Party and Social Structure

Peter M Siavelis. Handbook of Party Politics. Editor: Richard S Katz & William Crotty. 2006. Sage Publishing.

Introduction: Political Parties and Society

Political parties provide the primary links between society and the state in most contemporary democracies. However, until the turn of the last century political parties were painted in a very negative light, with most analysts portraying them as self-interested factions that interfered with potentially ‘purer’ forms of representative democracy (Daalder, 1983: 3; Sartori, 1976: 3-12). It was not until after World War II that parties approached near-universal acceptance as necessary, functional, and legitimate democratic actors. However, as democracy becomes the rule rather than the exception world-wide, trust in political parties has eroded. What is more, citizen attachments to parties are on the decline both in advanced industrial democracies and in the developing world. Scholars have pointed to the diminishing importance of parties, suggesting that the citizen-party nexus has been cut (Dalton et al., 1984; Lawson, 1988; Dalton, 1999: 65-6). The cleavages that defined political parties for most of the post-war period are much less relevant. Further, cross-national survey data show decreasing confidence in political parties. While this is part of a more general decline in confidence in political institutions, the low esteem in which parties are held is of special concern given their traditional roles as the main interlocutors between the governed and those who govern. Has our conception of parties come full circle? Once again, are parties to be considered the oligarchic expressions of self-interested elites that get in the way of governing, or is this new attitude towards parties simply a result of underlying social change, which has transformed the qualitative, yet still essential, roles that parties play?

This chapter asks where the study of society-party connections has been, and where it is going. Underlying this question is the deeper issue of whether parties still perform (or can reassume) the varied functions traditionally ascribed to them in modern democracies, or whether a new interlocutor between citizens and the state outside the traditional party model might be on the horizon. The chapter begins with an analysis of the earliest literature dealing with society-party relations. It analyzes the evolution in our understanding of cleavages, from the generative cleavages that emerged during the foundation of national societies to those rooted in post-industrial value change. In light of the deep changes wrought by the fall of communist systems and the birth of the ‘third way,’ the essay then analyzes work which asks whether there is ‘a left left,’ and, assuming there is, how the left is likely to look in the future. It then explores the literature on so-called new parties, and closes with a discussion of potentially fruitful avenues for future research on society-party connections. The chapter concludes that while parties may be in decline (at least in terms of their traditional roles), the richness of the literature on parties is not, and indeed, there has been a convergence in theoretical work on society-party relations in recent years.

Formalism in the Study of Parties

At the turn of the twentieth century, the study of politics was virtually indistinguishable from that of history. Political science (if it could be described as such) was prescriptive, descriptive and normative. Yet it is no coincidence that it was in the study of parties that political science initiated its long and contentious divorce from history. The work of Michels and Ostrogorski on political parties can arguably be considered some of the first that are distinguishable from the descriptive, historical tradition (see Michels, 1959; Ostrogorski, 1964). Nonetheless, neither hid their disdain for political parties, which were viewed as overt and negative manifestations of social conflict. What is more, even when parties become an object of study, theorists focused principally on party structures and formal organizations (Michels, 1959; Duverger, 1954; Neuman, 1956). While the role of society was implicit in terms of speaking of the functions that political parties performed, or the reality that ‘oligargarchies’ tended towards the domination of citizens, few early studies explicitly addressed the connection between society and political parties.

Despite the dearth of explicit discussions of the society-party connection in early political science literature, the seeds of the focus and methods used today were planted by important precursors to modern theorists. First, in empirical terms, Daalder traces the roots of the study of the social bases of politics to Hume and his distinction between ‘parties of interest vs. parties of principle.’ He also notes that, decades before the development of the modern cleavage literature, traditional analyses spoke of party conflict in terms of ‘town vs. country, church vs. anti-clericals, one estate against another; and later of classes which were thought to be inevitably in conflict with one another’ (1983: 16).

Social Cleavages and Party Politics

The modern study of the social origins of parties is rooted in the notion that rather than simple associations of interest, parties emerge organically from deep-seated divisions within society. An important normative shift also set the stage for the more serious study of the social bases of political parties. The late 19th and early 20th century saw an evolution towards a conception of parties as natural and legitimate, especially with movements toward mass suffrage expansion. Transformations in the substantive and methodological bases of political science also underwrote changes in the study of parties. The two world wars, the crisis of democracy in Europe, and particularly the fruits of the Weimar Republic, drove home the reality that the formalistic, institutionally based and normative focus of the discipline failed to capture the essence of politics. Throughout the 1950s deeper questions about the social basis of politics grew out of the behavioral revolution. The advent of survey research provided the tools to better understand complex society-party relations, and to more effectively advance and test arguments connecting parties to social divisions.

