Euroscepticism in Southern Europe: A Diachronic Perspective

Abstract

After laying out the rationale and framework of the issue, this introductory article offers a survey of party and popular euroscepticism in European Union member states, Italy, Greece, Portugal, Spain, Cyprus and Malta and candidate Turkey, over several decades. Leonard Ray's criteria of ideological extremity, electoral unpopularity and opposition are used to assess whether South European euroscepticism has been a marginal phenomenon. The article investigates whether Maastricht constituted a turning-point for the rise of euroscepticism and accession for its decline. Finally, it asks whether euroscepticism in Southern Europe is moving towards a new ‘constraining dissensus’.

Keywords:

• European Integration
• Party Attitudes
• Public Opinion
• Italy
• Greece
• Spain
• Portugal
• Turkey
• Cyprus
• Malta

Euroscepticism really emerged as a significant issue on the EU agenda after the early 1990s. The event generally regarded as responsible for ‘uncorking the bottle’ (Franklin, Marsh & McLaren 1994) was the agreement of December 1991 on European Union. The Maastricht Treaty, as it is popularly known, was uniquely qualified to arouse opposition, due to its multiple challenges to national sovereignty, its economic prescriptions with their implications for national redistributive policies, and the fears of the erosion of national identity aggravated by the project for European citizenship. The crisis over the treaty's ratification first shook the cosy belief in a ‘permissive consensus’ in public opinion, under which a broad majority offered passive support for integration, allowing it to advance without significant opposition (Lindberg & Scheingold 1970). Further referenda crises followed over three of the four subsequent European Treaties. Within a seven-year period in the 2000s, voters in one or more member-states successively rejected the Treaty of Nice in 2001, the European Constitution in 2005 and the Lisbon Treaty in 2008. These repeated upsets both decelerated and delegitimised the integration process. In addition to opposition at the mass level, a reduction in consensus was also apparent among national governing elites, including in core EU states. A striking example was the stirrings of dissent in founder member, the Netherlands, long regarded as ‘one of the most enthusiastic supporters of European integration’ (Harmsen 2005, p. 99). Enlargement then added new eurosceptic elites from Scandinavia and Central and Eastern Europe to the EU equation.

The rising salience of euroscepticism is indicated by its central role in a new theory of European integration. ‘Post-functionalist’ theory was developed by Hooghe and Marks (2006, 2009) in the aftermath of the Constitutional Treaty failure. While the neofunctionalists traditionally emphasised the role of interest groups as potential drivers of integration, Hooghe and Marks identify popular euroscepticism as the new brakeman. They argue that public opinion, ignored by earlier theorists due to the development of integration as a series of elite bargains, now restricts elites’ room for manoeuvre. The result, as Hooghe puts it (2007a, p. 5), is ‘a limited zone of acquiescence for policy choice’. Their analysis, that European integration has moved from the ‘permissive consensus’ to a new era of ‘constraining dissensus’, implies a pessimistic prognostication. This is far removed from the neofunctionalists’ positive predictions concerning the forward march of integration, back in the latter's early days.

But how much have things actually changed? Down and Wilson (2008, p. 46) examining data from the European Commission's Eurobarometer surveys, suggested that the overall level of popular support for integration in the early 2000s, while lower than in the 1980s, was ‘little different’ from the 1970s—the era when Lindberg and Scheingold first proposed the ‘permissive consensus’. This raises the question as to whether there is really more euroscepticism now than in the past—or whether perhaps it is simply perceived to have a greater presence because, for example, there has been much more popular consultation on integration in the post-Maastricht era. The only way to answer this question is through diachronic studies, focusing on how euroscepticism has changed over time. As Szczerbiak and Taggart (2008b, p. 26) have noted, this currently constitutes a gap in the literature—one which this issue aims partially to fill through a study of Southern Europe (SE), chosen for reasons that will be explained below. After outlining the framework of the collective project, the article offers an overview of euroscepticism in Southern Europe from a diachronic perspective.

Project Framework

Scope and Time Line

The goal of the present project is to contribute to our understanding of euroscepticism by examining how it has played out in different national contexts. This is by no means the first comparative collection of country case studies of euroscepticism (see, for example, Milner 2000; Harmsen and Spiering 2005; Taggart and Szczerbiak 2008a). However, this is the first collection systematically to examine both the pre- and post-Maastricht periods in order to provide a fuller picture of eurosceptic change. It thus covers the era of the European Community (EC) as well as the post-Maastricht EU (with the term EC/EU used when referring to the whole time period). Rather than setting a uniform starting date for the national case studies, each contributor was encouraged to identify the point that seemed most meaningful in terms of their own country's integration, including pre- and post-accession periods. In terms of a chronological cut-off point, it was considered better not to cover beyond the onset of the international economic crisis in 2008–09, which seemed likely to open a new era in terms of attitudes to European integration in Southern Europe. To date, the study of euroscepticism has developed along two often separate axes, party and popular euroscepticism, concerned with elite and mass attitudes respectively. While most of the literature focuses on either one or the other, ¹ the contributors to this project were asked to include both, again with the aim of providing a fuller picture.

Regional Focus: The Choice of Southern Europe

In terms of geographical scope, the collection edited by Milner focused on north-west Europe, while the other two comparative volumes cited above include a range of countries from both Western and Eastern Europe, of different sizes and relative power profiles, and with different ages of membership, ranging from founder members to 2004 entrants. The present project also aims to emphasise diversity, but has chosen to do so within one particular region. As a laboratory for the study of attitudes towards European integration, Southern Europe offers a particularly rich range of case studies. In terms of membership age, it includes a European Community founder member (Italy), three ‘second generation’ members (Greece, Spain and Portugal), two recent entrants (Cyprus and Malta), and one negotiating candidate with a long-term relationship with European integration (Turkey).

Moreover, for the purposes of a diachronic perspective, Southern Europe is likely to be particularly fruitful, as all seven states have been closely linked to the European integration project since the era of the European Community, with four members of this group being the first states to sign Association Agreements with the EC in the 1960s and 1970s. (See Table 1 for a time line of South European relations with European integration.) The choice of SE thus offers a good basis to study euroscepticism over several decades. Meanwhile, euroscepticism in SE has to date been under-researched and for some of our case studies, notably Cyprus and Malta, the articles in this issue are pioneers in examining the phenomenon in their respective countries.

Table 1 South European Countries: Association and Accession

Country	Association (date of signature)	Full Membership
Italy	–	1957
Greece	1961	1981
Portugal	–	1986
Spain	–	1986
Cyprus	1972	2004
Malta	1970	2004
Turkey	1963	–

Perhaps the most important reason to choose Southern Europe, however, is the region's reputation, in the words of Hooghe & Marks (2007a, p. 13) as ‘the EU's most pro-European’. According to Gómez-Reino, Llamazares & Ramiro (2008, p. 134), ‘the incorporation of Mediterranean countries into the European Union was preceded, accompanied, and followed by a wide consensus on the positive effects of European integration’. Examining popular euroscepticism in Greece, Spain and Portugal, Llamazares and Gramacho (2007, p. 212) note that respondents from these countries, together with the Italians, rank ‘among the most euro-enthusiast’ in Europe. In the case of Italy, Conti (2003) remarks that this ‘is a country that has for a long time been seen as one of the most Euro-enthusiastic among the member states’, while Quaglia (2008, p. 58) claims that ‘in the past, any reference to “Italian euroscepticism” would have been regarded as an oxymoron’, given the country's unremittingly pro-European image.

