Managing stake and accountability in Prime Ministers' accounts of the "refugee crisis": A longitudinal analysis.

Taking a (critical) discursive psychological approach, the present study explores the identity management of the Finnish and Swedish Prime Ministers (PM) in relation to the "refugee crisis" and their countries' asylum policies. By taking a longitudinal approach and analysing the PMs' accounts of the "refugee crisis" from 1-year period, we focused on the ways rhetorical devices related to ethos, logos, and pathos were used to manage the issues of stake and accountability, as well as on the ways in which categories were worked up to serve particular functions. Our comparative analysis demonstrated significant similarities in the Finnish and Swedish PMs' talk, especially with regard to the transfer from a discourse of pathos and ethos, describing refugees in terms of individualism and humaneness, to a discourse of logos, emphasizing rationality, justifying sharpened immigration policies, and homogenizing refugees. However, the different historical paths of the two countries' immigration policies and the specific political situation had implications for the PMs' discourse. The Swedish PM could feasibly scapegoat the Sweden Democrats and the political right in opposition, whereas the Finnish PM, with the populist radical right as a government partner, engaged more heavily in distinctions between "real, needing" and "false, undeserving" refugees. We argue for the longitudinal approach in the analysis of political discourse, as such an approach allows to identify the changes and continuities in the discourse, as well as to grasp the dialogical interplay between the discourse and its context.

INTRODUCTION

During autumn 2015, over one million refugees and asylum seekers mainly from the conflict zones in Syria, Afghanistan, and Iraq crossed the borders into Europe. This caused a crisis within the European Union, with member states being unable to reach an agreement about distributing the responsibility for receiving the people in need.

Even though Southern Europe for geographical reasons received the vast majority of the newcomers, the northern‐most part of the continent did not remain a passive bystander to the crisis. Rather, in 2015, the Scandinavian country of Sweden was, alongside Germany, the country to accept the largest number of refugees and asylum seekers (160,000) in relation to the population size. Also Sweden's neighbour Finland received a number of asylum applications 10 times greater (30,000) than during previous years.

Both Finland, led by a conservative government‐coalition, and Sweden with its red–green government, initially maintained an open and receptive policy towards the refugees and asylum seekers. As we shall see, however, these policies turned out to be short‐lived, not least due to the rising political pressures from populist radical right parties that had gained huge electoral successes in both countries in the 21st century.1 However, an important difference from the perspective of the present study is that whilst the populist Finns Party (FP)2 holds a position in the Finnish national government, the Sweden Democrats (SD) have remained, or rather been kept, in opposition. As autumn 2015 proceeded, the previously open national borders were replaced in both Finland and Sweden by heavy border controls and sharpened conditions for asylum. As a result of these policy shifts, both countries received criticism in Amnesty International's (2017) annual report on human rights violations across the world, and free movement within the Nordic countries was seized for the first time since its introduction over 60 years ago.

Taking a (critical) discursive psychological approach (e.g., Edley, 2001; Wetherell, 1998), the present study will examine the accounts of the Finnish Prime Minister (PM) Juha Sipilä and the Swedish PM Stefan Löfvén, concentrating on their political speeches, blog writings, and statements in mainstream media. The study will ask how the two PMs rhetorically managed matters of stake and accountability (cf. Potter, 1996) in relation to their respective countries' asylum policies. In analysing how the PMs' rhetoric developed during the time from summer 2015 to summer 2016, our analysis pays attention to changes and continuities in the rhetoric, asking whether and how it aimed to respond to surrounding (populist) political pressures to take a sharper stance on immigration and asylum. Ultimately, our aim is to provide new insights for social psychological research on the discourse and identity management of political leaders. In this paper, we argue for a longitudinal approach in the analysis of political discourse, as such an approach allows for the identification of changes and continuities in the discourse, as well as for the exploration of the dialogical interplay between the discourse and its context.

THE SOCIAL PSYCHOLOGY OF POLITICAL RHETORIC

The study of political rhetoric has its roots in ancient Greece. Central to Aristotle's perspective on rhetoric were the notions of ethos, logos, and pathos. In addition to taking a stance that was morally trustworthy (ethos) and providing proof in support for an argument (logos), Aristotle (1991) argued that a successful rhetorician should also be able to arouse feelings (pathos).

In social psychology, the study of rhetoric arose as part of the “turn to language” that involved a rejection of cognitivism and a commitment to approaching talk and text as communicative action rather than as expressions of inner psychological states (e.g., Condor, Tileaga, & Billig, 2013; Edwards & Potter, 1992; Potter & Wetherell, 1987). Much of the social psychological work on political rhetoric builds on Michael Billig's rhetorical psychology, which itself draws on studies of classical rhetoric, and stresses the argumentative and persuasive nature of language (e.g., Billig, 1987, 1991). Billig makes the connection between social psychology and classical rhetoric explicit by claiming that “it can be asserted with probably little exaggeration that all the major themes of modern social psychology can be found in classical rhetoric, and in particular in Aristotle's Rhetoric” (Billig, 1987, p. 54). Billig (1987) takes a dialectic approach to argumentation and thinking, which entails treating categories as rhetorical phenomena: People have the capacity to employ categories but also to engage in the opposite process of particularization. In addition, the “two‐sidedness of argumentation” indicates that an argument (logos) only makes sense in relation to alternative arguments (anti‐logoi). Thus, political discourse constitutes oppositions—dilemmatic elements—which are both explicit and implicit. Although Billig's rhetorical psychology provides more of a guide to scholarly enquiry than a strict method for the social psychological analysis of political rhetoric, it makes an important analytic contribution by underlining the role of the argumentative context of discourse, that is, the importance of taking into account not only what is said but also what is being argued against.

Discursive psychology shares these same premises with rhetorical psychology, beginning with the assumptions that thinking is rhetorical and categories are constructed and worked‐up flexibly in discourse (Billig, 1987; 1996, Potter & Wetherell, 1987). The central proposition is that in the analysis of political discourse one needs to analyse how politicians construct versions of the world, perform actions with their talk (e.g., explain, justify, blame, attribute responsibility), and use rhetoric to achieve particular social and political functions (Edwards & Potter, 1992; Potter, 1996).

One of the issues that has long interested discursive psychologists concerns speakers' discursive management of stake and interest (Edwards & Potter, 1992; Potter, 1996). According to Potter (1996, p. 110), “anything that a person (or group) says or does may be discounted as a product of stake or interest.” Politicians need to pay careful attention to how they manage such issues in their talk, because appearing to be personally invested or interested is seldom advantageous, in particular, when the aim is to present seemingly “factual” arguments, and especially for politicians. By contrast, appearing to have “no stake” or be “largely disinterested” is often a desirable position when speaking in favour of or against any certain stance or position (Edwards & Potter, 1992), for example, for or against immigration. The dilemma of stake and interest can be managed in a variety of ways depending on the ways facts and accountability are constructed in discourse. According to Potter (1996, p.121), there are two general ways of representing reality that differ in terms of how agency is treated. On the one hand, there are discursive practices that involve assigning agency to the speaker, and on the other, there are rhetorical strategies that involve externalizing agency to matters “out‐there.”

The former kind of rhetorical practices is in line with Aristotle's idea of ethos and serves to increase the credibility of the speaker. These strategies often concern the identity of the speaker; in our present case, a politician (Condor et al., 2013). The resources of identity work for a credible speaker include such strategies as “concessions” (explicitly acknowledging potential counter‐claims), “disclaimers” (explicit disavowing of the stance or opinion a speaker subsequently advocates), “stake confessions” (acknowledging vested interest), “stake inoculations” (rebutting the potential claim to having a stake even before one is challenged on it), “building category entitlements” (claiming membership of a category that legitimates one's claim to specific knowledge), and “footing shifts” (e.g., presenting information as someone else's message;e.g., Potter, 1996).

The latter kind of management of dilemma of stake mentioned by Potter (1996) relates to political discourse that persuades through reason and logic, or logos, in Aristotle's terms. In constructing factual descriptions or “out‐there‐ness” (Potter, 1996), this kind of rhetorical work distances the speaker from his/her claim and transfers agency to external factors (Potter, 1996). These strategies include the notion of empiricist discourse, constructions of impersonality, quantification, the construction of corroboration and consensus, extreme‐case formulations, manipulation of detail, and pronoun use (Edwards & Potter, 1992; Pomerantz, 1986; Potter, 1996).

