More than 200,000 migrants and refugees arrived at the Scandinavian (Denmark, Norway, Sweden) borders in 2015 (Eurostat 2016). The numbers steadily increased throughout the year, reaching a peak in October (UDI 2015; Bendixen 2016; Migrationsverket 2016). In the initial phase of this influx, widely characterised by the media as a ‘refugee crisis’, willingness to receive even more refugees increased in Scandinavia (Gallup 2015; Ipsos 2015), but as the crisis unfolded, the rhetoric shifted. From late 2015 to early 2016, the Scandinavian countries (all Schengen members) introduced the internal border control allowed by the Schengen acquis ‘in the event that a serious threat to public policy or internal security has been established’ (EC n.d.). The implementation was criticised by scholars and UN representatives alike, however, for restricting the right to apply for asylum, and thus undermining international conventions such as the Refugee Convention (Gammeltoft-Hansen 2018; Nielsen 2016; SVT Nyheter 2016). A handful of politicians claimed that the implementation eroded Scandinavian values such as equality and solidarity, often considered a key element of the so-called Nordic model (Trägårdh 2007). How exactly then were Scandinavian authorities able to legitimise implementing border controls and risk breaking the Refugee Convention? This is the basic question the article aims to answer through examining how different actors in the Scandinavian news media legitimised or rejected border control during the ‘refugee crisis of 2015’ and the contexts in which border control was discussed. Although challenged by constant change, the traditional news media continue to have a great influence on public perceptions of world events (Peters & Broersma 2017). It is therefore a highly pertinent place to study how border control was legitimised.
Border control was one of several measures implemented during 2015/2016 as part of a more restrictive immigration policy in all three Scandinavian countries. Others included offering temporary (instead of permanent) residency, limits on financial support and restrictions on family reunification (see Barker 2018; Brekke & Staver 2018). Border control implementation was, however, an important means of restricting immigration, both symbolically and in practice. First, according to the UN, the new border control posed a threat to the Refugee Convention (Nielsen 2016). Second, it had an immediately apparent effect as the number of migrants who crossed the Scandinavian borders dropped instantly (Bendixen 2016; Migrationsverket 2016; Østby 2015). This reduction in numbers might not have resulted solely from the Scandinavian border controls, but the timing does suggest that it is of vast interest to analyse the arguments that led up to this decisive moment.
When border securitisation, apparently aimed to limit the inflow of migrants and refugees, emerges in countries that seem to represent ‘Scandinavian solidarity’, I argue that the welfare state provides a particularly suitable arena to intertwine a securitisation discourse with a humanitarian discourse. Building on the three concepts, securitisation, humanitarianism and welfare state sustainability, I propose the term compassionate (border) securitisation to describe the overarching legitimisation of border control. This linguistic construction serves to maintain the public perception of generous, solidary welfare states, as well as the states’ self-image as guardians of humanitarianism and refugee rights. Humanitarianism and welfare sustainability thus become the very condition of securitisation while simultaneously reproducing the existing order of power. In the data, an alleged concern about states’ inability to provide asylum seekers with sufficient shelter and welfare benefits becomes the main argument for sealing the borders.
I begin by addressing the Scandinavian context and theories on borders, crises and the media. I then present the methodology and the analysis before outlining the discourses in the data (the contexts within which border controls were suggested) and the specific arguments used to legitimise the new controls. Finally, I summarise and discuss the findings.
The Scandinavian countries are similar in many respects, especially in their cultures, economies and policies. Having historically promoted human rights internationally, they have also upheld a self-affirming picture of themselves as ‘humanitarian superpowers’ (Löfgren 2017). The welfare state, often considered the hallmark of Scandinavian countries, relies upon the redistribution of wealth and resources to create a social safety net (Brochmann & Hagelund 2011). The concepts of equality, solidarity and trust are key elements in the Scandinavian welfare states (e.g., Andersen 2008; Kjølsrød 2003). A common concern in such states is the potential challenge posed by immigration (Freeman 1986), as more diverse populations are frequently suggested to decrease the solidarity and trust upon which their redistributive systems rely (Reeskens & van der Meer 2019).
Despite their common aims and concerns, however, the Scandinavian countries do differ on immigration policy. While Swedish immigration policy has until recently been regarded as liberal, Denmark is more conservative and has opted out of the common European Union (EU) immigration policy to retain full authority over immigration. Norwegian immigration policy is generally somewhere in between. While acknowledging the undeniable upheavals of recent years, Brochmann (2018) maintains that Sweden and Denmark usually occupy opposite poles in the field of Nordic immigration policy. These policy differences are reflected in the number of applicants for asylum in Scandinavia during 2015: around 215,500 arrived in total, of whom 21,316 applied for asylum in Denmark, 31,150 in Norway and 162,877 in Sweden (EC 2016).
This study is situated within two fields: comparative media studies and migration studies. Theoretically, it draws upon the concepts of humanitarianism and securitisation. Europe’s outer border policy has been dominated by two seemingly contradictory developments (e.g., Aas &Grundthus 2015; Chouliaraki, Georgiou & Zaborowski 2017; Walters 2011; Watson 2011). On the one hand, there has been the progressive securitisation, surveillance, and militarisation of the Schengen border, reflecting the EU’s efforts to identify and control the mobility of certain people, services and goods (see Vaughan-Williams 2015), including an asserted effort to combat terrorism and human smuggling (Horsti 2012). On the other hand, human rights and humanitarian forms of solidarity at the border have gained prominence, and led to the increased presence of aid organisations at the border, EU rescue operations to save migrants at sea and border personnel who also administrate shelter for refugees (Chouliaraki and Georgiou 2017; Musarò 2017). These developments are also embedded in EU policy documents that describe them as protective both against possible threats (European Parliament 2019) and in favour of refugees and their rights (European Parliament 2015). The intricate entanglement of these conflicting approaches – securitisation and humanitarianism – has been referred to as ‘humanitarian bordering’ (Walters 2011), in which practices, policies and rhetorics of aid and rescue merge with those of control and securitisation. For example, Horsti (2012) found that Frontex (the European border police) employs humanitarian rhetoric and concepts such as claiming to ‘save lives’ and ‘protect human rights’ when describing military operations, border control and migrant detention. In other words, both securitisation practices and communications are legitimised and concealed through humanitarian discourses (see also McMahon & Sigona 2020).
