the role of scientific research as a tool for societal engineering in this area has clearly declined (Scholten 2011). In fact, there appears to be a growing disenchantment about research-policy dialogues. The politicisation of migrant integration policies has at various moments led to public scepticism about the credibility of migration scholars (see also Scholten and Verbeek 2014), in Parliament as well as in the media, partly because of their deep involvement in previous policies that were so fiercely rejected later. The research-policy dialogue structure has become much more politicised and mediatised compared with earlier years when a strong mutual relationship pertained in official, but relatively private (non-mediatised) settings. Jan Rath coined the term ‘technocratic symbiosis’ to describe this earlier scenario (Rath 2001).

In the current literature on research-policy dialogues on migrant integration lots of attention has been devoted to politicisation but little to mediatisation. The DIAMINT project was also designed to address this gap. In DIAMINT, we took a broad perspective on research-policy dialogues: we did not only look at relatively direct communications between researchers and policymakers in formal and informal settings – but also at indirect communications via the mass media.

The Netherlands is a strategic case to explore the combined impact of politicisation and mediatisation of research-policy dialogues about migrant integration, because media attention for the topic reached unprecedented peaks in the 2000s, also in international comparison.Footnote 1 As Vliegenthart (2007: 139) states: ‘Research that tries to explain the recent changes in the Dutch debate on immigration and integration of minorities without taking the media factor into account is incomplete – at best’ (see Uitermark 2010). In his longitudinal (1990–2004), large-scale content analysis of the political and media debates on immigration and integration in the Netherlands, Vliegenthart (2007: 136) concluded that the results resonated most strongly with a ‘mediacratic conception of democracy’. In this conception, the media ‘assert influence on both politics and public and are in that way most powerful in shaping political decision-making processes and public opinion’ (Vliegenthart 2007: 135).

The next section gives the necessary context for the analysis by briefly describing the key issues in migrant integration and migrant integration policies in the Netherlands. The third section summarises the existing literature on research-policy dialogues on migrant integration in the Netherlands, which is mainly devoted to their politicisation: as the issue became politicised, did research-policy dialogues become politicised as well? The fourth section explores the hypothesised mediatisation of research-policy dialogues, about which very little research is available yet. It ‘zooms in’ on the roles taken by individual scientists in different research-policy dialogue structures (mediatised and non-mediatised) and the possible consequences of these roles for the utilisation of their knowledge by policymakers. In agreement with the other country cases, three specific key topics were chosen to strengthen the comparative analyses at the international level: Islamic schools, educational segregation, and naturalisation. The concluding section suggests that the combined politicisation and mediatisation of research-policy dialogues on migrant integration may have created a new boundary, i.e. a boundary between mediatised and non-mediatised research-policy dialogues. Consequently, the old boundary between the ‘two worlds’ of ‘research’ and ‘policy’ may have lost some of its relevance.

2 Migrant Integration and Migrant Integration Policies

The recent history of immigration to the Netherlands and the immigrant presence in the country are not drastically different from those in neighbouring European countries (Entzinger 2010, Entzinger et al. 2011). Currently, about 11 % of the Dutch population of 16.5 million people are foreign born and for that reason can be qualified as immigrants. One in five persons living in the Netherlands is either an immigrant or a child of an immigrant. Three communities (including the ‘second generation’) stand out in size: Turks, Surinamese and Moroccans, each numbering between 350,000 and 400,000. The Turkish and the Moroccan communities are legacies of the so-called ‘guest worker’ policies in the late 1960s and early 1970s. Most migrants from Surinam arrived in the 1970s, when this former Dutch colony acquired political independence. The end of the Cold War led to a significant growth of East European migrants, particularly since the accession of the new member states in 2004 and 2007, and of asylum seekers, some of whom later acquired refugee status. Besides, growing numbers of Dutch and foreign residents find their spouses in other countries. In recent years, the number of highly skilled migrant workers has also increased, although many of them do not settle for good. Meanwhile, family migration among the three largest communities, the Turks, the Surinamese and the Moroccans, is continuing, albeit at a much slower pace than before (Entzinger 2010).

