The Discursive Appeal to Solidarity and Partisan Journalism in Europe’s Migration Crisis

The article analyses the discursive appeal to solidarity in the mass media during the unfolding of Europe’s migration crisis. Solidarity was claimed by numerous actors in the public discourse to legitimise political decisions and mobilise public opinion. While it seems that the call for solidarity was shared bymany actors, media studies show the ‘partisan journalism’ of media outlets. Thus, the political orientation of media outlets influences their coverage of public debates. Hence, to what extent do different quality newspapers cover the same solidarity claims in times of crisis? In order to answer this question, the crisis coverage of two German and two Irish newspapers with centre-left and centre-right political orientations is examined via the discourse network methodology. Germany is selected due to high political parallelism and a strong affectedness by the crisis, while Ireland is selected because of low political parallelism and a weak affectedness by the migration crisis. The findings demonstrate that partisan journalism persists during Europe’s migration crisis. Especially German party actors are present in both countries, underpinning the central position of Germany. Regarding the appeal to solidarity, political solidarity claims prevail in all four newspapers, indicating the political-institutional asymmetry in the Common European Asylum System. The study contributes to the strategic framing of concepts in public debates and demonstrates that the left-right distinction of media outlets is hardly affected by the migration crisis.


Introduction
Solidarity was a buzzword in Europe's migration crisis in 2015.Numerous actors claimed solidarity with refugees and called for solidarity among member states of the European Union (EU).Commentators called the time period from August to October 2015 the 'long summer of migration' and appreciated volunteering by citizens across Europe (Della Porta, 2018;Hamann & Karakayali, 2016;Kasparek & Speer, 2015;Wallaschek, 2018).Other studies showed that the public discourse during the crisis rather focused on security and economic frames (Greussing & Boomgaarden, 2017;Rheindorf & Wodak, 2018).Hence, media outlets report about the crisis in dif-ferent ways.One reason for this is 'partisan journalism'.Media news coverage and commentaries are influenced by political ideologies which lead to a specific framing of issues in the respective newspapers (Brüggemann, Engesser, Büchel, Humprecht, & Castro, 2014;Hallin & Mancini, 2004;van der Pas, van der Brug, & Vliegenthart, 2017).Moreover, journalists quote and interview other actors to support their own opinions (Hagen, 1993).The question is then whether ideological differences between media outlets vanish or increase during the crisis?To what extent can a discursive contestation of solidarity and security frames be observed?
The article investigates the solidarity discourse in two German and two Irish daily quality newspapers in 2014 and 2015.One centre-right (Welt, Irish Independent) and one centre-left (Süddeutsche Zeitung [SZ], Irish Times) newspaper in each country is selected to analyse the different political orientations as well as different models of political parallelism.While Germany was strongly affected by the migrant crisis in 2015, receiving hundreds of thousands of asylum seekers within a few months, Ireland hardly received any refugees which is mostly due to the geographical location of the island (European Commission, 2017).Regarding political parallelism, Germany is sorted into the Northern European democratic corporatist media model that assumes a close relationship between party organisations and press outlets.Ireland, however, is included in the North Atlantic liberal model which assumes the press and party organisations to be rather distant (Hallin & Mancini, 2004).Accordingly, both countries have been affected differently by the migration crisis in 2015 and follow different versions of political parallelism.
I analyse the framing in the media outlets by applying the discourse network methodology.This method combines discourse analysis and social network analysis by examining the interdependent process of actor presence and concept formation in public discourses (Leifeld, 2016).I analyse which actors are present in the newspapers and which frames are deployed.Since solidarity is understood as 'contested concept' (Gallie, 1956), I focus on different meanings of solidarity and which concepts are most present during the migration crisis discourse.First, I demonstrate that party actors, national executives and EU representatives dominate the solidarity discourse.The partisan journalism persists in both countries during the migration crisis.Second, I show that political solidarity is a shared claim in the four selected newspapers despite the different political orientations of the media outlets.The study demonstrates that the actor constellation corroborates previous work on partisan journalism and 'opportune witnesses' while the omnipresent discursive appeal to solidarity challenges the expectation that solidarity is predominantly invoked by leftist actors.
The article proceeds as follows: In the next section, I briefly outline the research on partisan journalism and Europe's migration crisis.Thereafter, I describe the data and the discourse network methodology.The next section presents the results of the discourse network analysis, followed by a discussion of the findings and implications of the study.