This picture largely refers to the ‘old’ SE, i.e. the four former dictatorships which joined the EC pre-Maastricht, where integration acquired especially positive connotations through its promotion as the route to democratisation. It applies less to the ‘new’ Southern Europe of Turkey, Cyprus and Malta, which do not necessarily share a strong pro-integrationist stance. Nevertheless, this europhile image suggests SE might be a particularly useful weathervane when considering the future of a ‘post-functionalist’ European Union. If euroscepticism has developed into a force to be reckoned with even in this region, or has at least grown there significantly in comparison with the past, this might confirm the ‘post-functionalist’ diagnosis of a ‘constraining dissensus’ likely to limit the further development of the EU.

Defining Euroscepticism

In establishing a framework for the issue, an important matter is how to define euroscepticism, a topic on which there has been a certain amount of academic debate. For example, Oliver Daddow (2006, p. 64), writing from a historian's perspective, has suggested the possible meanings of the term could include ‘a broader-brush populist scepticism about anything to do with Europe at all’. However, in everyday language, notably in the media, euroscepticism has acquired the rather specific meaning of questioning European integration and it is this usage that has also become established in political science. It therefore does not seem useful to broaden our working definition beyond this.

Nevertheless, as Nick Sitter (2001) has commented, ‘euroscepticism is not a single coherent stance on the EU as a polity’. Instead it covers a broad range of positions in which opposition can concern the whole concept of European integration or be confined variously to its current form, to particular present aspects or policies, and/or to future deepening, either in general or with regard to specific competences. This has led some authors to attempt to further narrow the definition. Notably Kopecky & Mudde (2002) suggested that the term ‘eurosceptic’ should apply only to one specific category of integration opponents: that of ‘europhile europessimists’, i.e. those who are positive about European integration but negative about its present or pessimistic about its future development. But this excludes the key group of hardcore eurodoubters—the ‘europhobe europessimists’ whom these two authors classify in a separate category as ‘euro-rejects’. It does not seem particularly helpful to our understanding of opposition to integration to suggest the phenomenon should be studied without including its most determined opponents.

The problem that concerned Kopecky & Mudde, of capturing the category of those who are positive towards integration in principle but opposed to its current practice, has also occupied other authors, such as Wessels (2007), who proposed a three-point scheme of ‘adamant eurosceptics’, ‘eurosceptics’ and ‘critical Europeans’. Moreover, eurosceptics and integration supporters are not permanent, mutually exclusive categories. Instead, the borders between these two groups may be fluid, changing according to circumstances and over time. Conti (2003) in particular has highlighted this problem with his category of ‘functional Europeans’, whose support for European integration is not based on fundamental commitment and therefore may be changeable.

Despite considerable sympathy with these classificatory attempts and the reasoning behind them, it was felt the most useful approach for the issue would be to use a definition which, while limiting euroscepticism to European integration, would encourage the authors to opt for the broadest possible coverage of negative attitudes towards the EU. Hence, the project has adopted the definition of euroscepticism proposed by Paul Taggart (1998, p. 365) as ‘contingent or qualified opposition, as well as … outright and unqualified opposition to the process of European integration’.

Also utilised is the classificatory distinction between hard and soft euroscepticism subsequently developed by Taggart and Szczerbiak. Hard euroscepticism is defined as ‘principled objection to the project of European integration as embodied in the EU’, involving ‘outright rejection’ of both political and economic integration and ‘opposition to their country joining or remaining members of the EU’ (Szczerbiak & Taggart 2003, p. 12; Taggart & Szczerbiak 2008a, pp. 7–8). With soft euroscepticism, ‘there is not a principled objection to the European integration project or transferring powers to a supranational body such as the EU, but there is opposition to the EU's current or future planned trajectory based on the further extension of competences that the EU is planning to make’ (Szczerbiak & Taggart 2003, p. 12).

In the case of party euroscepticism, contributors were invited to examine only those parties that have been represented in the national or European Parliaments. The world of extraparliamentary politics often includes a colourful range of opposition to integration, but it was felt venturing into this domain risked entanglement in the potentially picturesque but politically irrelevant. In identifying eurosceptic parties, contributors were asked to provide information on their stands in national parliamentary ratification of EC/EU treaties. Richard Dunphy (2003, p. 3) has cautioned against a simplistic identification of opposition to an EC/EU treaty with opposition to European integration. As he rightly remarks with regard to the Maastricht Treaty, ‘it is perfectly possible to oppose [it] on the grounds that it was not sufficiently integrationist—that it did not go far enough’. Nevertheless, European Treaty ratification is a ‘history-making moment’, when political forces stand up to be counted concerning the direction that European integration is taking. The political statement made by abstaining or voting against a European Treaty would appear to express at least some soft euroscepticism and therefore appears relevant to our account.

The next section opens our overview of euroscepticism in SE by laying out the specific questions to be investigated. This will be followed by two sections examining party and popular euroscepticism in SE. The article will conclude by attempting to answer the question of whether a ‘constraining dissensus’ now seems to be emerging in Southern Europe.

Overview of SE Euroscepticism: Research Questions

The basic subject under investigation in this issue is how the pattern of South European euroscepticism has changed over time and, in particular, whether party and popular euroscepticism are stronger within national political systems than in the past. To examine this, two working hypotheses have been formulated. The first is that the Maastricht Treaty had a ‘system transforming’ effect on South European euroscepticism (Down & Wilson 2008, p. 38). This hypothesis thus seeks to determine the eurosceptic response to the deepening of integration. The second hypothesis is concerned with euroscepticism as characteristic of a particular phase of the national trajectory in European integration. Specifically, it proposes that there is an ‘accession effect’, with euroscepticism declining following full membership. The experience of recent and currently aspiring EU entrants (the Central and Eastern European Countries [CEECs], Croatia and Turkey) suggests that in a candidate state, the demanding nature of accession negotiations is likely to result in a popular reaction against EU entry. In contrast, the experience of membership and participation in the latter's benefits—including decision-making influence, enhanced international status and possible budgetary inflows—might be assumed to have a mitigating effect on euroscepticism.

Party euroscepticism has been famously characterised by Paul Taggart (1998) as ‘a touchstone of dissent’, a characteristic of protest parties. Following Taggart's original observation that ‘no major governmental party in an EU member-state is unequivocally anti-EU’ (Taggart 1998, p. 381), Sczcerbiak and Taggart (2000, pp. 5, 8) subsequently proposed a core–periphery distinction, with eurosceptic parties located on the periphery of the party system, but likely to modify their stance when mutating into parties of power. Hooghe, Marks & Wilson (2004) found euroscepticism to be concentrated in particular party families, which happen to be principally those located towards the two ends of the political spectrum. The picture of party euroscepticism which emerges from these authors is thus one of a marginal rather than a mainstream phenomenon. Following this literature, Leonard Ray (2007, p. 154) has identified the basic features of eurosceptic marginality as ‘ideological extremity, unpopularity and opposition’, which he suggests can be measured by ideological positioning, national parliamentary vote share and government participation. Ray's three criteria will be employed in this article to assess the marginality (or otherwise) of party euroscepticism. Meanwhile, in the case of popular euroscepticism, the measure is straightforwardly numerical, based on percentages of eurosceptic opinion poll responses or euro-referenda votes.

Party Euroscepticism before Maastricht

The first finding on examining party euroscepticism in Southern Europe is that there has been considerable national variation across the region. This can be illustrated by the crucial decision to enter the EC/EU as a full member. As Table 2 shows, of the six South European member states, accession was a consensual choice with unanimous parliamentary support in two cases—Spain and the Republic of Cyprus. Behind this consensus was the belief that European integration would provide the essential framework for, respectively, the stabilisation of Spanish democracy and the solution of the Cyprus problem. In Portugal and Italy, EC entry was opposed by one and two opposition parties respectively. But the last two cases, Greece and Malta, were polarised around the issue of participating in European integration, with opposition coming from parties of power. In Greece, just ten months after EC entry, almost 60 per cent of the vote was gained by two hard eurosceptic parties (see Verney in this issue). In the Maltese party duopoly, under which only two parties have been represented in parliament since 1968, one of the two poles of the system promoted non-alignment and opposed accession for over three decades. In both cases, it is clear that accession would not have happened—at least, not when it did—if the roll of the electoral dice had turned out differently.