Crises are often cited to justify securitising and sealing borders (Musarò 2017). Gatrell (2016) suggested that articulating the events of 2015 in terms of crisis failed to acknowledge that the refugee crisis had been going on for decades. Van Reekum (2016) also proposed that framing the events of 2015 as a crisis legitimated security responses. Bleiker et al. (2013) found that asylum seekers and refugees in Australia have primarily been portrayed as arriving in excessive numbers, and therefore the refugee crisis was seen not as a humanitarian disaster but as a potential threat (see also Horsti 2017). Framing events of 2015 in terms of threat, security and humanitarianism also seem to have dominated the media coverage of the refugee crisis of 2015 (e.g., Berry et al. 2015; Hovden, Mjelde & Gripsrud 2018; Triandafyllidou 2018). Chouliaraki et al. (2017) concluded that media coverage of the crisis in eight European countries was largely informed by the framework of humanitarian security in articles featuring the need for care and compassion towards refugees. Coverage that supported increased militarisation, such as sending refugees and migrants back, closing borders and increasing the presence of police or guards, portrayed such actions as ways to protect the country. An extended study by Hovden et al. (2018), including Scandinavian data from April 20–May 1, September 3–16 and November 16–27, 2015, found that Swedish media coverage was much more positive towards refugees than Danish.
Scandinavian scholarly literature shows that Scandinavian media generally portray asylum seekers as either threats or victims (Horsti 2008; Thorbjørnsrud 2015) and over the last decades have discussed immigration increasingly through problem-oriented frames and discourses (e.g., Brune 2004; Eide 2002; Horsti 2008; Hovden et al. 2018; Madsen 2004). For example, scholars have found that Danish media coverage of immigration has focused on the dominant themes of crime, social problems and ‘Danish values’ (Madsen 2004; Hussain 2000). Meanwhile, several other studies document how the ‘victim’ or solidarity frame prevails in particular contexts (e.g., Benson 2013; Figenschou and Thorbjørnsrud 2015; Van Gorp 2005).
This study examines articles published in two of the biggest newspapers in each of the three Scandinavian countries: Aftenposten (AP) and Verdens Gang (VG) in Norway, Politiken (PT), and B.T. (BT) in Denmark and Dagens Nyheter (DN) and Aftonbladet (AB) in Sweden. In 2014, the daily readership of the print edition of each newspaper was (in thousands): AP: 576; VG: 483; PT: 306; BT: 164; DN: 702; AB: 739 (Nordicom 2017). The methods applied are quantitative content analysis and critical discourse analysis (CDA). The articles are collected through the databases Retriever (Norwegian and Swedish sample) and Infomedia (Danish sample). The search string applied was *asyl* [asylum] OR *migra* [migrant] OR *flygtning*/*flyktning*/*flykting* [refugee] OR *innvandr* [migrant], which included all variations of the different words. The period investigated, 1 October 2015 to 31 January 2016, included the weeks leading up to the decisions to implement border controls (Sweden: 12 November and 4 January; Norway: 26 November; Denmark: 4 January). Because Scandinavians increasingly read news online (Nordicom 2017), those sites are also important to investigate. A search of Infomedia, however, returned only a handful of articles published exclusively online, whereas all print articles were available through both Infomedia and Retriever. I chose to limit the investigation to printed sources, although most of the print articles in this study are also published online. All genres, reporting genres as well as opinion pieces, were included. This broad search string and the ubiquity of the matter during the autumn of 2015 resulted in several articles that mentioned the refugee crisis is only one sentence or clause, and these were discarded. If more than half of the article reported or commented on the refugee crisis or Scandinavian border controls in 2015/2016, it was included in the study. This culminated in a total of 3176 articles and opinion pieces, 601 of which included specific arguments for or against border controls.
I then conducted the quantitative coding. The aim of the coding was first to identify the discourses present in the overall coverage of the refugee crisis, that is, the possible contexts in which border control was discussed. The next aim was to map the various arguments raised by different actors on whether or not to implement border control. This means the mapping of discourses is based on all 3176 articles, whereas the second part on border control is based on a smaller sample of 601 articles. I coded the articles according to the following variables: media outlet, section, size, genre, sources, human interest frame, word use and labelling of the crisis (e.g., refugee wave, refugee stream), overt themes and border control legitimatisation. The themes were identified after briefly skimming the headlines and first few lines of the articles. Through this skimming, I identified 36 frequently presented themes such as system collapse (referring to regions and institutions lacking the capacity or resources to handle the migrant arrivals), individual migrants as threats and personal stories highlighting refugees’ difficult circumstances. Other rarer themes were coded as ‘other’. Multiple themes could be present in one article; therefore, it was possible to code up to three different themes per article. The variable on border control legitimatisation was only applied to the 601 articles that contained arguments for or against border control. This category was rather simple to operationalise, as quotes that legitimised or rejected border control were explicit and occurred in more or less identical linguistic formulations (see Figure 1). To test for coder reliability, 5% of the articles were re-coded by a second coder. Most of the variables had an acceptable coder agreement (i.e., we coded the articles identically). Intercoder reliability scores were 82% for the theme variable, which is considered acceptable agreement (Østbye 2007) and 96% for the specific category of legitimising border control.