Dutch integration policy, however, focuses only on specific categories of migrants. Unlike many other immigration countries in Europe, citizenship is not generally considered as the primary distinguishing factor between migrants and the native population (Entzinger 2010). Rather, ethnic origin tends to be more relevant in the public perception as a means of differentiating between them and us. The Dutch have even constructed a term for this: the Greek-based word allochtonen refers to those people whose ethnic roots lie outside the Netherlands and who, for that reason, can be differentiated from autochtonen, the native Dutch. The statistical category ‘niet-westerse allochtonen’ (‘non-Western allochtones’, just over one-tenth of the population) includes both first-generation migrants and persons who were born in the Netherlands, but have at least one parent born in a ‘non-western’ country, i.e. in Africa, Central and South America, Asia (excl. Indonesia and Japan), or Turkey (Statistics Netherlands 2008). All other allochtonen are labelled ‘Western’ and are generally seen as more similar to the native Dutch – and therefore as less problematic. It should be noted that the term allochtonen is not uncontested and controversy surrounding the concept has led several municipalities to abolish it. One reason for this is that perceived integration problems no longer coincide so clearly with ‘non-Western’ origins (Ham and Van der Meer 2012).

The development of Dutch national integration policy is marked by discontinuity: the Minorities Policy in the 1980s had distinct multiculturalist traits; Integration Policy in the 1990s had more universalist traits; and finally the so-called New Style Integration Policy since 2002 has had distinct assimilationist traits (Scholten 2011: 75–76) (Table 12.1). In the latter frame, the dominant idea is that migrants are to blame for their slow integration, while efforts to step up this process should come from their side (Entzinger 2010). Some lip service is being paid to the idea that integration should be two-sided and that the established population should also leave some space to newcomers, but only a few concrete policy measures have pointed in that direction. Most of the new measures leave little or no room for public recognition of migrants’ cultural identity (Entzinger 2010). The Dutch case seems to be exceptional in the sense that it exemplifies a radical turn from multiculturalism to assimilationism (Poppelaars and Scholten 2008: 339). In fact, although the Netherlands has long been celebrated for its multiculturalist policies, nowadays this multiculturalist approach is widely dismissed as a failure in Dutch public and political discourse (ibid.).

The ‘assimilationist turn’ (Joppke and Morawska 2003) in Dutch integration policies was less outspoken in the period 2007–2010, when the Labour Party was part of the national government. As distinct from the New Style Integration Policy of the preceding period, migrant integration was now connected to urban policy and to neighborhood policies, away from the more symbolic facets of national integration policies and issues of national identity. However, the Centre-Right coalition led by Prime Minister Rutte that came to power in 2010 again discursively promoted assimilation, national unity and ‘Dutchness’, as was to be expected given its political composition and the fact that the minority coalition needed parliamentary support from the anti-immigrant Freedom Party. Actually, however, it did not achieve much in this realm. At present, policymakers are withdrawing from this policy area, focusing instead much more on limiting immigration, with the exception of some high-skilled flows. This, however, is easier said than done, since most immigration now stems from other EU countries, and these flows cannot really be controlled.

3 Research-Policy Dialogues: A State of the Art

This section is partly based on Entzinger and Scholten (2013).

3.1 The 1970s and the 1980s

Research as well as policymaking on migrant integration had been virtually non-existent in the Netherlands until the late 1970s, as the country did not consider itself a country of immigration (Entzinger and Scholten 2013). The settlement of growing numbers of post-colonial migrants since the 1950s and of ‘guest workers’ from Mediterranean countries since the 1960s was not recognised by the authorities as a permanent phenomenon. Thus specific policy efforts to promote their integration were not deemed necessary. Only the Ministry of Social Work (CRM) assumed a more proactive role, since it was the first to be faced with demands from immigrants and also encountered growing tensions between immigrant communities and the native population. The Ministry, however, had virtually no experience in these matters, and therefore turned to the social sciences. In those days the belief was widespread that social scientists could provide the necessary tools for social planning and engineering, and provide answers to the emerging challenges.

In this context several boundary organisations entered the scene. The Social and Cultural Planning Office (SCP) – established in 1973 and closely associated with the Ministry of Social Work – began to collect information on the position of immigrants in society. Mobilising data would be a first step towards raising awareness that new policy measures would be required. Taking this approach a step further, the Ministry of Social Work installed the Advisory Commission on Minorities Research (ACOM) in 1978. ACOM was meant to advise the Ministry on programming the research needed for the further development of policies in this area (Scholten 2011). This was not unusual in those days: so-called research-programming commissions were set up in several policy areas, in some cases because the area was too controversial for decision-making (e.g. demography, where birth control was still an issue), in other cases because it was too new and too little was known to enable sound policymaking (e.g. environmental issues, and also immigration). The ACOM membership was exclusively academic, and ACOM soon acted as a linchpin or archetypical boundary organisation between policymakers and the research community. After a few years, as other ministries began to be faced with the effects of continuing immigration, ACOM’s advisory capacities were expanded to the entire government.