Table 2 South European Parties against Accession Treaty Ratification

Country	Parties voting against	Parties abstaining
Italy (Treaty of Rome)	PCI Communist	PSI Socialist
Greece	None	PASOK Socialist
		KKE Communist*
Portugal	PCP Communist	None
Spain	None	None
Cyprus	None	None
Malta	MLP Socialist	None
Note: *The Greek socialists and communists, having previously declared they would vote against the accession treaty, both boycotted Parliament during the debate. In this case, the boycott was intended to indicate a stronger degree of opposition than voting against—in the words of the Greek socialists’ leader, it constituted a refusal to legitimise the procedure by the party's participation (Papandreou 1979).

Ideological Positioning

Examining party euroscepticism in the pre-Maastricht era, one of the reasons why SE has traditionally been regarded as a region of low euroscepticism becomes immediately apparent. In the 1970s and 1980s, in the two large member states, Italy and Spain, there was a pro-integrationist consensus, which embraced all parliamentary parties. Indeed, the Italian parliament during the 1980s manifested a rather striking pro-federalist activism, calling for the ratification of the European Parliament's Draft Treaty on European Union, criticising the Single European Act for not going far enough in a federalist direction, and pressing for a referendum to give the EP a mandate to draw up a new Treaty on European Union (Preda 1991). Thus, as shown in Table 3, when the Single European Act came up for ratification in the mid-1980s, the only parliamentary forces opposing it in the then four member-states were the two orthodox communist parties (CPs) in the two smaller countries. In both cases, these CPs were permanent opposition parties and minor players in their respective national systems.

Table 3 South European Parties against European Treaty Ratification

Treaty	Country	Parties voting against	Parties abstaining
Single European Act	Italy	None	None
	Greece	KKE Communist	None
	Portugal	PCP Communist	None
	Spain	None	None
Treaty of European Union	Italy	None	None
	Greece	KKE Communist	None
	Portugal	PCP Communist	CDS-PP Conservative
	Spain	EE Regionalist Left	IU Radical Left
Treaty of Amsterdam	Italy	None	LN Regionalist/right
	Greece	KKE Communist/DIKKI Left populist	SYN Radical left
	Portugal	PCP Communist/BE Radical Left	None
	Spain	IU Radical Left	None
Treaty of Nice	Italy	RC Communist	None
	Greece	KKE Communist	SYN Radical Left
	Portugal	PCP Communist/BE Radical Left	None
	Spain	None	IU Radical Left
Treaty establishing a European Constitution	Italy	RC Communist LN Regionalist/right	None
	Greece	KKE Communist/SYRIZA Radical Left	None
	Portugal	Not ratified	Not ratified
	Spain	IU Radical Left/BNG, ERC, CHA, EA, Na-Bai Regionalist/left	None
	Cyprus	AKEL Communist	KOP Green
	Malta	None	None
Treaty of Lisbon	Italy	None	None
	Greece	KKE Communist/SYRIZA Radical left/LAOS Radical right	None
	Portugal	PCP Communist
	BE Radical left	None
	Spain	ERC, BNG Regionalist/left IU/Radical Left	ICV Regional eco-socialist/Na-Bai Regionalist/left
	Cyprus	AKEL Communist	KOP Green
	Malta	None	None
Note: The author would like to thank Nicolo Conti, Antonio Egea de Robles, Ignacio Molina, Christophoros Christophorou, Marco Lisi, Sophia Michalaki and Kyriaki Kafyra for responding to queries concerning the information in this table. Responsibility for any errors remains the author's alone.

However, the Italian and Spanish pro-integrationist consensus was a new development: earlier, integration had encountered opposition in both countries, specifically from the communist parties. As shown in Table 2, in the 1950s the Italian communists voted against ratification of the Treaty of Rome. Meanwhile, in the early 1960s, their Spanish counterparts, then operating underground during the Franco dictatorship, campaigned against the regime's aim of ultimate Association with the EC (Story 1979, p. 154). The change in communist policy was thus crucial. Eurocommunism was largely a South European phenomenon, with the Spanish and Italian parties its major proponents. There was also a small eurocommunist party in Greece, which, while represented in Parliament, was electorally overshadowed by the orthodox CP. Adopting the strategy of a democratic road to socialism, these parties sought a framework for the establishment of a stable democracy in which the left could come to power and promote socialist change in a continent divided into Cold War blocs. In moving away from the Soviet embrace, they reassessed European integration as offering more favourable ground for the pursuit of this aim than a national road to socialism. Eurocommunism thus constitutes a major explanatory factor for the limited level of South European euroscepticism at this time.

Another significant factor was the complete absence of right-wing euroscepticism in all but one case. The exception was Turkey, which, as noted by Yilmaz (in this issue), displayed a quite different pattern, with euroscepticism located not only on the far left but also on the nationalist right and in the Islamist movement. In fact, Turkey was a rather special case because of the protracted and difficult nature of its relations with the EC (and, subsequently, the EU). As Günes-Ayata (2003, p. 205) notes, ‘since the country signed a treaty with the European Common Market in 1963, there has been no political tendency (right, left, centre, Islamist, nationalist) that has not gone to some degree through a stage of euroscepticism’. This applied particularly to the 1970s, when economic crisis and disappointment over the functioning of the Association Agreement combined with growing alienation from the EC over the Cyprus question and Greek accession to fuel euroscepticism. At this time, a soft euroscepticism was expressed by the Kemalist and etatist CHP (Cumhuriyet Halk Partisi—Republican People's Party), which entered the Socialist International in 1975. CHP supported Turkey's westernisation but, particularly in the 1970s, had increasing difficulties with the programme of market opening which lay at the heart of the Association.

In contrast, in the other SE states pre-Maastricht, euroscepticism took a hard form, focused around rejection of national participation in the European Community. Soft euroscepticism was generally expressed only by parties that were in the process of moving away from an initial hard eurosceptic stance, such as the Greek socialists in the 1980s.2 During this period, especially the early years when integration was a relatively new process, hard euroscepticism still appeared a meaningful position. Only one SE state was a founder member, so for the rest the issue of membership remained open. In the six SE states apart from Turkey, prior to the 1990s, opposition to integration could be found exclusively among parties of the anti-capitalist left: the communist parties in Greece, Portugal and Cyprus, as well as in Italy and Spain (in the 1950s and 1960s); a Greek left front party (pre-1967 when the CP was illegal); and the socialist parties in Italy, Greece and Malta. Among the socialists, for the Italians euroscepticism was a short-lived phase in the 1950s (see Quaglia in this issue). The Greek and Maltese socialists were rather different. Having little in common with West European social democracy, which they strongly criticised, they preferred a non-aligned and third worldist orientation to EC membership.

Marginal or Mainstream?

In these six countries, euroscepticism in the pre-Maastricht period thus fulfils one of our three criteria of marginality, through its limitation to the far left of the political spectrum. In contrast, in Turkey, Dosemici (2011 forthcoming) notes that, after 1967, eurosceptic sentiments were expressed across the political spectrum, unlike the early 1960s when, he claims, there was ‘near unanimous support’ for integration. Subsequently, when political parties began functioning again after the 1980 military coup, there was greater consensus around the EC orientation, including a new turn in economic philosophy towards the market economy.

