Have "consociational" forms of parliamentary government ceased to exist in Western Europe but is the majoritarian model on the way to becoming more

Have “Consociational” Forms of Parliamentary Government Ceased to Exist in Western Europe but is the Majoritarian Model on the Way to Becoming more “Cooperative”? JEAN BLONDEL European University Institute, Florence University of Siena F. BATTEGAZZORE University of Pavia WP núm. 223 Institut de Ciències Polítiques i Socials Barcelona, 2003 2 The Institute of Political and Social Sciences (ICPS) is a consortium created in 1988 by the Barcelona Provincial Council and the Universitat Autònoma de Barcelona, the institution to which the ICPS is officially linked for academic purposes. “Working Papers” is one of the collections published by the ICPS, specialising in the publication of work currently being undertaken by social researchers –in the author’s original language– with the aim of facilitating their subsequent scientific debate. Inclusion of work within this collection does not prohibit its future publication by the author, who maintains full rights over the article. The work in question cannot be reproduced without the permission of the author. Publication: Institut de Ciències Polítiques i Socials (ICPS) Mallorca, 244, pral. 08008 Barcelona (España) http://www.icps.es © Jean Blondel Design: Toni Viaplana Printer: a.bís Travessera de les Corts, 251, entr. 4a. 08014 Barcelona ISSN: 1133-8962 DL: 3 A number of developments which occurred since the 1990s in Western Europe suggest that those parliamentary governments traditionally referred to as “consociational” have been moving towards “majoritarianism”: this has been the case in all four countries which have long been regarded as embodying this form of government, together with Switzerland –which does not have a parliamentary government, however–. The Netherlands, Belgium, Austria and Italy. In rapid succession, from the middle of the 1990s, these four countries have come to be ruled by governments which have eschewed the kind of “power-sharing” arrangements which characterised “consociationalism” and have moved towards a “government and opposition” form of rule more characteristic of the majoritarian model. Meanwhile, but in a less dramatic fashion, changes which are occurring in some of the “majoritarian countries” suggest that there may be some “softening” of “majoritarianism” where it has traditionally prevailed. It is difficult to interpret these movements in the context of the only framework which has been elaborated to distinguish between types of parliamentary governments, that of Lijphart. This is despite the fact that that framework opened up in a pioneering manner a line of analysis on these governments. Perhaps the main reason is that the way the framework is structured does not allow for a mechanism to register dynamics. This is also because the distinction is based on the opposition between “majoritarianism” and “consensus democracy”, a distinction which leads to a contrast, not to dimensions. Moreover, there is no precise definition of the two contrasting types, while the criteria used to categorise these types are a series of dichotomised variables designed to account for the fact that a particular system of government falls within one or the other of the two types. These categorisations are thus in the nature of “independent variables” while the type of parliamentary government is the dependent variable. Given that some problems arise about the way the two broad types of systems are characterised, it is important first to revisit Lijphart's model in order to see where it might have to be somewhat modified in order to be better 4 adjusted to the conditions prevailing at least in Western Europe. Only then will it become possible to examine the extent to which Western European parliamentary systems appear to be all moving towards what might be described as a more “cooperative” form of majoritarian government. I Three aspects of the distinction between “majoritarian” and “consensus” governments given by Lijphart need to be examined closely. First, the two types must be more precisely defined. Second, there should not be a contrast between two types, but there should be dimensions and indeed two dimensions. Third, there is a need to have a further look at some of the variables used to determine the position occupied by a given parliamentary government as well as for a tightening of the operationalisation of these variables. Searching for Definitions of the Lijphart Types of “Majoritarian” and “Consensus” Parliamentary Governments Lijphart's undertaking aims at covering, especially in the 1984 volume, Democracies, but to a large extent also in the 1999 work, Patterns of Democracy, all types of democratic governments: as a matter of fact, however, these works have concentrated on types of parliamentary systems: and especially among types of parliamentary systems in Western Europe. 