Political parties, as Kaare Strøm observes, are generally 'the most important organisations in modern politics…. Students of political parties have commonly associated them with democracy itself'. 1 The importance of political parties in German politics is undisputed. In his influential account of (West) German politics, Peter J. Katzenstein considers parties not only to be 'an essential institutional node linking state and society', but-aside from co-operative federalism and parapublic institutions-one of the three crucial nodes in the (West) German policy network. 2 In German academic and political parlance, the term Parteienstaat ('party state') indicates that parties are considered to be one of the defining elements of German political life. Article 21 of the Basic Law and the Party Act of 1967 emphasise their legal recognition in Germany's constitutional framework. Their role is also recognised in the form of generous public subsidies supporting their activities. 3
The aim of this article is to assess continuity and change in the German party system since unification in 1990. From a methodological point of view, this presents a challenge, because there are at least three kinds of trends which are difficult to disentangle: (1) the direct effects of unification, (2) general longer-term trends independent of, but coinciding with, unification and, (3) indirect effects of unification (that is, for example, general trends accelerated or delayed by unification). In the absence of suitable statistical controls, it is not possible to determine the 'net contribution' of these different effects. Nevertheless, party and party system change in Germany since unification can be assessed against the backdrop of developments in comparable political systems in Western Europe in order to avoid conclusions based on spurious co-variation.
Political parties and party systems in Western Europe have been characterised by complex patterns of persistence and change over the past century. In the research process, stability has often been taken for granted, while the effort has tended to focus on the description and explanation of change. Yet, as Peter Mair contends, in the late 1990s, despite a large body of literature emphasising party-system change, Lipset and Rokkan's famous 'freezing hypothesis'-formulated in the 1960s and suggesting that the West
Thomas Saalfeld, University of Kent at Canterbury