You are on page 1of 45

ROOTS OF REALISM

CASS SERIES ON SECURITY STUDIES

Editor

Benjamin Frankel

Roots of Realism
edited by Benjamin Frankel
Realism: Restatements and Renewal
edited by Benjamin Frankel

ROOTS OF REALISM

Edited 1?J

Benjamin Frankel

LONDON AND NEW YORK

First published 1996 by Frank Cass Publishers


Published 2013 by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
711 Third Avenue, New York, NY, 10017, USA
Routledge is an imprint of the Taylor & Francis Group, an informa business
Copyright 1996 Taylor & Francis
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 Cataloging-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalogue record for this book is available from the Library of Congress
This group of studies first appeared in a special issue, titled Roots of Realism, of
Security Studies 5, no. 2 (winter 1995/96), published by Frank Cass & Co. Ltd.
ISBN 13: 978-0-714-64203-1 (pbk)
ISBN 13: 978-0-714-64669-5 (hbk)

CONTENTS

Preface
Introduction

Vll

Bnyamin Frankel

ix

Ashlry]. Tellis

Paul A. "Rahe

105

Mark V. Kauppi

142

!..Aurie M. Johnson Bagl?J

169

Thomas]. Johnson

194

Markus Fischer

248

Jan Willem Honig

283

Robert G. Kaufman

314

Patricia S. Wrij!,htson

354

I. RECONSTRUCTING REALISM

Reconstructing Political Realism:


The Long March to Scientific Theory
II. CLASSICAL REALISM

Thucydides' Critique of Realpolitik


Thucydides: Character and Capabilities
Thucydidean Realism:
Between Athens and Melos
The Idea of Power Politics:
The Sophistic Foundations of Realism
Machiavelli's Theory of Foreign Politics
III. MODERN REALISM

Totalitarianism and Realism:


Hans Morgenthau's German Years
E. H. Carr, Winston Churchill,
Reinhold Niebuhr, and Us:
The Case for Principled, Prudential,
Democratic Realism

Morality, Realism, and Foreign Affairs:


A Normative Realist Approach
Bibliography

389

Index

417

Page Intentionally Left Blank

PREFACE

articles gathered in this and a companion volume were


presented as papers at the Securiry Studies conference on "Realism:
Restatements and Renewal," held at the Miller Center, the University of Virginia, 6-10 October 1994. 1 The spirited discussions and the
camaraderie among the participants made the conference a success, proving
that scholarly gatherings can be serious and demanding without being dull
or tedious. The discussions at the conference made an important contribution to the quality of the two books which came out of it: Many conference
participants wrote detailed comments on the papers and shared the comments with the authors. These comments, and the criticisms offered by
conference participants, helped the authors improve their papers.
Special thanks go to Campbell Craig, Daniel Deudney, Peter D. Feaver,
Ted Hopf, Chaim Kaufmann, Douglas J. Macdonald, Jerome Slater, Shibley
Telharni, Phil Williams, James Wirtz, and Fareed Zakaria. Thanks also to W.
David Clinton, Gary Schaub, and Michael Siler. The participants at the conference owe a special gratitude to Paul W. Schroeder. Political scientists are
often criticized for creating models which, in their elegance and parsimony,
often give the impression of being removed from historical reality. Schroeder's admonitions surely kept honest the political scientists among the conference participants.
I am especially grateful to the many dedicated reviewers of Securiry Studies.
They toil in selfless, and often thankless, anonymity to make sure that the
quality of the articles published in Securiry Studies, whether in regular issues
of the journal or in the journal's special issues and books, meet exacting
scholarly standards. Our reviewers cheerfully shouldered the considerable
burden of reviewing successive versions of the articles collected in the two
realism volumes.
Kenneth W. Thompson, director of the Miller Center and an esteemed
scholar of realism, was generous with his time, wise counsel, and comments. The staff of the Miller Center could not have been more helpful in
making the conference run smoothly and without interruptions.

OST OF TI-IE

1. See the special issue of Securiry Studies 5, no. 3 (spring 1996); and Benjamin Frankel, ed.,
Realism: Restatements and Renewal (London: Frank Cass, 1996).

We are especially grateful to Marshal Zeringue, a fellow at the Miller


Center. Without Marshal's untiring work, without his dedication and industry, the conference would never have gotten off the ground.
We would like to express our appreciation to Zachary S. Davis, who
helped conceive of the project.
The conference was made possible by a generous grant from the
Compton Foundation.

Benjamin Frankel

Security Studies
Washington, D.C.
February 1996

INTRODUCTION
BENJAMIN FRANKEL

realist theory of international relations. Rather,


there is a family of realist theories and explanations, differing from
each other in the importance they assign to different variables (for
example, the polarity of the international system, or the role of domestic
institutions). The theories in the realist family, however, do have a common
center of philosophical gravity: they are all grounded in an understanding of
international politics, and politics more generally, as a constant struggle for,
and conflict over, power and security. From the beginning realism has offered explanations for how political units-today we call them statesprotect and preserve themselves in an anarchic environment in which dangers to security and welfare are always present, and even survival itself is
not assured. The pursuit by states of their own security and autonomy is
impinged upon and limited by other states' pursuit of their-these other
states'-security and autonomy. The relationship among states is thus fundamentally and inalterably a conflictual relationship, with states constantly
and continuously jostling with and elbowing each other as they try to improve their security and enhance their autonomy. This restless agitation is
made more dangerous because of the anarchic nature of the international
system: There is no superior arbiter of states' conflicting claims, and no
superior authority with the ability to enforce arbitration rules.
There are debates among realists whether it is mere security that states
seek; whether the reasons why the pursuit of security by one state impinges
upon and limits the security of another state are systemic-and therefore
not amenable to change-or domestic, and thus susceptible to modification
and change; and whether states adopt aggressive or defensive strategies to
enhance their security, and more. There is no debate among realists, however, that, at a minimum, states are worried about their security and that
they act vigilantly to enhance that security in an environment which offers
them no choice but to do so.
The essays in this volume address different aspects of this realist understanding of international politics.
HERE IS NO SINGLE

Benjamin Frankel is editor of Semrity Studies.

ROOTS OF REALISM

RECONSTRUCTING POLITICAL REALISM

Ashley J. Tellis, in his "Reconstructing Political Realism: The Long March


to Scientific Theory" (3-101), defmes the assumption that states are worried about their security, and that they take action to enhance that security,
as the "minimum realist program." This program has been articulated differently by different realists, and the purpose of Tellis's reconstruction of
the realist research program is to show how it has evolved over time. He
does so by analyzing the works of five theorists-Thucydides, Machiavelli,
Hans Morgenthau, Morton Kaplan, and Kenneth Waltz--each representing
a turning point in the evolution of political realism. Tellis thus shows "how
the realist program, which began as a philosophical reflection on the nature
and behavior of security-seeking entities, has gradually been transformedhowever imperfectly-into the abstract, deductive formulations which
modem social science demands" (4). His article is primarily a rational reconstruction, not a historical narrative. It is an interpretation which recreates the internal history of the realist research program, describing and
analyzing the different substantive and methodological problem shifts
which mark that history.
Tellis argues that since its metaphysical Thucydidean beginnings, political
realism has evolved in the direction of science, that is, toward producing
scientific formulations of its central assumptions. This evolution, however,
has been incomplete. To determine what a good theory is, Tellis appeals to
the methodological criteria offered by critical rationalism. When applying
these criteria even to contemporary political realism, he finds even the its
best scientific versions deficient. "It will not have crossed the threshold of
acceptability as a minimally adequate scientific research program until it
sheds the last vestiges of naive empiricism in favor of a rationalistdeductive system built around the construction of situationally-determined
exit models explicitly incorporating acting individuals as the theoretical
primates," Tellis writes (89-90). We can see the move away from the naive
empiricism of traditional realism in the construction of partial realist theories in the rationalist tradition, partial theories which address topics such as
alliances, polarity, and arms races. A general realist theory in the rationalist
tradition is yet to be developed.
Such a theory-a general rationalist realist theory based on a deductive
explanatory system-"cannot be produced unless the current analytically
primary units of international politics, 'states,' are 'reduced' into their constituent units, 'individuals,' so that the former may then be 'synthesized'
through a system of deductive logic," Tellis writes (91). This would mean