While Lipset’s early work points out that ‘lower income groups vote mainly for the parties of the Left, while the higher income groups vote for parties of the Right’ (1960: 223-4) and that ‘[i]n every democracy conflict among different groups is expressed through political parties which basically represent a democratic translation of class struggle’ (1960: 221), it is really later work by Lipset and Rokkan that provides the classic and most complete elaboration of the connection between social cleavages and political parties (Lipset and Rokkan, 1967; Rokkan, 1970, 1975). Based on a comparative study of Western European countries, these scholars argued that processes of early national development and the industrial revolution divided societies, and that the resulting social cleavages became politicized in the form of political parties with the advent of modern democracy. In particular, they point to two successive revolutions during Western Europe’s long march to democracy: the national revolution and the industrial revolution. These revolutions prompted four primary lines of cleavage that shaped the development of European party systems: center-periphery state-church, land-industry and owner-worker. What is more, they maintained that party manifestations of these cleavages remain relevant after the initial impetus for their formation has disappeared, leading them to their now famous assertion that ‘the party systems of the 1960’s reflect with few but significant exceptions the cleavage structures of the 1920’s’ (Lipset and Rokkan, 1967: 50). Early research in large part confirmed Lipset and Rokkan’s findings. Rose and Urwin (1969, 1970) found remarkably consistent levels of electoral support for parties born from the cleavages explored by Lipset and Rokkan from the end of World War II to the 1970s.

While the notion that cleavages can be frozen remained an undercurrent in the literature on society-party connections, scholars allowed for variation among party systems based on the particular constellation of cleavages within different countries (including language, ethnicity, race) and whether or not these cleavages were cross-cutting (that is to say, non-reinforcing) or coincident (see Lipset, 1983). In general, cross-cutting cleavages were said to tend toward the formation of parties with more heterogeneous bases, because differences among groups are dampened by multiple loyalties and other social characteristics that individuals share. Dahl (1966: 378) cautioned against a simple application of this theory, arguing that the strength of the cleavage also determines the type of party system that ultimately emerges as cleavages become politicized.

While the literature on crisis and cleavages developed primarily out of the Western European and US experience, these ideas have also been applied to the study of party development in other places. Many early theories of party system development (such as Duverger, 1954) had little application in the developing world. However, Lipset and Rokkan’s general contentions regarding the relationship of crisis to party development were quite relevant. LaPalombara and Weiner (1966) underscored that crises of legitimacy, integration, and participation and how they were handled by emerging democracies were determinative in defining the party systems of developing countries. They continued by arguing that these crises are often telescoped, and that their timing and sequencing were central to defining the nature of emerging party systems. While replete with normative preferences for two-party systems and biases tied to notions of ‘modernity,’ this work is one of the first that systematically applies theories developed out of the Western European experience to non-European areas.

Nonetheless, cleavage analysis was not without its critics. Some suggested that the cleavage literature tended toward sociological reductionism by ignoring the multidimension-ality of the determinants of citizen-party identification. Others pointed to problems with the definition and measurement of concepts such as social class and religiosity (Urwin, 1973). Shamir’s (1984) time series analysis underscored that Lipset and Rokkan were mistaken to contend that party systems had been frozen in the first place, pointing to volatility during the years on which their study was based. In addition, Mair (1983) convincingly summarized the many studies which showed a good deal of electoral volatility beginning in the 1970s. Others noted that the treatment of political parties as dependent variables in the cleavage literature created two sets of problems. In normative terms, traditional cleavage analysis removed the influence of the voter, by suggesting that the determinants of party support rested solely with voters’ characteristics, and underplaying voter agency in electoral choice. In empirical terms, the early cleavage literature overlooked parties as independent variables, whose activities also shaped society (Urwin, 1973: 195). Finally, the cleavage literature, and particularly that which underscored freezing in partisan structures, left theorists with little capacity to explain change. It was precisely party system change that prompted a shift in the study of society-party relations, bringing into question widely held notions about the solidity of well-developed party systems.

Despite these criticisms and problems, cleavage analysis remains at the core of work on party-society relations. Contrary to what its most strident critics would suggest, this literature became quite nuanced as it developed, and has been applied successfully to case studies in new, transitional, and consolidated democracies. Also, the substantial contribution of cleavage analysis is evinced in how its concepts and terminology continue to be used in contemporary studies of democracy. For example, cleavages remain at the heart of Inglehart’s (1997) analysis of the materialist and post-materialist dimensions of politics and at the core of Kitschelt’s (1994, 1995) work on new parties and party change. Indeed, cleavage analysis is inherently useful because politics in modern societies is fundamentally a struggle between different groups seeking to obtain resources and to promote their values and visions for society. These differences spur conflict, and parties have historically been best equipped to structure debate on these conflicts and to negotiate the terms of debate that lead to public policy decisions.

