This article was downloaded by: [University of Lund]

On: 30 November 2009

Access details: Access Details: [subscription number 731775989]
Publisher Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK

West European Politics

Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713395181

The transformation of Western European party systems revisited

Steven B. Wolinetz a
a
Associate Professor, Memorial University of Newfoundland,

To cite this Article Wolinetz, Steven B.'The transformation of Western European party systems revisited', West European
Politics, 2: 1, 4 — 28
To link to this Article: DOI: 10.1080/01402387908424224
URL: http://dx.doi.org/10.1080/01402387908424224

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
 The Transformation of
Western European Party Systems Revisited
Steven B. Wolinetz*

Otto Kirchheimer has argued that a transformation of Western

European party systems is under way. If Kirchheimer's assertions
about the emergence and success of catch-all parties are correct,
we should discover decreasedfragmentation in Western European
Downloaded By: [University of Lund] At: 14:30 30 November 2009

party systems. However, there is little evidence for this. In recent

years many party systems have become more, rather than less,
fragmented. Catch-all parties have succeeded only in countries
which experienced crises and disruptions in development.
Elsewhere the strength of partisan attachments prevent the
success of catch-all parties. Increased fragmentation in the 1970s
reflects the weakening of partisan attachments, the emergence of
new concerns, and growing reactions against goverment policies
and practices. Minor parties have gained support because they
were better able to mobilise discontent and fulfil the expressive
functions of political parties.

I. THE CATCH-ALL THESIS AND CHANGE IN WESTERN EUROPEAN PARTY

ALIGNMENTS
Twelve years ago, Otto Kirchheimer argued that a transformation of
Western European party systems was under way. According to
Kirchheimer, former class and denominational parties of mass inte-
gration were transforming themselves into ideologically bland catch-all
parties, more interested in winning the next election than the defence
of principle or the pursuit of more distant ideological goals. This was
occurring, Kirchheimer argued, because the advent of an affluent and
consumer-oriented society had loosened the bonds between parties of
mass integration and their clienteles. Ideology was no longer an
exclusive motive for voting behaviour.
Aware of the change and afraid of diminishing support, political
parties were responding by scuttling excess 'ideological baggage,'

*Associate Professor, Memorial University of Newfoundland. This paper was

written while on leave at the Institute of Political Science, Leiden University, the
Netherlands. The research was supported by a Fulbright fellowship and a Canada
Council Post-doctoral Research Award. An earlier version, presented at the 1976
Annual Meeting of the Canadian Political Science Association, has been published
in Italian in Rivista italiana di Scienza politica, no. 1, April, 1978.
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 5

stressing the qualities of their leaders, avoiding controversy, and seeking

the support of interest groups. This in turn transforms both the role
and function of political parties. Parties de-emphasise their 'expressive
functions'—the articulation of demands and grievances—abandon
'attempts at the intellectual and moral encadrement of the masses,' and
operate largely as electoral agencies, bidding for votes wherever they
can be found. As a result, the style and mode of political competition
are transformed. Because the catch-all party is a superior instrument
of competition, other parties imitate it, and the central features of
party competition cease to be 'the formation of action preferences,'
advocacy of causes or defence of a 'classe gardée. ' Instead, we find an
opportunistic battle among bland parties trying to achieve the best
Downloaded By: [University of Lund] At: 14:30 30 November 2009

marketing results.1
Kirchheimer's scenario is widely cited as a statement of what has
been happening to parties and party systems in Western Europe.
However, because of the complexity of the argument and the broad
sweep of changes which it embraces, the catch-all thesis has rarely
been tested.2 The argument is an intellectual tour deforce. Kirchheimer
links major themes in the literature on advanced industrial societies to
changes in voters' loyalties, couples these to changes in party strategy,
style, and presentation, and connects these to changes in party functions,
changes in the nature and style of party competition, and changes in
party alignments—the growth and success of catch-all parties. We are
confronted with a series of inter-related suppositions: that voters in
affluent societies are less motivated by ideology than before, that parties
have perceived this and have responded by de-emphasising ideology
and broadening their appeal, that these changes in strategy and style
are successful and win more votes, that these changes alter the relation-
ship between parties and their supporters, and that when one catch-all
party succeeds, others imitate it, transforming the party system.
The difficulties in testing such a multi-barrelled set of hypotheses are
immense. There are no necessary links between sequences in the argu-
ment, and an empirical test of one supposition tells us nothing about
the validity of others. One can easily mistake evidence for one part of
the scenario for evidence for the entire sequence. It is possible, for
example, to document changes in party ideologies, but this means
neither that these changes have occurred because of changes in voters'
loyalties, nor that changes in ideology will lead to electoral success or
imitation by other parties. As a result, we can neither accept nor reject
the argument without testing each proposition separately.
In addition, the concept of a catch-all party is exceedingly difficult to
operationalise. In view of Kirchheimer's emphasis on parties bidding
for diverse support, it is tempting to define a catch-all party as a party
with a heterogeneous base of support.3 However, this is inadequate:
Kirchheimer's argument says very little about the success or failure of
 6 WEST EUROPEAN POLITICS

heterogeneous parties, per se, but instead focuses on the ways in which
class or denominational parties adapt to weakened loyalties. Although
Kirchheimer implies that parties will become more heterogeneous, this
need not happen immediately: parties could gain support but remain
homogeneous on key characteristics. Minimally, any test of the catch-
all thesis should be couched in terms of changes in bases of support.4
But even this falls short of the mark: Kirchheimer suggests that parties
employing certain kinds of strategies are likely to win support at the
expense of others. However, few comparative studies of party strategies
exist, and we know very little about the ways in which different strategies
affect party strengths.
The difficulties of testing the catch-all thesis are compounded by the
Downloaded By: [University of Lund] At: 14:30 30 November 2009

way in which Kirchheimer excluded certain parties from his purview.

Although Kirchheimer purported to describe the transformation of
Western European party systems, his analysis focuses largely on parties
in Britain, France, Italy, and West Germany. Kirchheimer excluded
most parties in the smaller democracies because party competition
in these countries was more limited and direct—parties had either
majorities or near majorities or regular access to power and thus had
no need to adopt catch-all strategies—and because parties with very
specific group or regional claims were too narrowly based to become
catch-all parties.5 Eliminating the smaller democracies is reasonable
if the conditions of competition are different in these systems, but
doing so deprives the catch-all thesis of much of its richness, and brings
the relationship between affluence, weakened loyalties, and the adoption
and success of catch-all strategies into question. Three of the four
countries which Kirchheimer considers—Italy, France, and West
Germany—have experienced severe disruptions in their development.
It is possible that the emergence of large catch-all parties in two of
these countries is the result of either crises and their successful resolution
or changes in constitutions and electoral laws, while in Britain, catch-all
competition may simply reflect the dynamics of a two-party system.
Moreover, if the adoption of a catch-all strategy is a response to
decreasing loyalties to mass parties in affluent societies, then there is
no reason why the smaller democracies should be excluded. On the
contrary, it is in precisely these countries that the phenomenon should
be most advanced.
Kirchheimer's scenario is also limited because it was written in the
1960s and is grounded in the end of ideology literature. The end of
ideology debate has yet to be resolved, but in view of the experience
of the late 1960s, it is fair to say that what occurred was not an end or
disappearance of ideology, but rather changes in ideologies, changes in
priorities derived from ideologies, and changes in the locations in the
political system in which ideologies were expressed. Since the 1960s,
there has been a resurgence of ideology (ideological themes are more
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 7

