EU crises and integrational panic: the role of the media

ABSTRACT Even a casual look at the history of the European Union (EU) since its inception in 1957 shows that at numerous junctures through its development the EU (or European Economic Community/European Community [EEC/EC] in its previous incarnations) has been portrayed as being in severe crisis. Of course, the EU continues to exist today, and it is arguably stronger and more integrated than ever. This article focuses on the role of international media coverage in framing certain events as crises and seeks to draw out a pattern across three prominent case studies: the 2003 Iraq crisis; 2005 constitutional crisis; and 2010–12 eurozone crisis. Detailed media content analysis shows that the international media was not just reporting on crises, it was framing challenges and setbacks to EU integration as existential threats. The authors introduce the concept of integrational panic to conceptualize the media's role in potentially contributing to the social construction of EU crises.


INTRODUCTION
At numerous junctures since its inception in 1957, the EU (or European Economic Community/European Community [EEC/EC]) has been portrayed as being in severe crisis. These crises have been covered in the media with a heavy, negative slant, complete with 'end of Europe' rhetoric about the future disintegration of the European project. While acknowledging that numerous factors go into the construction and build-up of crises in Europe, this article focuses on the role of the international media, and seeks to draw out a pattern in its coverage across three prominent cases: the 2003 Iraq crisis; 2005 constitutional crisis; and 2010 -12 eurozone crisis. To date, these crises have been studied only on a case-by-case basis.
Through comparative media content analysis, we argue that in each case the international media was not simply reporting on events, but framing challenges and setbacks to EU integration as existential crises. This is potentially significant because the media, more generally, has been shown to play a strong role in shaping public debate and in determining how people think about issues (Jamieson and Cappella 1996; Price et al. 1997). We find that across all three cases examined here, the media rapidly escalates its coverage of events, anticipates negative outcomes and taps into 'us' vs 'them' undercurrents in European society in its coverage. The implication is that the media may be contributing to societal perceptions of existential crisis in the EU, and the phenomenon we describe as integrational panic. Our goal in this article is not primarily to explain why the media behaves in the way that it does (see Groeling and Baum 2009), or to establish its direct role in changing public perceptions, although these are important next steps. Rather, here we seek to first understand how the media covers these crises as an indication of the role it plays in interpreting events and advancing certain understandings (and not others) about crises.
The next section reviews a number of perspectives on the origins of crises in international relations, and explains our approach, which recognizes the socially constructed aspect of many crises. The purpose of this section is to provide a broader understanding of how we understand crises to justify the subsequent empirical analysis of the media. The subsequent section examines the three crisis case studies in turn, focusing on how the media covers events as crises through its heightened coverage, framing and amplification of certain tensions within society. Finally, in the conclusion we come back to the potentially strong role of the media in contributing to the build-up of crises.

WHAT'S IN A CRISIS?
An international crisis is defined as 'an extraordinary moment when the existence and viability of the political order are called into question' (Ikenberry 2008: 3). Across the social sciences, there are at least three broad perspectives -systemic, behavioral and sociological -that grapple with the origins of crises (Geva et al. [2000: 447-71]; see also Brecher and Wilkenfeld [1997]; Gamble [2006]; Neal [2010]; Schmitt [1922Schmitt [ /2004). The core assumptions of this article are in line with the third, sociological perspective. We anticipate that certain societal actors -especially the media -are well positioned to shape public perceptions and have the power to contribute to the social construction of crises. In light of this, we argue that certain parallels can be drawn between the well-known sociological concept of moral panic (Cohen 1972) and what we call integrational panic.
One major approach to understanding the origins of crises is the systemic/ structural perspective, which sees crises as resulting from 'unfolding events' that enable 'destabilizing forces' in the international system to disturb routine patterns, undermine institutions and ultimately threaten the nature of the existing structure (Young 1968: 6-15). The assumption is that crisis events simply occur, and then decision-makers must find a way to reinstate stability, which may require fundamentally changing the system in place. This approach takes as given that crises are objective, structural phenomena, and that human perceptions do not matter (Gourevitch 1984;Verba 1971;Young 1968). As Boin et al. (2005: 5) argue, 'the cause of the crisis lies in the inability of a system to deal with the disturbance ' (2005:5). It assumes that crises originate from factors built into the system 'that come to represent an undeniable threat to the system'.