Partisan Journalism
Previous studies show that media outlets have a political orientation which impacts their news coverage.The seminal work by Hallin and Mancini (2004, p. 21) refers to political parallelism as "the degree and nature of the links between the media and political parties or, more broadly, the extent to which the media system reflects the major political divisions in society".Hence, journalists do not cover news in a neutral way, but mostly fol-low the general political orientation of the media outlet (Hagen, 1993).Hallin and Mancini (2004) distinguish between a Northern European democratic corporatist media model, a Mediterranean polarised pluralist model, and a North Atlantic liberal model.Regarding the dimension of political parallelism, the first two models share the high alignment between party organisations and the press whilst the North Atlantic liberal model assumes a rather distant relationship between party and press due to a loose tradition of party presses.Brüggemann et al. (2014) revised the typology by Hallin and Mancini and added a fourth media system.Nonetheless, Germany and Ireland are prototypical cases for the democratic corporatist model (Germany) and the liberal model (Ireland) which offers the opportunity to compare both cases with regard to different modes of political parallelism.
Most studies analyse the appearance of specific parties in the respective newspapers and the issue framing in order to observe a close or distant alignment of party actors and the press (Berkel, 2006;Brüggemann et al., 2014;van der Pas et al., 2017).However, as others argue, this is a rather narrow understanding of political parallelism.Political orientation is not only reflected in party actors' statements in the media and party actors are not the only actor group in the media.Broadening the scope of the analysis shows that other actors also appear in the newspapers that are ideologically close to the political orientation of the media outlet (Allern & Blach-Ørsten, 2011;Kaiser & Kleinen-von Königslöw, 2019).Hagen (1993) describes this strategy as calling "opportune witnesses" since journalists do not report balanced or objectively about an event, but are rather biased in their selection of interview partners and the representation of quotes.Additionally, specific nationaloriented frames might guide the news coverage more than political ideology.Especially in times of crisis, the degree to which a country is affected by the crisis might be more relevant for the media coverage than the political orientation of the outlet (Salgado & Nienstedt, 2016).
Accordingly, I expect that more party actors appear in the German newspapers' coverage than in that of the Irish newspapers.Additionally, I expect that centreleft newspapers (SZ, Irish Times) tend to feature actors from the left political spectrum to a greater extent while centre-right newspapers (Welt, Irish Independent) give more actors from the right political spectrum a voice in their coverage of Europe's migration crisis.