The third of Aristotle's appeals, pathos, referring to persuasion through emotions, is not discussed as such as part of the literature on stake management. Yet it can be regarded as a powerful device in the discursive practices of politicians, such as in extrematization, which presents a matter as serious and threatening; in normalization and abnormalization of one's own and others' actions as natural or as deviant; and in particular, in the construction and use of categories (Potter, 1996). Metaphors are particular kinds of descriptive categories that can be powerful vehicles of persuasion in political communication. For example, Charteris‐Black (2009) has explored the systematic employment of metaphors in the speeches of major British and American leaders and shows how the fear of the unknown Other is often communicated by metaphors and symbols and how metaphors can also appeal to logos as they provide “proof” that support arguments.

To summarize the discussion above, Aristotle's three appeals are in many ways compatible with the rhetorical strategies of interest within discursive psychology. Yet, as Aristotle's appeals relate to broader patterns of rhetoric, they may be particularly useful when trying to identify changes and continuities in discourse, as the present study aims to do.

DISCURSIVE CONSTRUCTION OF ASYLUM SEEKERS AND REFUGEES

Previous discursive research on talk about asylum seekers and refugees has explored the ways that these groups of people become “othered” and excluded. This research has suggested that opposition to immigration often becomes framed as common sense instead of as an issue of race (Billig, 1988; Capdevila & Callaghan, 2008; Lyons, Madden, Chamberlain, & Carr, 2011). This finding is in line with the discussions of “new racism” (Barker, 1981; Billig, 1988) and “discursive deracialisation” of talk (Augoustinos & Every, 2007; Every & Auguoustinos, 2007), pointing to the laborious rhetorical work people need to engage in, in order to express discriminatory views and yet avoid being labelled as racist. Deracialization of political discourse has also meant that notions of race have been replaced by other constructs such as those of “culture” and “the nation” (Every & Auguoustinos, 2007). Research shows how in political discourse arguments about “national identity,” “national borders,” and “national values” are not only used to warrant exclusion of asylum seekers and refugees but also to present such practices as “not racist,” but legitimate (e.g., Capdevila & Callaghan, 2008; Every & Auguoustinos, 2007; Lyons et al., 2011; Van Dijk, 1993).

Goodman and Burke (2011) summarize three kinds of arguments that are commonly used to oppose asylum: economic reasons, religious reasons, and the associated threat of terrorism, and asylum seekers' lack of ability to integrate into society (see also Augoustinos, Tuffin, & Rapley, 1999; Capdevila & Callaghan, 2008; Lynn & Lea, 2003). The dilemmatic discourse between real and false, deserving and undeserving immigrants has been documented in many studies (e.g., Andreouli & Dashtipour, 2014; Capdevila & Callaghan, 2008; Lynn & Lea, 2003). For example, in their analysis in the British context of letters to newspaper editors on asylum seekers, Lynn and Lea (2003) identified the distinction between “genuine” and “bogus” refugees as a key discursive resource for denying asylum‐seekers access into and rights in Britain. This discursive device of differentiating between good (genuine) and bad (bogus) refugees allowed the writer to criticize immigration and yet maintain his or her credibility as reasonable and sympathetic towards “real” asylum seekers.

So far, only a few studies have analysed political discourse related to the “refugee crisis” of 2015. The focus in these studies has mainly been on discursive practices of inclusion and exclusion in elite and public settings (Sambaraju & McVittie, 2017). One study (Kirkwood, 2017) focussed on sympathetic categorizations of refugees, namely, on the humanization of refugees in parliamentary debates in the UK. Kirkwood found that the use of mundane categorizations allowed for expressions of sympathy towards people seeking asylum or refuge and thus supported their inclusion. However, sympathy invoking positive accounts could be expressed alongside accounts that promoted a restrictive and paternalistic inclusion of refugees. Similarly, Nightingale, Quayle, and Muldoon (2017), who examined talk in an Irish radio call‐in programme, showed how callers avow grief and sympathy for those seeking refuge, whilst at the same time withdrawing from offering them inclusion. Similarly, Sambaraju, McVittie, and Nolan (2017) found such “sympathy‐talk” in their study of political debates on the refugee crisis in the Irish parliament during 2015. They showed that speakers promoted the inclusion of refugees through specific forms of talk about Ireland and the European Union, making relevant particular “unique” aspects of Ireland to warrant further inclusion and to diminish responsibility by framing the refugee crisis as a European issue. Thus, in line with previous research on refugees, specific national constructions could be used to enhance or to mitigate responsibility for limited inclusion (Nightingale et al., 2017; Sambaraju et al., 2017).

Perhaps most relevant for the purposes of the present study is the research by Goodman, Sirriyeh, and McMahon (2017) on the recategorization of the refugee crisis in British media outputs. This analysis demonstrates how versions of the “crisis” evolved from constructions that presented it first as a problematic “Mediterranean crisis” approaching Europe from outside, then as a “Calais migrant crisis” threatening the UK, after which it became “Europe's migrant crisis” that was portrayed as a multifaceted problem for the whole of Europe and the UK. The authors show how a sudden rhetorical shift from a “migrant” to a “refugee” crisis and back again to a “migrant” crisis took place first after the publications of photographs of the drowned boy Alan al‐Kurdi that spurred positive and sympathetic descriptions of people in need; and again following the terrorist attack in Paris in November 2015, which invoked threatening images of the crisis and its associations with terrorism.

In the present study, and in line with Goodman et al. (2017), we take a longitudinal perspective in analysing the changes and continuities in the Finnish and Swedish PMs' discourse during the 1‐year period beginning from the refugee crisis of 2015. Alongside our focus on the PMs' stake management, our present interest lies in the ways in which they construct and deploy categories and position themselves and others within those categories, and what particular functions these categories play in reference to the refugee crisis. In order to study how the issues of stake and interest are displayed and managed in the PMs' accounts of the refugee crisis, we focus on rhetorical devices related to ethos, logos, and pathos, as well as on the ways in which categories are worked up and undermined to serve particular social and political functions.

As heads of the national government, PMs represent not only their own political party and its voters, but rather, they act as representatives of their entire nation. In other words, they are faced with the need to appeal to heterogeneous audiences with mixed expectations (Condor et al., 2013). Thus, it is interesting to analyse how the Finnish and Swedish PMs discursively negotiate the tension between these heterogeneous audiences when faced with the widely controversial topic of the refugee crisis.

METHOD

Material

Our material consists of the Finnish and Swedish PMs' accounts of the refugee crisis in mainstream media and in their personal blogs. In this section, before presenting the material in more detail, we briefly describe the context of the study and provide some information about the two PMs.

The Finnish PM Juha Sipilä has led the government since the liberal and agrarian Centre Party emerged as the winner of the Finnish parliamentary elections in May 2015, closely followed by the populist FP and the conservative National Coalition Party. These three parties formed the new government based on a centre‐right coalition. This was the first time that the populist FP entered government, and its leader, Timo Soini, was nominated as Minister of Foreign Affairs. The new Finnish PM Sipilä was largely unknown outside his hometown Oulu until he joined the Centre Party before the 2011 elections. Prior to his political activities, Sipilä graduated as an engineer and has had a long business career during which he became a millionaire. Sipilä's Prime ministership has been characterized by talk of the Finnish economic crisis and needs for both considerable public savings and major reforms—issues he has approached from the perspective of his background in engineering and business administration: for example, from the beginning of his Prime ministership he became well‐known for his “process charts.”

In Sweden, the leader of the Social Democratic party, Stefan Löfvén, became PM after his party won the parliamentary elections in 2014. The Social Democratic victory was meagre, however, and Löfvén only managed to form a minority government in coalition with the Environmental party. Löfvén's professional background lies in a career in welding, and he has been active in the Swedish Metalworker's Union. Having been politically engaged from an early age, he became leader of the Social Democrats in 2012. As a consequence of the government not having a majority in parliament, Löfvén's period as PM has been characterized by political turbulence, beginning with the so‐called December Agreement that six political parties reached in order to exclude the populist radical right SD from policy‐making and that secured the continued trust from the political right for Löfvén's government. However, the agreement was dissolved on October 9, 2015, and as we shall see below; this forced the political parties to find new ways of cooperating and minimizing the impact of the SD.