Introduction

xi

turning Waltz's methodological approach upside down: "Rather than attempting to recreate a new 'systemic' approach to international politics, it
requires the construction of a new 'reductionist' framework centered on the
security-seeking individual as both the unit of analysis and the locus of causation" (92). Realist Man will thus replace the State as the foundation for a
new deductive-individualist explanatory framework.
Tellis concludes that the critical rationalist methodology of science
promises to improve the realist research program in three ways. First, it will
help scientific realism explain political behavior at all three levels of analysis
in terms of a single explanatory principle: conflict. Second, it will help scientific realism refine existing explanations of politics by specifying how all
entities in an anarchic world are inescapably constrained to seek domination. Third, it will enhance the explanatory ability of scientific realism to
account for the production and maintenance of order across the political
spectrum by describing the formation of states, by explaining the behavior
of both stable and unstable states in international politics. "Critical rationalism thus takes scientific realism back to its roots in traditional realismespecially in the arguments of Machiavelli-and helps make it, for the first
time, not simply an explanation of 'international' politics but rather a scientific theory of egoist competition writ large" (94).1
THUCYDIDEAN REALISM

Political scientists and historians alike point to Thucydides as the embodiment of early realism. Paul Rahe, in his "Thucydides' Critique of Realpolitik" (105-41), agrees that this is a reasonable reading of Thucydides.
Thucydides was not a naive observer of politics and war, and his book
provides evidence for his unsentimental assessment of human motives and
conduct. There is, however, another side to Thucydides, a side which many
ignore. Thucydides' narrative does provide evidence that he did not believe
that religion or piety would place limits on human conduct, but the same
text also provides evidence that he considers dangerous the exclusive reliance on realism-what would later be called Machiavellian realism-for an
analysis of or as a guide to public policy. The purpose of Thucydides' history is not to study how to attain and retain power, and it is also not an
examination of why anarchy dominates international relations. Rather,
Thucydides is interested "in exploring the conditions essential for, the cir-

1. Markus Fischer shows how Machiavelli provides what he-Fischer--calls "the missing
microfoundation" of neorealism, that is, a theory of man. See discussion in xviii-xix.

xn

ROOTS OF REALISM

cumstances conducive to, and the fragile character of what we would now
call civilized life. His account of realism is subordinate to this theme-as is
his graphic depiction of the consequences in store for those who embrace
and publicly endorse the theory and practice of Realpolitik" (110).
The contrast Thucydides draws between the Melian dialogue and the
Sicilian expedition offers an insight into his views of realism. It is evident
from Thucydides' narrative, writes Rahe, "that the passionate pursuit of
unbounded, undefinable ends is incompatible with prudent, measured deliberation concerning advantage, for one cannot proportion means to ends
when the ends are indeterminate" (128). Indeed, Machiavellian realismthe kind of realism the Athenians exhibited-was a threat to prudent deliberation, because without moral limits to provide a sense of proportion,
human beings will likely pursue unlimited goals. Morality and the pursuit of
one's interest are thus not mutually exclusive; on the contrary: they are
mutually supportive: "There would appear to be a connection linking the
civilized capacity to respond to the claims of justice and human decency
with the sense of measure, of limits, and restraint necessary for a sober
consideration of self-interest" (128). The Athenians ignored the connection
between the observation of moral limits and the pursuit of their own interests. Their daring and audacity in war eventually deteriorated to hubris,
causing them to lose the ability to pursue their interests in a measured, balanced manner. In the end, their rejection of moral limits did not afford
them a more effective way to pursue their interests. Uninhibited in their
pursuit of their interests, they descended into stasis and barbarism.
Athens was a more sophisticated, open, and dynamic society than Sparta.
This openness and daring came at a price, however. It was "precisely because Athens gives relatively free rein to the potential for greatness inherent
in human nature, that it is Athens, not Sparta, that loses all sense of measure and falls apart" (139). Rahe suggests that Thucydidean realism is a realism that knows bounds. The lesson contemporary proponents of Realpolitik should draw from Thucydides is that total emancipation from moral
bounds, and the freedom to desire and pursue anything which is within
one's power to attain, will not lead to liberation and freedom. The opposite
is the case: such freedom will make one a slave to an attitude of reckless
daring borne of the notion that the only limits which matter are the limits
of one's ability. In the end, such freedom and such attitude will harm one's
interests rather than advance them.
Minimalist realism. Laurie M. Johnson Bagby, in her "Thucydidean Realism: Between Athens and Melos" (169-93), reaches similar conclusions.
Both fundamentalist and structural realists, Bagby writes, have misinter-

Introduction

xw

preted Thucydides by selectively borrowing from his narrative. Fundamentalists overemphasized Thucydides' discussion of human nature, while
structuralists overemphasized his discussion of the power relationship between Athens and Sparta as a cause of the war.
Both readings are overdeterministic. In a deterministic world, a world in
which what is cannot not be, a world in which the inexorable and inevitable
reign, there is no room for free choice and hence no room for moral considerations. Today's structural realism stresses the inescapable impact of
anarchy on the relations among states, to the exclusion of all other influences. In this, Bagby writes, it is similar to the Athenians' fundamentalism,
which emphasized the compelling force of the Athenian will to power, to
the exclusion of all other considerations (188-89). Because both theories
are vulnerable to deterministic interpretations in practice, both have been
the subject of the same type of moral criticism. Bagby argues that Thucydidean realism is not vulnerable to such interpretations and criticisms.
Thucydides' realism is a minimalist realism. "A prudent moderation is
often the closest thing to the exercise of morality in foreign policy," Bagby
writes.
To the extent that such moderation can be called moral, Thucydides can
be said to believe that morality and expediency can coincide .... those
who are too focused on justice, morality, and piety are not well equipped
for survival in the international realm. Those who are too focused on
their supposedly compelling passions, however, are also not well
equipped for survival. Ibis is the key to understanding the moral dimension of Thucydides' minimalist realism, which lies somewhere between Athens and Melos. (191)

The importance of character. In his essay "Thucydides: Character and Capabilities" (142-68), Mark Kauppi, examines the theoretical implications of
what he describes as the most famous sentence in Thucydides' history of
the Peloponnesian War-"What made war inevitable was the growth of
Athenian power and the fear which this caused in Sparta."
Realists of all kinds appeal to Thucydides for support of their theoretical
constructs. Power-transition realists find an especially strong case in
Thucydides' writing. The essence of Thucydides' explanation for the outbreak of the Peloponnesian War, after all, appears to revolve around the
shift in relative capabilities between two powerful countries. For powertransition theorists (and, for that matter, for structural realists), the identity
of the two countries, in this case Athens and Sparta, is irrelevant. A change
in the balance of power between any two countries would similarly breed