frequently expressed) and a resurgence of political conflict. We need to

know more about the ways in which these phenomena affect both the
adoption and success of catch-all strategies.
Because parties are the main vehicles through which citizens can
express their preferences, it is important to know how and why parties
change. Though there are difficulties in his argument, the changes which
Kirchheimer outlined are worth examining not only because his argu-
ment is widely accepted, but also because his work provides a useful
background for examining stability and change in post-war Western
European party systems. We can follow one of two strategies: either
focus on one party system and examine the extent to which the sequence
of changes which Kirchheimer described has taken place, or else focus
Downloaded By: [University of Lund] At: 14:30 30 November 2009

on only one aspect of his scenario but study it across a wider range of
political systems. Since we are interested in the extent to which a
transformation of Western European party systems has taken place, we
will adopt the latter strategy and attempt to determine the extent to
which changes in party alignments have taken place in thirteen European
democracies between 1945 and 1975.

I I . CHANGES IN WESTERN EUROPEAN PARTY ALIGNMENTS: A TEST OF

THE CATCH-ALL THESIS
One aspect of Kirchheimer's scenario lends itself to a simple test. If
Kirchheimer's argument is correct—if decreasing loyalties have caused
mass parties to adopt catch-all strategies, and if the adoption and success
of a catch-all strategy spurs imitation—then we should expect to find
substantial changes in the party of alignments of countries in which
catch-all strategies have been employed. Because catch-all parties are
superior instruments of competition, they should win votes from parties
which fail to adapt. As a result, party systems should become less
fragmented. In addition, because of weakened loyalties and the op-
portunistic style of party competition, we should expect increased fluc-
tuation in party strengths and greater traffic among political parties.6
In the absence of either detailed aggregate analysis or longitudinal
survey data, measuring traffic across party lines is difficult. We are
forced to fall back on national electoral data. Although this can never
give us a fully accurate measure of shifts among parties, it allows us to
make rough estimates and form impressions which may be helpful in
later research. If we consider electoral trends in Western Europe, the
impression which we get depends on the vantage point which we select.
If we choose the late 1960s, the dominant impression is one of stability.
Writing in 1967, Lipset and Rokkan contended that most West
European party alignments had been frozen since the advent of universal
suffrage some fifty years earlier.7 Examining persistence and change
in nineteen democracies from 1945-1968, Rose and Urwin came to the
same conclusion.8
 8 WEST EUROPEAN POLITICS

We can affirm this by considering the election outcomes.9 Except for

fluctuations in the immediate postwar elections—usually the result of
temporary surges by Communist parties—most party alignments were
stable. The principal exceptions were France and West Germany. The
former display sharp fluctuations during the Fourth Republic, followed
by the emergence of the Gaullists as a dominant party in the Fifth
Republic. In West Germany, the Christian Democrats and later the
Social Democrats enjoyed sustained growth at the expense of minor
parties. The steadiness of party percentages in most systems suggests
that there were relatively few switches and that there were few voters
available for the growth of catch-all parties.
The picture in the mid-1970s is different. Electorates have become
Downloaded By: [University of Lund] At: 14:30 30 November 2009

more volatile. However, there is little evidence for the growth or

ascendancy of catch-all parties. Instead, several systems have become
more fragmented as minor—and often newly created—parties have
gained votes at the expense of established parties.
Developments in Denmark, Norway, Belgium, the Netherlands,
and Great Britain are particularly interesting. In each of these countries,
'normal' party configurations have been stretched to include a host of
minor parties articulating protests against tax policies, communal
claims, left and right-wing socialism, and demands for furthergoing
démocratisation. In Denmark, Glistrup's Progress Party, organized in
protest against high levels of taxation, won 15*9 per cent of the popular
vote in 1973 and the Centre Democrats, a right-wing splinter of the
Social Democratic Party won 7-8 per cent. At the same time, the Social
Democrats, who had roughly 40 per cent of the vote in the 1950s and
1960s, dropped to 25-7 per cent. In Norway, drastic shifts occurred in
the aftermath of the Common Market referendum. The Labour party
fell from 46-5 per cent in 1969 to 35-3 per cent in 1973, while the Liberals
divided in two because of the E.E.C. issue, fell from 9-4 per cent to
2-3 per cent. The pro-market liberals, organized as the New People's
Party, won only 3-4 per cent. As in Denmark, the main beneficiaries
were minor parties: the Christian People's Party rose from 7-8 per
cent to 11'9 per cent while the Socialist Electoral Alliance, consisting
of the Communists, the Socialist Peoples Party, and independent groups
(anti-E.E.C. Labour factions) won 11-2 per cent. Anders Lange's Party,
a protest group modelled after Glistrup's Progress Party, won 5 per
cent. In Belgium, Flemish and Francophone federalist parties have
gained at the expense of the Christian Social Party, the Socialists, and
the Liberals. In the Netherlands, the confessional parties have been
losing strength. The decline of the Catholic Party—from 31-9 per cent
in 1963 to 17-7 per cent in 1972—has facilitated the entry and success
of several minor parties, and more recently the growth of the Socialists
and Liberals. In Britain, the resurgence of the Liberals and the growth
of the Scottish National Party in 1974 have cut into the strength of
 WESTERN EUROPEAN PARTY SYSTEMS REVISTIED 9

both Labour and Conservatives. Though developments in these

countries may prove to be temporary, they cast doubt on the validity
of the catch-all scenario.
We can measure increases and decreases in the fragmentation of
party systems by using Rae's index of fractionalization (F). The index
measures 'the proportion of pairs of members in a system which contain
persons who have voted for (or belonged to) different parties in the last
previous election.'10 The index varies from zero (complete agreement)
to one (complete disagreement). If catch-all parties were gaining
strength from other parties in a given system, we would expect to find
a decrease in the level of electoral fractionalization (Fe) from election
to election. Figure 1 graphs the level of electoral fractionalization for
Downloaded By: [University of Lund] At: 14:30 30 November 2009

parliamentary elections from 1945-1975 for thirteen Western European

democracies.
The graphs in Figure 1 provide little support for the catch-all
scenario. In only four countries—West Germany, France, Austria, and
Ireland—is there any evidence for a sustained decline in electoral
fractionalization. Of these, the decline is sharpest in West Germany and
Fifth Republic France. Neither the Austrian nor the Irish party systems
were highly fragmented to start with. Four other systems—Finland,
Italy, Sweden, and Switzerland—have had fairly constant levels of
fractionalization, while the remaining five countries—Belgium, the
Netherlands, Denmark, Norway, and the United Kingdom—have
become more fragmented. In some instances—Britain and Norway—
increased fragmentation is a recent development, while in others the
increases occur over a longer period. As is apparent from Table 1,
increased fragmentation is a phenomenon of the 1960s and 1970s
rather than the postwar period as a whole. While all but four countries
register some increase in fractionalization for the entire period, most
systems registered slight declines between 1945 and 1960. In any case,
there is very little reason to believe that catch-all (or any other kinds of)
strategies have produced consolidation in the majority of party systems.
On the contrary, there have been increases in fragmentation in several
countries and three of the four countries with steady levels of frag-
mentation have remained stable at relatively high levels of fraction-
alization.
 10 WEST EUROPEAN POLITICS

FIGURE 1.