Security and Solidarity Frames in Europe's Migration Crisis
Studies on Europe's migration crisis have mainly exhibited two aspects: Firstly, scholars demonstrated the incompleteness of the European migration policy, and secondly, they highlighted the public debate on a security and solidarity framing of the migration crisis.
The EU's lack of supranational authority in dealing with Europe's migration crisis has been noted (Genschel & Jachtenfuchs, 2018).The EU has developed a Common European Asylum System (CEAS) to harmonise the different national migration policies across Europe, but the European Council is still the main institutional body that decides on the rules and further political development of this policy area.Moreover, the member states are still the main regulators by setting norms and rules (Zaun, 2016).Europe's migration and asylum policy is based on three principles of the Dublin Regulations: First, the member state in which an asylum seeker first enters the EU has to deal with its asylum claim.This regulation gives the border countries of the EU, especially Spain, Italy, and Greece, a one-sided responsibility.Second, the Dublin Regulations try to avoid an ambivalent legal status of asylum seekers by attributing the responsibility to deal with the asylum claim to one country.Third, asylum seekers cannot claim asylum in more than one member state simultaneously or try to get asylum in another member state after the first rejection.However, this leads to the fact that claiming asylum in non-border EU countries such as Germany or Ireland is made almost impossible (Chetail, 2016).Or as Bast (2013) sums up the institutional discrepancy: "In the European treaties it is expressly mentioned that there must be solidarity among the EU member-states in terms of burdensharing-but this organizational task has not yet been implemented.There is no asylum system based on solidarity."Hence, while solidarity is a guiding principle in the Lisbon Treaty, the current EU asylum policies are hardly perceived as solidary.
Scholars investigated how the migration crisis was depicted in the public debate.They show that security and economic frames prevailed in the media while humanitarian claims were rather marginal (Greussing & Boomgaarden, 2017).Others demonstrate how security measures such as border surveillance were depicted as humanitarian acts and indicate a discursive shift towards right-wing populism (Colombo, 2018;Musarò & Parmiggiani, 2017;Rheindorf & Wodak, 2018).Furthermore, solidarity movements and pro-refugee protests were quite visible and active during the crisis.They influenced public opinion, mobilised citizens to help refugees and therefore shaped the perception of the migration crisis substantially (Della Porta, 2018;Karakayali, 2017;Lahusen & Grasso, 2018).The image of Alan Kurdi, a 3-year-old boy who was found dead at the Turkish coast received massive media attention and created a public outcry about the dangerous route to Europe and the lack of solidarity with refugees.The 'We can do it' ('Wir schaffen das') statement by German Chancellor Angela Merkel regarding the reception of hundreds of thousands of refugees stuck in Hungary created a very positive (but volatile) public opinion on receiving refugees who are in need of protection (Vollmer & Karakayali, 2018).While the call to solidarity strongly resonated with the public, there has been little investigation into how actors frame solidarity (Closa & Maatsch, 2014;TransSOL, 2018).Therefore, I build upon research that looks at the discursive construction of solidarity (Wallaschek, 2019) and differentiates solidarity into various meanings.For the present study, four meanings of solidarity are most relevant: Political solidarity refers to new institutional mechanisms and instruments that foster cooperation; cultural solidarity promotes supportive action for shared norms and identity; monetary solidarity is based on risksharing and financial support for others; and lastly, social solidarity claims to redistribute resources and refers to voluntary actions by individuals and social groups.
These four meanings are expected to be present in the migration crisis, because of the asymmetrical CEAS and the need for new institutional instruments (political solidarity).The EU asylum policy strongly rests on human rights and solidarity as a guiding principle in the Lisbon Treaty (cultural solidarity).Financial help as humanitarian aid or to other member states dealing with incoming asylum seekers is important during the crisis (monetary solidarity).The voluntary actions by citizens and activists helping refugees as well as social policy measures might feature in the public discourse (social solidarity).
Accordingly, I expect that meanings of solidarity are more prominent in centre-left newspapers than in centreright newspapers.Conversely, I expect that non-solidarity claims are more prominent in centre-right newspapers than in centre-left newspapers.