XlV

ROOTS OF REALISM

suspicion and anxiety and might well lead to war, as it did in the case of
Athens and Sparta.
Thucydides' insight that there is a relationship between shifts in the distribution of capabilities among states and the onset of war has become a
point of departure for power-transition realists. Works in the powertransition tradition address an important question: If we accept that there is
a relationship between changes in the distribution of power and the onset
of war, and if we further assume that the power of states relative to each
other is in constant flux, with different states continuously becoming more
or less powerful relative to other states, how, then, will it be possible to
achieve and maintain a peaceful international system?
According to Kauppi, the answer Thucydides gives to this question
highlights an important fact: the interpretation of Thucydides as being the
first power-transition theorist is correct, but it is incomplete. Unlike powertransition theorists, Thucydides did see the specific character of the countries involved-not only the balance of power between them-as an important cause of the war. "In other words," writes Kauppi, recalling that
most famous sentence of Thucydides, "the emphasis in the phrase 'growth
of Athenian power' should be as much on the adjective 'Athenian' as it is
on the noun 'power"' (143). The fear Sparta had of Athens was not the result solely of the increase in the power of a neighboring state, but more the
result of the actual or perceived special character of Athenian society. The
specific characteristics of the Athenian society created an Athenian citizen
whose daring and self-assurance were a driving force behind Athenian imperialism-and as such the major source of anxiety in Sparta. To use the
terminology of international relations theory, Thucydides was as much concerned with second-image variables as he was with third-image ones.
The importance Thucydides attached to second-image variables is evident in his discussion of the sources of Athenian imperialism. Thucydides
attributes Athens' imperialism to the Athenian political culture and the kind
of citizen it had created. The Athenians were restless, action-oriented people. Even more than action, they prized "daring" (ttilma). This specific
Athenian characteristic, which made their neighbors even more uneasy
about Athens' increased capabilities, was the result of two characteristics of
the Athenian polity. First, Athens' democratic political culture liberated
individual Athenians and allowed room for the exercise of their talents.
Second, the Persian threat, which made the Athenians a seafaring people
(151-52). The way the Athenians handled the Persian threat further contributed to their audacious foreign policy. "Out of the crucible of the Persian wars and the experience of abandoning their city," Kauppi writes, "the

Introduction

xv

Athenians gained a unique sense of self-assurance which allowed them to


dispose of traditional restraints on behavior, including notions of justice"
(158). The specific characteristics of the Athenian society, especially the
absence of a strong sense of limit and proportion, thus made Athens' accumulation of power that much more menacing.
Kauppi agrees with Thucydides' emphasis on domestic variables, saying
that "Thucydides got it right by emphasizing how domestic factors such as
the nature of state and society-as in the case of Athens-will determine
the extent of imperialistic ambitions. These ambitions will in turn influence
the threat assessments of neighbors and other interested powers" (167).

THE SOPHISTIC AITITUDE

Thomas Johnson, in his "The Idea of Power Politics: The Sophistic Foundations of Realism" (194-247), examines the sophistic origins of realism.
He notes that the criticism directed at the realists today is similar to the
criticism leveled against the fifth-century B.C. Greek sophists. The similarity
of the criticisms of sophism and realism, he says, is not a coincidence because the ideas articulated in the sophistic enlightenment form the philosophical basis for contemporary realist theory in international relations.
The sophistic attitude consists of three principles: first, truth and justice
in the world are relatively defined, and there is no one universally knowable
or accepted truth; second, an inherently pessimistic view of human nature
and its ultimate potentialities; and third, a recognition of the primary role of
power-from persuasion to threats to physical coercion-in enforcing parochial conceptions of truth (204-5).
Johnson then goes on to show how these sophistic principles underlie
the thoughts of Machiavelli, Hobbes, Carr, and Morgenthau. His discussion
of the sophistic sources of Morgenthau's world-view is a good illustration
of his discussion of the other realist thinkers. Morgenthau shared with Carr
an impatience toward British and American liberals who witnessed, with a
sense of horror and incredulity, the resurfacing of power politics in the
1930s. The liberals asserted that that resurgence was but one more piece of
evidence of the fundamental irrationality of the revisionist powersGerman, Japan, and Italy. Power politics, so the liberals argued, had not
reappeared because it was always present in history, always a feature of the
relations among groups of people, always an aspect of human nature.
Rather, the reemergence of power politics in the 1930s reflected the irrational aspirations and designs of specific men who held the leadership of

XV1

ROOTS OF REALISM

specific countries, and who, because of their irrationality, did not know
how to use their reason properly (238).
Morgenthau did not criticize the very principles of liberalism. He criticized instead the belief to which many liberals subscribed that liberalism
represented the eternal, essential truth of mankind, and that that truth had
been finally recognized-and realized-in the wake of the First World War.
For Morgenthau, liberalism was a product of specific historical conditions
and a particular culture. It was not a universal, transcendent truth. Morgenthau did not advocate principles of morality and ethics in politics that
would replace liberal principles, but rather asserted that such principles
must be rooted in the reality of politics in which the pursuit of power is the
dominant drive and cause (238-39).
Johnson's discussion of how the sophistic attitude permeates the thinking of Machiavelli, Hobbes, Carr, and Morgenthau leads him to conclude
that the end of the cold war does not mean--cannot mean-that realism
has become obsolete as a theory of international politics, or as a guide to
statecraft. New powers will rise and fall; and norms and standards will
change as the social conditions which give rise to such norms and standards
change. The sophistic attitude teaches us that these changes "are reflective
of the particular power arrangements prevailing at a particular time in history." True, the specific norms, standards, social conditions, power arrangements, and ways in which power is manifested may be different, but
politics will continue to be preoccupied with and driven by the ideas expressed by the sophistic attitude. "Truth will still be defined relative to the
interests of those who articulate it. ... power will retain its central and constitutive role in structuring all political interaction" (246).
For Johnson, the clear line linking sophism to realism shows that realism
is not a product of the European experience of the last two hundred years.
Contemporary realist theory may give the impression that it addresses issues and problems unique to the late nineteenth and twentieth centuries,
but it is overlaid with the timeless ideas central to the sophistic attitude.
Similarly, even if a realist theory of international relations has now to take
into account the historical and contingent elements specific to the postcold war period, such a theory, to be a viable theory, will have to be constructed on the firm foundations provided by the sophistic attitude

MACHIAVELLI: THE MISSING MICROFOUNDATION OF NEOREALISM

Markus Fischer, in his "Machiavelli's Theory of Foreign Politics" (248-79),


reconstructs Machiavelli's understanding of foreign politics in a systematic

Introduction

xvu

fashion. For political theorists, Fischer offers an inquiry into the foreign
aspects of the state of nature which arises from anthropological pessimism.
It shows how foreign anarchy constrains domestic order and impedes the
pursuit of virtue. For international relations scholars, he offers a contemporary analysis of Machiavelli's propositions on the behavior of political units
in a system of self-help. Fischer argues that Machiavelli provides neorealism
with its lost philosophical foundation-a theory of the individual rather
than the state as neorealism's constituent unit-while at the same time
challenging neorealism on the causes of war and security, the recurrent
formation of balances of power, and international institutions.
Fischer's reading of Machiavelli's discussion of the relations among
states supports the aggressive realist version of structural realism. 2 Machiavelli posits that the primary political order emerges from the original
condition of licence, when an autocratic ruler suborns a number of individuals. As other similar orders are formed as a result of the remaining individuals organizing themselves (or, more commonly, as they are suborned
by other autocratic rulers), the state of licence continues to exist in the relations among these different orders. This is the condition international relations scholars call anarchy (256). The state of licence which exists among
these orders creates other features which contemporary realists associate
with the international system: the security dilemma, self-help, and preoccupation with relative gains.
Because anarchy exists in the relations among political orders (or states),
these orders are constantly worried about their security, if not survival.
They find the military capabilities of other states threatening, regardless of
the professed or assumed intentions of these other states, because one can
never be certain that these capabilities will not one day be used to one's
detriment (256). 3 In the face of the immediate or potential threat which
another state's capabilities pose, a state ought, at a minimum, to acquire
enough military strength to deter would-be aggressors (258). An effort to
build a strong deterrence, however, may well trigger the security dilemma:
because states cannot be certain that the capabilities acquired for deterrence

2. For discussion of offensive and defensive realism, see Jack Snyder, Myths of Empire:
Domestic Politics and International Ambition (Ithaca: Cornell University Press, 1991 ), 10-13; John
J. Mearsheimer, "Back to the Future: Instability in Europe After the Cold War," International
Semriry 15, no. 1 (summer 1990): 5-56; Fareed Zakaria, "Realism and Domestic Politics: A
Review Essay," International Semriry 17 no. 1 (summer 1992): 177-98; and Benjamin Frankel,
"The Reading List," Semriry Studies 5, no. 1 (autumn 1995): 184-87.
3. This is a very different reading from the one Kauppi offers regarding the relationship
between capabilities and character, and how the two are related to whether or not a state
should be regarded with alarm by other states. Sec discussion in xiii-xv.