In terms of the second criterion, electoral popularity, in both Turkish elections which took place in the 1970s the eurosceptic parties were among the most important political forces. The Kemalist CHP was the first and the Islamists the third party, while the nationalists were a smaller force with three parliamentary seats in 1973 and 16 in 1977.3 In our other six countries, the anti-capitalist and eurosceptic left constituted a significant force in four. Under their respective two-party systems, the socialists were one of the two potential parties of government in Malta throughout the period and in Greece from 1977. The Greek socialists’ vote underwent a meteoric rise, from under 14 per cent immediately after the party's foundation in 1974 to a highpoint of 48 per cent in 1981, while under the Maltese party duopoly the socialists’ vote share in the 1970s and 1980s ranged from 48.9 to 51.5 per cent. The communists were the second party in Italy throughout the 1950s and 1960s, with around one-quarter of the vote. Also forming the second party were the Cypriot communists, apart from the elections of 1981 when they were first and 1985, when they came third. Only in Portugal and Greece were the communists the third or fourth party in every election, consistently scoring less than 15 per cent of the vote.

With regard to the third criterion of marginality—that of opposition—eurosceptic parties participated in power in three South European countries before 1991. In Turkey prior to the 1980 coup, a series of unstable coalition governments included both the Islamists (1974–75 and 1975–77) and the nationalist right (1975–77), while the CHP led a coalition government in 1978–79. Single-party governments were formed by the eurosceptic socialists in Greece (1981–85)4 and Malta (1971–87). The sole communist government participant—albeit very briefly—was the Greek CP, at a time when its traditional hard-line opposition to integration had been muted to a soft eurosceptic stance. In 1989–90, during a period of perestroika in the Soviet Union and of national political crisis in Greece, the Greek communists, as part of a left alliance, took part in two short-lived coalition governments. In contrast, the Italian, Spanish, Portuguese and Cypriot communists were in permanent opposition.

Eurosceptics in Power

Government participation by eurosceptic forces had a considerable impact on national relations with integration. Malta, despite an Association Agreement dating back to 1970, for almost two decades did not pursue closer relations with the EC, due to the 16-year tenure of the eurosceptic socialists—the longest-lasting eurosceptic government in SE. It was only after the socialists lost power in 1987 that the succeeding government actively pursued accession, submitting a formal application in 1990. The timing of the Maltese membership request meant it became enmeshed with enlargement to Central and Eastern Europe, resulting in a further 14-year wait before the country joined what was by then the European Union.

In 1970s Turkey, in a climate of economic crisis and increasingly violent political unrest, a succession of coalition governments including Islamists and nationalists do not seem to have raised obstacles to Turkey's EC links on grounds of principle, but pushed for greater flexibility from the EC in the implementation of the Association.5 In 1978 a CHP-led government announced a five-year unilateral moratorium on the fulfilment of Turkey's obligations under the Association. The Turkish action, followed by the military coup of 1980, was to result in what was essentially a lost decade for Turkey–EU relations. The latter were only formally reactivated in 1986, subsequently leaving Turkey at the back of the now rapidly growing enlargement queue.

For these states with Association agreements, the election of eurosceptic governments thus led to the distancing of the countries concerned from the EC and probably delayed their accession processes by decades. For a member-state, the consequences were rather different. The Greek socialists’ first administration (1981–85) was the sole example of a single-party eurosceptic government in a South European member-state not only in the pre-Maastricht era, but also up to the present. It may, therefore, be regarded as an exceptional case. It had a long-term impact on Greek–EC relations, shaping an image as a difficult partner which the country has not fully overcome almost three decades later. In addition, it had consequences for the integration process. In an EC of only ten members, before the first amendment of the Treaty of Rome, the Greek government had considerably more veto power than an average member-state today. It was thus able to dilute foreign policy cohesion and, often by lining up with two other dissenters, the UK and Denmark, to hinder the deepening of integration (see Verney in this issue). While the presence of the Greek government was not the main factor, it clearly contributed to the ‘constraining dissensus’ which seemed to be paralysing integration in the early 1980s.

Post-accession Decline

Of the four states that entered the European Community during this period, the hypothesis that opposition to European integration diminished after accession and before Maastricht is confirmed in two cases. These were Italy, due to eurocommunism, and Greece, where, after a few years of full membership, the socialists had become reconciled to EC membership and the communists were also shifting ground (see Verney in this issue). For the Greek socialists, the change occurred after a few years in power and for the two communist parties at a time when they aimed to become coalitionable. The cases of these three parties therefore support the Szczerbiak and Taggart thesis that parties which are either participating in government or aiming to do so will move away from euroscepticism. The hypothesis concerning post-accession change does not apply to the two Iberian countries. In the Spanish case, no eurosceptic parties were represented in Parliament before accession, so there could not be a reduction of party euroscepticism afterwards. In Portugal, where there seemed no prospect of communist government participation, the communist party—the country's sole eurosceptic force—did not change its stance in the five-year period between EC entry in 1986 and the Maastricht Treaty agreement in 1991.

Party Euroscepticism after Maastricht

The post-Maastricht era saw new patterns of euroscepticism take shape. European integration was now well established and undergoing rapid deepening, most of the states that had previously remained outside—the neutral members of the European Free Trade Area and the Central and East European countries—were queuing up to join, and four of our South European countries were already member states. Under these conditions, hard euroscepticism became a less sustainable position and, with a few exceptions, virtually disappeared from Southern Europe. In its place, a broad range of soft eurosceptic positions emerged, with considerably more ideological variation.

Ideological Positioning

On the right, in Turkey the 1990s found both the nationalists and the Islamists taking eurosceptic positions, including opposition to the customs union agreement signed in 1995. The nationalists retained a soft eurosceptic stance in the 2000s, declaring support for Turkey's EU membership so long as it did not conflict with what they defined as long-term national interests including foreign policy issues, notably Cyprus, and the domestic treatment of minorities (Avci 2004). In one of the most striking reversals of the post-Maastricht era, after 1999 the Islamists switched to support for EU membership as a central plank of their policy, a change attributable to the dynamics of domestic politics and the realisation that Europeanisation, as a route to democratisation, could create a framework allowing the formation of a stable Islamist government.

Meanwhile, right-wing euroscepticism was no longer confined to Turkey but now appeared in three more SE countries. Its emergence did not follow a common pattern, as the parties concerned had disparate ideological orientations. In Portugal, the conservative party (CDS-PP) briefly adopted a eurosceptic stance in the years immediately after Maastricht (see Costa Lobo and Magalhaes in this issue). In Greece, a party of the radical right (LAOS), belonging to the new trend of populist and anti-immigrant parties in Western Europe, was founded in 2000. In Italy, following the collapse of the previous party system in 1993, euroscepticism appeared among the parties of the new centre-right bloc (see Quaglia 2008 and in this issue). The European stand of the two larger parties, Forza Italia and Alleanza Nazionale, which merged in 2007 to form a single party of the centre-right, was differentiated from the bipartisan euro-enthusiasm of Italian parties in the past, having a more pronounced national orientation and a more pro-Atlantic foreign policy.6 However, they will not be included in this survey of party euroscepticism because, at the history-making moments when European Treaties were being ratified, they always voted in favour.7 This was not the case with their coalition partner, the Lega Nord, like LAOS a populist and anti-immigrant party, which adopted a more clearly eurosceptic stance.

The Lega's European policy was closely linked to its primary identity as a regionalist party demanding self-determination for northern Italy. The party's initial europhile stance drew on references to a Europe of the Regions as the most appropriate framework to realise this project, while the 1998 switch to euroscepticism seems to have been at least partly influenced by the belief that Italy's successful entry to Economic and Monetary Union would weaken support for the country to split. The Lega's eurosceptic stance has been anything but consistent, as indicated by the party's record with regard to European Treaty ratification: it abstained on the ratification of the Treaty of Amsterdam, voted for the Treaty of Nice, against the European Constitution and for the Treaty of Lisbon (see Table 3).