Another boundary organisation that played a pivotal role in the research-policy dialogue was the Scientific Council for Government Policy (WRR), a high-level governmental advisory body, set up in 1972 to advise the Dutch government on new developments in society and their potential relevance for policymaking (Scholten 2009). The WRR is an independent body, free to decide its own agenda, but administratively part of the Office of the Prime Minister. The government is obliged by law to respond to its recommendations. The WRR has reported on migrant integration four times in the 40 years of its existence. The first report, called Etnische minderheden (‘Ethnic Minorities’), was released in 1979 and set the scene for the initial multiculturalist approach to migrant integration adopted by the Dutch government. In fact, this report was taken as a direct basis for the first memorandum on migrant integration policies, the Nota Minderhedenbeleid (‘Memorandum on Minorities Policy’) of 1981, which was finalised in 1983.

These developments in the Dutch research-policy dialogue on migrant integration reflected the strong policy orientation of social scientists in this period. This was matched by a strong belief in rational societal steering amongst policymakers and politicians. Amongst policymakers there was little experience with migrant integration policies. Respondents in our fieldwork agree that a strong consensus prevailed that migrant integration should not be turned into a partisan issue. This led to a context of de-politicisation, which enabled social science research to provide an authoritative venue for policy formulation.

The research-policy dialogue also affected the development of migration research itself. By the early 1970s, only a handful of social scientists had been active in the area, largely inspired by the immense amount of knowledge that was developed over many decades by their US colleagues and, to a lesser extent, in the UK. In those days, dominant scholars in these two countries had begun to distance themselves from the classical idea that migrant integration would involve assimilation. The idea that immigration would lead to the formation of ethnic minorities, recognisable as such for at least several generations to come, had become the major paradigm in these more experienced immigration countries. The ‘ethnic minorities’ concept also appealed to many of the Dutch academics: several anthropologists, for example, were familiar with the concept of ‘ethnicity’ through earlier research in the Dutch colonies.

The construction of a direct dialogue structure between migration research and policymaking in the late 1970s and early 1980s not only led to the formation of an Ethnic Minorities Policy, but also facilitated the dominance of the ethnic minorities paradigm in early migration research. Rath observes that ‘the development of political-economic theory on guest workers in the Netherlands was quite suddenly interrupted, precisely at the moment that the state incorporated researchers into the bureaucratic apparatus and initiated wide-scale funding for politically relevant research’ (Rath 2001: 152). Similarly, Penninx, one of the key architects of the WRR’s Ethnic Minorities report, observes that significant criticism existed towards ACOM, which ‘in some circles of researchers (..) was seen as a biased group of advisors that functioned as gatekeepers for (..) research funding, and gave no or insufficient attention to certain topics or disciplines’ (1988: 37). Thus, support for the ethnic minorities paradigm coincided with the dismissal of contending knowledge claims. This happened, for example, to those who were doubtful about the effects of mother tongue teaching for migrant children and also to adherents of the racism paradigm. The same happened to research initiatives that emerged, albeit very slowly, from within the ethnic communities themselves (Essed and Nimako 2006).

3.2 The 1990s

Until well into the 1990s, the ethnic minorities paradigm remained the dominant paradigm in the Netherlands, both in policymaking and in research. Research in the area expanded rapidly, mostly financed by a growing number of Ministries that perceived a need for a sounder basis for their new policies. Slowly but surely the ‘technocratic symbiosis’ (Rath 2001) began to collapse. In an effort to strengthen the academic nature of immigration research, the Netherlands Organisation for Scientific Research (NWO) funded a special programme in which the need for more theory was emphasised. At several universities chairs in migration studies were established, while specialised research centres for migration and ethnic studies were set up at the Universities of Utrecht (ERCOMER) in 1993 and Amsterdam (IMES) in 1994. Gradually, other academic disciplines, less burdened with the minorities paradigm than sociology and anthropology, also took on an interest in this research area. In fact a second report by the Scientific Council for Government Policy (1989), on immigrant policy, had already punctuated the monopoly of the ethnic minorities paradigm. The Council’s main argument for this was that its use had reinforced tendencies towards separation and increased the migrants’ dependence on public services, rather than fostering their integration. As an alternative the Council proposed a policy of individual integration into mainstream institutions, such as labour and education, combined with facilities to improve the immigrants’ knowledge of the Dutch language.