Data and Methods
The study analyses the migration crisis coverage of four newspapers, two daily quality newspapers in each country (Germany and Ireland) in the years 2014 and 2015.While the analyses of Europe's migration crisis often begin in 2015 (Greussing & Boomgaarden, 2017;Vollmer & Karakayali, 2018), the first signs of the migration crisis emerged before 2015.The 'Mare Nostrum' mission of the Italian Navy started in October 2013 after a tragic shipwreck on the coast of Lampedusa in which more than 360 migrants died.Mare Nostrum lasted for a year and saved approximately 150,000 migrants.The Italian government urged the EU to establish a follow-up mission.While the Italian mission was a 'save and rescue' mission, the following Frontex-led mission 'Triton' was oriented towards 'sea border protection' (EPSC, 2017).Lasting conflicts and civil wars in the Middle East (Syria) and Northern and Eastern Africa (Libya, Somalia, Eritrea) prompted further migration movements towards Europe.In 2015, more than one million migrants reached EU territory while at the same time approximately 3,800 people died in the Mediterranean.Nonetheless, the EU did not agree on a substantial reform of the Dublin system or EU migration and asylum policy in general.The European Commission's proposal to relocate 160,000 registered refugees across the member states was approved by the European Council by majoritarian vote, but the current numbers of relocation show that hardly any member state completely fulfilled its obligations (European Commission, 2017).
The four selected newspapers are the German newspapers SZ and Welt, and the Irish newspapers Irish Times and Irish Independent.They have a large readership, a nationwide circulation and represent the different political ideologies of centre-left (SZ and Irish Times) and centre-right orientation (Welt and Irish Times).Choosing the Welt instead of the FAZ as the German centreright newspaper is based on previous work that demonstrates that SZ and Welt differ more on migration policy issues than SZ and FAZ (Eilders, 2002).Hence, partisan journalism on solidarity during the migration crisis might be more pronounced between SZ and Welt.
Quality newspapers still have an agenda-setting function and strongly influence the public debate with their comments, interviews and reports on political issues.Moreover, quality newspapers cover more political and policy-related issues than tabloids (Koopmans, 2007;Nossek, Adoni, & Nimrod, 2015;Reinemann, Stanyer, Scherr, & Legnante, 2012).The newspaper articles have been coded by applying the political claims analysis (Koopmans & Statham, 1999).Rather than coding the entire article, it focuses on the specific claims made by actors.Claims are defined as "the purposive and public articulation of political demands, calls to action, proposals, criticisms or physical attacks, which, actually or potentially, affect the interests or integrity of the claimants and/or other collective actors."(Koopmans, 2007, p. 189).This study is not only interested in the claimant (actor) who makes a statement on a specific issue, but also in the justification the claimant uses (de Wilde, 2013).By focusing on this aspect, the partisan journalism as well as the framing of solidarity are examined.
The selection of newspaper articles is based on a nominalistic approach and thereby the keyword search string includes the term 'solidar*' or close synonyms.I selected 967 articles in the four newspapers via the database Factiva and coded 633 claims.Using a nominalistic approach reduces the number of articles and sets a rather high threshold to select newspaper articles.Since the study is interested in how actors frame solidarity, using the term 'solidar*' in the search string increases the likelihood of retrieving the most relevant articles during the migration crisis.Table 1 summarises the number of claims in the four newspapers in each year.
This overview shows that German newspapers cover more claims than Irish newspapers, which reflects Germany's strong affectedness by the crisis and the public relevance and interest in the topic.Nonetheless, the number of claims increases in the Irish newspapers from 2014 to 2015 which shows that Europe's migration crisis gained media attention in peripheral countries not directly affected by the crisis.
The discourse network methodology is deployed to study the co-occurrence of framing and actor appearance in the selected media.It takes into account that actors refer to certain frames in their claims which might be shared by other actors in the public debate.These actor-frame relations are studied in discourse network analyses (Leifeld, 2016;Leifeld & Haunss, 2012).This perspective highlights agenda-setting processes and how actors frame public debates.Accordingly, the discourse network structure is a two-mode network, because two types of nodes (actor and concepts) appear in the network.Only those claims are used for the discourse network analysis in which actors justify their claim.This reduces the number of claims that are analysed, because claims without a justification have been coded as 'no justification' and are not considered in this study.
To account for the most present actors and concepts, the eigenvector centrality is calculated.It measures how central a node is in a network by counting the number of edges between two nodes and considering whether the node is linked to other central nodes in the network.The scale ranges from 0 to 1 and the closer the value is to 1, the more central the node is (Bonacich, 1987).The network statistics are calculated and the network figures are visualised with the R package igraph (Csardi & Nepusz, 2006).

Results
The findings are presented in the order of the articulated expectations in Sections 2 and 3. First, I turn to political parallelism and then focus on the framing in the four newspapers.The expectation was that more party actors appear in German than in Irish newspapers on the solidarity debate during Europe's migration crisis.Additionally, due to the political orientation of the newspaper, more left or right party actors are featured in the newspapers with a similar political ideology.In order to give an overview of the different discourse networks, Table 2 summarises the network statistics of the four newspapers.