Regionalist euroscepticism constitutes a second strand of the new euroscepticism in Southern Europe. It reflects the deepening impact of integration on centre–periphery relations within the nation-state, an impact that varies from case to case, resulting in a differentiated stance towards the EU (see De Winter & Gómez Reino 2002). Besides Italy, it has also emerged in Spain—the only other SE member state large enough to have regional autonomist movements. In the Spanish case, as the table shows, a series of regionalist parties from the Basque country, Catalonia and Navarre have opposed European Treaty ratification in the last decade. In contrast to the Italian Lega Nord, however, these have all been left-wing. Their stand has been partially influenced by the unwillingness of the EU institutions to support autonomist demands that would undermine the territorial integrity of the member-states. In contrast, in candidate country Turkey, the Kurdish political parties have not been eurosceptic, viewing European integration as offering a more favourable framework for recognition of their ethnic identity, and the EU institutions as potential allies in their struggle for minority rights.

A third new strand of SE euroscepticism has been Green. This was quite weak, first because post-materialist values have been less appealing in a region with a lower economic development level and so Green parties have not been strongly represented in South European parliaments. Second, the Green movement has been divided in its stance on European integration. While in the past Green parties tended to be eurosceptic, in recent years they have become more supportive of integration. Thus, for example, the Greek Eco-Greens are pro-integrationist. However, Europe, a eurosceptic environmentalist party was represented in the Greek Parliament in 1990–93 while in the 2000s, both the Cypriot and Catalan Greens abstained in the ratification of the Treaty of Lisbon.

Alongside these new entrants remained the traditional backbone of South European euroscepticism—the left. Within the South European socialist family, the Maltese socialists continued to oppose their country's EU membership in the post-Maastricht period. The Greek socialists, following the earlier change in their EC policy, in the 1990s became consolidated as a pro-integrationist party and the main architect of their country's entry to Economic and Monetary Union. Meanwhile, a short-lived Greek socialist splinter (DIKKI) accepted the country's EU membership but rejected the direction integration was taking post-Maastricht. In Turkey, the CHP, which had adopted an enthusiastic pro-integration stance in the 1990s, moved to a soft eurosceptic stance after 2002 when it became the official opposition to an Islamist government elected on a pro-integrationist platform. While the CHP still proclaimed its support for the EU membership project in principle, in practice its support, like that of the nationalists, was conditional on accession not clashing with national interests such as Cyprus and it strongly opposed many of the reforms required by the accession process (see Gulmez 2008).

The neoliberal deepening of integration inaugurated by Maastricht and strengthened by subsequent treaties also evoked new opposition from a group of post-eurocommunist parties, which formed the basis for a new family of the radical left. Maastricht of course coincided with the formal death of eurocommunism, marked by the dissolution of the Italian communist party in 1991 (preceded by the Greek eurocommunist party in 1986). The new parties were the Spanish IU, an electoral alliance formed in 1986 with the Spanish CP as its dominant component; the Greek SYN, formed in 1992 by former eurocommunists and cadres from the renewal wing of the orthodox CP; and RC (Rifondazione Comunista), an Italian communist successor party. While continuing to support European integration in principle and rejecting an exclusively national road to socialism, these parties became increasingly critical of the direction that integration was taking, moving to a soft eurosceptic stance which became more pronounced with the passage of time.

Among the orthodox communist parties, the Greek communists’ flirtation with soft euroscepticism ended with their decision to vote against the Maastricht Treaty, after which they returned to a more hard-line stance. The Portuguese CP softened its stance after Maastricht, but it too returned to a harder eurosceptic line by the end of the 1990s (see Costa Lobo & Magalhaes in this issue). Subsequently, the Greek and Portuguese CPs can be regarded as the basic rearguard of hard euroscepticism in Southern Europe. They are also the only two parties in Southern Europe which, throughout the decades in which their countries have participated in European integration, have consistently opposed European Treaty ratification at every opportunity on which they have been offered, from accession to Lisbon. In contrast, the Cypriot communists moved to a soft eurosceptic stance. Formally dropping their opposition to their country's EU membership in 1995, they retained a ‘eurocritical’ stance on the direction integration was taking, as reflected in their votes against the European Constitution and Lisbon Treaty (see Agapiou-Josephides in this issue). Finally, a new addition to the SE eurosceptic radical left was the Portuguese BE (Bloco Esquerda—Left Bloc), founded in 1999.

Marginal or Mainstream?

Turning to the three criteria of marginality, as a general rule all the South European eurosceptic forces listed above were located in a marginal position towards one of the two ends of the political spectrum. The sole exception was the Turkish CHP, which is usually regarded as centre-left. On the second criteria, electoral unpopularity, eurosceptic parties with strong electoral support were found in the two 2004 entrants. In Malta, under the party duopoly, the socialists’ lowest post-Maastricht vote share was 46.5 per cent, while in Cyprus, the communists generally gained 30–35 per cent of the vote, making them the second party in the 1990s and the first party in the 2000s. The third ‘new’ SE country, Turkey, also registered a substantial eurosceptic vote, with the Islamists emerging as first party with over 21 per cent of the vote in 1995, the nationalists as second party with 18 per cent in 1999, and the CHP during its post-2002 soft eurosceptic phase as second party with almost 21 per cent in 2007. In contrast, in the four ‘old’ SE states, there was only one occasion on which an individual eurosceptic party gained more than a 10 per cent vote share (the Spanish IU with 10.5 per cent in 1996).

With regard to the third criterion, eurosceptic parties took part in government in four of our South European states during this period: in Italy and Cyprus for the first time and also in Turkey and Malta. The Maltese socialists formed the only post-Maastricht single-party eurosceptic government in Southern Europe (1996–98). Eurosceptic partners in coalition governments included, in Turkey, the Islamists (1996–97) and nationalists (1999–2002) and in Italy, the Lega Nord (2001–06 and post-2008)8 and the ‘reconstructed communist’ RC (2006–08). The most striking instance of communist government participation occurred in Cyprus, with cabinet participation in 2003–07 followed in 2008 by the election of the party's leader as President of the Republic, the country's central political post. Thus, all three of the ‘new’ SE states experienced eurosceptic parties in government post-Maastricht, as opposed to only one of the four ‘old’ SE states. For the latter, Italy, this marked a rather major change, given this particular country's pro-federalist past. In two ‘old’ SE states, Spain and Portugal, eurosceptic parties have never participated in power.⁹

Eurosceptics in Power

In the two SE states that were non-EU members when eurosceptic parties took power in the 1990s, the impact was once again on national relations with the EU. In Malta, then a candidate for membership, the socialists suspended the accession application. As a result, Malta was initially excluded from the next round of membership negotiations, until the socialists lost power and the application was reactivated two years later. In Turkey, then an accession applicant, the 1996–97 government, including the Islamists, whose favourite foreign policy project was the foundation of the D-8 group of Muslim nations, marked a new low point in Turkey–EU relations. It contributed to the Luxembourg decision, six months after the coalition's fall, not to recognise Turkey as an EU candidate. Within the government of 1999–2000, a period when Turkey had been recognised as a candidate for EU membership and was working to meet the political criteria that would allow the opening of accession negotiations, the nationalists ‘frequently acted as a brake when it came to passing necessary reform legislation’ (Avci 2004, p. 209), thus potentially slowing down the accession process.

