ABSTRACT
Introduction Following the ‘deliberative turn’ in democratic theory, political scientists increasingly come to see an extensive infrastructure of political discussion as an indispensable element of a functioning democracy (Dryzek 2002). Accordingly, in their study of German and Swiss legislators’ political ‘talk culture’, Steiner et al. (2004: 5-6) claim that in order to understand the quality of contemporary democracies, the extent and nature of political debate must become the subject of thorough empirical analysis – at the level of political elites within formalised contexts which they analyse in their landmark study, but also at the level of ordinary citizens in their everyday lives. Obviously, in contemporary representative democracies, elections are the setting where such activities are of most immediate relevance. Do citizens discuss political matters with others when elections are approaching, or do they rather try to make up their minds in isolation, taking into account only the information they encounter in the mass media? If they discuss politics, with whom, how intensely and in which ways do they do so? Do patterns of interpersonal political communication differ – between citizens, but also across time? And, lastly, does political discussion matter for elections? In particular, does talking engage citizens with politics in ways that make them more prone to cast their ballots, and if so, does it assist them in making up their minds? These are the questions which the present chapter addresses. We will present analyses pertaining to the 2005 German Federal Election, using a unique dataset that allows for new insights concerning the appearance and relevance of political discussion at elections. In particular we will extend the existing literature by adding a dynamic perspective that has been neglected so far. We shall look at political conversations during the course of an election campaign: do political discussions become more widespread and more frequent during campaigns? And how does the level of homogeneity evolve in the run-up to elections? Furthermore, we will distinguish two basic forms of political conversation. In their classic studies of the 1940s and 1950s, Lazarsfeld and his colleagues were exclusively concerned with interpersonal communication within primary groups (Lazarsfeld et al. 1944). However, later studies have stressed that secondary relationships can also be of considerable importance when it
comes to talking politics (Huckfeldt and Sprague 1995). In many respects, it makes a difference whether politics is addressed in casual chats with other family members at home and with close friends, or whether such talk takes place at the workplace or ‘over the garden fence’ (Curtice 1995), as the quality of the respective underlying relationships differs. The ‘strong’ primary relationships between spouses, other family members and friends are loaded with positive affect, and characterized by ‘intimacy, trust, respect, access, and mutual regard’ (Kenny 1994: 718). Secondary relationships, in contrast, are ‘weak’, functionally specific, and rest on the principle of structural equivalence rather than emotions and mutual attraction (Granovetter 1973). While this does not necessarily imply that either one of these types of relationships is a priori more important as a context for political discussion than the other, they are probably quite different in terms of the frequency of their occurrence, but also their content and, presumably, their consequences. We begin our analysis with a short description of our data and key variables. In the subsequent sections we address the following questions: how did the incidence and frequency, but also the amount of disagreement experienced in political discussions evolve in the course of the 2005 German election campaign? What were the determinants of the intensity and contents of these conversations? And ultimately, what were their consequences on Election Day with regard to turnout and the easiness or difficulty of voters’ decision-making?
