If you were designing a political system from scratch, what decision-making procedures would you enshrine to guarantee fair and efficient outcomes? According to Jon Elster, all possible procedures are permutations of three ideal types: arguing, bargaining and voting.
Voting involves the aggregation of individual preferences. The typical example is the referendum by secret ballot. To be democratic, the principle of majority rule must apply. A fine procedure, we might say. But surely it is misplaced optimism to believe that some sort of invisible hand will guide the mass of probably uninformed voters towards mastery of a complex issue.
Bargaining, on the other hand, involves interaction between participants. The isolation and anonymity of the participants is removed, and decisions are arrived at after those that command the weightiest resources ( eg. money, control of the army, authority over demonstrators ) make an agreement in exchange for various concessions.
Arguing similarly involves participant interaction, but appeals are made to impartial reason rather than partisan interest. The deliberations of the jury room are the model for this procedure. If 'voting' has its roots in Rousseau's theory of democracy and 'bargaining' belongs with the liberal democratic tradition of Dahl and Schumpeter, 'arguing' is firmly rooted in the republican tradition. Elster cites Pericles' eulogy of Athens: instead of a stumbling block, discussion is "an indispensable preliminary to any wise action". The idea turns up throughout history: Burke's speech to the electors of Bristol implies a deliberative model of sorts.
Its most recent incarnation was partly a result of Habermas' influential theory of communicative action. Habermas' claim that speech 'does' things ( from Austin's speech-act theory ), and is primarily oriented towards understanding and consensus, was ideally suited to revamping the theory of deliberative democracy. Despite being an heir of Kant and Marx, Habermas does not really get away from the republican mindset inherent in the model. In this sense, the elitist implications of deliberative democracy worry me . . .
Elster's volume fleshes out some of these worries in a reasonably comprehensive way. Susan Stokes' essay 'Pathologies of Deliberation' is well worth reading, as is James Johnson's 'Arguing for Deliberation: Some Skeptical Considerations'. Elster makes the important distinction between deliberation in the making of a constitution and the level of deliberation in the final constitutional document.
Of the remaining essays, Cass Sunstein's 'Health-Health Trade-Offs' is the stand-out, managing to locate the debate in solid empirical examples. Sunstein's conclusion that must find a balance between 'voting' and 'arguing' struck a chord and reminded me of its applicability to the current hot political topic of GM food: how can we balance voter's 'gut feelings' against GM food with a vigorous scientific and public policy debate which is increasingly pointing to its advantages?
Deliberative democracy is not simply abstract theorising. It is very much located in the politics of modern societies. I strongly recommend this book.
