After International Relations

After International Relations articulates a systematic critical realist response to a

quest for more emancipatory methodologies in international relations. Heikki
Patomäki here establishes a way out of the international relations problematic
that has puzzled so many great thinkers and scholars for the last two hundred
years. After International Relations shows how and why theories based on the
international problematic have failed; expresses an alternative, critical realist
research programme; and illustrates how this research programme can be put
to work to enable better research and ethico-political practices.
Developing a critical realist methodology for international relations, peace
research and global political economy, this book resolves many of the theoretical
aporias of international relations. It explains in detail the way research should
be conducted in open systems and pluralistic contexts. It refines the logic of
explanatory emancipation beyond the confines of nation states, and responds
to the problem of violence. It goes further than any existing text in developing
critical theoretical methodologies in international relations, and showing how
they can be applied in research and acted upon in ethico-political practice.
The book begins with a critical genealogy of the emergence of the international
relations problematic—the irrealist foundations of which were laid down by David
Hume and Immanuel Kant. After spelling out the ontological and epistemological
underpinnings of the critical realist alternative, it explores the future role of
critical realism in the construction of non-violent and democratic world politics,
which overcomes the Kantian antimonies and dilemmas of the international
problematic. It will be of great interest to all those involved in international
relations, peace research and international political economy

Heikki Patomäki is Professor of World Politics and Economy at

Nottingham Trent University, and also the Research Director of the
Network Institute for Global Democratisation. His most recent books
include Democratising Globalisation: The Leverage of the Tobin Tax and The Politics
of Economic and Monetary Union (with [Link]).
Critical realism: interventions
Margaret Archer, Roy Bhaskar, Andrew Collier,
Tony Lawson and Alan Norrie

Critical realism is one of the most influential new developments in the philosophy of
science and in the social sciences, providing a powerful alternative to positivism and
post-modernism. This series will explore the critical realist position in philosophy
and across the social sciences.

Critical Realism
Essential readings
Edited by Margaret Archer, Roy Bhaskar, Andrew Collier, Tony Lawson and Alan Norrie
The Possibility of Naturalism, 3rd edition
A Philosophical Critique of the Contemporary Human Sciences
Roy Bhaskar

Being & Worth

Andrew Collier
Quantum Theory and the Flight from Realism
Philosophical Responses to Quantum Mechanics
Christopher Norris
From East to West
Odyssey of a Soul
Roy Bhaskar
Realism and Racism
Concepts of Race in Sociological Research
Bob Carter
Rational Choice Theory
Resisting Colonisation
Edited by Margaret Archer and Jonathan [Link]
Explaining Society
Critical Realism in the Social Sciences
Berth Danermark, Mats Ekström, Jan Ch Karlsson and Liselotte Jakobsen
Critical Realism and Marxism
Edited by Andrew Brown, Steve Fleetwood and John Michael Roberts
Critical Realism in Economics
Edited by Steve Fleetwood
Realist Perspectives on Management and Organisations
Edited by Stephen Ackroyd and Steve Fleetwood
Also published by Routledge

Routledge studies in critical realism

Edited by Margaret Archer, Roy Bhaskar,
Andrew Collier, Tony Lawson and Alan Norrie

1. Marxism and Realism

A Materialistic Application of Realism in the Social Science
Sean Creaven
2. Beyond Relativism
Raymond Boudon, Cognitive Rationality and Critical Realism
Cynthia Lins Hamlin
After International
Relations
Critical realism and the
(re)construction of world politics

Heikki Patomäki

London and New York

First published 2002
by Routledge
11 New Fetter Lane, London EC4P 4EE

Simultaneously published in the USA and Canada

by Routledge
29 West 35th Street, New York, NY 10001

Routledge is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2003.

© 2002 Heikki Patomäki

All rights reserved. No part of this book may be reprinted

or reproduced or utilised in any form or by any electronic,
mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data

Patomäki, Heikki.
After international relations: critical realism and the
(re)construction of world politics/Heikki Patomäki.
p. cm.—(Critical realism—interventions)
Includes bibliographical references and index.
1. International relations—Social aspects.
2. International relations—Methodology.
3. Critical realism. 4. World politics.
I. Title. II. Series.
JZ1249.P38 2002
327.1'01–dc21 2001048310

ISBN 0-203-11903-7 Master e-book ISBN

ISBN 0-203-16339-7 (Adobe eReader Format)

ISBN 0-415-25659-3 (hbk)
ISBN 0-415-25660-7 (pbk)
Contents

List of illustrations ix
Acknowledgements xi

Introduction 1

PART I
Overcoming international relations 19

1 How it all started? Can it end? A genealogy

of the international problematic 21

2 Learning from Alker: From quantitative peace

research to emancipatory humanism 43

3 How to tell better stories about world politics 70

PART II
Explicating a critical realist methodology 97

4 Critical realist social ontology 99

5 Double hermeneutics of iconic modelling 123

6 The normative logic of emancipation 143

PART III
Visions of world politics 165

7 Modelling Thucydides’ Melos episode 167

8 A global security community 193

vii
viii Contents

9 From security to emancipatory world

politics: Beyond ‘Nordic nostalgia’ 210

Epilogue 237

Bibliography 239
Index 257
Illustrations

Tables

1.1 The repetitious cognitive structure of international relations 38

7.1 The first seven moves of the Melian dialogue 175
7.2 The first seven moves quasi-formalised 176
7.3 Relevant causal components of the Melos episode 181
8.1 Amalgamation and integration 195
9.1 The Cold War interpretation of Norden 212
9.2 A political-economy interpretation of the Swedish model 216
9.3 Central government budget, total expenditure, as percentage
of the GDP 223

Figures

1.1 The solution set of the international problematic 37

1.2 Three Humes and two Kants 38
4.1 An example of a dialectical contradiction 107
5.1 The ampliative approach 124
5.2 The reductive approach 125
5.3 A taxonomy of iconic models 127
6.1 Stratification of agency and action 152
7.1 The strategic decision-making tree of Melos and Athens 171
7.2 A graphical presentation of the rest of the Melian dialogue 177
8.1 Generation of a security community 204
9.1 Structuration of the conditions of the decline of the Swedish model
1976–89, in the context of stagflation and indebtment 219

ix
Acknowledgements

During the good ten years that this book has been in the making, I have
accumulated debt to many persons. Any reader will soon notice the central role
of Hayward [Link]. He has not only inspired me, but also commented on
most parts of this book, on various occasions. Special thanks are also due to
Tuomas Forsberg and Colin Wight, with whom I have worked closely together,
and who are doing critical realism in their own ways. Collaboration with them
has sparked many stimulating discussions and debates. Also the following
colleagues have provided useful comments and irreplaceable support (I apologise
for any possible mistaken omissions): Jason Abbott, Osmo Apunen, Chris Brown,
Walter Carlsnaes, Stephen Chan, Mick Dillon, Chris Farrands, [Link],
Stefano Guzzini, Henrikki Heikka, Mark Hoffman, Jef Huysmans, Pertti
Joenniemi, Eerik Lagerspetz, Robert Latham, Richard Little, Iver [Link],
Hannu Nurmi, Lloyd Pettiford, Juha Räikkä, Magnus Ryner, Hidemi Suganami,
Teivo Teivainen, Teija Tiilikainen, Ole Wæver, Rob Walker (who has also
instigated the title of this book), Keith Webb, Alexander Wendt, Matti Wiberg
and Andy Williams. In some cases, the most valuable comments have come
from the persons most critical of my project!
Chapter 2 was previously published as ‘Hayward Alker: An exemplary voyage
from quantitative peace research to humanistic, late-modern globalism’, in I.
Neumann and O.Wæver (eds) (1997) The Future of International Relations. Masters
in the Making?, London: Routledge, pp. 205–35. Chapter 3 was previously
published as ‘How to tell better stories about world politics’, European Journal of
International Relations 2(1):105–33, reprinted by permission of Sage Publications
Ltd (© Sage Publications Ltd and the European Consortium for Political
Research (ECPR) Standing Group on International Relations 1996). Chapter 4
was published previously as ‘Concepts of “action”, “structure” and “power”’ in
“critical social realism”: A positive and reconstructive critique’, in the Journal for
the Theory of Social Behaviour (1991) 21(2):221–250, reprinted with permission.
Chapter 6 was published previously as ‘Scientific realism, human emancipation
and non-violent political action’, Gandhi Marg (1992) 14(1):30–51, reprinted
with permission. Chapter 7 was first published in Finnish as ‘Mitä Thukydideen
Melos-dialogi voi kertoa?’ [What Can Thucydides’ Melos-Dialogue Reveal?],
Rauhantutkimus (1990), 20(4):26–63; used here with permission. Finally, Chapter
xi
xii Acknowledgements

9 was published as ‘Beyond Nordic nostalgia: Envisaging a social/democratic

system of global governance’, Cooperation & Conflict (2000) 35(2):115–54, reprinted
by permission of Sage Publications Ltd (© 2000 NISA). All chapters have
been, however, rewritten for the purpose of this book. New sections have replaced
many previous ones, and, particularly, the shape of the pre-1997 papers has
been completely refashioned.
Pauline Eadie has helped to put this book together in a clean-cut Routledge
format. Many thanks for your persistence! Katarina has encouraged me greatly
and helped to simplify and clarify some parts of the text. Most importantly, she
has remained patient during the intensive months of rewriting the chapters and
editing the manuscript—despite the demands of our two small daughters, Anna
and Oona.
I dedicate this book to Katarina and her remarkable recovery from a very
serious illness.
Heikki Patomäki
Nottingham
21 May 2001
Introduction

What are international relations? This book argues that all social relations are
changeable. They are dependent on transient concepts, practices and processes.
What has once emerged can change and even become absent. So it is also with
international relations.
There is no closed list of transformative possibilities. After International Relations
claims that critical realist (CR) social sciences can help determine some of them.
Social sciences should proceed in three successive stages. First, in order to explain,
researchers have to engage with social meanings and assume a dialogical relationship
with relevant actors or traces and evidence of history, including texts and quantified
observations. Only with dialogically understood data can social episodes, practices,
relations and processes be studied adequately and in sufficient detail.
Mere understanding is not enough. Causally explanatory models can and
should rely also on holistic metaphors, historical and other analogies, and various
social scientific theories. The best available explanatory model is often different
from prevalent understandings. Consequently, in the second stage, explanatory
models may not only turn out to be critical of the understandings of the lay
actors but also take steps towards explaining them.
The critical moment evokes a need for social scientific re-signification of
practices, including edification by means of better causal stories and proposals
for concrete alternatives. If a change is argued for, as it often is, realist strategies
for transformation have to be spelled out as well. Constant openness to critique
and further studies is a condition of validity. In the third, and practical moment,
CR explanations should translate into practical action—often transformative.
Also, practical-political action should be a learning process, in turn enriching
theoretical discourses.
CR has emerged in a particular world historical context. As some
constructivists and post-structuralists argue, there are signs of far-reaching
transformations of international relations into something more akin to world
politics (cf. Ruggie 1998:174–75, 195–97; and, more explicitly, Walker 1993:183).
After International Relations makes a case for CR social sciences which are not
content with describing transformations but try to understand, explain and
critically assess contemporary practices, relations and processes, and, thereby,
contribute to making them more emancipatory and empowering.
1
2 Introduction

International relations

International relations emerged step by step, element by element, in Europe by

the early nineteenth century. Christianity was divided; rationalism emerged
and capitalism and sovereign states developed side by side. In the context of the
expansion of Europe and capitalist world economy, the international institutions
and problematic surfaced gradually. Some components of the inter-state order
were codified in the Treaty of Westphalia in 1648, some others in the Peace of
Utrecht in 1713. The term ‘inter-national’ first appeared in 1789 in two important
texts. The French Revolution Declaration of the Rights of Man pronounced that
‘the source of all sovereignty is located in essence in the nation’ (Beik 1970:95).
This manifesto was certainly more visible than Jeremy Bentham’s footnote in
Principles of Morals and Legislation, and it signalled the end of the inter-dynastic
and the beginning of international relations. What is important here, however,
is that Bentham’s discussion of ‘international’ in a footnote can be seen as
evidence that the notion had started to gain ground already before the events of
1789 (DerDerian 1989:3).
I argue in Chapter 1 that, although Immanuel Kant did not yet use the term
‘international’, the international problematic was articulated systematically by him
in the 1780s and 1790s.1 Kant’s metaphysical or political writings would not
have been possible without David Hume. With his immediate predecessors and
contemporaries, from Thomas Hobbes and John Locke to René Descartes and
George Berkeley, Hume laid down the basis for this problem field in the mid-
eighteenth century. A few decades later, Kant, a central Enlightenment
philosopher, together with his immediate predecessor Jean Jacques Rousseau,
articulated these conceptions into an acute social problem of war and peace.
Kant thought this problem can and should be overcome by an arrangement of
‘league of nations’, free trade, rule of law and republicanism. My emphasis on
Hume and Kant does not mean that I believe in the great men theory of history
of thought. Both Hume and Kant should be understood as participants among
many in intellectual debates of their time. Sociologically, they can be seen as
intellectuals of emerging social forces.
In this sense, Kant is the beginning of a familiar story. Since the mid-nineteenth
century, political realists have criticised Kantian ‘idealists’ or liberalists for
advocating dangerous actions based on Utopianism. The nature of human beings
or the world will make all international reforms futile, perverse or even dangerous
(of the ‘rhetoric of reaction’, see Hirschman 1991). Politics is plagued with
unintended consequences; life is often tragic. Wise statesmen know how futile
or perverse reforms tend to be and, more critically, how dangerous the world
so often is. Hence, the real statesmen, with ‘politics as their vocation’, assume
responsibility for the consequences of their actions within the given order of
things (Weber 1985 [1919]; see also Smith 1986:15–53). Morgenthau (1946:199)
blamed Kant for the general decay in the political thinking of the Western
world, and, in effect, for the catastrophes of the twentieth century. He argued
that the best way to preserve peace is to prepare for war; balance powers by
 Introduction 3

making good alliances; and cultivate diplomacy (Morgenthau 1946: Chapters

VII and VIII). Order and stability are the core realist values for prudent foreign
policy. When some time had passed since the dark days of the world wars, and
the twenty years of crises between them, political realists began to recognise
that international economic co-operation is possible, but usually only if it accords
with the interests of the great powers. A new factor may nonetheless have
surfaced. International regimes can be seen as ‘intervening variables’ between
structure and great power interests (Krasner 1982).
Liberals argue, however, that on the basis of late twentieth-century experiences,
there is in fact much more room for ideas, reforms and peaceful change than
political realists think. The United Nations can—and sometimes does—play an
important role in facilitating agreements and co-operation, and managing conflicts
(Haas 1986; Rochester 1995). Although built by two great powers, the USA
and Britain, ideas and transnational coalitions have shaped the Bretton Woods
institutions (Ikenberry 1992). Although it was the dissident movements that
eventually broke communist rule in Eastern Europe, the driving force behind
the end of the Cold War was the learning process of Mikhail Gorbachov and
his inner circle, and, consequendy, the new Soviet foreign policy (Risse-Kappen
1994; Evangelista 1995, 1999). New kinds of important actors have emerged
everywhere, from NGOs to multinational corporations. Liberal-minded reformers
argue, also, that it is possible to further strengthen international norms, rules
and principles, at least under the management of relevant great powers.
Many mainstream liberals maintain, however, that the real hope for a more
peaceful world lies in the fact that liberal democracies (or liberal states) do not
fight each other. This is the almost only significant empirical invariance that has
been, allegedly, found thus far, and it seems to confirm the Kantian theory of
peace. The advocates of democratic peace theory (Doyle 1995) maintain that
since liberal democratic states do not fight each other, a zone of peace has
emerged. This would seem to give a recipe for a peaceful world. The zone of
democratic peace can be extended. Political realism is applicable only to the
relations between the liberal democratic states and the others. In the future, by
making all states liberal democratic, preferably prudently, a perpetual peace
may well emerge.
What unites Hume, Kant and most contributors to debates within the
international problematic is their commitment to positivism. By positivism I
mean the set of abstract and closely inter-related ideas that causality is about
constant conjunctions (‘whenever A, B follows’); that the properties of entities
are independent; that their relations are external or non-necessary; that the
basic things of the world are therefore atomist, or at least constant in their inner
structure; and that being can be defined in terms of our perceptions or knowledge
of it. The main discussions in international relations have presupposed the
international problematic in the orthodox positivist fashion. Since the 1960s,
positivism has assumed the form of ‘neo-utilitarianism’ in the USA-centric
discipline of international relations (Ruggie 1998:9). Neo-litarians represent
economics as the exemplary social science and believe in the claim of Gary
4 Introduction

Becker (1976) that the rational-choice approach is applicable to all human

behaviours (for a systematic CR criticism, see Archer and Tritter 2000). Neo-
realists may disagree with the idea of free traders, that there is a natural harmony
in the capitalist world economy, but like mainstream economists they believe in
positivism and the self-interested instrumental rationality of the states. However,
some regime theorists have tried to show that regimes of co-operation can be
more just than ‘intermediate variables’; also, intellectual and entrepreneurial
leadership seem to have a role in establishing new forms of economic or
environmental co-operation (Young 1989, 1991)
For more than a decade now, the positivist orthodoxy has been under a
sustained challenge from critical theories, feminism and post-modernism. Often,
all parties in this debate are ill-defined and misrepresented by the others. Indeed,
mis-characterisation would often seem to be the essence of debate. For many
post-positivists, positivism is not only epistemologically and ontologically flawed,
it is also co-responsible for many of the social ills and political catastrophes of
the modern world. Yet, for many positivists (empiricists, scientists, believers in
the possibility of applying the rigorous standards of natural sciences also in
social sciences) the post-positivist assault amounts to advocating subjectivism,
irresponsible relativism and lack of standards, which work against conducting
proper research and the effort to make human conditions better.
Post-positivism presupposes positivism—and many of its assumptions. For
example, post-positivists think that Hume and his followers have been right
about natural sciences and their objects. They mistake mechanics for natural
sciences and constant conjunctions for causality. Whereas positivists have tried
to reduce being to perceptions and logical sense-relations in language, post-
positivists are susceptible to linguistic fallacy, according to which it is ‘our
language’ that constructs being. The linguistic fallacy gives rise to manifold
confusions between subject and object. Moreover, the more radical post-positivists
have tended to be as sceptical about values as Hume. Only recently have the
post-structuralists come up with any ethical notions (Campbell and Shapiro
1999). The more moderate constructivists share the reformism of the Kantian
liberals (Ruggie 1998; Wendt 1999). Many of these more radical Kantians
maintain, however, that we should continue to imagine and construct more far-
reaching institutional alternatives for the future (Falk 1975; Galtung 1996; Held
1995; Holden 2000). However, the clear majority of scholars seem to feel that
these ideals and blueprints appear detached from realities (if worth discussing at
all). As ever, Kantian moral reason seems cut off from the world of causes and
effects.2
Since the international problematic is based on irrealist and false
understandings, illusions and mystifications, CR can help to overcome the
international problematic. However, although it is a central idea of CR that
theoretical discourses are located in the world of causes and effects, and are
potentially efficacious, the practical effects of CR must be seen as contingent.
Asymmetric power relations play a crucial role in sustaining a structure of
(contradictory) meanings and related problematic. There is no automatic
 Introduction 5

correspondence between any particular theoretical discourse and the social worlds
they refer to and purport to understand, explain and, sometimes, change.
Nonetheless, since actions and practices are dependent on and informed by
concepts, theoretical innovations are potentially efficacious in changing realities.
Practices can be constituted or guided by misleading or partially false or
contradictory understandings. More adequate concepts—if given a chance—should
therefore enable and also encourage changes of practices, institutions and deep-
seated social relations. It is certainly true that achieving well-founded conceptual
changes require no less hard work than practical political activities.

Critical realism

Developments

Critical realism was preceded by realist developments in the philosophy of science

in the 1960s and early 1970s (Bunge 1959, 1963; Harré 1961; Harré and
Madden 1975; Hesse 1966; Bhaskar 1978 [1975]). In the 1970s, realist ideas
were brought also into social sciences; in many cases this was also the original
motivation to re-think natural sciences. Harré and Secord published their ground-
breaking The Explanation of Social Behaviour in 1972. Although this book is sceptical
of large-scale sociological theory,3 it laid the ground for Russell Keat and John
Urry’s Social Theory as Science (1975) and for some of the basic ideas of Anthony
Giddens’s New Rules of Sociological Method (1976) and Central Problems of Social
Theory (1979). These works developed a realist alternative to both positivism
and post-positivism. Although it is true that lay meanings are constitutive of
social worlds, realist social sciences do not have to give up the inter-related
notions of causality, relations and structures. Roy Bhaskar’s seminal The Possibility
of Naturalism (1979) built up on the themes of Keat and Urry, and developed an
original account of agency and structure on a par with, but also in contrast to,
Giddens’s (1976, 1979, 1981) theory of structuration.
The term ‘critical realism’ was coined in the late 1980s (Sylvan and Glassner
1985; Bhaskar 1989:180–92; Isaac 1990; Fay 1990). Why was this movement,
which also emerged in the late 1980s, associated so strongly with the works of
Roy Bhaskar? Under the influence of Harré, Bhaskar’s early books were well
written and made their case powerfully. Other early contributors denied the
possibility of large-scale social theory, political science or political economy
(Harré), or tended to leave realist themes aside in their later works (Keat and
Urry) or never articulated any form of realism in full (Giddens). Whereas Keat
(1981), for instance, turned to Habermas to find conceptual grounds for social
and political critique, Bhaskar outlined an original realist theory of emancipation
in The Possibility of Naturalism (1979:76–91), and developed this theory further in
Scientific Realism and Human Emancipation (1986:180–223).
Bhaskar also represented CR as the missing methodology of Karl Marx. For
many this made Bhaskar suspicious. In the 1980s, the bulk of social theorists
6 Introduction

were following Giddens, who was writing a multi-volume critique of ‘contemporary

historical materialism’ (see Giddens 1981, 1985). In some circles of British
Marxism, however, Bhaskar became a guru to be followed. In contrast to Bhaskar
(1979:83–91), I do not think it is plausible to read either the early or later Marx
as a critical realist. However realist Marx may have appeared to be about structures,
relations and laws, many of the conditions for CR did not exist in the nineteenth
century (about these conditions, see Bhaskar 1991:139–41). The non-realist
conceptual resources on which he drew substantially mis-represented Marx.4
Moreover, according to Manicas’s (1987) realist re-interpretation of the history of
social sciences, Marx not only lacked an adequate theory of science; he also
lacked a theory of state.5 Certainly, there were very few—if any—signs of
epistemological relativism in Marx’s often polemical texts.
More than Marx himself, it seems that the early Frankfurt School anticipated
CR. For instance, Max Horkheimer’s (1989 [1937]) classical essay on ‘traditional
and critical theory’ criticised orthodox Marxism alongside Western positivism
and capitalism; revived Kantian critical questions about the presuppositions of
our knowledge; made a rudimentary use of reformulated Kantian transcendental
arguments and a realist notion of causality; and envisaged a democratic theory
of emancipation.6 These are also central themes of CR. Indeed, Bhaskar
(1993:352) himself points out that his favourite twentieth-century Marxist is
Theodor Adorno, Horkheimer’s long-standing colleague and occasional co-
author.
Alas, since the mid-1980s, Bhaskar’s so-called Marxism notwithstanding, and
in some circles because of it, CR has been associated particularly strongly with
his works. Scientific Realism and Human Emancipation was followed by an introductory
collection of Bhaskar’s papers (Reclaiming Reality, 1989) and an attempt to make
an intervention in contemporary philosophical debates in the form of a book-
length critique of Richard Rorty (Philosophy and the Idea of Freedom, 1991). Neither
book developed CR much further. Rather, they aimed at making CR better
heard and more influential. Bhaskar’s dialectical (Bhaskar 1993; 1994) and
transcendental turns (Bhaskar 2000) may have had the contrary effect.
Dialectic: The Pulse of Freedom (1993) and its somewhat more accessible
accompaniment, Plato etc.: Problems of Philosophy and Their Resolution (1994) are
complex, difficult and, at times, esoteric or pompous works. Bhaskar’s
philosophical discourse is tacitly assuming new responsibilities and powers in
‘dialectical CR’. Whereas the earlier form of CR was a philosophy for and of
science, Dialectic is not only about enabling and transforming (critical social)
sciences, but also about ‘a general theory of dialectic’. Dialectic is brilliant in its
analysis of Aristotle, [Link] and Marx, among others, yet it seems that it
also attempts to develop a general theory of everything, particularly world
history, freedom and good society, but also of the constitution of the universe.
In another reading, however, one might find that Dialectic is a result of a systematic
de- and reconstruction of Western metaphysics, which naturally leads to ‘an
extreme complication, multiplication, explication of “precise and rigorous
distinctions”’ (Derrida 1988:128).
 Introduction 7

Plato etc.: The Problems of Philosophy and Their Resolution (1994) is represented as an
attempt to popularise Dialectic, although it has a slightly different focus and somewhat
less speculative appearance. Like Dialectic, it has fascinating and very useful discussions
on causality, time, space and causality/freedom, taking the notions developed in
earlier works further. The emphasis is on a totalising critique of Western philosophy.
Not unlike Derrida’s criticism of Western logo-centrism, Bhaskar claims that, since
Plato, the ontological monovalence7—including the absence of notions such as
‘absence’, depth and change—have generated a philosophical problematic that, from
a consistent CR point of view, is nonsensical. Yet this philosophical problematic—
also through the structures of positivism, which are causally efficacious in many
modern practices—has had far-reaching consequences in the structuration of societies.
Bhaskar asserts that ‘the net project of Dialectic and Plato etc. is to attempt an anti-
Parmidean revolution, reversing 2500 years of philosophical thought, in which
negativity (absence and change), ontology, structure, diversity and agency come to
the fore’.8
Perhaps quite logically then, Plato etc. was followed by Bhaskar’s post-colonial
turn to the East and, as it seems, partial retreat to mystical thinking. From East
to West, published in 2000, ‘aspires to begin to construct a dialogue, bridge and
synthesis’ between traditions of both radical libertarian Western thought and
mystical Eastern thought. Many fellow critical realists have found the outcome
to be a philosophically and historically weak attempt to bring in popular New
Age ‘philosophies’ (e.g. Hostettler and Norrie 2000). Despite the importance of
the project, it also seems to contradict many CR theses, including the idea that
philosophy must be broadly consistent with the substantive contents and
methodologies of the sciences (see Patomäki 2001c, for a discussion of, and
comparison to, the somewhat similar attempts by Friedrich Nietzsche, Jacques
Derrida and Johan Galtung to appropriate notions from Buddhism and develop
dialogical notions of global philosophies).
While Bhaskar has been either adding new layers and aspects to CR, or
testing and exceeding its limits (depending on one’s view on his later works),
scientific realism has gained ground in analytical philosophy (e.g. Margolis 1986;
Searle 1996). CR itself has been cultivated among a group of social, political,
legal and economic theorists. These theorists include Margaret Archer (1988,
1995, 2000), Andrew Collier (1994, 1999), Tony Lawson (1997), Alan Norrie
(1991, 1993, 2000) and Andrew Sayer (1992, 1995, 2000). Besides introductions
to CR, and applications of CR to the problem fields of various disciplines,
many of these works have developed original positions on a number of central
philosophical and social theoretical issues. It is also significant that there are
parallels to CR developments in continental Europe, in particular in France. At
times, Derrida (cf. 1998 [1967]:50; 1988:114–54) and Paul Ricoeur (1981:274–
96; 1984:78; 1992:300), for instance, seem to have been approaching a position
close to CR. However, it is Pierre Bourdieu (e.g. 1977; 1998:1–13), in particular,
who has been working, since the 1970s, on a social theory very similar to CR.
This compatibility has been increasingly acknowledged by people working on
CR social theory (see, for example, May 2000).
8 Introduction

Philosophical theses

In this book, I identify CR with three philosophical theses and a related theory
of emancipation. The three philosophical theses concern (1) ontological realism,
(2) epistemological relativism and (3) judgemental rationalism. Ontological realism means
that the world is real and—except for a very small part—independent of the
researcher’s (my or our) knowledge of it. Bhaskar (1986:7) argues that ‘every
philosophy, inasmuch as it takes science for its topic, is essentially a realism, or
at least has realism for its principle, the pertinent questions being on how far and
in what form this principle is actually implemented’.
CR claims to have taken the principle of realism further than any Western
philosophical system thus far. The claim is that the world is not only real but it
must also be differentiated, structured, layered and possess causal powers, otherwise
our knowledge of it—or our being—would not be possible. Natural science is a
matter of the conceptual and practical work of scientists, who create artificial closures
in laboratories in order to understand and explain mechanisms, which can be
causally efficacious across different factual contexts. With few exceptions, there are
no closed systems in nature or society. Therefore, a realist conception of causality
does not equate causality with constant conjunctions but with structured powers
capable of producing particular, characteristic effects if triggered.
In this regard, natural and social sciences are similar. There are real causal
powers both in natural and social systems, whether actualised or not. CR
encourages research to focus on identifying and analysing the structures of beings
and mechanisms (or complexes), which are causally efficacious. Moreover, it is
also a condition of science that scientists and other social actors themselves possess
causal powers. A corollary is that, although not all reasons are causally efficacious,
and actors can themselves be mistaken about the effective reasons for their actions
and practices, some reasons must be efficacious in generating actions.
The idea that reality is layered, and that different, emergent layers (such as
biological, human and social) are ontologically irreducible and causally efficacious
implies a further ontological notion. Absence and emergence must also be real. Not
everything that is has always existed. Whatever exists can become absent—and
already is absent in relation to most fields and spaces most of the time anyway.
A parallel of the notions of absence and emergence is that change is not only
possible but also ubiquitous. This is one of the senses in which CR is a dialectical
system of thought and action.
CR has thus cultivated ontological realism. The second main feature of CR is
its epistemological relativism. Epistemological relativism connotes that all beliefs
and knowledge claims are socially produced, contextual and fallible. CR knowledge
is structured and layered. There are strongly universalising philosophical claims
such as those for ontological realism and epistemological relativism; there are
rather stable scientific theories; and there are reflective interpretations of social
meanings and episodes. However, even the deepest and most universal
philosophical theses must be, in principle, open to criticism and change. Openness
and transfer inability are essential to any research. The work of scientists is based
 Introduction 9

on the pre-existing means of production, including both the available material facilities
and technologies as well as the antecedently established facts and theories, models
and paradigms. Models and paradigms must rely on analogies and metaphors.
Therefore, a rationally developing science uses both historically transient conceptual
resources and imagination in a systematic and controlled fashion. However, in the
social sciences, laboratory experiments and precise quantitative measurements
are impossible. Social sciences are conditioned by the fact that they have to gather
all their empirical evidence from open, historical systems. Moreover, social sciences
also involve attempts to understand lay meanings. Thus, social scientific models
and theories must be more reflectively interpretative undertakings than natural
sciences.
Epistemological relativism is consistent with the idea of rational judgements.
Judgemental rationalism means that, in spite of interpretative pluralism, it is possible
to build well-grounded models and make plausible judgements about their truth. CR
gives, in fact, the best available account of the grounds for scientific activities. It
explains the nature of the objects we study. It explains what we as researchers do and
why—and also why we should aim at amending and revising our theories and models.
It is easy to see, for instance, that ontological realism and epistemological relativism
are conditions for rational truth-judgements. On the one hand, without independent
reality, truth-claims would be meaningless. Truth-claims have to be about something. On
the other hand, with our theories already mirroring reality, no further research or
truth-judgements would be needed. Besides making sense of the rationale for research,
CR enables us to acknowledge that claims to truth have both antecedent conditions
and consequent effects. Some of these effects can be seen, in a Foucauldian manner,
albeit in realist terms, as effects of power.
A theory of emancipation is also an essential component of CR. The core of
this theory is the idea that lay meanings are both necessary for social practices
and relations, and can be criticised. This pertains to why this particular form of
scientific realism is called ‘critical’.

In what sense critical?

In what sense is CR ‘critical’? Kant’s philosophy was called critical, because Kant
problematised and analysed the presuppositions of our knowledge. Bhaskar
(1986:22) mentions that CR is post-critical in its form. That is, CR relies upon
transcendental arguments, yet it analyses historical practices and not universal
categories of human mind. CR is critical, first, in this post-Kantian sense. It asks
Kantian questions and deploys transcendental arguments. However, in contrast
to Kant, CR uses these arguments to establish a realist ontology. Further, in
contrast to naïve realism, CR sees beings as relational and changing, and, also,
argues that there is no unmediated access to these beings.
The second sense in which CR is ‘critical’ is related to the distinction between
ontology and epistemology. No knowledge simply reproduces reality or ‘mirrors’ it.
Correspondence in this sense would not make any sense (Sayer 2000:42). Any
10 Introduction

attempt to describe and explain involves complex mediations; that is, interpretation.
CR is a form of reflective philosophy, self-consciously interpretative and relativist.
The starting point for Kant’s epistemological reflections was an abstract bourgeois
individual, an allegedly representative example of the human species, a universal,
disembodied mind. For a critical realist, the point of departure is man as an
embodied historical being, the existence of which is made possible by antecedent
conditions and structures, both natural and social. Men and women have causal
powers to change some of these conditions and structures, but, as Marx (1968
[1852]:96) said, ‘not just as they please, not under circumstances chosen by
themselves, but under circumstances directly encountered, given and transmitted
from the past’.9 Adequate theory of knowledge is thus not possible without adequate
social ontology. Also in this sense ontology is primary. Ultimately, CR contends that
interpretative reflectivism must be grounded on ontological considerations.
Third, CR includes a theory of emancipation; it is also ethico-politically critical.
The original version of the CR theory of emancipation was based on the notion of
truth. Theories or discourses that are in some important regards false can, nonetheless,
be necessary for the reproduction of practices and relations. False understandings and
related structures are unduly limiting human possibilities. The basic idea is that we
can derive ethico-political judgements from truth-judgements. Hence, as virtuous
scholars, we may have a moral obligation to change those practices and relations,
which presuppose false theories and discourses. Although truth-judgements are
necessary for an emancipatory argument, they are not sufficient. A generalised
conception of emancipation strives for human flourishing, and also takes into account
other values such as pluralism, justice, democracy, good, virtue, economic efficiency,
ecological care and the like.

After international relations

This book is divided into three parts. The first part outlines the international
problematic and some attempts to overcome it. It explain why theories based on this
problematic have failed, and concludes with CR suggestions about how to tell better
stories about world politics. The second part deepens these suggestions and develops
them in detail into a CR account of social sciences. It is also my contribution to CR
social and political theory. The last section applies CR methodology to the study of
world politics. The different applications aim not only at making the methodological
discussions more concrete, but suggest—by way of explanatory criticism and concrete
Utopias—transformations from international relations towards world politics.

Part I: overcoming international relations

In the first chapter ‘How it all started? Can it end? A genealogy of the
international problematic’, the conventional problematic of international
relations (IR) is explicated. I argue that the IR problematic has usually been
 Introduction 11

mis-identified. Those within the theory/problem-field solution set have since

Hume and Kant reified the inside/outside—domestic order/inter national
anarchy—problematic as natural and unchanging, although it is based on very
specific geo-historical practices and related metaphors. It took centuries to
establish the practices of capitalism and modern states, and it was only in the
nineteenth century that we can really talk about ‘international’ relations with
highly regulated diplomatic practices. The bourgeois conception of man,
Hume’s positivism, the problem of order and Kant’s dichotomies between the
Humean world and moral reasons are at the heart of the international
problematic that surfaced with the new realities. The post-structuralists have
brilliantly exposed the specific and problematical presuppositions of this
discourse, yet they have also mis-identified some of the underlying premises
and failed to recognise others.
What typically goes unnoticed is the role played by irrealism in structuring
these debates, often stemming from reactions to Kant and, later, Hegel, Nietzsche
and Max Weber. In many respects, the post-structuralist critique is in fact
embedded within the same background discourse and is derived from a long
philosophical tradition of anti-realism/scepticism. The first chapter thus provides
a dialectical comment on the international problematic by situating it in a more
adequate way, and by uncovering its presuppositions in a more adequate way.
It also suggests, albeit only tentatively, possible historical and theoretical ways
to overcome this specific historical problematic. However, before further steps
can be taken, the CR alternative has to be articulated.
The discipline of IR was formally founded (at least in the Anglo-American
world) during the aftermath of the First World War, in 1918. The idea was to study
scientifically the conditions of peace and war, to overcome the problem of war. After
the Second World War, with the ascendance of Political Realism in the USA—which
also became the centre of research on international politics—the emancipatory task
was mostly taken over by Peace Research, which was prominent also in the Nordic
countries, Germany, the smaller English-speaking countries such as Australia and
Canada, as well as in India and other Third World countries. In the 1950s and
1960s in particular, the project of Peace Research was based on the positivist belief
in a science-induced progress; in the USA even more than anywhere else.
Since the 1960s, Hayward [Link] Jnr has been a key figure in the movement
from positivism to more critical and reflectivist approaches both in IR and
Peace Research. Alker started as a pluralist, who attempted to develop cuttingedge
mathematical and statistical techniques in modelling international political
interactions. The second chapter ‘Learning from Alker: From quantitative Peace
Research to emancipatory humanism’ discusses Alker’s development in terms
of four lessons drawn from his intellectual development.
The first lesson about ‘how to collect and analyse “data”’ marks not only
a movement from quantitative to qualitative but also an ontological shift in
understanding what data is: Alker started to argue that ultimately our understanding
of the nature of data depends on our social ontology. According to the ontology of
inter-subjectively constituted social entities we have to have multi-perspective
12 Introduction

descriptions of social events and characteristics. The second lesson concerns how
to analyse and help to solve collective dilemmas. The Prisoner’s Dilemma (PD)
is a game-theoretical model in which individual utility maximisation appears to
be self-defeating, or at least contradictory to some kind of social or collective
rationality. Not only systems of subordination have been analysed in terms of
this model, but also the ‘tragedy of global commons’ as well as inter-state insecurity,
arms races and crisis bargaining. After a long road towards more conversational
and dramaturgical models, Alker eventually came to the conclusion that there are
close connections between the world understandings of different research
paradigms—with their parallels in the actors’ world understandings—and of actors’
abilities to resolve PDs (acknowledging that any way of defining the problem and
its ‘solutions’ is based on the adoption of a perspective).
Lesson three is about ‘how to broaden and deepen scientifically relevant
modelling approaches’. The starting point is that ‘we must broaden and deepen
the universe of scientifically relevant modelling approaches appropriate for
the formal analysis of interpretative and theoretical world histories’ (Alker
1996:271). Alker makes a strong plea for more reflective approaches: ‘conflict
and crisis, as well as the steady state in which violence remains latent, can be
partially explicated in terms of dynamics which are approximated by the
interactions of artificial systems equipped with the story-building and story-
appraising characteristics’ (Alker 1996:270). Hence, an adequate analysis of
world histories must in fact incorporate the (perhaps ideological?) role of
narratives and their interplay. The argument can be summarised in a question:

Is there some improvement possible in the way scientific historical accounts

approach value questions, structural constraints and human choice possibilities
that seem to give all great world histories the reflective character and dramatic
force of a tragic morality play or the ironic happiness of a Russian fairy tale?’
(Alker 1996:270)

Indeed, the fourth lesson is about ‘how to theorise history?’ Alker concludes
that world history should be analysed (interpretatively) as an open process, in
which choices of actors do make a difference, sometimes also to the intended
direction. After having analysed the interplay of power-balancing, collective
security and the international socialist transformation through people’s wars,
he demands more contextually sensitive and historically reflective doctrines of
collective security, some less total, more realistic alternatives.
Unlike many post-positivists, including most post-structuralists, Alker has
continued to develop more adequate and emancipatory methodologies for the study
of peace, conflict and international relations. However, his post-modernist excursions
and exercises in ‘computational hermeneutics’ etc. have led to the criticism that he
is ultimately an ontological idealist. There are many passages and claims that would
point towards a more realist position, some of them made in explicit reference to
CR. However, it is true that the question of realism is left largely unthematised in
 Introduction 13

Alker’s texts, and there are also some noteworthy ambiguities. In conclusion, I
maintain that, although Alker’s methodological lessons are mostly valid and his
attempts to develop better methodologies highly valuable, there is a need to articulate
a consistently realist ontology and epistemology
The third chapter ‘How to tell better stories about world polities’ discusses some
key meta-theoretical problems of IR. The levels-of-analysis problem emerged in the
early 1960s as part of the positivist project to categorise the explanations of peace,
war, co-operation and conflict in terms of locating the main explanatory ‘factors’
into separate ‘levels’: individual or sub-state group or bureaucracy; state; and inter-
state system. In the 1980s, Richard Ashley and Alexander Wendt brought the
agency-structure problem into IR, following Giddens’s and Bhaskar’s earlier
contributions to social theory. In the debates of the 1990s, the levels-of-analysis and
agent-structure problems have often—and misleadingly—been conflated.
By analysing a well-known debate between Wendt vs. Martin Hollis and Steve
Smith, I make a case for a CR approach to grasp levels and depth, and to resolve—
at least tentatively—the agent-structure problem. In particular, I maintain that the
levels-of-analysis problem is nothing but an implicit verticalisation of the horizontal
inside/outside problem, the essential component of the international problematic. I
propose, first, that layers and depth should be analysed more realistically in terms of
deep assumptions of background discourses, historical sedimentation of social
structures and emergent power of various overlapping (or intra-relational) social
systems. Second, that a more adequate account of global spatiality would also enable
us to see non-territorially organised practices, including—despite their territorialist
pre-conceptions—many of the traditional international institutions, such as
diplomacy, international law, balance of power and collective security. Third, that
the agent-structure problem can be unpacked by means of analysing the elements of
causal complexes responsible for the effects we are interested to explain. And fourth,
that social scientific explanation is a form of understanding, based on a double-
hermeneutic process.
Moreover, I conclude the third chapter by arguing that any explanation
includes a genuinely critical moment, with real emancipatory potential.

Part II: explicating a critical realist methodology

In the previous part, the quest for a CR ontology and epistemology was brought
up. In response, the second part explicates the basic elements of a CR
methodology for the study of social relations and politics from a globalist
perspective. Since the subject matter determines its cognitive possibilities, the
fourth chapter ‘Critical realist social ontology’ starts this part by developing a
social ontology by a conceptual analysis of ‘action’, ‘structure’ and ‘power’. I
scrutinise critically the conceptions of Giddens and Bhaskar, and thereby pursue
further the fundamental categories of social scientific ontology. Accordingly, I
also spell out in detail the notions of causal complex and iconic model, which form
the core of the argument of the rest of the book.
14 Introduction

The fifth chapter ‘Double hermeneutics of iconic modelling’ elucidates in detail

the methodology of iconic modelling of causal complexes. By utilising ideas of Rom
Harré, Nicholas Reseller, Paul Ricoeur and Andrew Sayer, I explicate a CR
methodology for the study of social worlds. The systematic use of imagination—
metaphors and analogies—plays an essential role in constructing an adequate iconic
model. Instead of deriving propositions from ‘a secure starter set’, social scientific
research should start with a wide, pluralist set of interesting candidates and, step by
step, eliminate and refine them until only a contracted range of fully endorsed iconic
models is left. This is the reductive approach to model building. Iconic models are
projective and make references to the real world.
I also argue that, instead of armchair philosophising, research is about
encountering, collecting and analysing empirical evidence. Although quantitative
data is indispensable and sometimes useful, qualitative evidence is ontologically
primary. The process of collecting and analysing evidence is double hermeneutics.
Both the accounts of the lay actors and the iconic models of the researcher are
situated not only in space but also in time and have a structure of a narrative. In
anticipation of the more general emancipatory argument of Chapter 6, I argue,
following Ricoeur, that the third moment of mimesis in social scientific research re-
signifies the human practices when the new narrative of the researcher is restored back
to the time of action and practices.
The sixth chapter ‘The normative logic of explanatory emancipation’
discusses, first, the nature of truth. It makes an argument for a metaphorical
correspondence theory, which is consistent with ontological realism and
epistemological relativism, and criticises Bhaskar for not being sufficiently explicit
in his refutation of the metaphysical correspondence theory. Following Harré,
but developing his notion further, I argue that truth is a regulative metaphor,
which has normative force. Truth is a human judgement, which is based on a
metaphor of correspondence to the way the world really is. It suggests a
resemblance between theory-dependent statements, on the one hand, and the
essential relations of the area of reality under study, on the other. This conception
enables me to both justify and re-fashion Bhaskar’s logic of explanatory
emancipation. If some of the central beliefs of an ideology are false, and if one
possesses an explanation of the reproduction of this ideology, the quest to absent
or transform the corresponding practices follows.
Resisting the tendency to reify this logic as ‘naturalist’, I argue that its force is
based on the normative force of the notion of truth. Any truth-judgement must be
subject to the limitations and constraints of the condition of epistemological relativism.
This is particularly important in contexts in which the transformation of politics
into violence is more than a mere abstract and remote possibility. A consistently
relativist CR must attempt to avoid reproducing—or, for that matter, building—
structures of cultural violence. Any critical theory must be willing to analyse the
violence inherent in its own categories. When critical explanatory arguments are
turned into political praxis, first preference should be given to reciprocal
communicative action. Moreover, strategic political action has to be non-violent
and remain open to critical communication. I conclude this chapter by making a
 Introduction 15

case for explicitly normative visions. Positive arguments about concrete Utopias
should complement explanatory criticism, and guide transformative actions.

Part III: visions of world politics

After having explicated the fundaments of a CR methodology, in the last section

of the book I turn to demonstrating how these ideas can be put to work.
Simultaneously, I try to locate the position and task of critical theories in the
context of global politics of the turn of the twenty-first century.
Thucydides’ History is one of the constitutive texts of Political Realism. In this
sense, Thucydides is still very much a part of the existing power-political reality. The
basic aim of Chapter 7 ‘Modelling Thucydides’ Melos episode’ is to try out different
possibilities for the iconic modelling of the Melos episode and thereby to illuminate
the CR methodology of iconic modelling, suggested and developed in the previous
chapters. However, the analysis of the Melos episode allows me to show that, with the
help of CR methodology, better, less reificatory, more historical and more systematically
evidence-based readings of Thucydides are possible. This example allows me to take
part in the theoretical attempts to deconstitute political realism as a discursive, inter-
textual practice, possibly also to interfere in the ongoing constitution of the power-
political reality—without denouncing the perhaps lasting value of Thucydides’ story.
Chapter 8 ‘A global security community’ takes up the IR problematic again
for a new, reconstructive discussion, in light of the ideas developed in previous
chapters. The international problematic is constituted by a simple argument,
according to which the lack of a world state implies that international relations
are ‘anarchic’. The Kantians think that the solution to this problematic is to
establish a supra-national law and order. The CR notion of causality supports
Karl Deutsch’s historical and empirical claim that the existence of the state is not
a necessary or a sufficient condition for peace, nor is the non-existence of the state
a necessary or a sufficient condition for the prevalence of the acute threat of
political violence. Deutsch has also argued, on the basis of suggestions of a study
of a number of historical cases, that the imposition of a unified system of
enforcement of norms may well decrease rather than increase the chances of
peace. A pluralist security community is one in which there is a real assurance
that the members of that community will not fight each other, but will settle their
conflicts in some peaceful way, by means of peaceful changes. This helps to reveal
the reifications of the IR problematic, create the basis for a more fruitful problematic
and develop a vision for the role of emancipatory research of world politics.
History is always open. Past struggles can always be re-opened in new present
contexts that may be more favourable to the possibilities which were previously
suppressed; new combinations of the existing elements of social contexts can be
invented and innovated; new social forces can emerge; and also genuinely novel
elements may be innovated and fed into the processes of present and near-
future political struggles. The social forces constituted by the IR problematic
oppose the emergence of global public political spaces or are tempted to impose
16 Introduction

unilaterally the norms of the dominant states. It is argued that the project of
CR—and related forms of other critical theories—is to work for enhancing the
self-transformative capability of contexts, and thereby for creating the conditions
for global integration and opening up and sustaining public spaces for world
politics (in the wide sense of the term).
The last chapter ‘From security to emancipatory world politics: beyond
“Nordic nostalgia”’ is both an example of explanatory criticism of a contemporary
process of change and also indicates the urgency to conceive ethico-political
projects in globalist terms. Of the Social Democratic regimes of the Bretton
Woods era, the Swedish model was perhaps the most radical and universalist.
This model has been shared in most respects by all Nordic countries, which
together form a highly integrated pluralist security community, Norden. The
Swedish model was economically successful and sustained highly egalitarian
economic policies for more than forty years, and translated those aspirations
into a progressivist foreign policy of active neutrality of the Third Way.
Ole Wæver’s 1992 article ‘Nordic nostalgia: Northern Europe after the Cold
War’ seemed to articulate the widespread feeling of the end of Norden. However,
this argument is flawed and mistaken. Logically and conceptually it is confusing
and tendential, and the evidence to back the empirical claims is casual. This
explanation is contrasted to an alternative, more systematic iconic model of the
same process. The alternative explanation in terms of political economy starts
from the observation that it was not the end of the Cold War but rather the
end of the Bretton Woods system that gave rise to the re-defmition of the Third
Way already in the early 1980s, well before the end of the Cold War. The
problems of the Nordic model stemmed from insurrections against local relations
of domination at the workplace; transformations of the occupational structures
and class relations; the crisis of the Bretton Woods system of regulating the
global economy; and the liberalisation of the exit options of capital, among
other processes. Together with the end of the Cold War, this interplay has also
reinforced irrealist and economistic discourses, which had replaced the earlier
rather Marxist concepts of the theorists of the Social Democratic Party. The
various irrealist and economistic discourses generated accounts about the
requirements of ‘new times’ and gradual changes in the meaning of the notion
of Third Way. They also constituted neo-liberal framing of new social problems.
Starting with the CR idea that emancipation is ‘characteristically the transition
from an unwanted, unnecessary and oppressive situation to a wanted and/or
needed and empowering or more flourishing situation’ (Bhaskar 1993:397), it
is my argument that emancipatory, globally oriented political action is a condition
for the Nordic ideals to be realised and developed further under new conditions.
Tackling the real conditions of the decline of the Norden demands a new
orientation in time and space. The nation-state cannot provide a sufficient
framework for progressive political action any more, and the socially flavoured
Wilsonian ‘idealism’ appears also to be a somewhat anachronistic basis for
‘progressive internationalism’. Claims about democracy and social justice have
to be re-articulated.
 Introduction 17

This is not to argue that politics now occur only outside sovereign states.
Rather, local and national struggles are essentially connected to regional and
global struggles, and therefore they cannot be taken as separate spheres (‘first
progressivism at home, and then exportation of these universalist ideals to the
rest of the world’). Along these lines, I discuss two programmatic visions that
should take the Nordics beyond ‘Nordic nostalgia’, and provide inspiration and
hope for others. The Europeanist strategy is based on a new positive
understanding of the ‘logic of stateness vis-à-vis the free play of market forces’
(Hettne 1993), this time articulated as the ‘pursuit of regionness’. Yet, this strategy
can at best be complementary to a more globalist strategy of carrying out global
democratic and social reforms.10
The currency transaction tax is a particularly promising idea, given that it
would tackle a key issue of globalisation, namely the power of global financial
markets; enable the collection of substantial resources for global purposes
(including Keynesian demand-inducements and re-distribution); and constitute
a leverage for other democratic reforms. To really tackle the conditions of the
decline and the partial neo-liberal transformation of Norden, many further
programmatic ideas would be needed as well. However, particularly assuming
some success in empowerment and increased autonomy, these programmatic
ideas may also concern the organisation of very immediate local sites and related
national and regional developmental strategies, in the spirit of democratic
experimentalism.
Indeed, the argument is that a re-construction of world politics seems to be
a condition for successful emancipatory politics anywhere.

Notes

1 British and US editions have often translated Staat into ‘nation’ in Kant’s political
writings, but this is misleading (I am thankful for Stefano Guzzini and Ollie Kessler
for checking the original text). Kant was discussing a society formed by bourgeois
citizens, and he sometimes gave anthropomorphic characteristics to the resulting entity,
but in no way did he, unlike Hegel soon after him, consider nations as carriers of
moral conviction and historical progress. It is of course also true that, even for Kant,
the theme of war and peace was not the first and foremost concern. Manicas (1987:32)
argues that ‘after Machiavelli, no theorist in the “liberal” tradition put war at the
center of his analysis’. Kant is but a partial exception. Nonetheless, in his political
writings—which may appear marginal to his main output—he clearly argued that the
problem of war and peace is absolutely decisive to the future of mankind. Therefore,
he also put some effort into articulating the problematic systematically and also to
developing original solutions to the problem of peace and war. For comments on why
liberal political theorists have not been concerned with international relations, see
Wight 1966 and, for a very different account, Ashley 1988. More recently, since the
1980s, due to the prevalence of globalisation discourse, the situation has been rapidly
changing.
2 Paradoxically, however, Morgenthau’s accusation that Kantians are to blame for the
catastrophes of the twentieth century seems to affirm the causal efficiency of ideas.
3 This scepticism is more fully articulated in Harré 1979.
18 Introduction

4 A few remarks on Marx’s understanding of ‘laws’. Although this is not the place to
develop the argument further, I believe that, in his understanding of ‘laws’, Marx in
fact conflated at least three very different meanings:

(i) Laws in the sense of ‘natural laws’ of classical political theory, concerning particularly
the labour theory of value (which was also the normative basis of his immanent
critique of capitalism).
(ii) Laws in the sense of Hegelian laws of ‘logic’, explaining for example the ‘transition’
from (i) to (iii), and also forming the basis of his critique of ‘alienation’ and,
simultaneously, the assumption of ‘laws of history’.
(iii) Empirical realist laws a la Newton and Mill, who also both allowed for scientific
realist questions and inferences. These conflations have been an endless source
of confusions and debates within Marxism (cf. the problem of getting from labour
value to relative prices in the market, i.e. the ‘transformation problem’), and a
reason for many to reject the Marxist doctrine. An attempt to rescue this doctrine
by re-reading Marx as a critical realist is thus likely to mystify more than clarify.

5 Isaac (1987) has argued, however, that it is possible, to an extent, to make sense and
develop further some Marxist conceptions of power in realist terms. Marxism has
manifold gaps and lacunae and is not the correct theory of the world, but ‘deserves to
be examined and criticized on its own merits as a theory, and not simply discarded on
strictly formal (and specious) grounds because it fails to conform to the canons of
empiricist philosophy’ (Isaac 1987:10). This and other similar works give a more
interesting and plausible account of the relationship between CR and post-Marxist
theories of power, for instance.
6 Bhaskar (1989:141), however, criticises Horkheimer for introducing a contrast between
emancipatory and technical reason, which opened up the possibility, which Herbert
Marcuse and Habermas so vehemently exploited, of turning critical theory against
the ‘system’ (world). For Marcuse and Habermas, the ‘system’ (world) is based on
technical reason and science. Whereas Marcuse attempted a totalising critique of the
‘system’, Habermas in fact accepted the inevitability of this ‘system’, thus criticising
only its tendency to ‘colonialise’ the life-world. In this form, critical theory is derived
from (1) a misunderstanding of science and (2) reification of the ‘system’ (world).
Bhaskar’s criticism of these misunderstandings and reifications is correct. However,
these confusions hardly played any important role in Horkheimer’s original text.
7 Ontological monovalence is defined as a purely positive account of reality. If there
is no absence, there can be no emergence either. Without absence/emergence, there
can be no change. ‘It acts ideologically to screen the epistemological and ontological
contingency of being and to sequester existential questions generally, as ontological
actualism sequesters “essential” ones. The result is the doubly dogmatically
reinforced positivization of knowledge, and eternalization of the status quo’ (Bhaskar
1994:254–6).
8 In the first chapter, I apply a rather similar type of analysis to IR theory. However, I
try to eschew a totalising critique. Rather, I want to participate in the debates and
open up new possibilities for better debates and research on world politics/global
political economy, as well as better and more emancipatory practices. These would
not be possible without a dialogical attitude with others here and now, however
constituted.
9 These circumstances are not only external to human beings. In one of his more thrilling
metaphoric statements, Marx (1968 [1852]:96) argued further that ‘the tradition of all
the dead generations weighs like a nightmare on the brain of the living’.
10 Or one could well also talk about socialist reforms in a sense of questioning regulating
or transforming private property rights in democratic and pluralist terms.
Part I

Overcoming
international relations
1 How it all started? Can it
end? A genealogy of the
international problematic

Every problem, whether practical-political or more theoretical, has a set of

presuppositions. Put together—to the extent that there is some consistency—
these presuppositions form a theory and give rise to a characteristic set of
problems and plausible answers to them. Mainstream debates in
international relations stem from a particular theory/problem-field solution
set, or, in short, a problematic. 1 The presuppositions of the international
problematic include: (1) a positivist interpretation of the new natural
sciences, in particular Newton’s mechanics, and the related idea that it would
also be possible to find similar simple formulas to describe the laws of
human nature, as invariant operations of human understanding, or invariant
behaviour, perhaps expressible in terms of differential calculus or other
mathematical formula; (2) conceptions, opportunities and problems
associated with the rise and expansion of capitalism; (3) the practical and
theoretical problems stemming from the emergence and consolidation of
the modern state; and (4) the problem of order, which emerged with (1)–(3)
and the erosion of the authority of the Christian Church. As will be seen,
these components are very closely related.
In this chapter, I discuss the emergence of the international problematic by
focusing on two thinkers—David Hume and Immanuel Kant—and the lineage
that can be traced back to them. Hume was responsible for the positivist
interpretation of Newtonian mechanics. Hume was also among the first to
theorise balance of power, which is one of the mechanistic solutions to the
problem of order. It was also Hume who woke Kant up from his ‘illusionary
dreams’. In contrast to Hume, however, Kant defended the realm of autonomous
and critical reason. Reason says that war is detrimental to morality and progress.
Kant maintained that, in the long run, all human development, indeed human
survival, is dependent upon solutions to the problem of peace and war. By
applying Hobbes’s notion of state of nature to the sphere of relations between
nations and states, he argued for an international social contract. He proposed
that the problem of war could be overcome by an arrangement of rule of law,
republicanism, free trade and ‘league of nations’. This is the beginning of the
international problematic.

21
22 Overcoming international relations

First component: Humean positivism2

Many predecessors of Hume—such as Francis Bacon, René Descartes, Thomas

Hobbes, John Locke and George Berkeley—made arguments for abstract
rationalism and empiricism, and re-introduced the ancient doctrine of atomism.
Often, these bachelors and pensioners of wealthy households, or solitary
aristocrats, had to stay in exile for years in the troubled days of the seventeenth
century. Besides these traumatic political experiences, they were influenced by
the breakthroughs of Bruno, Copernicus and Galileo—so much so that Hobbes
wanted to explain man and society in terms of simple motions of bodies. These
philosophers were grounding their political theories on abstract individualism.
It was Hume, however, who gave an influential empiricist re-interpretation to
Newton’s mechanics. It was also Hume who developed more thoroughly the
sceptical and also political implications of this metaphysics, and explicitly
theorised capitalist market society and balance of power.
Isaac Newton made his revolutionary discoveries—theory of gravitation and
calculus—within eighteen months from 1665 to 1667, still in his early twenties.
At the time, he had been forced to flee to the countryside because of the outbreak
of plague in the city of Cambridge. Nothing was heard of these theories for
nearly twenty years. Newton’s theory of gravitation had been based on an
inaccurate measurement of the earth’s radius, and he realised the differences
between the theory and the facts. In 1684, Edmond Halley, an English
astronomer, persuaded Newton to publish his findings. Philosophiae Naturalis
Principia Mathematica came out in 1687. By then, Newton had learnt the true
value of the earth’s size. Principia was the first book to contain a unified system
of a few scientific principles explaining what happens on earth and in the heavens.
It deployed calculus to present and explain these principles. The result was a
revolution of thought.
It appeared that Newton showed a way from philosophical speculation to
experimental science. Yet Newton’s experiments were usually mere observations
and not always particularly systematic. Like Galileo a few decades earlier, Newton
tended to reduce scientific problems to very simple terms on the basis of everyday
experience, coupled with some methodical observations, and commonsense logic.
Then he analysed and resolved the problems according to novel mathematical
descriptions. Newton’s revolution invited numerous commentaries. Hume’s re-
interpretation was perhaps the most far-reaching of them. Hume was publishing
his thoughts in 1739–57, some fifty years after the publication of Newton’s
Principia. Hume, soon to be followed by Adam Smith and others, gave an
empiricist reading to Newton’s mechanics and also applied the consequent ideas
to social and political theory. Whereas Newton himself had allowed for scientific
realist questions and inferences, Hume reduced all knowledge to senseperceptions
and a few simple operations of mind (Manicas 1987:10–23).
Hume’s starting point was not only Newton, however, but Descartes’s method
of systematic doubt and the empiricism of Bacon, Locke and Berkeley. Hume’s
devastating critique of ontology was based on a few very simple sceptical
 Genealogy of the international problematic 23

arguments. Hume (1975 [1777]:§116) explored ‘how far it is possible to push

these philosophical principles of doubt and uncertainty’. In the end, Hume’s
own scepticism turned out to be moderate, ‘a country gentleman’s scepticism’.
He assumed that we can put some trust on sense-perceptions, although we
sometimes ‘must correct their evidence by reason, and by considerations’ (Hume
1975 [1777]:§117). The role of scepticism was to purge philosophy and science
from metaphysics and ontology, and leave it only with empirical perceptions
and a few simple operations of mind. This was in fact not commonsensical in
the eighteenth century—no more than it is today. Hume notes that almost
everybody in their everyday practices assumes that there is a reality, which
exists quite independently of them, or us, or, ultimately, me. They are wrong:

But this universal and primary opinion of all men is soon destroyed by the
slightest philosophy, which teaches us, that nothing can ever be present to
the mind but an image or perception, and that the senses are only the inlets,
through which these images are conveyed, without being able to produce
any immediate intercourse between the mind and the object. The table, which
we see, seems to diminish, as we remove farther from it: but the real table,
which exists independent of us, suffers no alteration: it was, therefore, nothing
but its image, which was present to the mind. These are the obvious dictates
of reason….
(Hume 1975 [1777]:§118)

Everything—God and angels, beings and their causal powers, society and
morality—have to be, therefore, reduced to a few general principles governing
our (or rather my own) sense-perceptions and how the human mind in general
understands them. Hume argued, for instance, that causation is a mere operation
of mind, ultimately based on regular sense-impressions of contiguity and
succession between A and B. These give rise to a habit of thinking, an association,
that A and B are always connected. Causation is therefore not real, but a mere
property—or rather a universal habit—of human understanding. Causation is
just one of the three ‘bonds that unite our thoughts together’ (Hume 1975
[1777]:§41). Also, Hume’s re-interpretation of Newton’s mechanics is based on
this radical empiricist theory of knowledge and causation. Hume’s scepticism
also gave rise to the famous problem of induction, later declared to be the ‘scandal
of philosophy’. The problem of induction is:how can we know that B will also
follow A in the future? Do we really know that the sun will also rise tomorrow
morning? (See Hume 1975 [1777]:§29).
Hume did not succeed in purging ontology from philosophy and science.
Rather, he posited a particular, narrow ontology as absolute, as the only ontology
that is beyond doubt. Hume persevered tacitly to realism of senses, perceptions
and mind, but denied reality otherwise. If we assume that only sense-perceptions
and mind are real, what will be the resulting social ontology? Society is reduced
to individuals (recipients of sense-perceptions and recorders of constant
24 Overcoming international relations

conjunctions), and these individuals become atom-like, constant in their inner

structure (there are a few universal principles governing the operations of human
mind). The other side of this ontological coin is, paradoxically, voluntarism.
Since there are social arrangements and institutions, such as private property
and sovereign state, they must be explained as outcomes of voluntary actions of
individuals (cf. Bhaskar 1978 [1975]:16, 57; Bhaskar 1986:250–59, 287–308)
In his moral philosophy, which he also called ‘the science of human nature’
(Hume 1975 [1777]:§1), Hume was looking for a Newtonian formula of human
nature, a few simple principles from which human behaviour, tempers and
sentiments, morality and political institutions can be deduced. Although Hume
(1975 [1777]:§163) believed that he was following ‘Newton’s chief rule of
philosophizing’, his project was Newtonian more in its aims than its means. To
use the terms of the ‘great’ 1960s methodological debate in IR (see Hoffman, S.
1959; Bull 1969; Kaplan 1969), Hume was a traditionalist engaging in armchair
speculation; he did not envisage a systematic experimental political science.
Newton’s example was not yet fully followed. The time for empirically controlled,
quantitative political science was to come later, in the 1930s.

Second component: capitalist conception of man

Modern capitalist enterprises emerged in Renaissance Italy. Soon the innovation

spread elsewhere in Western Europe. New practices and a new man were born.
Although book-keeping is as old as the art of writing, the modern practices of double
book-keeping and capital accounting were developed side by side with the new legal
forms of enterprise, property rights and contract (Weber 1978 [1956]:375–80). In
accounting, ‘“capital” is the money value of the means of profit-making available to the
enterprise at the balancing of books; “profit” and correspondingly “loss”, the difference
between the initial balance and that drawn at the conclusion of the period’ (Weber
1978 [1956]:91). For example, Hobbes, who was equally fascinated by geometry and
advances in natural sciences, seemed to pose book-keeping as the model for rationality
in general:

In summe, in what matter soever there is place for addition and substraction,
there is also place for Reason; and where these have no place, there Reason has
nothing at all to do…. As when a master of a family, in taking an account, eastern
up the sum of all bills of expence, into one sum; and not regarding how each bill
is summed up, by those that give them in account; nor what it is he payes for; he
advantages himself no more, than if he allowed the account in grosse, trusting to
every of the accountants skill and honesty: so also in Reasoning of all other
things, he takes up conclusions on the trust of Authors, and doth not fetch them
from the first items in every Reckoning. Loses his labour; and does not know any
thing; but only beleeveth.
(Hobbes 1986 [1651]:111–12)
 Genealogy of the international problematic 25

Rational man is possessive and calculative (as well as patriarchal), constantly

keeping books about his revenues and expenditure, or his capital account—or,
equally well, of his own appetites and aversions, or, as Hume phrased the idea,
calculative gains defined in terms of his passions. This conception of man was
the outcome of the development of a society dominated by market relations,
new division of labour and new technologies. As Marx quite sarcastically put
it, these thinkers from Hobbes to Hume and Bentham were taking ‘the modern
shopkeeper, especially the English shopkeeper, as the normal man’ (in Capital
vol. 1, cited in Keat and Urry 1975:98).
The sceptics and empiricists, impatient with impractical metaphysical
speculations, provided, paradoxically, the metaphysics of this new man. The
bottom line of this conception is social atomism, the idea that individual men
are already essentially excluded from each other and owe themselves and their
work etc. only to themselves. For instance, according to Hume, human beings
are selfish and have a love of gain, and although humans are also benevolent,
that benevolence is limited. This benevolence or, more generally, sympathy, is
in need of explanation, not selfishness and gain-seeking. Hume’s problem of
sympathy was: given individualism and the fact we can know only our own
feelings, how can we feel sympathy towards other people?
Although the radical implication of Hume’s empiricist epistemology—that we
can only know empirical regularities and should not draw normative conclusions
from them—was later canonised as ‘Hume’s guillotine’, Hume’s own moral and
political philosophy does by no means avoid making normative claims. Like
Hobbes, Grotius and Locke, Hume also outlined normative rules and principles
as ‘natural laws’. One of his central notions was justice. For Hume, justice has to
do with private property and ownership. A just individual abides by the legal
rules governing private property and ownership in his society. The three
fundamental rules of justice, namely, stability of possession, transfer by consent
and keeping of promises, are laws of nature.
There is a very close link between the atomist social ontology and the
justification of private property and the related ‘fundamental rules of justice’.
The essence of private property is an individual (or corporate) right to exclude
others from the use or benefit of something (MacPherson 1973:122). This right
was usually also presented as absolute, permanent and therefore inalienable,
but freely alienable by consent (by contract). Hume, for example, justified his
conception of justice in terms of a hypothetical history. Assume, with Hume,
that people are atomist and mostly selfish, that resources are scarce and that
these individuals have a limited amount of beneficial possessions. Imagine how
they would behave in a society without the enforced institution of private
property. They can easily resort to violence to get what they want. Life would
be, to use Hobbes’s (1986 [1651]:186) famous phrase, ‘solitary, poore, nasty,
brutish and short’. Slowly and painfully, people would have to learn and,
eventually, will learn. Thence the institution of private property—and the related
principles of justice and contract enforcement—will be created by a spontaneous
act of rational individuals.
26 Overcoming international relations

Positivism and capitalist conceptions of man are essentially connected. Both

emerged with the development of a fully autonomous capitalist market society
in Western Europe in the seventeenth and eighteenth centuries. The sceptical
method, the abstract arguments and the hypothetical, counterfactual histories
made the novel practices and institutions appear both self-evident and universally
valid, despite the complicated struggles, (civil) wars, revolutions and mass
dislocations of people in seventeenth- and eighteenth-century Western Europe—
and also the colonised Americas.3 These struggles and transformations were
soon to spread to Eastern Europe, Russia, Turkey and elsewhere.
In fact, it was easier for the supporters of the new bourgeois philosophy to
establish the institutions of autonomous civil society, such as private property
and liberal state, than to create rational man. Although the ‘English shopkeepers’
and capitalist landlords and entrepreneurs provided the model, in most contexts
the stable, calculable, cautious, reliable self-interested rational individual had to
be gradually fabricated by means ranging from civilising education to violent
enforcement (Connolly 1993:27). Positivism and the new conception of man
justified far-reaching transformations, yet, simultaneously, these theories pushed
the real historical struggles and transformations out of sight. As Bhaskar
(1986:307–08) has pertinently summarised: ‘Positivism at once naturalises and
normalises things and reflects in an endless hall of mirrors the self-image of
Bourgeois Man. It is, one might say, the house-philosophy of the bourgeoisie.’

Third component: problems of the modern state

The spatial imagery of a strict inside/outside dichotomy rose gradually with the
new cartography of the Renaissance, bursting into blossom by the time of the
Enlightenment (Harvey 1990:240–59; Walker 1993:125–40). Originally, modern
cartography was also developed to help establish property rights over land and
to serve commerce, particularly colonial commerce. It soon became connected
to statist and, later, nationalist myths. Private property provided the model. In
an important paper published in the early 1980s, Ruggie (1983:276–9) noted
that modern sovereign states have been constituted as analogous to possessive
individuals (see also MacPherson 1964). In the case of an increasing number of
states and treaties, external sovereignty was constituted as analogous to private
property rights in the seventeenth- and early eighteenth-century Western and
Northern Europe. This gave the sovereign body an absolute and exclusive
control or ownership over all things ‘inside’.
The idea of nation as an exclusive community rose much later, at the time of or
after the French Revolution. Nationalist ideas were subsequently cultivated with great
fervour and also used to oppose the rule of empires and great powers, and unequal
exchange in the capitalist world economy (cf. Linklater 1990:55–75; Linklater
1998:145–78). In essence, however, the notion of exclusive national community is a
metaphorical extension of the capitalist conception of man, who owes nothing to
society to which he belongs, and whose possessions exclude the others.
 Genealogy of the international problematic 27

It took centuries before sovereign statehood was established and consolidated

in the exclusive territorial sense. Property in the Middle Ages was simultaneously
conditional and not easily alienable (paradigmatically, a right to a stream of
revenues from inherited land, which was also controlled by lower vassals).
Similarly, membership in the European medieval diplomatic community was
also both conditional and informal, and assumed therefore a wide variety of
forms. Membership in the diplomatic community was conditional because it
carried with it explicit social obligations, for instance towards the Church and
papacy. Due to the lack of formal rules, cities, associations of trades, universities,
commercial leagues, empires and papacy all were considered legitimate actors
and could be granted, for instance, the right of diplomatic representation and,
later, embassy. Well into the eighteenth century, there were complicated trans-
European relations between ‘overlapping authorities and multiple loyalties’ (Bull
1977:254), instead of exclusive territorial spheres of centralised states.
Some of the modern state concepts and practices emerged in Rennaissance
Italy. After the widespread disappointment with the ancient republican ideals, a
new Tacitus-inspired reading of Machiavelli and other humanist authors gained
strength. ‘A term that Guicciardini first used became in time a watchword for
this new movement: it was ragion di stato, “reason of state”, which appears in the
titles of books and pamphlets all over Europe from about 1590 onwards’ (Tuck
1992:121). Perhaps encouraged by the increasingly rational ordering of space
in Renaissance maps, the theatre of state appeared to these thinkers as an
increasingly distinct sphere, with a logic of its own. At this time, state-structures
were developed in Spain and Portugal to also govern colonies in America and,
soon, in England, France, Sweden and elsewhere.
The Peace of Westphalia (1648) that ended the Thirty Years War in Europe
is often cited as the beginning of modern inter-state relations. It introduced or
consolidated a number of new notions from freedom of religion to state
sovereignty and great powerness. Yet, Westphalia was only a phase in a long
and complicated process. For instance, England became an exclusive and unified
territorial state, gradually, only after the Glorious Revolution of 1688. The
struggles between Protestants and Catholics, King and Parliament, old privileged
and the rising bourgeois classes and the capitalist landlords and the traditional
peasant communities took place in a trans-European setting of inter-dynastic
rivalries. King James II, a Roman Catholic, favoured Catholics and attempted
to restore monarchical autocracy. The Protestant majority of England disliked
his policies. In 1688, leading politicians invited William of Orange, who was
the husband of James’s Protestant daughter Mary, and also the ruler of the
Netherlands, to invade England with Dutch forces. William met little resistance,
except in Scotland and Ireland. William and Mary were crowned co-rulers of
England. The Revolution established Parliament’s right to control succession to
the throne and to limit the monarch’s power. In 1689, Parliament passed the
Bill of Rights, which banned Roman Catholics from the throne and made it
illegal for a monarch to suspend laws, keep an army in peacetime or levy taxes
without Parliament’s consent.
28 Overcoming international relations

In the course of the seventeenth and early eighteenth centuries, the capabilities
of (particularly Western) European states were greatly strengthened due to a
number of administrative and financial innovations. These innovations enabled
an increased mastery over activities on states’ ‘own’ territory and also
management of their overseas colonies. Colonialism in fact precipitated the
universalisation of European property rights and, also, legal regulation of the
‘high seas’. But for the emergence of the international problematic, the
development of state capabilities at ‘home’, as it were, was decisive. Seventeenth-
century France was a forerunner of many bureaucratic practices. External borders
were clarified—in 1718 a line was drawn on a map for the first time in a treaty—
and states assumed administrative control over the units inside the borders (see
Giddens 1985:84–102).
Socio-economic developments came to be carefully monitored by means of
state reports and memoranda. Gradually, the financial ties to the feudal system
were cut off. The new sources of state revenue included customs on economic
flows, taxes on transactions, establishment of trade companies and involvement
in economic developments. Moreover, states began to use lotteries—early form of
bonds—as a means of raising funds for wars or colonial projects, and they also
established central banks (Goede 1999). Diplomatic representation became both
more permanent and professional. By the eighteenth century, most European
states had large corps of diplomatic staff, at home and abroad. However, it was
not until 1796 that the terms ‘diplomacy’ or ‘diplomatic’ were used in English to
apply to the conduct and management of international relations, and not to the
study of archives. It was after 1815 that the status and rules of this profession
were established by an international agreement (Nicolson 1939:26–8).
The seventeenth- and eighteenth-century liberal philosophers dealt with the
problems of the modern European state. They justified the authority of the
sovereign state but also carefully circumscribed its rights and sphere. Internally,
the authority of the state was strictly limited both by the autonomous sphere of
civil society and by natural and positive laws. The rule of law was thus
established. Externally, the state emerged as analogous to private property. Hume,
for instance, lived in mid-eighteenth century Britain, in which the bourgeois
revolution had been completed and which was, evidently, a growing power and
empire. The relations between European states, or European powers and the
rest of the world, were not his first preoccupations. However, after having
completed his voluminous History of England in the early 1760s, Hume worked
at the British Embassy in Paris. In 1767–8, he was in London as Under-Secretary
of State, Northern Department. Also, in his essays, mostly drafted before his
practical-political experiences, Hume reflected upon the rise of England. He
argued for instance that Great Britain’s growing power was due to the prevalence
of ‘commerce and industry’ in Britain (Hume 1994 [1777]:219–30). Hume (1994
[1777]:135–42) also devoted one of his essays to the idea of the balance of
power. Like the diplomats and princes, who had written the principle down in
the Treaty of Utrecht in 1713, Hume saw balance of power as the solution to
the problem of European inter-state order.
 Genealogy of the international problematic 29

Fourth component: the problem of order

According to Toulmin (1992:17–44), the first modernity of Renaissance

humanism was boosted economically by the rise of new capitalist market practices
in the Northern Italian city-states and the colonial expansion of Spain and
Portugal. This period of growth was followed by a worsening climate (‘Little
Ice Age’), economic depression, a sequence of violent outbreaks of plague, and
religious strife and wars in the seventeenth century. Violence, destruction and
brutality went so far that, at times, groups of cannibals were roaming in parts
of Central Europe. Deep uncertainty, fear and anxiety made the new philosophers
of the ‘second modernity’ turn to those few things they thought would be
beyond any doubt: their own immediate sense-perceptions; geometry and
mathematics, and the new astronomy; calculative reason; and atom-like
individuals and their contractual relations. It was the religious and inter-dynastic
strife and violence that encouraged them to search for universally valid arguments
for political order. They also wanted to safeguard the peaceful everyday practices
of the rising bourgeoisie.
Albert Hirschman (1977) has shown, however, that in an important sense
there was continuity from the Middle Ages through Renaissance humanism to
the ‘second modernity’ of the seventeenth and eighteenth centuries. The notion
of fallen man was transformed first into the sceptical and relativist but potentially
virtuous and glorious man of humanism, and then into the passion- and interest-
driven rational, calculative man of Hobbes, Locke, Hume and Smith. A kernel
of the Christian conception of human nature was preserved through these
transformations. Humanists, including Machiavelli, denounced Christian
morality, and accepted St Augustine’s ‘three sins of fallen man’—lust for money
and possessions, lust for power and sexual lust—as objective descriptions of the
human condition, ‘how man really is’. With ancient republicanism, however,
the humanists stressed political power, glory and aristocratic virtues as more
important than the drive for money and possessions. The later philosophers,
many of whom indeed modelled man on the ‘English shopkeeper’, reversed
this hierarchy of importance. By the time of Hume, they could argue confidently
that commercial activities, even banking and financial money-making pursuits,
were not only beneficial and honourable, but also that the glory of the state was
dependent on them.
St Augustine, like other medieval Fathers of the Church, was wrestling with
the problem of moral and political order. Man is free and responsible for the
evil in the world God created, yet God is also inside every person. It was the
task of the authority of religion and Church to convince people to side with
God against evil in them. The first answer to the problem of moral and political
order was thus religious command. However, St Augustine did not trust man.
He entrusted to the secular authorities ‘the task of holding back, by force if
necessary, the worst manifestations and the most dangerous consequences of
the passions’ (Hirschman 1977:15). The religious command lost its plausibility
30 Overcoming international relations

by the time of the religious wars. ‘A feeling arose in the Renaissance and became
firm conviction during the seventeenth century that moralizing philosophy and
religious precept could no longer be trusted with restraining the destructive
passions of men’ (Hirschman: 14–15).
How is it possible to have an organised and relatively peaceful society in the
absence of God and His commands? This is the modern problem of order.
Hobbes made a novel, apparently secular argument for a solution, which
appeared at least somewhat similar to that of St Augustine. The Hobbesian
solution to the problem of order was coercion and repression, combined with
the idea of rationalising education, to replace religious command. However,
there was a flaw in Hobbes’s argument. Why should we assume the sovereign
to be different from other humans? Either the Hobbesian order would end up
in a full-fledged tyranny, or Hobbes would have to make the sovereign
accountable to God. In the first case, the whole argument would turn out to be
futile; in the latter, the system would have to rely, ultimately, on God’s commands,
i.e. religious belief.
The great innovation of the eighteenth century was to direct human passions
to the service of generally beneficial order. Hints at this solution can be found
in earlier texts, but it seems that the idea was articulated, first sporadically and
then more systematically, in the seventeenth and eighteenth centuries. Already
Bacon’s The Advancement of Learning (1605) mentions the practical-political idea
of countervailing passions within and between states. Later, for instance, Spinoza’s
Ethics (c. 1680), Mandeville’s The Fable of Fees or Private Vices, Public Benefits
(subsequent versions 1705, 1714 and 1729), Vico’s The New Science (1725) and
Hume’s A Treatise of Human Nature (1739), rewritten as Enquiries Concerning Human
Understanding and the Principles of Morals (1748 and 1751), discuss the idea of
setting one passion against another. Thus, Hume argued that in many cases
private passions, including the ‘passion of interest’, can best be controlled by
means of countervailing passions. Love of leisure can be countered with the
lust for gain. One (faction of) state can counter another (faction of) state, both
driven by passion for power and glory. It is therefore possible to engineer social
progress by setting up one passion to fight another.
Hume was ambivalent about the notions of interest and passion. Increasingly
since the early seventeenth century, prevalent discourses had contrasted ‘interest’,
which tended to acquire an economic meaning, against ‘passion’. Whereas the
aggressive pursuit of possessions by means of violence and power seemed a
source of misery and evil, the rational money-making of the bourgeoisie in a
stable and just setting, based on producing and exchanging goods, services and
assets, appeared beneficial to many. Thus rose the great civilisational quest to
fabricate the calculative self-interested man. In a particular social setting,
characterised by Hume’s three fundamental rules of justice—stability of possession,
transfer by consent and keeping of promises—calculative money-making appeared
to benefit everybody (i.e. property-owning men). In 1776, Adam Smith, who
had pleased Hume by summarising his Treatise so insightfully as a young student,
invented the metaphor of the ‘invisible hand’. Smith’s metaphor describes how
 Genealogy of the international problematic 31

individual self-interested actions can lead to order and progress. In order to

make money, people produce things that other people are willing to buy. Buyers
spend money for those things that they need or want most. When buyers and
sellers meet in the market, a pattern of production develops, which results in
harmony. Smith said that all this would happen without any conscious control
or direction, ‘as if by an invisible hand’.
There was also a demand to treat the problem of order more and more
scientifically, in terms of constancy and predictability of behavioural patterns. It
is striking that the problem of order, thus conceived, is in fact a homeomorph
to the Humean problem of induction (Bhaskar 1994:31). It is similar to, and
can be modelled in terms of, the problem of induction. If society can be known
only in its instances—individuals and their externally observable behaviours—
the problem of order is acute at all times: how is it possible that an organised
society could come about, knowing how chaotic, how much in conflict the
instances of society can be? The answer of a Humean sceptic is that indeed
there is no guarantee. However, by using the state to safeguard the institutional
underpinnings of justice and order, and by carefully engineering passions or
interests to counter each other or to yield unintended yet socially beneficial
results, we can achieve constancy, predictability and relatively stable social order.
In the course of the eighteenth century, three social ‘mechanisms’—besides
state coercion—were developed: the invisible hand; checks and balances within a
state; and balance of power between states. The problem of order in relations
between European states was usually discussed only in passing. However, by the
time of Hume, many of the basic assumptions of the international problematic
had surfaced. The first is the existence of a society of states, with states analogous
to persons, and persons conceived as possessive individuals. The second
presupposition is the problem of order. The third is the idea that it is possible to
achieve order by careful social engineering of human passions and interests. The
fourth assumption is the implicit social ontology of atom-like individuals, which
are constant in their structure (‘English shopkeepers’ as the standard), and which
are simultaneously subject to determinist regularities and, quite amazingly, capable
of creating the social order spontaneously. Given these assumptions, it is thereby
no surprise that in his study on balance of power, Hume concludes:

In short the maxim of preserving the balance of power is founded so much

on common sense and obvious reasoning, that it is impossible it could
altogether have escaped antiquity, where we find, in other particulars, so
many marks of deep penetration and discernment. If it was not so generally
known and acknowledged as at present, it had at least an influence on all the
wiser and more experienced princes and politicians. And indeed, even at
present, however generally known and acknowledged among speculative
reasoners, it has not, in practices, an authority much more extensive among
those who govern the world.
(Hume 1994 [1777]:139)
32 Overcoming international relations

Hume claims that there have always been balance of power policies.4 This
rational policy is relentlessly threatened by human passions. An eternal sceptic,
Hume did not quite share the increasingly common sentiment that reason
could tame passions and transform them into socially beneficial interests. He
is famous for his phrase ‘reason is, and ought only to be the slave of passions’.
It is also telling how Hume complains that the balance of power does not
have a practical effect to the extent it should. In Hume’s view, statesmen are
driven by their passions, not by rational considerations. This condition does
not change. Here we have the elementary formulation of the to-be Kantian
distinction between causal phenomena and the impotent world of moral reason.
What is most noteworthy in this passage is that Hume is repeating his
metaphysical conviction that there can be no real changes. If society consists
of atom-like individuals, constant in their inner structure; and if, despite the
ubiquity of ephemeral changes, there are invariant regularities between relevant
As and Bs in society; it is hard to see what could, really, ever change.
Everywhere people must be the same as in eighteenth century England and
France. If not, they must be becoming the same, if they are rational.