We focus on the PMs' interviews and statements in mainstream media during a 1‐year period starting in August 2015, when the media began to frequently report on the refugee crisis and more refugees started to enter the Nordic countries, and ending in July 2016, when media attention receded and the number of refugees decreased drastically.

Although our principal material consists of media articles, we completed it with the PMs' blog entries from the same time period. Neither PM Sipilä nor PM Löfvén are active bloggers, rather, they use their blogs as platforms for publishing their speeches and writings (e.g., columns) that are published in other media. However, as PMs, their blogs reach a large audience; and traditional media follow their writings. We included blog entries as we thought that, compared with mainstream media statements that are concise in their content and form, blog entries allow the writer to develop and manage his/her personal stance to a greater extent, and, importantly, circumvent the role of journalists.

Unlike blog entries then, media articles are created or selected and edited by editors or journalists and may thus reflect not only the agenda of the PM but also the political position of, for example, a particular newspaper. Although Finland's mainstream media has traditionally had a close relationship with the decision‐makers, it has strongly detached itself from party politics (e.g., Tommila & Salokangas, 1998). The traditional role of Finland's mainstream media has been to stabilize society and national unity, rather than to challenge existing power structures and power holders (Lounasmeri, 2016). Thus, PMs in particular have considerable impact on shaping the media discourse on a given subject, as they are granted more media time and space. The same holds true for Sweden, where, however, the nationwide newspapers are regarded as somewhat more explicitly ideologically and politically attached: The two big morning papers adhere to independently liberal (Svenska Dagbladet) and liberal‐conservative (Dagens Nyheter) positions, whereas the two biggest tabloids lean towards independently liberal (Expressen) and Social Democratic (Aftonbladet) positions (Hellström, 2016). The differences between these mainstream newspapers are, however, subtle, which makes the situation in Sweden compatible with that of Finland.

As our primary concern was not with the way in which the PMs' discourse was framed by the newspapers, but with the PMs' discourse itself, we selected different media (e.g., both dailies and tabloids, both mainstream and local), focussed on those interviews or statements that were circulated in several media, and concentrated our analysis on direct quotes by the PMs. Although we cannot entirely avoid the possibility that the media have left out something significant from these statements, we hope, however, that our analytic procedures have served satisfactorily to retain the authenticity of the PMs' voices.

We selected the media articles and blog entries on the basis that they contained the PMs' accounts of the refugee crisis. The relevant articles were selected by entering the PMs' names and refug*, immigrant*, or asylum* into the websites' search engines. The largest Finnish newspapers Helsingin Sanomat, Iltasanomat, Iltalehti, and the website of the Finnish Broadcasting Company YLE, as well as more local newspapers such as Kaleva and Kainuun Sanomat were included in the analysis. The Finnish corpus of material consisted of 146 media articles and 17 blog entries, most media articles were published by Yle (78) and Helsingin Sanomat (44). The Swedish media material was collected from the largest Swedish newspapers and tabloids, Dagens Nyheter, Svenska Dagbladet, Aftonbladet and Expressen, the Swedish National Broadcasting Company SVT, as well as from the Social Democrats' official website. The final corpus of material consisted of 124 media articles and 14 blog entries in the form of political speeches.

Analytic procedures

The analytical approach of this study relies on work in (critical) discursive (e.g., Edley, 2001; Edwards & Potter, 1992; Potter, 1996; Potter & Wetherell, 1987; Wetherell, 1998) and rhetorical psychology (Billig, 1987, 1991; Billig et al., 1988). Critical discursive psychology (CDP) pays attention to micro‐level details, supplemented with a macro‐level layer of analysis, the focus being on the historical, social, and political context of discourse. CDP investigates the relationship between broader discourses and the speaking subject. Thus, the speaking subject is considered an active agent in discourse production as well as a product of discourse. The perspective allows for analysing the content, form, and function of the discourse (Pettersson & Sakki, 2017; Sakki & Pettersson, 2016): what is being said, how and for what purposes. In comparison with discursive psychology, a critical discursive perspective entails a focus not only on the immediate discursive context but also on the broader societal, cultural, and historical one (e.g., Edley, 2001; Wetherell, 1998). Of particular importance for the aims of the present study is to analyse the ways in which the PMs position themselves and others, or become positioned, in certain ways within the discourse of the refugee crisis and the changing political and societal situation.

Interpretative repertoires, ideological dilemmas, and subject positions are central components of CDP. In this study, we are particularly interested in how subject positions appear within various interpretative repertoires or discourses (we use this latter concept in this study). Put differently, in this study, we adopt a CDP approach in order to enhance our understanding of how the PMs make sense of the emerging refugee crisis and construct identities for themselves and others in relation to the changing social and political situation.

Our analysis began with reading and rereading the entire material several times in order to identify the patterns, or consistency and variability, within and between accounts (Potter & Wetherell, 1987). We read the material in temporal order with close reference to the changing social and political context in order to identify the continuities and shifts in the PMs' discourse. This allowed us to identify four major rhetorical shifts in both countries in the 1‐year period of the refugee crisis. These discourses partly overlapped, yet also differed from each other, as did the timing of the shifts that occurred within them, depending on how the refugee crisis and the discussion surrounding it evolved in the two countries.

The very same statements and accounts given by the PMs were often quoted in several simultaneously published media articles. Thus, once we had identified the main discourses, we were able to further reduce the large corpus of material and to engage in a more detailed reading in order to identify the rhetorical strategies deployed in the management of stake and interest, the construction of categories, and the subject positions and their functions.

The extracts selected below are ones we consider typical and illustrative of the PMs' rhetorical work on the refugee crisis. Being fluent in Finnish and Swedish, we translated the material originally written in Finnish and Swedish into English. In so doing, we paid special attention in order to retain, or explain, idiomatic expressions and subtle meanings where these occurred (as,e.g., in Extracts 7 and 12 below).

ANALYSIS

The Finnish Prime Minister's rhetoric on the refugee crisis

In this section, we will investigate how the Finnish PM rhetorically manages his stake and accountability in relation to the refugee crisis from early autumn 2015 until summer 2016.

“Refugees in need” (August to early September 2015)

The starting point of the refugee crisis in Finland can be traced back to the beginning of August 2015, when the Finnish media started to report about European events (e.g., drowning migrants, Hungary building a barrier), the unforeseen growth of the number of refugees arriving in Finland, and the measures this required (e.g., establishment of new reception centres). By the beginning of August, about 3,000 refugees had arrived in Finland, and this number almost doubled in August. With the unexpected arrival of refugees, authorities and politicians began to react to the new situation.

In August 2015, the Finnish PM Juha Sipilä took a strong public position on behalf of refugees. Several newspapers reported the PM's statements in which he described refugees as humans in need and danger (see also Kirkwood, 2017). In most of these press‐interviews, the PM refers to his recent visit to Lebanon as having opened his eyes to the severity of the situation. He thus employs a personal narrative in order to make sense of the new situation and to argue for the facticity of his stance (Edwards & Potter, 1992; Potter, 1996). Put differently, he employs ethos to build his credibility as a speaker. The function of this narrative is to position refugees as people in need, as in the extract below:

The Finns must realise that the newcomers truly escape war and

unstable conditions, that it is not a question of living‐standard

refugees.

The category of refugees as escapers of war and devastation is constructed in opposition to a potential counter category (Billig, 1987)—that of “economic refugees,” or more precisely, “surfers seeking a higher standard of living,” a term that was used by the secretary of the populist FP Riikka Slunga‐Poutsalo a few days earlier (Suomen Uutiset, 2015). This counter‐arguing plays a rhetorical function as it allows the PM to distance himself from such immigration‐hostile, populist rhetoric.

The discourse of refugees in need is also interesting in terms of responsibility and accountability. As we see in the above extract, PM Sipilä emphasizes obligation using the modal verb “must” (Line 1), which expresses a high level of commitment. This commitment is addressed to “Finns” (Line 1). This imperative rhetoric seems to be typical for Sipilä. It may be that his position as a politician and leader of the country requires him to sound convincing to the audience about the future, to know what action to take. Indeed, imperative talk in politicians' rhetoric orients to a requirement to be seen to be providing a plan of well‐defined future actions, rather than a set of hypothetical abstractions (Charteris‐Black, 2013).

