00 Front_Walgrave 22/01/2010 3:35 PM Page i dfc

THE WORLD SAYS NO TO WAR

m

h$h

m

00 Front_Walgrave 22/01/2010 3:35 PM Page ii dfc

Social Movements, Protest, and Contention

Series Editor Bert Klandermans, Free University, Amsterdam

Associate Editors Ron R. Aminzade, University of Minnesota

David S. Meyer, University of California, Irvine
Verta A. Taylor, University of California, Santa Barbara

Volume 33 Stefaan Walgrave and Dieter Rucht, editors, The World Says No
to War: Demonstrations against the War in Iraq

Volume 32 Rory McVeigh, The Rise of the Ku Klux Klan: Right-Wing
Movements and National Politics

Volume 31 Tina Fetner, How the Religious Right Shaped Lesbian and
Gay Activism

Volume 30 Jo Reger, Daniel J. Myers, and Rachel L. Einwohner, editors,

Identity Work in Social Movements

Volume 29 Paul D. Almeida, Waves of Protest: Popular Struggle in
El Salvador, 1925-2005

Volume 28 Heidi J. Swarts, Organizing Urban America: Secular and
Faith-based Progressive Movements

Volume 27 Ethel C. Brooks, Unraveling the Garment Industry:
Transnational Organizing and Women's Work

Volume 26 Donatella della Porta, Massimiliano Andretta, Lorenzo Mosca,
and Herbert Reiter, Globalization from Below: Transnational
Activists and Protest Networks

Volume 25 Ruud Koopmans, Paul Statham, Marco Giugni, and Florence
Passy, Contested Citizenship: Immigration and Cultural Diversity

Volume 24 David Croteau, William Hoynes, and Charlotte Ryan, editors,
Rhyming Hope and History: Activists, Academics and Social
Movement Scholarship

For more books in the series seepage 000.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM F

H$H

THE WORLD SAYS NO TO WAR

Demonstrations against the
War on Iraq

-4

Stefaan Walgrave and Dieter Rucht, Editors

+

Preface by Sidney Tarrow

Social Movements, Protest, and Contention
Volume 33

University of Minnesota Press

Minneapolis • London

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page iv dfc

m

Copyright 2010 by the Regents of the University of Minnesota

Ml rights reserved. No part of this publication may be reproduced, stored in

retrieval system, or transmitted, in any form or by any means, electronic,

deal, photocopying, recording, or otherwise, without the prior

written permission of the publisher.

Published by the University of Minnesota Press

1 1 1 Third Avenue South, Suite 290

Minneapolis, MN 55401-2520

http://www.upress.umn.edu

Printed in the United States of America on acid-free paper
The University of Minnesota is an equal-opportunity educator and employer.

20 19 18 17 16 15 14 13 12 11 10 10 9 8 7 6 5 4 3 2 1

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page v dfc

Contents

Preface vii

Sidney Tarrow

Introduction xiii

Stefaan Walgrave and Dieter Rucht

1 . February 1 5, 2003: The World Says No to War 1
Joris Verhulst

2. Political Opportunity Structures and Progressive

Movement Sectors 20

I Michelle Beyeler and Dieter Rucht

| : 3. Politics, Public Opinion, and the Media: The Issues and J

Context behind the Demonstrations 42

Joris Verhulst and Stefaan Walgrave

4. Legacies from the Past: Eight Cycles of Peace Protest 61
Bert Klandermans

5. New Activists or Old Leftists? The Demographics of Protesters 78
Stefaan Walgrave, Dieter Rucht, and Peter VanAelst

6. Peace Demonstrations or Antigovernment Marches? The

Political Attitudes of the Protesters 98

Bert Klandermans

7. Paths to the February 15 Protest: Social or Political Determinants? 1 19
Donatella della Porta

8. Boon or Burden? Antiwar Protest and Political Parties 141

9. Open and Closed Mobilization Patterns: The Role of Channels

and Ties 169

Stefaan Walgrave and Bert Klandermans

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page vi dfc

10. Promoting the Protest: The Organizational Embeddedness
of the Demonstrators

Mario Diani

11. Crossing Political Divides: Communication, Political
Identification, and Protest Organization

W. Lance Bennett, Terri E. Givens, and Christian Breunig

12. The Framing of Opposition to the War on Iraq
Dieter Rucht andjoris Verhulst

Conclusion: Studying Protest in Context
Stefaan Walgrave and Dieter Rucht

Acknowledgments 273

Appendix A: Methodology of Protest Surveys in Eight Countries

Stefaan Walgrave and Claudius Wagemann 275

Appendix B: Media Content Analysis 285
/oris Verhulst

1 Contributors 289 1

I Index 291

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page vii dfc

m

Preface

Sidney Tarrow

The first decade of this new century witnessed a number of events of global
importance, from the World Trade Center bombing to the wave of terrorism
that followed, from the attack on the Taliban and al-Qaeda in Afghanistan
to the war on Iraq, and from the electoral revolutions in the old Soviet bloc
to the expansion into that area of the European Union, followed by its con-
stitutional failure; and from the launching of the Iraq war in 2003 to the
American withdrawal promised for 2010. These events and others like them
have fundamentally changed global politics.

With the exception of the electoral revolutions, however, most of these
episodes have been bad for humanity and destructive of human rights. The
World Trade Center bombing alone produced a militaristic turn in American
foreign policy and retractions of human rights, both at home and abroad. It
also brought a crisis in the Atlantic alliance and splits in the European Union,
a renewal of Islamist terrorism, and a diversion of the world's attention from
the grave problems of Darfur, HIV/AIDS, and global warming. And it led the
United States to fail to follow up on its initial victory in Afghanistan and to
a recrudescence of Islamist power in both that country and in Pakistan. From
the optimism that greeted the turn of the new century, there has been a turn
to profound pessimism for advocates of global progress and human rights.

Yet not all has been bleak for advocates of world peace, global solidar-
ity, and popular politics. On February 15, 2003, millions of people around
the world combined in the largest example of collective action in history to
protest against the impending war on Iraq. These demonstrations had three
main properties that marked them off from previous peace campaigns:

H$H

-4

00 Front_Walgrave 22/01/2010 3:35 PM Page viii dfc

• Their vast geographic reach

• The enormous numbers of participants they attracted, many of them
with little previous experience of contentious politics

• Their coordinated transnational nature.

Let us not exaggerate. These demonstrations, however vast, did not stop
the rush to war, which had been, in any case, decided long before. And they
did not draw on as broad a range of social groups as was maintained at the
time. Nor did the campaign against the American-led attack on Iraq lead to a
sustained social movement against a conflict that — at this writing — continues
to destroy lives and take treasure and lives from both proponents and victims.
But it was something new in the history of contentious politics: a concerted
campaign of transnational collective action — using the new tools of the Inter-
net and more conventional means of mobilization — which brought millions
of people together around common goals and against common targets.

Of course, there have been transnational campaigns of collective action
before. From the transnational network formed in support of the Chiapas
rebellion in the early 1990s (Olesen 2005), to the "Battle of Seattle" against
the World Trade Association (Lichbach 2003); from the campaign against
land mines in the late 1990s (Cameron, Lawson, andTomlin 1998) to the
wave of global justice protests that followed (della Porta et al. 2007); from
the justice cascade against ex-dictator Augusto Pinochet of Chile (Lutz and
Sikkink 2001a and b; Roht-Arriaza 2006) to the campaigns against global
warming from 2006 on: in the new century, contentious politics has begun
to escape the boundaries of the national state (Tarrow 2005).

But much of this activity has been organized by well-financed NGOs and
lacked a mass base. The most successful operations — like the landmine cam-
paign — depended on the support of allied states (in this campaign, Canada,
for example). Moreover, many of the activities that activists think of as global
have actually taken bilateral form — as in the "boomerang" campaigns doc-
umented by Margaret Keck and Kathryn Sikkink (1998). Much of these have
been carried out by domestic constituencies in the name of global values, as
in the Battle of Seattle, which drew largely on American trade union sup-
porters who quickly disappeared from the global justice coalition after 9/11
(Hadden and Tarrow 2007). Coordinated transnational campaigns on the
part of non-state actors, like the one that burst on the world on February
15, 2003, are still a relative rarity in global politics — thus the great impor-
tance of this book.

But that is not all. In addition to documenting a major new threshold
in the transnational coordination of contentious protest, The World Says No

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page ix dfc

to War is important for three other reasons: methodological, contextual, and
dynamic.

First, methodologically, the book demonstrates conclusively the poten-
tial for the use of coordinated survey techniques for the study of contentious
politics in action. Until recently, with the exception of a pathbreaking strand
of research in France (see Favre 1990), survey research on contentious politics
drew on surveys of the mass public, not activists themselves, or depended on
the memories of activists long after their actions ended (McAdam 1988).
Instead, the organizers of this study accessed the activists in the act of pro-
testing — or very soon afterward — when their images of participation were
still "hot" and not yet warped by selective memory or disillusionment (see
appendix A for details). It will be important to see whether other scholars
take up their method, applying it to other types of protest campaigns, per-
haps digging more deeply into the biographies of the activists and following
up with detailed studies of their itineraries after the protest is over.

Second, from the point of view of comparative contextualization, the
study's coordinators were neither too far from nor too close to their subjects.
They made a reasonable compromise between commissioning separate analy-
ses of the protestors in the eight countries they studied (which might have pro-
| duced in-depth but noncumulative case studies) and carrying out a context-

~W free examination of their massive dataset of roughly 4,500 respondents (which ~W

might have produced a context-free exercise in number-crunching). Their de-
cision was guided by the heuristic process model (see Figure 1. 1) that organizes
the book. They asked each of their authors to focus on either the independent,
the intervening, or the dependent variables in their model. This intermediate
strategy allowed them and their authors to use comparison creatively, relating
the findings of the study to their national contexts, as well as to the sociode-
mographics of their populations and the framing of the protest.

Third, although the study took place at only one point in time-one in
which American hegemonic overreach was at its height — the book provides
us with elements for a process analysis of a transnational movement against
it. It shows how the mobilization of a vast transnational protest campaign
was planned, organized, framed, and carried out. In the process, it shows
that the internet combined with more conventional means of mobilization,
but did not replace them. It also demonstrates the key role of activists from
the then-powerful global justice movement in founding the emerging anti-
war movement. And it shows how key factors in mounting the demonstration
were the European Social Forum in Florence, Italy, and the World Social
Forum in Porto Alegre, Brazil. Though taken at one point in time, the survey
provides precious information on the dynamics of contention.

H$H

-4

00 Front_Walgrave 22/01/2010 3:35 PM Page x dfc

The book is particularly valuable for the light it sheds on individuals.
It shows how key activists with "complex identities" were close to the core
of the mobilization (see W. Lance Bennett, Terri E. Givens, and Christian
Breunig's chapter in this book). It reveals the mix of common and diverse
factors producing the demonstration in the eight countries under its analyt-
ical lens. Finally, it reveals a workable model of cross-national collaboration
permitting the exploitation of a common dataset by scholars with diverse
research perspectives.

Critics might complain that there is little information here about the
diffusion of information in the weeks and months before the protest. And stu-
dents would also have profited from more information about the interaction
between protest organizers and institutional actors in the various countries in
which the demonstrations took place. This would have been particularly use-
ful in the United States, where the Bush administration did all in its power
to "sell" the war to the public, succeeding in stilling the voices of protest that
were so insistent before the first bombs fell on Baghdad. Future researchers
using protester-based survey techniques may wish to add a subsequent phase
of follow-up research on public opinion after the demonstrators go home.

Walgrave, Rucht, and their collaborators have given students of social
movements, antiwar movements, and activism a precious example of how
multicountry research can be carried out by many players, through theoret-
ically based empirical research on the formation and execution of a transna-
tional protest campaign. That alone is a significant achievement that deserves
praise and replication.

Works Cited

Cameron, Maxwell A., Robert J. Lawson, and Brian W. Tomlin, eds. 1998. To Walk
without Fear: The Global Movement to Ban Landmines. Oxford: Oxford Uni-
versity Press.

della Porta, Donatella, Massimiliano Andretta, Lorenzo Mosca, and Herbert Reiter.
Globalization from Below: Transnational Activists and Protest Networks. Minnea-
polis: University of Minnesota Press.

Favre, Pierre. 1990. La Manifestation. Paris: Presses des Sciences Politiques.

Hadden, Jennifer, and Sidney Larrow. 2007. "Spillover or Spillout: Lhe Global Jus-
tice Movement in the United States after 9/11." Mobilization 12: 359-76.

Keck, Margaret, and Kathryn Sikkink. 1998. Activists beyond Borders: Advocacy Net-
works in Interna! haca, N.Y.: Cornell University Press.

Lichbach, Mark I. 2003. "Global Order and Local Resistance: Structure, Culture,
and Rationality in the Battle of Seattle." Unpublished paper, University of Mary-
land Political Science Department.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xi dfc

Lutz, Ellen, and Kathryn Sikkink. 2001a. "International Human Rights Law and
Practice in Latin America." International Organization 54: 249-75.

. 2001b. "The Justice Cascade: The Evolution and Impact of Foreign Human

Rights Trials in Latin America." Chi

Olesen, Thomas. 2005. International Zapatismo: The Construction of Solidarity in the
Age of Globalization. London: Zed.

McAdam, Doug. 1988. Freedom Summer. Oxford: Oxford University Press.

Roht-Arriaza, Naomi. 2006. The Pinochet Effect: Trar, ? in the Age of

Human Rights. Philadelphia: University of Pennsylvania Press.

Tarrow, Sidney. 2005. The New Tn vism. Cambridge: Cambridge Uni-

versity Press.

m

h$h

00 FrontJ/Valgrave 22/01/2010 3:35 PM Page xii ^

4

-^-

00 Front_Walgrave 22/01/2010 3:35 PM Page xiii dfc

Introduction

Stefaan Walgrave and Dieter Rucht

On February 15, 2003, following the global time zones from Australia in the
East to Seattle in the West, a massive flood of protest conquered the streets
throughout the world. Millions of people in more than six hundred cities
worldwide protested against the imminent war on Iraq. This massive demon-

| worldwide protested against the imminent war on Iraq. This massive demon-

~W stration was the culmination point of a sustained protest wave against the ~W

Iraqi war starting in late 2002 and lasting for several months. At that ti
the United States and the United Kingdom were busy preparing for war with
Iraq, but it seemed as though war could be avoided, as the United Nations
Security Council was still debating a potential resolution. Some of the pro-
tests staged on February 1 5 were small-scale, only local marches consisting
of a few neighbors banding together; others were large-scale national pro-
test demonstrations. In Rome, Italy, for example, an estimated record crowd
of 3 million people expressed their anger about the imminent war, in Barce-
lona, Spain, more than 1 million demonstrators appeared, while in Warsaw,
Poland, only ten thousand people exhibited their resistance. Together, the
February 1 5 demonstrations were the largest protest event in human history.
Journalists, observers, politicians, social movement scholars, and political sci-
entists alike were stunned by the magnitude of, dispersion of, and apparent
similarities among the global February 15, 2003, events.

The astonishment on the part of social scientists regarding these protest
events reveals first that these antiwar protests, indeed, were extraordinary;
and second, that social science seems to lack appropriate models to grasp
transnational protest phenomena of their size. This book tackles both these
challenges.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xiv dfc

INTRODUCTION

First, the study aims to significantly enhance our empirical understand-
ing of what happened that day who took to the streets and why and how such
a protest event resulted in mobilizing as large a crowd as it evidently did.
Much has been said and claimed regarding the February 1 5 events; political
actors, movement leaders, journalists as well as social scientists have frequently
voiced opinions pertaining to the lessons to be learned from the,. February 15
did become a myth quickly For example, claims were made that people from
all walks of life participated, that February 15 essentially was a global social
justice action, that the internet was pivotal in coordinating the protest events
and mobilizing the people involved in them, and that local political conflicts
conflated with the antiwar issue. However, there was no empirical base for
these contentions. Drawing on an original protest survey directed at a great
number of February 15 participants, combined with a broad variety of other
materials and sources, our book offers the necessary solid empirical base.

Second, and more important, this book embarks on a quest to improve
general understanding of protest events in a comparative perspective. Con-
sidering the February 1 5 case, our goal is to enhance general knowledge of
the background, nature, and significance of protest politics. Many aspects of
protest are insufficiently understood. To raise only some of the questions: To
| what extent and why does the composition of protest events differ across issues

~W and nations? How do divergent mobilization patterns affect the composition ~W

of protest events? What is the relationship between the social movements'
infrastructures and the types of protesters that take to the streets? What deter-
mines a protest's dominant mobilization pattern? How and why do people
initially participate in protest? What is the role of the internet in persuading
people to participate in protest? We do not claim to provide conclusive answers
to all of these questions; our main contention is that contextualizing protest
can help us make important inroads toward better comprehending it.

One of the main limitations of the mainstream empirical literature of
protest is the lack of comparison. The bulk of collective behavior scholar-
ship, in particular, that of students of social movements, engages in single or
multiple case studies. This limits the researcher's capacity to thoroughly test
existing theories by confronting them with diverse cases and to generalize
findings. To not compare cases would rule out the possibility of systemati-
cally taking into account the very context in which contention takes place.
This leads to a second stream in the literature: political participation stud-
ies, which have been flourishing in the political sciences since Barnes and
Kaase's study on political action (Barnes and Kaase 1979). Essentially com-
parative in nature, these studies draw on general population surveys across
several countries, allowing for more rigorous tests of theoretical propositions

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xv dfc

INTRODUCTION XV

and broader generalization. However, political participation studies tend to
completely decontextualize protest behavior, as they do not contain any evi-
dence of participation in specific events and merely characterize protest par-
ticipation in general terms. While the case studies often lack a comparative
angle, the studies drawing on general population surveys are, in part, com-
parative, though they entirely disregard the context.

A massive wave of similar demonstrations in many countries, the events
of February 15 offer scholars a unique opportunity to go beyond single cases
and adopt a truly comparative perspective, taking context into account. In
this study, we systematically compare antiwar demonstrators and demonstra-
tions in eight Western democracies. On February 15, 2003, the authors of
this book, in a collective effort and using the same template questionnaire,
conducted protest surveys at eleven different locations. The surveys compare
demonstrations and demonstrators in the United States (New York, San Fran-
cisco, and Seattle), United Kingdom (London and Glasgow), Spain (Madrid),
Italy (Rome), the Netherlands (Amsterdam), Germany (Berlin), Switzerland
(Bern), and Belgium (Brussels). 1 These nations represent most of the largest
Western countries in the months leading up to the war and allow for a com-
parison of the mobilization against the different positions that Western gov-
| ernments adopted vis-a-vis the war: from active participation and initiation

"W" (United States and United Kingdom), to limited military collaboration and ~®~

verbal support (Spain, Italy, and the Netherlands), to silent opposition (Swit-
zerland) or raucous critique of the US (Germany and Belgium). The eight
countries are highly diverse in their protest cultures, public opinions, systems
of interest mediation, and in (involved) political institutions.

What makes the February 1 5 protests particularly interesting from an
analytical perspective is that a number of crucial variables can be held con-
stant when comparing the protests and protesters in the countries studied.
In all venues, demonstrators opposed the imminent war on Iraq; and all events
occurred on the same day (and thus in the same stage of the conflict's build
up), relied on the same protest repertoire of peaceful demonstrations, tar-
geted the same war-preparing countries (the United States and the United
Kingdom in particular), and appeared to shout the same slogans and carry
similar banners. While holding constant the issue, time, and action type, with
a focus on the variety in institutional structure — the diverging political
stances of government and opposition, the differences in political cultures,
protest repertoires, and social movement strengths in the eight countries under
study allow for a strong comparative design. The essential research question
that our study raises is: What are the core differences and similarities across the
February 15 protest events in the eight nations, and what factors account for them?

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xvi dfc

"4

INTRODUCTION

The main empirical claim upheld throughout this book is that — although
similar in timing, action repertoire, set-up, slogans, banners, atmosphere —
the February 1 5 demonstrations differed considerably across countries. Under
a surface of apparent similarity a host of differences was concealed The dem-
onstrators displayed substantial variation across countries; people taking to
the streets in Madrid, for example, did not exhibit the same characteristics,
attitudes, and convictions as those protesting in Glasgow, at the same time
and against the same war. Confronted with such findings, the predominant
theoretical claim relevant throughout this volume is that political context does
indeed matter for protest events. Distinct political contexts mold dissimilar
protest events. Protest correlates as the composition, target, amplitude, rad-
icality, and claims of the contention differ in relation to the diverging polit-
ical context.

That political context matters in terms of social movements can hardly
be considered a new assertion. In fact, it is the core claim of one of the most
influential accounts of social movements. More than two decades ago emerged
the political opportunity structure (POS) approach, stressing that social
movements are dependent on the political environment in which they oper-
ate (Kitschelt 1986, Tarrow 1998). The book fully endorses the core claim
of the POS approach, yet, simultaneously acknowledges that the classic ver-
sion of this approach was devised to study not specific protest events but
rather social movements at large. Movements have an organizational struc-
ture and a certain continuity based on an ongoing interaction with power
holders, while protest events, although often staged by social movements, can
be one-shot actions without any organizational backbone (Tilly 1994). The
protest wave against the war on Iraq that reached its apex on February 1 5
exemplifies this fact; it declined quickly after the war broke out, with only a
few protests in most countries just one month later. The dynamic of social
movements may be determined by the ebb and flow of protest events, the
movement itself can survive without collective action and remain alive for
years in a state of abeyance. In a similar vein, a social movement may be
weak — with only a few core activists, light structures, and hardly any visi-
bility — but its protest events may be large. Particularly with regard to issues
of war and peace, a distinction between the peace movement and the protest
events staged to support the peace cause seems crucial. Accounts reveal that
the relationship between peace movement and peace protest is complicated,
at least in Western Europe, as specific peace organizations may not be well
developed, while their demonstrations often receive ample support from the
entire associational spectrum of the Left resulting, occasionally, in huge mobi-
lizations (Rochon 1988).

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xvii dfc

INTRODUCTION

Devised to comparatively account for the strength of social n
their constituency, organizational structure, mobilization level, turnout, mil-
itancy. — the POS does not make claims regarding the features of individual
activists or demonstrators. Yet, this study focuses on the individual demon-
strators and asks who they are, why they participated and how they had been
activated. As we opt to concentrate on protest politics and individual par-
ticipants rather than on social movements and movement strength, we can-
not simplistically assume the classic explanatory factors associated with the
POS approach. We believe, for example, that social movements are only one
of the determinants of protest events. The POS approach states that political
opportunities determine the strength of social movements; we maintain that
a protest event is affected by the broader political context and that move-
ment strength is only one of the intervening variables between the political
context and characteristics of individual demonstrators.

What, therefore, is the theoretical argument buttressing the book? Our
focus is on similarities and differences among demonstrators in eight nations.
This is the dependent variable, that which we strive to fully understand and
explain. We focus on three features of the demonstrators: their sociodemo-
graphics, attitudes, and (political) behavior. The relevant context of protest
| events such as February 1 5, the independent and intervening variables, consist

W of five sets of explanatory factors, which are derived from key social move- v9)~

ments and political participation theories:

1 On an overall level, the sociodemographic composition of the popula-
tion — especially age, class, and education — predetermines the con-
stitution of the protest and accounts for cross-country differences and
similarities. This argument draws on theories about "social central-
ity" as suggested in general participation theory (Verba, Schlozman,
and Brady 1995). Trivial as it may seem, a generally high educational
level in a given country, is mirrored among the protesters.

2 Structural features of the political system play a role, more concretely
the channels of access to the decision-making system, the distribution
of decision-making power, and the prevalent mode of interest medi-
ation. This dimension corresponds to the central claim of the POS
approach, as developed in the social movement literature (Kriesi 2004).
Note that the classic POS approach focuses on permanent and stable
arrangements that structurally mold general protest behavior in a given
country; it is not meant to inherently explain variable traits of spe-
cific protest events or campaigns, let alone to account for the traits of
individual protesters.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xviii dfc

INTRODUCTION

3 More specific issue-related political context variables have an impact
on the protest. These are the concrete and volatile political opportu-
nities, such as sudden changes in political elites' support for a protest
cause. This specific "interaction context" will be operationalized in
three dimensions: (a) the position of government and opposition re-
garding the issue at stake, (b) mass media's coverage of the issue and
the protest, and (c) the supportive or nonsupportive attitude of the
general public (Kriesi 2004). This middle-range context dimension
can be traced back to several theories, among which are those of "rel-
ative deprivation" (Gurr 1970) and the framing literature on social
movements (Gamson 1992, Snow et al. 1986). We expect this con-
text to influence the composition of the demonstrators. In countries
where political elites are not supportive of the protest's cause, media
are ignoring or marginalizing the protest, and public opinion is un-
aware of or manifestly opposed to the protesters' goals, the protesters
are likely to differ from those in countries where elites, media, and
the public opinion endorse the respective protest.

4 On a more concrete level, social movements that organize an event
come into play. Having distinct constituencies, structures, goals, and

/fa strategies, these movement organizations affect the profiles of the pro- /is

I testers. This argument coincides with the central contention of resource

mobilization theory, that social movement organizations and the e:
ing structure of the social movement sector affect a protest's mobiliz-
ing force and the type of people that show up for it (McCarthy and
Zald 1977b). As applied to the February 15 protests, we expect that
the demonstrators will basically mirror the traits of the social move-
ments behind them that staged the events. Different social movement
sectors lead to different people on the streets.
5 Finally, and most closely linked to protest events per se, the actual
strategies and channels of mobilization utilized by organizations and
participants affect the demonstrators' sociodemographic characteris-
tics, attitudes, and behavior. This dimension, serving as the macro-
micro bridge, corresponds with a vast body of literature stressing the
importance of mobilizing agencies, motivations, and networks for both
social movements and, more generally, political participation (Klan-
dermans and Oegema 1987, Rosenstone and Hansen 1993, Diani and
McAdam 2003). In terms of February 15, we expect to find different
mobilization patterns across countries, which ultimately led to differ-
ent types of demonstrators.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM F

H$H

INTRODUCTION

m

Thus, we argue that protest politics can only fully be understood in a
multilayered and multifaceted context (such as the one we have described).
These five context dimensions are not entirely independent from one another.
Political system arrangements, for example, are a prior condition for the
issue-specific political context that, in turn, influences the structures and
strategies of social movements. An additional example: the presence and
strength of certain social movements similar to and supportive of peace
groups, combined with the media's stance and predominant public opin-
ion, has an effect on the type and scope of the mobilization processes. There-
fore, we conceptualize the first four dimensions as independent variables, as
shown in Figure 1.1 below. The one remaining dimension — mobilization
processes — can be considered as a set of intervening variables that are influ-
enced by the independent variables and have a direct effect on our threefold-
dependent variables — the sociodemographic characteristics, attitudes, and
behaviors of the protesters in the eight respective nations. These three blocks
of variables correspond to the three parts of this book.

As the reader may have inferred, our theoretical framework is twofold,
as it draws on two strands of research that have largely remained separate,
although often studying the same phenomena: the political participation
literature (see factors [1] and [5]) and the social movement literature (see
factors [2] to [4]). Methodologically, the individual actor-level evidence we
collected in surveys resembles the emblematic design of the participation
studies. The contextualization of the data in a movement-staged campaign or
event, by contrast, is typical for movement studies. Focusing mostly on move-
ment organizations (mesolevel) or country features (macrolevel), quantitative

Independent \ nruiMo.
Country Level

Intervening Variables
Country or Individual Level

l)qK*iMk'i!t Yari.ihk's

Sociodemographics
Political Attitudes

Sociodemographics of Populac

pohucal Xr2) structure

Organizations/Networks (chapter
10)

Recruitment (chapter 9)

Social Movement Sector
(chapters 2 and 4)

n

|lssue-SpecincContext ( chapter

Y^ "™

Figure 1. 1. Explanatory model and organization of the book

H$H

-4

00 Front_Walgrave 22/01/2010 3:35 PM Page xx dfc

XX INTRODUCTION

individual-level data (microlevel) are rare to find in social r
search. Our theoretical and methodological blending of both approaches is
warranted, as we deal directly not with individual participation or social
movements but with protest events found in the middle of the spectrum.
These are precisely the instances and mechanisms that connect movements
to participants. Protest events are a micro-macro bridge in action.

Obviously our simple argument cannot be considered as a fully inte-
grated and rigid theoretical model. For example, we do not specify the inter-
actions among the different independent variables or among the dependent
variables, nor do we arrange the independent variables in a distinct causal
chain. The model we present is rather a heuristic device; it serves as an orga-
nizing or sensitizing tool that the authors of this book will use selectively.
No chapters systematically integrate all context dimensions and factors of the
model. The reasons for this are twofold. First, as research on specific protest
events and more so research using surveys of protesters is rare, we do not have
much data to build on. Our study is explorative, aimed at detecting new
theoretical insights instead of testing precise hypotheses. Second, it does not
appear useful to design an elaborate and detailed theory, as we are unable to
test the latter systematically. The more numerous the independent and de-
pendent variables and the more precise and nuanced our propositions and
hypotheses regarding their mutual relationships and dynamics, the more cases
are necessary to test these hypotheses. Though our surveys cover thousands
of respondents, we compare on a country level and, as a result, have restricted
ourselves to eight cases only. Even our simple heuristic model with three inde-
pendent, two intervening, and three dependent variables cannot be tested
conclusively with eight cases. That is why, additionally, we organize this book
in chapters, each focusing on at least one of the eight variables or factors
proposed above. We concede that in this stage and with our available data,
it is impossible to fully weigh the different explanatory factors against each
other and to systematically explore all relevantly mutual interactions and
dynamics.

Our theoretical argument is essentially contextual as it is comparative.
Depending, however, on the target at hand and on the selection of the cases,
a comparison can take two forms. Applied to our particular case, if countries
are very different but the phenomena to be explained — protesters' sociode-
mographics, attitudes, and behavior — are similar, it appears to be of great-
est value to search for certain similarities within the dissimilar countries that
account for these similar "outcomes." If cases are generally similar but out-
comes different, the standard strategy is to search for certain case differences
that can be used to explain the differences in outcomes. The present study

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xxi dfc

INTRODUCTION

does not include either of these ideal typical comparative designs. It covers a
broad range of countries with large institutional, political, and cultural diver-
gences, but we did not select the countries based on any theoretical design.
Clearly, though, ours comes closest to a most-similar-systems design, as it
merely focuses on Western, postindustrial democracies — even if it compares
systems such as the United States, Italy, or Switzerland that differ in some
respects. Whether outcomes are similar or dissimilar cannot easily be estab-
lished in our study. Remember that the outcomes, or the aspects of them that
we are particularly interested in, are features of demonstrators, and these
form a continuum. We hardly have an external benchmark that allows us to
conclude that the demonstrators in a given country score high or low on any
certain variable.

In their chapters, the authors have employed different strategies. The
book's main angle is comparative — assessing and explaining dissimilarities
and similarities among the protesters in eight nations — yet not all chapters
deal with an explicit comparison to the same extent. Some authors stress sim-
ilarities, that is: they primarily search for commonalities across the demon-
strators. They use comparative evidence to submit existing theoretical, mostly
noncomparative, propositions to a test (Do they hold across all countries?)
/fa or to derive new theoretical claims (Do we obtain similar findings in all /is

| countries?). Authors following the similarity logic do not explain differences

but focus on similarities stemming from theoretical arguments (see chapte
7 and 1 1). Other authors depart from the differences across countries. Their
aim is to account for the discrepancies in outcome — namely features of the
demonstrators — relying on contextual differences between countries (see chap-
ters 6, 8, 10, and 12). These authors explicitly draw on comparative argu-
ments as those we presented above. Some chapters combine both logics;
they test general theories across nations to assess whether they hold under
all circumstances, and they focus on explaining the remaining differences
(chapters 5 and 9).

The chapters in this book make use of more than the heuristic model
described above. Depending on what needs explanation, they apply — and
in part further develop — major approaches in current social movement theory
by emphasizing aspects such as resource mobilization, structural opportuni-
ties, movement networks and cultures, and framing. Therefore, apart from
contributing to the exploratory model we propose, nearly all the chapters of
this book introduce their own theoretical arguments and test correspond-
ing propositions. However, as far as the dependent variable is concerned, all
chapters rely on the same core data set.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xxii dfc

"4

INTRODUCTION

The main empirical source for this book is a comparative protest survey
addressing participants in the eleven demonstrations in the eight nations.
We will present our methodology in more detail at the end of the book; in
this introduction, we will briefly sketch our approach. Protest data were col-
lected that rely on an innovative protest survey methodology consisting of
directly questioning participants at demonstrations, a rather uncommon re-
search technique. Favre and colleagues even speak of "a strange gap in the
sociology of mobilizations" (Favre, Fillieule, Mayer, 1997). To the best of our
knowledge, protest surveying has only been used in few studies (Jasper and
Poulsen 1995, Waddington 1988). Most elaborate is the work of the French
research team including Favre, Fillieule, and Mayer, who developed a method
designed to offer every participant an equal opportunity of being interviewed
(Fillieule 1998, Fillieule 1997). Stefaan Walgrave and Peter Van Aelst refined
their method further and tested it on seven Belgian demonstrations in the
1998-2001 period (Van Aelst and Walgrave 2001; Norris, Walgrave, and
Van Aelst 2005). The essence of the method is to take a random sample of
demonstrators, distribute postal questionnaires to the demonstrators selected,
ask them to fill out the questionnaires at home and send it back to the re-
searchers (postage paid by addressee). This technique has been also applied
in surveying protesters against unemployment and welfare cuts in four Ger-
man cities in 2004 (Rucht and Yang 2004). In general, protest surveys result
in quite high response rates and tests of response bias for the most part yield
encouraging results. We will report on the precise fieldwork method, sam-
pling strategy, response rates, and response bias tests at the end of the book.

Regarding the February 15 protest survey under study here, war against
Iraq was still an open matter in December 2002. The authors of this book,
an international team of social movement scholars, began preparing for com-
parative data collection and created the International Peace Protest Survey
(IPPS), should protests against a potential war occur. Drawing on an exist-
ing questionnaire and fieldwork method developed by the Belgian team
working with Stefaan Walgrave, they prepared themselves and waited for
the "Big One." The researchers did not expect it to be as big as it was. On
February 1 5, the authors and their students distributed about ten thousand
questionnaires in eleven different protest venues in eight countries, resulting
in 5,838 successful postal interviews, with response rates varying from 40 to
55 percent. We also managed to conduct 913 oral interviews with demon-
strators in Belgium, the Netherlands, Switzerland and the United Kingdom.
In the United Kingdom (Glasgow and London) and the United States (Seat-
tle, San Francisco, and New York), we covered several demonstrations. As
differences among the demonstrators in these different venues within the

H$H

-4

00 Front_Walgrave 22/01/2010 3:35 PM Page xxiii fa

INTRODUCTION XX111

same nation were minimal, we decided, unless explicitly mentioned other-
wise in specific chapters, to consistently draw on a collapsed data set on a
country level. Taken together, perhaps the IPPS is the largest data set avail-
able on actual protesters to date. The IPPS questionnaire ascertains key vari-
ables, such as sociodemographics, political attitudes, political behavior, and
organizational affiliation. The majority of available survey data do not focus
on one specific protest event, since they are measurements of general partic-
ipation. The IPPS, though, contains event-specific questions on attitudes,
mobilization processes, network structures, and organizational background.
Therefore, our data allow us to make significant progress in connecting pro-
test with its context and forerunners.

This book both draws on the core protest survey and utilizes as well as
relates it to a broad range of other data, such as general population surveys,
surveys on social movement activists, media reports, flyers and calls for
action, and speeches held during the event. Together, these put the core sur-
vey data into context and help answer all relevant questions initially raised.

Our study is organized in analytical chapters, since we are interested not
in individual country results but in comparison. All chapters are related to
the model that guides our book. The first chapter of the book introduces the
case to be analyzed: the worldwide February 1 5 protest event. Second, from
general to specific, the first several chapters present three layers of the con-
textual approach that form the theoretical core of the book. These inde-
pendent variables will be used throughout the book, serving as a theoretical
toolbox to help us better understand similarities and differences between
countries. From a POS approach perspective, general regime features regard-
ing elite-challenger relationships predetermine the existence, size, and out-
come of contentious political action. Therefore, the first contextual chapter
outlines the general regime features for the eight nations (chapter 2). At the
middle-range level, these general characteristics are supplemented with a
specific set of explanatory factors related directly to the Iraq War and to the
citizens who joined the protest. We chart the respective governments' posi-
tions on the Iraqi issues, illustrate the position taken by public voices and
the population at large in the eight nations vis-a-vis the war, and analyze the
media's coverage of the Iraqi conflict and of the protest against the war (chap-
ter 3). More concretely, from a resource mobilization perspective, previous
protest and its organizational remnants could potentially be an important
resource for renewed mobilization. This resource approach applies particularly
to the peace movement, a movement exemplifying oscillating mobilization
levels with long-lasting latent phases and sudden outbursts of massive activ-
ity. Hence, one chapter is devoted to previous peace protests in the eight

H$H

-4

00 Front_Walgrave 22/01/2010 3:35 PM Page xxiv (fo

INTRODUCTION

s (chapter 4). Chapters 2 to 4 are influenced by the fact that regimes,
issue specifics, and peace mobilization are mutually associated; the chapters
succinctly explore these mutual links. Not drawing on IPPS data but relying
on other primary and secondary evidence, these are mainly descriptive. But
they also raise questions and generate hypotheses about how and to what
extent "their" context layer might affect one of the dependent variables, be it
sociodemographic characteristics, attitudes, or behavior of the demonstrators.

Chapters 5 through 8 are devoted to the protesters. Who are they? What
are their backgrounds? To what extent did they participate in previous pro-
tests? What do they believe in, and why did they participate in the February
1 5 event? These chapters draw extensively on IPPS data and address the clas-
sic individual-level features that are central to all participation studies: socio-
demographics (chapter 5), political-attitudinal characteristics (chapter 6), and
political-behavioral characteristics of demonstrators (chapter 7). In other
words, these chapters concern the dependent variables: they describe them
and draw on the contextual elements presented in the first four chapters to
account for intercountry differences, testing the usefulness of the contextual
approach of protest events. Against the backdrop of certain theories, each of
these chapters tests the hypotheses revolving around diversity and normal-
ization of protest, the centrality of the peace issue versus government oppo-
sition in the demonstrators' motivational structure, and the determinants of
participation. Chapter 8 follows a slightly different logic and deals with the
possible political spillover of the protest by developing an informed guess
about the demonstrations' potential impact on subsequent elections in the
eight nations under study.

The remainder of the book focuses on mobilization. Which channels
and means were used to inform and motivate people? What was the under-
lying organizational infrastructure of the event? Were many protesters re-

cruited via social movements? In terms of the contextual model structuri

the book, these chapters examine the intervening variables, namely move-

ind mobilization processes. These variables can bi
ered both independent and dependent. Mobilization and movemei
be explained by more general context layers as well as they help expk
demographic characteristics, attitudes, and behavior. These chapters tackle
the mobilization puzzle, progressively pinpointing the traits of the February
15 events. In chapter 9, we begin by asking the general question of whether
people were mobilized via open (e.g., media) or closed (e.g., organizations)
mobilization channels. In chapter 10, we elaborate on the organizational track
by charting the associational microlevel networks that initially generated the
protest. We further explore the looser and informal mobilization track in

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xxv dfc

INTRODUCTION

chapter 1 1, by assessing the role the internet played in bridging diverse issue
networks. Finally in chapter 12 we disregard the structural aspect of the
mobilization puzzle and turn to the cultural aspect by examining the fram-
ing of the Iraq issue by the media, by movement spokespersons or organiz-
ers, and the "ordinary" demonstrators.

Ultimately, this book focuses on an extraordinary series of demonstra-
tions staged on one single day and serving one single purpose: preventing an
imminent war on Iraq. As many of the aspects of the protests were identical
or similar in the countries in our sample — action form, aim, slogans, organiz-
ers, timing — we can usefully compare the demonstrations and detect differ-
ences among the protests and the protesters in the eight countries. Guiding
this inquiry, we draw on a loosely structured set of theoretical propositions —
boiling down to the fact that context matters for the features of individual
demonstrators — that do not form a strict theoretical model but that rather
serve as a heuristic guide for the individual chapters. In summary, we believe
that these context dimensions enable us to make sense of what happened on
February 15, 2003, and, more generally, shed light on how protest events
interact with their environment.

1 "°* 1

~W 1 . As the London and Glasgow samples emerged to be similar — the political con- ~W~

text was identical — we systematically collapsed data on the protesters in London and
Glasgow. Unless specifically mentioned, we consistently utilize a collapsed UK dataset
in this volume. The three U.S. demonstrations, for obvious reasons, have been pooled.

Works Cited

Barnes, Samuel, and Max Kaase. 1979. Political Action: Mass Participation in Five
Western Democracies. London: SAGE Publications.

Diani, Mario, and Doug McAdam, eds. 2003. Social Movements and Networks: Rela-
tional Approaches to Collective Action. Oxford: Oxford University Press.

Favre, Philippe, Olivier Fillieule, and Nonna Mayer. 1997. "La fin d'une etrange
lacune de la sociologie des mobilisations: Letude par sondage des manifestants:
Fondements theoriques et solutions techniques." Revue Francaise de Science Poli-
tique 47: 3-28.

Fillieule, Olivier. 1997. Strategies de la rue. Les manifestations en France. Paris: Presses
de Sciences Po.

. 1998. "Plus ca change, moins fa change." Demonstrations in France dur-
ing the nineteen-eighties." In Acts of Dissent. New Developments in the Study of
Protest, ed. Dieter Rucht, Ruud Koopmans, and Friedhelm Neidhardt, 199-226.
Berlin: Sigma.

H$H

00 Front_Walgrave 22/01/2010 3:35 PM Page xxvi fa

XXVI INTRODUCTION

Gamson, William. 199 ics. Cambridge: Cambridge University Press.

Gurr, Ted Robert. 1970. Why Men Rebel. Princeton, N.J.: Princeton University Press.

Jasper, James M., and Jane Poulsen. 1995. "Recruiting Strangers and Friends: Moral
Shocks and Social Networks in Animal Rights and Anti-Nuclear Protests." Social
Problems 42: 493-512.

Kitschelt, Herbert. 1986. "Political Opportunity Structures and Political Protest: Anti-
nuclear Movements in Four Democracies." British Journal of Political Science
16: 57-85.

Klandermans, Bert, and Dirk Oegema. 1987. "Potentials, Networks, Motivations and
Barriers: Steps Towards Participation in Social Movements." American Sociolog-
ical Review 52: 519-31.

Kriesi, Hanspeter. 2004. "Political Context and Opportunity." In TheBlackwell Com-
panion to Social Movements, ed. David A. Snow, Sarah A. Soule, and Hanspeter
Kriesi, 67-90. Oxford: Blackwell.

McCarthy, John D., and Mayer N. Zald. 1977. "Resource Mobilization and Social
Movements: A Partial Theory." American Journal of Sociology 82: 1212-41.

Norris, Pippa, Stefaan Walgrave, and Peter Van Aelst. 2005. "Who Demonstrates?
Anti-State Rebels or Conventional Participants? Or Everyone?" Comparative
Politics 2: 251-75.
| Rochon, Thomas R. 1988. Mobilizing for Peace: The Antinuclear Movements in West-

~W ern Europe. London: Adamantine. ~W~

' Rosenstone, Steven J., and John Mark Hansen. 1993. Mobilization, Participation '

and Democracy in America. New York: Mai

Rucht, Dieter, and Mundo Yang. 2004. "Wer demonstrierte gegen Hartz IV?" Fors-
le Soziale Bewegungen 17: 21—27.

Snow, Robert D., E. Burke Rochford, Steven K. Warden, and Robert D. Benford.
1986. "Frame Alignment Processes, Micromobilization and Movement Partic-
ipation." American Sociological Review 51: 464-82.

Tarrow, Sidney. 1998. Power in Movement — Social Movements and Contentious Poli-
tics. Cambridge: Cambridge University Press.

Tilly, Charles. 1994. "Social Movements as Historically Specific Clusters of Politi-
cal Performances." Berkeley Journal of Sociology 38: 1—30.

Van Aelst, Peter, and Stefaan Walgrave. 2001. "Who Is That (Wo) man in the Street?
From the Normalisation of Protest to the Normalisation of the Protester."
European Journal

Verba, Sidney, Kay Schlozman, and Henry Brady. 1995. Voice and Equality. Civic
Voluntarism in American Politics. Cambridge, Mass.: Harvard University Press.

Waddington, David. 1988. Flashpoints of Public Disorder. London: Methuen.

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 1 —(g)—

l

February 15, 2003:
The World Says No to War

Joris Verhulst

A Historic Day of Worldwide Antiwar Action

On February 15, 2003, various slogans — "Not in my name!" "No war on
Iraq!" "Don't attack Iraq!" "No blood for oil!" "The world says no to war!" —
nifying mantras that echoed on the streets of more than six hun-

| were the unifying mantras that echoed on the streets of more than six hun-

~W dred cities throughout the world, on the marching cadence of ten to fifteen ~W

million protesters. Diehard activists shared the streets with citizens of all
kinds: students, teenagers, young couples with children, but also housewives,
doctors, university professors and senior citizens (Simonson 2003). February
1 5 was the day the world stood up against an imminent United States-led
invasion of Iraq in a simultaneous flood of protest demonstrations. Taken
together, these were the largest and most momentous transnational antiwar
protests in human history (Epstein 2003, 109), and that on one single day.
Some of the protests were small and only local marches, in which a few neigh-
bors sided with each other; others were national protest demonstrations of
exceptional size and showing unparalleled internal diversity. But in all of
them, the participants showed their aversion of the possibility of war. In the
United States, the February 15 protests were the largest antiwar demonstra-
tions since those against the war in Vietnam; in Europe they largely surpassed
the 1991 Gulf War protests. In many countries, they even outshone the
early 1980s demonstrations against the deployment of NATO cruise and
Pershing II missiles in Europe, which were then considered to have "dwarfed
all previous protest movements in Western Europe in the post-war period" and
were believed to have engendered a "wave of political protest unprecedented"
(Rochon 1988, xvi, 3). Apart from the West, protests were organized in

H$H

-4

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page ? fa

>ss all other continents (e.g., Lebanon, Syria, and Israel; Japan,
Malaysia, Thailand, South Korea; South Africa, Tunisia, and many more),
though in most cases turnouts were not half as spectacular. The largest non-
Western demonstration was probably the one in Syria, where some hundred
thousand people hit the streets; probably the smallest one took place in Antarc-
tica, where a group of scientists held a rally at their observation station. The
only region in the world where peace voices were silent was mainland China.
Altogether, friend and foe, especially in the Western world, were surprised
by the number of protests and protesters and by the diversity of the people
at these demonstrations. Social movement scholars and other observers were
startled by the transnational coordination: at first sight the different protest-
ers were driven by the same ideological beliefs, in a surge of demonstrations
that was alike concerning protest trigger, issue, and target. And the protests'
timing and action repertoire were similar, as well and, with only a single ex-
ception — in Athens and Thessaloniki in Greece — peaceful. A few days after
the demonstrations, many commentators, following the New York Times s
Patrick Tyler (2003), referred to them as the expression of a new "super-
power." Since February 15 there was talk of "two superpowers on the planet:
the United States and World public opinion" (Cortright 2004, xi).

This chapter describes the history, the political context, the setup and
coordination as well as the mobilization levels of the February 15 protest day.
The worldwide coordinated character of the protest makes scrutinizing the
organizational backbone most relevant: it's natural to ask how so many peo-
ple at a time were mobilized in these protests when the international peace
movements appeared to have reached a low since the mobilizations against
the Gulf War in 1991? How did their transnational coordination take place?
The chapter also accounts for the size of the protest by detailing how many
protests took place in how many countries.

War Talk: September 1 1, the Axis of Evil, and the Bombing of Baghdad

The Gulf region has a turbulent history, and the roots of the 2003 Iraq con-
flict can be traced back for many years (see Figure 1.1 for a summary time-
line). The Iran-Iraq war had swept the region between 1973 and 1988. After
a mere two years' breathing space, the Iraqi regime invaded the Emirate of
Kuwait for annexation, claiming that this oil-rich region was a former Iraqi
province. By mid January 1991, the international community, led by the
United States and backed by the United Nations, launched the military oper-
ation "Desert Storm" to set Kuwait free. This military confrontation would
last no longer than forty days. The Iraqi oil export was put under a severe
embargo and restricted by the "oil for food" programs; Iraq also had to allow

H$H

-4

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 3 afc

THE WORLD SAYS NO TO WAR 3

UN inspectors to search for weapons of mass destruction. For the next ten
years, a U.S-led military base kept control over the region and of the Iraqi
no-fly zones in which it would sporadically carry out bombardments.

On November 31, 1998, U.S. president Clinton signed the Iraq Liber-
ation Act. Because of several military maneuvers by the Iraqi army, and because
the Iraqi regime had ceased all cooperation with the International Atomic
Energy Agency and UN weapon inspectors, U.S. Congress wrote the act to
"support a transition to democracy in Iraq" through and after the "replace-
ment of the Saddam Hussein regime" (Congress 1998). The act was made
concrete through "Operation Desert Fox," led by then-president Clinton in
mid-December 1998, which was intended to "decrease the Iraqi capacity to
manufacture massive weapons of mass destruction" and essentially to "over-
throw the Iraqi regime." The plans to get rid of Saddam Hussein were thus
not intrinsically linked to the Bush administration, as would later be regu-
larly assumed, but can be traced to actions years earlier.

On September 11, 2001, several airplanes crashed into the New York
World Trade Center towers and the Pentagon, resulting in the death of thou-
sands of U.S. civilians. Although these attacks were attributed to (and later
claimed by) Osama bin Laden, the American government also connected
them to Saddam Hussein and his Iraqi regime. On October 7, 2001, a U.S.-
led coalition army invaded Afghanistan for the search for Osama Bin Laden
and to bring down the Taliban oppression. This war officially ended by mid-
November 2001. In his State of the Union address of January 29, 2002, U.S.
president George W. Bush used the expression "Axis of Evil." He pointed to
three other countries that were presumed to be sponsoring terrorist develop-
ment and activities and needed to be monitored with the utmost vigilance.
: North Korea, Iran and Iraq. The threat they posed was depicted as immi-
nent and immediate: "Time is not on our side," Bush said. "I will not wait
on events, while dangers gather. I will not stand by as perils draw closer and
closer" ( Bush 2002). The day after the one-year September 1 1 commemora-
tion events, in a dossier titled "A Decade of Lies and Deceit," Bush addressed
a request to the UN Security Council for the authorization of the removal
of Iraq's president Saddam Hussein. Barely two weeks after that, UK prime
minister Tony Blair presented a report — then suspected and later proven to
be exaggerated — on the Iraqi arsenal of chemical and biological weapons and
Saddam Hussein's ability to launch such weapons within forty-five minutes.
By the end of September 2002, the U.S. and British forces had resumed the
first bombing of the Iraqi no-fly zones.

In October 2002, U.S. Congress adopted a resolution authorizing an
attack on Iraq. The war preparations went full-speed ahead. Meanwhile, the

I

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 4 afc

^1 1 a si e

i ^ g -

8 § H S

o S 3S q E s S '

q & 6 X

2 H E

e || | || | 6 ee

loOOOO 6 2 22 Q

I

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 5 (&)

^ fft-

I si;

in

I

■M! Ill I.

5 2 .6 i

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 6 (&)

UN Security Council started debating a new Iraqi resolution. They agreed
on the fact that Iraq had undertaken "obvious," "severe," "flagrant" and "un-
acceptable" violations of the previous Gulf War ending resolution on the
national disarmament of weapons of mass destruction (Wouters and Naert
2003). The UN Security Council demanded the Iraqi government give the
UN weapon inspectors free reign so they could provide an "actual, accurate
and exhaustive" list of all available weapons of mass destruction, and to imme-
diately remove all of these from Iraq. If Iraq did not acquiesce, it would have
to face "serious consequences resulting from its ongoing violations" (ibid.).
At the explicit request of China, Russia, and France, three permanent mem-
bers of the UN Security Council, this phrase "serious consequences" replaced
the provision the United States had proposed earlier, United States in which
the nation proclaimed that it would use military force if Iraq violated any
of the UN demands. Still concerned that this more moderate expression
could clear the way for unilateral American invasion of Iraq, the three coun-
tries were very explicit: the resolution was by no means an authorization to
use violence in any cases of new violations; rather, in such cases, the UN
Security Council would immediately assemble to discuss further measures.
After eleven days of deliberation, UN Resolution 1441 was unanimously
| approved on November 8, 2002. Iraq accepted it within five days, and five

~W days later the first inspectors set foot on Iraqi soil. In the shadow of these ~W

official measures, the first armed skirmishes were already taking place be-
tween the U.S. -UK and Iraqi troops.

On December 7, 2002, in response to the UN resolution deadline, the
Iraqi government delivered a thirteen-thousand-page document on its weapons
arsenal. Late in January 2003, chief UN weapon inspector Hans Blix de-
clared before the Security Council that the Iraqi cooperation could be aug-
mented. On January 30, the leaders of eight European countries issued a
war-supporting statement to newspapers around the world:

The Iraqi regime and its weapons of mass destruction represent a clear
threat to world security. This danger has been explicitly recognised by the
U.N. All of us are bound by Security Council Resolution 1441 , which was
adopted unanimously. ... In doing so, we sent a clear, firm and unequiv-
ocal message that we would rid the world of the danger posed by Saddam
Hussein's weapons of mass destruction. We must remain united in insist-
ing that his regime be disarmed. . . . The combination of weapons of mass
destruction and terrorism is a threat of incalculable consequences. It is one
at which all of us should feel concerned. Resolution 1441 is Saddam Hus-
sein's last chance to disarm using peaceful means.

H$H

-4

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 7 fa

THE WORLD SAYS NO TO WAR 7

The eight countries justified their cooperation and urged others to join
them, referring to shared values proper to all countries in the Western world;
shared fears and threats, based on the September 1 1 terrorist attacks; an historic
debt toward the United States that has liberated the world from commu-
nism and Nazism; the fear for weapons of mass destruction, and the inter-
national justification of an attack on Iraq through the UN Security council.

Six days later, on February 5, 2003, U.S. minister of foreign affairs Colin
Powell presented new alleged evidence to the UN Security Council about
Iraq's disposal of weapons of mass destruction and of the link between Iraq
and al-Qaeda. Five days after this speech, France and Belgium, with Ger-
many's support, ratified their antiwar stance by using their NATO veto against
what they considered the premature protection of Turkey. The alliance
expected that Turkey would become involved in the war if Iraq was attacked
and wanted to start preparing for this. France and Belgium, though believed
a diplomatic solution was still possible for the Iraq crisis and, according to
Belgian minister of foreign affairs Louis Michel, complying with a NATO
decision to prepare for war in Turkey would make them "get stuck in war
logic and the message will be given that it is too late for diplomatic initia-
tives" (Beirlant 2003). France, Belgium, and Germany wanted to at least await
the new weapon inspectors' report to be presented in the UN Security Coun-
cil on Friday, February 14. This day, on the eve of the February 15 protests,
Hans Blix presented a much more mixed evaluation than he had previously,
stating that Iraq had undertaken several positive cooperative steps and that
a total disarmament of weapons of mass destruction would be possible within
a few months.

In spite of the increasing Iraqi cooperation and in the face of the im-
mense popular protest around the globe on February 15, governments from
the .United States, United Kingdom, and Spain handed in a new motion for
resolution in the UN Security Council on February 24, arguing that Iraq
had not seized its final opportunity for disarmament and that military con-
frontation was needed and justified. However, China, France, and Russia
did not support the new resolution, and the latter two countries were even
prepared to veto it. The resolution could not get approval without a two-
third majority, for which none of the five permanent members could have
used its veto. But the war machinery was already in motion. The United
States set up a search for a "moral majority": when nine of the fifteen Secu-
rity Council members supported the proposal, it would be backed by a broad
consensus within the Council, thus morally justifying war. Once again, the
United States and its allies were turned down, leading the United States to
abandon the path of a new UN resolution. On March 17, the United States,

H$H

m

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 8 afc

Spain, and the United Kingdom agreed that Resolution 1441 provided suffi-
cient justification for an armed intervention. On March 20, 2003, supported
by the "Coalition of the Willing" the United States gave the starting shot for
the attack on Iraq. 1

Peace Talk: Organizing against War in Iraq

By the time war was becoming unavoidable, peace activists and organizations
started joining their forces to set up large, worldwide mobilizations. The enor-
mous success of these actions would surprise not only commentators and
politicians but, in no small amount, the activists and organizers themselves:

It was clear by this time [late January 2003] that our movement had steadily
gained momentum. Despite our successes however, of which we were all
extremely proud, not even the most optimistic activists were prepared for
what we saw on February 15. It was a day that we will never forget. In a
worldwide show of unity and solidarity with the Iraqi people, we took to
the streets in the millions, demanding an end to the Bush admin
war plans . . . that this administration is hell-bent for a war. The build-up
in the Gulf during these days of demonstrations has been unceasing. I still
expect that war to come, and soon. Nonetheless, I find myself amazed by
the variegated mass of humanity that turned out yesterday. It felt wonderful.
A mass truly, but each part of it, each individually made sign and human
gesture of it, spoke to its deeply spontaneous nature. (Engelhardt 2003)

The gradual buildup toward an Iraq war was paralleled by growing antiwar
sentiments in all parts of the world and by a gradual organization of and
mobilization for protest against the idea of an upcoming war. Throughout
the entire inception of war, dissident voices were heard. One might ask how
these protests fit in the war race. Bearing in mind the astonishment of politi-
cians, commentators, and organizers about the scale of the protests, the key
question is: Where did these protests come from, and how were they set up?
Following an initial agreement made in a preparatory meeting in Bar-
celona in early October 2002, the idea to set up an international day of
demonstrations against an impending war was first publicly voiced at the
first European Social Forum in Florence, Italy, in November 2002. As this
was a European meeting, the idea originally remained confined to Europe.
In Florence, approximately forty thousand individuals and some six hun-
dred organizations were present: trade unions as well as environmental, global
justice, and peace organizations, among others. The forum was a four-day
event set up for the "democratic debate of ideas, formulation of proposals, free
exchange of experiences, and planning of effective action among entities and

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page9_^_

"4

THE WORLD SAYS NO TO WAR O

:s of civil society that are engaged in building a planetary society
centered on the human being" (Simonson 2003). The Florence European
Social Forum issued a joint antiwar call to "all citizens of Europe" to "start orga-
nizing enormous anti-war demonstrations in every capital on February 15."

Anti-war call

To all citizens of Europe

Together we can stop this war! We, the European social movements
are fighting for social rights and social justice, for democracy and against
all forms of oppression.

We stand for a world of diversity, freedom and mutual respect.

We believe this war, whether it has UN backing or not, will be a catas-
trophe for the people of Iraq — already suffering because of the embargo
and the Saddam Hussein regime — and for people across the Middle East.
It should be opposed by everyone who believes in democratic, political
solutions to international conflicts because it will be a war with the poten-
tial to lead to global disaster.

There is a massive opposition to war in every country of Europe.
Hundreds of thousands have already mobilized for peace.

We call on the movements and citizens of Europe to ;

1 . organising massive opposition to an attack on Iraq starting now

2. if war starts, to protest and organise actions immediately and call
for national demonstrations the next Saturday

3. to start organising enormous anti-war demonstrations in every
capital on February 15.

We can stop this war (European Social Forum 2002a)

The forum not only launched a call for future demonstrations, it also
staged one of the first large antiwar demonstrations. On November 9, 2002,
in the heart of Florence, between five hundred thousand and 1 million people
(according to police estimates) took to the streets to oppose war (Simonson
2003). This was the first large European protest against war on Iraq and a sig-
nificant precursor of what would follow. Chris Nineham from the UK Stop
the War Coalition said of the European Social Forum and its antiwar position:

At the last preparatory meeting in Barcelona, we agreed that the main slo-
gan of the demonstration in Florence would be "Don't Attack Iraq" and that
the meeting would issue a call for cross-continent anti-war action. These
ial decisions. They risked putting the forum on collision

H$H

m

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 10_^_

course with governments and social democrat organizations across Europe.
But they were decisively correct. When word got out that the demonstra-
tion at Florence would focus on stopping the war, the European Social
Forum became a magnet to activists. 1,300 people signed up to come from
Barcelona alone in the three weeks before the forum. People were deeply
relieved that such a mainstream project conceived on such a grand scale
was to take a principle stand on the big issue. It was a stand that had
eluded most politicians, and it showed that the European Social Forum
was going to be something different, something honest, something that
would make a difference. (Nineham 2002)

But the European Social Forum was not the first to set up internationally
coordinated protest against war: between the Barcelona preparatory meeting
and the Florence Social Forum, on October 26, 2002, the first internation-
ally coinciding protests against an eventual war took place. These were the
initial signs of the transnational efforts made by the antiwar campaign. Large
manifestations in the United States with some two hundred thousand peo-
ple hitting the streets were paralleled by more modest protests in Europe:
twenty thousand in Berlin, ten thousand in Amsterdam, thirty thousand in
Madrid. This first protest wave was coordinated by the U.S. -based Interna-
tional ANSWER (Act Now to Stop War and End Racism) Coalition, rooted
in the left-wing Workers World Party (Cortright 2004, 5). ANSWER unites
a broad spectrum of players in civil society, "including traditional peace
groups, students, global justice and anti-racist activists, and mainstream labor,
environmental, civil rights and women's organizations" (Simonson 2003, 7).
Meanwhile, other organizations in the United States had also started discus-
sing the coordination of future events. A new umbrella organization, "United
for Peace and Justice," was formed to take up the coordination role. It was
established before the October 26, 2002, protests and consisted of more than
fifty organizations: traditional peace organizations; new, Internet-based peace
groups (e.g., MoveOn.org); global justice groups (e.g., Global Exchange) and
major constituency organizations (e.g., National Organization for Women)
(Cortright 2004, 14). This new umbrella organization would become the
moderate pillar of the U.S. peace movement, and the catalyst for the Feb-
ruary 15 protests on U.S. soil.

In Europe, one month after the European Social Forum in Florence, an
interim preparatory meeting took place in Copenhagen in December 2002.
Present were delegates from peace movements from all over Europe: Den-
mark, Greece, Macedonia, France, the United Kingdom, Germany, Sweden,
Norway, Belgium, Italy, and Switzerland. These were joined by a delegation

H$H

-4

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 1 1 afc

THE WORLD SAYS NO TO WAR II

from the Philippine peace movement and one from the U.S. network United
for Peace and Justice. In retrospect, it's clear that this is where the first steps
toward the future transatlantic cooperation were made (Brabander 2004).
Here, the original antiwar call was further elaborated, and a platform against
war on Iraq was founded.

Platform against war on Iraq

Statement of the meeting to coordinate European-wide action
against war on Iraq

As agreed at the assembly of the social movements in Florence in
November, activists from 1 1 European countries, the USA and the Phi-
lippines have come together in Copenhagen to coordinate European-wide
action against war on Iraq.

We endorse the anti-war call launched at the assembly in Florence.
We believe that a war on Iraq, with or without UN support, would be a
disaster for the people of the Middle East and beyond.

It is clear there is majority opposition to war in almost every country
in Europe and across the world. That is why this war cannot be fought in
our name. This is also why we believe it is vital to build the broadest pos-
sible anti-war alliances everywhere around the demand No War on Iraq.

Our meeting showed that the movement against the war is gaining

All the countries represented have called action on the 1 5 February.

We reinforce the decision to protest in every country immediately
after war starts, to hold national protests the following Saturday and to
organize coordinated mass national demonstrations in capital cities on
February 15.

To this end we have decided to continue our coordination at a Euro-
pean level, to set up a European-wide anti-war website, and to have a com-
mon banner on each of our demonstrations demanding No War on Iraq.
We are committed to spreading anti-war coordination both inside and
beyond Europe, and to holding another enlarged meeting after the Febru-
ary 1 5 demo. We will continue to campaign until this war is stopped.

We urge the movements in countries not represented at our meeting
to join in our initiatives. We urge every organization that opposes this war
to work for a massive mobilization on February 1 5 . Together we can stop
the war. (European Social Forum 2002b)

Subsequent to the Copenhagen meeting, an intensive e-mail network
was set up, connecting all European peace movements. The Europe-wide

H$H

-4

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 12_^_

ir Web site that the Copenhagen text refers to did not get off the ground;
instead all national umbrella organizations and coalitions set up their own
sites but they linked to one another and to one of the above-mentioned U.S.
organizations listing all worldwide demonstrations, and/or to the UK Stop
the War Coalition. The idea of a common banner was a success: all over
Europe as well as the rest of the world, the same "Stop the War" logo would
be used (albeit in different colors and different styles) on movements' com-
munication outlets, websites, demonstration leaflets, and banners.

On January 1 8, 2003, a second wave of transnational protests took place,
its center again in the United States. These demonstrations were for the sec-
ond time set up by the ANSWER Coalition, and they coincided with the
birthday of Martin Luther King Jr., who had been murdered thirty-five years
earlier. In Washington, D.C, half a million protesters marched; in San Fran-
cisco a hundred and fifty thousand took the streets. Smaller protests were
organized in Belgium, the United Kingdom, Spain, Italy, the Netherlands,
Germany, and many other countries around the world (Simonson 2003).

Between January 23 and 27, 2003, the European Social Forum antiwar
call was further disseminated on the third World Social Forum in Porto Alle-
gre, Brazil. The World Social Forum Secretariat had set up a workshop ex-
clusively devoted to planning the February 1 5 international day of protest.
With some five thousand organizations present from every corner of the
globe, the call was spread throughout the world. The ANSWER Coalition
was present as well. It, along with two other U.S. antiwar coalitions — Win
without War and Not in My Name — would strongly support the February
15 event but would leave the role of main U.S. organizer to United for Peace
and Justice, as the latter group had attended the Copenhagen meeting. This
was not surprising, since organizing the February 1 5 actions did not begin
until mid-January, and turnout still was unpredictable.

February 15, 2003, was the first time in peace movement history that
so many organizations from all corners of the world joined forces on a sin-
gle action day. Earlier attempts by the peace movement to merge in a trans-
atlantic effort, more specifically in the struggle for nuclear disarmament in
the 1980s and 1990s, had failed, "partly because of the external constraints
and opportunities defined by different national political debates and con-
texts, . . . and important differences between the U.S. and European peace
groups" (Cortright and Pagnucco 1997, 159).

The February 1 5 mobilizations benefited from two relatively new and
entwined mechanisms, the dynamics of the social fora and the use of world-
wide electronic communication technologies. The European and World Social
Forums and the different respective preparatory meetings were the main

H$H

-4

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page13_^_

THE WORLD SAYS NO TO WAR 13

driving forces of the transnational coordination and mobilization. World-
wide, national peace organizations, increasingly alarmed by both the 9/11
terrorist attacks and the aggressive reaction on part of the U.S. government,
had since early 2002 all been active on a national level against the ii
of Afghanistan and the idea of war with Iraq. These national organi
now had the opportunity to meet each other at the different social forums
intrinsically linked with the transnational global justice movement. These
forums served as the operating base for the setup of February 15. Various
organizations belonging to the global justice movements started their own
(trans) national mobilizing campaigns and used these different occasions to
update each other on their national efforts as well as strengthen bonds with
their colleagues from other countries.

These important face-to-face meetings were complemented by a second
major mechanism favoring the massiveness of the February 15 protests,
namely the intensive use of the Internet and e-mail circuits. All national peace
movements and coalitions were linked to each other by joint mailing lists
and cross-referencing each other on the Web. On an international scale, the
exact same thing took place, allowing the different movements to act very
fast. In some countries, like the United States and Belgium, the effective mobi-
lization efforts actually got off the ground only by mid-January and reached
full force only after February 5, when Colin Powell presented the alleged
U.S. evidence of Iraq's possession of weapons of mass destruction (Brabander
2004). Through these new channels, established lines of movement interac-
tions, diffusion, were supplemented with new ties, such as brokerage (Tar-
row and McAdam 2003). The mechanisms of diffusion and brokerage made
it possible to agree on one international day of protest, using the same slo-
gans and banners and thus uniting all the people in the different streets into
one global protest. That the transnational character of the protests was clear
before the protests took place might have been appealing for the mobilizing
campaign and might have functioned as a self-fulfilling prophecy, mobiliz-
ing people who wanted to take part in this global day of peace action.

Action: February 15, 2003, the World Moving for Peace

As you watch the TV pictures of the march, ponder this: if there are 500,000
on that march, that is still less than the number of people whose death Sad-
dam has been responsible for. If there are one million, that is still less than
the number of people who died in the wars he started.

The February 15 protests were remarkable for their size. Although many
observers, scholars, and politicians intuitively regarded them as an isolated
event, coming out of the blue, we can now state that this was not the case

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 14_^_

14 VERHULST

at all. In the months preceding February 15, many other events were staged
to challenge the prospect of war, and many efforts were undertaken to pre-
pare this impressive transnational manifestation. Beside their overall magni-
tude, the February 1 5 mobilizations varied across countries. Several cities were
flooded by an unseen mass of protesters, whereas other protests were rather
modest and not exceptional at all. Let us take a look at the mobilization lev-
els in different countries and, in particular, at the turnout in the eight coun-
tries under study here.

Table 1 . 1 shows February 1 5 mobilizations in different countries; listed
are the national organizing organizations and coalitions, turnouts, and the
national mobilization levels as compared to the national populations. The
list is far from complete: some accounts claim that mobilizations took place
in six hundred cities, from the Danish city of Aalborg to the Spanish Zara-
goza. This is only a non-exhaustive list of the largest demonstrations in some
selected countries, to put the protests in the eight countries studied in this
book into perspective. Many smaller and more local marches are not repre-
sented in the table, which could pose a problem in the interpretation of the
U.S. turnout number, since there were activities in almost all U.S. states. 2

Taking a close look at the turnout numbers, one is immediately struck
I by the differences. In Italy, for example, an incredible one in twenty citizens

~W took to the streets; ten times more than in the Netherlands. Here are but a ~w~

few examples of variation in turnouts.

The highest mobilization levels were found in Spain and Italy, where
one in seventeen and one in twenty inhabitants joined the February 15 pro-
tests, setting participation records. In fact, the demonstration in Barcelona
has been chronicled in the Guinness Book of World Records as the largest
antiwar rally in human history (Guinness Book of World Records, 2003).
These countries' governments were the most conspicuously in favor of the
war in continental Europe.

Italy and Spain are followed by Australia and Ireland: in both of these
countries, about one in forty people took to the streets. In Australia, the pro-
test could be considered an event against the official national support for war.
In Dublin, this was not at all the case: since the Irish government did not
endorse war without UN backing, the Irish protest can be seen as an expres-
sion of disapproval of the position of the British government as well as a
statement of support of Ireland's government. Closing the top five ranking
is the United Kingdom, where 1.7 percent of the population was displaying
its disapproval of war. Other massive protests occurred in Greece (1.2 per-
cent), which did not officially support war and even had organized a summit
to reconcile the differences of opinion among the European states. Portugal,

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page15_^_

15 VERHULST

which officially supported the war and Norway, which opposed it, attained
mobilization levels of 1 percent and 1.3 percent.

In many countries, the February 1 5 protests reached unprecedented pro-
portions when compared to previous protests. Yet in others, previous record
levels were not met. In Belgium, for example, seventy thousand took to the
streets, versus the roughly three hundred thousand in 1983 that had pro-
tested the placing of the cruise and Pershing II missiles. In the Netherlands,
the difference was even larger: on February 15, about 0.4 percent of the pop-
ulation took to the streets; the number had been ten times greater in 1983.
In Germany, half a million protesters showed up, where there had been twice
as many, in several protests combined, in 1983 (Rochon 1988, 5-7). The
relatively low turnouts in Belgium and Germany, is not surprising, since the
government opposed an imminent war and, thus, the stakes involved were
lower. In the Netherlands, the official government position was pro-war, but
the government was resigning at the time of the protests. The overall rela-
tively small numbers in the United States are probably mainly due to the
rally-around-the-flag effect. Many troops were already encamped in the Gulf,
which lead many Americans to place support for their own troops above
their disapproval of war. Also, several marches in the United States, like the
I one in New York City, did not have approval of city officials, rendering them

~W less legitimate than others and possibly dangerous. ~W

The February protests will also be remembered for their truly transna-
tional character. In spite of all the differences among them on the national
level, it is beyond doubt that, taken altogether, they were exceptional: never
before had such a large-scale, global, carefully planned and coordinated day
of action taken place. It is this worldwide coordination that truly shows the
uniqueness of the event and distinguishes it from other worldwide simultane-
ous mobilizations around one unanimous theme — for example, the annually
recurring worldwide May 1 demonstrations and the International Women's
Day events (March 8).