Kant’s articulation of the international problematic

It was Hume who interrupted Kant’s ‘dogmatic slumber’. However, it was

Kant who finally, at the time of the French Revolution, put all the pieces together
and gave a familiar shape to the international problematic. Kant argued that
neither natural law nor the balance of power provides a basis for peaceful order
in Europe. Natural law theorists are mere ‘cold comforters’ (for a legal positivist,
natural law is just wishful thinking), and in fact they often provide justifications
for wars (Kant 1983 [1795]:116). Balance of power is like ‘Swift’s house, whose
architect built it so perfectly in accord with all the laws of equilibrium that as
soon as a sparrow lit on it, it fell in’ (Kant 1983 [1793]:89).
Domesticating the radical tenets of Rousseau, who had died a few years
before Kant wrote his political texts, Kant presumed that the external relations
among nations form a state of nature in the sense of Hobbes. In his famous
1795 essay ‘The perpetual peace. A philosophical sketch’, Kant saw this as a
normative problem to be overcome. Since international war and violence occur
repetitively, there is a reason to find a way out from this situation, to make an
international social contract to avoid ‘perpetual peace to occur only in the vast
graveyard of humanity as a whole’ (Kant 1983 [1795]:110). Kant continues:

Just as we view with deep disdain the attachment of savages to their lawless
freedom…and consider it barbarous, rude and brutishly degrading of humanity,
so also we should think that civilized peoples (each one united into a nation)
would hasten as quickly as possible to escape so similar a state of abandonment.
(Kant 1983 [1795]:115)
 Genealogy of the international problematic 33

As long as there are no external laws, states are living in a state of nature. Thus,
Kant advocates the rule of law and elementary principles of citizenship also in
the international realm.
Kant introduced a far-reaching dualism to Western political philosophy,
that of phenomena/noumena. At the level of phenomena, Kant’s world is
ultimately Humean. Kant was plainer than Hume himself; for Kant,
international relations are in a state of nature. This is, indeed, the origin of
what Ashley (1988) calls the anarchy problematic. As a transcendental idealist,
Kant also introduced the noumenal world, the site of reason and morality.
However, this world was cut off from the world of causal phenomena. To
demonstrate that his theory was also practically useful, Kant was at pains to
introduce all kinds of ‘secret mechanisms’ that would eventually establish and
then maintain the legal order of perpetual peace. These ‘mechanisms’ resonated
with the ideas of the late eighteenth century, and they included: the ‘invisible
hand’ of free trade and the civilising effect of commerce; the republican
constitution of states, guaranteeing freedom and peacefulness; and attempts
to curb the capabilities of states to finance and wage wars.
Already ten years before his famous ‘Perpetual peace’, Kant tried to
explain why history should develop into a cosmopolitan and peaceful
direction by introducing the notion that nature, too, is teleological and
that ‘all of a creature’s natural capacities are destined to develop completely
and in conformity with their end’ and that ‘man’s natural capacities are
directed toward the use of his reason’ (Kant 1983 [1784]:30). The main
argument is that there must be a simple and crude but, from a rational
perspective, necessary learning process. The impetus for learning comes
from the experience of pain and the high costs of the ever more destructive
wars. A lengthy quotation from Kant illustrates the logic of the argument:

In the end, even war gradually becomes not only a very artificial undertaking,
so uncertain for both sides in its outcome, but also a very dubious one, given
the aftermath that the nation suffers by way of an evergrowing burden of
debt (a new invention) whose repayment becomes inconceivable. At the same
time, the effect that any national upheaval has on all the other nations of our
continent, where they are all so closely linked by trade, it is so noticeable that
these other nations feel compelled, though without legal authority to do so,
to offer themselves as arbiters, and thus they indirectly prepare the way for
the great body politic of the future, a body politic for which antiquity provides
no example. Although this body politic presently exists only in very rough
outline, a feeling seems nonetheless to be already stirring among all its
members who have an interest in the preservation of the whole, and this
gives rise to hope that, finally, nature’s supreme objective—a universal
cosmopolitan state, the womb in which all of the human species’ original
capacities will be developed—will at last come to be realized.
(Kant 1983 [1784]:37–8)
34 Overcoming international relations

Since Kant’s world is dualistic, the moral reasons (noumena) can do very little
to change determinist chains of constant conjunctions (phenomena). The learning
process, republicanism and the civilising effect of commerce gave him some
hope. Ultimately, Kant had to rely on anachronistic teleological/Aristotelian
assumptions about the God-given telos of humanity. Often, he phrased this idea
in terms of the late eighteenth-century notion of Providence. Kant thought that
there was a guarantee for morality only in the Kingdom of Angels (absolutely
disembodied rational beings), that is, in Heaven (see Kant 1956 [1788], 121–4).
Repeating the dilemmas of St Augustine, Kant was doubtful about the prospects
of human progress on Earth. No wonder Kant’s theory appeared for later
German thinkers to be not particularly ‘realistic’.

The nineteenth-and twentieth-century set of solutions

Since Kant and the wars of Napoleon, different solutions to the international
problematic have been discussed, some of them in some regards original. For
Hegel, of course, Kant’s consciousness was unhappy. It was unhappy because,
ultimately, it had to turn to other-worldly solutions in its search for morality
and rationality. So Hegel wanted to show that the existing world is already
rational. At one and the same time, as a kind of a by-product, Hegel also
invented the post-Enlightenment solution to the problem of universalism vs
particularism: universal progress is carried out by particular nations, which often
fight against each other. The winner carries history forward. Later, the Hegelian
idea of Cunning of Reason assumed the form of social Darwinism.
It was in this intellectual context that the term realpolitik was coined in
Germany after the unsuccessful liberal revolutionary year of 1848. The
disillusioned liberals drew the conclusion that liberal thinkers should denounce
their ambitious programmes of change. Some of them developed the fundaments
of realpolitik (Palonen 1987:99–102).5 The contrasting Other of the realpolitik
theorists and practitioners was Kant and the ‘rationalist’ Enlightenment thinking.
They also read romantic, conservative and other anti-Enlightenment texts, as well
as Thucydides or Machiavelli and, later, Charles Darwin or Friedrich Nietzsche—
and used these texts to articulate their differences from liberalism and
Enlightenment ‘rationalism’. However, in fact their scepticism and Machtpolitik-
rationalisations never questioned Hume’s and Kant’s account of the phenomenal
world, only the Kantian possibility of overcoming it. In fact, this impossibility was
implicit in both Hume and Kant. Hume believed that ‘nothing can change’. Kant
thought that inter-state relations were in a Hobbesian ‘state of nature’ and that
moral reason was outside the phenomenal world and therefore causally impotent.
The ‘realism vs idealism’ debate thus started well before the formal Anglo-
American academic discipline of international relations was established in the
aftermath of the First World War. Although the native American Reinhold
Niebuhr came quite independently to similar kinds of conclusions in the 1930s
and 1940s, he was much better known as a theologian. Outstandingly, it was
 Genealogy of the international problematic 35

Hans Morgenthau who brought the German realpolitik discourse to the USA in
the late 1930s. Just listen to him in his most Nietzschean moment, contrasting
idealisations of ‘scientific man’ to the brute ‘realities of power polities’, when he
refers to Hume and, simultaneously, echoes Kant’s problem of the relationship
between phenomena and noumena:

Aristotle anticipated this modern problem, as so many others, when he

remarked in the Nicomachean Ethics: ‘Intellect itself, however, moves nothing’.
When rationalism was reaping its philosophic triumph, Hume could say:
‘Reason is, and ought only to be the slave of the passions, and can never
pretend to any other office than to serve and obey them’.
(Morgenthau 1946:154)

The nihilism and will-to-power metaphysics of Nietzsche was in fact already built in
Hume’s scepticism and notion of passion. In the twentieth-century political realism,
the standard reference point has been the synthesis of Max Weber (see Smith
1986). Weber combined ideas from Kant, Hegel, romantic and hermeneutic thinkers
such as Schleiermacher, nineteenth-century positivists such as Kelsen and Comte,
realpolitik, Nietzsche and the turn of the century positivism. Implicitly he adopted
the Humean account of phenomena and causality. In domestic politics, Weber was
a liberal, a sceptical and sometimes critical believer in modernisation as progress. In
international relations he articulated a version of political realism that denied any
progress and emancipation. Weber thought that outside of modern nations there
are no shared values. The result is a quasi-Nietzschean struggle of wills-to-power of
different national leaders. Thereby, Weber justified international ‘power polities’,
that is, the unilateral use of military force to achieve aims and to establish relations
of domination. Besides this Nietzschean, anti-Enlightenment strand, Weber’s
positivism also opened up the possibility of ‘scientific’ political realism.
We can name some of the twentieth-century solutions in terms of attitudes
towards relations of domination. These attitudes were originally identified by
Hegel, but have been further elaborated by Bhaskar (1993:2–4, 330–5; 1994:107–
8, 208–9). Whereas the stoic tries to be indifferent and neutral to the reality of
the world, the sceptic attempts to deny the world. The Beautiful Soul thinks it
is enough to pretend that Heaven is already on Earth, and act accordingly. The
Unhappy Consciousness seeks refuge in other-worldly asceticism or solace in
the projective duplication of another world, and after-life or beyond, where for
instance happiness will be in accord with virtue (as Kant believed). The existing
relations of domination are the problem. It is because of the lack of any notion
of emancipatory transformation in this world that they opt for the stoic, sceptical
or other-wordly solutions, to accommodate themselves with the world.
A twentieth-century stoic tended to adopt an attitude of quasi-scientific neutrality.
Here we have the later Morgenthau, who claimed that political realism is the
scientific theory of international politics, and the neo-realists, who have become
faithful followers of a modern, scientist Hume by borrowing their methodological
36 Overcoming international relations

ideas from positivist economics. However, the Morgenthau of Scientific Man versus
Power Politics was extremely sceptical about finding any scientific knowledge about
the world, because, ultimately, the world is romantically tragic and Nietzschean.
Nietzsche relied upon the Humean account of the phenomenal world. He believed
that Hume’s ‘English petit bourgeois’ scepticism could be re-written, more boldly,
in terms of the metaphysical voluntarism of ‘will-to-power’. Some post-modernists
are very close to repeating this pattern (for a close link between the Humean
notion of causality, scepticism, Nietzsche and forms of deconstruction, see Culler
1983:86–7). Finally, the Christian realists such as Niebuhr, and early Herbert
Butterfield and Martin Wight, seem to verge on an Unhappy Consciousness. To
justify their traditionalist Humean theories and deep pessimism, and their retreat
from the world of emancipatory politics, they have returned to medieval
metaphysics, to the ideas of St Augustine. The Christian realists see the origin of
tragedies of wars to lie in the original sin of humankind, in the nature of the
fallen man. Salvation must therefore be otherworldly.
Thus, the fault line of the international theory problem-field solution set
starts with Hume and Kant, as shown in Figure 1.1, and continues via Hegel,
realpolitik, Nietzsche and Weber to the twentieth-century self-declared realists.
However, as we now know, their ‘realism’ was in fact based on the Humean
rejection of realist ontology and the social metaphysics this denial served to
justify and naturalise. Moreover, Hume’s conception of human nature originates
in Christian theology. Most twentieth-century forms of political realism and
Kantian ‘idealism’ merely reproduce this theory/problem-field solution set.

The ‘great debates’ of international relations

According to conventional wisdom, there have been four ‘great debates’ in the
twentieth-century academic discipline of international relations. The inter-war
debate between Kantian ‘idealism’ and political realism led to the dominance of
the latter after the Second World War. The second, methodological debate
between traditionalists and positivists took place in the 1950s and 1960s.
Positivism won this debate, at least in the USA and Northern Europe. In the
1970s, liberals took up the notion of interdependence and attacked the
assumptions of political realism. In the 1980s, various forms of post-positivism
have risen to dispute the basic tenets of positivism.
Most aspects of these ‘great debates’ can be summarised in terms of a relatively
simply and highly repetitious cognitive structure depicted above and in Figures
1.2 and Table 1.1. To clarify the possibilities, it is useful to distinguish between
three Humes and two Kants. Hume believed in Newtonian science, but was in
his own political analysis a classical political theorist. Hume2 demands the
application of the experimental methods of Newtonian science to political science
and international relations, whereas the Nietzschean Hume3 radicalises Hume’s
scepticism and notion of passion. Similarly, there are also two Kants (a
Nietzschean Kant would be a contradiction in terms). Like Hume, Kant was a
 Genealogy of the international problematic 37

Figure 1.1 The solution set of the international problematic

classical political theorist. However, many scholars studying democratic peace

theory or, say, the effects of interdependence follow positivist research methods.
They want to find out whether the Kantian hypotheses hold.
Table 1.1 summarises the ‘great debates’ of international relations in these
simple terms. Obviously, there are other aspects to these debates. The point is
not to claim that this summary is exhaustive, merely that it captures surprisingly
well the essence of the ‘great’ debates of International Relations. Most of sovereign
nation-states have come into being after Hume and Kant, a few in the 19th
century, and many more in the 20th century. World War I triggered a series of
catastrophes, in the context of an unprecedented increase in productive and
38 Overcoming international relations

Figure 1.2 Three Humes and two Kants

destructive powers of modern organisations. Since the 1950s, these developments

threatened to make Kant’s gloomy prophecy true: ‘Perpetual peace will occur
only in the vast graveyard of humanity as a whole’.
The centre of international relations as a discipline has been in the ‘Lockean
heartland’ of globalising capitalism. The repetitious cognitive structure of the
international problematic has been maintained by a background theory, itself
essentially connected to a set of everyday practices. Particularly in the Lockean
heartland, it has been easy to conceive the calculative ‘English shopkeeper’ as
the normal man. Evoked by wars and violence, this bourgeois conception of
man has been surrounded by dismal discussions about original sin and irrational
passions. Until recently, the success of science and technology has been envisaged
from a Newtonian-Humean vantage point. Last but not least, the implicit social
ontology has tended to be atomist, with consequent strict inside/outside
dichotomies and the exclusion of outside beings as the norm.

Table 1.1 The repetitious cognitive structure of international relations

 Genealogy of the international problematic 39

When the Woodrow Wilson Chair was founded in 1918 in Aberystwyth,

Wales, the idea was to promote peace by means of research. The methods of
the inter-war era were classical, the approach legalist and the focus on the
League of Nations. Alfred Zimmern, [Link] Dickinson and James Shotwell
were replicating Kant’s arguments—and, implicitly, his antinomies—in a new
world historical context (Kant1). The story goes that in the 1930s, classical
political realists, such as Niebuhr and Morgenthau, challenged this orthodoxy
(Hume1 and Hume3). Powerfully boosted by the US financiers of the Cold
War, classical political realism began to dominate the study of international
relations. Kenneth Waltz (1959) and others, already in the 1950s, transformed
it into scientific neorealism (Hume2). It was in this context that peace research
came into the limelight. The ‘idealists’ were pushed to the margins in international
relations, but peace research continued the research agenda of the 1920s and
1930s by new scientific means (Kant2).
Disciples of classical political theory, such as Hedley Bull—whose 1977 The
Anarchical Society: A Study of Order in World Politics, also followed the substance of
Humean political theory—accused positivists of untenable methods and
irrelevance. Morton [Link] (1969), [Link] Singer (1969) and Oran Young
(1969) responded by maintaining that the traditionalists are in fact making grand
claims about empirical regularities but are unwilling or unable to test them
scientifically. Many positivists also found the retreat to metaphysical speculation
about human nature unacceptable. While traditionalism prevailed, for example in
Britain and France, positivism became the orthodoxy in the USA, Germany and
the Nordic countries. The interdependence school rose in the USA in the late
1960s and early 1970s to challenge political realism. In essence, the globalists were
arguing that the Kantian belief in the positive effects of free trade and spirit of
commerce was well founded. Others took up the idea that republicanism or liberal
democracy may make states peaceful, at least in their mutual relations (Kant2).
The debate between positivism and post-positivism is more difficult to
summarise in terms of this cognitive structure. Notions of critical theory, post-
structuralism and, in particular, critical realism take issue with the presuppositions
of the Hume—Kant theory problem-field solution set. By shifting the grounds of
the background theory, they also enable new ways of posing the central problems
and open up a new set of possible solutions. Nonetheless, many aspects of the
post-positivist challenge can be read back to the international problematic. Not
only may Nietzschean post-structuralism be easily married with classical political
realism (Wæver 1989) but radical postmodernism is often a mirror image of
positivism, presuming that nothing has any stable identity, including the sense-
perceptions and operations of mind, and therefore nothing can really be known
and very little, if anything, should be said (Dillon 1999, 2000). Moreover, many
less radical—or more commonsensical—Wittgensteinians, discourse analysts and
constructivists take for granted Kant’s dichotomy between noumena (reason)
and phenomena (world of causes and effects). They assume that social sciences
concern only reasons, which are embedded in language. Often this ontological
position is taken to support the claim that ideas matter (Kant1). Finally,
40 Overcoming international relations

post-positivism can also be a way of arguing for a return to classical normative

theorising (Hume1 and Kant1).

Conclusion: can it be overcome?

This chapter is not just a historical genealogy of the international problematic. It

is also a dialectical comment on it. I have written this genealogy from a perspective
that presupposes depth, differentiation and structuration, as well as absence and
emergence, of social realities. In effect, these are denied by the international
problematic. Every student of international relations may know what the Peace of
Westphalia (1648) symbolises. The modern ‘international system’ was born. Yet
they fail to acknowledge the gradual and complicated geo-historical processes
that took centuries to assume a form and shape of ‘international relations’. They
also fail to acknowledge the very specific ontological, epistemological and moral
assumptions that constitute the underpinnings of the international problematic.
Perhaps only a post-structuralist, such as Walker (1993:137–8), can appreciate
Kant’s—the arch-idealist’s—important role in putting together the components of
the international problematic. Moreover, once we have recognised the highly
repetitious nature of the resulting cognitive structure, stories about ‘great debates’
in international relations should appear in a very different light.
A dialectical comment can bring into light a general lack or invalidity of a
theory, or a theory/practice inconsistency. Both are abundant in the studies based
on the international problematic. Admiration of the Newtonian revolution
notwithstanding, most international theorists have eschewed systematic experimental
or empirical research. It is easier to make grand generalisations, based on a very
limiting set of ontological assumptions, in an armchair. It was only in the 1950s
and 1960s that a systematic scientific study of international relations became reality.
However, Bull (1969:26) was right to argue that ‘the scientific approach has
contributed and is likely to contribute very little to the theory of international
relations’. This is because the positivist conception of science is false and its tacit
ontology (atomism and its metaphorical and metonymical extensions to collective
entities) seriously misleading. Both are rendered plausible by a geo-historical context
dominated by commodification and practices of capitalist market society.
A dialectical comment can also assume a Derridean, Foucauldian or
Habermasian form (Bhaskar 1994:214–18). Along Derridean lines, I have already
showed (1) how it was the arch-idealist Kant who played a crucial role in
articulating the problematic, later reformulated as the opposition between ‘realism’
and ‘idealism’; and (2) that this ‘realism’ is in fact based on irrealism, secreting
an implicit realism of perceptions or ‘will-to-power’ or something equivalent,
depending on the particular manifestation of this discourse. A further Derridean
move could, for instance, focus on how and why reasons are presented as a
mere supplement to the causal phenomena. Why is there this binary opposition
that also forms a violent hierarchy, in which positive value is always accorded
to the way ‘things and people really are’ instead of reasons, although ultimately
 Genealogy of the international problematic 41

‘reality’ in this theory is mostly empty of content and structure? It is this order
of things that my dialectical comment aims to disturb. As Derrida puts it:

Deconstruction must, through a double gesture, a double science, a double

writing, put into practice a reversal of the classical opposition and a general
displacement of the system. It is on that condition alone that deconstruction will
provide the means of intervening in the field of oppositions it criticizes and which
is also a field of non-discursive forces…. Deconstruction does not consist in
passing from one concept to another, but in overturning and displacing a
conceptual order, as well as the nonconceptual order with which it is articulated.
(Derrida 1982:329)

Derrida (1982:329–30) argues further that there are always concepts and
meanings that are irreducible ‘to the dominant forces’ behind the international
problematic, and which have resisted that conceptual and/or non-conceptual
order. This is in fact the Foucauldian moment, entailing resistance to every
repression. Classics, such as Thucydides’ Peloponnesian Wars, can be re-read to
demonstrate how they are irreducible to the dominant understandings. Karl
Deutsch’s notion of pluralist security community, or Hayward Alker’s learning
and manifold methodological innovations, are examples of concepts and
resources that can be used for effective interventions in this theory/problemfield.
Derrida does not go far enough, however. There is also a quest for new
conceptual resources. Reality should be reclaimed, also by introducing negativity,
absence and emergence, and thereby real change. By positing reasons as causes,
we locate meaningful human activities in the differentiated, stratified and
structured world. Moreover, the intransitive and transitive dimension of research
should be distinguished. Thereby much confusion about the relationship between
subject and object of research can be clarified. By being able to identify the
internal and external relations between social/political theories and social realities,
without conflating the two, critical realism promises a new emancipatory
conception of social scientific research.
Finally, but crucially, the Habermasian aspect of a dialectical comment entails
a willingness to communicate with and listen to all. A comment would not be
dialectical if it did not engage with, and learn from, the others in the field.

Notes

1 About the concept of theory problem-field solution set, see Bhaskar 1994:10, 215–16
and 224.
2 Strictly speaking, the term ‘positivism’ is historically misleading, as it was invented by
Saint-Simon and Comte in the nineteenth century. In the eighteenth century, Hume
was quite rightly conceived of as an empiricist. With Bhaskar (1986:226), however, I
concede that ‘most of positivism is already contained [and] elegantly expounded in
the writings of Hume’, and that it is best to see positivism as a causally efficacious
movement of thought and cognitive structure, with a number of possible variations
42 Overcoming international relations

and permutations within it. Thus, in the twentieth-century economics (the positivist
social science par excellence), for instance, there has been a long-standing debate between
the rationalists, who have substituted empirical regularities with the illusionary certainty
of hypothetical mathematical-deductive connections, and empiricists, who are typically
involved in the much more messy and tricky business of statistical analysis (econometrics).
3 Two seminal accounts of these transformations are Karl Polanyi’s The Great
Transformation (1957 [1944]:33–102) and Barrington Moore’s Social Origins of Dictatorship
and Democracy (1969:3–39). England was a pioneering country, with the first phase of
transformations taking place in the seventeenth century and the second accompanying
the Industrial Revolution. According to Moore (1969:24–7), the consolidation of
capitalist property rights of the landlords and bourgeoisie went hand in hand with the
destruction of the medieval rights to land of the small peasantry. Polanyi (1957
[1944]:77) summarises the extent of transformations, enforced often by what he calls
‘inhuman methods’: ‘Eighteenth century society unconsciously resisted any attempt
at making it a mere appendage of the market. No market economy was conceivable
that did not include a market for labor; but to establish such a market, especially in
England’s rural civilization, implied no less than the wholesale destruction of the
traditional fabric of society.’
4 Polanyi (1957 [1944]:259–62) argues that England conducted something close to power-
balancing policies against the potential preponderance of any particular Continental state,
already 200 years before balance of power was established as a principle and system in
eighteenth-century Europe, and before Hume theorised it as a historical law or invariance.
Polanyi may, however, be a victim of the confusion of Hume. Hume’s claim about
balance of power as invariant conjunction confuses the practice of making allies (common
for instance in Ancient Greece, cultivated by modern England in the fear of the
Continental powers) and the empiricist understanding of the mechanical or Newtonian
‘law’ of balance of power. The idea of ‘balance’ is mechanical and presupposes familiarity
with the science of Galileo or Newton. Moreover, the term ‘states system’ occurred
only sporadically if at all before the Napoleonic period (Bull 1977:12), and the balance
of power system was not formalised before the Concert of Vienna in the nineteenth
century. Contrary to his own understanding, Hume was thus, in fact, a conceptual
innovator, and certainly not a mere passive observant of a universal ‘law’.
5 Although this is slightly out of the reach of my story, it is certainly not insignificant
that there were also quite different (pathological) reactions to these same developments.
By the time he had already renounced his academic ambitions as politically impossible,
in 1843, Karl Marx, in the face of censorship, had to move to Paris in order to continue
his journalistic activities. This was the turning point. Marx started to see radical
revolution as a necessary condition for any change in Germany. In 1845, he was
banished from Paris as a dangerous revolutionary. On the outbreak of the Revolution
of February 1848 he was banished from Belgium, and, in 1849, again from Germany,
and, in June 1849, again from Paris. After these traumatic events in his formative
years—he was thirty in 1848—he finally settled in London, where he wrote the bulk of
texts associated with ‘Marxism’. As a kind of mirror image of the conservative realpolitik
thinkers, Marx adopted the assumption that no peaceful changes are possible, only a
one-shot revolution. In 1895, [Link] (1968 [1895]:21) praised Engels by arguing
that ‘hatred for political despotism was exceedingly strong in’ Marx and Engels,
‘combined with a profound theoretical understanding of the connection between
political despotism and economic oppression’. Lenin, who was facing even more fierce
political oppression in Tsarist Russia (e.g. the killing of his brother), contended that
Marx and Engels ‘became socialists after being democrats’. However, it should be noted
that at this point Lenin, too, seemed to be also in favour of political liberties. See
Chapter 4 for remarks on the further developments, and Chapters 5 and 8 about the
political significance and far-reaching consequences of this multi-stage, pathological
learning process.
2 Learning from Alker
From quantitative peace research to
emancipatory humanism

Simple learning is goal-seeking feedback, as in a homing torpedo. It consists in

adjusting responses, so as to reach a goal situation of a type that is given once
and for all by certain internal arrangements of the [neural] net; these arrangements
remain fixed throughout its life. A more complex type of learning is the self-
modifying or goal-changing feedback. It allows for feedback readjustments of those
internal arrangements that implied its original goal, so that the net will change
its goal, or set for itself new goals….
(Deutsch 1963:92)

After the catastrophe of the Second World War, Kantian themes re-emerged,
in a positivist guise, within peace and conflict studies. Kenneth Boulding,
Karl Deutsch, Johan Galtung and many others hoped that positivist social
sciences would help us to overcome the international problems posed by the
threat of violence and war. Although the mainstream in the USA—powerfully
boosted by the US financiers of the cold war—believed in the eternal wisdom
of political realism, peace researchers thought that scientific research could
demonstrate that political realists were wrong. Even better, they envisaged a
more peacefully organised world based on scientific knowledge. Like modern
medicine, peace and conflict studies can help save lives. Given the spectre of
thermonuclear war, this task assumed the urgency of saving the whole of
human civilisation.
In the early 1960s, while international relations (IR) as a discipline was
dominated by the classical political realism of George Kennan, Hans Morgenthau
and Henry Kissinger, the behaviouralist movement was reaching its climax in
the social sciences (particularly in the USA and Northern Europe). Hayward
Alker shared the belief of positivist peace researchers that it was possible to
improve political conditions by means of knowledge produced by scientific
method. He also attacked the loose journalism of foreign policy analysis, as
well as the classical speculative style of writing IR, both of which were common
in the early 1960s. Despite the tenets he shared with the mainstream of peace
research, Alker has been exceptional in one crucial respect: he learned from his
trials and errors and, step by step, changed his direction.1
By showing how one can learn, indeed develop, radically new approaches,
Alker makes us see the false necessity of the rigidities of a given academic
43
44 Overcoming international relations

character and goal. Even more importantly, Alker provides critical resources for
overcoming the international problematic. Alker’s numerous writings—some of
which were collected together and published in 1996 as Rediscoveries and
Reformulations: Humanistic Methodologies for International Studies2–in fact come very
close to critical realism (CR), and in part are explicitly inspired by CR. At the
end of this chapter, I shall discuss some ambiguities concerning ontological
idealism/realism, and point towards possible further steps in this learning process.

Alker’s four phases

Alker’s scientific career began with measuring politics and analysing the United
Nations (UN). The angry idealism of Alker’s first phase had three major pillars:
the advocacy of modern, rigorous social sciences; a pluralist theory of conflicts,
according to which there are always many different actors and at least some of
the conflicts are based on misperceptions; and the belief in the potentialities of
the U N. Alker’s advocacy of rigorous, Newtonian social sciences
notwithstanding, even during this phase he always tried to take seriously the
claims of those who did not share his fundamental assumptions and concerns.
Alker saw open, critical discussions among representatives of different approaches
as a prerequisite for scientific progress. Reliance on a given scientific method—
whatever it may be assumed to be—was not enough for him.
The second pillar was his theory of conflicts—and particularly the cold war
conflict—that relied partially on the notion of misperception and partially on his
pluralistic account of social reality. Alker, together with Bruce Russett, asserted
for instance that ‘numerous uncertainties and misperceptions continue regarding
the nature of the distinct issues before the UN’. There are many different
political opinions, but they are often misperceived and their nature
misunderstood. Alker and Russett went on to ask whether ‘we can we find a
way of naturally and objectively summarising the issues before the General
Assembly without losing their specific content?’ (Alker and Russett 1965:9). It
is noteworthy that they indicated that the moral-political aim of their study was
the reduction of international tension.3 In arguing for this possibility, they built
on an underlying Grotian—although positivist and utilitarian—account of the
nature of international society, according to which rules and institutions matter:

Rules of order are often followed, in both local and international society,
because of various kinds of expected gains or losses. A reputation for morality
and law-abidingness can be useful, and the contrary reputation damaging.
(Alker and Russett 1965:146)

The third pillar of Alker’s first phase is related to the UN. The direct (even
if qualified) analogy between domestic party politics and world politics in the
General Assembly gives support to his trust in the positive potentialities of
 Learning from Alker 45

the UN. ‘The world could be thought of as a political system in which the
major blocs are analoguous to two parties that compete for the favor of the
uncommitted voters’ (Alker and Russett 1965:147). Further, Alker claims,
with Russett (1965:148), that ‘by providing a forum where the parties must
participate in a continuing electoral competition for the allegiance of the
neutrals, the United Nations performs a major function in preserving the
system’s stability’. To summarise, the moral-political idea behind Alker’s
behaviouralist phase was to try to overcome—or at least ease—the tension
between the blocs of the cold war with the help of a more adequately
functioning UN and by utilising the increased scientific understanding provided
by empirically oriented modern social sciences.
The problems of the post-colonial world come more deeply into this picture
only gradually. It was while teaching methodology as a visiting professor in
Santiago, Chile, in the early 1970s that Alker most fully opened his mind to
radically different epistemological perspectives, which he found to cut across
different geo-historical locations. In Salvador Allende’s Chile, Alker had many
discussions on the techno-rational domination aspects of rigorous and expensive
standards of scientific practice. Eventually, he made the observation ‘that
mathematical social science…distorts social relations to the extent that it
participates in and mirrors our alienation under “capitalism”’(Alker 1996:186).
At least in his later rationalisation, Alker refers to his lengthy 1974 article
Are there structural models of voluntaristic structural action?’ as summarising
a major turning point (but see also Alker and Christensen 1972). Nonetheless,
he, like Herbert Simon, still used, even if not exclusively, the criteria of validity
of the statistical tradition. That 1974 article ‘evoked considerable resistance’
among the causal modelling logical empiricists of the time and did not find a
publisher in the USA (it was published in the European journal Quality and
Quantity in September 1974). Alker argued that

humanistic scholars are often right (sometimes for the wrong reasons) in
objecting to statistical models of complex social behavior; but [I claim] also
that a faith in positivist approaches to explaining voluntaristic social action is
sustainable if an appropriate transformation of those approaches takes place.
(Alker 1974:199–200)

In other words, he argued that the faith in positivist social sciences has to be
reconciled with the social theories of more ‘humanistic, voluntaristic social action’
scholars, otherwise the faith in positivism is not justified. In his second phase,
culminating in the Chile years and lasting till 1976, Alker was gradually moving
towards thoroughly informational—and also hermeneutico-dialectical—ontologies
as well as towards his new, more late-modern (or post-modern) mode of
globalism. All the time, he was trying to account carefully for all of his
epistemological, methodological and substantial moves. In methodology, he
was mostly interested in artificial intelligence, qualitative-mathematical analysis
46 Overcoming international relations

of Prisoner’s Dilemma game play and global socio-economic and ecological

modelling. After having taken one more step towards a full-scale critical
reflectivism, he explained his position—in 1977, in the context of discussing
quantitative dependencia theories—of the mid-1970s as follows:

I usually argued that quantitative Marxian mathematical political

economy and qualitative, nonanalytic computer simulations like those by
Abelson, Boudon, and Brunner struck me as avoiding at least some of
these problems [of positivism].
(Alker 1996:187, n. 4)

When doing more concrete research, he still wrote about the evolutionary
possibilities of the UN’s conflict management system, but was also increasingly
concerned about ecological problems and large-scale violent conflicts such as
the War of the Pacific. By 1977, he had developed the basics of a new dialectical
approach to the study of world politics, an approach that was not only reflective
but also very original, even idiosyncratic. An outstanding paper from 1977, in
which he outlined many of his new ideas, was called ‘Can the end of “power
politics” be part of the concepts with which its story is told?’ Even though it
was merely an unpublished APSA paper,4 it was the first time he made public
the most radical results of his long rethinking process.
Alker’s third phase can be termed as ‘emancipatory humanism’ He
summarises his position by asking whether we can combine ‘the evident
ability of the greatest historical writers to catch our moral and political
imaginations, and comment profoundly on the choices before us’, with the
‘hard won professional commitment to falsifiable scientific theorizing’ (Alker
1996:270). Indeed, this is what he is now doing, trying to combine history-
based, moral-political visions with falsifiable scientific theorising. In
explicating his innovative position further, he also comments on Immanuel
Wallerstein’s challenging point that the heavily narrative accounts of most
historical research seems not to lend itself to quantification:

I do not think quitting is the right response to Wallerstein’s challenge…. Nor

do I think that ‘quantification’ is the only mode of formalization necessary
or appropriate for the logical and empirical rigor and tractability that
mathematical representations have given to so many of the natural and social
sciences. We must broaden and deepen the universe of scientifically relevant
modelling approaches appropriate for the formal analysis of interpretative
and theoretical world histories. Historical evidence, much of it textual, should
not a priori be reduced to quantitative time series, or otherwise ignored.
(Alker 1996:271)

This passage should give a good starting point for analysing Alker’s humanistic,
emancipatory approach to the study of world politics. In the following, I shall
 Learning from Alker 47

discuss Alker’s approach by suggesting lessons from his development. This

indicates two things. First, it reveals that, at the general level, I sympathise with
many of the ideas of the third-phase Alker. Moreover, I think that what Alker
has written—even if my interpretation of his ideas and development is a fusion
of horizons in a Gadamerian sense5—is a very important intervention in the
international theory/problem-field and its characteristic solutions.
There will be four lessons, and after them a more critical section on the
limitations and problems of Alker’s approach. The four lessons are not easily
expressible in one short sentence, and thus I have simply categorised them in
accordance with their topics. These topics are:

1 how to collect and analyse data;

2 how to analyse and help to solve collective dilemmas;
3 how to broaden, pluralise and deepen scientifically relevant epistemologies
and their corresponding modelling approaches; and
4 how to theorise history.

These lessons can and should be read both as intepretative guidelines to Alker’s works
and as methodological and theoretical arguments that are based on Alker’s works.

The first lesson: how to collect and analyse data

As Alker began his career as a mainstream empiricist it is appropriate to begin from

what is often assumed to be the basic operation of social sciences: empirical data
collection and analysis. It was Alker’s conviction in the 1960s that ‘statistics in
particular is an appropriate, reality-oriented, quantitative discipline for dealing with
many of the key international relations theory-building problems’ (Alker 1966:655).
Many considered Alker’s early studies to be among the most rigorous exemplars of
the scientific approach to the study of IR. Thus, the research designs of those early
studies of Alker were often also replicated. A few years later, this ‘self-critical
methodologist interested in what was “really” going on’ (see Alker 1976:46) found
himself dissatisfied with the practice of being replicated. Why? The replicators
seemed to lack any awareness of the vulnerability of simplistic data-theoretic or
modelling assumptions. ‘The more I have learned about the limitations of my early
work, the more attracted others became to it’, Alker (1976:50) commented ironically.
Should this kind of development be called scientific progress?6
What was wrong with his early work? In his Ph.D. thesis dealing with
voting behaviour in the General Assembly, Alker used multiple regression
analysis and factor analysis. Soon this approach turned out to be flawed, for it
could not answer such simple questions as ‘Did aid buy votes?’ or ‘Was aid a
result of a favourable voting behaviour?’ Furthermore, Alker realised that
‘separate regression equations might not add up causally’. His first response
was to develop more complex ways of analysing data, the core of which was
48 Overcoming international relations

constructed simply from the roll call votes of countries in the General Assembly
and from the (subjectively interpreted) situational characteristics of these
countries. As Alker learned that factor analysis, too, can have a causal
interpretation, he worked for a while on the assumption that empirical
phenomena are structured by some underlying causal models. Yet there seemed
to be something wrong with this approach, too, for the world political reality
was, at least apparently, able to transcend all attempts to specify a definite
causal model, with unchanging, lawlike co-efficients, of its causal interrelations.
This was the case despite the fact that the studies focused intensively on only
one functional area, collective security practices.
Perhaps there was something wrong with the most fundamental assumptions
concerning the nature of reality and data? Indeed, in Alker’s tenacious, self-critical
attempts to improve upon his past modelling exercises, there finally occurred an
explicit ontological shift from mechanical, Humean causality, first towards an
ontology of information-processing, characteristic of cybernetics and artificial
intelligence, and then, eventually, towards an ontology of conversational or discourse
analysis (see Alker 1986:2).
Most clearly, the ontological shift was evident in Alker’s new modes of
understanding what data are. For one thing, he began to rely on precedent logics
‘that are used like psychologies to interpret historical successes and failures’
(1976:51). The testing of hypotheses about precedent logics—what are the relevant
precedent cases and how do actors learn from them?—required a new kind of
data, namely verbal data. As a consequence of this and other similar
methodological shiftings, data was not seen any more merely in terms of
behaviour and (subjectively interpreted) situational characteristics. Eventually,
Alker generalised from his experiences—which made it rational for him to broaden
and deepen the notion of ‘data’—and concluded that ultimately our understanding
of the nature of data depends on our social ontology. According to the ontology of inter
subjectively constituted social entities, we have to have multi-perspective
descriptions of social events and characteristics:

Whether an act should be described as a threatening, promising, or mocking

action is determined by the perceptions, interpretations, judgments, commitments
and shared meaning conventions of the parties involved. The correct description
of a social action is thus more than a reliable convergence of coder and/or diplomatic
judgments; its meaning and identity is constituted by the multiple interpretative
perspectives of the principal actors in such events. Such interpretative
complexities, I believe, are an appropriate emphasis for an increasingly historical,
practical, institutionally-aware and internationally constituted political science.
(Alker 1991:5–6)

A related ontological and data-theoretical point is expressed in Alker’s (1996:186)

complaint that ‘mathematical social science…distorts social relations to the extent
that it participates in and mirrors our alienation under “capitalism”’. Alker
 Learning from Alker 49

refers here to the fact that the standard North American statistical practices
were—and still are—much too individualistic and mechanistic to account for
internal relations between entities. As Ollman puts it, positivist practices presuppose
that any social factor is always:

logically independent of other social factors to which it is related. The

ties between them are contingent, rather than necessary; they could be
something very different without affecting the vital character of the factors
involved, a character which adheres to that part which is thought to be
independent of the rest. [However,] in Marx’s view, such relations are
internal to each factor (they are ontological relations), so that when an
important one alters, the factor itself alters; it becomes something else.
(Ollman 1971:15)

Alker himself is not arguing that there are merely internal relations (as
some Hegelians—and even Ollman—might do). Rather, his point is that the
identities of social entities, such as rules, actors, institutions, are relational,
that social entities are interconnected at the ontological level. For one thing,
this view makes it possible to claim, as Bhaskar (1986:307–8) does, that
‘positivism at once naturalises and normalises things and reflects in an endless
hall of mirrors the [individualistic and atemporal] self-image of Bourgeois
Man’. Methodologically, the dialectical view evokes the claims that
quantitative, compositional statistical knowledge is no more objective than
the essentially qualitative, relational knowledge and that, rather, one should
accept that the standard statistical knowledge is conceptually and ontologically
secondary to the much richer relational knowledge (see also Chapters 4 and
5). However, let us hear how Alker himself explains and clarifies his position:

When both internal and external relations affect case relationships, case-
matching or ‘precedent-seeking’ efforts become more complicated and
historically more interesting. Similarity matching and dissimilarity contrasting
of procedurally specified data stories are ‘external’ comparisons unless the
attributes used in such efforts are in some sense essential, characteristic,
and case-identifying ones. Obviously one wants to relate essential
characteristics. The analytical problem, for which statistics can only be
moderately helpful, is to develop ontologically cogent distinctions between
essential and inessential properties of the case descriptions.
(Alker 1996:352)

The second major change in Alker’s understanding of the nature of adequate

data was more epistemological than ontological. Alker began to emphasise that
data—collected in accordance with a particular procedure—are always interpreted
by somebody from some particular perspective, and that typically the perspectives
of scholars are analogical with the perspectives of the involved political actors
50 Overcoming international relations

(with whom they more or less share some but not all formative perspectives).
Quite obviously, with the adoption of this kind of understanding, data lose
their seemingly innocent and neutral status. Consequently, one should get rid of
the reifying effects of the positivist social scientific practices. In line with Alker’s
new epistemological perspectivism and his lasting aim at constructing falsifiable
scientific theories, Alker advised, in 1977, scholars to follow the rule that

the findings of scientific experts with different doctrinal/value emphases must

be examined, checked, and compared at least to the extent likely regime
participants differ in their environmental assessments.
(Alker 1977:48)

Later on, he generalised and radicalised his position, without ever advocating
incommensurabilism or an absolutely relativistic ‘anything goes’ attitude. All of us
who are advocating modest relativism know that the incommensurability thesis, for
instance, is vulnerable to the problems of self-reference.7 Alker’s point has always been
to enable better scientific practices. In 1988, he stated his own ‘first lesson’ for empiricists as
follows:

Data-coding procedures should be considered key dependent variables in an

emancipatory peace research because they often (sometimes unconsiously)
reflect just those social and political forces affecting war and peace that are
supposed to be the objectives of investigation.
(Alker 1996:339)

He illustrates this thesis with a ‘true story’. In a pioneering North American

study on the quantification of levels of co-operation and conflicts in international
events data, out of the five student coders, one was a citizen of a Third World
country and a woman (the others were white men from the USA). Her codings
did not ‘reliably’ agree with those of the others and were discarded for most
purposes. She then went on to co-found a new important paradigm of conflict
research, lateral pressure theorising. The name of this woman is, of course,
Nazli Choucri (cf. Choucri and North 1975). This example illustrates how
apparently ‘unreliable’ ways of seeing may simply stem from different
interpretative perspectives that, moreover, typically correspond to the perspectives
of particular actors (and thus reflect different social forces). Moreover, to the
extent that the ‘unreliable’ ways of seeing are novel, they can also become
scientifically significant. Thus, lateral pressure theory, which was co-developed
by Choucri, ‘redefined the meaning and significance of imperialism as it had
been experienced by domestic populations within great powers, their unequal
allies, and Third World countries’ (Alker 1996:339).
According to Alker’s third-phase theory of data, relatively objective data
collection is possible if, and only if, the collectors are explicit and reflective
 Learning from Alker 51

about the case inclusion and exclusion rules for a data set. They should also be
explicit about the normative bases, procedural preferences and political allegiances
that inform coding practices. To emphasise: this kind of normative and
methodological explicitness is a prerequisite for adequate data construction. An
ontological parallel is that in the construction of a data set, one should include
the reasons actors give for the actions they take (or avoid) and the precedents
they cite (or avoid) for justifying their (in)actions. Moreover, one should also include
different narrative accounts of the relevant episodes, as presented by the actors.
The hermeneutically, by searching for a justified, sense-making account of the
whole, reasons, precedents, and narratives of actors can then be analysed both
and critically, by assessing the validity claims of the actors.
Finally, there is the question of deviant cases. Galtung (1977:72–95) has argued
that the study of the deviant cases should teach us the crucial variables of an
invariance and thereby help to break that invariance. Trying to go beyond this
formulation, Alker also demands that one should take the best and the worst cases
and use them to uncover the practical ‘grammars’ of action and habit that make
such outcomes possible. These ‘grammars’ should then be seen as in part changeable,
that is, as both potentially negotiable or debatable and therefore political.

The second lesson: how to analyse and help to solve

collective dilemmas

The Prisoner’s Dilemma (PD) is a game-theoretical model in which individual

utility maximisation appears to be self-defeating, or at least contradictory to, or in
conflict with, overall or longer-term public or collective rationality8. Many major
problems of political science have been scrutinised in terms of this model. These
include market ‘imperfections’, the ‘tragedy of global commons’, interstate insecurity,
arms races, crisis bargaining and also systems of subordination, which can be seen
as artificial and consciously structured systems of PD of the subordinates.
Alker, who entitled his 1976 autobiographical essay ‘Individual achievements
rarely sum to collective progress’, analysed the PD model for the first time in
his ground-breaking 1974 essay ‘Are there structural models of voluntaristic
social action?’ In this ‘Structural models’ essay, Alker tried to find less
reductionistic models than the linear or non-linear behaviouristic models, i.e.
he tried to find ‘multilevel models of voluntaristic social action systems’ that
could nevertheless be interpreted to be causally or ‘structurally’ deterministic.
The PD is so well known, at least in the context of mainstream political
science and IR, that there should be no reason to (re)present its general two-
person or no-person game matrices, game-trees or mathematical formulas. The
basic idea is that individual and collective rationality differ. In his 1974 article,
Alker’s main point was that it is possible to develop and partially to corroborate
more empirically adequate models of voluntaristic social action, in line with the
‘nearly pervasive concern of leading non-Marxian Anglo-American social science
theorists and philosophers’, but very much ‘unlike most social scientific statistical
52 Overcoming international relations

work that has emphasised simple specifications and often interpreted them
positivistically or behavioristically’ (1974:236). He argued, however, that these
‘empirically more adequate’ models of iterated or repeated games are necessarily
more complex than their predecessors. For instance, the artificial intelligence-
looking Emshoff model—in which a game has both a history and a future, and in
which actors are learning from the outcomes of their previous choices in a
probabilistic manner, according to a certain algorithm—is empirically more adequate
and a possible solution to the problematic of infinite reflective regress back to
mutual calculations. It is also a very different and much more complex solution
than the standard—but in most cases misleading—simple prescription of minimax
(guaranteed loss minimising) and maximin (secure-gain maximising) strategies.
In many real-world cases of collective goods, the assumption that PD games
have a history and a future, too, is adequate, but not in all cases. If the stakes
in a ‘game’ are extremely high, such as the loss of one’s job, years in prison,
death of a subordinate by the masters (perhaps in a concentration camp) or a
nuclear war [sic!], would not a single-choice ‘game’ be a more adequate model
than an iterated one? In his 1975 analysis of the descriptive foundations of
polimetrics, Alker turned, following Baumgartner and Burns, and Burns and
Buckley, to developing a new PD model by redefining it as a three-actor
domination system and not viewing it as an iterated game. In this way he was
able to analyse the logic as well as moral and causal limitations of ‘divide and
conquer’ or ‘divide and rule’ strategies of various rulers/masters of world history,
including those of the modern totalitarian nation-states.9
In contrast to the standard practice of presenting the PD in an abstract form,
Alker (1975:180–8) developed a three-party model in the context of the narrative;
it was probably developed after the original model with its numerical payoffs. In a
typical variant of the PD narrative, the position and power of the District Attorney
(DA) is not taken into account at all, although (s)he has the capability to get the
prisoners to do something they would not otherwise do. This capability is based
on the DA’s position in the system of legal and penalty practices, which enables
him or her to control communication. That control is crucial in this context in at
least two respects. First, the DA can use it (plus the absence of counsel) to decrease
the trust that Prisoner A has in Prisoner B. Second, he or she can use it to paint an
unconnected and rather black picture of Prisoner B, thus causing Prisoner A to feel
less loyalty to (which pre-exists contingently), solidarity with and sympathy for his
or her partner (see Alker 1975:180–8). The loyalty of prisoners might be, for
instance, due to the fact that they belong to the same community, movement or
organisation, which has particular normatively binding regulative rules of its own.
Seen in this way, the modality or technique of power on which the PD is based is
panopticism, described, modelled and explained by Michel Foucault:

Each individual, in his place, is securely confined to a cell from which he is

seen from the front by the supervisor; but the side walls prevent him from
coming into contact with his companions. He is seen, but he does not see; he
 Learning from Alker 53

is the object of information, never a subject of communication. The

arrangement of his room, opposite the central tower, imposes on him an
axial visibility; but the divisions of the ring, those separated cells imply a
lateral invisibility And this invisibility is a guarantee of order.
(Foucault 1979:200)

This socially produced invisibility of individual actors is also the crucial facility
for the power of the DA. This asymmetrical power would be diminished if
there were regulative rules such as ‘prisoners should be allowed to talk with
their attorney before interrogation’, or if the prisoners nevertheless considered
each other as loyal partners instead of strategic calculators. Indeed, even the
associated PD narrative is an extremely abstract interpretation, for it abstracts
this ‘game’ from the more contextual rules as well as from the questions such
as: Are the prisoner’s really guilty? Why are they committed to egoist means-
ends rationality? Why are they not allowed to talk to their lawyers before
interrogation (a talk with an attorney would provide the possibility of indirect
communication)? Consequently, can it be said that the legal system in question
is democratic or authoritarian, just or unjust? On the basis of these
considerations, Alker developed a complex model for estimating the power of
the DA and the factors giving rise to that power.
Alker did not stop developing his approach to resolving PDs to this still
somewhat positivist model of three-party system of domination (his criticisms
of positivism notwithstanding, he was still looking for a model of ‘rational’
incentives and, indirectly, for mathematically definable and analysable structural
invariances of social action). In Alker and Hurwitz’s 1980 ‘student manual’
called Resolving Prisoner’s Dilemmas, there is an explication of some of the substantial
and edificatory (cf. Rorty 1979:359) reasons for moving towards thoroughly
conversational and dramaturgical social ontologies. It is not only that empirical
and theoretical reasons seemed to be pointing towards more complex models of
intentional and voluntaristic—although in a sense also structurally determined—
social action. It is also the case that there are close connections between the
world understandings of different research paradigms—with their parallels in
the actors’ world understandings—and of actors’ abilities to resolve PDs. These
connections include the following (Alker and Hurwitz 1980:80–82):

1 Communication of co-operative intention increases an actor’s commitment

to it in ambiguous circumstances.
2 Players develop co-operative goals in the iterated PD as a result of a recognition
of each other’s behaviour as communication of motives, intention and expectations.
3 PDs are more often resolved when the game is redefined as a problem for
moral discourse, and even more in the line of Habermas: ‘PD becomes
problematic due to distorted, constrained, or suppressed communication.’

As these observations indicate, by making it clear that there is a form of

consciousness that is beyond ‘strategic rationality’, the reasons for moving
54 Overcoming international relations

towards more humanistic and late-modern, interpretative social ontologies can

also be explicated in terms of Hegelian dialectical development of consciousness.
The basic idea has been formulated by Frederick Olafson in one of Alker’s
favourite references:

The difficulty which mankind experiences and which it is able to solve only
very gradually and very slowly is one of achieving a conceptualisation of the
self that will permit that reciprocity to emerge as something more than a
kind of appendage to an already fully constituted self. The crucial insight on
which Hegel’s position rests has to do with interdependence of our concept
of the self and our concept of the other and it involves the claim that these
conceptions move forward pari passu until they reach a point at which an
underlying identity of the one with the other is grasped.
(Olafson 1979:241)

Olafson is very careful in pointing out that the underlying identity is not the
Hegelian absolute one, nor should it be seen as a deterministic outcome of an
evolutionary history. Rather it involves the constitutive reciprocity and potentialities
of change inherent in the Habermasian undistorted and unconstrained
communicative interaction.10 With these points as a background, one should be
able to understand Alker and Hurwitz’s concluding lines in 1980 on PDs:

As a formal noncooperative game or psychological experiment, the Prisoner’s Dilemma was

misdefined and/or underspecified. Surely this thesis follows from our own view of
PD experiments as moral dramas on scientific stages…. It [also] accepts at
least one implication of Plon’s critique as well: that formal PD games
underspecify or mispresent socially relevant class/domination relations. PD
avoidance and promotion are part of liberal capitalism’s self-justificatory efforts,
but they also transcend the politics of the contemporary era. One should really
study the interaction of rules of play, rules of the game and rules of the different
social orders in which the game is embedded. Our dramaturgical perspective
accepts the view that real life PD dilemmas need actively to be recreated.
Essentially they involve three or more unequally powerful actors following
incomplete scripts about their own moral choices, scripts whose lines are at
least partially revisable in the actual context of their recurrence.
(Alker and Hurwitz 1980:118)

In his 1980 article with James Bennett and Dwain Mefford (1980), which fuses
Alker’s interests in the UN’s collective security system and PD modelling, he
goes on to analyse security as a contradiction between individually reasonable
security-seeking practices and collective security. That is, Alker scrutinises
problems of traditional state security in terms of a PD situation. By building
upon an organisational model based on the Emshoff model, in which actors are
 Learning from Alker 55

learning from their previous choices, and by ending up with something that is
quite reminiscent of Deutsch’s well-known cybernetic model of foreign policy
making,11 Alker, Bennett and Mefford try to analyse not only the practical
intentionality of means-ends calculation but also the processes that define and
re-define political goals themselves.
By focusing their analysis on the cognitive processing within organisations,
Alker, Bennett and Mefford emphasise the importance of studying how
precedents or lessons from the past are initially acquired, and then subsequently
applied, in the practical activity of problem solving during the day-to-day conduct
of foreign policy. Their point is that an individual actor’s course of action is
principally determined by the prior construction of the set of possible initiatives
and consequences attributable to the parties involved. This is a cognitive process
that involves defining new situations on the basis of comparison with previously
experienced situations which exist as richly described incidents in individual
and collective memory. These incidents are typically structured as sequences of
action, as narratives, the elements of which are selected and linked together by
causal and intentional connectives to yield a plausible ordering or configuring.
From this perspective, there can be argued to be two ‘orders of constraint’ that
shape and select policy:

1 The first comprises factors that set a bound on the scope and diversity of the narratives
(or scenarios) the policy-making apparatus is capable of generating. In the
Foucauldian terminology, discourses are bound by the way they characterise
the field of possible objects, establish the way language functions with respect
to objects and position the speaking and acting subject. They are also bound
by the available set of narrative structures.
2 The second involves the mechanisms of choice that select and amalgamate the narratives
under consideration. In the Foucauldian terminology, this is a matter of the form
and way of localisation and circulation of discourse(s) within the foreign-
policy-making organisations. Power, also in the more traditional Weberian
sense—as the capacity of an actor positioned within a social relationship to
carry out his or her own will despite resistance—and organisational structures
come into the picture here.

Even though the original point of departure had been, a decade earlier, the
rather positivist study of collective security practices and iterated PDs, Alker is
here taking on board Foucauldian (Foucault 1991a:56–7; Foucault 1991b) and
Ricoeurian (Ricoeur 1984 and 1988) ideas in explaining changes and continuity
of foreign policy.
Also of great interest for an IR theorist is the result of Alker’s more recent
game-theoretical study entitled as ‘Fairy tales can come true: Narrative constructions
of prisoner’s dilemma game play’ (originally with Karen Rothkin and Roger
Hurwitz). The main claim is that the mechanisms generating reciprocity are not
the simplistic ‘tit-for-tat’ strategies discussed by Anatol Rapoport and [Link]
(1965) and, in an evolutionary fashion, by Robert Axelrod (1984).
56 Overcoming international relations

Rather, despite the physical communication barrier, the players attempt

practically to reason together. They reflectively explore their historically
developing situations, and frequently (re)construct their social relationships
and (eventually multi-move) strategies as the game proceeds, until a stable
pattern of play, based on mutually supporting interpretations, has developed.
(Alker 1996:320)

On the basis of Alker’s lessons, how should we analyse and help to solve collective
dilemmas? Acknowledging that there are many very different situations of collective
dilemmas, Alker’s methodological point is twofold. First, we should really study the
relation between the rules of play and the game, and the rules and the often
asymmetrical resources of different social contexts in which the PD game is
embedded, rather than abstract and reductionistic models of PD. Second, the focus
should be on studying the (logics of the) scripts about actors’ moral choices. The
precedential, narrative-embedded and dramaturgical lines of these scripts are up to
an extent revisable in the actual context of their recurrence—and thus potentially
political.

The third lesson: how to broaden and deepen

scientifically relevant modelling approaches

Recall now the response of Alker to Immanuel Wallerstein’s challenge, to the claim
that the heavily narrative accounts of most historical research do not seem to lend to
quantification. Alker argued, in accordance with the main results of his studies of the
late 1970s and early 1980s, that ‘we must broaden and deepen the universe of
scientifically relevant modelling approaches appropriate for the formal analysis of
interpretative and theoretical world histories’ (Alker 1996:271). From the early 1980s
onwards, Alker has followed his own suggestions and instructions rigorously. For
many in the mainstream of political science, IR and peace research, Alker’s two
hermeneutical-structuralist studies of this period may appear utterly incomprehensible,
or at least irrelevant to the concerns of ‘serious’ political science.12 In contrast to this
view, I would like to argue that these studies by Alker are among the most innovative
published IR works of the 1980s. In the ‘Two reinterpretations of Toynbee’s Jesus’,
Alker (originally with Lehnert and Schneider) poses the basic research questions as
follows:

In the light of both traditional and more recent scholarly approaches, what
are the appropriate, scientific ways to ascertain the meanings we and others
see and feel within the Jesus story? What motivating power or charisma
does this story, or ‘myth’, contain? What basic structure or structures, what
infectious, self-replicating, ‘viral’ qualities account for this power? Can we
replicably discover and empirically test some of the ways in which its deepest
structure has been rewritten into different ‘surface’ texts? Or do the ‘rewrites’
 Learning from Alker 57

that are discussed by many traditional scholars themselves differ in their

fundamentals? Which versions of the Jesus story, thought of as an imitable
hero story, have had more or less appeal in which personal, cultural, economic,
or political contexts, thought of in contemporary or historical terms?
(Alker 1996:105)

Alker, Lehnert and Schneider are looking for ‘analytically reproducible,

motivationally suggestive plot structures implicit in story texts’, with the help of
computer-assisted hermeneutical analysis. In other words, Alker, Lehnert and
Schneider ask: What are the cultural and/or universal factors that limit the
scope and diversity of the narratives that actors are capable of generating? The
fundamental aim is thus to find general, widely spread, emotion-generating
‘logics’ of narrative construction, or ‘scripts’, which are implicit in the way in
which elements of narratives are selected and linked together by causal and
intentional connectives to yield a plausible ordering or configuring, a story.
How should, and could, one answer the questions posed by Alker and his
colleagues? Assuming that motivationally productive, mimetic, plot-like structures
help to give meaning to children’s stories and religious and political myths
alike—and that they also structure the precedents constitutive of the problem
solving of the day-to-day conduct of foreign policy—the first scientific goal must
be an adequate and operational identification of these structures. Once identified,
their connections to the reflective monitoring of conduct and, more deeply, to
the lives of actors must be investigated. We need systematic coding procedures
for the identification of elements of narratives and their structures in texts.
Alker, Lehnert and Schneider state in the ‘Two reinterpretations of Toynbee’s
Jesus’ paper that ‘reasonably explicit and reliable heuristic and grammatical
coding rules exist for most of this process, but no completely programmed and
validated coding algorithms are available’ (Alker 1996:118). Although Alker,
Lehnert and Schneider use computers both in their coding and analysis of the
textual material, they make it clear that ‘computers can help focus, clarify,
criticize, explicate; they do not supplant the human interpreter’ (Alker 1996:137).
Model building in computational hermeneutics can become quite complicated.
Lehnert found 199 molecule-like structures, Alker and Schneider many more, of
which a computer search algorithm further identified about half as undominated,
‘top-level units’. Most of these were found to be connected with each other. Next,
central and transitional undominated plot units were organised as episodical
clusterings of the Jesus narrative. From the consequent clusterings schema, Alker,
Lehnert and Schneider formulated generalisations about narrative structures. They
also put forward some suggestions about how these—presumably universal—structures
are connected to the reflective monitoring of conduct and to the lives of different
actors. As fascinating as these suggestions and tentative generalisations may be,
there are only a few concrete, precise statements that an IR scholar or a peace
researcher could easily utilise. The 1985 essay on ‘Toynbee’s Jesus’ is best seen as
another innovative analysis of the research possibilities open to a post-positivist.
58 Overcoming international relations

However, for a peace researcher or an IR scholar, ‘Fairy tales, tragedies and

world histories’ is more concretely suggestive. This 1987 article—the idea goes
back to the mid-1970s—begins with the observation that, despite their obvious
differences, very similar configurative elements connect fictional stories and
traditional histories. Now, remember the claims (1) that ‘conflict and crisis, as
well as the steady state in which violence remains latent, can be partially
explicated in terms of dynamics which are approximated by the interactions of
artificial systems equipped with the story-building and story-appraising
characteristics’ (Alker, Bennett and Mefford 1980:195), and (2) that the stories
constituting foreign policies—as well the actions of political movements—are often
world histories. From this perspective, the importance of Alker’s attempt to
analyse and explain the appeal of great world histories should be obvious.
To delimitate his area of study, Alker chose three schools of writing world
histories (and futures) as his object of analysis. The first one is the global modelling
studies evoked by the Club of Rome’s amplification of debates occasioned by Jay
Forrester’s World Dynamics. The second one is the Annales studies of modern world
systems stimulated by Ferdinand Braudel’s exemplary writings. Finally, he also
analyses Marxist analysis of the long waves of capitalist development, as exemplified
by Ernest Mandel’s theories. Alker is not treating these studies merely as—potentially
policy-constituting—fairy tales, but, rather, he is trying to leave space for finding
out the relative sense-making and truthfulness of these three different styles of
history writing. Perhaps this also explains the choice of the textual objects of his
study, and the fact that he did not take up the neo-realist world historical fairy
tale of ‘hegemonic cycles’ for a textual analysis?13 At least he writes that:

I take it to be one of the reasons for the relative ‘success’ of such studies is the
extent to which the authors or their fellow workers have combined drama-
like readability with an impression of historical and/or scientific
trustworthiness and the quasi-metaphysical, almost inevitable grandeur to
which seekers after total truths (including myself) are especially susceptible.
(Alker 1996:268)

Alker is also perfectly aware that the determinism—and, yes, positivism—of the
studies of Jay Forrester, Ferdinand Braudel, and Ernest Mandel does not ‘leave
much room for individual or group responses to the enormous moral, economic
and political issues raised by their accounts’ (Alker 1996:269):

Life is not a myth or a fairy tale with a guaranteed happy ending; neither is
it an inevitable tragedy, one that encompasses all of Western civilization or
the human species. Nor are most political or cultural leaders successfully
heroic. Should one then refrain from attempting to give meaningful
interpretations to world history? Or should one try to refrain from being
‘ideological’ in making such efforts, if it is indeed possible to do so? Can we
 Learning from Alker 59

indeed refrain from mythical, poetic or moralistic and ideological elements

in writing scientific histories of the challenges, the limits and the potentialities
of our times?… If [some such argument] is accepted, then we may ask these
questions in another way: Is there some improvement possible in the way
historical accounts can be made testable as they approach value questions,
structural constraints and human choice possibilities? Is there a way of making
world historical accounts empirically revisable while at the same time allowing
them to have the reflective character and dramatic force of a tragic morality
play or the ironic happiness of a Russian fairy tale?
(Alker 1996:269–70)

The bulk of the paper is devoted to the analysis of narrative structures. The point
of departure is the way Vladimir Propp, in his analysis of Russian fairy tales,
reduces all of his analysed tales to a single structure, or schema, with two variations
of the middle-of-the-stories events. First, the setting, or the preparatory aspect of
a story, is defined. Then, the story develops because of villainy or because of
some kind of lack; through the completion of some difficult tasks, either the
villain must be overcome or the lack must be resolved, or both. In the middle of
the story, there are two possibilities: either (1) a ‘struggle, then victory’ sequence
of episodes, with the branding of the hero, a struggle between hero and villain,
the defeat of the villain (this is the peak of the narrative, perhaps with some initial
misfortune), the pursuit of the hero, as well as unfounded claims by false heroes;
or (2) a ‘solution to a difficult task’ sequence of episodes, with unfounded claims
by false heroes (the hero still at home), a difficult task, the branding of the hero
(by a princess, often in and/or before a battle), with a preliminary or almost final
solution possibly occurring (this is the peak of the narrative, perhaps with some
initial misfortune) and the pursuit of the hero—with the hero being rescued from
pursuit. Finally, the hero is recognised, he is given a new appearance or new
possessions, and false heroes, as well as any still unpunished villains, are dealt
with. The end scene is that the hero is married (to the princess) and/or ascends
to the throne, etc. This narrative structure can be generalised as a set of rewrite
rules for a simple story grammar, in terms of STATES and EVENTS, and THEN
connectives, and CAUSE connective.
This more abstract and formal way of expressing the generative structure of
narratives is argued to be better at conveying text composition syntax than the
semantic and pragmatic dependencies among story elements that are still present
in Propp’s formalisation. Nonetheless, Alker sticks to the formal rewrite rules
when he illustratively analyses the structure of Mandel’s Long Waves of Capitalist
Development. The point of this illustration is that we should measure ‘actual
historical sequences’ and ‘think of possible world historical developments in
terms of alternative story grammars’ because this kind of thinking and analysis
‘can enlighten and orient us toward a better future of constrained, but multiple
possibilities’ (Alker 1996:302). In other words, the analysis of narrative structures
should be brought back to the analysis of world historical episodes, tendencies
60 Overcoming international relations

and structures, in order to make models of these better, to make them more
imaginary and more sensitive to different world historical possibilities.
What else has Alker to say about how to broaden and deepen scientifically
relevant modelling approaches in the post-positivist era? In a paper from 1981,
when commenting more generally on the global modelling studies evoked by
the Club of Rome’s amplification of debates occasioned by Jay Forrester’s World
Dynamics, Alker (1981b:353) says that ‘too many people identify political/
administrative cybernetics with Forrester-Meadows world modelling’. They may
also identify it with sophisticated econometric modelling that includes, for
example, error feedbacks, or with the more complicated models of systems
theories. However, this perspective on global modelling is very narrow and
problematic for a number of reasons:

I want to correct such false impressions. Relevant contributions include

communications-oriented redefinitions of the essence of politics; the
reintroduction into logico-empirical inquiry of teleological ontology and
epistemology; emancipatory cybernetic hierarchies of knowledge; linguistic
understanding and action; scientific modeling of systemic reproductive and self-
productive processes, as well as pathological and healthy, even innovative and
creative transformations therein; and a non-economic vocabulary for evaluating
political successes and failures.
(Alker 1981b:353)

Alker not only points to the relevant social theoretical contributions for enriching
the potentialities of global modelling, and admitting that the relevant literature is
often ‘heavily historical and not easily formalizable’ (Alker 1981b:370), but also
outlines a new set of problems for those global modellers whose background is in
the cybernetic tradition. He terms this the dialectics of state formation in centre-periphery
systems. It is dialectical because the works within this problematic should also deal
with identity-involving or constitutive relations between centres and peripheries,
and between national parts and systemic wholes. The emphasis on state formation
comes already from Deutsch’s cybernetic- and communication-centred studies on
nation building. The idea that all this occurs in larger centre—periphery structures
came, naturally, from the Dependencia school and World System Analysis, and it
was supported by the (still popular in 1981) studies of dependency relations and
autocentric strategies for their reversal. However, Alker wanted to include other
themes as well, and to show how all these could be systematically modelled:

Bruno Fritsch’s concern with the increased capital formation requirements of

modernizing nation-states exploitatively interdependent with their environments
adds an important ecological moment to state formation problematique as
previously formulated…. The inner-relations of imperialistic powers and their
self-reconstituting, power-balancing systems also seem a particularly recalcitrant
 Learning from Alker 61

problem for the would-be narratively oriented global modeler…. If nations,

classes, or global systems can be identified in terms of their innermost,
constitutive production modes or organising principles, a fair approximation
to such identities can be obtained using incompletely prespecified metaprograms,
frames, demons, scripts, or production systems, as these procedural entities
are currently conceived in artificial intelligence research.
(Alker 1981a:372–3)

Despite his inspiring suggestions, Alker has never systematically shown, by doing an
exemplary study, how one could and should conduct more appropriate global
modelling. Perhaps Ashley’s 1980 book The Political Economy of War and Peace, which is
a brilliant work in the lateral pressure paradigm, comes closest to what Alker has in
mind.14 Nevertheless, his methodological works from 1977 onwards offer many valuable
hints, concrete suggestions and even detailed, although only partial, modelling exercises
for anyone interested in improving the past global modelling approaches. In addition
to the models of narrative structures, one can find among his works a refined, artificial-
intelligence-based analysis of the inner relations of imperialistic powers and their self-
reconstituting, power-balancing systems (Alker 1996:184–206); a case for a ‘neo-
classical polimetrics’, which is grounded in political argumentation about practical
choices in particular contexts, argued itself in the context of analysing and quasi-
formal modelling of the ‘dialectical logic of Thucydides’ Melian dialogue’ (Alker
1996:23–63; see also 1984); and exercises in linguistic, computer-assisted political
discourse analysis (for instance, Alker et al. 1991a; Alker et al. 1991b). These and
other related works may not amount to an articulation of a full-scale approach to
rigorous post-positivist global modelling, but they are important arguments about
how to broaden and deepen scientifically relevant modelling approaches.

The fourth lesson: how to theorise history s

Many of the points made in the previous three lessons are immediately relevant
from the point of view of the question ‘How do you theorise history?’ Next I
will now finally scrutinise Alker’s approach to theorising world history by
exposing the ‘Dialectics of World Order’ project, initiated by Alker and developed
in transnational co-operation with Tahir Amin, Thomas Biersteker and Takashi
Inoguchi.15 The ‘Dialectics of World Order’ project is an attempt to theorise
recent world history and possible futures in the dialectical and post-positivist
terms of Alker’s third phase.
The first outcome of this project was the publication of the article called
‘Dialectical foundations of global disparities’ (Alker 1981a). In that article, four
contending, inter-penetrating systems of global order are proposed as appropriate
units for an inquiry: capitalist power balancing, Soviet socialism, corporatist
authoritarianism and collective self-reliance. The fact that one can no longer be
content with some of Alker’s hypotheses does not invalidate his approach. On
the contrary, the far-reaching and unexpected transformations of the 1980s and
62 Overcoming international relations

early 1990s seem to confirm his fundamental—and well-grounded—assumption

that world history should be analysed (interpretatively) as an open process, in
which choices of actors do make a difference, sometimes also to the intended
direction. Furthermore, even though one should try to re-identify the relevant
world orders, many of Alker’s ‘hypotheses’ are still fruitful in investigating their
inter-relations. These include the following:

1 Every region of the world unequally reflects the mutually inter-penetrating

and (sometimes) opposed world orders.
2 Besides revolutionary class conflicts, two major determinants of the birth of
newer state types have been the world wars (the cold war included) and the
intense pressures for modernisation associated with relatively late development
in a world of more technically advanced, predatory and powerful actors.

