Scholarly article on topic 'Identity, Instrumental Self-Interest and Institutional Evaluations: Explaining Public Opinion on Common European Policies in Foreign Affairs and Defence'

Identity, Instrumental Self-Interest and Institutional Evaluations: Explaining Public Opinion on Common European Policies in Foreign Affairs and Defence Academic research paper on "Political Science"

0
0
Share paper
Academic journal
European Union Politics
OECD Field of science
Keywords
{""}

Academic research paper on topic "Identity, Instrumental Self-Interest and Institutional Evaluations: Explaining Public Opinion on Common European Policies in Foreign Affairs and Defence"

Tv\p

European Union Politics

DOI: 10.1177/1465116507085955 Volume 9 (1): 5-29 Copyright© 2008 SAGE Publications Los Angeles, London, New Delhi and Singapore

KEY WORDS

■ common defence and security policy

■ common foreign policy

■ European integration

■ public opinion

■ responsiveness

Identity, Instrumental Self-Interest and Institutional Evaluations

Explaining Public Opinion on Common European Policies in Foreign Affairs and Defence

^ Harald Schoen

Johannes Gutenberg University, Germany ABSTRACT

This paper addresses public opinion on common European policies in foreign affairs and defence. It proposes three models of support for common policies in these fields. Drawing on Eurobarometer data, the analysis shows that instrumental self-interest and territorial identities contribute considerably to explaining support for common foreign affairs and defence policies. Moreover, support for common policies is strongly driven by domain-specific evaluations of the European Union's performance. These findings suggest that popular support for common European policies in foreign affairs and defence has an experiential base. Thus, elites have an incentive to respond to public opinion when making policy decisions, so this analysis supports the liberal view in international relations rather than the realist account.

Introduction

After early ambitious attempts to establish common European policies in defence and foreign affairs failed, for a considerable period European integration in these domains advanced slowly, if at all. In the 1990s, however, integration proceeded at a considerably higher rate. The Maastricht Treaty established the Common Foreign and Security Policy (CFSP) as the second pillar of the European Union. In more recent years, with the establishment of a High Representative for the CFSP, decision-making was made more efficient, troops were assigned to the European Union (EU) and the EU began to engage in civilian and military crisis management (e.g. M. Smith, 2004). The Constitutional Treaty goes even further as it includes creating the position of a Foreign Affairs Minister of the EU, widening of the so-called 'Petersberg tasks', a mutual defence clause and a terrorism-related solidarity clause. Notwithstanding the Constitutional Treaty's failure in 2005, European integration has advanced in foreign affairs and defence and is likely to progress further in the future (e.g. Howorth, 2001; Ojanen, 2006).

Whereas political science has paid considerable attention to advances toward common European policies in foreign affairs and defence (e.g. H. Smith, 2002; K. Smith, 2003; M. Smith, 2004), only a few studies deal with public opinion. Lindberg and Scheingold (1970: 55-63) found that the permissive consensus among Europeans on European integration did not extend to foreign affairs and defence. Several more recent studies reported trends of public opinion on collective defence (Manigart and Marlier, 1993; Everts, 1995; Sinnott, 1997; Everts and Sinnott, 2002). Prior analyses also explored the location of foreign affairs and defence in the structure of citizens' preferences over EU policy (Sinnott, 2000; Gabel and Anderson, 2002; see, for party manifestoes, Gabel and Hix, 2002). Loveless and Rohrschneider (2006) showed that attitudes towards current defence institutions have a bearing on EU support in Eastern Europe. At the same time, only a few studies addressed the sources of Europeans' attitudes toward common foreign and security policy (e.g. Rattinger, 1996). In this line of research, Carrubba and Singh (2004) showed that Europeans have different goods in mind when thinking about a common defence policy, whereas Ray and Johnston (2007) explored Europeans' opinions about the EU's role in European defence as compared with NATO. In summary, though prior research has produced interesting results, we still do not know much about the sources of public opinion on common European policies in foreign affairs and defence.

Findings on these sources have considerable implications concerning the image of public opinion and the role it can play in policy-making. According to the traditional view (e.g. Lippmann, 1922; Almond, 1950), voters are not

well acquainted with policies as remote to daily life as defence and foreign affairs and they do not form genuine attitudes or give indifferent answers in surveys (Converse, 1964). As a result, public opinion is fickle and cannot serve as a reliable guide for policy-making (e.g. Slater, 1983). Moreover, because citizens do not follow events closely, they are not able to review elite behaviour in these policy domains. Consequently, political elites have no incentive to respond to public opinion in foreign affairs and defence, so that public preferences in this field do not matter for elite decision-making. However, this view, which is in line with realist accounts of international relations (Waltz, 1967: 14; Morgenthau, 1978: 558), has been challenged in recent decades. Revisionists put forward evidence suggesting that public opinion on foreign policy is not as ill informed and volatile as traditional accounts claim. Accordingly, public opinion on defence and foreign affairs possesses structure and responds reasonably to foreign policy events (e.g. Peffley and Hurwitz, 1992; Shapiro and Page, 1988; Isernia et al., 2002). Thus, elites have an incentive to respond to public preferences in this field. Indeed scholars have shown that public opinion plays a role in constraining elites' decisions on foreign policy and European integration (e.g. Carrubba, 2001; Sobel, 2001; Jacobs and Page, 2005). All in all, these findings back liberal strands of thought in international relations (e.g. Wendt, 1992).

Against this backdrop, this paper aims to shed light on the sources of European public opinion on common foreign and security policy. It starts by discussing competing models of support for common European policies in the foreign and defence domain. Drawing on Eurobarometer data, it examines the explanatory power of the three models. The paper concludes by discussing the findings and elaborating on some implications.

Theorizing support for common foreign and defence policies

Prior research has rarely addressed citizens' attitudes toward common European foreign and defence policies. This analysis thus cannot adopt a tailor-made model of support for common European policies in this area. However, establishing common policies clearly implies advancing European integration, and citizens' attitudes toward common defence and foreign policy appear to be related to general preferences on European policy (Gabel and Anderson, 2002). Therefore, factors that previous research has shown to shape citizens' attitudes toward European integration should go a considerable way toward explaining support for or opposition to common policies in defence and foreign affairs. As a consequence, this analysis starts by

outlining three models of support for European integration that are presumed to be applicable to explaining levels of support.

Utilitarian models of support for European integration build on the assumption that calculations of costs and benefits drive attitudes toward the European Union. Accordingly, these models of EU support claim that citizens rely on egocentric or sociotropic assessments of the (putative) consequences of European integration when forming opinions about it (e.g. Gabel, 1998a, b). Previous studies showed that economic evaluations play a role in shaping attitudes toward European integration, which for a long time chiefly dealt with economics (e.g. Eichenberg and Dalton, 1993; Anderson, 1998; Gabel, 1998b, c; Hooghe and Marks, 2005). As regards public attitudes toward foreign affairs and defence policy, putative consequences for the economic well-being of Europeans are presumably less influential than utilitarian evaluations of effects on other goods such as national security (see Carrubba and Singh, 2004).

National security benefits from common European policies depend on several factors. For one thing, nations with military capabilities to protect national security will benefit less from common policies than countries possessing small military forces. For another thing, nations actually facing threats to national security will benefit comparatively more from European integration in this domain (Carrubba and Singh, 2004; see Hutchison and Gibler, 2007). Accordingly, support for common foreign and defence policies should be increased by (perceived) threats to national security and it should be decreased by existing military capabilities.

The performance model builds on the presumption that evaluations of how the EU institutions perform have a bearing on support for European integration and common policies. For instance, Rohrschneider (2002) shows that concerns about the democratic process at the EU level decrease EU support considerably. Likewise, Karp et al. (2003) find that concerns about the EU's ability to be responsive to citizen demands and a lack of confidence in the EU institutions increase dissatisfaction with the European Union. Concerns about the EU's current institutional performance also have been shown to affect public support for EU enlargement (Karp and Bowler, 2006). In this view, citizens consider the EU institutions' performance in managing current affairs and rely on these evaluations when forming attitudes toward the EU and specific EU policies. The performance model thus predicts that favourable evaluations of the performance of EU institutions increase support for common policies in defence and foreign affairs.

Citizens' evaluations of the EU's performance can be general or domain specific. General evaluations are likely to affect attitudes toward European integration regardless of the policy domain. By contrast, domain-specific

appraisals respond to policy content. Evaluations of the EU's economic performance, for example, should exhibit large effects on attitudes toward further economic integration, whereas their effects on opinions about European foreign and defence policy should be minimal. Following this line of reasoning, support for common foreign and defence policies should be increased by favourable evaluations of the EU institutions' performance. This model also predicts that satisfaction with the EU's performance in this policy area exhibits particularly strong effects on popular attitudes toward European integration in foreign affairs and defence.

An influential strand of thought deals with territorial identities as factors influencing attitudes toward European integration. Identity is a complex and multidimensional concept (e.g. Risse and Engelmann-Martin, 2002). At the individual level, it is warranted to distinguish two dimensions. Whereas the intensity dimension captures the strength of the loyalty felt by an individual, the content dimension refers to the meaning of the group to an individual member, including goals, norms, interests and cognitive models (e.g. Abdelal et al., 2006).1 Identifying with a collective is powerful in shaping political perception, attitudes toward political objects and political behaviour (e.g. Sniderman et al., 2004). More specifically, attitudes toward European integration have been found to be affected by territorial identities (e.g. Carey, 2002; Hooghe and Marks, 2005).2

Following this line of reasoning, identifying with Europe should affect support for common foreign and defence policies, with the effect of strong European attachments depending on the subjective meaning of Europe. European identifiers believing in the idea of a European free-trade zone are not particularly likely to back a common foreign and defence policy. By contrast, citizens subscribing to the notion of Europe as a unitary actor in international politics will endorse integration in foreign affairs and defence. At the same time, given a specific content, the effect of European identity should increase with the intensity of European attachments. Still, there is another possibility. Because the EU appears to have achieved hegemony in the civic dimension of European identity (e.g. Bruter, 2003), identifying with Europe might imply agreeing on the goal of forming a unitary European actor. If European identifiers take this meaning of Europe for granted, strong European loyalties in themselves will increase support for transferring authority in the foreign and defence domain to the EU.

As with European attachments, the effect of strong national attachments on support for common European foreign and defence policy is conditioned by the meaning of the nation to its members. Previous research has demonstrated that a wide variety of conceptions of national identity exists (e.g. Marcussen et al., 1999). As regards creating a common foreign and defence

policy, those components of national identity that refer to statehood and the nation's role in international relations are particularly salient. The principle of neutrality in international affairs is a case in point (e.g. Kovacs and Wodak, 2003). Because a common European foreign and defence policy might lead to involvement in international conflict, citizens whose notion of national identity encompasses the principle of neutrality should refuse the creation of a common foreign and defence policy. Moreover, pooling authority at the European level implies curtailing national governments' ability to pursue distinct courses of action. Citizens subscribing to the principles of national sovereignty, national self-determination and non-interference in internal affairs thus should be particularly sceptical about transferring sovereign rights to supranational institutions. As with European identity, these effects should increase with the intensity of national attachments. Because national identities in Europe are closely linked to the idea of the nation state (e.g. Bloom, 1990), national identifiers may quite naturally subscribe to the notion of national sovereignty and self-determination. If this presumption holds, the intensity of national attachments by itself should decrease support for common policies.

Though seemingly paradoxical, a common foreign and defence policy could also be thought of as empowering, rather than weakening, the nationstate in maintaining self-determination and sovereignty (see Risse, 2003). In this vein, a common European foreign and defence policy might be a means for the nation-state to deal with external threats. As a consequence, the effect on support for European integration in foreign affairs and defence of national identity focusing on statehood and self-determination should be conditioned by the occurrence of external threats. Given external threats, attachments to the nation-state should be less effective in decreasing support for common foreign and defence policy.

The three models of popular support differ considerably in their implications for the image of public opinion about common European foreign and defence policies. According to the identity model, citizens ask whether European integration in this domain is in line with their notion of national or European identity, including goals, norms and values. Since identities are incorporated in institutions, symbols and cultural understandings (e.g. Oakes et al., 1994; Risse, 2003), they are, although potentially variable, quite stable over time. As a result, this model suggests that attitudes toward European integration in foreign affairs and defence are also not volatile. The utilitarian model suggests that citizens consider the putative consequences of policy decisions when forming attitudes toward the CFSP. Thus, they are assumed to be prospective instrumental voters. The performance model, in contrast, maintains that public attitudes toward European integration in foreign affairs

and defence are driven by evaluations of the EU's general or domain-specific performance. For instance, citizens form evaluations of whether political elites act in accordance with public preferences on common defence and foreign policies and rely on these assessments when making up their mind about further integration in this field. Accordingly, citizens are retrospective voters and public support has an experiential base. Thus, if the performance model applies, political elites have a strong incentive to respond to public opinion on foreign affairs and defence. However, the validity of this model does not guarantee policy responsiveness because elites might disregard or mis-perceive public opinion (e.g. Powlick, 1991; Kull and Ramsay, 2000).3

Data, measures and methods

The data for my analysis come from the Eurobarometer 62, a survey among Europeans conducted on behalf of the European Commission in November and December 2004. Respondents were randomly selected using sample points, random routes and Kish selection grids and were interviewed personally, using Computer Aided Personal Interviewing (CAPI) if available.

Attitudes toward common foreign and defence policy are measured by six variables. Two items address support for one common foreign policy and for one common defence and security policy at a rather general level. The remaining items tap more specific issues. To begin with, respondents were asked whether the EU should have its own Foreign Minister as proposed in the Constitutional Treaty (Article I-28). Another item taps the idea that the European Union should have its own seat on the United Nations Security Council, which had been proposed by pundits and several European politicians such as Commissioner for External Affairs Benita Ferrero-Waldner (e.g. Sylvester, 2004; Stelzenmüller, 2005). Whereas these two items refer to institutional issues, the remaining items concern the EU's handling of international crises. The first question concerns whether member states of the European Union should agree on a common position when an international crisis occurs. The second item concerns a European military rapid response force to be deployed when an international crisis occurs, which the EU proposed for the first time in the wake of the Kosovo crisis in 1999 (e.g. M. Smith, 2004: 233).

Single-item measures are problematic for several reasons. I thus explored whether the six items could be combined into one or two factors. Factor analysis suggests that the two general items form one factor, as do the four specific items. Because the results of the substantive analysis vary considerably across items, however, I report the results for the two factors in the

appendix4 and chiefly deal with the six single items in the remainder of this paper.

To examine how well the three models perform in explaining support for these six proposals, I run logistic regression models. The utilitarian model predicts that support for the CFSP depends on threats to national security and existing capabilities to protect national security. To test this model, I include a dummy variable that equals 1 for countries that have experienced military interference in internal affairs since World War II (threat). I expect a positive effect of threat on support for the CFSP. At the same time, I anticipate respondents living in countries possessing nuclear weapons to be less supportive of the CFSP (nuclear weapons) because these military capabilities reduce the benefit from a common European foreign and defence policy.5 These and other variables in my analysis are detailed in the appendix.

As regards the performance model, I include evaluations of the performance of EU institutions and satisfaction with democracy at the EU level (EU institutions, EU democracy). To explore the predictive power of domain-specific assessments, I include evaluations of the role the European Union plays in defence and foreign affairs in the respondent's country (EU defence, EU foreign policy). Moreover, favourable evaluations of the EU's role in global matters such as peace in the world, the fight against terrorism and protection of the environment should increase support for common European policies in the foreign and defence domain (EU global affairs). Finally, the analysis includes evaluations of the impact of EU membership on personal security (security). These variables are plagued with endogeneity problems because evaluations of EU performance are likely to be affected by generalized preferences over transferring authority to the supranational level. To deal with this problem, I use the residuals resulting from regressions of the original variables on the general preference for creating a European political union and for more EU competencies. Using 'purged' measures leads to conservative estimates that capture the effect of performance evaluations presumably more adequately than analyses using original variables.

Because the Eurobarometer data are rather ill suited to represent multidimensional territorial identities, the analysis chiefly employs second-best indicators. One captures how strongly the respondent feels attached to her nation (national attachment) and another one covers the strength of her European attachment (European attachment). Turning to the content of national identity, one indicator captures whether the respondent exclusively identifies with her nation, indicating a strong attachment to the ideas of statehood and national sovereignty (exclusively national identity). Because these measures of the intensity and the content dimension suffer from similar problems to the indicators discussed in the previous paragraph, they were purged using the

procedure described above. Another dummy variable measures whether a respondent lives in a neutral country (neutrality). To test whether the effects of content of national identity are conditioned by the intensity of national attachments, I also created two multiplicative terms (national attachment X neutrality; national attachment X exclusively national identity). The substantive dimension of European identity is captured by a dummy variable that equals 1 for inhabitants of the original six EC member states (original six), thereby assuming that these respondents are particularly likely to subscribe to the notion of Europe as a unified actor. As with national identities, I multiply the content dimension by the intensity dimension of European identity to explore interaction effects (European attachment X political union). Finally, to test whether the effect of the belief in national sovereignty responds to territorial threats both variables are multiplied (exclusively national identity X threat).

In addition to these variables I include several controls. Along with age and gender I employ formal education. Moreover, because left-right ideology has been shown to affect attitudes toward European integration and foreign policy, with people at the right-hand end of the ideological spectrum being sceptical about European integration (e.g. McLaren, 2002) and disproportionately favourable to the use of military force (e.g. Juhasz, 2001; Holsti, 2004: 191-6), I include a relevant variable.

The survey data were gathered from individuals living in 25 countries and are thus multi-level data. Using conventional methods of estimation would increase the probability of significance tests passing conventional levels. To deal with this problem, I estimate models using robust standard errors clustered by country. This method leads to more consistent standard errors even when some assumptions concerning variance are not valid.

As in previous studies of public opinion using Eurobarometer data, missing values are a considerable problem in this analysis.6 By and large, prior research has chosen one of two avenues to deal with missing values (e.g. Allison, 2002). In several studies, item non-response is treated using a listwise deletion option. Though often leading to acceptable results, in this analysis it is not applicable because in the data set on which my analysis draws item non-response cumulatively results in a tremendous reduction in cases. Another strategy replaces missing values by some reasonable guesses. For example, missing values on dichotomous variables are substituted by the mean of the variable's two values or by the mean for those respondents with valid values on the variable of interest. This strategy results in biased results (e.g. Little, 1992), so it is not employed. Rather, to deal with the non-response problem, this analysis employs multiple imputation; i.e. missing values are replaced with substitutes that are not set by the researcher but are estimated based on the distribution of the given data using an iterative procedure. To

do this I use MICE ('multiple imputation by chained equations') (Van Buuren et al., 1999) as implemented in Stata (Royston, 2004, 2005).7

Results

As the results in Table 1 indicate, by and large European citizens strongly approve of common policies in foreign affairs and defence. Roughly 84% support pursuing a common defence and security policy, and 77% agree that EU member countries should pursue one common foreign policy. Turning from general aims to more specific policies with institutional implications, about 82% support the idea of the EU possessing a seat on the United Nations Security Council, though this notion is not very popular with political elites. At the same time, 77% support the proposal to create an EU foreign minister, implying that a strong majority of European citizens regrets the failure of the Constitution when it comes to this policy proposal. As regards the handling of international crises, EU citizens endorse common action. Nine out of ten respondents support the proposal that member states should agree on a common position in international crises. However, EU citizens are somewhat less supportive of the idea of forming a rapid military reaction force to be deployed when an international crisis occurs. This difference might indicate that some Europeans subscribe to the idea of common reactions to international crises but do not want to use military force when a crisis occurs.

In sum, the six items are supported by large majorities of the European public. Thus, these descriptive results affirm prior research showing that there is considerable support for policy measures that make the EU a stronger actor in foreign affairs and defence (see e.g. Everts and Sinnott, 2002). By the same token, by being reluctant to advance European integration in foreign affairs and defence as well as in other fields such as economic regulation, political elites appear to be at odds with European public opinion. However, looking at the marginal distributions of popular preferences does not tell us whether

Table 1 Support for common European policies in foreign affairs and defence

Common Common Seat on Rapid

foreign defence UN Foreign Crisis military

policy policy Council minister management force

Support (%) 77 84 82 77 90 78

Note: N = 24,770.

political elites have an incentive to respond to public opinion on European policies in foreign affairs and defence. To answer this question I now analyse the sources of support for policies in this domain.

Tables 2 and 3 report the results of logistic regressions of support for common European policies in the foreign affairs and defence domain. To examine the effects of the predictor variables in more detail, I also calculate, predicted probabilities of supporting policy proposals from the models reported in Table 2, while setting all other variables to their mode, median and mean, respectively. Table 4 reports the differences in the probability of supporting the policy proposals at the high end and at the low end of the independent variables or, for non-dichotomous predictors, two standard deviations above and below the mean.

As the low values for pseudo-R2 in Tables 2 and 3 indicate, the variables included in the models explain only a small to moderate portion of the variation in Europeans' attitudes toward EU foreign and defence policies.8 In light of the rather low variance in the dependent variables and the 'purging' of independent variables, this finding comes as no surprise. Moreover, the independent variables are somewhat better in explaining opinions on the rather general issues of establishing common defence and foreign policies than attitudes toward more specific proposals. Interestingly, results for policies pursued by the EU, e.g. the rapid military reaction force, do not differ substantially from those for proposals that EU institutions do not officially support, e.g. an EU seat on the UN Security Council.

Utilitarian considerations play a considerable role in influencing attitudes toward European integration in this domain. In line with my expectations, being a citizen of a nation possessing nuclear weapons decreases support for a common foreign policy and a common defence policy. According to the results in Table 4, the effects are quite sizeable. In particular, respondents inhabiting countries with nuclear weapons are, ceteris paribus, by roughly 16 points more likely to approve of a common foreign policy than are other respondents. When it comes to more specific policy proposals, however, the effects are considerably smaller. Surprisingly, nuclear weapons increase, rather than decrease, support for creating a rapid military force. This finding suggests that citizens of nuclear powers deem a European rapid military force a suitable supplement to their nation's capabilities of pursuing the national interest.

The results also show that threats to national security are conducive to support for European integration in foreign affairs and defence. However, the effects vary considerably across policy proposals. External threats considerably increase support for a common defence policy, coordinated policies in international crises and a rapid military force. By contrast, they do not affect

Table 2 Explaining attitudes toward six issues referring to common European foreign and defence policy (logistic regression)

Nuclear weapons Threat

European attachment

Original six

National attachment

Exclusively national identity

Neutrality

EU democracy

EU institutions

Security

EU defence

EU foreign policy

EU global affairs

Common Common Foreign

foreign policy defence policy minister

-0.82** -0.64** -0.39

(0.15) (0.15) (0.22)

0.23 0.67** 0.10

(0.24) (0.16) (0.24)

0.59** 0.29 0.30

(0.20) (0.27) (0.23)

0.43** 0.54** 0.22

(0.16) (0.17) (0.20)

-0.19 -0.16 0.16

(0.25) (0.24) (0.28)

-0.52** -0.46** -0.19*

(0.09) (0.10) (0.09)

-0.69* -1.09** -0.86*

(0.33) (0.27) (0.32)

0.64** 0.28 0.09

(0.18) (0.38) (0.27)

0.39 0.51 0.43*

(0.20) (0.28) (0.21)

0.72** 0.43** 0.70*

(0.17) (0.16) (0.14)

0.30** 1.02** -0.07

(0.09) (0.15) (0.13)

0.66** 0.59** 0.35*

(0.16) (0.14) (0.11)

0.75** 0.66** 0.58*

(0.16) (0.16) (0.15)

Crisis Rapid military

UN seat management force

-0.56** -0.03 0.41*

(0.20) (0.20) (0.20)

0.01 0.62** 0.61*

(0.24) (0.19) (0.23)

0.01 -0.08 -0.35

(0.22) (0.29) (0.25)

0.22 0.46** 0.26

(0.15) (0.17) (0.16)

0.21 0.49 0.31

(0.23) (0.29) (0.22)

0.02 -0.16 -0.05

(0.11) (0.12) (0.09)

-0.72* -0.36* -0.03

(0.34) (0.17) (0.24)

0.06 -0.34 0.36

(0.19) (0.41) (0.19)

0.29 -0.08 0.31

(0.19) (0.23) (0.20)

0.56* 0.68** 0.67*

(0.27) (0.25) (0.19)

0.00 0.25 0.42*

(0.18) (0.14) (0.11)

0.50** 0.77** 0.24*

(0.16) (0.21) (0.12)

1.04** 1.25** 0.70*

(0.14) (0.17) (0.15)

Table 2 Continued

Gender Age

Education 1

Left-right ideology

Constant

Adj. Pseudo-fl2 N

Common Common Foreign Crisis Rapid military

foreign policy defence policy minister UN seat management force

-0.09 (0.10) 0.00 (0.00)

-0.21 (0.13) -0.28* (0.12) -0.06 (0.08) -0.33 (0.21) -0.26 (0.21) .09

24,770

0.07 (0.13) 0.00 (0.00)

-0.22 (0.16) -0.14 (0.15) 0.02 (0.12) -0.49* (0.21) 0.01 (0.29) .10

24,770

0.12 (0.11) -0.00 (0.00)

0.11 (0.16) 0.02 (0.19) 0.10 (0.08) -0.34* (0.17) -0.03 (0.24) .04

24,770

0.31** (0.09) -0.00 (0.00)

0.36 (0.19) 0.20 (0.19) 0.39** (0.09) -0.14 (0.20) -0.54* (0.26) .04

24,770

0.00 (0.11) 0.00 (0.00)

0.10 (0.16) 0.22 (0.18) 0.28** (0.09) -0.02 (0.17) -0.10 (0.44) .05

24,770

0.01 (0.09) -0.01** (0.00)

0.43** (0.07) 0.14 (0.12) 0.13 (0.08) 0.34* (0.13) -0.49* (0.20) .04

24,770

Notes'. Entries are unstandardized logit coefficients, with robust standard errors clustered by countries in parentheses.

* p< .05, ** p< .01.

Table 3 Interaction effects in explaining attitudes toward common European foreign and defence policy (logistic regression)

Common Common Foreign Crisis Rapid military

foreign policy defence policy minister UN seat management force

National attachment -0.04 -0.03 0.03 0.05 0.10 0.04

(0.05) (0.05) (0.06) (0.05) (0.06) (0.04)

Neutrality -0.22* -0.35** -0.28** -0.23* -0.12* -0.02

(0.10) (0.09) (0.11) (0.11) (0.06) (0.08)

Neutrality : : national attachment 0.00 0.01 -0.03 0.02 -0.03 -0.05*

(0.02) (0.02) (0.02) (0.02) (0.03) (0.02)

Exclusively national identity -0.19** -0.17** -0.08** -0.01 -0.06 -0.04

(0.03) (0.03) (0.03) (0.04) (0.05) (0.04)

Exclusively national identity 0.00 0.00 0.03 -0.00 0.04 -0.06*

: : national attachment (0.03) (0.03) (0.03) (0.03) (0.03) (0.02)

Threat 0.10 0.30** 0.04 0.01 0.28** 0.28*

(0.11) (0.08) (0.12) (0.12) (0.10) (0.11)

Exclusively national identity : : threat 0.07* 0.08** 0.07** 0.04 0.06 0.06*

(0.03) (0.03) (0.02) (0.03) (0.04) (0.03)

European attachment 0.10** 0.05 0.06 -0.02 -0.01 -0.06

(0.04) (0.06) (0.05) (0.04) (0.06) (0.05)

Original six 0.19** 0.23** 0.09 0.10 0.19* 0.10

(0.07) (0.08) (0.08) (0.06) (0.07) (0.07)

Original six : : European attachment 0.05* 0.02 -0.00 0.03 -0.02 -0.04

(0.02) (0.04) (0.03) (0.03) (0.04) (0.03)

Constant -0.28 0.01 0.14 -0.39 0.46 -0.17

(0.24) (0.26) (0.24) (0.34) (0.34) (0.22)

Adj. Pseudo-fl2 .09 .10 .04 .04 .05 .04

N 24,770 24,770 24,770 24,770 24,770 24,770

Notes: Entries are unstandardized logit coefficients, with robust standard errors clustered by countries in parentheses. The coefficients for the control variables (reported in Table 2) are not reported for the sake of clarity and brevity. Before creating multiplicative terms, the variables were standardized to avoid collinearity problems. As a result, the coefficients for the linear terms differ from those reported in Table 2. * p< .05, ** p< .01.

Table 4 The effect of predictor variables on the probability of supporting six policy measures (differences in probability)

Nuclear weapons Threat

European attachment

Original six

National attachment

Exclusively national identity

Neutrality

EU democracy

EU institutions

Security

EU defence

EU foreign policy

EU global affairs

Common Common Foreign

foreign policy defence minister

-15.8 -8.9 -6.9

3.3 5.8 1.5

7.3 2.5 3.8

5.8 4.9 3.2

-2.5 -1.5 2.1

-10.5 -6.7 -3.8

-12.9 -17.5 -17.0

7.3 2.3 1.1

5.1 4.8 5.6

10.7 4.6 10.5

5.0 12.5 -1.1

10.8 6.9 5.7

9.5 6.0 7.4

Crisis Rapid military

UN seat management force

-7.1 -0.3 7.1

0.1 4.2 10.0

0.1 -0.4 -5.4

2.2 3.4 4.8

1.8 3.8 4.9

0.3 -1.7 -1.2

-9.5 -3.5 -0.6

0.5 -2.4 -5.1

2.5 -0.5 5.1

5.5 6.0 12.2

0.1 2.4 8.8

5.4 7.1 4.8

8.7 9.1 10.9

Notes: Cell entries are differences in the probability of supporting the policy proposals at the high end and at the low end of the independent variables or, for non-dichotomous predictors, two standard deviations above and below the mean. Probabilities were gleaned from the models reported in Table 2 while setting all independent variables but the one noted in the first column were set to their mode and mean.

attitudes toward the three proposals relating to diplomacy rather than to the use of force. Accordingly, EU citizens appear to respond to issue content when forming attitudes toward European integration in foreign affairs and defence.

National identity plays a considerable role in shaping attitudes toward common foreign and defence policies at the EU level. EU citizens subscribing to the notion of national sovereignty, self-determination and noninterference in internal affairs are particularly hesitant about supporting a common foreign policy, a common defence policy and creating the position of an EU foreign minister. At the same time, the effect of this notion of the nation-state is only weakly conditioned by the intensity of national attachments, if at all (Table 3). Only the effect on support for a rapid military force gets stronger when moving from weak to strong national attachments. As a result, only among strong national identifiers does endorsing the principle of national self-determination decrease support for a rapid military force (b = -0.10; p = .006).

In contrast, the effect of a strict notion of national sovereignty and national self-determination on support for common European policies is considerably conditioned by external threats. According to the results in Table 3, in the absence of external threats EU citizens subscribing to the principle of national self-determination are particularly likely to reject integration in the foreign and defence domain. Given external threats, however, support for this norm does not affect attitudes toward European integration in foreign affairs and defence policy. This finding suggests that external threats make EU citizens more inclined to consider the common foreign and defence policy not only as a challenge to national sovereignty but also as a means to maintain it.

The analysis also reveals that subscribing to the norm of neutrality decreases support for common foreign and defence policy. As Table 4 shows, citizens of neutral countries are less likely, by about 18 points, to endorse a common defence policy than are other respondents. This finding is in line with the assumption that neutrality renders citizens particularly sceptical about involvement in military conflicts. Neutrality also plays a sizeable negative role in support for a common foreign policy and a European foreign minister. At the same time, it does not affect attitudes toward a rapid military force that could be sent to trouble-spots.

The results reported in Table 3 show that the effect of subscribing to neutrality is scarcely conditioned by the intensity of national attachments. The statistically significant coefficient for the interaction term in the analysis of support for a rapid military force hardly alters this conclusion because the effect of neutrality is indistinguishable from 0 irrespective of whether national attachments are very weak or strong. The intensity of national loyalties also

leaves support for European integration in this policy domain unaffected, as indicated by the statistically insignificant coefficients in Table 2. As a result, the analysis suggests that the content dimension of national identity is much more influential than the intensity dimension in shaping public support for European integration in foreign affairs and defence.

Similarly, European attachments are quite ineffective in shaping public attitudes toward European integration in this policy domain. Intense European loyalties considerably increase support only for a common foreign policy. As regards creating a rapid military force and, albeit less clearly, common reactions in international crises, European attachments even tend to decrease support. This result suggests that for some EU citizens the meaning of Europe encompasses the concept of a civilian power. At the same time, respondents subscribing to the idea that Europe should become a unitary actor in international politics are particularly likely to support common policies in foreign affairs and defence policy. As the results in Table 2 show, this effect is quite strong on attitudes toward the two general issues and the question of commonly managing an international crisis. Except for attitudes toward a common foreign policy, however, the effect of this substantive component of European identity is virtually unconditioned by the intensity of European loyalties (Table 3).

Europeans rely on evaluations of the existing European institutions when assessing proposals referring to European defence and foreign policy. They do not indiscriminately draw on retrospective assessments, however. Rather, general evaluations of institutional performance are considerably less effective in shaping attitudes toward common policies than are domain-specific assessments. The results reported in Table 2 show that satisfaction with EU democracy rarely affects attitudes toward common policies. Likewise, general performance evaluations are not powerful predictors of public support for European integration in this area.

Domain-specific evaluations of the EU are much more powerful in shaping public opinion on European policies in foreign affairs and defence. Favourable evaluations of the effects of EU membership on personal security considerably increase support for common policies. For example, respondents believing that EU membership contributes to their personal security are over 10 points more likely to endorse a common foreign policy and a European foreign minister than are persons not holding this belief. Assessments of the EU's impact on national defence policy affect support for pursuing one common defence policy, the rapid military force issue and, albeit less strongly, a common foreign policy, but they do not exhibit an effect on the remaining three proposals. As the predicted probabilities reported in Table 4 show, the effects of the defence-related assessments are quite sizeable: a positive opinion

about the EU's role regarding the national defence policy increases support for the rapid military force and one common security and defence policy by roughly 8 and 12 points, respectively. Likewise, evaluations of the EU's impact on national foreign policy exhibit effects on attitudes toward common European policies. As with retrospective defence evaluations, the effects of foreign policy assessments, albeit to a smaller extent, vary across policy proposals. In particular, they are more limited on support for a rapid military force than attitudes toward other issues such as a common foreign policy or commonly managing international crises. This pattern suggests that EU citizens respond to issue content when forming attitudes toward European integration in defence policy and foreign affairs.

Attitudes toward the EU's performance in global affairs consistently influence support for common European policies. As the results reported in Table 4 show, these effects are quite substantial. For example, a person with a good opinion about the role the EU plays in managing global problems is more likely by roughly 10 points to support a rapid military force, a common foreign policy, commonly managing international crises and an EU seat on the UN Security Council than a person who criticizes the EU's performance in global affairs. According to this pattern, these retrospective assessments appear to be particularly influential in affecting attitudes toward issues referring to the EU's role in world politics.9

The control variables play a minor role in directly shaping attitudes toward foreign and defence policy at the EU level. Gender affects only support for establishing an EU seat on the UN Council, with women being somewhat more inclined to endorse this proposal. With increasing age, support for the rapid military force declines somewhat. Education exhibits its largest effect on support for the rapid military force, with those at the low end of the educational ladder being most supportive of this measure. By and large, left-right ideology is not very effective in affecting attitudes toward European foreign and defence policy. However, moving to the right on the left-right dimension increases the likelihood of supporting the European rapid military force while decreasing support for a common defence policy and a European foreign minister. This finding is in line with prior research showing that a rightist view renders a person more likely to endorse the use of military force and to reject the transfer of sovereign rights to the supranational level (e.g. Rattinger, 1996; Juhasz, 2001).

Conclusion

This paper addresses public opinion on common European policies in foreign affairs and defence. It shows that Europeans are generally very supportive of

common European policies in these domains. Thus, political elites are somewhat at odds with public opinion, because in foreign affairs and defence EU member states are reluctant to cooperate or even to transfer authority to the supranational level. Showing that elite actions are not in line with public opinion does not imply that elites have an incentive to respond to popular preferences, however. As the traditional view of public opinion on foreign affairs and defence suggests, voters are not well acquainted with these policy domains so that they do not perceive elite behaviour and cannot review whether policy-making is in line with their policy preferences. To examine whether elites have an incentive for policy responsiveness, this paper studies the factors shaping support for common European policies in foreign affairs and defence. The findings suggest that utilitarian evaluations play a considerable role in influencing attitudes toward common European policies. National military capabilities decrease support for common European policies whereas external threats make EU citizens more inclined to endorse integration in this domain. National and European identities also affect these attitudes, with the content dimension being considerably more influential than the intensity dimension. For instance, citizens subscribing to the principles of neutrality and national self-determination are quite sceptical about common policies, whereas endorsing the notion of Europe as a unitary actor in international politics increases support for further integration in foreign affairs and defence policy. Accordingly, to some extent EU citizens ask whether European integration in this policy domain is in line with norms and values when forming attitudes toward it. Whereas general assessments of how EU institutions perform play a minor role, evaluations of the EU's performance in foreign affairs and defence are powerful in shaping support for common European policies in these policy domains. Thus, it appears that EU citizens form policy-specific evaluations of the EU's performance that in turn shape support for common European policies. As a result, citizens are able to examine whether elites act in accordance with popular policy preferences and to rely on these evaluations when forming attitudes toward common European policies. This analysis thus suggests that elites have a considerable incentive to respond to public opinion when making policy decisions.

It would be premature to conclude from this that elites are responsive to popular preferences, however. To begin with, politicians who are inclined to follow public opinion might misperceive the public's preferences. Furthermore, they might not perceive any incentive to be responsive because they might consider voters not to rely on evaluations of past performance when forming attitudes. Moreover, despite perceiving public opinion and its sources correctly, elites subscribing to the realist view of international relations might refuse to take public opinion into account. The institutional setting of the EU may also undermine the elites' inclination to respond to

public opinion in defence and foreign affairs. Compared with the national political arena, for voters it is more difficult to hold EU politicians accountable, so that the incentive for elite responsiveness might be too weak to be effective. Finally, elites could prefer to overcome problems arising from public opinion by strategically manipulating the information citizens rely on when forming evaluations of how the European Union performs in foreign affairs and defence. Regardless of these qualifications, however, this analysis shows that there is an incentive for elites to be responsive (see also Carrubba, 2001), thereby supporting the revisionist view of public opinion on these policy domains rather than the traditional view.

This analysis also points to fertile areas of future research. In the absence of better measures, it used many second-best indicators, notably single-item measures. Scholars could overcome this shortcoming by gathering data that are better suited to capture relevant concepts such as identity and perceived threats. As the small to moderate proportions of variance explained indicate, there is much room for additional factors to affect attitudes toward foreign and defence policy at the EU level. It appears worthwhile to include elite cues because elites interpret political events and thus play a crucial role in influencing the way ordinary citizens, most of whom have no first-hand experience of EU policies, perceive and evaluate elite behaviour. For example, the predispositions voters rely on when forming attitudes toward a specific policy may differ depending on public rhetoric. Future research on public opinion on common European policies in defence and foreign affairs could also benefit from adopting hierarchical models of attitude constraint. By exploring how core values and general preferences shape attitudes toward specific issues, scholars could shed light on the causal mechanisms underlying public opinion on European integration in foreign affairs and defence.

Notes -

I am very grateful to the anonymous reviewers and the editor of this journal for their very helpful comments and suggestions.

1 Accordingly, the utilitarian model is related to collective identities because the criteria people rely on when evaluating the comparative utility of different courses of action may be embedded in social identities.

2 The content of collective identities is not fixed and multifaceted, so that persons identifying with a collective can differ in the understanding of it and different components of identity can be given priority depending on the context (e.g. Risse, 2003). Prior studies on European integration hardly captured the complex nature of territorial identities. To be sure, several

studies have argued that the content of national identity is variable and subject to social construction, so that for some citizens national and European identities are compatible whereas for others they are competing (e.g. Diez Medrano and Guttierez, 2001; Citrin and Sides, 2004). Many empirical analyses of public opinion, however, have primarily dealt with the strength of territorial attachments, thereby assuming the content of territorial identities to be fixed.

3 Of course, the validity of this model is not a necessary condition for elites being responsive to public opinion because elites may feel obliged to respond to public opinion, regardless of its sources.

4 The results of the models with the two factors as dependent variables are reported in the appendix, which is available at http://www.uni-konstanz.de/ eup/issues.htm.

5 NATO membership could be considered another indicator, with non-members being more in need of additional military security (see Carrubba and Singh, 2004). Leaving aside collinearity problems, NATO membership reflects national identity at least partially, however, so that it is rather ill suited to test the utilitarian model.

6 In this respect, the items on foreign and defence policy dealt with in this present study differ only moderately, if at all, from items concerning other issues that were used in prior studies.

7 The results of the models with imputed values do not differ substantively from those of models without imputation (see the appendix).

8 Statistical tests confirmed that there were no serious collinearity problems.

9 Additional analyses show that this effect is particularly strong among respondents who became EU citizens in 2004.

References

Abdelal, Rawi, Yoshiko M. Herrera, Alastair Iain Johnston and Rose McDermott (2006) 'Identity as a Variable', Perspectives on Politics 4(4): 695-711.

Allison, Paul D. (2002) Missing Data. Thousand Oaks, CA: Sage.

Almond, Gabriel A. (1950) The American People and Foreign Policy. New York: Praeger.

Anderson, Christopher J. (1998) 'When in Doubt, Use Proxies. Attitudes toward Domestic Politics and Support for European Integration', Comparative Political Studies 31(5): 569-601.

Bloom, William (1990) Personal Identity, National Identity and International Relations. Cambridge: Cambridge University Press.

Bruter, Michael (2003) 'Winning Hearts and Minds for Europe: The Impact of News and Symbols on Civic and Cultural European Identity', Comparative Political Studies 36(10): 1148-79.

Carey, Sean (2002) 'Undivided Loyalties: Is National Identity an Obstacle to European Integration?', European Union Politics 3(4): 387-413.

Carrubba, Clifford J. (2001) 'The Electoral Connection in European Union Politics', Journal of Politics 63(1): 141-58.

Carrubba, Clifford J. and Anand Singh (2004) 'A Decision Theoretic Model of Public Opinion: Guns, Butter, and European Common Defense', American Journal of Political Science 48(2): 218-31.

Citrin, Jack and John Sides (2004) 'Can There Be Europe without Europeans? Problems of Identity in a Multinational Community', in Richard Herrmann, Marilynn Brewer and Thomas Risse (eds) Transnational Identities: Becoming European in the EU, pp. 161-85. Lanham, MD: Rowman & Littlefield.

Converse, Philip (1964) 'The Nature of Belief Systems in Mass Publics', in David Apter (ed.) Ideology and Discontent, pp. 206-61. New York: Free Press.

Diez Medrano, Juan and Paula Guttiérez (2001) 'Nested Identities: National and European Identity in Spain', Ethnic and Racial Studies 24(5): 753-78.

Eichenberg, Richard and Russell J. Dalton (1993) 'Europeans and the European Community: The Dynamics of Public Support for European Integration', International Organization 47(4): 507-34.

Everts, Philip (1995) 'NATO, the European Community, and the United Nations', in Oskar Niedermayer and Richard Sinnott (eds) Public Opinion and Internationalized Governance, Beliefs in Government Vol. II, pp. 402-28. Oxford: Oxford University Press.

Everts, Philip and Richard Sinnott (2002) 'Opinion publique, défense et intervention militaire', La revue internationale et stratégique, No. 48 : 147-58.

Gabel, Matthew J. (1998a) Interests and Integration. Ann Arbor: University of Michigan Press.

Gabel, Matthew J. (1998b) 'International Economics and Mass Politics: Market Liberalization and Public Support for European Integration', American Journal of Political Science 42(3): 936-53.

Gabel, Matthew J. (1998c) 'Public Support for European Integration: An Empirical Test of Five Theories', Journal of Politics 60(2): 333-54.

Gabel, Matthew J. and Christopher J. Anderson (2002) 'The Structure of Citizen Attitudes and the European Political Space', Comparative Political Studies 35(8): 893-913.

Gabel, Matthew J. and Simon Hix (2002) 'Defining the EU Political Space: An Empirical Study of the European Election Manifestoes, 1979-1999', Comparative Political Studies 35(8): 934-64.

Holsti, Ole R. (2004) Public Opinion and American Foreign Policy, rev. edn. Ann Arbor: University of Michigan Press.

Hooghe, Liesbet and Gary Marks (2005) 'Calculation, Community and Cues. Public Opinion on European Integration', European Union Politics 6(4): 419-43.

Howorth, Jolyon (2001) 'European Defence and the Changing Politics of the European Union: Hanging Together or Hanging Separately?', Journal of Common Market Studies 39(4): 765-89.

Hutchison, Marc L. and Douglas M. Gibler (2007) 'Political Tolerance and Territorial Threat: A Cross-National Study', Journal of Politics 69 (forthcoming).

Isernia, Pierangelo, Zoltân Juhâsz and Hans Rattinger (2002) 'Foreign Policy and the Rational Public in Comparative Perspective', Journal of Conflict Resolution 46(2): 201-24.

Jacobs, Lawrence R. and Benjamin I. Page (2005) 'Who Influences U.S. Foreign Policy?', American Political Science Review 99(1): 107-23.

Juhâsz, Zoltân (2001) 'German Public Opinion and the Use of Force in the Early

1990s', in Philip Everts and Pierangelo Isernia (eds) Public Opinion and the International Use of Force, pp. 57-85. London and New York: Routledge.

Karp, Jeffrey A. and Shaun Bowler (2006) 'Broadening and Deepening or Broadening versus Deepening: The Question of Enlargement and Europe's "Hesitant Europeans"', European Journal of Political Research 45: 369-90.