To conclude, the February 15 protests were unquestionably unique. They
were exceptional for their size, shared themes and shared timing, and similar
action repertoires. 3 They were, in the eyes of many, the foretelling of a new
superpower. Jiirgen Habermas and Jacques Derrida (2003) declared that the
February 1 5 demonstrations would "go into history books as a signal for the
birth of a European Public." But what about their participants? Were they
the same protesters in different countries? Or did country-specific oppor-
tunity structures, societal contexts and/or historical strengths, and peace-
movement development cause national differences regarding who took
to which streets? In other words: did these protests that shared so many

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 16_^_

m

^

o

5
§
1

jj

12

1

c

—

§

s.

U

2

_4)

C

o

1

J

E

•-E

D

!5

o

E

-o

c

o

o

c

2

B

<u

o

S»

o

.£

°C

£

■E

c

3

o

1

<~>

£.

$

2

Q-

lO

O

>

3

_Q

if!

^

^- -

x

JQ

|2

U

I 3

I I -

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 17_^_

I §1

n

I JlJJ

H$h-

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 18_^_

characteristics mobilize the same people ir
questions of this entire volume.

Notes

1. This "coalition of the willing," whose members were willing to actively or
passively support the forcible removal of the Iraqi regime, included Afghanistan,
Albania, Angola, Australia, Azerbaijan, Bulgaria, Colombia, Costa Rica, the Czech
Republic, Denmark, the Dominican Republic, El Salvador, Eritrea, Estonia, Ethiopia,
Georgia, Honduras, Hungary, Iceland, Italy, Japan, Kuwait, Latvia, Lithuania, the
Republic of Macedonia, the Marshall Islands, Micronesia, Mongolia, the Netherlands,
Nicaragua, Palau, Panama, the Philippines, Poland, Portugal, Romania, Rwanda,
Singapore, Slovakia, the Solomon Islands, South Korea, Spain, Tonga, Turkey, Uganda,
the Ukraine, the United Kingdom, the United States, and Uzbekistan. According to
a White House press release of March 2 1 , 2003, "contributions from Coalition mem-
ber nations range from: direct military participation, logistical and intelligence support,
specialized chemical/biological response teams, over-flight rights, humanitarian and
reconstruction aid, to political support." To further support its case, the White House
also stated: "The population of Coalition countries is approximately 1 .23 billion
people; Coalition countries have a combined GDP of approximately $22 trillion;

| Every major race, religion, ethnicity in the world is represented; The Coalition in-

W eludes nations from every continent on the globe" (White House, 2003). vS?"

2. For a comprehensive list of participating cities, see Chrisafis et al. 2003,
Simonson 2003, Cortright 2004, and many newspaper accounts. In cases where
different numbers are ascribed to the same demonstration, the most recurring, the
most official, or the median number is taken.

3. Since 2003, antiwar protests have been organized worldwide each year around
March 20, the date of the invasion of Iraq. Yet, as of this publication, turnout num-
bers have only been a fraction of those recorded on February 15, 2003.

Works Cited

Beirlant, Bart. 2003. "Brussel en Parijs blokkeren Navo-beslissing in Irak." De Stan-
daard, February 12.

Brabander, Ludo de. 2004. Interview by Joris Verhulst, January, Brussels.

Bush, George. 2002. State of the Union address. January 29, Washington D.C.

Cortright, David. 2004. A Peaceful Superpower. The Movement against War in Iraq.
Goshen, Ind.: Fourth Freedom Forum.

Cortright, David, and Ron Pagnucco. 1997. "Limits to Transnationalism: The 1980s
Freeze Campaign." In Trm bal Politics — Sol-

idarity Beyond the State, ed. Jackie Smith, Charles Chatfield, and Ron Pagnucco,
159-74. Syracuse, N.Y.: Syracuse University Press.

H$H

01 Chapter 1_Walgrave 22/01/2010 3:35 PM Page 19-tffe—

THE WORLD SAYS NO TO WAR 19

Chrisafis, Angelique, David Fickling, John Henley, John Hooper, Giles Tremlett,
Sophie Arie, and Chris McGreal. 2003. "Millions Worldwide Rally for Peace:
Huge Turnout at Six Hundred Marches from Berlin to Baghdad." Guardian,
February 17, 2003. http://www.guardian.co.uk/world/2003/feb/17/politics.

Epstein, Barbara. 2003. "Notes on the Antiwar Movement." Monthly Review 55:
109-16

Engelhardt, Tom. 2003. "The March That Wasn't to Be." TomDispatch, February
16. http://www.tomdispatch.compost/4l0/the_march_that_wasn_t_to_be.

European Social Forum. 2002a. "Anti-War Call." November 12. European Social
Forum, Florence.

. 2002b. "Platform against War on Iraq: Statement of the Meeting to Coor-
dinate European-Wide Action against War on Iraq." December 15. European
Social Forum, Copenhagen.

Guinness Book of World Records. 2004. "Largest Ami- War Rally." http://wwww
.guinnessworldrecords.com/contentpages/record.asp?records id =54365-

Habermas, Jurgen, and Jacques Derrida. 2003. "Nach dem Krieg: Die Wiederge-
burt Europas." Frankfurter Allgemeine Zeitung, May 31, 33-34.

Nineham, Chris. 2002. "The European Social Forum in Florence: Lessons of Suc-
cess." Znet. December 12. www.zmag.org/content/showarticle.cfmPSectionID
| =l&ItemID=2758. 1

~W Rochon, Thomas R. 1988. Mobilising for Peace. The Antinuclear Movements in West- ~W~

em Europe. London: Adamantine.

Simonson, Karin. 2003. "The Anti-War Movement: Waging Peace on the Brink of
War." Paper prepared for the Programme on NGOs and Civil Society, Centre
for Applied Studies in International Negotiation, Geneva.

Tarrow, Sidney, and Doug McAdam. 2003. "Scale Shift in Transnational Conten-
tion." Paper presented at the conference on Transnational Processes and Social
Movements. Bellagio, Italy. July 22-26.

Tyler, Patrick E. 2003. "Threats and Responses: News Analysis. A New Power in the
Streets." New York Times, February 16, 2003. http://www.nytimes.com/2003/
02/17/world/threats-and-responses-news-analysis-a-new-power-in-the-
streets.html?pagewanted= 1 .

U.S. Congress. 1998. Public Law 105-388, An Act to Establish a Program to Sup-
port a Transition to Democracy in Iraq, October 31. http://news.findlaw.com/
hdocs/docs/iraq/libact 103198.pdf.

White House. 2003. "Coalition Members," press release. http://www.whitehouse
.govnews/releases/2003/03/20030327-10.html.

Wouters, Jan, and Frederik Naert. 2003. "De oorlog tegen Irak schendt het Inter-
nationaal recht." De Financieel-Economische Tijd, March 22. https://www.law
.kuleuven.be/iir/nl/onderzoek/opinies/FNJWirakFET.pdf.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 20_^_

2
Political Opportunity Structures and
Progressive Movement Sectors

Michelle Beyeler and Dieter Rucht

m

It is widely argued that social movements are influenced by stable structural
features of the political systems in which they are embedded. This is our start-
ing point. We are interested in these nation-specific structures that, via a set
of intermediary variables, ultimately may have an impact on the size, forms,
and other properties of the antiwar protests that are at the center of this
book. While it may be impossible to find a direct link between general polit-
ical structures and specific protest incidents, we can at best explore a poten-
tial causal bridge between these general political structures and the structures
of a large social movement sector or even specific kinds of movements. These,
in turn, are expected to influence issue-specific protest campaigns or protest
events. In this chapter, we shed light on structural features of political sys-
tems and their possible influence on social movement sectors in the eight
countries under study.

The structural context of specific protests can be conceptualized at two
levels. First, every protest is a manifestation of an extended mobilization struc-
ture that, in the case of antiwar protests, has developed over a long period.
Such a mobilization structure encompasses two layers, that of the directly
involved movement and that of a set of affinity movements. The directly in-
volved movement, commonly referred to as the peace movement, is not at
the center of this chapter (but see chapter 4). Rather, here we will focus on
the broader set of movements (of which the peace movement is only one ele-
ment), which are sympathetic to and potentially supportive of the cause of
peace. Typically, large antiwar protests recruit from a cluster of movements
and, to some extent, even attract people who are not affiliated with any social

H$H

-4

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page21_^_

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 21

movement network but share the core protesters' worldviews and values. Such
a cluster of movements can be referred to as a social movement family (della
Porta and Rucht 1995, 230). Peace movements, notably in the societies we
are investigating, draw on a mobilization potential whose core for the most
part consists of progressive and leftist groups. Accordingly, one element we
are interested in is the size and character of the progressive left in the coun-
tries under study, hereafter labeled the "progressive movement sector."

This movement sector, in turn, is influenced by general political struc-
tures, the second level of the structural context of the antiwar protests. Even
in this era of internationalization and globalization, social movements and
their activities are still profoundly marked by nation-specific factors, such
as a regime type. We will turn first to the description of the nation-specific
political context and then to the corresponding movement sectors in an
attempt to explore the question of how the former might influence the pro-
gressive movement sector.

Political Structures as a Context for Protest Activities

The activities of protest groups and social movements are influenced ir
eral ways by a number of environmental factors, among which, for gooc
sons, so-called political opportunity structures have by far attracted the
attention (Tarrow 1983; 1998; Kitschelt 1986; Kriesi 1995; 2004). The
concept of political opportunity structures is not undisputed. Some schol-
ars criticize it on more principal grounds (Goodwin and Jasper 1999); oth
ers acknowledge its value but emphasize additional dimensions beyond
those included in it (Goldstone 2004). Yet most scholars agree that this con-
cept comprises a set of factors that potentially and actually shape movement
activity. Among these are formal institutional structures, informal procedures
in relation to a given challenge, and the configuration of power as regards a
given challenger (Kriesi 1995), or, according to McAdam's comprehensive
account: " 1 . the relative openness or closure of the institutionalized politi-
cal system, 2. the stability or instability of that wide set of elite alignments that
typically undergird a polity, 3. the presence of absence of elite allies, 4. the
state's capacity and propensity for repression" (1996, 27).

The concept of political opportunity structures was mainly used as a set
of independent variables at the national level to explain, commonly in a cross-
national comparative perspective, general features of social movements, for
example, the movements' strengths or prevailing strategies. 1 Over time, the
concept has been refined and amplified. For instance, dimensions beyond
the political opportunity structure dimensions have been considered: Brand
(1985) has pointed to the importance of deeply rooted societal cleavages,

I

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 22_^_

22 BEYELER AND RUCHT

Rucht (1996) has suggested a more encompassing concept of a societal con-
text structure, and Koopmans and Statham (2000) and Ferree et al. (2002)
have stressed the importance of discursive opportunities as independent fac-
tors shaping movement activity. Several social movement scholars have em-
phasized the role of both movement-specific and situational factors (Gamson
and Meyer 1996; Rucht 1998; see also chapter 3), while still others have
proposed the idea of transnational opportunities that, at least for certain
kinds of movements, come into play (Marks and McAdam 1999). The range
of dependent variables has also been extended, including both strength and
strategy as well as outcomes of social movements (Kitschelt 1986; Rucht
1999; Kolb 2007). Given this extension of the concept, it is not surprising
that this trend of concept stretching has raised critical comments on the part
of some observers, pointing to the "danger of becoming a sponge that soaks
up virtually every aspect of the social movement environment — political
institutions and culture, crisis of various sorts, political alliances, and policy
shifts" (Gamson and Meyer 1996, 275). Whatever the limits and flaws of the
political opportunity structure concept may be — all assumptions derived
from it remain at the macro level of general movement structures and refer
neither to individuals engaged in protest activities nor to specific protest

~W Because we are mainly interested in cross-national differences of move- ~W

ment sectors as a background for specific antiwar protests that predominantly
had a nationwide recruitment base, neither the subnational nor the trans-
national background is relevant. Accordingly, we concentrate on national
structures. Second, we deliberately focus on political opportunities, which,
with regard to the progressive movement sector in general and the peace
movement in particular, are crucial. After all, engaging in armament or war-
fare is primarily a political matter of which governments and/or parliaments
are the decision-making bodies. Accordingly, antiwar movements are gen-
uine political movements, as opposed to other kinds, which are more ori-
ented towards social or cultural issues.

Drawing basically on versions of the classical political opportunity struc-
ture approach, we will discuss two crucial dimensions in the eight countries
under study: formal access to the decision-making system and the configu-
ration of the left-wing power structures, that is, the strength of its parties
and trade unions in particular. We believe that it is useful to take a closer
look at the institutional context in which social movement sectors develop
and particular protests, such as those of February 15, occur. The description
of these contexts will contribute to a better understanding of the similarities
and differences of the protests exposed in the main sections of this volume.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 2 3 dfc

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 23

Formal Access to the Decision-Making System

Social movement scholars have rarely attempted to operationalize political
opportunity structures for more than three or four countries (but see Kolb
2007). Once we aim at comparing a larger set of cases, we have to turn to
highly aggregated indicators that only hint at the relative positions of the
different systems. Kriesi (2004, 71) suggests summarizing the degree of insti-
tutional accessibility of the political system by drawing on Lijphart's (1999)
distinction between majoritarian and consensus democracies. This concept
of comparison estimates the degree to which power is centralized (as in majori-
tarian democracies) or dispersed among different actors or institutions (as in
consensus democracies). Centralization of power has consequences for all
political actors, including protest groups. In political systems where the locus
of power is highly centralized and political parties are very strong, the rul-
ing party or party coalition has much discretion when it comes to making
decisions. In such systems, the role of oppositional political parties, dissent-
ing interest groups, and challenger movements is to raise their voice in oppo-
sition to the government's plans or decisions. However, these oppositional
voices are irrelevant unless they entail a reasonable likelihood of making a

, difference in future elections. Such systems can be classified as "closed." By

Afr- contrast, we count systems as "open" when governments are composed of .fa.

I large party coalitions and/or many parties exist, when power is decentralized

because of a strong federalist structure and when oppositional political parties
and dissenting interest groups rely directly influencing the policy-making
process, for example, via referenda or appeals to courts.

Lijphart also includes corporatism into his concept of consensus democ-
racy. Corporatism, however, has a differential effect on various types of socie-
tal groups. While providing access points for few and preferably hierarchically
structured societal actors (mainly the top representatives of capital and labor),
corporatism excludes a wide range of other actors who have little or no voice
in political decision-making. Due to these inconsistent effects on social move-
ments (of which trade unions can be part), we exclude corporatism from our
estimate of openness and closeness of the political system. Instead, we focus
on four items, that, in summary, will provide us with a rough idea about the
degree of openness or closeness of the decision-making systems in the eight
countries, namely the number of effective parties in parliament, an index of
federalism, an index of judicial review 2 (all indexes taken from Lijphart (1999,
312-13, based on the period from 1972 to 1996), and an index on the use
of nationwide referenda. This latter variable has been inspired by Huber,
Ragin, and Stephens (1997). We used a scale from 1 (never) to 5 (frequent).

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 2 4 dfc

BEYELER AND RUCHT

^

Table 2. 1 depicts the range of the respective countries in accordance with
these criteria. The last column presents a simple additive index of the four
variables. 3

According to the additive index displayed in the far-right column, Great
Britain stands out as the most closed political system or, using another ter-
minology, the country where power is centralized the most. Only two major
parties, each with a high degree of party discipline, are able to win national
elections. Together with the few channels of access to the decision-making
system, this creates a high degree of independence from the general public
and special interest groups on part the of the officeholders. On the other end
of the spectrum, Switzerland, for a variety of reasons, can be considered a
prototype of a very open polity. A weak central government; a fragmented
party system, with only loosely integrated national parties; a non-professional
parliament; and low partisan discipline in parliamentary votes foster a polit-
ical system very responsive to interest groups and social movements.

Within these two extremes, the other countries in our sample take inter-
mediary positions, with the United States and Germany tending more to-
ward the open pole and the Netherlands and Spain toward the closed. The
political systems of the remaining countries tend to be very closed for actors
beyond the political parties or (corporatist) interest groups. In these countries,
the parties clearly dominate the decision-making processes. With regard to
economic and social policies, strong economic interest groups are webbed into
the decision-making process through either corporatist or informal arrange-

What does this mean for the social movement sector? The more sophis-
ticated literature on political opportunity structures has postulated a curvi-
linear relationship between the degree of openness/closeness of the political

Table 2. 1 . Variation in the openness of political systems

I

effective judicial national

Federalism 3 parties" review referenda 6

United States 5 2.4 4 2

United Kingdom 1 2.2 1 1

Spain 3 2.8 3 1

Italy 1.5 5.2 3 2

Netherlands 3 4.7 1 1

Switzerland 5 5.6 1 5

Belgium 3.2 5.5 2 1

Germany 5 2.8 4 1

Note: indexes taken from Lijphart (1999, 312-13) based on the period from 1972 tc
inspired by Huber, Ragin, and Stephens (1997).

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 25_^_

"4

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 25

system and the strength of the movement sector (Eisinger 1973, 28; Meyer
2004). Accordingly we expected the strongest movement sectors in the coun-
tries ranging in the middle ground. As a corollary weaker movement sectors
were expected in both Switzerland (very open), and Great Britain (very
closed),. The comparative work of Kriesi et al. (1995) suggests that move-
ment mobilization was indeed weaker in Switzerland than in France and the
Netherlands from 1975 to 1989. The argument on curvilinearity would lead
us to expect strong protests in Germany, which, according to the findings of
Kriesi et al., was indeed the case in the 1970s and 1980s. It will be interest-
ing to see whether our data will provide further support for the hypothesis
on curvilinearity.

The structure of a decision-making system primarily hints at the degree
movements have direct formal access to the government. Direct access, how-
ever, is only one part of the greater picture. To raise their issues and pass them
through the decision-making system, social movements frequently rely on
allies within the decision-making bodies. Particularly in those polities that
offer few direct channels of influence, alliances with actors that do have for-
mal access to the decision-making bodies are highly relevant. Therefore, the
second aspect of the political opportunity structure that we consider is the
configuration of political actors that matter for challengers, that is, the alli-
ance and conflict structure (see Rucht 2004) in the eight countries. Given
our interest in antiwar protests, we focus specifically on the configuration and
relative strength of the institutionalized left, which traditionally has been an
ally of peace n

Configuration and Relative Strength of the Institutionalized Left

From the literature and various surveys conducted in the past, we know that
peace protests tend to predominantly recruit from left-wing groups. Some
of these, most notably peace groups, provide the natural constituency for
peace protests. Others, in particular left parties and trade unions, may con-
tribute significantly to peace protests because of their sizeable membership.
This reliance on left parties and trade unions was particularly salient in the
peace movements in Belgium and the Netherlands in the 1980s (for the
Netherlands, see Klandermans 1994; for Belgium, see Stouthuysen 1992 and
Walgrave 1994). More generally, however, left parties and unions perhaps
are vital not as recruitment pools but as allies of genuine peace groups in
putting the peace issue on the public agenda and/or in influencing decision-
making processes. After all, the media does not consider peace groups, at
least in their early phases of mobilization, to be main players, whereas it can-
not ignore left parties, especially when represented in parliament, and trade

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 26_^_

BEYELER AND RUCHT

unions. By taking available data into account, we will attempt to present
rough indicators for the relative strength of these two kinds of groups as well
as their framework and position within the overall power structure.

Left Parties

A crude impression of the strength of left-wing parties (Communist, Social-
ist, Social Democratic, and Labor parties) can be obtained by looking at
the degree to which they have been represented in the national government
(Lable 2.2). Since we are interested in structural features rather than in
short-term effects related to specific political constellations, we use averages
of approximately the last fifteen years. According to this indicator, the left
has been strongest in Spain (average 55.3 percent), followed by Belgium (45
percent). On the other end of the spectrum are the Netherlands and Ger-
many with 28.6 and 24.5 percent, respectively. It is only in the polities with
majoritarian voting systems (the United Kingdom, Germany, and Spain)
where these parties were able to govern on their own (see column 3).

Based on the numbers of cabinet seats in the period from 1985 to 2002,
progressive movements in Spain and Belgium stood the best chance of find-
ing influential allies within the party system, whereas the opposite holds
| for Switzerland, Germany and possibly the United States, where the Demo-

~sW cratic Party cannot be classified as "left." But likelihood may not necessarily ~W

turn into manifest support: consider that some moderate left parties, in par-
ticular social democrats, tend to be close to the center or compromise with
a coalition partner from the center or even the right, therefore tending to

Table 2.2. Configuration and strength of left parties (1985-2002)

Average

Maximum

United States c

-

_

United Kingdom

33.3

100.0

Spain

55.3

100.0

Italy

30."

49.9

Netherlands

31.7

50.0

Switzerland

28.6

28.6

Belgium

45.0

58.8

Germany

24.5

100.0

Competing leftist parties
Maximum Communist Party b Green Party c

Source: Our own calculations, based on data compiled in Armingeon et al. (2004).

Notes: a The average of the Left parties' strength in government is measured by averaging the

number of cabinet seats held by leftist parties as a percentage of total cabinet posts, 1985-2002.

^Maximum share or scars acquilvd during this period.

c The Democratic Parry in the United States was not classified as left.

H$H

-4

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 7 7 afc

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 27

develop a conflicting relationship with radical leftist groups. A moderate left
party in government is likely to attract strong criticism from the radical par-
liamentary and/or extra-parliamentary left. Even a (moderate) left-wing gov-
ernment may not necessarily be a blessing for the progressive movement,
and at the same time, a decidedly conservative government may unwillingly
contribute to strengthening forces on the left that to challenge the govern-
ment, tend to ally despite ideological differences. This, for example, was true
for Italy in the government led by Berlusconi. More generally, research based
on protest event data has shown that the overall volume of protest is greater
under right governments than left (Koopmans and Rucht 1995). Given
these differential potential effects of governmental constellations, we cannot
draw strong conclusions regarding the effects on social movements.

According to the political opportunity structure literature, the "division
of the left" is another determining aspect of the configuration of power on
the left, and it has implications for the progressive movement sector. In
those systems where social democrats compete with a strong Communist
Party, Kriesi (1995, 181) expects little action space for new social move-
ments, as the social democrats traditionally had to subordinate their support
of new social movement mobilization to their struggle for hegemony of the
left. In our country sample, this applies to Spain and Italy. Communist par-
ties were particularly influential in Italy, where the Partito Comunista Ital-
iano nearly received 30 percent of the votes as late as in the mid-1980s (see
Table 2.2, column 4). In Germany, (former) Communists only regained
some of their political relevance after reunion. Therefore, we should not
expect a strong effect on the strength of new social movements, which devel-
oped in large part before the political constellation changed. Also the pres-
ence of strong Green parties — that tend to be left-leaning and frequently
serve as mouthpieces of progressive movements — indicates that the social
democrats have to compete with other parties for left-wing votes. Unlike
strong Communist parties, the presence of significant Green parties in the
political system will not reduce the prospects for new social movements. On
the contrary, these parties often originated from these movements and can
be regarded as their strong allies. Green parties attracted sizeable numbers of
votes in Germany, Belgium, Switzerland and the Netherlands, but many fewer
in Italy.

Trade Unions

Apart from some political parties, the trade unions are also significant actors
in the left spectrum. When it comes to the respective infrastructure, we con-
sider union density, calculated as the proportion of union members in the

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 28_^_

BEYELER AND RUCHT

total of paid employees (see Table 2.3). Union membership indicates its orga-
nizational strength and, related, financial and personal resources. More mem-
bers seem to imply more potential power; however, even a large membership
may not be an asset in conflicts between unions and their opponents, as long
as the unions are compliant and willing to compromise. Consider, for exam-
ple, the trade unions in Germany: they are traditionally strong, but, since
World War II, they have been reluctant to engage in bitter fights. Thus, we
will utilize strike statistics as a second measure which, unlike union mem-
bership, does indicate actual engagement in protest.

An important characteristic regarding organized labor concerns the
degree to which trade unions get formal, institutionalized access to the polit-
ical system. While pluralist systems do not grant trade unions such privi-
leged access, corporatist systems do. As the "Corporatism" column in Table
2.3 shows, there are considerable differences regarding the system of inter-
est group mediation in the eight countries. In corporatist systems, the labor
conflict has been pacified and transformed into an institutionalized mode of
conflict resolution.

Union density also varies considerably. It is by far the highest in Bel-
gium, with Great Britain and Italy following suit, and clearly lowest in the
United States. As a matter of fact, in the 1980s and 1990s peace n

j United States. As a matter of fact, in the 1980s and 1990s peace movements

~W in Belgium enjoyed considerable support from trade unions. More generally, ~W

however, high union density does not necessarily translate into readiness for

action. As it can be seen in Table 2.3, it tends to be accompanied by low

strike activity. This is the case in the Netherlands, Switzerland, Germany, and,

Table 2.3. Indicators of the power configurations in the organization
of labor

United States

2.125

United Kingdom

2.000

Spain

2.000

Italy

3.000

Netherlands

4.125

Switzerland

4.000

Belgium

3.750

Germany

4.375

31 7

re: Our own calculations, based on ILO data compiled in Armingeon et al. (2004).
■s: a The indicator for corporatism is representative for the 1990s (Siaroff 1999).

b Trade union density in 1990 (OECD 2004).

The strike level is measured by the average of working days lost per thousand employees
reen 1985 and 2000.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 29_^_

"4

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 29

to some degree, Belgium. In the first three countries, additionally, the pres-
ence of strong independent central banks unwilling to accommodate high
wage demands further contributed to silencing trade unions. Among those
countries with pluralist interest group systems, labor activity is especially
high in Spain and Italy, where governments and monetary authorities have
frequently accommodated demands for higher wages, thereby encouraging
further labor conflicts. Readiness for engagement in labor conflicts tends to
be associated with a more general readiness for political action, so we may
expect unions to support other leftist movement causes, particularly in Italy
and Spain.

In compiling and consolidating the variety of information on the struc-
ture of the political context, we cannot arrive at straightforward conclusions.
Different factors are at work, and we cannot expect these to consistently
influence the progressive movement sector in one direction. Regarding the
strength of this sector, the hypothesis of the curvilinear effect of the politi-
cal opportunity structure, measured as distribution of power, leads us to expect
a relatively weak progressive movement sector in Switzerland and possibly
the United States, and, though for different reasons, in Great Britain as well.
The situation of the institutionalized left suggests a weak support for pro-
gressive movements in Germany and Switzerland, but strong support in Spain
and Italy concerning left parties in national government. Also, the high level
of strike activity in Spain and Italy may indicate the unions' readiness to sup-
port progressive movements. Considering organized labor measured as union
density, the unions appear to be the strongest potential partner for progres-
sive movements in Belgium and the United Kingdom, while the opposite
holds for the United States and Spain. In general, these results leave us with
two relatively consistent assumptions only: First, for Switzerland and, to a
lesser extent, for the United States, we expect unfavorable conditions for the
progressive movement sector, whereas the opposite can be expected for Spain
and Italy. Second, as to the composition of the progressive movement sec-
tor, the configuration of left power structures leads us to expect a dominance
of the Old Left in Spain and, to a smaller degree, Italy. By contrast, we expect
new social movements to be stronger in the Netherlands, Switzerland, Ger-
many, and possibly Belgium, where the labor conflict has largely been paci-
fied and the new social movement sector also enjoys support from strong
Green parties.

In the next section, we look both for indications of the strength of the
progressive movement sector as a whole and as for specific movements or types
of organizations. We deliberately include the left parties and the unions, which
we have so far considered a context variable for genuine social movements

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 30_^_

30 BEYELER AND RUCHT

but can also be regarded as a constitutive part of the expansively defined pro-
gressive movement sector.

The Strength and Shape of Progressive Movement Sectors in Eight Countries

The best measurement for the strength of social movements perhaps is their
effective mobilization as reflected in protest activities. These activities can be
registered in protest event analysis data based on newspaper reports or police
archives. Unfortunately such data are not available for the eight countries
under study 4 Therefore, we have to rely on survey data that give us at least
an approximate estimate on the mobilization potential that could be tapped
by the organizers of the February 1 5 protests. Though far from being ideal
for our purpose, the World Values Survey (WVS) can be used to infer indi-
cations of the strength of the progressive movement sectors in our eight coun-
tries. In population surveys, the number of respondents active in different
social movement organizations is relatively small. To have a reasonable num-
ber of cases at our disposal, we use the integrated data file covering all four
waves of the survey, from 1981 to 2004 (WVS 2006). Based on this data,
we obtain a rough picture of the size of the mobilization potential of pro-
gressive groups in the eight countries.
I We deliberately focus on the broader progressive movement sector be-

~W cause, as indicated, peace (or antiwar) protests typically recruit people far ~W

beyond the small core of committed hardcore peace activists. Mostly depend-
ent on contingent political decisions, such as introducing new weapons or
engaging in warfare, peace activities fluctuate considerably over time, being
almost completely absent in some periods while in others periods attracting
large numbers (Rochon 1988; Cooper 1996). In other words, there tends to
be a large gap between the organizational weakness of the more or less per-
manently existing peace groups and the high potential or actual mobilization
with regard to peace issues. Consider that in the period of the imminent war
against Iraq, in some countries more than four-fifths of the populations were
opposing the war (see chapter 3 of this book), thus providing a huge pool
for recruitment that only could be activated to a low degree for the February
15 protests. Moreover, earlier research has shown that those who actually
participate in peace activities are by no means representative of the overall
population. Rather, to a large extent, they are politically aware citizens who
are or have been active in a variety of groups not necessarily focusing on
matters of war and peace. As Marullo and Meyer rightly stated, peace move-
ments' successful mobilizations "rely on the co-optation of substantial re-
sources from other movements typically not involved in peace politics" (2004,
644; see also chapter 4 of this book) Antiwar campaigns generally can count

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 3 1 afc

"4

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS

on the support of, and recruitment via, a wide range of different groups en-
gaged in women's rights, environmentalism, third-world concerns, and other
issues commonly attributed to new social movements. In addition, trade
unions and leftist parties tend to participate readily in antiwar campaigns.

Left-Wing Orientation and Readiness of Protest Demonstration

As mentioned, the strength and other characteristics of the institutionalized
left in the eight countries studied differ. These variations are also reflected in
the percentage of people attracted by the clearly left political spectrum as
measured by their self-positioning on a left-right scale. 5 The greatest share
of people leaning to the far left can be found in Spain and Italy, the two poli-
ties where Communist parties have political weight and trade unions are
highly militant. Given this constellation, it comes as no surprise that more
of these respondents position themselves on the far left side of the political
spectrum than, for instance, do those in the United States or the United King-
dom, where leftist parties are clearly marginalized in institutional politics. In
particular the United States, where parties seldom mobilize on clear-cut left-
right issues, stands out for its comparatively low level of citizens with a definite

Although we expect the peace movement to recruit demonstrators pri-
marily from left-leaning people, we cannot simply assume that all people on
the left are ready to take part in demonstrations. Therefore, we try to get a
better estimate of the actual potential for political demonstrations by con-
sidering those respondents who did participate (or who at least were ready to
participate) in a legal demonstration. For this purpose, we use the item "at-
tending lawful demonstrations," combining the categories "have done" and
"might do." We assume that this is the measure that comes closest to assess-
ing the mobilization potential for the kind of protest of which the antiwar
demonstration on February 1 5 is just one example. Figure 2. 1 puts the demon-
stration potential in relation to the openness of the political system. The graph
on the left suggests that the openness or closeness of the political system is
related to the mobilization potential of left-oriented people. The readiness
of these people to demonstrate is highest in Germany and Italy, countries
with a semi-open system, and lowest in Great Britain, with its highly closed
polity. In this respect, the picture corresponds to the assumption of a curvi-
linear relationship. Yet the left-wingers in the highly open Swiss system also
have a great propensity to take part in demonstrations, while this is not the case
for those living in the other country with an open system, United States.

Figure 2.1 also shows peoples' readiness to participate in street demon-
strations, regardless of their position on the left-right scale (see the graph on

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 32_^_

BEYELER AND RUCHT

^

s
S

HA. .«

1-

/4* X

/ NKT. BiL ,\

ii

/ N

■ OB \

4 M

sw •,

Openness Openness

Soam?: World Value Surveys, 1981-2004.

Note: Openness of rhe political system is measured by the additive index presented in Table 2.1.
Demonstration potential is measured as a percentage of respondents who have participated or
might participate in a street demonstration. Those on the left are the respondents who placed
themselves on the values 1 to 3 on the left-right scale of 1 (far left) to 10 (far right).

Figure 2. 1: Variation in the openness of political systems

the right side). Switzerland has the lowest participation rate; in this respect,
the data support the curvilinear hypothesis. Hence, the overall picture is mixed:
while our assumption that the protest potential in our eight countries is re-
lated to openness of the political system is supported, other factors seem to
be at work as well.

What accounts for the differences between the two graphs in Figure 2. 1 ?
Overall, remarkable cross-country differences are apparent in the degree to
which protest is primarily an instrument of the left. In Switzerland and the
Netherlands, in particular, there is a clear bias toward the far left with regard
to the acceptance of protest. In the United States, Italy and Germany, though,
many conservatives also participate in demonstrations. 6 These differences
cannot be deduced systematically from variation in the political opportunity
structures. However, the greater reluctance to take part in demonstrations
in Switzerland can be explained in that the Swiss civil society organizations
have more institutionalized channels to put their demands on the political
agenda as well as directly influence political decisions by referenda. Such an
open political system is likely to be accompanied by a low level of political
mobilization outside institutionalized decision-making processes.

Involvement in Progressive Groups

In order to assess the strength of the progressive movement sector, we have
to take into account the degree of citizens' actual involvement in political
groups. Before we turn to group involvement of the left-wing activists, we
present overall involvement rates in different groups (Table 2.4). The upper
half of the table shows the percentage of respondents who declare to be

I

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 3 3 dfc

"4

z

£

O)

c

"5»

c

o

1

o

£

2

c

-s

z

c

8.

£

•*

o

c

1

3

5

A

|

o

1

□

1

o

0)

E

SI

%

o

E

"8

^5

8

13?

II

£

iH

£

O) Q.

S 3

•J= o>

1= c

*3

c

5 -o

|1

-£ 9-

Is

■*

CN

0)

2

£

h$h

-4

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 3 4 afe

34 BEYELER AND RUCHT

involved in trade unions, a political party or in different types of social move-
ment groups, including the peace movement. As we can see, there are quite
remarkable differences, both across groups and across countries. High levels
of involvement in trade unions can be observed in the Netherlands and Ger-
many, while involvement in political parties is particular high in the United
States and in Switzerland. Regarding the movements' thematic fields, groups
working on ecology attract more people than those focusing on women and
peace. Involvement in peace groups is highest in the United States and the
Netherlands, and lowest in Italy and Germany. Considering the larger set of
new social movement organizations, the degree of involvement is particu-
larly high in the Netherlands (followed by Belgium and the United States),
but amazingly low in Italy and Spain.

We argue that the peace groups primarily mobilize among a left-wing
constituency. Table 2.5 presents the percentage of leftists active in different
kinds of groups. Based on this indicator, we may identify different patterns
of mobilization structures in our eight countries. In Spain and Italy, where
the labor conflict has not yet been settled (see Table 2.3), voluntary work in
new social movement organizations is relatively rare, particularly when com-
pared to in the United States and Belgium. However, involvement of Span-
ish and Italian left-wingers is not high in trade unions or parties either.

Table 2.5. Left-wing activists in different groups, including
overlapping membership (percentages)

NSMO a NSMO a
Union Parly NSMO a and Union and Party N

United States

1.9

United Kingdom

2.2

Spain

4.1

Italy

4.9

Netherlands

2.3

Switzerland 15

10.6

Belgium

4.2

Germany

Averages

6.9
3.8

Source: World Value St

irveys 1981-2004.

Nota:Thc data include percentages of active group members, that is, those who "do unpaid

v x : . ^ :

on points 1 to 3 of the ten-point left-right-scale (see also notes to Figure 2.1 and Table 2.4).

b The data on Switzerland are not directly comparable to the other data because of difference
in the questions asked. Thus, we used "inactive/active membership" as a proxy to "belong to a
group / do unpaid work." We calculated averages without the Swiss data.

H$H

-4

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 35_^_

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 35

Thus far, we have not considered multiple memberships or overlapping
activity in different groups. We generally assume that mobilization structures
are especially strong and effective in countries with a high level of perme-
ability of, or overlap between, different social movement organizations. Sig-
nificant overlaps are an indicator for opportunities to engage in joint action.
While such links are extremely helpful for fostering mobilization, they do not
have to be strong, as Granovetter (1973) showed on more general grounds.
As argued above, this is particularly true for antiwar protests, which, due to
the small membership of peace groups, heavily rely on recruitment from other
political groups or social movements. We expect that antiwar mobilization
is achieved more effectively in countries with a high degree of unity among
the different movements of the left. As a rough measure for the extent of
unity or fragmentation of the progressive movement sector, we look at the
degree to which leftist people who are actively involved in new social move-
ment organizations are also active in either trade unions or political parties.
To correctly interpret this data, we have to take into account that the
lap is partially driven by the size of the activity rates in the different groups.
Furthermore, we should also take into account that these estimates ari
culated on a relatively low number of respondents. The link between new
social movement organizations and trade unions is by far strongest in Bel-
gium, followed by Spain, and the United Kingdom. The British case is par-
ticularly interesting because of the relatively low overall number of trade union
activists. In Germany, however, where unionization is relatively high, the over-
lap is small. Also in the Netherlands and Italy, the two sectors appear rather
disconnected; in the Netherlands, however, the parties and new social move-
ment organizations tend to be significantly linked. Even stronger links be-
tween them exist in the United States and Belgium.

Conclusion

Our aim here is to describe the nation-specific political opportunity struc-
tures and the corresponding progressive movement sectors in the eight coun-
tries under study. We relied on quantitative data; drawing on it, we explored
relationships between the opportunity structure and the movement sectors
relevant for protest mobilization. Using the premise of the political oppor-
tunity structure approach, we hypothesized a curvilinear relationship, that
is, that both relatively closed and relatively open political systems offer less
favorable conditions for the progressive movement sector than do semiopen
systems. Regarding especially left-wing parties and trade unions — as a poten-
tial alliance partner and mobilization pool for movement activities — we ex-
pected a weak established left to be unfavorable to the progressive movements.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 3 6 afe

36 BEYELER AND RUCHT

Furthermore, we expected to find a stronger position of new social move-
ments within the progressive movement sector in countries with pacified labor
movements and relevant Green parties.

In the first step of our analysis, we have selected and operationalized two
basic dimensions of the general political environment: the openness of the
political structures and the strength of the established left. Regarding the
first, we found that the United Kingdom is the least open system and Switzer-
land and the United States are the most open. According to the hypothesis
of curvilinearity, our data would suggest the least favorable structural con-
ditions for progressive movement sectors in these three countries. Regarding
the second dimension, we found strong left parties in Spain and Belgium
and a strong Communist Party in Italy. Hence we would expect strong pro-
gressive movement sectors in these countries. Yet significant Green parties in
Germany, Belgium, Switzerland, and the Netherlands also tend to be allies
for progressive movements. As far as organized labor is concerned, we found
union density the highest in Belgium and the lowest in the United States and
Spain. Yet perhaps more important, by far the highest levels of strike activ-
ity exist in Spain and Italy. Moreover, we found a large proportion of leftists
in Spain and Italy. Summarizing these findings, our structural data would
| suggest progressive movement sectors weak in the United States and Britain,

~W and presumably strong in Italy and Spain. ~W

In the second step of our analysis, we measured the strength and char-
acteristics of the progressive movement sector by looking at participation in
demonstrations, membership in groups, and activity in groups based on WVS
data. Considering the various measurements, we could not establish a clear
and consistent ranking order in terms of the strength of the progressive move-
ment sector in the eight countries. When it comes to the strength of new
social movement organizations in particular, the Netherlands, Belgium, and
the United States rank high, while the Old Left seems to be strong in Spain
and Italy, countries where the class conflict is still viable.

What do these findings mean in light of our hypotheses? Unfortunately,
we could not arrive at entirely conclusive results. On the one hand, at least
in our sample, we were unable to establish a clear link between the degree of
openness of the political systems and the overall strength of the progressive
movement sector. Thus, we could not confirm our hypothesis; however, we
found the degree of openness to be indeed related to the strength of the
demonstration potential (not to be equated with actual participation in
demonstrations). Very open and very closed political systems contribute to
lower overall potential participation rates in demonstrations. If we do, how-
ever, only look at the demonstration potential among the left-wingers, the

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 3 7 afc

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 37

picture is less clear again. Only regarding the composition of the progressive
movement sector can we confirm the expectations to a certain extent.

Why were we unable to consistently explain cross-national variations in
the strength of progressive movement sectors by political structures? One
reason may be that political structures consist of many dimensions whose
ranges and relative weights, let alone interaction effects, are unclear regard-
ing impacts of these political structures on social movements. Furthermore,
the progressive movement sectors are certainly influenced by additional fac-
tors as well, for example, political cultures, that is, factors that we could not
grasp in our analysis. In addition to relatively inert political context struc-
tures, more flexible and volatile structures may come into play in a given
country, whether, for example, a right or a left government is in charge, the
government is open to reforms, or countermovements are present.

Although we found variation regarding the independent variables, in par-
ticular the degree of openness among the eight political systems we com-
pared, this variation likely is too small to generate significant and robust
results. The countries studied here are liberal democracies with similar basic
structures. It was not a deliberate choice to rely on a "most similar systems
design" (see Przeworski and Teune 1970) in this book, but this is what we
/fa essentially have, as long as we disregard differences in more specific institutions /is

| such as electoral systems. We would probably identify clearer relationships

if we were to investigate a set of countries with greater variation of political
systems, for example, if we had included authoritarian regimes. In this case,
it may well be that both very closed (authoritarian, repressive systems) and
very open / responsive systems (such as those in Scandinavian countries)
would more clearly support the validity of the argument on curvilinearity.
Regardless of such a possibility, we have no reason to assume a direct
link between the stable national political opportunity structure and specific
events, such as the February 15 protests. While we maintain that such gen-
eral structures account for ample and general properties of movement sec-
tors, it may be more rewarding to search for domain-specific opportunities
that influence corresponding social movements. Why, for example, should
contextual factors that influence the strength of the gay movement or the
farmers' movement matter for the strength of the peace movement? In addi-
tion, when looking at particular policy domains, we can assume that the cor-
responding movements are shaped by more contingent factors, such as the
specific governmental position on the matter of conflict or precipitating
incidents, for example a nuclear accident that may spur anti-nuclear protest.
We agree with the assumption that "shifts in political opportunity structures

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 38_^_

38 BEYELER AND RUCHT

... are too broad to tell us much about the development or success of spe-
cific movements" (Goldstone 2004, 356; see also Rucht 1998).

Nevertheless, certain basic political context factors may well have some
impact on specific protest campaigns, though these are plausibly strongly
mediated by additional factors. For example, the general ideological line of
the national government may influence the composition of the protesters.
Accordingly, a conservative government favoring the war in Iraq is likely to
foster a broad and inclusive leftist antiwar coalition, while a leftist govern-
ment taking the same position is likely to be confronted with a less broad
alliance of protesters, because some groups critical toward the war are sup-
portive of the government at large. Such considerations will be the focus of
the remaining chapters in this volume, which specifically concern the anti-
war protests and their participants.

Notes

1 . Note that in the context of protest studies, the first empirical attempt to
apply the concept of political opportunities did not refer to nation-states but to the
"structure of political opportunities of a community" to explain the intensity of protest
in forty-three cities in the United States (Eisinger 1973, 11).
I 2. This index takes into account two aspects of judicial review: first, whether

~tyy the courts (or a specific constitutional court) have the power to invalidate parliamen- v®r

tary laws, and second, the degrees of activism in the assertion of this power by the
courts. (These degrees include: no judicial review, weak judicial review [formal right
of courts, but judges use it with caution and moderation], medium-strength judi-
cial review, and strong judicial review (Lijphart 1999, 225-28).

3. The maximum values of the four variables are not identical; therefore not
every variable carries exactly the same weight. We ignore this fact because the differ-

4. Earlier research covering four West European countries has shown signifi-
cant variation in the volume and composition of progressive protest (see Kriesi et al.
1995). This research, for example, suggested that unconventional protest activity of
labor and other left groups was high in France and the Netherlands but low in Ger-
many and Switzerland. In Germany, nearly three quarters of them could be attrib-
uted to the new social movements, whereas the corresponding proportion in France
was only 36 percent (ibid., 20). Also relative to the size of the population, partici-
pation in new social movement protests was much higher in Germany than in France.
However, participation in labor protests, again controlled for the size of the popula-
tion, was much higher in France than in Germany, the Netherlands, and Switzerland
(ibid., 22). It remains to be seen whether this picture, based on newspaper coverage
from 1975 to 1989, also holds for more recent periods.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 3 9 dfe

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 39

5. We classify as clearly left-wing those respondents who position themselves
on the values 1 to 3 on the ten-point scale from 1 ("far left") to 10 ("far right.")
Based on estimations from the World Value Survey 1981-2004, these are the per-
centages of left-wingers in our eight countries: Spain 29.3 (N = 10070), Italy 26.3
(N = 5366), the Netherlands 19.7 (N = 3241), Germany 18.2 (N = 8804), Belgium
14.3 percent (N = 5849), Switzerland 15.1 (N = 2612), Great Britain 13.3 (N =
4744), and the United States 10.6 (N = 6906).

6. According to the World Value Survey 1981-2004 data, 72.5 percent of those
on the right in Germany participated or might participate in a demonstration. In
the United States, that figure in is 61.7 percent, and in Italy it is and 62.9 percent.
However, figures are low in Switzerland and the Netherlands (33.7 and 38 percent).

Works Cited

Armingeon, Klaus, Philipp Leimgruber, Michelle Beyeler, and Sarah Menegale. 2004.

Comparative Political Data Set 1960-2001. Berne: Institute of Political Science,

University of Berne.
Brand, Karl- Werner. 1985. "Vergleichendes Resumee." In Neue soziale Bewegungen

in Westeuropa und den USA, ed. Karl- Werner Brand, 306-34. Frankfurt a.M.:

I Cooper, Alice Holmes. 1996. Paradoxes of Peace: German Peace Movements since 1945. I

~W Ann Arbor: University of Michigan Press. ~W~

' della Porta, Donatella, and Dieter Rucht. 1995. "Left-Libertarian Movements in Con- '

text: A Comparison of Italy and West Germany, 1965-1990." In The Politics of

Social Protest: Comparative I 1 1 Movements, ed. Bert Klan-

dermans and Craig Jenkins, 229—72. Minneapolis: University of Minnesota Press.
Eisinger, Peter K 1973. "The Conditions of Protest Behavior in American Cities."

American Political Science Review 67: 1 1-28.
Ferree, Myra Marx, William A. Gamson, Jiirgen Gerhards, and Dieter Rucht. 2002.

Shaping Abortion Discourse: Democracy and the Public Sphere in Germany and

the United States. Cambridge: Cambridge University Press.
Gamson, William A., and David Meyer. 1996. "Framing Political Opportunity." In

Opportunities, hi

Contemporary Movement Theory, ed. Doug McAdam, John D. McCarthy, and

Mayer N. Zald, 275-90. Cambridge: Cambridge University Press.
Giugni, Marco. 2004. Social Protest and Policy Change: Ecology, Antinuclear, and Peace

Movements in Comparative Perspective. Lanham, Md.: Rowman & Littlefield.
Goldstone, Jack A. 2004. "More Social Movements or Fewer? Beyond Political

Opportunity Structures to Relational Fields." Theory and Society 33: 333-365.
Goodwin, Jeff, and James M. Jasper, 1999. "Caught in a Winding, Snarling Vine:

The Structural Bias of Political Process Theory." Sociological Forum 14: 27-54.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 40_^_

40 BEYELER AND RUCHT

Granovetter, Mark. 1973: "The Strength of Weak Ties. nil of Sociology

18: 1360-80.

Huber, Evelyne, Charles Ragin, and John D. Stephens. 1997. "Comparative Welfare
States Data Set." Northwestern University and University of North Carolina.

Kitschelt, Herbert. 1986. "Political Opportunity Structures and Political Protest: Anti-
Nuclear Movements in Four Democracies." British Journal of Political Science
16: 57-85.

Klandermans, Bert. 1994. "Transient Identities? Membership Patterns in the Dutch
Peace Movement." In New Social Movements: From Ideology to Identity, ed.
Enrique Larana, Hank Johnston, and Joseph R. Gusfield, 168-84. Philadel-
mple University Press.

Kolb, Felix. 2007. Protest and Opportunities: The Political Outcomes of Social Move-
ments. New York: Campus.

Koopmans, Ruud, and Dieter Rucht. 1995. "Movement Mobilization under Right
and Left Governments: A Look at Four West European Countries." Discussion
Paper FS III, 95-106. Wissenschaftszentrum Berlin fur Sozialforschung.

Koopmans, Ruud, and Paul Statham. 2000. "Migration and Ethnic Relations as a
Field of Political Contention: An Opportunity Structure Approach." In Chal-
lenging Immigration and Ethnic Relations Politics: Comparative European Perspec-
| fives, ed. Ruud Koopmans and Paul Statham, 13—56. Oxford: Oxford University

m Press. &

Kriesi, Hanspeter. 1995. "The Political Opportunity Structure of New Social Move-
ments: Its Impact on Their Mobilization." In The Politics of Social Protest: Com-
parative Perspectives on States and Social Movements, ed. J. Craig Jenkins and
Bert Klandermans, 167-98. Minneapolis: University of Minnesota Press.

. 2004. Political Context and Opportunity. In The Blackwell Companion to

Social Movements, ed. David Snow, Sarah A. Soule, and Hanspeter Kriesi, 67-90.
Oxford: Blackwell.

Kriesi, Hanspeter, Ruud Koopmans, Jan Willem Duyvendak, and Marco G. Giugni.
1995. New Social Movements in Western Europe: A Comparative Analysis. Min-
neapolis: University of Minnesota Press.

Lijphart, Arend. 1999. Patterns of Democracy: Government Forms and Performance in
Thirty-six Countries. New Haven: Yale University Press.

Marks, Gary, and Doug McAdam. 1999. "On the Relationship of Political Oppor-
tunities to the Form of Collective Action: The Case of the European Union."
In Social Movements in a G ed. Donatella della Porta, Hanspeter

Kriesi, and Dieter Rucht, 97-111. London: Macmillan.

Marullo, Sam, and David S. Meyer. 2004. "Antiwar and Peace Movements." In The
Blackwell Companion to Social Movements, ed. David A. Snow, Sarah A. Soule,
and Hanspeter Kriesi, 641-65. Oxford: Blackwell.

H$H

02 Chapter 2_Walgrave 22/01/2010 3:36 PM Page 4 1 afc

POLITICAL OPPORTUNITY & PROGRESSIVE MOVEMENT SECTORS 41

McAdam, Doug. 1996. "Conceptual Origins, Current Problems, Future Directions."
In Opportunities, Mobilizing Structures, and Framing: Comparative Applications
of Contemporary Movement Theory, ed. Doug McAdam, John McCarthy, and
Mayer N. Zald, 23-40. Cambridge: Cambridge University Press.

Meyer, David S. 2004. "Conceptualizing Political Opportunity." Social Forces 82:
1457-92.

OECD. 2004. OECD Economic Outlook. Paris: OECD.

Przeworski, Adam, and Henry Teune. 1970. The Logic of "Compar,
New York: Wiley.

Rochon, Thomas R. 1988. Mobili merits in West-

ern Europe. Princeton, N.J.: Princeton University Press.

Rucht, Dieter. 1996. "The Impact of National Contexts on Social Movement Struc-
tures: A Cross-Movement and Cross-National Comparison." In Opportunities,
Mobilizing Structures, and Framing: Comparative Applications of Contemporary
Movement Theory, ed. Doug McAdam, John McCarthy, and Mayer N. Zald,
185-204. Cambridge: Cambridge University Press.

. 1998. "Komplexe Phanomene — komplexe Erklarungen. Die politische

Gelegenheitsstrukturen neuer sozialer Bewegungen." In Paradigmen der Bewe-
gungsforschung, ed. Kai-Uwe Hellmann and Ruud Koopmans, 109-27. Opladen,
| Germany: Westdeutscher Verlag.

W . 1999. "The Impact of Environmental Movements in Western Societies." v9)~

In How Social Movements Matter, eds. Marco Guigni, Doug McAdam, and
Charles Tilly, 331—61. Minneapolis: University of Minnesota Press.

. 2004. "Movement Allies, Adversaries, and Third Parties." In The Blackwell

Companion to Social Movements, ed. David A. Snow, Sarah A. Soule, and Hans-
peter Kriesi, 197-216. Oxford: Blackwell.

Siaroff, Alan. 1999. "Corporatism in Twenty-four Industrial Democracies: Meaning
and Measurement." European Journal of 'Political Research 36: 175-205.

Tarrow, Sidney. 1983. Struggling to Reform: Social Movements and Policy Change dur-
ing Cycles of Protest. Western Societies Program, Occasional Paper No. 15, Cen-
ter for International Studies, Cornell University, Ithaca. N.Y.

Tarrow, Sidney. 1998. Power in Movement: Social Movements and Contentious Poli-
tics. New York: Cambridge University Press.

Stouthuysen, Patrick. 1992. In de ban van de Bom. De politisering van het Belgisch
veiligheidsbeleid, 1945-1985. Brussels: VUB Press.

Walgrave, Stefaan. 1994. "Nieuwe sociale bewegingen." In Vlaanderen. Een sociolo-
gische verkenning. Leuven, Belgium: SOI.

WVS, European Values Study Group and World Values Survey Association. 2006.
European and World Values Surveys Four-Wave Integrated Data File, 1981-2004.
v. 20060423.

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 42_^_

3
Politics, Public Opinion, and the Media:
The Issues and Context behind the Demonstrations

Joris Verhulst and Stefaan Walgrave

m

February 1 5 was organized by a closely collaborating transnational network
of social movements. Demonstrations in all eight countries studied in this
volume shared the same action repertoires, frames, and goals (see chapter 1).
Yet, each country's protest was organized by specific national movements
against the backdrop of specific national opportunities. It goes without say-
ing that mobilizing against war in the United States, for example, was dif-
ferent than mobilizing in Germany. The protests were rooted in, or at least
affected by, different national political and societal contexts. The UK gov-
ernment supported the war and sent troops to help the Americans get rid of
Saddam Hussein, and the Belgian government strongly opposed the war —
the position of the government in each country must have had consequences
for its protest movement. Since the political and societal context in each of
these nations was substantially different, we expect the demonstrators in each
to be different too and to bear the traces of their respective milieus. In chap-
ter 2 we analyzed the general, non-issue-specific structural similarities and
differences among the eight countries in terms of access for challengers and
strength of the progressive social movement sector. The approaches to social
movements among these core elements of the political opportunity struc-
ture remained unrelated to the Iraq conflict. Since we are studying a single
protest event and not a social movement, and since we are interested here in
the individual features of the demonstrators and not, the levels of mobiliza-
tion in the different countries, we need to complement the classic opportu-
nity structure elements with more specific contextual factors. We accept that
protesters' engagement is determined not only by large overall structures but

H$H

-4

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 4 3 afc

POLITICS, PUBLIC OPINION, AND THE MEDIA 4J

also by specific political and societal contexts. As relevant context factors, we
take into account politics, media, and public opinion. As Rucht and Neid-
hardt (1991) state, political elites, mass media, and public opinion are among
the most important reference groups for social movements and protest. 1
Mutually influencing each other, political elites form the power center most
movements are trying to influence; the media can marginalize movements
or they can be an important ally affecting public opinion, and public opin-
ion support can boost a movements mobilization and subsequent political
impact.

By focusing in this chapter on politics, media, and public opinion, we
underscore our claim that protests such as the worldwide February 15
demonstrations cannot be fully understood within the general context of a
certain society with its inclination to nurture or discourage protest in gen-
eral. The protested issue itself matters, as do the stance of government and
opposition on the issue; the way the media handles it; and the resonance of
political positions and media coverage in the public. In other words, apart
from the long-term, general political opportunity structure, the specificity
of the February 1 5 events calls for a more specific political context.

Why do politics, media, and public opinion matter? Protest can, on the
one hand, be marginal, rowing against mainstream opinion and behavior in
society; on the other hand, it can also sail on dominant opinion and prac-
tice in a given society. In the first instance, protesters are a minority fighting
a conflictual issue with a clear domestic target; in the second, protesters are
representatives of a majority struggling for a valence issue mostly without
domestic target, since (almost) everybody seems to agree. We expect this
diverging context, apart from affecting the size of the mobilization, to dra-
matically affect the kind of people showing up to vent their discontent. In a
nutshell, our general argument runs as follows: if protesters stand up against
dominant opinion and practice in a given society, they will differ from the
population at large in terms of sociodemographic profile (higher education),
political attitudes (more political interest, stronger ideological stance), and
political behavior (more protest participation, more associational member-
ship). Protesting groups that go against the mainstream are often strong";.
The opposite applies to valence issue mobilization, in which we expect "weak"
groups also to be represented and, thus, a more representative sample of the
population will take the streets.

Government and Opposition on the War

Obviously, the official positions regarding the war differed dramatically among
the eight countries, which include the most war-favoring countries, like the

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 4 4 afc

VERHULST AND WALGRAVE

^

United States and the United Kingdom, and some of those most fiercely
against the war, Germany and Belgium. Yet, not only governments official
stance matters. The opposition counts too: it may support government or fight
it. In some of the studied countries, moreover, government was internally
divided; in others, the opposition parties were internally split. In short, the
alignment of government and opposition regarding the war is an important
context variable. For example, if the Left opposes war, against a right-wing gov-
ernment that backs it, we expect mobilization against the war to take the form
of antigovernment protest, predominantly populated by left-leaning persons
ind groups. Let us sketch in some more detail the government-opposition
ountries. Figure 3.1 summarizes our argument

-initiating country was, of course, the United

n terror" in the post-9/1 1 era, the U.S. govern-

lvade Iraq with the threefold objective of diminishing

igage in terrorist acts or acts of war in the region and

infiguration in the eight
and places the countries oi
The most eminent ws
States: framed by the "
ment was eager
the Iraqi threat

dispossessing the country of all resources to do so; bringing about a regime
change, leading to better life conditions for the Iraqi people; and effectuating
the first step in the democratization of the Middle East. The U.S. govern-
ment — led by Republican president George W. Bush, backed by a neo-
conservative administration consisting of Vice President Dick Cheney,
Defense Minister Donald Rumsfeld, and his deputy Paul Wolfowitz — was
supported by almost all Republican congress members. The Democratic oppo-
sition, conversely, was internally divided on the issue. On October 10, 2002,
Congress approved a resolution authorizing the American president to "use

I

Government Government parties

War-initiating countries
United States
United Kingdom

right/conservative (pro)
center left (divided)

Center Left
Conservative ai

War-supporting

Spain

Italy

Netherlands

right/conservative (pro)
right/conservative (pro)
right/conservative (pro)

War-opposing countries

Switzerland contra center left (contra)

Belgium contra center left/liberal (contra)

Germany contra center left (contra)

Figure 3.1. Position of government and opposition parties

Center and Far Left
Center and Far Left
Center and Far Left

Left (Greens)

Right/Conservative

Right/Conservative

regarding Iraq War

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 45_^_

POLITICS, PUBLIC OPINION, AND THE MEDIA 45

the armed forces of the United States as he determines to be necessary and
appropriate . . . against the continuing threat posed by Iraq." The resolution
was backed by 296 members of the House and opposed by 133. Of the
Democrats, 126 voted against it, while 8 1 of them supported it, whereas only
6 of 212 Republicans voted against the bill. In the Senate, the pro-contra
ratio was even more in favor of war: only one of the forty-nine Republican
senators voted against, and twenty-one of the fifty Democratic senators sup-
ported the war resolution, among them future Democratic presidential can-
didate John Kerry. Although war support seemed overwhelming, African
American, Latino, and female legislators voted in majority against the war
(Cortright 2004, 8—11). In short: the government was firmly pro-war and
the opposition did not really challenge government.

The United States' most staunch ally and war defender, especially active
in developing public arguments in favor of war, was the United Kingdom
represented by its Labour prime minister Tony Blair. The United Kingdom
would remain the only Western European country with a left-wing govern-
ment to endorse the war. In this perspective, Labour's internal struggle is far
from surprising: on February 27, 2002, 121 of 408 — nearly one in three —
Labour members of Parliament voted against war. This was the biggest revolt
a UK government party. The Tories supported Blair, but the Lib-

| ever within a UK government party. The Tories supported Blair, but the Lib-

~W eral Democrats fully opposed war, with 52 of their 54 members rejecting it. ~W

UK government, hence, was painfully divided on the issue: war supporters
found support among Conservatives, whereas Labour Party dissidents were
backed by the Liberal Democrats.

Spain and Italy were among the most overtly war-supporting countries.
The Spanish government, in particular, seemed to follow U.S. -UK war pol-
icy. Spain, in fact, sent (noncombat) troops, whereas Italy's support would
be limited to opening bases and airspaces to the coalition (though not for
direct military attacks). An almost equally large left-wing opposition chal-
lenged the Iraq policy of Spanish conservative Partido Popular prime minister
Aznar. The Italian case was very similar, with Prime Minister Silvio Berlusconi
of Forza Italia and his right-wing government coalition fully backing war
and a strong and united left-wing opposition ferociously against it. In both
es, the political polarization around the Iraq issue was huge.

Prime Minister Balkenende of the Netherlands and his right-wing gov-
vith Christian-democrats and liberals also supported the idea of
war: the Dutch government agreed to send (noncombat) troops to the region.
The social democrats and greens resolutely opposed to this involvement.
The situation in the Netherlands was a bit peculiar, though, because Balke-
nende I had resigned from office. Three weeks before February 15, general

H$H

-4

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 46_^_

46 VERHULST AND WALGRAVE

elections had been held but the new government, which would be called
Balkenende II, would not be formed until May 26. At the time of the Feb-
ruary 15 demonstrations, government negotiations were just starting; thus,
the Dutch government could not as clearly be situated in the pro-war camp
as its Italian and Spanish counterparts.

In Germany, chancellor and chairman of the German social democrats
Gerhard Schroder had been openly opposing a possible war during his fall
2002 election campaign. This stance had helped him and his green coalition
partner a new term. Later, however, the Schroder government would become
somewhat more temperate in its condemnation of war, granting the U.S.
troops clearance to use German airspace for materiel and troop transport and
not even ruling out a possible UN Security Council vote in favor of war. This
slightly more flexible attitude led Angelika Beer, the newly elected leader of
the government-participating Green Party, to condemn this clearance, argu-
ing that it would be a breach of the German constitution. In summary, the
German government was not really divided about potential participation in
a possible war; there were, however, minor frictions on the degree of non-
participation they should adopt. Both parties agreed that Germany would
not take part in any military action against Iraq, not even when this would
be endorsed by the UN. Meanwhile, opposition leader Angela Merkel
(Christian-democrat) had also turned her party's stance from one of com-
pliance with the United States to a cautious and moderate antiwar position.
Thus, in Germany both government and opposition ultimately rejected an
upcoming war.

In Belgium, all political parties simply (tacitly) agreed on the national
government's antiwar stance. In Belgium, although led by center-right (liberal)
Prime Minister Guy Verhofstadt seconded by center-right (liberal) Foreign
Minister Louis Michel, government fiercely and loudly opposed war. The
country even temporarily blocked a NATO decision about potential support
for Turkey, in case that country would have become engaged in the war. All
opposition parties, from Greens to Christian Democrats, opposed war as well.

In Switzerland too, all parties rebuffed the possibility of war on Iraq.
But in line with the country's long-standing neutrality tradition, Switzerland
opposed war only silently. The only exception was the Green Party, which
wanted the Swiss government to breach its silent opposition and make a clear
and manifest statement against war.

Figure 3.1 summarizes our findings; it shows that the number of dif-
ferent government-opposition configurations is limited. In the officially war-
opposing countries Switzerland, Belgium, and Germany, governments and
the challenging parties all rejected war; antiwar was a valence issue. In Spain,

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 4 7 afc

4

POLITICS, PUBLIC OPINION, AND THE MEDIA \~]

Italy, and the Netherlands, countries that supported the war but did not par-
ticipate in it, right-wing government was in favor of war but the left-wing
opposition vehemently opposed it. In these countries, the conflict corre-
sponded with the traditional government-opposition clash. In the United
States and the United Kingdom, the governments were, of course, pro-war,
but the opposition was divided: U.S. Democrats were split, while UK con-
servatives supported Blair and UK liberals rejected war. The most complex
configuration doubtlessly was found in Britain with the leading party, the
Labour Party — the only European left-wing party in power to support the
war — bitterly divided on the issue. Taking all this into account, we ordered
the eight nations, from most war-seeking to most war-opposing. Although
the Netherlands and Spain superficially had the same political configuration,
it is clear that the Spanish government went much further in defending the
war than the Dutch did. We will use this favoring-opposing order of coun-
tries throughout the book.

Mass Media and the War

Mass media are significant political actors; they intermediate between poli-
tics and the population, and their coverage affects both public opinion and
political actors' behavior. Especially when it comes to international affairs,
conflicts, and war, mass media are often the sole information channel people
can rely on. Therefore, international war and conflict are interesting cases for
those studying the relations among the elite, mass media, and public opinion.
The 1991 Gulf War, especially, received ample scholarly attention (Bennett
and Paletz 1994; Taylor 1992; Wolfsfeld 1997). By and large, the argument
goes that the American government effectively succeeded in steering and
manipulating the news flow to legitimize its military actions in the Gulf
(Hachten and Hachten 2000). The more general idea is that political elites
determine media coverage, be it completely and monolithically (Herman and
Chomsky 1988) or only to a limited extent and in combination with other
actors (Bennett 1990). Either way, the media take cues from political elites,
and their independence is limited, especially in war times (Entman and Page
1994). Entman conceptualized this top-down process as "cascading activa-
tion" (Entman 2003). The ruling administration feeds other political elites;
these affect the media and their news stories, which affect public opinion.
Entman acknowledges that a feedback mechanism exists and that lower-level
frames affect higher levels but this is not the rule.

The 2003 Iraq conflict increased attention to the interaction between
media coverage and war. One key difference between the 2003 conflict and
the 1991 Gulf War is that there were very few officially war-opposing voices

H$H

-4

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 4 8 afc

48 VERHULST AND WALGRAVE

to be heard in 1991. In 2003, many national governments opposed it and
tried to sell their point of view to national and world public opinion. A real
battle over the facts and their interpretation took place on the international
media scene, with the UN Security Council as a primary stage. As a conse-
quence, and corresponding with the elite-dominance hypothesis, we expect
substantive differences in war coverage across nations in 2003, with the
national media following their governments.

A few 2003 Iraq War studies were recently published focusing on the
American media (Calabrese 2005; Entman 2004; Lule 2004; Rutherford
2004). One study concluded that the American media, as expected, sup-
ported the bellicose president and hardly fostered any war-opposing sources
(Rendall and Broughel 2003). Comparative studies about the media cover-
age of the 2003 Iraq War are rare (Berenger 2004). Hooghe and Stolle (2005),
analyzing a week of TV news coverage in nine different nations in the run-
up to the war, counterintuitively found that differences between countries
were limited. Only the American and, to a certain extent, French TV stations
had diverging, more war-supporting or war-opposing coverage; in other war-
supporting nations (the United Kingdom, for example) the TV news coverage
was not different from that in war-opposing countries (Germany, Belgium).
The authors, hence, reject the idea that TV news tends to follow the national
government's position. Their analysis, though, is confined to only one weeks
media and a limited amount of news items per country. Moreover, coverage
in TV news may be much more mainstream and homogeneous than in
newspapers.

Regarding the war in Iraq in 2003, the question is threefold. First, did
the media emphasize the salience of the Iraq issue? Second, did mass media
support or oppose war on Iraq? And third, how did the mass media regard
the protests against war? The third question will be covered in chapter 12,
and the first two are largely addressed in this section. Governments in the
eight nations diverged fundamentally in their stances on Iraq, and the oppo-
sition parties in the different countries had differing opinions. Consequently,
we expect the national mass media to bear the traces of these political differ-
ences. Are political intercountry differences reflected in media differences?

It is clear that the Iraq issue was extremely prominent in all mass media
in all countries in the period preceding the February 15 protests. People's
attention was aroused, and media coverage of the imminent war was exten-
sive. Previous peace demonstration waves, like the protest against the deploy-
ment of cruise missiles in Europe in the early 1980s, had drawn on much
less media attention. Although foreign politics is not the primary issue in most
countries' media, the Iraq crisis was omnipresent. In a comparative analysis

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 49_^_

"4

POLITICS, PUBLIC OPINION, AND THE MEDIA 49

of two major newspapers in each of four countries — France, Germany, Italy,
and the United Kingdom — Kritzinger (2003) contrasted the amount of atten-
tion devoted to the Iraq conflict with the amount of coverage of the 1999
Kosovo crisis, both for the January-March period. Despite the fact that the
Kosovo crisis was happening geographically much closer to the countries
under study, differences in coverage are striking. In one newspaper, the Iraq
crisis got at least forty times the attention as the Kosovo crisis had four years
earlier, and in most newspapers Iraq was five times more written about than
Kosovo.

In terms of the framing of the imminent war, we engaged in an original
media analysis in the eight nations under study. In each country, we content-
analyzed three newspapers: the major left-leaning broadsheet, the major right-
leaning broadsheet, and the most popular national (or local) newspaper. Each
paper was scrutinized for Iraq conflict articles for two months — between
January 21, 2003, that is, three weeks preceding the February 15 protests,
and March 21, 2003, the day after the invasion of Iraq started (see appen-
dix B for more information).

A first step to assess the media's position on the war is to chart the dis-
cussion about the justness of the war. Were only arguments in favor of the
war mentioned, only arguments that dismissed a potential war on Iraq, or
was coverage fairly neutral? Table 3. 1 contains the results of this exercise per
country. First, it shows that the discussion about war and its justification was
at the heart of the media coverage in the run-up to the war. In well over half
of all (potential) Iraq War articles, at least one motive for or against war was
mentioned. Media did not just report about war preparation, its cost, the
new weaponry, the respective strategies, the likely course and consequences.
Overall, the press devoted a large amount of its coverage to the question of
why this war was necessary or unnecessary. We would need systematic com-
parative data about previous conflicts to substantiate this — for example, late-
stage coverage of the Vietnam War, which was also largely devoted to war's
justification — but it appears that this obsession with war's justification or dis-
qualification was exceptional. People who followed the media in the run-up
to the war on Iraq were, thus, constantly confronted with arguments about the
war and incited to take sides in the debate between supporters and opponents.

Yet, clearly, there are some striking differences among the c
The debate about the reasons for war was not equally strong in all c
In Spain, the United States, and the United Kingdom as well as in Belgium
and Germany, the articles mentioning one or more reasons for or against
war outnumbered the ones that did not bring up any of these. One possible
explanation could be the link with the national governments' stances on war:

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 50_^_

"4

s

5^2"

5k

s

£

<n cn ^

c>

0=

,_

S^S^

T

t

1

?2:S °

^

^

s

£

^^

VO

<u

a

A

£

■°'

£

c
'o

fc

Jo 22 S ^

"?

■* <?■

"?

in

o

c

"O

■5

O)

*

3

° ^j 2 =j

°°

ir>

*+

1

1

v-r

E

D
1

-a 1
1 II 1

I I s 1

55 1

g-a

<u

■a s

11

55

H$H

-4

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 51_^_

POLITICS, PUBLIC OPINION, AND THE MEDIA 51

the first three countries were most involved" in the build-up toward war, and
the two latter were the most fierce war opposers in our sample. In Italy, the
Netherlands, and Switzerland, this was exactly the other way round: here,
official war support or opposition was less explicit, possibly leading the media
to engage in more descriptive coverage of the eventuality of war.