‘From imperial power balancing to people’s wars: Searching for order in the
twentieth century’, a 1989 article written by Alker, Thomas Biersteker, and
Takashi Inoguchi, continues Alker’s interest in collective security practices, and
applies the same kind of thinking to analysing the development and relative
(in)validity of different modern models of peace and order. Alker, Biersteker
and Inoguchi (1989:135) limit themselves to an emphasis on three ‘partially
implemented, globally oriented security-seeking programs of the past century:
the Eurocentric balance of power, Wilsonian collective security, and international
socialist transformation through people’s wars’. The difference between the
analysis of ‘world orders’ and this analysis is that, while the former was also
concerned with the production modes that are also constitutive, in the latter
case only the organising principles are analysed.
The Eurocentric power-balancing system, which emerged in the eighteenth
century, became formalised in the nineteenth century and, in a modified
form, also constituted many cold war practices (cf. Patomäki 1992a:212–
14). The organisational principles of this model include the idea that states
are considered and treated as unitary actors, reciprocally recognising,
interacting with and differentiating among themselves. The system has
maintained itself by a mixture of force, the threat of the use of force, dynasty-
supporting marriages, alliances and diplomatic means. Further, unrestrained
power was seen as a threat to all other system members, hence the prescriptive
principles of action that collective power, through the use of alliances, should
be used against any state or bloc that became inordinately powerful
(‘balancing’ policies, in accordance with the central principle of astronomy,
mechanics and economics of the time). Both hegemonical aspirants and
their equilibrating antagonists were expected to stop fighting rather than
eliminate a core-state actor.
Alker, Biersteker and Inoguchi (1989) treat collective security simply as a
power-balancing alternative. They argue, quite provocatively, that ‘Wilsonian
collective security was not diametrically opposed to power-balancing, but rather
redefined and globalized an older Eurocentric power-balancing system’ (Alker
 Learning from Alker 63

et al. 1989:145). The only difference was/is that any threat to peace was/is
assumed to be of basic concern to all members of the international society.
Consequently, aggression was/is outlawed by a ‘balance’ of ‘all against one’.
However,

Wilson [also] called for a world of nationally homogeneous, self-determining,

market-oriented, trade-fostering, democratic states whose public opinion
would support the condemnation of future aggressors. This liberal image of
a cooperative world of rational, predictable people and states represents a
major change from the balance-of-power era.
(Alker et al. 1989:146)

However, this call for making the world homogeneous meant, in practice, that
war-avoiding doctrines and theories of collective security became coupled to war-
fighting practices and doctrines such as Wilson’s total war designed to end all wars.
Further, Wilson was not the only one who offered progressive, total alternatives
to traditional diplomatic practices; the proletarian internationalism of Vladimir
Lenin, Karl Kautsky, Rosa Luxemburg and other Marxists was at that time
Wilson’s major world political rival. While for Wilson ‘peace’ meant also the
continuation of the status quo, Lenin was advocating a major, world-wide socio-
economic change. Consequently, Leninism formed the background for those
from Mao and Lin Biao in China in the 1930s to the Sandinistas of the 1980s,
who have developed and propagated the conception of ‘people’s wars’. This
conception usually includes the belief that a future classless society will eventually
guarantee an eternal peace. Mobilising societal cross-class support to a—violent,
if necessary—revolutionary struggle for socialism would/should thus create the
basis for a lasting peace and order.
The analysis of Alker, Biersteker and Inoguchi has practical-political
implications. They demand more contextually sensitive and historically reflective
doctrines of collective security, ‘some less total, more realistic alternatives’ (Alker et
al. 1989:159). I would like to go further and suggest that we should also seek
alternatives that do not define or presuppose an eternal order that would amount, in
particular circumstances, to function as the status quo. Any status quo that would
be preserved with the help of Wilsonian ‘all against one’ powerbalancing policies—
even though this is perhaps possible only as long as somebody is able to orchestrate
and legitimate the actions of ‘all’—will eventually work against the basic underlying
principle, that is, peace. As will be argued in Chapter 8, contextual peaceful changes
should be made an institutionally guaranteed, always open possibility.
In any case, to the extent that one is willing to accept at least some of the
methodological notions that Alker has developed during his third phase, the
summary of these studies should illustrate how to theorise history in an
interpretative and heuristic manner. This concludes the fourth and final lesson
from Alker’s development. It is finally time to consider some limitations and
problems of Alker’s approach.
64 Overcoming international relations

Some problems and limitations of Alker’s approach

Up to this point, my story about Alker could have been read as a simple fairy tale.
However, as Alker has argued, life is not a myth or a fairy tale with a guaranteed
happy ending, nor is it an inevitable tragedy. We need more complex, more
interesting stories. On the one hand, Alker has already been recognised, even if his
arguments have been rarely followed; he was the President of the ISA in 1992–3.
On the other hand, although it certainly may still be plausible to envisage Alker as
a lonely, courageous and insightful figure, heroic in his attempts to re-think the
methodological foundations of IR and peace research, his constructive solutions to
the lack of convincing scientific, moral and political results are by no means
sufficient. Also, Alker’s stand on ontological realism/idealism is ambiguous.
An obvious limitation of Alker’s approach is the fact that he has been, after all,
mostly concerned with the problems of peace, war and (in)security, the only major
exception being ecological concerns (see Alker and Haas 1993). Those doing
international/global political economy or normative political theory may not be
able to find as many interesting things in Alker’s texts. Furthermore, in his third
phase Alker has also strongly emphasised communication, discourses and
narratives, which for some may indicate that Alker is not able to analyse hard,
‘material facts’ of life. This emphasis of Alker’s—and some contradictory
evidence16—notwithstanding, I think it is misleading to presume that Alker has
taken the side of the idealists in the debate between ‘idealists’ and ‘materialists’.
Yet, this is precisely the interpretation of Christopher Chase-Dunn (1989:31).
Chase-Dunn claims that Alker has ‘a cybernetic conception of systemic structure
in which social information becomes coded in cultural and symbolic systems’.
This, he claims, presupposes that Alker has adopted an idealist philosophy. The
accusation is that Alker may have ignored the role of work as material activity,
and also the role of material destructive powers of military systems. Alker qualifies
his metaphor or analogy in a manner that should make it clear that he is not an
ontological reductionist (‘idealist’):

‘Moments’ of social totalities might be thought of as aspects of societal

lifeworlds that are unified or ‘reciprocically’ and ‘internally’ related in historical
life-contexts by their essential (capitalistic?) generative modes of work, speech,
and control. These are not the ‘repetitive’or non-reflecting, lifeless systems
for which cogent empirical science is possible.
(Alker 1982:83, referring to the ideas of Habermas)

Despite the absence of political economy in Alker’s writings, I think it is fairer

to side him with the ‘critical scientific realists’ such as Rom Harré and Roy
Bhaskar than with ‘idealists’. Attempts to overcome the dualism between agents
and structures, ideas and matter, and superstructure and base structure, are at
the core of the CR social ontology. Consider, for instance, the following quotation
from Alker, which is an explicit commitment to this kind of thinking:
 Learning from Alker 65

One can take from Roy Bhaskar’s brilliant book, Scientific Realism and Human
Emancipation, a very clear statement of the guiding motivation of such a newer
conception [of the reality of things]: emancipatory peace research
epistemologically directs its ‘learning from the data’ toward the emancipatory
‘uncoupling (of) the present from the causality of the past’, replacing
‘depotentialising (disempowering, oppressive)’ psychological, social, and
ecological structures by ‘potentialising (empowering, enhancing)’ ones.
Conceiving of emancipation as a ‘special qualitative kind of becoming free’
that consists in the self-directed ‘transformation…from an unwanted and unneeded
to a wanted and needed source of determination’, Bhaskar argues that it ‘is both
causally presaged and logically entailed by explanatory theory, but that it
can only be effected in practice’.
(Alker 1996:354)

In a recent paper on ‘sequential prisoner’s dilemmas’, Alker sides explicitly with

realist philosophies of social enquiry (Alker and Hurwitz 2001). If Alker cannot be
adequately criticised because of his alleged ontological idealism (there will always
be determination and structures), one could certainly criticise Alker for having
never really gone through all the implications of his different methodological
propositions and game openings. Alker has worked like Foucault, who has happily
admitted: ‘My books aren’t treatises in philosophy or studies of history: at most,
they are philosophical fragments put to work in a historical field of problems’
(Foucault 1991b:74). Alker’s papers and articles, too, are philosophical fragments
put to work in a historical field of problems. Although there is clearly a family
resemblance between all of his post-1977 papers, articles, essays and books,
nevertheless there may be too much room for ambivalences, ambiguities and
contradictory interpretations. Of course, those comprehensive views that are able
to admit that there are holes, silences, contingent incommensurabilities,
ambivalences and ambiguities are more open to change and learning than those
that resist admitting this much (cf. Bhaskar 1989:155). The point is only to say
that there is a need to work out the ontological, epistemological and normative
underpinning more systematically.
From a wider perspective, perhaps the main problem with Alker’s approach
is that he has tied himself to such an extent to developing formalisable ways of
modelling social worlds. One may interpret this as a lasting trace of his early
belief in modern science, or one may see it as a fruitful commitment to thorough
and systematic analysis. In either case, the problem remains: as Alker (1981b:
370) himself admits, the relevant literature on world politics and history is often
‘heavily historical and not easily formalizable’. However, Alker is sensitive to
this problem. In one place he (1996:198) even defines himself as a ‘radical
traditionalist’ to indicate that he attempts a synthesis between the ‘traditional’,
epistemologically sceptical, and progressive, epistemologically optimistic (liberal
or Marxist), approaches. In addition to the ‘Dialectics of World Orders’ papers,
he has also written a non-formalistic, sympathetic paper on Hedley Bull’s world
66 Overcoming international relations

society problematic—a nice contrast to his 1966 article on the non-additivity

problem—arguing that Bull’s criticism of quantitative ‘cooperation under anarchy’
problematique has considerable force (Alker 1996:359).
In his 1992 Presidential Address called ‘The Humanistic moment in
international relations: Reflections on Machiavelli and Las Casas’, Alker (1992:347)
also advocates recovering ‘the humanist ideals and approaches which sometimes
get lost in our modern strivings for scientific rigor’. What does Alker mean by
‘humanist ideals and approaches’? He (1992:347) characterises them as the ‘critical,
interpretative, lesson-drawing disciplines centered around grammar, rhetoric,
history, and moral philosophy’. Furthermore, he connects the recovery of these
ideals and approaches to the beginning of the end of modernity and is willing to
anticipate subsequent eras of history. He should thus be seen as a late-modern
humanist who is nevertheless interested in developing rigorous methods.
However, the problem remains: there is a gap between what one can analyse with
the help of computer-assisted methods, (quasi-) formalities and rigorous models,
however humanistic these methods, formalities and models might be, and what is
‘heavily historical and not easily formalizable’. Alker’s partial answer to this problem
is his ‘Dialectics of World Order’ project, which shows a possible way of theorising
history in an interpretative and heuristic manner. In my view, this is not enough.
One should be able to go beyond this, to be able to find more general but pluralistic,
diversity-encouraging research methodologies for historical social sciences (as
Wallerstein or Giddens would call it). There are many hints at this kind of
methodology, like the following:

augmented by Rescher’s conception of dialectical reason: ‘Reasoning can

proceed not just inferentially’ (ampliatively from axioms) ‘but also dialectically’
(reductively, argumentatively) from a complex, contradictory set of plausible
initial positions. ‘There are two different sorts of cognitive disciplines—the
hard (e.g. physics), for which the mathematical (ampliative model [of
reasoning]) is doubtless optimal, and the soft (e.g. history), for which the
dialectical/reductive model is optimal’.
(Alker 1996:209, n. 3)

This might open a path towards a general methodology of iconic modelling. Iconic
models are based on the dialectical/reductive way of reasoning; concerned with
the explicit reflection on the metaphors and analogies used in formal and non-
formal social scientific explanations alike; able to incorporate the temporal,
narratively structured dimension of social worlds in them; and leave room for
both explanatory emancipation and imaginary edification. Although all evidence-
based interpretations of texts and explanations of social reality, when understood
in terms of iconic modelling, can only be trans-subjective and periodic, they
must also put forward much more general truth-claims, to be truly open to
critical dialogues with others. These ideas will be explicated in Chapter 6.
 Learning from Alker 67

Concluding remarks

As Alker’s development has been an intertextual journey through different

empirical research projects and theoretical texts, so we all are in the midst of
such a journey. Yet far too many seem to be afraid or incapable of travelling
very far. ‘It is safer to stay at home and in the neighbourhood’. It is thus not an
overstatement to call Alker’s voyage ‘exemplary’.
Alker’s development may be divided into three different phases. The
‘behaviouralist idealism’ of the first phase had three pillars: belief in science; a
particular theory of conflicts; and cautious belief in the potentialities of the UN.
Even at the risk of a teleological reading, Alker’s second phase is largely
understood here as a transitional one. Of course, at that time, Alker could not
foresee his future development. Yet he argued that the faith in positivist social
sciences had to be reconciled with the social theories of more ‘humanistic’
scholars, otherwise the faith in positivism was not justified. Finally, Alker’s
third phase may be called ‘emancipatory humanism’.
Alker’s long voyage shows how one can learn and develop new approaches.
Scholars should not see themselves as condemned to stick to doing research
within the approach they have once learned—often occupying a narrow position
in the highly repetitious international theory/problem-field solution set. There
is no reason to think that we are forced to stay where we already are, without
movement, without direction and occasional revisions of direction, without turns
and re-turns. Indeed, the story of Alker’s voyage should be able to reveal the
false necessity of building an academic home and then staying in the
neighbourhood for ever. Thereby this developmental story could even contribute
to inducing emancipatory changes in the field, to help to empower others to
start their own voyages, perhaps from where Alker has guided us.
In trying to build a sense-making narrative out of Alker’s writings, I classified
them in four groupings, each dealing with a question: (1) How to collect and
analyse data?; (2) How to analyse and help to solve collective dilemmas?; (3) How
to broaden and deepen scientifically relevant modelling approaches?; and (4) How
to theorise history? After having shown the way Alker developed, step by step,
new, humanistic, late-modern research methods and modelling possibilities, I also
discussed some of the weaknesses and limitations of Alker’s approach. Even though
it is of course true that Alker has not been able to resolve the methodological
problems of historical social sciences, he has been in the avantgarde of IR and
peace research. As argued in this chapter, the next task would be to articulate a
consistently realist ontology and epistemology, and a related methodology.

Notes

1 However, Galtung too has revised his ontological and epistemological assumptions;
see Galtung 1977 and 1996. For a critical realist attempt to develop these ideas further,
see Patomäki 2001a.
68 Overcoming international relations

2 Whenever possible, I refer to this book instead of the original papers, since many of
Alker’s papers were either circulated as unpublished mimeos or have been published
in places that are not easily accessible.
3 Alker wrote (1976:45): ‘My first “way of seeing” the UN political process was heavily
influenced by Dahl-Lipset-Haas type pluralism. And it was grounded in my personal
commitments. World cleavages might not be so cumulatively destructive if they did
not reinforce each other.’
4 Journal of Peace Research rejected it when Alker submitted it there! Eventually it ended
up in Alker 1996, as Chapter 5.
5 In the words of Gadamer (1977:94), the truth of hermeneutics is that it ‘allows
what is foreign and what is one’s own to merge in a new form by defending the
point of the other even if it be opposed to one’s own point of view’. It should be
noted that in the case of me interpreting Alker there are no really opposing
viewpoints, rather just differences of opinion and emphasis within a rather broadly
defined perspective.
6 In a later paper, Alker (1984:165) even points out that the technical advance in the
statistical programs of computers has been a disservice to the scientific development:
‘…it has left many with the false impression that experimentally oriented analyses of
variance, sample-based cross-tabulations, and regression analyses of non-experimental
data are—or should be—the analytical logics at the core of “polimetrics”’.
7 Neufeld (1993a:69) is thus simply wrong in his claim that Alker and Biersteker (1984)
‘accept that contending paradigms in International Relations are incommensurable’.
8 In the more technical terminology of the game theory, one says that in the Prisoner’s
Dilemma game the outcome of the individually rational choices is Pareto inferior (or
is not Pareto optimal). This means that there is an outcome in which both (or all)
players simultaneously could do better (for, moving towards a Pareto optimal state, it
would be enough for there to be one player who could do better while all others
would do so as well).
9 Note that, for Alker, totalitarianism is not a unique feature of Soviet or Chinese socialism
or of the Nazisms and Fascisms of Continental Europe (or of Latin America, Africa
and Asia, for that matter). Rather, Alker (1981a:91) is very explicit in saying that
‘more or less recessive variants of totalitarianism can be found in the state forms of
contemporary China, the Soviet Union, Germany, Japan, and the United States’.
10 Furthermore, the contextual Lebenswelt is in any communication situation as the ever-
present and taken-for-granted background that is ‘always already’ there when we act;
hence the attributes ‘undistorted’ and ‘unconstrained’ should be viewed to be
hermeneutically relative ideals, not absolute principles.
11 The ‘crude’—although rather complex—model of information flows in the foreign-
policy decision making can be found in the Appendix of the Nerves of Government
(Deutsch 1963:258–61); the basic assumptions of this model are discussed
sympatethically but reconstructively (from a more reflectivist position) in Alker
1981b.
12 Perhaps the bottom line for those not understanding Alker’s methodological, political
and moral development would be the following: ‘Our research activity has also been,
in a way, a Faustian romance: our entertaining commitment has been to unify a
Promethean science (Artificial Intelligence or computational text analysis) and the
ineffabilities of self-serving and self-transcending love’ (Alker 1996:141). For an
articulation of a similar, (late-) modernist moral vision that recasts the relation between
our ultimate ideals of love and empowerment, see Unger 1984.
13 Fortunately, this has been done by Grunberg (1990). She argues, after having tried to
show that ‘the theory of hegemonic stability fails to meet reasonable criteria of empirical
accuracy and analytic consistency’, that there are ‘mechanisms of mythic interference
with theory-building’ and that this interference can explain the rhetorical appeal of
the neo-realist hegemonic stability theory (Grunberg 1990:433–4).
 Learning from Alker 69

14 On the cover of the 1980 Ashley book, there is a brief introductory and evaluative
note by Alker: ‘If Choucri and North’s Nations in Conflict was arguably the best book in
the 1970’s produced by what Zinnes calls the Scientific Study of International Politics
(SSIP), Ashley’s major work may well secure that position in the 1980’s. At the same
time that it fulfills the promise of that school, it transcends its unreflective
epistemological narrowness, confronting and redefining the major political issues of
the day. Thus it confirms the frightening prospect that Choucri and North’s lateral
pressure perspective, developed in their analysis of the roots of World War I, also
applies to the current USA-USSR-PRC superpower competition. Changing modes
of powerbalancing are accounted for at the same time that Soviet “social imperialism”
is shown to be both imitative of earlier Western expansionism and contagiously
infecting the Chinese. Ecologically sensitive redefinitions of demographic-technological-
bureaucratic “growth”, not just clever power balancing realignments, are shown to be
prerequisities for a peaceful future.’
15 This is the name of the planned book or ‘an advanced text on theories of international
relations’, which, however, will probably never come out. Be that as it may, most of
the intended chapters have been published somewhere. In addition to Alker (1981a),
Alker and Biersteker (1984) and Alker, Biersteker and Inoguchi (1989), one should
mention at least Amin (1985).
16 In Alker (1986:2) it is stated that he shares ‘the ontological commitments of both
Plato and Leibniz’. There, Alker also talks about the ‘ideal reality of linguistic and
conceptual transformations and the ideal-material reality of social talk’. Talk is real
also in the critical realist understanding, as are linguistic and conceptual
transformations, but the ontological commitments of Plato and Leibniz are much less
easily reconcilable with realism.
3 How to tell better stories
about world politics

In the 1980s, there was nearly a consensus on at least one point, namely that
international relations (IR) contained three basic perspectives: (1) political realism;
(2) liberalism (seen also with varying emphasis as idealism, or pluralism or
economic liberalism); and (3) marxism (seen also, again with varying emphasis,
as structuralism or as a socialist model). Each of these holds a characteristic
view of international politics. In spite of a century of sustained effort, it is not
unfair to say that these perspectives continue to exist only in a rather speculative
outline. Also, as far as (1) and (2) are concerned, they mostly repeat the cognitive
structure of international problematic that was articulated by Hume and Kant,
and established in the nineteenth century.
Should we be happy with this ‘pluralism’ and continue the ‘great debates’ in
the field? Or is it possible to imagine the prospect that one of these rather
sketchy perspectives will eventually emerge as the accepted truth in this area of
analysis? In recent years there have been a decreasing number of self-proclaimed
Marxist scholars around, with the notable exception of the field of international
or global political economy.1 In the 1990s, after the short visit of Marxism in
the mainstream of IR, there has been, perhaps more than ever, a tendency to
reduce all problems of IR to an almost eternal dispute between political realism
and liberalism. Many aim at a synthesis of these two perspectives, a new mixed
perspective that could transcend some of the ‘great debates’ and contribute to
intellectual progress in the field.2 Postmodernists’ protests aside (cf. Ashley and
Walker 1990; George 1989), should we, as IR scholars, have this kind of
consensus as the ultimate telos of our research?
It is true, as Rob Walker (1987, 1989) has argued, that the IR perspectives
are also historical sites for a number of theoretical and metaphysical disputes.
These (meta)theoretical debates have, among other things, been about the
relationships between history and theory or history and structure (Buzan and
Little 1994; Schroeder 1994). IR scholars have discussed, for instance, whether
we should prefer the classical approach, emphasising history and conceptual
analysis, or the more strictly scientific approach, seeking to find behavioural
invariances by strictly controlled deductive or statistical means. Despite the
apparent depth of some of these conversations, mostly they have contributed
merely to the reproduction of the international problematic.
70
 Better stories about world politics 71

In the 1980s, however, the critical realist agency/structure debate was

introduced to IR. Ahead of its time, Richard Ashley’s celebrated essay ‘Poverty
of Neorealism’ (1984) discussed agency and structural determination in some
detail (cf. Patomäki 1989). Alexander Wendt’s article in International Organization
(1987) remains, for good reasons, one of the most often cited contributions. A
few years after this article, Wendt (1991) reviewed Martin Hollis and Steve
Smith’s book Explaining and Understanding International Relations (1990). The review
led to a debate between Wendt and Hollis and Smith.3 I use this debate as a
starting point for answering the question: Could these metatheoretical debates
open up new and more fruitful horizons for writing research-based, rational
and falsifiable, yet reflective stories about world politics (cf. Alker 1996:270)? In
other words, is it possible to tell better stories about world politics? How?
The debate of Wendt (1991, 1992b) vs. Hollis and Smith (1991, 1992),
continued later for instance by Smith (1994), Walter Carlsnaes (1994), and
Viviennejabri and Stephen Chan (1996), is an excellent case for my purposes.
It summarises the main points about agency and structure, and, also, the more
positivist levels-of-analysis problem. At the same time, it raises important questions
concerning the explanation vs. understanding controversy. There is also the
dialectical hope that contradictory claims in a critical dialogue could be genuinely
constructive. The result of a dialectical exchange does not consist of purely
negative or contradictory countermoves; it advances the discussion and shifts
the issue onto more sophisticated ground (see Rescher 1977:66–8). Up to a
certain, if limited, extent, this has also happened in the Wendt vs. Hollis and
Smith debate; at least it is recognised as a norm in it.
Two basic issues are raised in this debate. First, it is asked whether there are two
stories to be told in the study of IR, the one causal and the other interpretative, or just
one, a combination of these two. The second disagreement is about the relationship
between ontological assumptions and the basic concepts of causal explanation, conceived
in terms of the levels-of-analysis problematic. In the course of the debate, Wendt states
that he has become quite persuaded by Hollis and Smith that there often are two
stories to be told. However, he maintains that the levels-of-analysis problematic is
meaningful only as far as one is willing to take actors as constant in their inner
structure or as ‘exogenously given’ (Wendt 1992b:182). He argues that:

we can best avoid confusion if we reserve levels of analysis talk for questions
about what drives the behaviour of exogenously given actors, and agent-
structure talk for questions about what constitutes the properties of those
actors in the first place.
(Wendt 1992b:185)

Wendt claims that to construct system-level theories, one must take into account
how actors constitute themselves in interaction (cf. also Wendt 1992a). Thus,
Wendt wants to restrict the applicability of ‘levels of analysis talk’. Wendt assumes
that the actors in the international system are sovereign states. For Hollis and
72 Overcoming international relations

Smith, however, both system and state ‘levels involve questions about the nature
of agency’ (Hollis and Smith 1992:188). Indeed, at any given level of analysis,
the question of ‘what it means to be an actor’ emerges. On these grounds, they
argue that Wendt’s distinction between explaining the behaviour of the units,
on the one hand, and what, on the other, constitutes their properties, is essentially
empty. They use these points first and foremost to defend the general applicability
of the traditional levels-of-analysis thinking in IR.
After these two rounds, Carlsnaes and Smith (1994) continued the debate.
Whereas Smith (1994) tends to repeat his (and Hollis’s) earlier claims, Carlsnaes
(1994:277–87) defends Wendt’s original contention that we need to take into
account both inside (‘understanding’) and outside (‘explanation’) factors in our
analyses. Carlsnaes attacks the either/or character of the two-stories view by
distinguishing between philosophical theories of the relative weight of agential
and structural factors and empirical studies of the linkage between the two. He
combines these points in his call to the discipline ‘to integrate the explanatory
and understanding traditions within social science as well as to resolve the agency-
structure problem tout court’ (Carlsnaes 1994:282). Carlsnaes’s own contribution
to this goal is found in his argument that we should distinguish between three
explanatory levels: accounts of the ‘in order to’ intentions of foreign-policy actions;
explanations of why actors have these particular purposes (in terms of ‘because-
of accounts of the dispositions of actors); and explanations of how the structural
environment of actors (both domestic and international) affects the two levels.
He argues that if we view intentions, dispositions and external structures as a
reciprocal process of agency-structure interaction over time, we should be able
to build more adequate models for the analysis of foreign-policy behaviour (he
has explicated his own suggestions in Carlsnaes 1992 and 1993).
In what follows I shall continue and re-direct this debate. First, in line with
what Hollis and Smith as well as Wendt have indicated, I maintain that questions
concerning the existence of social entities and causal explanation are, indeed,
very closely related. However, I will go on to argue that it follows from the
critical realist notions of causality and social ontology, which I discuss tentatively
in this chapter, that the levels-of-analysis problematic cannot be plausibly
defended. It obscures more than it clarifies. Even though it is natural for a
critical scientific realist to think about levels in ontological terms, I contend that
‘international’, ‘inside-of-a-nation-state’ and ‘individuals’ are not levels in any of
the usual senses. Rather, we should talk about different kinds of interpenetrated,
relational contexts and look for depth in other directions.
Second, although Wendt states in his reply that he is now persuaded by
Hollis and Smith about the relationship between causal and interpretative
analysis, I shall defend and radicalise Wendt’s original position;
understanding and explanation are complementary and interdependent modes
of analysis. In fact, it is quite easy to show how explanation and understanding
are inter-related. Hermeneutic understanding is necessary for causal analysis,
because reasons are causes, yet that kind of analysis is not in itself enough.
Explanation presupposes holistic depth analysis of causes, including structural
 Better stories about world politics 73

causes. Yet, explanation is a mode of interpretation, and also aims at (better)

understanding.
This chapter suggests that the agency-structure debate can be a way out of
the increasingly outmoded ‘great debates’. Indeed, it is possible to envisage
novel interpretative possibilities by re-defining, ontologically, the way that IR
discourses characterise the field of possible objects, establish the way language
functions with respect to objects and position the speaking and acting subject.

Against the levels-of-analysis problematic

In his review essay, Wendt (1991:387) criticises Hollis and Smith for ‘conflating two
distinct problems: the levels of analysis problem and the agent-structure problem’.
Wendt goes on to claim that the levels-of-analysis problem ‘is a problem of
explanation: of assessing the relative importance of causal factors at different levels of
aggregation in explaining the behaviour of a given unit of analysis’ (Wendt
1991:387). Both in David Singer’s original formulation (1961) and in Wendt’s
reformulation of it, the given unit of analysis, as well as the ‘dependent variable’, is
the behaviour of the nation state—in particular its foreign policy. The explanatory
variables can be located at the level of ‘system’, ‘states’ or ‘individuals’, or viewed as
a combination of them. Wendt (1991:388) claims that Hollis and Smith have
changed the issue by taking the phenomenon to be explained to be either the
behaviour of the state actors or the behaviour of the international system, or even the
system or units themselves. By stipulating that the international system is real and
perhaps also able to behave, Hollis and Smith are effecting a shift from the original
behaviouralist research focus towards an ontology, towards the problem of emergent
collective properties and their explanation. Hollis and Smith reply as follows:

Briefly an explanation of state behaviour may or may not explain the state
away, depending on where one locates its sources—a theoretical as well as
empirical question involving ontology. But a general ontology is never handed
to the theorist on a plate, so to speak, and its justification is connected, although
indirectly, to explanatory merits. The level-of-analysis problem in international
relations cannot be tackled in isolation from the view taken of what is meant
by international system, and that is an ontological question.
(Hollis and Smith 1991:395)

What Hollis and Smith seem to be arguing here is that any attempt to explain
something in sworld politics necessarily implicates hypotheses concerning the
existence and nature of certain social entities, particularly the international system
and different ‘units’ (states, bureaucracies, individuals, etc.). Hollis and Smith are
not claiming that one cannot make different ontological assumptions in constructing
explanations at a given level-of-analysis, only that all explanations involve some
existential (or ontological) assumptions. This is a very important point.
74 Overcoming international relations

What then is Wendt’s complaint? For one, Wendt may think that Hollis and
Smith are in fact making unacceptable ontological or existential assumptions. Does
the international system behave? If it does not, perhaps it then has characteristics
that can be explained? But can we explain these properties of the international
system merely in systemic terms? Although a Waltzian would maintain a clear
separation between theories, respectively, of international politics and foreign policy,
the ‘dependent variables’, such as stability and coalition formations, seem to be
outcomes of foreign-policy making. As Wendt (1999:12) later explained, it ‘is
impossible for structures to have effects apart from the attributes and interactions of
agents’. For instance, ‘stability’ is often conceived as the opposite of violent conflicts
and war. Even in a step-by-step escalatory process, some foreign-policy actors must
decide and wage war. To avoid this conclusion, Waltz (1979:161–2) later qualified
his views by arguing that international stability means only two things, namely that
‘anarchy’ (=non-hierarchy) prevails and that the number of principal parties that
constitute the ‘system’ remains unchanged. This move makes the Waltzian ‘theory
of international polities’ either trivial or absurd. Besides coalitions, Waltz’s main
explanandum is stability. Waltz’s theory tries to predict the number of great powers
by referring, tautologically, to the number of poles in a system of power balancing.
I consider this to be the reductio ad absurdum of Waltz’s neo-realism. It also constitutes
a prima-facie case for Wendt’s contention that we should focus on foreign-policy
actions and their (perhaps constantly reproduced) outcomes, including war, peace,
state relations and regimes of co-operation.
For Wendt, then, what IR perspectives typically try to explain are the foreign
policies of states and their consequences. This is the case even though he takes
a ‘systemic view’, that is, he treats the identities and interests of states as analogical
to those of unitary, yet socially constructed, individuals who are continuously
reconstructing each other in the course of their systematically structured
interaction (Wendt 1992a; cf. Wendt 1994). Hence, given Wendt’s anti-Waltzian
ontological presuppositions and his ‘systemic’ theoretical aims, his complaint
might be simply that Singer’s original formulation of the levels-of-analysis
problem is superior to that of Hollis and Smith. This is due to the more plausible
existential hypotheses and a more clearly formulated research problem.
Yet there is something more to Wendt’s criticism, even though he does not
go very far in articulating it. Wendt (1991:388) might well be saying that he
prefers the agency-structure problematic to that of the level-of-analysis
problematic. If so, he should develop his case further. In particular, it is my
claim that, as a critical realist,4 Wendt should suspect that there is something
wrong with the whole levels-of-analysis problematic, since it is based on a
positivist (or empirical realist)5 account of science. It is not surprising that Hollis
and Smith adopt the standard positivist account of explanation and causality.6

The ‘outside’ way of accounting for behaviour is modelled on the methods

of natural science and is usually described as a search for causes. To explain
an event or state of affairs is to find another which caused it. This bold
 Better stories about world politics 75

statement conceals much dispute about the exact relation between a cause
and its effect, about the right way to define ‘cause’, and about the nature of
causality, both as a concept and in the world. So what follows is very
preliminary. But the broad idea is that events are governed by laws of nature
which apply whenever similar events occur in similar conditions. Science
progresses by learning which similarities are the key to which sequences.
(Hollis and Smith 1991:3)

In other words, Hollis and Smith agree with Singer that explanation is about
finding law-like regularities at work, and hence that explanation follows the so-
called deductive-nomological model of causal explanation.7 Hollis and Smith are
also of the opinion that explanatory theories should, above all, be parsimonious.
They criticise David Dessler for claiming that ‘the richer and more comprehensive
the underlying ontology, the better the theory’ (Dessler 1989:446). In this they
follow positivism, which has always emphasised simplicity. As Singer puts it:

Our model must have such analytical capabilities as to treat the causal
relationships in a fashion which is not only valid and thorough, but
parsimonious; this latter requirement is often overlooked, yet its implications
for research strategy are not inconsequential. It should be asserted here that
the primary purpose of theory is to explain, and when descriptive and
explanatory requirements are in conflict, the latter ought to be given priority,
even at the cost of some representational inaccuracy.
(Singer 1961:79)

These presuppositions explain why Hollis and Smith are willing to define the field
of IR in terms of the levels-of-analysis problematic. As far as the outsider’s story is
concerned, they think—with Singer—that the primary purpose of theory is to explain
phenomena in accordance with the deductive-nomological model, and in as simple
a manner as possible. They also seem to contend that this kind of explanation
amounts to assessing the relative causal importance of different levels of analysis.
Hollis and Smith do not question that empirical invariances are necessary
for causality. They should. Like in other social sciences, there are very few, if
any, uncontested invariances in IR.8 The fact that mainstream scholars go about
their business as if this were the case is no argument. As Friedrich Kratochwil
(1993:68) has pointed out, ‘activities in the profession are more often driven by
methodological fads and clientalism than by problems and sustained attention
to a research programme’. Moreover, these fads tend to remain within the highly
repetitious cognitive structure of the international problematic. The absence of
universal regularities does not, however, mean that there is no causality. Perhaps
there is something wrong with the prevalent notion of causality? Perhaps we
should qualify the Humean causality by making it more complex and contextual.
Instead of simple universal regularities, perhaps we should be searching for
more conditional regularities?
76 Overcoming international relations

In response to this [absence of universal empirical regularities], positivists

tend to argue that the social world is much more complex than the natural
world (‘interactionism’, already prefigured by Mill) or that the regularities
that govern it can only be identified at a more basic level (‘reductionism’,
prefigured by Comte), and that, in any event, concepts (or meanings), to the
extent that they are explanatorily relevant at all, can only be identified, or
hypotheses about them tested, empirically (that is, behaviourally). Neither
party doubts for a moment that empirical invariances are necessary for laws,
or that the conceptual and the empirical jointly exhaust the real.
(Bhaskar 1979:22)

Even after the end of the Cold War, some neo-realists have opted for the stubborn
Humean simplicity (cf. Mearsheimer 1990; for the criticism that neo-realism cannot
explain or even describe the end of the Cold War, see Gaddis 1992; Patomäki 1992a).
However, we know that neither the Millian (‘interactionism’) nor the Comtean
(‘reductionism’) strategy has been very successful during the past century and a half.
To import complexity and contextuality into IR is certainly a better solution than
sticking to the simplicities of failed Humean theories of the past. On the other hand,
after all these years of unsuccessful attempts of finding invariances, perhaps there is
room for more radical solutions? Perhaps empirical invariances and causality are two
totally distinct phenomena? Even more importantly, perhaps there are conceptions of
causality that explain this absence of universal empirical regularities?
It can be easily shown that the idea that causality involves empirical invariances
presupposes closed systems. Constant conjunctions of events obtain only in the absence
of outside intervention and qualitative variation (see Bhaskar 1979:160). Where
a genuine, non-changing causal mechanism has been isolated, can we say that ‘every
time {A, B, C…} occurs, X follows’. However, entities do change and systems are
more or less open except when carefully set up experimental conditions obtain.
In fact, ‘if we retain a Humean definition of causation as regular succession, we
will discover no causal laws outside astronomy, where the incapacity of other
mechanisms to deflect heavenly bodies from their courses approximates to a
natural closure’ (Collier 1994:34). In open systems, probabilistic empirical
invariances that seem to connect phenomena are often causally unrelated. This is
the well-known post hoc ergo propter hoc—fallacy.
A non-Humean, critical realist conception of causality would take as its
point of departure the idea that real causal mechanisms and complexes produce
effects in open systems. From this perspective, it is wrong to assume that causality
can be analysed in terms of simply necessary or sufficient conditions, yet this
is presupposed if one talks about empirical invariances. Causality should be
understood not only in terms of complexes, but also in terms of the subject
matter in question. The modified definition of cause as a so-called INUS-
condition (see Bhaskar 1979:165, 207 n. 23; Mackie 1976) is a leap forward
According to this definition, cause is an Insufficient but Non-redundant
element of a complex that is itself Unnecessary but Sufficient for the production
 Better stories about world politics 77

of a result. Consider the example [Link] (1976:308–9) discusses. Experts

investigate the cause of a fire that has partially destroyed a house. They
conclude that it was caused by an electrical short circuit within the house.
What do they mean?

Clearly the experts are not saying that the short-circuit was a necessary
condition for this house’s catching fire at this time; they know perfectly well
that a short-circuit somewhere else, or the overturning of a lighted oil stove,
or any number of other things might, if it had occurred, have set the house
on fire. Equally, they are not saying that the short-circuit was a sufficient
condition for this house’s catching fire; for if the short-circuit had occurred,
but there had been no inflammable material nearby, the fire would not have
broken out, and even given both the short-circuit and the inflammable
material, the fire would not have occurred if, say, there had been an efficient
automatic sprinkler at just the right spot.
(Mackie 1976:308)

In other words, they are saying that there were a number of conditions that were
themselves unnecessary but sufficient for the short circuit to produce the fire,
given certain natural mechanisms and their causal powers. I suggest that causality
always works like this. However, we should also make specifications in terms of
the subject matter in question. We need social ontology to specify the major features
of being we are interested in. Social ontology is an abstract model—or articulation—
of social being, not a metaphysical point of departure. Any social ontology is itself
a horizon, a hermeneutic standpoint that enables and limits the possibilities of
interpretation, and should thus itself be open to change. Richard Rorty (1991:38)
has rightly criticised the ‘metaphysical thinker’ who ‘thinks that if you can get the
right understanding of Being—the one that gets Being right—then you are home’.
In contrast to the claims of traditional metaphysicians, there is no safe ‘permanent
home’, just home-like stopping points. Social ontology can be criticised both on
the grounds of philosophical arguments showing the conditions for the possibility
of X and/or in terms of explanatory and other (dis)merits of different ontological
assumptions and existential hypotheses.
Critical realism is based on a combination of ontological realism,
epistemological relativism and judgemental rationalism. In the critical realist
view, social relations located in time and space form the relatively intransitive
dimension of the social sciences that, even in the case of research on the
contemporary era, to a large extent exist independently of the particular
knowledge that particular researchers may have of them. Discovery and cognitive
or discursive development, which include ontological theories and hypotheses,
belong to the epistemological dimension, which is relativistic and transitive.
Finally, the idea of judgemental rationality implies that it is possible to give
good, even if contextsensitive, temporal and contestable, reasons for their validity
of a judgement about truth (and truthfulness, rightfulness, good, etc.).
78 Overcoming international relations

From a critical realist perspective, many IR ‘paradigms’ and theories are

problematic in that they tend to define their ontology in terms of atomist states
as unproblematic individuals, which can then form ‘systems’; or in terms of
individual statesmen, that lead these unproblematic individual states in an
international society based on exclusive possessions (sovereignty). The world
becomes fixed, with the implicit assumptions of closed systems and Humean
causality, resulting in the reification of the hypothesised entities. A corollary of
these mistakes is to think that there are different ‘levels-of-analysis’ in the study
of IR. Social being is reduced to statistically measurable ‘factors’, with the
implicit assumption that the world is not real, nor differentiated, layered and
structured. There are only levels of analysis and hence perceptions or observations
that can be measured, counted and analysed.
Are there any schools of thought within IR that are based on more acceptable
social ontologies? In addition to the many strands of critical theory and post-
structuralism, the one that comes closest to satisfying the criteria of social
ontology advocated here is British Institutionalism. In brief outline, it is based
on the view that there are historical institutions such as great powerness, balance
of power, international law and diplomacy, which are based on the long-standing
rules, conventions and practices of modern, Western international society. These
institutions, as any other social institutions, can be studied by uncovering the
shared assumptions of those who act in the name of states, that is, the ‘how
things are to be done’ assumptions of statesmen and foreign-policy decision
makers (see Bull 1977; Suganami 1983).
However, British Institutionalism lacks an explicit notion of causality. To
generalise some of their ideas, and to remedy this lack of conception of causality,
we need first the idea that reasons are causes. Reasons are non-redundant elements
in any causal complex. However, the social complexes capable of producing
outcomes contain other components as well. It can thus be argued that there are
five necessary elements of social being in any relational causal complex (K) capable
of producing events, episodes and tendencies (and the like), namely: historically
constructed and socially positioned corporeal actors (AR); meaningful, historically
structured action (AN); regulative and constitutive rules implicated in every action
and constitution of actors (RU); resources as competencies and facilities (RE);
and relational and positioned practices (PRA). Together these form the sufficient
but unnecessary complex for the production of a result. Hence, there is never a
single cause but always a causal complex K= {AR, RU, RE, PRA, AN}. A result
X may be produced both by K1 and K(i>1), i.e. there may be many different sufficient
complexes capable of (co-)producing the same (kind of) result.
A methodological implication is that the description of a model of historical
causal complexes must be accurate and based on as open a process of empirical
evidence gathering as possible (for details see Chapters 4 and 5). The aim is to
build an explanatory model. Since this process also relies on methodical use of
analogies and metaphors, it can be adequately described as ‘iconic modelling’
(Harré 1970:33–62). An iconic model is a descriptive picture—which can and should
also be conceived of as an interpretation—of a possible world, a possible causal
 Better stories about world politics 79

complex, which is presumed to be responsible for producing the phenomena

we are interested in explaining. Iconic models are based on idealisations and
abstractions, and they always involve metaphors, metonyms and analogies.
Although they are interpretative and based on implicit and explicit conventions
of language, these conventions are also projective: the model describes and posits
existential hypotheses about entities and their relations.
If we view causality and explanation in these critical realist terms, the levels-
of-analysis problematic loses most of its edge. There is no longer room for the
three substantial ‘levels of analysis’; in their stead we are faced with qualitatively
different kinds of contexts of social action. Indeed, I agree with Walker’s
(1991:461, n. 9) suggestion that the levels-of-analysis distinction is nothing more,
nor less, than vertical articulation of a modern European view of horizontal
spatiality (state-territoriality). First, there is the immediate nexus, often of a
face-to-face kind, of interaction, experienced by all actors, including those acting
on behalf of the state, i.e. statesmen. This is the experience corresponding to
the ‘individual’ level. Second, there is the institution of the modern nation-state
that demarcates and governs an exclusive territory. In abstract, however, this
imaginative preconception of many legal and other practices says little about
any particular context of action or concrete social relations. Finally, there is the
world presumed to exist ‘beyond the borders’ with distinctive institutions of its
own such as diplomacy, alliances, international law and world markets.
These institutions do not exist ‘above’ nation-states and certainly not ‘beyond’
their borders, however, but are rather (re)produced in a manner similar to any
other social institution, in various spatial settings.9 There are also internal relations
between institutions. For instance, the institution of state sovereignty is internally—
albeit in a contradictory manner—related to the institutions of international society.
A sovereign state would not be what it is without being defined in international
law and in terms of such practices as power-balancing, collective security,
diplomacy or the multilateral governance of the world economy. In this sense,
‘sovereignty’ enables and constrains social action and constitutes certain kinds
of actors, such as statesmen and diplomats. In an analogous fashion, many
international and global institutions allow for, constitute and/or generate local
(and perhaps simultaneously transnational) processes, thus connecting global
governance to local developments. For instance, the organised practices of
governance—or ‘regimes’—of the capitalist world economy can constitute and
generate, in certain kinds of contexts, tendencies such as the internationalisation
of the state, the globalisation of consuming preferences and practices, and the
unification of political discourses via interpretations of selected events
disseminated by the few big multinational media companies. These are part
and parcel of hegemonic struggles in world politics. All of this suggests that we
should not use the modern commonsense notion of spatiality, reified in the
‘levels-of-analysis’ problematic, as the basis for constructing parsimonious and
deductive-nomological explanations of world politics. Rather, we should focus
on the relational causal complexes that are capable of producing potential and
actual effects—whatever the spatial settings.
80 Overcoming international relations

So far, so good. Nonetheless, a major question remains unanswered. Have I

not committed ignoratio elenchi, that is, put forward a line of argumentation that
does not refer to the standpoint under discussion? If we are interested in
understanding and/or explaining the foreign policies of nation-states, is not the
levels-of-analysis problematic relevant at least to this goal? It may be objected
that, although there are also social relations (which have been made possible by
emergent spatio-temporalities that connect local and global phenomena10), and,
although sovereignty is a constituted and relational social entity, this does not
mean that the levels-of-analysis problematic is not a useful analytical tool in the
study of foreign policies.
This is a germane question, in so far as the institution of the modern nation-
state governs exclusive territories and hence demarcates the boundaries between
the inside and the outside of a political community. If the levels-of-analysis
distinction is nothing more than a vertical articulation of horizontal spatiality in
modern Western thinking (which may well be changed), then commonsense
must tell us that it is easy to distinguish between ‘internal’ and ‘external’ social
forces. Perhaps this is particularly so for those actors, constituted by the rules of
the discursive practices of modern states and international institutions (see Ashley
1987; Bull 1977; Der Derian 1987, 1992), who reflectively monitor the foreign-
policy actions of such states. For these actors, the inside/outside distinction is a
question of two contexts of action that may, of course, be related in many ways.
On the one hand, these are the contexts of ‘domestic’ politics, state-civil society
relations, and of state bureaucracies. On the other, there is the ‘external’ world
of diplomatic and intelligence relations, multilateral negotiations, world markets
and world history. It may no doubt be sensible to ask—particularly if one is
willing to take for granted the understanding of these actors—whether the
‘internal’ or the ‘external’ is responsible for a particular outcome of a particular
process, and whether it is possible to generalise on the basis of these experiences.
Even in this case, however, the most reasonable way of proceeding is to create
a plausible iconic model out of a wide variety of different kinds of hypotheses,
and hence to construct (without reifying relational social existence) an evidence-
backed picture of a possible causal complex that is responsible for producing
the phenomena we are interested in explaining. This is in fact an argument for
something that is different from IR but includes it. I shall call it the study of
world politics.
In the levels-of-analysis problematic, there is agreement that there are always
at least two levels, that of the unit and that of the system. ‘Beyond this there is
no consensus’, Onuf writes laconically (1995:37). This is a telling agreement. It
tells us why the exclusion of non-state relations and social/political spaces is
constitutive of IR as a discipline. It is for this reason, I think, that International
or Global Political Economy (IPE) should not be viewed merely as a subfield
within IR. IPE is founded on a call for a new and more adequate starting point
for the field of analysis (see Gills and Palan 1994:3–8; Guzzini 1994:25, 33–4).
As a phrase, ‘world polities’ could well capture both these claims: there are
states and they have foreign policies and inter-state relations. However, these
 Better stories about world politics 81

are typically misidentified by the ideologies of the international problematic,

and there are also other kinds of relevant relations and spatial settings and
organisations. All these can be seen as at least potentially political. By ‘world’ I
simply refer to anything that is not merely local or national. Thus the title of
this chapter: how to tell better stories about world politics.

Realist conceptions of levels

Thus far I have made an argument for a realist conception of science. In such
conceptions, the metaphor of ‘level’ is widely used. This might suggest that
there is room, after all, for levels in such an approach to IR.11 However, levels
in these realist senses cannot be used to defend IR levels of analysis. There are,
it is true, emergent levels and society is such an emergent layer. We can also
talk about levels and depth within the emergent social layer. However, depth in
these senses has little to do with the ‘level-of-analysis’ problematic in IR theory.
Real emergent powers and properties do not seem to be pre-organised neatly in
accordance with the territorial logic of states.
With reference to depth we can, first, talk about temporal depth, in a sense of
the emergence and sedimentation of practices and social relations, which, if
systematically connected, may give rise to emergent powers and properties.
The genealogy of the international problematic, in Chapter 1, is a partial analysis
of such emergence, focusing on the preconceptions of practices and the
consequent problematic, but also discussing the emergent powers and properties
of capitalist market society and modern states. Second, there is the stratification
of human agency and action (see Giddens 1979:54; Bhaskar 1986:164). This
will be discussed in more detail in Chapter 6. Suffice it to say here that
(unconscious) motivation of, and real reasons for, action may be distorted in
self-understanding and public discourses, giving rise to various, sedimented
and institutionalised rationalisations. These distortions may be connected to
(systematic) unintended consequences of actions and practices.
Like motivation and reasons, knowledge in general is stratified. A distinction
can here be made between analytically and qualitatively layered depth, defined
as follows (see Bunge 1963:39–44).

1 Analytical depth: a piece of knowledge belongs to a degree of analysis An deeper

than another, An-1, if this accounts for a larger number of features of the
entities presumed to exist in both pieces of knowledge, or if it decomposes its
objects more thoroughly than An-1 does, or if it reveals a finer mesh of relations.
2 Qualsitatively layered depth: a piece of knowledge belongs to a degree of
layer An deeper than another An-1, if there are novel qualities in An, or if the
piece of knowledge explains some properties occurring in An-1, in terms
peculiar to An.
82 Overcoming international relations

How is it possible to achieve analytical depth? A dialectical synthesis, which

develops and qualifies previous positions, is a good example of achievement of
depth in this qualitative sense (see Rescher 1977:62–6). Depth is desirable in
social sciences also in the analytical senses of dialectical synthesis and accuracy
of relations. In Waltz’s neo-realism, there is hardly any depth at all. It is a flat
and superficial theory What about qualitatively layered depth? Discourse analysis,
for instance, takes a set of surface-level speech or writing acts and tries to
explain some of their properties in terms of deeper discursive structures (cf.
Alker and Sylvan 1994). Thus, the background discourses of actual
communicative exchange belong to the deeper strata of social realities.

Against scientific (re)conceptualisations of IR levels

Among possible levels of social realities are layers of practical and discursive
knowledge, and the stratification of agency and action. There are also other
realist conceptions of levels, some of them verging on positivism. These have
been summarised by Mario Bunge (1960; 1963: Ch. 3). Levels can be conceived
in terms of degrees, as when one talks about degrees of population density,
degrees of integration, adaptation, learning or orientation, or degrees of
importance. In this sense, IR levels do not have any meaning, and, in any case,
it would here be more appropriate to use the term ‘degree’ rather than ‘level’.
Levels can also be conceived with reference to complexity. Prima facie, it is plausible
to assume that international or world politics is more complex than national or
local politics. However, as the emphasis on the importance of simplicity and
parsimoniousness shows, this is not what Waltz or Singer had in mind when they
introduced this problematic; Waltz in fact takes the requirement of parsimony and
simplicity to its extremes. Thus, although it may be reasonable to talk about degrees
of complexity and even call them ‘levels’ (although I prefer not to do so), this is
something that goes radically against the underlying tenets of Singer, Waltz and
their followers. If it is complexity that is referred to, then the quest for accuracy and
depth should imply pursuit for highly complex iconic models of these realities.
Levels are often conceived of as layers, i.e. as superposed strata arranged
according to the order of their emergence in time, or in terms of their logical
precedence. A layer or stratum is a section of reality characterised by emergent
qualities. This is the essence of a common realist meaning of level. However,
‘international system’, as distinct from states, is not a layer in this sense. Modern
sovereign states and the European institutions of international (or inter-state)
society emerged at the same time. “‘International relations” is coeval with the
origins of nation-states’ (Giddens 1987b:4). Nation-states and IR are internally—
even if ambiguously—related to each other (see Bull 1977:36–7; Camilleri 1990;
Giddens 1987:257–8, 281–2; Koskenniemi 1989:55–98, 192–3). Neither would
be what it is without standing in relationship to the other. Sovereign states and
international relations emerged from the complex re-organisation of political
space in Europe. IR did not emerge from a lower ‘stratum’ of pre-existing
 Better stories about world politics 83

sovereign states but by way of conceptual and practical innovations, and this
made it possible to re-organise social relations and practices in seventeenth- and
eighteenth-century Europe in terms of sovereign states and diplomatic relations
between them. Sovereign states and IR were forged from the pre- and co-existing
materials. The conceptual innovations were built on the emergent conception
of capitalist man and metaphors drawn from the new sciences. Both the state
and IR were conceived in terms of the modern problem of order, which appeared
to have been intensified by the emergent powers of modern organisations.
It is possible to conceive of levels also as ranks. Ranks belong to asymmetric
dependence relations, that is, to hierarchies. Staircase pyramids exemplify such
hierarchies. States—even allegedly democratic states—are often conceived in terms
of hierarchy. Thus, the head of a state or a foreign minister is ‘above the ordinary
citizens or (wo)men on the street’. Consequently, perhaps we could talk about
two levels at least in this context (and add some other levels in between as
well). However, what about ‘international’? For a Waltzian, at least, this kind of
conceptualisation of levels would create a dilemma. It is, after all, the absence
of hierarchies that makes the international system anarchic for him. How could
anything be in this sense above states? If we, by analogy, extend the notion of
‘rank’ to include causal relations (the higher ‘levels as ranks’ causing the features
and actions of the lower ‘levels’), we would, first of all, dilute the meaning of a
‘rank’ considerably, and second, re-open the question as to what it is that produces
these causal effects. This question can be phrased in terms of the legitimate
realist question about emergent powers and properties of social systems.
In the case of an ideal-typical hierarchical organisation, it may seem obvious
what or who is responsible for the effects: the top-rank executives, officials or
managers. Waltz, for instance, can indeed be read as saying that (the leaders of)
great powers are such members of the hierarchy within the international system.
However, this would imply the admission that there is a formalised legal hierarchy
within the international system and that the international system is hence state-
like, not an anarchy. Thus, this understanding of IR levels would contradict the
basic thrust of the Waltz-Singer argument. I think, therefore, that any study of
world-wide asymmetric power and dependency relations would require an
ontology that is drastically different from that of Waltz.
Of all realist notions of levels, only two would seem possibly relevant to a
reconceptualisation of something akin to IR levels, namely hierarchical power
relations and the emergent powers and properties of systems. Social systems
have no essences independent of the contexts of action; they change together
with those relational contexts. Social systems can, nonetheless, have emergent
powers and properties. They must be analysable in terms of actions and
positioned practices occurring in concrete spatial settings. For instance, since
the seventeenth century, the rules and principles of international society have
defined, with increased depth and scope, not only the nature and rights of
sovereignty but also enabled capitalist world markets, through universal private
property rights and the ‘stability of possessions’ (cf. Bull 1977:5). It included
also the creation of labour markets and industrial organisations, at first in
84 Overcoming international relations

particular in England. The resulting complex precipitated the Industrial

Revolution, which soon led to new powers of modern social organisations,
both capitalist companies and sovereign states.
In the nineteenth century, after the era of the establishment of bilateral diplomacy
and sporadic multilateral conferences, multilateral regimes, such as the Concert
of Vienna and the Gold Standard, were created to govern the system of capitalist
world economy and states; also trans-European haute finance played a pivotal role
in this governance (Polanyi 1957 [1944]:10–15). Capitalist world markets have
made a world-wide division of labour possible—although in part the division of
labour was systematically organised by the colonial powers. This division of
labour has lagged or delayed causal efficacy, which has survived well into the
post-colonial age. Since the mid-nineteenth century, the world markets, evolving
within a particular changing legal framework, have also empowered multinational
corporations and given rise to, say, transnational intra-firm trade.
The European nation-states were able to organise larger, more methodical
and technologically more advanced armies than any social organisation before
them. This led to Kant’s concern over these destructive powers and, eventually,
in the wake of the Industrial Revolution, to the total wars of the twentieth
century. The emergence of air power, missiles and nuclear weapons was the
source of John Herz’s (1957; 1968) famous but wavering reflections on the
military viability and legitimacy of territorial state. Simultaneously with these
developments, Fordist (and Taylorist) organisations came to make possible the
mass production of goods, such as cars or Coca Cola cans or tanks or, later,
computers. Increased productive powers and consumption have also had far-
reaching effects on nature, creating—particularly given the characteristic responses
of the democratic welfareS states of the 1970s and 1980s—a new problematic of
world politics, with various regional and global responses. Moreover, state
sovereignty, itself an institution of international society, has made offshore financial
centres and tax havens possible, and safeguarded them (particularly in small
statelets, many with (post)colonial ties). The offshore centres have generated
new exclusive spaces for financial transactions, empowering transnational
criminals, international banks and multinational corporations, while dis-
empowering legal and taxation authorities of many, if not most, states.
These emergent powers and processes create not only new intra- and
interdependencies and relations of domination, they tend to restructure the pre-
existing structures and systems, give rise to new practices and relations, and
invite and encourage the formation of new social and political identities. It is
thus important to study the emergent powers and properties of social systems,
in contexts in which the past and/or outside may be said to be present, as in:
existential constitution; existential pre-existence (structures pre-dating the entrance
of any particular actors); co-inclusion (different processes interloping or clashing);
or lagged, delayed efficacy (past processes having cumulative effects now) (see
Bhaskar 1994:68–9). It is equally essential to avoid ordering these powers and
properties—and the consequent processes—in the simplistic and misleading terms
of IR levels, which merely reproduce and, ultimately, reify the international
 Better stories about world politics 85

problematic (cf. Chapter 1). Rather they should be analysed systematically as

they are:

1 as complex, layered assemblies of social relations between positioned practices,

some of which are themselves organised as collective actors or systems;
2 and as relatively integrated wholes (which are connected also through manifold
unintended consequences of action), the emergent powers and properties of
which may change the relational parts or lead to the emergence of new parts.

The IR levels would preclude open mindedly studying levels in these two possible
senses. As will be explained in more theoretical terms in Chapter 4, social
contexts and systems are not only open but are also, among other things,
overlapping and inter-penetrated in time and space. In parts, they can also be
intra-related, i.e. there are essential relations connecting parts of different systems
or wholes. Consider, for instance, the example of the Bretton Woods institutions
located in Washington DC, next door to the White House, linking, in the early
2000s, the US government and the treasuries and central banks of most states
on the planet, also through the discursive practices of orthodox economics. The
US government may, in parts, be essentially intra-related to the practices and
actors of Wall Street. And so on. Emergent powers and properties should not
be organised, in one’s mystifying imagination, to accord with the simple IR
levels just to yield the convenient possibility of continuing to reproduce the
international problematic. They should be analysed as they really are.
In sum, the scientific realist reconstruction of the traditional IR levels seems a
non-starter. Even the conventional practices of international society do not follow,
in terms of spatial settings of their (re)production, the neat territorial logic of horizontal
inside/outside distinction that is so often transformed into, and then reified as, the
vertical level-of-analysis problematic. Many, if not most, relevant components of
contemporary causal complexes may not be easily locatable in territorial statist terms.
The vertical metaphor of ‘above’ hardly adds to the explanatory power of the
territorial ‘outside’. More concrete and specific claims have to be made.

Responding to critics

My attempt to do away with IR levels has been criticised by Nicholas Onuf.

His remark is made at the outset of a long review of IR levels talk:

Throwing out the language does not necessarily clear things up. In place of
levels, Patomäki would have us ‘talk about different kinds of interpenetrated
contexts’. If contexts penetrate each other, it is hard to know what makes
them different in kind. To say that ‘we are faced with qualitatively different
kinds of contexts of social action’ is hardly a clarification.
(Onuf 1998:195)
86 Overcoming international relations

Onuf wants to ‘clear things up’. However, why should we restrain ourselves to
a mere clarification of the meaning of ‘IR levels’? If these levels are ontologically
and explanatorily unsound, should we not criticise the prevailing practices of
organising the field in terms of them? Onuf’s argument that it is difficult to
identify social contexts is no less shallow Contexts can be identified if social
practices and relations can be identified and their relevance for a given social
agency and action discerned. I am sure Onuf is not claiming that relations and
practices are unidentifiable, or that their relevance to agency and action cannot
be determined.12 What is his point then? Revealingly, it seems to be about
respect for conventions. Onuf (1998:218) defends IR levels in terms of
convenience of ‘treating the primary units [states] as unproblematic’. This verges
on the dogmatic. The closure is completed by ‘in our culture, as in the field of
International Relations, we would have difficulty getting along without the
language of levels’ (Onuf 1998:218).
Wendt published his magnum opus Social Theory of International Politics in 1999.
Wendt applies scientific realism to defend and articulate a sociological version of
state-systems theory. On prudential grounds, Wendt ignores arguments against the
levels of analysis and the related dichotomy of inside/outside. Wendt (1999:13)
argues that systems theory required that ‘domestic or unit and systemic levels can be
separated’.

Some might disagree. They might argue that international interdependence

is eroding the boundary between state and system, making domestic policy
increasingly a matter of foreign policy and vice-versa, or that the boundary
between state and the system is a social construction in the first place which
needs to be problematized rather than taken as given. For them, ‘levels’
thinking is a problem with IR theory, not a solution.
(Wendt 1999:13)

Wendt does not deny the force of these arguments. He ignores them. The legal
organisation of political authority distinguishes between the domestic and
international spheres. This is partly due ‘to the international institution of state
sovereignty, in which states recognize each other as having exclusive political
authority within separate territories’ (Wendt 1999:13). The inside rules are
thus different from the outside rules. Wendt thinks that this opens up the
possibility for a states system theory. The purpose of the states system theory is
to explain why and how states, conceived as analogical to persons, interact in a
systemic setting, and with what outcomes.
Although aware of the fact that ‘in not problematizing something we are
temporarily naturalizing or reifying it’ (Wendt 1999:89), Wendt nonetheless
points out that if we want to find out how the states system works, ‘we will
have to take the existence of states as given’ (Wendt 1999:246). This is the crux
of the matter. Not unlike Onuf, Wendt appeals to the convenience of taking the
states as given, in order to continue with the business of international theorising.
 Better stories about world politics 87

Wendt’s point is to replace the capitalist conception of man with a

socio-psychological conception of man, and leave other components of the
international problematic intact. This opens up some new possibilities for studying
mutually constitutive interactions of states.13
However, the basic anthropomorphic metaphor, coupled with the exclusion
of political economy, leaves a priori out both possible transformations of states
as social systems and related emergent structures. It precludes the possibility
that even states, their foreign policies and the relevant intra-, trans- and inter-
state relations are systematically misidentified by theories that reproduce the
international problematic. As Wendt points out many times—but ignores because
of his desire to build a theory of a closed system of states as ‘persons’—the quest
for depth would seem to imply a quest for delving deeper into the (re)production
and transformation of (1) states, (2) boundaries between states, (3) boundaries
between the separated but related spheres of economy and politics, and (4)
intra-and inter-systemic mechanisms of dependence and power structures.
A ‘convenient’ theory is detached from both history and systematic empirical
study, and refuses to move towards deeper layers of social realities. As will be further
elaborated in Chapter 4, it fails to study satisfactorily the dialectics of emergent
powers and properties of open historical systems. Instead of investing in this kind of
theory of closed history,14 we should develop concepts and metaphors that enable in-
depth explanations of changing world political realities, based on systematic
empirical research. Doing away with IR levels seems to be a necessary step towards
that direction.

Going beyond the ‘two stories’ dichotomy

Having discussed the levels-of-analysis problematic, let me now turn to the

question of whether there are always ‘two stories to tell’—the insider’s account
and the outsider’s account.15 More specifically, I want to make two claims here.
The first is that reasons can be causal, a notion in line with Carlsnaes’s
(1994:277–87) point that we always need to involve both ‘understanding’ and
‘explanation’. Second, I want to argue that the understanding of the meaning
that actors attribute their actions is always a productive enterprise. This holds
despite the transcendental argument for the real existence of social worlds that
are, for the most part, existentially and causally independent of any researcher’s
particular interpretation of them. In other words, in the process of such
understanding, something is always produced, namely a novel or conventional
interpretation that has potential to co-cause and/or co-constitute elements of
social worlds, and which at least shapes the mode of being and agency of the
researcher. Furthermore, I will also argue that a purely descriptive attitude
conceived in terms of an outsider’s ‘objective’ focus on such actor meanings is
impossible. Rather, as researchers we are necessarily drawn into the process of
assessing different validity claims circulated within social practices by virtue
and in the course of their (re)production.
88 Overcoming international relations

Reasons for a particular action in a given context are the causes of the
meaningful activities of the agent in question, leading in turn to various outcomes.
Another way of putting this is that the agent’s reasons are a necessary condition
for the particular bodily movements—including those needed for speaking and
writing—that occur in purposive action, in the straightforward sense that, had
the agent not possessed them, these actions would not have occurred in the first
place (unless over-determined). According to Harré and Secord (1972:86–7),
human powers of this kind include the following: the power to initiate change;
the capacity of being aware of things beyond oneself and knowing what they
are; and the power of speech. Furthermore, agents must know what they are
about to do, what they are doing and what they have done, in the sense of
being capable of describing their actions. Without these kinds of causal powers
of embodied social actors, there can be no social practices and institutions.
This solution seems so obvious that it may well give rise to immediate scepticism.
In Finnish, for instance, there is a word (syy) that refers to both causes and reasons
without distinguishing between the two. Can a scientific ontology be grounded in
such a commonsense notion? Although I cannot enter into a debate on this issue
here (but see Bhaskar 1979:102–46; Harré and Secord 1973), let me just very
briefly discuss some of the objections usually raised against this claim. In addition
to the obsolete behaviouralist insistence on the relevance of the concepts of stimulus
and response,16 there are basically two kinds of objections of relevance here, the
one philosophical and the other logical. The philosophical objection is that, if we
conceive of reasons as causes, we are either postulating temporally later state of
affairs X as teleological causes for an earlier Y, or we are committing a vicious
regress argument of a sort that ‘the causes for reasons R must be other reasons
R1, which in turn are caused by R2 etc., ad infinitum’. However, as Bhaskar has
noted, teleological explanations can be ‘most naturally construed as a species of
ordinary causal explanation, where the cause is, in the case of reasons, some
antecedent state of mind’ (Bhaskar 1979:107). Similarly, the question of whether
the reason R was the cause of A does not depend on an account, explication or
explanation of the process through which reason R has been (socially) generated
and constituted. Bhaskar thus notes the following:

possession of a reason should not in general be construed as an action, but

rather as a disposition or state. And the possession of a reason for an action
can itself be sufficient for that action, when appropriate circumstances
materialise. Thus if I am a German social democrat, I do not have to wait
upon a further reason to vote SDP on polling day—I just exercise my reason
in voting. So at no possible point does a vicious regress in fact arise.
(Bhaskar 1979:107)

This does not, of course, mean that actors do not have reasons for their political
identities—on the contrary. It simply means that a commitment to social
democracy (seen as a state or disposition) can, in certain circumstances, in itself
 Better stories about world politics 89

be a sufficient reason for voting for the SDP. In this example, a causal explanation
of a particular voting act is given in terms of this sufficient reason R, that is, in
terms of a commitment to social democracy. Naturally, the commitment itself
can also be explained, in which case we would have to move from a surface
account to a deeper account.
Finally, there is the logical connection argument, according to which (1)
causes must be logically distinct from effects, yet (2) reasons are logically
(conceptually, internally) connected to the actions they explain. Hence, it is
claimed, reasons cannot be the causes of actions (see Winch 1971 [1958]; von
Wright 1971). Although this tradition of (analytical) hermeneutics is right in
emphasising the existence of conceptual and internal relations in social worlds,
the problem is its total denial of causality in these worlds. The consequence of
this denial is the Kantian dualism between the causally deterministic natural
world, conceived in Humean terms, and the world of human (and perhaps
superhuman and divine) reason. This dualism makes it impossible, for instance,
to grasp the effects of human reason (through technology etc.) on nature or, vice
versa, the effects of the bodily mechanisms on human reasons or reasoning.
Moreover, logic connects statements, not events, actions and the like. Even if
there exists an internal connection between the statements ‘causing x’ and ‘x’,
sunburn is not a logical consequence of the sentence ‘prolonged exposure to the
sun results in sunburn’ (Bhaskar 1979:109–10). Statements do not cause sunburn.
In an analogous manner, neither thinking, talking or writing the statement
‘switching the computer off’, nor the sentence in itself, can switch off the
computer. A corporeal and reasoning human being has to do something for this
effect to occur. In other words, his or her reasons for doing something must be
causally efficient. This is the ground for being able to distinguish between an
agent’s knowledge about what he or she is about to do, what he or she is
actually doing, and what he or she has done, i.e. distinguishing between an
agent’s own understanding and what he or she is actually, in a causally efficacious
manner, doing.
Let us at this point return to the claim made by Hollis and Smith, namely
that the ‘understanding’ and ‘explanation’ cannot be synthesised into one
approach, and hence that one must perforce choose between the two. However,
reasons are causes, that is, non-redundant and existentially necessary elements
of any causal complex. The first stage in any explanation is thus to find these
causally efficient reasons. Regardless, what is the relationship between the reasons
and intentions of a person and inter-subjective meaning? In trying to give an
answer to this query, Habermas’s theory of understanding is, in my view, of
great help.
In order to understand something, Habermas (1987 [1981]:18–19) writes,
one has to make the assumption that there are inter-subjective standards in the light
of which human actors can decide whether they are following appropriate
discursive or practical rules in their interaction with others. Every (speech) act
presupposes a number of rules. Habermas (1984:112–13) notes that the social
scientist cannot treat human utterances as mere facts. He or she must be able
90 Overcoming international relations

to describe social rules, resources and practices in a language that, in principle,

can be used by social actors to reconstruct their linguistic self-understandings.
Furthermore, the process of understanding is bound up with a process of
bringing something about. An isolated social scientist reading texts in a library
and writing a paper only for his or her own amusement may perhaps be
producing something purely subjective and private. However, whenever he or
she comes out in the public domain—for instance by going out to do first-hand
data collection in the field—he or she will become a participant in a process of
actual communication.
This gives rise to the question of the theory-dependence of ‘data’. The data
against which explanatory models are tested cannot be described
independently of the theoretical language in use. This is the first level of
involvement. Second, the researcher cannot gain access to a symbolically pre-
structured reality through (outside) observation alone. He or she has to engage
in communication. Moreover, understanding meaning cannot be controlled in the
same way as can observation in the course of scientific experimentation.
Rather, there are two stages in the process. ‘Prior to choosing any theory-
dependency, the social scientific “observer”, as a participant in the process of
reaching understanding, through which alone he can get access to the “data”,
has to make use of the language encountered in the object domain’ (Habermas
1984:102–10). In this connection Habermas refers to Giddens’s notion of the
‘double hermeneutics’ of the social sciences:

The theory-laden character of observation-statements in natural sciences entails

that the meaning of scientific concepts is tied-in to the meaning of other terms
in a theoretical network; moving between theories or paradigms involves
hermeneutic tasks. The social sciences, however, imply not only this single level
of hermeneutic problems, involved in the theoretical metalanguage, but a
‘double-hermeneutic’, because social-scientific theories concern a ‘pre-
interpreted’ world of lay meanings.
(Giddens 1977:12)

In other words, language mediates the constitution of social ‘data’ or ‘facts’ in a double
way—as the linguistic and theoretical framework of the social scientist, and as the
language of acting subject-objects in their constitutive understanding of their
institutions and situations (Apel 1984:75–6). There is also interaction between these
two spheres, whether weak and indirect or strong and, perhaps, immediate. Thus,
meanings cannot be treated as mere facts in the positivist sense of the term, although
they are constitutive of the causal complexes capable of producing outcomes. The basic
thrust of my argument is, thus, that it is always necessary to understand the meanings
that actors attribute to their actions and practices if we are to acquire knowledge of the
causal complexes of social worlds. The consequence of this is to undermine the claim
that ‘sometimes one type of account makes more sense; sometimes another’. Both are
needed, always.
 Better stories about world politics 91

Conclusion: possibilities beyond the ‘great debates’

In his 1987 article, Wendt (1987:362) states that explanations are basically
answers to certain kinds of questions. Hence, the adequacy of an explanation
depends on the question posed. Wendt goes on to distinguish between two
types of questions: ‘How is action X possible?’ and ‘What made X actual rather
than Y?’ For him, structural analysis explains the ‘possible’, while historical
analysis explains the ‘actual’. Hollis and Smith are critical of this distinction,
using the Gulf War as an example:

Suppose one sets about explaining the occurrence of the 1991 Gulf War in
this spirit. How was it possible for war to be declared? This question might
invite an account of the rules which constitute the conditions needing to be
satisfied for it to be true that a state of war exists; or it might invite an
account of the recent history of the United Nations and the Middle East—
historical conditions which made this war possible; or some mixture of both.
These are very different sorts of account, as is clear when we enquire how
exactly the mixture is to blend theoretical analysis with historical trends. To
be brief, Wendt’s ‘how’? question encapsulates several running disputes about
explaining or understanding how a war could happen.
(Hollis and Smith 1991:406)

This criticism is convincing only in so far as one has accepted the criterion of parsimony
and the quest for deductive-nomological explanations. I hope that I have been able to
give some idea of the complexities involved in any geohistorical explanation, and thus
to give some plausibility to the alternative idea that the richness and depth of an
explanation is of higher value than its parsimonious nature. Consequently, it is not a
problem to pose different ‘how’ questions in one and the same study.
Indeed, we can distinguish between different ‘How is action X possible?’
questions. When we are concerned with the identity of X, we need an account
of relevant constitutive rules. This analysis freezes, so to speak, the social world
in question. Then we can move towards the world of historical, causally effective
interaction and ask: What are the INUS-components which made that action
possible? At this level we are analysing processual complexes in open systems,
and we cannot find general necessary conditions of X. Instead, we are interested
in the actual constellation of conditions that made X contingently possible. Other
constellations could have made it contingently possible as well. After this we
can ask: What made X actual rather than Y? When posing this question, we
are interested in reasons, justifications and the like, organised as discourses, as
well as in the interactive actions of various actors. Finally, we can pose this
genealogical query: How was X’s identity and the relevant constitutive rules
produced in the course of historical interaction?
This tentative list of questions is a plea for complexity and nuances, and for
patient, reflexively interpretative data-collection and model-building. For some
92 Overcoming international relations

it may sound like the more or less well-established practice of historical research.
However, it is much more than that. It is a plea for admitting the plurality of
possible interpretations and opening up the possibility for a genuine dialogue.
This plea is coupled with an appeal for causal in-depth explanations in terms of
relational complexes as they really are. It is not enough to stay at the level of
prima-facie reasons for actions; we must also analyse the conceptual and causal
relations of social action in their inexhaustible complexity, fullness and depth.
By systematic and iconic modelling it is possible to form pictures of the wholes
in question and to give deeper explanations by answering the questions ‘How
possible? and ‘Why actual’? Aspects of models can be generalised, given that
certain (deep) structures and relational mechanisms and/or complexes can be
found to be causally efficacious across many open-systematic contexts.
The prospect of avoiding reification opens up entirely novel possibilities concerning
the aims of story-telling about foreign policies and, much more generally, world politics.
It makes emancipation and edification possible. Interpretative in-depth explanations
try to make phenomena more understandable and intelligible. They have to enter the
processes of assessing truth-claims that take place within social practices. By virtue of
being holistic and aiming at explanatory power, these interpretations, based also on
truth-judgements, can be critical of actors’ understandings. Social relations that are
shown to render, in some important senses, anomalistic, false, reificatory, illusory and/
or superficial discursive formations should come to be criticised in being explained.
Novel interpretations of the social worlds and their inherent possibilities can
contribute to the process of re-signification of practices and thereby make possible
better, that is, more empowered, more ethical and more virtuous ways of being and
action. To enable this, we have to go beyond the ‘great’ debates of the international
problematic. It is time to transcend these limits by getting rid of at least parts of the
conceptual and metaphysical ballast weighing down the field. Instead of an eternal
debate or attempt at a synthesis between the sketchy perspectives called political
realism and idealism—amounting to the mere reproduction of the international
problematic—we can enable the possibility of genuinely novel interpretative possibilities.
By claiming that reasons are causes, and that, more generally, causality can be
understood in terms of INUS-components in open (and intra- and inter-related)
social complexes and systems, I have tried to open up a space for the study of
world politics in a more emancipatory and edificatory manner. Perhaps it is now
possible to overcome some of the limits of the contemporary political imagination,
as well as to improve on the way scientific historical accounts approach value
questions, structural constraints and human choice possibilities. However, first we
have to articulate the CR ontology and research methodology in more detail.

Notes

1 In standard textbooks (e.g. Gill and Law 1988; Gilpin 1987), Marxism is still taken
seriously as the third perspective; and (post-)Gramscian historical materialism (e.g. Cox
1987; Gill 1993) is often conceived as the nearly hegemonic movement within IPE.
 Better stories about world politics 93

2 Wæver (1994a:257) has been careful to acknowledge that the debate has changed:
‘During the 1980’s, realism became neorealism and liberalism became neo-liberal
institutionalism. Both underwent a self-limiting redefinition towards an anti-
philosophical, theoretical minimalism, and they became thereby increasingly
compatible. A dominant neo-neo synthesis became thereby the research programme.’
This interpretation not only confirms my point about the state of the art but also
nicely illustrates the phase of Western cultural development within which the adjective
new is used rather liberally. As Honnef (1994:14) has pointed out in analysing the
trends and fashions of contemporary art, the inflationary use of the term new never
actually appears on its own but always as a prefix (neo-) or as an adjective qualifying
a trend that already exists. The same goes for the fashions within IR.
3 In addition to the Wendt and Hollis and Smith debate, major IR contributions include
Wendt (1987, 1992a, 1994 and 1999), Dessler (1989), Carlsnaes (1992, 1993), Wæver
(1994a:260–71), Doty (1997) and Wight (1999, 2001).
4 However, there are different kinds of (critical) scientific realisms. For some, the emphasis
is on the word ‘realism’. Thus, Wendt (1991:397) writes, in distinguishing himself
from empiricism, that a ‘related dispute concerns the requirements of scientific
explanation, where empiricists seek to subsume phenomena under lawlike regularities,
whereas realists seek to identify the underlying causal mechanisms that generate the
phenomena’. In an e-mail conversation in 1992–3, Wendt often repeated the point
that he is a scientific realist precisely and only in this sense. As will emerge later on, in
my interpretation critical realist social sciences would sometimes do better without
the metaphor of mechanism. More generally, I define critical realism in terms of a
well-articulated ontological realism, epistemological relativism and judgemental
rationalism, and emphasise the role of CR theory of emancipation.
5 Scientific realist interpretation of a Humean phenomenology of constant conjunctions
is in no way a step forward from empiricism/positivism.
6 In fairness, it should be noted that in their reply to Wendt, Hollis and Smith (1991:397)
admit that there are other conceptions of causality. However, they never discuss or
take on board these other conceptions.
7 Even though they later, in Chapter 9, move in different directions, Hollis to a more
hermeneutic position and Smith to a scientific realist stand, they never distance
themselves from a Humean view of causality.
8 The best candidate would be the democratic peace theory, according to which liberal-
democratic states do not fight each other. As currently formulated, this theory is
reliant on the identification of what appears to be a strong and enduring constant
conjunction: liberal-democracies do not fight each other. That is, the argument is that
X (liberal-democratic regimes) explains the absence of Y (war, organised violence between
two states). The assumption must be that Y is a normal thing to occur, and that X is
the mechanism or thing that explains why it does not occur in this special situation
(or more strongly, prevents it from happening). The existence of a well-functioning,
on-switched lamp explains the absence of darkness in a room in the night, but would
be quite irrelevant in a bright sunshine. In a situation of constant sunshine, it is
darkness that should be explained (if it were to occur).
Moreover, as Barkaw and Laffey (1999:414–15) have argued, the correlation is
based, in part, on statist categories. Thus, for instance, the US covert actions against
democratically elected governments in the Third World resulted in significant
violence. However, they are not counted, and thus cannot refute the democratic
peace proposition. From a CR perspective, it is also the very idea of a constant
conjunction that is problematic. Constant conjunctions only obtain in closed systems.
However, the world of international politics has been, and is, anything but closed.
This suggests that it cannot simply be factors internal to democratic nation-states
that are responsible for this apparent constant conjunction. Even if a mechanism
(perhaps of a Kantian form) or, rather, a manifold structured complex were identified,
94 Overcoming international relations

it would still only be transfactually operative in an open system. This would mean
that there was something about the other relevant complexes and processes that
was also playing a causal role.

Hence, an adequate analysis must also focus on these other relations and processes,
forming a picture of the relevant parts of the geo-historical whole in question. Here,
the explanans and, up to an extent, also the explanandum of the theory changes. What is
the overall and possibly changing causal complex, which has tended to make the
twentieth-century relations between capitalist, Western-minded liberal-democracies
peaceful? Which of these components are changing in the early twenty-first century?
In effect, a CR analysis would suggest that if there is anything to explain, it must be
explained in terms of real tendencies and real working of spatio-temporal causal
complexes in open and therefore changing systems. See also Patomäki and Wight
2000; and Chapter 8 for an articulation of a rather different (partial) explanation in
terms of a pluralist security community.
9 Typically, main activities reproducing these institutions are located in the
governmental complexes or diplomatic settings of capital cities of different countries,
or in the diplomatic settings of the headquarters of various international
organisations. Spatially, these can be either inside or outside of any particular
territorial state, but always inside of at least one state (and usually many
simultaneously, given the modern communication technologies and the need of
diplomats to consult their governments).
10 Changing social practices may give rise to emergent spatio-temporalities. Long-distance
calls or computerised systems of information flows have disembedded, relatively
speaking, space from time. Immediate contacts have become possible over long
distances. A truly global multinational enterprise—which, although rare, can be taken
as an ideal case—presupposes the disembedding of space from place and relies in its
practices on the capability of disembedding space from time (cf. Bhaskar 1994:70).
These emergent spatio-temporalities may also enter into the discursive practices of
political identity—and interest—formation.
11 When I asked Bhaskar (personal communication, London, 22 July 1995) about the
plausibility of IR levels, his immediate reaction was: ‘Although I do not know the
subject matter very well, it seems, at least prima facie, plausible.’ Also Wendt, in his
comment on an earlier version of this chapter, asked whether it could be possible to
think about IR levels in terms of layered mechanisms. Thus, we could try to explain,
at least partially, higher-level laws in terms of layered mechanisms. We could also try
to explain, at least partially, higher-level laws in terms of mechanisms of lower levels.
For instance, Wæver (1994a:264) interprets IR levels in these terms.
12 The methodology of iconic modelling is devised for the purpose of identifying causal
complexes and social contexts; it will be explicated in detail in Chapter 5.
13 These new possibilities include: identity-constituting moves; learning-processes;
development of (dis)trust; altercasting; development of a sense of common fate; and
the formation of collective identities (see Wendt 1999: Chapters 6 and 7). In some
contexts, given particular research problems, Wendt’s ‘social theory of international
polities’ can thus provide highly useful and interesting hypotheses for concrete iconic
models.
14 Thus, in Wendt’s theory, structures are reduced to the ideas ‘states hold’. There is not
an unequivocal ‘logic of anarchy’, as in Waltz’s neo-realism, but three ‘cultures of
anarchy’—named after Hobbes, Locke and Kant—and processes of moving from one
to the other. This may be a richer and more nuanced or realist theory than that of
Waltz’s. It is nonetheless a theory of a closed list of interpretative and transformative
possibilities. These possibilities are derived from both philosophical theories and the
conventional international problematic rather than from any sustained empirical
research of concrete problems. Theory in this sense not only precludes transformative
 Better stories about world politics 95

possibilities, it also precludes interpretative possibilities, and substitutes iconic modelling

with armchair philosophising.
15 Note again the usage of the metaphor of inside/outside, this time in another context.
Again, this whole problematic of inside/outside seems to stem from binary thinking
(either inside or outside, there is no third alternative), from territorial understanding
of spatiality (given territorial demarcations, you can only be inside or outside), as
well as from individualism stemming from empiricism (a naïve view of human beings
as merely corporeal beings with well-defined and strict spatial borders).
16 Neufeld (1993b) talks about ‘meaning-orientated behaviourism’, which is less hostile
to Verstehen approaches and which dominates at least some areas of IR research.
However, the notion of causality of these ‘meaning-orientated behaviouralists’ is the
Humean one, with the standard (partially implicit) assumptions concerning regular
successions, isolated, closed systems and unchanging entities.
Part II

Explicating a critical
realist methodology
4 Critical realist
social ontology

The nature of objects determines their cognitive possibilities. This critical realist
(CR) idea, drawn from Aristotle, has also increasingly gained ground in
international relations (IR). Martin Hollis and Steve Smith have argued that
any attempt to explain anything in world politics necessarily implies hypotheses
about the existence and nature of entities. They contend that the justification of
ontology is connected to its explanatory merits. David Dessler (1989:446)
maintains that ‘the richer and more comprehensive the underlying ontology,
the better the theory’. Alexander Wendt (1999) seems to disagree with this
criterion, but has nonetheless written a remarkable book on the ontology of the
states system. Open-mindedly, Hayward Alker has learned, through trial and
error in his careful empirical studies, that ultimately our understanding of the
nature of data depends on our social ontology. Better social ontology is thus the
key to constructing more adequate sets of empirical evidence.
Individualism claims that ultimately there are only individuals or their minds,
preferences and decisions. The roots of individualism are in the capitalist
conception of man. This conception has never gone unchallenged. The
methodological debates between individualism and structuralism started some
time in the nineteenth century. These included the famous Methodenstreit in
economics; debates about Marx and Marxism; and manifold debates in sociology,
symbolised by the classic works of Auguste Comte and Emile Durkheim, among
others. In various guises, structuralism has claimed that there are emergent
social wholes that determine individuals and their behaviour or actions.
Anthony Giddens’s theory of structuration and Roy Bhaskar’s
transformational model of social activity (TMSA) are important attempts to
transcend this debate in systematic ontological terms. Giddens and Bhaskar
have many things in common. Both rely on a form of scientific realism, which
has incorporated basic insights of hermeneutics—as represented by Ludwig
Wittgenstein, Peter Winch and Hans-Georg Gadamer—and the critical theory
of the Frankfurt School. There are also differences between these two theories.
It has been claimed that Giddens’s theory is ultimately hermeneutic or voluntarist
in its implications (see, for example, Gallinicos 1985:133–66; Wendt 1987:335–
70), and that Bhaskar’s conception of social reality is a good example of Marxist
and structuralist thought (Shatzki 1988:239–60).
99
100 Explicating a critical realist methodology

In this chapter, I shall try to show some of the weaknesses and inadequacies
of both theories. My aims are, however, reconstructive. I try to clarify the
subject matter and elaborate further on the CR concepts of ‘action’, ‘structure’
and ‘power’. A central ontological theme is the nature of, and the relationshiph
between, internal and external relations. First, I develop the point that to explain
the (re)production of internal relations, we also need a theory of practical and
communicative action. Second, I argue that an adequate theory of external
relations requires an iconic model of different modes of social action. One of
these modes, it is argued, is indeed communicative action. Third, I attempt to
detach Giddens’s concept of power—which connects action essentially to power
as transformative capacity—from its Weberian undertone and incorporate into it
an important insight of Michel Foucault.
There is a far-reaching tension between Giddens’s and Bhaskar’s views on the
concept of structure. I criticise Bhaskar’s naturalist metaphors and analogies, but,
contra Giddens, maintain that a relational conception of structure is ontologically
more sound than Giddens’s ‘rules and resources’. As a final point, I elucidate
how the dialectics of emergent powers and properties of open historical systems
should be conceptualised and studied. Instead of the futile IR levels, I articulate
an alternative account of systemic emergence and stratification. In conclusion, I
collect the different strands of my argument together with the help of the notion
of a layered, relational causal complex. This ontological notion is central to the
whole of this book, to my explication of a CR research methodology.