Political Parallelism and Partisan Journalism
The differences in the representation of party actors between the newspapers is rather marginal.64 per cent of all actors in German newspapers are party actors while 60 per cent of actors have a party affiliation in the Irish newspapers.The centre-right newspaper Welt features, with a share of 76 per cent, the highest number of party actors.German newspapers predominantly cover claims from domestic party actors, although some EU actors (Juncker, Avramopoulos) are also present.The Irish newspapers feature Irish party actors, but also represent other European party actors like German politician Frank-Walter Steinmeier (SPD) or EU actors like Jean-Claude Juncker.Thus, the first expectation is not corroborated.Both Irish and German newspapers focus on party actors in the migration crisis discourse.
The representation of centre-left and centre-right actors follows the expected direction.The centre-left newspaper SZ features actors who are ideologically close to the SZ like the Catholic cardinal Marx or the SPD.The Welt covers more claims from centre-right actors like German Chancellor Merkel or Minister of Interior Thomas de Maizière (both CDU).Nonetheless, the most central actors in both German newspapers are conservative politicians (Merkel, de Maizière, CDU, CSU, Avramopoulos) while centre-left actors are less visible in the discourse.These conservative politicians have central policy-related competences and thereby get more public attention from the media outlets.Hence, the expectation in relation to 'opportune witnesses' is mainly corroborated for the German case, but it is also affected by the involvement of conservative politicians in government coalitions during the time period.
In the Irish case, the Irish Times focuses less on domestic party actors than the Irish Independent.However, the latter covers the solidarity debate in Europe's migration crisis to a lesser extent.If the Irish centre-right newspaper covers solidarity, party actors predominantly appear.Moreover, these party actors have a centre-right political orientation.More and different actors appear in the Irish Times, but there are hardly any left-leaning party actors.Nonetheless, more pro-refugee groups (e.g.Amnesty, Irish Refugee Council) are represented in the Irish Times.Hence, the expectations for the Irish newspapers can be almost completely corroborated.The Irish Independent represents more party actors than expected but features more claims by centre-right (party) actors.The Irish Times represents, as expected, not many party actors and covers more claims by centre-left actors.

Framing Pro-and Contra-Solidarity
The following section focuses on the framing in the solidarity debate.The expectation was that solidarity frames appear more likely and more pronounced in centre-left newspapers than in centre-right newspapers.Figures 1  and 2 show, however, that political solidarity dominates the discourse in the four newspapers and this is underlined by the eigenvector value of 1 in Table 2.This demonstrates the positive discursive appeal to solidarity.If actors refer to solidarity, then they seldom contest this concept, but support solidary actions in their claims.
Most actors in the four newspapers refer positively to solidarity and demand a reform of existing regulations.After a dozen migrants were found dead in a lorry on a highway in Austria in August 2015, the Irish Independent reports the political reactions and cites German Chancellor Angela Merkel: Ms Merkel told a news conference at the summit on the West Balkans in Vienna: "We are of course all shaken by the appalling news.This reminds us that we must tackle quickly the issue of immigration and in a European spirit-that means in a spirit of solidarityand to find solutions."(Irish Independent, 2015) Notes: SZ: Süddeutsche Zeitung, Welt: Die Welt, IT: Irish Times, IInd: Irish Independent.A threshold for the eigenvector centrality of the nodes is applied to focus on the most visible actors in the respective network.Only nodes with an eigenvector centrality of at least 0.1 appear in the discourse networks.Following De Nooy, Mrvar and Batagelj (2011), the network density for two-mode networks is computed as the mean degree centrality of the respective network.Mr Tusk, who has steered the EU towards a more hardline approach to the refugee crisis in recent weeks, reiterated the need for Europe to reinforce its external borders."Today, no task is more important for the moderate centre right than the re-establishment of Europe's external borders," he said."We can no longer allow solidarity to be equivalent to naivety, openness to be equivalent to helplessness, freedom to be equivalent to chaos."(Irish Times, 2015b) To sum up, the expectation that solidarity claims are more likely to be prevalent in centre-left newspapers than in centre-right newspapers is not corroborated in either country.Instead, the positive discursive appeal to (political) solidarity is observable in the German and Irish newspapers during Europe's migration crisis.