Fragmentation in thirteen Western European Party Systems 1945-1975:

electoral fractionalization (Fe) by election year.

FE—Y AXIS
AUSTRIA BELGIUM

1-0 10

0-9 09
Downloaded By: [University of Lund] At: 14:30 30 November 2009

0-R 08
al -

0-7 0-7 /
\ / \
0-fi / 06
"•™
/
0-5 0-5
45 49 53 56 59 62 66 70 71 75 46 49 50 54 58 61 65 68 71 74

13ENMARK FINLAND

•0 10

09 0-9

OR / 06 y S y
\ /
0-7 0-7

OR 0-6

0-5 0-5
45 47 5 0 53 53 57 60 64 66 68 70 73 75 45 48 51 54 58 62 66 70 72 75
A PR. SER
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 11
FIGURE 1—cont'd
FRANCE WEST GERMAN"!

1-0 10

0-9 0-9

0-8 0-8

0-7 0-7 \
\
0-6 n-fi
Downloaded By: [University of Lund] At: 14:30 30 November 2009

0-5 0-5
45 46 46 5! 56 58 62 67 68 73 49 53 57 61 65 69 72
JUN- NOV

IRELAND ITAC <

10 1-0

09 0-9

0-8 0-R

0-7 0-7 \ /
i i 1
\ \ /
0-6 0-6

0-5 O-fi
48 51 64 57 61 65 69 73 46 48 53 58 63 68 72
 12 WEST EUROPEAN POLITICS

FIGURE 1 —cont'd
NETHERLANDS NORWAY

1-0

0-9
<**
08 /
***

0-7

0-6
Downloaded By: [University of Lund] At: 14:30 30 November 2009

0-5
46 48 52 56 59 63 67 71 72 45 49 53 57 61 fc5 69 73

SWEDEN SWITZERLAND

1-0 — 1-0

0-9 09

0-8 0-8
;
07 |
0-7

06 0-6

05 0-5
48 52 56 58 60 64 68 70 73 47 51 55 59 63 67 71 75
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 13

FIGURE 1 — cont'd
Downloaded By: [University of Lund] At: 14:30 30 November 2009

UNITED KINGOOM

1-0

09

0 08

0-7
M B

0-6 /
/
05
45 50 51 55 59 64 66 70 74 74
FEB OCT

The observed pattern deviates from the predictions extracted from

the catch-all scenario. If Kirchheimer's assertions were correct, then
we should have discovered decreased fragmentation in a number of
party systems. Instead, increased fluctuations in the 1960s and 1970s
have been paired with the emergence and success of minor parties.
There are two questions which we must answer : first, why have catch-all
 14 WEST EUROPEAN POLITICS

TABLE 1
CHANGES IN ELECTORAL FRACTIONALIZATION ( F e ) IN WESTERN
EUROPEAN PARTY SYSTEMS, 1 9 4 5 - 1 9 7 5

overall*
number mean Fc net net" net3 change
of 1945-75 change change change 1940's-
elections 1945-75 1945-60 1960-75 1970's
Austria 10 •59 + 01 + •05 -•04 -•04
Belgium 10 •71 + •08 -•02 + •10 + 09
Denmark 13 •77 + •04 -•04 + •08 + •08
Finland 10 •81 +•03 + •03 0 + •03
France 0 10 •78 -01 + •05 -•06 -•04
Downloaded By: [University of Lund] At: 14:30 30 November 2009

West Germany 7 •65 -•21 -•14 -•07 -•16

Ireland 8 •68 -•11 —06 -•05 -•08
Italy d 7 •74 -•04 -05 + •01 + •03
Netherlands 9 •81 + •06 -•01 + •07 + •06
Norway 8 •74 + •05 -•04 + •09 + •02
Sweden 9 •70 + •01 -•01 + •02 + •01
Switzerland 8 •81 + •02 -•01 + •03 + •03
United Kingdom 10 •60 + •08 -•04 + •12 + -09c
a. based on election closest to 1960 (or election closest to the median election
year if there were two elections equally spaced before or after 1960: 1958 for
France, 1962 for Finland).
b. Overall change = mean F e for last two elections minus mean F e for first two
elections.
c. Gaullist and Republican Independents combined in 1973 election data.
d. variations in Italian scores reflect the difference in fractionalization for 1946
and 1948 elections (-79 in 1946, -66 in 1948).
e. based on the average between the February and October 1974 elections.

parties emerged and gained strength in some countries but not in others,
and second, why has increased fluidity led to increased fragmentation
rather than the growth of catch-all parties?
Neither question lends itself to a simple answer. Kirchheimer argued
that decreasing loyalties to mass parties would force parties to adopt
catch-all strategies and assumed that parties employing these strategies
would be more successful than others. But, although we can find
examples of parties moderating their doctrines or adjusting their
appeals,11 outside of France or West Germany, we have little evidence
for the growth or success of catch-all parties. To account for this, we
need to know more about the nature of voters' attachments to parties
and more about the ways in which these attachments condition changes
in party alignments.
In subsequent sections, we will argue that in most systems in the
first fifteen or twenty years after World War II, strong and persistent
attachments to political parties put a brake on the emergence of
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 15

catch-all parties, and that large catch-all parties succeeded only in

systems in which voters' attachments to parties were weak or non-
existent. These were primarily countries which experienced severe
disruptions in political development and countries whose common
characteristic was neither a decline of ideology nor a disappearance of
conflict, but rather crisis and its successful resolution. Elsewhere, few
loose votes were available for changes in party systems. In recent years,
though, voters' attachments to political parties have weakened. How-
ever, because the growth of a floating vote has coincided with the
emergence of new concerns and anti-government reactions, these votes
have been available not only to potential catch-all parties but also to a
host of smaller parties. One reason why this has occurred is that many
Downloaded By: [University of Lund] At: 14:30 30 November 2009

established parties have indeed undergone the changes in strategy,

style, and functions which Kirchheimer indicated. As a result, they are
less able to fulfil the expressive functions of political parties.