Internal relations and communicative action

According to Bhaskar (1979:52), the work of discovering the independently

existing and transfactually active ‘machinery of nature’ is not the aim of an
independent inquiry of metaphysics, but rather that of the sciences. Sciences
are therefore also concerned with existential hypotheses and presumptions.

Science employs two basic criteria for the ascription of reality to a posited
object: a perceptual criterion and a causal one. The causal criterion turns on the
capacity of the entity whose existence is in doubt to bring about changes in
material things.
(Bhaskar 1979:15–16)

The discovery of these entities always includes the formation of an iconic model,
which is basically concerned with qualitative analysis and picturesque description
of causal mechanisms, which are responsible for the phenomena we observe.
The formation of an iconic model must be done under plausibility control
(Harré 1970: Ch. 2).
On the other hand, philosophical reflection can reveal that reality must be
stratified, differentiated and structured if scientific activities are to be intelligible,
 Critical realist social ontology 101

because sciences have a characteristic pattern of description, re-description and

explanation of the phenomena identified at any one level of reality with a
mechanism consisting of differentiated elements (Harré 1970:16; Bhaskar 1986).
In this conception, reality consists of partially interconnected hierarchies of levels,
in which any element (e) at a level (L) is in principle subject to the possibilities
of causal determination by higher-order, lower-order and extra-order effects,
besides those defining it as an element of (L) (including those individuating it as
an (e)) (Bhaskar 1986:106). Moreover, the explanatory natural sciences are
typically moving towards deeper ontological levels.
Bhaskar’s naturalist argument is that ‘there is (or can be) an essential unity
of method between the natural and the social sciences’ and that ‘it is possible to
give an account of science under which the proper and more-or-less specific
methods of both the natural sciences and social sciences can fall’, notwithstanding
the ‘significant differences between these methods’ (Bhaskar 1979:3). This
argument is based on the conception of ‘society as an articulated ensemble of
relatively independent and enduring generative structures’ (Bhaskar 1979:3)
and on the analogy between natural and social structures (or mechanisms).
However, is the analogy between social and natural structures feasible?
Bhaskar (1979:57) uses magnetic fields as an illustrative and picturesque
example: social structures are theoretical and necessarily unperceivable in the
same sense that a magnetic field is. So, although social structures cannot be
identified independently of their effects, they are, in this respect, no different
from many objects of natural scientific inquiry. What really distinguishes social
structures is that not only are they incapable of being identified independently
of their effects, they do not even exist independently of these effects (Bhaskar
1979:57). Nevertheless, Bhaskar thinks that social structures are real and that
this can be shown with causal criteria.
The use of causal criteria for existence presupposes some notion of causality.
Bhaskar (1979:165) accepts in principle [Link]’s (1976) influential analysis
of a cause as an INUS-condition, which is quite well-suited for causal analysis
in open systems. The following is a modified definition of a cause as an INUS-
component:

Def. 1: Cause is an insufficient but necessary part of a condition that is itself

unnecessary but sufficient for the production of a result, i.e. the INUS-condition.

This definition implies, among other things, that there may be many different
sufficient conditions capable of producing the same result. Theodore Shatzki
(1988:24) doubts whether social structures can be treated as causes, because ‘talk
of causal efficacy, causal power, and being a generative structure requires that
anything that possesses these properties be something that actually brings about
changes’. There is some truth in Shatzki’s doubt, as we shall see. Nevertheless,
Bhaskar claims only that the dual character of society means that:
102 Explicating a critical realist methodology

Society is both the ever-present condition (material cause) and the continually
reproduced outcome of human agency. And praxis is both work, that is
conscious production, and (normally unconscious) reproduction of the
condition of the production, that is society. One could refer to the former as
the duality of structure, and the latter as the duality of praxis.
(Bhaskar 1979:43–4)

If social structures are always a necessary condition for any sufficient condition for
social action, as the ontological duality of structure implies, then they are also
INUS-conditions (def. 1), and hence causes by definition. Thus, social structures are
real. On the other hand, the problem with the magnetic-field analogy—and this may
have misled Bhaskar in some crucial points—is that it is quite reasonable to say that
a magnetic field sometimes produces or brings about changes (alone), whereas social
structures are never able to do so (alone), because social structures do not have
action-independent causal effects. Or to use Aristotelian terminology, social
structures are mere material causes, whereas magnetic fields can also be efficient
causes.
Are social structures always a necessary condition for action? In the TMSA
the ‘point of contact’ between human agency and social structures is the mediating
system of positions (places, functions, tasks, duties, commitments, rights, etc.)
occupied (filled, assumed, enacted, etc.) by individuals, and of the practices
(activities, etc.) in which, by virtue of their occupancy of these positions (and
vice versa), they engage. These positions and practices can be individuated
only relationally (Bhaskar 1979:51). To this must be added that social structures
exist only by virtue of the activities they govern, enable and constrain, and that
they are materially present only in persons and the results of their actions
(Bhaskar 1986:130–1).
In what sense are these kinds of structures always a necessary and pre-
existing condition for action? A distinction must first be made between internal
and external relations. These concern substantial relations of connection and
interaction, not formal relations of similarity or dissimilarity.1

Def. 2: A Relation RAB is internal if and only if A would not be what it essentially
is unless B is related to it in the way that it is.
(Bhaskar 1979:54)

Whereas external relations are conceptually contingent, internal relations are conceptually
necessary. Both can be found in society When Bhaskar claims that action (in
general) requires social structures as a necessary condition, it seems that he has
in mind only structures of internal relations. This claim stems from the
hermeneutic character of the social world: social world is meaningful; social
actors are intentional and intelligible; and their actions are always rule-enabled.
To have identity, actors and actions must be constituted by rules. There is a
difference between constitutive and regulative rules. The meaning of a particular
 Critical realist social ontology 103

action derives from the relevant constitutive rules, which in turn are related to
other constitutive and regulative rules. Social phenomena retain their identity
under regulative rule violation, but not under constitutive rule violation. Consider
the example of the social practice of trial by jury as presented by John Greenwood
(Greenwood 1987:171–204). The constitutive rules of this institution would
include something like the following (Greenwood 1987:188):

trial=def ‘the examination of evidence by a body of persons

comprising the judge, jurors, learned counsel, defendant,
and officers of the crown and the court, and the
determination by the jury of the guilt or innocence of the
defendant accused of some civil or criminal offence’.

juror=def ‘a member of a body of twelve men or women who comprise

a jury in a court of law, who take an oath to render a decision
on a judicial question’.

judge=def ‘a public officer appointed by the crown to hear and try

cases in a court of law’.

Because of these definitions or constitutive rules, the trial, juror and judge are
internally related to each other. In this sense social structures are the necessary and
ever-present condition of all social action. Only they make meaningful social action
possible.
It is important to be very careful about differences between ontological and
epistemological matters. We—as researchers—may define trial, juror and judge as above,
but then we are making existential hypotheses; we can then ask: ‘Is the world really like
this?’ (ontology). The attempt to answer this question will involve what Giddens calls
the ‘double hermeneutics’ of social sciences (epistemology): a theoretically reflective
scientist tries to investigate and interpret a world that is itself meaningful (Giddens
1977:12).
Epistemologically, as Sayer (1992:56) notes, the conceptual relations we can
plausibly posit between entities are theory-dependent. For example, whether it
is conceptually inconsistent to describe an individual actor both as a capitalist
and a proletarian depends on what we take to be the possible nature of the
person, means of production, etc. It is in fact a theoretical question whether we
should accept the categories of capitalist and proletarian at all. Ontologically,
the existence of internal relations in the social worlds—outside academia—depends
on the explicit or implicit ‘theories’ of relevant lay actors and thus it is not
possible to infer them purely theoretically, or objectively. This is probably the
reason why John Shotter (1983) characterises social processes as intrinsically
vague and incomplete. Bhaskar comments on Shotter’s point only briefly:

Vagueness is not the same as ambiguity. Vagueness inhabits the shading,

soft-edged, open-textured conceptual world of Wittgenstein and Waismann.
104 Explicating a critical realist methodology

Ambiguity points towards the conceptual spheres of ambivalence and

contradiction plumbed by Freud and Marx. Both seem to me indispensable
categories for social analysis, which should always be as precise as in Aristotle’s
words, ‘the nature of the subject permits’.
(Bhaskar 1983:91)

The spheres of ambivalence and contradiction are also real (ontology), not
merely indispensable categories for social analysis (epistemology). I do not,
however, mean the real and ‘objective’ dialectical contradictions inherent in
class relations, as Marx at least sometimes did—although in general the conception
of dialectical causality is important (cf. Bunge 1959:160–5)—but rather the
dialectical disputations and contradictions in the ‘shading, soft-edged, open-
textured conceptual world of Wittgenstein and Waismann’.
There is usually no universal and definitive consensus among lay actors about
the constitutive rules of institutions, or about the definitions of position-practice
systems. Hence the ‘essential internal relations’ (def. 2) in the social world are
always both ‘vague and incomplete’. Consider the following quotation from Winch:

If social relations between men exist only in and through their ideas, then,
since the relations between ideas are internal relations, social relations must
be a species of internal relation, too. This brings me into conflict with a
widely accepted principle of Hume’s: There is no object, which implies the
existence of any other if we consider these objects in themselves and never
look beyond the ideas we form of them.
(Winch 1971 [1958]:128)

The point of critical realism is, contra Winch, that there are both internal and external
relations—and that Hume seriously mis-specified the nature of external relations.
However, what exactly is the nature of internal relations? How do internal relations
as ‘ideas’ come into being? How can there be shared ‘ideas’ and internal social relations
in the first place? The crucial theoretical problem is thus: In what way are the constitutive
rules and social meanings of action produced, mediated, reproduced and changed?
How is the mutual understanding of social (constitutive) rules historically possible?
Giddens sees the first question as being at the core of the problem of order.
Innovatively, Giddens defines the problem of order in terms of realist spatio-
temporal causality rather than an analogy to the Humean problem of induction.
‘The “problem of order” is the problem of how it comes about that social
systems [practices, relations] transcend time and space’ (Giddens 1983:78). In
other words, what mechanisms are capable of producing the outcome, namely
that the ontologically critical internal relations, practices and systems transcend
time, place and, also, the biological existence of human beings?
According to Habermas (1984 [1981]:274–5), the actions of several actors
are linked to one another by means of the mechanism of reaching understanding.
This is the way actions are interlaced in social spaces and historical times.
 Critical realist social ontology 105

However, Habermas still works with the Hobbesian (Humean) problem of order
he has inherited from Talcott Parsons. Rather than ‘co-ordination of actions’,
the real problem is to explain the endurance and (re)production of ontologically
critical internal relations of social practices. However, Habermas (1987:295)
speaks also of ‘communicatively structured lifeworlds that reproduce themselves
via the palpable medium of action oriented to mutual agreement’, i.e. via
communicative action.
Reaching understanding seems to be the condition for human powers in
general. It is in this sense that we should talk about communicative action as
the (re)productive mechanism.

The capacity for a reflexive self-monitoring of one’s own causal interventions

in the world, to be aware of one’s own states of awareness during one’s
activity, is intimately connected with our possession of language, conceived
as a system of signs apt for the production and communication of information.
(Bhaskar 1979:104)

Both human powers and internal social relations are dependent on language
and systems of signs.
The social being is meaningful, and internal relations of social being must be
located at the level of understanding. However, there must also be lower-level
mechanisms capable of (re)producing language and ‘ideas’, or systems of signs.
These mechanisms include the genetically programmed, bodily powers of human
beings to learn practical skills and, thereby, work with and through meanings.
Practical bodily skills include walking, or making bodily gestures, or using the
human voice to make signs or handling a pen or knife and fork. Possession of
language requires practical skills, and practical skills are subject to natural laws
and determinations. ‘We never cease, and never can cease, to sustain relations
with all three orders of reality—natural, practical and social’ (Archer 2000:9).
However, in order to work with meanings one has to learn, via practical skills,
to understand meanings and follow social rules, both constitutive and regulative.
From the very early infancy onwards, embodied actors have the power to resist,
or take exception to, these rules. Later, with maturity, they can reflect upon,
criticise and ultimately change these rules.
The conception of structures of internal relations thus presupposes a theory—
or an iconic model—of both practical and communicative action. Hence, (internal)
social relations are intimately connected to action. This indicates that social
sciences cannot specialise in studying (internal) relations in isolation of
interactions, although Bhaskar (1979:56) seems to claim so. Social sciences have
two simultaneous sides: the analysis of action and the analysis of structures.
True, actions and structures are different kinds of things, irreducible to each
other, although ontologically dependent. They can be separated momentarily
for the analytical purpose of studying continuity and change, and the temporal
dialectics between agency and structure (cf. Archer 1995:168–70 passim). My
106 Explicating a critical realist methodology

point is, however, that historical social sciences must explain outcomes, which
occur only in shifting and changing open systems. Therefore, any explanatory
iconic model of concrete outcomes has to articulate a unified account of the complex of
relevant material and efficient causes at work in a well-specified area of time and space.

The vague and ambiguous nature of internal relations

Internal social relations are typically both vague and ambiguous. They are
vague because lay-actors have partially incoherent and somewhat wavering
practical and discursive theories of the social world in question. Internal relations
cannot be more exact and clear than the meanings circulated, stored and
transformed in the practical social world, in social practices. Internal relations
are ambiguous because actors hold conflicting theories about social institutions
and practices, and these conflicting views are connected to differentiated horizons
and horizon-constituted interests in human societies.
In other words, in a complex world, as contrasted to, say, a more
homogeneous and unchanging medieval village, there is neither any a priori
guarantee of the sameness of the meanings, nor any singular causal mechanism
capable of producing the same meanings for all actors (despite the existence,
since the eighteenth and nineteenth centuries, of centralised national systems of
education devised also for the purpose of homogenising language and meanings).
Because there can be no social reality without meanings and internal relations,
the thesis concerning ambiguousness can also be read as a thesis about the
social ontology of discordance (Connolly 1987:10–16). Because the mechanisms
producing meanings are essentially communicative, and because of the ubiquity
of ontological discordance, there tend to be constant dialectical disputations
and contradictions in the ‘shading, soft-edged, open-textured conceptual world
of Wittgenstein and Waismann’.
Dialectical contradictions, in the sense specified by Nicholas Rescher (1977:62–
3), do thus exist, and they generate, among other things, open disputations and
political argumentations. These conflicts are a central element of the political
aspect of the social world (see, for example, Alker 1988:805–20). Consider
Figure 4.1. The disputation begins with an assertion of a thesis (P) with one
grounding (Q), and a categorical assertion! Q (‘it is the case that Q’). The opponent
responds in this case with a provisoed denial ‘if both Q & R then ~P, and I
suspect that Q & R’. The first assertion in this kind of debate may, for instance,
be that ‘a trial should be defined, as I do, as a public examination of evidence…,
because otherwise there would be no guarantee of the fairness of trial’, and the
opponent might respond that ‘you may be partially right, but there are also, I
suppose, many occasions where publicity would only do harm to fairness, and
hence I can see no reason to define the trial in terms of publicity’. In the
dialectical logic the denial of ~P is not just to assert P, for there is also refinement
(amplification, improvement, introduction of new and deeper grounds) in the
transition from Q to (Q & R & S), for instance (Rescher 1977:66).
 Critical realist social ontology 107

Figure 4.1 An example of a dialectical contradiction

There can thus be dialectical disputations and contradictions in the ‘shading,

soft-edged, open-textured conceptual world of Wittgenstein and Waismann’.
Internal relations tend to be ambiguous, and they can become a matter of
explicit disputation and, in this sense, they can become political. Moreover,
rhetorical and dialectical disputations can enable and induce transformations of
concepts and theories (see Ball 1988). Since concepts and related theories
constitute the vague and ambiguous internal relations of historically positioned
practices, their transformation may lead to profound institutional changes.
Moreover, as Archer (1989:103–42 passim) has stressed, many meanings are related
and thereby form structural wholes. There are logical relations between components
of meaning structures. Components of meaning structures can also be contradictory.
Contradictions may be simply lived by for a long time, but they can also be taken up
as a basis of cultural and political criticism. In that moment, it depends on the dialectics
of power whether the critics can achieve changes or whether the contradictions are
simply suppressed by those who are interested in preserving the status quo.

External relations and modes of action

What exactly is the status of external relations in the CR social ontology? I

shall use the following rather general definition of external relations:

Def. 3: An external relation RAB is a contingent substantial relation between positions

of positioned practices or occupied positions of position-practice systems.

Consider the previous example of the social practice of trial by jury, which
might include regulative rules something like the following (Greenwood
1987:188):

1 ‘Jurors should not let their own prejudices influence their judgement.’
2 ‘Counsel should not make emotional appeals to the jury.’
3 ‘Judges should not intimidate witnesses.’
108 Explicating a critical realist methodology

For example, the second rule regulates the external relation between counsel and
jury engaged in the practice of trial by jury. Of course, they may be also
internally related. Generally, internal and external relations do not exclude
each other. Giddens claims that ‘these two senses of right and wrong always
intersect in the actual constitution of social practices’ (Giddens 1979:82). Onuf
goes even further and asserts that all rules are both constitutive and regulative:
‘a constructivist always sees this distinction as arbitrary and unhelpful’ (Onuf
1989:85). However, this contention would imply that a violation of a rule
such as ‘counsel should not make emotional appeals to the jury’ (2, above)
would change the meaning (identity) of trial, counsel, jury or legal argument,
which is clearly not normally the case. Hence, the two senses of right and
wrong may well intersect in the actual constitution of social practices, and yet
the distinction between constitutive and regulative rules is not only important
but also ontologically necessary.
Suppose that counsels, jurors and judges believe that these (customary,
moral or legal) regulative rules exist effectively. This does not imply that they
follow these rules. How, then, can structures of external social relations generate
action and changes? Let us first examine how Bhaskar conceptualises these
kinds of structures. Bhaskar (1979:51) uses the term ‘position’ in connection
with the concepts of ‘places’, ‘functions’, ‘rules’, ‘tasks’, ‘duties’, ‘rights’, etc.,
but he unfortunately does not specify their relationships. ‘Places’ and ‘functions’
seem to be near to the concept of ‘position’, and rules partially constitute
positions. These concepts do not, however, illuminate the matter very much.
Strikingly, terms like ‘tasks’, ‘duties’ and ‘rights’ seem to point to the
Durkheimian direction in social theory. According to Emile Durkheim (1969
[1938]:246–7), ‘social facts’ consist of ways of acting, thinking and feeling,
external to the individual, and endowed with the power of emotional, moral or
legal coercion. This characterisation notwithstanding, Durkheim is not able
to show that ‘social facts’ necessarily force individuals to conform to external
and/or regulative social rules. This holds true for Bhaskar’s TMSA as well:
regulative rules alone cannot produce changes. Nor are they a necessary
condition for action in general, for the relation between regulative rules and
action is by definition contingent (although, of course, a regulative rule is
typically an I NU S-condition for a particular action). Because of this
contingency, the analogy to a magnetic field does not hold at all in the case of
structures of external social relations, and thus it seems that this analogy
carries a misleading picture of social relations.
Any theory or analysis of external social relations requires a theory of social
action. A theory of social action should include taxonomy of homoeomorphic
iconic models of action. A homoeomorphic iconic model is one where the
subject of the model is the same as its source of analogy (picture, metaphor).
Teleiomorphic models are a sub-category of homoeomorphic models, and they
are constructed by means of selective idealisations (some of the properties of
the source-subject are made, according to some scale of values, more perfect
than the source-subject’s properties) and abstractions (some properties of the
 Critical realist social ontology 109

source-subject are chosen by some relevancy criteria) (see Harré 1970: Ch. 2,
esp. pp. 41–2). Weber’s ideal-types (Weber 1969) are a case in point, except
that Weber does not distinguish clearly between idealisation and abstraction.
According to Weber, an ideal-type:

offers us an ideal picture of events,…which brings together certain relationships

and events of historical life into a complex, which is conceived as an internally
consistent system. Substantively, this construct is itself like a Utopia which has
been arrived at by an analytical accentuation of certain elements of reality….
[Hence] we can make the characteristic features of [social] relationship[s]
pragmatically more clear and understandable by reference to an ideal-type.
(Weber 1969 [1949]:497)

The Weberian method can be re-interpreted along CR lines. An iconic model—

an ideal picture of events and relationships—includes existential hypotheses
concerning, say, the primary orientation of action. Weber did not want to use
ideal-types merely in distinguishing between different action-types. Rather, he
seems to have seen—at least intuitively—that action and social relations together
form a complex, which is responsible for the phenomena to be explained. The
problem is that we should not require, contra Weber, that an ideal picture of
events and relationships is internally consistent. There are inconsistencies and
contradictions in the real world. It is a condition of the adequacy of a precise
theory—or an iconic model—of an ambiguous (contradictory, vague,
indeterminate) phenomenon that it should precisely characterise that
phenomenon as ambiguous (contradictory, vague, indeterminate) (see Searle
1983:78). Moreover, the formation of an iconic model must rely on the idea
that the adequacy and accuracy of the model must be tested by double-
hermeneutic empirical means. The task of social scientists is to find out how
the real world works.
Ontology in the philosophical sense should be addressed to the most abstract
constitutive potentials of social life: the generic human capacities and fundamental
conditions through which the course and outcomes of social processes and
events are generated and shaped in manifold ways (Cohen 1987:279). There is
a strict limit to philosophical ontology. Does this limit contradict the idea that,
in order to grasp the nature of external relations and regulative rules, we need
a general theory of social action? No—rather it reveals that models of action
must be as historical as the modes of social action. In order to grasp thoroughly
the relationship between action and structures it is not possible to stay at a meta-level.
Social action-types are historically constituted. ‘No general, transhistorical or
purely philosophical resolution of these problems is possible’ (Bhaskar 1983:87).
Habermas (1984 [1981]:82–101) has developed a theory of social action. It
needs to be re-interpreted because it is based on the heritage of German idealism,
which insists on the irreducible contribution of an active, world-constituting
observer-subject. The idea of a ‘world-constituting observer-subject’ contradicts
110 Explicating a critical realist methodology

realist ontology. Habermas (1984 [1981]:82–4) ‘would like to replace the

ontological concept of “world” with one derived from the phenomenological
tradition’. He thinks these ‘phenomenological worlds’ can be differentiated
according to distinct validity claims of speech acts.2 A realist should interpret
Habermas’s ‘worlds’ as fallible and corrigible iconic models, as teleiomorphs of
historical social action. Habermas’s theory has been developed in twentieth-
century Europe, and may be too Eurocentric to be applicable outside the (late)
modern West. This cannot be decided a priori. Perhaps this theory is suitable
for the analysis of some or many other social formations? It is possibly in need
of important corrections, additions or simplifications even in the (late) modern
Western contexts.
Habermas distinguishes four types or modes of social action. First, in teleological
action the actor attains an end or brings about the occurrence of a desired state by
choosing (making a decision concerning) the means that offer the promise of
being successful in the given situation. If the decision is based on calculations that
take into account decisions on the part of at least one additional goal-directed
actor, then teleological action is called strategic. This model is usually interpreted
in utilitarian and rational choice terms. Second, the model of normatively regulated
action refers to members of a social group who orient their action to common
values, and the central concept of this model is ‘complying with the norm’, which
means fulfilling a generalised expectation of behaviour. Third, the model of
dramaturgical action describes a mode of action where participants in interaction
constitute a public for one another, before which they present themselves
(strategically or sincerely). Fourth, there is the model of communicative action, which
is, as was argued above, necessary for the explanation of the (re)production of
intersubjective meanings and internal relations. Communicative action refers to
the interaction of at least two subjects capable of speech/writing and action who
establish interpersonal relations, whether by verbal or extra-verbal means, although
language is given a prominent place in this model. The central concept of this
model is ‘interpretation’, which refers to the process of practical or discursive
learning, negotiating, criticising, disputing or debating definitions and conceptions
related to the relevant rule-constituted situation.3
This taxonomy of social action can be used to resolve the problem of the
‘generative powers’ of regulation. Why, for instance, would a counsel comply
with the norm ‘counsel should not make emotional appeals to the jury’?
Supposing that it is a legal norm, there are many possibilities. First, a counsel
might make calculations concerning sanctions, the utility of his or her good
reputation, etc., and comply with the norm for strategic reasons (and reasons
are causes). Second, a counsel may regard the norm as valid, either for traditional
or rational reasons, and comply with the norm because he or she is orienting
his or her actions to certain common values of his or her group. Third, a
counsel may try to present himself or herself as a certain kind of person, and
hence comply with the norm as a rule of dramaturgical action. Of course, a
counsel may also comply with the norm ‘mechanically’ (which is not the same
as ‘for traditional reasons’), as part of routine practice, or refuse to comply with
 Critical realist social ontology 111

it. This illustrates that while positioned practices always contain external relations
which are real, their existence and causes of existence are conditional. Their
causal effectiveness always depends on other necessary conditions, too (see def.
1), especially on the causally effective reasons and motivation for action. Models
of action are essentially models about the prevalent rules and principles of practical
reasoning. We are thus back to studying constitutive rules and internal relations
of different historical social worlds.
Consequently, models of social action must be falsifiable by means of double-
hermeneutic empirical research. We may assume that certain modes of action
are institutionalised into positioned-practices systems. For example, it is very
common to argue—or presuppose—that teleological/strategic action is internally
related to the capitalist market economy system. However, this is not
automatically or ahistorically so. For instance, the neo-liberalism of Thatcher
and Reagan can be seen as a political project, which attempts to make actors
comply with the strategic model of action in all of their everyday practices, thus
extending and intensifying the transformations already partially accomplished
by eighteenth- and nineteenth-century economic liberalism. Yet this
transformation cannot be completed. The transcendence of social systems in
time makes communicative and other non-strategic modes of action necessary.
In fact, these modes can be found even in capitalist practices.
It is possible to distinguish between different kinds of strategic action:
difference maximising, hasty egoism, far-sighted egoism and altruism, depending
on how actors understand their (external) relation to other actors (see Taylor
1976; Axelrod 1984; Frochlich and Oppenheimer 1984). These types occur in
a capitalist market society. Second, different types of strategic action can be
seen as orientations, which are based on deeper grounds, theories and motives (Hollis
1988:140–1). Thus, forms of strategic action may be essentially connected to
normative or expressive logic of action.
A capitalist market society consists not only of market transactions but also of
social relations inside hierarchical organisations, such as multinational corporations,
or between, say, privately owned organisations and state compartments.
Organisations facilitate co-operation among actors. This facilitation of co-operation
is to a varying extent due to the local degree of hierarchical surveillance and
control, but also because organisations create enduring contexts for social action.
In organisations there can develop a trustful atmosphere where the mode of
action changes, to some extent, towards that of normatively regulated action, at
least in the ‘high trust positions’ (Giddens 1987a:157–62). Also, communicative
action in organisations has better prerequisites for success than in the context of
markets (cf. Williamson 1975:37–9; also Perrow 1986). Furthermore, Galbraith
(1967) has shown how the actions of the technocratic staff of large firms can also
be interpreted in terms of the dramaturgical model.
So we should not be blinded by our definitions and analysis of the internal
relations between modes of action and position-practice systems, but at least allow
sufficient ‘flexibility and a certain amount of cross-checking of observation or
reflection under one group of concepts by another’ (Sayer 1984:54). For example,
112 Explicating a critical realist methodology

Habermas’s theory of the historical development of ‘System’ and ‘Lebenswelt’ is based on

a very simplistic assumption about the internal relation between the capitalist market
economy system and teleological/strategic action (see, for example, Tuori 1988:114).
To emphasise, the analysis of external and internal relations of positioned-practices
systems and modes of action must also be empirical and double-hermeneutic.
Philosophers and theorists should not think that they can envisage from their
armchairs how the world really works.

The relationship between action and power

In Giddens’s theory of structuration, there is also an internal relation between

action and power. Giddens (1979:55) defines action as follows:

Def. 4: First, the notion of action has reference to the activities of an agent. Second, it is
a necessary feature of action that, at any point in time, the agent ‘could have done
otherwise’. Third, action involves a stream of actual or contemplated causal interventions
of corporeal beings in the ongoing process of events-in-the-world, thus producing various
outcomes.

Intended action is important but nevertheless only one category of an agent’s

doing or his refraining; action may be unintended, and outcomes of action may
also be routinely unintended, although ‘agents virtually all the time know what
their actions are, under some description, and why it is they carry them out’
(Giddens 1987a:3). Giddens (1981:49) goes on to argue that action involves
intervention in the world, thus producing definite outcomes. From these notions
he concludes that power, in turn, can be defined as transformative capacity that
refers to agents’ capabilities (competencies and facilities) of reaching definite
outcomes.
As can be seen, I preferred in (def. 4) the expression ‘producing various outcomes’
to the Giddensian formulation ‘producing definite outcomes’. Giddens’s own conception
of power seems to be meant to resolve the problem of taking into account the reasonable
core of both individualistic and collectivist theories of power. However, Giddens’s
conceptualisation largely excludes those aspects of power that have been thematised
by Michel Foucault, namely the unintended existential consequences of different forms
of power. Giddens (981:49) thinks that Foucault has merely developed a modern
version of the notion of power, which regards power above all as a property of
collectivities. This is a partial misunderstanding, for Foucault writes that:

if we speak of the structures or the mechanisms of power, it is only insofar as

we suppose that certain persons exercise power over others [in a sense that
there is] an ensemble of actions which induce others and follow from one
another’.
(Foucault 1982:217)
 Critical realist social ontology 113

This is a very structurationist (or even individualist) formulation. What I consider

to be original in Foucault’s account of ‘productive power’ is the idea that power
also produces elements of the social world, although this often escapes the horizon
of actors. When actors utilise, invent or innovate resources—normally to grasp or
change the world—in the course of social interaction, they also produce discursive
knowledge, techniques, practical knowledge and skills, which constitute internal
social relations. Many of the consequent effects may be unintentional.
Consider two examples. First, Marxist attempts to grasp the historical meaning
of the First World War had effects of power on the subsequent Soviet society.
Although the Leninist idea of dictatorship of the Communist Party predates the
war, the war was a coincidence that made the Bolshevik Revolution possible.4 In
the period 1914–20, Russian Bolsheviks were first trying to comprehend the First
World War, and soon their own revolution and the Western intervention into
Russia. They, and Lenin in particular, transformed Marxist discourses on the
basis of these experiences. These transformed discourses had far-reaching effects
of power by constituting and legitimating practices of dictatorship. Although some
party members had been warning about the dangers of centralising power within
the party, the Bolshevik innovators mostly did not foresee the often very violent
and repressive effects of power resulting (also) from these redefinitions (for details
of these discursive transformations, see Roig 1982:49–64).
Second, when the Finnish political elite learned—a process that included both
individual learning and political struggles—from the experiences of the two lost
wars with the Soviet Union during the Second World War that they should be
political realists instead of ‘idealist nationalists’, they did not merely change their
foreign policy. They also constituted and legitimated new practices for governing
domestic political sphere, within which the transnational struggles of the Cold
War were both taking place and, to the extent possible, suppressed until the time
of Gorbachev (see Patomäki 1993). Novel ways of positioning emerged and new
rules and principles of governing ‘things political’—and dominating other actors to
comply with the wants and understandings of the ruling elites—were adopted.
Actors such as [Link] did not foresee many or most of the resulting effects
of power.

Actors more or less know what they are doing when they do it and can often
be quite clear in articulating it. But it does not follow that the broader
consequences of these local actions are coordinated, …[for actors] ‘don’t
know what they do does’.
(Dreyfus andRabinow 1982:187)

I think these examples, and the general Foucauldian point, indicate that there are
good reasons to widen the definition of power to include the longer-term
existential (ontological) effects of power. As will be seen, this wider definition can
liberate from the theoretical puzzles that Giddens has been struggling with. Thus
the simple def. 5:
114 Explicating a critical realist methodology

Def. 5: Power is the transformative capacity of agents’ capabilities (=resources

as competencies and facilities).

Action presupposes transformative capacity, i.e. power. To use power is to act

and to act is to use power. There are also many other innovative and interesting
features in this definition of power. As action and structures form an ontological
duality, so are power and structures also connected. Moreover, this non-zero-
sum notion of power is not tied conceptually to a conflict of interest; rather, the
relationship between power and conflict is contingent. Moreover, if power is not
defined in terms of ‘definite outcomes’, as Giddens does, the basic insight of
Foucault can be incorporated into this social ontology.

Social structure

Giddens (1979:64) defines structures as ‘structural properties…[which] can be

understood as rules and resources’. He recognises the existence of:

1 knowledge—as memory traces—of ‘how things are to be done’ on the part of

social actors (i.e. actors must be competent);
2 social (and positioned) practices organised through the recursive mobilisation
of that knowledge; and
3 the transformative capabilities, i.e. power, that the production of those practices
presupposes.

Giddens (1979:66) characterises structure as ‘rules and resources, organised as

properties of social systems’ and social system as an entity ‘produced and reproduced
in interaction…, via the application of generative rules and resources, and in the
context of unintended outcomes’. This raises the question of whether structures as
rules and resources do constitute, generate, organise and/or establish social relations
(as Giddens at least seems to argue, and as David Sylvan and Barry Glassner
(1985:90) claim that both Giddens and Bhaskar do), or whether structures should
rather be seen expressly as social relations. Bhaskar’s (1979:51) argument is, however,
that positions and practices can be individuated only relationally. What does this then
mean? Does it imply that social structures (and ‘mechanisms’) consist of relations? Is
it an ontological or an epistemological argument? Bhaskar maintains that:

the initial conditions in any concrete social explanation must always include or
tacitly presuppose reference to some or other social relation (however the
generative structures invoked are themselves best conceived). And it is, I suggest,
in the (explanation of the) differentiation and stratification, production and
reproduction, mutation and transformation, continual remoulding and incessant
shifting, of the relatively enduring relation presupposed by particular social
forms and structures that sociology’s distinctive theoretical interest lies.
(Bhaskar 1979:51–2)
 Critical realist social ontology 115

This suggests that Bhaskar makes an epistemological case for a special discipline
called ‘sociology’. (This argument may include ontological components, inasmuch
as it is grounded on the reality of internal relations.) Bhaskar, however,
distinguishes between generative structures and social relations, which therefore
must be different things. Since he too apparently views generative structures in
terms of ‘rules and resources’, then the claim that sociology’s distinctive
theoretical interest lies in the sphere of social relations seems to be somewhat
undermined. However, consider the following:

The social sciences abstract from human agency to study the structure of
reproduced outcomes, the enduring practices and their relations; the social
psychological sciences abstract from these reproduced outcomes to focus on
the rules governing the mobilisation of resources by agents in their interaction
with one another and nature. If the sphere of the former is social structure,
that of the latter is social (and socio-natural) interaction. They may be linked
by the study of society as such, where society is conceived as the system of
relations between the positions and practices (or positioned-practices) which
agents reproduce or transform….
(Bhaskar 1986:124)

After this, Bhaskar refers to Karl Marx, who wrote in the Grundrisse that ‘society
does not consist of individuals, but expresses the sum of interrelations, the
relations within which the individuals stand’. Bhaskar seems to argue that the
emergent properties at the level of society are social relations, that these relations
form structures, or mechanisms, which generate changes, and that these
mechanisms are analogous to natural mechanisms like a magnetic field. Giddens,
on the other hand, lays the emphasis on rules and resources, which constitute
and/or generate social relations through actions. For Giddens, social relations
seem to be by-products of the action, rules and resources nexus.
I have already argued that the analogy between a magnetic field and structures
of social relations does not hold. First, social rules and resources usually only
enable or allow certain possibilities and disable, disallow or exclude certain
others (Dessler 1989:453). Moreover, if it is true that rules and resources
constitute and/or conjunctively generate social relations, then rules and resources
are ontologically prior to external social relations, as Giddens seems to be arguing.
Consider two examples. Let RAB be (1) a relation between a wife (A) and a
husband (B), and (2) a relation between a modern statesman (e.g. a prime
minister, a foreign minister or a president A) and a national electorate (B). Both
relations can be analysed in terms of position-practice systems (a marriage;
liberal-democratic elections and the imagined community of a ‘nation’ in the
nation-state system), rules (constitutive and regulative), resources (connected to
powers or transformative capabilities) and related modes of action.
116 Explicating a critical realist methodology

Without these substantial qualities the relations (1) and (2) are empty, abstract
in almost the same way as formal deductive logic is. Moreover, these relations
cannot be understood or explained without explaining relevant action, rules
and resources, which are necessary for these relations to exist. Finally, as a
direct consequence of the dual character of action and structures (which implies
the need for a theory of action), there is a necessary—although perhaps loose—
connection between individual actors’ reasons and the meaning of social actions
and relations, although Bhaskar argues that:

the meaning of an action…[or] its correct identification as an act of a particular

type in a particular language and culture is always and in principle independent
of the intention with which it is on some particular occasion, by some particular
agent, performed.
(Bhaskar 1979:108)

This argument amounts to the same thing as Ferdinand de Saussure’s ‘langue’/

‘parole’ distinction. Its initial plausibility stems from the particular examples
utilised in this context. Saying ‘hello’ is a trivial example of an enduring,
repetitious and nearly universal practice. As such it may be a misleading example.
Even when saying ‘hello’ is merely a routine act, there must be practical
knowledge concerning the meaning of that act. Without that knowledge the act
would not have the meaning it has, and it would not be what it is. On the other
hand, the meaning of saying ‘hello’ may also, in relevant respects, be dependent
on particular reasons and understandings of actors. On some occasions there
may be a specific point in saying ‘hello’. In the beginning of a philosophical
essay it may convey a particular meaning. Also, in everyday practices, it may
assume, say, ironic or sarcastic connotations. In English, it can be pronounced
in various ways, to indicate local commonality or cultural distance between
actors. Other languages have their own variations and differences. Human beings
are not automatons, and cultures differ.
It is true that individuals engage in positioned practices that have been defined,
created and (re)produced historically before their entry, and which often are
reproduced as an unintended consequence of their actions. Yet there is a necessary
connection between actors’ knowledge and intentional reasons and the meaning
of social actions/practices/relations. In repetitious contexts, however, this necessary
connection may be so self-evident that it is easy to forget or, methodologically,
bracket.
On the basis of this discussion, how should we use the term ‘structure’?
Giddens’s definition of structures as rules and resources does not sound plausible
because the term ‘structure’ is normally used in connection with compositions
and/or relations. Moreover, as Archer (1995:93–101) has argued, Giddens’s idea
that structures are rules and resources, which only exist in the moment of
interaction, tends, ultimately, to reduce structures to action and reduce reality to
actual interaction. Yet structures and action are distinct and therefore analytically
 Critical realist social ontology 117

separable, and real powers and capabilities cannot be reduced to actual instances
of interaction. Hence, I would like to propose the following definition:

Def. 6: The term ‘social structure’ refers to internal and external relations of a
positioned practice; these relations are implicated and/or generated by the
components of the relevant causal complex.

The internal and external relations are constituted and generated by the relatively
enduring nexus of rules, resources and (re)productive actions. Constitutive rules
make action possible and also regulate it. Constitutive rules imply internal
relations, which form wholes with all kinds of logical relations between the
parts. They are also action-dependent in a sense that actors have the power to
reflect upon, criticise, deviate from and change the constitutive rules. External
relations are the outcome of action generated by (1) rules; (2) resources as
embodied capabilities and position-endowed facilities; and (3) intentional reasons
for actions. External relations are contingent. They are (re)produced in and
through social practices, with the possibility of ‘doing otherwise’ by the actors
as ever present. Intentional reasons, even if idiosyncratic, are embedded in the
contexts of interaction as well as in the background discourses (and related
practical skills and capabilities). Also, the logics of action are constituted by
historical rules and principles. In concrete empirical research, to uncover these
layers and components requires a lot of painstaking, even meticulous, work.
Def. 6 gives, clearly, a relational definition of structure. Is it also possible to
define social structure in compositional terms (e.g. ‘age structure’, ‘occupational
structure’, ‘structure of industry’)? Although this is certainly possible, and in
some contexts also legitimate, two essential things must be kept in mind. First,
from the causal analysis point of view it is necessary to examine substantial
relations of connection and interaction, but not formal relations of similarity
and dissimilarity. Thus, in order to perform causal analysis, compositions either
should be interpreted in relational terms, translated into them or explained by
means of relational conceptions. Second, all statistical information is the product
of a two-stage interpretation process. The first stage is the official process of
definition prescribing the criteria for inclusion in the category. The second stage
is the registration of instances, which is done by persons of varying competence
and variously biased (Maclver 1964:137; see also Alker 1996:332–54). Hence,
(1) the relatively quantitative compositional knowledge is no more objective
than the essentially qualitative relational knowledge; rather, (2) it is causally
and, therefore, ontologically secondary to the conceptually much richer relational
knowledge.

Dialectics of open, stratified systems

With def. 6, we both can see society as a relational entity (somewhat contra
Giddens) and at the same time deny that social structures are ontologically or
118 Explicating a critical realist methodology

causally analogical to natural structures or mechanisms (contra Bhaskar). What

then is a social system? In one place Giddens (1979:66) defines system as ‘reproduced
relations between actors or collectivities, organised as regular social practices’. On
another occasion, Giddens uses the expression ‘patterns of relationships between
actors or collectivities reproduced across time and space’ (Giddens 1981:26). The
first definition seems to be derivative from def. 6, but the other seems to carry a
somewhat more empiricist picture of the social world. Of course it is true that social
activities always occur in time-space, but ‘the timespace settings of interaction’
(Giddens 1987a:144) are organised as relational social practices. Most importantly,
the emergent powers and properties of systems cannot be reduced to actual activities
and interaction. Thus, ‘system’ should be defined as follows:

Def. 7: Social system is composed of time-space-situated relations between

positions of individual and collective actors; social system is organised as regular,
interconnected positioned-practices of individual and collective actors, but the
components of a social system are also connected through manifold unintended
consequences of action. Social systems have emergent powers and properties,
and these may constitute their raison d’être.

This is a relational definition, but one that does not deny that social activities
occur only in tim e and space. All collectivities are social systems, but
there are other wholes, which can be identified as systems as well, such as
international society or capitalist world markets. Social systems have no essences
independent of the contexts of action; they change together with those relational
contexts. Social systems can, however, have emergent powers and properties.
The problem is to avoid reification and mystification of systemic powers and
properties. They must be analysable in terms of actions and positioned practices
occurring in concrete spatial settings, as already outlined in Chapter 3.
The IR levels would preclude studying open-mindedly levels in these two
possible senses. Social systems are not only open, they are also overlapping
and, in parts, intra-related. States, for instance, are themselves open systems,
intra-and inter-related to various other systems consisting of, say, practices of
capitalist world economy, international institutions and systems of governance
of global political economy. In spatial terms, local developments within states,
as well as parts of the state apparatuses themselves, can be causally and
existentially both intra- and inter-related to a number of collective actors or
systems. Emergent powers and properties, as well as hierarchical power relations,
should be studied open-mindedly from this kind of realist perspective. Emergent
powers and properties should not be organised, in one’s mystifying imagination,
to accord with the simple IR levels just to yield the convenient possibility of
continuing to reproduce the international problematic. To repeat once more the
argument against armchair philosophising, the dialectics of these layered,
relational and open wholes can only be described and explained by means of
detailed and systematic empirical studies of concrete practices and relations.
 Critical realist social ontology 119

Causal complexes and the freedom of human action

The notions of context and complex form the core of CR methodology. According
to def. 1, cause is an insufficient but necessary part of a condition that is itself
unnecessary but sufficient for the production of a result, i.e. the INUS-condition.
There are five necessary social components in any causal complex (K) capable
of producing events, episodes, tendencies and the like, namely: actors (AR);
regulative and constitutive rules (RU); resources as competencies and facilities
(RE); relational and positioned practices (PRA); and meaningful action (AN).
Also, actors and their characteristic logics of action are historically constituted.
Together, these related and interdependent components, and the emergent powers
and properties of consequent social systems, create the sufficient but unnecessary
complex for the production of a result. Hence, there is never a single cause but
instead always a causal complex K={AR, RU, RE, PRA, AN}, perhaps including
emergent properties and powers. This formulation does not disclose the layers
of social beings and systems, or the unintended consequences and the spatio-
temporality of (interactions. These have to be studied as well. Nonetheless, the
notion of causal complex provides a useful way of systematising the elements
that will have to be studied in any given context.
A result X may be produced both by K1 and K(i > 1), i.e. there may be many
different sufficient conditions capable of producing the same result. On the
other hand, action is one of the elements, and actors can do things otherwise.
A context of action is relative to particular actor j, AR j and to AR j ’s and other
actors’ interpretations. As Giddens (1983:78) puts it, ‘context’ is always the
contextuality of action. A context must also be relative to a particular moment
in the continuous flow of time. Thus, CONTEXT=Kt _ {AN j t}, i.e. the causal
complex at moment t with the meaningful action of AR j at the moment t
excluded. This comes close to the difference between material and efficient
causes. CONTEXT forms the complex of material causes for AN j t,whereas
the real reasons for this action form the efficient causes of the actions of this
positioned agent ARj . His or her actions, in turn, are part of the CONTEXT
of other actors. Reasons for action are layered and may be circulated in
institutionalised practices where they can have a taken-for-granted nature, and
where many of the deeper assumptions—and the required practical skills—often
go unnoticed. Yet there is always a variety of possible actions in a given
context. This is just a consequence of def. 4, according to which it is a
necessary feature of action that, at any point in time, the agent ‘could have
done otherwise’. Two cases can be distinguished:

1 AN j t is consistent with all the elements of CONTEXT

2 AN j t is inconsistent with one/some of the elements of CONTEXT.

Although, because of the typically ambiguous and contradictory nature of social

contexts, (1) is never strictly possible, the analytical distinction between (1) and
(2) is important. In many respects, (2) is the interesting case here. With what in
120 Explicating a critical realist methodology

particular can action be inconsistent in a context? Action can break the constitutive
or regulative rules of the context. This may be due to a mistake (Winch 1971
[1958]:32), or it may be due to intentional deviance. In both cases, there may also
be some innovative implications, intended or unintended, learning-inducing or
pathological, leading to a closure. When innovative implications are unintended
they will reveal their innovative character only later in the flow of time, though
this is often the case for intended innovations as well. In a ny case, a careful
examination of the CONTEXT of AN j t and the nature of AN j t will disclose the
novelty and originality of that act(ion) (cf. Skinner 1988:79–118).
An individual actor can move from one context to another. It is possible that
an actor can act simultaneously in two or more different contexts, or at least
contexts can penetrate each other. An individual actor exists as a bodily being,
with embodied competencies. He or she is capable of remembering his or her
own past and reflecting, planning and ‘doing otherwise’. Through his or her
memory and embodied practical skills and capabilities, he or she is in fact also
a personified and personal ‘intersection’ of different contexts. For instance, an
individual actor can bring the generative rules and capabilities he or she has
acquired in one context to bear in another context. This may turn out to be,
say, inappropriate, curious, amusing, exotic, stubborn, immoral, rebellious or
creative. The ‘intersection’ is personified because the actor’s complicated and
(also always) ambiguous identity, as well as his or her stock of resources and
expectations concerning rule-following, is formed during his or her life path
through different relational contexts. Also, the ‘intersection’ is personal because
it is a unique composition, and it may include innovative features.
Freedom of human action is obviously not limited to the individual freedom to
‘act otherwise’. It is also a matter of changing realities: the constitutive and
regulative rules; the social relations; the generative structures and complexes. To
effect changes is typically a collective endeavour. The forging of transformative
social identity is a complicated and fragile process. If this process is successful, it
yields the parties emergent powers and properties—in the context of unintended
consequences of action. The realist grounds for transformative actions will be the
topic of Chapter 6.

Conclusion

The primary purpose of this chapter has been the explication of CR social
ontology by discussing the concepts of ‘action’, ‘structure’, ‘power’ and ‘system’.
The result is just an open-ended articulation, a flexible starting point for
systematic social scientific explanations. It is not a philosophical foundation.
Rather, it is—and should remain—open to further argumentation, experience,
criticism and change. That what is should be analysed and explained by social
scientists and lay actors, not by metaphysicians.
The main objects of criticism of this chapter included Bhaskar’s naturalist
metaphors and his vacillating notion of social structure. I used a re-interpreted
 Critical realist social ontology 121

version of Habermas’s theory of social action to fill gaps in Bhaskar’s ontology.

On the other hand, I argued that Giddens has introduced too narrow a concept
of power and an implausible and tendentially reductionistic definition of
‘structure’. These were corrected along relationalist lines. On that basis, I
explicated the dialectics of emergent powers and properties of open historical
systems.
Finally, a sketch of adequate social scientific explanation was developed.
In order to explain phenomena, a social scientist should build iconic models
of the causally efficient, historically formed complexes that produce the
phenomena he or she is interested in explaining. Causal explanation and
determination do not, by any means, deny human freedom. Human beings
possess real causal powers to reflect, debate and plan. Human beings can
choose to prefer the status quo or, alternatively, to resist, to act otherwise,
to imagine alternatives and, even, to forge new social identities. As will be
soon clarified, more adequate social scientific explanations can in fact increase
human freedom.

Notes

1 See Sayer (1992:82). A ‘good’ example of analysis of internal and external relations
without this distinction can be found in Elster (1978:22–5).
2 Habermas has, for sure, attempted to escape from the phenomenological philosophy
of subject/consciousness. Habermas (1987:295) speaks of ‘communicatively structured
lifeworlds that reproduce themselves via the palpable medium of action oriented to
mutual agreement’, i.e. via communicative action. Thus ‘lifeworlds’ are not, in
Habermas’s theory, situated in subject/consciousness. However, there are problems
in Habermas’s theory. First, he argues that the task of reconstructive and critically
Kantian philosophy is to articulate the always already presupposed normative validity
claims of speech-acts, and then he differentiates lifeworlds in accordance with these
validity claims. Second, Habermas grounds his theory of action on the theory of
differentiated lifeworlds. Finally, Habermas bases his theory concerning ‘system’ and
‘lifeworld’ on his theory of social action. The result is a philosophical-ontological
theory of modern societies, with mutually exclusive spheres. This theory exceeds the
reasonable limits of philosophical ontology. In contrast, an iconic model is proposed
in terms of fallible, corrigible and controllable existential hypotheses.
3 Habermas (1984 [1981]:86), speaks of ‘interpretation, which refers to negotiating
definitions of the situation which admit of consensus’. My re-formulation of this
indicates that I would like to substitute ‘a model of contextually structured and situated
communication, which presupposes merely the possibility of consensus together with
some normative prerequisites’ for the Habermasian ‘model of unconstrained consensus
formation in a communication community standing under cooperative constraints’.
The problem with the Habermasian model is that it is partially based on the notion of
ideal communication community. Furthermore, it curiously implies that language
and communication are used merely for ‘consensus formation’ or ‘negotiations’.
However, de facto this is not the case, and it is not the only normative ideal of language
use. Habermas (1987:326) seems to be of a different opinion, for he writes: ‘To the
extent that the lifeworld fulfills the resource function, it has the character of an intuitive,
unshakeably certain, and holistic knowledge, which cannot be made problematic at
122 Explicating a critical realist methodology

will—and in this respect it does not represent “knowledge” in any strict sense of the
word. This amalgam of background assumptions, solidarities, and skills bred through
socialization constitutes a conservative counterweight against the risk of dissent inherent
in processes of reaching understanding that work through validity claims.’
4 About the first, original nineteenth-century round of the pathological learning process,
articulated by Karl Marx and Friedrich Engels, which eventually led to Communist
dictatorship in Russia, see Chapter 1, note 5.
5 Double hermeneutics
of iconic modelling

In social sciences, a theoretically reflective scientist tries to investigate and interpret

a world, which is itself meaningful. This is what Anthony Giddens (1977:12)
calls the double hermeneutics of social sciences. There is, first, the movement
between theories and, second, the movement between the researcher, his or her
explanatory models and the lay meanings. But how exactly does this double
hermeneutics work?
The basic idea is simple: understanding, knowledge and social being exist
as a part of a relational, geo-historically formed context (even when its identity
may endure across a number of particular contexts). Hermeneutics is about
uncovering the relevant parts of the relevant context. Hermeneutics encourages
reflexivity. Since there cannot be any development of knowledge without fore-
knowledge, social sciences must explicate the contextual presuppositions of
their own understandings and devise new models, theories and stories on a
reflective basis. Social sciences must self-consciously be involved in hermeneutic
tasks of moving between theories. Hermeneutics is also about explicating the
lay meanings in relation to the appropriate context(s). As Richard Bernstein
(1983:141–2) puts it, the crucial task of empirical research carried out in the
double hermeneutic way is to find resources in our language and experience
to enable us to understand initially alien phenomena without imposing blind or
distortive prejudices on them.
The process of building iconic models of causal complexes is hermeneutic in
both senses. An iconic model—a picture of the relational components of a causal
complex—includes existential hypotheses, stipulations of internal relations and
action-possibilities, descriptive statements and causal hypotheses. The task is to
identify these correctly and, consequently, tell well-endorsed explanatory stories
about multiple determinations of concrete outcomes.
In this chapter, I shall elucidate the methodology of iconic modelling. First,
I discuss two different approaches to explanatory modelling, and make a case
for the reductive approach. Second, I analyse different types of iconic models
and argue that teleiomorphs, based on idealisations and abstractions, should be
central to social sciences. Third, I assess briefly the relative roles of quantitative
and qualitative evidence, making a case for the primacy of qualitative evidence.

123
124 Explicating a critical realist methodology

In the last section of this chapter, I shall discuss the role of time and space in
explanations. Following Andrew Sayer (2000:106–154), I contend that explication
of spatiality is usually relevant, but not always the key issue. However, temporality
is fundamental to explanations and actions alike. Drawing on Paul Ricoeur’s
theory of narratives, I conclude this chapter by clarifying the double hermeneutic
logic of temporality. This argument anticipates the logic of explanatory
emancipation.

Ampliative and reductive approaches to explanatory

modelling

Nicholas Rescher (1987:33–4) distinguishes between two profoundly different

approaches to looking for adequate explanations, the ampliative and the reductive.
The ampliative strategy has been common in modern Western science and
metaphysics. It searches for highly secure propositions that are acceptable as
‘true beyond reasonable doubt’. Given such a carefully circumscribed and tightly
controlled starter set of propositions, one proceeds to move outwards by making
inferences from this secure starter set. Figure 5.1 illustrates the resulting picture.
In the ampliative approach, the basic move is thus outwards from the secure
home base of unproblematic core beliefs. In stark contrast, the reductive strategy
proceeds in exactly the opposite direction. It begins not in a quest for accepted
truths but for well-qualified and interesting candidates or prospects for truths.
At the outset, it is enough to have interesting possibilities and somewhat plausible
candidates for endorsement. These candidates can also be mutually contradictory.
What the researcher has to do next is to impose a de-limiting and consistency-
restoring screening that separates, metaphorically speaking, the sheep from the
goats until he or she is left with something that merits endorsement. In the

Figure 5.1 The ampliative approach

 Double hermeneutics of iconic modelling 125

reductive approach, research proceeds by way of diminution and elimination as

Figure 5.2. illustrates. The reductive approach proceeds by narrowing down the
initial wide range of plausible prospects for endorsement.
While the paradigmatic instrument of ampliative reasoning is deductive
derivation, the paradigmatic method of contraction is dialectical argumentation.
To effect the necessary reduction, researchers do not proceed via a single
deductive chain. Instead, they move along complex cycles that criss-cross the
same ground from different angles in trying to identify weak points (of view)
and to look for the best candidates among competing alternatives. They are
seeking for a resolution for which, on balance, the strongest overall case can be
made. It is not ‘the uniquely correct answer’ but the strongest and most defensible
position that should be sought in dialectics. Moreover, the best available
explanatory model, comprising the most defensible components for endorsement,
always rests on relatively insecure ground. It is always possible to find a still
stronger or more defensible position, that is, a model and story that is truer.
For realist social sciences, which are interested in complex, stratified and
historical realities of multiple determinations and change, the ampliative approach
seems dogmatic. The ampliative approach aims at simplicity and security at the
expense of depth, complexity and alteration. In fact, it tends to preclude these
a priori, thus causing blindness to contrary evidence. It denies the need for
hermeneutics at any level. Paradigmatically, the ampliative approach aims not at
understanding initially alien phenomena but denying the unfamiliarity of
anything. The main purpose of the ampliative strategy is to reproduce the
secure starter set. Hence, its practitioners tend to impose blind and distortive
prejudices on the phenomena to be explained. The belief of David Hume, that
people are everywhere the same and that there can be no change, exemplifies
the consequences of the ampliative approach (although Hume’s belief was also

Figure 5.2 The reductive approach

126 Explicating a critical realist methodology

ontologically determined). For realist social sciences, the paradigmatic approach

must be reductive. The main aim is to maximise openness and learning capacity.

Types of iconic models

Both technical symbols and models depend upon a convention or a set of

conventions by which they become vehicles for thought about a subject matter.
The difference between them is that, while the conventions for a model are
projective, the conventions for technical symbols are arbitrary. Models are explicit
descriptive and explanatory models of something real, something that exists outside
the conventions of the model. Technical symbols can stand for anything, but
models cannot. A model can become a technical symbol when its source of
projection is lost or forgotten as, for instance, the bull’s head became alpha, but a
symbol such as ‘a’ cannot become a model. Projective conventions involve the
analogic or metaphoric use of the source of the model, in the act of construction
of the model. An iconic model is a descriptive picture of a possible real world.
Accordingly, a theory can be thought of as a statement—picture complex, which
refers to the world and makes truth-claims. (Harré 1970:36–40)
Models are built through complex hermeneutic and practical mediations.
Typically, a model tries to propose similarity or likeness to something that we are
already familiar with, or to something that has already been established in sciences.
Although we were interested in one and the same subject matter—say, the
determination of the foreign policy of late twentieth-century nation-states—we could
still use models drawn from different sources: laws of physics, evolution, biological
organisms, cybernetic machines, artificial intelligence, reading of texts, other familiar
bodily or social activities, episodes of Greek history, divine providence and so on
(cf. Harré 1970:33–5). In other words, the source of a model is usually a field that
is assumed to be better known, or to be more familiar, in the given context, and
hence of help in constructing in imagination a possible mechanism or structured
complex that can explain the phenomenon in question.
In a critical realist (CR) understanding, an iconic model stands in for a real
causal complex in the social world, of which social scientists are always at least
partially ignorant. The knowledge and competencies of actors are necessary for
the (re)production of social relations, practices and systems, yet their knowledge
may be partial, biased, short-sighted, reifying and mystifying, that is, partially
false. By the process of iconic modelling, both researchers and actors can learn
something. That is its function and importance. The iconic model is a picture
of a possible causal complex and mechanisms for producing the phenomena we
are interested in explaining.
There are different kinds of iconic models. Models can be classified into
different categories depending on their source and type. Figure 5.3 presents a
taxonomy of iconic models. Paramorphic models have a different source and
subject. A paramorph is constructed or imagined as operating according to
certain principles, often drawn from science and technology, but well-known
 Double hermeneutics of iconic modelling 127

Figure 5.3 A taxonomy of iconic models

genres of drama (tragedy, comedy, fairy tale) can also provide explicit sources
(possibly or usually in addition to science and technology). Paramorphs are
usually constructed to model processes. Paramorphs have two aspects: relations
to source and to subject. The relation of a paramorphic model to its subject
may be of three different kinds. In partially analogous models, the initial and
final states of a model and subject may be exacdy alike, while the processes by
which they were reached may differ. A complete paramorphic analogy models
all relevant details of its subject. In a partial paramorphic homologue the model
and subject would have an identical manner of working but only analogous
inputs and outputs. A singly connected paramorphic model draws on the
principles of one source, but a model can also be multiply connected to many
different sources (Harré 1970:44–5).
Examples of paramorphic models in social sciences include social physics,
sociological functionalism, evolutionary models and cybernetics. All of them
have been used in international relations (IR). Arms races have been modelled
in the Richardson—Rapoport tradition in accordance with the schemes of
Newtonian physics.1 World-System Analysis, functionalist theories of co-operation
and Waltz’s neo-realism have used sociological functionalism and evolutionary
models, albeit in very different ways.2 Traces of cybernetics can be found already
in David Mitrany, but it was Karl Deutsch who explicitly brought cybernetic
models to the field. It is noteworthy that, of all these schools of thought and
thinkers, only Deutsch has been really clear about the fact that he is constructing
models whose subject matter and sources differ. When the subject matter and
sources differ, models can be constructed only by way of analogy.
The field of the subject of the model can also be roughly the same as the
source on which the model is based. These models are called homoeomorphs.
Homoeomorphs can be divided into three subcategories. Only teleiomorphs
are interesting from the point of view of social sciences. The term ‘teleiomorph’
is intended to suggest that teleiomorphs are, in some respect or respects,
improvements over their subjects. Two kinds of teleiomorphs can be
distinguished, depending on just how they are related to their source-subject
(which may be a particular concrete entity, representing the type of all members
of some class of entities, or it may be some kind of an average). There are
128 Explicating a critical realist methodology

idealisations and there are abstractions. A pure idealisation has all the relevant
characteristics of its source subject, but at least some of them are, according to
some scale of value, more ‘perfect’ than the source-subjects’ properties.
Abstraction, in turn, can be analysed as follows. If the source-subject has the
relevant properties pl…pn, then an abstraction has the properties pj…pk (1<j<k<n);
that is, fewer than its source subject (Harré 1970:41–2).3 Max Weber’s ideal
types are a case in point. For Weber, ideal types offer an ideal picture of events,
types of action and social relationships by an ‘analytical accentuation’ of certain
elements of reality. In more Marxist terms, one can also talk about abstraction
as isolation in thought of a partial aspect of an object. It is important to keep in
mind what we abstract from, and whether the abstraction preserves essential
properties of the object (Sayer 1992:86–7).
However, teleiomorphs also involve analogies, metaphors and metonyms. In
the case of teleiomorphs, the typical analogies are geo-historical rather than
analogies to the (subject matter of) natural and technical sciences. Many of the
metaphors used may well go unnoticed. There is no simple way of ‘accentuating
reality’. Consider a few familiar examples of apparently literal ways of talking
about world economy and politics:

• ‘Germany is in the centre and Tanzania in the periphery of the world economy.’
• ‘There are three great powers in the world.’
• ‘The power of the USA was in decline in the 1960s and 1970s.’
• ‘Nixon bombed Hanoi.’
• ‘Finland felt that she had to be more active with respect to her policy in the
United Nations.’
• ‘Unilever has raised its prices of washing liquids again.’

By using George Lakoff’s and Mark Johnson’s (1980) elucidation, these

statements can be analysed and decomposed in terms of their metaphorical
structure. Centre and periphery are metaphors that stem from concrete
experience, from the fact that, when we look at some territory (land, floor,
space, etc.), our field of vision defines a boundary of the territory, namely the
part that we can see. The part we can see well is the centre; other parts of the
field belong to the periphery. Given that a bounded physical space is a ‘container’
within which one can be and that our field of vision correlates with that bounded
physical space, the metaphorical concept ‘visual fields are containers’ emerges
easily. Only by assuming this metaphor, it is possible to use expressions such as
‘Tanzania is in the periphery, Germany in the centre of the world economy’.
The second statement is based on the metaphor ‘significant is big’. Great
powers are ‘great’ because they are assumed to be significant. The power of the
USA can be said to have been ‘in decline’ because ‘having control of force is
up; being subject to control or force is down’. This metaphor clearly has a
physical, orientational basis: physical size typically correlates with physical
strength, and the victor in combat of two bodily persons is typically on top.
Consider then the expression ‘Nixon bombed Hanoi”. Although Nixon did not
 Double hermeneutics of iconic modelling 129

personally drop any bombs to Hanoi, this metonymic expression—a metonymy

uses one entity to refer to another that is related to it—substitutes controller for
controlled (as well as for other controllers). Also, Hanoi can stand in for the
whole country. Further on, if Finland is assumed to ‘feel’ something such as
being compelled to be more active in the United Nations, one is talking about
states as if they were persons. At the same time, and more implicitly, one is using
a metonymy that can substitute an institution such as a state—the whole of
which is seen as a ‘person’—for the concrete actors responsible for a particular
decision (to make Finland more active, or, in the Unilever example, the
corporation as a whole is thought to raise prices).
I hope these examples show how the language of world politics is thoroughly
metaphorical. There is no simple way of ‘accentuating reality’. This is an important
reason for talking about iconic models. The term ‘icon’ refers to an image as a
symbolic and metaphoric window between two worlds: the world of theories and
the real world we are trying to describe and explain, but which must remain
always partially alien or uncovered. By building a picture—statements complex
along the lines of the reductive approach, the aim is to create a well-endorsed
icon, a relatively credible window from the transitive sphere of social scientific
models to the intransitive world of a geo-historical causal complex (only the
future is open and changeable). Whether the model has been successful in this or
not has to be checked by means of ontological and theoretical plausibility control,
and, critically, by means of assessing available empirical evidence.

Modelling causal complexes

Research starts with a problem, or a question to be answered. There is an X

that is for some reason interesting, and seems to be in need of an explanation.
How was (is) X possible? Why did (has, could) it become actual? This X can
be an outcome of a particular episode, such as the Melos massacre, or a general
tendency, such as the tendency to adopt neo-liberal economic policies in the
late twentieth and early twenty-first century. A prevalent phenomenon can also
become absent (slavery) or emerge (politically active global civil society).
Moreover, a context-specific statistical correlation, or Tony Lawson’s (1997:204–
213) terminology, a contrastive demi-regularity, can also give rise to a research
puzzle. Examples of contrastive demi-regularities could include: ‘An increasing
portion of the world’s population lives in cities’; ‘Average unemployment rates
in Western industrial countries were higher in the 1990s than the 1960s’; ‘The
global economic growth has been steadily slowing down since the 1960s’; ‘Since
the late nineteenth century, wars between liberal-democratic countries have been
much less frequent than inter-state wars between them and others, or inter-state
wars between others’; ‘Since the Second World War, wars within states have
been much more common than wars between states’; ‘Measured in incomes in
constant dollar terms, global disparities have been growing exponentially since
the early 1960s’.
130 Explicating a critical realist methodology

Every research problem presupposes something, and this something belongs

to a wider system. For instance, the democratic peace hypothesis assumes statist
and legalist categories of significant violence. The first task of a research project
is thus to explicate these presuppositions, whether they concern categories of
data coding, implicit metaphorical structures, the validity of particular historical
descriptions, theoretical assumptions or implicit fairy tale or tragic narratives.
For instance, should we also allow for covert military operations or transnationally
constituted military force as an indication of non-peaceful relations between
liberal-democracies? (cf. Barkawi and Laffey 1999:414–15). If the research
problem can, upon close examination, be sustained, perhaps in a qualified form,
the next undertaking is to construct an explanation by building an explanatory
iconic model of the relevant causal complex. This follows the scheme of scientific
explanation in open systems (see Bhaskar 1979:165). The first step is the
resolution of a complex outcome into its components; the second, redescription
of component causes in terms of abstractions, idealisations and a CR social
ontology; the third, an analysis of possible material and efficient causes, some
of them perhaps characteristic of a number of geo-historical structures and
systems; and, last, empirical assessment of possible causal components, and
elimination of the weaker hypotheses.
The CR ontology is a hermeneutic standpoint. We cannot identify two
components (or complexes) with each other if we have not first identified them
as different, separate or specific. Also, we cannot identify something if we have
not identified it both as something specific and with other entities that are
essentially the same. We always need both operations, identification as…and
with…, and both presuppose each other (Israel 1979:103–7). The CR social
ontology provides a particular ‘as…’ and ‘with…’ to identify relevant components
of causal complexes, and to assess whether they also comprise relatively enduring
structures.
Moreover, analogies and metaphors are crucial in CR social sciences.
Identification is made possible by an understanding of differences and similarities.
Analogies and metaphors organise these differences and similarities. An important
aspect of the better understanding of something is to explain it in terms of
similarities between the alien object and familiar or established meanings. Realist
ontological concepts such as ‘powers’, ‘producing’ and ‘mechanisms’ are
themselves metaphorical, and should be assessed as such. For instance, the
metaphor of ‘mechanism’ is not always plausible in assigning elements of social
worlds; thus, the notion of causal complex. There is little that is strictly literal.
The ‘literal’ meanings are either routinely reproduced metaphors or are based
on immanent bodily and social experience, which ground metaphors.
As the notion of double hermeneutics reminds us, an understanding of the
meanings circulated within social practices is a prerequisite of adequate modelling.
Analogies and metaphors are part of the world itself We take into a model
world-constituting metaphors such as ‘visual fields are containers’, ‘significant
is big’, ‘having control of force is up; being subject to control or force is down’
or ‘states are persons’. However, when the world in question is modelled, one
 Double hermeneutics of iconic modelling 131

should not use these metaphors in describing the world uncritically, at least not
without being as explicit as possible about such a usage. Rather, we should
grasp and analyse metaphorically structured constitutive rules and definitions
for what they are, as constitutive, meaningful and intentional (even if contested)
features of the real world we are studying. Even if constitutive, they may
nonetheless be highly misleading as descriptive claims.
An iconic model stands in for real causal complexes and mechanisms. Actors
must have some familiarity with the social complex and related (or consequent)
mechanisms; their competencies and knowledge are the condition for social relations
to exist. However, we can assume that they are partially ignorant of at least some
layers and aspects of the worlds, in which they exist and act, and they may also
have false understandings of some or many of these realities. An iconic model
includes the following kinds of hypotheses (modified from Harré 1970:55–6):

1 Existential hypotheses. Existential hypotheses can be of the following kind: ‘There

are great powers’; ‘There is a centre and a periphery of a world economy’;
‘There are states that are analogical to persons’. However, more critically,
they can also explicate the relevant constitutive rules and metaphors: ‘There
are states that are considered by actors to be significant; these states are called
“great powers” because actors use the metaphor “significant is big”; the
significance of the states is a social construction, that stems from the constitutive
rules {x, y,…} and from the certain kind of resources that are constituted to be
significant’; A sovereign state has been constituted as a possessive individual
in international law and discursive practices’. On the other hand, existential
hypotheses can also concern emergent properties and powers arising in social
systems. All existential hypotheses, whatever kind they may be, must be
subjected to the scrutiny of a plausibility control and empirical evidence.
2 Stipulations of internal relations and action possibilities. We can distinguish between
different ‘How is X possible’? questions. When we are concerned with the
identity of X (an actor, an action, a practice, a relation), we need an account
of relevant constitutive and closely related regulative rules. For instance, how
is it possible to ‘declare a war’ or ‘to own and control property located in
various nation-states’? In a similar manner, we can also analyse the identities
of positioned actors. Any explication of identities and constitutive rules implies
a stipulation of internal relations. This kind of analysis freezes, in a sense, the
social world in question. Having done it, we can move towards the world of
historical, causally effective, dynamic interaction and ask: What are the INUS
components that made that action possible? At this level, the analysis is about
putting forward causal hypotheses concerning outcomes in the open systems
(see 4, below). The point is to grasp the actual constellation of conditions that
made X contingently possible. Other constellations could have made it possible
as well. Finally, we can ask a possibilistic causal question concerning the
existence of the actually existing elements: How did their existence come
about? In this manner, we can write genealogies of relevant components of the
layered causal complexes in question.
132 Explicating a critical realist methodology

3 Descriptions of the model. Descriptions of a model can be uncritically metaphorical

and metonymical: ‘The power of the USA was in decline in the 1960s and
1970s’. They can also be more reflective and contextual: ‘Given the constitutive
rules of the international society, the post-cold war hegemony of the Western
model of society, the development of the US resources, the world-historical
context of the end of the cold war, the dominance of haute finance and the self-
understandings of the Bill Clinton and George [Link] regimes, the US
hegemonic power can be argued to have been in its second height since the
early 1990s, but only in an ambiguous manner.’ Complex statements and
assumptions of this kind are, however, in need of further qualifications and
support, both from theory and empirical evidence; all its components should
be analysed further.
4 Causal hypotheses. These can be stated in such questions as ‘Was the decline of
the USA caused by a slow-down in its technical innovativeness, or by a political
crisis of legitimation?’, or ‘Will the second phase of the hegemony of the
USA, accompanied by its increasing unilateralism and arrogance, lead to the
erosion of the major co-operative institutions of the post-Second World War
world?’ However, they can be also put forward in a more critical way, by
trying to insulate the most relevant INUS components of the causal complexes
in question. We could ask, for instance, a counterfactual question such as:
‘Would the decision makers of the other G7 states have any reasons for
destroying the USA-led institutional arrangements of the world economy,
given the time—space situated CONTEXTI’? Or, to take another example:
‘Can we explain a particular treatment of MNCs in Zambia in terms of a
game-like bargaining situation between Unilever and Zambia’?; ‘What are
the crucial components of this situation, and how did they come about?’
Causal hypotheses are about insulating the pivotal material and efficient causes
of the relevant complex. What parts of the material causes could have been
different? Who could have acted differently, and how? What kind of context
could have made ‘acting differently in a particular way’ possible, or legitimised,
naturalised, encouraged or sanctioned it? Answers require a thorough analysis,
i.e. a dialectically developing process of iconic modelling, and an attempt to
specify deeper layers and, more thoroughly, the pertinent parts of the complex,
which forms a partial and open whole, and is causally responsible for the
outcome. A necessary requirement is methodical gathering and analysis of
empirical evidence, but counterfactual reasoning is also needed.

This scheme of iconic modelling clarifies the nature of many social scientific
practices. Yet it can also critically re-direct them and lead to more systematic
and open-minded practices. It can help to overcome many of the standard
dichotomies—such as the one between ‘explanation’ and ‘understanding’—which
mislead more than enlighten.
Richard Rorty has, however, rightly criticised ‘the metaphysical thinker’ who
‘thinks that if you can get the right understanding of Being—the one that gets Being
right—then you are home’ (Rorty 1991:38). Thus, we also need constant self-
 Double hermeneutics of iconic modelling 133

reflection, encouraged by concepts such as Martin Heidegger’s thematising.4

Thematising includes not only a positive attitude for truth but also a systematic
articulation of the understanding of being, the sketching out of the way of conceiving
which is appropriate to such entities and the de-limitation of an area of subject
matter. The articulation of the understanding of being means in the context of social
sciences that a theory of social ontology should be explicated, the question being:
What are the basic elements of the social world, and how are they interrelated?
The CR social ontology, discussed in Chapter 4, is such an articulation. It conditions
the appropriate ways of understanding and analysing social worlds. To acknowledge
this implies a moral requirement of constant openness to ontological criticism and
revisions, and a constant need to re-thematise the subject matter.

Empirical evidence

Assuming that after a critical scrutiny of its presuppositions, the research problem
has been sustained, even if only in a qualified form; and that at least a few
interesting candidates for endorsement as components of possible iconic models
have been specified; the next job is to assess and refine these candidates. In an
open-ended, realist research, empirical evidence must play a critical role in the
assessment and refinement of candidates. Although quantitative evidence is
indispensable, the reasons of actors and the meanings of actions and practices
can only be studied by qualitative means.

Quantitative data

Quantitative data-based techniques presuppose that the object can be analysed

in terms of magnitudes. Magnitudes can be expressed as integers or ordinal
numbers, or ratios. Some things can obviously be counted or measured. Some
relatively unproblematic cases would be the number of people, or the amount
of gold in a central bank or the number of nuclear warheads possessed by a
country. Economists are fond of thinking that all values can be measured in
terms of money (or ordinal preferences), since this is what happens in capitalist
markets constantly. Magnitude assignments are, in general, monotonic,
presupposing a linear continuum. In other words, there should be no contextual,
qualitative variations of values. This may look like a serious limitation on the
use of quantitative data in social sciences.
Advocates of quantitative methods usually appeal to the qualities of statistics
and mathematics as a precise and unambiguous language that extends our
powers of deductive and analytical reasoning far beyond that of purely verbal
methods (Sayer 1992:175). However, ‘quantitative’ does not imply ‘formal’ (and
vice versa). Statements linking ‘more’ of something to ‘less’ of something else
are quantitative, even when they are expressed verbally in essentially journalistic
or traditionalistic texts. Journalistic or traditionalistic studies are not ‘formal’; neither
134 Explicating a critical realist methodology

are they usually particularly hypothetical, theoretically reflective or systematic.