New institutionalist accounts can also be situated within this structural approach. Paul Pierson (2003), for example, argues that critical junctures are not only produced by exogenous factors, but could also stem from endogenous factors such as path dependence, threshold effects, cumulative causes, extended causal chains, and so on. The structural approach, however, encounters potentially serious difficulty in that there are numerous instances in which potential triggering events, and even systemic change, are neither seen as crises nor treated that way. In other words, the approach tends to discount the role of human perception and reaction to events in understanding the causes of crises, and thus may miss numerous cases in which the structure of the system would anticipate a crisis but it failed to materialize. Or, there was nothing in the structure to indicate a potential crisis, yet human interpretation defined an event as a crisis anyway.
A second major type of approach is the behavioral perspective. Rather than assuming that the ingredients for crises are somehow embedded in the structure of the system, it prioritizes the behavior of individuals as they relate to the system around them, particularly in how they are constrained by a set of utility-maximizing preferences. Charles Hermann argues that crises emerge when an event that threatens key goals of a state, provides only a short time to respond and comes as a surprise (Hermann 1972:187). Behavioralists see the study of crises as objective in that these are measurable events with certain thresholds that can be determined through analysis of the observable 'facts' involved. Billings et al. argue that the threshold depends on 'the perceived value of possible loss, probability of loss, and time pressure' (Billings et al. 1980: 303-4). They define 'loss' as the difference in one's state of being before and after the triggering event. So, from this perspective, there is a measurable point at which an event becomes a crisis based on the calculation of the costs involved. This emphasis on individual cost calculation as well as scientific measures of crisis triggers and thresholds still denies the importance of social processes that are so fundamental to the origins of crises. To understand cost or loss calculation, we argue that it is necessary to investigate what people value, which in the case of the EU may also include the idea of a supranational entity.
A third major perspective on crises draws upon sociology, and fully delves into the socially constructed and 'subjectively perceived' nature of crises (Hay 1999: 317 -44). Colin Hay, for example, argues that shared narratives about crises are what bring them into existence, not calculations of loss or cost. Similarly, Carroll L. Estes argues that social construction 'does not deny or ignore the existence of objective phenomena . . . Social action, however, is inseparable from the socially constructed ideas that define and interpret these phenomena' (Estes 1983: 446). This sociological approach, which takes human interpretation into account, has valuable explanatory power because it can tell us why various events have built up into crises, while others have not. Thus, Ikenberry's 'extraordinary moment' (2008: 3) of crisis occurs when society perceives that an event represents a threat to the current political or social order. We argue that this approach is most convincing, as it helps account for events that might have escalated into crisis but did not, as well as events that grew into major crises in part because of societal perception.
Within this sociological perspective, a central question becomes: which constituent of society is most central to the social construction of crises? Classic Marxist theories tend to emphasize the power of élites in engineering crises out of minor issues to serve their own selfish purposes, such as to divert public attention away from real issues of concern. At the opposite end of the spectrum, scholars who subscribe to a more grassroots model emphasize the importance of the general public in constructing crises (Brunvard cited in Goode and Ben-Yehanel, (2009: 55-7;Morin 1971;Victor 1993). The argument is that instead of élite engineering the terms of the crisis narrative, they are simply feeding off of the sentiments that already exist in the regular public. In between these two extremes, and relevant to our approach, is the argument that 'interest groups' in the middle of the social hierarchy matter most in constructing crises. Interest groups, in this sense, are defined as groups of opinion-makers, such as the media, epistemic communities, non-governmental organizations (NGOs), businesses, lobbying groups, professional organizations, and so on.