The different discourses (here also referred to as contexts) present in the material were identified through CDA. The premise for any discourse analysis is the rejection of universal truths, which implies that factual events are placed in particular contexts by the use of language (Fairclough 2013). CDA is often associated with qualitative text analysis, but this study is based on a quantitative content analysis. After coding the articles according to the 36 themes, the voices present, and the labelling and word use, I conducted an in-depth close reading of a selection of the coded articles to further investigate the most prominent themes. From the close reading, I drew out the three overarching discourses outlined as follows; each of the three discourses is based on multiple themes. The examples in the analysis were extracted during this close reading. Several discourses may be at work in each text, as there is a constant struggle among discourses to achieve hegemony (Fairclough 1995). Thus, the discourses outlined must not be seen as inseparable, but rather as different overarching discourses at work together and against each other in the texts.
The study aim is to identify the contexts in which border control, and its explicit legitimatisation or rejection, was raised during the refugee crisis. To investigate these contexts, I first map how Scandinavian news media portrayed the crisis during the autumn of 2015 and early 2016 through three overarching discourses: threat, solidarity (or humanitarianism) and (welfare) sustainability. After outlining the three discourses, I move on to describe the explicit legitimatisations of border control and discourses present in the articles specifically discussing border control. In the following, I provide a brief description of the sample before moving on to present the discourses in more detail. The examples of discourse presented throughout this section are drawn from the data.
As seen in Table 1, Sweden had the highest number of articles on the refugee crisis, but Denmark had the most in which border control was legitimised (or rejected). This is probably because Denmark was the last of the three countries to introduce border controls, but the Danish press also wrote frequently about the controls implemented in Sweden and Norway. The three discourses outlined in the following are the most common discourses present in the entire sample of 3176 articles across the three countries. Other and more marginal discourses were present, such as a ‘benefit discourse’ arguing that the society would profit from receiving more refugees, but I describe only the three dominant discourses.
|COUNTRY||OUTLET||ALL ARTICLES||ARTICLES BORDER CONTROL (ISOLATED)||TOTAL||TOTAL BORDER CONTROL (ISOLATED)|
Earlier studies have found that asylum seekers often are portrayed as threats (e.g., Benson 2013; Horsti 2008; Hovden et al. 2018). The threat discourse in this sample similarly discusses security measures to protect ‘us’ from asylum seekers that constitute ‘a potential danger for society’ (AP 17/11/15). The threat discourse is present in 33.5% of the entire sample, mainly in the Danish (51%) and Norwegian (34%) articles, and less so in the Swedish (22%). The most common theme coded, political measures to protect the country from high immigration, reported on intensified police presence or increased security measures (27.5% of the Danish sample, 17.5% of Norwegian and 14.5% of Swedish). For example, in an article on increased grants for the police, AP (27/11/15) reported the police as saying, ‘We can now allocate this towards intensifying the control of our territory.’ The second most common theme was individual asylum seekers as threats (terrorism, criminality, untrustworthiness; 14.5% of Danish sample, 11.5% of Norwegian, 7% of Swedish). Beneath the headline ‘Found: Pictures of Severed Heads on Mobile Phones Belonging to Asylum Seekers’, VG (01/12/15) reported that ‘[t]he Police Security Service has received hundreds of inquiries expressing concern about specific asylum seekers. Among the findings are pictures of severed heads and IS [Islamist State] symbols’. This article stresses the risks of providing asylum for refugees, implying that dangerous people might obtain asylum if no preventive measures are taken. Crimes or assaults are also widely represented within this theme. For example, a story titled ‘Asylum Seeker’s Touching Being Investigated’, about an asylum seeker in a discotheque who touched a woman between her thighs, made it to the front page of PT (16/01/16). There is no doubt that men who touch women against their will present a grave problem; however, not every sexual assault reported to police reaches the front page. When it did so in this case, it contributed to an increased focus on asylum seekers as threats. Other articles within this theme suggest different ways in which asylum seekers allegedly ‘trick’ the system. For example, BT contains several pieces alleging that asylum seekers convert from Islam to Christianity to ‘improve their chances of obtaining asylum […]’. The Danish right-wing politician making this claim continues, ‘It is absurd. This is a way to evade the legal asylum system’ (BT 09/10/15). Articles such as this contribute to the image of asylum seekers as untrustworthy people who do not respect the rules. Another frequently identified theme was migrant’s values as a threat to ‘our values’ and the risk of ‘parallel societies’ based on alien values (present in 9% of the Danish sample, 5% of Norwegian and 0.5% of Swedish). This focus on ‘values’ corresponds with previous findings (e.g., Hussain 2000). BT wrote that ‘[r]efugees are insulting innocent, Danish women’ and quoted a politician as saying
They [the refugees] should not walk around freely […]. Women are supposed to walk around freely, and asylum seekers have to respect that. They have to control themselves and leave their unpleasant views on women behind in their home country. (BT 13/01/16)
Asserting that men ‘have to control themselves’ implies that male asylum seekers are culturally conditioned to rape or sexually harass women, and they, therefore, threaten ‘our culture’. This excerpt also privileges Danes over ‘the other’ by assuming Danish women have the right to go wherever they want, whereas refugees do not.