14 of the 21 country-cases examined in 1984 were Western European parliamentary systems; in 1999, these still constituted 18 of the 36 country-cases, the bulk of the others (13) being parliamentary governments outside Western Europe, principally in the Commonwealth; only four presidential systems and Switzerland are non-parliamentary. It is thus permissible to concentrate the attention here on the applicability of the findings to parliamentary systems and essentially to Western European parliamentary systems. Indeed, the model has had the decisive impact of moving analyses away from the rather simplistic notion that parliamentary governments had all to be of the “Westminster” type if they were to function effectively. As this paper aims at examining changes which have been taking place among Western European 5 governments in the 1990s and in the early part of the twenty-first century, it confines itself to an examination of Lijphart's model in the context of these governments –Switzerland being naturally excluded from that analysisl, since it is not of the parliamentary type. Somewhat surprisingly, no precise definition of “majoritarian” and “consensus” government is given by Lijphart in either volume. The only reference to what might constitute a definition is provided at the beginning of Chapters 2 (at p. 21 in the 1984 earlier volume on Democracies which was primarily devoted to parliamentary governments) or 3 (at p. 31 in the 1999 volume on Patterns of Democracy), on "The Consensus Model of Democracy". In what is in reality merely a “presentation” of the two types, it is stated that "[t]he majoritarian interpretation of the basic definition of democracy is that it means ‘government by the majority of the people’" (1984, 21 and 1999, 31). This is scarcely a definition, would it only be because the “majority” which is relevant in the context of representative government is the majority in parliament (and typically in the lower chamber only if there are two chambers): such a majority does not necessarily coincide –indeed in practice rarely coincides– with a majority of the people and even of the voters, given the vagaries of electoral systems. Conversely, it is not stated either what “consensus” precisely refers to. The expression “consensus” presumably should not be taken literally, as it is unlikely to entail that all be represented in the government and agree with what that government does. There has to be a “cut-off point” somewhere. While never discussing what “consensus” precisely entails, Lijphart prefers it to the term consociational which he previously used, in part because it is "shorter – and easier to pronounce–" (1984, xiv), but more seriously for two reasons. One is that the majoritarian system is taken as the point of departure, not the “consociational” system, as had been the case in previous works: it is difficult to see why this change of standpoint makes the “other” system “consensual” rather than “consociational”. The other reason is that the “variables” now being adopted to account for the two models are "not coincident" with those previously used to define consociationalism: yet this might simply mean that 6 these new variables account better for consociationalism than the previous ones. Lijphart is indeed right, as we shall see, to use the term “consensus” rather than the term “consociational” to refer to the two types which he aims at describing, although he is not right to have jettisoned the term consociational altogether: we shall see that this term does indeed help to refer to a sub-set which needs to be considered. Yet there remains a major problem if one adopts Lijphart's terminology: we need to know what precise meaning is given to the word “consensus” in the context of the activities of a government. Is it simply that one is referring to governments which have a larger base than a “bare majority”? It manifestly means something “more” in Lijphart's interpretation since he quotes Arthur Lewis who speaks of "all who are affected ...(having) the chance to participate in making (the) decision" (31). Lijphart indicates later that the idea is to “include” rather than “exclude” (33). Yet exactly how many should be included or what should the process of decisionmaking be to warrant stating that a particular government operates on a “consensus” basis is not clarified. An attempt was made by Sergio Fabbrini, in his work Quale democrazia? (Rome: Laterza, 1994) to remedy to some extent to this lack of definition. The expressions used are not the same as Lijphart's: Fabbrini refers to “competitive democracies” instead of majoritarianism' and to “consociational democracies” instead of “consensus governments” but he does aim at covering the same distinction as Lijphart's. "Competitive democracies are those which tend to stress the opposition between alternative party options..... [consociative democracies are those] which tend to stress the agreement among the main party options" (1994, 15-16) (emphasis in the text; my translation). The aim in one case is to “exclude”, says the author who uses there the same formulas as Lijphart, while in the other it is to “include” (ibid.). One might tighten the definition a little and return to using Lijphart's expressions by stating the following. A majoritarian system of government is one in which democracy is conceived in terms of a conflict between “blocs”, typically of relatively equal size, the assumption being that there will be 7 alternance between these blocs over time. It is worth noting that such a system stresses the autonomy of politics with respect to the society in