Thus, based on the Swiss experience, the direct-democratic procedures turn out to be highly attractive. On the one hand, and in line with the results of public opinion research more generally (Sniderman 1993), the Swiss experience shows that the voters, in the large majority of the propositions submitted to them, are not really overburdened by the task expected of them. This results, above all, from the pre-structuring of their choices by the way institutions are set up in Switzerland and by the mobilizing and communicating strategies of the political elites. On the other hand, the Swiss experience also indicates that the citizens doubly benefit from direct-democratic procedures: such procedures not only increase public performance, but also the legitimacy of political decisions and, by implication, the general satisfaction of the citizens with their lives. Taken together, the Swiss experience amounts to a powerful empirical rebuttal of the arguments raised by sceptics like Sartori.
This does not mean that all is perfect in the Swiss variety of direct democracy. While the overall assessment of the Swiss experience with direct democracy lends itself to highly optimistic normative conclusions, several of the results I have presented depend on favourable conditions. Thus, citizens’ decisions in the case of votes on unfamiliar propositions with little elite mobilization (about a quarter of the cases in the period studied) rely much less on argument-based opinions. There are even a limited number of propositions where the majority of the individual voting decisions cannot be explained by either heuristic or systematic considerations (Kriesi 2005). In other words, even in Switzerland, with its long experience with direct-democratic votes, the procedures do not always work as intended. Moreover, from the point of view of normative democratic theory, the self-selection of the most incompetent is not unproblematic either. If it solves the problem raised by the conservative critics of direct democracy, it raises instead a problem of social justice that has to be dealt with by measures such as civic education, or the public support of the intermediaries responsible for direct- democratic campaigns.
In my view, the lessons to be drawn from the Swiss experience are twofold. On the one hand, this experience suggests that the political elites have a very important role to play in the pre-structuration of and the mobilization for the vote. If the elites do not engage in intense campaigning preceding the vote, as was the case in the Dutch referendum on the European Union Treaty in 2005, the electorate will not be able to cast an informed vote. The elite-led debate during the campaign preceding the vote proves to be decisive for the quality of the voters’ choice. On the other hand, in a properly designed direct- democratic system, the governing majority is not completely disarmed, but it does not have full control over the process either. Groups of citizens may set the agenda by imposing a vote, and the governing majority may not be able to convince a majority of the voters to adopt its preferred solution at the polls. Empowering the citizens by direct-democratic procedures implies the risk of defeat, for reasons that, most of the time, have very little to do with questions of citizen competence, but quite a lot to do with questions of pre-existing political preferences on the part of the citizens and the outcome of the political struggle among majority and opposition.