Of the articles coded with themes that indicated the threat discourse, 70% were coded either with only one theme or with several themes within the threat discourse. (e.g., some articles were coded both with the theme ‘values as a threat’ and with ‘individual asylum seekers as threats’). This means that although several discourses are often at work within the same text, most articles that thematised asylum seekers as threats drew primarily on the threat discourse.
The solidarity discourse thematises the circumstances and rights of individual refugees as the responsibilities of local communities or the nation-state towards them. Themes that constitute this discourse are present in 41.2% of the total sample (Danish sample: 32%, Norwegian: 42.5%, Swedish: 50%). The most frequently present theme, calls for solidarity/formal rights of refugees, appears in 18.5% of the Danish sample, 16% of the Norwegian and 13.5% of the Swedish. For example, the Swedish prime minister figured in several articles urging ‘[e]veryone to take responsibility’ (AB 28/10/15) and ‘exercise solidarity’ (DN 01/10/15). In an article on restrictive asylum policies, two professors of law stated that ‘[t]he way the government wishes to administer the asylum policy does not justify the violation of mandatory principles of the Refugee Convention or human rights’ (AP 22/01/16). Several of these articles included quotations arguing that history would judge the nation harshly if it did not accept the refugees. For example, an AB journalist asked, ‘Will we ever be able to look ourselves in the mirror again without being ashamed [if Scandinavia does not admit the refugees]?’ (AB 17/10/15). The second most frequent theme was refugees’ personal stories/human interest stories (Danish: 7% of the sample, Norwegian: 9%, Swedish: 9%). For example, VG (09/11/15) tells the story of a Syrian man and the consequences had he not left Syria. ‘I would have died, I would’ve never seen them [his family] again. I have lost everything by coming here, except the most important things, my wife and children.’ Several articles feature the stories of children fleeing alone. AP reported on the long journey of a young refugee who just arrived in Norway: ‘“[My dream is] that my mum, my dad, and my sister will meet me here,” says Ali (12). Then he starts to cry’ (AP 29/11/15). Offering a human face to a story often serves to evoke the public’s empathy or create an emotional angle (Figenschou & Thorbjørnsrud 2015). Here, the refugee is constructed as a vulnerable victim.
Another frequent theme (4.5% of Danish sample, 11% of Norwegian and 8% of Swedish) illuminated the poor conditions for refugees in general or in the asylum centres (e.g., BT 04/01/16). Another noticeable idea within this discourse, predominantly in Sweden (1.5% of Danish sample, 3.5% of Norwegian and 10% of Swedish), is the construction of the national self as a ‘humanitarian nation’. For example, in AB, the Swedish prime minister is quoted as claiming right-wing extremist values to be ‘un-Swedish’, because ‘[t]he DNA and soul of Sweden is warmth’ (AB 25/10/15). This construction resembles the notion of ‘humanitarian superpowers’ (Löfgren 2017). Moreover, the Swedish sample in particular contains several articles opposed to racism and discrimination against refugees (9.5% of Swedish sample versus 1.5% of Danish sample and 3.5% of Norwegian sample).
The solidarity discourse largely corresponds with previous findings that refugees from areas of conflict or wars, who tell their personal stories, are often portrayed as victims (e.g., Chouliaraki et al. 2017; Figenschou & Thorbjørnsrud 2015). The discourse, rooted in judicial conventions or the simple perception that solidarity is part of being human, argues that ‘we’ have a moral responsibility to help refugees. It is sometimes even argued that taking that responsibility is a fundamental Scandinavian virtue. Of the articles coded with themes within the solidarity discourse, 68% were coded either with only one theme or with several themes defined as constituting the solidarity discourse; most articles with themes of solidarity drew only on the solidarity discourse.
Migration concerning the welfare state and the sustainability of the nation has been thematised in various studies on immigration (e.g., Freeman 1986). The discourse of sustainability has rarely been presented in previous media research. In my sample, however, articles on the society’s and the welfare system’s ability to integrate immigrants, that is, the ‘sustainability of the welfare state’ (e.g., PT 23/10/15; AB 31/10/15; AP 23/11/15) are present in 47.3% of the entire sample of articles (50% of the Danish sample, 51.5% of the Norwegian and 41.5% of the Swedish). The most frequent theme coded within this discourse was system collapse (18.5% of Danish, 28.5% of Norwegian and 31% of Swedish), which implied that immigration systems were on the verge of collapsing because of the inflow of migrants, or that local or municipal services and institutions were overloaded and thus unable to provide migrants with sufficient housing, support, health services or schooling. The Swedish minister of foreign affairs, for example, expressed concern in DN (30/10/15):
They [a Swedish municipality] don’t know how to accomplish their budget, [they] are left with around 200 children that cannot attend school due to scarce capacity, social services are on their knees […] In the long run, the system will collapse.
The second most frequent theme was the welfare and economic consequences of high migration (17.5% of Danish sample, 17.5% of Norwegian and 5.5% of Swedish), implying that acceptance of a high number of immigrants would be at the expense of the welfare services. For example, a local Danish politician stated in BT (09/10/15) that the general costs of long-term integration mean that ‘we will have to cut down on core welfare services in the long run’. The author of an editorial in VG states that ‘Norwegian politicians are obliged to evaluate what is a sustainable number of newcomers. We cannot go the extra mile merely to put our welfare system at risk in the long run’ (VG 23/11/15). This draws heavily on the political notion of the welfare state as one in which the integration of newcomers into the labour market, and hence their contributions to the tax system, is pivotal. The third most common theme in this discourse was integration and integration policies (14% of Danish sample, 5.5% of Norwegian and 5% of Swedish). Within this theme, various actors, especially in Denmark, regularly argued that the allegedly failed integration of immigrants in the 1980s and 1990s demonstrated the importance of allowing residence permits to only a certain number of immigrants. A politician stated in BT (10/10/15) that
[T]he failed policy of the 1980s and 1990s proves that we need a sustainable distribution of refugees. [This failed policy] created several of the integration problems we see today: ghettos, waiting lists for social housing, people who drop out from school without being able to read and write, and […] all the people who receive economic support.