Partly as an effect of the second WRR report, awareness grew among policymakers that the minorities paradigm would not hold (see also Scholten and Timmermans 2010). Immigration continued and diversified, developing into a structural phenomenon following the fall of the Iron Curtain after 1989. Moreover, within the oldest migrant communities, a second generation born and raised in the Netherlands began to manifest itself. In 1994, Minorities Policy was officially relabelled as ‘Integration Policy’, and the migrant integration issue began to rise slowly on the political agenda. Policymaking had now reached a more mature stage, characterised by a growing number of policy instruments (e.g. the introduction of mandatory integration courses in 1998) and by stronger vested interests, also from civil society, including migrant organisations.

A gradual alienation between academics and policymakers occurred. ACOM, a strong defender of the minorities paradigm had rejected the 1989 WRR report as ‘unscientific’, thus trying to discredit the researchers involved (Scholten 2011). However, this effort to discredit the WRR failed largely, and ACOM was eventually dissolved in 1992. Migration research now became more fragmented, but also more pluriform. For the academic world this growing disconnection with policymaking did not mean that research funds became scarcer, as alternative funding sources gradually gained importance. The main change was that political agenda setting shifted from academia to policymakers, while the use of research outcomes became more selective. Governmental research organisations like CBS and SCP became more important partners for migrant integration policymakers.

The de-institutionalisation of the research-policy dialogue structure of the 1980s and the first signs of politicisation in the early 1990s coincided with an increasing demand for less conceptual and more ‘practical’ types of research. Policymakers needed large-scale data, preferably based on surveys or derived from existing registration systems. To borrow Peter Hall’s typology, this preference for quantitative data was oriented to first and second-order information (regarding policy instruments and settings of policy instruments) rather than third-order information (regarding the policy paradigm) (Hall 1993). Policymakers from the 1990s onwards were not inclined to use knowledge to reframe migrant integration policies. Several of our interviews indicate that the mobilisation of such data fulfilled a clear role in the interdepartmental coordination of migrant integration policies. The data produced by SCP and CBS allowed the coordinating Ministry of the Interior to organise its information position in relation to other ministries involved in migrant integration. This enabled it to monitor the implementation of policy measures by various departments, to identify potential deficiencies and to put issues on the agenda so that new policy measures could be developed. Also more evaluation studies were needed to assess the impact of earlier policies. Such studies were increasingly commissioned to research units within the respective ministries, like the Centre for Scientific Research and Documentation (WODC) of the Ministry of Justice, or to private consultancy firms.

3.3 The 2000s

In the context of the sharp politicisation of migrant integration in the early 2000s the role of research in policymaking declined even further (Scholten and Verbeek 2014). Some speak of the rise of ‘articulation politics’, or a form of politics aimed at gaining legitimacy by responding to ‘the voice from the street’ (Verwey-Jonker Institute 2004). This became manifest soon after the dramatic 2002 elections when the newly elected Dutch parliament adopted a motion asking for an investigation into ‘why immigrant integration had failed’, without even asking if it had failed. The Parliamentary Committee set up for this purpose commissioned the Verwey-Jonker Institute, an independent institute for social research, to study this state-of-affairs, mainly based on existing research (see also Chap. 5 on consultative commissions). Their final report was heavily criticised in parliament as well as in the media for being biased and overly sympathetic to left-wing ideologies. In these debates the credibility of the social researchers involved in this institute was openly put on the line. Social researchers were blamed for having been too involved in policymaking and for reflecting a multiculturalist bias. At the same time, policymakers and politicians who had been active in this area were blamed for lacking a clear political vision and for having delegated too much of the policymaking to researchers (Scholten 2011). The days of technocratic symbiosis were definitely over. Likewise, when the Parliamentary Committee itself came with its final report, concluding that immigrant integration had been relatively successful, it was also heavily criticised by many actors in both parliament itself and in the media.