A third general observation is that newspapers in all countries highlighted
both sides' arguments in the debate. In none of the countries did one of the
sides get all the credits. Even in the United States and United Kingdom, the
balance was fairly equal, with a comparable number of articles mentioning
no reasons at all, and articles mentioning arguments from both sides of the
pro-contra war debate. At first sight, hence, the media coverage in the news-
papers was fairly balanced, although we did not analyze how the arguments
were presented. Still we find interesting country differences. It is, of course,
difficult to weigh the arguments quantitatively against each other, but the
table shows that there were considerable differences among the countries. Sub-
tracting the percentage of positive arguments from the negative ones, we can
rank-order the countries' media from supporting to opposing war. It comes
as no surprise that the U.S. and UK press were most prone to war, with the
former having a +4.0 and the latter a+8.6 percent difference between pro-
and contra-war arguments. Newspapers in the United States and the United
Kingdom seem to have been inclined to follow their political leaders in en-
dorsing the war. In all other countries this difference is negative: the majority
of the press in continental Europe seemed to have been on the same antiwar
wavelength. Governments' antiwar stances were supported in the three war-
opposing nations (Belgium, Germany, and Switzerland). Yet, strangely, in the
three war-supporting countries (Italy, Spain, and the Netherlands), the press
was been even more inclined to counter the official governments' pro-war
arguments and to emphasize cases against war.

Finally, we asked our encoders to subjectively evaluate whether, accord-
ing to their personal judgments, the article explicitly displayed a pro- or
contra-war position. Each judgment, however, was only taken into account
when they could substantiate it by indicating a (part of a) sentence that made
clear their choice. Whereas the figures in the upper part of the table are more
an indication of how the debate about (not) going to war was held, here we
try to measure actual, explicit, and thus intended, media bias opposing or in
favor of war. The results of this evaluation exercise are shown at the bottom
of Table 3.1.

First, more than 10 percent of the articles analyzed displayed a bias for
or against war. When we look at the spread of biased articles over the coun-
tries, we see a very similar pattern as in the previous analysis. Though in most

H$H

-4

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 52_^_

52 VERHULST AND WALGRAVE

countries, many articles favoring or articles disapproving war were found,
the differences are clear. The British press featured the most overtly war-
endorsing articles, followed by the Spanish and U.S. Explicit antiwar articles,
however, were most found in Italy, the Netherlands, Belgium, and Germany.
Clearly, the Swiss press seems to have largely been joining that country's
ever-neutral status. In none of the countries were both types of articles bal-
anced. Overall, the press sided either with war endorsers or with war opposers.
When we subtract the number of pro-war articles and the number openly
opposing war, very similar results to those of the previous analyses appear:
the UK press was most biased toward endorsing the war (+5.2 percent). The
U.S. (+3.4 percent) and the UK media are the only ones to have positive bal-
ances between pro- and antiwar articles; in all remaining European coun-
tries, it is clearly negative.

Overall, the national newspapers produced a fairly balanced picture of the
run-up toward war. Since the international debate about going or not going
to war was very vivid, national media devoted a large amount their coverage
to it. Yet, in both war-initiating countries, the United States and the United
Kingdom, pro-war arguments clearly were more salient; and when the media
in these countries explicitly took sides, they most often did so in favor of war,
which was exactly the opposite for all other European countries in our sam-
ple. We conclude that there seems a causal link between official government
positions and the press coverage of (upcoming) war. But this link is weak,
and it is conditional. In war-leading nations the media seemed to follow the
government, but in the countries where government only verbally endorsed
the war, the press opposed the government viewpoint. In other words: only
when national governments took strong and rigid positions were they followed
by the press. Maybe the way national public opinion saw the eventuality of
war was been a determining factor for the direction of press coverage.

Public Opinion and the War

Most new social movements neither have strong organizational resources
nor are the beneficiaries of long-established loyalties that help them over-
come periods of invisibility. The public's opinion on an issue determines the
movements mobilization potential. Although the relationship between what
Klandermans (1984) called consensus mobilization and actual turnout is not
linear, because of diverging action mobilization capacities, a favorable public
opinion can boost protest turnout. Also, public opinion support is relevant
in terms of the protest's impact. Supportive public opinion is paramount
especially for countries waging war: Did the public in the eight nations favor

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 5 3 afc

POLITICS, PUBLIC OPINION, AND THE MEDW

Between January 2 1 and 27, 2003, just before the February 1 5 protests,
European Omnibus Survey (EOS) Gallup Europe (2003) conducted a com-
parative opinion poll in thirty European countries, covering 15,080 people
aged fifteen years and older. 2 The poll contains extensive evidence regarding
European public opinion about the potential war. We look first at the war's
salience among European populations and then at war support.

Salience was measured somewhat awkwardly. The respondents were con-
fronted with six pending international problems and were tasked which of
these most urgently needed to be solved. Apart from the imminent war on
Iraq, the respondents could choose among the Israeli-Palestinian conflict,
Indian-Pakistani tensions, international terrorism, the Chechnyan war, and
the North Korean nuclear crisis. Hence, issue saliency was not measured via
the traditional open "Most Important Problem" question. But since we are
interested in differences between countries and not in absolute levels, this is
not too problematic. 3 In Figure 3.2 the white bars illustrate the responses of
the EU states, with Switzerland added.

In all sixteen nations, except for Spain, public opinion considered a war

"4

I

Figure 3.2. Percentages of European nations' populations that found Iraq War
most important issue and that opposed it

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 54_^_

54 VERHULST AND WALGRAVE

in Iraq as the single most important international problem. Yet, there are large
intercountry differences. Our eight countries nicely cover the whole range
of opinions. The German population was most (53 percent) concerned with
the Iraq War, followed by the Belgian (47 percent). The Dutch, Swiss and
British populations each scored around 40 percent, while both the people of
Spain and Italy seemed to have cared considerably less about Iraq (both around
30 percent); the Spanish people, especially, did not perceive the possibility
of war as the most important international problem. Probably because of
their domestic terrorism problems, Spaniards, mention international terror-
ism as top priority (35 percent). Italians also had severe domestic terrorism
in the past and, likely consequently, attributed great priority to terrorism
(27 percent) but also to solving the Israeli-Palestinian conflict (19 percent).
Both the Italians and Spaniards considered the risk of terrorism in their own
country significantly higher than did the residents of the other European
countries, and both considered the United States as best capable of effectively
fighting international terrorism.

Issue salience is only half the story. The Gallup Poll also elicited people's

opinion on several statements regarding the Iraq conflict. We combined seven

statements about Iraq into one simple war opposition scale. 4 The results are

A^, shown in the dark bars in Figure 3.2. Again our eight countries span the /is

I whole spectrum of public opinion vis-a-vis the war. UK citizens, clearly, are

far least opposed to war; on average, only 40 percent of them agreed with
antiwar propositions. Of that in all European countries, public opinion in
the United Kingdom is most divided. In the Netherlands, Italy, Spain, and
Belgium (large) majorities oppose war. The most war-rejecting populations
are found in Germany and Switzerland. Only Greece and Austria, countries
not covered in this book, beat the Swiss in antiwar feelings. Saliency and opin-
ion direction seem only weakly related. A country like Spain, for example,
scored particularly low on saliency but contains a fair amount of war opposers.
The correlation between saliency and war opposition is positive, but mod-
est (r = 0.36).

Let us further specify the precise content of public opinion in the eight
nations, since the above scale draws only a raw picture. We focus on four
specific propositions: that Iraq poses a threat to world peace; that the respon-
dents' country can take part in a war when justified by a UN resolution; that
oil is the main motivation for the United States to invade Iraq; and that the
United States should intervene militarily in Iraq, even unilaterally. We choose
these because we presented the participants in the February 1 5 protests with
exactly the same statements and, later, we will compare public opinion with

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 5 5 dfc

POLITICS, PUBLIC OPINION, AND THE MEDIA 55

that of the demonstrators. Figure 3.3 contains the results for the seven Euro-
pean nations in our sample.

Most European populations agreed that Iraq was a threat for world peace;
this first diagnostic statement is not really causing much discord among
Europeans. The same applies to their account of the third element, the United
States' motivation to invade Iraq: a large majority concurs that a self-interest
driven search for oil is the main incentive. A large majority, as well had the
same response to the second proposition, rejecting the possibility of the United
States' unilateral invasion of Iraq without UN backing, which is what even-
tually did happened. The real divisive issue is the fourth one, the justification
of (the own country's partaking in) a potential invasion via the UN Security
Council. Here, opinions differ strongly. In the United Kingdom, an over-
whelming majority would approve war backed by such a resolution, while
in Switzerland even a Security Council endorsement would not convince a
majority of Swiss of the justness of war. No wonder Blair's UK government
did everything within its power to get UN backing. This divisive statement
goes to the heart of the debate about the war in Europe, especially the uni-
lateral and even illegal character of a possible Iraq war.

For the main purpose of this book — explaining why demonstrators in
| eight nations differ — divergences among countries are most interesting. Again,

"W" the UK population endorses war the most by far, followed by the Dutch and ~®~

Italian populaces, who moderately favor the war. German and Belgian peo-
ple are more skeptical about invading Iraq, while Swiss and Spanish citizens
seem absolutely opposed it. Put otherwise: while the United Kingdom's Tony
Blair was more or less successful in at least sparking doubts about Iraq in the
minds of the British people and while Italian prime minister Silvio Berlusconi
was less convincing, Spanish prime minister Jose Maria Aznar completely
failed to convince his population of the need to invade Iraq. Of course, the
fact that a majority of Britons would have supported war with UN backing
does not imply that they supported the actual war, which was being waged
without UN support; UN approval was crucial for British public opinion.
The fourth statement clearly shows that a majority of British citizens did not
support the actual war. And Swiss, Belgian and German public opinion con-
tested war, in line with their governments' positions.

Unfortunately, the EOS Gallup Poll is confined to Europe and contains
no evidence on U.S. public opinion. Where would the U.S. public stand?
U.S. polling evidence is widely available but not always comparable with
European surveys. We choose to focus on a poll conducted by the same poll-
ing company around the same time as the European EOS Gallup Poll. 5 At
this time, 56 percent of Americans would give the weapon inspectors more

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 5 6 dfe

"4

8 & .1 £

r j 3 - s '§ J §

1 <S "cr 3 .S | a <£

« e | r « S3 J

1 1| ||| { m

jf i .S ,° Hu J= Z

H$H

-4

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 57_^_

POLITICS, PUBLIC OPINION, AND THE MEDIA 57

time to conduct their inspections, versus 4 1 percent who believed that Iraq
had already had enough time to prove that there were no such weapons.
Relatedly, 39 percent of the surveyed answered that the U.S. should invade
Iraq as soon as the Bush administration decided on it; 56 percent of the peo-
ple could not favor an invasion without a new UN vote authorizing military
action. This last figure is particularly relevant, since it can directly be com-
pared with the European data. In Europe, between 68 and 87 percent
rejected war without UN backing, while "only" 56 percent did in the United
States. On the question "Which comes closer to your view? UN weapons
inspectors alone can eliminate the threat Iraq poses to other nations. Or, mil-
itary action is needed along with weapons inspections to eliminate the threat
Iraq poses to other nations," 71 percent of the people agree with the latter
proposition. Evidence from other polls at that time largely underpins the far
more war-supportive attitude of the U.S. people before to the war: approval
rate for the way the Bush administration handled the Iraq conflict was high,
military action "to remove Saddam Hussein from power" was favored by a
large majority, and more than half of the Americans said they would sup-
port war even if it was not approved by the Security Council, and more than
90 percent considered Iraq a threat to the United States. By and large, it is
safe to consider the U.S. public opinion the most supportive of war of all the
countries under study. A majority of the U.S. people believed in the neces-
sity for military intervention. Of the governments in all war-supporting coun-
tries, the Bush administration was doubtlessly most successful in convincing
its people of the need for war. However, this does not mean that the U.S.
public was not divided about the issue: a considerable minority of U.S. cit-
izens did oppose the war.

Conclusion

Political context, media coverage, and public opinion — the three relevant
dimensions of societal context possibly determining the features of protesters
in our eight countries — are not independent from one another. The direction
of the possible causal chain connecting these three is not straightforward, yet
we tend to believe that the government's initial stance on Iraq and the polit-
ical opposition's reaction to this position are essential factors influencing the
kind of media coverage and (subsequent) public opinion. Governments set
the agenda of the media. And, as — at least in the West — war on Iraq is an
unobtrusive issue not experienced directly by the people, but only interme-
diated by the media, we believe the media to impact public opinion. At the
same time, the commercial forces most mass media are subject to also make
them to cater to their public. Whatever the causal path, politics, media, and

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 58_^_

VERHULST AND WALGRAVE

on of political a~
regarding the Iraq War

'd public opinion

m

public opinion are associated. Figure 3.4 summarizes our findings. It presents
a very rough simplification of reality: government stance and public opin-
ion are complex phenomena, and summarizing them in a single pro- or anti-
war continuum cannot but oversimplify reality to some extent. The same goes
to a possibly even larger extent for national media, certainly when one takes
into account that they contain viewpoints from both broadsheets and tabloids
and both left- and right-leaning news outlets. Nonetheless, the scale in the
figure is very useful, as will be demonstrated later in this volume.

Figure 3.4 shows that the United States and United Kingdom, on the
one hand, and the officially war-opposing countries, on the other, had a very
clear and homogeneous position vis-a-vis the war: politics, media, and public
opinion fostered and supported war in both countries. The opposite applied
to Germany, Belgium, and Switzerland: all three agreed on their disapproval
of war. In Spain, Italy, and the Netherlands, a far more mixed pattern appears,
with the national government's war supporting position being challenged by
both media and public opinion in all three countries. National leaders clearly
lacked media support to convince their citizens of their Iraq policy.

Notes

1 . For reasons of convention we will use the term "public opinion" for the aggre-
gate of individual opinions usually measured in surveys. We are aware that these
opinions are mostly not publicly expressed and do not target the public as media and
the strict sense, these aggregated opinions are not "public

ind that

opinion."

2. All public opinion data :
mentioned otherwise.

3. The exact wording of the
solved as first top priority." Agrei

this chapte

are derived from this study, unless
'The potential war in Iraq should be

I

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 5 9 afc

POLITICS, PUBLIC OPINION, AND THE MEDIA 59

4. The scale was simply the mean (in percentage) of the aggregation of the war-
ders on the following statements: (1) Iraq represents a threat to world

peace (disagree); (2) Oil is the main motivation for which the United States wants
to intervene militarily in Iraq (agree); (3) The United States should intervene mili-
tarily in Iraq even if the United Nations does not give its formal agreement (disagree);
Do you consider that it would be absolutely justified, rather justified, rather unjusti-
fied, or absolutely unjustified that our country participates in a military intervention
in Iraq? (4) If the Iraqi regime does not cooperate with UN inspectors (unjustified)?

(5) If the UN inspectors discover weapons of mass destruction in Iraq (unjustified)?

(6) If the United States intervenes militarily in Iraq without a preliminary decision
of the United Nations (unjustified)?

5. It concerns a CNN / USA Today / Gallup poll, which asked a sample of a
thousand adults nationwide about their opinions on several Iraq-related proposi-
tions. The poll was conducted on January 23-25, 2003. Evidence can be read at
www.pollingreport.com/iraql7.htm (we accessed the Web site December 7, 2009).

Works Cited

Bennett, Lance. 1 990. "Toward a Theory of Press- State Relations." Journal of Com-
munication!: 103-25.
I Bennett, Lance, and David Paletz, ed. 1994. Taken by Storm: The Media, Public I

~^$7~ Opinion, and U.S. Foreign Policy. Chicago: University of Chicago Press. ~^7~

' Berenger, Ralph. 2004. Global Media Go to War: Role of News and Entertainment '

iring the 2003 Iraq War. Spokane, Wash.: Marquette.

Calabrese, Andrew. 2005. "Casus Belli: U.S. Media and the Justification of the Iraq
War. " Television and News Media 6: 1 53-75 .

Cortright, David. 2004. A Peaceful Superpower: The Movement against War in Iraq.
Goshen, Ind.: Fourth Freedom Forum. s

Entman, Robert M. 2003. "Cascading Activation: Contesting the White House's
Frame after 9/11." Political Communication 20: 415-32.

. 2004. Projections of Power: Framing News, Public Opinion, and U.S. Foreign Pol-
icy. Chicago: University of Chicago Press.

Entman, Robert M., and Benjamin I. Page. 1994. "The News before the Storm: The
Iraq War Debate and the Limits of Media Independence." In Taken by Storm:
The Media, Pub/: . Lance Bennett and David

Paletz, 82-104. Chicago: University of Chicago Press.

EOS Gallup Europe. 2003. International Crisis Survey, January 2003. www.eosgallup
europe.com/int_survey/index.html.

Hachten, William, and Marva Hachten. 2000. "Reporting the Gulf War." In
Media Power in Politics, ed. Doris A. Graber, 317-25. Washington D.C.: CQ

H$H

03 Chapter 3_Walgrave 22/01/2010 3:36 PM Page 60_^_

60 VERHULST AND WALGRAVE

Herman, Edward, and Noam Chomsky. 1988. Manufacturing Consent. New York:
Pantheon.

Hooghe, Marc, and Dietlind Stolle. 2005. "Kroniek van een aangekondigde oorlog.
Een vergelijkend onderzoek naar de verslaggeving in de aanloop naar de oorlog
in Irak." In Nieuws op televisie: Televisiejournaals ah venster op de wereld, ed.
Marc Hooghe, Knut De Swert, and Stefaan Walgrave, 175-93. Leuven, Belgium:
Acco.

Klandermans, Bert. 1984. "Mobilization and Participation: Social-psychological Ex-
pansions of Resource Mobilization Theory." American Sociological Review 49:
583-600.

Kritzinger, Sylvia. 2003. "Public Opinion in the Iraq Crisis: Explaining Develop-
ments in the UK, France, Italy and Germany." European Political Science 3:
30-34.

Lule, Jack. 2004. "War and Its Metaphors: News Language and the Prelude to War
in Iraq, 2003." Journalism Studies 5: 179-90.

Neidhardt, Friedhelm, and Dieter Rucht. 1991. "The State of the Art and Some
Perspectives for Further Research." In Research on Social Movements, ed. Dieter
Rucht, 421-61. Frankfurt a.M.: Campus Westview.

Rendall, Steve, andTara Broughel. 2003. "Amplifying Officials, Squelching Dissent:
| FAIR Study Finds Democracy Poorly Served by War Coverage." FAIR: Fairness

W and Accuracy in Reporting. http://www.fair.org/index.php?page=ll45. ~^9p~

Rochon, Thomas R. 1988. Mobilizing for Peace: The Antinuclear Movements in West-
ern Europe. Princeton, N.J.: Princeton University Press.

Rutherford, Paul. 2004. Weapons of Mass Pc ; ing the War against Iraq.

Toronto: University of Toronto Press.

Taylor, Philip. 1992. War and the Media: Propaganda and Persuasion in the GulfWar.
Manchester, UK: Manchester University Press.

Wolfsfeld, Gadi. 1997. Media and Political Conflict: News from the Middle East. Cam-
bridge: Cambridge University Press.

H$H

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page61_^_

4
Legacies from the Past:
Eight Cycles of Peace Protest

Bert Klandermans

m

Although the protest against the war in Iraq can be studied in its own right,
it is also a link in a much longer chain of protest events regarding issues of
peace and war. Large protest movements proceed in cycles; periods of mobi-
lization and demobilization alternate. That no mass mobilization takes place
does not necessarily mean a movement has disappeared, because between
periods of mobilization movements might continue to exist in abeyance
(Taylor 1989), and such abeyance structures appear important in later peri-
ods of mobilization (see Downton and Wehr 1998; Everts and Walraven 1984;
and Kleidman 1993 for examples regarding the peace movement).

The dynamics underlying such cyclical developments are in part built
into the very features of the movement. Obviously, peace movements respond
to issues of peace and war in national and international contexts. Therefore,
the peace movements in all eight countries of our study share a history of
comparable mobilizations. Indeed, each has a tradition that goes back a long
way and on occasion has shown to have a strong mobilizing capacity. Per-
haps more than any other, peace movements have reacted to international
developments, be they nuclear armament, immanent wars, or tensions be-
tween states, for example. This explains why the various national movements
have gone through similar cycles. At the same time, although the cyclical
patterns are comparable, the amplitudes can be very different: there can be
very big demonstrations in one country, while another witnesses only mar-
ginal ones. National circumstances are, of course, responsible for this variation.

In this chapter I will try to show how the interplay of general and country-
specific movement characteristics, international relations, and national politics

H$H

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page 62_^_

"4

62 KLANDERMANS

t for similar, yet different, dynamics of mobilization. In doing so, I
will concentrate on the last two protest waves — that against the deployment
of cruise missiles and that against the First War in Iraq. 1

The Movement against Cruise Missiles

The movements against cruise missiles in Europe and the movement for a
nuclear freeze in the United States developed against a background of a rel-
atively stable geopolitical situation that underwent an increased polarization
and remained largely unchanged over the course of the movement's life
(Klandermans 1991b). The Reagan years were characterized by marked ten-
sion between the superpowers, which were infused with new ideological zeal.
Reagan's rhetoric of the "Evil Empire" accompanied a massive escalation in
U.S. defense expenditures and assumed a conception of security that made
nuclear war seem possible in the European theater. This resulted in a gen-
eral perception of threat that fuelled the European peace movements (Rochon
1988). Moreover, that each national government had to decide on its terri-
tory's deployment brought the issue of nuclear armament home and consti-
tuted the basis for international alliances among national peace movements.
The peace movement of the 1980s constituted one of the first occasions of
significant interdependence among the social movement sectors of different
European countries. Activists traveled extensively not only within countries
but also between them, wherever the front of the struggle was perceived to
be. None of the national movements could prevent its government from
deciding to deploy cruise missiles, but this is not to say that the movements
were failures. They did mobilize large proportions of the population, and
kept the cruise missiles on the political agenda for many years. The unspo-
ken consensus on armament issues was definitely broken down.

The movement in the United States took a different direction. The call
for a nuclear freeze succeeded in creating a broad coalition that mounted a
formidable political force. The movement commanded enough resources and
allies that it could successfully challenge the president's planned deployment
of a new generation of missiles, the MX.

Despite the common grievance the movements in the various countries
differed sometimes considerably because of local circumstances. A brief over-
view of the movements in the countries where we conducted our research
underlines this point. 2

United States

In the United States the campaign against cruise missiles never really got off
the ground, primarily because the trajectories of the peace movements in

H$H

-4

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page 63_^_

LEGACIES FROM THE PAST 63

Europe and the United States were rather different. 3 Indeed, at some point
the goals that the European and U.S. movements had defined for themselves
were almost incompatible. While in Europe the peace movement focused on
cruise missiles, in the United States it centered on a nuclear freeze. Although
the movements on both sides of the Atlantic felt obligated to support each
other, providing this mutual reinforcement was fairly complicated. Essen-
tially a proposal for bilateral disarmament, the freeze did not coincide very
well with the more unilateral position of the movements against cruise mis-
siles. Moreover, as far as specific weapons systems were concerned, political
action against the MX was more likely to succeed than similar action against
cruise missiles. To the average American, cruise missiles were a European
problem rather than a U.S. one. Consequently, in the United States it was
difficult to mobilize protest against cruise missiles, and little action was under-
taken to prevent the deployment of the missiles in Europe, despite European
requests for such active support.

This is not to say that the U.S. peace movement was inactive. On the
contrary, the nuclear freeze movement was one of the major peace mobi-
lizations of the twentieth century. It followed earlier mobilizations against
atmospheric atomic testing and the war in Vietnam in the 1960s and 1970s
respectively. On the withdrawal of the United States from Vietnam the peace
movement dissipated, after an apparent victory, but also as in the earlier epi-
sode, the movement left behind an enlarged residue of peace movement orga-
nizations, trained activists, and sympathetic middle-class adherents who would
be mobilized more rapidly during the next cycle of antiwar protest. This cycle
exploded in 1980, as people unified behind the banner of the nuclear freeze
(Marullo 1992).

At the start of the nuclear freeze the U.S. peace movement consisted of a
wide range of groups and associations, ranging from moderate arms-control
associations to radical pacifists. These diverse segments of the peace move-
ment were largely invisible to the public and shared little with each othei
except their pursuit of peace. The call for a nuclear freeze changed that by
providing a short-term goal that most groups accepted, a set of tactics fs
iar to most, and a recruitment message that greatly expanded the base of adher-
ents (Kleidman 1993).

By 1984, however, the diversity of goals, strategies, and tactics of
ous segments of the movement had begun to reemerge, and new fissures de-
veloped. The range of problems confronting the movement, along with an
increasingly less bellicose president and the warming of the U.S. -USSR rela-
tions, led to the withdrawal of resources from the movement and its furthei
retrenchment and splintering.

H$H

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page 6 4 afc

64 KLANDERMANS

(/n/fei/ Kingdom

The British movement against cruise missiles had its roots in the Campaign
for Nuclear Disarmament (CND). Founded in 1958, CND was one of the
oldest peace movement organizations in Western Europe. Originally, CND
directed its protest against Great Britain's own nuclear force. Before the move-
ment against cruise missiles took hold, CND engaged in several protests
against nuclear armament. These protests reached their peak in 1962, when
a hundred and fifty thousand people participated in the annual Easter march.
CND became the backbone of the movement against cruise missiles. From
1979 until 1983 this movement figured prominently on the British politi-
cal stage, though the political opportunity structure during these years was
not very favorable. Prime Minister Thatcher's government was extremely un-
receptive to any of the movement's claim and, in fact, engaged in outright
repression. However, the movement was successful in persuading the Labour
Party to support unilateralism. After the government decided to deploy the
cruise missiles in 1983, the peace movement and CND declined. Apart from
Greenham Commons — the women's camp at the base where the missiles
were deployed — no major actions were organized. Yet CND did not disap-

I pear; it survived, although in a less spectacular fashion.

Afr- CND's alliance system encompassed the unions, some of the churches, .fa.

I and some of the local governments. Coalitions, however, existed primarily

at the local level. At the national level, churches were reluctant to enter into
coalitions, and affiliations with the parties of the Left only underscored the
government's claim that the movement was primarily a leftist affair. Although
at its peak CND contained 1,100 local groups with 250,000 members, its
organization was rather centralized compared to that of other peace move-
ments. Activists and members were typically middle class. Fifty percent of
the membership was employed in non-manual, professional jobs; the ranks
of the CND included almost no blue-collar workers.

Spain

At the time of the cruise missiles controversy, Spain was still to become a
member of NATO, and therefore the Spanish peace movement did not engage
in the struggle against cruise missiles. 4 The Spanish movement did launch,
however, with its Portuguese counterpart a campaign for nuclear-free zones.
In 1986 about four hundred town councils in Spain and Portugal had de-
clared themselves nuclear-free zones (Carter 1992).

The major issues in those years were, however, specific to the Spanish
>n, namely the presence of U.S. military bases in Spain and Spain's

H$H

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page 6 5 afe

LEGACIES FROM THE PAST 65

d NATO. In 1953 Francisco Franco signed a treaty with the
United States about military bases, and a new agreement was to be signed in
1982. In January 1981, the first big Spanish peace demonstration, aimed at
the military base in Torejon, took place as twenty thousand people marched
from Madrid to Torejon (Carter 1992).

A second issue the peace movement attempted to mobilize the public
against was Spain's membership in NATO. In 1982, the then-conservative
government took Spain into NATO. Afterward, the Spanish peace move-
ment's central aim became bringing Spain out of the Alliance as soon as pos-
sible. When the Partido Socialista Obrero Espana (PSOE; Socialist Labor
Party Spain) took over government, the countries moves toward military inte-
gration in NATO halted, but soon it became clear that the government was
reluctant to withdraw from NATO. In 1986, Spain held a referendum, and
to the great shock of the peace movement, not only did a large proportion
of the electorate abstain from voting, only one quarter of those who did vote
opted to leaving (Carter 1992).

Italy

Compared to other European movements against cruise missiles, the one in
_^k Italy was particularly weak and transient. It began later than the others, as _fo_

| an ad hoc body that met to organize a demonstration against cruise missiles

to be held in Rome in 1981. The Italian government decided rather early
(1983) to deploy cruise missiles at Comiso in Sicily, and most of the move-
ment's protest activities took place after this decision, aiming at the base where
missiles were to be located. A massive demonstration in Rome, in which five
hundred thousand people participated, did not make the government change
its mind. Shortly thereafter, the movement collapsed.

The movement against cruise missiles in Italy was a heterogeneous co-
alition of the Communist Party and the leftist parties and several nonpolit-
ical organizations. In addition to the traditional sociopolitical organizations,
groups of grassroots activists from such different backgrounds as student,
the feminist, and environmental movements and Catholic lay organizations
joined the campaign against cruise missiles. Together, the movements had a
very loose structure, and attempts to institutionalize it invariably failed.

The little that is known about the popular support of the movement
suggests that 25 to 40 percent of the Italian population sympathized with it.
Compared to the other European peace movements, the Italian group was
not very successful. Although it could mobilize massive demonstrations, it
did not really have much impact on Italian politics.

H$H

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page 66_^_

KLANDERMANS

The Netherlands

The movement in the Netherlands really acquired momentum with the
petition against the neutron bomb in 1977. At the same time the Interde-
nominational Peace Council (IKY) began its campaign against nuclear arms.
For more than four years it was uncertain whether a majority in the parliament
supported nuclear deployment. By exploiting this situation, the movement
could have the decision to deploy postponed several times. As a consequence,
the Dutch movement lasted longer than any other European peace move-
ment. As late as the end of 1985, it was still organizing a mass petition. Even
though the government did decide in November 1985 to deploy cruise mis-
siles, the 1987 signing of the INF treaty prevented implementation of this
decision. Thus, the Netherlands was the only NATO country in which cruise
missiles were never actually deployed. The movement succeeded in persuad-
ing the Social Democratic Party and the labor unions to join it, but despite
the movement's strong position within the churches, it failed to persuade the
Christian Democratic Party.

The movement developed an elaborate alliance system — leftist parties,
unions, women's organizations, youth organizations, and others coalesced in
3-called peace alliances at the national and the local levels. The n

^fc^ itself was fairly decentralized. Although it had national headquarters, its main _^

I strength was the local peace groups that did the actual mobilizing. The activists

were typical of the participants in new social movements: young, middle-
class professionals employed in social services.

After the government's decision to deploy cruise missiles, the movement
declined rapidly. Although it could not prevent the government from decid-
ing to deploy cruise missiles, the repeated postponements of the decision
were considered a success. Public support of the movement was impressive:
35 to 40 percent of the population was unconditionally against deployment
of cruise missiles; 30 percent was against deployment upon specific condi-
tions; and 2 5 percent was in favor of deployment (Oegema 1991). The move-
ment has been able to mobilize massive participation in two demonstrations
and a petition, but participants were primarily left-wing citizens. The move-
ment succeeded in keeping nuclear armament on the political agenda for
almost ten years.

Switzerland

The Swiss movement against cruise missiles — whose demands were not
directed at any Swiss authority — was a result of international opportunities
and the influence of mass peace protests in neighboring countries (Giugni

H$H

-4

04 Chapter 4_Walgrave 22/01/2010 3:36 PM Page 6 7 afc

LEGACIES FROM THE PAST 6j

2004). 5 Kriesi et al. (1995) argue that the protests against cruise missiles in
Switzerland to a large extent followed the early mobilization of the German
peace movement. Like Germany, Switzerland had a history of strong mobi-
lization for the Easter marches. The Swiss peace movement can be charac-
terized by a strong focus on internal issues and an extensive use of direct
democratic instruments to influence the political agenda. The most impor-
tant peace movement organization during the first decades after World War II
was the "Swiss peace council," founded in 1945. The protests against cruise
missiles in Europe spread to Switzerland by the end of 1981: In December
1981, the Swiss peace council together with some other groups, such as
Women against War, as well as trade unions and left-wing political parties,
mobilized thirty thousand people for a large national demonstration against
global nuclear armament. In 1982, thirty thousand participated in the tra-
ditional Easter march. In November 1983, a second national demonstration
on the issue of global nuclear armament took place. With fifty thousand par-
ticipants, this was one Switzerland's biggest-ever political gatherings.

The roles of the Swiss army defense industry, which were already a con-
cern before the cruise missiles protest wave in Europe, remained heavily on
the agenda of the Swiss peace movement thereafter. The Swiss army became
the most central target of the peace activists. One way of resistance was the
refusal of military service, and the numbers of young men going to jail for
their political convictions increased drastically. The most important means
to bring these issues to the political agenda were however not protest and other
unconventional forms of political participation but, rather, direct democratic
instruments. Since 1969, there have been a number of popular votes on issues
like arms trade, civil service, abolition of the Swiss army, and even the pur-
chase of warplanes. While in the beginning, Swiss peace council has been
heavily engaged in this political work, it lost most of its importance for polit-
ical mobilization of peace issues during the 1990s. During the 1980s, the
Group for a Switzerland without an Army (GSoA), became the most visible
peace movement organization in Switzerland. 6

Belgium

Belgium had a strong movement against cruise missiles. 7 Indeed, it was the
only country in which the movement continued to organize mass demonstra-
tions after the government had decided to deploy cruise missiles. The Belgian
peace movement differed from those in the other European countries by being
completely rooted in the new social movements. More specifically, it was
embedded in the network of environmental, women's, and third-world groups
and centers crisscrossing the country. This network became the structural

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 68_^_

KLANDERMANS

base of the peace movement. Although the new social movements provided
the infrastructure for the peace movement, they could not turn it into a mass
movement. Only when the peace movement entered a coalition with tradi-
tional political and social organizations — such as the labor unions, political
parties, and church organizations — did the true mass movement came into
being. It maintained a rather loose structure of local groups and organizations,
coordinated by a national umbrella organization. The activists in the Bel-
gian movement, too, were young, highly educated professionals employed
in civil service or education. Popular support for the movement ranged from
two-thirds to three-quarters of the population.

Despite the strong support, the movement was could not prevent the
government from deploying cruise missiles, although it was strong enough
to force the government to postpone its decision several times. Initially, the
political opportunities had seemed favorable. Members of Parliament in-
cluding Christian Democrats were divided on the issue, and the odds for a
parliamentary strategy looked good. But at the crucial vote, thirty Christian
Democratic members who had previously supported the movement stuck
with the party line and voted in favor of deployment.

I Germany

W Like the British peace movement, the German movement has a long history, ~w~

one that goes back to the early 1950s when Social Democrats, unions, Protes-
tant church members, communists, students, and university professors pro-
tested West Germany's entry into NATO and its subsequent rearmament. A
decade later, West German protesters, following CND, organized annual
Easter marches that, at their peak in 1968, would attract more than 350,000
participants. In the 1980s, the movement against cruise missiles grew — the
largest social movement West Germany had experience since World War II. In
this decade, a whole array of new groups, arising from the extra-parliamentary
opposition, joined the movement. These groups, to a large extent, deter-
mined the nature of the new German peace movement. Starting as an ad hoc
committee for the organization of large demonstrations in Bonn in 1981,
the West German movement against cruise missiles developed later than
parallel movements in other countries. Unlike the previous peace movements
in West Germany, the movement against cruise missiles did not initially
have the support of the Social Democratic Party (SDP) and the labor unions.
Later, when the SDP was no longer in office, it changed its policy and be-
came more sympathetic toward the peace movement. This shift, together with
the relatively closed German political system, produced a movement rooted
primarily in extra-parliamentary opposition. Most of the groups that played

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 6 9 afc

LEGACIES FROM THE PAST 69

a central role in the campaign against cruise missiles had their origins in
other movements — feminist, environmental, citizen, or youth. After 1983,
when the government decided to deploy, the movement declined, although
it did not disappear and continued to organize occasional protests.

The German peace movement had perhaps the most decentralized orga-
nization of all the ones in Europe. A steering committee was responsible for
coordinating events and activities on the national level, but the groups rep-
resented in this committee took great pains to ensure that it did not become
too powerful. For the most part, local peace groups were not related to any
overarching organization and were eager to keep their autonomy. The move-
ment was basically a grassroots organization with only a very loose superstruc-
ture, and the many different currents within it easily led to factionalism.

Public support of the peace movement was overwhelming. At its peak,
it had the backing of more than 60 percent of the people, from every sector
of the population (young and old, working- and nonworking class, left and
right, religious and nonreligious) (Cooper and Eichner 1991), while an esti-
mated 2 to 4 million people were active within the movement.

Entr'acte

| Soon the international situation changed radically. Gorbachev started a pro-

~W gram of reforms, which over time led to substantial reduction of the nuclear ~W

arsenal of both superpowers. When Gorbachev began a foreign policy inspired
by new values of openness and a search for dialogue with the West, percep-
tions of threat began to diminish, and former activists became increasingly
difficult to mobilize. Established goals of the movements became less rele-
vant and were actually achieved in the context of the East- West superpower
summits, which to many symbolized the end of the cold war. Then came the
events of 1989. One by one, Eastern European countries underwent radical
political change that the Soviet Union did not hinder and even, as in the cases
of East Germany and Bulgaria, actively promoted. These events created a
new social and political reality in Europe that significantly affected the peace
movement sector. Within a decade, dissolution of the Soviet bloc took place,
effecting a further reduction of the sustained international tension that had
characterized the cold war. Many peace activists point to the movement's de-
layed effect in achieving these changes in the security arrangements. Other
analysts viewed the change as the consequence of an untenable economic
situation in the eastern bloc.

In any event, after the collapse of the eastern bloc, the prospect for the
peace movement changed substantially. The next international crisis devel-
oped in a radically different environment — one in which there is only one

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 70_^_

KLANDERMANS

superpower and wars increasingly come to be presented as guided by moral
motives (Ruzza2004).

The Protests against the Gulf War

On August 2, 1990, Iraq invaded Kuwait. Washington, D.C., reacted by
immediately blocking Iraq's assets and obtaining international approval for
an arms embargo. Soon after, a multinational military contingent primarily
composed of American, French, British, Egyptian and Syrian troops was
assembled. A few days after the congressional elections on November 6,
President George Bush ordered a doubling of the 150,000 troops already
present in the Gulf. On November 29 the United Nations voted to author-
ize member states to "use all necessary means" to implement Resolution 660
if Iraq did not withdraw from Kuwait by January 15, 1991. On January 17,
air strikes began. A February 23 deadline for a ground offensive passed, and
a last-minute Soviet proposal for a negotiated solution of the conflict was
not accepted by the coalition. A "hundred-hours" war crushed the Iraqi mil-
itary and inflicted heavy Iraqi casualties, forcing that nations withdrawal from
Kuwait.

As Koopmans (1999) observed, no other war in the 1980s or 1990s
decades led to such immediate and massive protest as the Gulf War. Partic-

| decades led to such immediate and massive protest as the Gulf War. Partic-

~W ularly interesting was the growing importance of international diffusion of ~W

protest. What started with the movements against the cruise missiles further
expanded in the protests against the Gulf War, for example, the interna-
tionalization of protest and of protest issues. The importance of protest diffu-
sion is perhaps most clearly demonstrated by the phrase "No Blood for Oil,"
which became the peace movement's slogan across the globe. The build-up
to the war was accompanied by a short spell of intense protests between Jan-
uary 12 and 26 all over the world. Western European movements began to
grow and mobilize (Brittain 1991). As the deadline approached, activities
steadily increased in intensity and reached the high mobilization level of the
early 1980s. But the levels of mobilization differed for the various countries;
France, for example, which had hardly mobilized against cruise missiles, wit-
nessed large demonstrations, while countries such as the Netherlands, where
massive demonstrations against cruise missiles took place, had no protest of
any significance. What follows is a brief overview of the anti-Gulf War pro-
tests in the eight countries where we conducted our study.

United States

Thousands of American citizens protested their country's military interven-
tion. The first national demonstration in New York, on October 10, 1990,

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 71_^_

LEGACIES FROM THE PAST Jl

sponsored by the Coalition to Stop U.S. intervention in the Middle East, was
a modest one of: between 10,000 and 20,000 people. But the protest soon
escalated, and demonstrations reached much higher numbers: on January
19, 50,000 in Washington, D.C., and 100,000 in San Francisco; one week
later, during the National Campaign for Peace in the Middle East, between
100,000 and 250,000 demonstrated in the capital city and 100,000 marched
the streets on San Francisco. At the same time the protest became more dis-
ruptive. On January 19, a thousand people were arrested in San Francisco
for civil disobedience actions; in the following three weeks no fewer than
twenty-seven hundred were arrested.

United Kingdom

The Gulf War of 1 99 1 led to a brief revival of the peace movement in
Britain, but mobilization fell far short of the antinuclear protests of the early
1980s. One factor was the support of the war by the Labour Party, which
had been the main political basis of the movement against cruise missiles.
Only a handful of Labour dissidents opposed the Gulf War. With the Lib-
eral Democrats backing the war effort as well, only fringe parties, such as the
Greens and Plaid Cymru (Welsh Nationalists), and a range of far left group-
_^k ings supported the peace movement. CND remained the dominant player, _^_

| seeking to bring together a range of peace, church, and political groups in

the Committee to Stop War in the Gulf, which formed in 1991. Another
British initiative that gained some media attention was the Gulf Peace Team,
formed in London in 1990, establishing a protest camp with eighty-six vol-
unteers in Iraq near the Saudi border, all of whom were deported ten days
after the war started. In Britain, various protest events did attract thousands
of people, but support for the movement appeared limited to committed
peace activists and backing fringe organizations. Once the war had ended,
activities quickly dwindled.

Spain

The Spanish peace movement in the Gulf War era comprised two currents,
one more pacifist, with a long and successful history of fights against com-
pulsory military service and in support of conscientious objectors, and the
other current consisting of the antiwar movement that engaged in campaigns
against almost any then-current war. The major protest events in those days,
however, were the demonstrations against the Gulf War. Between 1990 and
1991 at least sixty demonstrations were held against the in Madrid alone,
with about 230,000 total protesters.

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 72_^_

KLANDERMANS

Italy

The threat of U.S. intervention in Kuwait generated a new wave of protest
in Italy. After Iraq invaded Kuwait, reinforced by the U.S. ultimatum for
military intervention, large demonstrations took place (Giugni 2004). For
example, on October 7, 1990, a hundred thousand people took part in a
demonstration in Assisi, the starting point of the traditional peace marches.
On January 12, and January 26, 1991, two hundred thousand and two hun-
dred thousand people, respectively, demonstrated in Rome. Furthermore, a
whole range of actions was held before and during the war. Obviously, the
peace mobilization was motivated by the Italian army's involvement in the
conflict. Even more so than the Gulf War, however, the military interven-
tion in Kosovo rallied the Italian peace activists, who staged an impressive
number of demonstrations.

The Netherlands

Surprisingly at least in view of the massive mobilization against the cruise
missiles, there were hardly any protests against the Gulf War in the Nether-
lands. While in Germany on February 17, some 120,000 people took part
in various protests, in the Netherlands no more than fifty protested in The
-Ofr Hague in front of the parliament and 200 in Amsterdam's central square. .fa.

I Altogether, only an estimated 19,000 people took part in various demon-

strations. There were several reasons for the low turnout. First, the Social
Democrats were in government and in favor of military intervention, as were
other major organizations, such as unions; only the small radical left parties
opposed it. Second, the peace movement was internally divided. The large,
church-based Interdenominational Peace Council seconded the Social Demo-
crats, and again only the radical Left was against the war.

Switzerland

In January 1991, The Swiss peace movement staged several demonstrations
against the Gulf War — the largest in Bern, which brought fifteen thousand
participants in the streets. These demonstrations were part of an ongoing wave
of protest regarding a variety of national and international issues organized
by the peace movement. In 1993, a new wave of demonstrations targeted
the war in Bosnia. In December of that year, a coalition of women's and peace
organizations called for a Women against War protest day, which mobilized
about thirteen thousand people in different Swiss cities. The aim of this action
day was to draw attention to the sexual abuse of women in Bosnia. Although
there were a number of demonstrations, compared to that of the early 1980s,

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 73_^_

LEGACIES FROM THE PAST

the size of this antiwar mobilization was quite moderate. Also, the partici-
pation rates at the traditional Easter marches had diminished considerably.

Belgium

In Belgium the protests did not reach the levels they had in the 1980s,
either. The largest demonstration on January 13 brought not more than
twenty thousand participants into the streets of Brussels. This was not in the
least due to discord in the peace movement. In 1990, two main umbrella
organizations of the movement of the 1980s had merged into one organiza-
tion Vlaams Aktie Komite tegen Atoomwapens/Overleg Centrum voor Vrede
(VAKA-OCV; Flemish Action Committee against Nuclear Arms/Consulta-
tion Centre for Peace). However, internal bickering about the distribution of
power between the member organizations immediately led to a split, prompt-
ing several important peace organizations to break away. It took a much
more critical stand toward the government, reproaching it for its lack of
action. This viewpoint was a reaction to the formation of another, even more
radical Maoist peace group. By the end of 1990, the peace organizations that
had left VAKA-OCV rejoined the umbrella to establish an Initiative Group
for Peace in the Middle East, which organized the January 1 3 demonstration.
_^k One week later, the same coalition organized another demonstration; again _^_

| the Maoist peace group was present, which made the two largest labor unions

withdraw from the coalition; soon the traditional-basis militants and the
constituency of the traditional peace groups would drop out as well. By early
February, the umbrella organization could only mobilize a few hundred in
just a few local demonstrations.

The scale of protest in Germany came as a great surprise. Altogether, close
to a million people took part in dozens of different protests and demonstra-
tions. Within a few weeks, there were protests all over the country, despite
that after the major mobilization against cruise missiles in the early 1980s
the peace movement seemed to have declined seriously, as suggested by the
gradual decrease of participation in the yearly Easter marches. This was be-
cause the organizational network of the movement remained operational as
the movement continued to protest various peace and war issues. Moreover,
the political situation in Germany was relatively favorable: the Social Demo-
crats were against the war, as were the unions. At the same, time a coalition
of other new social movements supported the antiwar mobilization; the
high schools were an important recruitment ground for these protests.

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 74_^_

74 KLANDERMANS

Conclusion

Peace movements existed in each of the countries where we conducted our
research; however, their individual histories differed considerably. To be sure,
in all countries, the movements were part of the same cycle: the protests against
nuclear armament (especially cruise missiles) and against the Gulf War.
Indeed, national movements responded to the same international phenom-
ena. But, I observed significant differences between the c
of the actual features of the campaign.

Figure 4.1 compares the two campaigns in the eight
dimensions: the organization of the campaign and the level of mobilization.
A difference between the two campaigns that catches the eye immediately
concerns the mobilization levels; with Spain as the exception, mobilization
levels were high in all countries during the cruise missile campaign, and I
observed significant differences between countries during the anti-Gulf War
campaign. Shifting coalitions are the second marked difference between the
two. Whereas most of the cruise-missile campaigns were organized by a coali-
tion of new social movements, the traditional Left, and the churches, the
latter two were absent in the anti-Gulf War campaign (with the exception of

, Germany). As far as the set-up of the campaigns is concerned, the cruise-

-Ofr missile campaign was always both locally and nationally organized. However, .fa.

I in the anti-Gulf War campaign, there was much less local activity, especially

in those countries where the mobilization levels were low. Finally, countries
differ in terms of how mobilization levels vary between the two campaigns.
On the one hand, I find countries such as the United States, Germany, and
Italy, with high mobilization levels during both. And Switzerland also reveals
stable, though lower mobilization levels over the two campaigns. On the
other hand, however, there are countries such as the United Kingdom, the
Netherlands, and Belgium, where mobilization levels during the anti-Gulf
War campaign were much lower than they had been during the cruise-missile
campaign. The differences in the Netherlands are especially striking. Spain,
finally, is the only country where the mobilization level in the anti-Gulf War
campaign was far higher than it had been during the cruise-missile cam-
paigns; it's worth noting, however, that in those early days, the peace move-
ment in Spain was still in its infancy.

The differences I observed between the two campaigns suggest that one
cannot predict from a previous campaign whether the next will succeed or
fail. In countries where I witnessed high mobilization levels and broad coali-
tions in one campaign, the movement failed to build viable coalitions and
realized negligible protest activity in the other. Indeed, from the one cycle

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 75_^_

". G .5* &"2 o S3 ft § -s S

i "i II 11 ft
II 1 1 |'| ||

-3.S- E«-

: S o §, 3 1- 3 o ' l -

§ 3 l-i

1 1 1§

I ill I III §. I J"j

H> .s'l IC "3 I 'I "3 el "3 j

e a .2 s g^ i ^ =

> a £ 3 | 3 « -a [

§ g/a -a c ~5 -a § ;;

I

H$H

-4

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 76_^_

76 KLANDERMANS

of protest to the other, networks and coalitions had to be reanimated. A strong
coalition in the one cycle does not guarantee that an equally strong coalition
in the next, nor does a high mobilization level in one cycle guarantee high
mobilization levels in the next. As Taylor (1989) maintains, movements go
into abeyance when the tide turns and must be reactivated and remobilized
for the next wave. Obviously, the conditions that prevail when and where
this revival takes place determine the form of the new campaign. This implies
that the characteristics not so much of the peace movement alone nor of the
general political opportunity structure, but rather the time- and space-specific
opportunities and the issue-specific context (see chapter 3) determine what
coalitions can be built and what mobilization levels can be achieved. A com-
plex interplay of general and country-specific factors accounts for the diverse
appearance of the movement at various points in time and space. The unique
opportunity offered by the demonstrations of February 1 5 is that they pro-
vide us with the possibility to disentangle the processes behind this diversity.

Notes

period the movement in the United States concentrated on nuclear freeze. As we will
see below, the two campaigns did not really collaborate.

2. Much of this section is based on Klandermans (1991a).

3. We are grateful to David Meyer for his help with the U.S. sections.

4. We are grateful to Ramon Adell for his help with the Spanish sections.

5. We are grateful to Michelle Beyeler for her help with the Swiss sections.

6. The GSoA was founded in 1982 to launch a popular initiative for the abol-
ishment of the army. Since then it has picked up a range of different issues and be-
came a central player in mobilizing for peace demonstrations.

7. We are grateful to Stefaan Walgrave for his help with the Belgian sections.

Works Cited

Brittain, Victoria. 1991. The Gulf between Us: The Gulf War and Beyond. London:

Carter, April. 1992. Peace Movements: International Protest and World Politics since
1945- London: Longman.

Cooper, Alice, and Klaus Eichner. 1991. "The West German Peace Movement." In
Peace Movement in Inter,, wcial Movement

Research?), ed. Bert Klandermans, 149-74. Greenwich, Conn.: JAI Press.

Downton, James Jr., and Paul Wehr. 1998. "Persistent Pacifism: How Activist Com-
t Is Developed and Sustained." Journal of Peace Research 35: 531-50.

H$H

04 Chapter 4_Walgrave 22/01/2010 3:37 PM Page 77_^_

LEGACIES FROM THE PAST 77

Everts, Philip P., and Guido Walraven. 1984. Vredesbeweging. Utrecht: Spectrum.

Giugni, Marco. 2004. Social Protest and Policy Change: Ecology, Antinuclear, and Peace
Movements in Comparative Perspective. Lanham, Md.: Rowman & Littlefield.

Klandermans, Bert, ed. 1991a. "The Peace Movement and Social Movement The-
ory." In Peace Movement in International Perspective. International Social Move-
ment Research 3, ed. Bert Klandermans, 1-42. Greenwich, Conn.: JAI Press.

. 1991b. Peace Movement in International Perspective. International Social Move-
ment Research 3. Greenwich, Conn.: JAI Press.

Kleidman, Robert. 1993. Organizing for Peace: Neutrality, the Test Ban, and the
Freeze. Syracuse, N.Y.: Syracuse University Press.

Koopmans, Ruud. 1999. "Globalization or Still National Politics? A Comparison of
Protests against the Gulf War in Germany, France, and the Netherlands." In
Social Movements in a Globalizing World, eds. Donatella della Porta, Hanspeter
Kriesi, and Dieter Rucht, 57-70. London: Macmillan.

Kriesi, Hanspeter, Ruud Koopmans, Jan Willem Duyvendak, and Marco G. Giugni.
1995. New Social Movements in Western Europe: A Comparative Analysis. Min-
neapolis: University of Minnesota Press.

Marullo, Sam. 1992. Endingthe ColdWar at Home: From Militarism to a More Peace-
ful World Order. New York: Free Press.
| Oegema, Dirk. 1991. "The Peace Movement in the Netherlands." In Peace Move-

*9)~ ment in Int. vement Research 3, ed. ^XJ

' Bert Klandermans, 93-148. Greenwich, Conn.: JAI Press. '

Rochon, Thomas R. 1988. Mobil: merits in West-

ern Europe. Princeton, N.J.: Princeton University Press.

Ruzza, Carlo. 2004. "Peace Movements." In Democrat Popular Protest

and New Social Movements, eds. Gary Taylor and Malcolm Todd, 290-306.
London: Merlin.

Stouthuysen, Patrick. 1995. "De Vredes- en antiracismebeweging, de ontmoeting
van oude en nieuwe sociale bewegingen." In Van "Mei 68" tot "Hand in Hand, "
eds. Staf Hellemans and Marc Hooghe, 69-88. Leuven, Belgium: Garant.

Taylor, Verta. 1989. "Social Movement Continuity: The Women's Movement in Abey-
ance." American

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 78_^_

5
New Activists or Old Leftists?:
The Demographics of Protesters

Stefaan Walgrave, Dieter Rucht, and Peter Van Aelst

This chapter analyzes the sociodemographic profile of the February 15 dem-
onstrators. Who are they in terms of age, sex, education, social class, and
religion? Since this cannot be answered without a comparative yardstick, we
| can narrow down our quest to the specificities of the February 15 protesters

~W when compared to other social groups. In more precise terms: Are the peace ~W

protesters typical new social movement supporters? Are they emblematic
Old Left activists? Or do they, in contrast, mirror the population as a whole
and, as such, represent an example of protest normalization? The compara-
tive design of the book begs for an additional descriptive question: To what
extent are the protesters' sociodemographic profiles different in the eight coun-
tries under study? The answer to this permits us to make headway with a
causal question: How can we account for the differences in demonstrators'
profiles in these countries? After all, these people demonstrated against the
same issue on the same day, shouting the same slogans, carrying the same
billboards, and relying on the same action repertoire.

The sociodemographic makeup of protest events may vary considerably.
Sometimes, protesters are rather homogenous. This is true, for example, when
distinct social groups, such as farmers, defend their economic existence by
means of protest. In other cases, protest groups express broad concerns that
potentially affect large segments of the populace, if not humankind as a whole,
for example, human-induced climate change or the risk of nuclear war. In
these cases, we can expect the protesters to be recruited from different social
backgrounds and to hold diverging ideological beliefs. As broad and hetero-
geneous as such protests may be — it is unlikely that they are representative

H$H

-4

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 79_^_

NEW ACTIVISTS OR OLD LEFTISTS? 79

of the population as a whole. From many surveys, we know that in the aggre-
gate people who are politically active — and more specifically those who par-
ticipate in protest actions — tend to be younger, better educated, and male
(Norris 2002). Of course, the degree of deviation from the average popula-
tion does vary greatly depending on the issue at stake. Also, the form of protest
has an impact on recruitment. It is well known that in violent protests young
men are strongly overrepresented. The organizers of protests, as well, may
have a differential effect on recruitment patterns. We can expect that pro-
testers following the call of broad alliances of diverse groups will be more
heterogeneous when compared to those organized by radical groups from the
political fringe. In addition, the location and timing of a protest action may
have an impact on its social composition; for example, apart from strikes,
workers rarely protest during the week, and elderly people do not tend to
travel long distances to participate in protests. Finally, we can also expect that
the tools and channels for mobilization affect the composition of the par-
ticipants: calls via the Internet will not, or will only indirectly, reach those
not hooked to the net (Internet users tend to be young and well educated).
In sum a set of structural and situational factors determine recruitment.
General assumptions and predictions are likely to be inadequate concerning
specific protest events.

First, in terms of the February 15 protest in 2003 against an imminent
war in Iraq, we can reasonably hypothesize the sociodemographic profile of
the protesters. Relative to other issues that directly pertain to specific social
groups, this is a matter that affects people in moral terms without having
immediate consequences on their daily lives. Hence, we expect protesters to
be relatively heterogeneous regarding age, sex, education, social class, and
religion. With regard to past peace protests, however, this general expectation
must be differentiated. From many studies on peace movements and peace
protests in the second half of the twentieth century, we know that partici-
pants tend to be male and younger and better educated than the rest of the
population (Norris 2002). Chapter 4 in this volume focuses on two previous
waves of peace protests and makes this point regarding the eight countries
in this study; demonstrators against cruise missiles in Europe in the 1980s and
protesters against the Gulf War in the 1990s were highly educated and typi-
cally middle class. As far as the period since the student revolt in the 1960s is
concerned, peace protesters in Western countries in general share the specific
features of what characterizes the activists and constituents of the so-called
new social movements. Young, well-educated people from the human-service
sector with liberal or leftist attitudes are strongly overrepresented among new
social movement constituencies. Compared to the population as a whole, this

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 80_^_

WALGRAVE, RUCHT, AND VAN AELST

i is predominantly male-dominated. Women are better represented,
though, compared to other kinds of political activism (e.g., within parties).
Previous peace activism research, thus, leads us to expect that February 15
activists are rather heterogeneous, as we may have anticipated, but still be
dominated by the "usual suspects."

Second, the story of February 15, is more complicated. Chapter 1 showed
that the events were coordinated and staged by an international network of
movement organizations, most of which originated within the so-called global
social justice movements. It was on the European Social Forum meetings of
these movements that the protest was set up and organized. Although the
global justice movements are closely associated with the typical new social
movement sector, they explicitly attempt to bridge these movements with the
Old Left, that is, the labor movement and the traditional left-wing political
parties with their working-class supporters (male, older, lower degree of edu-
cation, active in traditional industrial sectors). In many countries, indeed,
labor unions and traditional left-wing parties engage in the struggle for global
justice; they are represented in the movement's central agencies and contrib-
ute substantially to the movements' mobilization potential (see chapter 8 con-
cerning parties and chapter 10 concerning unions). Especially when it comes
n, the Old Left shows a significant track record, at least ii

| to peace activism, the Old Left shows a significant track record, at least in

~W Western Europe; in the United States, the peace movement had weaker links ~W

vith the Old Left. Chapter 4 elaborates this in more detail. Pre- 1960s peace
protest in Western Europe was basically carried by the Old Left and also dur-
ing the big marches against cruise missiles in the early 1980s labor unions
and left-wing parties did their part. As chapter 4 shows, in some of these
countries it was the Old Left's mobilization machinery — wherever it was in
the opposition — that contributed to the impressive peace protest wave of the
1980s. More than the environmental movement or the third-world move-
ment — two other emblematic examples of the new social movements — in
many countries the peace movement has been a coalition movement focused
on mass mobilization and drawing support from a broad range of movement
organizations. The movement's organizational backbone was most of the
time diverse, with different organizations (temporarily) joining forces to fur-
ther peace protest. In sum: contemporary and historical considerations lead
us to expect that the February 15 protesters will not only consist of the typ-
ical new social movement constituent but might also be characterized by a
strong presence of the Old Left. Both hypotheses boil down to the expecta-
tion that these protesters will not be representative of their respective popu-
lations but will consist of very specific segments of them.

However, in many countries, the political constellation regarding the war

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 81_^_

"4

NEW ACTIVISTS OR OLD LEFTISTS?

on Iraq — in particular that, as shown in chapter 3, in many countries oppo-
sition to the war was not confined to the Left but was dispersed throughout
the whole political landscape — suggests the opposite hypothesis: February
15 brought a fairly representative sample of citizens to the streets. As we saw
in chapter 1, many media accounts emphasized the protesters' high internal
diversity and portrayed them as coming from a wide range of political and
social backgrounds. This corresponds with the thesis of "normalization of
protest" that has gained scholarly support since the late 1990s. It states that
protest is becoming ubiquitous, just another strategy employed by citizens
to defend their interests, and that there is an ongoing evolution of increas-
ing diversity on the streets (Fuchs and Rucht 1994, Van Aelst and Walgrave
2001). The context in which the February 15 protest took place in many
countries might indeed have been conducive to such normalization. First,
this foreseeable war against the Iraqi regime was highly disputed among the
leaders and populace in the Western world and beyond. That — according to
various surveys — an overwhelming majority in most European countries
opposed the war implied that the mobilization potential was not only large
but also included people from different social strata and with different ide-
ological leanings (chapter 3). While antiwar mobilizations during the second
half of the twentieth century mobilized politically progressive groups, whereas
the mainstream was either undecided or tended to be on the other side, the
recruitment pool for this particular antiwar mobilization was extremely
large and thus, almost by definition, very heterogeneous in terms of its socio-
demographic profile. Moreover, that quite a few national governments were
skeptical or even frontally opposed to war leads us to expect a large degree of
heterogeneity among the protesters. In such a situation, protest was less likely
to be perceived as outsiders challenging the establishment; both opponents
and supporters of incumbent governments could be expected to participate.

Thus, three contradicting heuristic hypotheses can be established: first,
February 1 5 was, in terms of the participants' sociodemographic profile, a
typical new social movement event (young, better-educated people working
in the human services sector); second, it was a typical Old Left event (older,
less schooling, from industrial sectors); and third, it was neither of these but
rather a case of large internal diversity and normalization.

The comparative design of the present study might lead us to conclude
that all three hypotheses are true at the same time, though in different coun-
tries. The context in which the protests took place, in fact, differed extensively
among countries. Chapter 3 showed that there were substantial differences
among the eight nations in terms of the political stance of the incumbents
and the opposition, in public opinion vis-a-vis the war, and in the way the

H$H

-4

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 8 ? afc

82 WALGRAVE, RUCHT, AND VAN AELST

war was covered in the mass media. Chapter 2 established that the a
under study differ not only vary regarding issue-specific aspects but, more
fundamentally in terms of their openness to challengers and protest and the
strength of what was called the "progressive movement sector." For example,
the degree of pacification of the labor issue varies across countries, and so does
the strength of the new social movements. All these factors can be expected
to affect the demonstrators' sociodemographic profile.

Who Are the February 15 Participants?

Bivariately analyzing the composition of the February 1 5 demonstrators, we
see that men and women were present almost equally in the antiwar demon-
strations (see Table 5. 1). In most countries women were slightly in the major-
ity. In Belgium, however, there was a striking overrepresentation of men. On
the other side of the spectrum, In the United States, two- thirds of the protest-
ers were female. In Italy, Germany and, especially, Switzerland, the number
of young demonstrators is much higher than the average. In these countries,
about one-third of the participants were students. In Spain and in the United
States, the youngest cohort is underrepresented: only one out of ten partic-
ipants was a student. The U.S. demonstrations counted as many young peo-
ple as they did those over sixty-five, which is quite unusual for a protest
march. One thing is very clear for all of the countries we studied: the aver-
age antiwar protester is highly educated. In almost every case, the category
of participants with a university degree is the largest. Again, Switzerland and
the United States are the most extreme cases. Switzerland has a large num-
ber of demonstrators with lower educational degrees, many of these being
young people yet to receive high school diplomas. The United States, how-
ever, had a spectacular 93 percent of demonstrators with higher education
(non-university or university). The professional categories are more difficult
to compare. The Spanish protesters differ from the other countries because
far more of them are office and manual workers. The demonstrators over-
whelmingly worked in health, education, care, and research, and, to a lesser
extent, private services. Only a modest number of people worked in the in-
dustrial sector. Of course, these sociodemographics are correlated (not shown
in table). Female protesters, for example, have slightly higher educations, are
younger, and are more active in the service sector than men. It is not sur-
prising that the younger people among the demonstrators, those who had
not yet completed their studies, have a significantly lower degree of educa-
tion than other protesters.

Table 5.1 reveals that in all countries every major group in society was
represented to some extent. The February 1 5 protests unquestionably attracted

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 83_^_

1

<:

o

-_.

<D

I

*

<D

c

1

o

Jc

O)

'o

z

S2

£

o

i

o

-

<D

"O

if)

>.,

55

o

3

Jo

ui

o

£

S

J

3

Q.

s

cn

o

E

<D

"§

1

<§

to

<u

_Q

£

I £ d!Qi?S I^Sz^ lolllld ScSi "ouo

I

H$h-

-4

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 8 4 afc

84 WALGRAVE, RUCHT, AND VAN AELST

people from all pockets of society and can, overall, be considered heteroge-
neous events. Yet the table also illustrates that the internal diversity of the
protest differed from country to country. To account for these differences in
internal diversity and to synthesize this diversity in one measure, we calcu-
lated fractionalization indexes (Table 5.2). A fractionalization index conveys
the chance that two randomly drawn individuals fall in two different cate-
gories of a variable. The higher the index, the more diversity there is. Since
the index is dependent on the number of categories of a variable, it cannot
be used for comparisons between variables (with different categories) but
merely between countries.

Table 5.2 shows notable differences among countries. The U.S. protest-
ers, for example, are the least diverse concerning sex and education. Those
from Spain are least diverse in terms of age, those from Belgium in terms of
profession, and those from Switzerland are most likely to come from the same
economic sector. Aggregated internal diversity, averaging the five fractional-
ization indexes for each country, is more or less similar across countries, though
some differences exist. The U.S. and UK demonstrators seemed to have been
the most homogenous, leading to the lowest fractionalization indexes, while
the Swiss and Dutch were substantially more diverse internally.

Only by comparing the sociodemographic characteristics of the pro-
testers with those from the population in general can we assess whether the
February protesters were a fair sample of their respective populations. In the
United States, for example, protesters' educations may be higher simply be-
cause U.S. citizens in general are better educated than their European coun-
terparts. We relied on demographic statistics published by the Organisation
for Economic Co-operation and Development and the United Nations for
sex, age, and education and calculated differences in proportional presence
of the categories in Table 5.3.

Large proportions of men and women protested against war, with the
United States featuring a majority of female and Belgium a majority of male
protesters. In general, we find a slight overrepresentation of women, which
corresponds with previous findings that women have since long started to
catch up with their male counterparts in terms of protesting (Jennings and
Van Deth, 1990). All but the oldest age categories were overrepresented. Youth,
however, is not more overrepresented than young adults and adults. Younger
groups are most overrepresented in Italy and Switzerland, while older groups
were most represented in the United States. Reasons for the globally endur-
ing underrepresentation of older people on the streets are multiple: people
in their seventies probably have less protest experience (generation effect)
and have to overcome more physical barriers to join protest marches (age

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 8 5 afe

"4

3 _
o —
J

Q.

E

£

8 *
J 3

f 3 II £ z;

H$h-

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 86_^_

<t

S2

Q

£

<G

o

Q.

O

'o

c

<D

1

S

<D

□

-c

T5

1

g

i

z

J5

-

g

3

"O

0)

1

%

(D o

^f

x" °

^

« 0)

3

i- £

*e i

jj

1 c

3

S'c
!6 .2

£ 3

fc Q.

-S 8.

§T>

-o £

l-8

4> -O

<s §

co

to

V

_Q

|H

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 8 7 afc

NEW ACTIVISTS OR OLD LEFTISTS?

effect). A comparison with the population in terms of education shows clearly
that protesters in all countries had higher educations than the average citi-
zens. Individuals with tertiary (university and non-university) education de-
grees were especially strongly overrepresented among the demonstrators. In
terms of schooling, hence, the antiwar protesters are hardly representative,
least so in the United States and the United Kingdom. Averaging demon-
strators' representativity, we calculated average differences between the dem-
onstration and corresponding population categories in every country: the
lower this figure, the more demonstrators resemble their populations. Pro-
testers in the United States, the United Kingdom, Italy, Spain, and Belgium
were less representative of their respective populations at large than the those
in the Netherlands, Switzerland, and Germany.

In sum: people protesting imminent war were not representative of the
populations of their countries. They were somewhat younger, slightly more
female, and especially much better educated than the average citizen. This
hardly is surprising; social movements defend certain interests and are rooted
in corresponding population segments. Social movements that are able to
mobilize true cross-section of the population have yet to be invented. Since
movement support is not evenly present in all population segments, it is no
revelation that neither is movement participation evenly present. A n

| revelation that neither is movement participation evenly present. A movement

~W can only mobilize people who agree with its goals, and supporting the move- ~W

ment in word (attitudes) is something other than supporting it in deed (Klan-
dermans 1984). Our next question then is this: Were the demonstrators at
least exemplifying the war-opposing segments in their respective populations?
To compare antiwar activists with antiwar supporters in the respective
populations, we need population data. We draw upon a population survey
carried out by European Omnibus Survey (EOS) Gallup Europe between
January 2 1 and 27, 2003, only a few weeks before the February 1 5 protests
actually took place (EOS Gallup Europe 2003). All seven European countries
covered in this book are included in the poll; the United States, unfortunately,
is not. Chapter 3 presents the general results of this survey, suggesting mas-
sive antiwar sentiment among European citizens. In this chapter, however,
we are only interested in the sociodemographic profile of these war resisters.
To compare the populations antiwar segments sociodemographics with the
features of the movement activists, we must first determine movement sup-
port. What was the mobilization potential? What was the "official" stance of
the peace movements? This is difficult to determine. Peace movement officials
would probably reject war in any case. Since many EOS Gallup Poll ques-
tions were hypothetical and since some questions concerned motives for war
and not the justness of war, we decided that opinions about U.S. intervention

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 88_^_

"4

WALGRAVE, RUCHT, AND VAN AELST

in Iraq without UN resolution best captured the stake and aim of the anti-
war protest. We are aware that relying on one single question to delineate
mobilization potential is a problem, as it artificially reduces the scope of the
movement. In the seven European nations of this study no less than 79 per-
cent of the populace rejected war without UN approval. Consequently we
narrowed the movement's mobilization potential down to the people who
completely disagreed with an intervention without UN backing and consid-
ered this still very large group of 54 percent as the mobilization potential of
the movement. Who were those people?

The mobilization potential in the seven countries is characterized by a
slight overrepresentation of women; a remarkably older composition, with the
fifty- five and older group being the largest; having higher education although
not being overeducated; relatively small shares of professionals and manual
workers; especially high shares of non-professionally active citizens; and, finally,
citizens with clearly left-leaning political stances. Thus, at first glance, the
antiwar supporters look like the actual protest participants. Table 5.4 con-
tains the differences between these groups based on the subtraction of their
proportional presence. A negative figure means that this group is underrep-
resented among the actual protesters: this category was more present among
the antiwar supporters than among actual movement participants. Hence,
negative figures suggest a mobilization deficit: more people of this category
supported the movements claim than actually showed up at demonstrations.
Positive figures mean the opposite, namely that relatively more activists than
supporters were present.

The results in Table 5.4 must be interpreted with caution, since some
EOS-Gallup categories did not completely match ours. In addition, the Gallup
Poll samples were small and therefore prone to random errors. Only large
differences merit our attention. The general picture is clear and confirms the
literature about mobilization biases and political participation thresholds.
Advanced age seems to be a very tough barrier for mobilization: in the old-
est groups, it was not successful. Low education, too, is a formidable barrier;
people with a higher degree of education are strongly overrepresented, and
all other categories are underrepresented. Since the profession categories were
recoded on a less reliable base and the differences do not seem that large, we
must be extremely careful when interpreting these results in particular. The
gender factor did not have much impact either. The most striking mobiliza-
tion deficit, though is among right-leaning people. A considerable amount
of them supported the antiwar claim; yet they are very strongly underrepre-
sented at the actual protest events. This suggests that mobilization for Febru-
ary 1 5 mostly targeted left-leaning people and that supporters from the Right

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 89_^_

"4

! ^SS 2 III Suii

I 1 ! =

a- I

h$h

-4

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 90_^_

90 WALGRAVE, RUCHT, AND VAN AELST

were not reached and/or attracted. For a more in-depth analysis of the mobi-
lization process of February 15, we refer to chapter 9.