Yet, this does not mean that they are necessarily qualitative. To the extent that
they are quantitative by their very logic, the formal methods of empirical studies
can make them more methodical and precise. Positivists have always been right
about this. Analytical powers, clarity and preciseness are important virtues of
scholarly work.
Moreover, quantification is part of the object of study. For a long time, many
social objects have been systematically quantified, measured, for example, in terms
of money, or time or quantified inputs and outputs. Quantified flows and processes
play an important role in the governance of modern organisations. (Sayer 1992:178)
Capitalism and modern nation-states developed also by cultivating the practices
of book-keeping and data collection. In the seventeenth and eighteenth centuries,
statistics was explicitly a science of (and for) state (Foucault 1991c:96). In the
nineteenth and twentieth centuries, the role of quantified information has grown
ever stronger. However, with internationalised state structures and new forms of
governance, quantitative data has also become more global. In the twentieth century,
the World Bank, the IMF and various parts of the UN system, for instance,
collect, organise and also analyse vast amounts of quantitative data. The hope of
positivists has been that one day this quantitative data would reveal, under careful
scrutiny, invariant regularities in society, also possibly leading to an optimal theory
of moral and political interventions into societal flows and patterns.5 This hope
has been, for better or worse, in vain.
The fundamental question is: What must the object be like for it to be
quantifiable? Only qualitatively invariant objects and processes can be quantified
using interval scales. This is also the presupposition of assumptions such as
linearity, additivity, etc. Without these assumptions, the variables would not be
able to make stable reference and all the analytical tools would be, to a large
extent, useless (although in statistics there are some remedies e.g. to the problem
of nonadditivity; see Alker 1966). Social relations are, however, dependent on
concepts and meanings. As David Sylvan (1991a:268–70), among many others,
has pointed out, meanings are contextual. Social structures and relations are
thus not easy to quantify, at least not unambiguously. Moreover, as a language,
mathematics is acausal and astructural. It lacks the categories of ‘producing’,
‘generating’ or ‘forcing’ that we take to indicate causality. A function y=f(x) says
only that quantitative variation iny is formally related in some ways to
quantitative variation in x. The sign ‘=’ indicates only that the quantities are
equal. There is nothing about substantial causal relations here (Sayer 1992:179).
Two major problems of quantitative data are, thus, the bracketing (or
suppression) of the contextuality of social being and action, and the lack of
realist concepts of power and causality. The suppression of contextuality makes
standard sets of quantitative data problematic. As Hayward Alker (1996:339)
has argued, ‘data-coding procedures should be considered key dependent variables’.
The categories and concepts used in data coding may form deep layers of the
causal complex that explains x. Moreover, the lack of realist powers and causality
gives rise to a set of characteristic problems. Unlike temporal causal processes,
 Double hermeneutics of iconic modelling 135

processes in mathematical or statistic models can be reversed. For instance, given

information on such relatively unproblematic ‘variables’ as birth rates per 1,000
women in each age group, we can attempt to forecast the number of births in
future periods. For the purpose of ex post calculations, once the model has been
fitted, any variable can be treated as dependent, simply by re-arranging the
equation. A related problem is that these mathematical or statistical models aim
at prediction with no understanding of the causal relations involved (Sayer
1992:183–4). Once the components of a system have been quantified (more
often unreliably than not), it is always possible to fit a mathematical model to
the data ex post. However, given the nature of data and the condition of open
systems, the model is highly unlikely to be able to predict the future. In open
systems, outcomes do not remain stable even when the characteristic powers
and structures do.
There is also the problem of spurious correlation. This is the well-known post hoc
(ergo propter hoc) fallacy: simply because there is positive correlation between A and
B, it is thought that A causes B. Indeed, sometimes a correlation can be a reasonable
basis for expectation of a causal relationship. However, even when statistics or
mathematics are useful for recording some of the effects associated with particular
structures or causal powers, more qualitative languages are needed to show why
and how objects possess these powers (see Sayer 1992:179–80, 193–9).
Some statistical methods use, occasionally, causal language. For instance,
many econometricians think they are trying to identify Humean causal relations.
Econometrics not only requires that data is unambiguous and monotonic; it
also shares the fundamental assumptions of other forms of statistical analysis.
Tony Lawson (1997:69–85) has discussed in detail the consequent futility of
econometric developments. There is no point in repeating that discussion here.
Suffice to say that the underlying flat ontology of mere external relations of
association between atomist units seems to be the main obstacle to any significant
scientific development. Atoms possess no intrinsic conditions. An atom is
distinguishable only with respect to its position (or some higher order derivative
of position such as motion, acceleration, etc.) in space at a moment in time’
(Bhaskar 1978 [1975]:83). Methodologically, the plausibility of the assumption
of monotonity stems from this (largely implicit or secreted) atomist ontology.
Politically, it is in this sense that we should understand Alker’s (1996:186)
complaint that ‘mathematical social science…distorts social relations to the extent
that it participates in and mirrors our alienation under “capitalism”’. In contrast,
adequate social scientific data should not distort but, rather, help to identify
correctly the essential relations and structures of the objects we are studying.

Qualitative evidence

Quantitative data may be indispensable and can be useful in certain limited

ways, particularly in deriving research problems and estimating how common
particular structures and tendencies are. Ontologically, however, qualitative
136 Explicating a critical realist methodology

evidence is primary. Internal social relations exist as shared (even if often vague
and ambiguous) concepts. Internal relations are the necessary material causes
of social action. They are implicated in the causally efficacious reasons for
action, and involved in the outcomes of any generative social complex. Moreover,
regulative rules, as well as the external conditions and circumstances of action,
have to be given meaning by the actors, otherwise they could not enter into
their reasoning processes. Hence, to model a layered causal complex adequately,
one has to study the meanings of relevant actions and practices.
This is the second level problem of hermeneutics in social sciences, concerning
the movement between the researcher, his or her explanatory models and the lay
meanings. The mode of access to lay meanings is not external observation but
communication (cf. Habermas 1984 [1981]:102–13). If we study past worlds and
the relevant people are already dead, communication cannot be mutual. Rather, it
is about interpreting texts or text-like traces of the past. These texts might have
been constitutive of the meanings under study (first-hand evidence) or they may
make references to them (second-hand sources). If we study contemporary worlds,
and the relevant people are still mostly alive, mutual communication is possible.
Participatory ‘observation’, interviews, discussions and other forms of mutual
communication can complement the study of texts and text-like traces.
Despite the detachment of the researcher and his or her object of study in
terms of time, space and causal responsibility, the subject—object relationship is
complicated even when the object of study is historical. There may be (asymmetric)
internal relations between the being of the researcher and the constitutive meanings
and concepts of past objects. The researcher also has fore-knowledge and
anticipations about these meanings and related validity claims. How is it possible,
then, to achieve better understandings in a conscious and controlled manner? In
Hans Georg Gadamer’s view, it is actually impossible to find a method that would
guarantee objectivity in the traditional scientist sense. There is no interpretation
totally free of presuppositions and prejudices. While the interpreter may free
himself or herself from this or that situation, he or she cannot free himself or
herself from his or her own facticity, from the oncological condition of always
already having a finite temporal situation as the horizon, as the standpoint from
which the presumed possibilities and anticipations stem.6 It is this horizon that
should be put systematically at risk when it is confronted by new qualitative
evidence or experience (practical or communicative).
There are thus two sides to collecting and analysing qualitative evidence.
The onto-ethical relation of the researcher towards the unfamiliar or alien
meanings, actors and practices, and the methodological problem of classifying
types of evidence and analysing them systematically for the purpose of building
a well-endorsed iconic model. The onto-ethical relationship must be based on
virtues of openness and learning. ‘Historical consciousness is aware of its own
otherness’ (Gadamer 1975:306). The process of understanding meanings must
follow the logic of hermeneutic circle and occur through constant fusion of horizons
(see Gadamer 1975 [1965]:306–7 passim). The available evidence should be
classified in terms of what relationship they bear to the actions and contexts
 Double hermeneutics of iconic modelling 137

under study (primary, secondary, etc.). If there are as yet no interesting or

initially plausible candidates to assess, the analysis of qualitative evidence must
aim at identifying components of causal complexes; if some plausible candidates
have been specified, analysis must aim at assessing their empirical merits.
Detective work and the process of a trial provide particularly striking
metaphors for social sciences. Sometimes we must seek for any clues, even
remote and indirect ones, to establish elements of causal complex and to construct
a proper story to tell. There may be only a few traces available, perhaps because
of spatio-temporal distance, perhaps because of practices of secrecy Sometimes
both evidence and interesting candidates for endorsement are abundant, and
then a trial seems a more appropriate metaphor. The prosecutor and the counsel
for the defence have to be given sufficient time to investigate and bring in all
available evidence to make their case. Finally, a time of judgement will come: Is
candidate X plausible enough to be endorsed further? After the screening-out of
candidates, a consistent—as consistent as the subject matter allows—picture of
the causal complex as a whole has to be formed, and an explanatory story told.

Space and time in iconic models

Human beings are temporal and spatial beings. We cannot go back in time;
whatever we do is an intervention in irreversible causal processes; physically,
we cannot be in two places simultaneously; and the specific area of space we fill
in—‘this chair’—cannot be occupied by anybody else simultaneously. Bodily action
always involves a stream of actual or contemplated causal (and thus material)
interventions in the ongoing geo-historical processes, thus producing various
outcomes. Causal powers can also be activated or mediated across both spatial
and temporal distance, ‘provided that there are intervening media through which
they can exert their influence’ (Sayer 2000:112).
To what extent is it necessary to spell out the spatial and temporal location (or
distance) and organisation of the components of causal complexes, including actors
and actions? Interest in studying social space seems to coincide with the crisis of the
short-lived twentieth-century claim that territorial nation-states de-limit the space of
most social relations and processes. What is crucial is whether spatial constellations
and organisations make any difference to how concrete practices, powers and causally
efficacious complexes work. It cannot be decided a priori that spatial location, distance
and organisation are particularly important aspects of causal complexes. It has to be
left to the process of iconic modelling to determine how and why space matters.
Temporality seems essential to any social interaction and process. All
components of causal complexes are geo-historical, except the natural powers
and mechanisms that make society possible (and they are subject to long-term
processes of change too). Moreover, actors cannot change the past or intervene
in future processes they do not anticipate. In social interaction, the timing of
action is often crucial and, in some social contexts, it is also meticulously
organised. It is at least as important that in making something actual (present),
138 Explicating a critical realist methodology

actors unite reasoning about the past and possible futures in terms of narratives.
A social scientist, in his or her turn, tells stories and may thereby re-signify
practices. Also, the timing and media of his or her intervention makes a difference.

Space

In the early twentieth century, the science of geo-politics emerged with the
exhaustion of the geographical space for Western colonialists and explorers. It aimed
at viewing the planet earth as a whole and stipulating positivist ‘laws’ about the
statist or imperial governance of the limited global space (Tuathail 1996). However,
geo-politics has been an exception rather than the rule. In the good part of the
mainstream of twentieth-century economics, politics and sociology, it has been
assumed that economy and society mean or at least correspond to the nation-state.
Often, space has been abstracted away with little or no concern about the
consequences (Sayer 2000:108–9).
Henri Lefebvre, David Harvey and others have complained that the tendency
to abstract away from space—or to make simplistic assumptions about it—has kept
social scientists ignorant about the deep conditions of the social objects they are
studying. This ignorance has also caused blindness about the possibility of (the
ongoing) profound changes. For Lefebvre (1991; 1974:11), space has an active
operational and instrumental role in the organisation of social systems, including
production and exchange. Hegemonic forces or classes also make systematic use
of space in their attempt to gain or secure preferred practices and relations.
Following Lefebvre, Harvey (1990:201–39) argues that deep cultural conceptions
of space and time are necessarily created through material practices and processes
of production, which must be organised spatially. Command of time and space
are also important resources of power. In turn, practical organisations and cultural
conceptions of time and space create senses of belonging and identity, and also
render particular cultural and political projects worthwhile.
The term ‘geo’ comes from a Greek word for earth. Social processes are geo-
historical in the sense that, until very recently, human powers and social actions
have always occurred somewhere on the surface of the planet earth. However,
actions and their contexts can be independent of particular spatial settings.
Embodied competencies, available facilities, rules, actor identities, practices and
modes of action can remain unchanged across various locations of actors and
their facilities (Harvey 1990:118). Facilities such as long-distance mobile phone
calls, the Internet and trans-continental flights have further contributed to the
(relative) detachment of action from location or place, at least for those endowed
with sufficient resources (cf. Bhaskar 1994:70).
Space has no causal powers independent of the objects constituting it. It is
one of the problems of theories of ‘centre’ and ‘periphery’ in the capitalist
world economy that they offer a generalised spatial location as the crucial
component of explanation of under-development, failing to specify the relevant
causal powers (or their absence) and related geo-historical processes adequately
 Double hermeneutics of iconic modelling 139

(Sayer 2000:112–13). Conceptions and metaphors of space are thus not a

substitute for explanation in terms of a causal complex K={AR, RU, RE, PRA,
AN}. Rather, the problem is how to grasp the spatiality of the relational elements
of complexes and the consequent social systems.
Place and distance can make a critical difference to the properties and powers
of actors, and the spatial organisation of objects can also be relevant in many
other ways. Although the spatial setting is not necessarily the key issue, the
spatial forms of relevant objects should not be ignored, and it often has to be
described in detail. For instance, the ever-increasing space—time distance between
causes and effects gives new reasons for systematic studies on the complex
spatio-temporal chains of causal dependencies. In the late 1990s, the conditions
of peasants in East Timor were linked to the consequences of financial decisions
made in London, New York or Washington. Often, it is equally important to
specify the relations between power and space/mobility, occurring in the context
of areal differentiation in the cities, between regions or globally.
To take another example, new military developments may also necessitate the
study of, say, outer space systems. In the US Space Command Vision 2020, the
medium of space is envisaged as the key to ‘full spectrum military dominance’ to
secure US interests and investments globally. By combining ‘global surveillance
with the potential for a space-based global precision strike capability’, challenges
stemming from the globalising ‘world economy with a widening gap between “haves”
and “have-nots”’ can be countered militarily, if needed (US Space Command 2001).
The point is that the spatial forms of relevant objects cannot be theorised in
abstract. They have to be studied empirically by using all available evidence.

Time

All social phenomena, all social interaction, anything that exists in society, is
temporal. Time is not a simple succession of abstract ‘nows’. Time is involved
in social phenomena in a very complex manner. An adequate iconic model
should be able to capture the relevant aspects of temporality. The historical
nature of social being includes the geo-historical processes of formation of all
components of causal complexes. Hence the need to write genealogies of these
components and the fields and problematics they give rise to. Moreover, social
interaction occurs in time. Adequate iconic models have to account for the
irreversible temporal sequences of episodes and processes.
Temporality is also an essential part of social action in a deeper sense.
According to Paul Ricoeur (1984:68–84), there is a unity of having-been, coming-
towards and making-present, since these are thought and acted upon together by
the actors. This is the temporality of practical experience and action. The making-
present of practical action stems from the anticipation of possibilities of
transformative action producing outcomes on the basis of understanding of that
which has-been (history). The horizon of action is thus inherently temporal.
Moreover, particularly the appeal of more total narratives may be reinforced by
140 Explicating a critical realist methodology

awareness of one’s own mortality. Many political projects derive their motivation
from the sense they render to our/their lives. This opens up the way to the
questions posed by Alker:

What motivating power or charisma does this story, or ‘myth’, contain? What
basic structure or structures, what infectious, self-replicating, ‘viral’ qualities
account for this power? Can we replicably discover and empirically test some
of the ways in which its deepest structure has been rewritten into different
‘surface’ texts?
(Alker 1996:105)

Ricoeur (1984:52–87) makes a distinction between three moments of mimesis.

Mimesis1 represents the temporal signification enacted by the actors themselves.
In purposeful actions, actors organise factual and possible events imaginatively
in terms of plots. Narrated actions have symbolic aspects that stem the capacity
for poetic transposition: actions are evaluated in terms of characters and consequences
and on the basis of inherent norms and standards. At this level, time is prefigurated.
Mimesis2 is the moment effected by the narrative of the social scientist in his or
her attempt to explain social phenomena by constructing an iconic model. The
time of social scientific modelling is constructed to the nth level upon the already
constructed temporality. This time never ceases to refer back to the temporality
of practices described by mimesis1. A researcher draws a meaningful story of
his or her own from a diversity of temporal events of reality.
The transition from mimesis1 to mimesis2, which can be called the activity of
configurating, is an organic part of the reductive approach of iconic modelling. Like
any other hypothesis, a hypothesis concerning emplotment should be made
vulnerable to refutation and qualification, and open to the probative force of empirical
evidence. Emplotment combines two temporal dimensions. By stipulating causal
hypotheses one captures the episodic dimension of temporality and creates
components of explanation that go beyond mere chronicle. By grasping together
the whole of the episode one constructs a narrative or a story proper, a story that
has a (counterfactual; unreal) sense of ending. This sense of ending makes it possible
to read the story from the conclusion to the beginning, i.e. to ‘reverse’ the irreversible
real time. However, if history is always open, how can there be ‘a sense of ending’?

Interpretation here is compatible with explanation, to the extent that it is

confined to providing inquiry with the guidance of some seminal idea and to
indicating degrees of importance. It is no longer, as a consequence, the only
worthwhile interpretation, to the exclusion of all the others. But the most
interesting interpretation is the one that assigns itself the task of evaluating a
sequence of events or a set of institutions in terms of their terminal consequences,
themselves evaluated in terms of their value or lack thereof. The overall
meaning of a process is these very terminal consequences.
(Ricoeur 1984:119)
 Double hermeneutics of iconic modelling 141

A research project may thus also begin with a pre-understanding of some terminal
consequences—actual or potential—which are assumed to be important. The choice
of terminal consequences must itself be a responsible choice. In principle it is
possible that the same material of temporal actors, rules and resources, practices,
interactions, circumstances and intended and unintended outcomes can be
ordered according to different actual or potential terminal consequences. All
these temporal interpretations can be true with regard to the causal powers and
sequences upon which they are elaborated. There may thus be many stories to
tell on the basis of the same material.
Ricoeur (1984:70–82) argues that narrative has its full meaning only when
it is restored back to the time of action and human practices. Mimesis3 marks
the intersection between the world of social scientific modelling and the world
of the hearer and the reader. Although there may already be redundancy of
meaning in the social world itself, there is always potentiality for new and
better stories such as is assumed to be the case in psychoanalysis (by both
parties) or trial (by the judge). The main point of mimesis3 is that it re-signifies
human practices in their temporal dimension, to the extent that explaining and
narrating is to remake action. Ricoeur calls this process iconic augmentation of practices.
Iconic augmentation depends upon the symbolic structuring of the prefigurated
time. The practical question is: How plausible is the new and purportedly more
true and better explanatory story when compared to the prefigurated one(s)?
More structurally, it is also a question of whether the new stories have access to
the world of prefigurated practices.

Conclusion

It is not easy to build adequate social scientific explanations. The real difficulty
does not lie in any technical or formal details. It is all too easy to stick to the
ampliative approach and imagine that one can substitute complex thinking and
research with a few technical operations of deductive or theoretical reasoning,
or statistical analysis. In the reductive approach to iconic modelling, it is rather
the multiplicity of possibilities, layers and components that have to be taken
into account, which forms the real difficulty.
This chapter has given some methodical guidelines about how to construct
an explanatory iconic model. First there is a question and an X (some ‘terminal
consequences’) to be explained. The presuppositions of this question, and the
description of the X or ‘terminal consequences’, have to be scrutinised critically.
The next task is to specify available interpretative hypotheses and stories, and
possibly envisage new ones; and determine sources of available evidence, whether
quantitative or qualitative. Are there many interesting candidates for
endorsement? Is evidence abundantly available? If not, one has to use all possible
clues to construct a plausible iconic model and narrative. If yes, different
hypotheses have to be assessed systematically against available evidence, and,
eventually, a judgement about the truth of these hypotheses has to be made.
142 Explicating a critical realist methodology

The components of a layered, relational causal complex and all the relevant
material and efficient causes must be specified in sufficient detail. The researcher
has to decide the extent to which spatial location, distance and organisation are
important aspects of the causal complex. Moreover, she will also have to thematise
the temporal structure of the explanation at the levels of mimesis1 and mimesis2.
In every step of the double hermeneutics of iconic modelling, there is, first, a
movement between a wide variety of theoretical concepts, and, second, a
movement between the researcher, her explanatory models and the lay meanings.
In mimesis3, the explanatory story assumes its full meaning when it is restored
back to the time of action and human practices.
However, mimesis3 is not the only practical implication of social scientific
explanations. In Chapter 6, I shall explicate the logic of explanatory emancipation
and its relation to legitimate transformative actions. Social sciences also concern
normative issues.

Notes

1 In his highly acclaimed book Fights, Games and Debates, Rapoport (1960) argues that,
although social physics is a good starting point, the theory of games implies a partial
critique of that model, while the notion of debate implies a more thorough critique.
Thus, social physics can yield at best only a very partial insight into social processes.
2 For an explication of the metaphoric and analogical basis of these theories, and for an
analysis of the (lack of) explanatory power of these theories in the context of explaining
the end of the cold war, see Patomäki 1992a.
3 To speak of the characteristics of an entity in this manner implies that there are relevancy
criteria, which, however, are always partially relative to the task.
4 The notion of thematising comes from Heidegger, but I have modified and re-
interpreted it. Heidegger attempts to do justice to the possibility of positivist or scientific
realist natural sciences with the help of this notion; I am using it in developing
methodological tools for the social sciences based on critical realism. See Heidegger
(1990 [1962]):414–15, and Dreyfus 1991:82–3.
5 Peace research has also been elevated by this vision. See Chapter 2 for lessons of
Alker’s development; and Patomäki 200la for a reconstructive analysis of the
development of Galtungian peace research.
6 This idea is summarised by David [Link] in the ‘Editor’s introduction’ for Gadamer
1977:xlvii. The definition of ‘horizon’ can be found in Gadamer 1975 [1965]:269.
6 The normative logic
of emancipation

Truth and social scientific explanations have normative and political implications.
In contexts in which the transformation of politics into violence is more than a
mere abstract and remote possibility, the politics of truth is far from being innocent.
Grand narratives about the end of war have justified wars and violence. For instance,
the aim of people’s wars has typically been defined, as Mao Zedong said, by a
historical movement from capitalism to socialism: ‘Once man has eliminated
capitalism, he will attain the era of perpetual peace, and there will be no more need
for war’ (Alker et al. 1989:152). This view is based on a theory of society and a
story of history as a whole. It presupposes judgements about the truth of that
theory and the story. Similarly, the USA, led by Woodrow Wilson, took part in the
intervention against the Soviet Union in 1918. In accordance with the Kantian
liberalist theory of world politics, Wilson saw the anti-democratic Bolsheviks as a
general threat to peace and security, besides rendering more difficult the defeat of
Germany in the ‘war to end all wars’. Hence, in Kantian thought, wars may seem
justified in terms of making the world more peaceful and orderly. Modern
emancipatory projects have also confronted each other on the battlefield.
‘Isn’t the most general of political problems the problem of truth?’ asks
Michel Foucault (1991b:82). Indeed, there is a close connection between politics
and truth—truth understood socio-historically as a system of ordered procedures
for the production, regulation and operation of statements (Foucault 1986
[1971]:74). Truth has effects of power. Under suitable conditions, a particular
truth can produce, via actions, parts of social reality. Truth also has the effect of
including and excluding, legitimating and de-legitimating potential or actual
components of reality, including military actions. Foucault is not alone in seeing
a very close connection between politics or its failures and truth. Any notion of
truth implies criteria, standards and measures, which have social and political
consequences. Strict standards and measures have also lost much of their
credibility, because intellectuals would like to freeze and dismantle the guillotine,
as well as other modern machines of violence stockpiled all over the fragile
planet (see, for example, Schürmann 1984:175–6). Hannah Arendt (1973:82)
articulated this sentiment three decades ago: ‘At least we can learn from Rousseau
and Robespierre that absolute goodness [truth] is hardly any less dangerous
than absolute evil [untruth and deceit].’
143
144 Explicating a critical realist methodology

Critical realism makes a distinction between realist ontology, relativist epistemology

and judgemental rationality, and includes a truth-based theory of emancipatory
explanation. In the following, I shall discuss Roy Bhaskar’s theory of truth and the
related theory of explanatory emancipation. My argument is that because of the
normative nature of truth and epistemological relativism, there is an essential
connection between explanatory or any normative criticism and the non-violence of
political action. This is also an attempt to explicate a possible theoretical basis for a
peace culture of the critical social sciences.1 In other words, my twofold argument
is a suggestion of how a truth-based emancipation is possible without violent
consequences in a pluralist world.
In this chapter, I also argue for a rich normative discourse about alternatives.
Truth and non-violence do not exhaust moral and political problems. Besides truth,
there are many other normative notions, which can be brought into ethico-political
discourse about social relations and practices. Also, transformative action would be
futile without concrete and feasible ‘Utopias’ of alternative possibilities. The normative
logic of emancipation is not only precarious; it is also pluralist and multifaceted by
its very nature.

Epistemological relativism

Epistemological relativism means that all beliefs are socially produced. Knowledge
is transient, and neither truth-values nor criteria of rationality exist outside geo-
historical processes. Epistemological relativism should be distinguished from
the incorrect thesis of judgemental relativism, which asserts that all beliefs (or
statements) are equally valid, in the sense that there can be no rational grounds
for preferring one to another (Bhaskar 1979:73). This distinction is necessary.
Any kind of action presupposes some criteria:

it is clear that if one is to act at all there must be grounds for preferring one
belief (about some domain) to another; and that such activity is typically
codifiable in the form of systems of rules, implicitly or explicitly followed.
(Bhaskar 1979:74)

Anything does not go. To take a simple example (from Putnam 1989:54): if
someone really believed he could fly and were foolish enough to pick a conceptual
system that confirmed this, and then acted upon it by jumping out of a window,
he would, if lucky enough to survive, see the weakness of the conceptual system
at once. Social relations are similarly real, because they have causal effects via
actions. For instance, breaking regulative rules—say by failing to complete a census
form or by hauling your boss over the coals—tends to have definite consequences.
Bhaskar (1986:24) argues that judgemental rationality depends upon the
recognition of ontological realism and epistemological relativism. Explicit denial
of this condition would lead to a theory/practice contradiction (cf. also Habermas
1979:2 and 66–8). Action presupposes reality, which is independent of beliefs of
 The normative logic of emancipation 145

any particular actor, otherwise the whole idea of action would be unintelligible.
People could be jumping out of windows if they just believed that they could fly.
Yet relativism regarding explanations prevails. This is also a condition for rational
research and discussions. If there were absolutely true or self-evident claims or
statements, there would be no room for research or debates about their validity.
Moreover, because a choice between theories or explanatory models is a matter
of geo-historically situated human judgements, the choice also depends upon the
ethical responsibility of the cognitive agents concerned (Bhaskar 1986:24–5).

Duality of truth?

It is not possible to know the world except under certain particular, geo-
historically, transient descriptions, as Bhaskar (1986:99) argues. What, then, is
truth in Bhaskar’s theory? It seems to relegate the notion of correspondence
between epistemic statements and ontic objects to the status of a mere metaphor
for the goal of an adequating practice, in which ‘cognitive matter is fashioned
into a matching representation of a non-cognitive object’. However, Bhaskar
goes on to explain his notion of truth as follows:

truth effectively functions bivocally—to designate, on the one hand, claims,

judgements and values…which are, or could be, made or avowed within
social activity; and on the other, conditions, states of affairs and…composite
referents (truth-conditions or more simply truths) which exist quite
independently of us. (Thus, although truths are not things, laws, natural
kinds, etc., it would be misleading to deny that the latter were, in some
sense, truths.) Truth thus has a genuine ontological and not just an ontogenetic
(e.g. heuristic or regulative) use…; but its ontological use depends upon the
possibility of ontogeny (and hence human practice) in a way which the concept
of a natural law, for instance, does not.
(Bhaskar 1986:100)

Bhaskar maintains that his theory of truth can sustain the duality of the epistemic
and ontic aspects of truth instead of reducing these two aspects to the ‘oneness’
implicit in the (conventional) correspondence theories of truth.2 The self-
proclaimed pragmatist Richard Rorty (1987), for example, would strongly object
to Bhaskar’s views. For the pragmatist, Rorty argues, the notion of ‘truth’ as
something ‘objective’ is just a confusion between the following:

1 Most of the world is as it is whatever we think about it.

2 There is something out there in addition to the pragmatically known true
statements, which is called ‘the truth about the world’.

The pragmatist agrees with (1) but cannot make sense of (2). According to
Rorty, the realist tries to explain (2) with a further idea:
146 Explicating a critical realist methodology

3 The truth about the world consists of a relation of ‘correspondence’ between

certain sentences (many of which have yet to be formulated) and the world itself.

The pragmatist, Rorty (1987:41) says, can only retort that many centuries of
attempts to explain what that correspondence is have failed. What is Bhaskar’s
position here? There is no disagreement between Rorty and Bhaskar as far as
(1) is concerned, but there is certainly disagreement in respect of (2), for this is
one of the basic claims of Bhaskar’s transcendental realism. In effect, (3) is the
‘metaphysically realist notion of truth’—truth as ‘correspondence to the facts’,
or something of that kind. Hilary Putnam (1982:4), among many others, denies
this on the grounds that we do not have notions of the ‘existence’ of things or
of the ‘truth’ of statements that are independent of our conceptual constructions,
procedures and practices. Also, Bhaskar says clearly that it is not possible to
know the world except under particular geo-historically transient descriptions.
Where, then, does the difference between Bhaskar and the critics lie?
Bhaskar is not arguing for a metaphysically realist notion of truth (3). ‘The
supposition of the possibility of a (logocentric) correspondence, if interpreted as
a logical thesis, is meaningless or at the very best unverifiable’ (Bhaskar
1979:201). Rather, he maintains that there is ‘a difference which makes a
difference between (a) “it works because it’s true” and (b) “it’s true because it
works“’ (Bhaskar 1991:7). In open systems, where invariant regularities do not
occur, only (a) makes sense. Applications require knowledge about transfactually
efficacious mechanisms or structures. Therefore Bhaskar, unlike Rorty, accepts
claim (2). Bhaskar’s argument is largely based on the presuppositions of the
natural scientists’ historical practice of experimentation:

In an experiment scientists produce a pattern of events. There is nothing in

itself special about this. For, as causal agents, we are co-responsible for events
all the time. And scientists could, in fact, produce a vast array of events,
most of no conceivable significance. What is so special about the patterns
they deliberately produce under meticulously controlled conditions in the
laboratory is that it enables them to identify the mode of operation of natural
structures, mechanisms or processes which they do not produce. What
distinguishes the phenomena the scientist actually produces out of the totality
of the phenomena he could produce is that when his experiment is successful,
it is an index of what he does not produce. A real distinction between the
objects of experimental investigation, such as causal laws, and patterns of
events is thus a condition of the intelligibility of scientific activity.
(Bhaskar 1979:11–12)

Bhaskar argues explicitly against Rorty, in that from the argument ‘there is no
way to know a thing except under a particular description’ one cannot infer
that ‘there is no way to know that thing exists and acts independently of that
description’. In fact, one must know that the world has existed prior to science,
 The normative logic of emancipation 147

and one can know that it exists and acts independently of science, because this
is a practical presupposition of science (Bhaskar 1989:152). Philosophy, as a
fallible mode of investigation of the necessary presuppositions of geo-historical
practices, can tell us that it is a condition of the possibility of scientific activities
that the real world is stratified, differentiated and structured, and that change
and emergence are possible, independently of our knowledge concerning its
specific content (Bhaskar 1979:8). Hence, the claim (2) that there is something
out there in addition to pragmatically known true statements, or in addition to
what is called ‘the truth about the world’, which is not only possible but also
necessary for scientific practices (Rorty does not reject scientific practices or
their fallible results). Therefore, we can also speak about the manner in which
‘cognitive matter is fashioned into a matching representation of a non-cognitive
object’, and we can also make judgements on this (mis-)match.
So far, so good. Bhaskar nevertheless argues that there is also an ontological
aspect of truth. In Plato etc., Bhaskar (1994:63–4) has further distinguished between
truth as ontological (‘achievable when referential detachment occurs’) and as
alethic (the non-propositional truth of or reason for objects). What are these
aspects of truth? What do they add to Bhaskar’s scientific realism? If ‘many
centuries of attempts to explain what correspondence between statement/theory
and facts is have failed’, is it possible to explain what ontological and alethic
aspects of truth are (unless they amount merely to a restatement of ontological
realism)?
According to Rom Harré, a statement using the word ‘true’ is typically
used merely (1) to refer to a statement, and (2) to pass a well-justified judgement
on it. In some contexts what matters is whether a statement ‘corresponds’ to
the objects of knowledge, while in other contexts our assessment may depend
more upon whether the statement under review coheres with other statements
(see Harré 1970:190). In both cases, intersubjective evidence and reasons for
a judgement are required. This would seem to indicate that truth is merely an
evidence-based human judgement about descriptions and explanatory
statements, concerning how the differentiated, layered, structured and changing
world really is.
However, truth must be seen as partially independent of justification here
and now: truth is something more than the actually available evidence. In the
sense of ontological realism, there are independently existing conditions for
truths, i.e. the way the world really is (or has been). Thus, concrete truth-
judgements should never be seen as total or unambiguous.3 Rather, the emergence
of being into ‘unconcealment’ is best seen as the covering up of ‘truth in its
inexhaustible fullness’ (cf. Palmer 1969:245). Because of the condition of
relativism, reality can never be reduced to our current knowledge of it. We
should accept that even when we are right—given the available evidence—in
making a judgement that something is true, there are always more things left to
be learned: perhaps a finer network of relations and deeper layers of reality;
perhaps new perspectives and conceptions that would make it possible to see
new aspects of reality; perhaps new evidence, which would allow us to revise
148 Explicating a critical realist methodology

our descriptions and explanatory models. There is always an abundance of

potential evidence in addition to actual evidence, and potential theoretical
concepts in addition to the actually existing ones.
Do we need to talk about the ‘ontological aspect of truth’ to make these
points? The duality theory of truth merely adds to the complexity of critical
realism (CR), without improving its explanatory or illuminatory power or any
other characteristics. If other things are equal, simplicity is desirable. Second, the
ontological interpretation of truth may have unfortunate political consequences.
Bhaskar (1979:201) himself argues that the metaphysical correspondence theory
of truth is reactionary in that it reinforces the authority of dominant—or any other
accepted—interpretations by ontologising human judgements. However, this applies
to his own notion of truth as well. Third, if the argument about the political
consequences of ontologised truths is accepted, then a more general point follows:
by seeing truth exclusively as a product of actual and potential human activity (in
contrast to the real world to which true statements make references), we can best
sustain openness to new interpretations and evidence. For these reasons, Harré’s
theory of truth—truth as it is used in scientific practices—seems to be better grounded
than any ontologised theory. In Harré’s theory, truth is seen as a normative and
metaphorical principle, which guides conduct:

If we read the realist manifesto ‘Scientific statements should be taken as true

or false by virtue of the way the world is’ as a moral principle it would run
something like this: ‘As scientists, that is members of a certain community,
we should apportion our willingness or reluctance to accept a claim as worthy
to be included in the corpus of scientific knowledge to the extent that we
sincerely think it somehow reflects the way the world is.’ Put this way the
manifesto has conduct-guiding force. It encourages the good and the worthy
to manifest their virtue in trying to find out how the world is. Seeking
[ontological] truth is a hopeless epistemic project, but trying to live a life of
virtue is a possible moral ambition.
(Harré 1986:89)

Instead of ontologised truth, what is needed is a consistent distinction between

realist ontology, relativist epistemology and the possibility of judgemental
rationality. Judgemental rationality is a normative notion. Read in the normative
manner, Karl Popper’s fallibilism, for instance, would therefore mean something
like ‘However much personal investment one has in a theory, one should not
ignore contrary evidence’ or ‘One should seek harder for evidence and reasons
that would count against a theory than for those which would support it’ and
so on (Harré 1986:90). Thus, truth relegates the notion of correspondence
between epistemic statements and ontic objects to the status of a metaphor for
the goal of scientific practices, which aim at articulating a matching representation
of components of a complex and their essential relations.
 The normative logic of emancipation 149

Pluralist truth as a regulative metaphor

What is metaphorical truth? A metaphor testifies to the freedom of consciousness

and the power of imagination. A metaphor suggests a resemblance between two
domains. The positive function of a metaphor is that, with its help, our conceptual
grasp of the subject matter is enhanced (see Cooper 1985:43–58; Harré 1986:76–
7). In this manner, truth can be understood and explained metaphorically:
truth as a metaphor suggests a resemblance between theory-dependent statements
used in explanatory descriptions, on the one hand, and the imagined but also
presupposed reality, on the other.
A picture of an aspect or part of the world can be—among other things—
accurate, but it cannot be similar, or isomorphic, to the world. Arne Naess
(1972:120–32) argues that even if reality is one (at any particular moment), it
does not follow that there is one correct description of the reality (of that
moment). Two mutually inconsistent systems of statements may both correspond
to a particular domain of reality. This is because the relation between a system-
dependent statement/set of statements and the particular domain of reality is
not one of mirroring or copying, and not even one of structured isomorphism.
Moreover, in principle, there is no end to different characterisations and
descriptions, some mutually consistent, others inconsistent, which a researcher
can offer in responsible social research. Each description is conceived in its
relation to a wider and deeper theoretical system. In addition to systematic
analysis of available evidence, each description and explanation should thus
also be assessed in relation to the relevant background theories (which also co-
determine conceptions of adequate data).
Naess (1972:131) even suggests that all sufficiently consistent and
comprehensive (‘near-total’, ‘near-global’) points of view (frameworks, theories,
systems) have a non-zero status of validity (truth, correctness). This is because it
is untenable to proclaim that the validity of a comprehensive system is absent
altogether. A comprehensive system must explicitly include ontology, methodology,
logic and semantics—or a set of arguments to the effect that a body of rules and
assertions concerning one or more of these is unnecessary, nonsensical or has
some property justifying its absence from a maximal comprehensive view. It is
only within such comprehensive systems that concrete theories/models and precise
statements can live. As Ludwig Wittgenstein (1979:105) has put it, ‘all testing of
a hypothesis takes place already within a system. And this system is not so much
the point of departure, as the element in which arguments have their life.’
The problem is how to compare and assess statements, descriptions and
explanations that belong to different systems. This problem can be analysed with
the help of Naess’s notion of ‘preciseness’. A statement Ti is more precise than a
statement T0, if there is at least one interpretation that T0 admits, but Ti does not,
and if there is no interpretation admitted by Ti which is not also admitted by T0.
Preciseness thus defined is a transitive relation. Now, Naess argues that ‘it is
problematic to compare scientific theories or images of the scientific enterprise
which are made precise in different directions’ (Naess 1972:130). Furthermore,
150 Explicating a critical realist methodology

‘if one starts making an exposition of a theory more and more precise in a definite
direction, what is gained in preciseness is lost in neutrality and comparability,
and vice versa’. However, as Thomas Kuhn (1970:202) has noticed in responding
to his critics, new languages and theoretical systems can be learned, and the
meaning of the statements belonging to a particular ‘system’ can be translated—
though only contingently and without a resort to a neutral ‘third’ language. In
this sense, human beings have the power to break through languages and
theoretical systems. Furthermore, although Naess is right that theory-comparison
and choice is problematic, it does not follow ‘that there are no good reasons for
being persuaded. Nor does it even imply that the reasons for choice are different
from those usually listed by the philosophers of science: accuracy, simplicity,
fruitfulness, and the like.’ What is suggested, however, is that ‘there is no algorithm
for theory-choice, no systemic decision procedure which, properly applied, must
lead each individual in the group to the same decision’ (Kuhn 1970:199–200).
There are also differences between ‘comprehensive views’, or ‘systems’, in the
sense that they may be more or less open to different possibilities of making a
theory/model precise. The more precise a statement is, the easier it is to assess it
against available evidence. Thus, we need a plurality of initial hypothesis drawn
from different theories; constant hermeneutic mediations between theoretical
concepts; and the reductive approach to iconic modelling. Moreover, some
comprehensive views are also capable of admitting that there are holes, silences,
contingent incommensurabilities, ambivalences and ambiguities; and that the
open texture of all discourses is, in fact, a rich source for potential development
(Bhaskar 1989:155). By admitting this much, as I think CR does, a comprehensive
discourse can remain open to change and learning. Openness and learning enable
more adequate, yet geo-historically situated, truth-judgements.

Explanation and critique

The nature of objects determines their cognitive possibilities. The substances,

qualities and relations of social worlds are not independent of concepts or
intentional human agency. In the words of Bhaskar (1979:43–4), ‘society is
both the ever-present condition (material cause) and the continually reproduced
outcome of human agency’. This ontological condition makes the relationship
between a researcher and his or her object of knowledge more complicated
than in natural sciences. Although, usually, (also contemporary) social objects
exist and function quite independently of the researcher, subject and object tend
to be internally and externally related. In time, social scientific models and
stories may contribute to the re-signification and transformation of practices.
The task of social-scientific explanation is to form a plausible and empirically
confirmed iconic model of geo-historical components and their relations. The
researcher has to grasp the meaningful elements and internal relations of a
social complex; the relevant regulative rules; the resources and action-possibilities
of different positioned actors; the unintentional causal consequences of action;
 The normative logic of emancipation 151

the possible emergent powers of social system; and so on. It may well be that
there are many historical episodes, practices and/or systems that can be grasped
in terms of a particular iconic model. It is also possible to form higher-order
models Mn with wider range but with lesser accuracy than the concrete models
An–1 Bn–1 Cn–1, etc., which in some relevant respects are essentially similar.
The hermeneutic assumption of relative privilege of an agent’s own account
of his or her reasons depends upon the causal efficacy of his or her practical
competencies and reasons—for otherwise there would be no reason to prefer her
account to that of an outsider. However, causal efficacy establishes an ontic
distance between the agent’s competencies and reasons and any particular
expression or statement of them. Moreover, the description of even our own
sensations, competencies, beliefs and emotions is profoundly affected by a host
of conceptual choices. The ontic distance and interpretative openness mean
that any description of the agent’s real beliefs must be corrigible. (Bhaskar
1986:164). Also, a social scientific explanation can reveal that beliefs of actors
are in some regards false or illusory. An explanatory model can explain why
they must be false or illusory. A rationalisation of action does not necessarily
correspond to the real causes of the action, nor is it necessarily able to grasp the
wider consequences of action. Yet that very rationalisation can be necessary for
the (re)production of particular social relations and practices. The stratification
model of human action and agency captures this idea.
According to this model, illustrated in Figure 6.1, agency and action are
conditioned (i.e. caused materially) by social structures, which are, dialectically,
also the continuously (re)produced outcome of human agency and actions. The
consequences of actions include thus (re)production of particular structures.
This consequence can provide motivation M(r) both for reasons r and particular
rationalisation R(r) of these reasons; motivation M(r) may also be partially
unconscious; and rationalisation R(r) may conceal, mystify or reify the
consequence, which, at this level, is likely to be unintended.
Bhaskar (1979:74) distinguishes three ways in which interference between a
subject’s interest in an object and his or her knowledge of it could operate. It
could operate consciously (as in lying); it could operate semi-consciously, as in
the wishful thinking of the incurable optimist or the special pleading of a pressure
group; or it could operate unconsciously, whether or not it may subsequently
become accessible to consciousness. The conclusions of the unconscious mode
of interference can be either rationalisations of motivation, or they can constitute
mystifications or reifications of social structure (ideologies). This is the ontological
ground for the possibility of explanatory critique. Explanatory critique is directly
connected to emancipatory political change:

the object that renders illusory (or superficial) beliefs necessary comes…, at
least in the absence of any overriding considerations, to be criticized in being
explained; so that the point now becomes, ceteris paribus, to change it.
(Bhaskar 1979:67)
152 Explicating a critical realist methodology

Figure 6.1 Stratification of agency and action4

The force of explanatory criticism comes from the positive value of truth. In
the following, I shall explain in detail the necessary connections between
explanation and criticism, and take Bhaskar’s argument a few steps forward in
political theoretical terms.

The logic of explanatory emancipation

If interests are given an economistic interpretation, Bhaskar’s schema can be seen

as an articulation of Marxist ideology-critique. We do not have to assume, however,
that only one mode of action—strategic action—prevails, or that interests can be
defined narrowly in terms of money or property ownership. More generally,
interests can be layered—surface level interests are justified in terms of deeper
background discourses—and are explainable in terms of geo-historical processes.
For instance, repression and violence stemming from defending particular
established interests can be an outcome of a pathological learning process. For
Karl Deutsch (1963:248), the learning process is pathological if it leads to ‘a
change in inner structure that will reduce rather than increase the future learning
capacity of the person or organization’. He goes on to argue that:

will and power may easily lead…[even] to self-destructive learning, for they
may imply the overvaluation of the past against the present and the future,
the overvaluation of the experiences acquired in a limited environment against
the vastness of the universe around us; and the overvaluation of present
expectations against all possibilities of surprise, discovery, and change.
(Deutsch 1963:248)
 The normative logic of emancipation 153

Deutsch argues that pathological learning processes often occur in the context
of fear, stress and conflict. Background discourses, which naturalise, reify and
mystify social relations, can stimulate pathological learning. Layered interests
can also be related to aspects of structures such as the occupation of social
positions (‘power’ in the conservative sense) or resources for transformative
actions (‘power’ in the radical sense of political mobilisation). Once real, whether
fully conscious or not, interests can intervene in the formation of agency by
providing motivation M(r) both for reasons r and particular rationalisation R(r)
of them. Illusion can also become institutionalised and appear for a given actor
as a natural part of assuming a position in practices and organisations.
According to Bhaskar (1979:80), one is justified in characterising a discursive
formation (DF) as ‘ideological’ if both (1) some relevant elements of DF are
false; that is, one possesses a superior explanation T1 for the phenomena, and
(2) one possesses an explanation T2 of the falsity of the beliefs of DF in question
and why they are held. Where different theories are required for the satisfaction
of (1) and (2), they must at least be consistent with each other. (1) is the critical,
(2) the explanatory condition. (1) suffices for criticism and all intellectual criticism
satisfies this condition; (2) is required if the criticism is to be explanatory and
directed at emancipatory social change (Bhaskar 1986:179). Schematically, the
logic of explanatory emancipation can be summarised as follows:

Arg. 1: (1) T1 is more true (explains and illuminates more) than

TDF; it also covers at least some of the relevant areas of the
range of the TDF, which in turn is an essential part of DF, a
‘near-total’ system.

(2) T2 explains the (re)production of DF in terms of a causal

complex K i.

(3) Thus, one should evaluate negatively the relevant parts

of the causal complex (K i) that are responsible for the
(re)production of false beliefs.

(4) Consequently, one should also evaluate positively

political action directed at absenting or transforming the
relevant parts of complex Ki.

Bhaskar (1986:178) claims that his version of arg. 1 is naturalist, not moral, in
character. ‘The notion of false consciousness employed here simply involves in the
first instance that of disjuncture, mismatch or lack of correspondence (representative
adequacy) between belief and object.’ Bhaskar’s main target is the positivist fact/
value-distinction:

Might it not be objected, however, that the fact/value distinction only breaks
down in this way, because one is committed to the prior valuation that truth
154 Explicating a critical realist methodology

is a good, so that one is not deriving a value judgement from entirely factual
(natural) premises? But that truth is a good (‘ceteris paribus’) is not only a
condition of moral discourse, it is a condition of any discourse at all.
Commitment to truth and consistency apply to factual as much to value
discourse; and so cannot be seized upon as a concealed (value) premise to
rescue the autonomy of values from factual discourse, without destroying
the distinction between the two, the distinction that it is the point of the
objections to uphold.
(Bhaskar 1979:81)

The positivist fact/value-distinction fails. Nevertheless, Bhaskar cannot sustain

the claim that he is deriving value-judgements from purely naturalist premises.
However necessary and universal truth may seem for all discourses, commitment
to truth is a normative commitment (cf. Habermas 1979:2). Truth is therefore a
value among many values in normative discourse. Value-judgements are relative.
As Andrew Collier (1994:170) argues, ‘taking money from that Coca Cola
machine would be theft’ may imply ‘you should not take money from that
Coca Cola machine’. However, if it is ‘the only way to get coins to phone the
US President and stop nuclear war’, the case is drastically altered (as in the film
Dr Strangelove). Sometimes other normative considerations may, even, override
truth. Peter Winch makes a conservative argument to this effect:

question concerning the importance in human life of knowing the world as it

really is. I am thinking of the moral question which so exercised Ibsen in
such plays as ‘The Wild Duck’ and ‘Ghosts’: How far is it important to a
man’s life that he should live it in clear awareness of the facts of his situation
and of his relations to those around him? In ‘Ghosts’ this question is presented
by considering a man whose life is being ruined by his ignorance of the truth
about his heredity. ‘The Wild Duck’ starts from the opposite direction: here
is a man who is living a perfectly contented life which is, however, based on
a complete misunderstanding of the attitude to him of those he knows; should
he be disillusioned and have his happiness disrupted in the interests of truth?
(Winch 1971 [1958]:21–22)

In social sciences, this hesitancy may be reinforced by the fact that there are no
neutral criteria of truth to which to appeal. Epistemic criteria differ and are
dependent on the wider system to which they belong. Theories can be made
precise in many ways, and the hermeneutic movement between theoretical concepts
is always precarious. Moreover, in social sciences, subject and object tend to be
internally and externally related. These relations are also a matter of ethical concern.
Derrida’s concept of undecidability addresses these ethical problems. Undecidability
pertains that any judgement about truth or other normative issues has to assume
ethico-political responsibility. ‘A decision can only come into being in a space that
 The normative logic of emancipation 155

exceeds the calculable program that would destroy all responsibility by transforming
it into a programmable effect of determinate causes’ (1988:116). For Derrida
(1978:79–153), this ethico-political responsibility is also a question of violence.
Determinate decisions about truth of statements, models and stories can be made
on rational and pragmatic grounds. However, by talking about undecidability,
Derrida (1978:83; see also 1988:112 and 150–3) calls for responsibility of decision
and non-violent ethical relationships with the others.

The problem of violence

Science is a value-based social enterprise, relying on practical judgements, which

are in principle structurally similar to lay-actors’ judgements. Researchers have
no a priori privileged position—outside geo-historical processes—for making
normative truth-judgements. Hence, the following ethico-political problems:

1 Under what conditions is it possible to decide rationally, in relation to an

audience and situation, at some definite moment in the flow of time, that arg.
1 is valid?
2 Should we give the ‘accused’ a chance to answer the implicated accusation?
3 If communication does not suffice for a change, should some conditions be
set for strategic, transformative action?

The condition of epistemic conflict prevails. In most matters, we have different

possible, and contradictory, sources of knowledge. CR admits the possibility
that even a person opposed to all his contemporaries may turn out to be right
in matters of truth. However, an epistemology, which relies on inter-subjective
judgements and criteria, must seek rationality in dialogue. How should truth-
judgements be appraised? A rational judgement is not a matter of voting. The
requirement of openness implies that we cannot name any particular decisive
community; nor do we need a universal consensus for making rational
judgements. Rationality is about good reasons and learning, and implies openness
to criticism from whatever source. If a person is deaf to the arguments of others
he may either ignore contrary reasons or reply to them with mere dogmatic
assertions. Without openness, there would be no basis for learning or claiming
that the probative strength of all relevant evidence has been considered.5 What
matters are: good reasons for particular (falsifiable) iconic model and explanatory
story; engagement in dialogue in which the goodness of these reasons are tested;
and openness to critical debate. It is a condition for rational truth-judgements that
these requirements are not contradicted.
Bhaskar argues that, once we have also fulfilled the explanatory condition of
arg. 1, there is a moral obligation to remove or change the parts, which have co-
caused the false beliefs. This is too simple and strong, and therefore easily misleading.
False beliefs can be absented or revised by rational dialectical persuasion. In open
systems, some of the causally relevant components may survive revised beliefs.
156 Explicating a critical realist methodology

Moreover, because of the requirements of dialogical rationality, there is a strong

prima-facie case for a rational, open-minded dialogue instead of some form of
strategic action. Strategic political action may in fact violate these requirements.
For these reasons, and in order to explicate the positive value of truth, the
logic of explanatory emancipation should be modified as follows:

Arg. 2 (1) T1 is more true (explains and illuminates more) than

TDF; it also covers at least some of the relevant areas of the
range of the TDF, which is an essential part of DF, a ‘near-
total’ system.

(2) Truth is a positive value.

(3) Thus, one should evaluate negatively the relevant parts

of DF, and engage in a discussion about the merits and
problems of theories.

(4) T2 explains the (re)production of DF in terms of a causal

complex Ki.

(5) Thus, one should evaluate negatively the relevant parts

of the causal complex K i that are responsible for the
(re)production of false beliefs.

(6) Attempts (3) do not make any difference.

(7) T1 and T2 still hold, even if only in a qualified form.

(8) Consequently, one should evaluate positively public

communicative action directed at removing or changing the
relevant parts of Ki.

This explication also makes the criticised actors equal partners in a sustained
dialogue. It also counters Arendt’s (1973 [1965]:78) scepticism about politics based
on revealing sources of unconscious motivation: ‘Since [this demand to reveal
innermost motivation] actually demands the impossible, transforms all actors into
hypocrites; the moment the display of motives begins, hypocrisy begins to poison
all human relations.’ Indeed, is it possible to put forward critical arg. 1 without
destroying the trust, which is necessary for all communication? I think this is a
potential source of violence in CR, and therefore it has to be addressed explicitly.
Any social theory has to be aware of the violence inherent in its own categories,
and of the violence potentially justified in terms of its categories. Arg. 2 tackles
this problem.
What if public communicative action is not possible because of political
repression and violence, or has no success, perhaps because DF ignores criticism
 The normative logic of emancipation 157

or remains deaf and closed for other reasons? Should we conclude a positive
evaluation of strategic action directed at removing or changing relevant parts of
the complex Ki? Only a conditional conclusion to this effect is possible. Rational
judgements cannot, and the consequences of rational judgements should not,
contradict the notion of good reasons and openness to further critical debate.
Hence we could continue arg. 2 as follows:

Arg. 2 (9) Either despite the public communicative actions, TDF (and
DF) remains unchanged, or critical communicative actions
are not politically possible with respect to the relevant
political situation.

(10) Thus, one should evaluate positively strategic political action

directed at absenting or changing relevant parts of the complex
Ki on the condition that the rational persuasion that T1 or T2 or
both are false remains politically possible also with respect to
the changed political situation, which is co-caused by that
transformative action and its consequences.

Does (10) of arg. 2 exclude the use of any violent means to achieve the desired
end? I agree with Derrida (1988:112) that violence may ‘remain in fact (almost)
ineradicable’. The aim is, nonetheless, to argue for a peace culture of social
sciences and for reflective non-violence. Reflective analysis of violence and its
conditions in one’s own categories are among ‘the least violent gestures’ (Derrida
1988:112). The openness of social systems points towards the same conclusion
(10). In open systems, within which social actors possess generic powers,
qualitative changes and emergence are possible, but predictions are not. Changes
may also concern one’s own geo-historical, open-ended being. Moreover, in
open systems, unintended consequences of action are ubiquitous.
The requirement (10) can be read as a moral obligation to limit strategic action to
such modes that are conducive to preserving or building a sphere for public politics
(see Patomäki 2001a). Even in precarious and dangerous political situations, this may
be more feasible than it appears. Periods of relative stability, and widespread
reifications and mystifications of social structures, may make us forget the basic insight
of the theory of non-violent political action. Power is ‘fragile, always dependent for its
strength and existence upon a replenishment of its sources by the co-operation of a
multitude of institutions and people—co-operation which may or may not continue’
(Sharp 1984:8).

The role of other values

I have now argued that truth is a regulative metaphor of correspondence, a

virtue of scholarly practices and a condition of many discourses. Because truth
has normative force, an explanatory model, which is judged to be true by
158 Explicating a critical realist methodology

virtue of its explanatory statements being true, has normative political

implications. These implications are, however, conditioned by the relational
nature of human judgements and therefore by the ethico-political principle of
non-violence. My account of the logic of explanatory emancipation indicates
that Bhaskar has not succeeded to deduce ought from is. Rather, the discourse
about ‘what is’ is guided by regulative rules, and is therefore already normative.
There seems to be no reason why a normative discourse should be concerned
exclusively with truth and non-violence. It is one thing to say that practices and
systems should be evaluated only if we have adequate knowledge of them, and
another to say that explanation is the only possible source of ethico-political
critique. Social practices and systems can, in fact, be evaluated from a number
of normative standpoints, including economic efficiency and stability as well as
rights, justice, democracy, human flourishing and ecological well-being. Most
of these are not necessarily derivable from explanations (Sayer 2000:156).
In the 1990s, Bhaskar (1993:279–308; 1994:141–60) and Andrew Collier
(1999) have attempted to widen the sphere of CR normative discourse. However,
their aim is to derive values from fundamental realities such as (abstract) action
and (the independent existence of) being. Although some of their arguments can
be employed context-sensitively, I would like to defend a more concrete and
dialogical normative discourse. A philosophical preoccupation with the issue of
how an entire system of ethics and politics can be derived from a few simple
premises (see Bhaskar 1993:224–31; 1994:161–6) has simultaneously two effects:
(1) it detaches normative discourse from concrete realities; and (2) it makes
normative discourse monological and potentially violent by ignoring or excluding
the ethico-political concerns of different others. Bhaskar’s (1993:87) own term
‘speculative illusion’ describes these effects well. Preoccupation with the derivability
of philosophical systems implies the sublimation of both social life and research
into philosophy. This is, indeed, the speculative illusion of Dialectic and Plato etc.6
It makes more sense to start with a concrete geo-historical problematic and
see whether and how any philosophical arguments can be invoked in that context.
Philosophical arguments can be deployed in connection with statements that
attempt at description, explanation, iconic augmentation and evaluation of
practices and systems. There is no point in trying to reduce various explanatory,
normative and edifying arguments to philosophy, or making them all at once
and for good, independently of their geo-historical context. Moreover, judgements
about normative matters are structurally similar to truth-judgements, i.e. relational
and situated human judgements (see Patomäki 1992b).

The role of concrete Utopias

Clearly, a society would be improved if its illusions, contradictions, violence and

injustices were reduced. However, do we know how this could be achieved?
Mere negative criticism is not sufficient for concrete action. Criticism presupposes
the possibility of better practices (Sayer 2000:161). To address this problem, Bhaskar
 The normative logic of emancipation 159

(1993:286, 297) has introduced the term ‘concrete Utopia’.7 A concrete Utopia
does not currently exist, but should be politically possible to achieve, and feasible
as an alternative way of organising social practices and relations. The realisation
of a concrete Utopia involves practical wisdom, lessons drawn from past or
contemporary models, counterfactual reasoning about the possible effects of an
altered context, as well as thought-experiments about the consequences of the
transformed practices and systems (See Sayer 2000:160–5).
Practical wisdom is an Aristotelian intellectual virtue, a state that allows an
actor who attains it to be able to ascertain what is ethico-politically good, and
then to deliberate about how best to reach that good. This calls for knowledge
about the concrete political situation, and a systematic reflection on the conditions
of action. The more difficult part may concern, however, ascertaining what
would be a better, yet feasible set of practices or parts of a system. Lessons
drawn from the past or present practices may provide ‘drafts’ or models for
transformative projects, whether successful or not (Sayer 2000:162–3). The
lack of success of a particular practice or project could have been dependent on
the particular geo-historical context in which it originally occurred. What would
have been the possible effects of a somewhat different context? Is it possible to
build components of a complex that would make something similar more feasible
in the future? Roberto Unger emphasises the action- and context-dependency
of political projects:

Some of our projects may fail and by their failure may cast a critical light on
the descriptive and explanatory assumptions that informed them. But we
can never be entirely sure that these same projects might not succeed under
slightly different circumstances or with the help of slightly different methods.
(Unger 1987b:323)

Unger may over-emphasise the contingencies of failure and success. However,

this cannot be judged a priori. Practices or projects that have failed are thus
worth studying in great detail. It is, moreover, possible to produce at least
partially new and untried models or projects. How do we test their workability?
Knowledge about related existing mechanisms and contexts, coupled with
analogical and counterfactual reasoning, can help to determine some of the
potential problems of making them real. Although history is open and
unpredictable, at least some of the likely consequences of introducing novel
practices and relations can be scrutinised in terms of historical lessons, analogical
and counterfactual reasoning and thought-experiments. Transformed practices
and systems may also have emergent properties or powers, which will have to
be taken into account as well.
A concrete Utopia can remedy some problems while generating others. The
new problems could even be worse than the old ones. Social life may therefore
be frequently ‘dilemmatic’, as Sayer (2000:163) emphasises. Realist social sciences
cannot assume that any ideas or concepts can construct whatever social realities.
160 Explicating a critical realist methodology

To the contrary, it is necessary to study the viability of concrete Utopias by all

available means. In any given geo-historical context, there are thus limits to
programmatic imagination. However, by changing parts, or the nature of, the
wider context, new concrete Utopias may well become possible.

Conclusion: the tasks of critical social sciences

In the dialectical comment of Chapter 1, I argued that, to overcome the

international problematic, reality should be reclaimed. Causal powers, absence,
emergence and change are all real. I argued further that instead of the usual
conflation between subject and object, and the consequent confusions, their
relations should be analysed in more specific terms. Now we should know how.
First, subject and object of social sciences are separate geo-historically, causally
and ontically. Second, subject and object are also related internally in case-specific
ways—also through the double-hermeneutic process of understanding lay-
meanings. Third, social sciences are not only subject to causal determinations
(e.g. funding, research governance), but their products can also have geo-historical
causal effects on practices. Social sciences and their objects are both situated in
the differentiated, stratified and structured world of geo-historical determinations.
I have now concluded my articulation of a realist ontology and epistemology,
and related methodology, by locating social sciences in the world. In Chapter 5,
I argued that in mimesis3 social sciences can re-signify practices in their temporal
dimension, to the extent that explaining and narrating is to re-make action. In
this chapter, I have further elucidated how and why social sciences tend to have
critical ethico-political implications. The logic of explanatory emancipation is based
on the normative force of truth. Bhaskar’s basic insight is correct: ‘The object
that renders illusory beliefs necessary comes to be criticised in being explained;
so that the point becomes to change it.’ The problem is to specify the rules and
conditions of applying this critical argument without destroying trust and openness
to communication, which are necessary for adequate truth-judgements. I have
tried to accomplish this by rewriting the logic of the critical argument (arg. 2).
The relations between explanatory criticism and more general normative
discourse are manifold. Valid normative judgements depend on adequate
descriptions and explanations. Any normative criticism presupposes the possibility
of better practices, and their feasibility must be assessed in terms of realist
social-scientific models and thought-experiments. In these senses, descriptive
and explanatory conceptions are primary. However, truth is only a value among
many. There is no reason why normative discourse should be concerned
exclusively with truth and non-violence. To the contrary, practices and parts of
systems can be evaluated from a number of normative standpoints.
Moreover, as stated, criticism presupposes the possibility of better practices
and parts of systems. Social sciences are therefore also about Aristotelian practical
wisdom. They can contribute to ascertaining what is ethico-poiitically good and
how to reach that good—if it is achievable at all. In the ideal world of mutually
 The normative logic of emancipation 161

reinforcing and cumulative reforms, however, humankind may triumph over

anything we can now imagine.

Notes

1 Galtung defines ‘cultural violence’ as those aspects of culture, the symbolic sphere of
our existence—exemplified by religion and ideology, language and art, empirical science
and formal science (logic, mathematics)—that can be used to justify or legitimise
violence. In this sense, even peace research may be a form of cultural violence. Galtung
goes on to argue ‘that a major task of peace research, and the peace movement in
general, is that never-ending search for a peace culture…’ (see Galtung 1990:291).
2 Bhaskar 1986:100–1. Consider, for instance, Popper’s theory of truth, which is largely
based on Tarski’s account of the meaning of truth. According to Popper, it is ‘decisive
to realise that knowing what truth means, or under what conditions a statement is
called true, is not the same as, and must be clearly distinguished from, possessing a
means of deciding—a “criterion” for deciding—whether a given statement is true or
false’. In this theory, the meaning of truth is that ‘an assertion, proposition, statement,
or belief, is true if, and only if, it corresponds to the facts’. However, there are no
general criteria of truth, and hence ‘science is fallible, because science is human’, and
because ‘we may err’. Nevertheless, we can say ‘that a statement “a” gets nearer to the
truth than a statement “b” if and only if its truth content has increased without an
increase in its falsity content’. Moreover, ‘it should be very clearly understood that
the idea of one statement “a” getting nearer to the truth than another statement “b” in
no way interferes with the idea that every statement is either true or false, and there
is no third possibility’, for ‘it only takes account of the fact that there may be a lot of
truth in a false statement’. Finally, although Popper argues that in principle nothing is
exempt from criticism, he also maintains that ‘at least at the moment it seems to me
unlikely that criticism of the logical theory of truth and the theory of getting nearer to
truth will succeed’ (see Popper 1971:369–96).
This is a conventional version of the correspondence theory of truth, which, like the
other versions, is untenable. The problem with Tarski’s account of the meaning of truth
is that, as far as it is seen as philosophically neutral, it is tautological, trivial and non-
informative. As such, it cannot be used as a philosophical theory of truth. However,
Popper’s theory of getting nearer to truth, which utilises Tarski’s account, is neither
neutral nor trivial. The problem is that it presupposes that the truth is already known,
for, unless the truth is already known, how would it be possible to tell something about
the truth-content of an assertion, statement, belief and the like? In other words, in fact
Popper, if he is to keep his notions of truth and ‘verisimilitude’, slips back towards
absolutism, which he is so eager to refute, at least as far as Plato, Hegel and Marx are
concerned, (see Harré 1986:41–5; Margolis 1986:89–96). Hence, the consequence is
that there is no longer any difference between the best available ‘scientific’ theory or
interpretation of the world—the one decided to be the yardstick of truth, or the description
of the ‘facts’—and the world. This is the implicit ‘oneness’ of being and knowledge of it
of all traditional accounts of the correspondence theories of truth.
3 As Galtung notes, if the logical tool used is bivalent logic with its strict line between
valid and invalid, then no interference can be made with ambiguous truth-values for
the antecedents or the interference, and ‘truth’ becomes absolutist in nature (Galtung
1988:162–75; Galtung 1990:301). That this has been seen as a limitation is also clear
from the effort to develop the language of polyvalent logic with more than two truth-
values, by adding ‘indeterminate’, ‘undefined’ or ‘possible’ truth-value(s) to the
standard two-values system (see Haack 1974: particularly 30–5). Haack argues for
162 Explicating a critical realist methodology

‘deviant logics’ on the grounds that the positivist and logicist analytic/synthetic
distinction should be rejected: logic is not about ‘pure form’, as in Kant’s view, but
rather about more or less substantial rules of thought, and ‘no factual statement is
immune from revision’. However, if the classical laws of thought used in deriving
theorems in polyvalent logic are still based on bivalent logic, and/or if the ‘third
possibilities’ are of the nature ‘not yet decided to be true or false…some of the more
plausible interpretations of the intermediate values of many-valued systems are such
that “true” and “false” remain mutually exclusive and jointly exhaustive’ (Haack
1974:61), then it is hard to see that any real exception has been made to the following
‘classical’ rules:

(i) Excluded middle (tertium non datur): ~p v p.

(ii) No contradiction: ~(p & ~p).
(iii) Principle of identity: (p)p=p.

Many philosophers use the expressions tertium non datur and ‘principle of bivalence’
interchangeably, or take for granted that these principles are equivalent. However, it
is, in principle, possible to deny ‘bivalence’ and still hold that (~p v p) is a theorem of
the logical system. Nevertheless, if tertium non datur is a theorem of a system, and if
Tarski’s principle that ‘A’ is true if ‘A’ holds for that system, then the principle of
bivalence must also hold (see Haack 1974:67–8) (This seems to suggest, moreover,
that it has been very commonplace to assume Tarski’s principle implicitly, as a self-
evident truth). Moreover, although ‘tertium non datur’ does not exclude the possibility
that (p & ~p), it is thought, almost without exception, that ‘from a contradiction
anything is derivable, so that such a logic [which accepts (p & ~p)] would be useless
for the purpose of discriminating valid from invalid inferences’ (Haack 1974:36).
Now, if logic and mathematics are viewed as a formal bivalent game with one basic
rule, that a theorem T and its negation ~T cannot both be valid, then there may be
violent social and political consequences in the sense analysed in this chapter: truths
come to be seen as absolute and non-compromised (even though the possibility of
errors is admitted, in principle). Indeed, the praxis of scientific theory-choice and
criteria-choice requires and presupposes ‘alternative’ logical systems. Moreover, bivalent
logic is also inadequate in assessing the validity of a descriptive and picturesque model
concerning an aspect and part of the differentiated, structured and stratified reality
with its complex and ‘inexhaustible fullness’. From this point of view, it is easy to
agree with Joseph Margolis, who argues that relativism is a thesis according to which
‘it is methodologically advisable to retreat from insisting on a strong bipolar model of
truth and falsity’ (see Margolis 1986:111).
4 Modified from Giddens 1979:54; Bhaskar 1986:128.
5 These points rely on Habermas 1984 [1981]: Ch. 1. Habermas argues (1984 [1981]:
8) that ‘rationality has less to do with the possession of knowledge than with how
speaking and acting subjects acquire and use knowledge’. Further, he bases ‘the
rationality of an expression on its criticizability’ (Habermas 1984 [1981]:10). Further,
‘corresponding to the openness of rational expressions to being explained, there is,
on the side of persons who behave rationally, a willingness to expose themselves to
criticism and, if necessary, to participate properly in argumentation. In view of their
criticizability, rational expressions also admit of improvement; we can correct failed
attempts if we can successfully identify our mistakes. The concept of “grounding” is
interwoven with that of “learning”. Argumentation plays an important role in the
learning processes as well…. The medium in which these negative experiences (failed
attempts) can be productively assimilated is “theoretical discourse” that is, the form
of argumentation in which controversial truth claims are thematized’ (Habermas 1984
[1981]:18–19).
 The normative logic of emancipation 163

6 In fact, the later Bhaskar seems to oscillate between the speculative illusion of a
Hegelian philosopher and the romantic, radical democrat illusion about a free,
eudaemonic society, within which free, concretely singular individuals are capable
of emancipation that stems directly from ‘an inner urge that flows universally from
the logic of elementary absence (lack, need, want or desire)’ (Bhaskar 1993:299).
Both illusions would make CR social sciences redundant and therefore also negate
the original purpose of CR, which was to ground and show a place for critical
social sciences.
7 The oscillation mentioned in note 6 applies also to Bhaskar’s version of the idea of
concrete Utopia. The following explication is therefore based more on Sayer, Unger
and my own analysis than on Bhaskar.
Part III

Visions of world
politics
7 Modelling Thucydides’
Melos episode

To concretise the notion of iconic modelling, I shall explain the well-known

Melos massacre described by Thucydides in his The Peloponnessian War.1 I follow
upon Hayward Alker’s (1996:23–63) analysis of ‘the dialectical logic of
Thucydides’ Melian dialogue’2, and aim at developing Alker’s analysis into a
fully-fledged CR explanation. I shall assess different hypotheses about the Melos
episode, and show how an iconic model about the relevant causal complex and
development of its key elements can be built. Because there is little other readily
available evidence, other than Thucydides’ text and its interpretations, the task
of gathering and assessing evidence is made relatively simple.3
The chief point about the choice of this case is, however, that Thucydides’
History is constitutive of the international problematic and, thereby, of
contemporary political realism. David Hume (1994 [1777]:135) used Thucydides
to support his view about the universality of the balance of power system:
‘Thucydides represents the league, which was formed against Athens, and which
produced the Peloponnesian war, as entirely owing to this principle.’ Immanuel
Kant (1983 [1784]:38n) wrote that ‘the first page of Thucydides (says Hume) is
the only beginning of all true history’.4 Ever since, Thucydides’ History has
been taken as a corroboration of the wisdom of the tradition—and therefore as
evidence for the timeless inevitability of the international problematic. For
instance, Robert Keohane writes:

To explicate the research program of Realism, I begin with two classic works,
one ancient, the other modern: The Peloponnessian War, by Thucydides, and
Politics Among Nations, by Morgenthau. The three most fundamental Realist
assumptions are evident in these books: that the most important actors in
world politics are territorially organised entities (city-states or modern states);
that state behavior can be explained rationally; and that states seek power
and calculate their interests in terms of power, relative to the nature of the
international system they face…. In reconstructing state calculations,
Thucydides and Morgenthau both assume that states will act to protect their
power positions, perhaps even to the point of seeking to maximize their

167
168 Visions of world politics

power. Thucydides seeks to go beneath the surface of events to the power

realities that are fundamental to state action.
(Keohane 1986:163–4)

Keohane refers to one of the most contested passages in The Peloponnessian War (I:
23), according to which the real cause of the Peloponnesian war was ‘the growth of
the power of Athens, and the alarm which this inspired in Lacedaemon’. This
‘made war inevitable’. This was probably written two decades after its textual
context, as a post-war afterthought. Keohane takes it as evidence of Thucydides’
neo-realism.5 On this reading, it is also obvious to assume that the Melos massacre
is an indication of a universal invariance. This is the law of the stronger: great
powers naturally impose their will; weaker powers accommodate this or are crushed.
Athens acted inevitably as any great power would do, yet the Melians did not
submit. Hence, the Melian men were killed, and the women and children enslaved.
Readings of Thucydides legitimise or criticise contemporary practices. This is
why Thucydides remains important. I shall argue that available evidence points to a
very different explanation than that offered by followers of Hume. A crucial cause of
the massacre was that Athens as a collective actor changed in the course of the war.
The Athenians’ claim ‘that it is a necessary law of nature to rule whatever one can’
(V:105) is not a statement about a Newtonian law but an attempt to justify their
imperialism and, genealogically, a causal result of the pathological processes that
restructured the opinions and actions of Athenian citizens. The development that led
to these changes is part of Thucydides’ overall explanatory and moral narrative
about the fate of Athens and, also, Hellas. The Peloponnesian war was, to use
contemporary categories, also a civil war. Thucydides seems to have viewed history
as an open process, produced by actors capable of arguing and acting otherwise. A
few passages point towards a more necessitarian story, and they suggest that the
Peloponnesian war is best conceived as a tragedy. Methodical iconic modelling can
help to explicate relevant components of the causal complex and also to reflect
systematically upon different possible temporal structures of the explanation.

A rational-choice model

As is well known, the Melos episode—which occurred during the Nicias Peace
period 416 BC, in the middle of the 27-year-long Peloponnesian war—consists
of a long dialogue and a short statement of the subsequent events. The subsequent
statement explains laconically that Melos was attacked, all adult men were
killed and other Melians were enslaved by the Athenians (Thucydides V:84–
116). Why did the massacre happen? What caused it?
At the outset, it is enough to have interesting possibilities and somewhat plausible
candidates for endorsement. The first possibility is to articulate the neo-realist
logic in terms of a rational-choice model. The Melos massacre can be seen as an
outcome of strategic choices of two state-actors, Athens and Melos. Actually, this
 Modelling Thucydides’ Melos episode 169

interpretation can be effected in at least two different ways. The research

programme of situational determinism tends to see the necessitarian logic of rational
choice everywhere; it is based on the ampliative strategy of modelling. A rational-
choice model can also make sense from a critical realist (CR) viewpoint, but then
the preferences and (contingent) prevalence of strategic action, in a particular geo-
historical context, can be subjected to explanation too.
According to Spiro Latsis (1972), the research programme of situational
determinism endeavours to find, describe and analyse situations where the
‘obvious’ course of action is determined uniquely by objective conditions. These
situations can be called single-exit situations. Latsis explains them with the help
of an analogy:

to say that a seller under perfect competition deliberately chooses a course of

action to maximise profits is analogous to saying that a member of the audience
is maximising if he runs out of the single exit available to him in a burning
cinema.
(Latsis 1972:210–11)

According to Latsis, the temptations for the social scientist to employ such a
method are considerable, for it makes the application of mathematical techniques
easier—and the utilisation of mathematical tools is often equated with the quality
of being ‘scientific’. It also renders his field of study relatively autonomous and
hence immune to criticism from outside.6 Some general, regulative, philosophical
principles of this programme are (Latsis 1972:208–9):

1 The approach is individualistic: behaviour is explained in terms of

individual human agents, who are acting in a social situation, and their
exogenous preferences.
2 The strategically rational choices of the individual agents are so
constrained by their situation that only ‘minimal psychological
assumptions’ are required to explain their actions.
3 Agents are rational: they act appropriately to the ‘logic of the strategic
situation’; this implies that there are no chronic information problems,
although uncertainty may well be present.

How could we analyse the Melos episode in terms of the situational determinist
research programme? There are a few selected passages in Thucydides that are
usually taken up and presented to justify a political realist or a rational-choice
reading. For instance, ‘my habit has been to make the speakers say what was in
my opinion demanded of them by the various occasions, of course adhering as
closely as possible to the general sense of what they really said’ (I:22). This
passage is often combined with the view that Thucydides believed that brute
power politics ruled relations between states. It is assumed that Thucydides
made a distinction between relations within, and between, states. In the
170 Visions of world politics

inter-state relations, there is no justice—just the rule of brute force. It is the

strong who decide how they will treat the weak, and moral judgements are
virtually inapplicable. This is how we should also interpret the expression ‘what
was demanded by the various occasions’.
The adequate explanation must therefore start from the fact that given
‘situations’ make demands on actors and that the actors must speak in accordance
with these demands. However, what, then, did Thucydides mean by ‘the general
sense of what they really said’ or ‘what was called for by each situation’ (I:22)?
This is a translation of the Greek terms xympasa gnome. In accordance with the
tradition of political realism, G.E.M. de [Link] interprets these words as follows:

If I may offer my own conception of a xympasa gnome for what is undoubtedly

the set of speeches most often accused of having no basis in fact, namely the
Melian Dialogue, I would suggest that it is the Athenians’ assertion that the
Melians are far too weak to be able to resist Athens on their own and can
hope for no assistance from outside, and therefore would be wise to surrender
without further ado, and the Melians’ denial of this assertion.
(de [Link] 1972:10)

The ‘what was called for by each situation’ must therefore be based on
calculations of available force. Consequently, rhetorical moves stem from the
requirements of the power political situation. Or as Keohane puts the idea,
‘Thucydides seeks to go beneath the surface of events to the power realities that
are fundamental to state action.’
The Athenians sent envoys to Melos in 416 BC, at the time of a temporary
peace between Athens and Sparta, in order to persuade the Melians to be
subordinated under the power of the Athenian empire. The Athenians considered
the terms of submission to be moderate and reasonable, namely ‘alliance on a
tribute-paying basis and liberty to enjoy your own property’ (V:111). They
argued, moreover, that it would be in the interests of both parties if Melos
agreed. From this perspective, it seems that the following arguments reveal the
essence of the Melian dialogue:

MELIANS: And how, pray, could it turn out as good for us to serve as for you to rule?
ATHENIANS: Because you would have the advantage of submitting before
suffering the worst, and we should gain by not destroying.
(Thucydides V:92–3)

Here, actors express their preferences about possible states of affairs in these
statements. The only options open to Athens seemed to be either to destroy
Melos or to rule it. It would be much easier and more profitable to rule than to
destroy. For Melos, it would obviously be better to surrender than to be destroyed.
The statement by the Athenians can be read first of all as a statement of facts,
but simultaneously also as an ultimatum. From these, we can derive the strategic
 Modelling Thucydides’ Melos episode 171

Figure 7.1 The strategic decision-making tree of Melos and Athens

game-tree presented in Figure 7.1. (For the notation and general technique, see,
for instance, Ordershook 1986.)
Melos faces a choice between resisting and surrendering, while Athens may,
depending on the choice of Melos, face the choice between conquering Melos
with force and letting Melos be independent. For both parlies the preference
order is (a>b>c). Had Melos had perfect information about the situation, the
only real exit option would have been to ‘surrender’. Melos did not surrender,
however. Consequently, either Melos was irrational, or the model specified is
not a good one. Thus, further complications of the model are needed (given
that the rational-choice approach is not questioned, only the specific model).
There must have been uncertainty as to Athens‘s real capacities and intentions.
It can be assumed that when Melos was forced to make its choice, it did not
know what Athens was going to choose if Melos itself chose resistance. To
indicate this, an information set is drawn into the picture. Melians may have
given a high probability (p) to the Athenians doing nothing if they themselves
just resisted. Consequently, ‘resist’ is a rational choice for Melos if:

(7.1) [p(aM)–(1– p)(cM)]>bM

In other words, if the expected value of ‘resist’ was higher than the value of
‘surrender’, the strategically rational choice would have been to ‘resist’. Because
this actually occurred, surely this must have been the case. The massacre is, therefore,
best explained by an unfortunate misjudgement on the side of the Melians.