Research on immigration and particularly on integration is now being initiated in many more disciplines than before, including law, economics, psychology, and medical sciences. It is also much less policy-oriented than in the early days and theoretically richer and broader. The growing availability of databases and their improved accessibility have led to a much larger number of quantitative studies. Numerous PhD dissertations have been written in this field, which in less than four decades has become one of the core specialisms, certainly in the social sciences. The growing availability of European research funds has led to a dramatic increase in comparative studies, thus making Dutch researchers less dependent on national funding sources, many of which have been drying up anyway. Moreover, as policymaking is becoming more decentralised, local authorities have been taking up an interest in funding research. They often cooperate closely with local universities. On a critical note, it has been argued that the reliance not only of governmental researchers but also of university researchers on official ethnic data in the Netherlands contributes to the continuous substantiation of ‘the ethnic lens’, which sees socio-cultural explanations for every societal problem (Ham and Van der Meer 2012). Similarly, producing and using data on so-called ‘social-cultural distance’ between allochtonen and autochtonen of course confirms the idea that something like ‘social-cultural distance’ exists.

Finally, the Dutch DIAMINT study has found that civil servants in the Netherlands nowadays work with a ‘continuum of trustworthiness’. The trust worthiest source of expertise is considered SCP, because SCP investigates the questions that the Ministry is interested in by using the best-known sampling and statistical techniques. Somewhat halfway are researchers who are capable of meedenken, translatable as the ability to ‘think like a’ policymaker. The position of this middle group is a clear indication of how ‘policy-relevant expertise’ is constructed in the knowledge utilisation process itself: the expertise does not ‘exist’ beforehand (Caponio et al. 2014). The least trustworthy are scientists who in the perspective of the bureaucratic officials take extreme positions or who are unpredictable from a ministry’s perspective. Put simply, the positions of these experts are either known beforehand – which means that it is not necessary to invite them to hear them repeat their point of view – or completely unknown beforehand, meaning that it is risky to invite them to participate in research-policy dialogues, because the expert may propose ideas that are totally at odds with current political realities.

4 An Exploration of Three Subtopics in Research-Policy Dialogues

The analysis in this section proceeds in three steps. In the Netherlands political and media attention is often focused on the three I’s: immigration, integration, and Islam. Consequently, the subtopic ‘Islamic schools’ is an interesting case for exploring the combined politicisation and mediatisation of research-policy dialogues in migrant integration. With reference to Islamic schools, it is clear that policy dialogues have become not just strongly politicised but also mediatised, which will affect the nature of research-policy dialogues as well. The subsequent analysis of the research-policy dialogue on educational segregation, the second subtopic in this section, suggests that for that policy topic the impact of mediatisation has been more limited. The third subtopic, naturalisation, explores whether mediatised and non-mediatised research-policy dialogues on this politicised topic not only exist alongside each other, but whether they also constitute alternative communication channels between researchers and policymakers.

4.1 Islamic Schools

Most interviewees argued that after 9/11 the discussion about Islamic schools had changed – in fact, before 9/11 it was hardly an issue at all. It was simply taken for granted that Muslims in the Netherlands were using their constitutional right to establish their own schools. Like the Catholics in the early twentieth century, this would probably facilitate their societal emancipation. Prime Minister Ruud Lubbers (1982–1994, a Christian Democrat), for example, had encouraged Muslims to take this route. In 2003, however, then Prime Minister Jan-Peter Balkenende (2002–2010, also a Christian Democrat) warned that Islamic schools should not become ‘prisons of disadvantage’ (cited by Karsten 2006). Yet at the national level, there have been no official policies against Islamic schools. Their number is ‘not large’: in 2008, there were 44 Islamic schools for primary education in the Netherlands (Herweijer 2009: 43). Only 6 % of all pupils with a Moroccan or Turkish background were enrolled in them. Still, opponents have argued that Islamic schools worsen segregation and hinder integration (ibid.). Observations by some of the respondents confirm the argument by some scholars (Entzinger 2010; Sunier 2014) that the migrant integration debate in the Netherlands has ‘Islamised’, rather than the Netherlands itself (Fortuyn 1997). The mass media’s obsession with ‘Islam’ sometimes comes to the fore in surprising ways, as it can affect the use of expert knowledge even in seemingly unrelated domains.