We suggest that the sociological concept of moral panic is a useful point of departure for understanding the social construction of EU existential crises. 1 Moral panic 'seeks to explain a particular type of overreaction to a perceived social problem' (Rohloff and Wright 2010: 404), and shows how and why influential societal actors, especially the mass media, construct concern over a social issue that goes well beyond the reality of the problem (Cohen 1972;Krinsky 2013). Moral panic builds up to such an extent that it can change the rules and laws within society, and result in a perceived threat to the existing social order (Cohen 1972;Goode and Ben-Yehuda 2009). Even though the concept of moral panic was developed to explain how some actors within society come to be defined as threats to dominant social values, we argue that an adaptation of the framework is valuable in understanding how social actors, especially the media, can contribute to the creation of societal panic that leads to perceptions of crisis more generally. Integrational panic is a specific type of societal panic relevant to understanding the dynamics of EU crises. Diverging from the theory of moral panic, integrational panic arises from societal concern about a threat to the common European project of integration, a key characteristic of European social order today.
The central question that arises repeatedly during times of integrational panic is: will EU integration continue, or will it collapse? Thus, while just like in situations of moral panic various societal tensions such as East-West, North-South, élite -public, old -new become exaggerated in media coverage, in these cases there is less reason for outright moral indignation. In both types of situations, an 'us' vs 'them' dynamic is pushed to the surface with a mediadriven crisis build-up, but in the case of integrational panic it is the specific issue of EU integration that is at stake, rather than societal reaction to perceived threats to existing notions of morality. Our hypothesis is that if media coverage of EU events defines a crisis as such, frames it as potentially derailing EU integration and emphasizes a kind of 'us' vs 'them' conflict, it is likely to be a contributing factor to integrational panic and, subsequently, the social construction of crisis.
This overview of the debate surrounding the origins of crises situates our approach, and justifies our more specific analysis of the media as a potentially significant actor in the social construction of crises. As with the theory of moral panic, many societal actors are potentially involved in the social construction of crises, such as leaders, the public, interest groups, and so on, and there is often a synergistic effect among these various actors in the build-up of crises. The media is only one piece of this broader analytical framework, but given its centrality in shaping public perceptions, it can often play a strong role (Hunt 1997).

THREE CRISIS CASES
The three cases under examination are the 2003 Iraq crisis, 2005 constitutional crisis, and 2010-12 eurozone crisis as the most recent and prominent events in which the EU was perceived to be on the verge of breakdown. The aim is to establish a pattern in the role of the international media across time, especially since no study to date has systematically compared these EU crises to each other. Our evidence is based on content analysis of the news and op-ed articles for each case in four major international print media outlets: Time Magazine; The Economist; International Herald Tribune (IHT); and Financial Times (FT). 2 These are élite-oriented, highly respected among English-language news media, represent a variety of viewpoints and have a very high circulation globally. 3 While it would also be valuable to include local and other language media coverage within member states, this is beyond the scope of a single article, and so we choose here to focus on the role of the international English-language media as a good indication of how the media framed the crises or the public debates about them. It is reasonable to assume that élite decision-makers and opinion leaders within Europe also read these major news sources, and this fed into domestic narratives about crises.
The analysis for each crisis focuses on the volume of coverage, neutrality and framing. The neutrality of an article is counted as negative when the anticipated consequences of the crisis for the EU's continued viability are depicted as negative, positive when the EU is portrayed as being able to overcome current challenges, and neutral if the story expresses no judgment about outlook. In terms of media framing of the dominant tension, we focus specifically on: (1) scopewhether European or domestic; (2) players involved -whether among member states, between Brussels and member states, or within a particular member state; (3) nature -the substantive quality of the tension; and (4) sub-tensions within the dominant tension -the various reasons contributing to the major tension. We chose these specific categories because they encompass a range of most likely possibilities for where news stories might consistently draw out tensions or even assign blame for the difficulties the EU faces during times of crisis build-up, including whether this is a European versus national problem, the various actors involved who might be responsible, the main tension expressed and the causes of that tension. The analysis is restricted to the period of crisis build-up, with an aim to better illustrate the role of the international media in potentially contributing to the dynamics of a crisis at the outset.

The EU's crisis over Iraq 2003
The Iraq invasion was certainly a crisis in many respects -for the Iraqi people, American society and transatlantic relations -but well before the invasion this was first a crisis for the EU. After President George W. Bush addressed the United Nations on 12 September 2002 (U.S. Department of State 2012), calling for the international community to support an invasion of Iraq, it was clear that EU member states did not agree on how to respond. The international media quickly construed this as a major crisis for the EU. Figure 1 shows the volume of coverage. Thirty-six per cent of Time, 46 per cent of Economist, 40 per cent of FT, and 55 per cent of IHT stories were negative in their predictions, with only 1-3 per cent positive, and the rest neutral (see online appendix supplemental material).