The discourse of refugees in need culminates in PM Sipilä's announcement to give his own house in Kempele to the use of refugees. This news, reported both nationally and internationally, shows how Sipilä manages his stake and interest as PM and constructs for himself a “subject position of a humane and solidary helper.” The announcement appeared only 3 days after the heart‐breaking pictures of 3‐year‐old Alan al‐Kurdi were distributed in the media and can, thus, also be considered as a response to this, as illustrated in Extract 2 below (cf. Goodman et al., 2017; Kirkwood, 2017; see also the Swedish PM's rhetoric in Extract 13 below).

The idea of offering their own house to asylum seekers came from Sipilä's wife,

Minna‐Maaria.

“It was the wife's suggestion,” Sipilä says.

According to Sipilä the matter was thereafter discussed within

the family, and they concluded that the house in Kempele could

be opened for asylum seekers.

Placing asylum seekers in his own empty‐to‐be home is a very

personal matter for Sipilä. His own mother was a Karelian

evacuee from Suolahti. First she was placed in Tikkakoski, later

Kajaani. Sipilä says that the matter has been discussed a lot in

the family, especially when the grandparents were still alive.

“This was an issue that affected my own solution. We have

helped before”, Sipilä concludes.

According to Sipilä every Finn now has to consider how to help

refugees.

“This was our family's solution,” he states.

This extract shows how PM Sipilä produces an account that attends to interests without him being undermined as too interested (Edwards & Potter, 1992, p.158). One can note how Sipilä attributes responsibility to his significant “others,” to his wife (Line 3) and family (Lines 4–6, 16), which functions to create rhetorical distance from the decision to open the doors of his own house to refugees. These family references personalize the generous offer and constructs a position of a hero, a humane and solidary helper for Sipilä, without making him look calculated, and thus, allowing him to maintain his credibility as a politician.

Another rhetorical device used by Sipilä in the above extract is the one of narrative. Drawing on the events of his family's history allows him to create a personal relation to the matter at hand—asylum seekers (Potter, 1996). Here Sipilä engages in appeals of pathos: The narrative is used to make the emotional or mental states inferentially available, to draw on collective memories and emotions such as togetherness. The loss of Karelia and the resettling of the Karelian population can be considered one of the master narratives in Finland (Savolainen & Kuusisto‐Arponen, 2016). Thus, drawing on collective memory (Lines 8–9) and discussions with older generations (Line 11) functions rhetorically to appeal to shared experiences and memories. The story works as a past reference point, as an analogy, to the current refugee crisis. But it has other rhetorical functions as well. By saying that “we have helped before” (Lines 13–14), Sipilä can play with the category of “Finns” and portray himself as a representative of this positive national category—as one of the helpers. Tradition and family values are clues that help to build an association between the past positive events that the speaker has not personally experienced and the narrator in the present.

As we saw in Extract 1, the responsibility for the solution to the refugee crisis is attributed to the Finnish citizens (Line 1). Again in the above extract, Sipilä uses a modal verb together with the national category to emphasize the accountability of the ordinary Finns: “each Finn has to consider how to help refugees” (Lines 14–15).

“Refugees as numbers” (mid to end of September 2015)

The number of refugees arriving in Finland increased drastically in September 2015. The majority of people seeking asylum was now arriving through the city of Tornio in Lapland, which has an open border with Sweden. In mid‐September, the Finnish Interior Minister described “the flow of refugees and migrants into the country as out of control” (YLE September 14, 2015) and Finland's government announced plans for sharpening the immigration policy, including cutting the amount of money asylum seekers receive and tightening the criteria for reuniting families (YLE September 12, 2015). These rapid changes in the situation are also reflected in the PM's rhetoric.

The discourse of refugees in need that had dominated the rhetoric of PM Sipilä thus far shifts in a period of less than 2 weeks towards a discourse of refugees as numbers, where refugees and asylum seekers are increasingly described as a homogenous mass. The rhetorical shift from a “soft” emphasis on humanism and solidarity vis‐à‐vis asylum seekers, towards a “hard” rhetoric displaying numbers and statistics and portraying the refugee crisis as an actual threat to Finnish society characterize PM Sipilä's rhetoric after mid‐September 2015, as Extract 3 below exemplifies:

According to Prime Minister Juha Sipilä (Centre) the flood of

refugees is the number one problem for Finland. He estimates

that relatively speaking Finland is already as burdened by

refugees as Germany.

“I see the financial situation as a smaller problem than this (the

avalanche of refugees),” Sipilä stated at a press briefing on

Friday together with Interior Minister Petteri Orpo (Coalition).

“In relative terms the pressure is as big here (as in Germany),”

Sipilä said.

In the above extract, the analogy to Finland's economic situation (Line 5) provides a powerful rhetorical device as the entire Prime ministership of Juha Sipilä is characterized by his aspirations to save the Finnish economy. Another rhetorically powerful comparison is made in the reference to Germany (Lines 3–4). Germany's open‐door immigration policy was widely shared knowledge all over Europe by this time. Thus, the comparison works rhetorically to justify that “we have already helped more than we should have” and legitimizes the change in the rhetoric and hardened immigration policy.

References to the problems of refugee crisis (Lines 2, 5) characterize the reporting of the Finnish PM's discourse from now on. In the extract above (Line 2) and below (Line 4), one can note the use of extreme‐case formulations (Pomerantz, 1986). Extreme‐case formulations refer to the way in which justifications, arguments, accusations etc., are expressed through phrasing the relevant issues in extreme ways, for example, through the use of modalizing terms such as “completely,” “always,” or “never,” or in more subtle ways, for example, though quantification, as in Extract 4. In the present context, the extreme‐case formulation is used to strengthen the legitimacy of the PM's justifications for the severity of the threat caused by the arriving asylum seekers. Sipilä's rhetorical work now concentrates on the development of the accounts as factual, thus, reflecting appeals to logos, rationality of an argument. As one can see in the following extract, the dominant rhetorical device is that of quantification, which adds persuasive value to an argument by indicating objectivity and scientific certainty (Potter, 1996; Wetherell & Potter, 1992)

“In Finland the numbers are continuously growing, we are close

to 14,000,” Sipilä said in Brussels, where he participates in the

EU‐leaders' crisis meeting.

“We could reach really big numbers, 50,000 is possible this

year.”

According to Sipilä, considering the population‐sizes, the

number corresponds to if Germany would receive 800,000 asylum seekers.

Again, as in Extract 3, together with numbers the comparison to Germany ensures the threatening message of PM Sipilä (Line 6). Also, the above extracts demonstrate the way in which the rhetoric has shifted from personal and emotional in the discourse of refugees in need towards facts. Refugees are no longer referred to as persons, but as a number (Extract 4, Lines 1–2, 4), or as a natural force referred to as an uncontrollable “flood” (Extract 3, Line 1). As previous research suggests, such metaphors of water reinforce the belief that refugees are a threat to the nation‐state and its sovereign control because water is, as a liquid substance, difficult both to stop and to contain (Charteris‐Black, 2006; Pettersson & Sakki, 2017). Metaphors fulfil various functions: here evoking feelings, pathos, can be used to serve the rationality of the conclusion, logos (Charteris‐Black, 2009).

The blame for the refugee crisis is now attributed to the international community, or “the European Union,” and “other countries” that are portrayed as avoiding responsibility, as in Extract 5 below:

“The European Union should now direct money to the refugee

camps,” PM Juha Sipilä (Centre) remarks in YLE's interview.

According to PM Sipilä Finland will at the EU‐meetings next

week demand more resources for the EU's external borders and

for improving the conditions at existing refugee camps.

“We must get reasonable conditions at the camps, so that no

human smuggling would be necessary,” Sipilä said.

According to Sipilä there have been multiple discussions with

Sweden about the asylum seeker situation. He finds it wrong that

asylum seekers don't stay in the country where they first arrive.

“If one could trust that the system works and EU's external

borders hold, I would be ready to significantly raise the asylum‐

quotas,” Sipilä says.

According to Sipilä the possibility of rapid deportation of Iraqis

will be investigated next week.