The language of cleavages has also seeped into other areas of political inquiry beyond the study of party development. Cleavage analysis is central to Arend Lijphart’s (1977, 1980) work on consociational government, where cleavages such as religion, class, and language take center-stage. Finally, the tools of cleavage analysis are used in work on democratic transition and consolidation, where scholars have analyzed the salience of democratic/authoritarian and communist/post-communist cleavages (although not in as satisfying detail as they might, as noted below).

Notwithstanding the continued usefulness of the terminology and tools of cleavage analysis, by the mid-1960s it was clear that the widespread assumptions about Western political parties that grew out of it were experiencing fundamental change. First, in the United States, a dramatic increase in the number of independents in the 1960s and 1970s, and the later party realignment of the South, raised questions concerning the durability of the New Deal party system and the cleavages that spawned it. Theorists questioned whether these changes signaled the simple realignment and dealignment of the US party system, or were harbingers of its decomposition. Voters were increasingly independent, faction politics seemed to dominate, and voters lacked the clear identification with the issues that social cleavage analysis would predict (Burnham, 1970, 1975).

Second, from the 1960s Western European party systems were increasing volatile, with the appearance of new parties. Most almost immediately disappeared, but a substantial number survived. However, the diversity of party change across the continent made it difficult for theorists to systematically account for it, suggesting that the core assumptions underlying cleavage analysis and party system freezing were no longer valid. Once again, some noted that the age of the party had perhaps ended, and other modes of interest representation would prevail in post-industrial societies (Lawson and Merkl, 1988b: 3).

Third, growing affluence in the post-war period led theorists to predict an era of consensus politics and an end to ideology. However, the protest movements of the 1960s and 1970s shook up politics, and were accompanied by resurgence in ethnic and regional conflicts that confounded predictions of partisan stability in post-war politics.

Finally, the 1970s and 1980s unexpectedly ushered in a period where survey data and voting behavior showed that traditional assumptions about the connection between social class and party identification had been turned on their heads. The so-called ‘new right’ emerged as a conservative counterattack against the economic policies of traditional welfare states, but also, and perhaps more importantly, against the ‘new left’ which had grown out of the social movements of the 1960s. The ‘new left’ was new in its adoption of an increasingly ‘non-economic’ social agenda during the 1960s and 1970s in addition to its traditional orientation concerning social class and the preferred role of the state. Abortion, gender equality, gay rights, civil rights and environmentalism increasingly emerged as divisive issues that were more important than the traditional economic and social class cleavages that previously defined party politics. However, not only the salience of these issues, but also the kinds of party alignments they produced within mass electorates, were novel. Middle classes and highly educated professionals were more sympathetic to the social agenda of the new left, while it was blue-collar and lower-class voters who increasingly identified with the counterattack of the new right. As a result of these changes it became increasingly difficult to predict voter choice based on traditional social divisions. Franklin et al. (1992) found strong evidence that the analysis of social cleavages was less and less a useful tool to predict voter choice.

Demographic Shifts, Value Change, and Political Parties

During the 1980s these deep transformations prompted scholars to reassess the connection between social divisions and their party manifestations. While the language of cleavages remained an implicit tool of analysis, a consensus began to develop that the traditional connection between cleavages and parties (and primarily between social class and partisan identification) needed to be reassessed. Scholars began to analyze the sources of partisan change, and the consequences of partisan change in terms of the overall role that parties play in society.

With respect to the sources of change, scholars advanced two major streams of analysis. One viewed changes in partisan identification as a result of post-war demographic shifts (and primarily those produced by post-industrialism), while the other interpreted party transformation as a result of value change. While there is an analytical distinction between these arguments, they are certainly complementary and often overlap in their core assumptions and modes of analysis. Indeed, many tie value change to underlying demographic shifts.