One respondent (interview with educational researcher), for example, recalls how his research (with colleagues) on homeschooling was received initially. Homeschooling in the Netherlands mainly involves a small number of traditional Christians, who feel that no school adequately reflects their religious beliefs. The researchers basically concluded that ‘no dramatic’ situations existed (interviews with several scholars), i.e. that the parents involved educated their children well and that their children participated in society in an adequate manner. Parliament and the Minister of Education decided that no (additional) law was needed to regulate homeschooling. Two months later, however, one Islamic school had to close down amidst fierce media attention due to concerns over quality, and some parents said that they would rather start homeschooling than send their children to a non-Islamic school. Suddenly media interest in research on homeschooling exploded (interview). The minister responsible was quick to announce that he would scrutinise the existing regulations for homeschooling to prevent the Muslim parents’ homeschooling plans. In other words, the initial effect of the study was that policymakers concluded that they did not have to worry about the topic of homeschooling too much. No legislation process was started, even though the researchers had advised the introduction of additional laws to regulate homeschooling. Due to the publicity about the Muslim parents’ homeschooling intentions, however, in the second instance policymakers seemed to draw different lessons from the study.

Can we conclude that the political primacy in research-policy dialogues, which we noted earlier, has been replaced in some sub-domains by something like ‘media primacy’? In the next section, a key episode in the research-policy dialogue on educational segregation in the Netherlands is analysed to illustrate the fact that migrant integration researchers sometimes play a very pro-active role in mediatised research-policy dialogues themselves. Neither policymakers nor individual experts are merely playthings of media forces.

4.2 Educational Segregation

The issue of educational segregation in the Netherlands, often referred to as the issue of ‘black’ versus ‘white’ schools, has been on the public agenda since the 1980s (Herweijer and Van den Brink 2011, p. 1). Educational segregation in the Netherlands can largely be explained by residential segregation, but it is also related to the Dutch system of free school choice (Herweijer 2009). Parents are not obliged to choose for their children a school in the area where they live. National policies to combat educational segregation have been virtually absent. Since 2006, though, ‘school boards, municipal authorities and childcare providers have been required to consult with each other with a view to achieving a more balanced distribution of pupils across schools’ (Herweijer 2009, p. 14). In the period 2007–2011, the national government also supported pilots in 12 cities ‘intended to identify effective interventions at a local level to reduce segregation’ (ibid.).

Overall, educational research in the Netherlands is a well-developed area, with most of the funds being controlled by the government directly or indirectly (via the Netherlands Organisation for Scientific Research, NWO). Several established research institutes compete for funding. For these institutes, and the individual researchers affiliated to them, being part of this network is essential. Researchers outside of this dialogue structure find it much harder to get their projects funded and to get direct access to policymakers. Nonetheless, one of the best-known of these ‘solitary experts’, Professor Jaap Dronkers of Maastricht University (formerly European University Institute Florence, 2001–2009), pro-actively uses the media to inform the public and, by extension, policymakers.

In June 2010, Jaap Dronkers gave his inaugural lecture at Maastricht University, just after the success of Geert Wilders of the anti-immigrant Freedom Party in the national elections. Based on PISA data (in particular on reading skills of 15-year-olds), he argued that ethnic diversity in schools hampers educational achievements, also in primary education. In its press statement, Maastricht University emphasised certain ‘newsworthy’ elements of the lecture, but it did not mention that the data used pertained neither to the Netherlands nor to primary school pupils. The professor also pro-actively sought media attention himself and he received a lot of it (national TV news, several newspapers, including the front page of NRC Handelsblad). On its website, the largest newspaper of the Netherlands stated that ‘Many races in classes leads to poor achievement’ (de Telegraaf).Footnote 2 For the first time in years, Jaap Dronkers was invited to give a lecture at the Ministry of Education, where he summarised his inaugural lecture. A couple of months later, the new Minister of Education, in a cabinet that depended on support from Geert Wilders, announced that the national government no longer aimed at reducing segregation in education.

On the one hand, this episode could be depicted as a clear example of how social scientists can influence public policies on controversial issues via the media. Such a storyline would reveal how his ‘dramatic’Footnote 3 contribution silenced the argument in favour of fostering ethnically mixed schools. It became the ‘final blow’ – or at least it gave ammunition to the Minister (interview with educational scholar). On the other hand, jumping to this conclusion would be paradoxical, because to a considerable extent this research-policy dialogue centred on whether Jaap Dronkers had behaved unscientifically, i.e. whether he had made ‘rather definite and generalising statements on the basis of one study’, ‘especially in the media’.Footnote 4 The DIAMINT fieldwork made clear that even in this case, mentioned by virtually all respondents as an example of an extremely mediatised research-policy dialogue, the expert did not ‘force a political decision via the media’. Instead, his lecture was used to substantiate a pre-existing political preference, and party political dynamics offer a convincing explanation for this. In contrast, in 2007 – i.e. in a less contested political context – a similar critical research report was not used to abolish the national policy against educational segregation. Even more than before, fighting educational segregation was simply framed in an ‘integration and citizenship’ perspective instead of an ‘educational achievements’ perspective (interview). In other words: this extreme case shows that it is too simple to argue that ‘experts can choose to influence policymakers via the media’. Instead, actual knowledge utilisation always seems to be the result of a complex interplay between political factors and factors related to the media.