XV111

ROOTS OF REALISM

might not one day be used offensively, a military build-up, regardless of its
purpose, creates a threat to other states. These other states in turn would
begin to augment their own military capabilities.
Could states avoid such a spiral of arms-racing, Machiavelli wonders, by
acquiring capabilities strong enough to deter, but not so strong as to pose a
threat to others? Machiavelli says this is a practical impossibility (259). The
contingency of politics makes it impossible to strike a balance between inviting aggression through weakness and provoking war through strength.
Machiavelli's conclusion are logical, writes Fischer: since war cannot be
avoided, advantage must be gained by starting it. "To gain lasting security,
the adversary's war-making potential-its people, raw materials, and manufacturing capacity-must be brought under one's dominion or destroyed .... to seek preservation in a world where war is unavoidable is to
strive for domination and empire." (260-61).
Fischer then goes on the discuss the causes of expansion, alliance, and
the relationship between the international and domestic orders. He concludes his essay with a discussion of what he calls the missing link in neorealist theory. The two central elements of neorealism-the structure and
the state-are abstract concepts. What states do is determined by individuals, and the structure within which states operate is determined by state
conduct. Although Kenneth Waltz compares the operation of the international system to that of the market, he does not do what economists do.
They deduce the behavior of firms from the behavior of individuals; Waltz,
however, does not follow them in this fundamental step, creating instead a
theory without a microfoundation. Fischer argues that to strengthen neorealism, "we need to generate political units from assumptions about the
elementary properties and propensities of individuals" (273).
This is not an unimportant matter. A major element in the constructivist
challenge to neorealism is the fact that critical theory has an active subject
at its core. What neorealism takes for granted-state, structure, anarchycritical theory posits as intersubjectively constructed. Anarchy and the security dilemma, for example, are not a law of nature, but the result of
shared understandings and interpretations by human beings. When human
consciousness and values change, critical theorists argue, the reality-that
is, the shared understandings and interpretations-will change as well.
What theory of human nature does neorealism appeal to when it posits
the continued existence of sovereign states-and the resulting anarchy, selfhelp, security dilemma, and relative-gains concerns-as an inalterable feature of international life? Fischer argues that Machiavelli offers neorealism
its missing microfoundation-a theory of human nature: "Machiavelli's

Introduction

XIX

thought offers neorealists a deductive argument to ground their concept of


political unit in concrete assumptions on individuals; for he generates his
propositions from a well-developed psychology, and his propositions on
political order are broadly compatible with neorealist assumptions about
the state. This grounding consists in the generation of political order from
the state of licence and premises on human nature" (274). 4

MODERN REALISM

Jan Willem Honig, in his "Totalitarianism and Realism: Hans Morgenthau's


German Years" (283-313), offers an unorthodox interpretation of the
sources of Morgenthau's realism. While Johnson sees Morgenthau's realism
as but a contemporary manifestation of a realism which is based on the
sophistic attitude, Honig emphasizes the effects which intellectual and political developments specific to the twentieth century had on the evolution
of realism as a theory of international relations. He points in particular to
the intellectual debt realism owes to the totalitarian ideologies, especially
those which accompanied the rise of Nazism. The study of the development of Morgenthau's thought is suitable for this task because Morgenthau
attracted a wide following in the United States, and was a direct link with
German interwar theorizing on international relations.
Honig argues that there is a close affinity between Morgenthau's realism,
on the one hand, and German interwar thinking on international relations
which influenced or reflected Germany's foreign policy in the 1930s, on the
other hand. It should be noted that Morgenthau was not a realist during his
earlier years; it was the events surrounding the Second World War which
radicalized his thinking. He then borrowed important elements of the world
view from his erstwhile German intellectual opponents-those he opposed
during his prerealist days in the 1930s-as the fundamental building blocs
for his own theory.
The realism Morgenthau stands for thus represents a passing phase, tied
to a specific period and culture in history. This realism, because of its historicism and contextuality, does not afford more than a very limited insight
into the nature of the international system. The period important to the
development of Morgenthau's notions of realism was the interwar years in
Germany. During those years there was remarkable degree of consensus
among Germany's political leadership, the military, and academia regarding
the nature of international politics. The German elites accepted that the
4. Cf. the discussion of Tellis in x-xi.

xx

ROOTS OF REALISM

international system consisted of incompatible nations and that the foundation of international politics was the struggle for power and the defense
of the national interest. ''Whether the incompatibility followed from racist
or antiliberal convictions," writes Honig, "they agreed that the incompatibility was so fundamental that the struggle between nations had to assume a
total character and could only, if at all, be resolved by one power achieving
hegemonic status" (303). The acceptance of the notion that total war was
inevitable made debates over the nature of power and national interest irrelevant. Debate about the relationship between war and politics, war and
international law, and war and morality are, in effect, debates over the limits
which should be imposed on war in the name of politics, international law,
or morality. If total war is inevitable and limited war is merely wishful
thinking, then there is no point in debating limits to war.
Morgenthau's theory of international politics, however, is less terrifying,
though hardly less gloomy, than that of his paleorealist German forebears.
Honig finds no glorification of violence in it, but merely an acceptance,
tempered by Morgenthau's belief that in the end-and here Morgenthau
falls back on his prerealist days-humanity will find a better world.
In the end it can be argued that Morgenthau advocates a return to a
nineteenth-century brand of German conservative (that is, Bismarckian)
political philosophy, which also denied the necessity of adaptations in the
domestic political structures to cope with the pressures of growing popular
participation in politics (310). Making arguments which would later be echoed by George Kennan and Henry Kissinger, Morgenthau sought statesmen who, aloof from the passions of the populace, would keep the struggle
for power within relatively peaceful bounds and avoid total war. Rejecting
both the liberal notion that public opinion is a force for peace, and the national-socialist alternative that the popular will must be harnessed for the
coming struggle, Morgenthau also rejects the authoritarian implications of
the Bismarckian philosophy. His theory thus remains incomplete and selfcontradictory, and compromised by the peculiarly his-and the u.s.situation and experience (310).
Honig shows that many of Morgenthau's German contemporaries reacted differently to the national-socialist experience, and drew different
conclusions from it. "The same historical event thus led to very different
evaluations of the nature of international politics," he writes.
"Morgenthau's conversion from a youthful idealist to a mature realist, coincided with the rise of the United States to superpower status, whereas his
German mentors and colleagues changed their views in the opposite direction as a result of Germany's declining position in world affairs. In other