As for the specific content of this coverage, the dominant crisis narrative emphasized in this media coverage was mainly drawn from tensions among member states (41 per cent of Time stories, 61 per cent of Economist, 39 per cent of FT and 58 per cent of IHT stories), in particular the so-called East-West divide over Iraq (36 per cent), and competition among the 'Big Three' (the United Kingdon [UK], France and Germany) over leadership of the EU (27 per cent). With respect to the former, the media's narrative about East vs West amplifies a particular framing of the Iraq issue that was not necessarily self-evident because Western member states were just as significantly divided among themselves as the new member states were supposedly divided from them. The leaders in several major, Western European member states, such as the UK, Spain and Italy, supported United States (US) action, along with many Central and Eastern European countries (CEECs). Meanwhile, numerous other countries -such as Austria, Belgium, Finland, France, Germany, Greece, Ireland, Luxembourg and Sweden -vehemently opposed the invasion. Nonetheless, as an IHT story put it, 'An indirect French-German warning enters here as a statement that a European core exists and will not easily accept a new environment and a new membership ready to disregard its convictions' (Vinocur 2003). The Economist (2003b) put it even more strongly: 'With 25 members, the EU will be a different place, and once they have their feet under the table, the new boys might turn out to be less grateful and compliant than has long been assumed.' With respect to the narrative about the Big Three, this was intimately tied to the coverage of East-West division, and somewhat secondary to it. The EU was on the eve of welcoming the 10 CEECs into its fold, and the question of whether France, the UK or Germany would lead this newly enlarged EU was paramount. Typical of this coverage, an FT story stated, 'When it comes to foreign policy, Europeans rarely agree. Iraq has shown again that Britain, France, and Germany, Europe's three big powers, "nationalise" foreign policy when hard choices confront them' (Dempsey 2002). Even though France and Germany were on the same 'side' over Iraq, their differences in approach to security policy were also amplified in the media, emphasizing that German leaders did not like the French determination to build up the EU to balance against the US. As a story in the IHT put it: 'Are Mr Chirac and Mr Schroder making a mistake in placing so much emphasis on a bipolar relationship in Europe when enlargement increases the possibilities for multi-polar ties within the EU?' (Vinocur 2002).
We argue that it was not a foregone conclusion that this foreign policy disagreement over Iraq be described as an existential crisis for Europe. The EU did not even have a common foreign policy as part of its institutional structure until the 1992 Maastricht Treaty, and members had not been particularly successful at speaking with one voice in the past. The coverage of this crisis might have played out in far milder terms. But the high level of negative media coverage clearly defined this a major crisis for Europe. For example, as the IHT describes: There has rarely been a more dramatic illustration of the Union's failure to speak with a common voice, despite the fact that it officially has a common defense and security policy . . . This war is undermining the course of Europe toward integration. (James 2003) The FT portrayed the Iraq crisis as 'one of the darkest moments in the EU's history . . . The Union's divisions have torn apart its common foreign policy and hobbled its influence in reshaping the international order' (FT 2003a).
Other actors' tone-setting statements also provided further impetus, and were picked up repeatedly in the media. In January 2003, US Defense Secretary Donald Rumsfeld created more momentum behind the East-West narrative when he referred to 'old Europe', and argued that 'the center of gravity is shifting to the east' in Europe (BBC News 2003). Newspapers all over the world printed the famous Old Europe, New Europe quote, as well as French President Jacques Chirac's statement in February 2003 openly accusing the EU candidate countries of being, 'not very well behaved and rather reckless,' and that they had 'missed a great opportunity to shut up' (CNN World 2003). In response, Latvian President Vaira Vike-Freiberga said, 'Nowhere in the Copenhagen criteria does it say we cannot speak our minds' (Graff 2003). The international media used these remarks as further fuel in the narrative about member states at loggerheads.