Some problems of the rational-choice model

The rational-choice model predicts, after the event, the actual outcome. Is this
an explanation of the Melos episode? It is always possible to construct a
172 Visions of world politics

rational-choice model that predicts any outcome that has already occurred. If
‘prediction’ in this sense is the only criterion, then anything goes—at least as far
as it is articulated in rational-choice terms. In the case of genuine predictions,
the same immunity to criticism can also be effected with a ceteris paribus clausul.7
The question is whether the rational-choice model provides us with a true
explanation. If other hypotheses or evidence is given no chance, if there is no
accuracy and depth, and if there is no illuminatory power, can there be any
real explanation either? Of all possible hypotheses, does the evidence support
only the rational-choice model? What other hypotheses could have been
explicated and assessed? Driven by the scholarly virtue of truth, should not
we test other hypotheses and check the available evidence? Have all relevant
components of the causal complex been brought to the fore; all relevant aspects
of the story been told? Are we satisfied with the depth achieved in the model?
Any episode of social interaction can be re-described as a utility-maximising
decision-making problem of multiple interdependent actors. The first question
is: Is this re-description adequate? What does this re-description explain, if
anything? When rationality is understood as learning and openness to evidence,
how ‘rational’ is the model (7.1) as an explanation? For instance, historical
reasons cast a shadow of doubt on the accuracy of applying probabilistic, utility-
based choice-theoretical terms to explaining episodes in ancient Greece. There
is also direct textual evidence in Peloponnesian Wars that contradicts the rational-
choice model. By a careful analysis of the evidence, it is possible to move
towards deeper layers of social reality and also achieve depth epistemologically.
By specifying and writing a genealogy of the relevant key components, a different
story about the Melos episode can be told.
The conception of strategic rationality was alien to the Greeks. They were
not familiar with the Humean notion of causality or calculative, utility-maximising
action, which in seventeenth-century Europe was modelled on book-keeping
and capital accounting. For ancient Greeks, causes were rather defined in terms
of possibility, potentiality and teleological nature of entities rather than in terms
of empirical invariance.8 Thus, their strategic reasoning must have been different
from the modern economistic cost—benefit analysis.
Moreover, to understand the evidence Thucydides provides about the real
causes of events and processes, it is important to grasp what Thucydides was
doing when he was writing the History. It is not at all clear that Thucydides had
any idea of writing ‘scientific’ history in a modern sense. Although he seems to
have sought, quite methodically, accuracy and true descriptions, and although
he criticised the ancient poets for making up stories, he worked within the
genre of serious arts, particularly tragedy.

[…Thucydides] did not, as is commonly asserted, take a scientific view of

human history. Rather he took the view of one who, having an admirably
scientific temper, lacked the indispensable aid of accumulated and systematic
knowledge, and of the apparatus of scientific conceptions… Thucydides
 Modelling Thucydides’ Melos episode 173

possessed, in common with his contemporaries at Athens, the cast of mind induced
by an early education consisting almost exclusively in the study of poets.
(Cornford 1965 [1907]:ix–x)

Thucydides did not, for instance, make references to his sources or have his book
scrutinised by a scholarly community. He says only that his tried to check different
accounts to the extent possible, but this was difficult because ‘different eyewitnesses
give different accounts of the same events’ (I:22). More importantly, I take Francis
Cornford’s point to indicate that at the level of mimesis2 (see Chapter 5),
Thucydides worked with the poetic narrative structures available to him, given
his education. Moreover, like the authors of great tragedies, it seems that
Thucydides was writing moral philosophy as concrete lessons that he meant to
be drawn from his story, which he claimed to be true (see Farrar 1988:126).

The dialectical logic of the Melian dialogue, part I

Social-scientific iconic models, themselves interpretative, deal with a preinterpreted

world of lay meanings. The research process itself can be understood as a
dialogue between the researcher and the alien object to be grasped—and explained.
A paradigmatic case of object is social action in context. Social action can be
understood in terms of inter-subjective rules, which exist as parts of open, yet
structured, wholes. To a varying extent, rules are known only practically, not
discursively. Communication in general, and dialectical argumentation in
particular, can bring rule-commitments and parts of background discourses to
the explicit level of ‘know-that’. At least a dialogical process tends to make
commitments public.
I agree with Alker (1996:27) that it is illuminating to think ‘in contemporary
language, Thucydides’ polimetrics…as one of seeking to uncover—through
critical, partly formalizable, rational argumentation—the determinative structures,
the characteristic contrasts, the syntheses and transformations practically
constitutive of Greek polities’. Thucydides’ ‘polimetrics’ involves also depth,
since we can think ‘with Thucydides of reality as layered and define scientific
progress as the uncovering of the deeper, dialectically inner, determining layers
or characteristic dispositions’ (Alker 1988a:813).9 By first presenting a long
dialogue between Athenians and Melians, Thucydides reveals the most important
aspects and layers of reality as accurately as the subject matter and his sources
of information permit. (Classical scholars agree that the Melian dialogue was
Thucydides’ dramatic reconstruction, and not a reproduction of eye-witnesses’
reports.) If ‘the deeper, dialectically inner, determining layers or characteristic
dispositions’ have been explicated, it is quite reasonable merely to state, in a
few lines, what actually happened after the debate.
Dialectics is the realm of arguable opinions about contingent and plausible truths
concerning, for instance facts, identities, interests, Jaws of nature, justice or honour.
It is possible to reconstruct the dialectical logic of the Melian dialogue systematically
174 Visions of world politics

and quasi-formally Systematic formalisation can be of help in analysing and

uncovering the deeper layers and interconnections. Following Alker, I use two modes
of formalisation. First, a part of the Melian dialogue will be analysed with the help
of Nicholas Rescher’s formalisation of dialectics. Thucydides’ account of the
arguments used in the Melian dialogue is compatible with dialectical logic. Second,
graphical argumentation analysis permits a systematic analysis of interconnections
between arguments and the deeper layers of the Greek political realities.
Thucydides sets the scene of the Melian episode by explaining briefly its
background:

The Melians are a colony of Lacedaemon that would not submit to the
Athenians like the other islanders, and at first remained neutral and took no
part in the struggle, but afterwards upon the Athenians using violence and
plundering their territory, assumed an attitude of open hostility. Cleomedes,
son of Lycomedes, and Tisias, son of Tisimachus, the generals, encamping
in their territory with the above armament, before doing any harm to their
land, sent envoys to negotiate. These the Melians did not bring before the
people, but made them state the object of their mission to the magistrates
and the few; upon which the Athenian envoys spoke as follows.
(Thucydides V:84)

In Table 7.1, the first seven substantial moves of the Melian dialogue are
reproduced and in some cases summarised. For a brief introduction to Nicholas
Rescher’s (1977) formalised dialectics, see Figure 4.1 and the discussion thereof.
The basic idea is that the proponent takes the initiative and bears the burden of
proof. The opponent has three or four possible countermoves, including cautious
denial, provisoed denial and weak distinction or exception. The Melian dialogue
can be seen as a dialectical disputation—initiated by the Athenians—about whether
it is in the interest of the Melians to submit to the Athenian empire. In Table
7.2, the seven substantial moves of the Melos dialogue are presented formally
by using this notation. Flexible Rescherian rules of rational argumentation seem
to fit to this political debate quite well, even though the survival of the other
party is at stake. However, this formalisation is in itself nothing more than a
systematic reconstruction of the argumentation. Could the dialogue also help to
reveal deeper layers of the reality and truth about the Melos massacre?
The dialogue begins with what a situational determinist would consider as
the essence of the whole episode (moves 1–3). However, this is only the surface
level. The preferences and the prima-facie cases are stated here, but the reasons
for these (dis)positions are not yet given. Moreover, as the debate moves towards
deeper discursive levels, relevant components of the causal complex are disclosed,
such as partially shared practices, relevant actors and the dynamics of their
positioning. The debate also indicates that there are two different, yet equally
constitutive, perspectives about reality, the Athenian and the Melian. Reality is
not the same thing as the Athenian interpretation of it. Rather this reality involves
 Modelling Thucydides’ Melos episode 175

Table 7.1 The first seven moves of the Melian dialogue

at least two different and partially contradictory interpretations (and soon more
differences will be revealed).
Move 4 shifts the debate to qualitatively novel levels. The Melians ask:
‘Why could we not remain neutrals, friends instead of enemies, but allies of
neither side?’ The Athenian envoys answer in move 5: ‘Your hostility cannot
hurt us, but your friendship will be an argument to our subjects of our weakness,
and your enmity of our power.’ Finally, in their move 7, the Athenians make
public a number of still deeper theories and rules:

1 A constitutive rule, which establishes Athenians as rulers: Athenians define and interpret
themselves essentially as sea rulers and seem to think that others do so as well
176 Visions of world politics

Table 7.2 The first seven moves quasi-formalised

(‘we rule the sea and you are islanders’); this establishes an internal relation
between Athenians and all islanders.
2 A ‘realpolitik’ view about the dangers of any signs of Athenian ‘weakness’ as an imperial
ruler, coupled with the claim that morality is irrelevant: [Athens’s subjects think that]
‘those still independent do so because they are strong, and if we fail to attack
them it is because we are afraid’; [others, and subjects in particular] ‘see no
difference between right and wrong’; [it is therefore particularly important
that] ‘you should not escape’.

(1) and (2) explain the Athenian position, as well as their original claim. However,
to explain the Melian position in a similar way, it is necessary to analyse the
Melian dialogue further. Graphical argumentation analysis techniques have a
number of advantages over Rescherian formalisation. First, a graphical tableau
is relatively easy to read, because the natural language expressions are preserved.
Second, graphical techniques make it possible both to analyse the relations of
substantiation, grounding and criticism, and to present them in a visually
illustrative manner. Further, it is also possible to indicate how actors often make
each others’ implicit assumptions explicit (or at least claim to do that). In Figure
7.2, the rest of the Melian dialogue is presented from move 5 onwards.
There is a systematic pattern in this argumentation. Arguments M6, M10 and
M12 are all about equity, courage, honour, bravery and justice. M12 also supports
M10 while simultaneously attacking A11. M10 (‘does not this mean that you are
strengthening your enemies?’), in turn, partially supports MM10 (the Melian
statement that ‘we who are still free would show ourselves great cowards and
weaklings if we failed to face everything that comes rather than submit to
slavery’). This is because M10 indicates that other people, too, can act out of
ethical considerations. This gives reasons to believe that the Athenians are taking
a great risk by acting so aggressively. This claim contradicts the Athenian realpolitik
view about the dangers of being ‘weak’ ((2), above). Evidently, the crux of
Thucydides’ story is that, in the end, the Melians proved to be right. Athens’s
aggressive imperial actions had adverse, counterfinal consequences and
contributed to its final defeat.
Athenians never denied that the Melians were right, in principle, about
matters of justice and honour. Their arguments concerned whether there was
space for questions of right and wrong. ‘You should be sensible, this is not
Figure 7.2 A graphical presentation of the rest of the Melian dialogue
Sources: Thucydides 1900:V:84–116; cf. Alker 1988b.
178 Visions of world politics

about honour’ (A11b); ‘prophecies and oracles lead men to ruin by

encouraging hope’ (A13); ‘of all people that we know the Spartans are most
conspicuous for believing that what they like doing is honourable and what
suits their interest is just’ (A15). Finally, there is the ultimate argument, which
discloses that the Athenian realpolitik theory is not only about the potentialities
and nature of Athens’s subjects. Instead, it tends towards universalist claims
about the nature of the social world:

So far as the favour of gods is concerned, we think we have as much right to that
as you have…. Our opinion of the gods and our knowledge of men lead us to
conclude that it is a general and necessary law of nature to rule whatever one can.
This is not a law that we made ourselves, nor were we first to act upon it when it
was made. We found it already in existence, and we shall leave it to exist forever
among those who come after us. We are merely acting in accordance with it, and we
know that you or anybody else with the same power as ours would be acting in
precisely the same way.
(Thucydides V:105; my emphasis)

The Athenians acted upon a realpolitik theory (2). Now we can see that this theory
included naturalisation (reification) of social worlds: ‘it is a general and necessary
law of nature to rule whatever one can’. This alleged law is by no means Newtonian.
It is nonetheless necessitarian and used explicitly to justify actions. The Athenians
think that this ‘natural law’ makes their actions right, also in the eyes of gods.10
The Athenians’ realpolitik theory is a set of rules of interpretation and judgement
that they followed. At that time, it was a deep element of the Greek social reality.
Whereas the Melians were making ethical and moral arguments, the Athenians
tried to persuade them that this was not sensible, that there is not even a slightest
sign of hope that these considerations would help them. However, the Melians,
like many others, never accepted this claim. The Melians were ‘not prepared to
give up in a short moment the liberty which our city has enjoyed from its
foundation for 700 years’ (Thucydides V:112).
In the light of this analysis of evidence, what were the causes of the occupation and
massacre? Two at least tentatively plausible counterfactual possibilities can be
constructed:

CF1 If the Melians had accepted the Athenian interpretation of

the situation and the Athenian horizon of action, they would
have submitted (for they were rational).
CF2 If the Athenians had accepted the relevance of justice, equity,
courage, honour and bravery, as the Melians asked them to
do, they would have been forced to admit the illegitimacy,
injustice and potential insensibility of their expansionist
imperialism. Consequently they would have been compelled
to change some of their self-understandings concerning their
identity, aims and horizon of possibilities.
 Modelling Thucydides’ Melos episode 179

CF1 is the onlys counterfactual possibility that can be seen from the perspective
of political realism and the rational-choice model. Political realism is compatible
with the Athenian view. Political realism sees it as the only possibility and
therefore contributes to legitimising, long after the event, the Athenian imperialism
and acts of violence. Many have seen the second counterfactual (CF2) as more
important, either for prudential or moral reasons. However, there are multiple
contradictory perspectives in reality, and all of them are real and also products
of particular geo-historical processes. It is my argument that from a causal
explanation point of view, CF2 is decisive.
According to Thucydides’ narrative about the Peloponnesian war, the Athenians
acted quite differently earlier in the war. Pericles—the famous leader of the golden time
of the Athenian rule—warned, for reasons of honour and prudence, against adopting
a policy of expansionist imperialism and aggression. Hence, a different kind of
Athenian policy had already been actual, not only a possibility. Moreover, even at the
time of the Melos episode, in 416 BC, Athenian imperial expansionism was contested
within Athens.

The prophasis of the Melos episode, part I

Thucydides distinguished between two types of causes, between aitia and prophasis.
Aitia refers to the rationalisation of action, and prophasis to the underlying causes,
to the causes which are ‘behind’ or ‘under’ the level of rationalisations and
concrete events (see, for instance, Edmunds 1975:172–3). For political realists,
this means that ‘Thucydides seeks to go beneath the surface of events to the
power realities that are fundamental to state action’ (as Keohane put it). However,
there are other possibilities. For a critical realist, there are a number of things in
social reality that do not belong to the level of aitia, without there being any
fixed set of empirical invariances, which would determine the aitia.
From a CR perspective, the ‘going beyond’ of rationalisation can be done at
least in three directions. The first possibility is to move deeper into the discursive
formations and meanings, possibly towards explicating the implied internal
relations between social entities, possibly by moving analytically towards
unconscious levels (e.g. by analysing the mythical structuration of meanings,
perhaps in relation to the interplay between partially unintended consequences
of action and real reasons for a particular rationalisation). The second possibility
is to explain the concrete possibilities open to positioned actors—particularly the
possibilities taken seriously by the actors. These can be analysed in terms of the
relevant rule-constituted and rule-governed contexts of positioned practices and
by writing genealogies of the key elements of these contexts. The third possibility
is to analyse causal or existential, often unintended, consequences of action,
such as effects of power; the—often coincidental—timing of the occurrence of
these consequences; as well as the external relations of systemic
(inter)dependence, which may give rise to emergent powers and properties, and
which are (re)produced in the context of unintended consequences of action.
180 Visions of world politics

To contextualise adequately the counterfactual possibilities (CF1 and 2), the

next step is to move from aitia to prophasis by constructing an iconic model of
the material and efficient causes of the Melos episode. We have to specify a
model M = {A; R}, in which A=a1, a2, a3>,…is a set of elements, which themselves
can be relations or relational systems, and R=r1, r2, r3,…is the set of the relations
between these elements. One way to do systematic iconic modelling is simply
to explicate the causally effective components at a particular geo-historical
juncture. The problem with this procedure is that it freezes history. This deficiency
must subsequently be corrected with some countermeasures. Table 7.3
summarises a few hypotheses about some of the most relevant components.
These hypotheses stem from my analysis of the evidence provided by Thucydides
in the Melian dialogue in particular, but also, more generally, in the History as a
whole. They are open to criticism and revisions.11
The Athenians’ mode of action is a combination of the expressive and the
strategic. In the debate, the Athenians express themselves proudly as ‘sea rulers’
and as the imperial power. They acted strategically towards Melos. There is,
however, also the shared rule that Hellenes share a cultural and political identity
(WH), in contrast to barbarians, living outside Hellas. Thucydides even speaks
of ‘the country now called Hellas’ (Thucydides I:2).12 It is very ambiguous
whether the Peloponnesian war was, in modern terms, a civil or an international
war. It also involved strife between social classes: land-ownership-based oligarchy
versus trade-and-seafaring-based democracy of free, propertied men.
Many, including Thucydides himself and the ruthless Athenian leader
Alcibiades, had to flee from their polis in the course of the war, because of
violence between parties, and because of various suspicions and allegations.
Hellas was not a prototype of the modern European states system. Not only was
there no balance of power system, there was not even any formal recognition of
poleis as collective actors (Sylvan 1991a:7). Even the term ‘Athens’, as an
indication of a collective actor, as it is used in Table 7.3, may be misleading. In
Thucydides, polis is not usually referred to as a unitary actor. It was the citizens
of poleis that conducted actions, made treaties and could speak of ‘us’.
Exclusion from the citizenry was a major punishment, and could sometimes even
lead to violent enslavement (as emphasised by Sylvan 1991a). However, it was usually
possible, at least for the wealthy, to travel freely (unarmed) and, if necessary, settle
down in another polis, and, quite often, acquire a new citizenship.14 There were also
shared practices. Thucydides describes the development of ‘civilised customs’ in
‘modern’ Hellas:

It was the Spartans who first began to dress simply and in accordance with
our modern taste, with the rich leading a life that was as much as possible
like the life of the ordinary people. They, too, were the first to play games
naked, to take off their clothes openly, and to rub themselves down with
olive oil after their exercise. In ancient times even at the Olympic Games the
athletes used to wear coverings for their loins, and indeed this practice was
 Modelling Thucydides’ Melos episode 181

Table 7.3 Relevant causal components of the Melos episode

still in existence not very many years ago…. Indeed, one could point to a
number of other instances where the manners of the ancient Hellenic world
are very similar to the manners of foreigners today.
(Thucydides I:6)

The Olympic Games were an institution that brought the whole Hellenic world
regularly together and reinforced shared assumptions of Hellenic identity and
common ‘modern’ manners. From the Trojan War onwards, Hellas had also
been united many times against external enemies. Besides external enemies,
Hellenic co-operation was based on the partially shared linguistic, cultural and
political identity (e.g. Thucydides I:23 distinguishes between Hellenic and foreign
armies). Very importantly, the Hellenic unity was also based on pacification of
land and sea, which enabled trade and accumulation of wealth (Thucydides
I:4–9). According to Thucydides’ genealogy, ‘as Hellas became more powerful
182 Visions of world politics

and as the importance of acquiring money became more and more evident,
tyrannies were established in nearly all the cities, revenues increased, shipbuilding
flourished, and ambition turned to sea-power’ (Thucydides I:13). After the
Spartans had ‘put down tyranny in the rest of Greece’, the final obstacle to the
rise of Hellas was removed (Thucydides I:18). Soon after the Persian wars,
however, Hellas was divided into two competing camps:

‘Not many years after the end of tyrannies in Hellas the battle of Marathon
was fought between the Persians and the Athenians. Ten years later the foreign
enemy returned with his vast armada for the conquest of Hellas, and at this moment
of peril the Spartans, since they were the leading power, were in command of
the allied Hellenic forces. In face of the invasion the Athenians decided to abandon
their city; they broke up their homes, took to their ships, and became a
people of sailors. It was by a common effort that the foreign invasion was
repelled; but not longer afterwards the Hellenes…split into two divisions, one
group following Athens and the other Sparta.
(Thucydides I:18; my emphasis)

By the time of the Melian dialogue, the Athenians were losing the legitimacy of
their leadership. They had originally acquired this leadership in the Persian
wars. Ethical notions (mutually agreed regulative rules) played a decreasing
role. The Athenians attempted to replace legitimate leadership with brute force.
Party strife and inter-polis violence were thus making the shared practices and
common Greek identity less important in determining the course of actions.
However, the Nician peace treaty between Athens and Sparta, which was
beginning to break down in 416 BC, included rules regulating aggression between
the camps. The most important constitutive rule, depicted in Table 7.3, was the
self-definition of Athens as the sea ruler. Trade and imperial fees gave Athens
many resources; it was considered the greatest Hellenic power economically and
militarily. Moreover, at this moment, it was neither engaged in war with Sparta
(the Lacedaemonians), nor was its effective transformative capacity yet being
weakened due to the growing lack of legitimacy.
In Melos, the Athenian envoys represented Athenian citizenry. The constitutive
rule they acted upon was ‘we are Athenians’. The decision to adopt a policy
towards Melos was made in Athens. The envoys were normatively tied to the
given instructions. Nevertheless, they must have had an amount of autonomy,
also because of long communication distances. They were also committed to
the rules of argumentation. Debates, which in most poleis were usually public,
organised political life throughout Hellas. However, Melians pointed out that
the Athenian envoys may not been following the rules of dialogue:

MELIANS: No one can object to each of us putting forward our own views in a
calm atmosphere. That is perfectly reasonable. What is scarcely consistent
with such a proposal is the present threat, indeed, the certainty of you making
 Modelling Thucydides’ Melos episode 183

war on us. We see that you have come prepared to judge the argument
yourselves, and that the likely end of it will be either war, if we prove that we
are in the right, and so refuse to surrender, or else slavery.
ATHENIANS: If you are going to spend the time in enumerating your suspicions
about the future, or if you have met here for any other reason except to look
the facts in the face and on the basis of these facts to consider how you can
save your city from destruction, there is no point in our going on with the
discussion. If, however, you will do as we suggest, then we will speak on.
(Thucydides V:85–6; my emphasis)

Melians relied on the logic of normative and expressive action. They made points
about the requirements of equity and justice. However, Athenians were not
interested in these requirements; Melos was just one of the small islands with a
scarce population, oligarchic governance and limited military facilities. Melians
began to make expressive arguments: they are free and independent; and if they
have any sense of honour and bravery they are not going to submit, for it would
change the very nature of what they are. No matter what the consequences, they
wanted to stay independent, with great honour. In Hellas, the notion of honour
was particularly important (see Donnelly 1991: especially 32–7). This makes it
understandable why the Athenians were compelled to answer to the Melians in a
manner that implied the acceptance of the relevance of honour: ‘You will see that
there is nothing disgraceful in giving way to the greatest city in Hellas when she
is offering you such reasonable terms’. The Melians did not agree.
In the last instance, when a reference to heaven and fortune had proved futile,
the Melians pointed out that Sparta (the Lacedaemonians) might come to help
them. Melos did not belong to the Lacedaemonian alliance, and the Athenian army
was already encamped on the soil of Melos. The Spartans had the reputation of
acting slowly and cautiously, and their actions were restricted by the normative
requirements of the Nician Peace. The Athenian argument that ‘where danger is
concerned the Spartans are not, as a rule, very venturesome’ must have thus seemed
quite plausible. In ancient strategic terms, it is unlikely that the Melians made any
misjudgement about their hopeless situation (cf. the rational-choice model of Figure
7.1). It was for reasons of identity-constituting independence, courage and honour
that they refused to submit to the rule of the Athenian empire.

The prophasis of the Melian episode, part II:

the genealogy of Athenian aggressive imperialism

The Athenians initiated the Melian episode and also committed the massacre and
enslavement. Causally, the Athenians were responsible for the outcome. Although
the Melians could have acted differently, from a causal explanation point of view,
it is the aggressive Athenian imperialism that should be explained. In accordance
with CF2, it is possible to isolate from the causal complex the Athenian
aggressiveness as a particularly relevant INUS-component, and write its genealogy.
184 Visions of world politics

In contrast to the realpolitik justification of imperialism presented by the

Athenians in the Melian dialogue ((2), above), in the beginning of the war,
Pericles argued for an Athenian empire whose laws are just and should thus
appear legitimate—and whose actions stand up to the standards of honour. For
Pericles, justice and honour were both values in themselves and necessary means
for keeping the empire together. In a speech where he defended the decision to
make war against Sparta, Pericles stated: ‘What I fear is not the enemy’s strategy,
but our own mistakes’ (Thucydides I:144). Thucydides explains Pericles’ position
as follows:

For Pericles had said that Athens would be victorious if she bided her time
and took care of her navy, if she avoided trying to add to the empire during the course
of war, and if she did nothing to risk the safety of the city itself. But his
successors did exactly the opposite, and in other matters which apparently
had no connection with the war, private ambition and private profit led to policies
which were bad both for the Athenians themselves and for their allies. Such
policies, when successful, only brought credit and advantage to individuals,
and when they failed, the whole war potential of the state was impaired.
(Thucydides II:65; my emphasis)

Had Athens continued to ‘avoid trying to add to the empire’, or to observe the
principles or justice and honour, no Melos episode could have occurred. Why
did the policy of Athens change? There must be two aspects to this explanation.
First, at the collective level, the adoption of a particular policy can be explained
by giving an account of the decisions in terms of debates and voting in Athens.
These were conducted in accordance with certain agreed procedures. Second, a
change in the structure of the ‘seriously taken’ discourse can be explained in
terms of learning, power and interests in a changing context of action.
In the fourth year of the war, the Peloponnesians (Sparta and confederates)
invaded Attica (Athens and its environment). Immediately after this invasion, all
of Lesbos (except Methymna), led by the Mitylenes, revolted against the Athenians
and tried to ally with the Peloponnesians. After long and bitter military struggles,
the Mitylenes finally surrendered. ‘In the fury of the moment’, as Thucydides
puts it, the Athenians were ‘determined to put to death not only the prisoners at
Athens, but the whole adult male population of Mitylene, and to make slaves of
the women and children’ (Thucydides III:36). However, the next day ‘brought
repentance with it and reflection on the horrid cruelty of a decree, which
condemned a whole city to the fate merited only by the guilty’ (Thucydides
III:36). The case was re-opened for debate and vote. Cleon spoke for the original
decision, arguing in a manner similar to that of Athenian envoys in Melos:

Fears or plots being unknown to you in your daily relations with each other,
you feel just the same with regard to your allies, and never reflect that the
 Modelling Thucydides’ Melos episode 185

mistakes into which you may be led by listening to their appeals, or by

giving way to your own compassion, are full of danger to yourselves, and
bring you no thanks for your weakness from your allies; entirely forgetting
that your empire is a despotism and your subjects disaffected conspirators,
whose obedience is ensured not by your suicidal concessions, but by the
superiority given you by your own strength and not their loyalty…. I proceed to show
that no one state has ever injured you as much as Mitylene. I can make
allowance for those who revolt because they cannot bear our empire, or who
have been forced to do so by our enemy. But for those who possessed an
island with fortifications;…who were independent and held in the highest
honour by you—to act as these have done, this is not revolt—revolt implies
oppression; it is deliberate and wanton aggression…. Punish them as they
deserve, and teach your other allies by a striking example that the penalty of
rebellion is death.
(Thucydides III:37; my emphasis)

Cleon’s interpretation of the Athenian situation and possibilities was not accepted,
at least not by the majority of citizens. Diodotus, representing a Periclean point
of view, argued that the ‘two things most opposed to good counsel are haste
and passion; haste usually goes hand in hand with folly, passion with coarseness
and narrowness of mind’ (Thucydides III:42). Diodotus also protested against
using fear and intimidation as rhetorical devices. Instead, in his view, fair and
calm arguments were needed. Diodotus shared the opinion that the Mitylenians
were guilty of an unjust revolt. He argued, however, that the Cleonian policy
would be catastrophic in the long run. Many citizens accepted this argument.
In a nearly equal vote, the Athenians decided to save the majority of Mitylenians.
During the years following the Mitylenian revolt, the Cleonian perspective
became predominant in Athens. Athens as the ‘sea ruler’ was becoming increasingly
expansionist. A quick look at Cleon’s arguments can help to explain how and
why this occurred. In Cleon’s view, the world outside the city of Athens was full
of ‘fears’ and ‘plots’. All concessions and compassions were ‘full of danger’ to
Athenians. All Athens’s subjects were ‘disaffected conspirators’. Even partially
independent states like Mitylene were potential aggressors and allies of the enemy,
the Peloponnesians. Thus, Cleon concluded, ‘obedience is ensured by force and
fear, not by loyalty’. Gradually, the Athenians came to believe that Cleon was
right. It seems that continuous strife, fear and violence made Cleon’s points more
plausible. According to Thucydides, ambitious individual leaders—or leader-
candidates—also used these fears as a rhetorical (populist) power resource. The
result was the ‘realpotititk’ theory ((2), above), according to which ‘right is only in
question between equals in power, while the strong do what they can and the
weak suffer what they must’ (Thucydides V:89).
Power was at the core of this change. In its more narrow sense, power as
domination can be taken to mean, as Karl Deutsch (1963) does, that one does
not have to give in, because one has facilities and capabilities to force others to
186 Visions of world politics

comply. Power in this narrow sense is the priority of output over intake, the
ability to talk instead of listening. In a sense, it is also the ability to afford not
to learn and change what one is. ‘When carried to extremes, such narrow
power becomes blind, and the person or organization becomes insensitive to
the present, and is driven, like bullet or tornado, wholly by its past’ (Deutsch
1963:111). This kind of actor has a will in a sense that comes quite near to the
Nietzschean sense of a will-to-power.
The experiences of the long war and the plague caused the Athenians,
metaphorically speaking, to auto-program themselves to adopt the will of a ‘sea
power’, and to attempt to have an ability to speak and enforce their will but not
to listen. This was also because they did not trust others (including their subjects)
any more. According to this Cleonian perspective, if weaker parties are not
willing to listen to and to obey, it is necessary not only to force them to do so
but also to ‘teach others by a striking example that the penalty of rebellion [and
disobedience] is death’. This was the logic of the aggressive Athenian imperialism,
which eventually proved counterproductive and led to the defeat of Athens.

The prophasis of the Melian episode, part III:

the origin and ‘necessity’ of Athenian imperialism

Thucydides thus tells a story, according to which the Melos massacre was not
necessary. With a wiser, Periclean policy Athens could have refrained from
imperialist expansionism as well from the policy of teaching lessons by means
of the threat of death to others (CF2). However, there are also traits of
necessitarianism in Thucydides’ History. In one of the most contested passages
of the History, Thucydides argues that ‘the real cause of the [Peloponnesian war
was…[t]he growth of the power of Athens, and the alarm which this inspired in
Lacedaemon’. These, he argues, ‘made war inevitable’ (Thucydides I:23).
Thucydides himself was an Athenian general exiled after his strategic failure
in 424 BC, in the seventh year of the war. Thucydides was committed to the
Athenian position of the early war and he makes it quite clear that he agreed
with the policy line of Pericles. Afterwards he was in a position of a critical,
cosmopolitan outsider. He also communicated with the representatives of the
former enemy. For Thucydides, the Athenian policy possibilities ranged from
the prudent statemenship of Pericles to the aggressive excesses of Cleon and
others. (Roughly the same set of policy possibilities existed on the Spartan
side.) If this horizon of possibilities is combined with Thucydides’ proto-
sociological theory of the growth of the population, trade, navy and alliance
revenues of Athens,15 the war really was ‘inevitable’ because no counterfactuals
can be seen from this horizon.
As Thucydides describes it, the Peloponnesians were worried about the growth
of Athenian power. Many believed that most of Hellas would sympathise with
those trying to dissolve the Athenian empire, which had developed out of the
Delian League (an alliance against Persia). Before the final outbreak of the war,
 Modelling Thucydides’ Melos episode 187

‘both the Athenians and the Peloponnesians had already grounds of complaint
against each other’ (Thucydides I:66). In this situation, there were debates in
both camps about what to do. In Sparta, most shared the view that Athens was
already acting aggressively and oppressing large parts of Hellas. Therefore, war
should be declared without delay. Even the Athenians present in Sparta at that
time did not deny that there was some truth in these claims:

We did not gain this empire by force. It came to us at a time when you were
willing to fight on to the end against the Persians. At this time our allies came
to us of their own accord and begged us to lead them. It was the actual course
of events which first compelled us to increase our power to its present extent….
So it is with us. We have done nothing extraordinary, nothing contrary to
human nature in accepting an empire when it was offered to us and then in
refusing to give it up. Three very powerful motives prevent us from doing so—
security, honour, and self-interest. And we were not the first to act this way. Far
from it. It has always been a rule that the weak should be subject to the strong;
and, besides, we consider that we are worthy of our power.
(Thucydides I:75–6)

The main arguments of these Athenians were that they have done nothing
‘contrary to human nature’, and that they ‘are worthy of their power’. In Sparta,
the viewpoint of the King Archimadus prevailed. Accordingly, Sparta sent a
mission to Athens to complain about the grievances and continued war
preparations but did not (yet) declare war. It is noteworthy that Thucydides
does not present a full debate with different points of view about Athens’s
decision. Thucydides admits that ‘many speakers came forward and opinions
were expressed on both sides, some maintaining that war was necessary and others
saying that the Megarian degree should be revoked and should not be allowed
to stand in the way of peace’ (I:139; my emphasis). Yet, only the speech of
Pericles can be found in the History. In his speech, Pericles supported the policy
of ‘no concessions to Lacedaemonians’:

Athenians, my views are the same as ever: I am against making any

concessions to the Peloponnesians…. It was evident before that Sparta was
plotting against us, and now it is even more evident. It is laid down in the
treaty that differences between us should be settled by arbitration, and that,
pending arbitration, each side should keep what it has. The Spartans have
never once asked for arbitration, nor have they accepted our offers to submit
to it. They prefer to settle their complaints by war rather than by peaceful
negotiations, and now they come here not even making protests, but trying
to give us orders…. [I]f we do go to war, let there be no kind of suspicion in
your hearts that the war was over a small matter…. If you give in, you will
immediately be confronted with some greater demand, since they will think
188 Visions of world politics

that you only gave in on this point through fear. But if you take a firm stand
you will make it clear to them that they have to treat you properly as equals.
(Thucydides I:140)

If Thucydides maintained, as it seems, that the Periclean line of thinking and

policy was correct, there was no reason to report other opinions on this occasion.
Yet Pericles’ claims about Sparta do not cohere with the evidence that Thucydides
gives about the reflections and discussions in Sparta. It seems that Sparta was
made to respond to the grievances and claims of its allies, although it was very
cautious about taking any action against Athens. The war was inevitable only
given the insistence of Pericles that Athens should keep the empire and make
no concessions, although its actions were giving rise to grievances.
An interest to preserve the empire had been established in Athens. The
grandeur and luxury of many of the wealthy Athenian citizens had become
dependent on the fees of the subjects of Athens—besides exploitation of slaves,
women and weaker free men, such as the rowers. However, although deeper
emancipatory possibilities were inconceivable to an ancient poetic historian,
Thucydides could have considered the line of those in Athens who argued that
‘the Megarian degree should be revoked and should not be allowed to stand in
the way of peace’.

Moral, tragic and emancipatory aspects of the History

In ancient Hellas, it was common to think that some individuals and some
communities, by their moral qualities, were entitled to positions of leadership
and power. However, power is dangerous and corrupting, and in the wrong
hands it quickly leads to immoral behaviour, and then to civil strife, unjust war
and destruction. The development of Athens, as re-configured by Thucydides,
can be seen in these poetic terms.
There are also elements of tragedy in his text. For instance, the tragedy of
the Melos episode is that both the Athenians and the Melians were right. In the
short run, the Athenians were right: the Melians had no hope unless they
submitted, because ‘the strong do what they have power to do and the weak
suffer what they must’. Nonetheless, Melos did not submit, and they suffered.
In the long run, the Melians were right: by trying to replace moral legitimacy
with brute force, the Athenians eventually turned most of Hellas (and later also
Persia) against them. The result was the total defeat of Athens. This spectre of
suffering notwithstanding, Athens was set on a policy that was expansionist,
aggressive and harsh.
Speeches, dialogues and debates occupy a central place in the History, as in a
theatre drama. Any adequate reading of Thucydides should take into account
the drama-like structure and moralism of the History. In his Poetics, Aristotle
stated that every tragedy consists of a problem, which is presented in the
 Modelling Thucydides’ Melos episode 189

beginning of the spectacle, and the rest of the drama is about the settlement of
the problem (Aristoteles 1982 [335–22 BC]: section 18). In Thucydides’ work,
the ‘Great War’ is the problem, set out in the beginning. The rest is about the
settlement of this problem. In the History, there are a number of turnings
(peripeteias) of events, perhaps the most important of them being the internal
development of Athens and its consequences. In Aristotle’s view, tragedy is an
imitation of worthy action. Tragedy does not imitate by telling the narrative in
the voice of a third person but through the settings of the people active in the
tragedy There are two ‘natural causes’ of action: the ways of thinking and the
characters of actors, which are both expressed in action. Furthermore, Aristotle
equates thinking with the ‘capability of expressing with words that which is
possible and is called by the situation’ (or occasion). This capability belongs to
the sphere of politics and rhetoric (Aristoteles 1982 [335–22 BC]: section 6).
From this perspective, Thucydides’ method becomes quite understandable: ‘My
habit has been to make the speakers say what was in my opinion demanded of
them by the various occasions, of course adhering as closely as possible to the
general sense of what they really said.’ Thucydides aimed at accuracy in his
descriptions, but he was forced to reconstruct the speeches in accordance with his
knowledge of the characters of the participants. He was also structuring—particularly
during the final decade of the war—his narrative as a tragic drama, thereby trying
to make sense of the whole. The temporal structure of a moralist strategy can be
consonant with writing CR genealogies of causal components and with the scepticism
about closed schemes of history. The History also provides a narrative that fulfils all
three stages of mimesis, it aims at iconic augmentation of future practices:

It will be enough for me, however, if these words of mine are judged useful
by those who want to understand clearly the events which happened in the
past and which (human nature being what it is) will, at some time or other
and in much the same ways, be repeated in the future. My work is not a
piece of writing designed to meet the taste of an immediate public, but was
done to last for ever.
(Thucydides I:22)

Mistakes may be repeated, processes may resemble those of the Athenian

development and a somewhat similar tragedy may re-occur even in the twenty-
first century The forming of an iconic model always involves abstractions and
idealisations, and thus a model is necessarily more general than the concrete
context that is its source. There may be many historical episodes, practices or
systems, which can be reasonably identified with a model that is in some regards
similar to (say my reconstruction) of the History and the Melos episode in particular.
Violence, pathological learning and exploitative domination are possibilities open to
humanity. In this sense, Thucydides’ work has lasting value ‘for ever’.
It does not make sense to criticise ancient Athenian practices of
domination, more than 2,400 years after the Melos massacre. Past practices
190 Visions of world politics

and relations cannot be changed and nobody can now be prosecuted.

Criticism should be pointed towards practices that are based on theories,
which function similarly to the Athenian ‘realpolitik’ view, and which claim that
weakness is dangerous and morality is irrelevant. I mean many political realist
and rational-choice interpretations of world politics. It is rather alarming that
the quality of being ‘scientific’ can be equated with safeguarding oneself from
external criticism. This kind of closed-mindedness can constitute and
legitimate, in concrete contemporary contexts, tragic acts of aggression,
oppression and imperialism.16

Conclusion

To stop the process of iconic modelling of X is always risky. There may be a

new fruitful perspective, a fresh argument or a previously unknown or unnoticed
piece of evidence waiting just around the corner. Moreover, modelling can always
be turned towards different aspects of reality. New genealogies of other
components of a complex can be written, and they may stimulate the
development of rather different stories about the same X.
Nonetheless, at some point a judgement must be made whether a model is true.
Truth refers, of course, not only to the judgements concerning actual statements
and models, but also to how the world really is and thereby to potential evidence,
conceptions and theories, which can change our present views and judgements. So
it is also with my model of Thucydides’ Melos episode. This model explains in
detail the coming about of the Melos massacre and enslavement. The Athenian
citizens had concluded that, instead of moral leadership, their empire can be based
only on fear and brute force. Against the earlier policy of Pericles, they had also
decided to further enlarge their empire. This pathological learning process, which
ultimately led to the total defeat of Athens, was a partial consequence of war
experiences, which made it possible for ambitious leaders to inflame fear, suspicion
and a sense of archaic revenge also for their own benefit. It presupposed, however,
the violent and highly exploitative socio-economic system of Hellas.
Hume and Kant and their followers have mis-read Thucydides to support
their own ontological assumptions and ahistorical categories. Thucydides was
not the first positivist historian, and neither did he say anything about sovereign
states or balance of power. However, Thucydides was not a proto-critical realist
either. The point of this chapter is not to argue for a Thucydidean approach to
world politics. Rather I have tried to exemplify CR methodology and
demonstrate the historical specificity of the modern international problematic.
Having said this, however, there is a sense in which the value of the History
may ‘last for ever’. Violence, pathological learning, illegitimate domination and
exploitation are possibilities always open to humanity. We can learn not only
from the content of Thucydides’ noble tragedy, but also from his simultaneous
quest for truth, relevance and composition of an outstanding story, from which
the readers can draw ethico-political lessons.
 Modelling Thucydides’ Melos episode 191

Notes

1 When I refer to this text, I stick to the standard way of first giving the number of the
‘book’ (or part) of the text (there are eight of them altogether), and then the section.
Because of my inability to read Greek, I have had to content myself with English
translations. I have used particularly the translations by Richard Crawley (1952) and
Rex Warner (1954), but in some points I have compared these translations to the first
English translation by Thomas Hobbes (1628) and to the Finnish translation by J.A.
Hollo (1964), which have enriched my reading of many of the key passages.
2 Within classical scholarship there have been extensive debates about different
interpretations of Thucydides. The naïve rationalist interpretation, according to which
Thucydides followed the canons of Newtonian science and by careful observations
determined laws of history that, in turn, would make possible predictions in the
contemporary world, prevailed in the USA during the 1950s and 1960s, but has been
since then challenged and rejected (see Connor 1985:3–19). Alker was the first
international relations (IR) theorist to become interested in critical readings of
Thucydides. His paper was first presented in a conference in 1980 and finally published
in 1988 in the American Political Science Review, it was re-published, with amendments, in
Alker 1996. Others became interested in re-readings of Thucydides soon after Alker.
See in particular Garst (1989:3–27); Donnelly (1991); Bagby (1994); Kokaz (2001);
and Bedford and Workman (2001). It is perhaps worth noting that Machiavelli has
also become an object of similar reinterpretations by Walker (1993:26–49).
3 Thucydides is not the only possible source of evidence, but clearly without Thucydides’
readily available account, our knowledge about the Peloponnesian wars would be
very scant. Classical scholarship has nonetheless been able to piece together, for
instance, a picture of the intellectual setting of the Peloponnessian Wars and the class
structure of Athens at the time of the wars. Mostly, I will confine myself to the account
of Thucydides only.
4 Kant does not give his source; the relevant passage of Hume’s ‘On the populousness
of ancient nations’ is cited in Connor (1985:20).
5 For a similarly fetishist neo-realist statement on Thucydides, see Gilpin (1984:287–
304).
6 It is quite alarming that rational-choice theorists have the confidence to declare that
their aim is to avoid external criticism! However, it is in fact possible, up to an extent,
to relax the aim of finding a unique ‘exit’ without moving out of the situational
determinist programme. Is this relevant? In IR most discursive practices are simplified;
most sophisticated applications of rational choice are typically only of highly specific
academic interest. Moreover, although it is possible to relax one assumption, or a few
assumptions, in time, and then to develop a new strand of research around those
qualifications; and although new research can be opened towards different directions;
this does not, in general, change the core assumptions. The core assumptions unite the
different strands, and they are also taught in the basic courses and textbooks. Critical
debates about a research programme should be concerned with the core assumptions,
not so much with the peculiarities of a particular strand of the large and well-funded
programme. Besides, most modification of the assumptions of rational-choice theory
are modest in most respects. They tend to respond only to internal criticisms, not to
external ones. Even when the possibility of legitimate external criticism is admitted, it
is not taken seriously.
7 If the model fails to predict, as it usually does, it is assumed that the problem lies in
the false ceteris paribus assumptions. If it is, quite surprisingly, able to predict, it is seen
as correct, even though there could have been many other, different models capable
of doing the same thing.
8 At least, this was the view of Aristotle, and in any case the Galileo-Copernicus-Bacon
192 Visions of world politics

view was very revolutionary in its own time. No such accounts are known to have
existed before. About the Aristotelian notion of causality, see any philosophy textbook.
9 The 1996 version does not include this important quote; it seems to have been a
victim of length-control in the context of the tendency towards ever-expanding richness
of quotes and exploration of possible research paths, which has become ever more
evident in the 1996 version.
10 This English translation fails to capture the argument that the gods rule, according to
the Athenians, whenever they can. Also, gods are hierarchical rulers (masters), and
thus would not condemn the Athenians. See Thukydides (1964:V:105), the second
sentence).
11 Referring to an earlier version of this iconic model, Alker (1996:44) talks about
Patomäki’s ‘most ambitious misspecification [which] treats the entire Melos episode as
iconically modelled history’ [my emphasis]. I hope this version is a more adequate
specification of the major components of the relevant causal complex.
12 The term ‘country’ does not appear in the Finnish translation; it only speaks of Hellas;
see Thukydides (1995 [1964]:I:2, the first sentence).
13 There are hermeneutic risks of misunderstanding in applying these concepts to the
ancient Greek world (even if the mode of teleological/strategic action is partially re-
interpreted to fit with the ancient Greek contexts). Nevertheless, it seems that no
great injustice is done to the meanings expressed in the Melian dialogue if they are
reconstructed along these lines. On the other hand, this sense may well be due to my
inability to imagine alternative frameworks of interpretation.
14 This is how Athens developed in the first place, according to Thucydides (I:2). War
and disturbances drove people to Attica, which, as a land of poor soil, was relatively
free of violence and robbery. Immigrants could become citizens in Athens. Much
later, during the Peloponnesian war, Thucydides himself had to go in exile from
Athens in 424 or 423, but he never lost his revenues from the silver and gold mines in
Trachia. Alcibiades fled to Sparta in 414 and immediately assumed an important
position there. Although death and enslavement may have been considered as threats
to those outside their poleis, as David Sylvan (1991a) maintains, many well-endowed
Hellenes could in fact travel freely and often assume a new citizenship without great
difficulties, even during the war (see, for example, Thucydides V:4).
15 For Thucydides’ account of this, see (I:6–19; I:89–117). It is interesting that this part
of the narrative could be easily interpreted in terms of the lateral-pressure theory.
This theory is presented in Choucri and North (1975), and Ashley (1980). For a
general account, largely based on Thucydides, of the development of the Athenian
empire, see Meiggs (1972).
16 Compare this to the imperialistic logic of game theory, according to which—and from
the point of view of an iterated ‘prisoner’s dilemma’ in particular—there is a universal
need to make commitments and threats credible and to establish a ‘hard reputation’
by realising the threats every once in awhile. This may, then, require imperialist
operations such as the Vietnam War. See, for example, Axelrod 1984:154.
8 A global security community

International relations (IR) have been conventionally defined in contrast to the

order of modern political community, the nation-state (see Wight 1966; cf.
Ashley 1987:413). Inside, there is a peaceful social order, perhaps a good way
of life and shared security, or welfare, or justice or democracy—or at least the
possibility of progress towards achieving them. Outside, however, a state of
nature or war prevails. As long as there is no social contract, and a resulting
state (or something equivalent), the international sphere must be a struggle for
power, peace and survival. Put schematically:

Arg. (a) (Pr. 1) There are sovereign nation-states that have an

administrative staff and a legitimate monopoly over violence
in their own territory.

(Pr. 2) There is no world-state in the sense of (Pr. 1).

(Conclusion) Anarchical struggle for power and peace—and

the related ‘security dilemma’—prevails in the sphere of
international politics.

This view has been reinforced by the three world wars—including the Cold
War—that spanned the period from 1914 to 1989. International war, actual and
potential, occupied a central role in the world history of the twentieth century.
Yet, empirical evidence of large-scale violence and actual wars, particularly since
the Second World War, does not seem to support either (Pr. 1) or the conclusion.
Most political violence, and most wars, have concerned the ‘inside’ of states
rather than been wars between sovereign states. International wars have in fact
been relatively rare. Only about a tenth of armed conflicts have been inter-state
wars (see Alker 1989:21–4; Rummel 1995; Wallensteen and Sollenberg 2000).
Neither is a peaceful system associated exclusively with the confines of the
more stable states. The ‘security dilemma’, as conventionally understood, seems
to be a problem only in relations between some states. Many groupings of states
(regions) have rather developed into a community within which the ‘security
dilemma’ appears irrelevant. The democratic peace hypothesis, according to
193
194 Visions of world politics

which liberal-democratic states do not fight each other, covers some but not all
of these actual and possible security communities (see Adler and Barnett 1998a).
Karl Deutsch and his associates (1957) introduced the notion of security
community. They argued—at the height of the Cold War—that the existence of the
state is not a necessary or a sufficient condition for peace, nor is the non-
existence of the state a necessary or a sufficient condition for the prevalence of
the acute threat of political violence. These connections are contingent. The
imposition of a common government, with its capability of violent enforcement
of norms, may well decrease rather than increase the chances of peace. Besides
claiming that there is no axiomatic relationship between the absence of state
machinery of violence (or anything equivalent), and the condition of war, Deutsch
and his associates tried to disprove Arg. (a) as an empirical hypothesis. (For an
excellent summary of this innovative work, see Lijphart 1981.)
A group of eight scholars—six historians and two political scientists—studied
systematically the past experience of Germany, the Habsburg Empire, Italy,
Norway—Sweden, Switzerland, the UK and the USA, ‘to learn what history
might tell us about the problem’ (Deutsch et al. 1957:vii).1 Recently, this work
has been extended to cover the latter half of the twentieth century and, most
importantly, the world outside the North Atlantic area (Adler and Barnett 1998a).
In developing alternative and more processual hypotheses about the sources of
peace, Deutsch et al. (1957:5) distinguished between an amalgamated and pluralist
security community. Deutsch conceived both as political communities,
characterised by a process of political communication and the existence of some
shared rules and practices:

Amalgamated community: A single governmental unit with a process of political

communication, a single supreme decision-making centre, some machinery
for enforcement and practices of compliance.
Security community: A social group of people, which has become integrated:

• By integration is meant the attainment, within a territory, of a ‘sense of

community’ and of institutions and practices strong enough and
widespread enough to ensure, for a long time, dependable expectations
of ‘peaceful change’ among its population.
• By sense of community is meant that there is a belief on the part of actors
that they have to come to agreement on at least this one point: that
common social problems must and can be resolved by processes of
‘peaceful change’.
• By peaceful change is meant the resolution of social problems, normally by
institutionalised procedures, without resort to large-scale physical force.

A security community is one in which there is a real assurance that the members
of that scommunity will not fight each other physically, but will settle their
disputes and conflicts in some peaceful way. If the entire globe were an integrated
security community, wars and large-scale political violence would have been
 A global security community 195

eliminated. Amalgamated political communities may or may not be security

communities, and vice versa. Thus, two types of security communities can be
distinguished. Deutsch’s well-known 2×2 conceptual scheme, presented in Table
8.1, is based on these distinctions.
To use critical realist (CR) terminology, the notion of security community
was meant as a concrete Utopia, with the power to guide political practices (cf.
Deutsch et al. 1957:3–4, 7, 10–11). A concrete Utopia involves: lessons drawn
from past or contemporary models; counterfactual reasoning about the possible
effects of an altered context; as well as thought-experiments about the
consequences of the transformed practices and systems. Studies on the OSCE
(Adler 1998), the Gulf Cooperation Council (Barnett and Cause 1998; cf.
Patomäki 1994), South-east Asia (Acharya 1998), the Mercosur region in South
America (Hurrell 1998) and the NAFTA, including Mexico (Gonzales and
Haggard 1998) have thus shed new light not only on the current conditions of
these regions, but also of their possible, desirable futures.
The point of this chapter is twofold. First, I deploy philosophical and social
theoretical arguments to generalise and develop further the notion of security
community. The key point is that social contexts vary in their openness to
transformation. This provides a deeper, CR explanation of the conditions for a
security community. Second, I argue that the metaphor ‘complex social system
as a community’ has tended to mislead scholars to look only at regions based
on geographical proximity. I also make a case for studying global rules, practices
and processes. This is not just to reiterate the point that actions and their
contexts can be independent of particular spatial settings; neither is it only
about the relative detachment of action from location or place. What I have in
mind, first and foremost, is rules and institutions of international society and
global governance. Do they enable peaceful changes and, thereby, a sense of
global community? Or are we rather seeing steps towards amalgamation without
pluralist integration?
A global security community in the sense of this chapter does not presuppose
that every locality, state or region on earth must first be a security community.
Rather, it implies the opening up of global spaces for political communication

Table 8.1 Amalgamation and integration

196 Visions of world politics

and rule making. Public disagreements require conscious public airing. Politics
and violence both fall within the area of conscious, purposeful conflicts. Violence,
and threats of violence, tend to emerge with the rise of stubborn wills. In the
public sphere of politics, identities and interests are transformable and new processes
are initiated. Politics in this sense is non-violent; violence inevitably destroys the
preconditions of politics (Arendt 1970:54–6). Wæver (1996:127) says the same
thing when he notes that ‘for politics to emerge—not necessarily “democracy”,
but the precondition for it—we need to eliminate the logic of war’. He does not
speak generally of political violence but suggests, however, that ‘politics in this
sense demands the exclusion of security—the exclusion of security policy as well
as of insecurity’. A study of the conditions and nature of a global security
community thus has far-reaching implications for the possibility of world politics.

The conditions for a security community

Deutsch et al. (1957:12) suggested twelve essential background conditions for

an amalgamated security community to emerge and succeed. These include:
the mutual compatibility of main values; a distinctive way of life; capabilities
and processes of cross-cutting communication; high geographic and social
mobility; multiplicity and balance of transactions; a significant frequency of
some interchange in group roles; a broadening of the political elite; and high
political and administrative capabilities. They also include the willingness and
ability of the majority of the politically relevant strata:

1 to accept and support common governmental institutions;

2 to extend generalised political loyalty to them;
3 and to operate these common institutions with adequate mutual attention
and responsiveness to the messages and needs of all participating units.

These are the most essential background conditions.

Deutsch et al. (1957:12–13) feel that ‘success is extremely improbable in the
absence’ of the essential conditions. However, they argue that these are not
sufficient, for, in any concrete context, other conditions may be—and typically
are—needed as well. Moreover, ‘we cannot assume that because conditions in
one century led to certain effects, even roughly parallel conditions in another
century would lead to similar effects’ (Deutsch et al. 1957:11). There are no
timeless necessary and sufficient conditions.
Even if achieved, an amalgamated political community will remain contingent.
It is vulnerable to strife, secession and civil war (Deutsch et al. 1957:29–31).
Pluralistic security communities, however, are easier to establish and maintain.
Of the twelve conditions ‘that appeared essential for the success of an
amalgamated security community’, only the compatibility of major political
values, responsiveness to one another’s messages and needs, and partial mutual
predictability is required (Deutsch et al. 1957:66).
 A global security community 197

The pluralist security community is a path-breaking idea. At once it goes

beyond the mystifications of the international problematic. According to
Deutsch’s theory, the choice is not between diplomacy and balance of power, on
the one hand, and rule of law, supported by republican mechanisms, on the
other. There are variable social conditions for peace, war and politics, which
work beneath the appearances of the international problematic. However,
Deutsch blurs the distinction between internal and causal relations, and
vacillates between an empiricist and realist conception of causality. He is thus
led to confuse the features of a security community and the conditions of its
emergence and success. Consequently, the actors of geo-historical processes tend
to fade into the background, and there is only little space for politics of
identities, rules and practices.
Deutsch discusses causal relations in terms of empirical regularities and
conditions—not in terms of a productive relationship. Deutsch thinks that this
enables him to avoid a difficult problem: ‘Political scientists can very well seek
out and test possible regularities and probabilities without becoming entangled
in the metaphysics of any absolute causality concept’ (Deutsch 1963:13). There
are nonetheless hints of critical scientific realism. Deutsch argues that the accuracy
of a model and its match to ‘empirical processes’ must also be tested (Deutsch
1963:11). This is a call for realistic and geo-historically (sufficiently) accurate
assumptions. Moreover, Deutsch’s explicit use of models seems to indicate that
he is, up to an extent, an epistemological relativist:

In the end, we had to rely most of the time on nothing more scientific than
the use of analogy, occasional insight, and judgement. It is true that ‘events
are not affected by analogies; they are determined by the combination of
circumstances’. But analysis of events is certainly affected by analogies. To
throw analogies away would be prodigal, in the absence of any better
source of clues.
(Deutsch et al. 1957:13)

However, Deutsch’s qualifications of positivism are not enough. The dream of

Deutsch (and many others) of constructing a general theory of invariant political
processes cannot but fail. From a CR perspective, to say that A is necessary or
‘essential’ for B, is to say that B cannot exist without A. In other words, A
becomes an almost definitional feature of B. Is this logical relationship internal
or external? What could, then, be the difference between a necessary
characteristic and a causal condition? Indeed, for Deutsch, the background
conditions and the geo-historical processes tend to merge:

So long as this assembling of conditions occurs very slowly, we may treat the
status of each condition and the status of all of them together at any one time
as a matter of stable, seemingly unchanging background…. But as the last of
198 Visions of world politics

the conditions in each sequence are added to those whose attainment was
assembled previously, the tempo of the process quickens. Background and
process now becomes one…. Moreover, it is a process that may become
accelerated as a by-product of other processes of political and social change….
In this assembly-line process of history, and particularly in the transition
between background and process, timing is important.
(Deutsch 1957:70–1)

The basic idea is sound: no necessary condition is able to produce historical

outcomes alone. Many conditions together might suffice. The methodological
question is, however, would not the essential conditions that form a sufficient
condition imply logically the existence of a security community? What exactly
could the interplay between political events and actions and effects of background
conditions be? (cf. Deutsch et al. 1957:38). Where are the causal powers capable
of producing these effects? Where is agency and action? If the theory of the
emergence and existence of a security community is based on external necessary
and sufficient conditions in the positivist sense, it seems very hard to get
meanings, action and intentional making of history back.
Emanuel Adler and Michael Barnett (1998b:37–59) have taken a step forward
by distinguishing between three tiers of mechanisms and processes leading to
the development of a security community. Tier one, the precipitating conditions,
is largely exogenous, such as changes in technology, the rise of new world
views or external threats. They trigger states’ initial interest in co-operation
and, later, community building.2 Tier two, ‘factors conducive to the development
of mutual trust and collective identity’, involves both structures (conceived as
power and knowledge) and processes (actors and their actions and learning).
Tier three, ‘necessary conditions of dependable expectations of peaceful change’,
concerns the dynamics of mutual trust and collective identity. In this scheme,
actors, their knowledge/power and their actions matter. But how exactly?
Adler and Barnett (1998b:46) point out that trust comes logically prior to
collective identity Trust is based on past actions, and consequent expectations about
the future, but may be facilitated by third-party mechanisms. Beyond this, Adler
and Barnett have little to say about how power, knowledge, actors and their actions
may matter.

Learning and power

In a security community, problems and conflicts can be resolved or transformed

by peaceful means. This implies openness to the possibility of changes. The
basis is thus the development of dependable expectations of peaceful change
among the relevant actors. Trust and we-feeling, the formation of collective
identity, facilitate this dependable expectation. However, identity formation is
not a matter of a shared way of life:
 A global security community 199

The kind of sense of community that is relevant for integration, and therefore
for our study, turned out to be rather a matter of mutual sympathy and
loyalties; of ‘we-feeling’, trust, and mutual consideration; of partial identification
in terms of self-images and interests; of mutually successful predictions of
behaviour, and of co-operative action in accordance with it—in short, a matter
of perceptual dynamic process of mutual attention, communication, perception
of needs, and responsiveness in the process of decision-making. ‘Peaceful change’
could not be assured without this kind of relationship.
(Deutsch et al. 1957:36)

Homogenisation of ways of being and practices is not required; in fact, it can

be counterproductive. Deutsch illustrates this with the deterioration of English-
Irish relations between 1880 and 1914. As the Irish became more like English
in their way of life, practices and explicit political values, they became more
anti-English. To emphasise their anti-Englishness, some of their leaders then
sought deliberately to foster additional symbols and values, such as the use of
Gaelic language.3
Power is transformative capacity. Creating conditions for trust, fostering
particular symbols and values, and forging co-operation and collective identities
presupposes power—and may create new resources for subsequent actions. On
the other hand, also domination and organised violence presuppose power as
transformative capacity. How should we understand these different faces of
power and their connections to development of insecurity and violence or a
security community? In his Nerves of Government (1963), Deutsch introduces
important conceptual connections. For him, the key is the inner nature of actors,
which is constructed by mechanisms of learning.

Will is related to power. Hardening a decision—that is, closing the decision-

making system against any further messages by which that decision might
possibly be modified—is the key to the formation of a will. To have [narrow]
power means not to have to give in, and to force the environment of the
other person to do so. Power in this narrow sense is the priority of output over intake,
the ability to talk instead of listen. In a sense, it is the ability to afford not to learn….
When carried to extremes, such narrow power becomes blind, and the person
or organization becomes insensitive to the present, and is driven, like bullet
or torpedo, wholly by its past. The extolling of power by certain conservative
writers may not be unrelated to this pattern.
(Deutsch 1963:111)

Will in this sense implies the desire not to learn and change; and power as
domination may imply the ability not to have to do so. Will is constituted by
one’s understanding of the world, of one’s identity, of the right conduct of
action and one’s interests and preferences. ‘Both will and power may form
aspects of the pathology of social learning’ (Deutsch 1963:247). By pathological
learning Deutsch means ‘a learning process, and a corresponding change in
200 Visions of world politics

inner structure, that will reduce rather than increase the future learning capacity
of the person or organization. The result is an overvaluation of the present
expectations against all possibilities of surprise, discovery, and change’ (Deutsch
1963:248). One way to impose one’s ‘will’ upon others in a situation where
there is a conflict of interpretations or interests is to enhance one’s relative
power to others, by dominating others in terms of giving orders, making binding
rules hierarchically and manipulating with threats of violence. This tends to
result in an insecurity community with many related others (cf. Ashley 1987:428).
A lot depends thus on how open actors are to learning and changes. How
tolerable are they towards others and pluralism? And how responsive are they
to the wills of others? Tolerance, responsiveness and learning capabilities are
connected to the nature of the community—and therefore social context.

The self-transformative capability of social contexts

The assumption is that there will always be disputes and conflicts between
social forces. This implies that there can never be a stable ‘order’, an eternally
fixed set of practices and institutions. In more metaphorical terms, Deutsch’s
assumption is that it is not possible to tame or freeze history for a long time.
History is open. New interests and claims will emerge and new ‘messages’
demanding changes will be sent and made public. To simplify: if, despite
sustained efforts, there is no responsiveness from the side of status quo forces,
a pathological learning process may occur also among the advocators of changes.
This may lead to the escalation of conflict and (threats of) violence. The
preparedness to use large-scale violence inside and/or outside indicates non-
integration and is a sign of an insecurity community—inside and outside.
Security communities, in contrast, are characterised by the expectation that
future changes are going to be peaceful. Integration generates non-preparedness
to use violence. This can be explicated as follows:

A1 If a social system has become integrated, no relevant actor has any reasons
to prepare for the use of political violence.
A2 As actors know (A1), they do not expect anybody to use political violence
either to preserve the status quo or to foster changes.
B1 Non-preparedness becomes a generally followed and rarely, if ever,
questioned rule of action.
B2 In the course of social time, (A2) becomes an automatic, routine-like and
self-evident presupposition of political thought, argumentation and action.

(A1) is based on the thesis that actors are causally responsible for the outcomes
and that, at any point in time, actors could have done otherwise. Actors have
discursively or practically known reasons for acting in a particular way; they act
upon these reasons. Reasons are embedded in practices, narratives about one’s
identity and position in the world, world-understandings and related interests.
 A global security community 201

What matters, however, is not only whether actors really have any reasons for
violence but also whether they expect anybody else to think that way. Hence, a
security community presupposes both (A1) and (A2). If the practical experience
of actors confirms in the longer run (A1) and (A2), the security community
becomes sedimented in the longue durée of social time, i.e. becomes institutionalised.
(B1) and (B2) describe an institutionalised security community, such as the Nordic
countries from, at the latest, the early 1950s onwards.
Are there any general conditions for the generation of (A1 and A2) and (B1 and
B2), for the virtuous circle of the development of security community? The inner
structure of actors, communication and learning are socially conditioned. For
instance, the scope and diversity of the narratives (or scenarios) an actor is capable of
generating is the result of past learning of shared cultural meanings and forms the
basis for contemporary moral and strategic choices. However, available narratives
also depend on the mechanisms of choice that select and amalgamate them. A system
of domination can control or structure communication, and thereby shape not only
available narratives, but also trust and loyalty between actors. Moreover, actors and
actions are structured and stratified. Constitutive and regulative rules define and
position an actor, who has stratified and structured reasons for his or her actions that
occur in the context of unintended consequences of action. Reified and naturalised,
hierarchical or heteronomic, relations of domination may sustain particular practices
and related patterns of practical reasoning and rationalisation.
Following Roberto Unger (see 1987a:155–7; 1987b:35–6, 277–312), I think
it is possible to change not only the content but also the force of social contexts.
By force of contexts I mean their relative immunity to challenge in the midst of
everyday practices, the rigidity of positions of their positioned practices and the
consequent rigidity of social division and hierarchy. Social contexts can be
distinguished with respect to their openness to transformation. Unger refers to
this openness as the ‘negative capability’, but I think self-transformative capability of
contexts describes the idea better. It means the facility to challenge the context
in the midst of everyday practices and the disengagement of actors’ ‘practical
and passionate dealings from a pre-existing structure of roles and hierarchies’
(Unger 1987b:278–9). Since only agents in social relations can carry out context-
transformations and since this is social activity, the conditions for individual
self-transcendence and collective context-transformation are, in large part, social
conditions.4 On the basis of this reasoning, a general hypothesis emerges:

Hypothesis (b) The self-transformative capability of contexts

determines the possibilities for a pluralist and
amalgamated security community.

This is the hypothesis that the expectation of peaceful changes should be higher
in the more self-transformative contexts. It is compatible with the empirical
research of Adler and Barnett, Hayward Alker, Deutsch and others. For instance,
many of Deutsch’s background conditions for amalgamated security
communities—capabilities and processes of cross-cutting communication; high
202 Visions of world politics

geographic and social mobility; multiplicity and balance of transactions; a

significant frequency of some interchange in group roles; a broadening of the
political elite; and responsiveness of elites—indicate high self-transformative
capacity (but should not be confused with that capability per se). For pluralist
security communities, the requirements are less demanding, but the basic idea
is the same: contexts differ in their openness to change, and this is crucial for
the emergence and maintenance of a security community.
It should be noted, however, that a security community defined by the
existence of the sense of community and the expectation of peaceful changes is
neither a necessary nor a sufficient condition for the absence of large-scale
political violence. In principle, security as non-preparedness to use violence for
long periods of time is possible, in principle, without integration in the Deutschian
sense. On the other hand, mere dependable expectation of peaceful changes is
not usually enough:

Will the ‘minimalist’ or ‘maximalist’ view of international organization find

more support in historical experience [?] That is to say, can a security-
community be organised on the basis of agreement on only one point—the
necessity for peaceful change—or must agreement be achieved on many other
things as well?
(Deutsch et al. 1957:21)

Other agreements may be needed. It is also true that the idea of connecting the
development of social forces in time to the need for changes is characteristically
modern, which may limit its geo-historical applicability. Moreover, social systems
are open and hence there can be overriding mechanisms or processes at work
(for instance, think about the potential effects of an economic collapse or
depression in the midst of a politicisation process). The process towards increasing
self-transformative capacity of social contexts may also have unintended
consequences such as anomalies, pathological learning, socio-economic problems,
etc. Last but not least, the advancement towards greater disentrenchment is
never more than possible. It can be reversed or overridden by other factors.
Thus, hypothesis (β) does not imply that relative disentrenchment of contexts
will guarantee, permanently, the absence of political violence. Also, it is not
possible to generate a list of the institutional arrangements or imaginative
preconceptions that correspond to different degrees of disentrenchment. All
efforts to build more revisable contexts must work with the contemporary
contexts in or with the past or remote sequences of context making that we
study or remember (Unger 1987a:157).
These qualifications notwithstanding, hypothesis (β) seems to bear a great promise
for future research and practical actions. It is a CR re-interpretation of the nature of
mechanisms and processes underlying any security community. Hypothesis (β) can be
qualified, revised or refuted by means of empirical research and theoretical
argumentation.5 It can also be used in explanatory models and concrete Utopias.
 A global security community 203

The role of emancipatory research

Disentrenchment creates and enhances possibilities to transform social contexts.

This gives grounds for the dependable expectation of peaceful changes. A further
argument for the hypothesis (β) can be made. A security community may require
agreement on a number of things, and may lead to collective identity-formation.
Any sense of community can develop and harden into a will. The preparedness
to use violence is typically based on the necessitarian assumption about the
unchangeable essence of both oneself and the others (perhaps enemies). The self-
transformative capability of contexts has epistemic implications: it is not compatible
with illusions and mystifications about, or reifications and naturalisations of, social
realities. Conversely, the de-naturalisation of understandings can contribute to
the openness and responsiveness of the community.
Different strands of my argument are collected together in Figure 8.1. First,
there is the definition of a security community. A security community consists of
geo-historical social systems in which actors do not prepare for the use of
political violence against each other. Mere separate indifference is not enough for
a community. There must be some interdependence of elements of the systems.
Integration generates and helps to sustain a security community. Integration
consists of sense of community and expectation of peaceful changes.
Interdependence as such does not imply integration. Increasing
interdependence—or globalisation—can in fact reinforce existing incompatibilities
or induce the creation of new ones, particularly if connected to growing
disparities in the distribution of resources and wealth (Eberwein 1995:351–52).
Also, emancipatory research is situated in Figure 8.1. Various unjustified
psychological rationalisations and ideological mystifications of social structure
support the (re)production of those structures (Bhaskar 1986:194). Realist
explanations tend to work for enhancing the self-transformative capability of
contexts by criticising untrue naturalisations, reifications and fetishisations of
social being and related mystifications of knowledge, by making arguments for
peaceful transformations, and by creating mechanisms of reflective learning. At
a deep level, criticisms may also concern, say, the alienation and atomism
characteristic of capitalism—and the essentially related international problematic.
Any relevant relations and practices can be revised. More straightforwardly,
social scientific research can also propose concrete Utopias that either presuppose
or concern building or maintaining a security community.

A global security community

Adler (1997) argues that a security community does not have to be tied to a
geographical region. There can be also imagined security communities as
‘cognitive regions’. Thus, ‘security communities might emerge between non-
contiguous states’ (Adler and Barnett 1998b:33). For instance, Australia or Israel
can be part of the Western security community This point is important, but
204 Visions of world politics

Figure 8.1 Generation of a security community

does not go far enough. Adler (1997:252) himself maintains that multilateral
regimes and economic arrangements have created a non-territorial functional
space; that state authority in the realms of security, economic welfare, ecology
and human rights has been increasingly internationalised; and that a primitive
global civil society has emerged. Rules, relations of authority and systems of
domination do not have to follow the principle of territoriality. John Ruggie
(1998:192–7) describes the non-territorial economic ‘region’ of the world vividly.
He argues that states continue to engage with external economic relations with
each other. The terms of these engagements are largely set in systems of
multilateral governance.

In the nonterritorial global economic region, however, distinctions between

internal and external once again are exceedingly problematic, and any given
state is but one constraint in corporate strategic calculations. This is the world
in which IBM is Japan’s largest computer exporter, and Sony is the largest
exporter of television sets from the United States. It is the world in which
Brothers Industries, a Japanese concern assembling typewriters in Bartlett,
Tennessee, brings an antidumping case before the US International Trade
Commission against Smith Corona, an American firm that imports typewriters
into the United States from its offshore facilities in Singapore and Indonesia. It
is the world in which even the US Pentagon is baffled by the problem of how
to maintain the national identity of ‘its’ defense-industrial base.
(Ruggie 1998:196)
 A global security community 205

The non-territorial economic ‘region’ of the world is rule-governed. Some

fundamental rules, such as private property rights, originate in the seventeenth-
century expansion of the capitalist world economy. Others are sedimented in
various historical layers. The Bretton Woods system, which was created during
the Second World War, regulated economic activities in detail and also created
new organisations, the IMF, the World Bank and the GATT In the early 2000s,
the economic policies of many dozens of countries, particularly in the southern
hemisphere, are directly dictated by the IMF and the World Bank. The GATT
has been turned into the WTO, which has, despite widespread resistance and the
failure of Seattle, an ever-increasing mandate to regulate the terms of any economic
activities anywhere. The global financial markets re-emerged in the 1970s and
have consequently given rise to new heteronomic relations of domination, which
have been rendered also into the service of reviving the hegemony of the USA.
The global financial markets are also governed multilaterally, in part by the IMF,
but to a large extent by the Bank for International Settlements, BIS (see, for
example, Patomäki 2000b; 2001b: Chapter 3).
My (hypo)thesis is that this emergent sphere determines many parts of various
contemporary contexts of action across the globe. Thus arises the problem of global
security community. The conditions for collective context-transformation are,
perhaps increasingly, not only regional but also global. We should expect this to have
effects on the possibility of the emergence and maintenance of security communities
in various parts of the world. Even more far-reachingly, we can also talk about the
possibility of a global security community. To what extent do the emergent global
rules and relations have self-transformative capacity? In the opinion of relevant
actors, is there a firm belief in the existence of a global community? How large
preparations are made for war against (potentially) deviant or context-challenging
groups or states within the global community or by them?

Why the Kantian world order is not a security

community

To conclude this chapter, I shall briefly explain why the Kantian world order—
his solution to the international problematic—may not be a security community;
how Immanuel Kant was led to a dilemma he could not resolve; and how the
notion of global security community can help to overcome this problematic.
For Kant, all public rights are transcendental, i.e. universally valid. He claims
that the civil state is based a priori on the following principles (Kant 1983
[1793]:72):

1 The freedom of every member of society as a human being (every person

may seek happiness in the way that seems best to him or her).
2 The equality of each member with every other as a subject (everyone in a
nation who is subject to laws is a subject to the right of coercion).
3 The independence of every member of the commonwealth as a citizen
206 Visions of world politics

(= men who own property and are their own masters, not women, wage-labourers
or children); in the process of law-making, all citizens have one vote.

Kant’s federation of free nations must be partially different from a government of

a nation, because persons and nations also differ from each other. However, in
respect of (1) there is no difference. As human beings are free to seek their own
way to happiness within the state, so nations are free within the world federation
in a sense analogical to (1). In respect of (2), Kant was hesitant. He was not sure
whether everyone in a world federation should be subject to coercion should they
violate the universal cosmopolitan norms, including republicanism, openness to
free trade and the implicit assumption of globally valid private-property rights.
Moreover, it seems to me that the Kantian theory gives rise to a prima-facie
transcendental duty to make all nations similar (republican, based on free trade),
and, perhaps, but only ambiguously, also by means of enforcement. The basic
difference lies, however, in (3). In the world federation there is no—and, so Kant
assumed, should be no—real law-making activities.
Is this world order, envisaged by Kant, a security community? Within the
states, there are republican institutions, which make it possible to create and
transform laws by the citizens. Within the states there is also tolerance of difference
and pluralism, but only in the private sphere of persons. This means limited self-
transformative capacity6 For these reasons, the partially Kantian liberal-democratic
countries of the twentieth century may be argued to have, relatively speaking,
more disentrenched frameworks of social and political institutions and practices
than the seventeenth- and eighteenth-century European monarchies.
Internationally, however, the Kantian model seems fixed. Once the world
order has been created, no possibility of further changes is envisaged. Although
a high degree of interdependence and globally valid property rights are
presupposed, the only explicit norm is: ‘You shall not make war against another
free nation.’ In contrast to a security community, in the Kantian world order
there is no way to change anything. A peaceful status quo is the norm, whatever
this implies at the level of concrete practices. All political arguments presuppose
a number of contextual features, although these are usually left unexplicated.
This is also true of the contemporary Kantian arguments.

1 There is the formal separation between private (including property) and public,
which is equated with a particular kind of society/state relationship. Private
capitalist economy is privileged; the public state is controlled and checked by
regular elections and the system of ‘checks and balances’.
2 The model assumes free trade, interdependence and, in that sense, a globalising
laissez faire economy.
3 The model includes implicitly Western, and particularly US, hegemony and
dominance of global governance. The USA, the EU and Japanese state
managers and ‘private’ actors also control global investment decisions and
flows. The UN Security Council—dominated by the USA—is responsible for
identifying the violators of the norms that define one as ‘a peace-loving nation’.
 A global security community 207

The US army, supported by other Western states, enforces these norms, with
or without the UN Security Council (see Patomäki 1999a; 2001c).
4 At a thinly hidden level, there also is the US power projection, which consists
of extended deterrence, on the one hand, and the ‘idealist’ project of spreading
the US conception of genuine identity and right political conduct, on the
other (Klein 1988; see also 1990).

Even Kant, however, who is often taken to have advocated collective security, was
not arguing for a world government. Kant was suspicious of the idea of a world
government. In his view, a world government could be easily turned into a
tyranny:

a federative state of nations whose only purpose is to prevent war is the only
state of right compatible with their freedom…. The idea of international right
presupposes the existence of many separate, independent, adjoining nations;
and although such a situation is in itself a state of war (assuming that a
federative union among them does not prevent the outbreak of hostilities),
yet this situation is rationally preferable to their being overrun by a superior
power that melds them into a universal monarchy. For laws invariably lose
their impact with the expansion of their domain of governance, and after it
has uprooted the soul of good a soulless despotism finally degenerates into
anarchy. Nonetheless, the desire of every nation (or its ruler) is to establish
an enduring peace, hoping, if possible, to dominate the entire world.
(Kant 1983 [1795]:138 and 124–5)

For Kant, anything resembling world government is likely to lead to tyranny.

Because the pre-contract ‘state of war’ is not acceptable either, Kant attempted to
construct a third way. For him, the solution was a very loose federation of free nations,
whose peace-proneness is based on the shared understanding of the human and
economic costs of increasingly destructive war; on the economic interdependence of
nations; on the inherent peacefulness of republican states; and on the rule of law.
Kant’s dilemma was: if these (in themselves already homogenising) conditions are
enough for guaranteeing peace, why is there a need for a league of peace in the first
place? If these conditions are not sufficient for peace, is there then not a need for a
world government exercising a monopoly on the control of the means of violence?
However, if there is such a world government, is there not a danger of it being just an
expression of a desire of a particular nation (or its rulers) to dominate the entire
world? In that case, is it not possible that the end result of global integration may be
a form of tyranny that may also ‘finally degenerate into anarchy’?
To a limited extent, Kant was also asking some right questions—but could
not answer them. The notion of a global security community provides a fresh
and, it seems, more adequate perspective on contemporary developments. If
the self-transformative capacity of contexts is the key, then the focus should be
on attempts, both theoretical and practical, to challenge and revise global contexts
208 Visions of world politics

of actions. The crucial issue is whether these challenges and revisions could
lead—how?—to increased learning and self-transformative capacity. This is a
question, it seems to me, that can be posed only after the international problematic.
A new background theory gives rise to a characteristic set of problems and also
provides some plausible answers to them.

Conclusion

Like CR, Deutsch’s notion of security community presupposes that history is

open. Otherwise there would be no point in trying to disentrench social contexts
instead of entrenching the arrangements of the illusory end of history. Even a
non-innovating humanity would be constantly heterogeneous in its identities,
practices and understandings of proper political conduct. At some point these
differences may well give rise to efforts to change the current contexts of action.
Moreover, past struggles can be re-opened, perhaps in new contexts that may
be more favourable to the previously suppressed possibilities; new combinations
of the already existing elements of social contexts can be invented and innovated;
and also genuinely novel elements may be innovated and fed into the processes
of political struggles. These political possibilities are always open to humanity.
The possibility of a global security community depend on how open the
relevant social contexts are to social transformations. However, any attempt to
challenge and revise global contexts of actions must be studied concretely, to
assess the relative impact of various causal mechanisms and powers efficacious
in open systems. Any real tendency can be overridden by other causally
efficacious tendencies.
A security community can also be simply presupposed by an explanatory
model or a concrete Utopia. The relevant issues may have little if anything to do
with security per se. Public politics may be about anything; any aspect of context
can be politicised, although not all of them simultaneously (agreements are always
necessary). It is a sign of an institutionalised security community that security
does not persist as a major issue. Wæver (1998:69) calls this ‘desecuritisation’,
the progressive marginalisation of mutual security concerns in favour of other
issues. De-securitisation opens up the possibility of world politics. In the public
sphere of politics, identities and interests are transformable and new processes can
be initiated. If this becomes a widely accepted principle of world politics, humanity
has succeeded to establish a global security community.