general. A consensus-type system of government is one in which democracy is conceived in terms of wide agreements among the political parties, these being regarded as representing the various social forces in the society. The word “consensus” is kept, but it is qualified by the reference to “wide” agreements: it is not posited that everyone has to be (above all ostensibly) in accord with the line which is adopted, as this is almost never the case, when one is referring to the activities of a whole government, although this may occur for particular decisions. The Need for a Two-Dimensional Approach Lijphart is absolutely clear that few countries, if any, fall in one or the other of the two types, majoritarian and consensus. He notes this point from the start. At the outset of the Chapter on the majoritarian model, he states that Britain and New Zealand are the “prototypes” (1999, 10) and early in the Chapter on the “consensus” model, he states that the "examples used to illustrate" that model are Switzerland (not a parliamentary system), Belgium and the European Union (not a parliamentary system and indeed more akin to the Swiss system). The two models have therefore to be regarded as “ideal-types” but also as extreme positions in a continuum. The countries which constitute “intermediate cases” are therefore very important: they constitute the large majority. Yet it is frankly impossible to accommodate these cases unless there is indeed a “continuum” along which these cases can be distributed. The problem is that the distinction between “majoritarian” and “consensus” parliamentary government leads to a dimension only if the word “consensus” is regarded as meaning “very large majority”. Such a dimension would identify types of “power-sharing” ranging from “bare majority” to “unanimity”. A difficulty would still remain in relation to “minority governments” of which there are, as we know, a large number (one sixth of the total for Western Europe, as indicated by Lijphart (1999, 98)): this type of occurrence is too frequent to be dismissed as being “exceptional”; but one has then to explain how parliamentary governments can be of a minority character, 8 minority meaning operationally that the parties represented in the government do not have a majority of seats in parliament. Yet the “power-sharing” dimension which we have just identified not only does not explain why minority governments exist: it does not correspond either to the idea of “inclusion” which Lijphart considers to be a key aspect of the type of parliamentary government which he refers to as “consensual”. “Inclusion” means that as many as possible are involved in the process of decisionmaking: it is distinct from “power-sharing” as it refers to the extent to which the views of “others” are listened to, whether these “others” share power or not. This is why such a concept can help to cope with minority governments. If such governments exist, that is to say if there are parliamentary governments in which the parties represented in the government do not have the majority in parliament, this must be because an agreement exists between the parties in the government and some of the parties which are not in the government. The agreement may be temporary and precarious or well-structured and longlasting: but it has in any case to integrate in some fashion segments of the legislature which otherwise would not support the government. This may not result in “consensus” in the strong sense of the word; but it does constitute a move going beyond “pure” majoritarian rule, a move which might be minuscule if the support needed is tiny or be very large if the support needed is substantial. In this way, one comes to be concerned with something different from “power-sharing”: one comes to be concerned with the way the government behaves in relation to bodies other than itself. If one then enlarges the notion and considers the problem in terms of “inclusion”, as the word “consensus” invites us to do, what one comes to be concerned with is no longer just with those parties represented in the government, but also with the relationship between the government and the rest of the political public. One is concerned with a mode of behaviour and not only with the proportion of those who are in government. This mode of behaviour can be “adversarial”: it can simply ignore the others and state that what the government wants has to be the law; but it can be “cooperative” and try at least to find out whether some accord can be 9 found. It can also occupy one of an infinite number of positions between these two extremes, for instance if the government attempts, but only up to a point, to listen to others and take the comments into account. It is therefore right to refer to that activity as aiming towards “consensus”. Overall, this means that, alongside the “power-sharing” dimension, there is a “mode of behaviour” dimension and it is that second dimension which is concerned with the question of “consensus”. Only if one recognises that two dimensions are involved can one resolve the problem posed by the contrast between “majoritarianism” and “consensus” government. For, if there is only one dimension, “consensus” has to be regarded as meaning “large majority”: one can then accommodate