Eurosceptic government participation in two member states in the 2000s does not seem to have had a constraining impact on the course of integration, in contrast to the earlier experience with the Greek government of the 1980s. Of course, by this time the potential for individual states to influence integration had been reduced by both EU enlargement and European Treaty revisions. Meanwhile, in Italy, both Lega Nord and Refondazione Comunista were contained within coalitions. While Lega Nord representatives on occasion made incendiary eurosceptic statements, they did not push government policy in a eurosceptic direction. Cyprus's size—it is the third smallest member of the EU-27—reduces its potential for influence over the integration process. In government, the communists needed to retain support for the national priorities of the Cyprus question and EU–Turkey relations. They had no wish to risk this by alienating the country's EU partners over broader questions of integration.

Post-accession Decline

Of the two countries that entered the EU during this period, Malta constitutes a clear case of party euroscepticism melting away after accession. The change in fact began a little earlier, following the socialists’ defeats in the EU membership referendum and national election the year before EU entry. After this, the socialist party, having opposed accession for three decades, decided to accept the new situation (see Pace in this issue). The extent to which the socialists reoriented their policy was indicated by their votes in favour of the European Constitution, the Lisbon Treaty and euro entry. This turned Malta from a country that had always been polarised on the EC issue to one with an apparent consensus on the main choices related to integration. Indeed, Malta was the only South European country in which the last two amended European Treaties have been ratified unanimously. In Cyprus, the decline in euroscepticism also pre-dated accession—although, in this case, by almost a decade, dating back to the communists’ 1995 decision to support accession, mentioned earlier. The change in EU policy was part of the Cypriot communists’ slow transformation from a party of permanent opposition to potential government participant. It was also influenced by the hope that EU membership would strengthen the Republic of Cyprus internationally and provide a potential route to solution of the Cyprus problem. Meanwhile, by 2003, the Maltese socialists had been out of government for all but two of the past 16 years and now needed an integration policy that would allow them to appear as a credible potential government in what was about to become an EU member-state. Thus, in both cases, the policy change seems to have been linked to the approach to power, as suggested by Szczerbiak and Taggart (2000).

Popular Euroscepticism in Southern Europe

The main theme of this section will be the study of opinion polls, supplemented by a discussion of referenda on integration. The examination of public opinion will be based on the Eurobarometer (EB), conducted in the member states on a biannual basis since 1973 and therefore providing a basis for diachronic and intra-country comparison. Usually, analyses drawing on EB data focus on the level of support for European integration. In contrast, this presentation investigates the strength of opposition. Two basic questions have been chosen as indices of euroscepticism.

In the first, asked since 1973, respondents are asked to rate their country's EC/EU membership as ‘a good thing’, ‘a bad thing’ or ‘neither good nor bad’. The three-point scale, including the possibility of a non-committal answer, means that opting for the answer ‘a bad thing’ can be interpreted as an unambiguous statement of opposition to membership, thus expressing hard euroscepticism. The second question, asked since 1983, seeks an assessment of whether the country on balance has benefited from membership (or will benefit in the case of candidate countries). Because this question offers only two choices—‘benefited’ or ‘not benefited’—it requires respondents to take a clear-cut position either for or against (unless they opt for Don't Know/Don't Answer, which consistently scores higher than for the first question). The answer ‘not benefited’ thus captures a broader range of negative sentiment towards European integration, including individuals who might not go so far as taking an openly negative stance against membership. It therefore embraces both hard and soft euroscepticism. The differences in the options offered produce an interesting discrepancy between the answers to these two questions, with a significant proportion of respondents taking the apparently contradictory position of not describing membership as ‘a bad thing’ even though they consider their country has not benefited. From the first EB in 1973 to the 72nd survey in Autumn 2009, the number of membership opponents in the member-states has ranged from seven to 17 per cent, while benefits sceptics have been considerably more, from 22 to 36 per cent. Nevertheless, both opposition to membership and scepticism regarding benefits have always been minority positions within the EC∖EU as a whole.

Our examination of public opinion begins in the 1980s, the decade of EC enlargement to Greece, Spain and Portugal, and is divided into three time periods. The first cut-off point is, of course, the agreement on the Maastricht Treaty in December 1991. The second dividing line is the Fifth Enlargement in spring 2004, when the former four South European member states became six. For these years, the Eurobarometer also includes regular measurements for Turkey. During this third period, which coincides with the crises over the European Constitution and the Lisbon Treaty, the spectre of euroscepticism loomed particularly large over European integration.

Public Opinion pre-Maastricht

During the 1980s, in the EC as a whole, benefits sceptics ranged from one-quarter to one-third of respondents while membership opponents, from a high point of 17 per cent, were clearly on a downward trend, falling to single figures in the second half of the decade. In Southern Europe in both categories, the average level of euroscepticism was fairly consistently below the average for the EC as a whole in three of the four states (as shown in Figures 1a and 1b). However, in both cases there was one SE outlier, which not only diverged from the other SE countries, but was also generally above the EC average—sometimes spectacularly so. What is intriguing is that this is not, as might be expected, the same country for both questions.

Figure 1a Membership Euroscepticism in Southern Europe, 1981-1991.

Figure 1b Benefits Euroscepticism in Southern Europe, 1983-1991.

Greece, the SE member state with the highest electoral support for eurosceptic parties during this period, also had the highest proportion of membership opponents (see Figure 1a). Its results on this question followed the trend for the North European countries of the first Enlargement, which had higher numbers of membership opponents than the EC founder members. In particular, the Greek results were similar to those for Ireland. In contrast, Spain and Portugal showed low levels of opposition to membership similar to the original EC6. However, rather unexpectedly, it was Spain that had the highest level of scepticism concerning benefits, where for several years the Spanish results were as high as 25–30 percentage points above the EC average (see Figure 1b). It is rather striking that, in a country where, as we have seen, the parliamentary elite had unanimously favoured membership, for the first three years after entry significantly more than half the respondents had a negative evaluation of the benefits of membership. This also meant that, in the Spanish case, there was an enormous divergence—sometimes as high as 50 per cent of the Spanish sample—between benefits sceptics and membership opponents.

Investigating for the effects of membership on levels of euroscepticism is not possible for Italy, which had already been a full EC member for 15 years when the Eurobarometer polls were launched. In the 1980s, after over two decades of membership, Italy had exceptionally low levels of both opposition to membership and scepticism regarding benefits (although the latter could sometimes be up to four or five times as high as the former). Of the three 1980s EC entrants, it is not really possible to discern an accession impact in Spain and Portugal, because the overall level of opposition to membership was so low, remaining in single figures throughout the period. The only exception was a brief ‘pre-membership blip’ in Portugal in 1984–85, when opposition to membership reached a high point of 12 per cent (still just below the EC average at the time) before reverting to single figures immediately after EC entry. In Greece, however, membership opposition fell from 22 per cent immediately after EC entry to single figures from autumn 1988, seven years later. Benefits scepticism registered a dramatic post-accession drop in all three countries: in Greece, from one-third of respondents in 1983 to below 15% in spring 1988; in Portugal, again from one-third of the sample in 1986 to single figures by 1991; and in Spain—even more striking—from around two-thirds of respondents in 1986 to around one-quarter six years later.

Public Opinion post-Maastricht

In the new EU launched at Maastricht, the levels of support for integration registered by the Eurobarometer immediately began to drop. In 1996 a psychological barrier was breached, when those in the EU describing membership as ‘a good thing’ became a minority for the first time since polling began in 1973. Throughout 1996–2003 the declared supporters of membership hovered around the crucial 50 per cent mark (ranging from 46 to 53 per cent), while those who believed their country had benefited from membership were a minority (below 50 per cent) in all but three surveys. However, turning from support for integration to declared opposition, our two indices of euroscepticism showed only a limited rise. Moreover, comparing the pre- and post-Maastricht figures suggests that actually it was the late 1980s, with their very low levels of euroscepticism, that marked a deviation from the trend. To some extent, it might be argued that the 1990s saw continuity with the period before the exceptional years at the end of the previous decade, when integration had acquired a new momentum after the 1980s deepening and the Cold War appeared to be coming to an end.