People protesting against imminent war on Iraq may not have been rep-
resentative of the entire protest potential, but they may have been typical for
peace protesters in general. Therefore, we can compare their sociodemo-
graphic profiles with those of people active in previous peace protests. Unfor-
tunately, past major and cross-national surveys on specific peace protests are
not available. However, some population surveys have been carried out in
various countries and at various times — surveys that included questions on
participation in or support of peace movements. While most of these can-
not be compared across countries because the wordings of the questions
differ, four Eurobarometer surveys, conducted in 1982, 1984, 1986, and 1989
in Belgium, the Netherlands, Germany, Italy, and Britain, can be used for
our purposes. Although these surveys serve as a reasonable basis for compar-
ing peace activists in the 1980s and in the early twenty-first century, one
methodological caveat should be underlined. In 2003, we surveyed protest-
ers on the spot, regardless of whether they considered themselves part of a
peace movement. In the Eurobarometers, people were not approached as par-
ticipants in a specific peace protest but were asked questions, usually in their
homes, about their relationship with the peace movement (along with sev-
eral other movements). The survey asked whether people approved the "anti-
war and anti-nuclear weapons movement" and whether they were members
of or might join it. We classified those respondents as actual activists who
approved strongly or somewhat and considered themselves members of such
a movement. In addition, we classified as potential activists those who ap-
proved strongly or somewhat and said they might join the movement. We
compared the sociodemographic characteristics of actual activists (usually low
numbers) and potential activists (higher numbers) on a country-by-country
basis, combining the four Eurobarometer surveys. Because the results for
activists and potential activists were similar, we decided to put the two groups
into one single category. Table 5.5 compares these 1980s potential and actual
peace activists with 2003 actual peace activists.

The comparison yields substantial differences between the 1980s (poten-
tial and actual) and 2003 actual peace activists. Considering gender, there is
no univocal pattern, but the divergences in education and age are flabber-
gasting. Most remarkable, especially in the United Kingdom and Belgium,
is the great proportion of highly educated protesters in 2003 compared to
the earlier peace activists. This tremendous difference cannot be explained
sufficiently by the growing proportion of highly educated people in the inter-
vening two decades. It is likely not coincidental that the distribution of age

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 91_^_

"4

.c

£

I

8

i2

£

5

i

1

1

i

_2

o

s

°

S

1

o

,c

8

J2

§

£

O

1

1

E

1

E

8

-S

8

I

-|

1

z

£

1

^

o

•■=

E

°

%

f

6

11

^ s

j J,

~o CO

31

«

2

jO

i£ 3'^x £%o 3uS

H$h-

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 92_^_

WALGRAVE, RUCHT, AND VAN AELST

groups among the 1980 and 2003 peace activists differs strikingly; in all five
countries, the proportion of young peace activists dropped considerably Per-
haps many peace activists of the 1980s, who then were predominantly com-
ing from the young and middle generations, became reactivated in 2003 —
altering that year's age balance in favor of the middle and older groups. Our
data support this interpretation to some extent; there is a linear and signifi-
cant relationship between age and previous peace activism. The older Febru-
ary 15 protesters were much more likely to have participated in previous peace
protests. Another explanation, of course, is that we compare very different
groups from the 1980s and from 2003. Potential activism is something com-
pletely different than real activism.

Until now we have only indirectly answered the central question under-
lying this chapter. We have not given a comprehensible answer to the ques-
tion of how the February 15 demonstrators can be typified best: as the usual
new social movement suspects, as the typical Old Left activists, or as a rep-
resentative sample of the population? All analysis so far clearly shows that
February 1 5 may have been internally diverse, but that demonstrators were
not at all a fair sample of the population. Most analysis also suggests that
February 1 5 was, above all, a classic new social movement event with a high
| proportion of women and an overrepresentation of relatively young, highly

~W educated people working in the service economy. Although supported by ~W

the traditional Old Left in many countries, as chapter 3 shows, our antiwar
demonstrators globally did not share the typical characteristics of the Old
Left activists. Protesters who were male, older, less educated, and industrial
workers were strongly underrepresented. Unfortunately, a direct comparison
between other new social movement and Old Left protest activists based on
surveys similar to the one this volume draws upon does not exist, since sur-
veying people at demonstrations is relatively new. However, we have at our
disposal a similar dataset containing participants in six other large demon-
strations staged in Brussels, Belgium (Van Aelst and Walgrave 2001). The
Belgian surveys covered most of the biggest demonstrations held in Brussels
during the 1998-2001 period, which were staged by typical Old Left agen-
cies, new social movements, and difficult-to-classify organizers. The profile
of the Belgian February 1 5 demonstrators is clear: they are typical new social
movement protesters, strongly resembling, for example, the antiracist and
global justice demonstrators while systematically differing from all other dem-
onstrators in the Belgian dataset. In particular, the similarity between the
February 1 5 and antiracist activists corresponds strikingly. These Belgian data,
thus, strongly suggest that February 1 5 was a typical new social movement
event. Of course we cannot generalize this finding based on evidence for a

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 9 3 afc

NEW ACTIVISTS OR OLD LEFTISTS? 93

single country. The evidence presented earlier, though, suggests that the Bel-
gian February 1 5 demonstrators certainly were no outliers but resembled the
antiwar protesters in the other seven countries. Though this is not conclu-
sive proof, it corroborates our basic rinding that February 15 was not, as mas-
sive and impressive as the event may have been, extraordinary in terms of
the kind of people who participated. Rather, demonstrators resembled the
typical new social movements constituents in advanced industrial democra-
cies at the beginning of the twenty-first century.

How to Account for the Differences among Countries?

Having higher degrees of education, being younger and female, working more
in the health, education, care, and research sectors, the February 1 5 demon-
strators were not representative of their respective populations. As stated, they
very look alike the typical new social movement supporter. This very general
conclusion, however, has to be more nuanced when one turns to the demon-
strators in the eight countries under study. Although they had much in com-
mon, we detected differences among countries. All demonstrated on the same
day, for the same reason, making the same claim, but they did not share the
same sociodemographic characteristics. Why? The whole range of compar-
| isons pursued above shows that, ultimately, countries do make a difference.

~W Variations among countries cannot be accounted for simply by differences ~W

in population compositions or dissimilarities among the war-opposing seg-
ments in their publics. We will rely here on two of the factors introduced in
chapter 3: the eight countries' political constellations regarding the Iraq issue
and public opinion vis-a-vis war. The phenomenon to be explained here is
the degree of diversity of the demonstrators in the eight nations in contrast
to the degree of their correspondence to the typical new social movement
activist. Since diversity is easier to measure, we can reformulate our explana-
tory question as follows: Why was there more diversity among February 15
protesters in some countries than in others?

Tables 5.2 and 5.3 best capture the differences in diversity; they contain
comparative fractionalization indexes and deviations of antiwar protesters
compared to the entire population. Both tables roughly point in the same
direction: the Netherlands, Switzerland, and Germany witnessed more inter-
nally diverse protest than the United Kingdom, Italy, and the United States,
though there is no clear-cut dichotomy between these two groups of coun-
tries. Belgium and Spain show a less consistent pattern, with Tables 5.2 and
5.3 contradicting each other. The evidence, hence, suggests that the politi-
cal stance of the government, ami- or pro-war, might have played a role in
determining the protesters' profiles: in countries with governments explicitly

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 94_^_

94 WALGRAVE, RUCHT, AND VAN AELST

and strongly supporting the Iraq War, turnout is less diverse and dissembles
the population as a whole. This makes sense: participating in antiwar protest
in these circumstances might be considered as opposing the domestic gov-
ernment and challenging the incumbents, which might deter government
supporters from participating even if they do not support the government
on this particular issue.

Remarkably the two most internally diverse countries were the most dif-
ficult to classify in our crude and- or pro-war scheme: the Dutch government
was only halfheartedly in favor of war — it had dissolved, and new govern-
ment negotiations were underway — while the Swiss government was only
weakly it. The Netherlands and Switzerland certainly witnessed the most
heterogeneous crowds of all eight countries, given their highest fractionaliza-
tion, which pointed to the most absolute diversity (Table 5.3), the lowest
deviation showing large resemblances to the population as a whole (Table
5.3), and with the smallest mobilization deficit (Table 5.4). This is all the
more striking since the level of protest in these countries was relatively small,
as each had only modest turnouts, as chapter 1 reveals. This suggests that,
apart from the simple pro-war/antiwar scheme, a complementary explana-
tion might be valid: if governments do not express themselves clearly but
| take an intermediate position, participating or not participating in the dem-

~W onstration can be considered neither challenging nor supporting govern- ~W

ment. Consequently, opposition and government parties could mobilize for
the demonstration, bringing a highly diverse crowd to the streets.

How are peoples attitudes linked to the government's stance and, sub-
sequently, to diversity? Unity or discord among the population might play
a role. Especially when a country actively participates in a war and bears the
burden of war, people tend to be divided: patriotic feelings are in balance
with antiwar attitudes. This might lead to polarization. In such a situation,
it is not likely that a diverse constituency takes to the streets, because resis-
tance and support for the war are not randomly spread over the population
but concentrated in certain population segments. This certainly was the case
in the United Kingdom and the United States. Chapter 3, indeed, makes it
clear that — although largely opposing the war — UK public opinion was more
divided and much more in favor of the war than all the other six European
countries under study. U.S. public opinion supported the war but was divided
as well. So government position determines the protesters' profiles directly
by deterring or not deterring government's supporters, and indirectly, by
means of its polarizing effect on public opinion. This might be an explana-
tion for the more homogeneous protesters' profiles in the United States and
the United Kingdom.

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 95_^_

NEW ACTIVISTS OR OLD LEFTISTS? 95

Finally, the presence or absence of elite allies or the mobilizing role of
oppositional parties (and civil society) challenging the incumbent war sup-
porting party may also have a role in the population's relation to the gov-
ernment stance on the war. The question is whether the political opposition
is willing and able to mobilize against the governments support of war. Abil-
ity depends on the strength of the party's own mobilization machinery
while willingness depends on the party's intent to link the antiwar issue with
a more general criticism of the government. What happens if the opposi-
tion, in this case left-wing, goes ahead full force with mobilizing against war
and, thus, against government? The issue becomes yet another battlefield
between incumbents and challengers; it loses its capacity to unite people of
different political and societal leanings and, as a result, the people showing
up are not diverse but correspond to the constituency of the (left-wing) oppo-
sition. In other words: the war issue primarily becomes domestic politics,
with the usual cleavage between government and opposition parties. There-
fore, the Spanish population's relative nondiversity is due not to disagreement
within the majority of the populace but the mobilizing role of the opposi-
tion and the resulting reinforced abstention of government supporters from
protest. Note that in the United States, because its administration initiated
| the war and because the country would bear its costs in human lives, the

~W Democrats — the challenging party — did not take the lead of the antiwar pro- ~W

test but rather decided to rally around the flag (see chapter 3).

In summary, we believe three factors to be crucial in explaining the diver-
sity of the protest: government loyalty stopping some people from protest
against the government they support, division among the populace, and the
(in) capacity of intra- und extra-parliamentary opposition to reach beyond
its own constituency. If government loyalty is activated, public opinion is
divided, and the oppositional forces are unable to bridge traditional cleav-
ages, diversity will be lowest. Diversity is largest if government loyalty is only
latent, public opinion is converging, and the opposition is not perceived as
the driving force to mobilization. All three factors are ultimately caused by
the initial position vis-a-vis the war taken by the incumbents. Governments
could take four different positions regarding the war on three dimensions:
clear or unclear position, supporting or the opposing war, supporting in words;
supporting in deeds. Table 5.6 summarizes this logic.

The Netherlands and Switzerland belong to the first type, Germany and
Belgium to the second, Spain and Italy to the third, while the United States
and the United Kingdom represent the fourth type. The United Kingdom
is more difficult to classify, however, since there clearly was oppositional mobi-
lization, though situated within the governing Labour Party.

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 96_^_

"4

96 WALGRAVE, RUCHT, AND VAN AELST

Table 5.6. Government position and intermediary factors leading to
different levels of protest diversity

Government position Intermediary factors = Diversity

(1) Unclear Division in public opinion

Active government loyalty - High

Mobilizing opposition -

(2) Clear: Opposing war Division in public opinion -

Active government loyalty - Moderate

Mobilizing opposition -

(3) Clear: Supporting war Division in public opinion -

with words Active government loyalty + Low

Mobilizing opposition +

Division in public opinion +

Active government loyalty + Low

Mobilizing opposition -

Conclusion

We started with three heuristic hypotheses about the socio demographic pro-
file of the protesters on February 15. Are they largely diverse, thus coming
close to representing the overall population? Do they resemble the Old Left
constituency? Or are they emblematic of the new social movement adher-
ents? Based on various ways of comparing the protesters with other groups,
among which are potential peace activists and participants in other kinds of
protests, we conclude that the new social movement pattern strongly pre-
vails. Contrary to the newspaper reports cited in chapter 1 and the claims
by many organizing groups, the demonstrators did not mirror the popula-
tion at large. Still, we identified differences among countries that, as one could
expect, can be attributed to the specific political context, for example, the
position of the government toward the war and the reaction and composi-
tion of the oppositional forces in a given country. Thus, the very fact of the
identical point of reference and the similarity of claims basically brought to-
gether the same kind of people in the countries under study, though national
:t modified their profile.

Works Cited

EOS Gallup Europe. 2003. International Crisis Survey, January 2003. www.eosgallup
europe.com/int_survey/index.html.

Fuchs, Dieter, and Dieter Rucht. 1994. "Support for New Social Movements in Five
Western European Countries." In A New Europe? Social Change and Political Trans-
formation, ed. Chris Rootes and Howard Davis, 86-1 11. London: UCL Press.

H$H

05 Chapter 5_Walgrave 22/01/2010 3:39 PM Page 9 7 afc

NEW ACTIVISTS OR OLD LEFTISTS? 97

Jennings, M. Kent, and Jan Van Deth. 1990. Continuities in Political Action. A Lon-
gitudinal Study of Political Orientations in Three Western Democracies. Berlin: Wal-
ter De Gruyter.

Klandermans, Bert. 1984. "Mobilization and Participation: Social-Psychological
Expansions of Resource Mobilization Theory." American Sociological Review 49,
583-600.

Norris, Pippa. 2002. Democratic i al Activism. Cambridge:

Cambridge University Press.

Van Aelst, Peter, and Stefaan Walgrave. 2001. "Who Is That (Wo)man in the Street?
From the Normalisation of Protest to the Normalisation of the Protester." Euro-
pean Journal of Political Research 39: 461-86.

m

h$h

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 98_^_

Peace Demonstrations or Antigovernment Marches?:
The Political Attitudes of the Protesters

Bert Klandermans

m

Why did so many people bother to demonstrate? The obvious answer seems
to be that they were opposed to the war in Iraq, but is this all there is to it?
Was it solely opposition to the war, or was it also greater dissatisfaction in
general that made people protest against their government?

Questions about dissatisfaction are the subject of grievance theory, which
attempts to account for the grievances that motivate people to take part in
protest. How did such grievances develop, and how are they embedded in
the opinions and attitudes people maintain? For a long time, social move-
ment literature neglected the discussion of grievances; the existence of griev-
ances in a society had been taken for granted. It was assumed that grievances
abound and that their existence is not a sufficient reason for protest move-
ments to develop. More recently, the formation of grievances again gained
attention of social movement scholars (Neidhardt and Rucht 1993; Klan-
dermans 1997). Klandermans (2004) distinguishes between the demand side
of participation, which refers to the potential for protest in a society, and the
supply side of participation, which refers to the opportunities offered by the
protest organizers. Attempts to understand the thoughts and emotions that
fuel peoples willingness to take part in political action concern the demand
side of protest. Demand for protest begins with dissatisfaction, be it the expe-
rience of inequality, feelings of relative deprivation or of injustice, moral
indignation regarding some state of affairs, or suddenly imposed grievances
(Klandermans 1997). Protest organizations are — more or less — successful in
responding to demand for protest, and we may assume that organizations that
supply what potential participants demand obtain more support than those

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 99_^_

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES? 99

that fail to do so. Mobilization is the process that links demand and supply;
studies about participation tend to concentrate on this and to neglect the
development of demand and supply factors. To take either for granted is
unjustifiable.

In this chapter, I focus on the demand side of the protest against the war
in Iraq. I am interested in the opinions, attitudes, values, and ideologies that
characterize the participants; I maintain that motivations to engage in protest
are more complex than the proximate goals of a protest event suggests and
that this is to be found more so in the case of large-scale events (Oegema and
Klandermans 1994). I assume that attitudes in support of protest goals are
embedded in broader oppositional sentiments, which have their own histo-
ries, as antiwar sentiments have (Neidhardt and Rucht 1993). While they
frame the issue they are mobilizing for, social movement organizations try
to connect the issues to such broader sentiments. The more successful these
attempts to amplify the collective action frame are, the larger the demand
for protest in a society will be (Snow et al. 1986).

In Chapter 4 of this book, I described how politics of peace and war have
evolved since the early 1980s in the eight countries included in our study
(see also Ruzza 2004). Typically, antiwar sentiments and support for antiwar
| and peace protests more frequently are observed among people at the (far)

"W" left side of the political spectrum (Klandermans 1991). In chapter 5, it was ~®~

shown that, indeed, there was a clear mobilization surplus at the left side of
the political spectrum. Depending on the government in office, antiwar and
peace protests can turn into antigovernment coalitions quite easily (Oegema
and Klandermans 1994). Chances for such coalitions appear to be highest
if the government in office is right-wing. The game becomes more compli-
cated if a right-wing government takes on an antiwar position or if a left-
wing government takes on an anti-peace position (Kriesi et al. 1995), but
we maintain that the hypothesized mechanism holds in that political con-
text as well. Even in the case of a left-wing government with an antiwar posi-
tion, I contend that antiwar protest can become large, provided that a more
general antigovernment sentiment exists.

Hence, the hypothesis I will test more specifically in this chapter main-
tains that in addition to support for the proximate goal of the demonstra-
tion — against the war in Iraq — broader sentiments of opposition account for
successful mobilization. In other words, for the demonstration to grow big,
more general oppositional sentiments must exist in a country. The more these
exist, the larger the demonstration will become. As discussed in chapter 1,
standardized for population size, the size of the demonstrations in the eight
countries differed substantially. I will here test our assumption that these

H$H

-4

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1Q0_^_

100 KLANDERMANS

different levels of mobilization can be explained by differences in broader
oppositional sentiment.

I will begin exhibiting these findings with a discussion of the attitudes
directly related to the war: the degree of opposition to the war and the opin-
ion about each national government's policy regarding the war. To explore
the broader oppositional sentiments that these attitudes are embedded in,
I will first discuss the degree of opposition against neoliberal globalization.
The core section of this chapter, however, concerns people's attitudes about
their national government: the ways they perceive the functioning of democ-
racy within their own country, their feelings of political efficacy, as well as
their trust in their country's social and political institutions. I will assess the
extent to which the attitudes about the war are embedded in a broader oppo-
sitional sentiment, and then I will close the chapter with a discussion of the
roots of oppositional sentiment by relating it to political identity and demo-
graphic background.

Opposition to War

It will come as no surprise that the participants in the demonstrations were
opposed to the war in Iraq. We presented our respondents with eight opin-
ions about the war in Iraq that were frequently expressed in those days and
asked them whether they agreed or disagreed. The means for the eight coun-
tries are given in Table 6. 1 The first section of the table together with a
measure (Eta 2 ) indicating the degree of differentiation between the eight

The one opinion shared by almost all participants is that the United
States wants to invade Iraq to secure its oil supply. Note a large share of the
population in the eight countries also agreed on precisely the same statement,
as was shown in chapter 3. But that is where agreement ends. Two frequently
heard justifications of military intervention, namely, the potential threat of
Saddam Hussein and his regime to the world and the suffering of the Iraqi
people, are both rejected and accepted by substantial numbers of participants
in the demonstrations. However, such assessments — even to those who agreed
with these statements — did not justify a war, not even with the authoriza-
tion of the United Nations. In the eyes of the majority of the demonstrators,
"war is always wrong." Note, however, that participants from the two coun-
tries that were in fact preparing for war — the United States and the United
Kingdom — in that respect are less determined. If the United Nations were
to authorize intervention, they would be less opposed to the war, and they
were much less of the opinion that war is always wrong. Did Bush and Blair
persuade their people? Two other arguments against the war were much less

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 01(g)

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES? IOI

appealing to the participants: that it is racist and a crusade against Islam. Pre-
dominantly, the participants are opposed to war as a means to solve the con-
flict with Iraq.

We computed a scale consisting of all eight items. The resulting scale
indicates "opposition to the war in Iraq." A score of five indicates that some-
one is opposed to war as a political means, views the war against Iraq as a
war between civilizations, and does not believe that the allegations against
the Iraqi regime are sufficient to begin a war against the country. A score of
one indicates the opposite. Assessed with this scale, the opposition was strong-
est among the demonstrators in Italy and Spain and weakest among those in
the Netherlands, as shown in Table 6. Lb. The differences, though, are not
spectacular. On the whole, the opposition to war is strong among partici-
pants in all countries, and that, of course, is what one would expect.

Dissatisfaction with the Government's Efforts to Prevent War

How satisfied or dissatisfied were these people with how their government
dealt with the imminent war? This, of course, depends on the government's
stance and on how active it was — in the participants' view — in pursuing this
stance. Chapter 3 discussed the positions taken by the governments of the
| eight countries included in our study. Against that background, we may

~W expect significant differences between Belgium, Germany, and Switzerland ~W

on the one hand, and the United States, the United Kingdom, Spain, Italy,
and the Netherlands on the other. This is exactly what we found (see Table
6.1.b). Country alone can explain no less than 68 percent of the variance in
dissatisfaction with government policy. The means per country are what one
would expect: very high levels of dissatisfaction were found in the nations
where governments were preparing for war or took a supportive stand; low
levels of dissatisfaction in those where governments — more or less explic-
itly — declared their opposition to war.

In all countries, dissatisfaction was positively correlated to the "opposi-
tion to war" scale. In this respect Spain is somewhat of an outlier. Its rela-
tively low correlation is due to a lack of variance: 90 percent of the Spanish
participants are completely dissatisfied with the government's policy. By and
large, even in the countries whose governments opposed the war, demon-
strators who were more opposed to the war wanted their governments to
take firmer stands.

Widening the Blame

Negative opinions about the war in Iraq were the obvious motivators of par-
ticipation. Yet, as indicated, we believe that the situation is more complicated.

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1Q2^_

"4

I

!a^ | a | §

•^ t

8 si

§ 8., [

hi'

I ol

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 10Ji^_

m

6 -£

pi 1
) O O

H$H

-4

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 04^

104 KLANDERMANS

We hypothesized that demonstrations of the size that we witnessed on Feb-
ruary 15 do not have a single cause, which suggests that the driving force
behind demonstrations of that size is more complex than the proximate goal
suggests. I will test our hypothesis that greater dissatisfaction in general with
the government and how politics work generates protests to the extent that
we have witnessed on February 15.

Opposition to Neoliberal Globalization

A first test of that hypothesis regards the possible embeddedness of the
opposition to the war in Iraq in the opposition against neoliberal globaliza-
tion. After all, as described in chapter 1, the European Social Forum — the
global justice platform in Europe — played a significant role in the organiza-
tion of the demonstrations throughout Europe (see also chapter 10, which
describes the organizational backbone of February 1 5). In other words, aver-
sion to neoliberal globalization may have been on the agenda of many a par-
ticipant as well. In view of this possibility, we asked our respondents whether
the antiwar protest was a way for them to express their negative feelings
about neoliberal globalization. For quite a few participants, it obviously was,
especially those from Belgium and Italy (Table 6.1.c). In each country, more-
over, feelings against neoliberal globalization were significantly correlated
with the "opposition to war" scale. Opposition to war in Iraq was clearly
embedded in sentiment against neoliberal globalization, but there is more.
Indeed, this sentiment was much less associated with dissatisfaction of one's
government's policy toward the war; this observation underscores that this
opposition to neoliberalism is not directed at national governments in par-
ticular. That, of course, limits the potential of these sentiments as a viable
explanation for the success of the national demonstrations. Obviously, global
justice sentiment was not participants' only frame of reference.

A Voice against One's Government

This brings us to the central tenet of this chapter — that the success of the
demonstrations against war in Iraq must be explained by oppositional sen-
timent broader than opposition to the war or the national government's pol-
icy toward that war. To test that hypothesis, I assessed whether participants
in the demonstrations felt at home in the political landscape of their coun-
try in various ways: How did they feel about the working of democracy? Did
they experience some feeling of political efficacy? And did they trust the
social and political institutions of their society (Table 6.2)?

As for the working of democracy, we asked our respondents extent to
which they were satisfied with the functioning of democracy in their country.

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 05^

"4

!J|i

' 12 "b '»

N a 1 ] 1 '

5j o o o ■

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1Q6_^_

I06 KLANDERMANS

Their assessments differed substantially depending on which country they
were from (25 percent of the variance in satisfaction can be explained by
country alone). Italian and U.S. demonstrators were the least satisfied with
the working of democracy in their home countries. In all other countries,
means were on the positive side of our four-point scale. In Switzerland, Bel-
gium, and the Netherlands, people evaluated the political system more pos-
itively than in the five other countries.

Political efficacy was measured with a scale based on eight statements
about politics and politicians (Cronbachs alpha .80). Two examples are "I
don't see the use of voting; parties do whatever they want anyway" (reversed
coding) and "People like myself do have an influence on what the political
authorities do." "One" on the scale represents a low level of political efficacy,
and "five" represents a high level. The scores on the scale reveal this pattern:
relatively high levels of political efficacy were found in the Netherlands, Bel-
gium, and Switzerland, and relatively low levels were found in Italy. Respon-
dents from the United States, United Kingdom, Spain, and Germany occupy
intermittent positions.

In five countries — the United States, Italy, the Netherlands, Switzerland,
and Germany — we included a set of questions in the questionnaire about
/fa trust in social and political institutions. We constructed a scale (Cronbachs /is

| alpha .86) that measured trust in many social and political institutions. Exam-

ples are judiciary, police, employers, press, parties, banks, government, par-
liament, and political parties. On the whole, as the figures in Table 6.2 show,
the level of trust is the highest in the Netherlands and the lowest in Italy,
which may confirm the stereotype, but the story is certainly more compli-
cated than that. In the United States, anything that has to do with national
government and national politics is not trusted at all; this goes for the admin-
istration, the parties, national government, and, most of all, the president.
In Italy, though, people distrust the private sector, banks and employers, and
the government, but not the president. The four institutions that reveal the
largest differences among the countries are the president (Eta 2 .38), the national
government (Eta 2 .32), the administration (Eta 2 .18), and the national par-
liament (Eta 2 .16). To reiterate: the participants from the United States do
not trust their president at all, while those from Italy and Germany trust
theirs highly. The two kingdoms, Belgium and the Netherlands, generate
intermediate levels of trust in their head of state. With regard to the other
three (government, administration, and parliament) the demonstrators from
the United States and Italy always reveal the lowest levels of trust, while the
three other countries are at more or less identical levels.

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 07(g)

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES?

However, the evaluation of the working of social and political ii
although interesting, is not what occupies us here. Our main
whether these evaluations amplified the drive to participate in the demon-
stration against the war in Iraq.

Broader Opposition

We maintained that demonstrations of the size we witnessed on February 15
against the war in Iraq are embedded in broader opposition against the national
government. One way of testing this hypothesis is to assess whether opposi-
tion against the war was embedded in more general discontent about the
way social and political institutions work. Therefore, I conducted principal
component analyses with the two opinions about the war (opposition to the
war and dissatisfaction with the governments policy vis-a-vis the war) included
and the opinion about democracy, political efficacy, and trust in the core
social and political institutions in Western societies. As trust was only mea-
sured in some countries, I conducted separate analyses with and without trust
(Table 6.3).

The results are very consistent. Opposition to the war and the policy of
the national government is embedded in broader sentiments of resistance in
_^k all eight countries. The loadings on the principal component are invariably _fo_

| high, and the proportion of variance explained ranges between 40 and 50

percent. Spain deviates from this pattern; dissatisfaction with its government s
policy loads on a separate factor. This, however, is predominantly due to a
lack of variance. Indeed, more than 90 percent of Spanish demonstrators
were dissatisfied with their governments policy, irrespective of their own polit-
ical sentiments. The meaning of this is straightforward. The five attitudes
belong to a single underlying dimension: the more the participants are op-
posed to the war and the more dissatisfied they are with their national gov-
ernment's policy, the less happy they are with the functioning of democracy
in their country, the less politically effective they feel, and the less trust they
have in the social and political institutions of their country.

Note that our findings do not imply any causal relationship and, in fact,
this is not what we were after either. Our point is that the protests are clearly
embedded in broader opposition to national governments. Such embedded-
ness does not so much have a monocausal signification but rather is of a
recurrent manner. That is to say, opposition to the war and dissatisfaction
with the governments policy reinforce an existent oppositional sentiment
toward the national government and its institutions. In its turn, those sen-
timents work together to produce an ever-larger demonstration.

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1Qa^_

"4

x ll || 1| h $

H$h-

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1Q2^_

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES? IO9

Levels of Mobilization

If, indeed, opposition to war and dissatisfaction with the government's effort
to prevent war are embedded in a broader such sentiment, one would assume
that this oppositional sentiment accounts for the diverging mobilization lev-
els in the eight countries. Thus the questions arise: Does such a link between
oppositional sentiment and mobilization levels indeed exist? Are partici-
pants in the demonstration more negative about politics in general than
non-participants are? The answers to these questions, of course, are crucial
to the test of my assumption that massive protest reflects a broader opposi-
tional sentiment than the immediate issue of the demonstration does. To
answer the latter question, we exploited the fact that the question regarding
the functioning of democracy was included in several waves of the Euro-
barometer. 1 Although this poses the drawback that we have no information
on the United States and Switzerland, we assume that a comparison of the
remaining six European countries, nonetheless, will be relevant and reveal-
ing — and indeed it is. In Table 6.4, I compare the findings of two Euro-
barometers and our own survey. The question regarding the functioning of
democracy was not included in the surveys of 2003, yet it was asked in
November 2001 and again in April 2004. Hence, we have two standards of
JL comparison, one roughly a year prior and another a year later. ^1

T The figures in the table show the proportions of the respondents who

indicate that they are "not very satisfied" or "not at all satisfied" with the
functioning of democracy in their country. First, participants in the demon-
strations in all six countries appear less satisfied than the average citizen, cer-
tainly if one takes 200 1 as the standard of comparison. Second, the six coun-
tries included in the comparison clearly fall into two broad categories: the
Netherlands, Belgium, and Germany, on the one hand, where the differences

Table 6.4. Dissatisfaction with the functioning of democracy: percentage
not satisfied

UK SP IT NL BE GE

Eurobarometer November 2001 27 30 60 26 31

Eurobarometer April 2004 34 29 60 33 35

Iraq demonstration survey 70 62 93 40 37

Survey-Eurobaromenter 2001 discrepancy 43 32 33 14 6

Survey-Eurobarometer 2004 discrepancy 36 33 33 7 2

Mobilization level at Iraq demonstration 1.6 2.5 2.6 .04 .0

ts the percentage of responden

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 10^

IIO KLANDERMANS

are relatively small, and the United Kingdom, Spain, and Italy, on the other,
where the differences are great. The bottom row shows the mobilization level
of the six countries; the dichotomy in terms of satisfaction with the func-
tioning of democracy is strikingly reproduced in terms of the corresponding
mobilization level. In those countries where the discrepancy between par-
ticipants and citizens is the greatest, the mobilization level is the highest. That,
of course, makes sense. If the demonstrators were only as dissatisfied as those
who did not demonstrate, such discontent could not explain their partici-
pation. That the biggest demonstrations were in countries where, compared
to non-participants, participants were more dissatisfied with the ways their
democracy supports our argument that the demonstrations were about more
than opposition to war alone.

But the question, naturally, is whether this is a real or a spurious corre-
lation. One could argue that the opposition to war was very strong in Italy
and Spain and that the mobilization levels in those two countries therefore
were high, but opposition was strong in Belgium as well, though turnout was
lower, while negative attitudes toward war among the populace in the United
Kingdom was of the same level as that in the Netherlands and Germany,
where turnout was significantly lower. One could also argue that the level
of dissatisfaction with the national government's efforts ti

| of dissatisfaction with the national government's efforts to prevent war was

W very high in the United Kingdom, Spain, and Italy, but it was high in the v9)~

Netherlands too. Finally, one could argue that the discrepancy in Belgium
was so low because the Belgian government was against the war, but the Ger-
man government was against the war as well, and the Dutch government was
not. These and other considerations are addressed in Table 6.5, which pres-
ents multiple regression analyses, with the mobilization level of a country as
the dependent variable and the attitudes about the war and national politics
as the independent variables.

The number of cases in this analysis is very small, hence one must be
cautious regarding the interpretations and conclusions. Yet, the pattern is con-
sistent and allows for a clear conclusion. The differences in mobilization
levels among the six countries in the analysis nearly can be completely
accounted for by the combined influence of opposition to war and dissatis-
faction about the functioning of democracy. Note, however, that the discrep-
ancy between the level of dissatisfaction among the general population and
that among our respondents rather than the absolute level of dissatisfaction,
makes the difference. The presence of many people who are more dissat-
isfied than the average citizen about how democracy works in their country
is linked with demonstration size. Note also that being more dissatisfied
with the functioning of democracy rather than satisfied with the government's

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 11 afc

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES?

Table 6.5. Level of mobilization and attitudes about the war and
national politics

Multiple regression: standardized betas

Opposition to war
Dissatisfaction with

government's policy
action with

functioning of democracy
Discrepancy with Eurob;

November 2001
Discrepancy with Eurobt

April 2004
Political efficacy

N

Model 1 Model 2 Model 3 Model 4 Pearson's r
.56 .64** .54* .50

"4

efforts to prevent war accounts for the different mobilization levels. Indeed,
the lowering of the beta coefficient compared to the Pearson correlation sug-
gests that dissatisfaction with one's government's efforts feeds into more gen-
eral dissatisfaction about the functioning of democracy. In its turn, that
broader dissatisfaction made the mobilizat

Political Identity

The embeddedness of the resistance to the war in Iraq in oppositional sen-
timent is further evidenced by the political identity of the average demon-
strator. The vast majority of the participants in the demonstration are of a
leftist political orientation. On a self-placement scale from 0, "left," to 10,
"right," scores range from 1.8 in Italy to 2.8 in Belgium and the United King-
dom. Indeed, more than 90 percent of the participants occupy the left end
(0^) of the scale. This orientation is confirmed by the participants' self-
reported voting behavior — both in the past and in the future. The political
landscape clearly differs in the eight countries, but very few of our respon-
dents voted or intended to vote for parties at the right end or even in the
center of the political spectrum. Basically, the Social Democrats, the Greens,
and the Far Left populated the demonstrations, although the exact mix of
the three groups varied, depending on the national political configuration.
The partisan affiliation of the demonstrators will be much further elaborated
in chapter 8 of this volume.

I

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 1?(£)

112 KLANDERMANS

In the context of this chapter, neither political identity nor voting be-
havior per se interests me. My objective is to show that the participants in
the demonstration were opposed to more than the imminent war in Iraq. In
that respect, it is obviously relevant that the political identity of the major-
ity is oppositional; the demonstrators voted for parties in the opposition,
and if they did not, they planned to vote for them in the future (see chapter
8). Thus, even though we lack information on nonparticipants, it is safe to
conclude that compared to the remainder of the population in their coun-
try, the participants in the demonstrations were from a specific range of the
political spectrum.

Embedded Opposition

In a final step to demonstrate the embeddedness of the opposition to the war
in Iraq in more general oppositional sentiments, we conducted regression
analyses and multivariate analyses of variance reported in Table 6.6. 2 In
these analyses, we attempted to assess the extent to which opposition to the
war in Iraq and to the government's policy regarding it were influenced by
people's political identity and political attitudes. We assessed the influence
of these political attitudes alone and controlling for country and voting in
| the past, with all the countries included, and with only those countries in-

~W eluded in which the questions about trust were asked. ~W

Table 6.6 summarizes the results of several regression analyses (Models
1 and 3) and multivariate analyses of variance (Models 2 and 4). In Models
1 and 3, the opposition to war and dissatisfaction with one's government's
policy are regressed in a stepwise manner on left-right placement, the assess-
ment of democracy and political efficacy, and a rejection of neoliberal glob-
alization as motivation to take part in the demonstration. Models 2 and 4
report results from multivariate analyses of variance with country and past
vote as factors and in three separate steps left-right placement, the assessment
of democracy and political efficacy, in addition to anti-neoliberal globaliza-
tion sentiment as covariates. The table presents the cumulative R-squares for
the subsequent steps. 3 As we are predominantly interested in the question of
whether opposition to war and one's government's policy vis-a-vis the war
were embedded in a more general oppositional sentiment, we restrict our
presentation of results to these summary figures.

Models 1 and 3 show that even among the leftist-oriented people who
participated in the demonstration, left-right placement continues to have an
impact. Put simply: the more left the demonstrators, the more opposed to the
war they are, and the more dissatisfied with their government's policy. But
left-right placement does not exhaust the differences in opinion regarding

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page1l3^_

^

fr

1

a.

l

D,

£

£

<B

"o

b

5

2

f

-£

^

c*

c

£!

(D

C

J

1

o

•=

5

b

o_

*S

1

1

_>*

o

o

£

T5

g

2

o

.>

"5

E

1

"O

c
o

°

2

1

1

-OJ

o

~S

O

o

c

.g

$

S-

1

-D

1

c

g

§.

Q.

O

-a

<D

-o

"O

J

£

<o

-6

(D

-Q

•2

: '6 ts .5 '6 ta .1

li 'I f

^ c §

1 'b 1

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 14^

114 KLANDERMANS

national politics relevant in this context. Significant proportions of v;
explained are added when differences in opinion about democracy, political
efficacy, and trust in sociopolitical institutions are entered into the equation
for both opposition to war and dissatisfaction with the government's policy
The less satisfied the demonstrators were with the functioning of democracy
in their country, the less politically effective they felt, and the less they trusted
the sociopolitical institutions in their country, the more they were opposed
to the war in Iraq and the more dissatisfied they were about their govern-
ment's policy regarding the war. Opposition to neoliberal globalization added
significantly in the case of opposition to war but did not have an impact in
the case of dissatisfaction with the government's policy vis-a-vis the war. I
observed this when discussing the zero-order correlations (Table 6.4). I ex-
plained this finding, which is now reconfirmed in multivariate analyses, by
arguing that neoliberal globalization implies global dynamics that a national
government controls only to a very limited extent.

Models 2 and 4 show that the relationships between political identity
and attitudes and opposition to war hold, irrespective of which the respon-
dents' nations and of voting patterns. Left-right placement and political atti-
tudes continue to add significantly to the variance explained, beyond country
I and past vote. Even in the case of dissatisfaction with government policy,

~W political attitudes add beyond the large proportions of variance explained by ~W

voting patterns and country variation. The proportion of variance added is,
of necessity, modest, as 75 and 80 percent of the variance is explained by
country and voting alone, but both are statistically significant improvements
of the models. Opposition to neoliberal globalization, however, does not add
further. This was to be expected on the basis of previous analyses.

I also found significant interactions of country and voting patterns. This
indicates that the impact of voting patterns differed per country. For a fur-
ther exploration of these interactions, I will focus on the parties to the left:
the Far Left, the Greens, and the Social Democrats. As support for the re-
maining parties is too limited in all or most countries to generate a reliable
comparison, I collapsed those parties into one category, namely "others,"
which comprises the parties further to the right on the political spectrum.
To explore the individual interactions, I computed the means for opposition
to war and dissatisfaction with government's policy vis-a-vis the war for these
three parties and the remaining parties in the eight countries.

In most countries, opposition to war is strongest among supporters of the
Far Left, followed the Greens, and then the Social Democrats. Opposition to
war is weakest among supporters of the parties farther to the right. However,
the differences among the party supporters vary across countries. Although

H$H

-4

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 15^

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES? 115

they follow the general pattern, the differences in Italy and the Netherlands
are not significant. The impact of voting patterns on the opposition to war
is the largest in Germany and Switzerland. In Switzerland, because the oppo-
sition is significantly lower among supporters from the parties farther to the
right than among those of parties to the left. This holds for Germany as well.
But in addition, German supporters of the Far Left are significantly more
opposed to the war than those of the Greens and the Social Democrats. We
found a pattern similar to that in Germany in Spain, though slightly less
prominent. In the United States, supporters of Ralph Nader (27 percent) were
more opposed to the war than supporters of Gore. In the United Kingdom,
supporters of any but the leftist parties did not differ from supporters of the
social democrats.

As far as dissatisfaction with government's policy is concerned, the pic-
ture is more complicated. First, there are the demonstrators from the United
States, the United Kingdom, and Italy who are equally annoyed, regardless
of which party they choose. Second, there are the demonstrators from the
Netherlands and Spain who are annoyed as well, but at different levels, de-
pending on party preference; that is to say, the more left, the more annoyed.
Finally, there are demonstrators from Belgium, Germany, and Switzerland
who are much more satisfied with their governments policy. In Belgium,
however, supporters of the Far Left are far more dissatisfied with their gov-
ernment's policy than are those of the other parties. Also in Germany, sup-
porters of the two parties in office (Greens and Social Democrats) are the
predominantly satisfied ones. Members of the remaining parties (including
the Far Left) are significantly less satisfied.

The Demographics of Opposition

My analyses thus far supports the argument that opposition to the war in
Iraq and dissatisfaction with one's government's policy in that regard are part
of a broader sentiment of opposition — against neoliberal globalization and/or
how politics are working in the participants' country, or both. Such opin-
ions are embedded in a broader political identity, as reflected in left-right
placement and voting behavior. The last step in this chapter concerns the
demographic correlates of these political attitudes. How are they related to
such factors as gender, age, education, and religion? To assess these relation-
ships, I conducted multivariate analyses with the opposition to the war and
one's government's policy regarding the war as the dependent variables; coun-
try and religion as fixed factors;; and age, gender, and level of education as
covariates of country and religion.

Although the effects are never very strong, the analyses suggest that the

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 l£4fe

Il6 KLANDERMANS

two attitudes — opposition to war and dissatisfaction with government — are
distributed differentially among the categories of gender, age, education, and
religion. Women are more opposed to the war in Iraq than men are. More
educated people are less opposed to the war than less educated people. Age
did not influence the degree of opposition to war. These findings hold for
all eight countries. In most countries, participants from Protestant or Roman
Catholic backgrounds are less opposed to the war than those who adhere to
no religion or adherents of other religions, but in Spain and Switzerland we
found the opposite pattern in some instances. Dissatisfaction with the gov-
ernment's policy vis-a-vis the war increased if demonstrators were older, but
gender and level of education were unrelated to such feelings of dissatisfac-
tion. In Spain, participants belonging to Protestant or Catholic churches are
more dissatisfied with their government's war policy.

Conclusion

It's not surprising that we found strong opposition to the war in Iraq among
participants in the demonstrations in all eight countries, although we observed
differences in degree of opposition among the countries. Rejection of war as
a means to solving an international conflict brought the people to the streets
| even though many of the demonstrators believed that the Iraqi regime de-

~W served no sympathy. ~W

Dissatisfaction with the government's policy vis-a-vis the war was, to a
large extent, dependent on the actual policy of the government of someone's
country: very high levels of dissatisfaction in the countries that were prepar-
ing for war or that took a supportive stance and satisfaction in the countries
whose governments rejected war. The more opposed demonstrators were to
the war, the more dissatisfied they were with their government's policy, even
in those countries where governments declared to be opposed to the war.
For obvious reasons, the correlation between opposition and dissatisfaction
was the strongest in the two countries that were planning for war: the United
States and the United Kingdom.

The frames of reference of the demonstrators, however, were much
broader than opposition to the war and dissatisfaction with the government's
policy; these were clearly embedded in broader oppositional sentiments. On
the one hand, for many people, the demonstration was also a way of express-
ing their opposition to neoliberal globalization. On the other hand, the dem-
onstration signaled more general opposition to national governments expressed
by citizens opposed to the powers in office. The different levels of mobiliza-
tion in the eight countries can largely be explained by their opposition to
the war and by their dissatisfaction with how democracy functioned in their

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page 1 17(g)

PEACE DEMONSTRATIONS, ANTIGOVERNMENT MARCHES? UJ

country. The embeddedness of the opposition to the war in broader oppo-
sitional sentiments is further underscored by the political identity of the
participants, who invariably came from the leftist side of the political spec-
trum and disproportionately voted for opposition parties. The governments'
policies vis-a-vis the war reinforced existing oppositional sentiments and made
people take to the streets in unprecedented numbers.

Notes

1 . The Eurobarometer also includes questions about trust in the national gov-
ernment and parliament. Unfortunately, the 1999 survey was the last one that in-
cluded these questions. We did not feel that a comparison of our data with these data
would be valuable. However, our own data show that the level of satisfaction with
the functioning of democracy is strongly correlated with these two indicators of trust
and also feelings of political efficacy (Pearson r = .60, .55, and .53 for trust in gov-
ernment, trust in parliament, and political efficacy, respectively). Hence, we believe
that our findings regarding the functioning of democracy are a reliable indicator of
people's attitude toward national politics.

2. As country and vote were nominal variables, it was more appropriate to treat
them as fixed factors in a multivariate analysis of variance.

ziv 3. We chose to report R-squares rather than full models, which would have meant ,i\_

J including many more figures without necessarily providing more information. After

all, we were interested in the R-squares as indicators of variance explained and the
R-square changes as indicators of the increase of explained variance when a group of
variables was entered in the equation.

Works Cited

Klandermans, Bert. 1991. "The Peace Movement and Social Movement Theory."
In Peace Movement in International Perspective, International Social Movement
Research 3, ed. Bert Klandermans, 1-42. Greenwich, Conn.: JAI Press.

. 1997. The Social Psychology of Protest. Oxford: Blackwell.

. 2004. "The Demand and Supply of Participation: Social Psychological Cor-
relates of Participation in a Social Movement." I npanion to Social
Movements, ed. David A. Snow, Sarah A. Soule, and Hanspeter Kriesi, 360-79.
Oxford: Blackwell.

Kriesi, Hanspeter, Ruud Koopmans, Jan Willem Duyvendak, and Marco G. Giugni.
1995. New Social Movements in Western Europe: A Comparative Analysis. Min-
neapolis: University of Minnesota Press.

Neidhardt, Friedhelm, and Dieter Rucht. 1993. "Auf dem Weg in die Bewegungsgesell-
schaft? Ueber die Stabilisierbarkeit sozialer Bewegungen." Soziale WeltAA: 305-26.

H$H

06 Chapter 6_Walgrave 22/01/2010 3:40 PM Page1lB^_

Il8 KLANDERMANS

Oegema, Dirk, and Bert Klandermans. 1994. "Why Social Movement Sympathiz-
ers Don't Participate: Erosion and Nonconversion of Support." American Soci-
ological Review 59: 703-22.

Ruzza, Carlo. 2004. "Peace Movements." In Democracy and Protest: Popular Protest and
New Social Movements, ed. Gary Taylor and Malcolm Todd, 290-306. London:
Merlin.

Snow, David A., E. Burke Rochford Jr., Steve K. Worden, and Robert D. Benford.
1986. "Frame Alignment Processes, Micro-Mobilization, and Movement Par-
ticipation." American Sociological Review 51: 464-81.

m

h$h

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 13^

Paths to the February 15 Protest:
Social or Political Determinants?

Donatella delta Porta

m

This chapter analyzes the degrees and forms of participation among February
15 demonstrators. The questions of participants' previous experiences in anti-
war demonstrations is particularly relevant, because of the very characteris-
tics of the peace movements — small nuclei of committed pacifists — but the
capacity to at times mobilize very large and heterogeneous networks rooted in
various movements; and the involvement of citizens with mainly ethical or
religious stances, who are not (or not yet) politicized (see chapter 4). The
issues of diversity in participants' backgrounds as well as the presence of large
numbers of first-timers were particularly relevant for the 2003 demonstrations
against the war in Iraq, which did not confirm the paths of cross-national
strengths and weaknesses in mobilization capacity that had emerged in previ-
ous waves of political mobilization: for instance, the protest remained weak in
the Netherlands (where the protest against the deployment of cruise missiles
had been very strong in the early 1980s) but was impressively widespread in
Spain, where no previous large mobilization had existed (although the anti-
militarist movement was very active). In several countries, the mobilizations
against Iraq attracted not only already mobilized activists but also new gen-
erations of protestors as well as many "first timers." Called for by the Assem-
bly of European Social Movements, held at the end of the first European
Social Forum, the Global Day of Action was strongly supported by a broad
and at that time particularly visible global justice movement (della Porta 2007) .
An analysis of the social and political backgrounds of demonstrators could
help us understand the extraordinary mobilization capacity of that protest
(see chapter 2) and also its differential success in mobilizing "on the street."

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 120_^_

DELLA PORTA

In this chapter, I will focus on two dependent variables: frequency and
action repertoires of previous protest participation. Beyond the description
of the February 15 marchers, I also want to examine the various degrees and
forms of participation previously engaged in by those citizens who took to
the street on the Global Day of Action. In social movement studies, reper-
toires of protest have usually been explained by political opportunities, either
structural or contingent. Cross-nationally, the institutional closure or open-
ness toward challengers as well as toward the coalitions in government are
expected to have impact upon the degree and forms of mobilization. In gen-
eral, the presence of allies in government supports the use of "insiders' strate-
gies" (lobbying or codecision) rather than mass protest — unless those allies
look for public expressions of support, in particular through contentious
collective actions that address an international organization or foreign gov-
ernment/s. Since the peace demonstrations mobilized different social and
political groups in the countries under study (see chapters 5 and 6 in this
volume), the cross-national analysis has to be supplemented by an analysis
of individual paths of participation. Especially in research on political par-
ticipation, sociodemographic background and political attitudes are the most
often discussed sets of variables. In this analysis, I will single out the individ-
/fcs ual characteristics of more or less frequent demonstrators, as well as moder- sL.

| ate and more radical ones, in terms of their sociodemographic backgrounds

and their political attitudes, all of which converge in forming their political
identities. 1 In what follows, I will observe country characteristics and discuss
the extent to which they can be explained by national political opportunities.

In this chapter, I shall discuss these hypotheses and also take a step for-
ward, beyond causal explanation of individual choices and toward the analy-
sis of the dynamics of militantism, that is, the presence of long-term paths.
After introducing the explanatory model, I will devote a second section to a
cross-national comparison of degrees and forms of commitment. I shall then
discuss the hypotheses about the socioeconomic centrality of activists, look-
ing mainly at profession, age, gender, and education. In the fourth section,
I shall address the hypotheses about the construction of a militant career,
looking at the previous experiences of activists, in terms of both participa-
tion in organizations and activities promoted by different movements. In
the concluding section, I will reflect on the impact of environmental vari-
ables, filtered via activists' experiences, on the specific paths of militancy of
different generations. Throughout, I will consider political opportunity struc-
ture (measured by the country in which the demonstration was held) as an
intervening variable.

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 21 afc

PATHS TO THE FEBRUARY 15 PROTEST 121

What Do We Want to Explain: Degrees and Forms of Participation

Political participation has been defined as all citizens' behavior oriented to
influence the political process (Axford, Huggins, and Turner 1997, 109).
There are, however, many different degrees of political participation and var-
ious means for influencing the political process, from minimal forms, such
as voluntary exposure to political messages (reading newspapers, listening to
news) to activation in specific activities — from "ballots to bullets." Research
on political participation initially explicitly excluded protest, focusing on con-
ventional forms of participation, such as being exposed to political messages,
voting, carrying political buttons, contacting public officers, contributing
money for a party or candidate, assisting in a public speech, engaging in a
political campaign, occupying electoral public office (Milbrath 1965, 18).
Since the 1970s, however, surveys have included nonconventional forms of
political participation, such as letters to newspapers, boycotts, self-reduction
of tax and rent, occupations of public buildings, traffic blocks, petitions, sit-
ins, wildcat strikes, marches, damage to property, and violence against per-
sons (Dalton 1996).

Protest forms can be more or less radical in nature, ranging from con-

. ventional petitioning to more conflictual activities, including violent forms

Afr- of action. Russell Dalton has suggested several thresholds on the contin- .fa.

I uum, from the least to the most radical forms: "The first threshold indicates

the transition from conventional to unconventional politics. Signing peti-
tions and participating in lawful demonstrations are unorthodox political
activities but still within the bounds of accepted democratic norms. The sec-
ond threshold represents the shift to direct action techniques, such as boy-
cotts. A third level of political activities involves illegal, but nonviolent, acts.
Unofficial strikes or a peaceful occupation of a building typify this step. Finally,
a fourth threshold includes violent activities such as personal injury or phys-
ical damage" (Dalton 1996). Recent surveys of participants in large protest
events— such as the 2001 protest against the G8 in Genoa (Andretta et al.
2002; 2003), the 2002 European Social Forum in Florence (della Porta et
al. 2006), and the 2003 European Social Forum in Paris (Agrikoliansky and
Sommier 2004) — indicate that the activists had a rich and various repertoire
that combines many forms of action.

None of these studies has gone in depth into the analysis of the con-
textual, sociodemographic or attitudinal determinants of activists' choices of
protest repertoire nor systematically compared activists of different countries.
This is indeed what I want to do in what follows, by looking at the degree
and repertoires of previous experiences with political participation of those

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 122^_

122 DELLA PORTA

who mobilized on February 15. I will measure frequency of participation by
the question "Can you estimate how often in the past five years you took
part in a local, national or international demonstration or public protest?"
coded into three categories: "first time," "two to ten times" and "more than
ten times." I measure protest repertoires by an extensive set of questions refer-
ring to forms of action used in the past, which (on the basis of previous re-
search in the field as well as factor analysis, see below) I have recoded: First,
ial forms include contacting a politician, an organization or
a civil servant; wearing a pin or hanging a flyer, poster, or
sticker of a political campaign; making a donation; raising funds; contact-
ing the media or appearing in the media; and abstaining from an election
for reasons of protest. Second, demonstrative forms include signing a people's
initiative, a referendum, or a petition; setting up or gathering signatures for
a petition; and taking part in a strike. Third, political consumerism includes
activities such as taking part in a product boycott or buying a product for
political, ethical, or ecological reasons. Fourth, disruptive action is opera-
tionalized as occupying a public building, school, or university or squatting
in houses or abandoned areas. Finally, violent action is defined as the engage-
ment in forms of action that imply violence against property or persons.
| Research on social movements suggests that national political opportu-

~W nities influence national repertoires of action. As I shall argue in the next para- ~W

graph, however, the relevance of political opportunities for the explanation
of the country-specific dominant repertoire is not confirmed when one looks
at the activist backgrounds of those who participate in specific protest events.
Moving from aggregate to individual data, there are two classical explanations
for individual political participation. The first focuses on the individual
availability of resources, such as, in particular, money (and, consequently,
time) as well as education (and, consequently, the sense of having the "right
to speak up"), taking some sociodemographic characteristics as indicators.
These hypotheses have been put forward in the pioneering research on polit-
ical participation, especially stressing participation as a middle-class activity,
but also confirmed later on, particularly regarding level of education. However,
these studies referred in particular to conventional forms of participation.
Research on unconventional forms has already indicated some different pat-
terns. Regarding the labor movement, theories of political participation have
stressed the role of class consciousness more than class position. Analyses of
new social movements, although mentioning the high presence of segments
of the middle classes, have also emphasized the role of beliefs more than be-
longing to a specific social group. Research on social movements has empha-
sized the importance of political socialization — both primary socialization,

H$H

-4

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 123^_

PATHS TO THE FEBRUARY 15 PROTEST I23

as in the family, and secondary socialization, as in peer groups — in facilitat-
ing political activism.

Repertoires of Action and Political Opportunities

Looking at the political opportunities for protest, comparative studies have
considered the degree of centralization of the state apparatus, government
control over market participation, and the dependence of the judiciary on
other arms of government (Kitschelt 1986, 61-64); regime structure (Rucht
1994, 303-12); and territorial decentralization, the functional division of
powers, and direct democracy (Kriesi 1995). Institutional variables influence
the strategies adopted by national social movement families: social move-
ments tend to use the channels of access made available to them by "weak"
states, reducing their availability for more disruptive protest; and, in reverse,
closed political institutions trigger an escalation of political conflicts (della
Porta and Rucht 1995). In this volume, Beyeler and Rucht present some
indicators of power sharing and potential alliance for progressive movements
as important political dimensions. Additionally, Mario Diani links the mobi-
lization capacity of the peace marches to the position of their respective
national governments on the Iraq War with open opportunities in Germany
and Belgium (and in some ways also Switzerland), and, in contrast, very closed
opportunities in the United States, United Kingdom, Italy, and Spain (and,
although less so, in the Netherlands). But how do political opportunities
relate to frequency and repertoire of previous participation among the mobi-
lized activists? Do traditionally closed systems tend to produce more radical
demonstrators? And do open opportunities facilitate the participation of
"first timers," as well as unconventional participation in general? Or, is it
when a danger appears as more menacing (for instance, with an adversary in
government) that even previously nonmobilized people take into the street?
Our data confirm the difficulty, mentioned in chapter 2, of easily translat-
ing general political opportunities in terms of characteristics of mobilized
protestors at specific marches.

As for frequency of participation, answers to the question "How often did
you participate in protest activities in the last five years?" indicate that Ital-
ians are much more highly mobilized (average 10.56), followed by Belgians
(6.64), Spaniards (5.74), Swiss (6.66) and Germans (5.17) with moderate
values and Americans (4.79), Brits and Scots (3.55), and Dutch (3.22) on
the less mobilized side. 2 Relevant country differences are also visible as far as
"first timers" are concerned: against an average of about a third (30 percent),
the Netherlands presents a very high percentage of first timers (54 percent),
followed by the United Kingdom (49 percent), while first timers are rare in

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 124_^_

124 DELLA PORTA

Italy (9 percent, with an even slightly lower 8 percent for the postal sample).
The frequency of previous participation seems to reflect the strength of recent
mobilizations in the global justice movements in various countries (see della
Porta 2007) more than the position of each country in terms of distribution
of power (as reported in Table 2.1, this volume) . Hypotheses suggesting higher
levels of mobilization when the right is in government are also not clearly
confirmed — for example, look at the positions of the Belgians and the Ger-
mans, whose center-left governments had taken strong stances against the
war. Frequency of participation is explained neither by the influence of left-
wing parties in terms of access to power (Table 2.2) nor by indicators of the
strength of labor organizations (Table 2.3), while (with the exception of Bel-
gium) frequency tends to increase with the percentage of citizens that locate
themselves on the left of the political spectrum (Table 2.4), with scores high
in Spain and Italy and low in the United States and the United Kingdom.

To investigate the specific repertoires of participation belonging to the
protest repertoires of activists from different countries, we have recoded the
question regarding the use of various forms of action, singling out conven-
tional, semi-conventional, political consumerism, disruptive and violent forms
and comparing actual values with expected counts by country. A factor analy-
| sis we ran confirmed that our theoretically based aggregation in sub-types

"W" reflected the statistical structure hidden behind our set of data. 3 ~®~

According to our data (see Table 7. 1), most activists from various coun-
tries have a very high degree of experience with semi-conventional forms of
participation, with almost all U.S. activists, at the highest level, having expe-
riences in semi-conventional actions (98 percent) and Spaniards with the
lowest (although certainly not low in absolute terms), at 80 percent. The Bel-
gians and Dutch are slightly lower than average, with all other countries very
near the average of 88 percent. I can therefore state that, with some excep-
tions, almost all of those who took part in the February 15 marches have
experiences in semi-conventional forms of action as well. If the two outliers
seem to provide some leverage for the hypothesis that "weak" states (such as
the United States) facilitate more conventional forms of participation (and
"strong" state such as Spain do the opposite), the analysis of the patterns of
our countries does not confirm such a clear picture.

More differences emerge if we look at the demonstrative forms of par-
ticipation that have been used more than the average by the Italians and Swiss
(91 percent), followed by the Americans (89 percent) and then Spaniards,
Belgians, and British, while the Germans and the Dutch have values well
below the average. Also in this case, traditional political opportunity struc-
ture hypotheses are of little help (suffice it to say that the top position is

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 25^

"4

PATHS TO THE FEBRUARY 15 PROTEST

Table 7.1. Repertoires of participation by countries: Positive responses
("yes" in percent)

Semi-

Political

Disruptive

conventional

Direct forms

consumerism

action

Violent

United States

98

89

80

4

1

United Kingdo

m 90

79

4

1

Spain

80

87

62

7

1

Italy

90

91

■

4

Netherlands

66

5

1

Switzerland

91

92

91

7

2

Belgium

87

6

1

Germany

86

68

70

6

2

Total

88

84

78

10

2

N

4,552

4,315

3,998

495

96

occupied by activists from the "most different" Italy and Switzerland). The
same is true if one looks at a set of demonstrative forms that I have kept sep-
arate because of their increasing relevance for contemporary movements:
political consumerism, especially boycott and ethically driven consumption. In
this case as well, there are significant differences that cannot be understood
in terms of political opportunities as defined by distribution of power: boy-
cott is used more by the Swiss but also by participants in the February 15
marches in Italy, Belgium, the United Kingdom and the United States — sig-
nificantly less, again, in the Netherlands and Germany (della Porta 2003).

If we look at disruptive action (illegal but not violent, in Dalton's defi-
nition), the Italian case emerges as exceptionally different from the others
(with almost 28 percent of the marchers responding that they have used dis-
ruptive protest), while differences are much lower among the other countries
(ranging from 4 percent in the United Kingdom and United States to 7 per-
cent in Switzerland). 4 Very low are the values related to the use of violent actions:
1.9 percent of the total number of participants, with a higher (but still not
high) 4 percent in Italy.

We can account for the difficulty in using political opportunities to ex-
plain the previous protest experiences of the marchers if we consider that the

al and political base in the v
es could be expected to have an
ffering strategic propensities of

peace marches involved a quite different

ious cases. In particular, political opportuniti

effect ceteris paribus, that is, in explaining di

similar social and political groups. Here, the countries differ a great deal

terms of the capacity of the February 15 demonstration to mobilize activists

from various social movements. Looking at previous experiences in specific

types of demonstrations (Table 7.2), we can observe the relevant participation

H$H

-4

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 12fi^_

126 DELLA PORTA

of Italians in almost all types of demonstrations, of Belgians especially in new
global, antiracist and third-world events, and of Germans in pacifist and
antiracist, but also environmentalist protest. The Spaniards have especially
frequent experiences in demonstrations on social issues; citizens of the United
Kingdom on antiracism; and Americans on women's and peace issues; while
there is relatively less mobilization on the various issues among Swiss and
Dutch. Comparing these data with those from the World Values Survey, we
note that the differential presence of activists with experiences regarding var-
ious issues does not seem to be explained by the mere strength of member-
ships in related movements in the national population (Table 2.5).

Social Centrality and Participation

Moving from the macro to the micro, these are the first questions to inves-
tigate: What are the characteristics, within each country, of the activists with
different degrees of participation? What explains the choices among the differ-
ent repertoires of activism? Are the same explanations valid across countries?
A first set of variables necessary to consider refers to individual socioeco-
nomic resources. The early research on political participation, based on sur-
veys, reveals very low levels of participation (for a review, see Lagroye 1993,
312). 5 Moreover, the number of citizens involved diminished dramatically
for the more demanding forms of participation. The normative problems in-
volved in this selectivity were increased by the nonrepresentativeness of
those who participated: in fact, higher levels of participation were singled
out, ceteris paribus, for those who were socially central, that is, the better
educated, middle-class, male, medium-age cohort, married people, city res-
idents, ethnic majorities, and citizens involved in voluntary associations (Mil-
brath and Goel 1977). Similarly, in their research on participation in the
United States, Verba and Nie (1972) observed that the higher the social status
of an individual, the higher the probability that she or he will participate;
Verba, Nie, and Kim 1978) have confirmed this observation in a seven-
nation comparison (that concluded that social inequalities are reflected in un-
equal political influence. Usually, higher social status implies more material
resources (and free time) to invest in political participation, a higher probabil-
ity of being successful (via personal relationships with powerful individuals),
and especially a higher sense of personal achievement. Psychological disad-
vantages overlap with social disadvantages, reducing the perception of one's
own "right to speak up" (Bourdieu 1979, 180). If participation responds to
demands for equality, it does tend to reproduce inequalities, since "any indi-
vidual participates, at least potentially, with the differential (or unequal) influ-
ence (if we do not want to use the word 'privilege,' that would have an ancien

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 12Z^_

1 3

4

C "p2

1

8.

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 12B^_

DELLA PORTA

^

regime flavor) that characterizes his/her position in the system of private inter-
ests" (Pizzorno 1966 [1993], 90).

Research on social movements has also looked at the social characteris-
tics of activists from at least two points of view. First, it has often been ob-
served that the new social movements recruit in a specific social base, made
up mainly of some components of the middle class. Second, to account for
the overrepresentation of young and student activists, I used the conception
of biographical availability to point to the circumstances that increase free
time and reduce family responsibility, reducing constraints on participation
in movement actions (see McAdam 1988). Because of adolescent enthusi-
asm and the lack of social inhibitions, young people are also expected to be
more radical in their forms of action.

To what extent does an adapted version of social centrality explain the
degree and forms of participation in a highly active population such as the
one we have surveyed, which is, however, quite heterogeneous at least (and
not only) for gender and age? To answer this question, I have crossed our
dependent variables with gender, age, professional situation, education, and
religion. 6 Our data tell us that, in our highly mobilized population, socio-
biographical data have little explanatory value in accounting for the frequency
and repertoire of participation (see Table 7.3). Gender, class, and religion are
unable to tell us how much and in which forms demonstrators have mobilized
in the past. As for frequency of participation, the only significant correlation
we found is that the lower the age, the higher the frequency of previous par-
ticipation (rho = .04). Notably, however, not only the strength of correlation
but also its direction changes in different countries: participation increases
with age in Belgium but decreases in Germany and the Netherlands, and age
is irrelevant in the remaining countries. Education has some effect only in

I

Table 7.3. Correlations between socio-demographic characteristics
and forms of action (Cramer's v)

Semi-

Political

conventional

Demonstrative

consumerism

Disruptive

Violent

Sex

.01

.05

.07

.04

.06

Age

.06

.09

.15

.26

.10

Profession

.09

.06

.11

.25

.07

Working sectc

r .13

.11

.18

.16

.08

Education

.08

.06

.04

.12

.07

Religion

.05

.05

.04

.04

.04

H$H

-4

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 122^_

PATHS TO THE FEBRUARY 15 PROTEST I29

Belgium and Switzerland. Sociodemographic background has no explanatory
power as far as first timers are concerned: they have the same gender, age,
professional, and educational characteristics as other demonstrators.

Sociodemographic characteristics (with the exception of gender, which
seems to play absolutely no role) are slightly more useful for explaining the
preference for some forms of action — not so much conventional (used by
almost all marchers) or violent forms (used by almost none), but political
consumerism and disruptive forms. As expected, the last are more popular
among the young and students (especially among Italians and Americans,
less among Spaniards). Age and profession also have some explanatory power
as far as political consumerism is concerned.

Political Attitudes, Networks, and Participation

While sociobiographical data tell little about protestors' frequency of par-
ticipation and choice of repertoire, more relevant results emerge if one looks
at some specific attitudes, as well as membership in associations and protest
groups (a sort of social capital for protest) and previous experiences with
movement activism on causes congruent with the targets and demands of
the February 1 5 demonstrations.

Those who marched against the war in Iraq had frequent and intense
previous experiences with various forms of political participation — living tes-
timony for the "participatory revolution" that social scientists had already
identified in the 1970s and has been reassessed in the 1990s (Meyer and Tar-
row 1998; Norris 2002). After the path-breaking research of the 1950s and
1960s, more recent research started to note increasing degrees and reduced
selectivity of participation. Particularly in Western Europe, at least until 1990,
political participation increased and percent of passive citizens declined
(from 85 percent in 1959 to 44 percent in 1990) (Topf 1995, 68). Addi-
tionally, while conventional participation remained stable, unconventional
forms have drastically increased, with decreasing inequality as far as gender,
age, and education are concerned. When labor mobilization seemed to de-
cline, research on the global justice movement pointed at a remobilization
of the class cleavage, with increased participation from the labor movement
(della Porta 2003).

Especially with the growth of political participation and the increase of
research on its less conventional forms, the debate about the degree and sources
of selectivity reemerged. Alessandro Pizzorno (1966) has already noted that
characteristics of political participation are rooted in the systems of solidar-
ity at the basis of the very definition of interest: interests can in fact be iden-
tified only with reference to a specific value system; values push individuals

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 30^

130 DELLA PORTA

to identify with wider groups in society, providing a sense of belonging and
willingness to mobilize. From this perspective, participation is an action in
solidarity with others that aims to protect or transform the dominant values
and interest systems; it therefore requires the construction of solidarity com-
munities within which individuals perceive themselves and are recognized as
equals. Political participation aims at this identity construction: before mobi-
lizing as a worker, I have to identify myself as a worker and feel that I belong
to a working class. Identification as awareness of being part of a collective us
facilitates political participation that, stated Pizzorno, "increases (it is more
intense, clearer, more precise) the higher is class consciousness" (1966, 109).
In this sense, it is not the social centrality mentioned by Milbrath (1965)
but the centrality with respect to a class (or group) — as linked to identifica-
tion with that class (or group) — that defines an individual's propensity to
political participation. This explains why some groups composed of individ-
uals who are endowed with low status are able, under some conditions, to
mobilize more effectively than other groups. Participation is therefore ex-
plained by not only individual but also collective resources.

While the construction of a collective identity is a precondition for action,
it is also a consequence of it. In fact, participation itself changes individual
I identity, increasing the sense of belonging to some groups and weakening

~W other potential identifications. In collective action, identity is produced and ~W

reproduced (della Porta and Diani 2006, chapter 3). Revolutionaries' barri-
cades, workers' strikes, students' occupations are actions oriented to influence
public authorities and also have an internal function of strengthening class
consciousness — or, in more modern words, collective identification. Partic-
ipation, then, in a sort of virtuous circle, strengthens the sense of belonging,
which promotes more participation. During action, participants identify not
only themselves (the "we" they identify with) but also the targeted adversaries
("them"). It is indeed in action that the process of cognitive liberation devel-
ops (McAdam 1988, della Porta and Diani 2006). And it is in the course of
actions that the struggle for the recognition of emerging identities evolves —
a recognition that is part and parcel of identity building (Pizzorno 1991;
della Porta, Greco, and Szacolczai 2000). Specific experiences of intense pro-
test participation determine subsequent choices, and militant identities de-
velop during long-lasting activist "careers" (McAdam 1988; della Porta 1995).
Persistent activism has been explained as an effect of attitudinal availability
(deriving from socialization) and situational availability (flexible careers, part-
time jobs, no children), but also of the reflexive effects of movement partic-
ipation in terms of strengthening an alternative value system (Downton and
Wehr 1997).

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 31 (fo

PATHS TO THE FEBRUARY 15 PROTEST I3I

Our data can be useful in testing the extent to which the degree of par-
ticipation increases within the processes of development of specific political
values and in various types of movement organizations and activities. In this
direction, they can help us to determine the mobilization potential of the
peace demonstrations and identify how beliefs, networks, and previous activ-
ism influence patterns of participation. That peace demonstrations have a
quite heterogeneous social and political base could be related to the location
of the peace marches in various national social movement sectors — particu-
larly left-libertarian families and labor movements — with related effects on
the construction of collective identities. In particular, in some countries the
February 15 marches can be located within periods of intense protest, mainly
focused around global issues. In Italy and Spain, and also in Switzerland and
Belgium, the organization of transnational counter-summits contributed to
the identity construction of the global justice movement, which then provided
a basis for bloc recruitment in the peace movement. The February 1 5 protest
therefore inherited from the global justice movement mobilizations their
nationally specific characteristics (della Porta 2007): for example, the larger
presence of the labor movement in Italy and Spain, and of left-libertarians
in Switzerland and Belgium. In the United States and the Netherlands, in
st, the marches seem to have remained more within the tradition of

| contrast, the marches seem to have remained more within the tradition of

"W" national movements. ~®~

Looking at the various components of the development of (broadly de-
fined) collective identities as movement activists, we have therefore crossed
our dependent variables with two sets of variables indicating political atti-
tudes (political interest in politics; location on the left-right continuum; judg-
ment about parties, police, and the global justice movement; trust in various
actors); membership in various organizations; participation in protest events
by different types of movement (not the frequencies of participation in dem-
onstration, which is our dependent variable).