the many cases of parliamentary governments which are not of one type or the other along that dimension, but it is a vast exaggeration to refer to “consensus” as the “spirit” of “inclusion” is not taken into account. Or one sticks to the idea of “consensus” meaning really “consensus” –at any rate meaning real attempts at building consensus–, but one is then either confronted with a contrast between two incompatible types or one has to introduce a second dimension concerned with the manner in which the government deals with the political class. This last solution is the only realistic way out of the problem. As a first and simplified approximation of these two dimensions, one can distinguish among four quadrants in which real-world parliamentary governments can be located. These are a “bare majority adversarial” quadrant, a “bare majority cooperative” quadrant, a “large majority adversarial” quadrant and a “large majority cooperative” quadrant. The first of these quadrants corresponds to what might be referred to as “true Westminster” parliamentary governments. The second corresponds to governments, such as those of Scandinavia, of Finland in most cases, and to an extent of Germany, in which the “bloc” in power does attempt (or has) to build a degree of consensual arrangement with the forces of the opposition. The third and the fourth quadrants correspond to what are usually referred to as “consociational” governments: power-sharing is substantial but, in one case, there is a posture of imposition on the part of those in power, while in the second, the mode is at 16 after World War 2, a government led by a Socialist and composed principally of socialists and liberals run the country for four years, between 1950 and 1954, only to be replaced, after that coalition lost the election, by a single-party Christian Social party government. Admittedly, both before World War 1 and between the World Wars, the party system had become more complicated, in part, between the wars, by the emergence of a party whose leader was an admirer of fascists and nazis. Yet up to the 1960s “majoritarianism” remained a feature of Belgian parliamentary government. The situation was altered when the main issue ceased to be economic and religious and centred on the language-cum-community problem. The emergence of that issue led to demands for profound institutional changes, changes which could not be achieved without constitutional amendments and therefore without the agreement of a large majority in parliament. The three major parties –which meanwhile had split along language lines– had therefore to work together if reform was to take place, a process which ended in 1993 only, when the new federal constitution came into force. During the intervening period, political arrangements started to resemble the mode which was characteristic of The Netherlands: specifically, the Christian parties (both Dutch– and French-speaking) became the combined “pivots” of the system. Belgium was therefore naturally regarded as having a consociational type of parliamentary government. Yet it is as if the introduction of the new constitution in 1993 was closing the “parenthesis” of the consociational period. Only five years later, in 1998, a government headed by a liberal and composed of Liberals, Socialists and Greens –thus replicating, almost fifty years later, the Socialist-Liberal government of 1950– was constituted without the Christian parties, which were relegated to the opposition, after having been regarded as indispensable for the working of the Belgian parliamentary government for a generation and that government handsomely won the 2003 General Election: it was reconstituted, this time without the Greens.. Unquestionably the formation and subsequent success of the non-Christian coalition does indicate that Belgium has returned to majoritarianism. Whether the government will practice a cooperative form of 17 parliamentary government needs to be determined, as, in the past, consociationalism in Belgium was appreciably more adversarial than its counterpart in The Netherlands. Austria. More than any other Western European parliamentary country, Austria was the country of the “grand coalition”. The nearly two decades of single party government between 1966 and 1983 and the small SocialistLiberal coalition of 1983-1986 seemed to be a mere hiatus when the government returned to operate on the basis of the “grand coalition” formula which seemed to have served the country well during the early postwar decades. There had been a number of reasons for adopting that formula. The desire to avoid the huge interwar tensions between the parties which had indirectly led to the absorption of the country into Germany in 1938 played an important part, together with the need to show unity in the face of occupation in 1945, particularly in the face of Soviet occupation, as well as, once the State Treaty of 1955 had been signed, the need to maintain the neutrality of the country. The result was a form of consociationalism which, in the Austrian case, given the combined electoral strength of the two main parties in the 1950s and 1960s –over 90 percent– meant indeed something approaching complete “consensus”. The single party government of the Populist party from 1966 to 1970 