In the above extract, PM Sipilä speaks on behalf of the entire nation (Lines 3–7, cf. Augoustinos, LeCouteur, & Soyland, 2002). The use of imperative language and obligating modal verbs (“demands,” “must”; Lines 4, 6) allows Sipilä to present himself as an active agent that furthers the best interests of the Finns and knows what to do. Sipilä blames Sweden indirectly by saying that it is wrong that asylum seekers do not register in their first country of arrival. During September 2015, the Finnish press commonly accused Sweden of allowing, and even supporting, asylum seekers passing the border from Northern Sweden to Finland without requiring them to register in Sweden. The conditional formulation “if one could trust that … I'd be ready to raise” (Lines 11–12) further attributes responsibility to external forces, to the EU and other countries (Edwards & Potter, 1992, p.135). The last part of the extract (Lines 14–15) makes explicit the hardening of the immigration policy, although generally, Sipilä seems to be avoiding partaking in this kind of discourse. Silence can be a way of managing matters of stake when engaged in performing potentially offensive or sensitive actions that may not be supported by the entire audience—here, the Finnish people (Wetherell & Potter, 1992). As Potter (1996) concludes, such silences are not analytically easy to identify; however, in our present context, the European refugee crisis was being widely discussed, from the media and political debates to lay talk, in Finland as in the EU at large. Thus, the case of the PM avoiding the topic is arguably rather remarkable, and not that difficult to observe. Indeed, and as will become apparent in Extracts 9, 10, and 11 below, Sipilä's gerrymandering involves avoiding engaging in the heated issues of Finland's immigration and asylum policy, whilst instead focusing on politically more “neutral” topics, such as the broader context of the EU.

“Refugees as dishonest” (October to December 2015)

By October 2015, the number of asylum seekers arriving in Finland gradually decreased. In December, the number of asylum seekers was barely 1,800, whilst still in October more than 7,000 people were registered as asylum seekers. Immigration laws and criteria on asylum seekers were further tightened and the Interior Ministry warns that two thirds of the asylum applications will be turned down (YLE, November 11, 2015).

During the period October–December 2015, new elements started to dominate the Finnish PM's discourse on the refugee crisis. The two previous discourses of refugees in need and refugees as numbers continued to exist and intertwine to serve new rhetorical functions. Unlike in these previously dominant discourses, in which the narratives and personal experiences, on the one hand, and facts, numbers and analogies on the other, were used as main rhetorical strategies, categorization and “we”‐talk now became central rhetorical resources alongside the Finnish law and international conventions in the justifications to legitimate the immigration policy, as in Extract 6 below:

“If a person comes to the border or into the country illegally, he

can directly be deported, but if it is an asylum seeker the matter

has to be treated according to international agreements and the

Finnish law,” Sipilä continues.

Referring to international conventions and the Finnish law (Lines 3–4) allows PM Sipilä to manage his own stake and avoid responsibility by portraying himself as non‐agentic in terms of the concrete actions regarding the immigration policy. Sipilä's discourse shifts towards a law and justice position and towards a more demanding and critical position vis‐à‐vis immigration that resonates with the demands of the populist FP (see,e.g., Pettersson & Sakki, 2017; Sakki & Pettersson, 2016).

The previously dominating subject position of “humane and solidary helper” is intertwined with another subject position, as in the following extract from Sipilä's speech at the Centre Party congress.

Finland deals with the refugee‐crisis according to our country's

laws and our international commitments, but we are not naïve.

We investigate every asylum‐application and –applicant

separately. A large share of the asylum seekers escapes war and

persecution, and need asylum. Those who really need asylum

will get it, others receive a negative decision. It is clear that in

such huge migration there will be people who are not really in

need.

Now we Finns have to find within us the arch‐Finnish quality of turning difficulties into possibilities.

The extract above is taken from a speech concerning the terrorist attacks that had taken place in Paris 1 week earlier. As in most of PM Sipilä's accounts of the refugee crisis during this period, he draws upon Finnish law and international conventions to present himself as non‐agentic in terms of the handling of the refugee crisis and, thus, allowing him to take distance from the tightening immigration policy.

However, he makes his stance clear by saying that “we are not naïve” (Line 2). Sipilä uses this very same expression in different occasions to indicate that the Others cannot be trusted. On Lines 4–5, Sipilä says that “a large share of the asylum seekers escapes war,” which seems to work as a concession before the main argument and against potential counter‐claims from his audience. In this way, the PM manages to present himself as informed about each side before reaching his conclusion that not all refugees are in need (Lines 6–8). On the contrary, and in the same way as Capdevila and Callaghan (2008) show in their analysis of the campaign speech by the British Conservative Party leader Michael Howard, although not explicitly named or addressed, an image of a dangerous and threatening refugee is introduced here. The image emerges in particular through the rhetorical context of the extract, that is, the terrorist attacks in Paris (see also Goodman et al., 2017).

Again, we also see how the plural personal pronoun “we” is used by PM Sipilä to construct the category of “arch‐Finnishness” (perisuomalaisuus), associated with the auto‐stereotype of Finnish perseverance and resourcefulness. One can note that Sipilä not only makes a distinction between “real and deserving” and “false and undeserving” refugees (Capdevila & Callaghan, 2008; Lynn & Lea, 2003) but also creates a subtle division between “us,” the ingroup that is characterized by persistence, and “them” that are associated with dishonesty. Along the same lines, in another newspaper interview, Sipilä provides alternative motivations for asylum‐seeking.

According to Sipilä the travels to Europe of young African men

could be avoided through a reform of the military service in

Africa.

An example is the 10‐year military service. If it were changed

into 2 or 1.5 years, it wouldn't be a reason to leave the country.

The above extract is part of an interview given at the EU‐Africa summit for migration in Malta. In line with its rhetorical context, the function of the above statement is to search for solutions to the refugee crisis. However, by so doing, PM Sipilä implies that African3 men leave their countries to avoid military service (Lines 4–5), not war and threatening life conditions, which is in stark contrast to how he described the situation in August and early September (see Extract 1). A shift from a humane towards a suspicious approach to the refugee crisis is evident.

“Refugees away” (Spring 2016)

From the beginning of 2016, the number of asylum seekers arriving in Finland decreased drastically. The slowdown was related to the decision by the EU and Turkey in March 2016 to turn back asylum seekers trying to enter Greece via Turkey. Most of the statements given by the PM during spring 2016 were related to these negotiations. As an example, consider Extract 9 below:

The agreement between the EU and Turkey has raised criticism,

but Sipilä is initially satisfied.

“There are problems in the agreement, as in all solutions to big

challenges. But I think it concentrates exactly on the important

issues.”

“There is order at the borders, help is sent closer to the crisis‐

areas, and people are given hope on the spot of returning to their

home countries. Refugees are also taken from the camps in an

organized manner.”

10

The above extract starts with a concession in which PM Sipilä acknowledges that the treaty is not without problems, which allows him to present himself as informed and understanding all sides of the issue (Lines 3–5). The problematic nature of the treaty is then diminished by generalizing the attribute of problem to all treaties and challenges. After that Sipilä deploys the pronoun “I,” constructing himself as an active agent to evaluate the value of the treaty. The personal pronoun “I” is then shifted back to impersonal form to warrant the facticity of the value of the treaty (Line 4). The empiricist repertoire is complemented with a three‐part list (Lines 6–8) to further emphasize the facticity of the argument. All this rhetorical work is needed to portray the treaty as a success and to convince the audience that the refugee crisis is now being solved and that PM Sipilä has played an important part in this.

In spring 2016, there were fewer asylum seekers arriving Finland than in Autumn 2015, and this was accompanied by a reduction in PM Sipilä's accounts of the refugee crisis. When he did discuss the issue, he seemed to return to a softer approach in which refugees were portrayed as humans in need, as the following extract demonstrates:

“They have travelled from the midst of constant insecurity and

years of conflict. For those people leaving their homes it has

been a choice between life and death,” Sipilä told MPs on

Tuesday afternoon.

“We have a duty to help these kinds of people,” he added.

The PM pointed out that managing the influx of asylum seekers

requires effective action in Finland and the EU.

“The asylum‐process must be expedited and repatriation‐

decisions should be implemented without delay,” Sipilä remarked.