In general, our analysis confirms the role of some values and beliefs in
explaining prior mobilization. First, in all countries, the most committed hard
core of demonstrators is formed by committed leftists. If we look at the rho
value (which indicates the strength of correlation), in particular, high level of
participation is related to political position: The farther left the respondent,
the higher the previous participation in demonstrations (rho = .31). Partic-
ularly, the mobilizations on global issues have constituted the main reservoir
of committed activists. In fact — and, again, consistently for all countries —
a relevant explanation for a high level of participation is the demonstrators'
attitude toward the global justice movement: the higher the sympathy for the
movement, the higher previous participation in demonstrations (rho = .28).

H$H

-4

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 32^

132 DELLA PORTA

In the same direction, the more the respondent agrees with the s
"For me, this antiwar protest is another way to express my feelings against
neo-liberal globalization," the higher the amount of his or her previous par-
ticipation in demonstrations (rho = .22). The connection between trust in
social movements and participation in demonstrations is also confirmed by
the answers to a battery of questions addressing trust in civil society: the
higher the trust in civil society, the higher the previous participation in dem-
onstrations (rho =.16).

Previous participation is also strongly correlated with attitudes toward
police. In fact: the less necessary police attendance is considered, the higher
the previous participation in demonstrations (rho = .28); and the lower the
trust in the police, the higher the previous participation in demonstrations
(rho = .29). Demonstrators with high levels of participation had negative
judgments of police in all countries, confirming that the more frequent the
interaction with police, the less the confidence in them. It is also confirmed
that demonstrators are not antipolitical people; to the contrary, the more com-
mitted they are to demonstrating, the more they express interest in politics. In
fact, less strong, but still relevant (and relevant for all countries), is the cor-
relation of frequency of participation with interest in politics (the higher the
interest in politics, the higher the previous participation in demonstrations;
rho = .20). Confirming a tense relationship with institutional politics, the
more participation increases, the more trust in political institutions (rho = .15)
and other institutions (rho = .16) is lowered. 7 This is particularly true in Italy
and the United States, but also in the Netherlands, Belgium, and Spain.

Additionally, the more protest-experienced the marchers, the less they
expressed antipolitical feelings. There is no systematic correlation between
agreement with a statement such as "Political parties are only interested in
my vote, not in my ideas and opinions" and previous participation in demon-
strations (rho = .02). Items addressing sense of personal or collective efficacy
("When people like myself voice opinions to politicians, these are taken into
account"; or "Political parties are only interested in my votes") are unable to
explain the differences in frequency of previous participation in most coun-
tries (and sometimes the existing correlations have different directions in
different countries) — probably because they tend to mix criticism of "old
politics" and sense of efficacy of politics "from below."

The data tells that participation feeds on itself but also that partici-
pation in some types of demonstrations (such as the peace movement, but
also demonstrations on social rights and global justice) explains February 15
marchers' degree of participation especially strongly. To compare the impor-
tance of the type of previous demonstrations, it is sufficient to look at the

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page133^_

PATHS TO THE FEBRUARY 15 PROTEST I33

absolute z- values in the test statistics of separate Mann- Whitney U tests. Pre-
vious participation in peace demonstrations had the strongest effect on the
number of previous demonstrations (that is, peace demonstrators are frequent
demonstrators), whereas previous participation in regionalist demonstrations
has the least effect (that is, regionalists are not very frequent demonstrators).
Particularly interesting is the relevant role of participation in the global justice
and social or labor demonstrations. There are, however, telling differences
among countries: previous participation in demonstrations for global justice
is a very good predictor of high levels of participation in Italy and Belgium
(although in both countries antiracist demonstrations are even more correlated
with participation at the February 15 demonstrations), as well as in the United
Kingdom; but also in Switzerland, Spain, and the United States (where, how-
ever, the best predictor is participation in social or labor demonstrations); and
much less in the Netherlands (where antiracist, human rights, and environ-
mental demonstrations come first). Table 7.4 gives the order of importance of
the single types of demonstrations in each country as explanation of frequency
of participation in the February 15 protests. Similarly, together with member-
ship in peace movement organizations, membership in global justice organi-
zations is also relevant in explaining the degree of previous mobilization.
| In summary, the most committed demonstrators are leftist, with sym-

~W pathies for the new global movement and low confidence in the police, and ~W

with high interest in politics but mistrust of institutions. But can the same
variables also explain the choice of specific repertoires? As mentioned, while
almost all demonstrators used conventional forms of action and almost none
used violence, there are interesting differences in the use of disruptive forms
of action and of the specific repertoire of political consumerism (data not
shown but available on request). Those having experience with the most dis-
ruptive forms of action are, in general, more likely to be leftists and to have
participated in demonstrations with various goals (particularly global jus-
tice, but also third world and labor; the only exception are the women's and
regionalist movements). Moreover, in parallel with what has already been said
for degree of political participation, the use of disruptive forms of action is
also related to (negative) judgments about police attendance at demonstra-
tions and distrust in political institutions; this does not, however, reduce con-
fidence in the capacity of "people like myself" to have an impact on politics.
Participation in student organizations has a high explanatory value. A simi-
lar explanatory model emerges for the use of political consumerist repertoires,
which are also positively correlated with interest in politics and membership
in church, environmental, charitable, human rights, and peace organizations,
as well as participation in new global organizations.

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page134_^_

ri 2 §

"4

I

2

t- 1 -3 1 J

c ^c

^1

!8 £

[llj

1*

£ 8

- § 2 I"

ill!

Nil

H$H

-4

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 35^

PATHS TO THE FEBRUARY 15 PROTEST I35

Also in this case, however, differences emerge among countries, which
could be interpreted by looking at the differences in the social movement
sectors. For instance, in countries with (relatively) weak global justice move-
ments, Political consumerism seems to have remained a typical repertoire of
religious groups and voluntary associations active particularly on third-world
issues. In contrast, in countries such as Italy or Belgium, a very similar model
accounts for the use of both disruptive (but not violent) repertoires and boy-
cotts, or conscientious consumption. Also very dissimilar cross-nationally is
the impact of membership in environmental or labor organizations, which
in some cases increases and in others decreases (or has no effect on) the ten-
dency to use disruptive forms of action: for instance, in Italy, membership
in environmental organizations does not affect the propensity to use disrup-
tive forms of action, as it does in Germany; in Belgium, Switzerland, the Neth-
erlands, Spain, and the United Kingdom, membership in labor movements
has no effect on the use of disruptive repertoires, as it does in Italy or the
United States.

If one compares the explanatory capacity of the two models as a whole,
"political attitudes" (defined in terms of political attitudes and experiences
of mobilization, leaving out participation in specific types of demonstrations
in order to avoid the risk of endogeneity) appears much more effective than
"social centrality" in explaining both frequency of participation at demon-
strations and repertoire of political participation (see Table 7.5). After stan-
dardizing the number of variables included in the two models (which were
originally six for the first model and twenty-nine for the second) by running
a factor analysis for reducing the twenty-nine original variables in six latent
factors, to deal with mixed-scale-level models, ordered logit models and clas-
sical logit models were calculated. 8 Since the numbers of respondents were
highly unbalanced among the countries, every single analysis was conducted
by "blocking" for the countries (i.e., giving equal weight to each country).
In conclusion, in all analyses, the model based upon activists' attitudes and
networks was superior to the social centrality model, which has significance
only for the explanation of disruptive behavior; even in this case, however,
it performs worse than the second model. This is, above all, the case for the
dependent variable "frequency of demonstrations."

Conclusion

Our analysis of the degrees and forms of participation indicates that most of
those who demonstrated on February 15 had used a large repertoire of political
participation. While most participants had experiences with conventional
forms of participation, they often combined these with less conventional forms

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page13fi^_

"4

DELLA PORTA

Table 7.5. Comparison of the two models via ordered logit models
(pseudo R squares)

Frequency of demonstrations
Semi-conventional forms
Demonstrative
Political a

Violent a

Note: The adjusted R squares indicate how well the model explains. The cell entries can be read as
shares. All values are significant at least at the .01 level.

and sometimes with disruptive forms of action — but almost never with vio-
lent repertoires. They are neither troublemakers nor inexperienced protes-
tors. To the contrary, the national characteristics of the February 15 marches
(and marchers) seemed to be influenced less by political opportunities (dis-
tribution of institutional power and coalition in government) than by the (step
and degree of) development of other convergent movements, in particular
those on global issues.

The differences in the social and political base of the peace marches called
for an examination of the degrees and forms of participation in terms of par-
ticipants' social backgrounds and political attitudes. My analysis indicated
that the social centrality of the individuals is not particularly telling in terms
of frequency of participation (with the exception of age, which is inversely cor-
related with degree of participation) and only partially related to the forms of
action (again, with young and students more likely to use disruptive forms of
protest and religious behaviors sometimes related to "political consumerism").

The attitudinal model provided stronger explanations. High degrees of
participation emerged as significantly and strongly correlated with being left-
ist and identifying with the global justice movement as well as related mobi-
lization (on social and labor issues, third- world solidarity or antiracism) . Strong
identification with and deep embeddedness in movement networks are the
best predictors of degree of participation. Marchers are not antipolitical: they
have an interest in politics and believe in their capacity to affect public deci-
sions. However, they are very critical of institutional politics. Similar constel-
lations of variables — although with significant cross-national differences —
also explain the use of more innovative (political consumerism) and/or risky
(disruptive) forms of action.

The dynamics of collective activism cannot be understood simply in terms
of individual choices. If persistent activists keep their militantism during
periods of latency of protest, thereby nurturing movements' identities (Melucci

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page13Z^_

PATHS TO THE FEBRUARY 15 PROTEST I37

1989), individual paths are strongly influenced by macro trends in the evo-
lution of social movements. Experiences with social protest produce gener-
ations of activists, but the form of their commitment is influenced by the
shifting involvements of public versus private concerns, or, more directly, by
cycles of protest (Tarrow 1989).

Notes

1 . A methodological note is in order. In this analysis, I have used different cor-
relation indexes, according to the level of scale of the variables. In particular, I have
used the "Spearman Rank Order Correlation" (or simpler: "Spearman's rho") for
two ordinal variables. Usually, little variance in one of the two decreases the values
of correlation indexes; as in my an: ibles very often only take on about
three values, and the rho values tend to be quite low. However, the tests of signifi-
cance that have been developed for Spearman's rho work well, no matter how much
(or little) variance can be observed in the variables, and no matter how many cases
are involved. For nominal data (cross tables) I used "Cramer's v," a standardization
of the more commonly known "phi coefficient," to get values between -1 and 1 and
to make results comparable to each other. Respecting the ordinal character of most
of the variables, I have also used Mann-Whitney U tests (for dichotomous inde-

| pendent variables) and Kruskal-Wallis tests (for independent variables with more 1

*9)~ than two values). The classically known correlation, going back to Bravais and Pear- v®r

son (also known as "product moment correlation"), is of little use, since it should
only be applied if both variables have interval scale level, or, in other words, if both
variables are truly quantitative. This was hardly ever the case in our survey. I have
also abstained from other methods (such as analysis of variance or t-tests), as they
assume a meaningful calculation of means, which is not possible for ordinal values.

2. Values are estimates, based on assumptions of equal distribution within the
categories. Similar results come from the mean rank (Kruskal-Wallis test), which indi-
cates a quite varied picture, with Italy showing a very strong level of participation
(in first place, with a mean rank of 3928), followed by Belgium, Spain, Germany,
and Switzerland (with quite similar values, between 3219 and 3023), and then the
United States (with 2859) and then, by quite a distance, the United Kingdom (2283)
and the Netherlands (with 2137).

3. In fact, we have extracted four factors which largely confirmed our catego-
rization: Factor 1 largely corresponds to semi-conventional forms (including con-
tacting a politician, a civil servant; contacting/appearing on the media, contacting
an organization); in Factor 2 the main forms of political consumerism (buy a prod-
uct, boycott a product) converge; Factor 3 designs direct forms of action (with strikes
and sit-ins); Factor 4 the disruptive squatting and occupation of building.

4. To explain the Italian exceptionalism, I have considered the strategy of sampling:

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 1 38 afc

I38 DELLA PORTA

on the streets, in most countries; in part on the special trains leading to the march;
and in part at the march itself, in the Italian case. A comparison of the two Italian
subsamples (one with the questionnaires collected on the trains, one with those col-
lected at the demonstration in Rome) indicates, however, no significant difference
in any of the variables I have used in this chapter — with the only exception of par-
ticipation in environmental organizations.

5. In the United States, according to Lester Milbrath (1965), as many as 60
percent of the citizens had until the early 1960s just a minimal level of participation,
and 30 percent were totally uninterested in politics. Similar results on the United
States are presented in Verba and Nie (1972), and on the United States, the United
Kingdom, Germany, Italy, and Mexico at the beginning of the 1960s in Almond
and Verba (1963, 474).

6. Age groups have been recoded: 0-17, 18-24, 25-34, 35-44, 45-54, 55-64,
and 65+: religion in these: one group without faith, including declared free-
thinkers; one Christian group; and one religious, but non-Christian, group. In the
social centrality approach, belonging to the dominant religion should increase pro-
pensity to political participation.

7. In particular we recoded trust in political institutions (king, queen, or presi-
dent; public administration; local government; national government; national parlia-

| ment; European Union; United Nations); the police; other institutions (educational

~tyy system or legal system, political parties, church, press, banks); and civil society (trade ^S7~

unions, social movements or citizens' groups).

8. In the model of centrality, one variable was nominal dichotomous, four were
multinomial, and the sixth was interval. In the second model, eighteen nominal
dichotomous variables and eleven ordinal variables were brought together. This usu-
ally makes a multivariate estimation rather difficult, not least since the dependent
variables were not on an interval scale (which would have improved the situation),
but on an ordinal scale (frequency of demonstration activities) or a nominal dichot-
omous level (forms of participation, yes or no). Bivariate correlations between each
independent variable and each dependent one gave values very similar to those in
the aggregated model (data not shown, but available on request).

Works Cited

Agrikoliansky, Eric, and Isabelle Sommier, eds. 2005. Radiogmphie du mouvement

altermondialiste. Paris: La dispute.
Almond, Gabriel, and Sidney Verba. 1963. The Civic Culture: Political Attitudes and

rinceton, N.J.: Princeton University Press.
Andretta, Massimiliano, Donatella della Porta, Lorenzo Mosca, and Herbert Reiter.

2002. Global, Noglobal, New Global: La protesta contro il G8 a Genova. Roma:

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page132^_

PATHS TO THE FEBRUARY 15 PROTEST I39

. 2003. Global— New Global: Identitat und Strategien der Antiglobalisierungs-

bewegung. Frankfurt a.M.: Campus Verlag.

Axford, Barrie, Richard Huggins, and John Turner. 1997. Politics: An Introduction.
London: Routledge.

Bourdieu, Pierre. 1979. La distinction. Paris: Minuit.

Dalton, Russell. 1996. Citizen Politics in Western Democracy: Public Opinion and
Political Parties in Advanced Western Societies. Chatham, N.J.: Chatham House.

della Porta, Donatella. 1995. Social Movements, Political Violence, and the State: A Com-
parative Analysis of Italy and Germany. New York: Cambridge University Press.

. 2003. I New Global. Bologna: II Mulino.

Perspectives. New York: I

niliano Andretta, Lorenzo Mosca, and Herbert Reiter.
2006. Globalization from Below: Transnational Activists and Protest Networks. Min-
neapolis: University of Minnesota Press.

della Porta, Donatella, and Mario Diani. 2006. Social Movements: An Introduction.
Oxford: Blackwell.

della Porta, Donatella, Monica Greco, and Arpad Szacolczai, eds. 2000. Identita,
riconoscimento e scambio: Saggi in onore di Alessandro Pizzorno. Roma: Laterza.
| della Porta, Donatella, and Dieter Rucht. 1995. "Left-Libertarian Movements in Con-

■W- text: Comparing Italy and West Germany, 1 965-1 990." In The Politics of Social W

Protest: Comparative Perspectives on States and Social Movements, ed. J. Craig
Jenkins and Bert Klandermans, 229-72. Minneapolis: University of Minnesota
Press.

Downton, James Jr., and Paul Wehr. 1997. The Persistent Activist: How Peace
Commitment Develops and Survives. Boulder, Colo.: Westview.

Kitschelt, Herbert. 1986. "Political Opportunity Structures and Political Protest:
Anti-Nuclear Movements in Four Democracies." British Journal of Political Sci-
ence 16: 57-85.

Kriesi, Hanspeter. 1995. "The Political Opportunity Structure of New Social Move-
ments: Its Impact on Their Mobilization." In The Politics of Social Protest: Com-
parative Perspectives on Sta Movements, ed. J. Craig Jenkins and
Bert Klandermans, 167—98. Minneapolis: University of Minnesota Press.

Lagroye, Jacques. 1993. Sociologie politique. Paris: Presse de la Fondation Nationale
de Sciences Politiques.

McAdam, Doug. 1988. Freedom Summer. New York: Oxford University Press.

Melucci, Alberto. 1989. Nomads of the Present: Social Movements and 1
Needs in Contemporary Society. London: Century Hutchinson.

Meyer, David S., and Sidney Tarrow. 1998. "AMovement Society: Contentious Pol-
itics for the New Century." In The Social Movement Society: Contentious Politics

H$H

07 Chapter 7_Walgrave 22/01/2010 3:41 PM Page 140_^_

m

140 DELLA PORTA

for a New Century, ed. David S. Meyer and Sidney Tarrow, 1—28. Lanham, Md.:

Rowman & Littlefield.
Milbrath, Lester W. 1965. Political Participation: How andWhy Do People Get Involved

in Politics? Chicago: Rand McNally.
Milbrath, Lester W., and Madan Lai Goel. 1977. Political Participation: How and

Why Do People Get Involved in Politics? 2nd ed. Chicago: Rand McNally.
Norris, Pippa. 2002. Democratic Phoenix: Reinventing Political Activism. New York:

Cambridge University Press.
Pizzorno, Alessandro. 1966 [1993]. "Introduzione alio studio della partecipazione

politica." Repr., in Allesandro Pizzorno, Le n ica assoluta, 85-

128. Milano: Feltrinelli.
. 1991. "On the Individualistic Theory of Social Order." In Social Theory for a

Changing Society, ed. Pierre Bourdieu and James S. Coleman, 209—31. Boulder,

Colo.: Westview.
Rucht, Dieter. 1994. Modernisierung und neue soziale Bewegungen. Frankfurt a.M.:

Campus Verlag.
Tarrow, Sidney. 1989. Democracy and Disorder: Protest and Politics in Italy. 1965-

1975. New York: Clarendon.
Topf, Richard. 1995. "Beyond Electoral Participation." In Citizens and the State, ed.

Hans-Dieter Klingemann and Dieter Fuchs, 52-91. New York: Oxford Uni-

Verba, Sidney, and Norman H. Nie. 1972. Participation in America: Political Democ-
racy and Social Equality. New York: Harper and Row.

Verba, Sidney, Norman H. Nie, and Jae-On Kim. 1978. Participation and Political
Equality: A Seven-Nation Comparison. New York: Cambridge University Press.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 1 41 afc

Boon or Burden?:

Antiwar Protest and Political Parties

m

The enormous size of the demonstrations on February 15, 2003, gives rise
to the question of whether this marks a turning point in the relationship
between protest and parties. Political protest has sometimes been seen as a
challenge to representative democracy and the dominant role of political
parties in aggregating political demands. Others regard protest behavior as
just one of a range of forms of political behavior that also includes voting
and working within political parties. The question thus arises of whether the
unprecedented mobilization over the Iraq War has provided political parties
with a new challenge or a new opportunity: did the demonstrations signify
more alienation from party politics, or did they provide a fertile recruiting
ground for parties in search of new supporters?

Peace protests have traditionally been associated with parties of the Left.
Past debates have, in particular, seen conflicts between the main parties of the
Left, such as Labor, Social Democratic, or Socialist parties, and smaller par-
ties competing for left-of-center voters: this includes Communist and former
Communist parties as well as various parties of the New Left that emerged
out of the student and ecological protest movements of the 1960s and 1970s.
In a few cases, peace issues were of key importance for the formation and
development of small parties, such as in the Netherlands and Germany. Lib-
eral parties also supported key demands of the peace movement, such as, in
Britain and the Netherlands, opposition to the deployment of medium-range
nuclear missiles in the early 1980s. Overall, the political spectrum of parties
that potentially seize on "peace" issues is thus quite wide, including liberals,
greens, social democrats, socialists, and communists (Rochon 1988, 156-78).

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 142^_

142 RUDIG

The events surrounding the campaign against the Iraq War could be
seen as just the latest installment of a long-running saga of party competi-
tion on peace issues. However, the political constellation in some cases has
changed rather radically. Of particular interest must be the reversal of the sit-
uation between Britain and some continental European countries. While the
British Labour Party spearheaded the antinuclear campaign of the early 1980s,
Social Democratic governments in Germany and elsewhere supported the
stationing of nuclear missiles. In 2003, Socialist and Social Democratic par-
ties in continental Europe were critical of the Iraq War, while the Labour
Party supported the U.S. -led war movement.

In terms of party competition, the main issue is thus the extent to
which small, leftist parties have been able to attract supporters who other-
wise would associate themselves with the main party of the Left. But apart
from choosing between one party and another, the question arises as to what
extent political parties feature at all in the political world of the peace
demonstrator. In this chapter, I will address two choices: that between party
and nonparty political activity and that between support for one party or
another.

The first specific questions to be addressed concern the role of political
/fa parties before and during the demonstration. What was their role for the /is

I mobilization of demonstrators? How important were parties — as compared

to nonparty organizations — in the process of people turning up for the
demonstrations?

The second question concerns the impact of the antiwar movement in
general and the February 1 5 demonstrations in particular on party politics.
What choices about their association with political parties did demonstrators
express in terms of future electoral participation, voting preference, and pos-
sible future party membership? And to what extent are these choices reflected
in the actual development of party preferences and membership following
the 2003 antiwar protest?

These questions raise a number of theoretical issues that will be ex-
plored in the next section. After this I briefly discuss the data I will use to
examine these questions and then present the results in two main empirical
parts: The first will look at the background of demonstrators and their pre-
vious political affiliations. The second one will analyze any changes of party
political orientation the antiwar movement has precipitated and look at the
impact of these processes on the electoral fortunes of antiwar parties as well
as their possible success or failure in attracting demonstrators as new mem-
bers and activists.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 143^_

BOON OR BURDEN? I43

Theoretical Reflections

Political parties are said to be in crisis: Party membership is declining in most
Western European countries. Moreover, the attachment of voters to the major
political parties has declined sharply. What people think about political
issues is less and less dependent on the views held by "their" parties. Parties
are thus losing their grip on the political process. But how correct is this
assessment? Are we really seeing a general decline in party affiliation or, rather,
a reshaping, a realignment of political action, including party politics?

The evidence on the current state and future of political parties is rather
contradictory. On the one hand, there is a lot of evidence that suggests a
process of political alienation is in progress, with decreasing electoral turn-
out (Franklin 2004), decreasing party membership (Scarrow 2000; Mair
and van Biezen 2001), and reduced trust in governments and parties (Dal-
ton 2004). On the other hand, there is also evidence that this decline is by
no means even. Some elections attract high turnouts; some parties have man-
aged to increase members. The thesis of a linear road to apathy regarding
party politics might thus be premature (Norris 2002).

Two different aspects of the alienation literature must be distinguished.
. The first concerns alternative forms of political participation that bypass or

Afr- challenge the dominance of parties. The theme of "antiparty" sentiment be- .fa.

I longs here (e.g., Seyd 1995; Poguntke 1996) as well as the issue of party link-

age (Lawson and Merkl 1988; Poguntke 2002). Do parties lose their ability
to link the concerns of citizens with the decision-making process? Is there a
fundamental rejection of political parties that challenges not just individual
parties but the whole system of decision-making based on elections con-
tested by political parties? Faced with greater competition from interest groups
and social movements for the attention of activists, parties may lose out in
such a "market of activism" (Richardson 1995).

The second dimension has to do with party competition. Much of the
literature on alienation from parties is essentially concerned with the major
parties, which face challenges from within the party system in the form of
small and/or new parties. Disillusionment with the large parties should pro-
mote the rise of new and small ones, either to challenge the large ones to
reform and respond to popular demands or to replace them altogether. In this
context, protest movements are not primarily challengers to the dominance
of parties per se but they act as a conduit for the spawning of new political
parties and/or the promotion of small parties to provide the basis of a new
party system and/or a new relationship between parties and civic society. The
protest movements of the 1960s and 1970s, and the rise of new parties like

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 144_^_

the Greens in the 1980s and 1990s were the early foci of this literature (see
Dalton and Kuechler 1990; Muller-Rommel and Pridham 1991)

Considering the publication dates of some of the literature referred to
so far, it is clear that this debate is not particularly new. Successes of small
parties in more recent elections have revived interest in their study in some
countries, for example, in Germany (Jun, Kreikenbom, and Neu 2006), but
the trend of reduced turnout and declining party membership started some
time ago. New movements and parties have emerged at fairly regular inter-
vals since the 1960s or 1970s. Despite facing all these challenges, the estab-
lished parties have not crumbled — small parties on the whole have remained
small; many new parties have risen, fallen, and disappeared.

One explanation may lie in the way the large parties have responded. In
fact, the major parties have been fighting back, reorganizing and restructur-
ing to stabilize their operations in the face of new challenges. Katz and Mair's
(1995) theory of the "cartel party" is perhaps the best known analysis of this
fighting back. The flipside of this is how stable an electorate the challenging
parties can rely on. At least for the "new politics" parties that have arisen
since the 1970s, Inglehart (1990) predicts problems in securing a stable elec-
torate on the basis of the notion of "cognitive mobilization" (see also Dal-
| ton 1984). The highly educated "new politics" voter sees no need to form

~W firm ties with one party but prefers to decide election by election, issue by ~W

issue, policy by policy, which party to vote for. The result appears to be a sys-
tem of party competition where the large parties do not have the security of
old and need to fight harder to defend their position and the smaller parties
have more to gain but remain on fairly shaky ground, with little inherent
stability.

Inglehart's basic analysis has changed little over the years. The spectac-
ular upsurge of antiwar protest accompanied by a loss of support for mass
parties and increasing challenges from new and small parties in the early
twenty-first century seems to follow the same basic principles first used to
diagnose the student protest of the 1960s and the rise of new social move-
ments in the 1970s and 1980s (Inglehart 1977, 1990, 1997). How well do
protest movements against the Iraq War fit into this scheme?

The failure of previous movements to challenge the predominance of
parties in political life should make one cautious about readily embracing an
"antiparty" argument. Looking at the issue in question here, party systems
could be expected to be well positioned to absorb this new wave of activism.
While most previous movements — such as the student movement, the anti-
nuclear movement, and also the feminist movement — faced a phalanx of
hostile parties when they first emerged in the 1970s in most countries, the

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 1 45^

BOON OR BURDEN? I45

r position in 2003 found a ready home in several parties in all countries
covered. In Western Europe, "peace" and "war" have historically been linked
to the Left, and have divided the Left on numerous occasions. Throughout the
twentieth century, different parts of the Left located themselves on the con-
tinuum between radical pacifism and support for war in a series of different
contexts, with the antiwar position usually associated with the left-wing part
of the spectrum and the pro-war position linked to the right of the Social-
ist/Social Democratic spectrum. The peace movements of the 1950s were
strongly based on the traditional left; the New Left that arose with the student
movement and took on a strong antiwar message in opposition to Vietnam
added a second dimension. As a result of these developments, the antiwar issue
could be expected to be picked up by at least two party families: One is the
Old Left, which covers both the left-wing part of any Socialist or Social
Democratic party and any parties farther to the left, in particular parties that
stand in the tradition of Communist party organizations. The second party
group open to antiwar campaigning are the New Left parties, arising from
the student movement and new social movements of the 1960s and 1970s.

As the party system is thus better placed in the 2000s to absorb peace

mobilization than it was in earlier protest cycles the focus of attention then

| shifts to the question of which parties will benefit from this phenomenon.

W The key battleground is, of course, between the main party of the Left and v9)~

the "challenger" parties of the Old and New Left. What factors are likely to

play a role here?

I broadly distinguish between external, factors outside the control of
Old or New Left challenger parties, such as electoral systems and the policy
positions of the main parties, and internal factors, consisting of attributes of
the parties themselves.

To start with external factors, the position of the main party of the Left
clearly could be expected to be of major importance. We should expect that
Old or New Left challenger parties would be particularly successful in the
United Kingdom, where the Labour Party has not only moved sharply to the
right in recent years over economic and social issues but also provided strong
political and military support for the U.S. -led war in Iraq. In contrast, the
main center-left parties in the other countries invariably opposed the war,
either in government (Belgium, Germany) or in opposition (Spain, Italy, the
Netherlands, and the United States). Small challenger parties should find it
more difficult to mobilize effectively on the antiwar issue if the main left-
wing party is also committed to the same agenda.

While Green and Far Left parties could thus be expected to benefit, we
have to ask the extent to which antiwar mobilization has transcended the

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 14fi^_

146 RUDIG

realm of left-wing politics. Unlike opposition to stationing nuclear missiles
in the early 1980s, which was supported only by a minority (Flynn and Rat-
tinger 1985), the Iraq War was unpopular among large sections of the pop-
ulation throughout Western Europe. While those radically opposed to any
kind of military action might have been a minority, the perceived lack of
United Nations sanction for the war potentially broadened the range of recruits
to the antiwar movement across the entire left-right spectrum. This is shown
empirically in chapter 3, where Joris Verhulst and Stefaan Walgrave establish
that even among right-wing voters war opposition was widespread. There-
fore, if the antiwar movement involves a broader political spectrum, mem-
bers of other parties might enter the picture as potential players. In Britain,
it was the Liberal Democrats as well as Welsh and Scottish Nationalist par-
ties who expressed opposition to the war and took part in the demonstra-
tions. Protesters and voters thus had the choice between no fewer than four
left-of-center antiwar parties in Scotland. Among the other countries, only
in Spain could we expect a stronger role for regional parties, but as separate
demonstrations in regional centers were not covered; their influence is unlikely
to show in the Madrid data.

Does it matter whether the main party of the Left, taking an antiwar
position, is in government or in opposition at the time of the demonstration?
This is quite a tricky issue: as far as antiwar voters are concerned, a rational
choice would be to support the antiwar party most likely to be able to recap-
ture government and change government policy. This should benefit the
major party of the Left, and thus we might expect less support for Old or
New Left challenger parties where the main party of the Left is in opposi-
tion to the government. However, calculations are made more complex if we
consider the process of government formation: small parties that are likely
coalition partners of the major party of the Left are less likely to experience
a loss of support to the major party. The Italian Greens fall into this cate-
gory while the Spanish United Left as well as the Dutch Socialist Party and
Green Left party were not seen as likely future coalition partners.

In both cases where the main parties of the Left were in government
opposing the war, Green parties were also part of the governing coalition
(Belgium, Germany). One might argue that a vote for the smaller a
parties might send a strong signal to the main party of the Left to n
its antiwar position. At the same time, without a government to oppose, the
stakes of the antiwar issue for left-wing politics could be expected to be much
lower. Indeed, demonstrations were largest in countries where the govern-
ment supported the war (Britain, Spain, the Netherlands, Italy) as was shown
in chapter 1 of this volume.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 14Z^_

BOON OR BURDEN? I47

Calculations that might be sensible for voters, however, do not neces-
sarily apply to demonstrators. For a low-saliency protest issue with no strong
domestic polarization, in particular, one might expect that only those very
committed to the issue would make the effort to come out on a demonstra-
tion. This would lead one to expect that demonstrators in countries whose
government opposed the war are more likely to represent the "usual suspects"
of committed peace activists, that is, those partisan or at least sympathetic
to the parties of the Old or New Left. On the other hand, in high-saliency
protest demonstrators, one would expect a far larger cross-section of the
population to be mobilized. In Britain, Spain, Italy, and the Netherlands, we
could thus expect the political-party profile of demonstrators to be less likely
to be dominated by small parties of the Old or New Left.

Turning to matters of the parties themselves, differences in the posi-
tioning of parties on the war issue seem to be a poor predictor of attracting
support. Antiwar parties were generally opposed to the Iraq War under any
circumstances, with or without a UN Security Council resolution. For anti-
war protesters, there was little to choose here in most cases, the only excep-
tion is the British Liberal Democrats, who were rather reluctant to take part
in the demonstration and were less vociferous in their condemnation, focus-
| ing very much on the UN Security Council issue in their arguments.

"W" With little or no policy differences between and Old Left and Green "®"

parties on the war issue, what could account for their relative fortunes? One
possible factor is the strength of each party group. Green parties have been
relatively weak in Southern Europe (Italy and Spain), leaving the Far Left in
a better position to benefit. Greens have played a very important role as chal-
lengers to the main party of the Left in Belgium, Germany, and Switzerland,
with antiwar positions being significant, and thus we could expect the Greens
to be high-profile beneficiaries of antiwar mobilization in these countries.
Finally, neither Greens nor Far Left parties have been particularly strong
forces in either British or U.S. party politics. However, the strong rightward
movement of the British Labour Party combined with constitutional and
electoral reform offering new opportunities for small parties in European
and regional elections has triggered increased support for Green parties as
well as a revival of the Far Left, in particular with the Scottish Socialist Party.

Beyond taking a specific position on the war, other political choices could
affect the attractiveness of these parties to antiwar protesters. While some
small parties seem happy to be perpetual challengers, others have sought to
pass through the party lifecycle and move from outsider status to be con-
sidered potential coalition partners in government (see Rihoux and Riidig
2006). The German Green Party is perhaps the most obvious example; it

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 14S^_

148 RUDIG

grew in prominence at the height of the antinuclear movement of the early
1980s and made nonviolence part of its political identity. By the late 1990s,
however, it had gone through a transition process that saw it change from an
opposition party rejecting any form of German military intervention abroad
to a government party that backed the first combat mission of the German
military since World War II in the 1999 Kosovo conflict (Riidig 2002).
Having, in the view of some, thus disowned its roots in the peace move-
ment, it will be interesting to see to what extent the party can command the
support of antiwar activists on the Iraq War issue, which sees Greens back
in the antiwar camp.

What, then, can we expect? First, members associated with the Green
and Far Left parties can be expected at demonstrations, as these parties gen-
erally try to play prominent roles in the organization and mobilization of
antiwar protest in all countries concerned. Data on the presence of party
members will give us some clues as to their relative importance. Their national
membership numbers are normally quite low, so we cannot expect huge fig-
ures. As they are traditional mass parties, it will be interesting to see how
well antiwar social democratic or socialist parties can mobilize members to
attend demonstrations.

Looking to future political action, how successful will these small par-
ties be in attracting voters and members from the pool of potential new activ-
ists that present themselves in these demonstrations? Much of the media
hype has been about how people who never protested before were moved to
turn out on this occasion. The results presented in chapter 7 show that the
number of protest novices was not quite as high as some media reports sug-
gested, and quite a few people who had demonstrated years before were
mobilized again. But despite this, small parties clearly must have seen the
demonstrations as a major recruiting ground: how successful were they?

This is a crucial question in the context of the debate over the future of
party politics. If people are motivated to come out and demonstrate in the
hundreds of thousands, are they happy to leave it at that, or is this a step-
ping stone for a reinvigoration of party politics with a generation of activists
politicized in the antiwar campaign going beyond one-off single-issue politics
to become immersed in politics involving both unconventional and con-
ventional forms of action? In the absence of panel data on the political pref-
erences of peace protesters over time, there are two ways one might examine
their party political impact. First, to what extent are the changes in party
preferences expressed by peace demonstrators typical of the electorate as a
whole? Did antiwar protesters form a vanguard that led the way in terms of
electoral realignment? Second, to what extent were parties able to attract new

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 142^_

BOON OR BURDEN? I49

members from the ranks of the peace movement? The decision to join a polit-
ical party is likely to have a greater medium- to long-term effect than just
changing voting preferences. Based on membership data from 2002 and 2003,
I will assess the role antiwar protesters played in this regard.

Data

The main focus of this chapter is voters and party members. Excluded from
this analysis are thus demonstrators who were too young to vote in the last
general election. Also excluded are those ineligible to vote because they do
are not citizens of the countries in which they are demonstrating. As the
electoral choices are different in England and Scotland, in this chapter, unlike
in most other chapters in this volume, they are considered separate coun-
tries, for the purposes of electoral analysis. About 10 percent of demonstrators
in Glasgow came from England, and these were reallocated to "England."
In addition to the survey data on the demonstrators taking part in the
protest marches on February 15, 2003, I use a range of other data sources
(e.g., European Social Survey, World Values Surveys, general election data,
party membership data). One particularly important aspect is the comparison
of the demonstrators with the general population, and I use survey evidence
| from a variety of polling organizations to identify how typical demonstrators

~W~ were of a general trend in public opinion and voting choice. Party member- ~W~

ship data have been gathered mainly from party reports and accounts.

The Role of Parties in Mobilizing Demonstrators

A first indication of the involvement of political parties is the number of
party members that took part in the demonstration. We asked demonstra-
tors about their present and past involvement with political parties, offering
four options: active member, passive member, previous member and, lastly,
not a party member at present or in the past. In order to compare demon-
strators with the general public, we also compiled the results of a general
public opinion survey carried out shortly before the demonstration, the 2002
European Social Survey, which asked about current party membership as well
as the result of an attempt to compile national membership data on the basis
of membership information gathered from the parties themselves. The results
are displayed in Table 8.1.

Demonstrators are substantially more heavily involved with "conven-
tional" party politics than the general public. In Western Europe, between 1 1
and 27 percent of demonstrators were members of a political party at the time
of the demonstration. 1 The European Social Survey data as well as the data
compiled by Mair and van Biezen (2001) suggests that party membership is

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page150_^_

1

3
Q.

8.

■E

"4

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page15l^_

BOON OR BURDEN? I5I

an activity involving a rather small minority of the public at large. There are
some interesting individual results. Taking into account past party member-
ship reveals that about one third of demonstrators have been a member of a
party at one time or another. There is relatively little cross-national variation.
The only striking difference is that past party membership is rather more
pronounced in the United Kingdom than elsewhere. This might be an indi-
cation of a particularly severe decline in party membership in Britain, but
also of a high degree of mobilization on the Iraq issue there.

While we have no data on the parties that demonstrators were members
of, we did ask specifically about links demonstrators had to groups involved
in the organization of the demonstration. Most demonstrators were not mem-
bers of any of these groups, but a substantial minority was — and political
parties were one of the most frequently cited groups (see Table 8.2.).

Upon analyzing the importance of different types of groups, some inter-
esting patterns emerge. In the Netherlands, 17 percent of the respondents
were members of political parties that (co)organized the demonstration. In
no other country are political parties the most important type of organizing
group demonstrators were involved in. Parties play a more marginal role in
other countries: they are practically not mentioned at all in England and the
| United States. In Belgium, Switzerland, Germany, Spain, and especially Italy,

~W trade unions play the most dominant role, with parties less important but still ~W

mentioned more frequently than in the English-speaking countries. Chap-
ter 10 in this volume elaborates the organizational membership of the demon-
strators in more detail.

Parties may, of course, have not only mobilized members and those with
a strong sense of party identification, but also elements of the general pub-
lic with clear preferences or attachments to specific antiwar parties. While
we did not ask a question about party identification or the degree of attach-
ment to particular parties, we can use information collected about past vot-
ing behavior to gauge the relative mobilizing power of antiwar parties. A
look at previous voting behavior also should tell something about the degree
of alienation from the political process (see Table 8.3.).

If we exclude those who were too young and those who simply refuse
to reveal anything about their previous electoral behavior, it's clear that com-
paratively few demonstrators abstain from the electoral process on principle:
in total, only about 4 percent of demonstrators said they did not vote because
there was no political party available they wanted to vote for, and just 1 percent
indicated they did not vote out of a fundamental objection to representative
democracy. There is some limited cross-national variation: Twelve percent of
Swiss, 9 percent of Spanish, and 8 percent of English demonstrators cite a

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page152^_

"4

o

<u

O)

1

m

<u

I

■5

1

I

<n

S

1

Z

>

c

_D

-°^

t

c

.g

c

Si

£

55

c

o

o

s.

b>

o

Q.

3

s

z

j

C

3

o

E

<D

-o

*o

o

1

1

O)

c

•^

o

>

CN

i

E

od

1

25
•2

£■

4 i«5Slsj

8 3 § 'I I
2 1-23 j

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page153^_

i a.

>£

o

^2

"I

£

£

>§:

£

o

1

z

>§:

<u

L^^

"5

■c

o

^a

1

^

"O

e

8.

E

n_

>&

o

o

1

U

>£

<u

>»

L^^

S

<u

E

z

>a

a.

""

_D

>§:

o

Z5

jis

>

-o

_4)

"5

v

Of

co

_>>

'e

<u

^

-Q

>

£

£

8 Rs | ^&

Hill Ml

§j ill iM
i § •* 1 1 i-* §

lilll 111

1 1 1 1 o ■£ I c

I II El all

ll 1 It {

a g ^? u EL £ £
i||-ss.|j

H$h-

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 1 54^

154 RUDIG

lack of a suitable political party. A lacking belief in representative democracy
is indicated by 3 percent of Spanish and English demonstrators. Five per-
cent of Spanish and 4 percent of English demonstrators also said they had
spoiled their vote. There is a similar pattern in the responses to the state-
ment about not taking part in election for reasons of protest.

Very few demonstrators have not taken part in an election for reasons of
protest in Belgium, Netherlands, and Germany; Swiss voters with 6 percent
are already somewhat more inclined to "protest" by abstaining, on roughly
the same level as voters in the United Kingdom and the United States. Spain
and Italy top the bill with 10 percent electoral protesters. When one com-
pares past turnout differences between demonstrators and the general pub-
lic, it is generally much higher among demonstrators, suggesting that these
are people more politically active than the average population, even consid-
ering that recalled turnout in any survey is always somewhat higher than the
actual turnout. But, apart from Belgium and Italy, the differential turnout
is substantial: there is thus no evidence that demonstrators are alienated from
the electoral process: to the contrary, they are more involved in traditional,
conventional political participation that the general public (see also chapter
7 about previous activism).

Overall, there is little evidence here to suggest that antiwar demonstra-
tors display what Poguntke (1996) calls generalized antipartyism. Is "specific
antipartyism" as expressed in the rejection of major parties more common?
To gain a picture of the main pattern in party preference among demon-
strators in different countries, let us look at the vote in the last election (as
recalled by our respondents) and compare it with the actual election results
(Table 8.3.).

These data first of all show that the demonstrators were not a representa-
tive cross-section of the population. Voters of center-right parties are heavily
underrepresented. Very few demonstrators indicated having voted for parties
to the right of the main socialist or social democratic parties. Spain did have
a significant number of votes (more than 10 percent) for the conservative
party, the Peoples Party (PP); and in Belgium, 6 percent recalled having voted
for a Christian Democratic party. 2

Demonstrators were associated mostly with the main parties of the Left,
such as the Socialist or Social Democratic party, the Greens and parties of
the Far Left. The only major exception is the Liberal Democrats in England
and Scotland, and the Nationalists in Scotland. Both Liberal Democrats and
the Scottish National Party can be considered to stand to the left of the rul-
ing Labour Party, which sets them apart rather sharply from some far-right
wing regional parties in Belgium and Italy, for example, as well as most liberal

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 1 55^

BOON OR BURDEN? I55

parties in continental Europe, which generally have moved to the right (see
Benoit and Laver 2006).

There are a number of important differences in the respective impor-
tance of the main parties of the Left, the Greens and the Far Left. Social Dem-
ocratic voters are most dominant in Switzerland, followed by England, Spain,
Italy, and Scotland. Green Party supporters are most strongly represented in
Belgium, the Netherlands, and Germany. Far Left parties are most dominant
in Italy, the Netherlands, Spain and Germany.

In summary, demonstrators are thus clearly not individuals alienated from
party politics; to the contrary, they display very high levels of conventional
participation. Also, as shown in data on political interest (not reported here
in detail), demonstrators generally extremely pay attention to politics, again
far more than by the general public by a substantial margin in all countries
(except the Netherlands).

The Impact of February 15 on Party Politics

What happened after demonstrators had gone home? Did the experience
have any impact on their post-demonstration behavior? Ideally, I would dis-
cuss here the results of a follow-up of demonstrators months or years after
| the event, trying to trace their behavior and possible link to the demonstra-

~W tion. In the absence of this type of individual-level data, I will try to do two ~W

things: first, interpret the results of questions we asked the demonstrators
about their likely future political behavior, and second, turn to aggregate data
on party membership and voting behavior to gauge the possible impact of the
demonstrations. 3

First, did the demonstration lead demonstrators to be more alienated
from conventional party and electoral politics, or is the opposite the case: did
taking part in the demonstration strengthen the involvement in conven-
tional political participation? The data does not suggest that demonstrators
have been turned off the electoral process: just 2 percent overall indicated they
would not vote or issue an invalid vote. More of a constraint on my analy-
sis are those who said they did not know whom to vote for in the future: they
amount to 7 percent of demonstrators eligible to vote, with a particularly high
incident of voters uncertain about their future party choice in Spain (18 per-
cent), England (16 percent), Scotland (15 percent), and Belgium (13 percent).

Second, what were the voting intentions of demonstrators at a future
general election? Compared with their past voting record, there are some
changes but also some continuities. The main continuity perhaps is that also
on this score, demonstrators are not representative of the population as a
whole. Comparing the voting intention of demonstrators with those of the

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page15£^_

I56 RUDIG

general population, I find huge discrepancies in most cases (see Table 8.4).
Even more than in the case of past voting, voters of center-right parties are
almost completely absent. In some countries, this contrasts quite starkly with
the then-current trend of voting in favor of conservative parties, for exam-
ple, in Germany. Even in Britain, Labour's vote held up very strongly: there
is no sign that the Iraq War adversely affected its vote. Clearly, in the popu-
lation at large, the Iraq War was not necessarily the dominant issue.

Among demonstrators, the general picture is one in which the main party
of the Left loses support and smaller antiwar parties — such as the Greens,
the Far Left, and, in England and Scotland, the Liberal Democrats — pick up
additional support. However, the scale of the loss differs quite markedly; it is
highest in England and Scotland. But, importantly, Social Democratic par-
ties are also losing much support in other countries, namely in Switzerland
and in Germany. Losses are much smaller in Italy and the Netherlands, while
the balance is positive for the Belgian and Spanish Socialists. This might sug-
gest that the position of the Social Democrats on the war is not necessarily
a determining factor: not supporting the war has not protected the German
and Swiss Social Democrats from losing support to smaller parties among
demonstrators.

What, then, determines whether a demonstrator turned his or her back
on the main party of the Left and embraced smaller parties like the Greens
and the Far Left? I can try to predict demonstrators' intended votes with ref-
erence to a variety of attributes: one might expect that young, cognitively
mobilized voters might be more willing to turn to smaller parties; demon-
strators with some previous engagement in social movement activities might
be expected to be more easily persuaded to vote Green or Far Left. Perhaps
most interesting would be a look at the influence of ideology and the strength
of feeling about the justification for the war in Iraq.

I conducted a series of multinomial logistic regressions to answer these
questions, with both the overall data and individual national datasets, com-
paring voters of Green, Far Left (and, in Britain, the Liberal Democrats) with
those expressing a voting preference for the main party of the Left (Socialist/
Social Democratic/Labor parties). To reduce the complexity of the model,
I removed variables that did not make any impact in any country or in the
overall analysis of all countries. In terms of party choice, some countries did
not have well-established Green or Far Left parties, and the number of dem-
onstrators expressing a preference for these parties was too small for a reliable
analysis; in these cases, I only considered the main small party alternatives. 4
The results of a simplified model with just six independent variables are pre-
sented in Table 8.5. What did I find?

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page15Z^_

m

* q- si

I fil 111 111

I'ilblllll!

"-slip -|1 -s S | J
§ §t 2 'a £ 1|§^|

1 5 iff ii ii«j §

sill! 1 1$ I § § "i

f llfll Hi | alii

H$h-

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 1 58^

,i

5

i
e
o

j.

O

1

O)

o

c

•j=

5

- %

3

\o

_8
.3

^ I

1/5 1(5

o

o

E

£

z

"■

g

|

1

o

i/i

■ j.

i- -^

O)

,c

o

>

.3

J

£ §

J.

-o "a

C 0)

° 0>

3

£

S 22

CD u

o 1

I

•s ?

o

J2 _

c a

= 1

E.i

4-f

<SI

iq

od

a>

_Q

|H

I

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page152^_

BOON OR BURDEN? I59

Inglehart's notion of the cognitively mobilized voter could not substan-
tially be confirmed, as education and interest in politics failed to account for
party choice. Of course, demonstrators constitute a subgroup of the popu-
lation distinguished by very high levels of education and political interest,
and thus, within a group so homogeneous in these terms, these variables may
not have the expected effects. Age, however, was quite an important predic-
tor in most countries. A preference for small antiwar parties was consistently
associated with younger voters, with the single exception of the Party of
Democratic Socialism in Germany. Relative youth was a particularly pro-
nounced factor in Italy and Great Britain, with demonstrators aged eighteen
to thirty-four particularly ready to vote for small parties. Overall, this is con-
sistent with the long-established idea that party identification is particularly
weak among younger voters, and thus such voters are more likely to be at-
tracted by protest and challenger parties.

Previous protest frequency and other indicators of political activity failed
to make any difference regarding variables associated with previous political
activity. The only variable that appeared fairly consistently associated with
party choice was having participated in specific demonstrations that had left
or green socialization: in all countries but Belgium, having taken part in a
demonstration on issues of social justice, such as a trade union march, made
it less likely for a demonstrator to express a preference vote for a Green party.
By contrast, having participated in such an activity did not generally provide
a great boost for voting for a Far Left party: it made it slightly more likely
in most cases, but nowhere is the effect statistically significant. By contrast,
having participated in an environmental demonstration provided a good pre-
dictor of Green party choice in all countries.

Turning to ideology, sympathy with the global justice movement fairly
consistently was a factor for both Green and Far Left voting; positioning on
the left-right scale was particularly important for Far Left parties. Opposi-
tion to the idea that the war might be justified if sanctioned by the UN Secu-
rity Council was more important for Far Left than Green parties.

In the overall analysis — using weighted data, giving each country the
same number of cases, and using country dummies to ensure that country-
specific relationships do not skew the overall model — the picture emerges
that Green voters tend to be younger than supporters of the main party of
the Left, less likely to have attended demonstrations on social issues but far
more likely to have attended environmental ones, and have more sympathy
with the global justice movement. For Far Left party voters, youth is less sig-
nificant than for Greens, as is previous involvement in protest, while sym-
pathy for the global justice movement has a similar effect. What sets Far Left

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 160_^)_

160 rudig

voters apart in particular is a stronger ideological left c
stronger opposition to UN-approved military action in Iraq.

But what about the country-by-country picture? Are there differences
in terms of whether the main party of the Left is in government or in oppo-
sition? When one compares the models on a country-by-country basis, the
U.K. case stands out as fundamentally different from the rest. Here, two vari-
ables clearly dominate the model, namely age and opposition to a war justified
by the United Nations. Strangely, perhaps, the model is more or less the same
for all three antiwar parties. There are differences in terms of previous pro-
test behavior and left-right positioning, which go broadly into the expected
directions.

If I compare the countries with governments supporting the war (Nether-
lands, Spain and Italy) with those who do not (Belgium, Switzerland, Ger-
many), I do not find any major differences. Certainly ideology as measured
by left-right position and sympathy with the global justice movement play
far more important roles than they do in the United Kingdom, while the
issue of UN sanctioning of the war is fairly marginal. With the exception of
Italy, age also plays a rather minor role in the other countries.

One possible interpretation of this finding is simply that, except in the
case of Britain, the results are a manifestation of how voters of Green and
far-left parties are fairly settled and stable in their orientation. Demonstra-
tors who express a future voting preference for these parties are similar to the
voters who would generally be expected to vote for these parties in any case,
and, thus, the Iraq issue makes relatively little difference in this context. In
Britain, by contrast, the justification of the Iraq War has a dominance that
in particular affects younger voters who have no settled party preference. In
that sense, the British picture is one of greater change and instability.

One problem with the analysis presented so far is that it does not try to
measure changes in voter preferences. We also conducted a detailed analysis
of what determined whether previous voters of the main party of the Left
either stayed loyal to their party or changed to one of the smaller parties. A
series of multinomial logistic regressions were conducted (coefficients not
shown) with the variables used in Table 8.5. The main problem encountered
with this analysis is the small number of cases; very few coefficients reach the
basic level of statistical significance (p<.05). Thus, the results could at best
be seen as indicators of trends. Essentially, the analyses of changes in voter
preferences do not lead to any different results. In terms of the direction and
size of coefficients, all of the conclusions reached on basis of the previous
analysis are confirmed. In other words, new voters of Green and Far Left
parties display essentially the same attributes as all voters of these parties.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 161^_

"4

BOON OR BURDEN?

This result thus confirms a general trend in our analysis that deempha-
sizes the potential change associated with the events of February 15: it is the
"usual suspects" who attend these demonstrations and make the expected
voting choices. Looking beyond the data gathered on demonstrators, how-
ever, is there any evidence of changes in the electoral fortunes of ai
ties as a result of the protest events?

The answer, in short, is "no": an analysis of the poll rating of a
parties (detailed results not shown) demonstrates that while some parties
experienced brief increases in their ratings after the February 1 5 demonstra-
tions, this was only a blip — in case was the political standing of any of the
antiwar parties transformed by the Iraq issue.

The actual elections results following the Iraq War generally confirm
this trend. In some countries, the Iraq issue was completely marginal to elec-
tions; in others, elections were heavily influenced by the debate on Iraq. The
only national elections in which the Iraq conflict played a major role were
those in Spain and the United Kingdom. The Spanish parliamentary elec-
tions of 2004 were strongly influenced by the terrorist attack immediately
preceding election day, which brought the Spanish Socialist policy on Iraq
back to the fore (Ordeix i Rigo 2005); in the British general election of 2005,
the Iraq War also was a major issue. In Britain, there was a realignment of
the Far Left with the Socialist Alliance dissolved to be replaced by a new
Respect Party mainly associated with the former Labour member of Parlia-
ment and prominent antiwar campaigner George Galloway. While Galloway
won a seat in the House of Commons, Labour's losses to antiwar parties and
candidates were not major enough to deny the party a renewal of its gov-
ernmental mandate (Kavanagh and Butler 2005). Small antiwar parties might
have expected to do particularly well in the European elections of 2004
(Lodge 2005), as second-order elections normally provide a good basis for
protest voting, but even on this occasion, the results were not entirely posi-
tive for the antiwar camp. Some antiwar parties did quite well, but others
had disappointing, and in some cases, catastrophic results (i.e., the Greens
in Belgium). Overall, it is difficult to escape the conclusion that the antiwar
issue was, at best, of marginal importance in terms of influencing the elec-
toral fortunes of antiwar parties.

Finally, did the demonstration lead to an increase in party membership
of antiwar parties? To assess the possible impact of the antiwar campaign on
the recruitment of party members by antiwar parties, I compiled the national
party membership figures of all major antiwar parties at the end of 2002 and
2003, that is, the years before during which the main antiwar mobilization
took place. While this is admittedly a rather crude measure, it allows a first look

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 162^_

at what could be expected to be a major membership increase of these parties
following the 2003 demonstrations. The results are compiled in Table 8.6.

Yes, there is a major change in membership for quite a few parties. In
Scotland, parties of the Far Left (SSP and Socialist Alliance) and the Greens
appear to have benefited strongly; the Liberal Democrats and the Scottish
National Party, in contrast, were not among the winners of 2003. Other coun-
tries in which antiwar parties appear to have had increased memberships are
England, Spain, Italy, and the Netherlands. In Germany, Belgium, and Switz-
erland, however, membership developments were mostly negative.

Overall, the figures support the interpretation that in countries where
the antiwar issue attained a high salience, namely those where the government
supported the Iraq War, antiwar parties benefited from a significant mobiliza-
tion process. Virtually all Old and New Left antiwar parties in these countries
managed to increase their membership, in some cases substantially, between
the end of 2002 and of 2003. In contrast, countries with a low salience of
the issue, the mobilization potential in terms of attracting new members was
obviously low. Here, membership levels seem virtually unaffected by the anti-
war issue.

While there is thus some evidence suggesting that the antiwar demon-
| strations of 2003 have not been completely without impact on party poli-

~W tics, further research is needed to assess how important the Iraq issue really ~W

was for the new members who joined during 2003. Another open question
is how stable an addition to party membership the antiwar mobilization
cycle produced has been. Even if the impression of a mobilization effect in
countries whose government supported the Iraq War can be confirmed, it is
important to note that while antiwar parties benefited by quite substantial
margins in relative terms, the actual numbers involved are quite small. In
Britain, the high percentage changes involve just a few hundred additional
members, as opposed to the hundreds of thousands who demonstrated. In
Italy and the Netherlands, some antiwar parties had net increases of a few
thousand, and in Spain, the United Left could add at best a few thousand
members over a longer time period. Only the Spanish Socialists could mobi-
lize new members in the tens of thousands. However, most of these appear
to have joined after their election victory in 2004 and thus could hardly be
claimed as antiwar movement related.

If one compares the net membership increases with the number of dem-
onstrators, then the mismatch is particular strong in Britain, Italy, and Spain.
In the Netherlands, however, a comparatively tiny mobilization is associated
with a quite sizeable net increase in antiwar party membership of several thou-
sand. The Netherlands also stands out here, thus reconfirming the status of

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 163^_

Table 8.6. Parly Membership: December 2002 to December 2003

Membership — Membership — Difference
Country Party end of 2002 end of 2003 (percent)

Scottish Labour Party
Scottish Natic
Scottish Liberal Democrats
Scottish Socialist Party
Scottish Green Party

22,153

13,382

4,171

"4

England

Labour Party

227,248

192,789

-15.2

and Wales

Liberal Democrat

67,284

69,134

+2.7

Green Party

5,268

5,858

+ 11.2

Socialist Alliance

2,000

2,400

+20.0

Germany

Social Democrats (SPD)
Party of Democratic

693,894

650,798

-6.2

Socialism (PDS)

70,805

65,753

-7.1

Alliance '90/Greens

43,881

44,052

+0.5

Spain

Socialists(PSOE)

484,321

546,746 (2004)

(+12.9)

United Left (IU)

65,000 (2000)

69,000 (2004)

(+6.2)

Italy

Democrats of the Left (DS) 534,358

549,372

+2.8

Federation of Greens

16,000

20,000

+25.0

Party of Italian

Communists (PdCI)

23,747

30,932

+30.3

Communist Refoundation

(PRC)

89,124

85,770

-3.8

The

Labour Party (PvdA)

57,374 (2001)

60,062

(+4.7)

Netherlands

Green Left

18,469

20,503

+ 11.0

Socialist Party (SP)

36,406

43,389

+ 19.2

Belgium

Flemish Socialists (SP.a)

63,898 (2001)

61,637

(-3.5)

Walloon Socialists (PS)

82,470

83,105 (2004)

(+0.8)

AGALEV

5,348

5,955

+ 11.4

ECOLO

4,463

3,751

-16.0

Switzerland

Social Democrats (SPS)

35,150

34,509

-1.8

Greens

5,000 (1998)

4,757 (2005)

(-4.9)

Note: a The numbers are in parentheses if t: o not match exactly the annual

figures immediately pre- and postdating the demonstration.

Sources: Great Britain: International Peace Protest Survey (IPPS) 2003; Annual Accounts submitted
to Electoral Commission, http://www.electoralcommission.org.uk; except for Scottish Labour
Party in 2002: Annual Report 2003 figures supplied by Scottish Labour Party, September 5,
2005; 2003: Scotsman, February 24, 2005 (Figures for England and Wales for Labour and Liberal
Democrats were calculated by subtractiir il membership.);

Germany: Niedermayer 2005; Spain: Mendez, Morales, and Ramiro 2004; Italy: Archivio dati
to Carlo Cattaneo, http://www.istcatta-
. n. http://www.rug/
nl/dnpp; Belgium: Agalev/Groen: Personal communication, Jo Buelens, February 2006; ECOLO:
Van Haute 2005; Socialist Parties: Personal communication, Stefaan Walgrave, December 2006;
Switzerland: SPS, personal communication Rosmarie Burki, SPS, 2 October 2006;Greens 1998:
Ladner and Brandle 2001; 2005: Swiss ( irecn Part) : only figures available for 2005 and 2006;
personal communication, Simon Grossenbacher, Sekretariat Les Verts-Griine, Bern; 5 September

I

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 164_^_

164 RUDIG

a country that was already marked by the importance parties played in the
organization of the demonstration (see chapter 10). Exploring the reasons
for this may raise some interesting research questions. The country has not
stood out so far in terms of the size of party membership: comparisons here
place it just slightly above average (see Table 8. 1). In one other area, namely
environmental NGO membership, the Dutch are a clear outlier and display
levels of membership that far outstrip any rival (Dalton 2005). If Dutch polit-
ical activists are thus generally "joiners," then this might provide a nation-
specific cultural basis where small political parties are better placed in using
new issues to mobilize new members.

Conclusion

The antiwar demonstrators of February 2003 displayed political allegiances
and preferences that indicated increased support for Green and Far Left
parties. Demonstrators were not turned away completely from the electoral
process and party politics; rather, they were prepared to vote and express pref-
erences. Beyond this, however, there was very little evidence of a major increase
in party political activity generated by the protest events.

Despite the extraordinary size of the February 1 5 demonstrations, it is
quite obvious that they were not pivotal in terms of changing the political
fortunes of many political parties. In some countries, the issue was simply
too marginal to affect party development in a more sustained way; in oth-
ers, there was an "Iraq effect," which had taken hold in late summer or autumn
of 2002; some cases can be identified in which antiwar parties received a
boost in February and March, but in most instances, this was just a tempo-
rary result. No major political earthquake effect on parties can be detected.

Also in terms of general political behavior patterns, such as participa-
tion in elections and joining political parties, the evidence suggests that the
antiwar movement was not an earth-shattering event that changed the face
of politics. Some parties did experience a new influx of members that may,
at least in part, have been associated with the events of February 1 5, 2003,
but the evidence available suggests only fairly minor changes. The antiwar
movement apparently did not constitute an experience that politicized sub-
stantial percentages of the population not previous involved in politics.

Were the February 15 events of thus a boon or a burden for the parties?
There is no evidence to suggest that alienation from electoral politics played
a major role in the protests; demonstrators in all countries had been active
voters and were also eager to take part in future elections. There is more evi-
dence that the demonstration was associated with a move away from the major
parties to smaller Green and Far Left parties. But even here, the benefits

H$H

-4

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 1 65^

BOON OR BURDEN? 165

experienced by these parties in terms of additional voters and members were on
quite a small scale. They were a boon, but a whimper rather than a big bang.

What I thus find is a discrepancy between the unprecedented degree of
protest mobilization, on the one hand, and the feebleness of the impact of
such mobilization on elections and party politics, on the other. Why did the
events of 2003 not have a more profound effect? One interpretation is that
demonstrators predominantly were engaged in some form of identity or ges-
ture politics and did not regard electoral or party political activity to be par-
ticularly relevant in this context. If, indeed, even such a major issue cannot
revive interest in political parties among at most a small minority of those
mobilized to demonstrate, then this episode might provide further evidence
for theories of a post-party politics.; however, there are some indications that
an instrumental element was present in some cases, namely in these coun-
tries where there was realistic prospect of a change of government leading to
a change of policy on Iraq. In Spain, the Socialists were the main vehicle of
change and they received overwhelming support from demonstrators, while
the United Left, an outsider party that was not part of the governmental proj-
ect, benefited only weakly. In Italy, though, the Far Left was part of the broad
left-wing alliance, and a vote for parties belonging to the alliance indicated
support for an alternative government, not just a protest vote for an outsider
party. The membership figures, significantly, suggest that the Party of Italian
Communists (PdCI) rather than the Communist Refoundation (RC) bene-
fited most. This would be compatible with an instrumentalist explanation
of party choice, as the PdCI had remained faithful members of the left-wing
bloc while the RC had a more ambivalent attitude to government. 5 The ges-
ture politics interpretation thus predominantly applies to countries where
either the government already was opposed to the Iraq War or, as in Britain,
both major parties supported the war effort with no practical prospect of a
change of government that would be linked with a change of policy.

A note of caution might be appropriate, however: I have analyzed data
gathered on one specific political event that, to those who witnessed the un-
believable number of people who moved through the streets of cities around
the world that day, seemed to be special. Being so close to the event might
raise expectations in one's mind of an appropriate impact on party politics
that might exceed what could reasonably be expected. Some past protest
events that were seen as significant, such as the May 1968 protests, did not
immediately produce major political realignments, new parties, or the like
but nevertheless had a profound political effect. The current assessment of
the February 15 protests not having a major impact on party politics may
thus be premature.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 16fi^_

Notes

1 . The very high figure of 55 percent for the United States most likely indicates
that "party membership" is understood in very different terms in the U.S. context.
Political parties in the United States are structured very differently than those in
Europe, and the concept of party "membership" in a European sense does not exist
in the United States, at least as far as the main parties are concerned. Some state
green parties have established "membership that is comparable to that in European
green parties, but practices vary widely among states. The respondents most likely
understood the question to relate to being a registered supporter of a particular party.

2. We obviously have to consider that voting recall is likely to be affected by
rationalizations of previous behavior as well as social desirability factors: it thus
could be expected that these figures not only overestimate the number of people
who voted but also the number of demonstrators who voted for antiwar parties.

3. The U.S. surveys did not include a question on intended future voting choice,
and as no data are thus available, the United States is not included in the subsequent

4. The dependent variable used for this analysis was future voting preference,
with the categories "Green,""Far Left, "other," and "Socialist/Social Democrat/Labor

I Parties," with the last as reference category. For the United Kingdom, the data of

-rtft- England and Scotland were combined, and a separate category for the Liberal Demo- .0£\-

I crats was used. The coefficients for the "other" parties are not displayed because of

the small number of cases renders them largely meaningless. The small number of

ables. The independent variables used were (a) Age, with three categories: 18-34,
35-54, and 55 or older, with the last used as reference category; (b) Previous atten-
dance of demonstration: Social/Labour, recoded with the nonparticipants as refer-
ence category; (c) Previous attendance of demonstration: Environmental, coded as
b); (d) Sympathy for antiglobalization movement, recoded to make those without
sympathy the reference category; e) Far Left: those who scored 0, 1, or 2 on the 0-10
left-right scale are categorized as Far Left, with everybody else as reference category;
d) Justification of Iraq War: those who "completely disagree" that war against Iraq
would be justified if sanctioned by the UN Security Council are pitted against every-
body else (i.e., those who only moderately disagree with this statement, are unde-
cided, or agree with this position).

5 . I am grateful to Daniele Albertazzi (University of Birmingham) for this point.

Works Cited

Benoit, Kenneth, and Michael Laver. 2006. Party Policy in Modern Democracies. Lon-
don: Routledge.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 16Z^_

BOON OR BURDEN? l6j

Dalton, Russell J. 1984. "Cognitive Mobilisation and Partisan Dealignment in

Advanced Industrial Democracy." Journal of 'Politics 46: 264-84.
. 2004. Democratic Challenges, Democratic Choices: The Erosion of Pol,

port in Industrial Democracies. Oxford: Oxford University Press.
. 2005. "The Greening of the Globe? Cross-National Levels of Environmental

Group Membership." Em hies 14: 441-59.

Dalton, Russell J., and Manfred Kuechler, eds. 1990. Challenging the Political Order:

New Social and , stern Democracies. Cambridge: Polity.

Di Virgilio, Aldo. 2002. "Le elezioni politiche del 13 maggio 2001." Quaderni del-

I'Osservatario Elettorale 46: 1 55-82.
Flynn, Gregory, and Hans Rattinger, eds. 1985. The Public and Atlantic Defense.

Totowa, N.J.: Rowman & Allanheld.
Franklin, Mark N. 2004. Voter Tu Electoral Competition in

Established Democracies since 1945- Cambridge: Cambridge University Press.
Inglehart, Ronald. 1977. The Silent Revolution: Changing Values and Political Styles

among Western Publics. Princeton, N.J.: Princeton University Press.
. 1990. Culture Shift in Advanced Industrial Society. Princeton, N.J.: Princeton

University Press.

. 1997. Modernization and Postmodernization: Cultural, Economic, am

| Change in Forty-three Societies. Princeton, N.J.: Princeton University Press.

W Inglehart, Ronald, Miguel Basanez, Jaime Diez-Medrano, Loek Halman, and Ruud ~W~

' Luijkx, eds. 2004. Human Belief and Values: A Cross-Cultural Sourcebook Based '

on the 1999-2002 Values Surveys. Mexico City: Siglo XXI Editores.
Jun, Uwe, Henry Kreikenbom, and Viola Neu, eds. 2006. Kleine Parteien im Aufwind:

A.M.: Campus.
Kavanagh, Dennis, and David Butler. 2005. The British General Election of 2005-

London: Palgrave Macmillan.
Katz, Richard S., and Peter Mair. 1995. "Changing Models of Party Organization and

Part Democracy: The emergence of the Cartel Party." Party Politics 1: 5-28.
Ladner, Andreas, and Michael Brandle. 200 1 . Die Schweizer Parteien im Wandel: Von

Mitgliederparteien zu professionalisierten Wahlerparteien? Zurich: Seismo.
Lawson, Kay, and Peter H. Merkl, eds. 1988. When Parties Fail: Emerging Alterna-

Princeton, N.J.: Princeton University Press.
Lodge, Juliet, ed. 2005. The 2004 Elections to the European Parliament. London: Pal-
grave Macmillan.
Longchamp, Claude, and Lukas Golder. 2003. SVP-Aufstieg erstmals gebremsf. Medi-

enbericht zum 3. SRG SSR Wahlbarometer '03. Bern: GfS-Forschungsinstitut,

Politik, und Staat. Online at: http://www.polittrends.ch/wahlen/wahlbarometer
Mair, Peter, and Ingrid van Biezen. 2001. "Party Membership in Twenty European

Democracies, 1980-2000." Party Politics 7: 5-21.

H$H

8 Chapter 8_Walgrave 22/01/2010 3:42 PM Page 16S^_

168 rudig

Mendez, Monica, Laura Morales, and Luis Ramiro. 2004. "Los afiliados y su papel
en los partidos politicos espanoles." Zona Abierta 108/09: 153-207

Morgan, Bryn. 2001. General Election Results, 7 June 2001. (Research Paper 01/54).
London: House of Commons.

Miiller- Rommel, Ferdinand, and Geoffrey Pridham, eds. 1991. Small Parties in West-
ern Europe: Comparative and National Perspectives. London: Sage.

Niedermayer, Oskar. 2005. "Parteimitgliedschaften im Jahre 2004." Zeitschrift fur
Parlamentsfragen 36: 382-89

Norris, Pippa. 2002. Democratic Phoenix: Re, :il Activism. Cambridge:

Cambridge University Press.

Ordeix i Rigo, Enric. 2005. "Azn.i ire or Punishment for Supporting

the Iraq War? Hypotheses about the Causes of the 2004 Spanish Election Results.
American Behavioral Scientist A3: 610-15.

Parker, Simon, and Paolo Natale. 2002. "Parliamentary Elections in Italy, May 2001,"
Electoral Studies 21: 649-80.

Poguntke, Thomas. 1996. "Anti-Party Sentiment — Conceptual Thoughts and Empir-
ical Evidence: Explorations into a Minefield." European Journal of Political
Research 29: 319-44.

. 2002. "Party Organisational Linkage: Parties without Firm Social Roots?" In

| Political Parties in the New Europe: Political and An, res, ed. Kurt

■W- Richard Luther and Ferdinand Muller-Rommel, 43-62. Oxford: Oxford Uni- W

' versity Press. '

Richardson, Jeremy. 1995. "The Market for Political Activism: Interest Groups as
Challenge to Political Parties." West European Politics 18: 116-39.

Rihoux, Benoit, and Wolfgang Riidig. 2006. "Analysing Greens in Power: Setting
the Agenda." European Journal of Political Research 45: S1-S33.

Rochon, Thomas R. 1988. Mobilizing for Peace: The Antinuckar Movements in Western
Europe. Princeton, N.J.: Princeton University Press.