Introduction

XX1

words, it could be argued that one's perception of the nature of the international system is directly related to the position of one's own country in
that system" (312). 1bis may be read as a confirmation of what Thomas
Johnson called the sophistic origins of Morgenthau's realism.
Can; Churchill, and Niebuhr. In his essay "E. H. Carr, Winston Churchill,
Reinhold Niebuhr, and Us: The Case for Principled, Prudential, Democratic Realism" (314-53), Robert G. Kaufman argues that classical realism
is an essential but incomplete framework for understanding international
politics. Realism correctly emphasizes the importance of power, geopolitics,
man's ineradicable flaws, and the constraints of anarchy, but it "needs supplementing with additional factors drawn from the idealist tradition: the
necessity for transcendent moral standards by which to judge relative degrees of moral and geopolitical evil; a recognition of the pacifying effects of
stable liberal democracies vis-a-vis one another; the importance of the domestic determinants of foreign policy for promoting cooperation or exacerbating international conflict" (315).
Kaufman contrasts Carr's realism to the realism elaborated by Churchill
and Niebuhr. For Kaufman, Carr's realism is excessively structuralist, collectivist, and morally relativistic. It resembles neorealism in that it exaggerates the realm of necessity in international politics. The versions of realism
which Churchill and Niebuhr advance offer "a more appealing middle
ground, which may help us escape the equally grave dangers of either an
unrealistic realism or an unrealistic idealism" (325).
Like Churchill's, Niebuhr's transcendent realism distinguished revisionist
regimes from status-quo regimes, and limited revisionism from unlimited
revisionism. These distinctions were based on the internal characteristics of
the key states in the international system. Like Churchill, Niebuhr took
ideology seriously. It was not merely a rationalization for the will to power,
but a partially autonomous force which could either restrain the ambitions
of regimes or make them even more ruthless and dangerous.
For policymakers, a Churchillian-Niebuhrian realism, according to
Kaufman, offers the most promising approach for seeking freedom and
security as complementary objectives in a still-dangerous world, a world in
which the interests of states clash and the correlation of forces remains
critical. Churchillian-Niebuhrian realism avoids the errors of idealist utopianism, without succumbing to the moral relativism of pessimistic realists
such as Carr. 1bis realism admits "the powerful constraints of international
anarchy, without denying, as structural realists do, the importance of ideology, regime type generally, and liberal democracy particularly" (351).

xxu

ROOTS OF REALISM

Niebuhr's and Churchill's realism remains relevant today. The relative


anarchy of international politics and human nature will continue to constrain, but not make meaningless, moral choice in international politics.
"Rejecting either utopian moralism or cynical realpolitik, Niebuhr and
Churchill largely succeeded in advocating the significantly lesser of geopolitical and moral evils," Kaufman writes. "Both evaluated moral choice in
international politics not just by its intentions, but by its consequences"
(352). Churchill and Niebuhr thus offer a richer version of realism which
synthesizes systemic and domestic explanations of international politics.
Their realism also shows that there is no need to choose between a morally
relativistic realism, on the one hand, or utopian idealism, on the other hand.
"What we need is a sober, not somber realism, which incorporates liberal
democratic ideals, a transcendent moral standard by which to judge the
lesser evil, a recognition of man's capacity for good as well as his immutable capacity for evil, and a modest faith in the possibility, albeit contingent,
of a slow progress to a more peaceful but eternally imperfect world. This is
why Churchill and Niebuhr always will remain so important for us" (353),
Kaufman concludes.
Normative realism. Patricia Stein Wrightson, in her "Morality, Realism, and
Foreign Affairs: A Normative Realist Approach" (354--86), addresses the
accepted convention that realism excludes moral concerns from questions
of foreign policy. In fact, she writes, the realist approach to morality and
foreign policy is more ambivalent. It is true that realists argue that morals
are too subjective a standard with which to judge politics, but they also insist that realism is the most moral approach to foreign policy because it
promotes stability among political adversaries.
Wrightson first examines how cold war realists such as Hans Morgenthau and George Kennan addressed the moral question in foreign affairs.
She then turns to a discussion of Reinhold Niebuhr's approach to international politics, which she calls "normative realism." In defining normative
realism, Wrightson shows how Niebuhr avoids the confusion and cynicism
that seem to characterize the traditional realist approach to the moral question in foreign policy.
Morgenthau's answer to a world in which there are no simple moral
choices is to reject conventional moral standards, and instead infuse concepts such as the national interest and the balance of power with ethical
value. His effort to incorporate a moral component into realism is constrained by his bleak view of human nature and by his insistence that the
moral sphere is autonomous from the political sphere (385).

Introduction

xXlll

Niebuhr, however, asserts that policymakers can and should address the
moral ambiguities inherent in political action. He sees the balance of power
and the pursuit of the national interest as means to an end-justified if they
promote moral ends and unjustified if they do not. The difficulty of conducting oneself in a moral way on the world scene is a challenge to be embraced, not an insurmountable obstacle in the face of which statesmen
should give up any hope of moral choice. "For Niebuhr, man's will-topower, his insatiable self-interest, are the givens of collective life. As such,
they thwart the possibility of achieving a simply virtuous collective life, but
they do not destroy it altogether" (386). Niebuhr is not naive. He does not
believe that greater attention by policymakers to the moral aspects of foreign policies will bring utopia to earth or put an end to war. "What he does
believe, rather, is that such conduct can make the world a more livable
place" (386). This is not a small achievement.
The essays in this volume do not represent a comprehensive or systematic discussion of realism. They rather highlight and give expression to the
major debates within realism: should the analytical foundation of realist
theory be a theory of the individual or a theory of the state (or what Robert
Gilpin calls the "conflict group'')? Are domestic variables (national character, political culture) important in shaping state conduct? Do states seek
security, or are they engaged in a continuous effort to accumulate more
power? What is the role of moral considerations in the conduct of states?
These essays offer different answers to these questions, but in addressing
these issues they show two things: first, realism is not a single theory but a
family of interpretations. While realism has an identifiable center of philosophical gravity, it accommodates different readings; second, the diversity
and richness of realism are a testimony to its vitality and robustness. It is
now fashionable to proclaim that the end of the cold war and the demise of
the Soviet Union herald the obsolescence of realism. Because the root
causes of human and societal interaction have not changed since Thucydides, the insights realist readings have offered over the years are as relevant and valid today as they have always been. They provide a sturdy foundation on which to base an analysis of politics among nations, and the
practice of such politics.

Page Intentionally Left Blank

I. RECONSTRUCTING REALISM

Page Intentionally Left Blank

RECONSTRUCTING POLITICAL REALISM


THE LONG MARCH TO SCIENTIFIC THEORY

ASHLEY

J. TELLIS

differences in the approaches, methods, and


formulations of various individual realists, there is little doubt that
political realism constitutes a coherent tradition of explaining political behavior. Centered on an understanding of politics as a permanent
struggle for power and security, political realism has consistently sought to
explain how entities seek to preserve themselves in an environment characterized by pervasive egoism and the ever-present possibility of harm. The
presence of egoism implies that all entities value only themselves; the interaction of many such entities creates a situation in which each becomes a
limitation on the security, freedom, and ambitions of the others; and the
competition which results is characterized by each entity constantly jostling
with other entities in an attempt to preserve its own power and enhance its
own safety. Realist approaches thus perceive politics primarily as a conflictual interaction. Consequently, their analyses of political behavior center,
at least in the first instance, on a positive description of how political entities cope with ubiquitous threats in the face of unending security competition. This positive description of political actions aimed at enhancing security constitutes the "minimum realist program." This program has been
articulated differently in diverse formulations beginning with the antiquarian philosophic-historical reflections of Thucydides and culminating in
contemporary social-scientific theories, the most prominent of which is the
systemic-structuralism of Kenneth Waltz.
This article reconstructs the realist research program with a view to understanding how it has evolved over time. By analyzing the works of five

ESPITE SUBSTANTIAL

Ashley J. Tellis is associate policy analyst at RAND, Santa Monica, California.