The principal proponents of arguments based on demographic change pointed to widespread affluence and dramatic shifts in the occupational and social structures of advanced industrialized societies to account for change in the Western European and American party systems. They noted that previous analysts (Lipset and Rokkan, 1967; Campbell et al., 1960; Converse, 1976) were correct in underscoring long-term and stable party alignments among the electorate in the first decades following World War II. However, Dalton et al. (1984: 8) note that with the growth of affluence ‘[partisan change – rather than partisan stability’ became the ‘common pattern’ in advanced industrial democracies. The result of these changes was a trend toward the decomposition of electoral alignments and the fragmentation of the ‘socio-psychological bonds between voters and parties.’ As advanced industrialism eroded traditional relationships and community identification, interpersonal connections and institutional attachments became more fluid, with increased competition among competing social networks that divided citizen loyalties. These processes played out in the decomposition of electoral alignments in Western countries, which Dalton et al. argue is evinced in the fractionalization and volatility of party systems in the 1960s and 1970s.

Dalton et al. point to two principal potential avenues of change for these ‘decomposed’ party systems. The first is realignment, where there is a shift in the social bases underlying particular parties and/or party coalitions. With realignment, groups previously unaffiliated with a partisan option choose one, or those who have abandoned one party decide to associate themselves with another. Such realignments have been well accounted for in the American literature, but they are also a fundamental feature of European politics (Key 1959; Butler and Stokes, 1974; Rose, 1974). The other pattern for party systems in transition is one of dealignment, which occurs when a significant portion of the electorate dissociates itself from traditional parties. Dealignment is a process or an end stage, signifying either the fist stage in an electoral realignment, or the decline of political parties as the basic organizational units of politics (Inglehart and Hochstein, 1972).

A related strain of the literature on party transformation also recognizes post-industrial demographic and social changes, but focuses more on values and how value change results in shifts in partisan alignments (Inglehart, 1977, 1997). Inglehart, the most influential analyst of value change, argues that generational and concomitant value change account for the transformed social bases of parties. In particular, he argues that those who came to age during the post-war period did not experience the deprivation, depressions and the economic scarcity of their parents. This, along with enhanced educational opportunities, has transformed the fault lines of European societies. Among younger generations, who Inglehart contends have ‘post-materialist’ values, there is less concern with ideology and the economic role of the state, and much more concern with non-economic social issues such as abortion rights, equality, participation, the environment, and personal morality. These issues have displaced the typical ‘old politics’ that was important to the war generations, who had a ‘materialist’ orientation, and were much more concerned with economic stability and growth, domestic order, and military and social security. In essence, Inglehart argued that class-based political polarization had been replaced by value-based political polarization. Inglehart’s formula helps account for the new popularity of the ‘right’ among contemporary working classes, where the materialist message resonates more, better reflecting their value orientations.

Inglehart has built a large edifice of theory on the materialist-post-materialist distinction, with extensive analysis of cross-national survey data, which has now been undertaken in over 40 societies. Inglehart’s surveys, and particularly those that apply to major European countries, have consistently demonstrated a trend towards post-materialism.

Flanagan underscores a different kind of value change, though one could reasonably consider it a subdimension of the larger issues with which Inglehart deals (Inglehart, 1997: 122). In particular, he argues that along with sociocultural and economic transformation has come a decline in respect for authority, religion, and the work ethic as they have traditionally been understood (Flanagan, 1982). Traditional values have in large part been replaced by values more related to self-actualization – quality of life, leisure activities, a tolerance for distinct lifestyles, openness to new ideas, and nonconformity. Also, Flanagan builds on this work to more squarely criticize Inglehart’s argument and elaborate a new theory of value-driven politics. Flanagan suggests that while he generally shares many of the assumptions and findings of Inglehart’s work, a reliance solely on an analysis of the materialist-post-materialist distinction is insufficient. Flanagan introduces another libertarian-authoritarian axis that differentiates the new right from the old right, arguing for what amounts to three sets of value orientations. His category ‘libertarian’ is essentially the same as the one identified by Inglehart as ‘post-materialist.’ However, while some of the ‘old right’ is simply materialist, Flanagan contends that we must differentiate a new and distinct cluster of values that differentiates the old right – with its concern for strictly material values – from the new right -concerned with ‘security and order … respect for authority, discipline and dutifulness, patriotism’ and characterized by ‘intolerance for minorities, conformity to customs, and support for traditional religious and moral values’ (Inglehart and Flanagan, 1987: 1304).

Inglehart’s arguments and other analyses of realignment and dealignment were not without their critics. Clark and Dutt (1991) argued that rising levels of unemployment actually contributed to post-materialist values, a contention at odds with Inglehart’s arguments. Others took issue with the essence of Inglehart’s work, challenging his interpretation of the phenomena that led to an exploration of value change in the first place. Several scholars questioned whether dealignment has occurred in particular cases, or, more seriously, whether dealignment is even a real trend in broader terms. Keith et al. (1992) contended that the rise of independent voting in the United States was grossly overstated by proponents of the dealignment argument, and that most voters who identified themselves as independents actually leaned strongly towards one of the two established parties. Other scholars similarly challenged claims about partisan change in Europe (Mair, 1993; Bartolini and Mair, 1990).