Riidig, Wolfgang. 2002. "Germany." In Gn\,

dinand Muller-Rommel and Thomas Poguntke, 78-1 1 1 . London: Frank Cass.

Scarrow, Susan. 2000. "Parties without Members? Party Organization in a Chang-
ing Electoral Environment." In Par, litical Change in
Advanced Indus,: r, ed. Russell J. Dalton, and Martin P. Watten-
berg, 79-101. Oxford: Oxford University Press.

Seyd, Patrick. 1995. "Anti-Party Politics in Britain: An Assessment." Jahrbuch fur
PolitiklYearbook of Politics 5:21-41.

Van Haute, Emilie. 2005. "After Governmental Participation: Still the Age of Cohesion
between Green Members and Activists? The Case of Ecolo, Belgian French-
speaking Green Party." Paper presented at 3rd ECPR General Conference, Buda-
pest, September.

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 162^_

Open and Closed Mobilization Patterns:
The Role of Channels and Ties

Stefaan Walgrave and Bert Klandermans

m

Mobilization can usefully be discussed in terms of the demand and supply
metaphor. "Demand" refers to the will of (a segment of) the population to
protest and show its discontent, while "supply" refers to the offer of a cer-
tain collective action event staged by organizations and social movements.
Mobilization brings demand and supply together. To be sure, this economic
metaphor does not apply entirely to collective action events — both "parties"
are not exchanging different types of goods, they essentially want the same
(collective action), and there can be demand without supply, in the case of
spontaneous collective actions. Yet, it is a very useful way of speaking about
collective action in general and mobilization in particular. Mobilization, then,
is one side of the demand, supply, and mobilization triangle that accounts
for the ebb and flow of social movements (Klandermans 2004). Mobiliza-
tion is the process that gets the movement going and links a certain demand
for protest among the population in a country to a supply or offer from the
protest market in that country. Demand and supply would remain potentials,
if processes of mobilization were not to bring the two together. This makes
it understandable why so much of the literature on social movements is de-
voted to mobilization processes. Yet, it would be a mistake to neglect demand
or supply factors. Mobilization is only possible on the basis of some demand
and supply being developed over the course of time; if neither of them ex-
isted in a society, mobilization would be inconceivable and in vain. In this
chapter, after a theoretical introduction developing two mobilization typolo-
gies, we will focus on the assessment of mobilization patterns in the eight
nations studied. Next, we will assess the extent to which mobilization patterns

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Z0_^_

170 WALGRAVE AND KLANDERMANS

make a difference and affect the features of the demonstrators. Finally, we
will attempt to account for these differences in mobilization patterns by asso-
ciating them to diverging patterns of supply and demand in the eight coun-
tries under study.

Open and Closed Channels, Strong and Weak Ties

According to Klandermans, social movements face two separate mobiliza-
tion challenges: consensus mobilization, that is, persuading people of the
good cause, and action mobilization, that is, actually bringing people to the
streets (Klandermans 1984; Klandermans 1997). Our focus here will be on
action mobilization and what can be learned about from the demonstrations
of February 15.

Action mobilization is a funneling process with different stages in which
generalized preparedness to act in support of a cause must be transformed
into specific preparedness to participate in a particular protest event, and
finally into actual participation (Oegema and Klandermans 1994). This
activation process has three aspects: getting in touch with people to get the
message about the event across (informing them about what, where, and
when), motivating people to participate (making sure that the pros of par -
| ticipation outweigh the cons), and helping motivated participants overcome

~W barriers to participation (Klandermans and Oegema 1987; Oegema and Klan- ~W

dermans 1994). A whole range of social movement organizations, political
parties, associations, and NGOs staged the protests of February 15. In chap-
ter 1, we described the organizers in each country, and chapter 10 shows
how the actual protesters were organizationally embedded. Obviously, the
organizational makeup of the antiwar coalition affected the process of mobi-
lization, if only because organizations differ in how they mobilize. Our aim
here is not to compare mobilization strategies employed by the various orga-
nizations in the countries of our study; rather, in this chapter, we are rather
interested in the opposite end of the mobilization chain — the people targeted.
As we only interviewed people who actually showed up for the demonstra-
tion, we cannot contrast successful and unsuccessful mobilization attempts
but can only sketch action mobilization attempts that succeeded, as it was
participants who reported back. In this context, our focus is on information
dissemination. Certainly, being targeted with information is not a sufficient
precondition for participation, but for obvious reasons it is a necessary con-
dition. For example, research by Klandermans and Oegema (1987) showed
that in the specific case they investigated, simply none of those who a move-
ment organization failed to target participated in the event. Our respondents
were necessarily reached by information about the demonstration. Our core

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Zl^_

OPEN AND CLOSED MOBILIZATION PATTERNS I7I

question in this chapter is of how they were reached: which people were suc-
cessfully targeted for mobilization through what kind of information channels?

Since the first formulation of resource mobilization theory (McCarthy
and Zald 1976), social movement literature has acknowledged the organized
character of mobilization. Numerous studies have since documented the cru-
cial significance of social networks for understanding mobilization and differ-
ential recruitment (see Diani 2002; Diani and McAdam 2003; Gould 1993;
Kitts 2000; Tindall 2004). However, most of the available research focuses
s not focus on the specific information channels but rather on the general
sociostructural embeddedness of mobilization. Resource mobilization theory,
for example, emphasizes the necessity of formal organizations for mobiliza-
tion (McCarthy and Zald 1976). People have to be more or less integrated
in social movement organizations to enable large and successful mobilization
processes. Via learning processes that develop participatory skills, via the gen-
eration of social networks reducing information costs and expanding weak
ties, and via intentional mobilization does membership in formal associa-
tions foster protest participation (Morales 2003; Tindall 2004). Movement
organizations are mobilization machines specialized in helping people elim-
inate the barriers that lie between willingness to participate and actual par-
ticipation. Some scholars, in contrast, claimed that informal networks could
do the mobilization job as well and that, rather, informal preexisting networks
are the sociostructural requirement for collective action. McAdams's concept
of a micromobilization context proved especially influential in this respect
(McAdam 1988; McAdam, McCarthy, and Zald, 1996).

Although the role of formal and informal networks is widely documented,
it is perfectly conceivable that the actual targeting and information dissem-
ination are not embedded in a social network. Research in Belgium, for
example, showed that the media, not formal organizations or informal net-
works, brought the demonstration message across and thus played a central
role making the White March, the biggest demonstration ever held in Bel-
gium. At least in targeting the population, but probably even in motivating
potential participants and lowering the barriers for participation, the mass
media played a major role (Walgrave and Manssens 2000). Thus, mobiliza-
tion, certainly the information dissemination aspect of it, can happen out-
side of organizational networks or, better, reach beyond them. Based on this
general idea, we propose two different dimensions of mobilization: strong
versus weak ties and open versus closed mobilization channels.

Granovetter first launched the idea of the strength of ties in a seminal
article (1973) elaborating the importance of strong and weak ties for social
movement mobilization (see also Putnam [2000] who speaks about "bridging"

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Z2^_

172 WALGRAVE AND KLANDERMANS

and "bonding" social capital). The strength of weak ties, so writes Granovet-
ter, is that they connect cliques (that is, groups with strong ties). Therefore,
it is the weak ties that are crucial in spreading information among cliques.
Without weak ties, information would not travel beyond group boundaries.
Strong ties, though, while not very effective in disseminating information,
are more effective than weak ties when it comes to activation (McAdam 1986;
see also Marwell and Oliver 1993).

The open-closed distinction is based on various typologies we encoun-
tered in the social movement literature. Klandermans and Oegema (1987),
for instance, discern four routes to target individuals for mobilization: mass
media, direct mail, and organization and friendship ties. They distinguish
between formal and informal mobilization attempts, the first being deliber-
ate mobilization efforts of the movement (via flyers, posters, stands, adver-
tisements, and so on), and the second being personal links with someone who
planned to participate in the upcoming demonstration. They conclude that
although more people are reached by formal channels, the informal links are
especially crucial for stimulating actual participation, since they yield much
more participation motivation than formal information channels (see also
Gould 1991, 1993). Snow and colleagues distinguish face-to-face dissemi-
nation from mediated dissemination (Snow, Zurcher, and Ekland-Olson,
1980). Face-to-face dissemination requires the physical presence of a source
of information, and mediated dissemination implies the use of media — like
newspapers, television, mail, the Internet, and telephone. Furthermore, Snow
and colleagues distinguish private channels for mobilization and recruit-
ment from public channels. Snow, Zurcher, and Ekland-Olsohs taxonomy
of recruitment channels suggests that some channels can reach the public at
large, while others only reach certain segments of the population.

Elaborating on these ideas, we distinguish also their openness versus
closedness. The characteristic that defines the openness or closedness of a
communication channel is the target: open mobilization channels have no
restriction regarding whom they target, while closed mobilization channels
only target people with certain characteristics, for instance members of an
organization. The broader the target groups, the less specific personal char-
acteristics, the more open the mobilization channel. An organization that
primarily targets its own members is an example of closed mobilization. Labor
union demonstrations are typical examples of recruitment via closed chan-
nels. As a rule, such demonstrations are crowded with union members, and
hardly any nonmembers show up at these protests. Closed channels are typ-
ically employed in so-called en-bloc recruitment (McAdam 1986), where
members of a group or organization are recruited as a whole. The mass media,

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Z3^_

OPEN AND CLOSED MOBILIZATION PATTERNS I73

on the other hand, are probably the best example of an open mobilization
channel. Although there are some notable biases in media use — coincid-
ing largely with organization membership — mass media can be considered a
ubiquitous mobilizer because a vast majority of the population is confronted
with mass media outlets.

Theoretically the two dimensions — kinds of ties and openness or closed-
ness of channels — are independent from one another, as they refer to differ-
ent facets of the mobilization process. This is not to say that they are not
correlated in practice. Obviously, membership in an organization (strong tie)
makes it more likely for one to be targeted by mobilization attempts from
that organization (closed channel). Thus, we expect both dimensions to be
associated to some extent. To complicate matters further, the same channel
might function differently in different countries and under different circum-
stances. For example, we considered newspapers as open mobilization chan-
nels, but they can be explicitly partisan, owned and controlled by a social
organization or political party. Hence, the readership of such newspaper prob-
ably tends to be selective, offering certain social and political features. The
decision to read this specific newspaper most likely is based on ideological
beliefs. As a consequence, mobilization via this newspaper is less open than
would be the case via more neutral ones. In a similar vein, friendship net-
works or links with colleagues can be more open depending on the issue. For
work-related and bread-and-butter issues, colleagues act as closed channels
referring to specific interests. Yet for universal issues, such as the protest against
the war in Iraq, colleagues are more likely to act as an open mobilization
channel. Here is another example of the contingent nature of mobilization
channels: some people are informed about a demonstration by flyers, ads, or
posters. Are these open or closed mobilization channels? The answer depends
on the strategy of the organizers. If they decide to distribute flyers in train
stations or on the streets, this technique is fairly open. If, in contrast, they
focus their distribution on political meetings, obviously only some specific
groups are targeted. The same applies to posters: are they suspended on the
streets or, rather, in specific places like universities, schools, and clubs?

During a mobilization campaign, a movement organization minimally
tends to target its own members. This is the pattern of mobilization via closed
channels and through strong ties Marwell and Oliver (1993) identify as the
pattern most likely to be adopted by organizers. Yet, unless an organizer is
satisfied to stay within the boundaries of the organization's constituency,
weak ties and open channels must be employed. After all, not every citizen has
strong ties, or any ties at all, to the organizations staging the events. If the
movement is engaged in a battle for universal values — such as the antiwar

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 1 74(g)

174 WALGRAVE AND KLANDERMANS

demonstrations of February 15 — chances are that the message gets across to
a broader segment of the population. In this process, the mass media play a
crucial part. Although their normal role is not a mobilizing one and the media
seldom take sides in a controversy or urge people to participate in any pro-
test, they do convey the message that there will be a protest event, certainly
when it is anticipated to be a large-scale one, such as a mass demonstration
in the country's capital. Therefore, we expect open channels to have played
a substantial role in the making of February 15. In chapter 3, indeed, it was
documented that there was massive attention for the war in all covered coun-
tries, and chapter 12 will show that in many countries the protests against
war also got a lot of attention in the media.

Patterns of Mobilization

What kind of links did people have to the organizations staging the demon-
stration? We asked our respondents whether they were members of one of
those organizations or whether they knew people who were members. If the
answer to either question was affirmative, we asked them how they maintained
contact with these organizations — indirectly, through such media as the Inter-
net or newsletters, or directly, by attending meetings. We used the answers
to these three questions to construct an indicator of the strength of the ties
that an individual has with relevant movement organizations. We distin-
guished five levels of strength: no ties and weak, moderate, strong, and very
strong ones. 1 Table 9.1 shows the findings with regard to the five different
levels of connectedness for the eight countries.

Half of the participants had no ties with organizations that staged the
demonstrations in their country. The remaining half had ties ranging from
weak to very strong. Overall, we observed considerable variation in the
strength of ties among countries. For example, in the United States and the
United Kingdom — the two countries that went to war — close to two thirds
of the demonstrators had no ties to the protest-staging organizations. In the
Netherlands and Germany, also relatively few participants were tied to the
mobilizing organizations. In Belgium and Switzerland, the degree of embed-
dedness was almost identical and, in contrast, shifted towards stronger ties.
Spanish demonstrators predominantly had weak or moderate ties. In Italy,
no fewer than 70 percent of the participants had some ties to antiwar orga-
nizations, while more than one third even had strong or very strong ties. These
very strong ties respondents deserve a special commentary. We first sus-
pected this might be an artifact of the sampling procedure employed in Italy,
where respondents to the survey were approached during the train ride to
Rome from the north part of the country. (This is explained in full detail in

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Z5^_

"4

E

■8

nip

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Z£^_

I76 WALGRAVE AND KLANDERMANS

the methodological appendix at the end of this volume.) As a consequence,
one might expect these participants to be mobilized more in advance and to
take part in a more organized way than those from the other countries who
were surveyed on the demonstration venue. This is what our data show. 2 Yet,
tests specially aimed at discovering method biases for the Italian survey showed
no systematic differences between postal and face-to-face surveys in Italy
(see appendix A).

To compare the degree of openness of the mobilization in the various
e asked our respondents how they came to know about the dem-
l. We recoded their answers into four categories ranging from more
open to more closed: via television, radio, and newspapers; family, friends,
school, and work; Web sites, ads/flyers, mailing lists and posters; and organi-
zations. Table 9.2 presents our findings regarding these channels.

Close to one tenth of the participants learned about the demonstration
via some organization, while the remaining nine tenths were equally distrib-
uted over the three other types of channels. The utilized mix of channels var-
ied per country. In the United States, interpersonal networks and ads, flyers,
et cetera had equal importance and were more important than the other
channels. The same holds for Belgium, though less prominently. In the
United Kingdom, the vast majority of the participants were reached by ads,
et cetera, and in Spain the mass media served that purpose. In Italy and Ger-
many, either mass media or interpersonal networks informed equal propor-
tions of the participants. Italy also had highest proportion of the participants
informed via organizational networks. In the Netherlands, mass media and
ads and so forth were approximately equally prominent. Switzerland was the
only country where interpersonal networks were the most important. In terms
of openness versus closedness of mobilization channels, our data suggest that
Spain experienced the most open campaign, followed by Germany and the
Netherlands, whereas the campaigns in the United States, the United King-
dom, and Belgium seem to have been the most closed.

Having presented our focal variables grasping the two dimensions of
mobilization, we explore the relationship between the two. We argue that
both dimensions, although probably statistically associated, are theoretically
independent of one another. We expected to find every possible pattern among
the five levels of ties and the four levels of openness; this turned out to be
the case. The moderate correlation (Pearson's r = .27) suggests that, although
correlated, the two factors are not identical.

However, the dominant pattern is of people without ties to movement
organizations who have learned about the demonstration via open communi-
cation channels (be they mass media or interpersonal networks). Substantial

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1ZZ^_

"4

|

§s^s

£

o

N

"°° 2

1

£

c

s

-o

fc

_Q-

£

_,

\qmmo

2

z

S

in

c

o

i

mva^m

<u

!s

o

E

2

"35

c

D

_c

o

"

D

E

3

rn oq q cn

z>

2

8.

II s

1

1^ e
Ji jj

(>

_«

~S 2>c^'3

_Q

13 '§ !3 fiP

•2

££36

Z

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 1Za^_

I78 WALGRAVE AND KLANDERMANS

proportions of the participants without any ties to any movement organiza-
tion learned about the demonstration through such relatively closed chan-
nels as ads, flyers, posters, and the Internet. Apparently, these are not closed
to the extent that they exclude those not tied to the organizations that are
the sources of information. This is understandable to a certain extent. Ads,
flyers, and posters may be closed, but chances are that people other than
those connected to the senders will come to see and read them. Participants
who have strong ties to movement organizations are more likely to have
learned about the demonstration through closed channels, the more so if they
have very strong ties; in fact, among those with very strong ties are partici-
pants who mostly used closed channels.

Is this general pattern present in all countries? Apparently not, as evi-
denced by the figures in Table 9.3. This table presents the country breakdowns
for participants who had no ties and were informed about the demonstration
through open channels (mass media or interpersonal channels) and for those
who had strong or very strong ties and were informed through closed channels
(ads, et cetera, and organizational channels). The breakdowns reveal interest-
ing differences in mobilization patterns. Spain, the Netherlands, and Ger-
many display a relatively open pattern of mobilization with an emphasis on
I mass media channels, while in the United States, the United Kingdom, and

W Switzerland, mobilization is relatively open as well, but the emphasis is on ~w~

interpersonal communication channels. And Italy and Belgium have the high-
est proportions of participants mobilized via closed patterns of mobilization.

The previous discussion confirms that mobilization patterns are context-
dependent. Obviously, the mobilization campaign in one country has been
more open than that in another. It is equally interesting to know whether
the mobilization campaign for the specific February 1 5 demonstrations have
been more open or closed than those of other demonstrations. Comparative
evidence for all eight countries is lacking, but we had at our disposal com-
parable data involving six other demonstrations staged in Belgium (Brussels)
between 1999 and 2002 (Van Aelst and Walgrave 2001; Norris, Walgrave,
and Van Aelst 2004). This Belgian evidence was gathered in an identical
manner but covers strongly diverging issues and movements: from typical
new social movement events like antiracist and global justice demonstrations,
to classic bread-and-butter actions staged by trade unions (nonprofit, social
sector, and education), to a new right demonstration against tolerant Belgian
drug abuse laws. Analysis shows that the differences among the seven Belgian
demonstrations are certainly much larger than those found on February 1 5
among countries. Protests staged by new social movements in Belgium are,
on average, characterized by a substantially more open mobilization process

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page1Z9_^_

"4

E
8.

.8 .8 ^

ZZ^

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 180_^_

ISO WALGRAVE AND KLANDERMANS

than those staged by the other social movements. Apparently, issues matter
a lot, much more than country differences. Compared to the other demon-
strations in Belgium, the Belgian protest on February 1 5 against the war on
Iraq was an evident and very strong case of open mobilization. Of course,
this is no conclusive proof that the February 15 demonstrations in general,
and not only in Belgium, were characterized by open mobilization. But since
the Belgian antiwar demonstration, compared to the seven other February
1 5 demonstrations, was by no means an extremely open case — in fact, Bel-
gium ranked among the more closed mobilizations — it is at least plausible
to assume that the mobilization processes leading up to February 1 5 were in
general relatively more open than for other demonstrations in the other seven
countries under study as well.

Consequences of Mobilization Patterns

Do mobilization patterns make a difference? Do they influence the behav-
ior of the participants? For example, do they affect when participants decide
to participate, whether they go with others or alone, how far they travel, and
so on? Or do mobilization patterns influence who is going to demonstrate
in terms of age, gender, education, political attitudes, and opposition to war?
In other words, do different patterns of mobilization produce different dem-
onstrations? Oegema and Klandermans believe that mobilization processes
make a difference: "Mobilization attempts, incentives, and barriers do not
occur randomly throughout a population, but coincide with characteristics
of movement organizations, campaign characteristics, specific actions, char-
acteristics of individual communities, and social categories" (1994, 705). Most
of the factors associated with different mobilization patterns are held con-
stant in this study: it concerns the movement, campaign, and action form.
This gives us the opportunity to focus entirely on the characteristics of the
participants possibly associated with mobilization patterns as such. In this
section, we will compare the characteristics of people mobilized through the
six different patterns. We will do so by means of multivariate analyses of vari-
ance with the strength of ties, degree of openness, and country as factors.
Such analyses enable us to disentangle the main effects of ties, channels, and
ind the interaction of these factors.

Going to the Demonstration

People may go to demonstrations or with family, friends or acquaintances.
They may decide to participate long before the actual demonstration or at
the very last moment. The decision to take part is, of course, also influenced
by how far someone lives from the actual venue of the demonstration. In all

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 1814b)—

OPEN AND CLOSED MOBILIZATION PATTERNS l8l

these matters it is conceivable that the process of mobilization affects people
as it takes place. To investigate this matter, we asked our interviewees with
whom they came to the demonstration, when they decided to participate,
and how far they traveled.

In all countries the majority of the participants came with family or
friends. Colleagues and comembers of movement organizations were com-
panions in much smaller proportions. One tenth of the participants came
alone. Although the patterns in the eight countries are very similar, we observe
some interesting variation among them. For example, almost twice the num-
ber of participants came alone in the Netherlands as in other countries. In
Spain more than anywhere else, the demonstration resembles a family fair.
In Italy and Germany, though, friends were the most common companions.
Members of movement organizations relatively frequently accompanied par-
ticipants in Italy, Belgium, and the United States.

In Italy, the United States, and the United Kingdom, a higher propor-
tion of the people decided to participate in the protest at an earlier stage
than in other countries. Belgian and Dutch participants made up their minds
late, most frequently only in the last week or even the very last day before
the demonstration. The farthest distance traveled on average was in Italy. In
all other countries, we found the typical pattern of a relatively large propor-

| all other countries, we found the typical pattern of a relatively large propor-

W tion from within a few kilometers, smaller numbers from intermediary dis- v9)~

tances, and again larger numbers from longer distances.

Were such matters as participants' accompaniment, when they decided
to participate, and the distance they traveled to the demonstration related to
the way people were mobilized? To sort that out, we conducted multivariate
analyses of variance, with the strength of ties, the openness of mobilization
channels, and country as factors. 3

On the whole, one can conclude that accompaniment to the demon-
stration is influenced by mobilization patterns. However, the influence is not
the same for each type of group. The analyses reveal a main effect of ties for
colleagues and comembers only. Stronger ties more often make people come
to the demonstration with colleagues and comembers of movement orga-
nizations. The strength of ties does not have an influence on whether peo-
ple come with family or friends or on their own. The mobilization channel
used has a significant impact on all five possible social arrangements. Follow-
up analyses (not in tables) show that people who are mobilized via closed
channels are more likely to come with comembers, while people mobilized
through open channels are more likely to come alone or with family or friends.
The size of the F-values indicates that the impact of channels of mobiliza-
tion on accompaniment to the demonstration is much larger than that of

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 182^_

l82 WALGRAVE AND KLANDERMANS

the strength of ties. The significant interaction of ties and mobilization chan-
nels implies that the two factors reinforce each other. We found significant
interactions especially in the case of coming alone or with comembers: hav-
ing no ties to movement organizations and being mobilized via mass media
makes it much more likely for people to come alone, while having strong ties
and being mobilized via organizations makes it much more likely for people
to come with comembers of movement organizations. Obviously these effects
do not fully explain the country variation. First, the modest but significant
country interactions suggest that some variation exists among the eight coun-
tries in how ties and mobilization channels influence companionship at the
demonstration. However, the patterns are too unsystematic to interpret. Finally,
the size of the main effects of country implies that a substantial proportion
of the variation in accompaniment among the countries is due to factors
other than mobilization patterns.

Mobilization patterns have a substantial impact on the timing of the
decision to participate and the distance traveled (Table 9.4). We found ties
a main effect, which implies that people with strong ties decided earlier to
participate and traveled farther than people with weak ones; the same holds
for people with weak ties compared to people with none. We also found a
I main effect of degree of openness of mobilization channels: people who were

~W mobilized through closed channels decided earlier. Mobilization channels did, ~W

however, not influence the distance traveled. The effect of ties and channels
reinforced each other. Some variance in the dependent variables is due to the
differential impact of mobilization patterns in the various countries. The
significant main effects for a country again signify that factors other than
mobilization patterns are responsible for the variance among the countries. 4
In sum, diverging mobilization patterns did, indeed, generate diverging pat-
terns of participation. Open channels and weak ties or the absence of ties
bring to the demonstrations more people who come on their own or with
family and friends; these people decided more recently and traveled less far.
Those who came with comembers, who decided longer ago and traveled far,
were much more tied into an organizational field that staged the demon-
ind were more often mobilized via the organizations.

Differential Recruitment

Did such diverging patterns result in differential recruitment, that is to say,
were the people who showed up different depending on the mobilization
patterns through which they were recruited? Our answer to this question
begins with the distribution of demographics. With regard to gender, age, and
education, we found main effects of the strength of ties and the openness of

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 183^_

OPEN AND CLOSED MOBILIZATION PATTERNS 183

channels for all three variables. Male demonstrators were more often mobi-
lized through ties and open channels than females. As for age, we found a
curvilinear relationship: people without ties or with strong ties were older
than those with weak ties. Similarly, we found that people mobilized through
mass media channels and people who were mobilized through closed chan-
nels were older than those mobilized through interpersonal channels. Finally,
we found no systematic pattern with regard to education.

Not only demographics such as age or gender were influenced by pat-
terns of mobilization; the various mobilization patterns resulted in differen-
tial recruitment in terms of ideology and social and political participation as
well. We assessed whether participants differed in terms of involvement in
civil society organizations, interest in politics, protest frequency, left-right self-
placement, and voting behavior. Table 9.4 contains the results, which are
fairly straightforward, for involvement in civil society organizations, interest
in politics, and protest frequency:: main effects of ties and channels and an
interaction of ties and channels on all variables. Participants with stronger
ties are more often involved in civil society organizations, are more inter-
ested in politics, have more frequently participated in protest, and are more
leftist-oriented in terms of both self-placement and past vote. Mobilization
channels have less impact. Only mobilization via closed channels had an
impact on these social political attitudes and behavior: participants mobi-
lized this way are more actively involved in civil society organizations and
protest, more interested in politics, and more leftist-oriented. As for the inter-
action of ties and channels, we found the same pattern for all five variables:
the effect of the openness of mobilization channels was the largest among
people without ties or with weak or moderate ones. Among those with strong
and very strong ties, the channels through which they were mobilized did
not matter.

In a final attempt to allude to differences among participants contin-
gent on mobilization patterns, we discuss the attitudes about the war (Table
9.4). We found main effects of ties on top of the main effects of country.
Participants with ties to movement organizations and/or who were mobi-
lized through closed channels were more opposed to the war and to their
own government's policy regarding the war than participants who had no
such ties or were mobilized through open channels. Overall, the pattern is
clear. The organizations that were staging the demonstrations, obviously, were
opposed to the war. People who were tied (directly or indirectly) to these
organizations and/or were mobilized through closed channels by those orga-
nizations were more opposed to the war in Iraq and to their government.
This is not to say that others were not opposed to the war, but it emphasizes

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 184_^_

"4

o

£

K

•o

^

£

V

o

oo

o

"*

z

(D

8»

fc

5

t

1

c

o

•^

s

C/J

JQ

o

E

&.

3

-o

c

o

i

c

o

8.

SJ

a

J

c

o

o

15

i

•*

(>

<u

-Q

•2

I

J

2 ■*

-Ssl

h^hh QZZPh iJZZPP iZZPP

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 185^_

m

§

izzPP co zzpp OzzPP

1 § % | £ s f

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 18£^_

l86 WALGRAVE AND KLANDERMANS

that recruitment networks are reproduced in the composition of the body of
participants.

Obviously, mobilization patterns differed in the eight countries included
in our study. In some countries, the demonstrators' ties to organizations that
staged the demonstration were much stronger than in others; in some coun-
tries, people were mobilized significantly more via open channels than in
others. Such differences matter; they resulted in differential recruitment. The
various modes of mobilization brought different people into the streets. Peo-
ple, depending on how they were mobilized, not only differed in terms of
age, gender, and education but also in terms of political identity and social
and political participation. Moreover, a person's accompaniment was contin-
gent upon the mode of mobilization employed. As a consequence, the con-
figurations of participants in the various demonstrations were diverse. All
this naturally raises the question of why mobilization patterns differ among
countries. We will devote the last section of this chapter to this question.

Causes of Mobilization Patterns

Mobilization patterns are neither an a priori given nor completely exoge-
nous, but integrated parts of the societal configurations that enable protest
events. The aim of this book is precisely to map those configurations and
carefully investigate how they have influenced the events of February 1 5 in
the eight nations under study. As mobilization patterns appear to have affected
the recruitment of the demonstration, the question arises as to what are the
causes of the observed differences in mobilization patterns. In the introduc-
tion to this chapter, we stated that mobilization is the process which links
demand for protest to supply. Protest organizations stage events that offer
people who feel aggrieved the opportunity to protest. Mobilization, we argued,
is necessary to bring demand and supply together. It is, so to say, the sales
and marketing mechanism of the movement industry. Obviously, mobiliza-
tion can only succeed if demand and supply are tuned into each other. The
people we interviewed were brought together; after all they took part in the
demonstration. However, not all these people were mobilized in the same
manner. Such differences in mobilization patterns among countries, we be-
lieve, are partly determined by the configuration of demand and supply fac-
tors in a country and by the extent to which they are in sync.

In terms of demand factors, the public opinion in a country vis-a-vis
the cause is especially important. Mobilization, obviously, is a completely
different matter when the majority of the people support the cause than it
is when support is only marginal and goes against the mainstream. In the
case of a supportive public opinion, dissemination about the actual location

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 18Z^_

OPEN AND CLOSED MOBILIZATION PATTERNS 187

of a demonstration suffices to get people to take to the streets. In such a sit-
uation, demand for protest is high and people want to vent their anger some-
how. Thus, any offer on the protest market will fall on fertile soil. People do
not need to be motivated, nor do they need help to tackle any barriers. Under
such circumstances, open patterns of mobilization can be extremely success-
ful — as the White Marches in Belgium demonstrate. Indeed, strong ties to
movement organizations and en masse mobilization of members of organi-
zations are not needed in recruiting participants. Therefore, we expect that
in countries where public opinion is prevailingly antiwar, the mobilization
patterns will be more open, that is to say, more frequently employing open
channels and less dependent on ties to movement organizations.

The mass media coverage of an issue is related to the public opinion
within a country; this was established in chapter 3. We expect that mobi-
lization becomes increasingly open the more the mass media sympathize with
the cause of a protest campaign. This is due to the mobilizing role of the
media as such. Since media are most effective in disseminating or spreading
the protest message, they contribute to open mobilization processes. More-
over, media coverage and its tone are associated with public opinion: media
might play a part in creating a sentiment among the population and/or rein-
/fcs force public opinion. Either way, sympathetic coverage of a protest issue, we /is

| suspect, is linked to open mobilization. The same holds for the run-up to

the Iraq War: the more the media show sympathy for antiwar protests, the
more open the mobilization will be.

Supply factors refer to the set of organizations staging and supporting
protest events and to their numbers, strength, activity level, history, and action
repertoire. In general, we expect closed mobilization to be more important
in countries where the social movement sector is dense. We believe that this
density accounts for the significance of ties and closed channels in the mobi-
lization process. Only a limited number of organizations are able to generate
massive street mobilization by relying on their own networks. Traditionally,
three types of actors belong to this category: labor unions, political parties,
and new social movements (including the global justice movement). The
antiwar coalitions in the various countries differed in the scope, composi-
tion, and strength of the organizations included and the extent to which
these organizations have a track record of organizing mass demonstrations.
We will use several measures that tap the density and strength of the social
movement sector in a country and employ those in analyses that link density
and strength of the movement sector to the openness of the mobilization
process. More specifically, we expect that ties to movement organi

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 18B^_

WALGRAVE AND KLANDERMANS

^

become more important and mobilization channels more closed the more
the density and strength of the movement sector increase.

Table 9.5 contains simple Pearsons correlations between the two depen-
dent variables (openness or closedness and strength of ties) and six inde-
pendent variables at the country level. The two dependent variables are the
average strength of ties and the average openness of mobilization channels
in each country. The independent variables are a demand-side factor, namely
the degree of opposition to the war in the public opinion, and five supply-
side factors, namely the activity level of new social movements, union mem-
bership, and the level of party membership in a country as indicators of the
density of the movement sector; the demonstration culture and strike activ-
ity levels in a country serve as indicators of the level of contentiousness of
the movement sector in a country 5 Because the number of cases is very small
(N = 8), coefficients have to be very high to be significant. Therefore, we use
a lower threshold than usual.

We found support for our assumption that mobilization patterns are
affected by supply and demand factors — quite a few coefficients are (mod-
estly) significant — but the evidence is complex. First, not all correlations pass
the set significance threshold. Moreover, there appears to be a significant vari-
ation in the correlation patterns. Sometimes the strength of ties is predom-
inantly affected, sometimes the mobilization channels used, and sometimes
both. We found, indeed, that mobilization patterns tend to be more open
in countries where the opposition to war was stronger: participants there
were more often mobilized via open channels. Furthermore, as expected, the
denser the movement sector was in a country, the more closed were patterns
of mobilization. In the case of new social movement activity, this was re-
flected in the mobilization channels used, whereas in the case of union and
party membership it reflected more in the significance of strong ties for the
mobilization. Finally, if the movement sector in a country was more con-
tentious, mobilization for the Iraq demonstration tended to increasingly work
via open mobilization channels.

In summary, mobilization patterns appear to vary among countries. In
some countries, the antiwar protest was characterized by an open pattern,
in others it was more closed. As we expected, the process of mobilization is
formed by the configuration of supply and demand factors in a country; at
least this is what our findings suggest. The pattern of correlations is com-
plex, however. Demand factors appear to influence which channels are used
but not the significance of ties; some supply factors affect both indicators of
the openness of mobilization, others only what channels are used or on the
importance of ties. The involvement of citizens in new social movements

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 189_^_

"4

5

>

1 e

§1

S R

(S

.5- c

fll

O CN

°- E 9-

1 a.

■ij

" CO

^ i

E

1l*

C, ^

c3 c

oS SR

ill

'

o 1

e

1

8 S

2 " s

a -f S

£ u

H$H

-4

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 190_^_

I90 WALGRAVE AND KLANDERMANS

seems to increase the usage of open channels but the significance of ties does
not, while involvement in traditional organizations such as labor unions and
political parties seems to generate the opposite pattern. The level of con-
tentiousness of the movement sector again seems to reinforce the use of open
mobilization channels. Obviously, in order to understand the dynamics of
supply, demand, and mobilization, more systematic research is called for,
but this much is clear thus far: they interact in a complicated manner to gen-
erate protest.

Conclusion

In this chapter, we concentrated on patterns of mobilization and distinguished
among more open and closed patterns of mobilization. Openness and closed-
ness were defined in terms both of the communication channels employed
and of the prominence of strong ties in the mobilization campaign. Closed
patterns of mobilization depend more on strong ties and closed channels of
communication; open patterns do not depend on strong ties or ties at all
and employ open channels of communication, such as mass media or inter-
personal networks.

We demonstrated that the openness or closedness of a mobilization cam-
paign affects aspects of the mobilization process such as the companionship
of participants at the demonstration, the moment they decided to take part,
and the distance they traveled. People who were mobilized in a closed rather
than an open manner are accompanied by colleagues or comembers of orga-
nizations rather than with family or friends and decide earlier to participate
and travel farther distances. Moreover, open and closed patterns mobilize dif-
ferent sorts of participants. This holds for such demographics as gender and
age, political attitudes and behavior, and for attitude toward the war. We ob-
served several striking differences between channels and ties in their impact
on these participant characteristics and discovered some crucial interactions
between the two aspects of openness. We found that, on the whole, partici-
pants who were tied to organizations that staged the demonstration were
more involved in politics and leaned more to the left. Of the mobilization
channels, only movement organizations had the same effect. The other chan-
nels did not tap into politically specific populations. It is interesting to see that
the effects of channels only exist among participants without ties or with
weak to moderate ties. Among people with strong ties to movement organ-
izations, mobilization channels no longer made a difference. This, of course,
makes sense, as the effect of channels only holds for organizational channels.
Chances are, of course, that people who have strong ties to movement orga-
nizations are mobilized through those organizations. For obvious reasons,

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 1 91^

OPEN AND CLOSED MOBILIZATION PATTERNS I9I

people who were tied to movement organizations that were staging a demon-
stration or who were mobilized direcdy through those organizations were more
opposed to the war and dissatisfied with their governments policies.

We also demonstrated that countries differ in terms of mobilization
patterns employed and that these differences are related to a variation in the
configuration of demand and supply factors in a given country Strong de-
mand seems to reinforce open mobilization patterns, as does a contentious
social movement sector, whereas a dense social movement sector seems to
reinforce closed mobilization patterns. The evidence from our small sample
of countries is far from conclusive, but it does suggest that different con-
figurations of demand and supply generate different mobilization patterns.
This is an important finding and certainly worth pursuing in further research,
as it alludes to the dynamic relationship among three key factors in the emer-
gence of protest events.

Notes

1 . Weak ties imply that someone only knows people who are members of a
movement organization. Moderate, strong, and very strong ties imply an increas-
ingly dense combination of knowing someone, being a member oneself, and main-

| taining contact with movement organizations.

~tyj~ 2. The Italian respondents were mobilized in a much more organizationally v®r

embedded manner: they traveled much farther than their colleagues in other coun-
tries, they most frequently mentioned organizations as their main source of infor-
mation for the demonstration, more of them attended the demonstration in the
company of other organization members or colleagues, and they made the decision
to participate fairly early.

3. We ran the same analyses with and without Italy, and we ran separate analyses
on the Italian sample to assess whether the diverging sampling strategy in Italy has
biased the conclusion. As this was not the case, we included Italy in the analyses.

4. The large main effect of "country" is due to the Italian sample. If we omit Italy,
the F-value reduces to 28.24. In this analysis, the remaining F-values barely change.

5. The degree of opposition to war in public opinion is based on the EOS
Gallup poll on the war carried out in Europe just before February 15. General activ-
ity levels of new social movements in countries is derived from the European Social
Survey (no data on the United States or on Switzerland) asking for participation in
an activity of a humanitarian or an environmental/peace organization during the last
1 2 months. Figures for general union membership are derived from the World Labour
Report 1997-1998 of the International Labour Organization (www.ilo.org). Gen-
eral demonstration culture in a country is based on the World Values Survey answers

is about participation in a lawful demonstration. Strike activity levels are

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 192^_

192 WALGRAVE AND KLANDERMANS

based on figures of the ILO, taking the yearly average of the absolute number of
demonstrations recorded in a country between 1998 and 2003 divided by the pop-
ulation size. Statistics about general partisan membership in a country are based on
Mair and van Biezen (2001).

Works Cited

Diani, Mario. 2002. "Network Analysis." In Methods of Social Movement Research,
ed. Bert Klandermans and Suzanne Staggenborg, 173—200. Minneapolis: Uni-
versity of Minnesota Press.

Diani, Mario, and DougMcAdam, eds. 2003. Social Movements and Networks: Rela-
tional Approaches to Collective Action. Oxford: Oxford University Press.

Gould, Roger V. 1991. "Multiple Networks and Mobilization in the Paris Com-
mune, 1871." American Sociological Review 56: 716—29.

. 1993. "Collective Action and Network Structure." American Sociological Review

58: 182-96.

Granovetter, Marc. 1973. "The Strength ofWeakTies." American Journal of Sociol-
ogy 78: 1360-80.

Kitts, James A. 2000. "Mobilizing in Black Boxes: Social Networks and Participation
in Social Movement Organizations." Mobilization 5: 241-57.
I Klandermans, Bert. 1984. "Mobilization and Participation: Social Psychological Ex-

W pansions of Resource Mobilization Theory." American Sociological Review A3: ~W~

' 583-600. T

. 1997. The Social Psychology of Protest. Oxford: Blackwell.

. 2004. "The Demand and Supply of Participation: Social Psychological

Correlates of Participation in a Social Movement." In Blackwell Companion to
Social Movements, ed. David A. Snow, Sarah A. Soule, and Hanspeter Kriesi,
360-79. Oxford: Blackwell.

Klandermans, Bert, and DirkOegema. 1987. "Potentials, Networks, Motivations, and
Barriers: Steps toward Participation in Social Movements." American Sociolog-
ical Review 52: 519-31.

Mair, Peter, and Ingrid van Biezen. 2001. "Party Membership in Twenty European
Democracies, 1998-2000." Party Politics 7: 5-21.

Marwell, Gerald, and Pamela Oliver. 1993. The Critical Mass in Collective Action: A
Micro-Social Theory. Cambridge: Cambridge University Press.

McAdam, Doug. 1986. "Recruitment to High-Risk Activism: The Case of Freedom
Summer." American Journal of Sociology 92: 64-90.

McAdam, Doug. 1988. f, New York: Oxford University Press.

McAdam, Doug, John McCarthy, and Mayer N. Zald, eds. 1996. Comparative Per-
spectives on Social Movements: Political Opportunities, Mobilizing Structures, and
Cultural Framing. Cambridge: Cambridge University Press.

H$H

9 Chapter 9_Walgrave 22/01/2010 3:39 PM Page 193^_

OPEN AND CLOSED MOBILIZATION PATTERNS I93

McCarthy, John D., and Mayer N. Zald. 1976. "Resource Mobilization and Social
Movements: A Partial Theory." American Journal of Sociology 82: 1212—41.

Morales, Laura. 2008. Joining Political Org/; utions, Mobilisation, and

Participation in Western Democracies. Colchester, England: European Consor-
tium for Political Research Press.

Norris, Pippa, Stefaan Walgrave, and Peter Van Aelst. 2005. "Who Demonstrates?
Anti-State Rebels or Conventional Participants? Or Everyone?" Comparative Pol-
itics 2: 251-7 '5.

Putnam, Robert. 2000. Bowling Alone: The Collapse and Revival of American Com-
munity. New York: Simon & Schuster.

Oegema, Dirk, and Bert Klandermans. 1994. "Non-Conversion and Erosion: The
Unwanted Effects of Action Mobilization." American Sociological Review 59:
703-22.

Snow, David A., Louis A. Zurcher, and Sheldon Ekland-Olson. 1980. "Social Net-
works and Social Movements: A Microstructural Approach to Differential Re-
cruitment." American Sociological Review 45: 787-801.

Tindall, David B. 2004. "Social Movement Participation over Time: An Ego-Network
Approach to Micro-Mobilization." Sociological Focus 37: 163-84.

Van Aelst, Peter, and Stefaan Walgrave. 2001. "Who Is that (Wo)man in the Street?
| From the Normalisation of Protest to the Normalisation of the Protester." Euro-

■W- pean Journal of Political Research 39: 461-86. W

' Walgrave, Stefaan, and Jan Manssens. 2000. "The Making of the White March: The

Mass Media as a Mobilization Alternative to Movement Organizations." Mobi-
lization 5:217 '-40.

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Pagp. 13&fe

10
Promoting the Protest:
The Organizational Embeddedness of the Demonstrators

Mario Diani

m

Although they have been frequently associated with peace movements (and
understandably so), the February 15 demonstrations were, first of all, spe-
cific protest events, as large and impressive as they were. As we know, the
relationship between protest events and social movements is a complex one.
Regardless of whether we define them as "sustained interactions between
power holders and authorities," a la Tilly (see, e.g., 1995, 369; 1994), or as
"informal networks linking individuals and/or organizations, engaged in a
conflict on the basis of a shared collective identity" (Diani 1992, 13; Diani
and Bison 2004), social movements are usually associated with collective
action, displaying substantial continuity over time. Specific events, no mat-
ter how big, do not necessarily make a movement. Rather than the peak of
long-term mobilization efforts conducted by peace campaigners, the Febru-
ary 15 demonstrations might have been instead the outcome of one-off efforts
conducted by ad hoc coalitions, set up precisely for those events.

Therefore, looking at the composition of the coalitions that mobilized on
February 15 may provide valuable insights about the nature of those marches.
In general, those coalitions were highly heterogeneous, as their members
ranged from established interest groups and political parties to environmen-
tal, human rights, development, women's, and ethnic minority groups; global
justice organizations; churches; and so forth. But what was the relative pres-
ence of peace activists and peace organizations in the different marches? Did
these mainly attract people with long-lasting commitment to peace activism?
Or, regardless of their numbers, did those people and their organizations play
a central role in promoting the demonstrations? If that was the case, then

H$H

-4

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 12S^_

PROMOTING THE PROTEST I95

the February 1 5 events could be most plausibly seen as peaks in longer waves
of peace-movement mobilization, following in the wake of the campaigns
originated by the attack on Afghanistan in fall 200 1 . If, however, the dem-
onstrators came overwhelmingly from other sectors of social and political
activism, then an interpretation of the demonstrations as a major expression
of broader political conflicts within the specific countries might be in order.
Similar conclusions might be reached if one discovered that organizations
without peace profiles were most prominent among those promoting the
marches. For example, if one found this the case in Italy, it would not be
difficult to link the February 1 5 demonstration in Rome to the larger protest
wave against Silvio Berlusconi's right-wing government (della Porta and Diani
2005).

To address these questions, I will draw on our dataset from two differ-
ent if integrated perspectives. First, I will look at the individual demonstra-
tors' background in collective action and in associations. By doing so, it will
be possible to present hypotheses about the sectors of civil society that con-
tributed to the demonstrations, and their interdependence. Then, I will focus
more specifically on demonstrators' involvement in the organizations pro-
moting the demonstrations. This will enable me to assess what types of orga-
nizations played the most central roles in which country, as well as the extent
to which organizations central to the promotion of the marches were actu-
ally connected to peace activism, and thus were expression of a longer term
peace activist agenda, rather than actors linking the peace issues to other types
of collective agendas.

I shall also relate our findings to the characteristics of the political con-
text in which the different mobilizations occurred. Since the 1980s, theories
explaining the success and traits of social movements as a result of favorable
"political opportunities" (della Porta 1995; Tarrow 1998[1994]; Kriesi 2004)
have addressed at least three dimensions of the political context: the most
permanent, structural features of a given polity. Among these are the insti-
tutional arrangements available to facilitate citizens' access to the political
system, the established cleavages, or the cultural traits of a society that may
facilitate or hinder the spread of certain interpretations of reality and/or of
certain identities (Kitschelt 1986; Rucht 1990; Diani 1996; Koopmans and
Statham 1999; see also chapter 2 in this volume). These properties are the
ones most closely associated with the concept of "political opportunity struc-
ture" and are often differentiated from more volatile political opportunities
(see, e.g., Gamson and Meyer 1996), such as those offered by changes in the
configuration of political actors potentially supporting or opposing a certain
movement (their alliance and conflict structure: Curtis and Zurcher 1973;

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page tl

"4

Klandermans 1990), as well as in social n
This expression refers to the opportunities created by the interplay of elites'
and policy-makers' strategies and social movements' strategies (Kriesi 2004,
70). The explananda for these theories have been similarly varied, ranging
from the reasons accounting for specific episodes of collective action to large-
scale changes in the levels of participation in contentious politics or in the
spread of certain action repertoires (e.g., Tilly 1995). Chapter 3 in this vol-
ume presents an empirical assessment of the specific interaction context con-
cerning the February 1 5 demonstrations in the eight countries under study.
In this chapter, I stay on the parsimonious side in my choice of both
explanandum and of explanatory model. More specifically, I look at the com-
position of the coalitions challenging the attack on Iraq and at how it might
be affected by a distinctive element of the "interaction context," namely,
national governments' policy toward the war (see chapter 3 for details on
these positions). As argued in chapter 2, it would make little sense to try to
explain inherently variable traits such as the composition of coalitions con-
verging on a specific event, in the light of the most permanent features of
the polity. For example, while the United States could in general be regarded
as a weak state offering many points of access to citizens' organizations (Kit-
schelt 1986; Kriesi 2004), such broadly structural factors are unlikely to have
created a favorable opportunity structure for the 2003 peace protest. In con-
trast, one might plausibly suggest that policy options and political discourse
combined to create most unfavorable opportunities in the countries that
actively promoted the war and whose forces took the heaviest toll (the United
States and the United Kingdom). We specified this in chapter 3 when we
showed that politics, media, and public opinion in the United States and the
United Kingdom were, indeed, most supportive of the war compared to the
other countries. In war-initiating countries, only the organizations most com-
mitted to the peace cause might be expected to play a significant explicit
role. The others might be discouraged from actively engaging in the demon-
s for fear of being somehow labeled as "antipatriotic." In contrast, in
whose governments opposed the war, such as Belgium, Germany,
or Switzerland, the demonstrations might have attracted a broader range of
organizations. These might be expected to be broadly representative of the
full spectrum of groups and associations, active in those center-left sectors
of civil society most likely to criticize the military intervention. Finally, in
countries where governments supported the war but with a limited military
commitment, such as Italy, Spain, or the Netherlands, there might be the
highest chances for conflicts opposing government and opposition about
the war issue. Accordingly, a more substantive role within the coalitions for

H$H

-4

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page lflSfe

PROMOTING THE PROTEST V)~]

established left-wing political actors such as political parties and unions might
be in order.

Demonstrators from Associations and Protest Communities

February 1 5 was the result of the mobilization of highly diversified sectors
of civil society in the different countries. Therefore, to understand the posi-
tion of the demonstrations in their national politics, it is necessary to explore
their members' overall involvement in other types of collective action. Pre-
liminary attempts to map civil society structures from this perspective have
largely focused on organizational multiple memberships (Knoke and Wood
1981; Diani 1995; Carroll and Ratner 1996). However, one must recognize
that participation in collective action participation takes at least two differ-
ent forms: through associations and in the context of loosely structured protest
communities (Staggenborg 1986; Taylor and Whittier 1992; Melucci 1996;
Doherty, Plows, and Wall 2003; Diani 2005). The latter are usually re-
garded as less stable and more occasional forms of interaction, often associ-
ated with specific protest events or campaigns, unable to show the same
capacity to last as formal organizations. When continuity in relation to infor-
mal networks is spoken of, it is often in reference to counter- and subcultural
networks, seen as preconditions rather than as instances of collective action
in their own right. However, even sustained participation in protest com-
s may be expected to create both the shared frames and the social
s that normally result from participation in associational life.
Accordingly, I look at participation in associations and at sustained par-
ticipation in protest activities as two different practices that, for all their dif-
ferences, nonetheless generate ties and solidarity among the people involved.
For this reason, both may be expected to secure the continuity required for
collective action to develop over time. For the same reasons, they may also
be expected to have had a role in shaping the February 1 5 demonstrations.
I will refer to these two sets of experiences as the associational field and the
protest community. In this chapter, I consider part of the associational field
those demonstrators that in February 2003 were active in at least one of fif-
teen different types of organizations. Apart from peace organizations, these
included organizations active in transnational and global issues (antiracist or
immigrant's rights organizations; global social justice organizations; third-
world organizations; human rights organizations); political parties; labor
unions and professional organizations; sociocultural organizations (art, music
or educational, charitable, and church or religious organizations); new social
movement organizations (associations for women's rights, environmental orga-
nizations); sport, leisure, and youth associations. As for the protest community,

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page tl

I looked at people who in the previous five years had taken part in more than
ten demonstrations regarding issues of peace, antiracism, human rights, the
third world, social inequality and labor, the environment, global justice, and
women's rights.

Combining high intensity of participation in demonstrations with active
participation in associations at the time of the march, we get a more com-
plete picture of the collective action settings in which the February 1 5 marches
were embedded (Table 10. 1). Overall, levels of active participation were very
high. Only one quarter of the demonstrators were neither actively involved
in associations nor in protest communities. At the other extreme, slightly
more than one in ten demonstrators was involved in both. Over half were
active in associations, while a mere 3 percent were participating in protest
communities only. This profile changed significantly across countries:

1 . In four countries (the United States, the United Kingdom, the Neth-
erlands, and Switzerland), demonstrators came disproportionately
from associations, while they had below- average levels of participation
in protest communities.

2. Two countries (Spain and Germany) had the largest shares of demon-
strators not involved in either type of collective action (52 percent

_^L and 49 percent). _^L

| 3. The remaining two countries (Belgium and Italy) combined above-

average levels of participation in associations (if slightly, as in Italy)
with above-average involvement in protest communities.
4. All in all, Italy and the United States showed the most different
profiles, with Italian demonstrators highly involved in protest com-
munities (36 percent versus an average 16 percent) and with only
average participation in associations, and Americans displaying excep-
tionally high levels of associational participation (85 percent versus 71
percent) and below the average involvement in protest (10 percent).

That the associational field and the protest community were actually dis-
tinct is demonstrated by the links created by overlapping memberships, that is,
by those demonstrators who were active in more than one type of as
or protest field. Grouping together types of associational and protest a<
on the basis of the number of demonstrators simultaneously engaged in them
always identified two main sets, one corresponding to associations and the
other to protest communities. This suggests the convergence on the demon-
strations of two fairly distinct sectors of civil society. This was by no means
an inevitable outcome. In principle, one could have also expected stronger
recurrent links between associations and protest communities focusing on

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page tl

"4

-5

VO oo rn CN

CN

i

>/<

i

<u

O

E

J

|

1

CL

c

I

£

SQ3 M 3

s

1

•^

1

z

^)°- >*

to

c

J

22

i

~~

M ^ M

fc

1

r£

Q_

M ^„ ^

m

*

W

<* -*

5

o

1

.g

o

'§

o

3

<* VO ~ 0>

o

.=

':

1

o

1

2 s

8

Q-

"c § "c a

^J

°s«:2

o

§ 1 § §

<u

a"l S'§ 1

2

£

Z < (5 <

Z

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page

similar issues (for example, between ecology organizations and ecology pro-
r between women's organizations and women's protest
:s). Instead, data reduction procedures robustly showed that links
were stronger among associations on one side and protest communities on
the other, regardless of the issues addressed, than between associations and
protest communities interested in similar issues.

"4

Promoting the Protests

Within that broad range of protest and associational activities, it is of par-
ticular interest to look at the demonstrators' involvement in the organiza-
tions that promoted the march. While innumerable organizations signed up
to march, their relative weights and contributions varied drastically. A closer
look at the relationship between protestors and the organizations promot-
ing the marches may provide a better understanding of who played what
role in what country. Almost one fifth of respondents claimed membership
in such organizations, and one quarter reported acquaintances in some of
them; slightly over half reported no ties whatsoever (Table 10.2). One could
question the validity of these figures, as respondents might not have been
that informed about the involvement of their organizations in the promo-
tion of the marches. This might be the case; however, the differences between
some membership figures, depending on whether we look at promoting orga-
nizations (Table 10.4) and organizations at large (Table 10.5), suggest oth-
erwise. For example, while 27 percent of the Dutch demonstrators belonged
to unions or professional associations, only 1 percent of those claiming to
belong to one promoting organization did (consistent with the fact the unions
in the Netherlands were hardly involved as march organizers). Other possi-
ble examples include membership in parties and unions in the United States
(Tables 10.4 and 10.5).

The discrepancy between general organizational membership and in-
volvement in specific promoting organizations suggests that the demonstra-
tions attracted a population that was not restricted to the membership and
constituency of their backing organizations. At the same time, however, such
a profile was not evenly spread across the different countries. The British
demonstrations attracted the highest number of people without any tie to
the organizers and also by far the fewest members of promoting organiza-
tions. At the other extreme, Italy was the only country with a very high pro-
portion of demonstrators who were members and also the only one to have
a balance between members and people with no ties. The United States also
had a high number of people with no ties, while Spain was the closest to the

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Pagp. ?0^

"4

|

£^ M <N

■"

£-

=1

•* vo O 00

<

if\ (N tS C4

8

o

^o c^ -h oo

8.

c

1

s

£

£ 8 S fe

A

g

D

<

■* mN 3

'E

o

2>

o

O)

c

8

o

E

o

Q.

1

=

aa.oti

(D

tE
o

1

t=

£££§

g

o

£

"*

D

■z

<!

R2 m S

■a

o

E

(D

V

S 2 """ Si

-o

E

if)

O

£

Z>

S <N 2 o

3

l

«3

CN

d

a

« 1 S

_Q

■an

H$H

-4

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page

Italian figure, as more than half of its demonstrators had sc

to the organizers. The other countries were closer to the average.

A clearer pattern emerges in looking at the distribution of these figures
depending on the national governments' attitudes toward the war (Table 1 0.2) .
The percentage of people who participated without the incentives provided by
close ties to some of the promoting organizations was highest among demon-
strators from the nations that promoted the war (United States and United
Kingdom), lowest among those whose governments supported the war (Italy,
the Netherlands, and, at the time, Spain), and average in countries that op-
posed it (Belgium, Germany, and Switzerland) . As I argue in the next section,
these differences are likely to stem from the narrower range of promoting orga-
nizations in the United States and the United Kingdom than in other coun-
tries and from the greater range of organizations promoting the marches in
countries where governments supported the war, such as Italy or Spain.

If this approach is extended approach to participation in associations
and in the protest community at large, one might ask the extent to which
the organizations behind February 1 5 managed to act as bridges between the
different sectors of civil society that mobilized on that day. In all countries,
the demonstrators connected with promoting organizations also showed much
stronger connections to both the world of associations and protest commu-
nities (Table 10.3): the percentage of demonstrators active in both sectors was
three times higher among members of promoting organizations than among
people who knew somebody there and nine times higher that of people with
no ties to the demonstrations promoters whatsoever. Beyond promoting
and coordinating the marches by establishing coalitions of organizations,
the organizations behind the demonstrations also linked the different sec-
tors of civil society concerned with the Iraq War through the multiple in-
volvements of their individual members.

Table 10.3. Current participation in associations and protest

communities by contact with promoting organizations
(percent)

None Know member Member Total

None 37 24 26

Associations only 57 58 63 58

Protest communities only 3 6 3
Associations and prt

N

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Pagp. ?oa^

PROMOTING THE PROTEST 203

What Organizations Are Behind the Demonstrations?

What organizations were more frequently indicated as promoters of the
marches by the demonstrators? We recoded the names of the organizations
to which demonstrators belonged under seven headings, consistent with the
types introduced earlier: political parties; unions and professional associations;
peace organizations; organizations active on global justice issues; new social
movement organizations; sociocultural organizations; sport, leisure, and youth
associations. Parties and unions — the latter more than the former — were by
far the most represented among demonstrators. The countries where the role
of "old politics" actors in promoting the marches were most conspicuous
included the Netherlands, Spain, Germany, and Italy. In particular, political
parties were most prominent in the Netherlands (Groen Links and SP), even
though Rifondazione Comunista in Italy and the socialist PSOE in Spain
stood out in a context where parties were otherwise fairly minor players. As
for unions, they are very important in both Mediterranean countries and in
Germany (CGIL in Italy, the Comisiones Obreras and the UGT in Spain,
GEW and Ver.di in Germany).

In contrast, the role of interest representation organizations seemed neg-
gible in both countries directly promoting the war, United States and United

, llglUlC 111 UUU1 CUUlllilCS UllCCUy piUlllUUllg UlC Well, W111LCU OLdLCO dlld V^lllLCU

Afr- Kingdom. There, peace organizations and peace coalitions, such as CND .fa.

I and the Stop the War Coalition in the United Kingdom, and SNOW in the

United States, were the main players in promoting the marches. This seems
related not to the shortage of party or union members among demonstrators
(Table 10.5) but to those organizations' reluctance to engage directly with
promoting the marches. In the United Kingdom, 28 percent of the demon-
strators were actually current or former members of political parties, in line
with the 30 percent recorded for the total sample, while in the United States
the share was as high as 54 percent. Likewise, current and former union mem-
bers amounted to 44 percent in the United States and 43 percent in the
United Kingdom, versus a general figure of 37 percent.

Moving to other organizations, transnational organizations were strongly
represented in Belgium and Germany, whereas they were hardly noticeable
in other countries. The role of groups mobilizing on migration, human rights,
and global inequality issues was particularly prominent (11.11.11 and Oxfam
in Belgium, ATTAC in both countries). The presence of environmental and
religious organizations appeared minimal across the board, with Belgian and
(curiously, given the weakness of Spanish environmentalism) Spanish dem-
onstrators slightly overrepresented among the former (particularly Greenpeace
in both countries), Germans and Dutch among the latter. 1 Other organizations

H$H

-4

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 2fl&fe

204 DIANI

with recurring mentions included Amnesty International and — in Italy —
ARCI, the umbrella cultural association of the Italian Left. 2

If we look at countries' different approaches to war (Table 10.4), the pic-
ture is very clear: in countries promoting it, where government's opportunity
to play the patriotic card was easiest to seize, specific peace organizations had
to take up most of the organizing roles in the marches and to mobilize the
largest share of demonstrators (this is also reflected in the highest share of
demonstrators in those countries with previous experiences in peace orga-
nizations: 28 percent, versus 12 percent in the other two types of countries).
The strength of patriotic feelings and of the rhetoric about "supporting the
troops" may have discouraged many organizations from actively promoting
the demonstrations, and participation in these may have been guided prin-
cipally by strong personal motivations. For example, while many Labour
Party members in the United Kingdom actually attended the demonstrations,
their party was officially hostile to them, even though some local branches
actually were involved in the organization of the marches. 3

In countries that supported the war but were unable to play the patri-
otic card because of their lowest commitment, like Italy or Spain, interest
organizations played the biggest organizing role. This further confirms the
war issue in those countries being embedded in broader conflicts between
the ruling coalition and its political opposition, as was detailed in chapter 6.
This might have resulted in more demonstrators with links to the political
organizations promoting the march (in particular, the organizations of the
Left). This accounts particularly well for the Italian figure, showing a very
high involvement with promoting organizations (even when controlled for
different styles of data collection, as the figures in parentheses in Table 10.2
demonstrate). In Italy, the intensity of political conflict and mass protest had
been particularly high in the two years preceding the march, that is, since
200 1 , when Berlusconi had taken over as prime minister.

Finally, in countries where government and opposition were both against
the war and, thus, no domestic conflict about the war was present, actors
that were difficult to locate with respect to the traditional left-right cleavage
played a major organizing role. These included organizations close to the new
social movements as well as operating on transnational issues or on moral
and social issues, which, once again, could not be easily associated with
left-right divisions (Table 10.4). In those countries, the chances of the dem-
onstrations taking a more distinctively super- or transnational dimension
may have been highest.

Of course, all the findings discussed above might simply be due to the
fact that in different countries demonstrators were in general more involved

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 2QS^_

"4

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page

with certain types of organizations than with others, regardless of whether
those organizations participated in the promotion of the demonstrations.
However, a look at the distribution of organizational memberships at large
among demonstrators suggests this not to be the case (Table 10.5). With the
only exception of sport, leisure, and youth associations, U.S. demonstrators
were significantly more involved than the others in all types of organiza-
tions; yet only peace organizations were overrepresented among organizers
of the U.S. protests. As for the UK demonstrators, they were no more involved
in peace organizations than the others, yet only peace groups, again, seemed
to be playing the organizers' role.

Conversely, parties and unions had figures around the average in coun-
tries supporting the war, despite their overwhelming role as promoters of the
demonstrations. Among the countries whose governments opposed the war,
Switzerland and Belgium actually fit the pattern, as global justice, new social
movement, and sociocultural organizations were as overrepresented among
demonstrators in general as they were among those linked to promoting orga-
nizations. Germany, however, does not fit it, as the same organizations were
consistently underrepresented among those who took to the streets. All in
all, global justice, new social movement, and sociocultural organizations were
| not as present among the general demonstrators as they were among those

~W~ demonstrators who were members of promoting organizations. v9)~

Explaining Membership in Promoting Organizations

I will now explore more in detail the relationship between membership in
certain types of organizations and membership in the specific organizations
that promoted the demonstrations. The latter may surely depend on the dis-
tribution of different types of organizational memberships in general, although
the paragraphs above do not suggest this. It is also important to see the extent
to which members of the promoting organizations differed from the others
on a number of important dimensions.

As gender, education, and age have proved to have an impact on pat-
terns of political participation at large, it might be worth exploring the pos-
sibility that they influenced involvement for the organizers of the February
15 marches. On the one hand, the organizations promoting the marches
might be more established than the others (a reflection of their capacity to
carry on a remarkable organizational effort), and this might result in their
members having the traits usually associated with more established political
participation: prevailingly male, with higher levels of education (i.e., in this
context, holding degrees), and relatively old (i.e., in their thirties or forties
rather than in their twenties). On the other hand, the strong association of

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 2flZfc)

"4

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page

the peace campaigners with the global justice movement might lead us to
expect exactly the opposite profile, with relatively young demonstrators, a
greater proportion of females, and a greater role for working-class and deprived
people (hence, a smaller presence of degree holders than, say, in the typical
new social movements of the 1980s and 1990s). Table 10.6 below illustrates
results separately for each country. Sociodemographic traits only played roles
in Italy, where members of promoting organizations were older and more
frequently male than nonmembers, and in Switzerland, where older demon-
strators were also overrepresented.

There are also reasons to expect differences between members of pro-
moting organizations and the rest when looking at the demonstrators' atti-
tudes and ideological orientations. The self-placement on the left-right scale
could either reflect a more moderate position of the members of the pro-
moting organizations vis-a-vis the other demonstrators or the capacity of
more radical sectors to promote a broader agenda, capable of attracting rel-
atively more moderate sections of the public. Similar remarks could apply to
sympathy for the global justice movement or for the level of opposition to
war (measured in terms of higher scores on the "opposition to war scale," see
chapter 6 in this volume).

re, sympathy for the global justice movement only played a

| As it were, sympathy for the global justice movement only played a

"W" modest role in two countries (see Table 10.6), the United States and the ~®~

Netherlands, while the strength of antiwar feelings did not at all distinguish
members of promoting organizations from the other demonstrators. Previ-
ous recourse to radical protest was most important in the United States (a
further confirmation of the very distinctive nature of the public mobilized
in that country); it was also significant, but to a much smaller degree, in Italy
and Switzerland. Consistently across continental Europe, members of pro-
moting organizations were more left-leaning than the other demonstrators,
despite the relative ideological homogeneity of the demonstrators, with Spain
the only exception. Other than Switzerland, these were countries with "con-
servative welfare regimes" (Esping-Andersen 1990), where traditional polit-
ical cleavages were more salient and the boundaries between institutional
and protest politics seemed less stable. By contrast, the left-right divide seems
less relevant in liberal democracies such as the United States and the United
Kingdom, characterized by more individualistic styles of participation. Spain
and Switzerland represented significant exceptions to this overall pattern.

Looking at the broader organizational memberships provides a consis-
tent picture, yet one more articulated than the one I presented earlier. Being
a member of a peace organization drastically increased one's chance of being
associated with a promoting organization in the United States and the United

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page

.9- 'rt

0<2d£3zJ

[ll ®

H$h-

-4

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 2Hjj j fc)

Kingdom. It also did so, if to a more limited extent, in the Netherlands, Switz-
erland, and Germany but not in Spain, Italy, or Belgium — a further proof
that the association between peace demonstrations and peace organizations
should not be taken for granted. By contrast, members of traditional parties
and unions were more likely than nonmembers to be associated with the
promotion of the marches in all countries but the United States. However,
those chances were highest where the peace issue was mostly controversial
given the governments' position, namely, Spain, Italy, and the Netherlands
(although the impact of parties and unions was also strong in Germany, where
the issue was not controversial). Global justice organizations' members were
more frequently involved organizing marches where I would expect them to
be, according to my previous exploration, that is, in Belgium, Germany, and
(if to a smaller measure) Switzerland. However, they were also heavily involved
in both Italy and Spain.

Finally, membership in other organizations had no significant role in
promoting the demonstrations, and when it did, it was a negative one. Mem-
bers of sport, leisure, and youth associations in Spain and of sociocultural
(i.e., cultural, charitable, and religious) organizations in Italy were indeed less
likely than the other demonstrators to be connected to an organization pro-
moting the marches. In the case of the organizations less directly associated
with political action, like charities, cultural associations, or sport clubs, mem-
bership may have had an impact on individuals' predispositions to attend
the demonstrations, but not on those demonstrators' involvement with the
promotion of events. Remarkably, membership in the organizations active
on environmental and women's issues, that is, those more directly associated
with the new social movements, seemed to have hardly any effect (except in
the United States, and even there, modestly so): surely important in terms
of the political socialization of the individuals, it did not facilitate a more
direct association with the demonstration promoters.

Conclusion

The February 1 5 demonstrations not only attracted substantial numbers of
people with no previous experiences of collective action — which is made even
more remarkable by their being national demonstrations, located in the cap-
itals or in other major cities, and therefore more demanding to attend (see
chapter 9 in this volume). They also mobilized very broad sectors of civil
society, at least of its leftist sections. The organizers showed a complex rela-
tionship to peace activism. On the one hand, many organizations were not
primarily oriented to peace issues; on the other, their members had a substan-
tial background in peace activism. Not only that, they also had above-average

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 21^_

"4

PROMOTING THE PROTEST

o both the world of associations and to the protest c<
operating within civil society. In this sense, members of promoting orga-
nizations acted as links between specific peace issues and much broader sec-
tors of collective action, and they did so from a perspective that was only
occasionally more radical than the other demonstrators' in terms of action
repertoires, strength of opposition to war, or links to global justice move-
ments. Members of promoting organizations, however, were more left-wing
than the other participants in most European continental countries. In any
case, the promoting organizations did not stand out for their moderation. If
anything, when differences were significant, they suggested that promoting
organizations were more radical rather than more moderate than the others.

Generally, the organizations that promoted the marches attracted quite
different sectors of civil society in different countries, with political parties
and unions underrepresented among their members (although not among
the demonstrators in general) in the United States and the United Kingdom,
and overrepresented in Italy and Spain, peace organizations dominant only
in the Anglo-Saxon countries, and left-wing affiliation being a predictor of
membership in those organizations in most of continental Europe but, once
again, not in the United States or the United Kingdom. Although it would
be difficult to identify specific explanations for all such differences, it seems
plausible to suggest a link between governments' policies and the configura-
tion of the coalitions that promoted the marches. This stresses the impor-
tance of the more contingent elements of the political context, in particular,
of the interaction context in which social movement actors and their oppo-
nents (here, mainly national governments) engage in a struggle that involves
both specific political resources and broader symbolic elements (Kriesi 2004).
While coalitions were very heterogeneous in all countries, the political oppor-
tunities determined by the governments' approaches to the war, and by their
variable inclinations to adopt specific patriotic discourses, contributed to shape
the configuration of those coalitions. At the very minimum, they seemed to
affect the relative weight of promoting organizations in the different coalitions.
In countries that directly promoted and actively participated in the attack
on Iraq, the strength of the call to national unity (as well as, in the United
Kingdom, the fact that the government was a "left" one) made it very difficult
for organizations other than those focusing on peace issues to engage explic-
itly in promoting the protests, even though their members were massively
involved in the marches on an individual basis.

In contrast, in countries where governments supported the intervention
but from a more cautious, and noncommittal position, the issue became
heavily politicized along traditional lines, which led to the peace protestors

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 212^_

relying on much larger alliance structures. Accordingly, the coalitions promot-
ing the demonstrations in countries such as Italy, Spain, or the Netherlands
saw a huge role for oppositional forces, including both unions and parties.
Finally, where there was no particular need to pressurize national governments
who were already hostile to the war, promoting coalitions emerged that were
more representative of the overall spectrum of organizational backgrounds
from which the demonstrators came. If anything, the organizations whose
broad issues were most directly linked with peace issues, such as those active
on global justice themes, consistendy played a more significant promoting role
in Germany, Belgium, and Switzerland. These conclusions should not be auto-
matically generalized to explanations of coalitions promoting social movement
activity on a longer and more sustained basis. They do offer, however, valu-
able insights for our understanding of mobilization dynamics in the context
of specific large-scale protest events like the February 1 5 demonstrations.

Notes

1 . One might be puzzled by the minimal role played according to our data by
environmental or women's organizations in promoting the marches in countries like the
United States or Germany, where those organizations are notoriously strong. As far

| as the United States are concerned, a possible explanation focuses on the central role

~tyy played by ad hoc peace coalitions in promoting the marches; in Germany, however, the ^S7~

strong association of radical environmentalism with the global justice movement may
have encouraged activists to define promoting organizations as such rather than as envi-
ronmentalist — hence the low presence of the latter among promoting oi

2. Here are the frequencies for the most frequently mentioned o
across all countries: Amnesty International 25, ARCI 15, ATTAC 59, CCOO 29,
CGIL 83, CND 12, GEW 64, Greenpeace 27, Groen Links 53, GSOA 22, Oxfam
18, PSOE 26, Rifondazione 30, SNOW 20, SP 76, UGT 28, Ver.di 19.

3. This finding might also be the result of the UK (and, indeed, the U.S.) tra-
dition of putting up specific umbrella organizations to coordinate specific campaigns.
As a result, the UK and U.S. respondents might have felt encouraged to identify those
peace (umbrella) organizations as promoters of the marches and to underestimate
the role of other actors.

Works Cited

Carroll, William K, and Robert S. Ratner. 1996. "Master Framing and Cross-
Movement Networking in Contemporary Social Movements." Sociological Quar-
terly 37: 601-25.

Curtis, Russell L., and Louis A. Zurcher Jr. 1973. "Stable Resources of Protest Move-
ments: The Multi-Organizational Field." Social Forces 52: 53-61.

H$H

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 21Xk)_

PROMOTING THE PROTEST 213

della Porta, Donatella. 1995. Social Movements, Political Violence, and the State. Cam
bridge: Cambridge University Press.

della Porta, Donatella, and Mario Diani. 2005. '"No to the War with No Ifs or Buts':
Protests against the War in Iraq." In Italian Politics Yearbook 2004, ed. Sergio
Fabbrini and Vincent Della Sala, 200-218. New York: Berghahn.

Diani, Mario. 1992. "The Concept of Social Movement." Sociological Review 40:
1-25.

. 1995. Green Networks: A Structural Analysis of the Italian Environmental Move-
ment. Edinburgh: Edinburgh University Press.

. 1996. "Linking Mobilization Frames and Political Opportunities: Insights from

Italian Regional Populism." American Soc'i A: 1053— 69.

. 2005. "The Structural Bases of Movement Coalitions." Paper presented at the

ASA Annual Meeting, Philadelphia, August.

Diani, Mario, and Ivano Bison. 2004. "Organization, Coalitions, and Movements."
Theory and Society 33: 28 1-309.

Doherty, Brian, Alex Plows, and Derek Wall. 2003. "The Preferred Way of Doing
Things: the British Direct Action Movement." I 'fairs 56: 669-86.

Esping-Andersen, Gosta. 1990. The Three Worlds of Welfare Capitalism. Princeton,
N.J.: Princeton University Press.
/i, Gamson, William, and David S. Meyer. 1996. "Framing Political Opportunity." In _/i,

| Opportunities, Mobilizing Structures, and Framing, ed. Doug McAdam, John D.

McCarthy, and Mayer N. Zald, 275—90. Cambridge: Cambridge University Press.

Kitschelt, Herbert. 1986. "Political Opportunity Structures and Political Protest: Anti-
Nuclear Movements in Four Democracies." British Journal of Political Science
16: 57-85.

Klandermans, Bert. 1990. "Linking the 'Old' and New': Movement Networks in the
Netherlands." In Challenging the Political Order: New Social and Political Move-
ments in Western Democracies, ed. Russell J. Dalton and Manfred Kuechler, 122-
36. Cambridge: Polity.

Knoke, David, and James R. Wood. 1981. Organized for Action: Commitment in Vol-
untary Associations. New Brunswick, N.J.: Rutgers University Press.

Koopmans, Ruud, and Paul Statham. 1999. "Ethnic and Civic Conceptions of
Nationhood and the Differential Success of the Extreme Right in Germany and
Italy." In How Movements Matter: Theoretical and Comparative Studies on the
Consequences of Social Movements, ed. Marco Giugni, Doug McAdam, and
Charles Tilly, 225-51. Minneapolis: University of Minnesota Press.

Kriesi, Hanspeter. 2004. "Political Context and Opportunity." In The Blackwell Com-
panion to Social Movements, ed. David A. Snow, Sarah A. Soule, and Hanspeter
Kriesi, 67-90. Oxford: Blackwell.

H$H

m

10 Chapter 10_Walgrave 22/01/2010 3:42 PM Page 21A^_

214 DIANI

Melucci, Alberto. 1996. Challenging Codes: Collective Action in the Inforr,

Cambridge: Cambridge University Press.
Rucht, Dieter. 1990. "Campaigns, Skirmishes, and Battles: Anti-Nuclear Movements

in the USA, France, and West Germany." Indus: rerlyA: 193-222.

Staggenborg, Suzanne. 1986. "Coalition Work in the Pro-Choice Movement: Orga-
nizational and Environmental Opportunities and Constraints." Social Problems

33:623-41.
Taylor, Verta, and Nancy Whittier. 1992. "Collective Identity in Social Movement

Communities: Lesbian Feminist Mobilization." In Frontiers in Social Movement

Theory, ed. Aldon D. Morris and Carol McClurg Mueller, 104-32. New Haven:

Yale University Press.
Tarrow, Sidney. 1998 [1994]. Power in Movement: Social Movements, Collective Action

and Politics. Cambridge: Cambridge University Press.
Tilly, Charles. 1994. "Social Movements as Historically Specific Clusters of Political

Performances." Berkeley Journal of Sociology 3%: 1-30.
. 1995. Popular Contention in Britain, 1758-1834. Cambridge, Mass.: Harvard

University Press.

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page ?1Sfc)

ll

Crossing Political Divides: Communication,
Political Identification, and Protest Organization

W. Lance Bennett, Terri f. Givens, and Christian Breunig

The February 1 5 anti-Iraq War protests mobilized demonstrators from a wide
variety of political backgrounds. For quite a few people, it was the first dem-
onstration they had ever attended. Others were associated with single issues
| or social movements, most notably peace organiSzations and related protest

W activities. Large numbers of demonstrators had been involved in global social ~w~

e protests against organizations such as the World Trade Organi
(Walgrave and Verhulst 2003, and see chapter 10 in this volume). Within
this broad context, there were many demonstrators with complex political
histories that crossed multiple issue and movement lines; these activists may
have been instrumental in establishing both the scale and the speed with
which the February 15 demonstrations occurred. Chapter 10 presented some
evidence making this point, and the idea in this chapter is that by partici-
pating in multiple political networks, these complex activists expanded the
diffusion paths by which information, coordination, and encouragement to
participate were passed through large numbers of people.

Conventional understandings of the diffusion of protest affiliations and
repertoires of action suggest that the underlying protest diffusion mecha-
nisms generally involve social relationships based on singular organizational
or political identities. Diffusion along dominant identification paths results
in mobilization that is typically bounded in scale along organization, ideol-
ogy or issue lines, often requiring brokerage mechanisms to bridge those lines
of division (Tarrow and McAdam 2005; McAdam 1988). A new wrinkle in
the idea of diffusion may be appearing in the form of individuals in various
transnational protest contexts who display identifications with multiple issue

H$H

-4

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 21&fc)_

2l6 BENNETT, GIVENS, AND BREUNIG

and movement networks, permitting diffusion at a fine-grained individual
level across a broad spectrum of issue networks. For example, della Porta
describes the movement for global justice as cohering at the individual level
via "heterogeneous, multiply faceted identities that reflect social complex-
ity" with the result that an "identity shift from single-movement identity to
multiple, tolerant identities has helped the movement in dealing with its het-
erogeneous bases" (2005, 186).

This conception of activists with complex or flexible political identities
helps resolve a conceptual dilemma concerning whether the complex array of
issues and organizations that have appeared in globalization demonstrations
(and that, in some countries, played important roles in the antiwar demon-
strations studied here; see chapter 10) can be reasonably called a movement, or
whether they are so fragmented that they might better be thought of as repre-
senting multiple movements. The concept of multiple identities suggests that
such diverse, large-scale mobilizations may be less fragmented than the multi-
plicity of issues and organizations make them appear. The formation of mul-
tiple or complex identities may also explain how individuals and organizations
within networks can readily "transpose" their objects of protest or solidarity
without losing their local issues and roots (Tarrow and McAdam 2005).

In the case of the February 15 protests, many of the diffusion paths for
protest participation seem to have emerged from the transposition of global

r network. Chapter 1 describes the role of global social justice
i the European Social Forum and the World Social Forum in the
early coordination and planning of the antiwar demonstrations. We propose
in this chapter that some part of this networking capacity is related to the
communication practices of particular types of activists. The basic idea is that
activists with complex political identifications somehow communicate more
broadly across different issue networks and are more likely than activists with
single-issue or organizational identifications to build bridges across different
issue communities. Our primary concern is to understand what communi-
cation practices enable this network-bridging to trigger mobilizations on the
scale witnessed in these demonstrations.

Identification, Communication, and Protest Organization

To build and assess a model of complex political identification, commu-
nication practices, and the scale of protest mobilization, we define and ex-
plore several sets of related variables. First, we developed three separate mea-
sures of complex political identification, as stated here and explained more
fully below: strength of affiliation with the global social justice movement,

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 2ISfe_

CROSSING POLITICAL DIVIDES XYJ

reported number of memberships in different kinds of organizations
(including churches; human rights, peace, and labor organizations; and par-
ties), and the number of diverse issues in the activist's demonstration history.

Next, to assess whether those with more complex identifications are closer
to the core diffusion paths of protest organization, we measured whether or
not each respondent was a member or personally knew a member of an orga-
nization responsible for the demonstration. Those who answered "yes" to
either question were designated as being in the "organizing circle" of the
demonstration and then assessed for the complexity of their political iden-
tifications. It is clear that organizations continue to matter in the production
of large transnational demonstrations, helping to publicize, mobilize, coor-
dinate, and otherwise support the participation of others (Fisher et al. 2004).
However, less is known about the mechanisms that ripple out from these orga-
nizations to diffuse participation at the finer-grained levels of political iden-
tification and network-bridging communication that we have measured.

Finally, we assessed how those with complex political identifications
communicated about both the demonstration in particular and politics in
general. We also assess how the communication practices of those with more
complex identifications differ from those with less complex identifications,
| such as those claiming membership in few different kinds of organizations,

~W those with less sympathy for the global social justice movement, or those with ~W

thinner histories of demonstrations on different types of issues. We hypoth-
esize that those with more complex political identities (as measured above)
manage those identities and share information across their various networks
with digital or electronic media channels such as e-mail, lists, text messages,
or interactive political Web sites operated on appliances such as computers
and mobile phones. (Hereafter, we use the term digital media to refer to these
communication modes, while using the shorthand e-media in our data tables).

Our surveys measured more conventional digital media applications
such as e-mail, lists, and Web sites that are typically accessed via computers.
However, we note that in some national and political contexts, other digital
technologies such as text messaging using hand-held devices may be more
important organizing tools (Rheingold 2002). The basic idea here is that the
flexibility, density, hyperlinked, and network-bridging capacities of digital
media enable complex activists to manage their multiple political identifica-
tions in ways that can ramp up the scale of protest activities via participa-
tion in more diverse political networks. These communication formats also
enable activists to maintain personal coherence in what might otherwise seem
a fragmented and stressful set of social and political relationships (which, to
outward appearances, may seem burdened with information overload).

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 21&k)_

2l8 BENNETT, GIVENS, AND BREUNIG

The strength of these communication and political identification pat-
terns may vary from country to country depending on the resources and
membership strengths of conventional social movement (e.g., labor, church,
and issue) organizations and the general accessibility and cost of digital media
services, from computer-based Internet to text messaging on mobile phones.
Despite these national differences, we anticipate that the general patterns
predicted here will hold for activists in most of the countries surveyed in the
February 1 5 demonstrations, because all of them are more or less similar
postindustrial democracies.

Communicating Complex Identities

The proliferation of interactive digital media technology enables the spread
of political news, mobilizing information, and action options, often packaged
within the same e-mails, lists, and Web sites. Much has been written on how
digital communication technologies — both hardware and software applica-
tions — facilitate activist network bridging in ways that go beyond the familiar
claims that the Internet mainly reduces the speed and costs of communica-
tion, or merely amplifies the capacity of existing organizations (Bimber 2003;
Flanagin, Stohl, and Bimber 2006; see also Center for Communication and
| Civic Engagement 2004). In addition to reducing costs and accelerating the

~W speed of mobilization, electronic networks may also constitute loose organi- ~W

zational structures linking actors in decentralized campaigns, providing com-
munity protest calendars and planning sites, enabling activist news reporting
and social forums, and otherwise linking diverse and often widely dispersed
activists (Bennett 2003, 2005).

The distributed (flat, nonhierarchical) design of many activist communi-
cation networks — referred to by Coopman (2003) as dissentworks — also
helps to explain their capacity to recombine and reconfigure around differ-
ent issues, producing a steady stream of large-scale coordinated protests that
have overcome impressive obstacles of geography, time, culture, resources,
and ideology. For example, the importance of nonhierarchical digital net-
working practices and technologies in bridging issue networks may help
account for the sustained transposition of organizational purpose and com-
mitment noted byTarrow and McAdam (2005) in describing the Zapatista
solidarity network. In our case, these digital networking practices may help
explain how initiatives that diffused through various world and regional social
forum networks helped organize a wave of transnational antiwar protests.
At the core of this analysis is the idea that applications of digital tech-
nologies create network structures that are attractive to activists with com-
plex or "tolerant" identities (della Porta 2005). These fluid, interactive, and

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 21£k)_

CROSSING POLITICAL DIVIDES 219

individually calibrated interpersonal communication networks may explain
a good deal about the scale and sustainability of recent transnational protest
politics (Bennett 2005; Tarrow 2003). The importance of interpersonal net-
works here is not new; it has been well established in social movement research
(Snow, Zurcher, and Ekland-Olson 1980; McAdam 1986; della Porta and
Diani 1999). However, recent research by Shah and colleagues on digital media
and activism suggests that these communication technologies and the net-
works they enable can transform both the substantive scope and the scale
of interpersonal political relationships as they move people to interact more
widely both on and off line (Shah et al. 2005; Nah, Veenstra, and Shah
2006; Hwang etal. 2006).

New communication technologies would not likely be used to bridge
diverse activist networks unless the underlying political identities of many
activists were suited to such flexible political exchanges. Recent analyses sug-
gest that political identity patterns in contemporary Western societies continue
to shift away from ideologically based identifications anchored in mass social
organizations (e.g., party, class, or church) and toward more self-directed polit-
ical affiliations driven by lifestyle values (Giddens 1991; Beck 2000; Inglehart
1997; Bennett 1998). Putnam (2000) finds in the United States that tradi-
| tional civil society relations (e.g., group memberships) continue to decline, a

~W trend that is most pronounced among younger citizens who are least engaged ~W

by conventional group membership and political participation patterns.

These broad political trends indicating more complex political identity
formation raise a series of questions about the relationship between individual-
level consciousness and social organization. How do multi-issue activists —
particularly those operating in transnational movements — manage complex
political identities on their own with fewer cues and less leadership from
conventional groups, leaders, and institutions? Of equal interest is the ques-
tion of how these relatively autonomous activists engage in sustainable col-
lective action with others. We propose that a place to start answering these
questions is to explore whether, and how, activists with more complex polit-
ical identifications differ in their communication behaviors from citizens with
simpler, or more conventional, political profiles.

Identification, Organization, and Media Use: Hypotheses and Measures

As explained earlier, we suspect that activists with multiple political identi-
fications are likely to be key diffusion links in large-scale protest networks
both because they operate close to the organizing circles of demonstrations
and because they are more likely, compared to the other activists, to acquire
and exchange information, maintain commitments and identifications, and

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22ffifc)_

BENNETT, GIVENS, AND BREUNIG

vith diverse social networks through various uses of digital
media. In particular, we predict that high levels of reliance on digital media
(our survey measured e-mail, lists, and Web sites) will be positively associated
with various measures of complex political identity, including identification
with the global social justice movement, participation in past demonstrations
for diverse issues, and affiliation with diverse organizations. Conversely, we
predict that reliance on conventional mass media channels (radio, television,
newspapers, and magazines) will be more characteristic for first-time dem-
onstrators and for more experienced activists whose political profiles are de-
fined by fewer issue or demonstration types and organizational memberships.

Measures of Complex Political Identification

Identification with the global social justice movement was measured in two
ways, which are reported in the analyses below as either sympathy or iden-
tification with the movement. The sympathy measure was a simple yes-no
response to the question "Do you sympathize with the movement against
neo-liberal globalization?" Identification strength was assessed by asking how
much the demonstrators identified with the movement along a five-point
scale: not at all, a little, somewhat, a lot, very much. We created two mea-
a binary, yes/no sympathy or support measure with values and 1 based

| sures: a binary, yes/no sympathy or support measure with values and 1 based

~W on the sympathy question, and a three-point strength of identification scale ~W

where 1 = no identification (either no on the sympathy question or not at
all on the identity question), 2 = a little or somewhat and 3 = a lot or very
much. The resulting identification patterns for the different countries are
presented in Table 11.1.

Demonstrators were also asked whether they were active, inactive, or
former members of sixteen kinds of organizations, ranging from political
parties; labor, environment, and human rights groups; to charitable organi-
zations. Our organizational diversity measure is based on the summation of
active membership in various kinds of organizations. Using this raw data, we
created a scale where all missing values are coded as no active membership
and organization diversity is divided in three categories (0, 1-3, and 4+).
The resulting organizational diversity patterns across the investigated coun-
tries are also displayed in Table 11.1.

To measure protest diversity, the questionnaire asked the respondents
who had participated in other demonstrations in the last five years to indicate
what different types of demonstrations they had participated in by checking
as many as applied from the following list: peace demonstrations, antiracism,
human rights, third world, social issues (including labor), environmental,
antiglobalization, women's rights, regionalist, and other. The level of protest

H$H

-4

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22Kfe

CROSSING POLITICAL DIVIDES 221

diversity is determined by the number of different types of demonstrations
a respondent checked. All blank, unchecked types are counted as zero (i.e.,
not attended). Utilizing these answers, we constructed a scale measure of pro-
test diversity with the following four groups: 0, 1, 2-5, and 6+. Table 11.1
reports the levels of demonstration diversity across our cases.

As indicated in Table 11.1, the overall levels of identification with the
global social justice movement were fairly high, but substantial differences
existed across the nations in the study. Sympathy and identification with the
movement appear to be especially prevalent among Italian demonstrators (.93)
but are still widespread among demonstrators in the "least sympathetic case,"
the United States (.63). For most countries, the average level of organiza-
tional diversity involves being an active member in between one and three
different kinds of organization. In Belgium, Spain, and the United King-
dom is the average person was active in less than one type of organization.
Regarding protest diversity, it appears that across all cases respondents had
participated in roughly two different types of protests. Nevertheless, variation
regarding protest diversity exists among the cases. Italian demonstrators are
the most diverse, with an average of more than three types of protests, and
those in the United Kingdom are the least diverse. On average, respondents
had engaged in a considerable number of protests in the last five years. The
cross-national average of participation in protest is between two and five.
Italians on average attended six to ten protests and thus are the most active
protestors, whereas the antiwar demonstrations in the Netherlands drew a
considerable number of people who were protesting for the first time in the
last five years (54 percent of all protestors).

Measuring Proximity to the Core Demonstration Organizations

To assess whether those with complex political identifications occupied strate-
gic positions in the protest diffusion network, we asked whether respondents
were members of or personally knew members of a group responsible for
organizing the demonstration. Answering "yes" to either question was scored
as 1, and no was scored as 0, forming a variable that we have labeled "asso-
ciation" in the "organizing circle" for the demonstrations. Scores on this mea-
sure (proportion belonging to organizing circle) varied from one national
sample to another, as follows: United Kingdom (.38), United States (.38),
Netherlands (.41), Germany (.41), Sweden (.50), Belgium (.51), Spain (.55),
Italy (.72). It is of course impossible to tell how much these differences
reflect either different national organizational structures or sampling differ-
ences, such as the existence of multiple protest sites in some national sam-
ples and single sites in others. In any event, the tests we impose on the data

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 222fe

M 1

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22Xfc)_

CROSSING POLITICAL DIVIDES

are designed to assess patterned relationships and are not likely to be affected
by these differences.

Measuring the Media and Communication Variables

The core questions of how those with relatively more and less complex polit-
ical identities manage the flow of information and political communication
required asking a number of questions about media use and communication
practices. Most generally respondents were asked a battery of questions about
how they received general political information, information about the Iraq
crisis and how they learned about the February 15 demonstration. These
items covered a range of information sources: television, newspaper, maga-
zines, radio, other people such as family or friends, Web sites, e-mail lists,
and others. For the general political information measure, respondents were
asked how often they utilized each source (on a four-point scale). These sources
are displayed separately, as they were measured in the top portions of the
core data tables in the analyses below and then combined into broader infor-
mation or media categories to simplify the analyses. The bottom portions of
the media and information source tables contain these summary measures
of mass and digital media, which were constructed by combining TV, radio,
| newspaper, and magazine into a mass media category variable and combin-

~W ing Web sites and e-mail lists into a digital or e-media variable. Friends and ~W

family as information sources are identified as social networks.

As we expected, media use patterns varied from nation to nation. For
example, there was wide variation in reliance on television for general polit-
ical information (recall that 1 = never, 2 = monthly, 3 = weekly, and 4 =
daily): United States (2.7); Sweden (3.1); Belgium (3.4); Germany, Italy, and
the/United Kingdom (3.5); Spain (3.6); Netherlands (3.7). These differences,
particularly in the case of the United States, may well reflect the quality of
available television news sources and the degree to which political issues are
presented in terms that interest activists. By contrast, newspaper use for gen-
eral political information was fairly similar, ranging from a low of 3.3 in Bel-
gium to a high of 3.7 in the Netherlands. As should be expected, e-mail use
for general political information was generally low (in the 1.5—1.8 range)
and only exceeded an average of 2 in Italy and the United States (which was
highest at 2.6, perhaps reflecting the popular trend of using automated news
services that come through e-mail). By contrast, a far larger portion of the
entire sample (.45) reported using e-mail for purposes of achieving social
change. The national differences on this measure were considerable: Germany
(.30); United Kingdom (.36); Sweden and Spain (.41); Netherlands (.42);
Italy (.45); Belgium (.50); United States (.73).

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22£fe

224 BENNETT, GIVENS, AND BREUNIG

Analyzing the Different Measures of Complex Identification

We begin by examining the basic relationships among the different n
of complex political identities. Since all variables of interest are measured on
an ordinal scale, we rely on Spearman's rho as our test statistic. Global social
justice movement identification is strongly related (.29) to protest diversity
in the cross-national sample. This correlation is also significant at the .001
level when each country is examined individually. The test statistic for this
relationship is the lowest in Italy (.20) and the highest in Switzerland (.33).
Although there is some overlap among the three measures of complex
political identification, the three variables clearly measure different types of
orientation that may operate differently. For example, the relatively weak asso-
ciation between global social justice identification and organizational diver-
sity (.06) suggests, as many observers have noted, that many activists who
support the global social justice movement are not strongly identified with
organizations. Even though the relationship between globalization movement
identification and organizational diversity is significant at the .05 level in the
Belgian, Dutch, Swiss, Spanish, and German cases, the strength of the asso-
ciations is not great. The Belgian case displays the strongest relationship, at
: .20, and no significant associations between global social justice identi-

. just .z-u, d.iu iiu aigimu^uii. djjuuduuiu ucuwccn B iuu<ti SU ci<u juaucc iuciiu-

Afr- fication and organizational diversity were found for the United States, Italy, .fa.

I or the United Kingdom. By contrast, organizational diversity is more strongly

related (.33) to protest diversity, with all the national samples registering
significance levels of .05 or greater. There was, however, considerable varia-
tion in the strength of the relationship, ranging from .53 in Belgium to .09
in Spain.

There was also an anticipated high correlation (.64) between scores on
protest diversity and protest frequency, which is measured as the respondent's
estimate of total times she or he had participated in local, national, or inter-
national demonstrations or public protests in the last five years. However, as
we expected, the correlations between protest frequency and organizational
diversity (.16) and global social justice identification (.34) are far lower than
the protest frequency-protest diversity correlation, indicating that the theo-
retical explanatory value of frequency, alone, would be confounded with the
measure of protest diversity and not with the other two measures of com-
plex political identification. We point the reader to the strong correlations
reported below between all three measures of complex political identity and
the communication variables in the later analyses, which suggest that the
interpretation is consistent with our theoretical construction, and not a sim-
ple function of protest frequency. In general, the different levels of association

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22Sfc)

CROSSING POLITICAL DIVIDES 225

among our complex political identification measures enable us to consider
media use through a rich set of activist identification measures.

The proposition that activists with more complex political identifica-
tions will be close to the organizing circles of the antiwar demonstrations
was supported overwhelmingly for all measures of complexity in all national
samples, as shown in Table 1 1.2. There were, of course, national differences,
due to both the organizational structures of national demonstrations and
properties of the distributions of scores on various measures. For example,
in the Italian case, the relatively lower associations between all three mea-
sures of complexity and organizing circle membership were caused by a com-
bination of Italy having by far the highest percentage of respondents in the
demonstration organizing circles (.65) and the highest scores on two of the
three measures of complex political identifications.

Complex Political Identification and Communication Practices

We can now assess our major prediction that activists with more complex
political identities (by all three measures) are more likely to use digital media
to manage their political communication. The following analyses examine
several different media use patterns: which media sources different types of
demonstrators use for general political information, which sources different

| demonstrators use for general political information, which sources different

~W~ political types used for information about the Iraq War in particular, and v9)~

what kinds of activists use digital media for creating social change.

General Political Information

We know that activists in recent years have developed impressive digital net-
working capabilities, and in some cases, have even succeeded in reaching
broader publics with news and political messages of their own design through
such Internet platforms as Indymedia (Pickard 2006; Peretti and Micheletti
2004; Jordan, Houser, and Foster 2003). This said, the framing of protest
activities by mass media outlets clearly remains an important concern in the
strategic calculus of many activists, and thus we do not expect electronic media
use to displace reliance on mass media for information and communication,
although as indicated in Table 1 1.3, digital media comes close to displacing
conventional media for high global social justice movement identifiers.

In general, our findings show that reliance on Web sites and e-mail lists
is significantly and positively associated with all three measures of political
complexity in all countries (with the exception of e-mail lists in the United
Kingdom, which contained a high number of missing values on these vari-
ables). Diagnostic regression analyses show that the described patterns hold
when controlling for a host of sociodemographic variables (age, sex, education,

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page ??K^

"4

§ i

£ |!g

a a a

£ I 8

H$H

-4

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22Zfe

CROSSING POLITICAL DIVIDES 227

etc.). These findings represent perhaps the most difficult test of our main
hypothesis, since they pertain to the media sources used for general political
information, rather than for information about specific issues such as the Iraq
War, or for promoting change.

In the interest of saving space, we report in Tables 11.3 and 11.4 the
relationships between media information sources and just two of the iden-
tification complexity measures, global social justice identification and orga-
nizational membership diversity, since the protest diversity findings strongly
resemble the organization diversity findings. They both clearly indicate that
in most countries digital media are more important than conventional mass
media sources for general information among demonstrators with complex
identifications. As a guide to interpreting the data, the various information
sources we measured are displayed down the left column in the top half of
Tables 1 1.3 and 1 1.4, with the results based on significance levels of the cross
tabs assessing the relationship between the complex identification scale and
the four-point media-use scale. The bottom half of each table offers sum-
mary measures of mass and digital media by combining TV, radio, newspa-
per, and magazine into a "mass media" variable and combining Web sites and
e-mail lists into an "e-media" variable. For example, daily use of an infor-
mation type (mass, digital, or social networks) is based on respondents indi-
cating daily reliance on at least one of the component information sources
that make up the type. The summary "e-media" measure detects whether
respondents indicated any use of either Web sites or e-mail as sources of
political information. E-media dominant means weekly or daily reliance on
any combination of digital media sources with only monthly or no use of
mass media sources. Similarly, mass media dominant means weekly or daily
reliance on any combination of mass media sources combined with only
monthly or no digital media use.

As Table 11.3 indicates, high global social justice identifiers not only
rely disproportionately on digital media for general political information,
but in some countries they are even averse to mass media use in their gen-
eral information-seeking behaviors. (The test statistic for both the direction
and significance of these associations is Somer's d.) By contrast, our other
two measures of complex identification reveal higher reliance on both mass
media and digital media sources, suggesting that there is some difference in
whether the basis of complex political identification is transnational or national
activism (see Table 11.4 for comparison of digital and mass media use by
those with high levels of organizational diversity). We suspect national media
simply do not feature the issues or perspectives global social justice identifi-
ers care most about, while they are aware of online information sources that

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 22Kfe

4

£lslH B

h$h

-4

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page ??afo

CROSSING POLITICAL DIVIDES 229

can satisfy those concerns. Also interesting is the finding that social networks
play less central roles in the general information networks of global social
justice activists than for demonstrators who scored high on the organiza-
tional membership (shown in Table 1 1 .4) and protest diversity measures,
suggesting, not surprisingly, that the face-to-face experiences implied in par-
ticipation in organizations and demonstrations involve enduring personal
information networks.

As Table 1 1 .4 illustrates, demonstrators with diverse organizational mem-
berships use a mix of digital and conventional media sources for their gen-
eral political information, with the notable general exception of television.
Those with complex organization and demonstration profiles tend to be mul-
timedia information seekers. This said, our main hypothesis is supported for
all three types of complex political identification: Web sites and e-mail lists
top any single conventional media source of general political information
for complex activists in most nations in our sample. Moreover, for general
political information daily e-media use is stronger among complex identifi-
ers in more national samples than daily mass media use.

Complex Political Identification and Information about the Iraq Crisis

Since there is remarkable similarity across nations in the patterns of media
and information associated with complex political identities, we can step back
and take a different look at these relationships by pooling and weighting
the country data and displaying two striking trends across the entire set of
demonstrators. This time we will look not at general political information
habits but at specific sources of information on the Iraq crisis itself. First, we
calculated the distribution of responses across the entire eight-nation sam-
ple for a question about all activists' (not just those with complex political
profiles) most important sources of information on the Iraq crisis. As com-
mon sense might lead us to think, mass media sources are far and away the
most important for the demonstrators as a whole: more than 35 percent of
the respondents say that newspapers were the most important source of infor-
mation regarding the Iraq crisis, TV follows with 27 percent, and then the
radio, with 20 percent. However, electronic media sources are most impor-
tant for those scoring highest on all of our measures of complex political
identification. Put differently, those who listed e-mail lists (2 percent) and
Web sites (7 percent) as the most important sources of information on the
Iraq crisis scored highest on all three measures of complex political identifi-
cation. (By contrast, those who found TV most important the lowest scores on
all three measures of complex political identification.) Those who cited e-mail
lists as most important scored highest on organizational and demonstration

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Pagp. 23Jjj j fc)

E

a. =?

4

I

3 a
•33 J

£zllJ>^

H$h-

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 2i^_

CROSSING POLITICAL DIVIDES

^

diversity, while Web sites were most important for high global social justice
identifiers. Since space prohibits displaying all of these relationships, Figure
11.1 illustrates the most important Iraq information sources across the range
of demonstration diversity scores found in the eight-nation sample.

Complex Political Identification and Use of the Internet for Social Change

The inclusion of multiple information, communication, and media variables
in the questionnaire enabled us to develop multiple indicators of electronic
media and mass media use. Table 11.5 shows relationships between all of our
complex identification measures (including both sympathy and strength of
identification with the global social justice movement) and a yes/no answer to
a question about whether the respondent had used the Internet in various activ-
ities aimed at promoting societal change in the past twelve months. The list of
change-oriented activities included, among other things, contacting a politi-
cian, organization, or local or national official; displaying a pin, poster, flyer, or
sticker; signing or gathering signatures for a petition, initiative, or referendum;
strike; boycott; fundraising; donation; contacting media; sit-in, occupation,
squatting; violent action; and "use the Internet for any of the above activities."

E-mail lists Websites Maga?i

Figure 11.1. Average level ofder
source of info.

diversity for each most important
the Iraq crisis

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 232^_

232 BENNETT, GIVENS, AND BREUNIG

As shown in Table 11.5, there are strong associations in all countries be-
tween high scores on all of our measures of complex identification and the use
of the Internet for some activity aimed at social change. We also note for com-
parison that those who were attending their first demonstration on February
1 5 indicated that they had not used the Internet for social change purposes
in the past twelve months. This pattern held for all nations in the sample.

Conclusion

Our data support a simple model of how the transnational coordination and
the immense scale of the February 1 5 demonstrations may have occurred.
Activists with complex political identifications may occupy critical positions
in diffusion networks that connect the organizations responsible for mobi-
lizing and supporting protests with larger numbers and types of issue and
organizational networks. This is not to deny the importance of the many
social movement groups (labor, peace, women's organizations) that surely
mobilized their constituents through conventional means. Nor do we over-
look the remarkable number of first-timers who found the demonstrations
through more conventional media channels and sought out local demonstra-
tions to express their opposition to war. What we do suggest is that to differ-
_^k ent degrees in each nation, there was an additional factor at work: activists _fo_

I who bridged organizational affiliations with their own diverse networks, which

they managed through digital media channels. These complex activists man-
age the information, and communication flows across their networks through
substantial reliance on digital media. Indeed, the activist types who are most
reliant on digital media and least reliant on conventional media are the global
social justice identifiers whose movement played an early role in planning
and spreading the word about the demonstrations. This capacity for fine-
grained network bridging may help explain how the scale of transnational
protest such as the February 1 5 demonstrations is possible to be organized
within such short periods of time and to occur nearly simultaneously across
such global distances.

The processes observed here may also help explain how transposition of
purpose occurs in activist networks, as in the relatively fluid repurposing
of social forum networks into antiwar networks, which, in turn, connected
readily with local and national peace organizations, trade union and political
party grassroots social movement networks, and large numbers of first-time
demonstrators. Such transposition may well entail brokerage between orga-
nizations, as Tarrow and McAdam (2005) observe, but it also seems to involve
network bridging at the individual level among activists who demonstrate

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 23Sfe)_

"4

1 II,

I § -i I j

f I I i i ;

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Pagp. ?34&)

234 BENNETT, GIVENS, AND BREUNIG

the capacity to identify and communicate across issues, organizations, and
geographical space.

The political communication styles observed here among activists with
these complex network identifications or affinities are important for under-
standing both the diversity of their political action repertoires and the capac-
ities of their networks. In particular, our findings suggest that digital media
applications are not just more convenient for some activists than conven-
tional communication channels or that digital media use merely reduces the
cost of communicating across diverse networks or great distances, although
these properties of digital communication are surely true. Digital commu-
nication technologies are also important because they enable activists to con-
stitute new organizational forms (e.g., large-scale, weak-tie network bridges)
that reflect — at the collective level — the principles of political consciousness
inscribed at the individual level: inclusiveness, diversity, and commitment to
radical democracy — with fewer leaders, fewer binding group memberships,
and less obligation to share specific ideological collective identity frames.
Again, this is not a generalization about all the demonstrators. It applies to
notable subsets in each nation that appear to operate with somewhat differ-
ent mechanisms for mobilization and participation and who rely on digital
/fa media to manage often surprisingly complex personal political networks. /is

| Their behavior reflects principles of inclusiveness and diversity that repre-

ingly visible meta-ideology in global protest — an ethos that
has been adopted in the charter of the World Social Forum, for example
(World Social Forum Bulletin 2004). It is also clear that the large populations
who demonstrated against the Iraq War reflected highly diverse positions,
demands, and political histories (Bennett 2005), findings that concur with
previous chapters in this volume (see in particular chapters 6, 8, and 10).
If rethinking the nature of communication among these activists (i.e.,
as organizational, not just informational) helps to explain the scale, the diffu-
sion speed, and the coherence of transnational protest, some conventional
understandings about social movements and political action may have to be
adjusted. For example, in his impressive historical analysis of social movements,
Charles Tilly (2004) has raised compelling concerns that these activist networks
may be less sustainable or effective because of their thin organizational ties
and seemingly weak activist commitment levels. As the story of transnational
protest unfolds, these concerns may eventually prove well-founded. How-
ever, there is also a possibility that such concerns about sustainability and
effectiveness reflect perspectives on social movement organization that have
been forged through observation of more conventional social protests focused

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 23S^_

CROSSING POLITICAL DIVIDES 235

within national cultural and institutional contexts. To the extent that observers
view the protests such as the antiwar demonstrations studied here (or the
globalization protests to which they were loosely linked) through the lenses
of conventional movement organization and collective identity frameworks,
something important may be missed: there may be considerable intention-
ality and strength in the design of flexible communication networks that nest
loosely within larger, dense, multi-issue networks in which individual activists
frequently shift commitments and action repertoires (Bennett 2005; Sur-
man and Reilly 2003).

A core organizational principle running through these repertoires of
transnational protest is the creation of self-organizing networks based on
personal relationships that are sustained at least partly by online information
exchanges and interactions (Jordan, Houser, and Foster 2003). The idea of
self-organizing networks in which each new technology release reflects the
political learning from the last release puts communication in a constitutive
relationship to social movement organization. To cite just one example, what
we conventionally think of as news (e.g., about the Iraq crisis) is now com-
monly produced, edited, rated, and distributed by activists themselves. The
movement slogan for this kind of information relationship was coined by
A^, Indymedia: "Be the media." /is

| While conventional mass media communication remains important, as

nportant, i

indicated in our data, new communication repertoires enable the integration
of information and action, marking a departure from the idea of a two-stage
mass communication model in which information comes, first, from distant
sources to passive individual receivers, who may then use that information
in developing action plans at a subsequent stage of the political process. It is
now common for information to become integrated with the action process
in real time, as activists report on events as they are happening and incorpor-
ate those reports in the evolving coordination of the event itself. For example,
demonstrators commonly report on their own actions via mobile phones or
digital video cameras linked to streaming technologies, enabling informa-
tion to be interpreted and fed back to participants who are able to then make
tactical decisions in real time that can shape the course of a protest action.
Rheingold (2002) has described these digitally assisted collective intelligence
and organizational capacities as smart mobs.

If it is time to begin to think differently about movement organization
and effectiveness, a place to start is with individual-level consciousness, action
repertoires, and communication preferences. The activists with complex polit-
ical identifications in this study seem able to sustain multiple commitments

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Pagp. ?3K^

236 BENNETT, GIVENS, AND BREUNIG

precisely because of the flexible, interactive communication systems that facil-
itate their creating durable (yet fluid) organizational networks. Even if access
to technologies is neither widely nor well distributed in the general popula-
tions of most societies, our look at eight nations suggests that there seems to
be enough digital communication capacity to reach enough people in enough
different places with common cues about times, locations, themes, and prac-
tices, to make a difference in how protests are organized and how activists
themselves think about their places in them. With political organization and
identification codes embedded in communication technologies, activists with
complex political identifications can maintain self-organizing networks ad-
vancing diverse causes, all with little apparent conflict of commitment.

Works Cited

Beck, Ulrich. 2000. What Is Globalization? Cambridge: Polity.

Bennett, Lance W. 1998. "The Uncivic Culture: Communication, Identity, and the
Rise of Lifestyle Politics." Political Science and Politics 31: 41-61.

. 2003. "Communicating Global Activism: Strengths and Vulnerabilities of

Networked Politics." Information, Communication, and Society 6: 143-68.

. 2005. "Social Movements beyond Borders: Understanding Two Eras of

I Transnational Activism." In Transnational Protest and Global Activism, ed.

W Donatella della Porta and Sidney Tarrow, 203-26. Lanham, Md.: Rowman & W

' Littlefield. T

Bimber, Bruce. 2003. In Democracy: Technology in the Evo-

lution of Political Power. New York: Cambridge University Press.

Center for Communication and Civic Engagement. 2004. "Democratic Applica-
tions of Internet Technology." http://depts.washington.edu/ccce/digitalMedia/
demonet.html.

Coopman, Ted. 2003. "Dissentworks: Identity and Emergent Dissent as Network
Structures." Paper presented at the Association of Internet Researchers, Octo-
ber. Toronto, Canada.

della Porta, Donatella. 2005. "Multiple Belongings, Flexible Identities, and the
Construction of Another Polities': Between the European Social Forum and
the Local Social Fora." In Transnational Protest and Global Activism, ed.
Donatella della Porta and Sidney Tarrow, 175-202. Lanham, Md.: Rowman &
Littlefield.

della Porta, Donatella, and Mario Diani. 1999. Social Movements: An Introduction.
Oxford: Blackwell.

Fisher, Dana R., Kevin Stanley, David Berman, and Gina Neff. 2005. "How Do
Organizations Matter? Mobilization and Support for Participants at Five Glob-
alization Protests." Social Problems 52: 102-21.

H$H

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 23St)_

CROSSING POLITICAL DIVIDES 237

Flanagin, Andrew J., Cynthia Stohl, and Bruce Bimber. 2006. "Modeling the Struc-
ture of Collective Action." Communication Monographs 7 '3: 29-54.

Giddens, Anthony. 1991. Modernity and Self Identity: St the Late Mod-

ern Age. Stanford, Calif: Stanford University Press.

Hwang, Hyunseo, Michael Schmierbach, Hye-Jin Paek, Homero Gil de Zuniga, and
Dhavan Shah. 2006. "Media Dissociation, Internet Use, and Anti-War Politi-
cal Participation: A Case Study of Political Dissent and Action against the War
in Iraq." Mass Communication and Society 9: 461-83.

Inglehart, Ronald. 1997. Modernization and Post-Modernization: Cultural, Economic,
and Political Change in Forty-three Societies. Princeton, N.J.: Princeton Univer-

Jordan, Ken, Jan Houser, and Steven Foster. 2003. "The Augmented Social Net-
work: Building Identity and Trust into the Next Generation Internet." White
Paper for the Planetwork Linktank. http://asn.planetwork.net/whitepaper.html/.

McAdam, Doug. 1986. "Recruitment to High Risk Activism. The Case of Freedom
Summer." American Journal of Sociology 32: 64-90.

. 1988. Freedom Summer. Chicago: University of Chicago Press.

Nah, Seungahn, Aaron S. Veenstra, and Dhavan Shah. 2006. "The Internet and
Anti-War Activism: A Case Study of Information, Expression, and Action."
| Journal of Computer i unication 12: 230-47.

W Peretti, Jonah, and Michele Micheletti. 2004. "The Nike Sweatshop Email: Political W

Consumerism, Internet, and Culture Jamming." In Pi ti id Markets:

icheletti, Andreas
Follesdal, and Dietlind Stolle, 127-44. New Brunswick, N.J.: Transaction.

Pickard, Victor W. 2006. "Assessing the Radical Democracy of Indymedia: Discur-
sive, Technical, and Institutional Constructions." Critical Studies in Media Com-
munication 23: 19-38.

Putnam, Robert D. 2000. Bowling Alone: The Collapse and Revival of American Com-
munity. New York: Simon & Schuster.

Rheingold, Howard. 2002. Smart Mobs: The Next Social Revolution. Cambridge,
Mass.: Perseus.

Shah, Dhavan, Jaeho Cho, William P. Eveland, and Nojin Kwak. 2005. "Informa-
tion and Expression in a Digital Age: Modeling Internet Effects on Civic Par-
ticipation." Communication Research 32: 531—65.

Snow, David A., Louis A. Zurcher, and Sheldon Ekland-Olson. 1980. "Social Net-
works and Social Movements: A Microstructural Approach to Different Recruit-
ment." Aiih, tl Review 45: 787—801.

Surman, Mark, and Katherine Reilly. 2003. "Appropriating the Internet for Social
Change: Towards the Strategic Use of Networked technologies by Trans-
national Civil Society Organizations." Version 1.0. New York: Social Science

H$H

m

11 Chapter 11_Walgrave 22/01/2010 3:43 PM Page 23&fc)_

238 BENNETT, GIVENS, AND BREUNIG

Research Council, http://mediaresearchhub.ssrc.org/appropriating-the-internet-
for-social-change-towards-the-strategic-use-of-networked-technologies-by-
transnational-civil-society-organizations/attachment.

Tarrow, Sidney. 2003. "The New Transnational Contention: Social Movements and
Institutions in Complex Internationalism." Revision of a paper presented at the
2002 Annual Meeting of the American Political Science Association, Boston.

Tarrow, Sidney, and DougMcAdam. 2005. "Scale Shift in Transnational Contention."
In Transnational .1 Porta, and Sid-

ney Tarrow, 121—50. Lanham, Md.: Rowman & Littlefield.

Tilly, Charles. 2004. Social Movements, 1768-2004. Boulder, Colo.: Paradigm.

Walgrave, Stefaan, and Joris Verhulst. 2003. "Worldwide Anti-War-in-Iraq Protest:
A Preliminary Test of the Transnational Movement Thesis." Paper presented at
the European Consortium for Political Research, Marburg, Germany, Septem-
ber 17.

World Social Forum Bulletin. 2004. March 19. http://www.modkraft.dk/phorum/
read.php?f=7&i=7222&t=7222.

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page ?3ijifc )

12
The Framing of Opposition to the War on Iraq

Dieter Rucht and Joris Verhulst

Wording matters when it comes to influencing people's hearts and minds.
Many words and catchphrases are value-loaded. Because they have or evoke
positive, negative, or ambivalent connotations and feelings, they are often care-
| fully chosen by actors in a political struggle, thus becoming part in a contest

W over naming, blaming, and framing (Gamson 1992). It seems that the more v9)~

morally loaded a conflict is and the higher the stakes, the more the actors
engage in a framing contest to win support and discredit their opponents.
Just consider the discursive struggles over abortion, in which each side delib-
erately chooses terms and slogans: "unborn baby" or "fetus," "pro-abortion"
or "pro-choice," "anti-abortion" or "pro-life," "abortion is murder" or "my
belly belongs to me," and so forth (Ferree et al. 2002). When it comes to the
protest against an imminent war in Iraq, we can also expect that the actors
on both sides place much emphasis on the words and slogans they want to
communicate to both their constituency and the audience at large. In part,
they have control over the phrasing of their messages. This applies, for exam-
ple, to the protesters' banners and speeches during a rally. Another — and
politically probably more relevant — part of the communication, however, is
beyond the direct control of authors and speakers: their messages, if covered
at all, are mediated by the mass media (Bennett and Entman 2001). As has
been demonstrated repeatedly, the media are highly selective in what and
how they report (Hilgartner and Bosk 1988), whom and what they cite lit-
erally or indirectly, whether they explicitly express their own views in com-
mentaries and side remarks. In short: naming, blaming, and framing are
crucial in the case we are analyzing. Accordingly, this chapter concentrates on

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page ?4(jfe

24O RUCHT AND VERHULST

the communicative dimension of the demonstrations against the ii
war on Iraq. To this aim, we are using three kinds of materials: First, since
we are interested in how the "ordinary" participants in the demonstrations
expressed their views and motives, we analyze mainly quantitatively their
responses to the open-ended survey questions. Second, we investigate qual-
itatively and illustratively how organizers, representatives of movement groups,
and speakers expressed their views and claims during the demonstrations.
Third, in addition to the analysis provided in chapter 3 dealing with mass
media reports and the governmental positions, we look more specifically at
the media coverage of the expected war on Iraq. In methodological terms,
depending on the material and the specific question in which are interested,
we use both a more conventional, quantitative content analysis and a more
qualitatively oriented frame analysis, as has been developed in social move-
ment studies in the 1980s and 1990s (Snow et al. 1986; Snow and Benford
1988; Gamson and Modigliani 1989; Gerhards and Rucht 1992).

The Voices of "Ordinary" Protest Participants

In a public protest, mere participation is a message in itself. Therefore, not
all participants feel the need to explicitly voice their views during the event.
Moreover, there are forms of protest — for example a silent march or a silent
prayer — where words are deliberately avoided. However, in the mass dem-
onstrations against the war in Iraq, explicit messages in various forms were
abundant. Many people were carrying handmade signs, standardized placards,
or large banners. And they were shouting slogans or singing peace songs,
giving interviews to journalists, and discussing the issue of war in small
groups during the marches and rallies. Whereas a look at the many signs and
banners unveils in an extremely condensed form what the most committed
people in the crowd wanted to say (e.g., "Bush — Terrorist No. 1"; "No blood
for oil,"), analyzing the International Peace Protest Survey questionnaire, as
done in several chapters in this volume, can give a more systematic picture
of the participants' views. While the questionnaire mainly included lists of
preformulated, that is, "closed" answers, we also asked a key question that
had no standardized answer options: "Why did you participate in the dem-
onstrations on February 15, 2003? (Write down your answer in the box
below)." The box provided space for about seven lines. Some demonstrators
responded by writing down only one single word; most gave one or two
short sentences; a clear minority completely filled the space provided. The
answers to this open-ended question are, in our view, the most comprehen-
sive and authentic data on the motives of the antiwar protesters on this par-
ticular day.

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page ?4fcfc)

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 24I

While it was already amazing to see that relatively few protesters refused
to accept the questionnaire (see appendix A), we were even more surprised
to learn how many of those who had completed their questionnaires had also
answered to the open-ended question. After all, this usually takes more time
and reflection than to tick a box and is therefore often omitted by respon-
dents in other research contexts. In our case, however, out of the 5,772 ques-
tionnaires that we received, a total of 78 percent included answers to this
particular question. The response rates ranged from 44 percent in London
to 97 percent each in Spain and the United States. This readiness to speak
out confirms our general impression based on many conversations held while
distributing the questionnaire: people were strongly motivated to express their

Exemplary Voices

We will first look at a few examples of how demonstrators have expressed
their motives. Quite a number of protesters categorically rejected war ("War
is no means to solve problems"; "Because I am convinced that war is always
the wrong instrument to solve conflicts, regardless when and where"; "I am
a pacifist"; "violence only breeds counterviolence").

Others rather rejected this particular war under the given circumstances
("I am opposing this war because it serves the hegemonic interest of the
USA"; "This war is for the control of oil"; "I don't want my country, the
U.S., to become the bully of the world"; "This war is absurd"; "Think this
is about global domination"; "Because war is the last resort, not a first resort.
After four years without inspections the inspectors have only been at work
for 11 weeks"; "Because a war of pre-emption is immoral and illegal under
international law and I'm worried we're setting a terrible precedent, not only
for us, but the world"; "I feel a diplomatic solution is appropriate. I am not
a pacifist"; "Because I think U.S. pre-emptive war is morally wrong, illegal,
unconstitutional and un-American. Our government is betraying our ideals
in a much more flagrant way than ever before").

A minority of respondents framed their opposition to war as part of a
broader political struggle. For instance, a London demonstrator stated the
"need for [a] vibrant and imperialist movement in the metropolis of capital."
Others defined the war as an act of "neo-colonialism," "imperialist politics,"
and the like. Still others expressed their solidarity with the Iraqi people or
more generally with those who suffer from war and exploitation ("To pre-
vent the killing of innocent Iraqi civilians"; "To save the life of Iraqis").

A substantial number of protesters made clear that they equally opposed
the United States government and the government of Saddam Hussein ("No

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 24^_

242 RUCHT AND VERHULST

doubt that Saddam & his sons are worse than animals but killing ii
people will not remove them"). Most European protesters who criticized in
particular the U.S. position emphasized that they were not against "Amer-
ica," the "USA" or "the American people" in general, but against Bush ("Bush
is an asshole"), the U.S. government, the U.S. elites. Contrary to the views
of many commentaries to be found mainly in conservative newspapers, very
few respondents expressed views that could be interpreted as outright anti-
American (see also Table 12.1), such as "USA is shit," "Because I am a
confessing opponent of America," "Hatred toward America and war"). Dem-
onstrators were also referring to the particular situation in their home coun-
try ("I want to support Chancellor Schroder and Foreign Minister Fischer
in their clear 'no' to this war"; "To protest the UK being the USA's lapdog").
A small minority commented on their personal situations to explain
their opposition to war: "Because my nine-year old son asked me whether
there will be war and what we could do against it"; "I have experienced World
War Two and know about the consequences"; "Because I am a mother. Be-
cause I am a teacher . . . Because I am a human being . . ."; "I was represent-
ing my six small grandchildren"; "To show her [my daughter] an example of
how to be an engaged citizen. To show her the beauty of being active in a
I community effort"; "I fear the draft").

~W Often along with more specific reasons, many demonstrators stated that ~W

they felt a moral obligation to express their dissent, that they didn't want to
be counted among those who were indifferent, or that participation in the
rally was the only means they had ("I felt a need to express my dissent. It's
my right"; "I feel I have a civic duty to express my hope for a peaceful reso-
lution"; "A need to 'stand up and be counted' in the effort to deflect our
administration's policy versus Iraq"; "I felt I had to swell the numbers in the
hope that PM [Prime Minister] would take notice of public opinion"; "Be-
cause it's the only action, within my reach, to try and stop this war"; "I dem-
onstrated in hopes that this would send a message to other countries via news
media"). These and many similar statements underscore that people not only
protested for instrumental reasons but also wished to express their feelings
of anger and frustration — and hope.

Indeed, in some responses hope was articulated that protest could even-
tually be successful ("We made a difference when we marched against war
in Vietnam — we can do it again"; "Because I believe that the pressure of mil-
lions of people in the streets can have an impact"; I felt it was time to put
my feet where my mouth is. I feel strongly and truly believe one can make
a difference"). Yet a few respondents expressed doubts over whether protest
could prevent war but still felt a need at least to try it, or, as one of them

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 24Xfc)_

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 243

noted, "to be counted as a voice + body against the war and the policies of
Bush and his war-mongers." Another demonstrator wrote: "To forget my des-
perateness because of our powerlessness and to make clear: This war, which
cannot be prevented, shall not be fought in my name."

In addition, a marginal proportion of respondents said they were pri-
marily or only present not to protest against war but for more mundane rea-
sons: "to take pictures," "to spend time with friends"; "because my boyfriend
organized a trip from Boston to take part in the rally," or on professional
grounds ("Freelance journalist, interested in social movements"; "To docu-
ment this event, I am a photographer").

While these selected statements give us a taste of the range and kind of
individual reasoning behind participation in protest, they do not provide a
systematic picture of the relative weight of different reasons and how these
are related to certain characteristics of the protesters.

Quantitative Results

To process the total of 4,5 14 non-standardized answers, we categorized them
according to our research interests. Because of different language capacities,
we conducted the categorization at three different places. The coding was
done the basis of general explanations but without a joint training of the
coders, let alone a test of intercoder reliability. Nevertheless, we believe that
the data are acceptable, as long as we refer to the big picture and ignore both
small numbers and small differences across categories.

In a first step, we inductively developed a classification of the various rea-
sons to take part in the protest, ending up with seventeen categories. The
results are displayed in Table 12.1. Because respondents often provided more
than one reason to protest, the percentages add up to more than one hundred.

As it can be seen in the far right column of Table 12.1, in the aggregate
of all countries, pacifist beliefs were the most frequent reason to protest (24
percent), followed by general criticisms against this war (22.1 percent), and
a number of more specific reasons. Almost 14 percent mentioned the desire
to express their views in front of the public. Anti-Americanism was entirely
marginal, with an average of 0.3 percent across all countries. Substantial cross-
national differences exist regarding attitudes toward the protesters' own coun-
tries' governments. Almost 10 percent of the Belgian demonstrators and almost
8 percent of those in Germany mentioned support of their government as a
reason to take to the streets. 1 Yet substantial cross-national differences in the
motives of protest also become apparent with regard to some other categories.
Pacifism, for example, is rarely mentioned in the United States and the United
Kingdom, with only 8.2 and 8.5 percent, respectively, but frequently in Italy,

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 24Afc)_

n e "* E |"§ .6 8

B it'll 11 I l^Hl SUtU el I

£u<^ £u^u < <; Q 3 £ $> & 3 < O

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page ?4Sb )

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ

^

with around 50 percent. Considerable differences also become apparent re-
garding the category "War breaks international law," brought up frequently
in Spain (26.6 percent) but rarely in Italy (1.7 percent) and Switzerland (2
percent). In Italy and Belgium, opposition to war was most often linked with
a more general struggle against the capitalist economic order (6.7 percent and
6.2 percent), while in the United States, this was not the case (1.5 percent).

In a second step, we recoded applicable categories listed in Table 12.1
according to one dimension that we found to be of particular interest. We
call this "type of reasoning," based on a "moral-affective" versus "cognitive"
dichotomy 2 While some items could not be classified according to this cri-
terion, there were also the instances that the same respondents gave two or
more reasons, of which one was "moral-affective" and another "cognitive."

Considering this "type of reasoning" dimension, we find that "cognitive"
reasons to protest war have a significantly greater weight (33.5 percent) when
compared to "moral-affective" reasons (16.0 percent; see Table 12.2). The
gap is particularly big in the United States, with a difference of about 37 per-
cent. Only among the Italian demonstrators did moral-affective reasons out-
number the cognitive ones.

Moreover, we can also study the correlation between the type of reason-
ing and several other characteristics of the protesters (figures not displayed).
The relative weight of cognitive reasons increases with the level of education
and age. Also, men mentioned cognitive reasons to a greater proportion than
women did. But, contrary to common expectations, religion seems to be un-
related to the kind of reasoning. Also, there appears to be no significant cor-
relation between the kind of reasoning on the one hand, and the position in
the working sector, and the satisfaction with democracy on the other hand.
Fewer protesters who claimed having taken part (only or also) in disruptive

Table 1 2.2. Type of reasoning, by country (percent)

I

United States

8.5

45.4

United Kingdor

n / London

4.1

26.5

United Kingdor

n / Glasgow

4.4

30.2

Spain

22.1

45.-1

Italy

20."

9.0

Netherlands

26.4

34.3

l-i.-i

33.9

Belgium

21.2

42.0

Germany

22.^

35.1

67.9
28.2

4-.4
25.3

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page ?4K^

246 RUCHT AND VERHULST

actions in the past, provided cognitive reasons when compared to those pro-
testers who had participated only in moderate actions.

The Voices of Protest Organizers and Speakers in Public

In a second step, we shed some light on how the groups that organized the
protest framed the antiwar cause in leaflets, calls for action, brochures, and
speeches of group representatives and other individuals during the demon-
stration. While the above answers to the open-ended question had to be
brief due to limited space, the written texts and the speeches were, of course,
much longer and more elaborate.

The constituencies of groups and movements participating in the anti-
war protest of February 15 were exposed to a flood of written material before
and during the event. Among this material were various texts distributed
within protest groups such as the "Anti-War Call" launched by the European
Social Forum, and the Platform against War on Iraq issued at the preparatory
meeting in Copenhagen (see chapter 1), public advertisements such as the
one initiated by the U.S. -based Artists United to Win without War, the Busi-
ness Leaders for Sensible Priorities, the Internet-based network MoveOn.org,
and the speeches held where ever masses took the streets. Because we only
I have selective material from these various sources and cannot assume repre-

~W sentativeness, we aim neither at a quantitative analysis nor at a comparison ~W

across countries. Rather, we present some examples we find interesting and
noteworthy (see also Cortright 2004).

Many written signs of protest had straightforward messages. The shortest
could be seen on a banner presented at London's Trafalgar Square: "No!" Fly-
ers and calls for action often were headlined by "No war on Iraq," "Europe
against the war," and the like. Such brief messages were occasionally accom-
panied by expressions of urgency as directly stated, for example, in an
"urgent world-wide call to action from citizen to citizen" distributed by
the U.S. -based group United for Peace. This sense of importance comes from
the perception of being "on the verge of a major disaster" that can only be
prevented by "the world public." Another example was an advertisement en-
titled "Prime Minister Blair, It Is Two Minutes before Midnight," published
in five UK newspapers: the Guardian, the Independent, the london Times,
the Telegraph, and the Financial Times. Combined with this dramatic ges-
ture, activists often assured that protest would matter, as indicated in both
the first and the final sentence of the "And- War Call" by the European Social
Forum: "Together we can stop this war" and "We can stop this war."

A further characteristic of such written documents was signaling to the
public that this war is "Not in our name," a message that could be made

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page ?4^

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ XAfJ

specific by adding a particular addressee, such as "Not in our name, Mr. Blair"
(Scottish Campaign for Nuclear Disarmament) or sender "Not in Iraqis'
names: Exiles thank the worldwide peace movement " (petition).

Besides slogans and catchy headlines, many texts stressed specific reasons
for opposing the war in Iraq. The three-column advertisements sponsored
by the New York-based Business Leaders for Sensible Priorities presented four
arguments in bold letters: "Warning — War Will Wreck Economy"; "Warn-
ing — War Will Breed Terrorism"; "Warning — War Will Discredit America";
"Warning — War Will Take a Terrible Toll on Human Life."

While many texts exclusively focused on the war and its immediate con-
sequences, others related the opposition to war with other kinds of struggles.
This is illustrated by a flyer of the ANSWER (Act Now to Stop War and
End Racism) Coalition for the United States. The group announced a Week
of Anti-war Resistance, starting on February 13, with "youth and student
action, teach-ins and forums" on the twelfth anniversary of the Gulf War
bombing of a shelter in Iraq and ending with a "Students and Youth Action
on the anniversary of the assassination of Malcolm X, a Coordinated Day of
Resistance — including student walk-outs from hundreds of high schools and
colleges — and other acts of non-compliance" on February 2 1 .

Quite a number of flyers and other texts emphasized the massiveness
and worldwide character of the resistance to war. This is stressed in both the
European Social Forum's "Anti-War Call" — "There is a massive opposition
to war in every country in Europe. Hundreds of thousands have already
mobilised for peace" — and the December 15, 2002, Copenhagen declaration:
"It is clear that there is majority to war in almost every country in Europe
and across the world. This is why this war cannot be fought in our name."
In the advent of the February 15 protest, organizers further encouraged poten-
tial participants by confronting them with lists of countries or cities in which
protest activities had been scheduled. For example, at the bottom of a call
for action distributed by the UK Stop the War Coalition, some eighty cities
are listed, starting with Adelaide and ending with "Wollongong and many
more (total over 600)." Thus, contrary to Mancur Olson's theory that indi-
viduals would be reluctant to participate in mass collective action when their
own contributions would not make a difference (Olson 1965), the organiz-
ers of the February 15 protests were right in assuming the opposite: the big-
ger the forthcoming event seemed to become, the more it would attract
additional people who otherwise would be reluctant to participate — as the-
oretically expected in the threshold model proposed by Granovetter (1978).
In other words: beyond the substantive arguments to oppose the war, the mere
size of the expected protests was also considered a motivator for participation

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 24&fc)_

248 RUCHT AND VERHULST

in protest, as expressed in such answers to the open-ended questio
"I want to be counted" or "To be another body in a mass of bodies." In a sim-
ilar vein, organizers not only pointed to the (expected) mass of people but
also to the broad variety of different groups and social strata opposed to war,
ranging from U.S. veterans of the Vietnam War to the UK "Dog Owners
against War Stunt" (as mentioned in a flyer of the Stop the War Coalition)
to the Marxist-Leninist Bolshewik Partizan, a group of Curds aiming at build-
ing a Bolshevist Party in Germany.

Furthermore, the various speeches and performances of the February 1 5
demonstrations were deliberately chosen to exhibit the variety of protest groups
and reasons to protest. Consider, for example, the program of the Berlin
event, which consisted of two gatherings earlier in the day (in old East and
West Berlin) from which marches led to the main site of the rally. The whole
event was coordinated by the Netzwerk Friedenskooperative (Network of
the German Peace Movement), which was part of a much broader Action
Alliance February 15. Besides a variety of music (e.g., drums, Soul and Gospel
Choir Berlin, well-known singers), poems, theater, and recitation, the pro-
gram included speeches by representatives of the East German citizen move-
ments that had toppled the communist regime and from members of trade
I unions and, among others, peace, Christian, immigrant, youth, and medical

~W groups. Their speeches, whose transcripts were distributed on the Internet ~W

in the aftermath of the event, offered a broad range of arguments against the
war: moral, social, political, juridical, social, and environmental. Also, they
indicated hidden agendas behind the official declarations of politicians ready
to engage in warfare; pointed to disastrous consequences; appealed to the
demonstrators and the worldwide public to take action; linked this struggle
to struggles in other periods, places, and policy areas; emphasized the unique
character of this protest as part of a worldwide event; supported — in the Ger-
man case — the position of their own government; expressed hope that war
could be prevented if millions of people said "no"; and all shared the belief
that war in principle, or at least this war, under the given circumstances, is
immoral and wrong. The declaration from the Netzwerk Friedenskoopera-
tive read by one of its speakers at the very end of the main rally concludes
by stating: "Peace is not everything, but without peace there is nothing!"

On the eve of the antiwar demonstration in Berlin, several more-or-
less prominent critics presenting themselves as the Coalition against Anti-
Semitism, in an open letter to the media, had accused the antiwar groups of
being "anti-American" and "politically naive." Probably in reaction to this,
several speakers at the antiwar demonstration expressed their sympathy with
"peace-loving Americans" while at the same time opposed not only the Bush

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 24&fc)_

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 249

administration but also the repressive regime of Saddam Hussein and his Bath
Party. The critique of being "anti-American" is also rejected in a written state-
ment entitled "The Peace Movement Is Anything but Anti-American." The
authors maintain their opposition to the war plans of the Bush government
while defining their protest even as pro-American insofar as they endorse "one
of the greatest achievements of the United States: The rule of law" (Bunde-
sausschuss Friedensratschlag, 2002).

Reviewing numerous written and oral statements by the antiwar pro-
testers, it is obvious that the lowest common denominator is that this war is
morally and politically wrong. However, the various motives of and reasons
for this opposition vary considerably, so that we cannot identify a single over-
arching ideological master frame (for this concept, see Snow and Benford
1992) shared by all participants. Whereas outspoken leftist participants tended
to interpret the war as an expression of imperialism, other groups perceived
it more as the consequence of misperception of a number of factors, such as
Iraq's weapons arsenal, its contribution to international terrorism, and the
chances of establishing a peaceful and democratic order after the war. Clearly,
the imminent war offered platforms to a broad variety of issue-specific groups
for relating their cause to the opposition to war, thereby providing many
/fa examples of frame-bridging and frame-extension, as discussed in the litera- /is

| ture (Snow at al. 1986). Also, we found clear examples of the three compo-

nents of collective action frames that Gamson (1992, 7) has identified, namely
injustice, agency and identity.

For the most part, however, mass protest is not an end in itself. As ex-
pressed by many participants, it is meant to send a strong signal to a large, pos-
sibly worldwide, audience that, ultimately, may have an impact on decision-
makers. To this aim, it is necessary to attract media attention and, if all goes
well, positive media coverage (Molotch 1979; Walgrave and Manssens 2000;
Verhulst and Walgrave 2006).

Media Coverage of the Iraq Crisis and the February 15 Demonstrations

Most of our knowledge of current affairs in the world stems from television,
radio and/or newspapers. This news is not an unbiased reflection of reality
but is selected according to news values and news routines, framing pref-
erences and political positions. Studies have documented the U.S. media's
agenda-setting, priming, and framing impact on mass opinion during the
1991 Gulf War (Iyengar and Simon 1994). With regard to the 2003 Iraq
War, Kull, Ramsay, and Lewis (2004) showed that a large segment of the U.S.
public, during and after the war, held misperceptions. These people thought

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25ffi^_

250 RUCHT AND VERHULST

weapons of mass destruction had actually been found in Iraq, that there was
indisputable proof of the link between Saddam Hussein and Osama bin Laden,
and that the majority of world public opinion supported the U.S intervention.
Kull and his colleagues found these misperceptions to be strongly associated
with media use. For example, Fox News viewers held more misperceptions
than CNN viewers, regardless of their educational or professional back-
grounds. Clearly, news sources do matter. And, as seen in chapter 3, they
also matter at the national level. In different countries, the Iraq crisis was
framed differently, which was likely to have affect mobilization potential and,
ultimately, the kind of people that actually took the streets. This brings us
to social movements. For them, mass media coverage can be a matter of life
and death (Walgrave and Manssens 2000; Verhulst and Walgrave 2005). Media
can validate and legitimize social movements' existence (Gamson and Wolfs-
feld 1993; Smith 1999). Mass media link social movements to their mobi-
lization potential and can help movements recruit new adherents and attract
additional financial resources (Barker-Plummer 2002). Positive coverage of
movements can enhance their impact on political decision-making. But the
relationship between social movements and mass media is fundamentally
asymmetrical, with the former much more dependent on the latter (Gam-
L son and Wolfsfeld 1993; Verhulst and Walgrave 2005; Rucht 2005). Move- ^L

I ments have a tough job in competing with other societal actors and events

to gain media attention (selection bias). And once a message or event is in
the media, the question remains of how the media report on it (description
bias) (Smith et al. 2001; Verhulst and Walgrave 2005; McLeod 1995; Cooper
2002). This is what we want to find out in the remainder of this chapter: Did
the antiwar movements manage to get some of their arguments and frames
on the Iraq conflict into the press? How were the February 1 5 protests cov-
ered, and how did the nationwide newspapers portray the demonstrators?
Chapter 3 explored the relationships among national governments' posi-
tions on the Iraq conflict, public opinion on the eventuality of war, and how
the media reported on the Iraq crisis. In summary, in those countries where
the governments were trying to engage in war, both national media and
public opinion were basically supportive of the government's stance. Where
the national governments strongly opposed the war, the media and public
opinion tended to share this position. Finally, in countries where the gov-
ernment's position on the eventuality of war was limited to lukewarm and/
or merely oral support, the newspapers and (eventually) public opinion tended
to not support the government's stance. In all countries, national press cov-
erage and public opinion on the eventuality of war converged.

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25|&)_

"4

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ

Framing of the Iraq Conflict

As we did in Chapter 3, we will use original media data. In each of the eight
countries, we analyzed three newspapers: the major left-leaning broadsheet,
the major right-leaning broadsheet, and the most popular national (or local)
newspaper. All papers were scrutinized for articles on the Iraq conflict, using
the neutral search term "Iraq," between January 21, 2003, that is three weeks
preceding the February 15 protests, and March 21, 2003, the day after the
invasion in Iraq began. We aimed at obtaining three articles per day and per
newspaper. In sum, 3,968 articles were selected and coded. Additional method-
ological information can be found in appendix A.

An initial question is whether the antiwar sentiments of the February
1 5 demonstrators were reflected in the different national newspapers and to
which degree these papers report on reasons why a war should not (yet) be
waged. Out of all articles, 1,521 (38 percent) mentioned at least one reason
not to go to war. Similar to the variety of responses of the protesters to the
question why they took part in the demonstration, the reasons not go to war
mentioned in the newspapers were quite diverse (see Figure 12.1). The dis-
regarding of the UN procedures, and the work and views of its weapon
inspectors, constituted the lion's share of the reasons mentioned why war
should not be waged (yet). When all UN-related reasons are cumulated, the

48.6

UN inspectors doing a good job; no evidence of weapons of mass destructioi
Lack of UN war mandate; damage to United Nations; illegality of preemptrv

Pacifism: no war as a political tool

Economic interest or oil

War will fuel (religious) terrorism or be perceived as West-Islam aggression

Destabilization of Middle East

Lack of link between Iraq and al-Qaeda

Double standards (invasion only of Iraq, i

Iraqi regime no threat to world peace

Cost of war to tax payers or to country's

ntry)

War fought for religious re

No hope for democratization

Lack of after-war strategy

War as dangerous precedent

Bush administration or family wanting to finish job of Gulf War

(Future) casualties among own troops

War as distraction from bad domestic policy

N

I

Figure 12.1. Reasons for not going u,

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 252fe

252 RUCHT AND VERHULST

second most important reason is pacifism, in the sense that war can never be
accepted as a political tool. Remarkably, this reason is more often mentioned
than, e.g., the economic interests in war (control over oil) or the fact that
war will raise the chance of religiously based terrorism. This is consistent with
the motivations made explicit by the February 1 5 demonstrators, who men-
tioned cognitive reasons more frequently than moral- affective reasons, includ-
ing pure pacifism.

Regarding the newspaper reports, pacifism is the most frequently men-
tioned reason not to go to war (as it was for the February 15 protesters). In
cross-national comparison, pacifism was least mentioned in the United States
and the United Kingdom (as was also the pattern for the protesters from
those countries) and most frequently stated in Spain. In general, the rank
order for pacifism across countries is the same for newspapers and protest-
ers (see Figurel2.2).

When looking from a cross-national perspective at the weight of vari-
ous topics and issues as reported in the newspapers under study here, the
first striking finding is the differential reference to the domestic government
in the context of the Iraq conflict (figures are not displayed). By far the most
references to the domestic government were found in the U.S. press, followed
by the UK and German press. This is understandable for the war-leading
countries for which the government plays a crucial role. However, we have
no explanation for why Germany and Belgium as war-opposing countries
differ so widely in terms of references to the own government. Divisions
within the press's own government were most salient in the United King-
dom, and, accordingly, were reported in almost every fifth article. German
newspapers mentioned a divided Europe most often; they also covered the

Percent of articles

United States 6.5

United Kingdom 7.2

Spain 27.2

Italy 16.0

Netherlands 10.0

Switzerland
Germany

Figure 12.2. Mentions of pacifism ("war is no political
tool") in newspapers per country

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25Xfc)_

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 253

divided opinions within the United Nations (as did U.S. papers). In Switzer-
land and the United States, overall opposition to war was mentioned least
frequently. In sum, the German press was most frequently referring to various
kinds of actors opposing the war, while the papers in Spain and the United
Kingdom reported most frequently on opposition among the populace. The
costs of war were most often referred to in the United States and in the United
Kingdom, the two countries that were the most determined to engage in
warfare. The idea that war cannot be prevented, however, was not as wide-
spread in the news outlets of these two countries, as it was in the Italian press.
The opposite idea, that war could still be prevented, was most frequently
found in the Spanish and German presses. In the United States, which can
be considered the most crucial player, the press highlights the domestic gov-
ernment's role and aspects of war preparation and war costs but remained
relatively silent on dissident voices.

Coverage and Framing of the February 15 Protests

In the two-month period around February 15, protests were mentioned in
270 articles (about 7 percent of all articles), with the Belgian and German
press far above average (16.5 and 13 percent, respectively) and all other coun-
| tries at about 4 percent. The actual protests mentioned in the articles add

W up to 414 different events, of which most were referred to in Germany and v9)~

Belgium, followed by the United States and the UK. Hence, protest is of
greatest interest to the press in the countries that range at the far ends of the
pro-antiwar scale. Nonetheless, the way protests were covered might have been
different, more critical on the one end and more supportive on the other.

Because we are especially interested in how the February 15 protests
were covered, we will only focus on protests mentioned between February
14 and 22 and only those in the Western countries or that were reported as
"worldwide." We counted 130 protests mentioned in the selected period,
the most in the United Kingdom (36) and the fewest in Switzerland (2). 3 In
the protests or public opinion on war were referred to in the titles of most
of the articles, especially those in the United States, and in many an antiwar
voice was featured: "French See Bush as the Ugly American" ( USA Today,
February 14); "From New York to Melbourne, Cries for Peace" {New York
Times, Februaryl6); "US Builds War Coalition with Favors — and Money"
( USA Today, February 15), and "Blair Under Siege Over Stance on Iraq. Oppo-
sition Up, Poll Numbers Down" {Washington Post, February 19).

In the UK press, the titles were more upfront, with some of them point-
ing to the dwindling popularity of the country's prime minister: "Blair's
Popularity Plummets" {Guardian, February 18); "Blair Hasn't Even Won the

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25fcfe

254 RUCHT AND VERHULST

Establishment Round to War" {Times, February 20), or "This Go^
no Longer Speaks for Me. Voices of Protest" {Guardian, February 17). In
other titles, the prime minister is mentioned as holding the line in spite of
the loud antiwar voices: "Blair to Defy Anti-War protests: Million-Strong
Demonstrations Will not Deflect Iraq Policy as Ministers Rally around Prime
Minister" {Guardian, February 17); "The Peace Marches: Blair's Warning:
'Weakness Will Be Paid in Blood.' Impassioned Prime Minister States the
'Moral Case' for Ousting Saddam" {Guardian, February 16). And finally,
some titles directly referred to the expected mass protests. For example, two
articles in the Times predicted: "Glasgow March May Be Biggest since World
War 2" and "Protests Will Reach Every Continent" (both February 14). Sim-
ilar titles can be found in other papers: "The Peace Marchers: One Million"
and "People Power Takes to the World's Streets" (both in Guardian, Febru-
ary 16). The Sun, the tabloid that openly backed the government's position
on war (e.g., "Blair Can See Big Picture" [February 28]; "Selfish Chirac
Shows France's True Colors" [March 03]) does not mention any kind of
protest in articles we found for this period, providing a stunning example of
a selection bias.

For Spain, we only have El Mundo reporting on the protests in this
period — for example, "The Old Europe Does not Want a War" (February 16).
La Stampa, in Italy, provides a similar picture: "Five Kilometers of Protest"
(February 16). These papers exhibit a positive attitude toward the antiwar
voices. As said, in Switzerland, only two protests, both in the Tages-Anzeiger,
are mentioned in the articles we selected for the period from February 14 to
22. In Belgium, protests mentioned in this period were always reported in a
neutral way, for example "Human Chain against War in Iraq" {Het Laatste
Nieutvs, February 19). In Germany, except for "Scholar Does not Believe in
Continuation of Protests" {Frankfurter Rundschau, February 18) and a neg-
ative report, "Who Demonstrated against Saddam" {Bild, February 14), all
relevant article titles are neutral or supportive of the protest.

Moving beyond the titles, we looked at how the protests were portrayed
in the text body. The total of 130 articles in the nine-day period does not
really allow for a meaningful quantitative analysis. Yet some quotes and
references provide us with an idea of how the February 15 protests and pro-
testers were presented in the press. Demonstrators' characteristics were re-
ferred to in one out of five articles in this period. With one or two exceptions,
all stressed the participants' amazing diversity: "college students, middle-aged
couples, families with small children, older people, environmental, and reli-
gious groups, business and civic organizations" {New York Times, February
16). In the UK press, with the exception of one reference to "anarchists," the

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25S^_

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 255

demonstrators were described as "people with an astonishing variety of back-
grounds and political viewpoints" {Guardian, February 17); or "nuns, CND,
social workers, anarchists, Walthamstow Catholic Church" members (ibid.,
February 16) or "a mixture of pacifists, nuclear disarmers and hard-leftists.
Many of the people were not pacifists or ultra-leftists (though they naively
gave credibility to the latter impressions) but part of the United Kingdom's
liberal conscience" {Times, February 20). In the Spanish El Mundo (Febru-
ary 16), the February 15 demonstrators were portrayed as being a mixture
of "Catholics, pacifists, immigrants, members of the Green Party, religious
minorities, parliamentarians of all the parties of the left-wing." Papers in
Italy, the Netherlands, Belgium, and Germany provided similar descriptions.
Yet, the German press also disproportionally referred to the presence of min-
isters and members of parliament "[German politicians] Thierse, Trittin,
Wieczorek-Zeul" {Frankfurter Allgemeine Zeitung, February 17), or "[minis-
ters] Wieczorek-Zeul, Kiinast, Trittin, and [head of parliament] Thierse"
{Frankfurter Rundschau, February 15). It seems that German politicians were
well aware of the popularity of their antiwar stance. In all countries, the media
tended to stress the astounding diversity of the February 1 5 protesters, thereby
painting a picture that did not match reality, as the results of our survey have
shown (see, in particular, chapter 5).

By and large, the demonstrations received newspaper coverage of which
most organizers only can dream of: protests were said not just to be the largest
ever seen, but in addition, to have attracted ordinary people from all parts
of the country and all layers of society. It seemed that the world had said no
to war on February 15.

Our final question concerns the ultimate challenge for protest organiz-
ers: In addition to positively framing the demonstrations, did they also suc-
ceed in getting their substantive message across? The answer, again, is mostly
positive. Slogans such as "the world says no to war," "no blood for oil," "we
can stop the war," "don't attack Iraq" and "not in my name" were mentioned
in almost all countries. In Italy, Spain, and the Netherlands, these general
catchphrases were complemented by those targeting the leaders of both war-
initiating and war-opposing countries. In Spain, for example, one could read
"USA Stay Home" {El Mundo, February 17); "Schroder Be Strong. Don't go
to War" and "The old Europe doesn't want a new war" (both in El Mundo,
February 16). Italian and Dutch papers echoed such messages. In Belgium,
the demonstrators were described as applauding their minister of foreign
affairs: "Thanks Michel — Merci Michel" {Het Laatste Nieutvs, February 17)
for his efforts to block a war-favoring NATO decision (see chapter 1). Yet
in Germany, apart from more general statements for peace, only one cited

H$H

-4

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25Bfe

256 RUCHT AND VERHULST

slogan indirectly addressed the UK government: "Make Tea, Not War!" {Frank-
furter Allgemeine Zeitung, February 17).

To summarize, the newspaper coverage of the Iraq crisis was far from uni-
form. Again, this proves that national context matters. The UK and especially
the U.S. press regarded the war as a national issue, with much attention paid
to its "practical" side, in terms of preparation, tactics, and costs, and with
relatively little emphasis on pacifism as a reason not to go to war. In the UK
press, considerable attention was given to the antiwar opposition within Prime
Minister Blair's Labour ranks. German press, however, reported on the Iraq
crisis as an international issue in which many countries were involved and
where many war-disapproving voices could be heard. A reflection of this atti-
tude is the importance given the concept of pacifism in German newspaper
articles. In all other countries' presses, too, the crisis was portrayed as an
internationalist issue, with Italian and Spanish newspapers paying much atten-
tion to opinions that disapproved the war and to pacifism as a reason not to
go to war. Yet, regarding the coverage of the February 1 5 protests in partic-
ular, differences are much smaller. In all countries, the protests tended to be
portrayed as positive events that brought onto the streets masses of diverse
people that could not be ignored. And, finally, the protesters also got their
messages and main slogans across.

Conclusion

In this chapter, we have used three kinds of sources — the voices of the "ordi-
nary" demonstrators, the organizers and speakers of the protests, and the
newspapers — to see how opposition to the (then imminent) war against Iraq
was communicated and framed. We highlight three general findings. First,
opposition to war was based not on a one-dimensional view but on a broad
variety of arguments, ranging from principal opposition anchored in uncon-
ditional pacifism to very specific reasons. Pacifism was the reason against war
most frequently mentioned by both the protesters and the newspapers. Al-
though all protesters agreed that the war was unjust and unfounded, we were
unable to identify an overarching interpretative ideological master frame for
this opposition to. Rather, we found a broad array of reasons to protest, among
which, contrary to the views of some observers, anti-Americanism was ex-
tremely marginal.

Second, the various arguments and frames regarding the war were not
evenly distributed across countries. Context mattered even in cases where
countries seemed similarly affected by the (potential) war. The single factor
with the biggest impact on the pattern of communication was whether gov-
ernments were actively willing to, or would eventually, engage in warfare

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25Zfe

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 257

(the United States and the United Kingdom) or reject it outright (Germany
and Belgium). Countries somewhere in between these extreme positions
tended to exhibit a more inconsistent pattern.

Third, while there seems much congruence in the views of the "ordi-
nary" protesters and the protest organizers and speakers, the media, not sur-
prisingly, were more selective in what exactly they reported about reasons foi
objecting to the war. Again, variation across countries is significant, reflect-
ing not only the position of the governments and public opinion toward
and protest, but also specific political constellations, for example, a divided
government in the United Kingdom. By the same token, the newspapers in
the United States, the country bearing the brunt of the costs of war, tended
to downplay the scope of opposition in other countries and international
institutions as well as in their own country. Overall, however, the protesters
were taken seriously by most newspapers and covered in a neutral or even
positive manner. Also, unlike in many other instances of protest, where media
tend to focus on side aspects rather the protesters' views and motivations,
the protesters were able to get their basic messages across. In part, this might
be explained by media and public opinion that sympathized with their cause.
To some extent, however, this relatively favorable way of reporting may also
/fa be a function of the protest's sheer size and perceived — but actually much sL.

| more limited — social heterogeneity. While relatively small groups of pro-

testers, even when covered by the mass media, can be easily presented in a
disrespectful and distorted way, such a characterization is hardly advisable
when millions of supposedly ordinary citizens (and media consumers!) take
to the streets. By and large, through media, the public in the eight countries
under study could get a clear idea of the scope and reasons of opposition to
war, though some analyzed newspapers tended to ignore or misrepresent the
protest altogether.

Notes

1 . Note that in the German case, for example, protesters even criticized their
own war-opposing government, because this government tended to tolerate logistic
support to the war by allowing U.S. troops on German soil.

2. Items 1,12, and 13 were classified as "moral-affective." Items 2 and 5 to 10
were classified as "cognitive."

3. Note that this is not a protest event analysis. We did not systematically
gather press articles reporting on protests. Rather, when any protest was mentioned
in an article we had found using search word "Iraq," our encoders filled out a sepa-
rate encoding form for it.

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25Kfe

258 RUCHT AND VERHULST

Works Cited

Barker-Plummer, Bernadette. 2002. "Producing Public Voice: Resource Mobiliza-
tion and Media Access in the National Organization for Women." Journalism
and Mass Communication Quarterly 79: 188-205.

Bennett, Lance W., and Robert M. Entman, eds. 2001. Mediated Politics: Commu-
nication in the Future of Democracy. Cambridge: Cambridge University Press.

Bundesausschuss Friedensratschlag [Federal Comittee on Peace Discourse]. 2002.
"Die Friedensbewegung ist alles andere als 'antiamerikanisch.'" AG Friedens-
forschung an del

ruary 18. Kassel, Germany, http://vvrww.uni-kassel.de/fb5/frieden/bewegung/
antiamerikanismus.html.

Cooper, Alice H. 2002. "Media Framing and Social Movement Mobilization: Ger-
man Peace Protest against INF Missiles, the Gulf War, and NATO Peace En-
forcement in Bosnia." European cal Research 41 : 37-80.

Cortright, David. 2004. A Peaceful Superpower: The Movement against War in Iraq.
Goshen, Ind.: Fourth Freedom Forum.

Ferree, Myra Marx, William A. Gamson, Jiirgen Gerhards, and Dieter Rucht. 2002.
Shaping Abortion Discourse: Democracy and the Public Sphere in Germany and
• the United States. Cambridge: Cambridge University Press.

-rtft- Gamson, William A. 1992. Talking Politics. Cambridge: Cambridge University Press. _^_

I Gamson, William A, and Andre Modigliani. 1989. "Media Discourse and Public

Opinion on Nuclear Power: A Constructionist Approach." American Journal of
Sociology 95: 1-37.

Gamson, William A., and Gadi Wolfsfeld. 1993. "Movements and Media as Inter-
acting Systems." Annals of 'the AAPSS 28:1 14-25.

Gerhards, Jiirgen, and Dieter Rucht. 1992. "Mesomobilization: Organizing and
Framing in Two Protest Campaigns in West Germany." American Journal of
Sociology 98: 555-95.

Granovetter, Mark. 1978. "Threshold Models of Collective Behavior." American
Journal of Sociology %?>■. 1420-43.

Hilgartner, Stephen, and Charles L. Bosk. 1988. "The Rise and Fall of Social Prob-
lems: A Public Arenas Model." American Journal of Sociology 94: 53-78.

Iyengar, Shanto, and Adam Simon. 1994. "News Coverage of the Gulf Crisis and
Public Opinion." In Taken by Storm: The Media, Public Opinion, and U.S. For-
eign Policy in the Gulf War, eds. W Lance Bennett and David L. Paletz. Chicago:
University of Chicago Press.

Kull, Steven, Clay Ramsay, and Evan Lewis. 2004. "Misperceptions, the Media, and
the Iraqi War." Political Science Quarterly 118: 569-98.

McLeod, Douglas M. 1995. "Communicating Deviance: The Effects of Television

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 25Sfc)

THE FRAMING OF OPPOSITION TO THE WAR ON IRAQ 259

News Coverage of Social Protest." Jo Electronic Media

39: 4-19.

Molotch, Harvey. 1979. "Media and Movements." In The Dynamics of Social Move-
ments: Resource Mobilization, Social Control, and Tactics, eds. Mayer N. Zald and
John D. McCarthy, 71-93. Cambridge, Mass.: Winthrop.

Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of
Groups. Cambridge, Mass.: Harvard University Press.

Rucht, Dieter. 2005. "Appeal, Threat, and Press Resonance: Comparing Mayday
Protests in London and Berlin." Mobilization 10: 163-81.

Smith, Jackie, John D. McCarthy, Clark McPhail, and Augustyn Boguslaw. 2001.
"From Protest to Agendabuilding: Description Bias in Media Coverage of Protest
Events in Washington, D.C." Social Forces 79: 1397-424.

Smith, Paul. 1999. "Political Communication in the UK: A Study of Pressure Group
Behaviour." Politics 19: 21-27.

Snow, David A., and Robert D. Benford. 1988. "Ideology, Frame Resonance, and
Participant Mobilization." In From Structure to Action: Comparing Social Move-
ment Research Across Cultures, ed. Bert Klandermans, Hanspeter Kriesi, and
Sidney Tarrow, 137-96. Greenwich, Conn.: JAI Press.

. 1992. "Master Frames and Cycles of Protest." In Frontiers in Social Move-

I merit Theory, eds. Aldon D. Morris and Carol McClurg Mueller, 133-55. New I

■W- Haven: Yale University Press. W

' Snow, David A., E. Bourke Rochford, Steven K. Worden, and Robert D. Benford.

1986. "Frame Alignment Processes, Micromobilization, and Movement Partici-
pation." Ami 51: 464—81.

Verhulst, Joris, and Stefaan Walgrave. 2005. "Gezien worden of Gezien zijn? Over
Nieuwe en Oude sociale bewegingen in de Vlaamse pers." Mens en Maatschap-
pij 80: 305-27.

Walgrave, Stefaan, and Jan Manssens. 2000. "The Making of the White March."
Mobilization 5: 217-40.

H$H

12 Chapter 12_Walgrave 22/01/2010 3:43 PM Page 2fiffifc)_

1

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page 2£|&)_

Conclusion:

Studying Protest in Context

m

Stefaan Walgrave and Dieter Rucht

On February 15, 2003, an unprecedented mass of people publicly expressed
their indignation in hundreds of cities around the globe. About one month
later, the United States and its allies did what the demonstrators had sought
to prevent: they invaded Iraq because of its alleged possession of weapons of
mass destruction. At least to the invaders, it seemed that this war would
soon come to an end. On May 1, 2003, aboard the aircraft carrier Abraham
Lincoln, President Bush declared "one victory." Behind him hung a banner
reading "Mission Accomplished." However, as we write these lines in April
2008, the war in Iraq is still not over. The large contingent of troops mainly
from the United States and United Kingdom seems incapable of controlling
the situation even after the Bush administration increased the contingent of
in Iraq. In hindsight, it appears that the fears of the February 1 5 demon-
strators were largely warranted — no wonder the fifth anniversary of the be-
ginning of the war provoked another wave of protest in many countries.

This volume, however, does not discuss whether the reasons for war were
legitimate, which course the war took or which consequences it had. Rather,
we have engaged in analyzing who those demonstrators against imminent
war were, why they took to the streets, and how they were mobilized. We drew
on exceptional empirical evidence based on thousands of answers to a ques-
tionnaire that was distributed in eleven cities in eight Western countries.
Our main aim was to compare the demonstrators in the eight nations to
make headway in understanding how the protest events on February 15 were
set up with relation to a specific environment, and how they were determined
by their specific context. The apparently homologous nature of the protest

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page

262 CONCLUSION

events under study — timing, slogans, action type, platform, and so forth were
almost identical — gave us the unique opportunity to investigate the simi-
larities and differences of these protests and the reasons that may account for
them.

Our Ambition, Theoretical Framework, and Assumptions

The central idea for our endeavor was that the protests on February 1 5 must
be studied in their political, social, and cultural context. We set ourselves
apart from the classic political participation studies, as they entirely decon-
textualize protest participation and do not allow the examination of how pro-
test motivation and eventual actual participation are molded by the structural
and situational environment. We also argue that most social movement stud-
ies, even when taking a comparative perspective, have difficulties in identi-
fying and weighing causal factors because too many variables come into play
(see Kolb 2007). In our cross-national research setting, however, the issue,
time frame, and form of protest is held constant. In this regard, we had an
ideal research setting — a kind of "natural experiment" that allows us to study
the impact of contextual factors.

The available macro- and meso-level theoretical perspectives do not offer
I clear clues as what to expect of individual protesters, their sociostructural fea-

~W tures, attitudes, and behaviors. The political opportunity structure approach, ~W

for example, which seems to be well-suited for comparative analysis, does
not make inferences about individual protesters. Rather, it was designed to
explain meso-level variables (social movements, social movement organiza-
tions, and campaigns and their successes) or even macro-level variables (the
amount and cycles of protest in a given society). Therefore, in the introduc-
tory chapter, we suggested five different context layers that may determine
or predetermine what individual participants may look like.

First and very generally, we assumed that the sociodemographic com-
position of the population at large has an impact on the composition of the
protester populations. If people in a country are generally highly skilled, we
expect a large proportion of protesters in that country to be highly skilled as
well, at least in a protest that does not recruit a particular social group such
as farmers or workers. Second, stable structural features of the political sys-
tem may also play a role, as they may determine the strength of the general
movement sector on which specific protest events can build. We hypothe-
sized that certain political structures foster a strong progressive movement
sector, which, in turn, is likely to produce a high turnout in peace protest.
Third, we introduced the concept of the issue-specific context, referring to
the particular political and social environment in which a certain protest event

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page 2fiXfc)_

CONCLUSION 263

is staged. How do government and opposition position themselves on the
issue? Is there support on part of the mass media? What does the public at
large think about the issue? Probably more so than general political oppor-
tunities, the issue-specific context matters in terms of the size and kind of
protest and the characteristics of the protesters. Fourth, we assumed that,
additionally, social movements and their structures, goals, and strategies affect
the profiles of the protesters. In countries with a strong peace movement, for
example, we expected the February 15 protests to be populated to a greater
extent by peace movement activists than in other countries. Finally, we also
expected demonstrators' characteristics to be influenced by the specific mobi-
lization strategies and processes that the organizers used to incite people to
take to the streets.

Main Descriptive and Explanatory Findings

In this volume, we were not able to systematically test the preliminary model,
which, at any rate, should be considered more of a toolbox or heuristic device
than the result of an integrated theory. In particular, we could not assess the
relative weight of these five contextual layers. Yet, looking at the empirical
evidence presented in this volume, we believe we can safely state that at least
I parts of the model seemed to have worked quite well and therefore offer valu-

~W able insights for better understanding the factors and mechanisms of protest ~W

participation.

One of our key findings is that quite substantial differences exist among
the February 1 5 protesters in the eight countries studied. This holds in spite
of that we chose to investigate protesters from similar countries (regarding
the logic of a most-similar-systems design, see Przeworski and Teune 1970)
focusing on the same issue, responding to same call issued in a prior social
movement meeting, and, finally, protesting on the same day. Yet it was only
at a first glance that the protesters may have appeared very similar — with the
same slogans, appearance, et cetera. As a matter of fact, contrary to our ini-
tial expectations, the protesters in the eight countries exhibited considerable
cross-national differences in their sociodemographic profile, their attitudes,
and their behaviors.

Regarding the sociodemographic profiles, of course, the average dem-
onstrator, in the aggregate of all countries, resembled the typical new social
movements' activists with high levels of education, a relatively large propor-
tion of whom were women, belonged to the younger age cohorts, and pre-
dominantly worked in the human service sector. But beneath this aggregate
profile we found notable differences. Just two examples: demonstrators were
much younger in Switzerland, and in Belgium, an extremely large percentage

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page

264 CONCLUSION

of demonstrators had high educational levels and a disproportionately large
number of them were male (see chapter 5).

In a cross-national perspective, attitudes varied as well. To be sure, all
who protested on February 1 5 were strongly opposed to the war. Nearly all
participants thought that it was fought for political control and/or control
over oil and that war without UN approval was unjustified. But apart from
this general rejection of war on Iraq, we came across significant differences
among countries in the degree of unconditional pacifism, the embeddedness
of antiwar positions in broader opposition to the government, or general feel-
ings of political dissatisfaction (see chapter 6). The protesters, when asked
in an open-ended question about their reasons and motives to participate,
offered a broad range of arguments, again showing significant differences
across countries (see chapter 12). Neither in their answers nor in written
pamphlets and speeches held during the demonstrations did we find evi-
dence of an overarching ideological master frame as a driving force to pro-
test, apart from general statements, as expressed in the joint call for action,
to "fight ... for social rights and social justice, for democracy and against
all forms of oppression." Interestingly, anti-Americanism, an important rea-
son underlying the protest according to some commentators, was almost
I completely absent in these many answers.

W We also found significant variation in the extent of membership in polit- v9)~

ical groups and previous protest activity. In general, many respondents to our
questionnaire had considerable protest experience, and many were active in
voluntary associations and social movements. But in some countries — the
Netherlands, for example — the clear majority of the demonstrators had taken
to the streets before. And in other countries, the United Kingdom for in-
stance, protesters were much less engaged in civil society groups (see chap-
ters 7 and 10). Participation in the antiwar protest is strongly correlated with
a leftist orientation and identification with the global justice movements
and similar groups focusing, for example, on labor and social issues, or third-
world solidarity (chapter 7). However, many of the organizations running
the protest were not primarily oriented toward peace issues (chapter 10). In
other words, the protesters, characterized by "complex political identifica-
tions," according to the analysis in chapter 11, were recruited from a broad
spectrum of political and social groups. Again, the range of these groups
varies across countries, with broader alliances, for instance, in Italy, Spain,
and the Netherlands, where governments were supporting the war from a
more cautious and less committal position, and smaller alliances in the war-
leading United States and United Kingdom (chapter 10). In line with ear-
lier research findings (e.g., Dalton 2002), the demonstrators, when asked

H$H

-4

13Conclusion_Walgrave 22/01/2010 3:43 PM Page

CONCLUSION 265

about their previous political activities, tended to mention both unconven-
tional and conventional forms of action. As chapter 8 shows, they were not
turned away completely from the electoral process and party politics. Yet, this
chapter also provides evidence that the protest had little effect on subse-
quent voting behavior.

Most of the chapters tried to explain the many cross-national differences
among the protesters found in our data, by drawing on one or more of the
five context layers presented above. Given certain problems of measurements
on the side of independent variables as well as the complexity of the task,
some authors could not fully explain the cross-national differences they found.

The most valuable tool by far to account for cross-national differences
in sociodemographic characteristics, attitudes, and behavior was what we
have called the issue-specific context of the protests — which Kriesi (2004,
70) has labeled the interaction context. Regarding this context, several more
specific dimensions come into play. First, the position of established politi-
cal actors is of crucial importance. Whether the national government par-
ticipated in, supported, or opposed war, and how the oppositional parties
reacted to that official stance, made a sizeable difference regarding the profile
of the (potential) protesters. The classification of countries presented in chap-
ter 3 by rank-ordering countries on a simple pro-war to antiwar scale proved
to be the best overall predictor of demonstrators' cross-national differences.
Although an analysis based on only eight countries limits the reach of our
conclusions, in many cases the simple fact that demonstrators lived in a coun-
try participating or not participating in war had a considerable impact on
their profiles when it comes to their sociodemographic characteristics, range
of motives to participate in the demonstration, and embeddedness in civil
society groups.

More or less the same applies to the second dimension of the issue-spe-
cific context: public opinion on the issue of war. Whether the public in a
country rejected or approved of war — to be precise, the extent to which it
rejected war — affected the protests and the protesters. Though there is strong
evidence that both factors of the "issue-specific context" — the stance of gov-
ernment or opposition and the populations stance on war — do matter, we
can assess neither the differential effect of these factors on the protesters nor
how the factors are related. On theoretical grounds, we believe that the polit-
ical positions of governments or oppositional parties affect public opinion
rather than vice versa. Accordingly, public opinion can be seen as an interme-
diary variable (strongly influenced by the stance of the government, yet not
necessarily in accordance with this stance) that affected the composition of
the protesters.

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page

266 CONCLUSION

"4

The media, the third dimension, act as intermediaries connecting estab-
lished political actors and opinions of the populace. After all, almost all the
information people receive about the war is presented via mass media. More-
over, we can assume that both the kind and distribution of speaker mentioned
in the media and the media's own positions regarding war, as expressed, for
example, in commentaries, also influence peoples views and attitudes.

At a closer look, it appears that the main factor affecting the sociostruc-
tural characteristics, attitudes and behavior of the demonstrators was not
merely governmental (or oppositional) stances on a continuum from par-
ticipation in war (United States and United Kingdom) to support (Spain,
Italy, and the Netherlands) to opposition (Switzerland, Belgium, and Ger-
many). Rather, it was whether or not the country was or would be militarily
engaged in fighting the war. Indeed, the United States and United Kingdom
demonstrators stood out in most analyses and differed most from those in
all the other countries. We found that, in some respects, the Spanish, Ital-
ian, and Dutch demonstrators — their governments supporting the war only
with words — differed from the rest as well. Yet, the actual participation in
war clearly was most consequential for the composition of the protesters. It
appears that "rally around the flag" mechanisms and "support our boys" dis-
courses substantially affected which kinds of movements, organizations, and
people publicly mobilized against war. Moreover, in both countries with war-
promoting administrations, even the main opposition parties were not able
to raise a strong and consistent argument against war; instead, they were mar-
ginalized in a spiral of pro-war discourse. Consequently, in these two coun-
tries, the media and public opinion were also by far the least opposed to the
idea of ridding the world from Saddam Hussein and his supposed "weapons
of mass destruction" by engaging in warfare. In the United States in partic-
ular, the newspapers, and perhaps the media in general, tended to downplay
the extent of opposition to war in other countries and in international institu-
tions such as the United Nations (see chapter 12). The situation was entirely
different in the merely war-supporting — but not war-leading — countries
(Spain, Italy, and the Netherlands) where the left-wing opposition noisily and
forcefully argued against war, and the entire left-wing social movement sec-
tor unambiguously mobilized against war. In these countries, demonstrators,
for the most part, did not differ that much from the protesters present in the
war-opposing countries (Switzerland, Belgium, and Germany). Also, the media
in these countries tended to provide a more complete and fair account of the
opposition to war than the United States did.

However, our argument that especially active engagement in warfare
made a difference to people who ended up demonstrating may be challenged

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page 2fiSt)_

CONCLUSION 267

by the observation that both belligerent countries also differed in other re-
spects: they are Anglo-Saxon nations not belonging to continental Europe,
unlike the rest of the nations in our sample. How can we be sure that the
systematic differences we found are a result of their governments engaging
in war and not an expression of Anglo-Saxon political culture? Again, in oper-
ational terms, we cannot separate the impact of these factors. But we believe
it is difficult to conceive that public opinion and media in these two coun-
tries also dislike war less than they do in the six other countries. Unfortu-
nately, we do not have at our disposal systematic evidence regarding general
war-and-peace attitudes of the population in our sample of a

Let us briefly summarize when and how the position of g
and political opposition (and likely linked media coverage and public opin-
ion) made a difference, as shown in various chapters of this book. Consid-
ering the sociodemographic traits of the demonstrators, chapter 5 showed
that their "normalization" — the extent to which they were heterogeneous and
resembling the population at large in a given country — was by far the least
obvious in the United States and the United Kingdom. In these two coun-
tries, protesters formed a more homogenous crowd, while the in Switzerland
and the Netherlands, they were the most diverse. Chapter 6 studied the atti-
| tudes of the demonstrators and, not surprisingly, established that those in

~W the bellicose countries were more opposed to their governments. More sig- ~W

nificant, perhaps, this chapter showed that these protesters also were more
skeptical about the general functioning of democracy in their countries and
that they were more politically dissatisfied. In short, in countries engaging
in or rhetorically supporting the war, resistance against war was embedded
in broader oppositional sentiments, bringing different kinds of people to the
streets. Chapter 8, focusing on political parties, revealed that although the
protesters were mainly left-wing party supporters, there were interesting cross-
country differences in their partisan makeup depending on the stances of the
countries' government and opposition. Regarding mobilization patterns, chap-
ter 9 demonstrated that the official governmental position, translated into
public opinion vis-a-vis the war, affected the dominant type of mobilization
in the different countries. The more the public endorsed the antiwar cause,
the more the mobilization relied on open channels. Chapter 10 presented
evidence of the significance of issue-specific political context. Different kinds
of organizations promoted the marches or rallies in the eight sampled coun-
tries, and, consequently, participants had quite different organizational rela-
tionships with the protest organizers. In the United States and the United
Kingdom in particular, the ties of participants with organizing groups were
weaker than in the other countries.

H$H

-4

13Conclusion_Walgrave 22/01/2010 3:43 PM Page

268 CONCLUSION

Dramatically different issue-specific national political contexts — the gov-
ernment's position on the war — almost consistently affected most dependent
variables. This finding raises questions about the prevalence and importance
of national versus international political contexts for protest participation.
Since for the majority of the eight countries, the war on Iraq was a "foreign"
issue, and they were not actively engaged in waging it, one might assume
that their national contexts are of little relevance. In fact, however, the spe-
cific national context seemed to have molded the protests most. We conclude
that international political opportunity structures do not directly affect pro-
tests and protest participants, but, rather, the latter are mediated by national
opportunity structures, power structures, conflict constellations, media cov-
erage, and public opinion. If the international context had been primary, we
would not have found as many substantial cross-national differences among
the protesters as we did.

Various chapters showed that the issue-specific context explains quite a
few differences among demonstrators. Thus it is a useful analytical tool.
What about the other explananda or "layers" in our heuristic model? Chap-
ter 5 documented that the composition of the demonstrators in terms of age
and education is associated with a country's general sociodemographic pat-
tern, the most general factor in our model. None of the chapters found a
clear link between the protesters' features and the general properties of the
national political context — the stable institutional arrangements and the sub-
sequent strength of the progressive social movement sector. While this is not
surprising, because the political opportunity structure approach was never
meant to explain the characteristics of individual participants in specific pro-
test events, chapter 2 did not locate a consistent link between structural access
for all kinds of challengers and the strength of the progressive movement
sector. Thus, based on general properties of political systems, it was impos-
sible to generate clear and unambiguous hypotheses about the number and
kind of demonstrators in the different countries.

Also the massiveness of previous antiwar demonstrations and the power
of the peace movement in the past, as explored in chapter 4, did not prove
to be a good predictor of the composition of the February 15 protesters. Com-
paring the protest against the deployment of nuclear missiles in the 1980s
with the resistance against the Gulf War in the early 1990s, we found major
discrepancies, with some countries mobilizing strongly in the 1980s and much
less so in the 1990s, and vice versa. This, also, made us uncertain as to the
turnout we might expect on February 15. However, as documented in chap-
ter 10, general levels of political participation, protest patterns, and features
of the social movement sector seem to have affected the organizational basis

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page

CONCLUSION 269

of the demonstrations. In countries that had strong global justice mobiliza-
tions prior to February 15, 2003, like Italy and Belgium, antiwar protesters
appeared to exhibit the traces of this past. In countries with strong and active
unions, the protesters more frequently were a trade union member. And in
countries with strong mass political parties, the protest participants were more
active in parties than in countries with fewer party members.

The final layer in our heuristic model comprised the specific patterns and
channels of mobilization adopted by the organizations staging the protest.
Did mobilization affect the differential organizational composition in the
eight countries? This clearly was the case. The open or closed character of
the mobilization had significant consequences on recruitment, as chapter 9
demonstrated. Sociodemographic characteristics, attitudes, and political be-
havior all were affected by the mobilization pattern. When laying out our
heuristic model in the introduction to this volume, we considered mobi-
lization pattern an intermediary variable. Indeed, while directly affecting the
dependent variables, that is, the characteristics of the protesters, the typical
mobilization pattern in the eight countries was, in turn, affected by two inde-
pendent variables of the model. The issue-specific context — in terms of the
public opinion regarding war and the strength of the social movements in a
I country — membership in organizations and patterns of protest were directly

~W associated with the openness or closedness of the mobilization process in the ~W

eight nations. Mobilization seemed to work as an intermediary factor, trans-
lating more structural dimensions of the political context into the specific
composition of the protesters.

Concluding the evaluation, we think our heuristic model proved to be
a theoretically reasonable and empirically plausible tool. Admittedly, we are
unsure about precise interaction of the independent and intermediary factors.
And we cannot, at this stage, examine the net effect of the various factors.
Yet we have evidence that four of the five proposed factors that hypotheti-
cally influence the specific composition of the protests and the protesters did
play roles and determined or codetermined the sociodemographic profile,
attitudes, and behaviors of the participants. Above all, the issue-specific con-
text, in particular the stance of government and opposition on the issue of
war, seems to make a great difference regarding the kind of protesters that
took the streets. The sociodemographic composition of the population appears
less influential. Also, the strength of the social movement sector seems to
affect the makeup of the demonstrations. Mobilization processes do have
the expected effect: they turn structural context factors into a specific com-
position of protesters. Only the political system variables did not yield sub-
stantial findings; we could not even use them to make concrete predictions

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page 2Zffi^_

CONCLUSION

about the strength and shape of progressive movement sectors. Though we
cannot rule out that the most general and stable characteristics of a given
political system might ultimately have some effects on the individual-level
composition of protest events, we assume that such effects are very indirect
and probably mediated by many additional variables. Political opportunity
structure approaches are simply too abstract and too general to account for
specific protest events and even more so for profiles of protesters. Therefore,
we advocate for the elaboration of middle-range theories that link macro
structures to micro behavior. In the social sciences in general, macro-micro
bridges (see Dogan and Rokkan 1969) are one of the weakest and most
underdeveloped areas. This is also reflected in studies on social movements
and, more generally, political participation. We hope we highlighted some
factors and variables that might be helpful in building such a bridging the-
ory, though we are fully aware that this is no more than a first step.

Lessons for Further Research

In this volume we had not only theoretical aspirations; we also sought to test
whether large-scale surveying of protest participants is feasible and, more im-
portantly, whether it is useful. Surveys of protesters on the spot are not very
_^k common, and they have never been used on such a large scale as in our com- _^_

| parative endeavor. Conducting simultaneous surveys in eight countries and

eleven protest venues is not an easy task. Yet it is possible, as our collective
effort shows. It is, of course, up to the reader to judge whether protest surveys
are useful to better understand the dynamics of protest events and protest
participation. We hope to have shown that surveying protesters by ques-
tionnaire offers valuable and hard data that can put existing theories to the
test and contribute to developing new theoretical insights.

Surveys of protesters necessarily imply a shift from the thus far prevail-
ing focus on the meso- and macrolevel aspects of social movements and
political protest to the micro level. The individual and his or her individual
participation are put on center stage. For social movement research, this is
quite a change, asking for better linking of meso- and macrolevel contexts
on the one hand with individual characteristics and behaviors on the other.
In fact, this entire volume has been a constant struggle to adjust existing the-
ories and insights to the individual level of analysis. Methodological inno-
vation, thus, has theoretical consequences. The main limitation of surveys
of protesters is that they cover only people who were successfully mobilized.
Accordingly, we cannot directly compare participants with nonparticipants. In
a sense, such surveys sample on the dependent variable. One way of dealing

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Page 2Z|g)_

CONCLUSION

with this inherent problem of the research design is to resort, as we did, to
comparison of the participants in a given event and in different nations.

Our focus was to explore the links among three groups of variables: first,
the relatively inert political structures; second, the — thus far largely ne-
glected — issue-specific context, including issue-specific stances of political
actors and the populace, but also channels of mobilization; and, third, the
characteristics of those participating in a specific act of protest. We were lucky
to rely on a quasi-experimental comparative research design: held constant
was not only the very specific issue but also the day of the protest. Moreover,
we studied one of the largest coordinated single protest actions in history by
using the same questionnaire distributed on the spot, which produced a large
data set that we analyzed according to analytical questions rather than offer-
ing country-by-country chapters. Taken to together, this resulted in a unique
enterprise that involved dozens of researchers and their assistants. Never-
theless, when we look at the wider field of research of social movements and
collective protest, it is obvious that we covered only limited ground. Only
two of the many limitations: We did not study in much detail the planning
of the protest, its logistics, or the concrete mobilization process as it occurred
in the eight countries and eleven cities where we distributed our question-
| naire. Nor did we study the impact of these protests on public opinion, pub-

"W" lie debates, and policy makers. ~®~

Though we consider cross-national comparison, as applied extensively
in this volume, to be a fruitful research strategy, we also think that surveys
of protesters yield even more insights when different protest issues are com-
pared within a country or across countries. On such a basis, we also expect
to find further evidence for the crucial relevance of issue-specific political
opportunities, mobilization structures, and collective action frames.

Works Cited

Dalton, Russell J. 2002. Citizen Politics: Public Opinion and Political Parties in Ad-
vanced Industrial Democracies, 3rd ed. New York: Chatham.

Dogan, Mattei, and Stein Rokkan, eds. 1969. Quanti: Analysis in the

Social Sciences. Cambridge, Mass.: MIT Press.

Kolb, Felix. 2007. Protest and Opportunities: The Political Outcomes of Social Move-
ments. New York: Campus.

Kriesi, Hanspeter. 2004. "Political Context and Opportunity." In The Blackwell Com-
panion to Social Movements, eds. David A. Snow, Sarah A. Soule, and Hanspeter
Kriesi, 66-90. Oxford: Blackwell.

Przeworski, Adam, and Henry Teune. 1970. The Logic of Comparative Social Inquiry.
New York: Wiley.

H$H

13Conclusion_Walgrave 22/01/2010 3:43 PM Pagp. ?7^ ± \

4

-^-

4 Acknowledgments_Walgrave 22/01/2010 3:44 PM_E

Acknowledgments

This book studies collective action and it largely draws on collective action.
The volume is the result of efforts by many people, not only those author-
I ing or coauthoring chapters. They were helped by a host of other people in

~W their respective countries. Although not authoring a chapter here, Manuel "W

Jimenez coordinated the data collection in Madrid; we are grateful t<
for letting us use the Spanish data. In the United States, main researchers
Lance Bennett and Terri Givens had a hand from people who helped get this
study into the field in Seattle and San Francisco: Victor Pickard, David Iozzi,
Lisa Horan, and Edith Manosevitch. Lars Willnat organized the team that
went into the field in New York. In the United Kingdom, Catherine Eschle
organized the fieldwork in Glasgow, and Derek Wall in London. In other
cities, numerous other people, most of them students, were also engaged in
collecting data.

Similarly, media data were not collected only by the authors of the media
chapters. More concretely, we thank Claudia Schmidt for collaborating on
the design of the coding scheme and letting us use the German media data
she collected.

We are grateful to the Forum Trentino per la Pace, and its president
Roberto Bombarda, for generously funding a public event at the Faculty of
Sociology in Trento on October 4, 2004, when some of our preliminary find-
ings were presented. We thank Maria Fabbri and Enzo Loner for assisting
with the organization of our meetings there.

Two anonymous reviewers and Sidney Tarrow read earlier versions of

H$H

4 Acknowledgments_Walgrave 22/01/2010 3:44 PM_E

274 ACKNOWLEDGMENTS

the manuscript and provided useful commentaries. We thank them for thei
suggestions.

Finally, the European Consortium for Political Research deserves ou
gratitude for the funding it provided the authors and editors to organiz
working meetings and finalize the book.

1

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page ZZBfc

Appendix A:

Methodology of Protest Surveys in Eight Countries

Stefaan Walgrave and Claudius Wagemann

This book largely draws upon a central dataset comprising data collected in
the eight countries under study. The main source is a survey conducted on
the same day February 15, 2003, based on a sample of almost six thousand
| demonstrators participating in massive protest events against the upcoming

~W war on Iraq. This survey was conducted by eight local country teams using ~W

(nearly) identical questionnaires and a similar fieldwork method. The Belgian
team, directed by Stefaan Walgrave, took the initiative for this endeavor.
This team also provided the funding for data input and preparation. When
the United States and the United Kingdom started to make clear in Decem-
ber 2002 that they would wage war on Iraq, the Belgian team began forging
a network of social movement and political participation scholars to survey
the potential demonstrations against the war. At that point, February 1 5 was
still relatively far away. The international team adapted the questionnaire
that the Belgian members had developed for previous protest surveys (Van
Aelst and Walgrave 2001; Norris, Walgrave, and Van Aelst 2005), as well as
their sampling procedure. Surveying is not a conventional or widely used
method for analyzing protest events; rather, techniques such as participant
observation, (qualitative) in-depth interviews, protest event analyses, life his-
tories, text analysis, and so forth have been applied. So the application of
survey techniques — widespread in other subdisciplines of political science
and sociology — is rather novel for social movement research.

Opting for such a method also has quite a substantial effect on the logic
of our research. Whereas previously applied methods mostly point to the moti-
vations, behavior, and attitudes of individual participants in demonstrations,

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page 22St)_

276 APPENDIX A

the individual person here disappears in an N survey and becomes a row in
a data matrix. At the end of the day, variables interest us much more than
cases. The methodological literature has called this a "variable-oriented ap-
proach" rather than a "case-oriented approach" (Ragin 2004). The positive
effect of such a method is that a concentration on variables, using large data-
sets, helps us increase causal leverage and obtain broader generalizations of
our results. Indeed, we might be able to draw conclusions about how differ-
ent variables work in different countries (or in our aggregated dataset as a
whole). However, we also risk the negative effect that correlations lead us to
conclusions based only on covariations instead of a rigidly performed in-
depth causal analysis of a few cases. If we find, for example, that first-timers
in demonstrations are more reluctant to use violence as a means of express-
ing their political opinions, then this is nothing other than a (possibly non-
randomly appearing) covariation of two distinct variables, and we can include
the causal nexus only with other analytical techniques. However, since the
February 1 5 demonstrations offered us the (perhaps) unique opportunity to
collect data from large samples in comparatively similar settings (similar slo-
gans and aims in all countries; same date and, therefore, same current state
of international politics; same issue, namely, the war in Iraq; etc.), we decided
I to expose the domain of protest events analysis — which has long been case-

~W oriented in its methodology — to a rather new methodological perspective, ~W

namely, the variable-oriented one.

As the following discussion will show, this did not work entirely neatly.
However, this was the lack of previous experience with survey techniques in
protest research that led to problems in the application of the method, but
rather the inherent characteristics of our object of research, that is, mass
demonstrations. Indeed, mass demonstrations are embedded in different con-
texts (usually highly emotionalized ones, with fluid issues, and combining
rather heterogeneous sets of actors, and so forth) than consumer behavior or
voting patterns. Therefore, they are analyzed in very specific circumstances,
so we first had to adopt strategies for coping with these specificities. In the
following, we will explain our procedure and sampling method in more
detail. Note that the method we have sketched out here has not been fol-
lowed in great detail in all countries; sometimes, specific circumstances made
it almost impossible to stick to the fieldwork method guidelines explained
below, in particular.

As mentioned, interviewing participants at protest demonstrations can-
not build on much previous methodological knowledge. Favre and colleagues
even speak of "a strange gap in the sociology of mobilizations" (Favre, Fil-
lieule, and Mayer 1997). To the best of our knowledge, protest surveying has

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page ZZZfc

APPENDIX A 277

only been used in a few studies (see, for example, Jasper and Poulsen [1995]
and Waddington [1988]). Most elaborate is the work of the French research
team including Favre, Mayer, and Fillieule, who developed a technique de-
signed to offer all participants an equal opportunity of being interviewed
(Fillieule 1997). Stefaan Walgrave and Peter Van Aelst refined their method
further (2001).

The actual process of surveying demonstrators applied in this study,
based on a random sample, was twofold. First, two fieldwork supervisors
(each accompanied by a team of questionnaire distributors/interviewers)
counted the rows of participants in the moving cortege, selecting every «-th
row, to ensure that the same number of rows was skipped throughout the
demonstration, and that the whole procession was covered. This should guar-
antee that all groups, no matter whether their members preferred to walk in
the first or last part of a march (this issue is also linked to questions of visi-
bility of a group in a march), have an equal chance to be part of the sample.
One of the two groups of fieldwork supervisors and distributors started at
the first row of demonstrators in the march and then gradually moved to the
back, counting and skipping rows until the last one. The other group of
fieldwork supervisors and distributors, the athletic ones (as they had to over-
| take the entire marching crowd), started at the end and gradually worked

~W their way to the head of the march. Each time a row was selected by the ~W

fieldwork supervisor, the distributors selected every w-th person in that row
and handed out questionnaires to these individuals. Ideally, the distributors
alternated among demonstrators at the left side, at the right side, and in the
middle of a row, again taking into account that some participants would
prefer to march at the margins or in the center, respectively, of the crowd.
Questionnaires were distributed in an addressed envelope, with postage paid
by addressee. This sampling method, as simple as it may seem, aimed at
guaranteeing an even distribution of the questionnaires over the demonstrat-
ing crowd and at giving every participant the same chance of being selected,
no matter where and with whom she or he marched.

The selected participants were asked to complete a ten-page question-
naire at home and to mail it back. The questionnaire used for February 1 5
in each of the countries had a common core, including questions on the par-
ticipants' profiles, the mobilization context, and political attitudes and behav-
ior, plus a few specific items in some of the countries. The answers had to be
short and succinct and were very often formulated in a multiple-choice for-
mat, in yes/no responses or in coded evaluations of opinions. In addition to the
mail survey, a random sample of other demonstrators in Belgium, the Neth-
erlands, Switzerland and the United Kingdom was interviewed face-to-face

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page ZZSfe

278 APPENDIX A

before the start of the demonstration. To do this, before the start of the
demonstration the gathering crowd was divided into sectors, and each inter-
viewer randomly, often following a specific procedure, selected a set number
of respondents in his or her sector. These short interviews were used as a
crosscheck to evaluate how far the different attitudes in responding to the
mail survey might have biased the result of the postal interviews. Confi-
dence in the reliability of this comparison was strengthened by the fact that
hardly anyone refused to participate in a face-to-face interview: Riidig assessed
this systematically in the Glasgow part of the survey and found that the
response rate there was no less than 95.3 percent (2006). However, in this
volume we have only drawn on respondents from the postal surveys, since
these questionnaires cover more variables, we did not carry out face-to-face
interviews for all the countries under research, and face-to-face data are only
used to test the representativity of the postal answers.

As all surveys, those used here for demonstrations also raise important
questions about the reliability and the representativity of sampling proce-
dures. Here, in since this is unknown territory, this is even more so: we do
not have any information whatsoever about how sociodemographic factors
or political attitudes might influence the response behavior. An undesired
| effect could be, for example, that demonstrators with more radical opinions

~W and/or attitudes would be more likely to refuse a response than others (alter- ~W

natively, they might be even more motivated to participate in a survey — but
also this would cause a bias). This is further exacerbated because we had rea-
son to believe that the February 1 5 demonstrations were not routine events.
Indeed, this was many demonstrators' first protest experience (see chapter 8).
However, there are also practical difficulties in conducting a truly ran-
dom sample: first, if a demonstration is large and fairly static and the streets
are congested with people, it becomes difficult for the interviewers to get
through the whole march and cover all kinds of groups, since they are sim-
ply immobilized. Second, it is impossible to get a good sample of respondents
in violent and/or irregular demonstrations (or in violent sectors of an oth-
erwise peaceful demonstration), even if we know that these forms of protest
are usually small in number. Third, in some exceptional cases, extremist groups
of demonstrators within a peaceful event refuse to accept the question-
naires — something that, of course, lowers the degree of representativity and
biases the result. Yet again, this is rare, and, in general, demonstrators who
have deliberately chosen to express their political opinion on the street are
likely to collaborate and may even be anxious to share their views with re-
searchers. Fourth, the weather conditions also play a role when one is con-
ducting research literally "in the streets": when it is raining cats and dogs,

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page ?73fc)

APPENDIX A 279

for example, the distributed questionnaires can become soaking wet and can
no longer be filled out and returned. Additionally, if it only starts raining
when a demonstration has already started, questionnaires that were distrib-
uted before the rain shower are much more likely to be sent back than the
others.

In terms of the February 1 5 sampling, the weather in the cities we sur-
veyed was mild. We can assume that the differences in temperatures did not
play any important role. Furthermore, the demonstrations we covered were
not irregular nor were they violent, nor had they violent sectors. Instead, they
were for the most part orderly processions without running or accelerating
segments. There were hardly any extremist demonstrators around, so this
did not pose a problem, either. However, it turned out that the unexpectedly
high numbers of demonstrators in some of our city cases made it simply
impossible to strictly follow our field survey method. For example, some
interviewers got stuck on the stairs in subway tunnels, or the crowds in the
streets did not move at all and the interviewers became completely stuck
because of the enormous attendance. When either the interviewers or the
crowd were static, the sampling procedure could not always be applied This
was the case in London, Madrid, and, for the most part, Berlin: in these
I cities, the static sampling procedure of dividing the standing crowd in sec-

~W tors was adopted also for the distribution of the postal questionnaires, which ~W

implies fewer guarantees that a representative sample of demonstrators was
drawn. Furthermore, not all interviewers and supervisors had previous expe-
rience with such the fieldwork method developed by the Belgian team; this
also caused some confusion and misunderstandings here and there. Although
in many instances we managed to follow the fieldwork procedure and sam-
pling method quite in detail, this was not always the case. We are unsure about
the effects different modes of application of the method had on our sample,
and we cannot test them. Yet, since we know that the disturbances to the
strict application of our sampling methods were not very big, we have reason
to believe that our samples in the different demonstrations are, by and large,
comparable. Social science often is forced to look at dynamic, spontaneous
forms of behavior and activity — not systematically evolving ones — so that
the methods it uses cannot be "exact," as in the natural sciences or as in more
experimental settings of social science research. This is even more the case
for demonstrations, which are, by definition, fast moving, hardly controllable,
quickly changing objects of research: that even the strictest method has to
be applied flexibly. Saying this, we believe that the distortions in the sample
should not make us worry too much, since the high number of questionnaires
and the dominance of cases in which we were able to apply the method in

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page 2BSfc)_

280 APPENDIX A

an ideal setting might outweigh the practical problems that occurred in some
instances.

In one case, the national team opted for a different sampling method.
In Italy, the team, led by Donatella della Porta and Mario Diani, decided to
draw mainly on face-to-face interviews, conducted before the Rome demon-
stration. Because the Italian team feared that response rates for the postal sur-
veys would be too low, they interviewed the demonstrators on trains on their
way to the protest in Rome. Of course, this might pose serious challenges to
the comparability of the Italian sample with the other national samples.
Interviewing on the spot might indeed lead to different answers than would
filling in a questionnaire quietly at home. Moreover, on packed trains, it is
very difficult to conduct interviews in private. Hence, social desirability and
group pressures may affect the answers. Interviewing on trains might also
lead to a bias since only demonstrators who travel by train have a chance to
be part of the sample. However, whether a demonstrator opts for the train
or another means of transport (or whether she or he simply lives close enough
to the venue) might also correlate with other sociodemographic and attitu-
dinal factors that could influence the representativity of the sample. For ex-
ample, we expect people traveling by train to be more likely to be members
| of an organization and to come from farther away (see chapter 9), which,

W consequently, might be an indicator of their high motivation. ~W~

In short, the differing sampling procedure in Italy might have substan-
tially affected the Italian sample. However, the team also distributed a lim-
ited number of postal questionnaires at the Rome demonstration. In fact, a
comparative analysis of differences between the train face-to-face surveys
and the Rome postal ones did not reveal any statistically significant differ-
ences between the two groups with regard to either their sociodemographic
characteristics or other variables. Thus, although it is still to some extent a
leap of faith, we are confident in including the Italian sample in our sys-
tematic comparison with the other countries. However, especially when it
comes to comparing the mobilization patterns of the Italians with those of
the Belgians, Germans, and so forth — as we do in chapter 9 of this vol-
ume — we think it is important to take the differential recruitment of re-
spondents into account and to be very cautious, as our sampling is quite
likely biased on the dependent variable (mobilization pattern).

We conducted surveys in eleven major cities in the eight covered coun-
tries. Apart from including the major demonstrations in the capitals of these
countries, in most cases (London, Rome, Madrid, Berlin, Amsterdam, Brus-
sels, and Bern) or in a major metropolis, in the United States (New York),
we also fielded our survey in Seattle and in San Francisco (United States) and

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page 2BS^_

APPENDIX A 28l

in Glasgow (United Kingdom). As we wanted to analyze these data at the
country level, the variation of the political and societal context is our main
independent variable. Therefore, we checked whether we could detect any
significant differences in the main variables between the different cities in
the United States and the United Kingdom. This proved not to be the case:
the Seattle demonstrators are very much like those in New York and San Fran-
cisco, and the London and Glasgow protesters hardly differed, either. That
is why we consider it safe to merge the respective city datasets into one sin-
gle set for the United States U.S. and one single set for the United Kingdom.
In most chapters we use these (merged) country datasets, but is some chap-
ters, for specific reasons, the authors split up the UK sample and deal with
the London and Glasgow demonstrations separately.

We relied on a double sampling strategy in some countries: we distrib-
uted postal questionnaires using our strict sampling method during the march
and asked people to send them back (we had paid the postage beforehand);
and we selected a smaller number of people, using on a less strict sampling
procedure, which gave the interviewers more leeway as they conducted a short
interviews on the spot. In this volume, we only use the completed postal
questionnaires. We used the face-to-face interviews to test the representativ-
| ity of the questionnaires that were sent back. As mentioned, we used face-

"W" to-face interviews in Belgium, the Netherlands, the United Kingdom, and ~®~

Switzerland. In Italy, almost all surveys were conducted face-to-face but since
that was been done in an entirely different way, we do not include the Ital-
ian data in our test of representativity.

What can be learned from the systematic comparison of both interview
types? Overall, the results tell us that there are, indeed, a few differences be-
tween the two interview types (even statistically significant ones) but that
these differences remain limited. They are, of course, not surprising. The
associations between the interview type and the responses can be traced to
three diverging methodological biases: mail survey non-response bias, face-to-
face survey selection bias, and face-to-face versus mail survey social desirabil-
ity effects. As we cannot determine for sure which method effect is causing
which kind of bias, we can only undertake an informed guess about what
causes these differences.

First, men and younger people are significantly overrepresented among
the face-to-face interviews. It seems that our interviewers, also young and
predominantly male, tended to select more male and younger conversation
partners, or, perhaps, that female and older respondents sent back the ques-
tionnaire more reliably. Thus, the observed gender and age differences be-
tween interview types might be due either to a response (mail) or selection

H$H

-4

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page 2B§^_

282 APPENDIX A

(face-to-face) bias. Second, some variables regarding the company in which
the interviewee attended the demonstration yielded significant differences,
all pointing toward a selection bias in the face-to-face interviewing: directly
interviewed people attended the demonstration to a lesser extent with family,
colleagues, and co-members than mail respondents did. It makes sense that
interviewers tended to address participants who came on their own rather than
those who were part of a group. Third, we noticed that the face-to-face inter-
viewees were more likely to agree with the Likert items they were confronted
with; this is commonly known as the acquiescence effect of face-to-face
;. This was especially observable for four statements in the ques-
e ("War is always wrong"; "War on Iraq is waged to secure national
oil supply"; "I am satisfied with my government's efforts to prevent war on
Iraq"; "War to bring down a dictatorial regime is justified"). Finally, we did
not find any significant differences between the two interview types for many
variables. There is no difference, for example, in the level of political interest,
previous type or frequency of protest participation, political left-right self-
positioning; perceived political efficacy of the demonstration. By and large,
we did not find a strong evidence of a systematic response bias in the mail
sample. However, we found traces of a probable selection bias and of a social
desirability effect in the face-to-face sample. Hence, we think we can rather
safely consider the mail survey respondents a fair sample of the population.
Similarly, Wolfgang Riidig, the leader of the UK team, has carried out an
extensive in-depth analysis comparing postal and face-to-face questionnaires
in the Scottish (Glasgow) subsample (2006). His results, also, show that there
are no substantial differences between the two interview types and reinforce
our confidence in the reliability of the mail survey data.

Response rates for the February 1 5 mail survey vary between countries,
but an overall response rate of almost 47 percent is clearly quite satisfying for a
postal survey for which no reminders were sent out. In previous protest surveys
covering a variety of demonstrations on very diverse sets of issues in Belgium,
we got, on average, a response rate of nearly 44 percent. Thus, the February 15
rates are by no means exceptional or disturbing for this kind of research. We
can say that peace demonstrators or anti-Iraq War demonstrators tend to send
back their questionnaires after they got home to about the same extent as
:i-drugs, global social justice, education, or social security dem-
. Figure A.l contains the February 15 response rates by country.

As we see, differences among countries appear substantial. In the
United Kingdom only a bit more than one in three selected demonstrators
sent back a completed questionnaire. At the same time, more than half of the
sampled protesters answered our questions in the Netherlands, Switzerland,

H$H

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page 2flafc)_

appendix a 283

Number of

Response rate

Number of

mail survey

for mail survey

face-to-face

respondents

(percent)

respondents

United States

705

47.0

_

United Kingdom

1,129

35.9

503

Spain

452

37.7

-

Italy

1,016

-

-

Netherlands

542

^4.2

100

Switzerland

by

53.1

181

Belgium

510

46.4

196

Germany

"81

52.1

-

Total

5,772

46.6

981

Figure A. 1. Responst

' rates in the eight

nations

-4

and Germany. At this point, we have neither a clue as to where these inter-
country differences come from nor a way to look systematically for an expla-
nation. It could well be that the sociodemographic composition of the marches
in the different countries led to different levels of conduciveness to this kind
of survey research; as shown in chapter 5, we did find significant differences
in the sociodemographic profile of the demonstrators across countries. Also,
the mood and radicality of the demonstrators and the political position of
government and opposition in the eight countries might have made a differ-
ence. More in general, political cultural differences between the countries
could have caused diverging response rates. Residents of the United Kingdom,
for example, might in general be less willing to participate in survey research
than Dutch citizens. Still, we believe that the response rates in the different
countries are similar enough that we can systematically compare the covered
demonstrators across nations.

Finally, we are aware that the method employed in this study raises
questions and that we might not have been able in this appendix to answer
all of them satisfyingly After all, survey techniques are a fairly new method
for the analysis of protest events, and they must be explored further and
tested more profoundly than this has been the case until now. The Belgian
team that developed the protest survey method is working, at present, ion
developing a much more thorough test of the reliability and representativity
of the fieldwork method. We hope to make further headway with this ap-
proach in the very near future.

Works Cited

Favre, Philippe, Olivier Fillieule, and Nonna Mayer. 1997. "La fin d'une etrange
lacune de la sociologie des mobilisations: L'etude par sondage des manifestants:

I

H$H

m

15 Appendix A_Walgrave 22/01/2010 3:44 PM Page 2B4^_

284 APPENDIX A

Fondements theoriques et solutions techniques." Revue Francaise de Science Poli-
tique 47: 3-28

Fillieule, Olivier. 1997. Strategies de la rue. Les manifestations en France. Paris: Presses
de Sciences Po.

Jasper, James M., and Jane Poulsen. 1995. "Recruiting Strangers and Friends: Moral
Shocks and Social Networks in Animal Rights and Anti-Nuclear Protests." Social
Problems 42: 493-512.

Norris, Pippa, Stefaan Walgrave, and Peter Van Aelst. 2005. "Who Demonstrates?
Anti-State Rebels or Conventional Participants? Or Everyone?" Comparative
Politics 2: 251-7 '5.

Ragin, Charles. 2004. "Turning the Tables: How Case-Oriented Research Challenges
Variable-Oriented Research." In Rethinking Sol erse Tools, Shared

Standards, ed. Henry E. Brady and David Collier, 123-38. Lanham, Md.: Row-
man & Littleheld.

Rudig, Wolfgang. 2006. "Assessing Nonresponse Bias in Activist Surveys: A Study
of 2003 Anti-Iraq War Protesters in Scotland." Unpublished paper, Glasgow.

Van Aelst, Peter, and Stefaan Walgrave. 2001. "Who Is That (Wo)man in the Street?
From the Normalisation of Protest to the Normalisation of the Protester." Euro-
pean Journal of Political Research 39: 461-86.

Waddington, David. 1988. Flashpoints of Public Disorder. London: Methuen.

H$H

6 Appendix B_Walgrave 22/01/2010 3:45 PM Page 2SS^_

m

oixb:

Media Content Analysis

Joris Verhulst

We decided to concentrate the media analysis on the two and a half months
prior to the break of war: from January 1 5 till March 2 1 , 2003. In that period,
the worldwide war debate was most vivid; it focused on the justification for
the imminent war and on the potential approval of war by the UN Security
Council. Also, the February 1 5 protest took place somewhere in the middle
of this time. This time span is well fit to test our propositions, since in
exactly these few months countries fought a bitter battle about the justifica-
tion of war, and most were forced to take a clear stance regarding the U.S. -UK
war plans. Therefore, we expect country variations to be high in this period.

For the media analysis we selected two or three newspapers per coun-
try. For each, we tried to include, if applicable, three articles per day, using
the broad search term "Iraq" in online search engines. In total, we used 3,968
news articles, which were encoded by trained researchers. We included
twenty-two different papers. Their attention to the Iraqi crisis differed, but
we had at our disposal sufficient articles per newspaper to carry out reliable
analyses for all twenty- two. Figure B. 1 lists the newspaper titles and the num-
bers of articles used for each.

The selection of so few newspapers per country probably is the trickiest
aspect of the research design. We always took the major left-land right-leaning
broadsheet newspapers, then, in most cases, we added the most popular news-
paper, often a tabloid. Are these two or three newspapers representative of
the written press in the eight countries? For some countries, we are rather
confident that they are; for others, we are much less sure. Our U.S. sample,
especially — with the New York Times, the Washington Post, and USA Today —

H$H

6 Appendix B_Walgrave 22/01/2010 3:45 PM Page 2BS^_

m

286 APPENDIX B

Left/right/
popular

Number of
articles

United States New York Times
Washington Post
USA Today

left
popular (right)

189

193

United Kindom Times
•
Sun

left
popular (left)

206
210

72

Spain ElMundo
El Pats

left

209

178

Italy // Giornale

La Reppublica
La Stampa

right
left

299
239

77

Netherlands NRC Handelsblad
De Volkskrant
De Telegraaf

left
popular (right)

96

402
141

Tages-Anzeiger

right (center)
left (center)

170
132

Belgium De Standaard
De Morgen
Het Laatste Nieuws

left
popular (right)

206
166

142

Germany Frankfurter Allgemeine
Bild

left
popular (right)

171
165

Total 3,968

Figure B. 1. Description of newspaper sample

might be not entirely representative. It might indeed be too liberal and left-
leaning to represent the average U.S. written press outlet. In Italy, we took
La Stampa as a third paper, which is not the most popular paper in the coun-
try and has a centrist position. We were not able to get electronic access to
the archives of the Switzerland's most popular tabloid, Blick. In Spain, the
most popular national newspaper is the high-quality El Pais, and the selec-
tion of a third, more tabloid-like, newspaper was hindered because of a lack
of electronic database access.

A final methodological remark concerning the media dataset relates to
the actual coding of the media content. Our encoders read each article in its
entirety and did not rely on electronic indexes provided by the news com-
panies. Our encoders were all Belgian graduate students able to speak the
language of the newspapers they read. Yet, because we encoded newspapers

H$H

6 Appendix B_Walgrave 22/01/2010 3:45 PM Page 2BS^_

APPENDIX B 287

in five different languages, overlaps between encoders could not be easily orga-
nized and intercoder-reliability measures could not systematically be calcu-
lated. To undercut possible problems, the graduate students were briefed
and trained extensively beforehand, and their work was thoroughly controlled
and checked by the supervisor of the encoding project.

m

h$h

16 Appendix B_Walgrave 22/01/2010 3:45 PM Page 2BSfc)_

1

H$H

17Contributors_Walgrave 22/01/2010 3:45 PM F

Contributors

w. lance bennett is professor of political science and Ruddick C. Lawrence
Professor of Communication at the University of Washington.

I michelle beyeler is lecturer and research fellow at the Institute of Politi-

~W cal Science, University of Berne, Switzerland. "W

christian breunig is assistant professor of political science at the Univer-
sity of Toronto.

Donatella della porta is professor of sociology in the Department of
Social and Political Sciences at the European University Institute, Florence.
She is coauthor of Globalization from Below: Transnational Activists and Protest
Networks (University of Minnesota Press, 2006).

mario diani is professor of sociology at the University of Trento in Italy.
He is coauthor of Social Movements: An Introduction (Blackwell, 1999).

terri e. givens is vice provost and associate professor in the
department at University of Texas, Austin. Her books include
Right in Western Europe (Cambridge University Press, 2005).

bert klandermans is professor of applied psychology at Free University in
the Netherlands. He is coauthor of Methods of Social Movement Research (Uni-
versity of Minnesota Press, 2002).

~^~

17Contributors_Walgrave 22/01/2010 3:45 PM Page_2^_

4

CONTRIBUTORS

dieter rucht is professor of civil society, conflict, and democracy at Wis-
senschaftszentrum Berlin fur Sozialforschung.

wolfgang rudig is reader in government at University of Strathclyde, Glas-
gow. His books include Anti-Nuclear Movements (Longman, 1990).

Sidney tarrow is Maxwell Upson Professor of Government and Sociology
at Cornell University. He is coeditor of Silence and Voice in the Study of Con-
tentious Politics (Cambridge University Press, 2001).

peter van aelst is assistant professor of political science at the University
of Antwerp.

joris verhulst is assistant professor of political science at the University
of Antwerp.

stefaan walgrave is professor of political science at the University of
Antwerp.

~^~

8 lndex_Walgrave 22/01/2010 3:45 PM Page 291 ^

Index

4

-^-

19 Series page (continued)_Walgrave 22/01/2010 3:45_E&L_Eage 1

Series page continued from p. ii.

Volume 23 David S. Meyer, Valerie Jenness, and Helen Ingram, editors,

Routing the Opposition: Social Movements, Public Policy, and
Democracy

Volume 22 Kurt S chock, Unarmed Insurrections: People Power Movements

in Nondemocracies

Volume 21 Christian Davenport, Hank Johnston, and Carol Mueller,
editors, Repression and Mobilization

Volume 20 Nicole C. Raeburn, Changing Corporate America from Inside
Out: Lesbian and Gay Workplace Rights

Volume 1 9 Vincent J. Roscigno and William F. Danaher, The Voice of

Southern Labor: Radio, Music, and Textile Strikes, 1929-1934

Volume 1 8 Maryjane Osa, Solidarity and Contention: Networks of Polish
Opposition

~vlgr Volume 1 7 Mary Margaret Fonow, Union Women: Forging Feminism in the "vw~

T United Steelworkers of America ' "

Volume 1 6 Bert Klandermans and Suzanne Staggenborg, editors, Methods

of Social Movement Research

Volume 15 Sharon Kurtz, Workplace Justice: Organizing Multi-Identity
Movements

Volume 14 Sanjeev Khagram, James V Riker, and Kathryn Sikkink,
editors, Restructuring World Politico Tia io So
Movements, Networks, and Norms

Volume 1 3 Sheldon Stryker, Timothy J. Owens, and Robert W. White,
editors, Self, Identity, and Social Movements

Volume 1 2 Byron A. Miller, Geography and Social Movements: Comparing
Ann nuclear Activism in the Boston Area

Volume 1 1 Mona N. Younis, Liberation and Democratization: The South
African and Palt-.' iia> Vaiional Movements

H$H

19 Series page (continued)_Walgrave 22/01/2010 3:45_E&L_Eage 2

Volume 1 Marco Giugni, Doug McAdam, and Charles Tilly, editors,
How Social Movements Matter

Volume 9 Cynthia Irvin, Militant Nationalism: Between Movement and
Party in Ireland and the Basque Country

Volume 8 Raka Ray, Fields of Protest: Women's Movements in India

Volume 7 Michael P. Hanagan, Leslie Page Moch, and Wayne te Brake,
editors, Challenging Authority: The Historical Study of
Contentious Politics

Volume 6 Donatella della Porta and Herbert Reiter, editors, Policing Protest:
The Control of Mass Demonstrations in Western Democracies

Volume 5 Marco G. Giugni, Doug McAdam, and Charles Tilly, editors,
New Social Movements in Western Europe: A Comparative Analysis

Volume 4 Hank Johnston and Bert Klandermans, editors, Social Movements
and Culture

M- Volume 3 J. Craig Jenkins and Bert Klandermans, editors, The Politics of W

' Social Protest: Comparative Perspectives on States and Social

Movements

Volume 2 John Foran, editor, A Century of Revolution: Social Movements

Volume l Andrew Szasz, EcoPopulism: Toxic Waste and the Move in* ntfoi

Environmental justice

~^~