I I I . MASS ATTACHMENTS AND CHANGES IN PARTY ALIGNMENTS

Mass attachments are crucial to the study of changes in party align-
ments. Party alignments change when there are either significant
changes in the number of parties competing or when there are sub-
stantial changes in the proportion of the electorate won by different
parties. If voters have no links to parties—i.e., if every election is a
new and random event, then the likelihood of changes in party align-
ments is high: the market is wide open and voters are available to all
comers. On the other hand, if voters have deep and abiding loyalties
to parties, then the number of voters available for either shifts in the
party balance or the entry of new parties is necessarily more limited:
the stronger and more widespread the attachments, the more structured
the electorate and the less the likelihood of change. The weaker the
attachments, the more fluid the electorate, and the greater the likelihood
of change.
Unfortunately, there is little agreement on the nature of voters'
attachments to political parties. The most commonly held view derives
from the work of the Survey Research Center: voters develop party
identifications which predispose them to vote for the same party in
election after election. Recently this has come under attack. Some
students of voting behaviour in Western Europe have discovered
surprisingly low levels of party identification, while others have found
that party identifications change as frequently as voting preferences.
The role of party identification in American voting behaviour is seen
as an artifact of the number of elected positions : voters are said to need
underlying predispositions to simplify the choices confronting them.
This is not the case in Western Europe. Elections are less frequent and
there are fewer decisions to be made. Parties are more closely tied to
social groups, and voting decisions are said to reflect underlying class
 16 WEST EUROPEAN POLITICS

or group loyalties, ideological preferences, or exposure to the cues and

pressures of churches and trade unions, or the influence of segmented
social structures.12
It is neither necessary nor possible to resolve this dispute here. For
the purpose of the present analysis, we will assume that the deter-
minants of voting behaviour vary within and among political systems
and examine the ways in which different determinants affect the like-
lihood of changes in party systems. The impact of party identifications
is well known: party identifications are said to intensify with age or
length of psychological membership. If these same identifications are
passed from parents to children, then party alignments are likely to
continue from generation to generation. Change will occur only if the
Downloaded By: [University of Lund] At: 14:30 30 November 2009

salience of candidates or issues causes voters to deviate from their

underlying predispositions or if new voters enter the electorate with
substantially different identifications and loyalties.13
The consequences of other types of attachments are not all that
different from those of party identifications. If voters decide on the
basis of class or group interests, and if there is a clear connection
between the class or group and support for a particular party, it is
likely that loyalties to parties will develop, be reinforced over time,
and be passed down from generation to generation.14 The same could
be said of voters motivated primarily by ideological considerations.
However, in this instance, we should expect somewhat greater variation
in party choice because these voters will be more likely to measure party
performance against their own ideals.15 Even if voters are not self-
motivated, but are mobilized by the pressure and influence of organ-
izational networks, we can still expect a high degree of continuity:
voters will remain loyal to their parties as long as the organizational
networks and pressures persist and as long as voters remain exposed
to them. Should the pressures or cues cease, however, such voters are
likely to be available to all comers.
We can now apply these suppositions to Western European party
alignments. In most countries, party alignments crystallized in the
late 19th and early 20th centuries and reflected the class and religious
differences salient at that time. Because substantial proportions of
these electorates were mobilized by mass parties and trade unions or
by churches or conservative forces competing for the support of newly
enfranchised voters, we can assume that party preferences originally
derived from the combined effects of organizational pressures, class or
group loyalties, and, in some cases, ideological predispositions. To the
extent that these pressures and predispositions were consistent—and
pointed to a vote for one and not for several parties—we can also assume
that voters developed durable commitments to political parties.
Repeated in successive elections, these attachments should harden over
time and be passed down from generation to generation. If so, then as
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 17

long as elections continue to be held, and as long as the same choices

continue to confront voters, then party alignments should continue
with very few changes over time.16 Change will occur only if voters
entering the electorate hold substantially different loyalties than voters
dying out, or if voters change their preferences. The latter will depend
on the availability of alternatives: if there is more than one party
representing a particular group or class or more than one party rep-
resenting an ideology, then we can probably expect more frequent
shifts. Nevertheless, because many voters are likely to be exposed to
consistent pressures and influences, we can still expect a high degree of
continuity in party alignments.
We can now return to our earlier discussion. In our survey of Western
Downloaded By: [University of Lund] At: 14:30 30 November 2009

European alignments, we found catch-all parties emerging and gaining

strength in only a few countries while party alignments elsewhere
remained fairly stable. If the foregoing is correct, then the explanation
for the stability of party alignments in most countries in the 1950s and
1960s is fairly simple: if catch-all parties are to succeed, votes must be
available for changes in the party alignment. Kirchheimer thought that
the advent of an affluent and consumer-oriented society would weaken
attachments to parties, but in the short run, attachments already
formed should be reinforced rather than eroded.
There are several reasons for this: first of all, although increased
prosperity, and changes in income or occupation may alter class
structures, this will not happen overnight, nor will it have immediate
effects on voting behaviour. Voters are likely to retain earlier pre-
dispositions and remain exposed to the same organizational pressures
and cues. Second, parties which have shared in power should be able
to retain followers' loyalties by adjusting their appeals and claiming
credit for prosperity and well-being. Even if they are not in power,
parties should be able to take a different tack and argue that things would
be better if they were. In short, although changes in party style may
occur, if existing parties are able to evoke underlying loyalties, then
few changes in party alignment will occur because few loose votes will
be available. Only in systems in which underlying loyalties to parties
never crystallised, or in systems where these loyalties have somehow
been eroded, are catch-all strategies likely to succeed or produce
changes in party configurations.
Let us look more closely at the countries in which Kirchheimer
claimed that the catch-all phenomenon was most advanced. Kirch-
heimer names four: Britain, France, West Germany, and Italy. All of
these countries have broadly based parties, but only three—France,
Germany, and Italy—have experienced substantial changes in their
party alignments in the postwar years. However, what these three have
in common is not a record of widespread affluence or a decline of
conflict, but rather common legacies of disrupted political development.
 18 WEST EUROPEAN POLITICS