Irrespective of differing in interpretations of their source, what do these changes mean for the future role of parties in Western democracies, and indeed, in new and developing democracies? Scholars answer this question in different ways, but three sets of responses tend to predominate. First, parties have ceased to be the dominant or most effective instruments of interest representation. Second, the role of ideology has been transformed, with consequences for the nature of the connection between societies and parties. Third, new sets of parties, both in traditional democracies and in new democracies, either play an increasing role or have the potential to displace traditional parties.

Parties, Society, and Representation

Transformations in the connections between society and parties have led some scholars to surmise that parties, given advancing technology, changing loyalties, and social change, may have become less than optimal agents of representation. Lawson and Merkl (1988b: 3) sum it up most directly by arguing that ‘it may be that the institution of party is gradually disappearing, slowly being replaced by new political structures more suitable for the economic and technological realities of twenty-first century politics.’ Scholars then suggest that neo-pluralist or neo-corporatist forms of representation may be the wave of the future. Indeed, for some, the predominance of the individual, and citizen capacity, trumps group representation altogether. For example, Bartolini and Mair (2001: 333) find that ‘citizens have an apparent capacity for direct action and no longer seem reliant on political mediation.’ Schmitter (2001) argues that parties have lost or abandoned their role in interest representation and aggregation. While parties continue to structure campaigns and elections and maintain some symbolic importance, they are much less imbedded in the overall governing process and in interest representation than they were in the past. This is particularly the case in new democracies where parties cannot rely on the habit of performing traditional roles, and the legitimacy that comes with time and success. Therefore, parties in new and reconstituted democracies must perform all of the standard functions assumed by political parties and face the simultaneous task of institutionalizing their own organizations (Montero and Gunther, 2002: 3).

While parties’ social bases are less identifiable, their functions transformed, and a good deal of party volatility exists, those who sound the death knell of parties overstate their case. Parties and other social organizations and forms of interest representation can coexist. Parties can take on varied and new functions, not necessarily competing with other social organizations. In addition, party organizations and party elites have the capacity to respond and adapt to social change, belying the image of parties as inflexible dinosaurs sometimes suggested in the literature. Indeed, scholarly consensus is emerging that the literature on party decline and disappearance was alarmist and inaccurate. In no democracies have parties been displaced as the major agents of interest representation. Several studies also suggest that parties have actually done a pretty good job of adapting to change, assuming new roles that allow them to function and often prosper (Tarrow, 1990; Aldrich, 1995). Despite voter cynicism, dealignment, fractionalization, and instability in their ranks, parties continue to be central representative actors, and will remain so.

Ideology, Parties, and Society: Is There a Left Left?

The fall of the Berlin Wall and the purported triumph of market capitalism profoundly transformed the left. Indeed, many scholars argued that ideology itself had ceased to be significant to the social bases of politics, a reality that would eventually be reflected in party systems. Across Europe, leftist parties have become less ideological, more pluralistic, and have accepted competitive markets as the key to growth. Statism is on the decline and the patterns of trade unionism and industrial production that provided the social bases for left-wing parties have been transformed. Lipset and Rokkan long ago established the left-right cleavage as the most important for understanding party systems in Western democracies. With the transformation of leftist politics around the world, and with class no longer the primary cleavage differentiating parties, the obvious question is whether there is a left left.

While macro-political processes and the fall of the Soviet Union certainly contributed to the left’s transformed role, deeper social and economic changes were also at play. Post-industrialism, increasing affluence, and the growth of the service and information economy eroded the left’s underlying social bases. Indeed, Debray (1990: 26) notes that the ‘[l]eft lost its coherent social base when it could no longer define itself as the mouthpiece of the “working masses”.’ Therefore, the question of whether there is a left left is intimately tied to the debate on the significance of social class.

Clark and Lipset (1991) argue that social class stratification has indeed weakened, resulting in a shift in how politics is organized. In large part echoing the arguments of Inglehart, they find that other value-based issues have become increasingly important. However, they also add individualism, technological shifts, and changes in the family and other sociocultural factors as additional elements that have led to decline in the importance of social class. They outline how new forms of stratification have replaced social class, and argue that indices of class voting have decreased in all of the countries they analyze.