Conclusion
The article analysed the solidarity debate in German and Irish newspapers in the midst of Europe's migration crisis.It highlighted the actor constellation in the public discourses and how actors frame solidarity.It is the first study to focus on the influence of partisan journalism on the coverage of the solidarity debate in Europe's migration crisis.
The article provides three novel insights into the study of political parallelism, the research on solidarity and Europe's migration crisis.Firstly, the article highlights that the left-right orientation of newspapers persists.Although a crisis might be understood as a 'critical juncture' that changes institutional settings and leads to a redefining of positions, centre-right and centre-left newspapers feature 'opportune witnesses' in their news coverage.Moreover, both Irish and German newspapers predominately feature party actors, mainly domestic government actors, in their news coverage.Hence, the study corroborates the 'structural bias' of newspapers towards the representation of government actors as well as the fact that government actors are the winners of Europeanised public debates (Koopmans, 2007;Van Dalen, 2012).
Secondly, solidarity seems to be everybody's preferred concept.If political actors refer to this concept, it is in a rather positive valence.It is rather unusual to openly reject the call for solidarity and claim that it is not necessary or falsely claimed.This prompts the question as to why this is the case, especially if the recent crises are strongly linked to the conflict over solidarity in the EU (Hutter, Grande, & Kriesi, 2016).One reason might be the historical legacy of the concept.Solidarity is an important concept in the labour movement but also in Catholic social teaching.Moreover, it is one of the fundamental norms in European countries (Featherstone, 2012;Sangiovanni, 2013;Stjernø, 2009).Hence, publicly rejecting solidarity seems rather unlikely.Instead, the discursive battle is about the interpretation of solidarity, how solidarity is framed and with whom solidarity is (not) expressed.
Lastly, the study demonstrated that political solidarity is the common denominator in the public debate.Re-forming the existing CEAS and supporting border countries are fiercely debated among actors.However, the (non-)reform of the CEAS and the failed implementation of an EU-wide quota to relocate refugees across member states (Biermann, Guérin, Jagdhuber, Rittberger, & Weiss, 2019;Zaun, 2018) demonstrate that political solidarity failed to be institutionalised.It created public awareness about the pitfalls of the policy area, but the opposition of several member states led to the failure of political solidarity after 2015.Nonetheless, these political conflicts are crucial for the future development of the Home and Justice Affairs of the EU as well as in the societal handling of refugees and migrants in European countries.
The study also bears some limitations.Focusing only on a rather short time period might exaggerate the general level of public solidarity claims.Hence, future studies might take a more extensive historical perspective on the framing of solidarity which goes beyond a crisis period.Moreover, further analyses could focus on online and social media in order to study the appeal to solidarity in media outlets beyond the mainstream media.Quality newspapers have a selection bias towards certain frames and powerful actors.Lastly, the study only considered articles and claims that contained solidarity or close synonyms for the subsequent analysis.Using a more general approach to investigate norms and values in public debates can illuminate the framing structure and use of such concepts in public discourses.

Figure 1 .
Figure 1.Discourse networks in the German newspapers (SZ, Welt).Notes: The graph depicts concepts as circles and actors as squares.The size of the nodes indicates the eigenvector centrality.The bigger the node, the more central it is.The thickness of the edges shows how often an actor refers to a concept.The bigger the edge, the more often the actors uses the concept.

Figure 2 .
Figure 2. Discourse networks in the Irish newspapers (Irish Independent, Irish Times).Notes: The graph depicts concepts as circles and actors as squares.The size of the nodes indicates the eigenvector centrality.The bigger the node, the more central it is.The thickness of the edges shows how often an actor refers to a concept.The bigger the edge, the more often the actors uses the concept.

Table 1 .
Number of claims in German and Irish newspapers, 2014-2015.
Notes: SZ: Süddeutsche Zeitung, Welt: Die Welt, IT: Irish Times, IInd: Irish Independent.For the German migration crisis debate, I used the following keyword string in the database Factiva: '(Flüchtling* or Flucht* or Migrant* or Einwander* or Zuwander* or Asyl*) and Solidar* and (EU or Europ*)'.For the Irish migration crisis debate: '(Refugee or escape or Migrant* or Migration* or Immigrant or Immigration* or Asyl*) and (solidar* or mutual w/1 support* or cooperat*) and (EU or Europ*)'.An asterisk controls for multiple endings of a word.Duplicates of articles were excluded from the article population.

Table 2 .
Network statistics about the four discourse networks.
Political solidarity is in particular linked to solidarity among EU member states.These claims show that the national governments are aware that the current Dublin Regulations do not properly work if hundreds of thousands of people try to claim asylum in the EU.They attribute too much responsibility to EU border countries in dealing with migrants and asylum seekers.Moreover, Greece and Italy had been hit hard by the Euro crisis and did not have the capacities to host and deal with the large number of refugees.Therefore, many actors demanded more (political) solidarity with Italy and Greece.In 2014, Dimitris Avramopoulos, the designated European Commissioner for Migration, Home Affairs and Citizenship at the time, claimed: