Downloaded By: [University College Dublin] At: 18:24 18 April 2008
International Integration and National Beliefs: A Psychological Basis for Consociationalism as a Model of Political Unification TOBIAS THEILER The model of consociational democracy has at its core the assumption that different culturally distinct communities are most prone to achieve political and economic integration if the dominant lines of social and cultural division between them remain unchallenged. This article seeks to corroborate consociationalism's 'good social fences make good political neighbours' postulate from a psychological angle. It relies on the belief systems model by Milton Rokeach. It first outlines the consociational and the belief systems model respectively. Then it anchors consociationalism's call for the preservation of social and cultural divisions during political and economic unification in Rokeach's proposition that individuals are more liable to adapt to a change in their environment if this change violates beliefs that are few in number and located at the periphery rather than at the core of their belief systems. It concludes by examining a range of Eurobarometer survey data on popular attitudes towards the European Union. These findings support the article's hypothesis.
Since Arend Lijphart first proposed the model of consociational democracy in the 1960s, a debate about its usefulness has involved two schools of thought. To some consociationalism represents a manual for social engineering, a set of organizational principles which, once properly implemented, enables two or more socially and culturally divergent communities to join into a unified and stable political and economic system.1 From this perspective the consociational model has been applied (albeit primarily on paper) as a potential conflict-solving or prevention device to cases as diverse as Canada, Sri Lanka and the European Union (EU). For a second group of writers consociationalism is of little more than typological value. Convinced that 'consociational devices cannot create the conditions for political accommodation where these do not exist',2 they contend that societies that have adopted such devices and are politically stable can trace their stability to some reason other than the fact that they
Tobias Theiler, doctoral candidate, University of Oxford Nationalism & Ethnic Politics, Vol. 5, No. 1, Spring 1999, pp.46-81 PUBLISHED BY FRANK CASS, LONDON
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
47
are consociational. For them, in short, consociationalism reflects rather than creates stability as they deny a causal relationship between the two. In this article I argue that a causal link between consociationalism and stability can indeed be inferred. I seek to infer it by focusing on the repercussions of integration along consociational lines for the belief systems of the individuals that are affected by it. For my discussion of the belief system I rely on the work by the psychologist Milton Rokeach. I first discuss the consociational and the belief systems model. Then I anchor consociationalism's call for the preservation of social and cultural cleavages during political and economic unification in Rokeach's proposition that individuals are more prone to adapt to a change in their environment if this change violates beliefs that are few in number and located at the periphery rather than at the core of their belief systems. In the final section I review a range of Eurobarometer survey data. It shows that, across the EU, support for integration in areas such as cultural and educational policy has consistently been much lower than for integration in culturally less sensitive areas, ranging from economic to foreign policy and defence. Moreover, it suggests that once the Maastricht Treaty had been drafted and was set to give the EU a limited mandate in the former type of policy areas, and once many opponents of the Maastricht Treaty had started to question the compatibility of further integration with the continued social and cultural autonomy of the member states, popular support for the EU declined sharply. Taken together, these findings support the notion that individuals' acceptance of European integration is in part contingent on whether they perceive it as compatible with the continued social and cultural integrity of their respective member states, and thus offer empirical backing for the article's hypothesis. Attempts to explain political preferences and behaviour with the help of psychological models that highlight suspected variations in the centrality of individuals' different beliefs or attitudes are not new.3 However, when it comes to assessing the plausibility of contending modes of political integration and multicultural democracy, such belief-system-centred explanations (as indeed psychological explanations per se) have so far only rarely been invoked. In particular, I know of no prior attempt to relate psychological models to consociationalism in a systematic manner. In light of this, the present article aims to propose a new and worthwhile angle from which to look at consociational theory, and thereby to stimulate further research, for example along the lines of what is suggested in the concluding section. Some preliminary remarks are in order. They help define more clearly what this article does, and does not, seek to accomplish. First, as the brief exposition in the first section will show, a consociational system has a
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
48
NATIONALISM & ETHNIC POLITICS
variety of characteristics. I am not concerned with all of them. Instead, I concentrate on the one characteristic which I believe accounts most for consociationalism's ingenuity, namely the proposition that in the context of political and economic integration the participating populations should be left socially and culturally 'encapsulated'. Narrowing the scope of the article still further, I examine the 'social fences' principle only with regard to its psychological implications, and in so doing only focus on mass rather than elite attitudes. In fact, throughout my discussion I assume a scenario in which a willingness by all relevant political elites to economically and politically integrate has, for whatever reason, already come into being and can thus be treated as an independent variable. In the light of those assumptions, and given that I apply consociationalism to a setting where the different constituent segments are states (a point to which I come back shortly), the article's central theme is captured by the following scenario: Let us imagine a group of states whose governments have decided to economically and politically integrate leading to, say, a single market and a common currency, and to overarching political institutions in charge of areas such as foreign policy and defence. Let us further assume that the states in question are democracies, and that their governments thus need to maintain sufficient public support or at least toleration for their integrative project, lest it falter over elections and referendums. Against the backdrop of such a scenario, most prevailing models of international integration themselves often inspired by models of intra-state integration in the 'nationbuilding' tradition - hold that to sustain such mass support over the long term the affected populations must be made to develop some form of overarching regional identity. This would be marked, if not by outright socio-cultural amalgamation, by at least a proliferation of shared social and cultural reference points and cross-cutting cleavages. As a potential means to bring this about, these models typically focus on cultural and educational policies, the creation of overarching myths and symbols, and the boosting of social and cultural contacts across national boundaries. If one applies the consociational 'good social fences make good political neighbours' principle to the above scenario, in contrast, a very different set of prescriptions emerges. They centre on the demand that the status of the participating states as the primary units of identification for their respective populations should be preserved; the formation of cross-cutting social and cultural cleavages be restricted; and areas of high social and cultural significance, such as cultural policy and education, be exempted from the integration process. At first glance this latter strategy can easily appear counterintuitive and implausible. But is it nonetheless possible to advance a
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
49
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
coherent and theoretically well-founded case for why, compared to the former strategy, it might be more successful in securing popular acceptance of economic and political integration in the long term? Evidently, my scenario bears strong resemblance to the European Union. And it is indeed the EU that provides the empirical background to my argument. Nonetheless, just as this article does not explore all facets of the consociational model, it does not discuss the topic of 'consociationalism and the EU' in all its possible dimensions either. This latter theme has attracted much scholarly attention in recent years4 and is bound up with a large array of issues: the EU's actual and potential institutional makeup, its mode of elite bargaining, the organization of interest groups, and many more. Those issues lie beyond the scope of this article. Instead, I focus on the EU primarily inasmuch as it offers itself as a test case for my hypothesis. More specifically, by analysing Eurobarometer data on the shift of popular attitudes towards European integration in relation to particular developments in the EU, I obtain some empirical support for my claim that popular support for economic and political integration declines in proportion to which it is perceived to infringe on the social and cultural autonomy of the participating communities. In principle, to be sure, those findings could serve as a basis for evaluating different options for the EU's future development, and lead one to argue in a cautiously prescriptive vein: Other things remaining equal, the EU would be more likely to thrive in its economic and political dimensions if it sought to leave the social and cultural boundaries of its member states intact, and stayed out of those policy domains on which the latter most rely for the maintenance of those boundaries. Yet this assertion, too, is linked to many more questions. To some of those I briefly allude in the conclusion, though I cannot hope to answer them in this article. They pertain, for instance, to the separability in practice of the social and the cultural from the political and the economic dimensions of integration, and thereby to the implications of a potential socio-cultural 'encapsulation' strategy in the EU for the Union's development at large. One final remark: even though in principle my hypothesis can be applied to a domestic setting as well, I treat in this essay consociationalism primarily with reference to integration between states. This is in line with my empirical focus on the EU, and in some respects it also helps to bring out my theoretical argument more clearly. And while the segments-arestates assumption contrasts with the emphasis of most consociational 'founding fathers' - such as Lijphart, McRae and others - on internally fragmented domestic systems, it is not particularly new or eccentric. As already mentioned, attempts to apply consociational theory to the EU have
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
50
NATIONALISM & ETHNIC POLITICS
attracted much scholarly attention in past years. Underlying all of them is the assumption, to which I subscribe, that consociationalism should be seen as a set of organizational principles that can potentially apply (and be applied) to a variety of formal and legal contexts. On this account, the actual or potential presence of consociational elements in a given political system does not depend on that system enjoying - or necessarily being in the process of acquiring - the attributes of sovereign statehood.5 The Consociational Model According to Lijphart's often-cited definition, a consociational democracy has four major features. First, the overall political system contains several population segments (which, in the context of international integration, are of course states). These are separated from each other by deep social and cultural cleavages and each enjoy a large degree of autonomy. Second, the overarching political system is governed by a grand coalition in which every segment is represented. Third, political decision making occurs through negotiation and compromise among the representatives of the different segments rather than by majority vote. Each segment holds a (formal or de facto) veto. Finally, the principle of proportionality applies as the standard of political representation, bureaucratic and legal appointments, and the allocation of public resources. The defining characteristics of a consociational system hint at its enabling conditions, i.e., the underlying factors that must be present for a consociational democracy to function. On an elite level this obviously includes above all a willingness by the leaders of all participating segments to compromise and cooperate with each other so as to maintain the overarching political system. For Lijphart, such a willingness may be induced by, among other things, the presence of an external threat and/or the prospects of economic rewards to be had from integration. As regards the condition of the segments at large, their suitability as consociational building blocks benefits from their maintaining a high degree of internal cohesion and homogeneity, and from their remaining separated from each other by cleavages that are sufficiently deep and reinforcing to help their members sustain a distinct in-group/out-group consciousness. Such cohesion at the inside and demarcation from the outside in turn help a segment to aggregate and define its political interest, and to designate representatives to articulate this interest in relation to the other segments that comprise the wider consociational system. Finally, if the social and cultural integrity of its segments is thus a condition for a consociational system to function, then this system must be conceived in such a way as to preserve that integrity and allow the segments
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
51
to 'maintain their separate identities and their mutual isolation'.6 This requires, first, that the political centre does not try to usurp the role of the segments as the primary units of mass identification, for instance by keeping its own visibility low and by granting the segments far-reaching powers in socially and culturally sensitive policy areas such as cultural policy and education. Moreover, it calls for the curbing of contact and interaction between the segments by subjecting their relationship to some form of 'separate but equal' doctrine.7 Examples are the rule of linguistic territoriality in Switzerland and Belgium and, in a setting where the different population segments are not geographically separated, the Dutch strategy of verzuiling ('pillarization') which led to the creation of religiously segregated parallel institutions in many areas of social and cultural life such as education and broadcasting.8 Other techniques that might be employed to the same end include the exemption of cultural 'goods and services' from free trade and anti-subsidy rules, and policies to reduce the inter-segmental mobility of persons. The Search for Theoretical Foundations As has often been suggested, it is above all its claim that the preservation of strong and reinforcing social and cultural cleavages is reconcilable with and - under the conditions outlined above - even beneficial to the maintenance of a unified and durable political system that accounts for consociationalism's ingenuity. It puts consociational thinking at odds with one of the most influential propositions of post-war analytical political theory: the proposition that cross-cutting cleavages equate with moderation and stability, while mutually reinforcing ones are at the source of conflict and centrifugal pulls.9 Consociational assumptions run equally counter to many modernization and development models of the 1950s and 1960s,10 as well as to most models of inter-state integration in the neofunctionalist tradition and its various offshoots that rose to prominence during the same period." They, too, asserted that strong and reinforcing internal cleavages are ipso facto detrimental to political stability within a state or an economically and politically integrated union between states. Consequently, they presumed that the supra-segmental centre of political authority needs to attenuate such divisions and thereby foster the rise of some form of overarching 'socio-psychological community'12 among members of the different segments involved in the integrative project. So as not to be self-referential, those who seek to provide a theoretical corroboration of consociationalism's 'good social fences make good political neighbours' proposition must do more than to show that, once established, a consociational system thrives on the ability of its segments to aggregate and articulate their respective political interest and therefore must
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
52
NATIONALISM & ETHNIC POLITICS
keep them socially and culturally 'encapsulated'. Instead, they must offer a rationale for why, in so far as the maintenance of democratic stability and cohesion of the overarching political system is adopted as the principal objective, a consociational strategy of integration can outperform one that aims at social and cultural amalgamation - or at least cross-cutting cleavage creation - and 'socio-psychological community' building in the first place. Potential rationales for the alleged stability-enhancing properties of the consociational 'encapsulation' dictum have already been proposed from a range of different conceptual corners. These include systems theory,13 and an analysis of elite motivations.14 Yet given that the fate of most integrative projects depends not least on them retaining some degree of popular consent, it is consociationalism's impact on a mass level which represents an equally plausible measure by which to judge its merits. From this angle, the question to consider becomes this: what if anything renders a consociational strategy of integration more likely to sustain overall popular support than alternative strategies? By so asking one inevitably touches on the perceptions, motivations and identifications of individuals in relation to changes in the position and status of the groups of which they are members. And thus, an obvious field in which to start searching for answers is that of social psychology. Social Psychological Foundations It is from the particular area of social psychological research concerned with inter-group relations that, in the first instance, there have emerged a range of findings that would seem to lend themselves rather conveniently to support consociational thinking. For example, from his famous summer camp experiments, Muzafer Sherif inferred that a boosting of contact and interaction between different groups can easily drive their members to greater hostility and self-differentiation along group lines rather than lead to improved mutual perceptions and eventual group merger.15 More recently, the social identity approach in social psychology has sought to place such findings on a coherent theoretical basis.16 Briefly summarized, it proposes that humans have a strong inclination to categorize themselves and others into groups. To identify with one's in-group means to have internalized it to the point where it has become intrinsic to the self-concept. Further, given the link between perceptions of the in-group and perceptions of the self, individuals can bolster their sense of self-worth and self-esteem by enhancing their image of the in-group. This they can achieve through, among other things, 'social comparison' with out-groups. Such 'social comparison' does not necessarily lead to a worsening of out-group perceptions (positive differentiation through a mere in-group enhancement is among the conceivable outcomes), but a deterioration of inter-group
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
53
perceptions is one of the possible consequences of social comparison. From this perspective, two mechanisms can account for a possible link between inter-group contact and inter-group hostility. First, to the extent that individuals perceive such contact to threaten the boundary of their in-group, it is concomitant with an 'attack on the self and can trigger attempts to defend that boundary through heightened self-differentiation along group lines. Second, as a greater proximity to out-groups presents their members with enhanced opportunities for social comparison, it can lead to a further accentuation of perceived inter-group differences and thus, potentially, to greater hostility between the groups.17 Such reasoning, if one accepts it, adds ammunition to consociationalism's theoretical armoury. If group identifications are 'sticky' and a boosting of interaction between groups has at least the potential to foster hostility and conflict, then a mode of integration that does not rely on a change of group identifications and entails a lesser intensification of intergroup contact has a comparatively greater chance of attaining popular acceptance and stability. In what follows, though, I seek to place the consociational 'encapsulation' principle on a broader - albeit by no means incompatible psychological footing. My argument builds on the assumption that integration can be seen as analogous to any other type of large-scale social and political transformation in that it is a source of numerous changes in individuals' everyday social and physical environment. As the discussion of the belief system in the next section will show, those changes are bound to challenge beliefs which individuals have developed and sustain in relation to this environment, and thereby cause dissonance within their belief systems. The ease with which individuals adapt their beliefs and thus come to accept integration depends on the number and centrality of the beliefs which it violates, and, by extension, on the amount of effort they must invest into carrying out the necessary belief modifications and adjusting their belief systems in response. Based on those premises, I set out to answer the following questions: first, what kind of changes does the formation of a consociational union between different communities produce in the environment of the individuals that inhabit them (and how, on this count, does consociationalism compare to alternative strategies of integration)? Second, how do the changes which consociationalism inflicts on their environment affect the belief systems of those individuals? Finally, and still by comparison to non-consociational modes of integration, what inferences can be drawn from this regarding individuals' likely reaction to the integrative process along consociational lines? Before taking on those questions, however, I need to discuss the nature of the belief system and the factors that determine, when individuals' beliefs
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
54
NATIONALISM & ETHNIC POLITICS
are violated, whether they resist or adapt to this violation. For this discussion I rely on the belief systems model by Milton Rokeach. Rokeach's work is relatively old but in important ways it is not outdated. Many of its key premises, as they pertain to the varying centrality of an individual's beliefs, their interconnectedness and the consequences of dissonance between different beliefs, are in their essence echoed by a good number of more recent approaches, notwithstanding the many differences that exist between the various models.18 Rokeach presents these premises in a way that is at once parsimonious and comprehensive. By so doing he lays out a solid theoretical basis on which I will proceed, in the subsequent section, to assess the impact of integration along consociational lines on individuals' mental state and, ultimately, on their political behaviour. The Belief System An individual's beliefs could be seen as disconnected and mutually oblivious entities - rather like fish in an aquarium. When any one fish falls out of favour with the aquarium owner it can be removed, and a more beautiful fish inserted in its place without repercussions for other fish or the aquarium. A different approach - and one which I follow instead - holds that beliefs are interconnected and organized into complex systems. These, according to Rokeach, have 'describable and measurable structural properties'.19 The belief system 'may be seen as an organization of beliefs varying in depth, formed as a result of living in nature and in society, designed to help a person maintain, in so far as possible, a sense of ego and group identity stable and continuous over time.'20 Rokeach identifies four categories of beliefs which he locates along a centre-periphery axis. The position of a belief on this axis is determined by the number of other beliefs to which it is functionally connected. The more central is the position of a belief, the more numerous are the beliefs to which it is linked. Core beliefs form the nucleus of the belief system. Accounting for only a fraction of an individual's beliefs, they represent '"basic truths" about physical reality, social reality, and the nature of the self'.21 Core beliefs are supported by a 'unanimous social consensus among all of one's reference persons and groups'22 and they are not normally subjected to conscious reflection by the individuals that hold them. Many core beliefs pertain to the presence, form and function of physical objects encountered in everyday life. Others are about social practices and institutions that have become so naturalized as to have attained the same taken-for-granted status and the same embeddedness in unanimous social consensus as physical objects proper.23
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
55
The acquisition of core beliefs is an integral part of an individual's integration into his or her social and physical environment, and of the positioning of the self in relation to it. Their importance can be illustrated through the concepts of object-constancy, person-constancy, and selfconstancy: Even though I see this rectangular table from many angles, I continue to believe ... that it remains a table and that it remains rectangular. ... [O]bject constancy is also a social phenomenon, built up in childhood side-by-side with person constancy, both object and person constancy being necessary prerequisites for developing a sense of selfconstancy. Not only does a child learn that objects maintain their constancy, but also that other people constantly experience physical objects as he does.... Object constancy and person constancy ... build up within ... [a child] a basic minimum of trust that the physical world will stay put. ... It is as if nature and society had conspired to provide the child with a minimum guarantee of stability on which to build his own sense of self-constancy.24 Second order beliefs occupy a more outward position on the coreperiphery axis. They differ from core beliefs in that, regardless of how passionately an individual holds on to them, their objects are not subject to a unanimous social consensus and cannot hence be taken for granted. This category is likely to encompass, for example, religious beliefs in multireligious societies. It also includes authority beliefs, about which persons, groups and institutions one relies on to help one interpret the world. Different spheres of experience call for different referents: they can include encyclopaedias, mothers, rifle associations and God. Third order beliefs are more peripheral still. They are derived 'secondhand through processes of identification with authority rather than by direct encounter with the object of belief'.25 Derived beliefs thus are a function of second order beliefs. When individuals switch their allegiance from one referent to another, many of their derived beliefs will change, too, as they acquire a 'package' of beliefs from their new referent. The final category contains inconsequential beliefs which typically pertain to matters such as aesthetic preference. Located at the outer boundary of the belief system, they are functionally connected to the fewest number of other beliefs. They are inconsequential because, as is argued below, their change has few repercussions for other beliefs and for the belief system as a whole.
56
NATIONALISM & ETHNIC POLITICS
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
The Origins of Belief Changes and their Repercussions for the Belief System Shortly, I will examine how the beliefs from the different categories just identified are affected by the formation of a consociational union. And I will show how, on that score, consociationalism compares to alternative strategies of integration. First, though, the discussion of the belief system must be somewhat expanded. In particular, one must investigate how beliefs from the different categories are related to each other, how they interact, and how they change. To shed light on those questions I have derived from Rokeach's model five propositions. The first proposition is that the belief system resembles other systems social and ecological, for example - in that its ideal state is one of equilibrium, a condition in which there is no friction (or dissonance) between its components. Second, this equilibrium is upset when, following the change of one or more beliefs, inconsistency and hence dissonance between several concurrently held beliefs arises. Such dissonance causes anxiety. The belief system can be restored to equilibrium and the anxiety eliminated by modifying or discarding the beliefs that cause it, by acquiring new beliefs with a bridging function, or through a combination of these. The third proposition holds that the magnitude of the disequilibrium caused by a belief change (and thus the number of beliefs that need to be altered as part of the subsequent re-equilibration process) depends primarily on the status of the belief whose change first triggers the disequilibrium. The change of inconsequential beliefs - which occurs frequently, often as the result of external manipulation such as advertising - has few reverberations for the belief system as a whole, for inconsequential beliefs are functionally connected to few other beliefs. But an alteration of core beliefs may lead to serious disruption of beliefs about self-constancy or selfidentity, and from this disruption other disturbances should follow...; it would lead one to question the validity of many other beliefs within one's belief system; it would produce a great deal of inconsistency within the belief system, that, to eliminate, would require [a] major cognitive reorganization in the content and in the structural relations among many other beliefs within the system.26 This can be illustrated further by drawing an analogy between the nucleus of the belief system and the nucleus of the atom. 'It is in the nucleus that the vast energy of the atom is contained; when the energy is released ... the structures of the nucleus and of the atom itself are dramatically changed. ... [If core beliefs] can be made to change, the entire [belief] system will be altered'.27
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
57
The fourth proposition deals with the causes of belief change. As was argued, one way for a belief to be altered is as part of a chain-reaction process which is initiated when the belief system must be re-equilibrated so as to eliminate dissonance after another belief has changed. Initial belief changes, in contrast, can be attributed to two different sources. Derived beliefs, for one, are by definition bound to change if their change is promoted by the referents from which they are derived.28 For a more central belief to become altered, however, its holder must typically encounter the belief object directly, and this encounter must suggest to him or her a discrepancy between the content of that belief (i.e., what one has thought something to be like) and the nature of the belief object as experienced firsthand (i.e., what one discovers it to really be like). Therefore, given that an individual's consciously apprehended social and physical environment is an aggregation of belief objects, every transformation in this environment has ipso facto the potential to bring about belief violations and (under the conditions outlined below) lead to belief change. The final proposition is that not every belief violation is inevitably followed by a belief change. In cases where the violation emanates from efforts by a positive authority to alter the belief, an individual will either change his beliefs and behaviour so that they conform with what he believes positive authority expects of him, or, if he cannot or will not change, he will alter his beliefs about the positive authority itself; he will become more negative or more disaffected with the authority and, in the extreme, he will even formulate new beliefs about new authorities to rely on.29 Similarly, if the belief violation results from a discrepancy between the content of the belief and the nature of the belief object as revealed during an encounter with the latter, individuals' resistance to carrying out the belief change, if it occurs, is bound to lead them to contest those figures, institutions, events or processes they hold responsible for the other-thanexpected nature of the belief object. This in itself, however, entails the change of existing and/or the acquisition of new beliefs. What determines whether the violation of a belief is followed by adaptation (i.e., a change of that belief) or resistance? Here an individual's strategy of coping with belief disruptions must be seen as one that seeks to minimize strain. As was shown, the magnitude of the reorganization of the belief system that is required to restore the belief system to equilibrium (and hence the amount of effort and strain imposed by it) rises in proportion to the centrality of the belief whose change causes the disequilibrium. It follows that a strain-minimizing individual tends to adapt to a belief violation by changing the belief in question as long as it is more peripheral
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
58
NATIONALISM & ETHNIC POLITICS
than the belief that binds her or him to the authority from which the belief violation emanates (and which would need to be changed instead if that authority were disavowed so as to protect the initial belief). In cases where the belief violation results from a direct encounter with the belief object, individuals are prone to change the belief in question as long as it is more peripheral than that which would come under pressure if they chose to contest the source responsible for the other-than-expected nature of the belief object (and hence of the belief violation) rather than to change the belief.30 Consociationalism and the Belief System I have reviewed both the consociational and the belief systems model and can now relate the two. Through so doing the consociational proposition that political integration is easier to achieve and to maintain if the participating populations remain socially and culturally 'encapsulated' will, it is hoped, become more plausible. The discussion proceeds as follows: first, comparing consociationalism to modes of integration as part of which existing social and cultural fences come under duress, I examine which types of beliefs are most liable to be violated by the formation of a consociational union. Then I look at the centrality of those beliefs, and at the extent to which they are affected. From this, finally, I assess the chances that, still by comparison, individuals' reactions to integration along consociational lines are more likely to be adaptation and less likely resistance. The subsequent section will then proceed to offer some preliminary empirical illustration by examining the development of public attitudes towards the EU. Like any transformation in their consciously apprehended environment, the formation of a consociational union between different communities (be they ethnic groups within a state or, as in the case of the EU, formally sovereign states themselves) is bound to disrupt and, in so far as it comes to enjoy popular acceptance, lead to the change of some beliefs that are held by the individuals who inhabit those communities. These belief disruptions can be attributed to two distinct origins. First, some beliefs come under pressure as the result of efforts by integration-prone elites (which may reside at the emerging supra-segmental centre of authority but also inside the individual segments) to alter them through education, propaganda, the creation and manipulation of symbols and so forth in order to secure a necessary level of popular backing for their project. Beliefs thus targeted likely include those whose objects are images of other populations with which the consociational union is being formed, if they are so hostile as to threaten popular acceptance of this union.
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
59
Similarly, the establishment of overarching political institutions brings about the need to secure a partial reorientation of political expectations at the mass level, and thus the belief changes which lie at the basis of such a reorientation. A second source of belief disruptions in the context of consociational union building can be traced to the likelihood that it exposes individuals to a multitude of integrative 'signs' and 'noises'. A growing inflow into the participating segments of foreign goods, immigrants, food labels and languages, the rise of system-wide political parties and so forth may be the likely or even the inevitable by-products of political and economic unification. In so far as they are apprehended by individuals as changes in their social and physical environment, they ipso facto represent changes in the nature of belief objects. As such they have the potential to cause belief violations. However, while consociationalism is hence liable to have some impact on the belief system, this impact is, compared to modes of integration as part of which the social and cultural fences between the participating communities become challenged, nevertheless a more modest one. This becomes evident by examining the overall number of initial belief violations it inflicts, the likely number of core beliefs among those disrupted, and, finally, the extent to which those beliefs that are affected come under pressure. To begin with, the overall number (i.e., comprising beliefs from all categories) of initial belief violations caused by integration along consociational lines is comparatively small. The reason for this should be readily apparent. In the ways discussed, consociational union builders strive to keep the participating populations socially and culturally 'encapsulated' and thereby prevent, as far as possible, processes of political and economic integration from 'spilling over' into the social and cultural realms. In so doing they cause fewer changes in individuals' environment and thus fewer initial belief violations and less overall need for belief change than would be the case if economic and political integration were accompanied by large-scale social and cultural transformations. Consociationalism's comparatively modest repercussions for the belief system are further brought to light when one turns to its impact on core beliefs in particular. (As was shown, because of the large number of secondary belief violations triggered by a core belief change, this can be a more important criterion than the overall number of initial belief violations inflicted by integration.) To capture this impact one must keep in mind, first, that a comparatively large proportion of those beliefs which individuals acquire during their primary socialization - i.e., as they become first integrated into and internalize the basic patterns of their social and physical
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
60
NATIONALISM & ETHNIC POLITICS
environment - are in fact core beliefs, and, second, that their communal (e.g., as in the case of integration in the EU, national [primarily in the sense of state-related]) situation is usually an important codeterminant of this environment. While strategies of integration in the 'psychological nationbuilding' tradition require individuals' partial 'de-socialization' from their existing communal contexts as the basis of their resocialization into a wider 'socio-psychological community', consociational union builders have no comparable ambition. Instead, the very strategy of safeguarding the integrity of the constituent communities by socially and culturally 'encapsulating' them helps to preserve their role as frameworks of reference and repositories of meaning for their inhabitants, and thereby to protect the integrity of many core beliefs which individuals have developed and sustain in relation to their particular communal milieus. There is a further indication of consociationalism's gentleness to core beliefs. As was shown earlier, among the beliefs that are most liable to come under pressure during the formation of a consociational union are those that pertain to perceptions of other communities with which this union is being formed. These beliefs are likely to become the target of integration-prone elites if they are so hostile as to endanger popular acceptance of integration. Yet at the same time, in many instances those beliefs are not bound to be core beliefs. Owing their existence to efforts by belief-transmitting agencies (such as the school system, conscript armies, religious authorities, etc.) to disperse and sustain them among the population at large, these beliefs may well not have been transmitted to individuals at a sufficiently early point in their socialization process and/or be promoted with enough intensity to acquire the status of core beliefs. This is especially likely in an integrative setting such as the European Union, where the different member states have pluralistic domestic systems and where there has been a relatively long absence of manifestations of acute hostility between them. Though negative stereotypes about other populations may still be promoted by a variety of domestic agents, this often occurs in a 'soft' and inadvertent way (note the accusations of lingering 'nationalistic biases' in history and civics textbooks frequently levelled in many EU member states) rather than in a deliberate and concerted fashion. In so far as beliefs about other communities are indeed of a non-core belief status - and their objects thus remain outside the body of 'basic truths' which shape individuals' fundamental understanding of the world - they are liable to change with comparative ease once they are no longer transmitted by their referents. The prospects for the rapid alteration of those beliefs are even greater when, as in the example about to be discussed, those referents come to promote new beliefs of an opposite content. The preceding paragraph will benefit from some illustration. Even though the claim that mass-based perceptions between populations are
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
61
comparatively easy to change may seem implausible at first glance, there are in fact many examples which show that hostile perceptions between different populations can improve relatively rapidly. In particular, they show how negative images fade once belief-transmitting agents (such as the school system) work towards their elimination rather than their promotion. A classic example of this is the Franco-German reconciliation programme after the Second World War. It stimulated a wide range of private, semiprivate and governmental initiatives to diffuse positive mutual images between the two populations and end the transmission of negative ones. These included the 'correction' of history textbooks, school-twinning schemes, television programmes in which the two countries were presented to each other in a favourable light, friendship festivals, and many similar measures.31 And as has been shown repeatedly, since the end of the war mutual perceptions between the two populations have indeed improved significantly - to the point where for French respondents Germany has become the second most 'trusted' country on earth.32 Similarly, there are many instances in which an (even more) rapid worsening of perceptions between populations occurred, and where this was again accompanied by what one can broadly describe as an ^lite-driven strategy to promote such a deterioration.33 It is admittedly difficult to prove a direct causality between such belief modification strategies and the actual belief changes that occurred. At the very least, however, the fact that these changes did transpire, and often at a relatively rapid pace at that, allows for one wellsupported inference: if perceptions of other populations are so fluid in either direction, the beliefs that are at their basis are less likely to be at the core of the belief system. For as was shown, core beliefs stand out by a strong resistance to change. Finally, besides violating comparatively few beliefs in general and few core beliefs in particular, consociationalism can also be found to affect those beliefs which it does violate to a comparatively modest extent only. This, for example, is the case with beliefs whose change is sought by integrationprone elites so as to secure a partial shift of political expectations at the mass level towards overarching political institutions. As we saw, lest it violate an attendant condition to the socio-cultural 'encapsulation' principle, integration according to the consociational blueprint has only comparatively modest repercussions for the internal political structures and processes of the different constituent communities which remain largely intact. By implication, it needs to be accompanied by only a relatively modest shift of political expectations towards the emerging supra-segmental centre of authority. This implies a need for correspondingly modest belief alterations in this regard. The same applies to beliefs which dlites may seek to alter in order to improve mutual perceptions between different
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
62
NATIONALISM & ETHNIC POLITICS
populations. For the necessary quality of mutual perceptions that must be attained for a consociational system to be viable is kept at a minimum by the fact that in such a system contact between the constituent communities - and hence the opportunities for friction between them to be generated - is restricted by the principle of social and cultural 'encapsulation'. In short, compared to strategies of integration as part of which the social and cultural fences between the participating communities come under attack, the formation of a consociational union has a more modest impact on the belief systems of the individuals that inhabit those communities. It leaves intact the taken-for-granted status and the object constancy of a greater number of belief objects. And even those beliefs which it does disrupt - and whose change is hence required if popular reaction to integration is to be one of adaptation rather than resistance - are, on average, of a more peripheral status and affected to a lesser extent. All this, in turn, suggests that individuals can adapt to the formation of a consociational union (and hence carry out the belief alterations that are at the basis of adaptation), yet do so at a relatively modest cost. The extent to which it requires their belief systems to become re-equilibrated, to which it endangers their overall sense of object- and self-constancy and to which it creates anxiety is less than would be the case if integration entailed the erosion of the social and cultural boundaries between the participating communities and thus more substantial belief violations (and if, by extension, adaptation implied a need for more extensive belief changes). Because of this, finally, the probability that individuals' reaction to the formation of a consociational union will indeed be one of adaptation is comparatively greater, while the chances that they will seek to protect their belief systems by resisting this process and the belief violations it entails are reduced. Some Empirical Evidence: Popular Attitudes Towards the European Union Applied to inter-state integration in the context of the European Union,34 the argument outlined above can be translated into two related hypotheses. First, individuals are least inclined to accept integration in those policy areas that are bound up most tightly with the ability of their respective states to maintain their social and cultural distinctiveness - i.e., the very areas that in a consociational scenario remain sheltered from the integrative process. Likely examples are cultural policy, social policy and education.35 The second hypothesis, which follows logically from the first, is that if individuals do perceive the European Union to pose a threat to the continued social and cultural autonomy and integrity of their respective member
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
63
states, for example by seeking to encroach into relevant policy areas, they become more inclined to oppose it. In this final section I examine a range of opinion poll data obtained from Eurobarometer surveys in order to compare them against those two assumptions and, by implication, against the consociational claim that popular acceptance of integration is more likely to be forthcoming if supranational institutions stay out of the social and cultural realms and if the participating populations remain socially and culturally 'encapsulated'. Yet before doing so I should reiterate the point made at the outset. My discussion is not concerned with the wider predicament of the European Union or the broader implications of the Maastricht Treaty, institutional, policy-related or otherwise. For instance, while I argue that in one important respect the Maastricht Treaty moved the Union away from consociational principles (and that this in turn helped cause a slump in mass support), the emphasis is on 'in one respect' - i.e., in so far as the EU became more widely perceived as a potential usurper of the 'national identities' of its member populations, for the reasons and in the ways I will discuss. This entails no claim that, say, the pre-Maastricht EU amounted to a full-fledged consociational system, and it is indeed compatible with the potential notion that in some other respects the post-Maastricht EU featured a greater number of consociational elements (for instance as regards its predominant mode of elite bargaining). In short, in this section I concentrate on developments in the EU and the shift of public attitudes in response to them only in so far as this has a direct bearing on the central theoretical concern of the article. A note on the uses and limitations of Eurobarometer data is equally in order. As is argued below, Eurobarometer suffers from a range of shortcomings. All the same, it is the only available data set suitable for the present purpose. There are two reasons for this. First, in comparing the two hypotheses spelled out above against public opinion data, one should seek to focus on the Union as a whole rather than merely on any one individual member state. Such a broad focus is required because, within individual member states, public attitudes towards the EU can at times be affected by factors peculiar to the domestic situation of the country in question (e.g., the unpopularity of a domestic government or of some of its policies). By concentrating on public opinion towards the EU on an EU-wide scale, one can dilute as far as possible the overall impact of such country-specific factors on one's findings. Second, since what I am interested in are shifts in public opinion in relation to developments in European integration over a period of time I require time-series data. Eurobarometer is the only survey of its kind that satisfies both these requirements: its data are compiled at regular intervals and simultaneously in all EU member states.
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
64
NATIONALISM & ETHNIC POLITICS
At the same time, Eurobarometer has a range of drawbacks that limit its usefulness for what is attempted in this section. For instance, while each Eurobarometer edition contains a batch of questions on different aspects of European integration and the EU, not all of them belong to the set of 'core questions' that are asked regularly. Many others, among which are some that would be the most relevant for the evaluation of my two hypotheses, have been one-off questions or are asked infrequently. Of those questions that Eurobarometer does ask regularly, many amount to little more than slight variations on the basic 'are you for or against integration?' theme. Furthermore, potentially important subtleties and distinctions - pertaining, for example, to affective versus cognitive dimensions of support and to the relative importance individuals attach to their preferences on different EUrelated issues - are only marginally reflected in Eurobarometer data or not at all. Because of these and some other limitations, in short, Eurobarometer does not provide a sufficient basis for a systematic and rigorous empirical testing of my hypotheses. Nonetheless, I believe that what one can derive from it is a preliminary hint of sorts that these hypotheses may well have some empirical foundation, pending the conduct of more refined and 'made to measure' inquiries into popular attitudes towards European integration. Support for Integration in Different Policy Domains The first hypothesis thus states that individuals are more likely to oppose integration in policy areas that are of a high social and cultural significance. And this proposition can be compared against Eurobarometer data in a straightforward manner. For at regular intervals Eurobarometer has asked the EU's citizens to separate policy areas in which they want the European Union to be involved from those which they prefer to remain sheltered from the Union's interference. As it turns out, this has produced results which have remained remarkably clear-cut and consistent over the years. Taking the findings obtained in the spring of 1994 as an example (see Figure 1), there was high 'net support' (i.e., percentage of approval minus percentage of rejection) for the Union assuming a role in giving aid to developing countries (58 per cent), the promotion science and technology (49 per cent), foreign policy (46 per cent) and environmental protection (33 per cent). Even the prospect of a common currency enjoyed relatively solid, though less decisive, public backing at 10 per cent. In contrast, Union involvement was sharply rejected in cultural policy (-20 per cent), social policy (-33 per cent), and education (-35 per cent).36 These findings thus show that a majority of Union citizens favours integration in many different policy areas. Remarkably, this even includes domains that have traditionally been associated with 'high politics' and/or located at the core of state sovereignty and independence such as foreign
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
65
FIGURE 1
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
SUPPORT FOR INTEGRATION BY POLICY DOMAIN (SPRING 1994)
|——^M
Foreign aid Science/ techn. Foreign policy Environment Currency Value added tax Defence Media policy Culture Social Policy
-
M^
Education -40
-30
-20
-10 0 10 20 30 40 Net support for Union involvement (%)
50
Source: Eurobarometer No.41 (Brussels: European Commission, 1994)
policy, defence and monetary policy (though for some qualifications on the currency question, see Note 37 below). At the same time, the data bear out the first hypothesis, in so far as most respondents want those fields that are most intimately intertwined with the member states' ability to keep their social and cultural fences beyond the Union's reach. Remarkably, moreover, this desire prevailed in defiance of the utilitarian arguments which the Union frequently invoked at the time in order to justify its attempted involvement in those fields (e.g., the claim that a partial equalization of educational standards between the member states was necessary to facilitate the movement of labour inside the single market and thereby maximize the economic gains to be had from integration, or that the Union's involvement in the audio-visual sector would strengthen European 'cultural industries' against US competition).37 What has been shown thus far is that popular preferences clearly go against Union involvement in policy areas of high social and cultural
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
66
NATIONALISM & ETHNIC POLITICS
significance. Yet by itself this does not indicate what conditioned those preferences. For example, it does not rule out the possibility that people rejected Union involvement in culture, education and social policy merely out of a (potentially more or less easy-to-change) expectation that it would fail to produce utilitarian benefits, rather than, as is the present hypothesis, in part out of a desire to protect the social and cultural boundaries and integrity of their respective member states, and thereby many of their most centrally located beliefs. Furthermore, the above findings do not suggest how strong and deeply held these preferences were in relation to people's preferences on other EU-related issues, and thus the extent to which their potential violation is bound to influence attitudes towards the EU and European integration at large. Eurobarometer itself is far from ideally suited to shed light on those questions. While it registers individuals' views on a variety of EU-related issues and their development over time, it does not, for the most part, seek to probe into the relative importance which individuals attach to those views or their conditioning factors. However, as is argued next, there is an indirect way of inferring that popular preferences for non-integration in domains of high social and cultural significance are likely to have been strong rather than weak, and also that they were conditioned at least in part by 'national identity'-related concerns. It is grounded on the observation that once these preferences had been violated, a substantial worsening of popular attitudes towards integration at large ensued. And as I will show further, this drop in popular support was at least in part brought about by fears that the Union had come to pose a threat to the social and cultural integrity of its member states. Support for the EU During the Debate on the Maastricht Treaty If a large majority of its citizens want the European Union to stay out of those policy areas that are bound up most closely with preserving the social and cultural integrity of their member states, and provided they hold this wish strongly, then a widespread perception that those very fields have become vulnerable to interference by the EU - and more generally that the Union has come to pose a social and cultural threat to its member populations - should be expected to lead to a decline in its popular standing. This, I propose below, is precisely what happened during the debate on the Maastricht Treaty in the early 1990s, when the Union suffered a substantial fall in mass support. A detailed exposition of the contents of the Maastricht Treaty (formally named 'Treaty on European Union') or the process that led to its eventual ratification by a whisker is beyond the scope of this article. Important for the present purpose is to recall that the treaty took the Union into a range of policy areas that had hitherto been the exclusive or near-exclusive preserves
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
67
of the member states (or, regarding some competences in some federal and/or bi- or multinational member states, of the federal and/or national sub-units). In addition to monetary, foreign and defence policy, these included culture, education, justice, and social policy. The actual scope and 'depth' of the competences which the Union was to receive in those domains was for the most part relatively limited. But the proposed widening of the Union's powers was nonetheless significant. For it promised to take the European integration project beyond the (largely economically oriented) Common Market premises on which it had originally been established and on which it had rested ever since. As the Maastricht negotiations continued throughout 1991, the prospect of an expansion of the Union's powers beyond its original economic mandate unleashed an intense public debate in many member states - a debate which intensified in the period between the formal signing of the treaty in early 1992 and the end of the ratification process in the last member state in the autumn of 1993. This was mirrored not least in rapidly rising awareness levels. Public awareness of the Maastricht Treaty rose from 44 per cent in the spring of 1992 to 85 per cent half a year later. By the spring of 1993, it attained, a for any EU-related issue exceptionally high, 94 per cent.38 The Maastricht debate did not unfold according to a uniform pattern in the different member states. Yet one can make out some widespread similarities. Not least these concerned the preferred types of argument which supporters and opponents of the treaty used in an attempt to rally public support behind their respective causes.39 While the treaty's advocates tended to concentrate on the economic, security-related and other utilitarian benefits allegedly to be had from it, many of its adversaries - from all parts of the political spectrum - expressed fears that an expansion of supranational powers would make the member states vulnerable to an expansionist and intrusive Brussels-based bureaucracy, with pressures for a social and cultural levelling and a corresponding loss of 'national identity' to follow. Significant for the present discussion is that even in many of those member states in which European integration had hitherto enjoyed a strong elite consensus, such opposition to the Maastricht Treaty came to reach relatively far into the political mainstream. Often it caused internal divisions in governing and major opposition parties alike. In conjunction with the very high level of public attention which Maastricht debate attracted, this meant that many EU citizens became, for the first time, confronted with an at least conceivably credible (i.e., not just emanating from the traditionally integration-sceptic political fringe) portrayal of the Union as a potentially intrusive power and usurper of the 'national identities' of its citizens. This depiction contrasted sharply with the hitherto almost unchallenged
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
68
NATIONALISM & ETHNIC POLITICS
representation (in many member states) of the Union as a benign guarantor of economic prosperity that was complementary to and respecting of the social and cultural particularities of its member states. To assess how this change in the Union's image affected its standing in popular opinion, I singled out three questions which Eurobarometer asks at regular intervals to gauge public support for integration.40 By subtracting the anti-integration from the pro-integration answers, I obtained three indices of 'net support'. These I averaged out into a single indicator of 'average net support' which is depicted in Figure 2. FIGURE 2 SUPPORT FOR INTEGRATION DURING THE MAASTRICHT DEBATE 60
Maastricht Treaty ratified
1988b 1989b 1990b 1991b 1992b 1993b 1994b 1995b 1989a 1990a 1991a 1992a 1993a 1994a 1995a
Notes:
(1) The 'Maastricht negotiations begin' date refers to the launch of the two parallel Intergovernmental Conferences (IGCs) on European Monetary Union and European Political Union in December 1990. (2) the figures for 1995 exclude Austria, Finland and Sweden which had joined the Union that year. (3) For the first part of 1994 Eurobarometer did not publish respondents' views on 'efforts to unify Western Europe' (i.e., one of the three questions I used to calculate average net support). For the purposes of computing the latter I presumed a result unchanged from the previous term.
Sources: Eurobarometer Trends 1974-1993 (Brussels: European Commission, 1994); Eurobarometer NosAl-44 (Brussels: European Commission, 1994-1996.
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
69
Support for European integration increased throughout the late 1980s, peaking in early 1991 at a record 57 per cent average net support. This occurred at a time when EU politics was dominated by an economic agenda, notably the implementation of the single market programme that had been adopted in the mid-1980s and was scheduled to be completed by the end of 1992. Yet in the second half of 1991, when the Maastricht negotiations entered their decisive phase and the debate on the treaty intensified in many member states (which, as we saw, became reflected in rapidly rising public awareness levels), the Union's popular standing began to fall. Net support for integration was down to 53 per cent in late 1991, 46 per cent in early 1992 and then plunged to 37 per cent by the second half of that year. After the Maastricht Treaty had been ratified by all the member states and public attention devoted to it began to decline somewhat,41 support for integration recovered slightly, but only to continue to slip thereafter at a slower pace. The very high rates of public approval which the Union had enjoyed into the early 1990s were not regained. Alternative Explanations One can make out a range of potential alternative explanations that might be invoked in an attempt to account for the waning support for European integration at the time of the Maastricht Treaty. In what follows I shall briefly consider the two that suggest themselves most readily. The first alternative explanation (AE1) in some ways resembles the account offered above. For it, too, attributes popular opposition to the Maastricht Treaty to anxieties that can broadly be construed as somehow 'national identity'-related. Yet in contrast to my preferred explanation, AE1 interprets such anxieties as having been triggered not so much by perceived threats to the continued internal social and cultural integrity of the member states but, instead, by the prospect of a loss of their respective sovereignty, independence and influence in relation to the outside world (brought about, for example, by one's country's participation in a common foreign, monetary and defence policy as it is stipulated in the treaty). To use imperfect but (for the following discussion) convenient labels, whereas the preferred explanation postulates 'national identity'-centred anxieties of an 'internal' kind, AE1 interprets them as having had primarily an 'external' dimension. A second potential alternative explanation (AE2) builds on the insight that the advent of the Maastricht Treaty coincided with an economic slowdown across Western Europe, and thus with a traditional spoiler of public enthusiasm for integration. According to this explanation, opposition to integration at the time of the Maastricht Treaty was propelled by fears of economic losses as opposed to fears of a loss of 'national identity' however defined.
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
70
NATIONALISM & ETHNIC POLITICS
Before addressing those potential alternative explanations, the earlier word of caution regarding Eurobarometer should be reiterated. Just as the available Eurobarometer data set does not lend itself to anything amounting to a systematic testing of the preferred explanation, it does not allow this to be done for the two alternative explanations either. What one can derive from the data, I believe, is some indication that those alternative explanations are either less likely than the preferred explanation (as in the case of AE1) or, as with AE2, may well be plausible but at the same time not prone to obliterate the preferred explanation. Starting with the first alternative explanation, one way to put its validity into doubt builds on the observation that the primary connotation of the term 'national identity' as it was used during the Maastricht debate throughout the member states was indeed 'internal' rather than 'external'. When opponents of the Maastricht Treaty claimed that it would harm 'national identities', the images most commonly invoked in conjunction with this claim were those of 'intrusion', 'cultural levelling', 'Europe forcing us to give up who we are' and so forth. Readily as I think most observers of the Maastricht debate would concur with this assessment, it is difficult to further substantiate it here. To do so would require the kind of 'deep' interpretative analyses of the dominant anti-Maastricht discourses at the time that have, for the most part, remained lacking.42 However, one can also find some more solid evidence against AE1. In particular, this explanation seems inconsistent with the survey results compiled in Figure 1 as they were discussed earlier, read in conjunction with the data presented in Figure 2. For these show that even at a time when support for the European Union per se was in the doldrums (i.e., spring 1994), there continued to be strong public support for EU involvement in such national sovereignty- and independence-sensitive areas as foreign policy and defence. In other words, it is not likely that the decline in overall public support for European integration at the time was triggered to a significant extent by respondents' concerns over Union encroachment into the areas just listed, given that support for integration in those very areas remained strong throughout. To investigate the second alternative explanation (only, however, to eventually be brought back to discussing the first) I turn to another Eurobarometer survey which is reproduced in Table 1. It asked those who professed to be anxious towards integration to list up to three different grounds for their anxiety. These answers then were grouped by Eurobarometer into the different categories shown. Many of the answer categories compiled in Table 1 are not easy to make sense of. Indeed, some ('too many changes', 'uncertain future', 'not ready yet') are so vague that one could read almost anything into them. This
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
71
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
TABLE 1 WHY AFRAID OF INTEGRATION? (AUTUMN 1992) GROUNDS FOR FEAR
FREQUENCY OF RESPONSE
Less [sic] jobs Too much immigration Loss of national identity Uncertain future Drugs, crime Higher prices Too much competition Not ready yet Pay for others Country no say Too many changes
33% 30% 29% 28% 26% 25% 23% 22% 20% 18% 11 %
Source: Eurobarometer No.38 (Brussels: European Commission, 1992).
problem is compounded by Eurobarometer's failure to reproduce the exact wordings of the different answers it grouped into those three answer categories respectively. For the present purposes, these responses should thus be set aside. Those answers that are left allow for two main inferences. First, economic concerns did help foster scepticism towards integration, as two sizeable categories of responses reflect economic grounds for anxiety ('less [sic] jobs', 'higher prices'). This helps corroborate AE2. But, and this is the second conclusion drawn from Table 1, economic concerns were not the only reasons for anti-EU scepticism at the time. Instead, they existed alongside answers that can broadly be interpreted as having been at least in part 'national identity'-related. They are 'loss of national identity', 'too much immigration', and (though, as I argue below, only at first glance) 'country no say'. Yet why assume that such 'national identity'-centred anxieties were of an 'internal' rather than 'external' kind as I defined it, and hence that Table 1 primarily supports, in addition to AE2, the preferred explanation rather than AE1? One answer draws on the earlier observation that the predominant connotations of the 'national identity' rhetoric during the Maastricht debate were in fact 'internal' rather than 'external'. In so far as one subscribes to this interpretation, one can infer that when respondents invoked the fear of a 'loss of national identity' as a reason for their scepticism towards integration, they meant it in the 'internal' rather than 'external' sense, and assumed that it would be understood in this way. As was suggested, however, while I hold this claim to be very much valid, it cannot be further substantiated here.
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
72
NATIONALISM & ETHNIC POLITICS
In any event, the contention that those who opposed the Maastricht Treaty on the grounds of fearing a loss of 'national identity' did so primarily on 'internal' rather than 'external' grounds can also be bolstered with some harder evidence derived from Eurobarometer data proper. First, in Table 1, responses that would have mirrored 'external' (i.e., AEl-type) anxieties in a direct and unambiguous way - along the lines of, say, 'our country will surrender its influence in the world' or 'once we have given up exclusive control over our defence policy other nations will no longer respect us' - are entirely absent. (At first sight the 'country no say' answer might appear as an exception. Yet in the context of the Maastricht debate it is bound to have had more to do with one of the preferred arguments advanced by Maastricht's opponents; an argument that was not really 'national identity'related at all but more of a utilitarian nature. It denounced allegedly insufficient provisions for individual member states to influence, and if necessary veto, policy decisions within the framework of EU processes and institutions, and by extension the danger of a member state becoming subjected to policies that run against its proper interest [e.g. in the realm of economic policy43]. This argument did not in the first instance pertain to the member states' broader standing in relation to the outside world.) Admittedly, my preferred explanation does not have much to show for it in the way of unambiguous and direct responses either, should one choose to ignore the above interpretation of the 'loss of national identity' answers on the grounds that it is largely impressionistic. An exception are the 'too much immigration' responses. It is plausible to assume that they were driven in part by the kind of anxieties postulated by the preferred explanation, given that throughout Western Europe the debate on immigration policy at the time was marked as much by social and cultural as by purely economic concerns.44 But the strongest indication against AE1 and in favour of the preferred explanation flows once more from the data presented in Figure 1, read in conjunction with what is shown in Figure 2. As was argued, it shows that at the same time that public backing for integration per se declined, support for EU involvement in such sovereignty and independence sensitive areas as foreign policy and defence remained strong. This makes it appear less likely that most of those who opposed the Maastricht Treaty on the grounds of fearing a loss of 'national identity' were driven by 'external' considerations. By implication, it makes it more likely that their 'national identity'-related anxieties were indeed of an 'internal' kind, dominated by concerns that the Union could come to threaten the social and cultural integrity and autonomy of its member states. Eurobarometer's potential to help illustrate my theoretical argument has been somewhat stretched to the limit by now. But taken together the data
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
73
reviewed in this section do, I believe, lend some backing to the two hypotheses stated at its beginning, and by implication to the central claim of this article. It shows that a majority of the Union's citizens supports integration in many areas, including economic policy and even such sovereignty-sensitive domains as foreign policy and defence. At the same time, they are strongly opposed to integration in those fields that are the most closely linked to the ability of their respective member states to defend and reproduce their social and cultural distinctiveness. And when the EU began to move into the domains of culture, education and social policy, and, most importantly, when an alternative discourse in which it was portrayed as an usurper of 'national identities' began to be disseminated more widely, its popular backing declined sharply. In short, the further the Union's perceived appearance deviated from the consociational demand that the political centre be seen as respecting of the social and cultural fences of the different constituent units, the less popular it became. Conclusion This article proposed a psychological justification for the consociational claim that popular acceptance of integration is more likely to be forthcoming if the participating communities remain socially and culturally 'encapsulated'. Compared to strategies of integration which aim for the attenuation and cross-cutting of social and cultural cleavages that divide the affected populations, integration along consociational lines violates fewer beliefs in general and fewer core beliefs in particular. This allows individuals to adjust to it yet minimize the repercussions it has for their belief systems, and in so doing ensures that political and economic integration attract comparatively greater levels of public acceptance. I concluded by examining a range of survey data on public attitudes towards the European Union. It shows that while popular backing for integration in areas ranging from economic to environmental and foreign policy remained relatively strong throughout, support for the Union declined once it had signalled an ambition to move into more socially and culturally sensitive policy areas, and once the compatibility of further integration with the continued social and cultural distinctiveness of the member states had been thrown into question in much of the popular debate on the Maastricht Treaty. Taken together, this provides some empirical evidence for the central theoretical claim of this article. A call for more refined empirical data is in order here. For instance, it remains to be shown more conclusively that those who professed scepticism towards the Maastricht Treaty on the grounds of fearing a 'loss of national identity' were not merely paying verbal homage to the by then popularized
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
74
NATIONALISM & ETHNIC POLITICS
identity discourse while being driven, in reality, by ostensibly less 'respectable' or less fashionable motives. Arguably, these could have included a refusal to channel more funds into poorer member states, lingering anti-German animosities, economic fears related to the planned common currency, and so forth. An indication that 'national identity'centred anxieties of the type postulated by the preferred explanation were indeed genuine flows from the fact that, as was shown, the Union's citizens have traditionally rejected integration in such areas as culture, education and social policy whereas they have been, on average, much more positively disposed towards integration in culturally less sensitive areas (including, for example, a common currency and foreign policy). It also seems supported by the observation that when those preferences were violated, support for integration at large declined significantly. Nevertheless, by itself this is still insufficient to demonstrate the strength of those preferences and thus their relative contribution to the formation of attitudes towards the Union in a conclusive fashion. The central problem remains this: to show that people have certain preferences is one thing, but to demonstrate that these are actually important to them - and a significant influence on the formation of attitudes towards broader policy issues and, ultimately, on political behaviour - remains quite another.45 To gain more clarity on this point, attitude measurements more subtle and more refined than those offered by Eurobarometer would be needed. Also, focus group-oriented research might be a promising way of circumventing the limitations of the traditional opinion poll and probe deeper into respondents' motivations. Yet even those who, despite the imperfect quality of the available empirical evidence, accept my theoretical argument, and by extension the notion that the EU would be more likely to retain mass support if it sought to leave the social and cultural boundaries of its member states intact, must grapple with many more issues. They bear on the viability of such a potential 'encapsulation' strategy in the EU, and to its wider implications. Of obvious importance is a variable that has been bracketed throughout this article, namely the behaviour of political elites. As we saw, at the very core of consociational reasoning lies the assumption that consociational arrangements are built and maintained by elites who consider them their best chance of overcoming or preventing political fragmentation. Seen in relation to the EU, this must lead one to ask whether, and if so why, the governments of the member states might opt to seek to socially and culturally 'encapsulate' their populations. There are signs that in the years since Maastricht they have increasingly veered in this direction. This is evident for instance in their frequent resistance to attempts to augment the Union's involvement in cultural and educational policy to match the legal
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
46
75
mandate it received in those fields in the Maastricht Treaty. At the same time, such opposition risks pitting national governments against non-state political elites associated with the EU itself, such as the European Parliament and the Commission. These two bodies have remained the most vocal proponents of cultural and educational policies aimed at guiding Europeans into a 'socio-psychological community'. Beyond this, one needs to examine more closely how a potential commitment by the EU to a social and cultural 'encapsulation' strategy might affect the broader teleology of integration. For instance, could the EU adhere to that particular consociational principle without ultimately taking on board the other institutional and policy-related elements that are part of the consociational blueprint? They include adherence to the proportionality rule and the retention of a veto for every member state - neither of which are unproblematic in an enlarged Union of 20 or more member states. The answer hinges in part on whether one sees the consociational model as an indivisible 'package' of organizational principles, or instead as a collection of 'pick and mix' options from which a given political system may adopt some while ignoring others. But the repercussions of a potential social and cultural 'encapsulation' strategy in the EU might become even more significant, as it could eventually undermine one of the central tenets on which the Union has evolved. From Jean Monnet to Jacques Delors, many of the EU's most influential 'movers' aspired to more than to the creation of a common market and a range of overarching institutions. Instead, they sought to root their project far deeper down the societal level by fostering a gradual 'denationalization' of many areas of economic and political life. Over time there would emerge pan-European political parties, interest groups, corporations, trade unions, consumer and business organizations and so forth, organized along functional, ideological or sectoral lines while cutting across national boundaries and standing divorced from any particular national context. Most observers agree that some 'denationalization' along those lines has materialized in the EU, though there is much debate about the extent to which this has occurred and about whether it is set to intensify in the future. In our context the most central question to which this gives rise is whether such a state-transcending reconfiguration in the political and economic realms could coexist with the kind of state-reinforcing social and cultural 'encapsulation' which the consociational logic demands. By so asking one becomes soon confronted with a more fundamental question that has equally been bracketed throughout this article: Just how neatly can, in integrative practice, the 'social' and the 'cultural' be disentangled from the 'political' and the 'economic', and would not the maintenance of high fences in the former realms in the end require the EU to retain at least
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
76
NATIONALISM & ETHNIC POLITICS
moderately high fences in the latter?47 Those who answer in the affirmative are thereby led to a paradox: because of the psychological mechanisms discussed in this article, the preservation of social and cultural boundaries between the member states would facilitate political and economic integration in the EU. But because of the intrinsic interconnectedness of the political and economic with the social and cultural dimensions of integration, it would at the same time set limits to the level of political and economic unification which the Union could hope to attain. There is a further issue that is brought into sharper focus by our discussion of consociationalism, the belief system, and the EU. As was argued, the viability of a consociational system depends in large measure on the existence and strength of broad-based loyalties that are directed towards the individual segments rather than towards the overarching political system. Nevertheless, with varying degrees of emphasis consociational theorists have argued that some segment-transcending loyalties and identifications are required to sustain the formation and maintenance of a consociational arrangement. And it is the situation in existing consociational democracies which is often evoked to support this claim. Although they have in place mechanisms to curb contact and interaction between their constituent segments, and although these segments have remained divided by unabated and reinforcing lines of cleavage, their members nevertheless maintain a shared sense of belonging to a segmenttranscending community and a responsiveness to overarching myths and symbols. This, for example, is the case in Switzerland where, as Schmid48 has shown, a shared sense of 'Swissness' has come to encompass members of all linguistic groups to a similar extent. It is transmitted early in the process of political socialization and includes beliefs that are, in terms of the earlier discussion, likely to be at the core of the belief system. Pondering consociationalism's applicability to European integration thus raises the question of how such overarching sentiments can emerge and be sustained despite adherence to the consociational imperative of social and cultural separation. To tackle it one would need, for one thing, to draw a sharp conceptual distinction between processes of social and cultural amalgamation and/or cross-cutting cleavage creation on the one hand and the emergence of loyalties and identifications that are shared by members of different socially and culturally divergent groups on the other. Moreover, it calls for an intergenerational perspective. For while central beliefs are relatively resistant to change within single individuals, they can change more easily - and in many instances obviously do change - across generations. The subject of intergenerational belief and attitude change has so far received little attention in the literature on political unification. Yet especially where, as is the case in the European Union, integration is itself
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
77
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
a generation-transcending process, those who want to explore its psychological implications must consider the living as well as the dead. ACKNOWLEDGEMENTS I am grateful to Jean Laponce for, a long time ago, first getting me interested in the theme of this essay. Thanks also to Matthijs Bogaards, Lars-Erik Cederman, Anne Deighton, David Glass, Susan Henders and William Wallace for their comments on earlier drafts.
NOTES 1. See, for example, Kenneth McRae, Conflict and Compromise in Multilingual Societies (Waterloo, Ontario: Wilfrid Laurier University Press, 1983). 2. Brian Barry, Democracy and Power: Essays in Political Theory, Vol. 1. (Oxford: Oxford University Press), p.5. 3. For instance the classic essay by Phillip Converse, 'The Nature of Belief Systems in Mass Publics', in David. E. Apter (ed.), Ideology and Discontent (New York: The Free Press, 1964). 4. See Dimitris N. Chryssochoou, 'Democracy and Symbiosis in the European Community: Towards a Confederal Consociation?' West European Politics, Vol.17, No.4 (1994), pp.1-14; Simon Hix, 'Approaches to the Study of the EC: The Challenge to Comparative Politics', West European Politics, Vol.17, No.1 (1994), pp.1-30. Much of this interest was stimulated by the innovative theoretical work of Paul Taylor, in which he employed consociational theory to arrive at a re-conceptualization of the integrative process in Europe that is at variance with classical accounts in the intergovernmentalist, neofunctionalist, and federalist traditions alike. See Paul Taylor, 'The European Community and the State: Assumptions, Theories and Propositions', Review of International Studies, Vol.17, No.2 (1991), pp.109-25. 5. At first glance my assumptions appear at variance with those of the consociational 'founding fathers' in yet another respect. In line with their focus on the domestic situation in existing states, they typically cast their model as a means of preserving political and economic institutions that had already come into being. In the EU, by contrast, the task is often depicted to be still primarily one of institution-building. Yet this difference starts to look merely like one of degree once one conceives of international integration as an incremental process. Then, the 'keeping together' and 'bringing together' objectives overlap: As soon as the first step towards integration has been taken, the task ahead is not just one of initiating further steps but also one of preserving what has already been achieved, not least of course by securing a necessary measure of public acceptance for it. 6. Arend Lijphart, The Politics of Accommodation: Pluralism and Democracy in the Netherlands (Berkeley, CA: University of California Press), p. 186. 7. Ibid. 8. Lijphart, The Politics of Accommodation; also Lijphart, Democracy in Plural Societies: A Comparative Exploration (New Haven, CT: Yale University Press, 1977). 9. Gabriel A. Almond, 'Comparative Political Systems', Journal of Politics, Vol.18, No.3 (1956), pp.391-409. 10. M. Elaine Burgess, 'The Resurgence of Ethnicity: Myth or Reality?' Ethnic and Racial Studies, Vol.1, No.3 (1978), pp.265-85. 11. Paul Taylor, 'The Concept of Community and the European Integration Process', in Michael Hodges (ed.), European Integration: Selected Readings (Harmondsworth: Penguin, 1972), p.205. 12. Donald J. Pachula, 'The Resurgence of Ethnicity: Myth or Reality?' Ethnic and Racial Studies, Vol.1, No.3 (1978), pp.265-85. 13. Jean A. Laponce, 'Canada, Switzerland, and Talcott Parsons', Queens Quarterly, Vol.99, No.2 (1992), pp.267-79.
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
78
NATIONALISM & ETHNIC POLITICS
14. Taylor, 'The European Community and the State'. 15. Muzafer Sherif, Group Conflict and Co-operation: Their Social Psychology (London: Routledge & Kegan Paul, 1967). 16. Henri Tajfel, Human Groups and Social Categories: Studies in Social Psychology (Cambridge: Cambridge University Press, 1981); see also Michael A. Hogg and Dominic Abrams, Social Identifications: A Psychology of Intergoup Relations and Group Processes (London: Routledge, 1988). 17. In terms of their implications those findings bear some affinity to Freud's concept of the 'narcissism of minor difference'. In Civilization and its Discontents Freud argued that proximity between groups could foster hostility and increased self-differentiation along group lines among their members (Sigmund Freud, 'Civilization and its Discontents', in The Penguin Freud Library, Vol.12, [London: Penguin, 1991], p.305.) In addition, Freud saw the shared pursuit of enmity towards out-groups to invigorate 'libidinal ties' among the members of the in-group and thereby to strengthen the in-group's cohesion (see Freud's 'Group Psychology and the Analysis of the Ego', in The Penguin Freud Library, Vol.12, pp.95-178). For a discussion of group boundary maintenance techniques from an anthropological perspective see Anthony P. Cohen, The Symbolic Construction of Community (London: Routledge, 1989). 18. For a good overview see the contributions in Robert P. Abelson et al., Theories of Cognitive Consistency: A Sourcebook (New York: Rand McNally, 1968). For a more recent example see Claude M. Steele, 'The Psychology of Self-Affirmation: Sustaining the Integrity of the Self, Advances in Experimental Social Psychology, Vol.21 (1988), pp.261-302. Steele combines dissonance theory with self-perception and self-affirmation models. 19. Milton Rokeach, Beliefs, Attitudes, and Values: A Theory of Organization and Change (San Francisco, CA: Jossey-Bass, 1968), p.1. 20. Milton Rokeach, The Three Christs of Ypsilanti: A Psychological Study (London: Barker, 1964), p.26. 21. Rokeach, Beliefs, Attitudes, and Values, p.6. 22. Rokeach identifies a second category of core beliefs which, however, is not directly relevant to the present discussion. It contains beliefs which run counter to widely accepted reality and are thus often treated as delusions or hallucinations. Though these beliefs are at the core of the belief system, they are grounded in zero (rather than unanimous) social consensus. 23. If related to the subject of social reality maintenance, this understanding of core beliefs has a strong bearing on the more sociologically oriented notions of social consensus and reification (or naturalization). See Pierre Bourdieu, Raisons pratiques: sur la théorie de faction (Paris: Éditions du Seuil, 1994). Also see Peter Berger and Thomas Luckmann, The Social Construction of Reality: A Treatise in the Sociology of Knowledge (London: Penguin, 1991). 24. Rokeach, Beliefs, Attitudes and Values, pp.6-7. 25. Ibid., p. 10. 26. Ibid., pp.7-8. 27. Rokeach, The Three Christs of Ypsilanti, p.20. A further but secondary determinant of the impact which a belief change has on the belief system as a whole is likely to be the extent to which the belief in question becomes altered. A slight adjustment to a belief inflicts a lesser jolt to the belief system than its complete overhaul or replacement. The extent to which a belief becomes changed depends primarily on the magnitude of the belief violation which induces the change. 28. Most derived beliefs, however, are unlikely to change at the suggestion of any one positive authority. For rather than being linked to a single referent these beliefs are generated and sustained through continuous interaction with what Berger and Luckmann call a 'chorus' of reality-confirming agents which extends into the 'totality of the individual's social situation' (Social Construction of Reality, p.171). Hence, for those beliefs to change the social base in which they are anchored must itself become transformed, or, alternatively, the belief holder must become alienated or cut off from his or her social base. 29. Ibid., p. 194. 30. Further, there can be situations in which the violated belief is of the same centrality as that
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
31.