As we see in Extract 10 above, which is part of Sipilä's speech given at the opening of the Finnish annual legislative session (valtiopäivät), very similar accounts of refugees in need are constructed by PM Sipilä as in the beginning of the crisis in August 2015. However, this rhetoric contrasts with the last sentence of the extract, formulated as an imperative demand to speed up the asylum process and the removal of asylum seekers from Finland (Lines 8–9). This, again, implies that there are two kinds of refugees in Sipilä's rhetoric, those who are in need and those who are dishonest (Lynn & Lea, 2003). As previous research has shown, this kind of distinction allows the speaker to present himself as empathetic, whilst simultaneously expressing suspicion and assigning negative attributes to refugees as a category (Lynn & Lea, 2003).

As previously, responsibility is addressed both to “us,” the Finns (Line 5), as well as to the EU (Line 7) and to countries that do not share the burden. The same line of argumentation is pursued in the following extract:

“In the corridors of Brussels there has already been talk about

withdrawing the EU's funding from those who don't share the

burden. Those who aren't solidary should in one way or another

do their part.”

This extract is a part of an interview that Sipilä gave to the media after having acted as PM for 1 year. Thus, the content of the article is not related to any specific event or topic, but rather constitutes the PM's general statements about the current situation and the crisis. In this particular example, he shifts his footing (Goffman, 1979) to the vague category of “corridors of Brussels,” on the one hand to appear disinterested, and on the other hand, to construct the account's objectivity by the construction of consensus. The passive voice further decreases his accountability and increases the rhetorical distance between himself and his utterance (Line 1).

In sum, two kinds of refugees, those who are real and in need and those who are false and dishonest, continue to exist in Sipilä'’s rhetoric in spring 2016, although compared to autumn 2015, his rhetoric returns to a softer tone. As there were fewer asylum seekers, the refugee crisis becomes less of a Finnish problem and more a matter that is dealt with far away. Accordingly, the PM's discourse in spring 2016 is devoted to finding a permanent solution to the crisis in the corridors of Brussels and at the borders of Southern Europe. Here, PM Sipilä presents himself as an active agent—as one of the problem solvers (e.g., his process charts to solve the crisis are presented in the media, YLE March 17, 2016 and in HS March 19, 2016).

The Swedish Prime Minister's rhetoric on the refugee crisis

We will now turn to examining how the rhetoric of the Finnish PM's colleague in the neighbouring country of Sweden evolved during the time period of late summer 2015 to spring 2016.

“Refugees in need” (August to mid‐October 2015)

Similarly to what we saw in the Finnish rhetoric above, during the beginning of autumn 2015 talk of humanism and solidarity dominated the discourse of the Swedish PM (see also Kirkwood, 2017; Nightingale et al., 2017; Sambaraju et al., 2017). Löfvén emphasized that receiving and helping refugees in need, people who are fleeing for their lives, is a moral responsibility of the entire Swedish nation. Similar to his Finnish colleague, the Swedish PM positioned himself as a humane and solidary helper of refugees in need, as in Extract 12 below:

“On August 27th 1957 it was decided that 570721G, G as in

gosse (Swedish for boy), would receive social care. Today G

would probably stand for gubbe (Swedish for old man),” Löfvén

said from the scene.

“I say this because it is so easy for people who flee here now

from oppression and terror to become a mere number in

bureaucracy; also they risk becoming a number, staples,

diagrams or curves now that the staples skyrocket.”

…

“This is why we gather here today, I thank you for wanting to

share this with me, 570721G.”

In the quote above, which is a report of the PM's speech at the launching of the programme “Sweden Together” that aimed to aid the integration of newcomers into society, Löfvén deploys a discourse of humanism and solidarity with those in need. Drawing analogically upon his own background as an orphan (Lines 1–2) functions to increase the ethos, or credibility, of Löfvén when he seeks to emphasize the individuality of refugees and asylum seekers. The personal narrative accomplishes an alignment between his own experience and that of the newcomers (Lines 10–11), which in turn functions to increase the pathos, or emotional appeal of his argument (Potter, 1996): Refugees and asylum seekers are like anyone of us, and we need to help them. The individualized discourse of asylum seekers and refugees continues in the following extract:

Not everyone survives. Alan, 3 years, was washed up on the

beach. The horror, the fright that that little boy experienced no‐

one should have to endure. I grieve for him and all other children

who die fleeing or at war. I grieve with his parents and relatives.

I grieve with humanity, when this happens before our eyes.

…

Agnes, you make me so proud. I am so proud to be Prime

Minister in the land you live in. Your letter shows that everyone

can do something, become a contact‐person for an

unaccompanied refugee‐child, donate money for clothes and

food

.…

But friends, receiving refugees can never be a task for

individuals, or individual organisations. It is a national task. As a

nation we shall take it on. Sweden's reception of refugees is the

responsibility of all of Sweden.

…

I am happy to see representatives and leaders from parties from

both so‐called blocks, and that the youth‐organisations have

come together. Today is no time for finger‐pointing. Today is a

time for mutual agreements.

The extract above is drawn from a speech that Löfvén gave at a manifestation in Stockholm in support of receiving asylum seekers. Similar to what we saw in Extract 12 above, this quote is dominated by references to individuals and the need to help them, whether they are refugees (Lines 3–5) or Swedes (Lines 7–10). Löfvén refers to Alan Kurdi, a Syrian boy who drowned in the Mediterranean on his way to safety in Europe (Lines 1–3). Images of his dead body circulated widely in traditional and social media and came to symbolize the tragic fate of many of those who sought to reach Europe in 2015. The reference to Alan serves efficiently to humanize refugees and asylum seekers (Kirkwood, 2017). Welcoming refugees to Sweden is depicted as a responsibility of all Swedes, yet, most importantly, as a moral responsibility of the entire Swedish nation (Lines 12–15). Thus, and similarly to what we saw in Extract 2 above in the rhetoric of the Finnish MP, humanity and solidarity with refugees and asylum seekers are depicted as inherent parts of “us,” the Swedes, and thus as defining features of Swedishness. The unity of Sweden and its people is emphasized further through Löfvén's stress upon the need for political cooperation and mutual respect across ideological and party lines (Lines 19–20).

Refugees as a dilemma (mid‐October to end of November 2015)

The PM's discourse of humanity and solidarity with refugees in need took a gradual, yet clearly detectable turn as autumn 2015 proceeded. This time was characterized by turbulence in Swedish politics, as the failure of the December Agreement became apparent. As a result of the agreement's collapse, Löfvén was pressured from the political right to implement restrictions on the Swedish asylum policy that—due to the large influx of refugees and asylum seekers—was considered to be on the brink of collapse. In this situation, Löfvén announced on October 23 that a number of restrictive measures, including temporary residence permits, would be introduced. In Extract 14 below, Löfvén seeks to manage his stake and interest when defending this policy change for the Swedish press:

Many of the other parties say that this agreement will lead to less

people finding their way to Sweden. Do you share this view?

No, if one looks at the world, which is what determines whether

people come to Sweden or not, it doesn't look like there will be

less refugees. The world is burning and it will be noticed in

Sweden. But it (the agreement) means that we can make sure

that more countries take their responsibility and then we can help

more.

…

“If we had had a normal situation we would have abided by what

we said. But Sweden is now in a very, very special situation. We

have a dramatic increase in people seeking asylum in Sweden. In

exceptional circumstances one cannot think normally, one has to

think outside the box,” he says.

Extract 14 above illustrates well the change that occurred in the PM's discourse around this time: It becomes increasingly dilemmatic. The policy change, according to the PM, is not meant to repel refugees from Sweden (Lines 4–6), rather, it aims to increase the responsibility of other countries. The reasoning, albeit seemingly illogical, goes: if more countries take more responsibility, so can Sweden (Lines 6–8).

In the extract above, talk of individual asylum seekers and refugees has disappeared. Instead, the PM refers to the “dramatic increase in people” seeking asylum and the current “exceptional circumstances” (Lines 12, 13). These extreme‐case formulations (Pomerantz, 1986) function rhetorically to add pathos, emotional weight to the justifications for the policy change. The quote above also instils an air of logos or reason: Löfvén protects himself from personal accusations of abandoning his and his party's formerly open approach to asylum‐seeking by presenting the policy change as a consequence of “out‐there,” external factors (Potter, 1996) and as a script and breach formulation (Edwards, 1994), that is, something abnormal that goes counter to how he would act if the conditions were normal (cf. Potter, 1996; Lines 10–14).