As evidence of this increasingly tense climate, the candidate member states that were about to join the EU were shut out of the summit in February 2003, and the French government openly threatened to veto their membership. There was actually serious talk of not going through with enlargement of EU membership to include the Central and Eastern European countries, and it was assumed that any aim to establish a substantive common foreign policy for Europe would have to be cast aside. There was a real sense that the EU had reached a breaking point, both in terms of its membership and in terms of integration.
The crisis reached its high point on 15 February 2003, the day when major anti-war protests broke out across Europe. In the immediate wake of this, EU leaders held a summit in Brussels on 17 February 2003 to try to craft a common reaction to the US stance on Iraq. They left the meeting with a joint statement that 'Iraq has a final opportunity to resolve the crisis peacefully' and that 'inspections cannot continue indefinitely' (The Economist 2003a). They agreed in principle that use of force could come into play as a last resort. In sharp contrast to this common stance, The Economist's verdict on the summit was that Europeans were 'United in theory, divided in practice' (2003a). Time Magazine further added to the negative bias, describing it as 'Europe's Family Feud' (Graff 2003).
By this time, there was no question that the international media was framing this as an existential crisis for EU integration. The Commission President warned that the 'total lack of a European common foreign policy' was bringing the EU closer to disaster. At a February 2003 European parliament meeting in Strasbourg, he said: If Europe fails to pull together, all our nation states will disappear from the world scene . . . Unless Europe speaks with a single voice, it will be impossible to continue working closely with the US on a longstanding basis while retaining our dignity. (Financial Times, 2003b) There was a clear argument repeated in the international press that the Iraq crisis could mean the failure of Europe's historic enlargement to include the CEECs (Fray 2003). But such a failure in the case of Iraq was equated in the international media with a more general inability for the EU to move forward according to its own goals and values. The overall trajectory and build-up of crisis in this case shows that the media had adopted a kind of 'conflict frame' that downplayed other ways of perceiving these events as they unfolded. This type of coverage -with its amplification of negative interpretations, dire predictions and focus on tensions among member states -closed off alternative and milder understandings of the crisis, likely contributing to a climate of integrational panic.

Constitutional crisis 2005
On 29 October 2004, all 25 heads of state had signed the new Treaty Establishing a Constitution for Europe, which was designed to streamline and simplify the legal basis of the EU, make some institutional adjustments in light of enlarged membership and codify new areas of co-operation, particularly in the foreign policy area. The next stage was for each member state to follow its own ratification procedures, which had to result in unanimous approval. A few months before, on 20 April 2004, British Prime Minister Tony Blair had unexpectedly promised a referendum on the Constitutional Treaty, which would make it much more difficult to approve, and was a procedure he had previously rejected. Then, another seven member states also announced that they would hold referenda on the Constitution. Thus, once the heads of government had signed the Constitutional Treaty the ground was set for the subsequent ratification period.
As evidenced by its frequent and negative predictions of the upcoming ratification process, media coverage characterized this process as a crisis for Europe. At this stage, the possibility of a failed Treaty was viewed as a setback in the minds of many, but not necessarily an event that threatened the existence of the EU. Nearly every major EU/EC treaty that had come before had resulted in initial negative referenda, including the 1992 Maastricht Treaty and the 2001 Nice Treaty. And autumn 2004 opinion polls showed strong support for the Constitutional Treaty (for example, 73 per cent of the Dutch and 70 per cent of the French were in favor) (Qvortrup 2006: 91).
Again, we argue that the volume of media coverage (see Figure 2) and negative spin (78 per cent of Time stories, 81 per cent of the Economist, 57 per cent of FT, and 100 per cent of IHT, with 0-6 per cent positive and the rest neutral) clearly indicate that these newspapers framed the episode as a crisis for the EU's very survival. Indeed, the four major media outlets in this study each had high predictions of failure before the French referendum had even taken place -89 per cent of Time stories, 79 per cent of Economist, 100 per cent of IHT and 68 per cent of FT. 4 Thus, the media likely contributed to a period of integrational panic, focusing almost exclusively on what would happen if the treaty failed.