In this discourse, social problems are frequently blamed on the unproportionate distribution of immigrants over time or in certain districts or countries. The arrival of too many migrants has been framed as a threat in earlier studies (e.g., Bleiker et al. 2013), but the main concern in this sample is not the number of asylum seekers, per se, it is the system’s internal capacity and its ability to provide adequate services for the refugees.
Contrary to scholars who describe the refugee crisis as a phenomenon that has been ongoing for decades (e.g., Gatrell 2016), media statements about the refugee crisis of 2015 (DN 29/01/16; BT 02/01/16) portray it as a crisis primarily for the Scandinavian system rather than for the arriving refugees as a temporary phenomenon, which may be solved introducing temporary political measures. The majority (76%) of articles coded to indicate sustainability were coded with that theme only or with several themes within the sustainability discourse.
The three discourses outlined previously are based on the themes present in the entire media sample covering refugees and asylum seekers from 1 October 2015 to 31 January 2016. The following analysis looks exclusively at the 601 articles that contained explicit arguments for or against implementing border control. According to Fairclough (1995), several discourses may be at work in a text. Although the three discourses outlined cannot be seen as inseparably attached to each article in the sample, many of the articles coded previously drew mainly on only one of the discourses. As I demonstrate in the following, the articles that discussed the legitimacy of border controls tended more than the larger sample to draw on all three discourses simultaneously.
Several of the 601 articles on border control implementation contained various arguments for and/or against its implementation; therefore, it was possible to code up to three arguments per article. Therefore, the total number of arguments, 1041, does not match the number of articles analysed in the part. As seen in Table 1, the Danish outlets published the highest number of articles on border control in general (232), followed by Sweden with 193 and Norway with the fewest (176). As seen in Figure 1, the same pattern holds for the number of arguments for or against border control: Denmark, followed by Sweden, and finally, Norway.
Looking at the explicit arguments in Figure 1, it appears that arguments 1–10 can be more or less clearly linked to the three main discourses: threat, solidarity and sustainability. However, in addition to these explicit arguments, there were several contextual arguments. For example, the reason for Argument 3 (security is threatened) was often that an asylum system on the verge of collapse (sustainability) made it harder to act in solidarity towards refugees (humanitarian discourse; e.g., DN 10/11/15; DN 12/11/15). Each of the articles on border control tended to involve all three discourses simultaneously. Consequently, the discourses merge when border control is discussed, and often one discourse is used to legitimate the other. In the following, I demonstrate how this occurs in the three countries.
As seen in Figure 1, the Danish legitimisation of border control is most commonly rooted in Argument 5 that it is necessary to reduce the number of asylum seekers (14% of the total Danish sample). This argument often arose in a context similar to the second and third most common explanations: national protection (11.5% of Danish sample) and the need for Denmark to have better control (9.5%). These arguments were linked to safeguarding so-called ‘Danish values’ or to economic factors such as the sustainability of the welfare state. Following this reasoning, it would not be fair to the Danish people to welcome too many refugees, as it may threaten the Danish economy and the welfare system. For example, the Danish Prime minister states:
It challenges our economy when we have to spend several extra billions on asylum seekers and refugees. This is money that could be spent on health, education and more jobs. It challenges our social cohesion when many people arrive in Denmark from completely different cultures. [These challenges] may create a situation where we must implement border control. (PT 02/01/16)
Arguments 6–9 often came up in the context of Denmark’s effort to uphold laws or conventions that it was bound to such as providing shelter for refugees but not for other migrants. These arguments were often seen in articles drawing on the solidarity discourse, mixed with arguments that expressed sustainability. For example, the author of an editorial in BT argues for border control:
[It is a] Ragnarok scenario that is threatening us. […] We cannot sacrifice our welfare society in the name of humanitarianism. […] If we want to continue to have a welfare society that is continuously in development, and simultaneously take care of the weakest, [we must only] receive the real refugees, those who actually need help. (BT 03/01/16)
Referring to Ragnarok, the Norse version of the end of the world, the author situates the crisis in a Nordic context, implying that it is a crisis for Denmark, rather than for the people seeking asylum. Furthermore, (s)he pits Danes against the refugees, contending that compassion and solidarity towards the nation and the Danes is at stake.
The phrase ‘Danish citizens’ often explicitly includes minorities in Denmark. In PT, for example, it is argued that those who will suffer from high immigration are Danes who already live in poverty.
[P]eople with an ethnic background other than Danish are already living in rough ghettos all over Denmark. There are already large-scale problems concerning integration and parallel societies in these ghettos, and of course these problems will grow bigger with all the people currently arriving. (PT 09/01/16)
In general, the Danish coverage was more oriented towards solidarity with Danish citizens than the coverage in the two other countries.
Of the arguments in the Danish sample rejecting border control implementation (38%), most argued that border control would lead to more people applying for asylum (9.5%) as it would oblige asylum seekers to register and apply in line with the Dublin regulation. This regulation would force transiting migrants to apply for asylum in Denmark rather than continuing to Sweden, Finland or Norway, illustrating the ambiguous effects of border controls.
