Chapter One The Decline of Working Class Party Representation

1 Social Cleavages

Before delving deeper into the problem addressed, it is necessary to present a precise definition of what social cleavages are and to illuminate how they relate to parties and organized interests. The idea of something like cleavages as a politics-structuring set of sociological divisions within the polity has existed since the beginning of political science, particularly in the field of political sociology. Indeed, the origins of the cleavage concept can be traced back to authors as Tocqueville, Marx, Michels, Ostrogorski, Weber, Coser and others. (Coser 1956; Lipset 1959b; Allardt and Littunen 1964; Lipset and Rokkan 1967) It is then all the more ironic that a subfield building on a basis of authors whose works focused on the sociology of politics largely lost sight of some of the sociological aspect in their writings, in particular the implications for mass interest groups. However, it was not until Lipset and Rokkan published their seminal piece that the cleavage concept began to attract significant attention within the field of psephology. (Karvonen and Kuhnle 2001; Franklin 2010) Earlier works in political sociology paved the way, as well as some of Lipset and Rokkan’s work, yet many of these works either did not receive as much attention as the 1967 article or lacked broader theoretical discussion of the cleavage concept. In other publications, the cleavage concept was taken as a given without much elaboration on defining it. Exceptions that unfortunately did not get as much attention were follow-up pieces by Rokkan that addressed some crucial criticisms on the 1967 piece and deepened the theoretical debate of the concept. (1970; 1975) Unfortunately, despite their very ambitious attempt, Lipset and Rokkan still did not do a thorough job of specifying exactly what they meant with the concept “cleavage.” Throughout the years, a range of related, yet slightly different, descriptions are used, which makes it difficult for the reader to grasp what they meant by the term, and nowhere did they devote a summarizing section to this, either. (1967; also cf. Flora et al 1999; Karvonen and Kuhnle 2001; Bornschier 2009)

Bartolini and Mair criticize the literature employing the cleavage concept for only providing truncated versions of the cleavage concept. (1990; Mair 2006, 373-374) Instead, Bartolini and Mair emphasize the holistic structural foundations lying at the root of the cleavage concept in the 1967 piece. The aspect that was often focused on in the body of literature that followed was the dependent variable it appeared to discuss: the stability of party systems. In a similar fashion, the reorientation of traditional mass based parties with a focus on group identity toward a focus on the individual voter—the catch-all party—was taken as sufficient evidence to argue for a decline of social cleavage alignments. Future authors applying the cleavage concept, sometimes for a single case, tended to attempt to explain the opposite: the party system stability, instability, or structure would be examined either to find or to deny the existence of cleavages. (Franklin 2010) The often diligent efforts put into such research, unfortunately, did not save it from the inherent logical fallacy. Given the fact that Lipset and Rokkan, and by extension many of the fathers of political sociology, at least mentioned other forms of organizational translation of cleavages, such as labor unions in the case of the class cleavage, reversing the relationship of dependent and independent variables could only make sense if such forms of organizational translation would also be included. Lipset and Rokkan explicitly recognized the importance of non-party political expressions of cleavages, yet were mostly interested in explaining contemporary party system structures. In particular, they write about “movements” or “social movements,” referring to groups prior to obtaining representation through parties. (1967: 17, 22, 23, 30, etc.) The conclusions drawn from such a reversed relationship by other authors are at best incomplete, and at worst incorrect. Consequently, follow-up research pronouncing the death of cleavage politics, or critics of the cleavage concept doubting the relevance of cleavage politics, are in no position to make such claims. [1]

Numerous follow-up pieces have been written that take the cleavage concept of Lipset and Rokkan more or less for granted, thereby avoiding a discussion of exactly what it is. Currently, the most widely accepted definition, and the one I’ll be using, is the one proposed by Peter Mair. (Bornschier 2009; Mair in Katz and Crotty 2006; also see Bartolini 2005 and Bartolini and Mair 1990) Mair defines cleavage by pointing to “three distinct characteristics” of the concept; in fact he mentions four characteristics, but the fourth comes later in the text. (373):

In the first place, a cleavage involves a social division that distinguishes between groups of people on the basis of key social-structural characteristics such as status, religion, or ethnicity. A cleavage is therefore grounded in distinct social reality. Second, there must be a clear sense of collective identity involved, in the sense that the groups on which the cleavage is grounded must be aware of their shared identity and interest as farmers, workers, Catholics, or whatever. … Third, a cleavage must find organizational expression, whether through a political body, a trade union, a church, or some other body. … one additional property of cleavages: they are deep structural divides that persist through time and through generations. (Mair in Katz and Crotty 2006; my emphasis).

As I mentioned above, Mair’s definition of cleavage seems to implicitly include a fourth element, which he later mentions on the same page. Mair argues that cleavages persist because they continue to be framed in party politics (373). In line with what I argued above, I think Mair’s point here is substantially incomplete, if not wrong. To be fair, this description by Mair of cleavage was based on earlier works he and Stefano Bartolini had done, and was written as part of a brief article in a party encyclopedia. Thus, I would not go so far to claim that this was indeed Mair’s complete stance on the subject, but instead an ill-worded sentence unintentionally revealing a lot about the empirical literature about the subject. The sentence contradicts the third element of the definition; stating that not just parties can be deemed cleavage expressions, but also seems to assume that the type of organizational expression determines the fate of a cleavage. This assumed importance of organizational expression seems to contradict the first part of his definition. If a cleavage is indeed grounded in distinct social reality, how could a particular choice of the organizational expression type be capable of changing this distinct social reality, let alone threaten the persistence of a cleavage as a whole? The inclusion of party in the organizational element of the definition appears to be observational rather than analytic. It adequately describes the turn the literature has taken: a strong emphasis on political parties but a lack of inclusion of other forms of political organization, and a lack of a thorough exploration of the relationship between parties, organized interests, and cleavages. Indeed, the concept has almost exclusively been used in party political research so much that it is sometimes unjustifiably mistaken as a necessary element of the definition of cleavage. Also, although the analysis of cleavages and party systems may appear to be more easily quantifiable for comparative empirical studies, they are certainly not free from methodological issues with measurement. (cf. Stoll 2008; Harrop and Miller 1987) Bornschier writes: “Going beyond the three constituting elements of a cleavage, then, the term cleavage is usually reserved for durable patterns of political behavior linking social groups and political organizations.” (Bornschier 2009) Political parties are thus indeed part of the organizational expressions cleavages may entail, but they are certainly not the only kind of political organizations that have this quality.[2]

What can be surmised from the above discussion is that the concept of cleavage, itself, has been shrouded in ambiguity since it has become popular in the field; the seminal piece Party Systems and Voter Alignments lacked clarity, and most theorizing of other scholars did not succeed in clarifying that much more either (1967).[3] A consequence is that future researchers defined the concept in a party-centric way, ignoring the wider theoretical implications of the Lipset and Rokkan piece. For the purpose of this project, I will continue to use the most widely accepted definition within the field, that being the Bartolini and Mair definition, which emphasizes that the organized forms cleavages are expressed by political organizations, yet are not limited to political parties. The reasons for doing so are that this definition is widely accepted in the field, that it is intelligible to students not familiar with this body of literature, and that it is analytically useful for this project.