The DIAMINT research also suggested that the success of Jaap Dronkers in getting media attention comes with certain costs. As noted above, some other researchers found his behaviour ‘unscientific’: they did not trust the validity of his claim. Furthermore, by taking a quite extreme position – which of course can partly explain why he received so much media attention – Jaap Dronkers also ran the risk of alienating rather than pleasing policymakers.Footnote 5 These themes will be picked up in the following section on naturalisation.

4.3 Naturalisation

In the 1980s, access to formal citizenship was still seen as a useful step in the integration process. Nowadays most parties define it as its pinnacle, or the ‘first prize’ as Rita Verdonk, Minister for Immigration and Integration in two centre-right cabinets (2003–2007), called it. In the past decade, research-policy dialogues on naturalisation mainly concerned the attempts by subsequent governments to make access to Dutch nationality more difficult. Several plans were introduced publicly, but were actually never put to a vote in parliament. Other attempts, however, succeeded (Van Oers 2013), for example when standardised naturalisation tests were introduced in 2003 (interviews).

Research on naturalisation in the Netherlands is scarce. It is mainly done by legal scholars, some of whom have been watching – and partly influencing – the developments in this domain for decades. The policy positions of these experts are generally less strict than those of politicians. They argue, for example, that immigrants should not have to denounce the nationality of their country of origin. Dual nationalities for them simply reflect the fact that people can have durable bonds with more than one state. However the manner in which they defend these positions (i.e. the tone of their message) is quite different. Furthermore – and crucially – different experts tend to use different channels to try and influence the policymaking process in this domain. Some have a very pro-active role in mediatised research-policy dialogues and regularly criticise the national government in public debate. This often involves using strong language. Others keep quite a low profile and hardly perform in the media at all. This does not mean, however, that the former have a stronger influence on policymakers (via the media) then the latter. ‘Behind-the-scenes’, that is in relatively unknown national and international legal advisory bodies or in private settings like informal lunches, the impact of individual experts can be ‘quite substantial’ (interview). However, it is important to distinguish between ‘details and major issues’: ‘The political climate (…) is very important. (…) But okay, within that politicised framework one tends to listen to scientists and there is still some space, especially for legal advice’ (interview).

This view on mediatised and non-mediatised research-policy dialogues on naturalisation is not only advanced by experts themselves. As a former top civil servant notes, some experts ‘can be very cynical and sarcastic’, always wanting ‘to take some further steps’. Yet experts who are too ‘vigorous’ are ‘a bit difficult to cite’ (interview). Other experts give priority to taking ‘into account the political field surrounding the issue at hand’.

Summing up, the fieldwork on naturalisation confirmed the hypothesis derived from the fieldwork on educational segregation that there are relations between the tone and content of the messages of experts, their media presence, and the usefulness of their expertise as perceived by policymakers. Crucially, direct (formal or informal) access to policymakers emerges as an important factor, rather than indirect access via the media. There are indications that experts who have no alternative channels choose more active roles in mediatised research-policy dialogue structures.

The roles taken by individual scientists in mediatised research-policy dialogues may also be connected with knowledge utilisation outcomes. Experts whose positions in the public debate are considered ‘too radical’ by policymakers may no longer receive commissioned research assignments, formal invitations to join advisory committees, or informal invitations to meet top officials and ministers. As Engbersen (2009) notes, the ‘loyalty-option’ implies some degree of ‘political realism’, i.e. conformity to the parameters perceived by the powers in office.