The most common argument in favour of border control in the Norwegian sample was Argument 9: the need to distinguish migrants from ‘real refugees’. The relative presence of this argument was 26%. This was often set in the context of the capacity of the reception system (i.e., the system could handle only a limited number of migrants at once). In an editorial, AP wrote that a stricter asylum policy, including border control, was an important political step towards justice for the ‘real’ refugees.
The high number of asylum seekers whose right to protection will be rejected is the direct reason that people with the right to protection must wait for a longer period to get a final answer on their asylum applications. (AP 11/11/15)
The second most frequent argument was Argument 4: to reduce the number of asylum seekers (14.5% of all Norwegian arguments). This often appeared in the context of distinguishing migrants from refugees to reduce the number of applications from people who were not likely to qualify. These arguments were largely justified by reasoning that large-scale immigration challenges both the premise of a sustainable welfare state and the level of equality, considered a fundamental value in such a state. For example, a politician wrote in a letter to the editor,
[t]he worst thing we can do is to create a new permanent underclass consisting of refugees who will have underprivileged lives in Norway, and in the long run they will be a burden on the welfare state and on society. [To prevent this], we must give those with the right to protection a quick answer, and those without this right must be deported. (VG 07/11/15)
Here border control is legitimised based on being just to the ‘real’ refugees, partly by distinguishing them from other migrants, and partly because of the welfare state’s alleged capacity to incorporate only a certain number of migrants.
In addition, actors quoted in Norwegian media promoted security arguments related to seeing the arriving asylum seekers as a potential threat. The third most frequent argument was that border control would help prevent criminality or terrorist attacks (9.5% of all arguments). In the context of the terror attack in Paris on 13 November 2015, a letter to the editor said, ‘Considering the threat from terrorism […] we should implement a strict border control’ (VG 19/11/15). Norwegian border control was implemented a few weeks after the terror attack in Paris, making Norway the first of the three countries to do so. This is probably why this argument is more common in Norway than in the other two countries. Rejections of border control (13.5%) in the Norwegian sample (the fewest in the three countries) mainly argued that it was inhumane. This was the fourth most common argument.
The most common legitimatisation of border control in the Swedish press was Argument 8: for the safety of the refugees (13.5% of all Swedish arguments). Several of the articles refer to international obligations and suggest that border control will be necessary if Sweden is to guarantee a functioning asylum system and provide shelter for asylum seekers. A Swedish minister stated, ‘We must face the situation at our border to be able to maintain an asylum system that corresponds with our legal obligations and provides shelter for those who are forced to flee’ (DN 24/10/15). As in Norway, both the sustainability discourse and the humanitarian discourse were used to justify sealing the borders, and border control was articulated as being in the interest of the asylum seekers themselves. The second most common legitimatisation of border control often occurs in the same context as the first: the fear that children would disappear (9% of all Swedish arguments), argued solely as a probable result of so-called system collapse. An employee in the Swedish migration directorate worried that the system was so stressed that arriving children were not being registered ‘[s]everal unaccompanied children disappear, at stations and around ferry landings, and it is not fair. That is the main argument for introducing border control’ (DN 14/11/15). Thus humanitarian arguments, resting on the premise of system collapse, were used to justify securing the border. Arguments 1 (to obtain better control; 9% of all Swedish arguments), 5 (reduce the number of asylum seekers; 8.5%) and 6 (avoid system collapse; 8.5%) all built on the premise that the Swedish system was ‘on the verge of collapsing’ (e.g., AB 10/11/15, DN 29/11/15). For example, a Swedish politician from the biggest right-wing party argued to introduce border control ‘for Sweden’s sake, we must introduce measures now to uphold our exiting [asylum] system’ (DN 10/11/15). Several additional reasons were also given, most often that Sweden had taken on an overwhelming responsibility in the refugee crisis, drawing heavily on the idea of Sweden as a morally good nation. For example, an editorial in DN states the following:
During this autumn, Sweden has mounted the greatest effort in history when it comes to refugees. Now it is over. [We must] create a breathing space in the Swedish refugee reception system. […] Sweden cannot be the only country that […] recognises that it has a humanitarian responsibility. (DN 25/11/15)
The portrayal of Sweden as a nation of humanitarianism merges with the sustainability discourse when border control is introduced: border control is introduced because Sweden is a solidarity nation on the verge of collapsing.
The second most common argument in the Swedish sample is a rejection of border control, based on Argument 10, that border control is inhumane or illegal. This constituted 13% of all arguments in the studied Swedish sample and illustrates that the actors in the Swedish data were more concerned with the humanitarian aspects of implementing border control than were those in the Danish and the Norwegian sample.
At the beginning of this article, I asked how border control was legitimised in Scandinavian news media, and in what contexts was the issue of border control brought up. The question is important because Scandinavian authorities risked undermining the Refugee Convention by implementing border control. To determine the contexts in which border control was discussed, I identified three overarching discourses under which the media wrote about the arriving refugees: threat, solidarity and (welfare state) sustainability. The threat and solidarity discourses resemble those found both in previous media studies of immigration journalism in general (e.g., Horsti 2008; Hovden & Mjelde 2019), studies of European border policy, and studies of border guards’ practices at the outer Schengen border (Aas & Grundthus 2015; Chouliaraki & Georgiou 2017). Lastly, I identified a sustainability discourse in the sample, whose premise is that there is a limit to how many immigrants the Scandinavian society can manage while remaining able to sustain the welfare state and the capacity of the asylum system. In general, the articles were largely dominated by one of these three discourses. However, in articles arguing for border control, the discourses merged. By framing border control as a way to prevent the collapse of the asylum system and thus keep asylum seekers safe (Sweden), to guarantee justice for ‘real refugees’ in the face of challenges to the asylum and welfare systems (Norway) or to protect impoverished citizens (often explicitly refugees or immigrants background) and their welfare benefits (Denmark), border control was articulated as a compassionate move towards both citizens and the refugees themselves. I term this merging as compassionate border securitisation and it describes the overarching legitimisation of border control.