What we see above is an illustration of the ambivalent direction in which the PM's discourse turned in late October 2015: His management of stake and interest became characterized by the ideological dilemma (Billig et al., 1988) between solidarity with people in need, derived from his own ideological home of social democracy, on the one hand, and the responsibility for those who are already here, derived from practicalities and political pressures, on the other. Through a defensive discourse (Potter, 1996), accordingly, the PM shifts his own position from humane and solidary helper to that of a humane and solidary, yet responsible helper. Echoing the rhetoric of the Finnish PM (see Extracts 3 and 4), the highly individualistic positioning of refugees that characterized the previous period disappeared, and becomes increasingly replaced with the language of quantification (e.g., Rapley, 1998; Verkuyten, 2013), that is, with describing refugees in terms of homogenized volumes (Line 12).

Refugees become a problem (end of November to December 2015)

The change in Swedish asylum policy from open to restricted reached its culmination on November 24, when the government announced a number of new measures that introduced ID border controls, restricted conditions for family unification, removed certain former grounds for asylum, and broadened and retained the temporary residence permits agreed upon in October. In sum, the policy was framed as aiming at the EU's minimum level. Extract 15 below is taken from the government's official announcement of the new policy:

“It pains me to inform you that Sweden can no longer receive

asylum seekers at the current high level. We have to create

breathing‐space for Swedish refugee reception.” He continues:

“This requires that we drastically decrease the amount of people

seeking asylum and receiving residence‐permit in Sweden, and

instead get them to go to other counties,” says Stefan Löfvén,

who emphasized that the measures are necessary in order to

further pressure other EU‐countries to establish a permanent

distribution‐system.

In Extract 15 above, the ambivalence that characterized the former discourse seems to continue, as illustrated in the PM's formulation that the policy change pains him (Line 1), yet is unavoidable due to prevailing circumstances (Lines 2–3, 4–5). Indeed, the policy shift was regarded by the Swedish press as Löfvén's response to increasing pressures from the right and from polls that indicated the Swedish population's displeasure with the prevailing asylum policy (Aftonbladet, 2015). The SD, in turn, celebrated the policy that they defined as “SD‐politics” (Sveriges Radio, 2015), a claim that Löfvén strongly strove to distance himself from: We dismiss the Sweden Democrats' extreme‐right persecution of immigrants, where they do everything to spread prejudice against Muslims, and incite people against people, as they have done ever since they left the White Power Movement (Löfvén, Speech at the Municipality conference, Örebro, March 12, 2016). The new policy was, however, severely criticized, by the political left, by NGOs, and not least from within Löfvén's own party.

In the extract above, Sweden's responsibility for asylum seekers and refugees is externalized to the EU. In sharp contrast to his own statement only 1 month earlier (see Extract 13 above), Löfvén now explicitly declares that the policy aims to reduce the number of people coming to Sweden (Lines 4–5). The metaphor (“breathing space”) on Line 3 serves to add logos, or reasons for and support of the argument that the policy is indispensable to the survival of Sweden. The positioning of refugees and asylum seekers as de‐individualized masses continues (Lines 2, 4–5). The PM's own management of stake here takes a further step from a position of responsibility for people in need to responsibility for Sweden and “our own ones.”

“You – the EU and the political right – are the problem” (Spring 2016)

The new, restricted asylum policy remained in place, and in the months that followed the PM's discourse remained dominated by justifications for this state of affairs. At this stage, the dilemmatic and defensive character of his previous talk transformed into an increased offensive and aggressive tone (cf. Potter, 1996). In particular, Löfvén sought to avoid criticism of the sharpened asylum policy by shifting the blame for it upon the EU, and, importantly, onto his antagonists on the political right.

Naturally the reception of over 160,000 people in one year has

been an enormous effort for Sweden, and will lead to even

bigger tasks in the future. Naturally this task becomes no easier

by the years we have had of under‐investments, tax‐reductions

and cuts. But as naturally the solution cannot be to saw off the

strongest branch we sit on, and to dismantle the model that has

made Sweden strong and united.

…

And even if the bourgeois parties have done everything to delay

it, this government has now started to build the new society.

In the extract above from Löfvén's speech on the Olof Palme Memorial Day in February 2016, the PM discusses the large influx of asylum seekers and the concomitant policy shift in autumn 2015. As we can see, and in sharp contrast to the humanized discourse of asylum seekers and refugees in Extract 12 above, Löfvén depicts these in terms of numbers and figures (Line 1). The asylum‐policy‐shift is warranted by a rhetoric of logos and “reasonableness” (Billig, 1995, p. 82): “self‐evident,” extreme circumstances had brought about the change (Lines 1–2). The three‐part list (Lines 1, 2, 5) increases the rhetorical power of this argumentation. Further, Löfvén connects the pressures brought about by the influx of newcomers with the economic policy of the previous—right‐wing—government. What this rhetorical move accomplishes is that the hotly debated asylum‐policy‐shift, brought about by Löfvén's government, becomes depicted as a desperately needed, highly reasonable measure to strengthen and save Swedish society (Lines 9–10). Through this discourse, in sum, the PM manages his stake and interest by implicitly rebutting the criticism that had been directed towards himself and his party of abandoning their social democratic, humane policy towards people in need, and instead phrasing the policy‐shift in terms of a patriotic welfare‐project that seeks to correct the errors of his political antagonists. This discourse becomes even more aggressive in Extract 17 below:

Last autumn our reception was put under enormous pressure in

an extreme situation, when 80,000 people came during two

months. You in the municipalities have done a remarkable job.

You should be honoured for that. But even your capacity has its

limits. You ran out of beds. The social services went on their

knees, and they still are.

In such a situation every responsible government must act. We

have determinedly worked for a more just distribution within the

EU, but it has not sufficed. Too many member‐states have failed,

not only in action, but also in will.

…

It is time for the EU to go from chaos to control. There has to be

some damn order in a union of cooperation.

…

Much can be said about the Swedish right‐wing's different

messages. But in their resistance towards the Swedish model

they are at least consequent!

…

The Moderates (right‐wing party) don't seem to want to see the

reality. Instead they glance at the Christ Democratic youth

organisation, and let champagne‐drinking young‐conservatives

decide that sealed agreements shall be broken and the stability of

Sweden endangered.

In Extract 17, the warranting of the asylum‐policy‐shift through references to the extremity of the situation (Line 2, 4–6, 7) and the large number of arriving people (Line 2) continues. In order to manage stake and interest, Löfvén takes the position of speaking—not as himself—but on behalf of the government, whom he depicts as “responsible” (Line 7). Löfvén engages in what Capdevila and Callaghan (2008, p. 11) call a “discourse of discipline and regulation,” emphasizing the need for measures to “take control of the situation,” which allows him to himself and his government's policy vis‐á‐vis asylum as both warranted and adequate. The government's decision to sharpen the asylum policy is indeed framed as a consequence of the lack of action and willpower of the rest of the EU (Lines 9–10). Through metaphorical language and strong formulations (Lines 12–13), Löfvén attributes responsibility for the current situation—the restrictions on the asylum policy—to the EU.

Subsequently, the offensive rhetoric continues, this time to be directed once more at political opponents on the right. The emphasis on political cooperation and mutual respect in Extract 13 above is nowhere to be seen. On the contrary, the political right here becomes accused of abandoning the Swedish model—the welfare state that has become the primary symbol of the entire Swedish nation (cf.,e.g., Hellström, 2016; Sakki & Pettersson, 2016). In the final paragraph Löfvén—referring to the Christian Democrats and especially their youth organization that was seen to have caused the party's abandoning of, and the subsequent collapse of, the December agreement—engages in an anti‐elitist rhetoric (Lines 20–23). Indeed, the formulation “champagne‐drinking young conservatives” (Line 21) bears strong populist connotations and serves efficiently to position the political right as elitists, detached from the reality of “normal” Swedes (cf.,e.g., Mudde, 2004). In combination with the insinuations that they have abandoned the Swedish model (Lines 16–17), these political antagonists become charged with even stronger accusations: of being hostile towards the Swedish nation and its people. Similar rhetoric and similar accusations have been found to be characteristic of Swedish populist radical right political discourse (Sakki & Pettersson, 2016).