In Britain, the decline of the Liberals and the rise of the Labour party
was an interwar phenomenon and one which reflected changes in the
composition of the electorate (the completion of suffrage extension)
and splits within the Liberal party. The existence of broadly based
parties in the postwar period is more a reflection of two-party com-
petition than changes in class structure or postwar affluence. In Italy,
substantial changes in the party alignment took place immediately
after World War II, but these reflected the Vatican's decision to cast
its lot with democracy, accept a secular Italian state, and mobilize
within it. The rapid growth of Christian Democracy in the late 1940s
was a direct result of this decision. Changes in the party alignment
reflected the completion of long delayed electoral mobilization.17 Since
Downloaded By: [University of Lund] At: 14:30 30 November 2009

the 1940s, the Italian party alignment has been frozen. Until recently
the only real change has been the slow but steady growth of the
Communist party as younger voters entered the electorate.
Changes in France and West Germany coincide with rapid economic
growth and the rise of consumer-oriented societies. Nevertheless,
although prosperity undoubtedly contributed to the growth of the
Gaullists in Fifth Republic France, and the Christian Democrats and
Social Democrats in West Germany, other factors are more telling.
Both countries share a common legacy of crises, disruptions in political
development, and changes in the rules of electoral competition. In
France, party attachments in the Fourth Republic were weak and the
electorate was extremely volatile.18 The rise of the UDR reflected not
only the prosperity of the Fifth Republic, but also constitutional
changes, the political magnetism of de Gaulle and the boon of political
stability. Crises and constitutional changes were especially important.
The collapse of the Fourth Republic provided de Gaulle with an
opportunity to assert his leadership, while the change from a parlia-
mentary to a presidential regime and the introduction of a directly
elected Presidency altered the conditions of competition. The lure of a
powerful, indivisible, and directly elected Presidency was sufficient to
encourage bloc formation. The coalescence of the Gaullist family in
turn encouraged alliances on the left.19
In West Germany, the growth of the CDU and the SPD and the
consolidation of the party system reflect not only the economic miracle,
but also disruptions in political development, the reconstruction of the
political system, and changes in the constitution and the electoral law.
The party systems of both Imperial Germany and the Weimar Republic
were extremely fragmented. The change from a polarized multi-party
system to the present three-party configuration took place in the early
1950s in a divided and truncated country. Twelve years of Nazi rule
had destroyed much of the old social structure. Though a multi-party
system re-emerged in 1949, the 5 per cent threshold successively
eliminated smaller parties. The three parties which survived were among
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 19

the four originally licensed by the occupying powers. Two of these were
built on the bases of former mass parties—the prewar SPD and the
Centre Party.20
The political systems in which catch-all parties have succeeded
differ from other Western democracies. France, Germany, and Italy
have each experienced major crises which hindered either the formation
of the nation state or the transition from oligarchy to liberal democracy.
Disruptions in the development or operation of the party system either
eroded voters' attachments to parties or prevented their formation in
the first place, leaving sizable portions of the electorate available for
the growth of catch-all parties. The histories of the Low Countries,
Scandinavia, Switzerland, and Britain are different: both the formation
Downloaded By: [University of Lund] At: 14:30 30 November 2009

of the nation-state and the transition to liberal democracy proceeded

more smoothly. As a result, politics has been less marred by the deep
cleavages, scars, and recriminations that we find in France, Italy, or
prewar Germany. Because political systems have been more stable,
party systems have had greater continuity, and voters have had more
chance to develop strong attachments to political parties. Because many
parties had blocks of loyal supporters, and because these loyalties
could be transferred from generation to generation, there were few
votes available for either the growth of catch-all parties or the entry
of newer parties. In the last few years, however, electorates have
become fluid, but so far, this has led not to the growth of catch-all
parties, but rather to increased fragmentation.21 How can we explain
this?

I V . PARTY SYSTEMS IN THE 1970s: SOURCES OF FRAGMENTATION

Earlier, we argued that in the short run, economic growth and prosperity
should reinforce partisan loyalties. The long run effects are likely to be
different. Electorates are constantly renewed by the death of older
generations and the entry of younger cohorts. Because younger voters
often share parental loyalties, this usually produces only gradual change
in party alignments. However, in advanced industrial societies, new
generations entering the electorate are likely to be available to a wider
range of parties. There are several reasons for this: because of greater
physical and social mobility, younger voters are likely to find themselves
in different locations in class and occupational structure. They are also
less likely to be exposed to the pressures of organizational networks.
Shifts into white collar and service sector occupations are likely to
reduce the influence of trade unions, and decreasing religiosity weakens
the influence of churches. Instead, electronic media are likely to play a
greater role in disseminating political information and providing cues
for voters. At the same time, voters raised in an affluent and materially
secure society are likely to have different values than voters who came
of age during the depression, war, or immediate postwar years. Because
 20 WEST EUROPEAN POLITICS

basic material needs have been assured, such voters are less likely to
respond to traditional economic or religious appeals.22
Thus far, we have been describing conditions which should facilitate
the growth of catch-all parties. However, both the emergence of new
cleavages and the very strong possibility of reactions against the
politics and policies of postwar welfare states make fluid electorates
equally available to minor parties. As we noted earlier, Kirchheimer's
scenario is based on the assumption that the principal change in the
postwar world has been a decline of ideology. This is not true. Increased
consensus on the desirability of a welfare state or a managed economy
has not led to a decline of political conflict, but rather a more complex
pattern of cleavage. In many countries, the decreased salience of class
Downloaded By: [University of Lund] At: 14:30 30 November 2009

and religious issues has permitted the emergence of secondary cleavages

and new concerns. In some instances, communal issues, based on
language or ethnic differences, have come to the fore. Elsewhere,
uneven economic development has exacerbated tensions between centre
and periphery and generated regional demands. In other cases, changes
in party ideologies or changes in party positions have spurred the
formation of factions or the organization of new parties: for example, in
several countries, either the abandonment of Marx or support for
NATO has produced schisms in Social Democratic parties. More
recently, issues such as Common Market membership have divided
parties in Britain, Denmark, and Norway. Where there were no latent
issues, new ones have sometimes been created—e.g. demands in the
Netherlands for improvements in the quality of life or quality of
democracy. At the same time, other voters continue to respond to more
traditional class or religious appeals. The result is an exceedingly
complex pattern of cleavage, difficult to aggregate into only two or
three political parties.
This has been compounded by growing reactions against the politics
and policies of the postwar welfare state. Since World War II, govern-
ments in Western Europe have been increasingly responsible for the
management of the economy. Providing full employment and ensuring
a minimum level of economic security has produced considerable
expansion in government activities and placed an increasingly high load
on decision-makers. Governments are now held responsible for the
well-being of the entire society. Although governments and governing
parties can be praised for good times, they can also be blamed for any
shortcomings or reversals which occur. The affluence and prosperity
of the 1960s generated high expectations—voters began to regard
material security, increased prosperity, and continuing improvements
in the quality of life as natural and unending. However, the costs of
existing programmes have sky-rocketed, and governments find it
increasingly difficult to expand or maintain existing programmes without
unacceptable increases in taxation. At the same time, the inability of
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 21

governments to control inflation and maintain full employment has

generated additional strains. The combination of high expectations,
increased load on governments, and disappointments can lead to sharp
reactions against those in power.
The emergence of new concerns and growing reactions against
government policies facilitate the entry and growth of minor parties.
Kirchheimer assumed that catch-all parties would thrive because they
could appeal to a wide range of voters. However, although parties
may be able to modify their positions somewhat in order to retain
doubtful supporters or win new voters, few parties can stretch their
appeals indefinitely. Appeals aimed at one group may alienate others.
The emergence of new cleavages and anti-government reactions places
Downloaded By: [University of Lund] At: 14:30 30 November 2009

established parties at a severe disadvantage vis-à-vis minor parties.

Although established parties may want to bid for additional support,
they are constrained by past habits, and by the need to retain existing
supporters and placate factions within the party. They must also defend
past policies.23 As a result, established parties may not be able to react
quickly or credibly enough to pick up new demands or mobilize dis-
content. New or minor parties are neither burdened by the weight of
existing organization nor compromised by participation in cabinets
or complicity in government policy. Free of the constraints of office,
they are better able to articulate new issues and concerns, better able
to mobilise discontent, and better able to fulfil the expressive functions
of political parties. Lack of organization is no obstacle: the party's
message can be spread easily and rapidly via the mass media.
It is this greater ability to fulfil the expressive functions of political
parties which accounts for the fragmentation of European party systems
in the 1970s. If we survey the political systems which have changed, it
should be obvious that the immediate causes of fragmentation vary
from system to system. In Belgium, increased fragmentation in the 1960s
reflects the increased salience of the language issue and tension between
two peripheries.24 In the Netherlands, changes in the party system in
the late 1960s and early 1970s are the product of reactions to con-
sociational practices and the perceived lack of opposition among the
five major parties, the emergence of demands for furthergoing demo-
cratization of political and social life, and changes in the Dutch Catholic
Church and in the Catholic subculture which created a large pool of
voters available for changes in the party system.25 In Denmark, the
sources of change are protest against high levels of taxation, dis-
gruntlement over the similarity of policies pursued by different coali-
tions, and disagreements arising out of the Common Market issue.26
In Norway, the principal factors were divisions within established
parties in the aftermath of the Common Market referendum and the
emergence of a protest dimension, smaller in scale than in Denmark.27
In the United Kingdom, fragmentation reflects economic downturns,
 22 WEST EUROPEAN POLITICS

the apparent failure of essentially similar policies tried by both major

parties, and the growth of Scottish nationalism.
The parties which have profited from the unfreezing of party align-
ments are as diverse as the sources of fragmentation. The beneficiaries
include Poujadist-style protest parties (the Farmers' Party in the
Netherlands, the Progress Party in Denmark, Anders Lange's Party
in Norway), fundamentalist parties in Scandinavia (Christian People's
Parties in Denmark and Norway), parties expressing communal claims
(the Flemish and Francophone parties in Belgium, Scottish
Nationalists), centre or 'progressive' parties (Democrats '66 in the
Netherlands, the Liberals in Britain), right-wing Socialist parties
(Democratic Socialists '70 in the Netherlands, the Centre Democrats
Downloaded By: [University of Lund] At: 14:30 30 November 2009

in Denmark), and parties of the left (the Radical Political Party in the
Netherlands, the Socialist People's Party in Denmark, the Socialist
Electoral Alliance in Norway). This diversity reflects the variety of
complaints in advanced industrial societies and the openness of the
process under way. There is no automatic relationship between changes
in social structure, and increased support for particular kinds of parties.
Instead, the changes which occur depend on the structure of the
electorate and on political activists, the concerns they articulate, the
strategies they choose, and the ways in which parties and politicians
respond.
Although we cannot predict the type of parties which will benefit,
we can specify the circumstances under which changes are likely to
occur. Two factors are crucial: the role of opposition parties within
the political system and the electoral law. If opposition parties are able
to articulate grievances and serve as a channel for discontent, dis-
gruntled voters will probably shift to the 'normal' opposition parties
rather than gamble on untried alternatives. However, this will depend
on the speed with which opposition parties pick up new themes and the
degree to which opposition parties are seen as complicit in existing
policies and practices. The more that opposition parties have shared
in power and moved in and out of coalitions, the less successful they will
be in articulating new concerns or channelling discontent. The impact
of opposition is amply illustrated in the contrasts among the Scandi-
navian countries: in Denmark and Norway, bourgeois coalitions have
been in power from time to time but produced no visible changes in
government policies or rates of taxation. In Sweden, Social Demo-
cratic hegemony left the bourgeois parties untainted by participation
in cabinets. As a result, they were better able to absorb and channel
the kinds of discontent which led to fragmentation in Denmark and
Norway.28 Significantly, in several of the systems which have become
more fragmented, either lines between government and opposition
have been blurred by shifting coalitions (Netherlands, Belgium, and to
some extent Denmark) or else opposing parties and coalitions have
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 23

ended up carrying out similar—and not always successful—policies

(e.g., Britain, Denmark, the Netherlands).
Electoral laws are important because they regulate the entry of new
parties. We can classify electoral laws as permissive if they facilitate the
entry of new parties or as restrictive if they hinder their entry or
success.29 Intuitively, we should expect that the more permissive the
electoral law, the greater the likelihood of increased fragmentation:
the more permissive the electoral law, the more likely that minor
parties will be available as alternate receptors for loose votes, and the
more likely that disputes within parties lead to schisms and the for-
mation of new parties. As Table 2 demonstrates, there is a relationship
between the permissiveness of list systems of proportional representation
Downloaded By: [University of Lund] At: 14:30 30 November 2009

and changes in electoral fractionalization: countries with either very

permissive or moderately permissive electoral laws are more likely to
become fragmented than countries with very restrictive electoral laws.

TABLE 2
THE RELATIONSHIP BETWEEN CHANGES IN ELECTORAL FRACTIONALIZATION AND THE
PERMISSIVENESS OR RESTRICTIVENESS OF ELECTORAL LAWS:

overall
net net net change
mean F e change change change 1940's-
1945-75 1945-75 1945-60 1960-75 1970-s

I. List systems of proportional representation:

very permissive
Netherlands •81 + •06 -01 + •07 + •06
Italy •74 -•04 -•05 + •01 + •03
Finland •81 + 03 + 03 + •03
moderately permissive
Belgium •71 + •08 -•02 + •10 + 09
Denmark •77 + 04 -•04 + •08 + •08
Norway •74 + 05 -04 + •09 + •02
Switzerland •81 + •02 -01 + •03 + 03
restrictive
Austria •59 + •01 + •05 -•04 -•04
Sweden •70 + 01 -01 + •02 + •01
West Germany •65 -•21 -•15 -•06 -•16
11. Other electoral systems :
Ireland
(single transferable vote) •68 -•11 -•06 -•05 -•08
France (double ballot) •78 -•01 + •05 -•06 -•04
United Kingdom
(single member, plurality) •60 + •08 -•04 + •12 + 09

List systems of proportional representation classified according to district magnitude

and presence or absence of formal legal thresholds. For details, see note 29. For
explanation of net change and overall change, see table 1.
 24 WEST EUROPEAN POLITICS