Representative of the media rhetoric during this period is a 25 September 2004 Economist article stating, 'Gentlemen, you are trying to negotiate something you will never be able to negotiate. But if negotiated, it will not be ratified. And if ratified, it will not work.' Similarly, Wolfgang Munchau, a regular FT columnist, wrote that, 'The bandwagon against the European constitutional treaty is rolling . . . it is worth asking whether there is such a thing as a Plan B. The answer is no, at least not officially' (Munchau 2004). This not only shows oft-repeated negative predictions about the prospects of the Treaty, but also a strong belief that the EU would not be able to find a way forward after its failure. And this was many months before the actual negative referenda occurred.
During this period of crisis build-up, the dominant crisis narrative emphasized in media coverage was public -élite tensions (89 per cent of Time articles, 60 per cent of Economist, 80 per cent of FT and 100 per cent of IHT). For example, a Time Magazine article stated: There's no trust in the future, no capacity for risk taking . . . The French just aren't happy about the way the world turns right now, and the referendum gives them a chance to say no to many things: to Turkey in the E.U., to Chirac, to enlargement, to offshoring of jobs, to globalization (Graff 2005) An FT article describing French unemployment, debt, economic slowdown and decline in social protection as 'a profound crisis of society which is transforming itself into an immense crisis of confidence ' (18 April 2005) was also representative of the tone in the international media, which had little to do with the terms of the new treaty. A Time Magazine article captured it well in March 2005: 'There's a growing desire to punish leaders, and blame Paris and Brussels alike for everything going wrong . . . Most people are basing how they'll vote on anything but the constitution itself' (Crumley 2005). Like in the Iraq case, as the crisis grew the media emphasized societal tensions and drew strong connections between the fate of the new Treaty and the future viability of the EU as a whole.
This media focus on a growing problem between publics and élites was comprised of several strains of tensions, depending on whether the focus was on EU élite or national élites. At the national level, tensions often invoked the fear of losing national identity and sovereignty (18 per cent), citizens' unhappiness with the current government (25 per cent) and unemployment and the role of migrant workers (6 per cent). At the European level, tensions reflected a feeling that the EU was remote (18 per cent), that the eurozone and single market would take precedence over citizens' rights (5 per cent), and the fear of what future enlargements would bring (19 per cent). Significantly, only 2 per cent of the news coverage actually dealt with the substantive terms of the Constitutional Treaty itself.
The crisis reached its height with the failed referenda votes in France and the Netherlands. The French voted 'no' on 29 May 2005, with 55 per cent opposed and 45 per cent in favor (voter turnout was 70 per cent). Three days later, 61 per cent of Dutch voters said 'no', compared to 39 per cent who voted 'yes' (voter turnout was 62 per cent). In the words of one former EU official with decades of experience in Brussels, the negative referenda resulted in immediate 'seismic shocks' to the system, and represented 'without a shadow of a doubt, a turning point in European integration' (Sutton 2005: 2). The Economist intensified its 'end of Europe' rhetoric. 'The Europe that died' was on the cover of the 4-10 June 2005 issue, and the article itself stated, 'The EU should move in the direction of being a looser, less federalist and more decentralized club . . . the club must pass more powers back to its members ' (2005a). Later that month, on 19 June 2005, The Economist published a cover image of a garbage can with the title, 'Where to file Europe's new constitution'. The cover of the 13 June 2005 issue of Time Magazine described the constitutional crisis, as 'The Great Crackup', and the article stated, 'In Brussels, there is still a reluctance to admit the obvious: that the constitution is dead' (Time Magazine 2005).
Thus, during the buildup of the constitutional crisis, media coverage emphasized the severity of the crisis. The narrative was that the EU had reached the end of the road for integration and enlargement, even possibly leading to departures of some member states. Some insisted that the EU must become looser, member states would take back power from Brussels, and there would not be a more integrated core, as some federalists had suggested. There was a sense that the 'no' votes were a sign that the EU should focus more on economic liberalization and growth, not social or political Europe (The Economist 2005a). If the international media played a role in shaping public opinion, this type of coverage would close off alternative, especially milder, interpretations of the treaty ratification process, and likely encourage integrational panic. The limited nature of the media's coverage provides insight into how the public might understand the crisis, as well as the ideological and political implications of this.