32.
33.
34. 35.
36.
37.
79
which would need to be altered if a change of the initial belief were to be avoided. From what has been discussed thus far it can be inferred that in those situations the individual's reaction to the belief violation likely depends on secondary criteria such as the presence of related and reinforcing beliefs and their position on the centre-periphery axis. See Otto Letze, Deutsch-Französischer Jugendaustausch: Organisation und Interaktion (Doctoral diss., University of Tübingen, 1986); John E. Farquharson and Stephen C. Holt, Europe From Below: An Assessment of Franco-German Popular Contacts (London: George Allen & Unwin, 1975). In this context one must distinguish between, on the one hand, the objective of improving mutual perceptions between different populations (as it guided for instance the Franco-German programme), and that of initiating a process of cultural amalgamation and/or cross-cleavage creation on the other. The former objective is compatible with the consociational blueprint and may even be required for its successful implementation, while, as we saw, the latter aim is not. See Ronald Inglehart, 'Trust Between Nations: Primordial Ties, Societal Learning and Economic Development', in Karlheinz Reif and Ronald Inglehart (eds), Eurobarometer: The Dynamics of European Public Opinion: Essays in Honour of Jacques-Rene Rabier (London: MacMillan, 1991), pp.145-85. A good example of such 'instant animosity' is the deterioration of public attitudes in Britain towards 'the Argentineans' during the Falklands (Malvinas) War. These moved from a condition of previous indifference to one of acute hostility in a matter of weeks. In this instance, it was the British tabloid press which assumed an important and rather legendary role as 'belief transmitter' by helping to 'psyche up' popular sentiments before and during the British reclaiming of the islands. See Robert Harris, Gotcha! The Media, the Government and the Falklands Crisis (London: Faber and Faber, 1983). With the entry into force of the Maastricht Treaty, what had commonly been referred to as the 'European Community' or the 'Common Market' became part of the newly created 'European Union'. For reasons of consistency I use the latter name throughout. To keep this discussion as simple as possible I do not here distinguish between 'state' and 'nation' (or 'state' and 'cultural community'). In reality, of course, those EU member states that are themselves divided into different ethnic, linguistic or national sub-groups (such as Belgium and Spain) should not be treated as unified socio-cultural units. The way in which the members of those sub-units react to the prospect of Union involvement in policy areas such as cultural policy and education is likely determined by the interplay of a range of factors. They include the extent to which competences in those areas have been devolved to their respective sub-unit level authorities (where those exist) by their central governments, to which they see those powers as critical in defending the social and cultural integrity of their respective sub-units, and not least the extent to which those sub-units are in fact their primary units of identification. All these factors help determine how much the sub-unit members 'stand to lose' as the result of a potential Union involvement in socially and culturally sensitive policy areas. In the present context I cannot further pursue this question. Compared to four years prior, there was a general decline in support for Union competences in most areas. This drop was disproportionately strong with regard to education, social and media policy, i.e. the very fields in which there had been a rejection from the outset. See Eurobarometer No.33 (Brussels: European Commission, 1990). From this perspective the relatively ambivalent state of popular opinion towards the single European currency noted earlier appears in a different light from that in which it is often portrayed. Rather than necessarily mirroring Europeans' alleged inability to make up their collective mind on whether a single currency will yield the kind of utilitarian benefits promised by the EU and many of their own governments, such an ambivalence in attitudes may also have stemmed in part from the ambivalent status of money in terms of the two issue categories. On the one hand, money has a strong utilitarian dimension, as do the policy issues to which it is linked (e.g., interest rates, inflation). On the other hand, as has often been noted, in its actual physical manifestation in the form of notes and coins (which in turn may carry images of national flags, poets, kings and queens, heroes, landscapes, and so on) money can also constitute an important 'national identity'-related artefact. In this capacity it can represent a symbolic marker of the national community in question which, like very few
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
80
38. 39.
40.
41.
42. 43. 44. 45.
46.
47.
NATIONALISM & ETHNIC POLITICS others, is apt to penetrate into the everyday lives of most citizens. This dual character of money, I believe, shone through in the level of popular backing for the introduction of a Euro-currency (and by implication of course for the abolition of national currencies). Weaker than for Union involvement in policy areas that do not posses a comparable symbolic and 'identitive' dimension (such as R&D and environmental protection), it was nevertheless stronger than support for supranational involvement in policy areas whose 'identitive' dimension was also strong yet not accompanied by a plausible utilitarian dimension of similar strength, such as cultural policy and education. Eurobarometer No.43 (Brussels: European Commission, 1995) A flavour of the Danish debate is provided by some of the essays in Morten Kelstrup (ed.), European Integration and Denmark's Participation (Copenhagen: Copenhagen Political Studies Press, 1992). For a comparative treatment see Ulf Hedetoft, National Identities and European Integration From Below: Bringing the People Back In, Working Paper No.54 (Cambridge, MA: Minda de Gunzburg Center for European Studies, 1994). Also see Mark Franklin, 'Uncorking the Bottle: Popular Opposition to European Unification in the Wake of Maastricht', Journal of Common Market Studies, Vol.32, No.4 (1994), pp.455-72 for a general overview. The three questions were as follows: (1) 'In general, are you for or against efforts being made to unify Western Europe: If for, are you very much for this, or only to some extent? If against, are you only to some extent against or very much against?' (2) 'Taking everything into consideration, would you say that [your country] has on balance benefited or not from being a member of the European Community (Common Market)?' (3) 'Generally speaking, do you think that [your country's] membership of the European Community (Common Market) is a good thing, a bad thing, or neither good nor bad?' Note: to compute net support for question 1, the 'very much' and the 'to some extent' answers were added up on each side respectively before the total of the 'not benefited' was subtracted from that of the 'benefited' answers. As for question 3, the 'neither good nor bad' answers were left out of the calculation. In the autumn of 1993 only 74 per cent still professed to be aware of the Maastricht Treaty. One year later this had fallen to 67 per cent. Source: Eurobarometer No.43. It is plausible to assume, however, that many of those whose specific memories of the Maastricht Treaty had faded nevertheless retained some of the more diffuse images and impressions of the Union which they had acquired during the Maastricht debate, and that those impressions continued to influence their stance towards the integrative process. But for some work in this direction see Hedetoft. In particular, this argument singled out the increase in the scope of qualified majority voting in the Council of Ministers and the delegation of national powers to European institutions with dubious accountability, such as the planned European Central Bank. See, for example, Martin Baldwin-Edwards and Martin A. Schain (eds), The Politics of Immigration in Western Europe (London: Frank Cass, 1994). On the link between attitudes and behaviour see Charles M. Judd and Jon A. Krosnick, 'The Structural Bases of Consistency Among Political Attitudes: Effects of Expertise and Attitude Importance', in A. R. Pratkanis et al. (eds), Attitude Structure and Function (Hillsdale, NJ: Lawrence Erlbaum, 1989); Jon A. Krosnick, 'Attitude Importance in Social Evaluation: A Study of Policy Preferences, Presidential Candidate Evaluation, and Voting Behavior', Journal of Personality and Social Psychology, Vol.55, No.2 (1988), pp.196-210; Russell H. Fazio and Carol J. Williams, 'Attitude Accessibility as a Moderator of the AttitudePerception and Attitude-Behavior Relations: An Investigation of the 1984 Presidential Election', Journal of Personality and Social Psychology, Vol.51, No.3 (1986), pp.505-14. Tobias Theiler, 'The European Union and the "European Dimension" in Schools: Theory and Evidence', Journal of European Integration (forthcoming). On the question of élite motivations in relation to social and cultural 'encapsulation' see Taylor, 'The European Community and the State'. This imperfect disentangleability is implicit in the dual status of money as discussed earlier (Note 37), and in the long-lasting debate on how to distinguish 'goods and services' that are of a cultural nature from those that are not. It also underlies frequently expressed doubts as to whether an intensification of the pan-European democratic process — for instance through
Downloaded By: [University College Dublin] At: 18:24 18 April 2008
CONSOCIATIONALISM AS A MODEL OF POLITICAL UNIFICATION
81
the introduction of direct Europe-wide referendums - would be a realistic option, given the relative dearth of shared social and cultural reference points among Europeans. To be sure, the consociational model itself has features that in part seem designed to make allowance for this interconnectedness of the different dimensions of integration. The most obvious example is its demand that the different segments retain not only a high degree of social and cultural but also political autonomy, the two being depicted as attendant conditions. 48. Carol L. Schmid, Conflict and Consensus in Switzerland (Berkeley, CA: University of California Press, 1981).