Nevertheless, the relationship is weak (some countries with highly

permissive electoral laws have become more fragmented while others
have not) and we should not overestimate the impact of electoral laws.
No matter how restrictive the provisions of the electoral law. there are
invariably 'incoercible minorities' which compete despite the dis-
advantages and remain available as alternate receptors for loose votes.30
Only in countries with very restrictive electoral laws—formal legal
thresholds of 4 per cent in Sweden and 5 per cent in West Germany—
does the electoral law present a formidable barrier to new parties. Even
so, the 5 per cent threshold in Germany did not prevent the National
Democratic Party from trying to leap the barrier, and it is possible that
the absence of new parties in Sweden and Germany reflects not the
Downloaded By: [University of Lund] At: 14:30 30 November 2009

electoral law, but rather clearer distinctions between government and

opposition. In several countries, the share of the vote won by new
parties has been sufficient to leap electoral barriers several times over:
the threshold in the Netherlands is only 2/3 of one per cent, but neither
the 2 per cent threshold which the Danes use, nor the 4 per cent threshold
which the Swedes have implemented would have blocked the rise of
Democrats '66 (4-5 per cent in 1967), the Farmers' Party (4-8 per cent
in 1967), or Democratic Socialists '70 (5-3 per cent in 1971). In Denmark,
neither a 4 per cent nor a 5 per cent threshold would have excluded the
Progress Party (15-9 per cent) or the Centre Democrats (7-8 per cent) in
1973. However, higher thresholds might have discouraged the formation
of new parties in the first place, and changes in electoral laws—the
adoption of more stringent provisions—might encourage consolidation.

V. CONCLUSION: ALTERNATING FRAGMENTATION AND CONSOLIDATION?

We began this essay by raising questions about the validity of
Kirchheimer's catch-all thesis as a statement of change in Western
European party alignments. From the foregoing discussion, it should
be apparent that the catch-all scenario has greater validity as a statement
of changes in party style and strategy and functions than as a statement
of changes in party alignments. Catch-all parties have succeeded only
in countries with low levels of partisan attachment which in turn reflect
crises and discontinuities in political development. The electoral
successes of catch-all parties in countries like France or West Germany
reflect not only the advent of affluent and consumer-oriented societies
but also constitutional changes, political crises, and their successful
resolution. Elsewhere, few catch-all parties emerged because few loose
votes were available. Electorates have now become more fluid, but
increased fluidity has been paired with increased fragmentation. This
has occurred not only because new sources of cleavage have emerged
but also because the politics of present-day welfare states can generate
sharp anti-government and anti-establishment reactions. Precisely
because some of the changes which Kirchheimer outlines have occurred—
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 25

because parties have de-emphasised ideology and pay more attention

to winning office than articulating demands—minor parties are often
better able to express new concerns and mobilise discontent.
Whether or not this will be a permanent development remains to be
seen. Although established parties may be slow to react, they are likely
to prove more durable than new or minor parties. There are a number of
reasons for this: established parties possess greater organisational
resources and are likely, even in decline , to retain some loyal support.
Moreover, electoral decline may well encourage changes in strategy
and tactics. Although minor parties may be better able to express new
demands or mobilise protest, their ability to survive is doubtful. Minor
parties have burst into the limelight propelled by the enthusiasm of
Downloaded By: [University of Lund] At: 14:30 30 November 2009

activists and the attention of the media, but sustaining momentum is

difficult. After a while, volunteers lose their enthusiasm and the interest
of the press and public wear thin. In the long run, minor parties are
likely to lack the resources and the personnel needed to carry out
successive campaigns or to staff the party at local, regional and national
levels. Finally, if minor parties translate their recently acquired strength
into political power they risk becoming compromised by their partici-
pation in policy-making. If so, minor parties will find it difficult to
mobilise discontent and will lose the main advantage which they possess.
This does not mean that party systems will become less fragmented.
Older new parties can easily be replaced with newer ones. In addition,
parties competing in successive elections may acquire small but loyal
followings. Because a large number of parties retain some support, the
market is obstructed and fragmentation persists. For catch-all parties
to succeed, what is needed is a substantial change—either further
erosion of previous loyalties or the elimination of several parties—
clearing the way for the growth of the remaining parties. This is not
likely to occur without either changes in the electoral law or crises which
erode voters' attachments and provide the opportunity for the
emergence of new leaders and new coalitions. But it is precisely this
kind of realignment, prevalent in North America, which has been
missing in Western Europe.
In the United States, the emergence of new cleavages and new
issues—and particularly the rise of third parties—has produced several •
rearrangements of the party system. Realignment has occurred when
issues could no longer be contained within the existing two-party
system. Realignment was possible because attachments to parties were
weak enough that they could be changed and necessary because the
winner-take-all character of Presidential elections required broad coali-
tions which could not be rearranged without realignment.31 In Western
Europe, the strength of class and religious attachments and proportional
representation inhibits comparable realignments. Divisive issues which
split parties in North America—parties which are essentially little more
 26 WEST EUROPEAN POLITICS

than coalitions in any case—can be dealt with either by forming new

parties or by shuffling cabinet coalitions. These options are not available
in the United States. Realignment becomes a crucial device for re-
arranging coalitions and accommodating demands for change.
What of present-day Western Europe? Class and religious ties are
now weaker and voters are available for changes in party systems, but
whether issues sufficiently divisive to force changes in party alignments
will arise remains to be seen.32 The use of proportional representation
and the absence, except in France, of directly elected heads of state or
government, encourages schisms rather than the American pattern of
splits and rearrangements. Barring deep systemic crises or changes in
electoral laws which encourage consolidation, we can expect either
Downloaded By: [University of Lund] At: 14:30 30 November 2009

continued fragmentation, or more likely, alternation between periods of

fragmentation and periods of consolidation.33 A good deal depends on
events and how parties and politicians respond to them. Neither are
very predictable.

NOTES
1. Otto Kirchheimer, 'The Transformation of Western European Party Systems'
in Joseph LaPalombara and Myron Weiner, eds., Political Parties and Political
Development (Princeton, N.J.: Princeton University Press, 1966), pp. 177-200.
See especially pp. 181-195, 198-200.
2. Attempts to test different aspects of Kirchheimer's argument include Sidney
G. Tarrow, 'Economic Development and the Transformation of the Italian
Party System,' Comparative Politics, vol. 1, no. 1 (April, 1969), pp. 161-183,
Richard Rose and Derek Urwin, 'Persistence and Change in Western Party
Systems Since 1945,' Political Studies, vol. XVIII, no. 3 (1970), pp. 299-300,
and Alan Zuckerman and Mark Irving Lichbach, 'Stability and Change in
European Electorates,' World Politics, vol. 29 (October, 1977), pp. 523-551.
3. See Richard Rose and Derek Urwin, 'Social Cohesion, Parties, and Regime
Strains,' Comparative Political Studies, vol. 2, no. 1 (April, 1969), pp. 10-11, 16.
4. This is the approach of Zuckerman and Lichbach. See Zuckerman and Lichbach,
op. cit., pp. 524-525ff.
5. Kirchheimer, op. cit., pp. 187-188.
6. Ibid., pp. 192-195.
7. Seymour Martin Lipset and Stein Rokkan, 'Cleavage Structures, Party Systems,
and Voter Alignments' in Lipset and Rokkan, eds., Party Systems and Voter
Alignments (New York: Free Press, 1967), pp. 50-54.
8. Rose and Urwin, 'Persistence and Change,' op. cit., pp. 296-297.
9. Data on election results through 1972 are available in Thomas J. Mackie and
Richard Rose, eds., The International Almanac of Electoral History (New York :
The Free Press, 1974). Data for subsequent elections can be found in the annual
updates by Mackie and Rose in the European Journal of Political Research.
10. Douglas Rae, 'A Note of the Fractionalization of Some European Party
Systems,' Comparative Political Studies, vol. 1, no. 3 (October, 1968), pp. 414.
See also Douglas Rae, The Political Consequences of Electoral Laws, rev. ed.
 WESTERN EUROPEAN PARTY SYSTEMS REVISITED 27

