all actors involved. In the ‘‘real world,’’ politicalWghts and disagreements ensue
and, in the academic world, analysis of electoral systems establishes who wins and
who loses under various systems. In a general sense the winners from any electoral
system are political parties as organizations. As Schattschneider wrote, ‘‘the polit-
ical parties created democracy and modern democracy is unthinkable save in terms
of the political parties’’ (Schattschneider 1942 , 1 ). What helps makes parties so
prominent is that they have toWght elections. The functionality of parties and
party-like organizations as vehicles forWghting elections means that, in general,
political parties are encouraged by elections and voting. Even non-partisan
elections in US local governments see party-like organizations devoted to getting
out the vote and endorsing candidates. Of course the more pointed question
becomes which parties win and which lose under various electoral systems. Over
and above that, electoral systems shape the internal cohesion and discipline of
parties. Some systems—such as the single transferable vote (STV)—encourage
factionalism and intraparty competition, while others—list PR—reinforce party
discipline. Electoral systems, too, shape the relationship between voters and
representatives: some systems, especially those that allow voters to cast a ballot
for individual candidates, encourage constituency service and the cultivation of a
‘‘personal’’ vote. Other systems do not encourage such a relationship and so shift
the incentives of candidates to focus more upon party than upon individual voter
concerns (Bowler and Farrell 1993 ; Carey and Shugart 1995 ).
But the main focus of electoral systems research has been upon which kinds of
parties win and which lose under various schemes. Studies of electoral systems
show repeatedly that diVerent electoral arrangements privilege or discriminate
against diVerent kinds of parties or candidates (Lakeman 1954 ; Rae 1971 ; Grofman
and Lijphart 1986 ; Taagepera and Shugart 1989 ; Lijphart 1990 , 1994 ; Farrell 2001 ). 1
With few exceptions, discussion of electoral system eVects have considered the
question of which parties beneWt largely without reference to the ideological or
programmatic component of parties and the discussion within this literature has
tended to consider how many parties are produced under various systems.
2 Duverger’s Law
.........................................................................................................................................................................................
Duverger’s Law occupies pride of place in this series of studies as one of the major
statements in electoral studies research and the canonical statement of the role of
1 Farrell 2001 is an excellent and accessible overview of the electoral systems literature.electoral systems 579