Hout et al. (2001) disagree. They explore the multiple declarations of an end to class politics that have been advanced during the past thirty years by post-materialist, functionalist, and new social movements theorists. They argue that while class structures have changed in the post-industrial era, class stratification is an enduring reality and new forms of inequality will not simply replace or erase previous ones. They specifically challenge Clark and Lipset who, they contend, confound class and social hierarchies. The simple reality that social hierarchies are less salient does not translate into the disappearance of class. Further, Hout et al. argue that Clark and Lipset’s distinction between manual and non-manual labor may no longer be a valid way to conceptualize social classes, and that they confuse debates over class issues with the question of whether class continues to be a determinant of voting behavior. Hout et al. conclude that class continues to have an important impact on party politics, but that this influence varies across contexts and over time. Indeed, they argue that in some cases social class has actually become more important. Weakliem (2001) adds to these criticisms, contending that collapsing parties into the categories of ‘left’ and ‘right’ (as is usually done for convenience in studies of class voting) ignores the significant ideological differences that can exist between parties lumped into the same categories.

In light of these realities, where do the parties of the left stand in terms of their social bases and electoral possibilities, irrespective of the source of change? Eley (2002: 483) notes that defeats and disappointments at the hands of capitalism have led socialists to accept the status quo and settle for the ‘more modest aims of civilizing capitalism, stressing democracy, social citizenship and rights at work.’ Democracy and the defense of civil rights, human rights and the welfare state have become central concerns for the left. In addition, though initially overlooking the demands of new social movements, socialist and social democratic parties eventually reached out to them, and provided an environment in which they could thrive. Sasoon (1997) notes how the fall from favor of the Keynesian and Marxist ideals that underwrote the left’s success have forced it into a defensive position. Debray agrees, though he maintains that the redefined left is best poised to step in and defend the old left’s victories, and by doing so, serve as an antidote to the incivility and ethnic and religious strife that capitalism in its current form has spawned. To do so, however, it must abandon its utopian pretensions, and develop socialism as a ‘moral sense and a civil method’ (1990: 28).

Lipset argues that rather than disappearing, European leftist parties are moving or have moved towards the development of a more US-style, non-socialist model. According to him, European economic and class structures have come to ‘resemble those of the United States’ (1997: 76). The advent of post-industrial societies and concomitant decline in the size and power of European labor movements have undermined the class bases of politics. Lipset analyzes changes in the social bases of politics and their party consequences using an ‘apolitical’ Marxist lens. Fundamentally, he accepts that social class and the structure of production indeed determine ‘political superstructures,’ including political parties. While he recognizes variations based on differences among cases, he contends that we should view the European left as becoming increasingly American. In country after country, parties on the left have accepted capitalism, and debates revolve around distribution of resources and ‘post-materialist’ political issues rather than the essential structure of the economy. ‘New social democrats’ and ‘third-wayers,’ for Lipset, are the quintessential symbols of this transformation.

However, the extent of this Americanization is questionable, and probably overstated. While there are cross-regional commonalities that respond to deep social changes across the developed world, the European and American left differ profoundly in nature and character. First, the American left has been less class-based, much more centrist, and much less of a dominant historical force than its European counterparts. Second, the American left is shaded by the deep individualistic and libertarian character of American society, which contrasts sharply with the egalitarian and communitarian European left. Finally, in terms of concrete expression of a new left agenda, European parliamentary, and usually multiparty, systems still have more of an inherent capacity to allow for the expression of a multidimensional left than the hypermajoritarian US presidential system. In short, elements of the traditional leftist tradition are likely to remain significant to Western European party systems, and to counteract their Americanization.

Rather than the left disappearing or simply morphing into a more American-style left, Sassoon (1997) points to a convergence between traditional communist and socialist parties and the social democratic left. This convergence has also been accompanied by a more general convergence between the left and the right in terms of their core guiding principles, making for more ‘centrist’ party systems. He traces this convergence not just to value change, but also to the universalizing forces of globalization, the increasing homogeneity of electorates, rapid communication, and the necessity to present succinct and more universally acceptable political messages in an era of sound-bite politics.

Despite continuing ruminations, consensus has emerged that there is, indeed, a ‘left left,’ but that its definition and its role have changed. Nowhere in the major countries of Western Europe is the left unelectable, and in the late 1990s and early 2000s there has been resurgence in the popularity of parties of the left, with socialist or social democratic parties assuming power in the UK, Sweden, Spain, and Germany. What is more, while the utopian ideals of eliminating capitalism and fundamentally transforming the social order have been abandoned, leftist parties retain a central commitment to defend past achievements, to promote an agenda of social reform, and to protect human and civil rights. Indeed, one could argue that the left has successfully adapted to change by pirating the agenda of the so-called ‘new’ post-materialist parties and, in essence, derailing the challenge they were presumed to pose.