and the Socialist government led by Kreisky between 1970 and 1983 seemed to be a mere hiatus in that, despite the fact that party conflict was severe on occasion, for instance over the question of nuclear energy, much of political life continued to be regulated on the basis of the “proporz” principles which had been elaborated in the early postwar period and which included in particular a complex but comprehensive “division of the spoils” among the supporters of the two main parties. The government came to be in the hands of one party only, but the rest of the political system continued to be operated on a consociational basis. In a sense, the first sign of change resulted from the experiment with the “small” Socialist-Liberal coalition of 1983-1986, both because this arrangement 18 meant that the Socialist party was looking for a possible alternative ally and because the Liberal party, hitherto considered as “beyond the pale” as a result of its original dubious credentials –it was regarded as the party of the “fascists”– had become viewed as worthy of participating in the government. A breach in the “proporz” system had been made: party conflict was beginning to be played differently. The experiment ended when the Liberal party seemed once more, under Haider, to be “unacceptable”: this led to the re-establishment of the “grand coalition” in 1986. Yet, by then, given the increased electoral successes of the Liberal party as well as given the emergence of small parties, notably the Greens, the support for the “grand coalition” at the polls shrank markedly while the “proporz” system was coming increasingly under attack. What Haider's Liberal party was demonstrating was not only that the coalition of the two main parties constituted an attempt to keep indefinite control on a large variety of spoils but also that a substantial proportion of the population was opposed to the very idea of consociationalism as the basis of political life. The result of the 1999 election was so favourable to the Liberals that it became impossible for the two traditional parties to continue on the basis of the classical formula. The formation of a Populist-Liberal government led to an outcry in many quarters, both in and outside Austria; it may or may not have been “politically correct” to allow Haider's party a share in the government. What the formation of that government did achieve, however, was to establish for the first time since 1945 the existence of a “bloc” system: it therefore marked the end of consociational rule, a development which was confirmed in 2002-3, when the government was based once again on the Populist and Liberal parties only. Whether Austrian majoritarianism will be adversarial or cooperative has to be determined gradually: Austrian consociationalism was indeed adversarial in that, under the “grand coalition” formula, the Liberal party was never regarded as constituting a “normal” party. With the main opposition being constituted by the Socialist party, it may well be that the governmental mode of behaviour will be less adversarial, but the presence of the Liberal party in that government is unlikely to constitute a factor of cooperation. 19 Italy. Although the Italian parliamentary system was almost as old as that of Belgium, its development in the nineteenth century was markedly less structured. The party system was fluid and practices such as trasformismo made it extremely common for elected representatives to switch from the opposition to the government side. A number of political leaders, Depretis and Giolitti in particular, were masters at the manipulation of parliament in the years preceding World War 1. The system ended brutally a few years after the end of the conflict, as the traditional political class proved unable to deal with a marked frustration in many quarters about the peace settlement and with a near-revolutionary situation in many industrial areas. The parliamentary system had few real supporters when Mussolini destroyed it. The re-establishment of parliamentary government two decades later in the aftermath of defeat took place in a climate of profound political and social divisions: these were restrained only as a result of the determination of all the political leaders and of the Communist leaders in particular, the latter having probably realised that the Western allies and especially the Americans would not allow Italy to be taken over by the Communist party. A mode of political behaviour somewhat reminiscent of the Dutch formula at the time of World War I came to emerge as a result. What could be regarded as the “pillars” in Italian society, especially the two main ones, catholic and communist, were made by their leaders to accept to “live together”: a “broad-based” consociationalism' had been established. That “broad-based” consociationalism was to be quickly shattered by the emergence of the cold war, however, and only some of its elements remained and were able to re-surface temporarily in the wake of the waves of right-wing and even more left-wing terrorism which shook the country thirty years later, in the mid and late seventies. In between, a “limited” form of consociationalism developed, its elements being constituted by the Christian democrats, by far the largest component, and