The author wishes to thank Cara Cantarella, Dale Copeland, Daniel Deudney, Markus
Fischer, Benjamin Frankel, Henk Goemans, Stacy Haldi, Mark Kauppi, Craig Koerner, John
Mearsheimer, Duncan Snidal, Dhun Tellis, Bradley Thayer, Eduardo Velasquez, and Stephen
Walt for their numerous comments and helpful suggestions.

ROOTS OF REALISM

individuals who represent distinctive turning-points in the evolution of


political realism, it will reconstruct how the realist program, which began as
a philosophical reflection on the nature and behavior of security-seeking
entities, has gradually been transformed-however imperfectly-into the
abstract, deductive formulations which modem social science demands.
Given this objective, it is important to recognize that this article is primarily
a "rational reconstruction" 1 and not a historical narrative. As such, it is
fundamentally an interpretation which recreates the internal history of the
realist research program. The term "internal history," in tum, refers to the
various substantive and methodological problem shifts marking the evolution of that research program.2
This emphasis on internal history does not make a rational reconstruction identical to a chronological narrative in which the evolution of a theory
or the growth of a discipline is described simpliciter. It is also not a history of
ideas, in which the development of a given "unit-idea" is systematically
traced out with respect to both time and usage in order to make manifest
the "exclusively logical progress in which objective truth progressively unfolds itself in a rational order."3 A rational reconstruction is an attempt to
explain, as rationally as possible, the growth of objective knowledge in
terms of the normative methodology provided by a philosophy of science.
Far from aiming for a theoretically uncontaminated chronology (as is often
thought to exist in popular conceptions of narrative history), or for the
systematic exposition of individual concepts (as is specifically sought out in
the history of ideas), a rational reconstruction settles for a more modest
objective: interpreting the significant problem-shifts defining the growth of
a particular research program when viewed from the perspective of both
substantive formulation and methodological approach. Thus, this essay
includes elements of narrative history and elements derived from the history of ideas, without being reducible to either or both of these disciplines.4
1. The term "rational reconstruction" has been borrowed from Imre Lakatos, "History of
Science and its Rational Reconstructions," Bos/on St11dies in the Philosopl[y of Saena 8 (1970):
91-136.
2. This focus on "internal history" implies that the question of how various problem
shifts are linked to the broader social conditions from which they emerge--an issue tackled
almost exclusively by "external history"--lies outside the province of this inquiry and is,
therefore, neglected. The distinction between internal and external history has been discussed
in Thomas S. Kuhn, "Notes on Lakatos," Bos/on St11dies in the Philosopl[y of Saene 8 (1970):
140--41.
3. Arthur 0. Lovejoy, "Reflections on the History of Ideas," ]011mal of the History of Ideas 1
(1940): 21.
4. This definition of rational reconstruction varies from lmre Lakatos' usage of the term.
Lakatos sought to use rational reconstructions to compare various competing methodologies
of science. That objective cannot be replicated here because the discipline of international

Political Realism: The Ung March to Scientific Theory

CRITICAL RATIONALISM AND RATIONAL RECONSTRUCTION OF REALISM

reconstruction of political realism scrutinizes sequentially


the contributions of three traditional realists-Thucydides, Niccolo
Machiavelli, and Hans Morgenthau-and two scientific realists-Morton
Kaplan and Kenneth Waltz-with a view to understanding how their work
comports with the methodological criteria for good scientific theory as
identified by the critical rationalist philosophy of science. The critical rationalist tradition is based on the founding work of Karl Popper. Developed further by Joseph Agassi, Paul Feyerabend, Imre Lakatos, David
Miller, John W. N. Watkins, and others,5 it offers contemporary realists
methodological solutions which avoid either the bankrupt inductivism of
logical positivism or the abandonment of objectivity by postmodernism. 6
Thus, it is particularly congenial to the scientific reformulation of political
realism because it considers objective social scientific knowledge to be both
possible and desirable. Furthermore, it encourages the creation of such
knowledge without pretending that possession of it would lead either to
HIS RATIONAL

relations and the research program of political realism are relatively young as social sciences
and, therefore, do not lend themselves to testing competing methodologies of science.
Hence, the technique of rational reconstruction used here is intended primarily to trace the
growth of the realist research program when viewed against Popperian prescriptions for the
construction of scientific theories.
5. The chief sources of the critical rationalist tradition are Karl Popper, The Logic of Scientific Discovery (New York: Basic Books, 1959); Karl Popper, Conjectures and Refutations (New
York: Harper & Row, 1965); Karl Popper, Oijective Knowledge (Oxford: Clarendon Press,
1986); Karl Popper, The Open S ociery and Its Enemies, vols. 1 and 2 (Princeton: Princeton University Press, 1971 ); Karl Popper, The Poverry of Historicism (London: Ark Paperbacks, 1961 );
Karl Popper, Realism and the Aim of Science, ed. W.W. Bartley III (Totowa: Rowman and Littlefield, 1983); Joseph Agassi, Science in Flux, Boston Studies in the Philosophy of Science
(Dordrecht: D. Reidel, 1975); Hans Albert, Treatise on Critical Reason, trans. Mary Varney
Rorty (Princeton: Princeton University Press, 1985); W. W. Bartley III, The Retreat to Commitment (New York: Knopf, 1962); David Miller, Critical Rationalism (La Salle: Open Court,
1994);]. W. N. Watkins, Science and Scepticism (Princeton: Princeton University Press, 1984);
lmre Lakatos, Philosophical Papers, vols. 1 and 2, ed. John Worrall and Gregory Currie
(Cambridge: Cambridge University Press, 1978); Paul Feyerabend, Philosophical Papers, vols. 1
and 2 (New York: Cambridge University Press, 1981).
6. Although sometimes identified as two distinct schools, critical rationalism actually encompasses those concerns usually associated with scientific realism. The conventional wisdom often identifies the former school with the deductive-nomological conception of science and the latter school with the search for causal mechanisms or underlying structures.
This distinction is superficial at best because the scientific laws sought by critical rationalism
are not isolated universal statements but part of a theoretical lattice which describes certain
observable behaviors deduced from a conception of underlying (perhaps unobservable)
structures. Therefore, the attempt to describe scientific realism as essentially different from
critical rationalism is unsustainable, and the former is treated as subsumed by the critical
rationalist philosophy of science for the purposes of this article. For a good discussion of
how critical rationalism is scientifically realist in both intent and structure, see Popper, Realism and the Aim of Science, 11-158.

ROOTS OF REALISM

epistemic certainty or to social and political utopia. The methodology proffered by critical rationalism thus enables scientific realists to produce theoretical formulations that are consistent with the deepest intuitions of their
historical predecessors, the traditional realists.
The three principal components of critical rationalist methodology, especially as applied to the social sciences, are conjectural knowledge, deductive systematization, and methodological individualism. 7
Cof!J"ectura/ knowledge. Conjectural knowledge was Karl Popper's celebrated
solution to the "problem of induction." The problem of induction, a traditional epistemological conundrum, arises from the fact that there appears to
be no logical way in which a discrete number of singular observations can
be used to justify the truth of any universal statement. Induction as a methodology, therefore, constrains the very activity which is of most interest to
science: the discovery and enunciation of true universal laws. Most philosophers now acknowledge that induction is a problematic procedure.
Popper's solution is an improvement because it offers the hope of enunciating true universal statements even as it affirms that "the ideal of certainty is quite barren"8 and, therefore, ought not to be pursued by science.
The basic idea underlying conjectural knowledge is. simple and straightforward. The worth of a scientific hypothesis is judged not by the empirical
base from which it is derived, but by its ability to pass rigorous tests. Any
hypothesis-no matter what its source or inspiration and no matter how
outlandish-is admissible. It is accepted for scientific consideration so long
as it can be empirically tested and, by implication, falsified. Consequently,
the value of a hypothesis or conjecture is not dependent on a scientist's
ability to justify it a priori through confirming procedures (as the inductivist
methodology of logical positivism traditionally demanded), but rather by its
ability to survive severe and prolonged testing.
Deductive rystematization. The objective of expelling false conjectures by
severe criticism and testing, in turn, generates the second requirement that
all conjectures be deductively systematized. Deductive systematization is
necessary for three reasons:
First, it enables the scientist logically to draw out the maximum number
of inferred, non-obvious, consequences from a given conjecture. This is
important because, other things being equal, the greater the number of test7. The following discussion of critical rationalist methodology is based on the more detailed exposition which appears in Ashley J. Tellis, "The Drive to Domination: Towards a
Pure Realist Theory of Politics" (Ph.D. diss., University of Chicago, 1994), 18-80.
8. David Miller, "Conjectural Knowledge: Popper's Solution to the Problem of Induction," in In Pursuit of Truth, ed. Paul Levinson (Atlantic Highlands: Humanities Press, 1982),
22.