Eurozone crisis 2010-2012
Like in the Iraq case, the crisis trigger for this third case was initially external in origin. The sub-prime mortgage crisis that brought down Wall Street in 2008 served as a kind of pre-crisis trigger, but did not yet signify the beginning of Europe's third existential crisis of the 21st century. Prior to the impact of the Wall Street crash, the eurozone was actually doing well, having maintained a steady growth rate since the introduction of the euro. Indeed, it took around two years for the Wall Street crash to really be interpreted as a full-blown existential crisis for the EU and its currency. Eurozone growth overall was negative in 2009, as the entire global economy contracted, but talk of the possible break up or end of the eurozone did not really reach a serious level until 2011.
Thus, a more immediate event was the realization of the sheer scale of Greece's debt, which had been hidden for several years. As far as the media is concerned, May 2011 marks the point at which most coverage assumed that eurozone problems would not just be confined to Greece. It became apparent that Portugal required a bailout, and that Greece would need a second bailout to avoid default. Still, even at this point, these two countries only represented 3-4 per cent of the whole eurozone economy (Greece alone represented around 2 per cent of the eurozone economy and Portugal an even smaller portion). Nonetheless, news coverage shifted from debate over what to do to help Greece to fears that the European project itself was in jeopardy. The crisis reached its peak in November 2011 with Greek Prime Minister George Papandreou's announcement that he would put austerity to a popular referendum.
A significant increase in media attention signaled the buildup of the crisis from May to October 2011 (see Figure 3). The socially constructed dynamic of crises is perhaps especially hard felt in the case of an economic or financial Figure 3 Eurozone crisis, volume of news coverage, 1 January 2010 to 31 October 2012 crisis because markets are fundamentally shaped and affected by perceptions and investors' confidence in the economy. Thus, while the structure of the eurozone did have some elements of weakness -all economies do -the media put forward a more limited and pessimistic interpretation of the debt problems, repeatedly defining these events as constituting a eurozone-wide crisis (77 per cent of Time stories, 78 per cent of Economist, 57 per cent of FT and 66 per cent of IHT, with 0 -7 per cent positive and the rest neutral). This arguably contributed to sharp reactions in international markets, with investors betting against other countries in the eurozone, leading to a contagion effect (News Wires 2011).
The covers of The Economist were particularly striking: Greek solvency problem into a E1,000bn existential crisis for the European Union' (Miliband 2011). Similarly, former French President Valery Giscard d'Estaing said in a speech: Greece is a great culture but a small country in our financial sphere and there is no basis for it rocking the euro at this point, had it not been constantly prodded by speculators and by banks from outside the euro area. (Rettman 2011) And former German Chancellor Helmut Schmidt added to this, stating, 'This talk of a crisis of the euro is merely hot air emanated by journalists and politicians' (Rettman 2011). Many media reports and op-eds argued that either Greece would have to leave the EU or it would bring the rest of Europe down with it. News stories like, 'It's Time to Admit the Euro has Failed' and '10 Reasons the Euro was a Dumb Idea' became commonplace (Sivy 2011;Gersemann 2011). Others predicted that the only way Greece could repair its own economy and save itself was to leave the EU. Despite the media's strong crisis narrative, at no point in this period did the European Commission or any EU head of state suggest that Greece (or any other country) might have to leave the EU.
As with all crises, there are multiple tensions at stake, but the one that stood out in the international media was that of hostility between northern and southern member states (41 per cent of Time stories, 36 per cent of Economist, 24 per cent of FT and 38 per cent of IHT -there were a wide range of other tensions mentioned less frequently, see online appendix supplemental material). Europeans (except for those in the South) had long talked about these tensions through the language of stereotypes. Northern Europeans were assumed to be 'hard-working, law-abiding people who live within their means', while Southern Europeans were 'work-shy, rule-bending, and profligate' (Mahony 2012). The media sharpened these stereotypes in its coverage through its focus on this as the dominant tension underlying the growing problems in the eurozone. Other tensions about the crisis that arguably would have been just as plausible (if not more so) in this context were: blame for US neglect in upholding financial regulation; criticism of Wall Street; criticism of the global banking sector for greed and corruption; questioning of capitalism in general; or complaining of growing inequality between the rich and the poor. However, these were either far less prominent or non-existent in media coverage, again demonstrating their role in potentially limiting public understandings about the crisis and contributing to a sense of integrational panic.