by a number of small parties, Republicans and Social democrats in particular, as well as eventually by the somewhat larger Socialist party. The Christian democrats did not have a majority; their electoral strength was indeed gradually declining: but, as the Communist party could not 20 be an alternative partner for the Christian democrats if Italy was to remain part of the Western alliance, the group of the parties which shared power had to remain broadly speaking the same and elections could not modify fundamentally the political equilibrium. “Limited” consociationalism was therefore a necessity; it was an adversarial form of consociationalism, at least until the late 1970s when the “historic compromise” was struck with the Communist party. Although “broad-based” consociationalism was then able to make a comeback, this was not for long, as, in a sense in a somewhat similar manner to what was occurring in Austria at about the same time, the “spirit” of the Italian political system began to be altered by the reassertion by the Socialist party of its autonomy under Craxi, a reassertion which was facilitated by the formation of the first governments not led by a Christian democrat in the early 1980s. Craxi succeeded in rendering credible the view that, as Mitterrand had done in France a few years earlier, the Socialist party could overcome the Communist party and, at that point, become a political force able to challenge the Christian democrats. Although this scenario was not to succeed, the seeds of majoritarianism has been sown: when the old party system collapsed in 1992, it became increasingly accepted by the political class that one of the reasons for the ineffectiveness and corruption inherent in that system was consociationalism. That view was propounded in a strident manner by the leader of the separatist Northern League; it was strongly put forward by Berlusconi, who had founded Forza Italia on the ruins of the Christian democrats, the Socialists and a number of small parties; it was adopted by many leaders of the Centre-left and Left, whether ex-Communists having reconverted themselves into the Party of the Democratic Left or ex-Christian democrats. The 1994 general election did indeed for the first time being about a sharp division between what was officially described as two “poles”. Berlusconi won, but only with the help of the Northern League which proceeded to defect six months later. Despite various palinodies which resembled the trasformismo of the past, the 1996 and 2001 general elections reinforced the move towards a 21 “two-bloc” situation: more strongly than the political class of The Netherlands, Belgium or Austria, the Italian political class had turned to “majoritarianism”, a majoritarianism which seemed likely to remain adversarial, in part because of the controversial figure of Berlusconi but also in part because of the communist past of many leaders of the Centre-left alliance. By the beginning of the twenty-first century, all the Western European parliamentary governments had thus moved away from power-sharing and had come to adopt the politics of “blocs”, on the understanding that general elections would be the mechanism which would decide which of the “blocs” would be in power for the duration of the subsequent parliament. If powersharing thus appears to have been abandoned, does it also mean that the move towards consensus politics is also on the way? An Uncertain Move Away From Adversarial Politics and Towards Cooperation We noted that the move away from consociationalism and towards majoritarianism which has taken place among Western European parliamentary systems at the end of the twentieth century does not appear to be necessarily linked to a move towards greater cooperation. Moreover, it might seem highly exaggerated, if not plainly wrong, to suggest that a move towards cooperation should be taking place among the three Western European parliamentary systems which have traditionally been majoritarian, Britain, Ireland and France, as well as among the three democracies of Southern Europe which emerged from authoritarian rule in the 1970s, Portugal, Spain and Greece. To take only one example, Thatcher's Britain in the 1980s was markedly more adversarial than Macmillan's Britain in the early 1960s. Yet there are both a general and a highly specific reason which appear to indicate that a move towards cooperation is in the offing, both in these six countries and in the “new” majoritarian countries. The general reason has to do with the current characteristics of “advanced democracies”. Somewhat paradoxically, at the same time as the “new” “robust” form of conservatism took roots in Britain and elsewhere, the move towards the “end of ideology” 22 was also strengthening: what Downs had predicted in 1957 and Kirchheimer (1966) had concretely noticed seemed to be occurring. Elections may have been based on strong conflicts, but these conflicts were no longer about the nature of the overall social system. Indeed, as the conservative wave made its way in the course of the late 1980s and the early 1990s, it seemed losing some of its strength and its supporters seemed losing some of their loyalty to the basic principles which the ideology aimed at implementing. Meanwhile, a series of “cross-bloc” conflicts, somewhat technical in character, in particular problems relating to the environment and to “globalisation” and its range of effects seem to be taking precedence over the more traditional topics of socioeconomic divisions. It may not be that an era of “good feelings” is emerging among the “blocs”; it is seemingly no longer evident, however, that, as in the 1980s, political conflicts in Western European countries are about the longterm goals which the conservative and socialist “blocs” may have once believed in. In the opening years of the twenty-first century, they seem to be about the way in which the “blocs” are able to manage their society, to render it economically more efficient while maintaining the fundamental features of the traditional features of “social justice”. Alongside what is a general move in “advanced political systems” in the direction of common –or at least less distinct– goals, developments more specific to Western Europe force greater cooperation between governments and the parts of the political class which do not belong to the governmental “bloc”. Such a trend is clearly the consequence of the existence of the European Union. As Switzerland, the European Union is a consociational body in which cooperation and indeed consensus are not just a fact but a necessity: countries as diverse as those which compose Western Europe –without mentioning those of Eastern Europe– could not be expected to forge a Union without, as Switzerland has done and does continues to do, being “constrained” to function on a cooperative basis. Yet a direct consequence of having to cooperate at the level of the Union is to have to cooperate at the national level as well: countries cannot present “their” case for a given policy in Brussels unless they are relatively united 23 internally. The government of the day has therefore to search periodically for proposals likely to be acceptable to, even if not formally accepted by, at least the most important fraction of the opposition. As a matter of fact, both the movement towards majoritarianism and the movement towards cooperation are in large part the consequence of the membership of the European Union. Majoritarianism is essential for the governments of the member-States as it enables these governments to make their case more forcefully since it provides the basis for strong leadership recognised as such beyond the borders of the country concerned. Cooperation within the country is also essential as it enables each government to make clear that the other “bloc”, the one which is in the opposition, were it to come to power, would adopt the same policy stand. This is institutionally manifest in the French case, where, especially since the late 1990s, “cohabitation” has made a strong mark on the approach of the country to the European Union as well as internally in relation to a number of key policies: but this is also the case in the other member-States, albeit in a less structured manner. The long-term tendency of Western European parliamentary governments may therefore be in the direction of “cooperative” “majoritarianism”. The move towards majoritarianism at the end of the twentieth century is undeniable: the only question is how quickly it will become fully established. It has taken place, though in a more or less “deliberate” manner, in all four countries in which consociationalism prevailed throughout all, most or at least much of the post World War 2 period. The existence of the European Union suggests that the move will continue, as it gives governments and their leaders the opportunity to see their authority being better recognised. The trend towards greater cooperation between government and opposition is less marked. Yet there is already a decline in the extent to which adversarial politics prevails in the “majoritarian” countries in comparison with the 1980s and early 1990s. The membership of the European Union can only have the effect of reducing hitherto strongly-felt divisions within the individual States and thus lead the government of each country to discuss with the opposition in a meaningful manner many of its key proposals. This may not 24 result, in the short-term at least, in the kind of cooperation with nongovernmental forces which characterises politics in Sweden or in Denmark. This does suggest, however, that a significant move will take place in that direction, the overall consequence being that, in terms of the manner in which parliamentary government operates concretely within the member-States, practices will be more similar than they were at any time among the countries of the “Old World”. Note 1. An earlier form of this paper appeared in Quaderni di Scienza Politica, (9), 2002, p. 226-251. References DOWNS, A.: An Economic Theory of Democracy. New York, NY, Harper, 1957. KIRCHHEIMER, O.: “The Transformation of the Western European Party Systems” in LA PALOMBARA, J.; WEINER, M. (eds.): Political parties and Political Development. Princeton, NJ, Princeton University Press, 1966, p. 177-200. LIJPHART, A.: Democracy in Plural Societies. New Haven, CT, Yale University Press, 1977. LIJPHART, A.: Democracies. New Haven, CT, Yale University Press, 1984. LIJPHART, A.: Patterns of Democracy. New Haven, CT, Yale University Press, 1999.