the relationship between the party system and the set of parties that make up that
system. One theme has been the importance of a party system for the eVective
functioning of democracy. In general we deWne ‘‘party system’’ practically by the
(eVective) number of parties. The eVective number tells us something about
the case such as Israel, but perhaps better is to add to the eVective number
consideration of parties that serve as generators of prime ministerial candidates,
or candidates for the major portfolios. In either case, the example above implies a
sort of path dependency on the particular parties that make up the party system
and on the height of barriers to entry to new parties and perhaps to achieving
major party status.
Key’s party-in-three-parts organized this chapter, but the reader may note that
the third part, the party-as-organization, appeared on stage only brieXy. Here it is
appropriate to observe that one of the major components of the party organiza-
tion, the activists and the resources in time, money, and eVort that they control, is a
critical component for synchronizing the party in the public’s mind and the party
in the legislature (see especially Aldrich 1995 ; Kitschelt 1989 , 1999 ). There is an
important regularity about party activists that cuts across the various types of party
systems. In majoritarian and proportional, in two- and multiparty systems, in the
US and European archetypical cases of this chapter, activists have turned out to be
more extreme than the electoral members of their parties. Recently, Kedar ( 2005 )
has developed and tested a theory of this process, arguing that voters support
parties with activists more extreme than they are, so that actual policy will be able
to be moved in the direction of the activists, but, through the inertia created by the
rest of the political system, almost assuredly less far than the activists would desire.
The result is a change in policy much like the more-moderate voter actually desires.
Aldrich and McGinnis ( 1989 )oVer a diVerent but complementary story based on
the US parties. Party activists can induce candidates and oYceholders to move
policy in their direction, but not as far in their direction (in this instance balancing
their need for extremity to gather resources from activists to win votes and their
need for moderation to retain support in the electorate). In either case, relatively
more extreme activists are motivated to connect public and politician, pushing
both to aVect policy changes more to the activists’ liking. Whatever the details, the
activists are central party organization members for aggregating and articulating
public desires and tying politicians to policy outcome. And, if Erikson, MacKuen,
and Stimson ( 2002 ) have the dynamics right, they are the source of the swing of the
policy pendulum.
Let me close with a fourth area which appears ripe for research breakthroughs.
This chapter pointed towards a fully comparative political parties project. Instead
of distinguishing between American political parties and the political parties of
other (advanced, industrial, and postindustrial) democracies, we are beginning to
see more clearly that political parties are common to all democracies, and they are
so because democracy is, indeed, unthinkable save through the agency of the
572 john h. aldrich