312 Jeff Manza and Nathan Wright
Individual-Level Factors: Religion and the Alignment of Voters
and Parties in Western Europe
The persistence of religious parties in Europe suggests an avenue for political expres-
sion of religiosity that is more explicitly connected to the party system than in the
United States. At the same time, there is a much wider consensus that secularization
processes at the individual level that are weak or nonexistent in the United States have
proceeded much farther in Europe (Dobbelaere and Jagodzinski 1995; Jagodzinski and
Dobbelaere 1995; Berger 1999). Yet secularization need not imply declining levels of
religious voting: New cleavages, such as those between secular and religious voters, may
replace older Catholic/Protestant divides; and voters with religious identities may be
more likely to act on the basis of those identities even if there are fewer of them. The
robustness of the religious cleavage, in those countries where one exists, has frequently
been proclaimed (as we noted earlier). So what has been the impact of these factors for
individual voting behavior?
The existing literature suggests two paradoxical findings. First, where a religious
cleavage has been embedded in the political system, religious identities continue to
exert a significant impact on individual voting behavior; at the same time, there has
been a general (but not universal) weakening of the religion-vote association in Europe.
The most carefully studied case by far is the Netherlands, and we consider some of the
evidence from that country first. Dutch society has long been characterized by what
have come to be known as “pillars,” reflecting stable, lasting, and loyal connections
between religion and voting. The four pillars consisted of Catholics, Protestants, and
nonconfessionals divided into Labour and Liberal constituents. Each pillar developed
its own political organization, with the Catholic and Protestant parties consisting of
their followers regardless of their social class, the Liberal party consisting of middle- and
upper-class nonconfessionals, and the Labour party consisting of working- and lower-
middle-class nonconfessionals (Eisenga et al. 1994). Party loyalty was fierce among all
four pillars, particularly among Catholics and Calvinists. From 1922 to the 1960s, Dutch
Catholics were considered among the most loyal voting bloc in the world, consistently
giving more than 85 percent of their votes to the Catholic party. By 1973, however,
Catholics were giving less than half (48 percent) of their votes to the Catholic party, and
shortly thereafter the Catholic party merged with the two largest Protestant parties to
form the Christian Democratic CDA. This new combined party’s first electoral showing
in 1977 was a mere 31.9 percent of the overall popular vote in the Netherlands, less than
what the Catholic party alone had received in 1963, and it has declined further since
then (Eisenga et al. 1994). The breakdown of pillarization has largely been attributed
to the forces of modernization and secularization, and these forces are widely believed
to have completely eroded what was once the strongest religious voting cleavage in the
world (Eisenga et al. 1994; Andeweg 1982; Becker and Vink 1994; Irwin and Dittrich
1984; Miller and Stouthard 1975). The emerging party system has been characterized a
number of different ways: As a new left-right political ideological continuum (van der
Eijk and Niemoller 1987), as reflecting a postmaterialist cleavage (van Deth and Geurtx
1989), or along new political party lines united by ideological views rather than class
and religious makeup (Middendorp 1991).
Yet changes in class structures and secularization processes do not necessarily pro-
duce a decline in the actual religious (or class) cleavages. Studies that have focused