Notes

1 Deutsch et al. (1957:16) explain the selection of their cases as follows:‘In selecting a
limited number of cases for intensive investigation, we concentrated on the area of
Western European and North Atlantic civilization. There are obviously other cases
that we should have liked to include. Considerations of time, resources, availability
 A global security community 209

of data, or apparent comparability with contemporary problems, excluded some of

the earliest cases as well as those located in Asia, native Africa (south of the Sahara),
and most of Eastern Europe.’ They go on to state ‘this collection of cases…supplies
reasonably good samples of most of the important cultural traditions and institutional
patterns of Western Europe and the North Atlantic area’.
2 Adler and Barnett, however, seem to focus on the co-operation and community building
between states, whereas Deutsch studied social formations much more generally. I
follow Deutsch in this regard.
3 Deutsch et al. (1957:42) in fact point out that a particular political event—the failure of
the Parliament to respond quickly and adequately to the disastrous Irish famine of
1846—led to the rise of movements that tried to distance Ireland from England. More
generally, the shared way of life and values notwithstanding, nationalism gave an
increasingly legitimate ideology for the Irish to oppose the asymmetric and often
exploitative relationship with London.
4 Gould (1978: particularly 109–10) argues that this kind of view of social action can be
attributed to the Marx of Gründrisse.
5 I will have to leave aside the methodological problems of studying hypothesis (β) by
empirical means. Suffice it to say that like any qualitative research on open systems, it
has to rely on the systematic use of analogy, occasional insight, available evidence
and judgement. The rules and practices of different contexts have to be judged in
terms of their openness to challenge and transformation, and then compared
systematically (even if a closed list is not possible, some relevant comparisons are).
Since it is a matter of interpreting geo-historical evidence in rather broad strokes,
particular care has to be taken about the coding procedures and practices (see the
Alker chapter for some details).
6 In addition to all the fixed background assumptions and implicit exclusions, there is
also the Foucauldian argument that, despite the rights of individuals, there may be
little real freedom and pluralism in Kantian liberalism. Rather, discourses of normalcy
and order, and the related practices, have tried, despite resistance, to make life uniform
(‘normal’) also in the private sphere. Thus, everyone ‘has his own Gulag’,
independently of his or her time-space location. Against this, Walzer’s point is partially
apt: ‘[Foucault’s] account of the carceral archipelago contains no hint of how or why
our own society stops short of Gulag. For such an account would require what Foucault
always resists: some positive evaluation of the liberal state’ (Walzer 1986:62). However,
some positive evaluation of the liberal state does not imply that Foucault is entirely
wrong. The state is indeed non-relevant to whole series of power relations, as Foucault
correctly asserted—yet failed to grasp the global aspects of this claim.
9 From security to
emancipatory world politics
Beyond ‘Nordic nostalgia’

This final chapter is a call for making the Nordic model a basis for emancipatory,
regional and global political action. My case is only superficially connected
with recent attempts by Tony Blair and Bill Clinton—with the assistance of
Anthony Giddens (1998)—to resuscitate the Third Way.1 Rather, the argument
of this chapter is based on a causal model, which explains the decline of the
Nordic model. Moreover, I also make a case for re-defining many of the ideals
of Norden in the context of the legitimation problems—if not crisis—of the so-
called Washington consensus.
At the outset, the question is: Are the principles of the Swedish model2
sustainable in a changing world that tries to re-formulate and re-define the
public sphere and rules for economic policy in accordance with orthodox market-
monetarist postulates in macro-economics (fiscal and monetary policy) and micro-
economics (e.g. trade, labour market and industrial policy)?’ However, the issue
of the future of the Nordic model also has a significance that goes far beyond
the immediate concerns of economic policies of states. Norden has also been a
model of an enlightened and anti-militaristic society that follows the principles
of distributive justice and which was supposed to be morally superior to two
alternative models of modernisation: the USA and the Soviet Union. After the
collapse of the Soviet Union, it thus appears that the question of whether there
is a future for the Nordic model also means ‘Is there any socially oriented,
democratic alternative to the Washington consensus-led neo-liberalist
globalisation?’
My answer is affirmative. It is of course true that, due to fundamental
contextual changes, attempts to preserve the Nordic model within the confines
of a nation-state have been in difficulties for a quarter of a century. However,
there are many features in these models and related moral and political ideals
that also carry weight in the globalising world at the dawn of the century. In
light of this, I discuss democratisation, solidarity, de-commodification and
universalistic social policy as moral ideals; and the activist politics of neutrality
that gave rise to globalist attempts at building a more just and peaceful world.
At the threshold of a new century, these are potential elements in the formation
of new identities and in telling new stories and creating new myths for counter-
hegemonic purposes.
210
 Beyond ‘Nordic nostalgia’ 211

Explaining the decline of the Nordic model

There are at least two main ways of understanding the Nordic model in terms of its
content and place in world history. This contextual specification also determines the
identification of the main features of the Nordic model. First, Norden can be seen as
a political ideal and model of sovereignty-based active internationalism in the Cold
War context. With the sudden fading away of the Cold War, there were constitutive
reasons and real tendencies for the decline of this model (Wæver 1992).
Second, it can just as easily be seen as a model of organising production and
exchange relations and income distribution within the Bretton Woods system
of embedded liberalism. In this interpretation, the Nordic model and the social
forces driving it were vulnerable to the processes of gradual disembodiment of
‘embedded liberalism’ that started to make a qualitative difference at the turn of
the 1960s and 1970s (see Ruggie 1982, 1995). In this reading, the active
internationalism of Norden stemmed more from the shared societal values that
were translated into foreign-policy orientation than from the Cold War context.

Norden in the context of the Cold War and

European integration

I shall start with the Cold War contextualisation, following Ole Wæver’s (1992)
interpretation of Norden’s fate after the end of the Cold War. Wæver claims that
in the early 1990s there emerged a ‘Nordic nostalgia’, a longing for the certainties
of, and Nordic privileges in, the Cold War world. Originally, as a political ideal
in the Cold War context, the Nordic model was defined as representing the
Third Way: neither US capitalist liberalism, nor Soviet-style state socialism, but
social democracy. In this sense, the Nordic identity was dependent on the
competition between capitalism and communism. Moreover, the neutrality of
Sweden—and also that of Finland—stemmed from the Cold War struggle over
models of organising society, and from the related military competition (even if
sometimes the Cold War seemed to have assumed for its participants a form of
an eternal and politically neutral Waltzian system of bipolar power-balancing).
The presumed superiority of the rational, enlightened and (mostly) anti-
militaristic Nordic society was in part based on the fact that in Norden the
military tensions were much lower than in Central Europe. This was despite
Norway and Denmark being members of NATO and Finland having a Treaty
of Friendship, Co-operation and Mutual Assistance (signed in 1948) with the
Soviet Union. Further, Norden was a passport-free zone of common labour
markets and shared social security systems. For Wæver, however (1992:84–5),
this regime was partially independent of the Nordics, and only partially
stemmed from ‘rather grandiose schemes and declarations about issues
somewhat intangibly related to the Nordic countries themselves’.
It is true that there were also Cold War-related military-strategic reasons for
creating a pluralist security community and for deepening co-operation among
212 Visions of world politics

Table 9.1 The Cold War interpretation of Norden

the Nordic countries.3 Nonetheless, even in the Cold War context, it was the
social democratic Sweden that led Norden, by showing an example of social
reforms and by exercising ethico-political leadership in building the idea of
actively internationalist Norden. This ideal was widely accepted for decades,
until the end of the 1980s. Even in this security-oriented interpretation, Norden
was not only about security, but also incorporated a particular socio-economic
system and a particular theory of modernisation. The main features of this
Norden, as seen through the lens of the Cold War competition, are summarised
in the first row of Table 9.1. The second row sets forth Waever’s account of
how the end of the Cold War undermined the identity and idea of Norden.
Wæver claims that, contrary to what many people might have expected, the
Nordic countries reacted to the end of the Cold War with scepticism, frustration
and attempts to limit the impact of change (Wæver 1992:77). Why? According
to Wæver, it is a matter of nostalgia both for a secure identity and for the
Golden Age of Social Democratic vision of world historical progress and the
building of a just society and peaceful world. For Wæver, the way that the
vision of Norden as an alternative has functioned was visible, for instance, in all
the EC debates in all the Nordic countries, where the ‘Nordic card has regularly
been played by opponents of EC membership’ (Wæver 1992:94). Nordic values
have been contrasted with the European market; Nordic co-operation has been
put forward as an alternative to European integration.
 Beyond ‘Nordic nostalgia’ 213

Problems with Wæver’s account

There is reason to believe, however, that this Cold War and security-centred model
of Norden and its prospects is at least partially inaccurate and inadequate. Consider
the explanations for changes from the Cold War Norden to the post-Cold War
Norden. The logic in each explanation is similar and always inadequate. The
argument about the impossibility of the Third Way after the end of the Cold War
is conceptual only—and incorrect as such. It seems to be based on the idea that there
are only relational differences, and that the world is about language only (cf. Wæver
1992:79, 86):

The old system had two competing points—the planned economy and
capitalism—so that any distance naturally led to the privilege of being third.
The new system has one dominant centre—the EC—and all distance from
this becomes simply peripheralism.
(Wæver 1992:86)

However, even in its own terms, this argument is questionable. Why should the
third be automatically ‘privileged’ (whatever that means)? Many critical cultural
theorists would argue on the contrary that there is little room for the third in the
binary logic of Western—and many other cultures’—metaphysics. (About the violent
implications of strict either/or thinking, see, for example, Beardsworth 1996:54–97;
Galtung 1990). During the Manichean struggle between the USA and the Soviet
Union, there certainly was very little room for any third possibilities in many parts
of the world.
Also, the essentialist argument about the EC as the dominant centre has
only little grounding, at least at logical and conceptual levels. It presupposes the
pseudo-Hegelian master—slave argument, which says that if there is a master,
all attempts by the slaves to keep a distance are simply ‘peripheralism’. Hence,
the only worthwhile strategy for the slaves is to try to gain recognition from
their master and participate in his practices. This reading finds support in claims,
for example, that the previously anti-militarist Nordic countries should now
start to ‘carry power-political responsibility’ (Wæver 1992:82). Obviously, there
could be more emancipatory readings of the master-slave dialectics.
However, the logic of the argument itself is problematic. Suppose we talked
not of the Third Way but instead about the Nordic model. Would we not still be
making reference to roughly the same real entity X (i.e. a particular social complex,
consisting of systematically organised relational, institutionalised practices, and to
the associated values of democracy, solidarity and egalitarianism)?
The question is: Why is it that this X had to change? Why is it that X could
not compete with the US model of modernisation without any problems even
after the collapse of the Soviet Union? There is no compelling logical, conceptual
reason why it could not have done so. Only if we can demonstrate, by way of
systematic empirical research, that this fallacious way of thinking was in fact
214 Visions of world politics

prevalent in Norden after 1989, would it be possible to claim that this was a
cause of the decline of the Nordic model.
Wæver does not provide any systematic empirical evidence. However, at the
level of constitution of discursive practices, there is also the possibility that
Waever’s article has, after 1992, reinforced the acceptance of this fallacious
inference and thereby contributed to the relative decline of Nordic values,
particularly among foreign-policy elites in these countries.
Consider, next, the explanation of change in the socio-economic system.
Besides simple descriptions of changes in economic policies in Sweden in
particular, the main explanation seems to be that the social democratic welfare
state is in an identity crisis because it is essentially connected to the notions of
modernisation and rationalisation, which, in these post-modern days, are in
crisis (Wæver 1992:85). However, even if we accepted this argument, and I
think there is something to it (cf. Leonard 1997), it would be far from obvious
why the US welfare system or neo-liberalism would be less in crisis. Even the
EU is typically represented as an exemplar of modernisation, although its
complexity and divergent nature has also given rise to more post-modernist
interpretations (cf. Ruggie 1993). Ruggie and others notwithstanding, the
modernity of social democracy cannot be a cause for the crisis of the social
democratic form of welfare state as contrasted to the ‘key ideas’ of standard,
modern liberalism, namely ‘political freedom rights, the free market and
integration’, as Wæver claims. Notably, Wæver only mentions the main
characteristics of the social democratic form of the welfare state in a footnote
(Wæver 1992:85, n. 13).
Finally, there is a further argument: Norden was not only defined as the
Third Way between capitalism and communism, but also in contrast to Europe.
That is, particularly in security political terms, it was supposed—by whom?,
with what effects?—to be better than Europe. After the end of the Cold War, the
key positive dynamics, in terms of movement towards peacefulness, were taking
place in Europe, and this supposedly turned the whole setting upside down. It
was Europe that proved to be somehow better than Norden. However, the problem
in this argument is the ambiguity in the terms ‘better’ and ‘Europe’.
To be a causal argument (that could, possibly, constitute a partial explanation
of the change of the Norden), Wæver’s point has to be seen as a claim about the
prevalent political rhetoric that was also efficacious in practice and action. Yet,
Waever’s point seems to be at least as much, if not more, a normative argument
that, in contrast to Norden, Europe after the end of the Cold War is now morally
and/or politically better. Simultaneously, the term Europe is used in different
senses in this argument. Sometimes it is the ‘common European home’ of Mikhail
Gorbachev (1989 as the key dynamics), and sometimes it is the Europe of the
EU (1992 as the key dynamics). However, exactly in what sense could it be
argued that Europe, in either of these two senses, suddenly became more peaceful
than Norden? There is no justification for this presumption, or for the significance
of the comparison itself. Yet, we do recall the argument that the reluctant Nordics
should now partake in the new master’s power political practices.
 Beyond ‘Nordic nostalgia’ 215

Towards an alternative explanation

Above I shed a shadow of suspicion over the Cold War and European
integrationoriented interpretation of the Nordic model. However, false beliefs
can be causally efficacious. The Cold War and security-centred model of Norden
together might explain the decline of Norden to the extent that the arguments
discussed above were de facto playing a role among the relevant political actors
by constituting their actions and practices. It is the task of further research to
determine the extent to which this was significant in the late 1980s and 1990s.
Even if it were true to a significant degree, this fact would simply become the
next object or tendency to be explained in depth.
There are reasons to believe that the causal processes bringing about difficulties
for Norden—and the Swedish model in particular—were largely unrelated to the
Cold War. In Sweden there had been adjustments to the Nordic model for a
decade before Gorbachev. The economic problems of Sweden started as early
as the 1970s, with some of them having their origins in the 1960s. In different
phases, they continued over the Second Cold War of the early 1980s and
became more severe again after 1989. The responses to these crises further
changed the context of political action and favoured particular solutions at a
later stage. They had already brought about a re-definition of the Third Way
by the early 1980s. The Third Way began to mean, increasingly, a compromise
between pure social democracy and neo-liberalism, rather than the Third Way
between capitalism and communism (Ryner 1994a:389 and 423, n. 2).
The problems and related re-definitions stemmed mostly from the internal—
external dynamics of the Swedish political economy, entangled in the globalisation
of social relations. Since the early 1970s, this globalisation has, in turn, taken place
in the context of the partial disruption and transformation of the Bretton Woods
system. However, to make the case for a political-economy explanation of the
decline of the Swedish model, we first need an adequate identification of its major
features.

Identifying Norden in the global

political-economy context

Norden is most frequently associated with the universalist welfare state. Indeed,
the already classic power mobilisation thesis of Gøsta Esping-Andersen (1990)
makes Sweden a representative of the institutional re-distributive model of social
policy, which operates externally to market logic. This model presupposes
poverty to be a permanent condition of industrial societies if not counteracted
by decommodified social institutions. For Esping-Andersen (1990:44), de-
commodification is emancipation from the dependency of capitalist labour
markets—and thereby also from employers’ control—by means of universalist,
high-level social benefits. Also, education, Healthcare and insurances were
organised collectively and outside markets, that is, in a de-commodified way.
216 Visions of world politics

Finally, this social democratic model was coupled with a full employment
policy.
However, there was more to the Swedish model than this. The Nordic model
also implied a certain way of organising production and exchange relations,
and presupposed a particular world order. Table 9.2 summarises the major
features of this model, again also by contrasting two eras (here I follow closely
Ryner 1994b; 1997; but see also Weiss 1998:86–100; and in particular the
more detailed historical account by Pontusson 1992:37–123).
The Rehn-Meidner programme, named after its key proponents Gøsta Rehn
and Rudolf Meidner, emerged with the Saltsjöbaden capital-labour Accord of
1938 and was developed in the 1940s and 1950s. This model advocated
solidaristic wage policy but also encouraged firms to make technological
innovations, by squeezing low-productivity firms and industries. It also helped
adjustments to the technological dynamism of the capitalist world economy by

Table 9.2 A political-economy interpretation of the Swedish model

 Beyond ‘Nordic nostalgia’ 217

means of an active labour market policy: unemployed were re-trained and moved
to new employment by the state.
The Social Democratic Party (SAP) rose to power in 1932 by winning
parliamentary elections. It remained in office—either alone or by forming coalition
governments—uninterruptedly for forty-four years, until 1976. SAP provided
ethico-political leadership in developing and consolidating the institutional
framework. The Saltsjöbaden Accord formed the basis for centralised, corporatist
wage negotiations between the Swedish Union Federation (LO) and the Swedish
Employers’ Association (SAF). However, the Accord also guaranteed absolute
autonomy for the capitalist firms and their managers in deciding upon finance
and investments, intra-firm relations and the allocation of returns and profits.
Since the Accord, this capitalist laissez faire aspect of Swedish social democracy
remained unchallenged until the late 1960s.
The Swedish model also presupposed a particular world order. The era of
Pax Americana between 1945 and 1973 was the ‘Golden Age’ of the Swedish
model—and the original Bretton Woods system of embedded liberalism (Ruggie
1982). The essence of the embedded liberalism compromise of the original
Bretton Woods system was to devise a framework that would safeguard and
even aid the quest for domestic stability and legitimisation, ‘without, at the
same time, triggering the mutually destructive external consequences that had
plagued the inter-war period’ (Ruggie 1982:393). Moreover, the Bretton Woods
system also represented a partial victory of productivism over financial capital:
the Bretton Woods systems imposed significant constraints on the freedom of
movement of financial capital including fixed exchange rates and capital controls
(Gill 1997a:7).
In the embedded liberalism compromise, it was up to nation-states to design
their own variations of domestic stability and legitimation in terms of liberal
democracy and the welfare state, but Pax Americana set definite limits to possible
variations. Although its foundations were laid before the Second World War, the
Swedish model was, in fact, a possible variation. It was also more pluralistic than
it might have appeared at the outset. Particularly the Cold War-centred
interpretation of the Third Way would see Norden as competing with the USA
and the Soviet Union as the symbol of modernity and enlightenment, and, in its
own self-understanding, as a model of a ‘more developed society’. Yet, the Swedish
model allowed internally for competing understandings of its meaning. It was
seen either as a modernisation of capitalism—rendering class struggle and ideology
irrelevant (capitalists, industrialists)—or it was seen as a middle way of extreme
tendencies (mainstream social democracy). It was also envisaged as a stage on the
way to democratic socialism (radical social democracy, the New Left).
Many industrialists and most mainstream social democrats would argue that
Sweden was able to establish a relatively egalitarian society by means of an
exceptionally generous and comprehensive welfare state. In contrast, the more
radical social democracy and the New Left would emphasise that social
democracy had neither changed the underlying capitalist economic structure
nor gained control over investments (Rothstein 1996:4–5).
218 Visions of world politics

Outline of a political economy explanation

By the 1960s, the Swedish model had become the basis of the identity of Norden.
However, despite its long-lasting success in fulfilling its own promises, it started
to come under pressure from a number of directions. The turning point for the
Swedish model came in 1976. The Social Democrats had to step down from the
Government for the first time in forty-four years, giving way to the Centre
Party and Conservatives. That, in turn, took place six years after the US
declaration that the dollar was no longer convertible to gold, and two to three
years after the oil crisis, in the context of a deep Swedish economic depression.
Figure 9.1 illustrates how, in open systems, both internal and external
conditions tend to change, and how the basis of an institutional framework
may erode. This also happened to the Swedish model, despite its sustainability
and achievements between 1938 and 1976. Figure 9.1 summarises the underlying
socio-economic dynamics in terms of changes in occupational structure and
related class identity, as well as in the identity, interests and options of private
capital. These processes co-explain many of the difficulties that emerged by the
1970s, and also some of the first transformations of the Swedish model.
The first signs of trouble emerged in the 1960s: expansionary macro-
economic policies triggered inflation, and in-built tendencies for technological
dynamism were compromised for political reasons. Increasing alienation
between the trade union leaders and the rank and file led to wildcat strikes,
quickly followed by a leftist revolt against the decision-making monopoly of
capitalist owners and managers over investments and workplace practices.
Backed up by the global New Left movement, the labour movement in
Sweden introduced the ideas of active industrial policy, industrial democracy
and collective share ownership, and thereby it also challenged the
fundamentals of the Saltsjöbaden Accord (see Pontusson 1992).
Continuous wage-push inflation became a recurrent problem in the Swedish
economy Productivity growth declined dramatically in the 1970s, particularly
after 1973.4 Indeed, the global economic crisis of Fordism of the 1970s hit
Sweden more severely than any other OECD country. Stagflation, declining
productivity growth after decades of high productivity growth, led to Sweden
accumulating substantial budget and payment deficits and a foreign debt during
the period 1976–82 (Ryner 1994a:387–92). Simultaneously, a new world order
began to emerge. The new world was dominated by globalisation and trends
towards both more orthodox economic liberalism and the dominance of financial
capital over production.
In these circumstances, organised business and the bourgeois parties of Sweden
began a massive mobilisation against industrial democracy and, more concretely,
against the wage-earner funds. Consequently, the labour movement retreated from
the more radical features of the plans to co-control and democratise investments.
Despite these political developments—a turn to right favourable to capital—Swedish
firms began an investment boom abroad in the 1980s, exploiting the new exit
options. The new Social Democratic government, elected in 1982, also tried to use
Figure 9.1 Structuration of the conditions of the decline of the Swedish model 1976–89, in the context of stagflation and indebtment
220 Visions of world politics

financial liberalisation as a means to find a new discipline. It increased the

constraint of global capital markets on Sweden’s balance of payments and interest
rate (Ryner 1994a:392), and eventually decided to de-regulate financial markets.
Figure 9.1 also shows how the welfare state was, ultimately, fairly ambiguous
in its capacity to generate power mobilisation. Assuming consumerist culture, the
improvements in living and working conditions suggested that there were less
reasons for social democratic or leftist struggles. Outside the public sector, the
labour market changed and fragmented. In sum, the LO’s constituency declined
between 1960 and 1988 from 75 to 58 per cent of organised labour. On the other
hand, changes in the composition and positioning of the blue-collar wage labour
and the middle class, as well as the new exit options for firms, created new
conditions for the mobilisation of social forces behind the interests and visions of
the trans-nationalising capital. Between workers and employers in sectors exposed
to world markets, there in fact emerged a shared concern with containing the
upward pressure on domestic costs generated by large public sectors (Clayton
and Pontusson 1998:97). In the context generated by these transformations, the
new Third Way based on introducing new forms of market discipline soon
appeared to be contradictory by further undermining the legitimisation principles
and industrial relations presupposed by social democracy.
In the 1980s, the employers’ federation SAF—at that time already representing
mainly the ideas and interests of trans-nationalised firms—abandoned all major
elements of the traditional Swedish capital-labour accord. SAF started to campaign
for welfare cutbacks, privatisation, reduced taxes and employers’ contributions
and EC membership.
It is therefore not surprising that the Social Democratic economic policy of
1982–91, which the SAP government called the Third Way, was represented as
an alternative to both the neo-liberal doctrines of the newly elected regimes of
Margaret Thatcher and Ronald Reagan, on the one hand, and to traditional
Keynesian policy, on the other hand. Already in 1983, the Swedish government
explicitly defined their policy as departing from the Rehn—Meidner model,
although the government also claimed that it was ‘adapting the Rehn-Meidner
model to new times’. This policy led to a partial recovery in the 1980s, followed
by a new series of major economic troubles that began in 1989–90.
In the 1980s, and even more so in the 1990s, the initiatives came from the
SAF and neo-liberal conservatives. SAF was insisting on decentralisation of
wage negotiations, and had found support from governmental committees of
economists (see, for example, Lindbeck et al. 1994). Their SAF position had
also changed because the structural power of capital was greatly enhanced owing
to the trans-nationalisation of production and finance (cf. Gill and Law 1989:479
ff.). Also, economic policies appeared to be more strongly conditioned by the
rapid reactions of short-term financial capital (cf. Hayat 1999). In this era,
which had already started to emerge in the late 1960s, embedded liberalism
compromise became dis-embedded, i.e. more akin to or orthodox economic
liberalism, driven and sustained by the power of the re-emerging global financial
markets (see Harmes 1998; Patomäki 2001b: Chapter 3).
 Beyond ‘Nordic nostalgia’ 221

Globalising and thoroughly liberalised financial markets began to discipline

the Swedish government and economic actors particularly thoroughly in the
late 1980s and early 1990s. In the context of the banking crisis of 1990–1,
which was co-caused by the liberalisation of financial markets, and experienced
at least as severely by Finland and Norway, Sweden decided to apply for EC
membership in October 1990, thus causing a lot of alarm and irritation in the
other Nordic countries, not least in Finland. The membership that was put into
effect in 1995 implied that Sweden accepted the Maastricht Treaty in its entirety,
including the Economic and Monetary Union and its convergence criteria and
tendencies towards monetarist economic policies.
It seemed that Sweden was again leading Norden—with the partial exception
of Denmark, which had already joined the EU in 1973—but this time away
from the Nordic model. Finland and Norway liberalised their capital movements
and financial markets more or less simultaneously with Sweden in the 1980s.
As far as European integration was concerned, Finland followed Sweden by
applying for EU membership in 1992 and having its application accepted in a
referendum in 1994. The leadership in Norway attempted to follow suit, but
was defeated in the 1994 referendum on membership by a close majority. These
changes were brought about by interplay between (interconnected) internal and
external forces, examples and options.
Obviously, the weakness of this explanation is that it tends to silence the
concrete history and choices of many actors. Things could have been otherwise,
and choices do matter. Moreover, despite the economic problems, social democratic
corporatism was still doing relatively well in the 1980s (Katzenstein 1985). Also in
the 1990s, there remained many arguably sustainable responses to globalisation,
and some of them were more in accordance with egalitarian and democratic values
and highly capable democratic states than others. The fact that the particular
features of the Swedish model—and those of the other, lately more successful
variations of the Nordic model—have been under pressure does not mean that there
have been, are or will be no national alternatives (see Garrett 1998; Weiss 1998; cf.
also the more radically democratic argument of Unger 1998).
My argument is not that there have been or are no alternatives, but that
many of the conditions for the sustainability and, possibly, partial demise of the
Nordic model have been beyond the reach of a territorial nation-state. Because
of the political choices made, and because of the deepening of the multi-
dimensional and multi-layered process of globalisation, the choices may be more
limited now than they were in the 1970s, 1980s and even in the early 1990s.
To repeat, globalisation—including the previous choices of nation-states to
privatise and liberalise areas of the economy that were previously controlled by
the state—has obviously meant new exit options for private, productive capital
and new room for manoeuvre for financial capital. However, globalisation has
also transformed important aspects of culture (Tomlinson 1999); technologically
mediated public and expert discourses (Bourdieu and Wacquant 1999; Herman
and McChesney 1997); and European and global relations of domination
(Minkkinen and Patomäki 1997; Patomäki 2000b; and 2001b and c). Economic
222 Visions of world politics

policies are now largely determined within multilateral systems of governance,

both European and global. Developments associated with globalisation have
also conditioned the electoral politics of the Nordic countries.
Hence, as a complement to the search for plausible national alternatives, it is
increasingly crucial to re-orient and re-define the political project of Norden in
time and space.

Relative persistence of the Nordic model

In terms of the moral and political ideals, the Nordic (or Swedish) model has
been in decline for a quarter of a century. Even still, these ideals are shared by
the majority of the population in all Nordic countries, although they have been
on the defensive, and are often blurred with themes of ‘necessary adjustments’
and ‘European integration’.
In Sweden, the ideal of a Third Way was re-defined and compromised as
early as the 1980s. The reforms of the welfare state that have been taking place
since then have had very little to do with traditional social democratic ideals.
They are much more in line with the neo-liberalist political programme on state
re-structuring and economic liberalisation in accordance with the Washington
consensus (for a recent articulation, see World Bank 1997; and for a slightly
contrasting view, Stiglitz 1998). Consequently, Sweden has been losing its ethico-
political leadership in the Nordic countries.
The unilateral decision by the Swedish government to apply for EC
membership in October 1990 was, in one perspective, perhaps the last nail to
the coffin of Norden. Finland and Norway followed suit. Finland has become an
exemplary EU member that has in many respects renounced its Nordic values,
as well as its association with Swedish ethico-political leadership. The Norwegian
population, however, voted that Norway should stay outside the Union, largely
on the grounds that membership would compromise its Nordic values.
So, was this the end of the Nordic model? Not necessarily. Although the model
itself has been compromised by neo-liberalism, there are signs and indicators of the
persistence of the power of social democratic institutions and values. My argument is
not only about the size and welfare benefits of the state or the average public opinion
in these countries, but also concerns deeper processes. Institutional transformations
are conditioned by public will-formation, which in turn is the outcome of
complicated processes of formation of ways of life, dominant forms of representation
and, indeed, institutions, which structure the decision-making situation faced by
actors and influence trust (Rothstein 1998:134). There is reason to believe that, given
a more adequate understanding of the changing context of Nordic politics, support
for revived Nordic values should clearly outnumber the current—hesitant, confused
and ambiguous—approval of neo-liberalism.
Now, the quantitatively measured role of the state has in fact been rising in
Norden since 1980, despite the partially implemented efforts at welfare state
retrenchment. Table 9.3 gives the figures for total expenditure of central
 Beyond ‘Nordic nostalgia’ 223

government budget as a percentage of GDP in 1980 and 1995 (World Bank

1997:241). In Finland, in particular, there has been a huge leap from 25.2 per
cent to 42 per cent. Obviously, Esping-Andersen (1990:19) is correct in
arguing that a focus on spending can be very misleading. Indeed, this growth
of the state reflects more the costs of the banking crisis, unemployment,
government debt and widening income inequalities than any real rise in the
government final consumption expenditure or welfare entitlements. With the
significant exception of Norway, which is not a member of the EU, there has
been zero or negative growth in real welfare spending, coupled with rising
labour market inequalities and commodification of welfare services, since
1989 (see Clayton and Pontusson 1998).
The outcome of the trends in the labour markets and transformation of the
welfare states has been a significant rise in inequalities, particularly in Sweden
(which remains nonetheless more egalitarian than most countries in the world).
It is equally telling that the former model Sweden has fallen behind Denmark,
Norway and Finland in GDP/capita figures in the late 1990s, although it is still
well above Finland and Denmark in the Human Development Index (see UNDP
1999: particularly Tables 1 and 5, on pages 136 and 149 respectively).
Despite these major qualifications, the very rapid growth of the relative size of
state expenditure to the GDP indicates the relative persistence of the institutions of
the welfare state in a situation that has been characterised by high unemployment
and major financial problems. This persistence of welfare institutions explains the
paradox that, measured in terms of employment, the tradable, open sector has
halved in size since the 1950s, although the exports/GDP ratio has continued to
grow. The explanation is that employment has risen only in the public, non-tradable
sector, particularly in the 1980s and 1990s (see Weiss 1998:95–8).
Equally importantly, it also seems that universalist, egalitarian welfare values
remain shared by a majority of the populations of Norden. It is exactly its
universalism that has made it so appealing and strong:

This formula [of social democracy] translates into a mix of highly

decommodifying and universalistic programs that, nonetheless, are tailored
to different expectations. Thus manual workers come to enjoy rights identical
to those of salaried white-collar employees or civil servants; all strata are
incorporated under one universal insurance system, yet benefits are graduated
according to accustomed earnings. This model crowds out the market,

Table 9.3 Central government budget, total expenditure, as percentage of the GDP
224 Visions of world politics

and consequently constructs an essentially universal solidarity in favour of

the welfare state. All benefit; all are dependent; and all will presumably feel
obliged to pay.
(Esping-Andersen 1990:28)

As the drafters of neo-liberal political programmes also know, unlike the situation
in the nineteenth century when capital also had global reach—even if without
the technological devices of the network society (cf. Castells 1996)—global political
innovations today must confront mass democracy and the broad-based desire
for representation. Democracy, at least in the sense of periodic voting in free
elections, is more institutionalised on a world-scale than ever before (cf. Gill
1997a:1–2). In Norden, national democracy is sedimented in the longue durée of
history: all Nordic countries have been liberal democracies since the early 1920s,
in Sweden and Finland without any interruptions.
The democratic, solidaristic and egalitarian values and interests seem to have
been deeply institutionalised. Measures to lower the level of social benefits, at
high levels of unemployment and with increasing pressure to deregulate labour
market and relations, have tended to be unpopular and hard to implement. In
Sweden, they cost the Conservative-led coalition government of 1991–4 its
support, and the Social Democrats have since been losing support as they
implement welfare cutbacks. After the September 1998 elections, SAP is only
able to form a government with the support of the Greens and the Left Party
(former Communist Party) (Widfeldt 1998).
In Finland, practically all parties have been integrated, to varying degrees,
into the neo-liberal programmes of state restructuring and economic liberalisation,
so there is no real electoral alternative. Apparent consensus among parties
notwithstanding, there are critical discussions within parties about the current
imperatives and options, and, most notably, a clear majority of citizens seem to
have been against the present (more laissez faire and commodified) kind of capitalist
market economy. The popular resistance against deepening integration seems
to have been stronger than in any other EU country (see, for example,
Europinion 1995). Finns have been against a European single currency and
EMU, yet Finland entered the third phase of EMU in 1999. In general, it can
be argued that, given the common sentiment, the acceptance of government
policy seems to indicate a false understanding of the EU processes.
Danish and Norwegian popular resistance to the European integration process
indicates a similar adherence to Nordic values. Denmark first rejected the
Maastricht Treaty and then had it accepted in the second referendum but with
major qualifications. Danish support for European integration has been primarily
confined to its function in facilitating trade. The opposition to the Maastricht
Treaty was based not just on Danish nationalism but also on the defence of the
Scandinavian model of democracy and welfare state (see Lawler 1997:574–8).
In Norway, the majority of voters have twice rejected EC/EU membership,
first in 1972 and then again in 1994. In the latter Norwegian anti-EU campaign,
the debate and referendum was represented as struggles between ‘commonsense
 Beyond ‘Nordic nostalgia’ 225

and capitalism’. This ‘commonsense’ was typically, either directly or indirectly, the
discourse of Nordic social democratic values, exceptionalism and progressive
internationalism, although authentic voices of the countryside and nationalism
were also strong (Lawler 1997:578–81).
In the Nordic democracies, which used to be highly participatory, a cleavage
between the political elite and large segments of the population has emerged.
The mainstream of the elite has been convinced of the necessity of neo-liberal
adjustments and European integration, yet the population tends to be much
more supportive of social democratic values. The conviction of the political
elite has been at least partially based on false beliefs about the situation and the
existing choices. For instance, it is a mistake to try to understand the popular
resistance to the EU only in terms of societal security, identity and nationalism,
as Wæver et al. (1993:24–40) do.5 Given a linear temporality, this mis-
identification tends to reinforce the elite belief that their societies are simply
lagging behind in the evolution of social organisation. This, of course, justifies
attempts to educate and lead or manipulate the bulk of citizenry.
It has indeed been claimed that political elites manipulated the acceptability of EU
membership. Not only did the elites of the parties in office make the decision first,
expecting their political parties, and ultimately, the wider public opinion to join them
later (preferably just before the referendum) (Lawler 1997:583), but they also
manipulated the conditions for public will-formation. For instance, the currencies of
Norway, Sweden and Finland were linked to the ECU in the course of the financial
crisis of 1990–1. This was a clear step towards membership, yet it was neither
supported by the expert opinion of economists nor discussed democratically (Moses
1997). Similarly, the order of the 1994 referendums clearly maximised the
bandwagoning effect in favour of the membership. Had the first referendum been
held in Norway, Sweden, too, would very likely have rejected membership. This, in
turn, would have strengthened the NO (to the EU membership) side also in Finland,
the country most favourable towards membership.
Joining the EU cannot be reduced to enforcement of neo-liberal reforms, but
given the nature of the EU of the 1990s, these are concomitant and mutually
reinforcing aims (as recognised also by Wæver 1994b:7–8). Although the EU is
facing severe legitimation problems everywhere, in Norden, it seems to be particularly
difficult for anti-state, commodifying, free-market neo-liberalism to find sufficient
support for free-market and privatisation reforms. Neo-liberal reforms are therefore
usually conducted under the rubrics of economistic necessity and ‘pragmatic’ or
‘realist’ acquiescence to the new globalised and post-Cold War world. However,
sometimes a reference is also made to more elevating stories that justify neo-
liberalisation and the European integration in terms of a quasi-Hegelian Zeitgeist.
Times are now different; social democratic Norden is mere nostalgia for the past.
Wæver’s article about Nordic nostalgia is a case in point.
Elite discourses notwithstanding, Norden abides in terms of both
institutionalised practices of the welfare state and, in particular, popular values,
partially enabled and sustained by the very institutions created by the social
democratic movement. Finally, it should be noted that Nordic co-operation
226 Visions of world politics

remains more or less intact. Common labour markets, freedom of movement

without passports and common social security systems—already achieved in the
1950s and early 1960s—continue to exist. The Nordic Council still functions,
albeit with a partially transformed form and substance. Norden is still a part of the
identity of these countries. There may have been a crisis of the Nordic model and
identity for some time, but Norden is far from being dead. Certainly, Baltic Sea
identity and co-operation, as was envisaged by Wæver (1992:96–102), have not
replaced it. It is worthy of note that Wæver later qualified his vision:

Identity building is only a pre-condition, it is not a guarantee for success.

Whether the region, in phase two, takes off from this platform is largely
dependent on economic success stories. This part of the story is still open.
Security developments, too, might cause a rupture ending the Baltic venture.
(Wæver 1994b:9)

It is very hard to make any sound judgements on the relative strengths of different
identities (are we talking about opinion polls, widely circulated public discourses,
presuppositions of practices or what?). It nonetheless seems to me that the Baltic
identity has remained much weaker and more vague than the Nordic one by all
accounts. Leaving aside Norway and, more ambiguously, Finland, there have been
few if any economic success stories around the Baltic Sea in the late 1990s.
Despite the relative perseverance of Norden, there is a widespread confusion
and few non-neo-liberal visions about the direction to be taken.

Beyond Nordic nostalgia: globalising

social/democratic vision

It is in the real interest of Norden—the majority of Nordics—to struggle for changing

the conditions, which have made it so difficult to sustain, reconstruct and develop
social democratic ideals. The real interest in this argument is defined in terms of
the existing Nordic identity and values, coupled with an accurate understanding
and plausible explanation of the relative decline of Norden. Moreover, Norden,
understood as a set of ideals constituting an emancipatory process, should be re-
oriented in terms of time and space. More concretely, this process should be at
least partially detached and freed from the notion of a sovereign state. Instead,
the focus should be on making world politics more empowering regarding national
alternatives as well as universally more democratic, socially responsible and
egalitarian. These are indeed existential conditions for the Nordic project itself.

The real interest of Norden

Tautologically, people act in accordance with their perceived interests. However,

are these perceived interests their real interests? When defining their interests, if
 Beyond ‘Nordic nostalgia’ 227

people are misinformed and misled by a false account of the reasons for
occurrence of unwanted things or the absence of wanted or desired things, it is
accurate to say that their perceived interests are not in their real interests.
Although the only way to test a hypothesis about real interests is a sustained
dialogue (under sufficiently ideal conditions), it is both possible to examine the
reasons for illusions, reifications and mystifications, and develop visions informed
by better accounts of the real-world processes.
Actors can be misinformed and misled at different levels. Meta-theoretically,
abstract ontological, epistemological and axiological assumptions—many of which
go unnoticed most of the time—can make concrete interpretations susceptible to
stoicism (indifference to the world), scepticism (denial of the world) or unhappy
consciousness (compensation for the frustrations of the real world found in a
fantasy world), or, in general, to different discursive there-is-no-alternative (TINA)
formations (see Bhaskar 1993, 1994; and, in the IR context, Figure 1.1 of
Chapter 1.
In elite discourses, the problems of the Nordic model since the 1970s have
made sense in terms of theories presupposing irrealist philosophies giving rise
to TINA-formations. These have included positivist economics/rational-choice
theories and sceptical post-modernism, both typically imported from the USA
(although the latter originates in France).6 These have been coupled with more
substantial theories that assumed that neo-liberalist reforms are necessary
adjustments to the irrevocable processes of Europeanisation and globalisation.
Theoretical concepts can also be vastly misleading simply by being rooted in
historical realities other than the one under scrutiny, and by being applied in an
unreflective and uncritical way (see Bourdieu and Waqcuant 1999).
Whatever the variation, the result is always the same: tacit acceptance of the
existing relations of domination and other constraints and absences, which are
jointly causing the unwanted and undesired situation that prohibits further
flourishing of human possibilities. These irrealist elite discourses are then fed to
popular commonsense, particularly through endless, repetitive discussions and
images of the increasingly privatised and commercial media intimately tied to
the globalising—mostly US—networks and corporations (see Herman and
McChesney 1997). Yet, although many understandings have changed because
of these powerful forces, some of the institutions and most of the values of the
‘commonsense’ have tended to remain characteristically social democratic and
Nordic.
What, then, is the real interest of Norden? My argument is that it is in the real
interest of the Nordics to struggle for changing the conditions, which have
made it so difficult to sustain and further develop social/democratic ideals.
Consider the model I have suggested for explaining the relative decline of Norden.
There are two sets of conditions here. First of all, there are those glocal conditions
and tendencies—essentially local and national processes having global conditions—
which have made labour-based power mobilisation, corporatist joint central
regulation and the welfare state itself more difficult to sustain. They have also
turned out to work against attempts to re-introduce ideas about democratic
228 Visions of world politics

control over investments and production process. Second, there are the related,
mostly external global processes that have undermined the embedded liberalism
of the original Bretton Woods system and have made the liberalism of the
Washington consensus appear so indisputable.

Reorienting the Nordic model in time and space

The Nordic model has been based on an extraordinary reliance on the sovereign
state and its capabilities to produce welfare and democratic, internationalist
progress. In an intellectual climate dominated by attempts to overcome the state
sovereignty, Lawler (1997:566) quite rightly calls this Nordic preoccupation
with the state as an ‘intellectually unfashionable attachment to a positive
understanding of the sovereign state which a generic term such as nationalism
cannot fully capture’.
Moreover, the reliance on the state in this way has not produced the typical
exclusive self—other patterns to which the notion of state sovereignty is connected
in political realist/idealist discourse, as depicted by post-structuralist theories
(see, for example, Ashley 1989). Since the early 1950s, the Norden has been the
exemplar for a pluralist security community, where it would be totally absurd
even to imagine any security threat coming from other Nordic countries (i.e.
threat of violence is simply a non-issue), and where all the social problems have
been and are resolved by means of peaceful changes (see, for example, Joenniemi
1996:11–22). These peaceful changes have been in accordance with the social
democratic ideology. At the end of the 1960s, the Nordic countries attempted
to set up NORDEK, a Nordic economic union. Even though the NORDEK
project failed in the early 1970s, Nordic integration remains significantly deeper
in social areas, labour markets and freedom of citizens than that of the EU.
Given the conditions that have been undermining the Nordic model, there is
a need for a new spatio-temporal self-understanding. The neat model of inside
progress and universal welfare, both within the Nordic countries and in the
Norden as a whole, combined with outside foreign policies, might have worked
for a while, under the very specific geo-historical circumstances of 1938–68
(partially disintegrated post-war capitalist world economy, the Bretton Woods
system, etc.). However, ethically, whether or not Nordic foreign policy has been
concerned with security and wealth in a traditional way or with the world’s
social problems by means of active state-led progressivism, it has, at least at
some level, implied an external other (cf. Neumann 1994). Existentially, it has
never come to terms with its own conditions, whether glocal, regional or global.
Starting with the critical realist (CR) idea that emancipation is ‘characteristically
the transition from an unwanted, unnecessary and oppressive situation to a wanted
and/or needed and empowering or more flourishing situation’ (Bhaskar 1993:396;
1994:253), I conclude this chapter by arguing that any meaningful Nordic
emancipation must be connected to regional and global social democratic reforms.
The explanation of the decline of the Nordic model shows that the causes were
 Beyond ‘Nordic nostalgia’ 229

to a significant degree related to globalisation, disembodiment of the embedded

liberalism model of global governance and global relations of domination. Hence,
the transformation to a more wanted, needed, empowering and flourishing situation
can best be achieved by way of attempting to transform the rules and principles
of European and global governance.
The call to transform the rules and principles of global governance should
not be mistaken as a plea to look to the outside. Rather, the best way to go
beyond Nordic nostalgia—longing for those days when there was a direction,
movement and moral leadership of Nordic social democracy—at the dawn of
the twenty-first century is to come to terms with the realisation that the totality
in question is not a nation-state-based society only, with its neat inside/outside
distinctions. Instead, there have been, and increasingly are (again, after the
catastrophic interruptions of the world wars), larger spatio-temporally processes
at work, many of which are not characteristically territorial at all.
This idea is not entirely new. Olof Palme, for example, expressed the idea that in the
long run the difference between national and world politics would disappear. In this
sense, Palme also advocated ‘international democracy’ (see Jerneck 1990:128–9).
However, this traditional social democratic orientation was not based on an analysis of
the glocal causal processes that have co-constituted and co-transformed what Norden
itself is and have conditioned what it can be and how it has been transformed. It is my
argument that emancipatory, globally oriented political action is a condition for the
Nordic ideals to be realised and developed further under new conditions.
However, many of even the more abstract ethico-political principles of the
Nordic model require re-thinking. In particular, the ideal of democratisation
and ‘the passion for equality’, in particular, will also have to assume novel
forms in the regionalising and globalising world. New political innovations are
needed if these kinds of ideals are to be enabled and/or implemented through
regional or global systems of governance.

Beyond Nordic nostalgia 1: acting through the EU

For many Nordic social democrats, nation-states are no longer sufficient as the
framework for political action. Perhaps the EU means more than simply
necessary adjustments and neo-liberalisation? Perhaps it may also open up new
possibilities for regulation and redistribution? Thus, European regionalism has
become a possible solution to the problem of controlling and regulating capitalist
markets and creating more democratic and solidaristic arrangements. On this
basis, Hettne (1993) has argued that there is a new positive understanding of
the ‘logic of stateness vis-a-vis the free play of market forces’, this time articulated
as the ‘pursuit of regionness’.
This strategy addresses at least one of the crucial conditions for the decline
of Norden, namely the trans-nationalisation of production and markets within
Europe. For the Nordic countries, the single market of the EU is the most
important one, whether defined in terms of imports and exports or the activities
230 Visions of world politics

of most trans-national firms. It is also possible to argue that the EMU is a

European response to some of the problems caused by the post-Bretton Woods
system of flexible exchange rates and the globalisation of financial markets.
Yet, there are only few explicit articulations of the content and steps of this
strategy. It has persisted as a somewhat vague emphasis on the ‘social dimension’
and transparency of decision-making of the EU, etc. Therefore, it comes as no
surprise that the Europeanist strategy of Hans-Peter Martin and Harald Schumann
(1997) has been widely discussed in the Nordic countries (also critically). Their
‘ten ideas against the 20/80 society’ of global laissez faire capitalism comprise the
following programmatic notions (Martin and Schumann 1997:289–92):

• Democratising and empowering the EU.

• Realising the European Monetary Union, in particular as a leverage to gaining
control over tax havens and making private capital incomes taxable.
• Extending the EU’s legislative powers to the area of taxation.
• Realising the currency transaction tax (Tobin tax) by the European Central Bank.
• European ecological tax reform.
• The realisation of a European luxury tax.
• Development of European trade unions.
• Strengthening and Europeanising civil society.
• Stopping de-regulation without a social safety net.
• Materialising social and ecological minimum norms for world trade.

These reforms would not amount to transforming the EU into a democratic welfare
state, but the direction is clear. These are steps towards democratising
decisionmaking and empowering the EU with state powers to control and regulate
capitalist market forces, and to take some steps towards European re-distribution.
In fact, to make this vision even more Keynesian, it could also include ideas
about European demand management, based, for example, on devaluation of
the euro, and employer’s contributions and taxes on households. To make it
more egalitarian in a Greenish manner, it has embraced ideas about cutting
down the working hours a person is allowed to work and distributing the
remaining hours more evenly (cf. Lipietz 1996).
The programme could go further in tackling problems in localities (in the
workplace in particular) through the EU by complementing these reforms with
democratic reforms in the labour market and within work organisations. This
could be done, for instance, in accordance with the model of negotiated
environment, which seeks efficiency not by means of ‘flexibility’ and
individualising competition (i.e. by commodification of human beings and social
relations), or by discipline (i.e. by neo-Taylorist means), but by means of
involvement and participation of both trade unions and blue- and white-collar
workers (for some details, see Lipietz 1996; also 1994; and Ryner 1997).
To address the quest to re-think political community and democracy, the
vision should also include considerations about European citizenship and identity
(cf. Minkkinen and Patomäki 1997:111–47). The simultaneous weathering of
 Beyond ‘Nordic nostalgia’ 231

national social and political citizenship and the emergence of the possibility of
post-national citizenship could be taken as a step towards reconstructing political
spaces and communities in Europe. Currently, European citizenship is based
merely on principles of economic liberalism (for criticism, see Habermas 1992).
Any sensible programmatic vision about Europe should include a strategy for
resuscitation of political citizenship in post-national terms.
Finally, to avoid the violence of building a new centre in the world and
excluding alternative possibilities, the new European identity should be able to
incorporate a moment of indecision and to consciously play against the tendency
of creating a new inside/outside dichotomy (cf. Derrida 1992). However, these
Derridean remarks already point towards the limitations of the Europeanist
strategy. Europe is in fact only a small part of the globalising world. A central
puzzle is the EU’s positioning vis-à-vis other parts of the world.
The Europeanist strategy of overcoming the Nordic nostalgia has in fact two
shortcomings. Obviously, it may cause Norden as a distinct identity to disappear
unless it will make Europe appear Norden writ large (cf. Therborn 1997:32–3).
However, more importantly, the conditions for the decline of Norden are not
only EU-European (cf. Figure 9.1). The relevant cultural and discursive relations
of representation and mediation are also global. The basic regulatory framework
of the world economy is devised in Washington and multilateral systems of
governance. Last but not least, the options of productive capital and the spaces
of financial markets are global. Hence, it does not suffice to address these issues
merely by pursuing European regionness.
Indeed, the European integration process has been and is conditioned by
globalisation (see Castells 1998:310–34). Gill (1997b:209) may simplify to the
extreme when he argues that EMU can be seen as a ‘structural adjustment
programme for Europe’, but he has an important point to make. The (neo-
)liberalist rules and principles of global governance are constitutive of the core
principles of the EU. The Treaty of Rome was based on the free-trade principles
of the Bretton Woods system; the Maastricht Treaty and its amendments clearly
reflect the concerns of the post-Bretton Woods Washington consensus. This is
not a coincidence. Global relations of domination and hegemony are at play.7 A
vision of global reforms is thus required as well.

Beyond Nordic nostalgia 2: emancipation by transforming

systems of global governance

Norden could have a potentially important role to play in advocating and realising
global reforms. In contrast to the values of Norden, the globalising world is
characterised by:

• tendencies towards deepening and intensifying commodification of human

beings, social relations, culture and nature (Esping-Andersen 1990:36–7; Amin
1994:31; Harvey 1996:138, 199, 300–2);
232 Visions of world politics

• rapidly growing inequalities both within countries and globally (UNDP 1999);
• increasing de-limitation of the space of democratic self-determination (Dryzek
1996: particularly Chapters 2, 4 and 6; Teivainen 1994; 1997);
• reinforced dominance of the US and non-democratic systems of global
governance (Crawford and Marks 1998; Held 1995:74–120; Patomäki 1999a;
2001b: Chapter 3; and forthcoming);

These trends, coupled with slackening conditions for economic growth, caused
notably also by financial globalisation (Felix 1995:15–32) and consequent
outbursts of major financial crises (see Michie and Smith 1999), imply a deep
and growing legitimisation problem for the Washington consensus-led systems
of governance. Adherence to the values of democratisation, solidarity, de-
commodification and universalistic social policy imply a commitment to major
changes.
Nordic countries have a long tradition of activism in advocating global changes,
often together as in the UN (however, for their differences, see Wiklund 1996).
During the Cold War, the issues dominating the agenda were related to disarmament
and development. Since the mid-1980s, ecological issues (the Brundtland
Commission) and global governance (Commission on Global Governance, co-chaired
by Ingvar Carlsson) have assumed a central place on stage.
The Commission on Global Governance has proposed that there should be
a new pillar of the UN system, namely an economic security council, and that
the UN system should be funded by means of a global tax (see, for example,
Commission on Global Governance 1995; Carlsson 1995). A few days’ revenue
from a global tax on financial transactions would suffice to cover all present
UN debts and annual expenses. The tax would also enable many new global
socio-economic and political projects. Although the UN may already be
domesticated and impoverished by the USA and a new universal organisation
may be needed (see Patomäki forthcoming), steps towards a global tax seem to
be a prerequisite for more adequate principles of governance.
To be justifiable, social/democratic values and purposes must always be
dialogically tested: Are they sufficiently universalisable? They must also gain
concrete ethico-political support to be legitimate (but this in itself is an argument
for democratisation). The following sketchy programmatic vision of global reforms
builds upon the basis of the past and ongoing projects. Of course, there is a
multiplicity of possible strategies of materialising some of the values of Norden in
the emerging global political space. An ideal strategy, based on sufficiently
universalisable values, would tackle a core issue of economic globalisation, include
means to collect major revenues for public political purposes (which must be
determined democratically) and enable possibilities for peaceful and democratic
changes of the rules and principles of systems of global governance.
The re-emergence of global financial markets is perhaps the clearest instance
of economic globalisation. In the Keynesian tradition, it can also be argued as
being the key issue. Unfettered financial markets create collectively irrational
outcomes, increase uncertainty, create conditions for financial crises and compel
 Beyond ‘Nordic nostalgia’ 233

states to follow similar, orthodox economic policies, with strong tendencies for
a deflationary bias and decreasing equality (see Kirshner 1999:315–16, 326;
Patomäki 200 1b, Chapter 3). ‘When the capital development of a country
becomes a by-product of the activities of a casino, the job is likely to be ill-done’
(Keynes 1961 [1936]:159). Like most IMF members, Finland, Norway and
Sweden have had first-hand experience of these effects. They experienced a
series of major banking and currency crises in the early 1990s.
To curb the power of the financial markets and seize some control over
them, at least four measures are needed. First, the horizon of investment decisions
have to be made more long-term; second, indebtedness, inter-dependencies and
actions of the investors have to be regulated; third, some control over financial
off-shores and tax havens must be seized; and, finally, the power of financial
flows has to be curbed by raising transaction costs.
Investment horizons can be lengthened at global, regional or national level,
simply by applying controls over short-term capital inflows (e.g. by minimum
non-remunerated reserve or holding requirements, applied in the 1990s by Chile
and Malaysia). Although, also, national and regional regulatory measures do
help, at least as temporary measures, re-regulation can best be implemented by
global arrangements. In particular, financial off-shores and tax-havens can only
be controlled by means of wide-scale collective action.
The best way to curb global financial flows and raise transaction costs is a
small global tax on all transactions. The idea, suggested by James Tobin in the
1970s (Tobin 1978), is simple: by implementing a low-rate tax on currency
transactions, many speculative movements would become unprofitable and the
financial system more stable. Simultaneously, the tax would yield huge revenues
for the international or global community (see ul Haq et al. 1996).
Realisation of the Tobin tax requires collective political action. As mentioned, the
Tobin tax is also an element in the Europeanist political strategy, as a unilateral tax
imposed by the European Central Bank. The premise is that ‘London and New York
are always going to prevent it’ from being implemented globally (Martin and
Schumann 1997:104). However, Hans-Peter Martin and Harald Schumann have
not explained how the EU could implement the Tobin tax unilaterally. A recent
Working Paper of the European Parliament summarises the standard view:

There would need to be agreement on its [the Tobin tax’s] application in

every financial centre in the world—otherwise foreign exchange markets would
move to ‘tax-free’ jurisdictions. The problems of introducing a with-holding
tax on interest even within the EU alone do not augur well for the chances
of such an international agreement.
(Patterson and Galliano 1999:6)

It thus seems that the minimum requirement is for at least all major financial
centres to be within the sphere of taxation, and that there is a system of global
governance of the tax (see Tobin 1996; Garber 1996).
234 Visions of world politics

However, it is, indeed, possible to modify the Tobin tax so as to enable any
grouping of relevant countries, preferably covering a substantial part of the
global financial markets, to start it at any time. (The proposal is made in Patomäki
1999b, and is developed further in Patomäki 2001b: Chapters 5–7.) Norden
could play a role in initiating this, by setting up the system and also finding
partners in Asia, Russia, non-EU parts of Europe and Latin America. Only for
Finland—the only Nordic country that has entered the third phase of the EMU—
would it be a prerequisite to have the EMU-Europe included from the outset.
The Tobin tax regime can be realised in two phases. In the first, a grouping
of countries would establish an open agreement—any state can join at any time—
and a supra-national body orchestrating the tax and collecting the revenues of
a small underlying transactions tax (10 basis points, at most); much bigger
exchange surcharge triggered by exceptional volatility (1–3 per cent or even
more); and a relatively high tax, up to 20 per cent in the case of non-cooperative
tax havens, on domestic currency lending to non-residents (only to non-residents
who are not yet within the tax regime). This arrangement would solve the tax
evasion problem and is economically sound. In the second phase, which would
be carried out either when all major financial centres and most other countries
have joined the first-phase system, or at latest by, say, year 2010, a universal
Tobin tax at a higher—yet absolutely low—0.5–1.0 per cent rate would be applied.
This arrangement is devised in such a manner that it would build up pressure
for the outsiders to join. In other words, since this model empowers actors to
work for the Tobin tax also without the support of the USA, the UK, the IMF,
etc., it is emancipatory in its implications.
Financial markets’ re-regulation and the establishment of the Tobin tax
constitute tackling a core issue of globalisation. The Tobin tax would also mean
the possibility of collecting major revenues for public political purposes. These
revenues could be used as leverage for other reforms. This, in turn, opens up a
new global ethico-political problematic. A global tax regime with sanctions and
surveillance systems, and potential for huge revenues—hundreds of billions of
US dollars—raises the whole problematic of political theory in the global context:
What are the principles of legitimation of collective organisations? How should
they be assessed in terms of material benefits and their distribution, rightful
authority, justice and democracy? There is a quest to re-think democratic
accountability, representation, participation and citizenship.
The strategy centred around the Tobin tax is both for the autonomy of
states and for new global institutional arrangements. The new institutional
arrangements would not hinder processes of globalisation but would rather
politicise, democratise and re-direct them (see Patomäki 2000a; 2001b). They
would also yield new resources for public political action. Some of these could
be used for global demand inducements to counteract deflationary tendencies.
Like in the Keynesian theory (see Keynes 1961 [1936]:372–84), demand
inducements should also contribute to global redistribution.
However, it is likely that financial re-regulation and the Tobin tax would not
suffice to reverse the tendencies towards increasing de-limitation of the space of
 Beyond ‘Nordic nostalgia’ 235

democratic self-determination, rapidly growing inequalities, deepening and

intensifying commodification and aggravation of global ecological problems.
Simultaneous actions, preferably in collaboration with transnational political
actors, to transform the content of the Multilateral Agreement on Investments
(which was re-discussed in the failed WTO Millennium Round in Seattle, but
is likely to come up again in one form or another) or reforming the UN system,
among other things, would help. To really tackle the conditions of the decline
and the partial neo-liberal transformation of Norden, further programmatic ideas
would be needed, perhaps building upon the leverage of the Tobin tax. However,
particularly assuming some success in empowerment and increased autonomy,
these programmatic ideas may also concern the organisation of very immediate
local sites and related national and regional developmental strategies, in the
spirit of democratic experimentalism (cf. Unger 1998).

Conclusion

In this chapter, I have argued that the Cold War and security-oriented
interpretation of Norden is, if not entirely incorrect, at least partially inadequate
and misleading. The global political-economy dynamic provides a better
explanation of the relative decline of Norden.
There has been a complex interplay of changes in the relations of production;
the dynamics of class identification and welfare state development; and
globalisation as the emergence of new exit options, the erosion of the embedded
liberalism of the original Bretton Woods system and the re-emergence of global
financial markets. These, together with the related cultural and socio-economic
changes, have transformed the structural conditions for labour- and capital-
based power mobilisation in Sweden and, mutatis mutandis, in other Nordic
countries. Together with the end of the Cold War, this interplay has also given
rise to irrealist and economistic stories about the requirements of ‘new times’;
of the gradual changes in the meaning of the notion of the Third Way; and the
neo-liberal framing of new social problems.
Tackling the real conditions of the decline of Norden demands a new social
democratic orientation in time and space. The nation-state can no longer provide
a sufficient framework for progressivist political action, and the socially flavoured
Wilsonian ‘idealism’ appears also to be a somewhat anachronistic basis for
‘progressivist internationalism’.
This is not to argue that politics occurs now only outside sovereign states. Rather,
local and national struggles are essentially connected to regional and global struggles,
and therefore they cannot be taken as separate spheres any more (‘first progressivism
at home, and then exportation of these universalist ideals to the rest of the world’).
Along these lines, I discussed two programmatic visions that should take
Nordics beyond the ‘Nordic nostalgia’. The Europeanist strategy is based on a
new positive understanding of the ‘logic of stateness vis-à-vis the free play of
market forces’, this time articulated as the ‘pursuit of regionness’. Yet, this strategy
236 Visions of world politics

can at best be complementary to a more globalist strategy of carrying out

global social democratic reforms.
Within the confines of this chapter, I could only outline some relevant global
reforms. The Tobin tax is a particularly promising idea, given that it would
tackle a key issue of globalisation, enable the collection of substantial resources
for global purposes (including Keynesian demand-inducements and re-
distribution) and constitute a leverage for other democratic reforms.

Notes

1 Obviously, there is a closer connection to Giddens’s (1998) attempt at ‘the renewal of

Social Democracy’ than to Clinton’s or Blair’s Third Way. While the former takes up
issues—even if vaguely—such as ‘democratising democracy’ and ‘cosmopolitan
democracy’, the latter appears to be little more than an attempt to revitalise elements
of the traditional Anglo-Saxon liberal welfare state (cf. Blair 1998) in the context of
continuing neo-liberalist transformations and deepening commodification of all areas
of social life.
2 Despite many major differences, I assume that Sweden’s welfare state and, at least in
the case of Finland, also neutrality have been the model for the rest of the Nordic
countries. Because of the lack of space, I explicate these differences only when it is
absolutely necessary.
3 For instance, in the early 1950s, Prime Minister Kekkonen justified Nordic co-operation
on the grounds that ‘an alliance of neutrality between the Scandinavian countries
could be seen as a logical continuation of the Pact of Mutual Assistance between
Finland and the USSR. Its significance would lie in the fact that it would remove even
the theoretical threat of an attack on the USSR via Finland’s territory’ (Kekkonen
1970:55–6).
4 There is no room to discuss the causes of this decline. Suffice it to mention that the
new safeguards for the weaker firms and workers may have reduced the incentives to
compete, whereas the leftist challenge to the control over investments might have led
to drastic shortening of the time-horizon of the managers of the corporations (their
positions and future became uncertain). These developments took place in the context
of the global crisis of the early 1970s. The political response of the managerial class
was a right-wing revolt, not proposals for a new Accord.
5 The paradox is that while Wæver is theoretically arguing for de-securitisation, his
explanatory categories tend to reduce everything to security issues.
6 Ryner (forthcoming) has shown how the rather realist and/or Marxist understanding
of the Social democratic leaders of the 1930s and 1940s, concerning the real crisis
tendencies of capitalism, gave way gradually in the 1950s and 1960s to the more
positivist understandings. Many of these later leaders were educated in economics,
not infrequently in the USA. These changes in the background discourse also paved
the way for the characteristically more orthodox solutions of the 1980s.
7 The free-trade core of the EU, based on the principles of GATT, WTO and the IMF,
and the related economic interests of the EU are discussed in Patomäki 1996:22–36.
For a more thorough analysis of the global relations of domination and hegemony,
see Patomäki 1999a; 2000b; 2001b; and forthcoming.
Epilogue

The point of After International Relations is not that relations between states (or
nations) would have disappeared. Many states remain important actors. A few
may be bigger and more capable than ever. Yet two major changes justify the
evocation of the idea about ‘after international relations’. First, one hundred
years after the exhaustion of geographical space by the European empires, and
the height of the pre-1914 capitalist economic inter-dependence, the world has
become even smaller and more vulnerable. The epoch of territorial empires
and geo-politics in the original sense is over. There are now about 200 formally
independent states, and a majority of them claim to be democratic. Globalisation
talk is often geo-historically myopic and ideologically biased. It is nonetheless a
sign of the times. It has become increasingly hard to sustain the illusion of the
inside/outside—or, worse, society/state of war—dichotomy.
Second, the epistemological and ontological ground has shifted. The problematic
articulated by Hume and Kant has become increasingly anachronistic. There is a
realist alternative to the positivist interpretation of the natural sciences. Although
the conceptions, opportunities and problems associated with the rise and expansion
of capitalism may be part of everyday practices more than ever, a majority of social
scientists, outside orthodox economics and its extensions, have renounced the
capitalist conception of man as the basis for social and political theory. Also, the
problem of building a modern state cannot appear the same any longer. Whereas,
for instance, in Africa this project turned out to be disastrous, the European states
have become involved in the long-term process of building an integrated security
community and partial union. Many states’ structures, also outside Europe, have
become regionalised or globalised. Last but not least, as has been argued in this
book, there is now an alternative to the problem of order. Besides criticising the
theoretical underpinnings of this problem, I have proposed that a security
community, whether pluralist or amalgamated, is essentially dependent on the
self-transformative capacity of relevant contexts.
Every problem, whether practical-political or more theoretical, lays on a set
of presuppositions. The aim of After International Relations has been to establish a
way out of the problematic that has puzzled so many great thinkers and scholars
of the past 200 years. By showing how and why the theories based on the
international problematic have failed (Part I), by articulating an alternative,
237
238 Epilogue

critical realist research programme (Part II) and by illustrating how this research
programme can be put to work (Part III), my hope is that this book will enable
better research and other practices. Although this may sound overly ambitious,
it, in fact, builds upon the contribution of numerous philosophers, social theorists
and critical IR and GPE researchers. Simultaneously, it is at best only an
opening.1
Marx (1968 [1852]:96) suggested that everything in history repeats itself.
The first time is a tragedy, the second a farce. Perhaps there will be an epilogue
to international relations too? The problem is that ‘in the past, the worst disasters
could not kill mankind’ (Jaspers 1961:318). Would a replay as a farce be possible?
History is open, and we can only guess what story will fit. Meanwhile, we can
try to learn from past mistakes, past trials and errors, and try to build, on that
basis, better models and stories. We know that only learning and change can
make a global security community possible.

Notes

1 Where this book ends, other books start. Democratising Globalisation. The Leverage of the
Tobin Tax (2001b) is an analysis of the consequences of the re-emergence of global
finance. It explains the instability of financial markets and the role of finance in global
relations of hegemony and domination. It makes a normative case for a particular
concrete Utopia—the Tobin tax—linked to a regulated currency regime, governed by a
democratic organisation. Democratising Globalisation will be followed by Critical Realist
Global Political Economy, a book that attempts to remedy the absence of power in
economics and the absence of economics in GPE. Time and Space in Republican Theory.
Rethinking Participation and Representation in the Global Age has been in the making for
years. It aims to re-think the nature of public political spaces and political participation
and representation under the conditions of the late-modern globalising world. Most
importantly, it purports to do this concretely, not just abstractly.
Bibliography

Acharya, A. (1998) ‘Collective identity and conflict management in Southeast Asia’,

in [Link] and [Link] (eds) Security Communities, Cambridge: Cambridge
University Press, pp. 198–227.
Adler, E. (1997) ‘Imagined (security) communities: Cognitive regions in international
relations’, Millennium: Journal of International Studies 26(2):249–77.
——(1998) ‘Seeds of peaceful change: the OSCE’s security community-building model’,
in [Link] and [Link] (eds) Security Communities, Cambridge: Cambridge
University Press, pp. 119–60.
Adler, E. and Barnett, M. (eds) (1998a) Security Communities, Cambridge: Cambridge
University Press.
——(1998b) A framework for the study of security communities’, in [Link] and
[Link] (eds.) Security Communities, Cambridge: Cambridge University Press,
pp. 29–65.
Alker, Jnr, H. (1966) ‘The long road to international relations theory. Problems of
statistical nonadditivity’, World Politics 18(4):623–55.
——(1974) Are there structural models of voluntaristic social action?’, Quality and Quantity
8(3):199–246.
——(1975) ‘Polimetrics: its descriptive foundations’, in [Link] and [Link] (eds)
Strategies of Inquiry. Handbook of Political Science, Volume 7, Reading, MA: Addison-
Wesley, pp. 139–210.
——(1976) ‘Individual achievements rarely sum to collective progress’ (Autobiographical
Reflections), in [Link] (ed.) In Search of Global Patterns, New York: Free Press,
pp. 38–58.
——(1977) A methodology for research design on interdependence alternatives’,
International Organization 31(1):29–63.
——(1981a) ‘Dialectical foundations of global disparities’, International Studies Quarterly
25(1):9–98.
——(1981b) ‘From political cybernetics to global modeling’, in [Link] and B.M.
Russett (eds) From National Development to Global Community, London: Allen & Unwin,
pp. 353–78.
——(1982) ‘Logic, dialectics, politics: Some recent controversies’, in [Link] Jnr
(ed.) Dialectical Logic for the Political Sciences, Amsterdam: Rodopi, pp. 65–94.
——(1984) ‘Historical argumentation and statistical inference: Towards more
appropriate logics for historical research’, Historical Methods 17(3):164–73.
——(1986) ‘Bit flows, rewrites, social talk: Towards more adequate informational
ontologies’, paper presented in centennial symposium of Tokyo University in
October 1986.
239
240 Bibliography

——(1988a) ‘The dialectical logic of Thucydides’ Melian dialogue’, American Political Science
Review 82(3):805–820.
——(1988b) ‘Politics as political argumentation: Towards regrounding a global
discipline’, paper presented at the XIVth IPSA World Congress, 31 August,
Washington DC.
——(1989) ‘Uppsala fireworks: Data-based thoughts on the origins and possible
obsolescence of the European state system’, in [Link] (ed.) States in Armed
Conflict 1988, Report No. 30 (July), Uppsala University: DPCR, pp. 19–34.
——(1991) ‘Making peaceful sense of the news: Institutionalizing international conflict
management events reporting using frame-based interpretative routines’, paper
presented for the panel of DDIR project at ISA, Vancouver, BC, 19–23 March.
——(1992) ‘The humanistic movement in international studies: Reflections on Machiavelli
and Las Casas’, International Studies Association, 1992 presidential address,
International Studies Quarterly 36(4):347–71.
——(1996) Rediscoveries and Reformulations., Cambridge: Cambridge University Press.
Alker, Jnr, H., Bennett, J. and Mefford, D. (1980) ‘Generalized precedent logics for
resolving insecurity dilemmas’, International Interactions 7(2):165–206.
Alker, Jnr, H. and Biersteker, T. (1984) ‘The dialectics of world order: Notes for a
future archeologist of international savoir faire’ , International Studies Quarterly 28(2):
121–42.
Alker, Jnr, H., Biersteker, T. and Inoguchi, T. (1989) ‘From imperial power balancing to
people’s wars: Searching for order in the twentieth century’, in [Link] Derian and
[Link] (eds) International/Intertextual Relations. Postmodern Readings of the World Politics,
Massachusetts: Lexington Books, pp. 135–62.
Alker, Jnr, H. and Christensen, C. (1972) ‘From causal modelling to artificial intelligence:
The evolution of a UN peace-making simulation’, in [Link] and [Link]
(eds) Experimentation and Simulation in Political Science, Toronto: University of Toronto
Press, pp. 177–224.
Alker, Jnr, H., Duffy, G., Hurwitz, J. and Mallery, J.C. (1991a) ‘Text modeling for
international politics: A tourist’s guide to RELATUS’, in [Link] (ed.) Artificial
Intelligence and International Politics, Boulder, CO: Westview Press, pp. 97–126.
Alker, Jnr, H. and Haas, P.M. (1993) ‘The rise of global ecopolitics’, in [Link] (ed.)
Global Accord: Environmental Challenges and International Responses, Cambridge, MA: MIT
Press , pp. 133–71.
Alker, Jnr, H. and Hurwitz, R. (1980) Resolving Prisoner’s Dilemmas, Cambridge, MA:
APSA.
——(2001) ‘Reasons, causes, games, lessons and evolution in inequitable sequential
prisoner’s dilemmas’, paper presented at the 42nd Annual Convention of the
International Studies Association in Chicago, 20–4 February.
Alker, Jnr, H, Lebedva, M., O’Donnell, S. and Polanyi, L. (1991b) ‘Retelling cold war
stories: Uncovering cultural meanings with linguistic discourse analysis’, prepared
for IVth US—Soviet Workshop on Models and Concepts of Interdependence Between
Nations, Moscow, 26–8 June.
Alker, Jnr, H. and Russett, B. (1965) World Politics in the General Assembly, New Haven,
CT: Yale University Press.
Alker, Jnr, H. and Sylvan, D. (1994) ‘Kuinka diskurssianalyysi voi edesauttaa
politiikkatieteen kehittymistä’ [Political discourse analysis], transl. into Finnish by
[Link]äki, Kosmopolis 24(3):5–25.
Amin, A. (1994) ‘Post-Fordism: Models, fantasies and phantoms of transition’, in [Link]
(ed.) Post-Fordism. A Reader, Oxford: Blackwell, pp. 1–39.
Amin, T. (1985) Nationalism and Internationalism in Liberalism, Marxism and Islam, Islamabad:
 Bibliography 241

International Institute of Islamic Thought.