Our analysis shows that these North-South tensions had four main aspects to them: (1) criticism of the South for being fiscally irresponsible; (2) criticism of Germany for being too iron-fisted; (3) tension between net detractors and net contributors to the EU budget; and (4) tension between euro and non-euro member states. But when the crisis reached its height in November 2011, the dominant narrative expressed in the media was the problem with Germany's behavior. Forty-five per cent of the coverage in Time, 54 per cent in The Economist, 48 per cent in the IHT and 39 per cent in the FT focused on this. There was a shift from blaming the South for being fiscally irresponsible to blaming the North for not offering its support and, in particular, blaming Germany for being too iron-handed.
This emphasis on Germany as the main problem may be somewhat surprising, given that the German economy itself was not suffering directly in light of the eurozone crisis, nor were Germans destabilizing the euro directly. But again, it points to the socially constructed nature of crises and the sorts of interpretations that arise in light of particular media frames: the story was that the real source of the crisis was not so much the financial hardships experienced by countries in the South, but Germany's unwillingness to act quickly enough, the austerity measures it required, and the control from Brussels that was a condition for any bailout.

CONCLUSION
Across the three cases, there is a pattern in the way that the media's coverage effectively closes down certain meanings or interpretations of events in a direction that emphasizes threats to the EU's continued viability. First, in each of the three case studies, the media quickly frames these events as serious crises, reflected in the volume of coverage as the crisis builds from its initial starting point. Second, the media puts a negative spin on this coverage, whether or not there is a basis for this. Third, there are specific dominant tensions reflected in the narratives about the crises in international news coverage, revealing the integrational panic dynamic. In other words, there is a pattern in media coverage of repeated use of a 'conflict frame' in describing European crises. Ultimately, we find that the media has consistently portrayed what might have developed as stumbling blocks in the ongoing process of EU integration as existential crises that seemingly threaten to derail the European project entirely.
An alternate possibility is that the media is simply reactive -dutifully reporting the news -and not proactively contributing to the ratcheting up of crises. After all, in each of the crisis cases discussed here, at some point in the crisis build-up phase social dialogue becomes entangled with the tensions expressed in the media. It is inherently difficult to separate the causal effect of the media versus the European social narratives at work. In this respect, we have shown that the timing in crisis trajectories is crucial. The key phase in which the media sharply frames events as potential existential threats is immediately after the crisis trigger. In the Iraq case, the media construed disagreement over foreign policy -which often occurs -as something novel that threatened the possibility of enlargement to the Central and Eastern European countries. In the case of the Constitutional Treaty, the media construed even the possibility of a negative referendum as the end of the road for integration, even though nearly every treaty before had at first experienced negative referenda. And in the case of the eurozone crisis, the media portrayed a Greek debt crisis that represented only 2 per cent of the eurozone economy as a threat to the very survival of the entire euro, and possibly even the EU itself. Thus, there is a clear point at which the media closes the gap between reality and worst-case-scenario thinking through its volume of coverage and negative spin, seemingly playing a proactive role in building up perceptions of crisis as an integrational panic framework would anticipate.
At the same time, it must be noted that our research focuses on the role of the international media, and it may be difficult to extrapolate beyond this without further investigation. Research of local and national media within Europe could reveal whether they frame these events similarly, and the extent to which they feed off and feed into international coverage, as well as how public opinion is specifically shaped by them. We can also examine weaker cases, and potential instances in which the 'dog didn't bark' to shed light on instances in which the international press may have reported events differently. In particular, there appears to be a kind of meta-narrative that the EU is somehow always dysfunctional, and a tendency to focus mainly on negative developments: bad news sells. The evidence presented in this article suggests that this is a fruitful direction for future exploration. Indeed, we argue that the media has played a strong role in framing challenges and setbacks to EU integration as existential crises, and this is central to the social construction of crises in Europe.
Biographical notes: Mai'a K. Davis Cross is assistant professor of political science and international affairs at Northeastern University and senior researcher at the ARENA Centre for European Studies, University of Oslo. Xinru Ma is a Ph.D. candidate in political science and international relations at the University of Southern California.