In modern political systems policy issues are rarely decided directly by the vote of citizens. As I argued at the end of the previous chapter, referenda, where they exist, are severely limited in the decision load that they can effectively deal with. General elections, by contrast, can rarely be interpreted as decisions on substantive policy; they should not primarily be considered as a mode of arriving at collectively binding decisions but rather as institutional arrangements for the legitimation and control of hierarchical government authority. From the citizen's point of view, therefore, modern constitutional democracies are not fundamentally different from their nondemocratic predecessors. Decisions can be imposed without the individual citizen's consent and are backed by the superior capacity of the state machinery to inflict severe deprivations. In comparison with absolutist and totalitarian systems, it is true, constitutional democracies are characterized by more, and much more effective, limitations on state power, defined by the constitutional protection of basic rights and by general laws that are binding upon the state as well and enforced through an independent judiciary. But within the domain so circumscribed, decisions of "the state" are still imposed on the citizen by hierarchical authority and enforced by superior force.
For the legitimacy of state authority it is of the utmost importance how these decisions are arrived at through interactions among the active participants in policyrnaking processes and how these interactions are connected to processes in which citizens are directly involved. But before I turn to these questions, which will continue the exploration begun in the previous chapter, it will be necessary to explicate the characteristics of the hierarchical authority relationship itself and of "hierarchical direction," which, in our understanding of actor-centered institutionalism, is one of the four basic modes of interaction -- along with unilateral action, negotiated agreement, and the majority vote. The present chapter will therefore have to deal with two related but analytically distinct themes. I will first discuss the policy-relevant characteristics of hierarchical direction, and I will then return to the issue raised at the end of the previous chapter: the role of voting and elections in the legitimation and control of hierarchical authority.