Karp, Jeffrey, Susan A. Banducci and Shaun Bowler (2003) 'To Know It Is to Love It? Satisfaction with Democracy in the European Union', Comparative Political Studies 36(3): 271-92.

Koväcs, Andräs and Ruth Wodak (eds) (2003) NATO, Neutrality, and National Identity. Vienna: Böhlau.

Kull, Steven and Clay Ramsay (2000) 'Elite Misperceptions of U.S. Public Opinion and Foreign Policy', in Brigitte L. Nacos, Robert Y. Shapiro and Pierangelo Isernia (eds) Decisionmaking in a Glass House. Mass Media, Public Opinion, and American and European Foreign Policy in the 21st Century, pp. 95-110. Lanham: Rowman & Littlefield.

Lindberg, Leon N. and Stuart A. Scheingold (1970) Europe's Would-Be Polity. Englewood Cliffs, NJ: Prentice-Hall.

Lippmann, Walter (1922) Public Opinion. New York: Macmillan.

Little, Roderick J.A. (1992) 'Regression with Missing X's: A Review', Journal of the American Statistical Association 87(420): 1227-37.

Loveless, Matthew and Robert Rohrschneider (2006) 'Attitudes toward European Integration and NATO in Hungary, Bulgaria, and Ukraine', in Robert Rohrschneider and Stephen Whitefield (eds) Public Opinion, Party Competition and the European Union in Post-Communist Europe. New York: Palgrave.

McLaren, Lauren M. (2002) 'Public Support for the European Union: Cost/Benefit Analysis or Perceived Cultural Threat?', Journal of Politics 64(2): 551-66.

Manigart, Philippe and Eric Marlier (1993) 'European Public Opinion on the Future of Its Security', Armed Forces and Society 19(3): 335-52.

Marcussen, Martin, Thomas Risse, Daniela Engelmann-Martin, Hans-Joachim Knopf and Klaus Roscher (1999) 'Constructing Europe: The Evolution of French, British, and German Nation-State Identities', Journal of European Public Policy 6(4): 614-33.

Morgenthau, Hans J. (1978) Politics among Nations. The Struggle for Power and Peace, 5th edn. New York: Alfred A. Knopf.

Oakes, Penelope J., S. Alexander Haslam and John C. Turner (1994) Stereotyping and Social Reality. Oxford: Oxford University Press.

Ojanen, Hanna (2006) 'The EU and Nato: Two Competing Models for Common Defence Policy', Journal of Common Market Studies 44(1): 57-76.

Peffley, Mark and Jon Hurwitz (1992) 'International Events and Foreign Policy Beliefs: Public Response to Changing Soviet-U.S. Relations', American Journal of Political Science 36(2): 431-61.

Powlick, Philip J. (1991) 'The Attitudinal Bases for Responsiveness to Public Opinion among American Foreign Policy Officials', Journal of Conflict Resolution 35(4): 611-41.

Rattinger, Hans (1996) 'Einstellungen zur europäischen Integration in der Bundesrepublik: Ein Kausalmodell', Zeitschrift für Internationale Beziehungen 3(1): 45-78.

Ray, Leonard and Gregory Johnston (2007) 'European Anti-Americanism and

Choices for a European Defense Policy', PS: Political Science & Politics 40(1): 85-91.

Risse, Thomas (2003) 'European Identity and the Heritage of National Cultures', in Robert Shannan Peckham (ed.) Rethinking Heritage. Cultures and Politics in Europe, pp. 74-89. London: Tauris.

Risse, Thomas, and Daniela Engelmann-Martin (2002) 'Identity Politics and European Integration: The Case of Germany', in Anthony Pagden (ed.) The Idea of Europe. From Antiquity to the European Union, pp. 287-316. Cambridge: Cambridge University Press.

Rohrschneider, Robert (2002) 'The Democracy Deficit and Mass Support for an EU-wide Government', American Journal of Political Science 46(2): 463-75.

Royston, Patrick (2004) 'Multiple Imputation of Missing Values', Stata Journal 4(3): 227-41.

Royston, Patrick (2005) 'Multiple Imputation of Missing Values: Update of Ice', Stata Journal 5(4): 527-36.

Shapiro, Robert Y. and Benjamin I. Page (1988) 'Foreign Policy and the Rational Public', Journal of Conflict Resolution 32(2): 211-47.

Sinnott, Richard (1997) European Public Opinion and Security Policy. Chaillot Paper 28. Alencon: Institute for Security Studies of WEU.

Sinnott, Richard (2000) 'Knowledge and the Position of Attitudes to a European Foreign Policy on the Real-to-Random Continuum', International Journal of Public Opinion Research 12(2): 113-37.

Slater, Martin (1983) 'Political Elites, Popular Indifference, and Community Building', in Loukas Tsoukalis (ed.) The European Community: Past, Present, and Future, pp. 69-86. Oxford: Basil Blackwell.

Smith, Hazel (2002) European Union Foreign Policy. What It Is and What It Does. London: Pluto Press.

Smith, Karen E. (2003) European Union Foreign Policy in a Changing World. Cambridge: Polity Press.

Smith, Michael E. (2004) Europe's Foreign and Security Policy. The Institutionalization of Cooperation. Cambridge: Cambridge University Press.

Sniderman, Paul M., Louk Hagendoorn and Markus Prior (2004) 'Predisposing Factors and Situational Triggers: Exclusionary Reactions to Immigrant Minorities', American Political Science Review 98: 35-49.

Sobel, Richard (2001) The Impact of Public Opinion on U.S. Foreign Policy since Vietnam: Constraining the Colossus. New York: Oxford University Press.

Stelzenmüller, Constanze (2005) 'A U.N. Seat for Europe', New York Times, 20 February.

Sylvester, Rachel (2004) 'Europe Should Have UN Seat', Telegraph.co.uk, 20 November. URL (consulted July 2007): http://www.telegraph.co.uk/news/ main.jhtml?xml=/news/2004/11/20/weu20.xml.

Van Buuren, S., H. C. Boshuizen and D. L. Knook (1999) 'Multiple Imputation of Missing Blood Pressure Covariates in Survival Analysis', Statistics in Medicine 18(6): 681-94.

Waltz, Kenneth N. (1967) Foreign Policy and Democratic Politics. Boston: Little Brown.

Wendt, Alexander (1992) 'Anarchy Is What States Make of It: The Social Construction of Power Politics', International Organization 46(2): 391-425.

About the author

Harald Schoen is Assistant Professor, Johannes Gutenberg University, Department of Political Science, Saarstraße 21, D-55099 Mainz, Germany.

Fax: +49 6131 39 22996

E-mail: schoen@politik.uni-mainz.de