Political &a/ism: The Long March to Scientific Theory

able consequences, the easier it is to falsify a particular conjecture. A more


fecund conjecture is preferable to one which is less: it possesses greater
"potential satisfactoriness"9 relative to its competitors. Other things being
equal, therefore, it is judged preferable to them.
Second, it helps establish internal coherence within a theoretical framework. This is necessary because, as Popper notes, while coherence does not
establish truth, "incoherence and inconsistency do establish falsity; so if we
are lucky, we may discover inconsistencies and use them to establish the
falsity of some of our theories." 10
Third, deductive systematization is indispensable because it provides the
means by which a theoretical conclusion can be falsified through the application of the logical rule of modus to/lens. This implies that although theories
can never be conclusively verified, scientific knowledge can nevertheless
grow because, whenever certain evidence secures provisional acceptance by
the scientific community, syllogistic logic can be used to demonstrate that a
theory is false.11
Methodological individualzsm. In addition to conjectural knowledge and deductive systematization, critical rationalism also argues that methodological
individualism is necessary for the growth of genuine social scientific knowledge. Methodological individualism is the stipulation that all social behavior
and all social institutions be explained by reference to the behaviors of in9. Popper, Conjectures and Refutations, 217.
10. Ibid., 226.
11. Ibid., 64. Because falsification thus becomes the only means by which scientific
knowledge can truly grow, even if only negatively, formulating deductively coherent theories
is of great importance. The foundations of such theories are, no doubt, conjectural and may,
in fact, even be utterly arbitrary. If, however, these conjectural foundations give rise to deductively coherent theoretical systems whose conclusions may, in turn, be falsified by severe
testing, it is possible that our knowledge may increase as a result. If these conclusions are
falsified when matched against accepted observation statements, we may know that at least
some hypothesis is not true. If, on the other hand, the conclusions are not falsified when
matched against accepted observation statements, nothing happens. That is to say, the unfalsified hypothesis is retained as one as-yet unrefuted explanation whose intrinsic truth we are
not-and may never be--certain of. This as-yet unrefuted explanation may never be considered true in any serious epistemological sense, "for no theory has been shown to be true, or
can be shown to be true" (Popper, Oijective Knowledge, 21). If it has survived more severe tests
than its competitors, however, we may rationally prefer it as a basis for practical action. The
critical rationalist tradition thus asserts that while we may occasionally discover that some of
our conjectures are false, we will never be certain that even our most successful conjectures
are true. Yet, because they are continually exposed to criticism, testing and possible revision,
the growth of objective knowledge is possible. Such knowledge-even if it refers only to
knowledge of those conjectures shown to be false-is objective nonetheless because the
process of criticism is a public activity open to inter-subjective examination and discussion.
Through such activity, it is possible that certain speculatively proffered hypotheses will survive long enough to provide us with some plausible universal laws explaining certain phenomena. If so, we will have secured these laws through a method not vulnerable to the debilitating defects of induction.

ROOTS OF REALISM

dividuals in specific situations. These behaviors, in tum, are a function of


an individual's preferences. When one such individual interacts with another, a specific structure of interactions arises. This structure, then, proceeds to constrain all individuals into exhibiting certain behaviors. These
behaviors can also give rise to particular social aggregates and if so, these
aggregates (or social "wholes'') can thereafter be explained simply as a result of the specific interindividual interactions arising from a given situational constraint.
Using the individual as the unit of analysis in this fashion precludes the
need for either psychological or biological explanations, and this methodology is justified by critical rationalism on two counts: the ontological and the
epistemological. The ontological justification derives from the fact that
critical rationalism denies the genuine existence of wholes beyond the
properties and interactions of their parts. 12 This denial is conditioned,
among other things, by a desire to avoid the pitfall of reification, which is
the pretense that social wholes exist a priori as unproblematic entities. The
critical rationalist asserts that social wholes like "markets," "firms,"
"states," and the "international state system," to offer just a few examples,
do not exist in fact. Rather, they are epiphenomena! entities-economizing
abstractions-constituted merely for theoretical purposes and, hence, are
not real in the sense that only individuals can be. Therefore, if an analysis
involving these social wholes is desired, the wholes must-in principle or in
practice-first be derived as products of the unintended consequences of
purposive interactions among conscious individuals.
Since social wholes do not meaningfully "exist"-and this is the essence
of the ontological argument-it follows that we cannot know them as such.
This is the essence of the epistemological argument. As Friedrich A. von
Hayek noted, "what of social complexes are directly known to us are only
the parts ... the whole is never directly perceived but always reconstructed
by an effort of our imagination." 13 Reconstruction by deductive logic is,
12. The best discussion of the critical rationalist position on wholes and parts can be
found in D. C. Phillips, Holistic Thought in Social Science (Stanford: Stanford University Press,
1976).
13. F. A. von Hayek, The Counter-Revolution ofScience (Indianapolis: Liberty Press, 1979), 93.
What is also entailed by this claim of methodological individualism is that, whereas we may
be eternally agnostic about what the true units of the natural sciences are (agnostic because
our knowledge of the fundamental units is constantly revised from generation to generation),
acting individuals alone are the true explanatory units of the social sciences. This essentially
means that while it may be possible to explain social outcomes by further reducing even
individuals to more primitive constituents, like faculties, genes, or atoms, such explanations
would not be, properly speaking, social scientific explanations but, rather, philosophical,
biological, or physical explanations. If, in fact, a genuine reduction of this sort is possible,
Popper would wholeheartedly support it (though he is quite skeptical both of the possibility

Political Realism: The Long March to Scientific Theory

therefore, fundamental because all social wholes are theoretically constituted entities. In the absence of an accepted constituting framework, it
would be impossible either to recognize their existence or to agree on
whether any given whole is to be considered an appropriate unit of analysis.
This simple insight has profound consequences for the growth of social
scientific knowledge-a point that becomes clear through the following
example. It is not at all uncommon to find Marxist theorists denying that
states are critical to the explanation of international political behavior. On
the other hand, realist scholars, concerned with states as actors, often deny
that ideologies, social forces, or classes are relevant to international political
explanations. Because the debate is framed in terms of such macroscopic
entities-entities lacking an unambiguous ontological status-all such deliberations are bound to be inconclusive. As a result, the dispute over which
units of analysis are legitimate (for purposes of explanation) remains unresolved and, to that extent, the growth of knowledge itself is retarded. The
only way out of this logjam, the critical rationalist argues, is to reduce all
social wholes to their elementary parts-the individual-and then attempt
to synthesize them by deductive logic. Those wholes which cannot be generated in terms of the rational behaviors of the elementary entities would
then be jettisoned; those wholes which can be so generated would be retained as legitimate units of analysis. Deductive causal explanations, utilizing these wholes, could thereafter be created.