Apel, Karl-Otto (1984) Understanding and Explanation. Transcendental-Pragmatic Perspective,
trans. [Link], Cambridge, MA: The MIT Press.
Archer, M. (1989) Culture and Agency. The Place of Culture in Social Theory, Cambridge:
Cambridge University Press.
——(1995) Realist Social Theory: The Morphogenetic Approach, Cambridge: Cambridge
University Press.
——(2000) Being Human, Cambridge: Cambridge University Press.
Archer, M.S. and Tritter, J.Q, (2000) Rational Choice Theory. Resisting Colonization, London:
Routledge.
Arendt, H. (1970) On Violence, New York: Harcourt Brace Jovanovich.
——(1973 [1965]) On Revolution, Middlesex: Penguin Books.
Ashley, R. (1980) The Political Economy of War and Peace, London: Frances Pinter.
——(1984) ‘The poverty of neorealism’, International Organization 38(2):225–86.
——(1987) ‘The geopolitics of geopolitical space: Toward a critical social theory of
international polities’, Alternatives XII:403–34.
——(1988) ‘Untying the sovereign state: A double reading of the anarchy
problematique’, Millennium:Journal of International Studies 17(2):27–262.
——(1989) ‘Living on border lines: Man, poststructuralism, and war’, in [Link] Derian
and [Link] (eds) International/Intertextual Relations. Postmodern Readings of the World
Politics, Massachusetts: Lexington Books, pp. 259–322.
Ashley, R. and Walker R.B.J. (eds) (1990) ‘Speaking the language of exile.
Dissidence in international studies’, special issue of International Studies
Quarterly 34(3).
Aristoteles (Aristotle) (1982 [335–22 BC]) Runousoppi [Poetics], trans. [Link],
Helsinki: Otava.
Axelrod, R. (1984) The Evolution of Cooperation, New York: Basic Books.
Bagby, L. (1994) ‘The use and abuse of Thucydides in international relations’,
International Organization, 48(1):131–53.
Ball, T. (1988) Transforming Political Discourse. Political Theory and Critical Conceptual History,
Oxford: Basil Blackwell.
Barkawi, T. and Laffey, M. (1999) ‘The imperial peace: Democracy, force and
globalization’, The European Journal of International Relations 5(4):403–34.
Barnett, M. and Gause, III, F. (1998) ‘Caravans in opposite directions: Society, state,
and the development of community in the Gulf Cooperation Council’, in [Link]
and M. Barnett (eds) Security Communities, Cambridge: Cambridge University Press,
pp. 161–97.
Beardsworm, R. (1996) Derrida and the Political, London: Routledge.
Becker, G. (1976) The Economic Approach to Human Behaviour, Chicago: Harvard
University Press.
Bedford, D. and Workman, T. (2001) ‘The tragic reading of the Thucydidean tragedy’,
Review of International Studies 27(1):51–67.
Beik, P.H. (ed.) (1970) The French Revolution, New York: Harper & Row.
Bernstein, R.J. (1983) Beyond Objectivism and Relativism: Science, Hermeneutics and Praxis,
Oxford: Basil Blackwell.
Bhaskar, R. (1978 [1975]) A Realist Theory of Science, Brighton: Harvester Press.
——(1979) The Possibility of Naturalism. A Philosophical Critique of Contemporary Human Sciences,
Brighton: Harvester Press.
——(1983) ‘Beef, structure and place: Notes from a critical naturalist perspective’ Journal
for the Theory of Social Behaviour 13:82–95.
——(1986) Scientific Realism and Human Emancipation, London: Verso.
242 Bibliography

——(1989) Reclaiming Reality. A Critical Introduction to Contemporary Philosophy, London:

Verso.
——(1991) Philosophy and the Idea of Freedom, Oxford: Blackwell.
——(1993) Dialectic: The Pulse of Freedom, London: Verso.
——(1994) Plato etc.: The Problems of Philosophy and Their Resolution, London: Verso.
——(2000) From East to West. Odyssey of a Soul, London: Routledge.
Blair, T. (1998) ‘A “third way” to achieve modern social democracy’, International
Herald Tribune, 28 September 1998.
Bourdieu, P. (1977) Outline of a Theory of Practice, trans. [Link], Cambridge: Cambridge
University Press.
——(1998) Practical Reason, various translators, Cambridge: Polity.
Bourdieu, P. and Wacquant, L. (1999) ‘On the cunning imperialist reason’, Theory,
Culture and Society 16(1):41–58.
Bull, H. (1969) ‘International theory. The case for classical approach’, in [Link]
and [Link] (eds) Contending Approaches to International Politics, Princeton, NJ:
Princeton University Press, pp. 20–38 [first published in the April 1966 issue of
World Politics].
——(1977) The Anarchical Society: A Study of Order in World Politics, New York: Columbia
University Press.
Bunge, M. (1959) Causality, Cambridge, MA: Harvard University Press.
——(1960) ‘Levels: A semantical preliminary’, The Review of Metaphysics 13(3):
398–406.
——(1963) The Myth of Simplicity: Problems of the Scientific Philosophy, Englewood Cliffs:
Prentice-Hall.
Buzan, B. and Little, R. (1994) ‘The idea of “international system” theory meets
history’, International Poitical Science Review 150(3):231–55.
Callinicos, A. (1985) ‘Anthony Giddens. A contemporary critique’, Theory and Society
14: 133–66.
Camilleri, J. (1990) ‘Rethinking sovereignty in a shrinking world’, in [Link]
and [Link] (eds) Contending Sovereignties. Redefining Political Community,
Boulder, CO: Lynne Rienner, pp. 13–44.
Campbell, D. and Shapiro, M.J. (eds) (1999) Rethinking Ethics and World Politics,
Minneapolis: University of Minnesota Press.
Carlsnaes, W. (1992) ‘The agency structure problem in foreign policy analysis’,
International Studies Quarterly 36(3):245–70.
——(1993) ‘On analysing the dynamics of foreign policy change: A critique and recon-
ceptualization’, Cooperation and Conflict 28(1):5–30.
——(1994) ‘In lieu of a conclusion: Compatibility and the agency-structure
issue in foreign policy analysis’, in [Link] and [Link] (eds) European
Foreign. Policy. The EC and Changing Perspectives in Europe, London: Sage, pp.
274–87.
Carlsson, I. (1995) ‘The United Nations: A time to reform’, Foreign Policy Magazine,
fall. Available online at: [Link]
Castells, M. (1996) The Rise of the Network Society. The Information Age: Economy, Society
and Culture, Volume I, Oxford: Blackwell.
——(1998) End of Millennium. The Information Age: Economy, Society and Culture, Volume III,
Oxford: Blackwell.
Chase-Dunn, C. (1989) Global Formation. Structures of the World-Economy, Oxford: Basil
Blackwell.
Choucri, N. and North, R.C. (1975) Nations in Conflict. National Growth and International
Violence, San Francisco: [Link] & Company.
 Bibliography 243

Clayton, R. and Pontusson,J. (1998) ‘Welfare-state retrenchment revisited. Entitlement

cuts, public sector restructuring, and inegalitarian trends in advanced capitalist
countries’, World Politics 51(1) (October):67–98.
Cohen, I. (1987) ‘Structuration theory and social praxis’, in [Link] and [Link]
(eds) Social Theory Today, Oxford: Polity Press, pp. 273–308.
Collier, A. (1994) Critical Realism. An Introduction to Roy Bhaskar’s Philosophy, London: Verso.
——(1999) Being and Worth, London: Routledge.
Commission on Global Governance (1995) Our Global Neighbourhood. The Report,
Oxford: Oxford University Press.
Connolly, W.E. (1987) Politics and Ambiguity, Madison: The University of Wisconsin Press.
——(1993) Political Theory and Modernity, Ithaca: Cornell University Press.
Connor, W. (1985) Thucydides, Princeton: Princeton University Press.
Cooper, D.E. (1985) ‘Metaphors we live by’, in [Link] (ed.) Philosophy and Practice,
Cambridge: Cambridge University Press, pp. 43–58.
Cornford, F. (1965 [1907]) Thucydides Mythohistoricus, London: Routledge & Kegan Paul.
Cox, R.W. (1987) Production, Power and World Order: Social Forces in the Making of History,
New York: Columbia University Press.
Crawford, J. and Marks, S. (1998) ‘The global democratic deficit: An essay in
international law and its limits’, in [Link], [Link] and M.Köhler (eds) Re-
imagining Political Community. Studies in Cosmopolitan Democracy, Oxford: Polity Press,
pp. 72–90.
Culler, J. (1983) On Deconstruction. Theory and Criticism after Structuralism, Ithaca, NY:
Cornell University Press.
Der Derian, J. (1987) On Diplomacy, Oxford: Basil Blackwell.
——(1989) ‘The boundaries of knowledge and power in international relations’, in
[Link] Derian and [Link] (eds) International/Intertextual Relations. Postmodern Readings
of World Politics, Lexington: Lexington Books, pp. 3–10.
——(1992) Antidiplomacy. Spies, Terror, Speed, and War, Oxford: Blackwell.
Derrida, J. (1978) Writing and Difference, trans. [Link], London: Routledge & Kegan
Paul.
——(1982) Margins of Philosophy, trans. [Link], London: Harvester Wheatsheaf.
——(1988) Limited Inc., various translators, Evanston, IL: Northwestern University
Press.
——(1992) The Other Heading. Reflections on Today’s Europe, trans. P.-[Link] and M.B.
Naas, Indianapolis: Indiana University Press.
——(1998 [1967]) Of Grammatology, corrected edition, trans. [Link], Baltimore: Johns
Hopkins University Press.
Dessler, D. (1989) ‘What’s at stake in the agent—structure debate?’, International
Organization 43(3):441–73.
Deutsch, K.W. et al. (1957) Political Community and the North Atlantic Area. International
Organization in the Light of Historical Experience, Princeton, NJ: Princeton University
Press.
——(1963) The Nerves of Government. Models of Political Communication and Control, New
York: The Free Press.
Dillon, M. (1999) ‘The scandal of the refugee: Some reflections on the “inter” of
international relations and continental thought’, in [Link] and [Link]
(eds) Moral Spaces. Rethinking Ethics and World Politics, Minneapolis: University of
Minnesota Press, pp. 92–124.
——(2000) ‘Poststructuralism, complexity and poetics’, Theory, Culture and Society 17(5): 1–26.
Donnelly, J. (1991) ‘Thucydides and realism’, paper presented at the 1991 ISA
Conference, Vancouver, 20–3 March.
244 Bibliography

Doty, R.L. (1997) ‘Aporia: A critical exploration of the agent-structure problematique in

international relations theory, European Journal of International Relations 3(3):365–92.
Doyle, M.W. (1995) ‘Liberalism and world Politics revisited’, in [Link] (ed.)
Controversies in International Relations Theory. Realism and the Neoliberal Challenge, New
York: St Martin’s Press, pp. 83–106.
Dreyfus, H. (1991) Being-in-the-World. A Commentary on Heidegger’s ‘Being and Time’, Division
I, Cambridge, MA: The MIT Press.
Dreyfus, H. and Rabinow, P. (1982) Michel Foucault. Beyond Structuralism and Hermeneutics,
Hertfordshire: Harvester Wheatsheaf.
Dryzek, J. (1996) Democracy in Capitalist Times. Ideals, Limits and Struggles, Oxford: Oxford
University Press.
Durkheim, E. (1969 [1938]) ‘Social facts’, in [Link] (ed.) Readings in the Philosophy
of Social Sciences, New York: The MacMillan Company, pp. 245–53.
Eberwein, W.-D. (1995) ‘The future of international warfare: Toward a global security
community?’, International Political Science Review 16(4):341–60.
Edmunds, L. (1975) Chance and Intelligence in Thucydides, Harvard: Harvard University
Press.
Elster, J. (1978) Logic and Society, London: John Wiley & Sons.
Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism, Cambridge: Polity.
Europinion (1995) ‘Results of monthly surveys of European public opinion’ 6 (October).
Available online at: [Link]
[Link].
Evangelista, M. (1995) ‘The paradox of state strength: Transnational relations,
domestic structures, and security policy in Russia and the Soviet Union’, International
Organization 49:1–38.
——(1999) Unarmed Forces: The Transnational Movement to End the Cold War, Ithaca: Cornell
University Press.
Falk, R. (1975) A Study of Future Worlds, New York: The Free Press.
Farrar, C. (1988) The Origin of Democratic Thinking. The Invention of Politics in Classical
Athens, Cambridge: Cambidge University Press.
Fay, B. (1990) ‘“Critical” Realism?’, Journal for the Theory of Social Behavior 20(1): 33–
41.
Felix, D. (1995) ‘Financial globalization versus free trade: The case for the Tobin tax’
(November), Geneva: UNCTAD Discussion Paper No. 108.
Foucault, M. (1979 [1975]) Discipline and Punish. The Birth of the Prison, trans. from
French by [Link], London: Penguin Books.
——(1982) ‘The subject and power’, in [Link] and [Link] (eds) Michel Foucault.
Beyond Structuralism and Hermeneutics, Hertfordshire: Harvester Wheatsheaf, pp. 208–28.
——(1986 [1971]) ‘Nietzsche, genealogy and history’, trans. from French by
[Link] and [Link], in [Link] (ed.) The Foucault Reader, London:
Penguin Books.
——(1991a) ‘Politics and the study of discourse’, in [Link], [Link] and [Link]
(eds) The Foucault Effect. Studies in Governmentality, London: Harvester Wheatsheaf,
pp. 53–72.
——(1991b) ‘Questions of method’, in [Link], [Link] and [Link] (eds)
The Foucault Effect. Studies in Governmentality, London: Harvester Wheatsheaf,
pp. 73–86.
——(1991c) ‘Governmentality’, in [Link], [Link] and [Link] (eds) The Foucault
Effect. Studies in Governmentality, London: Harvester Wheatsheaf, pp. 87–104.
Frochlich, M. and Oppenheimer, J. (1984) ‘Beyond economic man: Altruism, egalitarianism
and difference maximising’, The Journal of Conflict Resolution 28(1):3–24.
 Bibliography 245

Gadamer, H.-G. (1975 [1965]) Truth and Method, trans. [Link] and [Link]
(eds), USA: The Seabury Press.
——(1977) Philosophical Hermeneutics, [Link] ed. [Link], Berkeley, LA: University
of California Press.
Gaddis, J.L. (1992) ‘International relations theory and the end of the Cold War’,
International Security 17(3):5–58.
Galbraith, J. (1967) The New Industrial State, Boston: Houghton-Mifflin.
Galtung, J. (1977) Methodology and Ideology. Theory and Methods of Social Research Vol.1,
Copenhagen: Christian Ejlers.
——(1988) Methodology and Development. Essays in Methodology, [Link], Copenhagen:
Christian Ejlers
——(1990) ‘Cultural violence’, Journal of Peace Research 27(3):291–305.
——(1996) Peace by Peaceful Means, London: Sage, with PRIO.
Garber, P. (1996) ‘Issues of enforcement and evasion in a tax of foreign exchange
transactions’, in [Link] Haq, [Link] and [Link] (eds) The Tobin Tax. Coping with
Financial Volatility, Oxford: Oxford University Press, pp. 129–12.
Garrett, G. (1998) Partisan Politics in the Global Economy, Cambridge: Cambridge
University Press.
Garst, D. (1989) ‘Thucydides and neorealism’, International Studies Quarterly 33(1):3–
27.
George, J. (1989) ‘International relations and the search for thinking space: Another
view of the third debate’, International Studies Quarterly 33(3):269–79.
Giddens, A. (1976) New Rules of Sociological Method. A Positive Critique of Interpretative
Sociologies, London: Hutchinson.
——(1977) Studies in Social and Political Theory, London: Hutchinson.
——(1979) Central Problems in Social Theory, Los Angeles: University of California Press.
——(1981) A Contemporary Critique of Historical Materialism, Vol.1, Berkeley: University of
California Press.
——(1983) ‘Comments on the theory of structuration’, Journal for the Theory of Social
Behaviour 13:75–80.
——(1985a) ‘Marx’s correct view on everything’, Theory and Society 14:167–74.
——(1985b) The Nation-State and Violence. Volume Two of the Contemporary Critique of Historical
Materialism, Oxford: Polity Press.
——(1987a) Social Theory and Modern Sociology, Oxford: Polity Press.
——(1987b) The Nation-State and Violence. Volume Two of a Contemporary Critique of Historical
Materialism, Cambridge: Polity Press.
——(1998) The Third Way: The Renewal of Social Democracy, Cambridge: Polity.
Gill, S. (ed.) (1993) Gramsci, Historical Materialism and International Relations, Cambridge:
Cambridge University Press.
——(1997a) ‘Analysing new forms of authority: New constitutionalism, panopticism
and market civilization’, paper presented at the conference Non-State Actors and
Authority in the Global System, Warwick University, 31 October–1 November
1997.
——(1997b) ‘An emu or an ostrich? EMU and neoliberal globalisationlimits and
alternatives’, in [Link] and [Link]äki (eds) The Politics of Economic and
Monetary Union, Dordrecht: Kluwer, pp. 207–31.
Gill, S. and Law, D. (1988) Global Political Economy. Perspectives, Problems and Policies, New
York: Harvester Wheatsheaf.
——(1989) ‘Global hegemony and the structural power of capital’, International Studies
Quarterly 33(4):475–99.
Gills, B. and Palan, R. (1994) ‘Introduction: The neostructuralist agenda in
246 Bibliography

international relations’, in [Link] and [Link] (eds) Transcending the State—Global

Divide: A Neostructuralist Agenda in International Relations, Boulder CO: Lynne Rienner,
pp. 1–13.
Gilpin, R. (1984) ‘The richness of the tradition of political realism’, International
Organization 38(2):287–304.
——(1987) The Political Economy of International Relations, Princeton: Princeton University
Press.
Goede, M. de (1999) ‘Finance, gambling and speculation: The politics of risk and
uncertainty’, paper presented at the LSE Conference ‘The Politics of Financial
Globalisation’, 20 September.
Gonzales, G. and Hagg ard, S. (19 9 8) ‘The United States and Mexico:
A pluralistic security community?’, in [Link] and [Link] (eds)
Security Communities, Cambridge: Cambridge University Press, pp.
2 9 5–3 3 2.
Gould, C. (1978) Marx’s Social Ontology. Individuality and Community in Marx’s Theory of
Social Reality, Cambridge, MA: The MIT Press.
Greenwood, J.D. (1987) ‘Scientific psychology and hermeneutical psychology:
Causal explanation and the meaning of human action’, Human Studies
10:171–204.
Grunberg, I. (1990) ‘Exploring the “myth” of hegemonic stability’, International
Organization 44(4):431–77.
Guzzini, S. (1994) ‘Power analysis as a critique of power politics: Understanding
power and governance in the second Gulf War’, unpublished Ph.D. thesis
(December), Florence: European University Institute.
Haack, S. (1974) Deviant Logic, Cambridge: Cambridge University Press.
Haas, E.B. (1986) Why We Still Need the United Nations: The Collective Management of
International Conflict, 1945–1984, University of Berkeley: Berkeley Institute of
International Studies.
Habermas, J. (1979) Communication and the Evolution of Society, trans. [Link], Boston:
Beacon Press.
——(1984 [1981]) The Theory of Communicative Action. Volume One, trans. [Link],
London: Heinemann.
——(1987 [1981]) The Theory of Communicative Action. Volume Two, trans. [Link],
Cambridge: Polity Press.
——(1987) The Philosophical Discourse of Modernity, trans. [Link], Cambridge: Polity
Press.
——(1992) ‘Citizenship and national identity: Some reflections on the future of Europe’,
Praxis International 12(1):1–19.
Harmes, A. (1998) ‘Institutional investors and the reproduction of neoliberalism’,
Review of International Political Economy 5(1):92–121.
Harré, R. (1961) Theories and Things, London: Sheed & Ward.
——(1964) Matter and Method, London: Macmillan.
——(1970) Principles of Scientific Thinking, London and Basingstoke: Macmillan.
——(1979) Social Being, Oxford: Basil Blackwell.
——(1986) Varieties of Realism. A Rationale for the Natural Sciences, Oxford: Basil Blackwell.
Harré, R. and Madden E.H. (1975) Causal Powers, Oxford: Blackwell.
Harré, R. and Secord, P. (1972) The Explanation of Social Behaviour, Oxford: Blackwell.
Harvey, D. (1990) The Condition of Postmodernity. An Enquiry into the Conditions of Cultural
Change, Oxford: Blackwell.
——(1996) Justice, Nature and the Geography of Difference, Oxford: Blackwell.
Hayat, K. (1999) ‘International financial liberalisation: Implications for social
 Bibliography 247

democratic economic policy’, paper presented at the Politics of Financial

Globalisation Conference at the London School of Economics, London, 20
September.
Heidegger, M. (1990 [1962]) Being and Time, trans. [Link] and [Link],
Oxford: Basil Blackwell [first published in German in 1927].
Held, D. (1995) Democracy and the Global Order. From the Modern State to Cosmopolitan
Governance, Oxford: Polity Press.
Herman, E. and McChesney, R. (1997) The Global Media. The New Missionaries of Corporate
Capitalism, London: Cassell.
Hertz, J. (1957) ‘Rise and demise of the territorial state’, World Politics IX:473–93.
——(1968) ‘The territorial state revisited: Reflections on the future of the nation-state’,
in [Link] (ed.) International Politics and Foreign Policy, New York: The Free Press,
pp. 76–89.
Hesse, M. (1966) Models and Analogies in Science, Notre Dame: University of Notre
Dame Press.
Hettne, B. (1993) ‘Neo-mercantilism: The pursuit of regionness’, Cooperation and Conflict
28(3):211–32.
Hirschman, A.O. (1977) The Passions and Interests. Political Arguments for Capitalism Before
Its Triumph , Princeton: Princeton University Press.
——(1991) The Rhetoric of Reaction. Perversity, Futility, Jeopardy, Cambridge, MA: The
Belknap Press of Harvard University Press.
Hobbes, T. (1986 [1651]) Leviathan, [Link] (ed.), Harmondsworth: Penguin
Books.
Hoffmann, S. (1959) ‘International relations: The long road to theory’, World Politics
11: 346–77.
Holden, B. (ed.) (2000) Global Democracy. Key Debates, London: Routledge.
Hollis, M. (1988) ‘Say it with flowers’, in [Link] (ed.) Meaning and Context. Quentin
Skinner and his Critics, Cambridge: Polity Press, pp. 135–46.
Hollis, M. and Smith, S. (1990) Explaining and Understanding International Relations,
Oxford: Clarendon Press.
——(1991) ‘Beware of gurus: Structure and action in international relations’, Review of
International Studies 17(4):393–410.
——(1992) ‘Structure and action: Further comment’, Review of International Studies 18(2):
187–8.
Honnef, K. (1994) Contemporary Art, Köln: Benedict Taschen.
Horkheimer, M. (1989 [1937]) ‘Traditional and critical theory’, trans. from German
by M.J.O’Connell, in Critical Theory. Selected Essays, New York: Continuum, pp.
188–243.
Hostettler, N. and Norrie, A. (2000) ‘Do you like soul music?’, Alethia 3(2):2–8.
Hume, D. (1975 [1777]) Enquiries Concerning Human Understanding and Concerning the
Principles of Morals, ed. [Link], Oxford: Clarendon Press [reprinted from the
1777 editionfirst editions published in 1748 and 1751].
——(1994 [1777]) Political Writings, ed. [Link] and [Link], Indianapolis:
Hackett Publishing Company [includes extracts from the Treatise and nineteen
essays from the 1777 edition of Essays and Treatises of Several Subjects].
Hurrell, A. (1998) ‘An emerging security community in South America?’, in [Link]
and M. Barnett (eds.) Security Communities, Cambridge: Cambridge University Press,
pp. 228–64.
Ikenberry, J.G. (1992) ‘A world economy restored: Expert consensus and the Anglo-
American postwar settlement’, International Organization 46:289–321.
Israel, J. (1979) The Language of Dialectics and the Dialectics of Language, Harvester Press:
248 Bibliography

Brighton.
Isaac, J.C. (1987) Power and Marxist Theory. A Realist View, Ithaca: Cornell University
Press.
——(1990) ‘Realism and reality: Some realistic reconsiderations’, Journal for the Theory
of Social Behaviour 20(1):1–31.
Jabri, V. and Chan, S. (1996) ‘The ontologist always rings twice: Two more stories
about structure and agency in reply to Hollis and Smith’, Review of International
Studies 22(1): 107–10.
Jaspers, K. (1961) The Future of Mankind, trans. [Link], Chicago: The University of
Chicago Press.
Jerneck, M. (1990) ‘Olof Palme—en international propagandist’, in [Link] and
[Link] (eds) Socialdemokratin of svensk utrikespolitik. Från Branting till Palme,
Stockholm: Utrikespolitiske institut, pp. 121–42.
Joenniemi, P. (1996) ‘Norden: A community of asecuriry?’, Copenhagen: Working
Papers 6, Copenhagen Peace Research Institute.
Kant, I. (1956 [1788]) Critique of Practical Reason, trans. [Link], New York: Macmillan.
——(1983) Perpetual Peace and other Essays, Indianapolis: Hackett Publishing Company
[‘Idea for a universal history’, 1784‘May be true in theory, but is of no practical
use’, 1793‘Perpetual peace’, 1795].
Kaplan, M. (1969) ‘The new great debate. Traditionalism vs. science in international
relations’, in [Link] and [Link] (eds) Contending Approaches to International
Politics, Princeton, NJ: Princeton University Press, pp. 39–61 [first published in the
October 1966 issue of World Politics].
Katzenstein, P. (1985) Small States in World Markets, Ithaca, New York: Cornell University
Press.
Keat, R. (1981) The Politics of Social Theory. Habermas, Freud and the Critique of Positivism,
Oxford: Basil Blackwell.
Keat, R. and Urry, J. (1975) Social Theory as Science, London: Routledge & Kegan Paul.
Kekkonen, U. (1970) ‘The common neutrality of the Nordic countries’, a speech
presented in Helsinki on 23 January 1952 (the pyjama pocket speech), in
[Link] (ed. [Link]) Neutrality. The Finnish Position, London: Heinemann,
pp. 53–6.
Keohane, R.O. (1986) ‘Theory of world politics: Structural realism and beyond’, in
[Link] (ed.) Neorealism and Its Critics, New York: Columbia University Press,
pp. 158–203.
Keynes, J. (1961 [1936]) The General Theory of Employment, Interest and Money, London:
Macmillan.
Kirshner, J. (1999) ‘Keynes, capital mobility and the crisis of embedded liberalism’,
Review of International Political Economy 6(3):313–37.
Klein, B.S. (198 8) ‘Hegemony and strategic culture: American power
protection and alliance deference polities’, Review of International Studies,
14:133–48.
——(1990) ‘How the West was one: Representational politics of NATO’, International
Studies Quarterly 34:311–25.
Kokaz, N. (2001) ‘Moderating power: A Thucydidean perspective’, Review of
International Studies 27(1):27–49.
Koskenniemi, M. (1989) From Apology to Utopia. The Structure of International Legal Argument,
Helsinki: Finnish Lawyer’s Publishing.
Krasner, S. (1982) ‘Structural causes and regime consequences: Regimes as intervening
variables’, International Organization 36(2):185–205.
Kratochwil, F. (1993) ‘The embarrassment of changes: Neo-realism as the science of
 Bibliography 249

realpolitik without polities’, Review of International Studies 19(1):63–80.

Kuhn, T. (1970) The Structure of Scientific Revolutions, 2nd. enlarged edn, Chicago: The
University of Chicago Press.
Lakoff, G. and Johnson, M. (1980) Metaphors We Live by, Chicago: The University of
Chicago Press.
Latsis, S. (1972) ‘Situational determinism in economies’, British Journal of the Philosophy
of Science 23:207–45.
Lawler, P. (1997) ‘Scandinavian exceptionalism and European union’, Journal of Common
Market Studies 35(4):565–94.
Lawson, T. (1997) Economics and Reality, London: Routledge.
Lefebvre, H. (1991) The Production of Space, trans. [Link]-Smith, Oxford: Blackwell
[first French edition 1974].
Lenin, V. (1968 [1895]) ‘Frederick Engels’, in KarlMarx and Frederick Engels. Selected
Works in One Volume, edited by Progress Publishers, Moscow, London: Lawrence &
Wishart, pp. 15–22.
Leonard, P. (1997) Postmodern Welfare. Reconstructung an Emancipatory Project, London:
Sage.
Lijphart, A. (1981) ‘Karl [Link] and the new paradigm in international relations’,
in [Link] and [Link] (eds) From National Development to Global Community. Essays
in Honour of [Link], Boston: Allen & Unwin, pp. 233–51.
Lindbeck, A., Colander, P., Persson, T, Peterson, O., Sandmo, A., Swedenborg,
B. and Thygesen, N. (1994) Turning Sweden around, Cambridge, MA: MIT
Press.
Linge, D.E. (1977) ‘Editor’s introduction’, in H.-[Link] Philosophical Hermeneutics,
Berkeley: University of California Press, pp. xi–lviii.
Linklater, A. (1990) Beyond Realism and Marxism. Critical Theory and International Relations,
London: Macmillan.
——(1998) The Transformation of Political Community, Cambridge: Polity.
Lipietz, A. (1994) ‘Post-Fordism and democracy’, in [Link] (ed.) Post-Fordism. A Reader,
Oxford: Blackwell, pp. 338–58.
——(1996) ‘Social Europe: The post Maastricht challenge’, Review of International Political
Economy 3(3):369–79.
MacIver, R.M. (1964) Social Causation, New York: Harper Torchbooks.
Mackie, J.L. (1976) ‘Causes and conditions’, in [Link] (ed.) The Nature of Causation,
Urbana: University of Illinois Press, pp. 308–4–4.
MacPherson, C.B. (1964) The Political Theory of Possessive Individualism. Hobbes to Locke,
Oxford: Oxford University Press.
——(1973) Democratic Theory. Essays in Retrieval, Oxford: Clarendon Press.
Manicas, P.T. (1987) A History and Philosophy of the Social Sciences, Oxford: Basil Blackwell.
Margolis, J. (1986) Pragmatism without Foundations. Reconciling Realism and Relativism,
Oxford: Basil Blackwell.
Martin, H.-P. and Schumann, H. (1997) Globaliseringsfä[Link] på demokratti och
välfärd, trans. from German to Swedish [Link], Stockholm: Symposion.
Marx, Karl (1968 [1852]) ‘The eighteenth brumaire of Louis Bonaparte’, in Karl
Marx and Frederick Engels. Selected Works in One Volume, edited by Progress Publishers,
Moscow, London: Lawrence & Wishart, pp. 96–179.
May, T. (2000) ‘A future critique? Positioning, belonging and reflexivity’, European
Journal of Social Theory 3(2):157–73.
Mearsheimer, J. (1990) ‘Back to future: Instability in Europe after the Cold War’,
International Security 14(4):42–64.
Meiggs, R. (1972) The Athenian Empire, London: Oxford at the Clarendon Press.
250 Bibliography

Michie, J. and Smith, J. (eds) (1999) Global Instability. The Political Economy of World
Economic Governance, London: Routledge.
Minkkinen, P. and Patomäki, H. (1997) Yhdentymisen ja rahaunionin politiikkaa.
Suomija Emu globaalissa poliittisessa taloudessa [Politics of Integration and
Monetary Union Finland and E M U in the Global Political Economy],
Helsinki: Like.
Morgenthau, H. (1946) Scientific Man versus Power Politics, Chicago: University of Chicago
Press.
Moore, Jnr, B. (1969) Social Origins of Dictatorship and Democracy. Lord and Peasant in the
Making of the Modern World, Penguin: Harmondsworth.
Moses, J. (19 97) ‘Trojan horses: Putman, E C U linkage and the E U
ambitions of Nordic elites’, Review of International Political Economy
4(2):382–415.
Naess, A. (1972) The Pluralist and Possibilist Aspect of the Scientific Enterprise, London :
Allen & Unwin.
Neufeld, M. (1993a) ‘Reflexivity and international relations’, Millennium: Journal of
International Studies 22(1):53–76.
——(1993b) ‘Interpretation and the “Science” of international relations’, Review of
International Studies 19(1):39–61.
Neumann, I. (1994) ‘A region-building approach to Northern Europe’, Review of
International Studies 20(1):53–74.
Nicolson, H. (1939) Diplomacy, London: Oxford University Press.
Norrie, A. (1991) Law, Ideology and Punishment, Dordrecht: Kluwer.
——(1993) Crime, Reason and History, London: Butterworths.
——(2000) Punishment, Responsibility and Justice. A Relational Critique, Oxford: Oxford
University Press.
Olafson, F.A. (1979) The Dialectic of Action. A Philosophical Interpretation of History and the
Humanities, Chicago: The University of Chicago Press.
Ollman, B. (1971) Alienation. Marx’s Conception of Man in Capitalist Society, Cambridge:
Cambridge University Press.
Onuf, N.G. (1989) World of Our Making. Rules and Rule in Social Theory and International
Relations, South Carolina: University of South Carolina Press.
——(1995) ‘Levels’, European Journal of International Relations 1(1):35–58.
——(1998) The Republican Legacy in International Thought, Cambridge: Cambridge
University Press.
Ordershook, P.C. (1986) Game Theory and Political Theory. An Introduction, Cambridge:
Cambridge University Press.
Palmer, R.E. (1969) Hermeneutics. Interpretation Theory in Schleiermacher, Dilthey, Heidegger
and Gadamer, Evanston: Northwestern University Press.
Palonen, K. (1987) ‘The art of the possible on the periphery. [Link] and Urho
Kekkonen in the realpolitik tradition’, in [Link], and [Link] (eds) Transformation
of Ideas on a Periphery. Political Studies in Finnish History, Helsinki: Finnish Political
Science Association, pp. 98–115.
Patomäki, H. (1989) ‘A metatheoretical and methodological critique of neorealism:
Towards a fourth debate in international relations research’, Occasional Papers 8,
1989, Turku: Political Science, University of Turku.
——(1992a) ‘What is it that changed with the end of the cold war? An analysis of the
problem of identifying and explaining change’, in [Link] and [Link] (eds)
The End of the Cold War. Evaluating Theories of International Relations, Dordrecht: D.
Reidel, pp. 179–225.
——(1992b) ‘From normative Utopias to political dialectics: Beyond a deconstruction of
 Bibliography 251

the Brown—Hoffman debate’, Millennium: Journal of International Studies 21(1):53–75.

——(1993) ‘Kyllä tämä on niin saatanan surkeaa? Tutkielma Suomen 1990-luvun
poliittista identiteettiä ja intressejä rakentavista Kekkos-tarinoista’ [Kekkonen-
narratives: The way they constitute the political identity and interests of Finland
in the 1990s], Kosmopolis 23(2):49–78.
——(1994) ‘From power balancing and collective security to a pluralist security
community in the Middle East’, in [Link], [Link] and [Link] (eds) The
Search for Peace in the Mediterranean Region. Problems and Prospects , Malta: Mireva, pp.
72–109.
——(1996) Vain kauppakumppaneita? EU, Venäjä ja EU: n ulkosuhteiden rakenteistuminen [Only
Trading Partners? EU, Russia and the Structuration of the External Relations of
the EU], Helsinki: Haasteita 11, UPI.
——(1999a) ‘Good governance of the world economy?’, Alternatives 24(1):119–42.
——(1999b) ‘The Tobin tax: How to make it real’, Helsinki: UPI Working Papers 13, 99.
——(2000a) ‘The Tobin tax: A new phase in the politics of globalisation’, Theory, Culture
and Society 17(4):77–91.
——(2000b) ‘Republican sphere and the governance of global political economy’, in
[Link] and J.-[Link] (eds) Value Pluralism, Normative Theory and IR, London:
Macmillan, pp. 160–95.
——(2001a) ‘The challenge of critical theories: Peace research at the start of the new
century’, Journal of Peace Research 38(6):723–37.
——(2001b) Democratising globalisation. The leverage of the Tobin tax, London: Zed Books.
——(2001c) ‘From East to West: Emergent global philosophies—beginnings of the end
of the Western dominance?’, Theory, Culture and Society (18):6.
——(forthcoming) ‘Kosovo and the end of the UN?’, in [Link] Ham and [Link]
(eds) Mapping European Security after Kosovo, Manchester: Manchester University Press.
Patomäki, H. and Wight, C. (2000) ‘After postpositivism? The promises of critical
realism’, International Studies Quarterly 44(2):213–37.
Patterson, B. and Galliano, M. (1999) ‘The feasibility of an international “Tobin
tax”’, Luxembourg: European Parliament Working Paper 107, Economic Affairs
Series.
Perrow, C. (1986) ‘Economic theories of organization’, Theory and Society 15:11–45.
Polanyi. K. (1957 [1944]) The Great Transformation. The Political and Economic Origins of
Our Time, Boston: Beacon Press.
Pontusson, J. (1992) The Limits of Social Democracy. Investment Politics in Sweden, Ithaca,
New York: Cornell University Press.
Popper, K. (1971) ‘Addenda’, in [Link], Open Society and Its Enemies. Vol. 2, Princeton:
Princeton University Press, pp. 369–96.
Putnam, H. (1982) ‘Why reason can’t be naturalized?’, Synthese 52(1):3–23.
——(1989) Reason, Truth and History, Cambridge: Cambridge University Press.
Rapoport, A. (1960) Fights, Games and Debates, Ann Arbor: University of Michigan Press.
Rapoport, A. and Chammah, A. (1965) Prisoner’s Dilemma. A Study in Conflict and
Cooperation, Ann Arbor, MI: University of Michigan Press.
Rescher, N. (1977) Dialectics. A Controversy-Oriented Approach to the Theory of Knowledge,
New York: State University of New York Press.
——(1987) Forbidden Knowledge, Dordrecht: [Link].
Ricoeur, E (1981) Hermeneutics and the Human Sciences, ed. and trans. [Link],
Cambridge: Cambridge University Press.
——(1984) Time and Narrative, vol. 1, trans. from French [Link] and [Link],
Chicago: The University of Chicago Press.
——(1988) Time and Narrative, vol. 3, trans. from French [Link] and [Link],
252 Bibliography

Chicago: The University of Chicago Press.

——(1992) Oneself as Another, trans. by [Link], Chicago: The University of Chicago
Press.
Risse-Kappen, T. (1994) ‘Ideas do not float freely: Transnational coalitions,
domestic structures, and the end of the Cold War’, International Organization
48:185–214.
Rochester, J.M. (1995)‘The UnitedNations in anewworld order: Revivingthe theory
andprac-tice of international organization’, in [Link] (ed.) Controversies in
International Relations Theory. Realism andtheNeoliberal Challenge, New York: St Martin’s
Press, pp. 199–221.
Roig, C. (1982) ‘Leninist dialectic seen from the point of view of Kenneth Burke’s
dialectic: Theory, analysis and ideology in Lenin’s speech’, in [Link] (ed.)
Dialectical Logic for Political Science, Amsterdam: Rodopi, pp. 49–64.
Rorty, R. (1979) Philosophy and the Mirror of Nature, Princeton, New Jersey: Princeton
University Press.
——(1987) ‘Pragmatism and philosophy’, in [Link], [Link] and [Link]
(eds) Philosophy. End or Transformation?, Cambridge, MA: The MIT Press.
——(1991) Essays on Heidegger and Others. Philosophical Papers Vol.2, Cambridge: Cambridge
University Press.
Rothstein, B. (1996) The Social Democratic State. The Swedish Model and the Bureaucratic
Problem of Social Reforms, Pittsburgh: University of Pittsburgh Press.
——(1998) Just Institutions Matter. The Moral and Political Logic of the Universal Welfare State,
Cambridge: Cambridge University Press.
Ruggie, J. (1982) ‘International regimes, transactions and change: Embedded liberalism
in the postwar economic order’, International Organization 36(2):195–231.
——(1983) ‘Continuity and transformation in the world polity. Toward a neorealist
synthesis’, World Politics 35(2):261–85.
——(1993) ‘Territoriality and beyond. Problematizing modernity in international
relations’, International Organization 47(1):139–74.
——(1995) ‘At home abroad, abroad at home: International liberalisation and domestic
stability in the new world economy’, Millennium: Journal of International Studies 24(3):
507–26.
——(1998) Constructing the World Polity. Essays on International Institutionalization, London:
Routledge.
Rummel, R. (1995) ‘Democracy, power, genocide, and mass murder’, Journal of Conflict
Resolution 39(1):3–26.
Ryner, M. (1994a) ‘Assessing SAP’s economic policy in the 1980’s: The “third way”,
the Swedish model and the transition from Fordism to post-Fordism’, Economic and
Industrial Democracy 15(3):385–428.
——(1994b) ‘The political economy of globalisation and the crisis of the Swedish model’,
paper presented at the Annual Meeting of the BISA, University of York, York,
UK, 19–21 December.
——(1997) ‘Maastricht convergence and post-Fordist transition in Sweden and
Germany’, a mimeo, published in 1998 in International Journal of Political Economy
28(2): 85–123.
——(forthcoming) Capitalist Restructuring, Globalisation and the Third Way: Lessons from the
Swedish Model, London: Routledge.
Sayer, A. (1992) Method in Social Science. A Realist Approach, 2nd edn, London: Routledge.
——(1995) Radical Political Economy. A Critique, Oxford: Blackwell.
——(2000) Realism and Social Science, London: Sage.
Schroeder, P. (1994) ‘Historical reality vs. neo-realist theory’, International Security 19(1):
 Bibliography 253

108–48.
Schürmann, R. (1984) ‘Deconstruction is not enough: On Gianni Vattimo’s
call for “weak thinking”’, Graduate Faculty Philosophy Journal 10(1)
(spring):165–77.
Searle, J. (1983) ‘The world turned upside down’, The New York Review of Books, October
27:74–9.
——(1996) The Construction of Social Reality, London: Penguin.
Sharp, G. (1984) The Politics of Nonviolent Action. Part One: Power and Struggle, Boston:
Porter Sage Publishers.
Shatzki, T. (1988) ‘The nature of social reality’, Philosophy and Phenomenological Research
XLIX:239–60.
Shotter, J. (1983) “‘Duality of structure” and “intentionality” in an ecological
psychology’, Journal for the Theory of Social Behaviour 13:19–43.
Singer, J. (1961) ‘The level-of-analysis problem in international relations’, in [Link]
and [Link], (eds) The International System, PrincetonPrinceton University Press,
pp. 77–92.
——(1969) ‘The incomplete theorist: Insight without evidence’, in [Link] and J.N.
Rosenau (eds) Contending Approaches to International Politics, Princeton, NJ: Princeton
University Press , 62–86.
Skinner, Q (1988) Meaning and Context. Quentin Skinner and his Critics, ed. [Link],
Cambridge: Polity Press.
Smith, M.J. (1986) Realist Thought from Weber to Kissinger, London: Lousiana State
University Press.
Smith, S. (1994) ‘Foreign policy theory and the new Europe’, in [Link] and
[Link] (eds) European Foreign Policy. The EC and Changing Perspectives in Europe, London:
Sage, pp. 238–73.
de [Link], G.E.M. (1972) The Origins of the Peloponnessian War, New York: Cornell
University Press.
Stiglitz, J. (1998) ‘More instruments and broader goals: Moving toward the post-
Washington consensus’, Wider Annual Lectures 2, UNU WIDER (World Institute
for Development Economics Research Helsinki), available online at: http://
[Link]/[Link].
Suganami, H. (1983) ‘The structure of institutionalism: An anatomy of
British mainstream international relations’, International Interactions
7(5):2363–81.
Sylvan, D. (1991b) ‘Insides, outsides, and walls in Classical Greece’, paper presented
at the XVth World Congress of the IPSA, 21–5 July, Buenos Aires.
——(1991a) ‘The qualitative—quantitative distinction in political science’, Poetics Today
12(2):267–86.
Sylvan, D. and Glassner, B. (1985) A Rationalist Methodology for the Social Sciences, Oxford:
Basil Blackwell.
Taylor, M. (1976) Anarchy and Cooperation, London: John Wiley & Sons.
Teivainen, T. (1994) ‘Politics of economism and the containment of changes in the
global political economy’, in [Link]äki (ed.) Peaceful Changes in World Politics,
Tampere: Tapri Research Report No. 71, pp. 81–121
——(1997) ‘The independence of the European Central Bank: Implications for
democratic governance’, in P Minkkinen and [Link]äki (eds) The Politics of Economic
and Monetary Union, Dordrecht: Kluwer, pp. 55–75.
Therborn, G. (1997) ‘Europe in the 21st century: The world’s Scandinavia?’, Irish
Studies in International Affairs 8:21–34.
Thucydides (1843 [1628/c. 400 BC]) The History of the Grecian War, [Link], in
254 Bibliography

‘The English Works of Thomas Hobbes’, collected and edited by Sir

[Link], vol. VIII–IX, London: John Bohn.
——(1952 [c. 400 BC]) The Peloponnesian Wars, trans. [Link], Chicago, London and
Toronto: Great Books of the Western World 6, Encyclopaedia Britannica.
——(1964 [c. 400 BC] Peloponnessolaissota, trans. into Finnish by [Link], Juva: WSOY.
——(1972 [1954/400 BC]) The Peloponnessian War, trans. [Link], London: Penguin
Books.
Tobin, J. (1978) ‘A proposal for international monetary reform’, Eastern Economic Journal
4: 153–9.
——(1996) ‘Prologue’, in [Link] Haq, [Link] and [Link] (eds) The Tobin Tax. Coping
with Financial Volatility, Oxford: Oxford University Press.
Tomlinson, J. (1999) Globalization and Culture, Oxford: Polity Press.
Toulmin, S. (1992) Cosmopolis. The Hidden Agenda of Modernity, Chicago: The University
of Chicago Press.
Tuathail, G. (1996) Critical Geopolitics, London: Routledge.
Tuck, R. (1992) ‘Hobbes’, in [Link] et al. Great Political Thinkers. Machiavelli, Hobbes,
Mill, Marx, Oxford: Oxford University Press, pp. 107–238.
Tuori, K. (1988) Oikeuden rationaalisuus. Max Weber ja Jürgen Habermas oikeuskehityksen
tulkitsijoina [Legal Rationality. Max Weber and Jürgen Habermas as Interpreters of
Juridical Development], Helsinki: Helsingin yliopiston julkisoikeuden laitoksen
julkaisuja, D:4.
ul Haq, M., Kaul, I. And Grunberg, I. (eds) (1996) The Tobin Tax Coping with Financial
Volatility, Oxford: Oxford University Press.
UNDP (1999) Human Development Report 1999, Oxford: Oxford University Press.
Unger, R.M. (1984) Passion. An Essay on Personality, New York: The Free Press.
——(1987a) Social Theory: Its Situation and Its Task, Cambridge: Press Syndicate of the
University of Cambridge.
——(1987b) False Necessity. Anti-Necessitarian Social Theory in the Service of the Radial Democracy,
Cambridge: Cambridge University Press.
——(1998) Democracy Realized. The Progressive Alternative, London: Verso.
Wæver, O. (1992) ‘Nordic nostalgia: Northern Europe after the Cold War’, International
Affairs 68(1):77–102.
——(1994a) ‘Resisting the temptations of post foreign policy analysis’, in [Link]
and [Link] (eds) European Foreign Policy: The EC and Changing Perspectives in Europe,
pp. 238–73.
——(1994b) ‘Balts, books and Brussels. Nordic identity and cooperation after the Cold
War’, Copenhagen: COPRI Working Papers 11.
——(1996) ‘European security identities’, Journal of Common Market Studies 34(1):103–
32.
——(1998) ‘Insecurity, security and asecurity in the West European non-war
community’, in [Link] and [Link] (eds) Security Communities, Cambridge:
Cambridge University Press, pp. 69–118
Wasver, O., Buzan, B., Kelstrup, M. and Lemaitre, P. (1993) Identity, Migration and the
New Security Agenda in Europe, London: Pinter.
Walker, R.B.J. (1987) ‘Realism, change and international political theory’, International
Studies Quarterly, 31(1):65–86.
——(1989) ‘History and structure in the theory of international relations’, Millennium:
Journal of International Studies 18(2):163–83.
——(1991) ‘State sovereignty and the articulation of political space/time’, Millennium:
Journal of International Studies 20(3):445–61.
——(1993) Inside /Outside: International Relations as Political Theory, Cambridge: Cambridge
 Bibliography 255

University Press.
Wallensteen, P. and Sollenberg, M. (2000) ‘Armed conflict 1989–98’, Journal of Peace
Research 37(5):635–49.
Waltz, K. (1959) Man, the State and War, New York: Columbia University Press.
——(1979) Theory of International Politics, New York: Addison-Wesley.
——(1986) ‘Reflections on “theory of international politics”: A response to my critics’,
in R. Keohane (ed.) Neorealism and Its Critics, New York: Columbia University Press,
pp. 322–46.
Walzer, M. (1986) ‘The politics of Michel Foucault’, in [Link] (ed.) Foucault. A Critical
Reader, Oxford: Basil Blackwell, pp. 51–68.
Weber, M. (1969 [1949]) ‘Ideal types and theory construction’, in [Link] (ed.)
Readings in the Philosophy of Social Sciences, New York: The Macmillan Company, pp.
496–507.
——(1978 [1956]) Economy and Society. An Outline of Interpretative Sociology, ed. [Link] and
[Link], Berkeley: University of California Press [original published in German
in 1921].
——(1985 [1919]) ‘Politics as vocation’, in [Link] (ed.) Weber: Selections in
Translation, trans. [Link], Cambridge: Cambridge University Press, pp. 212–
25.
Weiss, L. (1998) The Myth of the Powerless State, New York: Cornell University Press.
Wendt, A. (1987) ‘The agent-structure problem in international relations theory’,
International Organization 41(3):335–70.
——(1991) ‘Bridging the theory/meta-theory gap in international relations’, Review of
International Studies 17(4):383–92.
——(1992a) ‘Anarchy is what states make of it’, International Organization 46(2):391–
425.
——(1992b) ‘Levels of analysis. Agents and structures: part III’, Review of International
Studies 18(2):181–5.
——(1994) ‘Collective identity formation and the international state’, American Political
Science Review 88(2):384–96.
——(1999) Social Theory of International Politics, Cambridge: Cambridge University Press.
Widfeldt, A. (1998) ‘Swedish election report. The Swedish parliamentary election of
1998’, ECPR News 10(1):18–19.
Wight, C. (1999) ‘They shoot dead horses don’t they? Locating agency in the agent—
structure problematique’, European Journal of International Relations 5(1):109–42.
——(2001) ‘Linking agents and structures to levels of analysis’, paper presented at the
42nd Annual Convention of the ISA, Chicago, 20–4 February.
Wight, M. (1966) ‘Why is there no international theory’, in [Link] and [Link]
(eds) Diplomatic Investigations. Essays in the Theory of International Politics, London: George,
Allen & Unwin, pp. 17–34.
Wiklund, L. (1996) ‘Norden 1 FN’, Internasjonal Politikk 54(1):75–83.
Williamson, O. (1975) Markets and Hierarchies, New York: Free Press.
Winch, P. (1971 [1958]) The Idea of Social Science and its Relatio n to Philosophy, London:
Routledge & Kegan Paul.
Wittgenstein, L. (1979) On Certainty, ed. [Link] and [Link] Wright,
Oxford: Basil Blackwell.
World Bank (1997) World Development Report 1997. The State in a Changing World, Oxford:
Oxford Univeristy Press.
von Wright, G.H. (1971) Explanation and Understanding, London: Routledge & Kegan
Paul.
Young, O. (1969) ‘Aron and the whale: A Jonah in theory’, in [Link] and
256 Bibliography

[Link] (eds) Contending Approaches to International Politics, Princeton, NJ:

Princeton University Press, pp. 129–43.
——(1989) ‘Politics of international regime formation’, International Organization 43(3):
349–75.
——(1991) ‘Political leadership and regime formation: On the development of
institutions in international society’, International Organization 45(3):281–308.
Index

absence 7, 8, 18, 41, 129 project 61, 65, 66; and ‘emancipatory
abstractions 127–8, 130 humanism’ 46–7; four phases of 44–7;
action: communicative 105, 110, 111; on history writing 58–9; and
complex causes for 81, 91–2, 119; and importance of openness 44; on Jesus
context 119–20; dependent on reality story 56–7; and learning process 43–4;
144–5; dramaturgical 110, 111; limitations of 64–6; on mathematical
freedom of 119–20; and internal and social science 135; and modelling
external relations 117; meaningful 119; approaches 56–61; as President of the
need for theory of 116; normatively ISA 64, 66; on social ontology 99; on
regulated 110; possibilities 131; and temporal motivation 140; theory of
rules 102–3, 108; rationalisation of conflicts 44 on Thucydides 167, 173,
151; relationship with power 112–14; 191; United Nations studies 44–6,
relationship with structures 109, 47–8, 54, 68; work on collective
116–17; social 108–12, 116, 136, 173; dilemmas (Prisoner’s Dilemma) 51–6,
strategic 111, 152; teleological 110, 65; and world history theorising 61–3
111; transformative 120 amalgamated community 194, 195, 196,
actors: as components of a causal 201–2
complex 87–8, 119 and context 119, ambiguity 103–4, 106–7
120; co-operation among 111; geo- ambivalence 103–4
historical constitution of 91, 137–8, Amin, Tahir 61
139, 140; ‘internal’ and ‘external’ 80; analogy 127, 128, 130, 197
knowledge of 126, 131; linked by analysis: historical vs. structural 91
understanding (Habermas) 104–5; and anarchy 33, 74, 191
misinformation 227; and practical Archer, Margaret 7
wisdom 159; in security community Archimadus, King 187
200–1; social 102; states as 71–2 Arendt, Hannah 143, 156
Adler, Emanuel 198, 203–4, 209 Aristotle 99, 102, 159, 188, 191
Adorno, Theodor 6 arms races 127
Africa 237 artificial intelligence 48, 61
agency 151, 152 Ashley, Richard 13, 33, 61, 69, 71
agency-structure problem 72, 73, 74 astronomy 22
aitia 179 Athens: aggressive imperialism 179, 182,
Alcibiades 180, 192 183–8, 190; conduct of Peloponnesian
Alker, Hayward R. 11–13, 41, 43–69 War 168, 179; debates and voting
collection and analysis of data 47–51; 184–5; development of 192; eventual
and critical realism 44; criticism of defeat 188; and the gods 192; loss of
scientific and statistical practice 45–6, legitimate leadership 182; and Melos
47–51; on data-coding procedure s episode 170–1, 174–9, 180, 181,
134; ‘Dialectics of World Order’ 182–8; as misleading concept 180
257
258 Index

realpolitik of 176, 178, 185, 189; Bull, Medley 39, 40, 65–6
resources 182 Bunge, Mario 82
atomism 22, 29; entities 3, 135; Butterfield, Herbert 36
individuals 29, 31, 32; social 25; states
78; world 78 cannibals 29
Attica 184 capital accounting 24
Augustine, St 29, 34, 36 capitalism 21, 133, 237; alienation and
Axelrod, Robert 55 atomism of 203; central to IR 38;
control strategies (EU) 229, 230;
Bacon, Francis 22, 30 development of 24, 25–6, 29, 83–4,
balance of power 21, 28, 31–2, 42, 167; 134; and globalisation 38, 79, 84, 118,
Alker on 62–3 224; link with positivism 25–6; and
Bank for International Settlements 205 problem of order 30–1; as social
banking crises 221, 223, 233 system 118; and strategic action 111;
banks, international 134, 230, 233 Swedish model 217
Barnett, Michael 198, 209 Carlsnaes, Walter 71, 72, 87
Becker, Gary 4 Carlsson, Ingvar 232
behaviouralism 43, 45, 73, 88, 95 cartography 26, 27
Bennett, James 54–5 Catholics, Roman 27
Bentham, Jeremy 2, 25 causal complexes 78–9, 119–20; and
Berkeley, George 2, 22 iconic models 123, 126, 129–33; and
Bernstein, Richard 123 time and space 137
Bhaskar, Roy 5–7, 9, 14, 35, 162; on causal efficacy 151
action 116; Alker and 64, 65; on causal hypotheses 132
atoms 135; on communicative action causal powers 88
105; concepts of structure 100, causal stories 87–9
114–15; on empirical invariances 76; causality: Alker’s ontological shift from
explanatory emancipation concept 16, mechanical model 48; critical realist 5,
144, 152–5, 156, 158, 160; on 8, 15, 76; Deutsch and 197; dialectical
Horkheimer 18; on interference with 104; and empirical invariances 75, 76;
object by subject 151, 152; on IR and hierarchies 83; Humean 48, 75,
levels 94; and normative discourse 76, 78; Kantian moral reason cut off
158; on philosophy and realism 8; from 4; and levels-of-analysis problem
on positivism 26, 49; on reason 88; on 72–3, 74–5; positivist 3; and post hoc ergo
science 100; on social structure 101–2, propter hoc-fallacy 76, 135; and
108; on society 150; TMSA 99, 102, quantitative data 134–5
108; on truth 144, 145, 146–7, 148, causation: Hume’s scepticism about 23
153–4, 158, 160; on vagueness and cause: ancient Greek attitude to 172; as an
ambiguity 103–4 INUS-condition 76–7, 91, 92, 101,
Biersteker, Thomas 61, 62, 63 102, 108, 119, 131, 132
Bill of Rights (1689) 27 centre-periphery structures 60, 128, 138–9
Blair, Tony 210, 236 Chammah, A. 55
Bolshevik Revolution 113, 143 Chan, Stephen 71
book-keeping 24, 134 change 43–4
Boulding, Kenneth 43 Chase-Dunn, Christopher 64
Bourdieu, Pierre 7 Chile 45, 233
bourgeois man 10, 17 Choucri, Nazli 50, 69
Braudel, Ferdinand 58 Christianity 21, 29–30, 34, 36
Bretton Woods system 3, 16, 85, 205, citizenship 230–1
211, 215, 217, 228, 230, 231 Cleon 184–5, 186
British Institutionalism 78 Clinton, Bill 210, 236
Brundtland Commission 232 Club of Rome 58, 60
Bruno, Giordano 22 coding practices 50–1, 57
 Index 259

Cold War 113, 193, 194, 232; Alker and de-commodification 215–16
44; end of 3, 16, 76, 212, 213, 214; deconstruction 41
Norden and 211–13, 214, 215, 217, 235; deductive-nomological model 75, 91
and political realism 39, 43; second 215 Delian league 186
collective identity 198–9 democracy: global 224, 229; liberal 224
collective security 62–3 democratic peace hypothesis 130, 193–4
Collier, Andrew 7, 154, 158 democratic peace theory 3, 93–4
colonialism 27, 28, 29, 84 Denmark 211, 221, 223, 224
Commission on Global Governance 232 Dependencia school 60
commodification 231see also dependency relations 60; as levels 83–4
de-commodification dependent variable 73, 74
communication 136, 173 depth 81–2; of explanation 91
Communist Party 113 Derrida, Jacques 7, 40–1, 154–5, 157, 231
comprehensive views 150 Descartes, René 2, 22
computational hermeneutics 57 desecuritisation 208
computers 57, 66, 68, 94 Dessler, David 75, 99
Comte, Auguste 35, 76, 99 Deutsch, Karl 15, 55, 60, 127; on learning
configurating 140 43, 152–3; on power 185–6; and
constant conjunctions 76 security community concept 41,
constructivism 4, 39, 108 194–5, 196–200, 201–2, 208, 209
context:of action 119–20; and deviant cases 51
quantitative data 134; see also social dialectical argumentation 125
context dialectical causality 104
contradictions 104, 106–7 dialectical exchange 71
contrastive demi-regularity 129 dialectics: definition 173; development of
Copernicus, Nicolas 22 consciousness 54; formalised (Rescher)
Cornford, F. 172–3 106–7, 174–6; and Melian dialogue
correspondence: as metaphor 145; and 173–9; of open, stratified systems
theory of truth 145–6 117–18
critical realism (CR) 1, 4–5; agency/ ‘Dialectics of World Order’ project 61, 65, 66
structure debate 71; components of 77; dilemmas, collective 12, 51–6
core notions of context and complex Diodotus 185
119; the ‘critical’ aspect diplomacy 28
9–10; development of 5–7; and discourse, normative 158, 160
epistemological relativism 8–9; and discourse analysis 39
judgemental rationalism 9; domination, systems of 199, 201, 204,
methodology 13–15; and ontological 205, 232
realism 8; see also specific aspects double hermeneutics 14, 103, 109, 111,
criticism 158 160; of iconic modelling 123–42
Cunning of Reason 34 dualism 64, 89
currency transaction tax 17, 230, 233–4 Durkheim, Emile 99, 108
cybernetics 48, 60, 127
East Timor 139
Darwin, Charles 34 EC see European Community
Darwinism 34 ecology 232, 235
data: coding procedures 50–1, 130; econometrics 135
collection and analysis 47–51, 134; economic cooperation 4
and epistemology 49–50; nature of elite discourses 225, 227
11–12; and ontology 48, 51; primary emancipation 9, 10, 16, 228–9; and
14; qualitative 135–7; quantitative security community 203; see also
133–5; theory-dependence of 90; explanatory emancipation
verbal 48 emancipatory explanation 144
Declaration of the Rights of Man 2 emancipatory humanism 46–7
260 Index

emergence 8, 18, 41, 129 feminism 4

empirical data see data Finland 221, 222, 223, 224, 225, 233,234,
empirical evidence 132, 133 236; and Cold War 113, 211; and
empirical invariances 3, 75, 76 political realism 113
empirical studies 72 First World War 11, 37–8, 113
emplotment 140 Fordism 84, 218
Emshoff model 52, 54 foreign policy: and agency-structure
EMU see European Monetary Union problem 72, 74; changes and
Engels, Friedrich 42 continuity in 55; and history 55, 58;
Enlightenment 34 and levels-of-analysis problem 80;
epistemological relativism 8–9, 14, 77, Nordic 228
144–5, 197 Forrester, Jay 58, 60
epistemology 9–10, 103, 104: and data Foucault, Michel 55, 65, 100, 209; on
49–50 panopticism 52–3; on power 112–13,
Esping-Andersen, Gøsta 215–16, 223–4 114; on truth 143
EU see European Union France 7, 28, 39
Europe: contrasted to Norden 212, 214; Frankfurt School 6, 99
development of modern states 27–8; free trade 39, 206, 231, 236
emergence of international relations French Revolution 26
2; empires 237; increasing functionalism, sociological 127
destructive powers 84; power- fusion of horizons 136–7
balancing system 62–3; problem of
definition 214; regionalism 229; Gadamer, Hans-Georg 99, 136–7, 142
seventeenth century breakdown of Galbraith, John Kenneth 111
order 29–30 Galileo 22
European Central Bank 230, 233 Galtungjohan 7, 43, 51, 67, 161
European Community: Nordic countries General Agreement on Tariffs and Trade
and 212, 213, 220, 221, 222 (GATT) 205
European Monetary Union 224, 230, geo-historical processes 147, 158, 159–60,
231, 234 197; dependence of truth on 144, 145,
European Union: capitalism control 146; see also spacetime
strategies 229, 230; citizenship and geometry 24
identity 230–1; demand management geo-politics 138
230; as exemplar of modernisation Germany 39
214; free trade 231, 236; and global Giddens, Anthony 5, 6, 210, 236; concept
investment 206; integration 231; of power 100; on context 119;
labour reform and trade unions 230; definition of structures 114, 116–17;
Nordic countries and 221, 222, 223, definition of system 118; on double
224–5, 229–31; tax reform 230; hermeneutics 90, 103, 123; on internal
taxation 233 and external relations 108; on problem
evolutionary models 127 of order 104; on social relations 115;
existential hypotheses 131 theory of structuration 99, 100, 112,
explanation 79, 132; causal 72–3, 74–5; 113, 114
emancipatory 144; social scientific Gill, Stephen 231
150–l; teleological 88 global political economy 118, 218–22; see
explanatory critique 151–2 also International Political Economy
explanatory emancipation 16, 144, 152–5, globalisation 227–8, 237: and capitalism
156, 158, 160; see also real interests 38, 79, 84, 118, 224; commodification
external relations 102, 104, 107–12, 117 trends 231; de-limitation of self-
determination 232; ecology 232, 235;
failure 159 economy 84, 204–5, 231; European
fairy tales 55, 58, 59 integration 231; finance 232–3;
false beliefs 155, 215 financial markets 205, 233–4;
 Index 261

governance 229, 231–5; inequalities Hollis, Martin 13, 71–5, 89, 91, 99
232; political innovations 224; ‘horizon’ 142
regulation 233; relations of domination Horkheimer, Max 6, 18
204, 205, 232; and security community Human Development Index 223
concept 203–5; social relations 215; humanism 29
strategy 17; surveillance 139; taxation Hume, David 2, 21, 25, 29; ampliative
232, 233–5, 236 approach of 125–6; as an anachronism
Glorious Revolution (1688) 27 237; on balance of power 28, 31–2, 42;
Gold Standard 84 on causality 48, 75, 76, 78; Christian
Gorbachev, Mikhail 3, 113, 214 realists and 36; conception of human
government debt 223 nature 36; critique of ontology 22–4;
graphical argumentation techniques 174, on external relations 104; and ‘great
176, 177 debates’ of international relations 36–8;
Greenwood, John 103 on growing power of Britain 28; on
Grunberg, Isabelle 68 justice 25, 30; liberals and 34;
Gulf Cooperation Council 195 Morgenthau on 35; Nietzsche and 36;
political career 28; posivitism 11, 22–4;
Habermas, Jürgen 5, 18, 41, 54; on post-positivism and 4; and problem of
rationality 162; theory of social action order 30; on reality 23; on reason 32;
109–10, 121; theory of ‘system’ and scepticism about causation 23; on
‘Lebenswelt’ 112; theory of selfishness of man 25; on Thucydides
understanding 89–90, 104–5 167, 190
Halley, Edmond 22 Hurwitz, Roger 54, 55, 65
Harré, Rom 5, 14, 64, 88; on truth 147, 148 hypocrisy 156
Harvey, David 138
haute finance 84 ‘icon’ 129
Hegel, Georg 34, 35, 37 iconic augmentation 141
Hegelian theory: development of iconic modelling 14, 66, 78–9, 80, 100,
consciousness 54; master-slave 109, 123, 173, 189, 190; ampliative
argument 213 and reductive approaches 124–6;
Heidegger, Martin 133, 142 analogies, metaphors and metonyms
Hellas 168, 180–2, 186–7, 188, 190; 128–9, 130–1, 132; and causal
notion of honour 183 complexes 123, 126, 129–33; double
Hellenes 180, 192 hermeneutics of 123–42; empirical
hermeneutic circle 136 evidence 133; homoeomorphic 108,
hermeneutics 35, 68, 89, 99, 102, 123, 127; hypotheses of 131–3; idealisations
150; computational 57; second level and abstractions 127–8, 130; of Melos
problem 136; understanding 72; see episode 179–80, 189, 190, 192;
also double hermeneutics paramorphic 126–7qualitative
Herz, John 84 evidence 135–7quantitative data
Hettne, Björn 229 133–5; space and time in 137–41;
hierarchical organisations 83, 111, 118 teleiomorphic 108–9, 110, 123, 127–9;
Hirschman, Albert 29, 30 types of 126–9
historical analysis 91 idealisations 127–8, 130
historical research 46, 56 idealism 2, 33, 34–5, 36, 39, 40–1
history: and causality 10; and foreign illusion 21; speculative 158
policy 55, 58; lessons of 159; openness IMF see International Monetary Fund
of 200, 208, 238; problem of ending imperialism 237; see also colonialism
140; theorising 12, 61–3, 66, 70; individualism 99
writing 58–9 induction, problem of 23, 31
Hobbes, Thomas 2, 21, 22, 25, 29, 105; Industrial Revolution 83–4
on book-keeping 24–5; and problem Inoguchi, Takashi 61, 62, 63
of order 30 inside/outside dichotomy 94–5, 229
262 Index

institutionalisation 153, 201, 224, 225 36, 37, 143, 209; moral reason 4, 11;
integration 203, 204 theory of peace 3, 21, 39;
interaction 76, 79, 116–17 transcendental arguments 9; world
interdependence 36, 39, 203 order 143, 205–8
internal relations 79; and communicative Kaplan, Morton A. 39
action 100–6; and iconic modelling Kautsky, Karl 63
131, 136; relationship with action 117; Keat, Russell 5
vague and ambiguous nature of 106–7 Kelsen (positivist) 35
‘inter-national’: first use of term 2 Kennan, George 43
International Monetary Fund 134, 205, 233 Keohane, Robert 167–8, 170
International Political Economy (IPE) Keynes, John Maynard 220, 230, 232–3,
80–1; see also global political economy 234, 236
international relations: four ‘great Kissinger, Henry 43
debates’ of 36–40 knowledge: of actors 126; quantitative,
international relations, discipline of (IR) qualitative and relational 117
10–13; foundations 11; in 1960s 43; in Kratochwil, Friedrich 75
1980s 70; related to emergence of Kuhn, Thomas 150
states 82–3; Wendt vs. Hollis and
Smith debate 71–5; see also levels labour, division of 84
inter-state wars 193–4 labour markets 83, 220, 223, 224, 227, 228
invariance 3, 75, 76, 134 Lacedaemon see Sparta
investment 233 Lakoff, George 128
Ireland 199, 209 language 90, 105, 116, 150
lateral pressure theory 50, 192
Jabri, Vivienne 71 Latsis, Spiro 169
James II, King 27 law, rule of 28
Japan 206 Lawson, Tony 7, 129, 135
Jesus story 56–7 League of Nations 39
Johnson, Mark 128 learning process: and change 43–4; and
journalistic studies 43, 134 Gorbachev 3pathological 152–3,
judgemental rationalism 9, 77, 144, 148, 199–200
155, 157 Lebenswelt 112
Lefebvre, Henri 138
Kant, Immanuel 2–3, 9, 10, 11, 21; belief Lehnert, G. 56–7
in Kingdom of Angels 34; as classical Leibniz, Gottfried 69
political theorist 36–7; concept of Lenin, V.I. 42, 63, 113
‘nation’ 17; concern over Europe’s Lesbos 184
destructive powers 84; importance of levels: against scientific
40; on international problematic 32–4; (re)conceptualisations of 82–7; as
‘mechanisms’ for peace 33; on complex assemblies and integrated
perpetual peace prospects 38; and wholes 82, 85; convenience of 86–7;
phenomena/noumena dichotomy as degrees 82; as layers 82–3; as ranks
33–4, 35, 39–40; and positivism 3; 83; realist conceptions of 81–2; social
suspicious of world government contexts 85–6; and study of social
concept 207; on Thucydides 167, 190; systems 118
on war and peace 17 levels-of-analysis problematic 72, 73–81
Kantian theory: as anachronism 237; liberalism 3, 70; economic 231;
critical questions 6, 9, 10; dualism 89; embedded 211, 217, 220, 228, 229;
and ‘great debates’ of international Kantian 2–3, 4, 34, 36, 37, 143, 209;
relations 36–40; Hegel’s view of 34; see also neoliberalism
idealism 2, 33, 36, 39; influence on Lin Biao 63
20th century thinking 39, 43; and IR Linge, David E. 142
problematic 15; liberalism 2–3, 4, 34, linguistic fallacy 4
 Index 263

LO see Swedish Union Federation 124–6; descriptions 132; Deutsch’s use

Locke, John 2, 22, 29, 38 of 197; mathematical and statistical
logic 89bivalent 161–2 135; as projective 126; source of 126;
Lowes Dickinson, G. 39 see also iconic modelling; Norden;
Luxemburg, Rosa 63 Prisoner’s Dilemma
Moore, Barrington 42
Maastricht Treaty 221, 224, 231 Morgenthau, Hans 17, 35, 36, 39, 43;
Machiavelli, Niccolò 27, 29, 34 Keohane on 167–8
Mackie, J.L. 77, 101 Multilateral Agreement on Investments 235
Malaysia 233 multilateral regimes 84
Mandel, Ernest 58, 59 multinational corporations 84, 111
Mandeville, Bernard 30
Manicas, Peter T. 6 Naess, Arne 149–50
Mao Zedung 63, 143 NAFTA 195
Marathon, Battle of 182 narratives 56–7, 59, 140, 141; in security
Marcuse, Herbert 18 community 201; see also stories
Martin, Hans-Peter 230, 233 nation, concept of 26
Marx, Karl 5, 6, 25, 42, 209; on history natural law 32, 178; and justice 25, 30;
10, 238; on ‘laws’ 18; on social and private property 25
relationships 115 natural sciences 4, 5, 8, 9, 101
Marxism 63, 70, 99; Bhaskar and 5–6; negotiated environment model 230
conceptions of power 18; and Social neo-classical polimetrics 173
Democratic Party 16; and Soviet neo-liberalism 111, 220, 222, 224, 225, 227
power 113 neo-realism 4, 35–6, 39, 76; Thucydides
mass production 84, 218 and 168; Waltzian 74, 82, 127
mathematics 134, 135, 169 neo-utilitarianism 3–4
meanings 105, 107; contextual nature New Age philosophies 7
134; problem of objectivity 136–7 New Left 217, 218
mechanisms: human 105; as metaphor Newton, Isaac 22
130; for peace 33 Newtonian mechanics 21, 22, 23, 24, 36,
Mefford, Dwain 54–5 40, 127
Megarian degree 187, 188 Nicias, Peace of 182, 183
Melos episode (according to Thucydides) Niebuhr, Reinhold 34, 36, 39
15, 61, 167–92; dialogue 173–9, 182–3; Nietzsche, Friedrich 7, 34, 35, 36, 37
iconic modelling of 179–80, 189, 190, NORDEK 228
192; prophasis of 179–88; as Norden (Nordic model) 16–17, 210–36;
rationalchoice model 168–73; as Cold War and European context (as
tragedy 188–90; see also Athens; Wæver) 211–12; decline of 211,
Thucydides 229–30; and emancipation 228–9,
Mercosur 195 231–5; and European integration
metaphors 128–9, 130–1, 132, 137; of 229–31; in global political economy
correspondence 145; of mechanism context 215–17; political economy
130; pluralist truth as 149–50 explanation 218–22; problems with
metaphysics 133 Waever’s account 213–15; real interest
Methodenstreit 99 of 226–8; relative persistence of 222–6;
Methymna 184 reorienting 228–9; see also Swedish
metonyms 129, 132 model
Mill, John Stuart 76 Nordic Council 226
mimesis 14, 140, 141, 142, 160, 173, 189 Nordic countries 201; co-operation within
Mitrany, David 127 225–6; government expenditure 223;
Mitylene 184–5 nostalgia 212, 225, 229, 231;
models: Alker and 12, 51–2, 56–61; positivism 39; potential split between
ampliative and reductive approaches elite and population 225; see also
264 Index

specific countries legitimate aggression 189–90

Norrie, Alan 7 Popper, Karl 148, 161
Norway 211, 221, 222, 223, 224–5, 233 position-practice systems 104, 111–12
noumena see phenomena/noumena positivism 3–4, 5, 7, 11, 13, 21; and
dichotomy capitalism 25–6; conception of science
40; debate with post-positivism 36,
objectivity 151 39–40; debate with traditionalism 36,
objects, nature of 99, 150 39; Humean 22–4, 25–6; and levels-of-
offshore financial centres 84, 233 analysis problem 74–5, 76; as
Olafson, Frederick 54 misleading term 41–2; and Nordic
Ollman, Bertell 49 model 227; post-Second World War
Olympic Games 180–1 43; and quantitative data 134; rational-
ontological realism 8, 77 choice theory 168–73, 191, 227; and
ontology 103–4; and data 48, 51; Hume’s social factors (Ollman) 49; Weber and
critique of 22–4; and levels-of-analysis 35
problematic 73–4; necessity of 9–10; post hoc ergo propter hoc-fallacy 76, 135
philosphical (limit to) 109; social 49, post-modernism 4, 36, 39, 70, 214, 227
77, 78, 99–122, 130; and truth 147, 148 post-positivism 4, 5, 36, 39–40
Onuf, Nicholas 85–6, 108 post-structuralism 4, 11, 39, 40, 228
order, problem of 3, 21, 30–1, 237; Alker power 153, 185–6, 188; and command of
on 63; Giddens on 104; inter-state time and space 138; as domination
193–4 199, 201, 204, 205, 232; fragility of
organisations 111, 134, 137; hierarchical 157; relationship with action 112–14;
83, 111, 118 as transformative capacity 199; see also
OSCE 195 balance of power
outcomes 106, 112, 135 power politics 169–70
practices: positioned 116, 119; relational
Paasikivi, J.K. 113 119; social 114
Palme, Olof 229 precedent logics 48
panopticism 52–3 Prisoner’s Dilemma (PD) 12, 46, 51–6,
Parsons, Talcott 105 65, 68
passions: in Hume 30 problem of induction 23, 31
Pax Americana 217 problematic, international 2, 3–4, 15–16,
PD see Prisoner’s Dilemma 21, 31, 34–6, 203
peace: democratic theory of 3, 93–4, 130, property: private 25, 26–7rights 83, 206;
193–4; Kantian theory of 3, 17, 21, 33, see also capitalism
38, 39; research 11, 39, 43, 142; see also prophasis 179; of Melos episode 179–88
violence; war Propp, Vladimir 59
Peloponnesian War see Melos episode; Protestants 27
Thucydides Putnam, Hilary 146
Pericles 179, 183–4, 186, 187–8, 190
Persian Wars 181–2, 187 Rapoport, Anatol 55, 127, 142
phenomena/noumena dichotomy 33, 34, rational-choice theory 168–73, 191, 227
35, 39–40 rationalisation 151; see also aitiareasons
philosophy 8–9, 22–4, 147, 158; of rationality: book-keeping as a model
science 5 (Hobbes) 24–5; as learning 172; see also
physics, social 127 judgemental rationalism
Plato 7, 69 Reagan, Ronald 111, 220
poetic narrative structures 173 real interests 226–7
poetic transposition 140 realism 5; Alker and 64; Christian 36;
Polanyi, Karl 42 conceptions of levels 81–2; vs.
political realism 11, 15, 39, 43, 70; and idealism 34–5, 36, 40–1; ontological 8,
Melos episode 178–9; used to 77; scientific 7, 93, 99; see also critical
 Index 265

realism; neo-realism; political realism 203–4; conditions required for 196–8;

reality 100–1, 149, 174–5 definition 203; global 203–5; and
realpolitik 34–5, 37; Athens and 176, 178, Kantian world order 205–8; learning
185, 189 and power 198–200; and Nordic
reasons:as causes 41, 72, 78, 87–90, 92; countries 211–12; pluralistic 196–8,
human 151, 200 202; role of emancipatory research
‘reductionism’ 76 203; and self-transformative capability
Rehn-Meidner programme 216–17, 220 of social contexts 200–2, 203, 204,
reification 11, 178 205, 206, 208, 237
relations 79; dependency 60; see also ‘security dilemma’ 193
external relations; internal relations; security practices 62–3
international relations; social relations Shatzki, Theodore 101
relativism see epistemological relativism Shotter, John 103–4
religion 21, 29–30, 34, 36 Shotwell, James 39
Renaissance 24, 26, 27, 30; humanism 29 Simon, Herbert 45
repression 152 Singer, [Link] 39, 73, 74, 75, 82
Rescher, Nicholas 106–7, 124, 174 single-exit situations 169
research 129–30; choice of terminal situational determinism 169, 174
consequences 141; as dialogue 173; slaves 213
problem of objectivity 136; subject and Smith, Adam 22, 29–30
object relationship 41 Smith, Steve 13, 71–5, 89, 91, 99
resources 119 social action 108–12, 116, 136, 173
Ricoeur, Paul 7, 14, 55, 124; on social atomism 25
temporality and narrative 139, 140, 141 social context 85–6; force of 201;
Rome, Treaty of 231 self-transformative capability 200–2,
Rorty, Richard 6, 77, 132–3, 145–7 203, 204, 205, 206, 208, 237
Rothkin, Karen 55 social democracy 211, 212, 214, 222,
Rousseau, Jean Jacques 2, 32 223–4, 228, 229, 236; and corporatism
Ruggie, John 26, 204, 214 221
rules 173; constitutive 102–3, 104, 108, Social Democratic Party 16, 217, 218–20,
111, 117, 119, 131, 201; regulative 224
102–3, 108, 119, 144, 201 social engineering 31
Russett, Bruce 44, 45 social ontology 49, 77, 78, 99–122, 130
social policy: institutional re-distributive
SAF see Swedish Employers’ Association model 215
Sainte Croix, G.E.M. de 170 social relations 104, 115–16; difficulty of
Saltsjöbaden capital-labour Accord (1938) quantifying 134; see also external
216–17, 218 relations; internal relations
Sandinistas 63 social sciences 5, 8, 9, 101, 115, 160; as
SAP see Social Democratic Party analysis of actions and structures
Saussure, Ferdinand de 116 105–6; and hermeneutics 123;
Sayer, Andrew 7, 124 inadequacy of mathematical methods
scepticism 23, 35, 36, 37, 227 (Alker) 45–6, 47–51; and nature of
Schleiermacher, Friedrich 35 truth 154–5; see also behaviouralism
Schumann, Harald 230, 233 social structure 101–3, 114–18
science 155; Bhaskar on 100–1; social systems 118emergent powers and
philosophy of 5; and truth 146–7, 148, properties 83–5
150; see also natural sciences; ssocial world: hermeneutic character 102
Newtonian mechanics; social sciences sociology 115
Second World War 11, 43, 113, 193, 205 South-east Asia 195
Secord, Paul F. 5, 88 Soviet Union 210; Finland’s treaty with
security community 194–209; 211, 236; influence of Marxism 113;
amalgamated 196, 201–2; cognitive see also Cold War
266 Index

space: in iconic models 137–9; and time belief in brute power politics
94, 118; see also geo-historical processes 169–70; on causes of Peloponnesian
Sparta 180, 181, 182, 183, 184, 186–8 war 186–8; on development of Athens
spatiality, horizontal 79, 80 192; different interpretations of 191;
Spinoza, Baruch de 30 on Hellas 180–2; and history as
state: and assumptions of space 138; tragedy 172–3; on lasting value of his
behaviour 73–5; and hierarchies 83; work 189; moralism in 188; on
modern (development of) 21, 26–8, Pericles 184; on Persian Wars 182;
84, 237; Nordic preoccupation with personal background 186, 192;
228; as open system 118; as person 87, ‘polimetrics’ of 173; tragedy in 172–3,
129; private property analogy 26; and 188–90; use of dialogue 168, 169–70,
problem of order 31; and quantitative 189; see also Melos episode
data 134; relationship with institutions time: in iconic models 137–8, 139–41;
79, 82–3; relationship with peace and and space 94, 118; see also geohistorical
violence 15; relationship with war processes
193–4; as unitary actor 71–2 TINA-formations 227
states system theory 86–7 TMSA see transformational model of
statistics 117, 133–5; Alker on 45–6, 49, social activity
68 Tobin, James 223–5, 236
stoicism 35–6, 37, 227 totalitarianism 68
stories 57–8; ‘two stories’ dichotomy Toulmin, S. 29
87–90; see also narratives traditionalism: vs. positivism 36, 39; texts
structuralism 91, 99 134
structuration, theory of 99, 100, 112, 113, tragedy: Aristotle on 188–9; history as
114 58; Marx on 238; in Thucydides
structure: definitions of 114–15, 116–18; 172–3, 188–90
nature of 100; social 101–3, 114–18 transformational model of social activity
Sweden 224, 235, 236; and Cold War (TMSA) 99, 102, 108
211, 212; economy 214, 215, 223, transformative action 120
233; and EU membership 222, 225; trans-nationalisation 229–30
inequalities 223; 1980s 222; see also trials 137; by jury 103, 107–8, 110–11
Swedish model truth: as metaphor 149–50; nature of 14,
Swedish Employers’ Association 217, 220 143–50, 153–4, 155–6, 157–8, 161–2;
Swedish model 16, 215–22; decline of ontological and alethic aspects 147,
215; see also Norden 148; politics and 143; Popper’s theory
Swedish Union Federation 217, 220 of 161
symbols, technical 126
systems: comprehensive 149; definitions undecidability 154–5
of 118; open 76, 117–18, 130, 135, understanding: of being 133; explanation
146, 155–6, 157, 202, 218; as 72–3, 132–3; Habermas’s theory of
positionpractice 104, 111–12; social 89–90, 104–5; as productive process
83–5, 118theoretical 150 87
unemployment 223, 224
Tarski, Alfred 161, 162 Unger, Roberto 159, 201
tax havens 84, 233 unhappy consciousness 34, 35, 36, 37, 227
taxation: EU 230; global 17, 232, 233–5, United Nations: Alker’s studies of
236 General Assembly 44–6, 47–8, 54, 68;
Thatcher, Margaret 111, 220 collection and analysis of data 134;
thematising 133, 142 new proposals for 232, 235; Nordic
Third Way 16, 210, 211, 213, 214, 215, countries in 232; Security Council 206
217, 220, 222, 236 Urry, John 5
Thucydides (The Peloponnesian Wars) 15, US Space Command Vision 2020 139
34, 41; aitia and prophasis concepts 179; USA 210, 214; action against Soviet
 Index 267

Union (1918) 143; covert activities 93; 207; people’s 63, 143; role of states
dollar no longer convertible to gold 193–4; to end all wars 63, 143;
218; hegemony of 205, 206–7, 232; twentieth century 193; see also violence
interdependence school 39; networks Washington consensus 222, 228
and corporations 227; political realism Weber, Max 35, 37, 109, 128
11; positivism 39; see also Bretton welfare state 212, 214, 215, 220, 222, 225,
Woods system; Cold War 227; spending 220, 223
Utopias 2; concrete 158–60, 195, 202, Wendt, Alexander 13, 86–7, 90–1, 93, 94,
203, 208 99; debate with Hollis and Smith 71–5
Utrecht, Peace of 2, 28 Westphalia, Peace of 2, 27, 40
Wight, Martin 36
vagueness 103–4, 106–7 will, hardening of 199–200
value judgements 153–4 will-to-power 36, 186
Vico, Giovanni Battista 30 William of Orange 27
Vienna, Concert of 42, 84 Wilson, Thomas Woodrow 62–3, 143
violence 155–8; critical theory and Winch, Peter 99, 104, 154
avoidance of 14; ‘cultural’ 161as Wittgenstein, Ludwig 99, 104, 106, 107,
outcome of pathological learning 149
process 152; politics and 196and Wittgensteinians 39
security community 193–6, 200–1, Woodrow Wilson Chair (Aberystwyth)
202, 203, 205; state and 15, 194; see also war 39
voluntarism 24, 36, 51, 99 World Bank 134, 205
world government 207
Wæver, Ole 92–3, 196, 208, 225; and ‘world polities’ 80–1
Norden 16, 211–14, 226 World System Analysis 60, 127
Waisman, Friedrich 104, 106, 107 World Trade Organization 205, 236
Walker, R.B.J. 40, 70, 79
Wallerstein, Immanuel 46, 56 xympasa gnome 170
Waltz, Kenneth 39, 74, 82, 83, 94
Waltzian neo-realism 82, 127; reductio ad Young, Oran
absurdum of 74
war: and Kantian world order 21, 206, Zimmern, Alfred 39