Essentially, the same picture applies to Southern Europe. As the figures show, levels of euroscepticism in Southern Europe showed a moderate rise but generally to around the levels of the early 1980s. Average levels of euroscepticism mostly remained below the EU average. However, there were some significant variations among the SE countries. Particularly for the earlier part of this period, there continued to be higher levels of euroscepticism in Spain than in the other three countries. In contrast, Greece, where opposition to membership had been high in the early 1980s, now had singularly low levels of euroscepticism.

In three of our four countries, in the 23 EB surveys conducted during this period, opposition to membership remained in single figures in two-thirds of the surveys in Portugal (where the highest level of euroscepticism was 12 per cent), all but three in the case of Greece (with a maximum level of 14 per cent), and all but two in Italy (with a high point of 13 per cent). However, in Spain, where opposition to membership had previously been low, it jumped to double figures immediately after Maastricht and stayed there for all but one of the next ten surveys, with a high point of 18 per cent. It is thus the Spanish and. to some extent, the Portuguese cases that explain the brief point in the early 1990s, shown in Figure 2a, when membership opposition in SE was higher than the EU average.

Figure 2a Membership Euroscepticism: Southern Europe and the EU 1992-2004.

Benefits scepticism remained considerably above the EU average in Spain until 1996 (see Figure 2b), when it dropped below the EU average to converge with the rest of Southern Europe. At the same time, benefits euroscepticism was rising in Italy. While Italian benefits sceptics had never exceeded 20 per cent of the sample in the 1980s, they did so fairly consistently throughout 1992–2003. This still left Italy significantly below the EU average, apart from at the very end of this period. (The Italian case explains the sudden convergence of the EU and SE averages after 2002.) This rise in Italian benefits scepticism did not translate into increased opposition to membership: throughout this period, Italian membership opponents remained consistently around half the EU average. What it suggests, therefore, is a rise in soft euroscepticism whose emergence parallels the first appearance of euroscepticism among Italian parliamentary elites.

Figure 2b Benefits Euroscepticism: Southern Europe and EU 1992-2004.

Public Opinion post-Enlargement

In the enlarged EU of 25/27 following the 2004 and 2007 enlargements, membership supporters and those believing their country had benefited from membership were consistently a majority, although membership support did not approach the levels of the late 1980s. Both euroscepticism indicators remained around the levels of the previous period, with membership opponents in the EU never exceeding one-sixth of the total and benefits sceptics around one-third. In Southern Europe, following enlargement to Cyprus and Malta in 2004 and the opening of Turkey's accession negotiations the following year, a differentiation appeared between the ‘old’ and ‘new’ countries.

In the four existing member-states, the average number of membership opponents was consistently below the EU average, although the gap tended to be closer than in the past (see Figure 3a). Among the ‘old’ SE-4, however, the number of Italian membership opponents, insignificant in the past, had more than doubled in comparison with the preceding period, to reach the EU average. A similar picture applies to scepticism concerning benefits: Greece, Portugal and Spain all remained consistently below the EU average (except for one survey in the case of Greece). In Italy, however, the level of benefits scepticism was now slightly above the EU average. It would seem that public opinion in the largest SE member state, once regarded as particularly resistant to euroscepticism, was now more closely aligned with the EU average than with the other ‘old’ SE countries.

Figure 3a Membership Euroscepticism : “Old” and “New” Southern Europe, 2004-2009.

In the ‘new’ Southern Europe, the average level of opposition to membership was six to 10 percentage points higher than the average for the ‘old’ SE-4, and four to five percentage points higher than the EU average (see Figure 3a). Cyprus had a strikingly high proportion of benefits sceptics, who, for the first three years after accession, numbered around 50 per cent. This differentiation may well be linked to the fact that in Cyprus accession had been promoted as a route to the solution of the Cyprus problem—which, however, failed to happen. In contrast, benefits scepticism in Malta and Turkey was slightly below the EU average and slightly above the average for the ‘old’ SE-4 .

Figure 3b Benefits Euroscepticism in ‘Old’ Southern Europe, 2004-2009

In Malta, membership opponents in the early post-accession years were less than 20 per cent—considerably less than half the proportion of the electorate who voted against EU entry in the 2003 referendum. This was also several percentage points below Greece, our other case of a country divided over accession, at a similar point in its integration history. Malta also seems to have registered a mild downward trend in popular euroscepticism post-accession, which, however, was reversed in 2008 and does not seem to apply to Cyprus. In any case, the few years since EU entry may be too short a period to decide whether there has been a ‘membership effect’ reducing opposition to integration in these two countries. In Turkey, there was a significant increase in euroscepticism in the spring 2006 survey, the first after the official opening of membership negotiations. Although levels of euroscepticism subsequently showed some fluctuations, they did not drop back to the levels of 2004–05, confirming the expectation that the demanding accession process is likely to nourish euroscepticism in candidate countries.

Figure 3c Benefits Euroscepticism: ‘New’ Southern Europe and EU 2004-2009.

Referenda

In Southern Europe there has been minimal popular consultation on issues of European integration. In all, three referenda have been held, all resulting in clear defeats for euroscepticism. Of the 14 EU member-states to have held membership referenda, the sole SE example was Malta. (Meanwhile, Cyprus was the only one of the ten 2004 entrants not to consult its people.) The Maltese referendum of March 2003 and its sequel—the following month's general election, which to a considerable extent functioned as a second vote on the EU—marked a national turning point (see Pace in this issue). As was to be expected, given the Maltese party duopoly in which one party was pro-integrationist and the other eurosceptic, the referendum campaign was highly polarised. The result was a 46.4 per cent ‘yes’ vote on a 91 per cent turnout, the only one of the nine EU entry referenda that took place in 2003 in which less than two-thirds of votes were cast in favour. Because the official opposition did not accept the outcome, a general election was called, with the socialists promising that in the event of victory they would hold a new referendum to choose between EU membership and ‘partnership’. Following the six per cent lead of the ‘yes’ side in the referendum, the socialists’ election defeat by a four per cent margin—a landslide by Maltese standards (Fenech 2003)—definitively closed the decades-long debate on whether the country should participate in the EC/EU. This was thus the only one of the three South European referenda that had a significant impact, determining the contested question of Maltese participation in integration.

Of the 15 referenda held by member-states to decide on European Treaty ratification, again only one took place in Southern Europe. The Spanish European Constitutional Treaty referendum in 2005 produced a vote of 76 per cent in favour, 17 per cent against and six per cent blank/invalid ballots. Although this sounds like an overwhelming vote of confidence, participation was only 42.3 per cent—‘the lowest turnout figure of any of the 22 electoral consultations in recent democratic history’ (Torreblanca 2005, p. 2)—meaning that less than one-third of the electorate actually voted in favour. While the openly eurosceptic vote, i.e. those voting against the Treaty, thus amounted to only seven per cent of the total electorate, the real winner seems to have been ’Euro-indifference‘, i.e. the non-voters, with 57.7 per cent. Following in the wake of the French and Dutch ‘no’ votes, this referendum had no consequences either for the Constitutional Treaty, which was dropped and subsequently replaced by the less ambitious Lisbon Treaty, or for integration more generally.