The term is a linguistic construction that illustrates how politicians and other actors argue to implement border control to maintain both the welfare system and solidarity towards refugees (and other inhabitants) while simultaneously implementing security measures that risk breaking the Refugee Convention. This means that security practices are hidden under the language of solidarity and welfare state sustainability. The term serves to maintain the perception of a generous, solidary welfare state as well as Scandinavia’s self-image as guardians of humanitarianism. When solidarity and welfare sustainability become the very reason for securing the borders to limit the influx of migrants, they help to maintain positive humanitarian images while reproducing the existing order of power. This merging resembles the notion of the humanitarian border (Walters 2011); however, in the Scandinavian context, sustainability of the welfare state becomes the very catalyst for both the threat and the solidarity discourse.
Yet, this compassionate securitisation is not a definite description of how border control is legitimised in Scandinavian news media. Several articles contain arguments for or against border control that draws on only one discourse, for example, articles that argue that border control can prevent criminality or articles that reject such control as inhumane. However, in most of the articles in this study, arguments for implementing border control are based on the three discourses, threat, solidarity and sustainability, in which the first two discourses build and support the third. This contributes to articulating border control implementation as a compassionate move towards both refugees and the nation itself.
The author has no competing interests to declare.
Aas, KF and Grundhus, H. 2015. Policing humanitarian borderlands: Frontex, human rights and the precariousness of life. British Journal of Criminology, 55: 1–18. DOI: https://doi.org/10.1093/bjc/azu086
Andersen, TM. 2008. The Scandinavian model—prospects and challenges. International Tax and Public Finance, 15(1): 45–66. DOI: https://doi.org/10.1007/s10797-007-9022-3
Bendixen, MC. 2016. How many are coming, and from where? Available at http://refugees.dk/en/facts/numbers-and-statistics/how-many-are-coming-and-from-where/ [Last accessed 24 April 2021].
Benson, R. 2013. Shaping immigration news. New York: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9781139034326
Bleiker, R, Campbell, D, Hutchison, E and Nicholson, X. 2013. The visual dehumanisation of refugees. Australian Journal of Political Science, 48(3): 98–416. DOI: https://doi.org/10.1080/10361146.2013.840769
Brekke, JP and Staver, A. 2018. The renationalisation of migration policies in times of crisis. Journal of Ethnic and Migration Studies, 44(13): 2163–2181. DOI: https://doi.org/10.1080/1369183X.2018.1433026
Brochmann, G. 2018. Immigrant policies of the Scandinavian countries. In: Nedergaard, P and Wivel, A (eds.), The Routledge Handbook of Scandinavian Politics. London: Routledge. DOI: https://doi.org/10.4324/9781315695716-18
Brochmann, G and Hagelund, A. 2011. Migrants in the Scandinavian welfare state. Nordic Journal of Migration Research, 1(1): 13–24. DOI: https://doi.org/10.2478/v10202-011-0003-3
Chouliaraki, L and Georgiou, M. 2017. Hospitability: The communicative architecture of humanitarian securitization at Europe’s borders. Journal of Communication, 67(2): 159–180. DOI: https://doi.org/10.1111/jcom.12291
European Commission (EC). (n.d.). Schengen Area. Available at https://ec.europa.eu/home-affairs/what-we-do/policies/borders-and-visas/schengen_en [Last accessed 25 April 2021].
European Parliament. 2015. Migrants in the Mediterranean: Protecting human rights. Available at https://www.europarl.europa.eu/RegData/etudes/STUD/2015/535005/EXPO_STU(2015)535005_EN.pdf [Last accessed 15 July 2019].
European Parliament. 2019. Protection of EU external borders. Available at: https://www.europarl.europa.eu/RegData/etudes/BRIE/2018/630316/EPRS_BRI(2018)630316_EN.pdf [Last accessed 27 April 2021].
Eurostat. 4 March 2016. Asylum in the EU Member States. Record number of over 1.2 million first time asylum seekers registered in 2015: Syrians, Afghans and Iraqis: top citizenships. Available at https://ec.europa.eu/eurostat/documents/2995521/7203832/3-04032016-AP-EN.pdf/790eba01-381c-4163-bcd2-a54959b99ed6 [Last accessed 24 April 2021].
Figenschou, TU and Thorbjørnsrud, K. 2015. Faces of an invisible population. American Behavioral Scientist, 59(7): 783–801. DOI: https://doi.org/10.1177/0002764215573255
Freeman, G. 1986. Migration and the political economy of the welfare state. The Annals, 485: 51–63. DOI: https://doi.org/10.1177/0002716286485001005
Gallup. 2015. Norske velgere vil ta imot flere flyktninger. TV 2 News. https://www.tv2.no/a/7362497/ [Last accessed 25 April 2021].
Gammeltoft-Hansen, T. 2018. Tid til et paradigmeskifte i flygtningepolitiken? [Time for a paradigm shift in refugee policy?]. In: Omdal, SE (ed.), Skandinavisk Uro [Scandinavian Unrest] Oslo: Forlaget Press. pp. 21–43.