Society and New Parties

The term ‘new parties’ is used in two principal ways in the contemporary literature on party systems, referring to both parties that emerged after an authoritarian or totalitarian regime, and those that emerged in long-established democracies with the aim of displacing existing parties or breaking into the party system. However, as other chapters in this volume cover the emergence of political parties in the wake of non-democratic regimes more explicitly, this chapter will focus primarily on new parties in existing democracies. ‘New’ parties are characterized as such for chronological reasons, but also because their social bases do not correspond to the traditional left-right dimension of politics. Lane and Ersson (1991) distinguish between ‘structurally’ based parties which emerge out of traditional cleavage dimensions and give rise to class-based, religious, regional or ethnic parties, and those that emerge from ‘non-structural’ issues. There is also an implicit and sometimes explicit assumption that new parties emerge when other parties fail to assimilate emerging popular movements. Indeed, Lawson and Merkl’s edited volume dealing with new parties is entitled When Parties Fail (1988a). Parties based on non-structural issue dimensions usually are centered around one or a few national policies. The environmentally oriented European Green parties represent a quintessentially ‘new’ party movement, as do feminist parties and anti-nuclear, anti-EU and anti-NATO parties. Though the majority of new parties have emerged on the left (either in its ‘new’ post-materialist or ‘old’ materialist form), fascist and extreme right-wing parties also fall under the rubric of ‘new’ parties, as they address issues that for some voters traditional parties have ignored.

Theorists have advanced a number of explanations for the emergence of new parties. Smith (1989: 360) argues that partisan dealignment and a marked decrease in party attachments provide ‘windows of opportunity’ for new parties to enter closed party systems. Harmel and Robertson (1985) contend that explanations for the emergence of new parties take one of three forms: explanations based on social factors (new cleavages or issues), political factors (ideology, party behavior, the availability of leaders, or the salience of new issues), or structural factors (type of electoral system, the freedom to organize, the extent of government centralization, or whether the system is presidential or parliamentary). They go on to test a multiplicity of variables advanced by scholars and find that the propensity to form new parties is related strongly to sociocultural diversity. They find that while there is little relationship between structural variables and the propensity to form new parties, the eventual success of new parties is related to the permissiveness of the electoral system.

While most analyses of ‘new’ parties have focused on the left, and particularly the post-materialist left, new rightist parties have also emerged across Western Europe. Kitschelt (1995) employs a framework that mirrors his previous treatment of the transformation of social democratic parties (1994) to explain the emergence of these ‘new’ extreme right parties, rejecting the notion that they simply represent the re-emergence of parties of the traditional right. In explaining their emergence Kitschelt argues that there has been a shift in the competitive space of Western European party systems. While the well-worn left-right cleavage dimension persists, superimposed upon it is a ‘libertarian-authoritarian’ divide. Competition has shifted away from the purely economic cleavage as the main axis of competition toward the libertarian-authoritarian axis, providing space for the emergence of new right-wing parties. He does not argue that the class cleavage is insignificant or has been completely displaced. Rather, it has been combined with a new emerging cleavage that has transformed the competitive space for political parties and the nature of what have been traditionally understood as ‘left’ and ‘right’ politics.

Despite the quantity of literature devoted to studying the emergence of new parties, scholars question the existence and significance of the phenomena as a symptom of fundamental change. First, many point to the dearth of durable new parties, underscoring the obstacles to their emergence and subsistence even in the face of dealignment and the decomposition of established parties (Rose and Mackie, 1988; Harmel and Robertson, 1985). Even parties that overcome these obstacles are usually ephemeral and have difficulty in entering the realm of ‘relevant’ parties.

Second, Mair (1991: 63) finds that while the growth of ‘new parties’ has been a significant and measurable trend in Western Europe, it has probably been overstated. If anything, Mair argues that ‘new’ small parties probably have simply replaced ‘old’ small parties, making for limited substantive effect on the overall competitive dynamic of post-war European party systems.

Finally, despite the widespread emergence of new parties and predictions of profound electoral realignment, the political expression of underlying cleavage structures, though not identical, remains strikingly similar to the past. While social change, value change, and declining partisan attachments certainly have affected party competition in Western democracies, voters frequently continue to identify with the same ‘political family’ of parties. For these scholars, the oft-predicted ‘unfreezing’ of partisan alignments has yet to occur (Bartolini and Mair, 1990).