5 Conclusions

As noted in the theoretical framework of this book, orthodox approaches to research-policy dialogues take the ‘two worlds metaphor’ as their starting point. The idea that one can somehow separate ‘science’ from ‘policy’ is implied by the very concept of research-policy dialogues. One of the key findings of the project, however, is that research-policy dialogues on migrant integration are characterised by extreme ‘role convergence’, to use Hoppe’s (2005) phrase (also see Wittrock 1991). Put simply, migrant integration policymakers do their own research and migrant integration researchers ‘make policies’, in various ways (Caponio et al. 2014). Very few scientists nowadays ‘lock themselves up in the ivory tower’. In other words, the border region between science and policy is not a ‘no man’s land’. Furthermore, we can also see a very strong role convergence when it comes to behaviour in the mass media: there are both scientists and policymakers who try to use the media to their advantage and they use similar strategies. It is one of the hallmarks of contemporary politicised research-policy dialogues on migrant integration that public officials have most trust in those researchers who depend on them most. And they mistrust so-called ‘independent scientists’, notwithstanding peer-review mechanisms, because they feel that these people have a (hidden) political agenda and are interested in promoting their political ideals.Footnote 6 The scientists after all may even propose certain measures on the basis of empirical research and call them ‘evidence-based policies’, while in reality the evidence is not so strong and the recommendations are mainly based on normative preferences. Of course, due to the direct financial links between for example SCP and the ministry it advices, the argument could also be made that SCP researchers are less ‘objective’ than seemingly untouchable professors at universities.

This chapter contended earlier that the current literature on research-policy dialogues in the Netherlands is primarily concerned with politicisation. The section on the three subtopics explored the role of mediatisation and tentatively concluded that more attention to this aspect is warranted. The empirical analysis suggests that mediatised dialogue structures differ from non-mediatised ones as opportunity structures for influencing migrant integration policy. In fact, there may be a new Boundary, which is the one between non-mediatised and mediatised research-policy dialogues. By this we mean that scientists who wish to have policy impact can take two routes nowadays. The first, more traditional route is to communicate with policymakers directly. In practice this mostly happens in private, i.e. without media coverage, in so-called ‘boundary organisations’ like committees, advisory bodies and think-tanks, at conferences, informal workshops, dinners, etc. The second route is to communicate with policymakers indirectly, via the media. In the project we have found numerous examples of public intellectuals who have developed media strategies to influence migrant integration policymaking. One interesting avenue for future research, we believe, is to find out whether the two routes ‘add up’, i.e. whether researchers have to choose or not. There is some evidence (also see the British case) that both routes are to a certain extent alternatives. This is because communicating via the media has a ‘cost’: put simply, scientists who choose the public route tend to take more critical positions and this decreases the likelihood that they also interact with policymakers directly in more private networks. There are relations between the tone and content of the researchers’ messages, their media presence, and the usefulness of their expertise as perceived by policymakers. For example, some researchers see the world of newspapers as ‘a totally different reality’ (interview). Scientists who do not have direct, private access to policymakers feel that they have no other option but to write letters to newspaper editors or to appear on radio or television. They also want to have an impact – on policy positions, public opinion, or both – and believe that writing only academic publications will not suffice.

In summary, up until the early 1990s the Netherlands was characterised by a strongly institutionalised research-policy dialogue structure that prevented politicisation and promoted a monopolisation of the ethnic minorities paradigm. Politicisation in the 1990s, and even more so in the early 2000s, punctuated this symbiosis, allowed for more pluriformity in knowledge claims and also led to more open knowledge conflicts. In situations characterised by a multiplicity of frames, actors with different frames can have very different selections or interpretations of ‘evidence’. The growing fragmentation in the field of migration research and also the more open manifestation of knowledge conflicts – in the mass media – seems to have contributed to this trend. Perhaps this trend reflects a more general trend in Dutch politics away from policymaking ‘behind closed doors’ (and thus in a non-mediatised setting), which has been fiercely criticised by populist politicians since the turn of the millennium. Hence the evolution of Dutch dialogues on migrant integration may have largely kept pace with more macro-institutional developments in research-policy dialogues in the Netherlands. Research-policy dialogues on migrant integration appear to have been significantly affected as Dutch political culture has been moving away from its traditional ploy of consensus building in private settings towards a more confrontational style of politics in public, mediatised settings.

6 A Note on Methodology

This chapter is based on a review of the literature, document analysis, analysis of newspaper articles, and semi-structured interviews with 33 respondents. The interviews (carried out between late 2011 and early 2013) lasted about 1.5 h on average. They were all recorded and transcribed. Most respondents were prominent researchers from universities (19) and (former) senior civil servants from the national government (6). Some respondents’ key affiliation was with boundary organisations such as SCP and WRR (5). Finally, we also interviewed some key figures from the media realm (3). We would like to thank all respondents for their participation. Due to the sensitivity of the topic, however, they shall remain anonymous.