Gatrell, P. 2016. Refugees—What’s wrong with history? Journal of Refugee Studies, 30(2): 170–189. DOI: https://doi.org/10.1093/jrs/few013
Horsti, K. 2008. Overview of Nordic media research on immigration. Nordicom Review, 29(2): 275–293. DOI: https://doi.org/10.1515/nor-2017-0191
Horsti, K. 2012. Humanitarian discourse legitimating migration control. In: Wodak, R, Schroeder, R and Messer, M (eds.), Migrations: Interdisciplinary Perspectives. Vienna: Springer-Verlag. DOI: https://doi.org/10.1007/978-3-7091-0950-2_27
Horsti, K. 2017. Global mobility and the media: Presenting asylum seekers as a threat. Nordicom Review, 24(1): 41–54. DOI: https://doi.org/10.1515/nor-2017-0296
Hovden, JF and Mjelde, H. 2019. Increasingly controversial, cultural, and political: The immigration debate in Scandinavian newspapers 1970–2016. Javnost–The Public, 26(2): 225–240. DOI: https://doi.org/10.1080/13183222.2019.1587702
Hovden, JF, Mjelde, H and Gripsrud, J. 2018. The Syrian refugee crisis in Scandinavian newspapers. Communications, 43(3): 325–356. DOI: https://doi.org/10.1515/commun-2018-0013
Hussain, M. 2000. Islam, media and minorities in Denmark. Current Sociology, 48(4). DOI: https://doi.org/10.1177/0011392100048004008
Löfgren, O. 2017. Scandinavia: Small countries with big egos. In: Hannerz, U and Gingrich, A (eds.), Small Countries: Structures and Sensibilities. Philadelphia: University of Pennsylvania Press. pp. 83–102. DOI: https://doi.org/10.9783/9780812293791-004
McMahon, S and Sigona, N. 2020. Death and migration: Migrant journeys and the governance of migration during Europe’s “migration crisis.” International Migration Review, 55: 1–24. DOI: https://doi.org/10.1177/0197918320958615
Migrationsverket. 2016. Årsredovisning [Annual report]. https://www.migrationsverket.se/download/18.2d998ffc151ac3871593f89/1485556210405/Årsredovisning%202015.pdf.
Musarò, P. 2017. Mare Nostrum: The visual politics of a military-humanitarian operation in the Mediterranean Sea. Media, Culture & Society, 39(1): 11–28. DOI: https://doi.org/10.1177/0163443716672296
Nielsen, N. 4 January 2016. Refugee fears prompt Swedish border clampdown. Euobserver. Online access at https://euobserver.com/migration/131699 [last accessed 25 April 2021].
Nordicom. 2017. Newspaper daily reach. Available at https://www.nordicom.gu.se/en/statistics-facts/media-statistics [Last accessed 25 April 2021].
Østby, L. 2015. Refugees in Norway. Available at https://www.ssb.no/en/befolkning/artikler-og-publikasjoner/refugees-in-norway?tabell=253399 [Last accessed 22 April 2021].
Peters, C and Broersma, M. 2017. Rethinking Journalism Again. New York: Routledge. DOI: https://doi.org/10.4324/9781315716244
Reeskens, T and van der Meer, T. 2019. The inevitable deservingness gap: A study into the insurmountable immigrant penalty in perceived welfare deservingness. Journal of European Social Policy, 29(2): 166–181. DOI: https://doi.org/10.1177/0958928718768335
SVT Nyheter. 2 January 2016. FN-organ riktar kraftig kritik mot id-kontrollerna [UN agencies strongly criticize ID checks]. Available at https://www.svt.se/nyheter/lokalt/skane/fn-organ-riktar-kraftig-kritik-mot-id-kontrollerna [Last accessed 25 April 2021].
Thorbjørnsrud, K. 2015. Framing irregular immigration. American Behavioral Scientist, 59(7): 771–782. DOI: https://doi.org/10.1177/0002764215573255
Trägårdh, L. 2007. State and Civil Society in Northern Europe. New York: Berghahn Books. DOI: https://doi.org/10.2307/j.ctt1x76dkp
Triandafyllidou, A. 2018. A ‘refugee crisis’ unfolding. Journal of Immigrant & Refugee Studies, 16(1–2): 198–216. DOI: https://doi.org/10.1080/15562948.2017.1309089
UDI [Norwegian Directorate of Immigration]. 2015. Asylsøknader etter måned (2015) [Asylum applications by citizenship and month (2015)]. Available at https://www.udi.no/statistikk-og-analyse/statistikk/asylsoknader-etter-statsborgerskap-og-maned-2015/ [Last accessed 26 April 2021].
Van Gorp, B. 2005. Where is the Frame? European Journal of Communication, 20(4): 484–507. DOI: https://doi.org/10.1177/0267323105058253
Van Reekum, R. 2016. Migration corridor, border spectacle, ethical landscape. Mediterranean Politics, 21(2): 336–341. DOI: https://doi.org/10.1080/13629395.2016.1145828
Vaughan-Williams, N. 2015. Europe’s border crisis. New York: Oxford University Press. DOI: https://doi.org/10.1093/acprof:oso/9780198747024.001.0001
Walters, W. 2011. Foucault and frontiers: Notes on the birth of the humanitarian border. In: Bröckling, U, Krasmann, S and Lemke, T (eds.), Governmentality: Current Issues and Future Challenges. London: Routledge. pp. 138–164.
Watson, S. 2011. The human as reference object? Humanitarianism as securitization. Security Dialogue, 41(1): 3–20. DOI: https://doi.org/10.1177/0967010610393549