Conclusions: Multiple and Exaggerated Reports of Death

The relationship between society and parties is characterized as much by change as continuity. In terms of change, collective identities are waning, and political preferences are more individualized. The social bases of party support are more complex, and less predictably aligned along the class cleavage. The meaning of right and left has been transformed, with important consequences for party competition. However, the overall electoral balance between parties of the ‘left’ and ‘right’ (however defined) in Western democracies has not changed much, and there has been little vote redistribution between the two major blocs. Major parties have proven relatively resilient (with the exception of Christian democratic parties) and have often absorbed the post-materialist agendas of their new party competitors. What is more, there is evidence that new parties often simply inherit the supporters of old parties that have disappeared.

While there remains disagreement in the literature with respect to the nature of society-party connections, over the last ten years there has also been a good deal of scholarly convergence. Debates in the literature customarily began with a death report of one kind or another. Parties were declared irrelevant, social class pronounced dead, cleavages said to have ceased to be significant, or the left to have breathed its last. Nonetheless, recent literature has been less inclined toward strident death declarations, and has better recognized the complex interaction of the old with the new. There is a consensus in the literature that social and economic changes have transformed parties. However, scholars also agree that parties continue to be the central and most widely recognized agents of representation. Similarly, while cleavages and their party manifestations are in flux, most scholars now recognize that neither social class nor class cleavages have ceased to be central organizing concepts for understanding party systems. While the meaning of the left may have been transformed, leftist politics (perhaps defined differently) is alive and well.

This convergence is certainly a positive development. However, there are lacunae in the literature. First, the literature on the relationship between society and parties is least developed for new and transitional democracies. Scholars must untangle whether and how the assumptions and theoretical conclusions developed for the USA and Western Europe apply in the developing world. Inglehart (1997: 7) argues with respect to value change, and particularly post-materialist values, that ‘across many societies, once given processes are set in motion, certain important changes are likely to happen.’ This statement smacks of the teleologies that characterized developmentalist literature in the 1950s and 1960s and seems to assume that social and economic changes will automatically lead to value change, and similar partisan effects in the developing world. Is this the case? Also, even without deep changes in social class or levels of industrialization, we see post-materialist values taking hold in the developing world. Is this a result of social change, or have post-materialist values taken hold as the result of a contagion effect?

In addition to this theoretical concern, we also lack cross-national empirical studies of society-party relations in the developing world. It is no surprise that most of this chapter deals with European party systems, which have been the primary object of theorizing. There are a number of case studies that underscore the transformation of society-party relations in particular or a few cases, and a smattering of articles that assume that transformations similar to those underway in the USA and Western Europe are also taking place in the developing world. However, with few exceptions (see Mainwaring and Scully, 1995), we still lack systematic cross-national and cross-regional studies of these phenomena. The literature is similarly underdeveloped when it comes to the role of political parties in democratic transitions. There is certainly more written on the development of parties in post-communist systems, including analyses of the multidimensional interaction of pre- and post-communist cleavages. However, while the centrality of parties is always seen as important in transitions from non-communist authoritarian regimes, there is little comparative theorizing on the precise role that parties play in structuring social relations in processes of democratic transition, democratic consolidation, or how and whether authoritarian/democratic cleavages assume significance following transitions—Moreno (1999) is an exception. We must analyze the longer-term effect of the new pro- and anti-authoritarian cleavages that often emerge in democratizing societies.

Second, theorists have dealt insufficiently with parties as autonomous actors. Most of the literature treats parties as dependent variables that react to structural changes in economies and social relations. While analysis of political parties as dependent variables is certainly a valid enterprise, parties are also independent agents that frame issues and elaborate party platforms, affecting how cleavages translate into values, beliefs, and political behavior. This is increasingly important with the advent of mass, centralized, and professionally orchestrated campaigning. On a related note, we need better accounts of the differences between deep value change and how short-term issues cycle through the electorate with the help of party advertising and publicity. For example, in the United States to what extent does the promotion of issues such as gay marriage, the role of religion in society, and anti-immigration rhetoric by the parties and candidates themselves help to set the political agenda that shapes how political values develop, with which party citizens identify, and how political beliefs are expressed?

Finally, the literature on the society-party nexus should do more to analyze internal party processes. The most analyzed point of contact between citizens and parties is in the voting booth. We have extensive studies of voting behavior, but little on the other potential connections between parties and citizens. Neither individual connections nor those mediated through other groups or the media are well accounted for in the literature. Citizens interact with party organizations, finance campaigns, are influenced by party publicity, and potentially play a role in the recruitment and selection of candidates. To understand these complex interactions, more serious study of internal party processes is essential in order to uncover the nuts and bolts of society-party relations beyond simply measuring voting as the determinative indicator of citizens’ ties to parties.