In summary, through moving from a defensive to an offensive rhetoric, that is, through putting responsibility for the restricted Swedish asylum policy on the EU and the political right, Löfvén manages to portray his and his government's decision as a consequence of the failures of these actors. Unlike the Finnish PM (e.g., Extracts 6 and 8), Löfvén does not succumb to making distinctions between “true” and “bogus” refugees (see also: Capdevila & Callaghan, 2008; Leudar, Hayes, Nekvapil, & Turner Baker, 2008). Rather, the restricted asylum policy is phrased, not as endangering the lives of people in need, but rather, as an act of patriotism and welfare–protection. This rhetoric is consolidated through depicting refugees and asylum seekers not as individuals in need but as plain numbers and extreme masses (Extract 16, Line 1; Extract 17, Line 2). The subject positions that are constructed through this discourse are, thus, that political antagonists and the EU become scapegoats; that refugees and asylum seekers become an undefined, homogenous mass; and, finally, that Löfvén manages his stake and interest by assuming the role of responsible father of the nation.

CONCLUDING DISCUSSION

In this section, we discuss the conclusions we draw from the analyses above and elaborate upon their implications for the discursive study of political rhetoric on refugees. Our analyses of the two PMs' discourse about the 2015 refugee crisis demonstrates not only how the crisis and the different agents involved in it—asylum seekers, Finnish/Swedish people, the EU, other countries, and the PMs themselves—were constructed but also how these constructions were oriented to perform particular actions. In both countries, in the beginning of the refugee crisis a discourse of refugees in need was constructed to portray the PMs in a favourable light and as examples of humane and solidary helpers and, thus, to attribute responsibility for the crisis to all Finnish and Swedish citizens—including the PMs themselves. Humanization (Kirkwood, 2017) and avowing sympathy (Nightingale et al., 2017; Sambaraju et al., 2017) have been shown to characterize parliamentary debates about the refugee crisis in the UK and in Ireland. As in this line of research, in the present study, sympathy for refugees was contingent upon the context in which it was located. In the Finnish PM's discourse, for example, sympathy was expressed mostly at the beginning of the crisis, before the refugees arrived in Finland, and on the other hand, after the crisis in spring 2016, when the number of newcomers decreased to normal levels. Thus, the more distant the crisis, the more positive accounts of “refugees” seemed to become (see also Goodman et al., 2017). The Finnish and Swedish PMs' discourse brings forth the same dilemmatic pattern that has been found in previous research: It is possible to treat others as humans, whilst nevertheless refraining from providing support for or inclusion of those regarded to be from “beyond the national borders.” As previous research (Leudar et al., 2008) has emphasized, discursive constructions about refugees and asylum seekers—especially those articulated in the mass media or by influential political actors—are not “merely” discourse. Rather, these constructions have implications for the ways in which refugees and asylum seekers construct their own identities and ultimately, thus, for their wellbeing as well as for their possibilities to be included in society.

Very few discursive studies have taken a longitudinal approach (see Goodman et al., 2017, for an exception). In the present study, such a perspective allowed us to analyse the changes and continuities in the discourse during the 1‐year period after the beginning of the refugee crisis in Finland and Sweden. As the analysis shows, some changes occurred abruptly, and some evolved more slowly. For example, in both countries, although somewhat earlier in Finland as compared to Sweden, following the drastic increase in the number of asylum seekers in September 2015 the rhetoric took a sudden shift in a harsher direction, focusing on the costs and problems of the crisis for the nation‐state. By contrast, in the Finnish PM's rhetoric, the division between “real” and “bogus” refugees developed more gradually. This division did not appear in the discourse of the Swedish PM, where, rather, a gradual shift towards talking of refugees in numbers and masses, rather than as human beings, occurred.

In both Nordic countries, the blame for the refugee crisis was attributed to the international community, to the European Union, and “other countries” that were portrayed as avoiding responsibility. Thus, in the same way as in Sambaraju et al.'s (2017) study of the Irish political debate on refugees, a supranational category—specifically the EU—served as a rhetorical resource that allowed speakers to argue both for the inclusion of refugees and for them to be kept outside the national borders. Indeed, supranational identities, alongside the constructions of nations (e.g., Augoustinos et al., 2002; Every & Auguoustinos, 2007; Reicher, Haslam, & Rath, 2008; Reicher & Hopkins, 2001a, 2001b), may be used in the rhetorical management of stake and accountability.

Both Nordic PMs aimed to present themselves in a positive light and as active agents—as helpers, protectors of law, problem solvers, and as fathers of their nations. But it seems that it was not always beneficial to come across as agentic—especially when faced with a controversial topic such as a harder asylum policy. In such cases, they tried either to silence the entire topic or attribute the responsibility to other agents, to the EU or to the countries who “do not share their responsibility.” One can also avoid responsibility for “bad politics” by hiding behind law and judicial systems, as we saw in the Finnish PM's rhetoric. Moreover, our analyses demonstrated that PMs may flexibly deploy the rhetorical strategies of “out‐there‐ness” (Potter, 1996), and thus, externalize agency when they engage in negative evaluations of the Other.

Our comparative analysis demonstrated significant similarities in the Finnish and Swedish PMs' talk, especially with regard to the transfer from a discourse of pathos and ethos, describing refugees in terms of individualism and humaneness, to a discourse of logos, emphasizing rationality, justifying sharpened immigration policies, and homogenizing refugees. However, as our longitudinal approach allowed us to illuminate, the shifts in the discourses occurred at different points in time in the two PMs' talk, reflecting the historical, social, and political contexts in which the talk was produced. In other words, the different historical paths of the two countries' immigration policies, with Sweden having a history of liberal immigration policy, thus requiring more defensive and offensive rhetoric when arguing for its sharpening, as well as the specific political situation; for example, what the position of the populist radical right was, had implications for the PMs' discourse. The Swedish PM could feasibly scapegoat the SD and the political right in opposition, whereas the Finnish PM, with the populist radical right as a government partner, engaged more heavily in distinctions between “real, needing” and “false, undeserving” refugees (e.g., Capdevila & Callaghan, 2008; Lynn & Lea, 2003), and in the double rhetorical strategy of positive ingroup‐representation and negative outgroup‐representation—the good Finns against the problematic refugees—that is a common feature of populist rhetoric (e.g., Reicher et al., 2008; Sakki & Pettersson, 2016; Van Dijk, 1993). Together, these findings highlight the importance of studying political discourse as part of its surrounding historical, social, and political context. As this is the central emphasis of CDP (cf.,e.g., Edley, 2001), the approach enabled a particularly rich exploration of the political discourse of the PMs' at both the micro and macro levels. As we hope to have demonstrated in our analyses, Aristotle's perspective on rhetoric and CDP are compatible approaches, with partly overlapping notions: CDP provides fine‐grained, sophisticated tools for the analysis of discourse, whereas Aristotle's notions concerning broader patterns of rhetoric were particularly useful for our analyses of changes and continuities in the discourse. These tools, together with our longitudinal approach, permitted us to analyse how the PMs' discourse evolved in relation to the changing social and political situation.

Intriguingly, the two Nordic PMs gave statements and interviews about the refugee crisis only when the situation so demanded—otherwise the topic was either left to the Interior Minister or to the populist voices. Regarding the topicality of the issue, it was striking to us how little rhetoric by the PMs about the refugee crisis was available. Our corpus may seem large, but it is important to note that usually the same interview or statement was circulated in several media. We conclude that the refugee crisis was a topic that the PMs did not want to engage in. One of the major challenges of any qualitative analysis, not only of discourse analysis, is precisely the question of how to study such silences and gaps. However, CDP is not devoid of tools, as Billig's (1987) idea of the two‐sidedness of rhetoric suggests considering also those arguments that are argued against or left implicit. If a topic is discursively mitigated or completely avoided, it may be a sign of its great significance for the speaker—such as the measures taken for hardening immigration or asylum policies. Provided the amount of such silences in our material, as well as the dilemmatic character of the two PMs' discourse, and given the significant political influence of any country's PM, we strongly encourage further social psychological research into political leaders' discourse on asylum and immigration.