(New Haven: Yale Press, 1971), pp. 53-58. The simplified computing formula
for the statistic is :
m
F = 1 - S ti2
i =l
where ti = the proportion supporting party i and m = the number of parties.
11. See John Clayton Thomas, The Decline of Ideology in Western Political Parties :
A Study of Changing Policy Orientations (Beverly Hills, Calif: Sage Professional
Papers in Comtemporary Political Sociology, 1975), pp. 26-28, 41-43.
12. See for example W. Phillips Shively, 'Party Identification, Party Choice, and
Voting Stability,' American Political Science Review, vol. 66, no. 4 (December,
1972, pp. 1204-1206, 1222-1225, or Jacques Thomassen, 'Party Identification
as a Cross-National Concept: Its Meaning in the Netherlands' in Ian Budge,
Ivor Crewe, and Dennis Farlie. Party Identification and Beyond (London:
John Wylie & Sons, 1976), pp. 63-80. See also David Butler and Donald
Downloaded By: [University of Lund] At: 14:30 30 November 2009

Stokes, Political Change in Britain, 2nd. ed. (London: Macmillan, 1974),

pp. 39-47.
13. See Angus Campbell, Philip Converse, Warren Miller, and Donald Stokes, The
American Voter (New York: John Wiley and Sons, 1960), pp. 120-146, 523-531,
and Philip Converse, 'Of Time and Partisan Stability,' Comparative Political
Studies, vol. 2, no. 2 (July, 1969), pp. 141-148ff., 165-167. See also Butler and
Stokes, op. cit., pp. 48-67.
14. See for example Butler and Stokes, op. cit., pp. 58-66. The authors downplay
the importance of party identifications (partisan self image) but nevertheless
place considerable emphasis on life cycle effects.
15. See Ole Borre and Daniel Katz, 'Party Identification and Its Motivational
Base in a Multiparty System: A Study of the Danish General Election of 1971,'
Scandinavian Political Studies, vol. 8 (1973), pp. 94-97.
16. See Lipset and Rokkan, op. cit., pp. 50-54.
17. Lipset and Rokkan, op. cit., pp. 50-52.
18. Philip Converse and Georges Dupeux, 'Politicization of the Electorate in the
United States and France' in Angus Campbell, et. al., Elections and the Political
Order (New York: John Wiley and Sons, 1966), pp. 277-283.
19. See David Cameron and Richard Hofferbert, 'Continuity and Change in
Gaulism: The General's Legacy,' American Journal of Political Science, vol. 17,
no. 1 (February, 1973), pp. 78-83.
20. Gerhard Loewenberg, 'The Remaking of the West German Party System,'
Polity, vol. 1, no. 1 (Fall, 1968), pp. 90-109.
21. Lipset and Rokkan, op. cit., pp. 50-51.
22. Ronald Inglehart, 'The Silent Revolution in Europe: Intergenerational Change
in Post-Industrial Societies,' American Political Science Review, vol. 65, no. 4
(December, 1971), pp. 991-993ff., 1009-1017.
23. Kirchheimer recognized the problems which parties may face. See Kirchheimer,
op. cit., pp. 189-190, 192-193.
24. See André Philippart, 'Belgium: Language and Class Opposition,' Government
and Opposition, vol. 2, no. 1 (November, 1976), pp. 63-82.
25. For elaboration, see Steven B. Wolinetz, Party Realignment in the Netherlands
(unpublished Ph.D. dissertation, Yale University, 1973), pp. 62-78, 91-108.
26. See Ole Borre, 'Recent Trends in Danish Voting Behavior' in Karl H. Cerny, ed.,
Scandinavia at the Polls: Recent Political Trends in Denmark, Norway, and
Sweden (Washington, D.C.: American Enterprise Institute, 1977), pp. 3-38.
27. See Henry Valen and Willy Martinussen, 'Electoral Trends and Foreign
Politics in Norway: The 1973 Storting Election and the EEC Issue' in Cerny,
op. cit., pp. 39-72.
 28 WEST EUROPEAN POLITICS

28. See Stein Rokkan and Henry Valen, 'Norway: The Election to the Storting in
September, 1973,' Scandinavian Political Studies, vol. 9 (1974), p. 217.
29. The classification is based on Rae, The Political Consequences of Electoral Laws,
pp. 42-44, 114-124. In the absence of formal legal thresholds, district magnitude
has been used to classify electoral laws. The larger the district magnitude, the
more permissive the electoral laws. To avoid confusion, the British single-
member plurality system, the French double ballot system, and the Irish single
transferable vote system have been grouped in a residual category. On the
complexity of measuring thresholds of representation, see Douglas Rae, Victor
Hanby, and John Loosemore, 'Thresholds of Representation and Thresholds
of Exclusion: An Analytical Note on Electoral Systems,' Comparative Political
Studies, vol. 3 (January, 1971), pp. 479-488.
30. See Giovanni Sartori, 'European Political Parties: The Case of Polarized
Pluralism,' in LaPalombara and Weiner, op. cit., pp. 165-171.
31. See Walter Dean Burnham, Critical Elections and the Mainsprings of American
Downloaded By: [University of Lund] At: 14:30 30 November 2009

Politics (New York: W. W. Norton & Co., 1970), pp. 6-10, 175-183, and
James L. Sundquist, The Dynamics of the Party System: Alignment and Re-
alignment of Political Parties in the United States (Washington, D.C. : Brookings
Institution, 1973), pp. 29-33, 275-298.
32. The question of Common Market entry in Britain and Norway is an example
of such an issue, but this has proved to be of relatively short duration in both
systems.
33. There is some evidence for greater consolidation in elections since 1975. Older
parties have recovered some of the ground lost earlier in recent elections in the
Netherlands, Denmark, and Norway. In the Netherlands, the merger of the
Protestant and Catholic parties has reduced the level of fragmentation : however,
the potential for further divisions persists. In Denmark, the Progress Party has
retained 13-15 per cent of the electorate in elections in 1975 and 1977. Because
of divisions within the majority family, the French party system has become
more fragmented.