In Portugal, a planned Constitutional Treaty referendum was cancelled after the French and Dutch voted against and the idea of a Lisbon Treaty vote was dropped following pressure from other EU members (The Times 2008). As a result, four of our South European countries—Greece, Portugal, Cyprus and Turkey—have never held a referendum on European integration. This is not particularly surprising in the case of non-member Turkey, or of Greece, where there has been no tradition of referenda since 1974. Perhaps more surprising is that Italy, a country that has made systematic use of referenda, holding 19 since 1974 on topics ranging from constitutional reform to artificial insemination or hunting, has only once consulted its people on European integration. This was the consultative vote on conferring a treaty-drafting mandate on the European Parliament, already mentioned. The referendum took place in 1979, simultaneously with the first direct elections to the EP, and resulted in an 88 per cent vote in favour (Preda 1991). This had no practical result as the Italian initiative found no response in the rest of the European Community. However, that what appeared to many non-Italians as a rather radical pro-federalist project was opposed by less than 12 per cent of those voting suggests that at this point euroscepticism had very little appeal to Italian public opinion.

Conclusions: Towards a New ‘Constraining Dissensus’ in Southern Europe?

Our diachronic study reveals that euroscepticism in Southern Europe has been less marginal than expected. In terms of electoral strength and government participation, in specific periods in several of our countries it has sometimes appeared mainstream. This suggests that the view of Southern Europe as a region of traditional pro-integrationist consensus needs to be somewhat nuanced in light of the evidence. This view refers largely to a period in the second half of the 1980s—a period when both party and popular euroscepticism reached their nadir but which constituted an exception rather than the rule.

In fact, the crucial choice of accession encountered substantive party euroscepticism, not only in the ‘new’ SE but also in two of the four ‘old’ SE states. In terms of public opinion, hard euroscepticism measured by opposition to membership was limited, reaching its highpoint with over one-fifth of Greeks in the early post-accession years. However, the soft variety expressed by benefits scepticism has attracted support from significant numbers of South Europeans at different times—most spectacularly from around two-thirds of Spaniards in the mid-1980s, over one-third of Italians in the 2000s, and 40–50 per cent of Cypriots post-accession. Meanwhile, of our seven countries, there are only two—Spain and Portugal—where eurosceptic parties have not participated in power. In considering whether Southern Europe is moving towards a new ‘constraining dissensus’, it should thus be recognised that the idea of a South European historical consensus around European integration may be closer to myth than reality.

Turning to our working hypotheses, the ‘system-transforming’ nature of Maastricht is confirmed with regard to party euroscepticism. Pre-Maastricht party euroscepticism in Southern Europe, except in the rather exceptional case of Turkey, was of the hard variety and limited to the left. By the eve of Maastricht, it had essentially shrunk to just three parties: the Maltese socialists and Portuguese and Greek communists, with the Greek party softening its stance. Post-Maastricht party euroscepticism became less intense but also less marginal, largely mutating into a softer version but spreading across the political spectrum. While now encompassing a broad range of ideological positions, it generally remained confined to the flanks of the party system, outside the central core represented by the centre-left and centre-right. Contrary to what might be expected, however, this new eurosceptic pluralism was not reflected in significantly increased electoral strength. In ‘new’ SE, eurosceptic parties continued to gain the same kind of vote share as in the past, while in the ‘old’ SE there were no longer individual eurosceptic parties of any significant size.

In the case of public opinion, the picture is somewhat complicated by national eurosceptic surges at different times. In the 1980s, hard euroscepticism was pronounced in Greece and the soft variety in Spain. In the late 1990s, as soft euroscepticism declined in Spain it began to rise in Italy. The latter emerged in the 2000s as the country with the highest levels of soft euroscepticism and therefore one to watch for the future. However, with regard to the overall picture in SE, Maastricht does not appear to have been ‘system-transforming’ in terms of changing the picture of popular euroscepticism, suggesting that national variations may have more to do with national conditions than with the deepening of integration per se. As elsewhere, enlargement has led to the introduction of new eurosceptic publics, but in the case of SE its impact is likely to be very limited. This is because the two new members, Cyprus and Malta, are both micro-states, representing a tiny proportion of the total EU population.

Turning to our second hypothesis, there does indeed seem to be evidence for an ‘accession effect’. Of our six EU member-states, party euroscepticism declined rapidly and dramatically in Greece and Malta and over a longer period of a decade and a half in Italy. In Cyprus, the main eurosceptic party made a pre-emptive strike, softening its euroscepticism before EU entry, while in Spain and Portugal, where party euroscepticism was either non-existent or very limited, there was little scope for reduction. Popular euroscepticism fell after a few years of membership in Greece, Spain and Portugal, while it is probably still too early to tell with regard to Cyprus and Malta. Italy remains an unknown, due to the absence of pre-accession data. If the pattern in Southern Europe is repeated elsewhere, then it would that suggest average levels of euroscepticism within the EU might increase after each enlargement, due to the presence of new members, but then fall a few years later as they ‘acclimatise’ to EU membership. The likelihood of an ‘accession effect’ should therefore be borne in mind when assessing the ‘constraining dissensus’ within the EU as a whole.

To date, euroscepticism remains, as it always has been, relatively marginal in South European public opinion, with hard euroscepticism representing a small group, while party euroscepticism is now usually soft and expressed by opposition parties. Even if both party and popular euroscepticism were to become stronger than at present, they would impact on European integration only if they succeeded in influencing government. In terms of eurosceptics in power, Southern Europe's major contribution to a ‘constraining dissensus’ occurred with the Greek socialist administration of the early 1980s. This sole example of a single-party hard eurosceptic government in an SE member-state seems unlikely to be repeated, while in the case of a coalition cabinet, a eurosceptic participant is likely to be constrained by the other members. A new ‘constraining dissensus’ in Southern Europe is therefore not currently visible and its emergence does not seem very likely on the basis of the trends discussed in this article. Nevertheless, Southern Europe is now at the heart of the European economic crisis and, as mentioned above, this has opened a new era in terms of the regions relationship with the EU. It remains to be seen how this will impact on South European euroscepticism in the future.

Acknowledgements

This collective volume began life as a panel convened by the author at the conference of the ECPR Standing Group on European Union Politics at Istanbul Bilgi University in 2006. The author would like to thank Gary Marks for kindly chairing the panel and Liesbet Hooghe and Anne Faber for their useful comments.

A substantial part of this paper was written during an all-too-brief stay as a Visiting Fellow at the Hellenic Observatory, London School of Economics. The author would like to thank the HO and particularly its Director, Kevin Featherstone, for offering her the necessary time and space to complete this project.

Notes

[1] Although, for example, the collection edited by Hooghe & Marks (2007b) investigates the links between the two.

[2] Possible exceptions—such as the Italian socialists’ initial opposition to the European Monetary System in the 1970s, mentioned by Quaglia in this issue—seem to have been short-lived and did not translate into constraining votes at crucial moments.

[3] The far left (TIP), which gained 14 parliamentary seats in 1965, reduced to two in 1969, was dissolved by the regime in 1971. Although subsequently re-founded, it did not re-enter parliament.

[4] The Greek socialists were returned to power in 1985, but by their second term had moved away from euroscepticism.

[5] On the problems in the functioning of the Association, see Arikan, 2006, chapter 3, including p. 68 on the Turkish governments’ demands.

[6] The main ‘eurosceptic’ stand of the second Berlusconi government (2001–06) was the European arrest warrant—but this seemed motivated more by concern with the Prime Minister's own judicial adventures than by euroscepticism.

[7] They are, however, included in the Italian case study by Quaglia in this issue.

[8] The Lega also participated in an earlier coalition government in 1994–95, but at this point was not a eurosceptic party.

[9] In Portugal, although the CDS/PP has been a coalition government participant (in 1980–83 and 2002–05), its periods in power did not coincide with its short-lived soft eurosceptic shift in the early 1990s.