SINGLE- AND MULTIPLE-EXIT MODELS

The ontological and epistemological arguments taken together, therefore,


boil down to the claim that because individuals alone, not social wholes, are
recognized as existent by all social scientists irrespective of their theoretical
or ideological affiliation, only a deductively systematized theory, which
bases its explanations on a priori ascribed properties of individuals, can
survive as a good model of social scientific explanation. As far as the formal requirements for such an explanation are concerned, the principles of
deductive systematization and methodological individualism combine to

of complete physical reductionism and its ability to provide explanatory completeness). Even
if such reduction were possible, however, the social sciences would nonetheless still have
great utility though at an intermediate rather than at an "ultimate" level of analysis. The social
sciences, therefore, by definition must explain social outcomes in terms of acting individuals
and asserting this does not in any way preempt the possibility that one day even acting individuals could be further decomposed into more primitive units by other disciplines.

10

ROOTS OF REALISM

demand the creation of single- or multiple-exit models. 14 These models are


intended to isolate the single or multiple rational choices open to any actor
in the face of some generated situational constraint. The simplest such
models are static models. These are relatively easy to create, even if a particular discipline is in its infancy. As a discipline develops, however, it is
possible to create more complex models of actor behavior, like comparative
static and possibly even dynamic models.
Regardless of which kind of model is actually developed at any given
time, it is important to recognize that the single-exit version represents the
preferred ideal from an explanatory point of view. This is because singleexit models incorporate sufficient information about the situational constraints facing a given actor such that only one decision is seen as rational
from the perspective of that actor. Single-exit models are thus attractive
because they make for determinate explanations. When creating such single-exit models is not feasible, however, multiple-exit models become necessary. These models represent the second-best, because they succeed only
in delineating a range of possible actions, rather than being able to identify
the one single and unique rational choice facing a given actor. Multiple-exit
models, therefore, introduce indeterminacy in an explanation and, to that
degree, are less preferable in comparison to single-exit models.
While the type of model that can be developed at any point in time is
often a function of the relative sophistication of a discipline, it is important
that some kind of exit model be created. It would not be an exaggeration to
say that the goal of social science consists primarily of developing theoretical exit models which expose the characteristic behavior of actors in some
defined situation. Developing these models is crucial because they serve as
"ideal types": they enable the observer to systematically identify the kinds
of constraining forces at work in some situation and, thereafter, to understand and predict the typical, appropriate, rational, behavior which any hypothetical actor would exhibit when confronted by the situational compulsions encoded in the model. 15 Exit models thus serve two functions: first,
they capture the causal mechanics at work in some hypothesized situation;
14. The logic and structure of various exit models has been discussed in Spiro Latsis, "A
Research Programme in Economics," in Method and Appraisal in Economics, ed. Spiro Latsis
(Cambridge: Cambridge University Press, 1976), 1-41.
15. It is not difficult to perceive that this "situational deterrninist" methodology proffered
by critical rationalism incorporates, with some modifications, the methodological legacy of
Max Weber with its emphasis on creating "ideal types" in the service of "ideal-typical"
analysis. The relationship between Weber's methodological program and critical rationalism's
view of social scientific analysis is explored in J. W. N. Watkins, "Ideal Types and Historical
Explanation," in Modes of IndividNalism and Collectivism, ed. John O'Neill (London: Heinemann,
1973), 143--65.

Political &a/ism: The Long March to Scientific Theory

11

second, they uncover the rational behavior of agents in those situations in a


formal way. The isolation of such law-like regularities through situational
logic provides the very stuff of which empirical testing and, it is hoped,
falsification, are made.
To sum up, the critical rationalist tradition suggests that an ideal social
scientific theory will have three characteristics. First, it will not be inductively justified; second, it will be an internally coherent deductive system
taking the form of static, comparative static or dynamic, single- or multipleexit models; third, it will be methodologically individualist in form.

APPLYING CRITICAL RATIONALIST METHODOLOGY

These three characteristics will now be used as yardsticks to assess the work
of Thucydides, Machiavelli, Morgenthau, Kaplan, and Waltz. Each of them
will be interpreted "synthetically," meaning that their individual works are
understood as interrelated parts of a larger, unified argument. This approach presumes that each theorist possesses a coherent weltanschauung
from which various given works radiate like spokes from the hub of a
wheel: any single piece of writing can be used to illuminate the others.t6
Such a presumption is legitimate for purposes of a reconstruction since
what is attempted is not an exegesis of the various texts per se, but an interpretation of how they encode the realist program when scrutinized in
terms of the following considerations.
First, what is the principal substantive claim offered by the theorist in
question? The inquiry here will focus on establishing how the realist program, focused as it is on power and security, is embodied in the principal
written contributions of the theorist concerned and what its role is in terms
of the evolving growth of scientific knowledge.
Second, what is the methodological form used by the theorist to advance
his substantive claims? The analysis here will focus on the manner in which
the substantive claims are derived-inductively or deductively-and
whether these claims are presented in the form of a unified theoretical system or merely as an exposition of pragmatic insights drawn from and useful
to the practice of statecraft.
Third, what are the units of analysis utilized by the theorist to frame his
theoretical claims and how does reliance on these units cohere with the
theorist's larger understanding of the causes of conflict? The investigation

16. This metaphor has been borrowed from Mark Hulliung, Citizen Machi01Jtl!i (Princeton:
Princeton University Press, 1983), 230.

12

ROOTS OF REALISM

here will seek to establish whether individuals, states, or state systems have
either analytical or theoretical primacy17 in the author's work, and whether
the causes of conflict are located in the nature of man, the internal organizations of societies, states, and governments, or in the character of the political system as a whole-a classificatory structure now identified as first-,
second-, and third-image explanations of political behavior by Waltz. 18
In addressing these questions, this reconstruction will view the progress
of political realism as a long-but yet unfinished-march from philosophical-metaphysical reflection to modem social-scientific theory. It will conclude with an evaluation of the present state of political realism as science,
an analysis of its limitations, and suggest how the critical rationalist methodology of science can provide the means by which realist theory can be
reconstituted in a way which rids it of its present limitations and significantly expands its explanatory scope.

THUCYDIDES: THE FOUNTAINHEAD OF THE POLITICAL REALIST TRADITION

written for The Enryclopedia of Philosophy, Thucydides is described as a "political observer, not a speculative philosopher," whose
"work builds self-consciously on a clearly defined philosophical position, an
antimetaphysical naturalism and positivism which he probably learned from
the practitioners of Hippocratic medicine and from the Sophists, who
dominated the higher education of fifth-century Athens." 19 This reading of

N AN ESSAY

17. The distinction between "analytical" and "theoretical" primacy is important to methodologically individualist approaches in social science. This is because all individualist theories are faced with the problem of explaining how rational individual action gets transformed
into collective social outcomes or particular social wholes, many of which may not even have
been intended by the acting agents. To solve this problem satisfactorily, a particular conception of individuals and individual action is required such that both social wholes and individual choices are integrated seamlessly. If this integration is satisfactorily achieved, it is possible
for a social scientific discipline to accord analytical primacy to some social whole of its
choosing (for example, "states," "classes," etc.), while still being able to demonstrate that
these preferred wholes are derived from individuals who continually have theoretical primacy
in the explanatory system. In other words, successful social scientific explanations recognize
individuals to be the "efficient causes" of social outcomes (and, hence, are accorded theoretical primacy), while the social wholes that arise are viewed primarily as emergent outcomes
which are economical for purposes of large-scale analysis (and, hence, are accorded analytic
primacy). For a good discussion of this question, see Reinhard Wippler and Siegwart Lindenberg, "Collective Phenomena and Rational Choice," in The Micro-Macro Unk, ed. Jeffery
C. Alexander, et al. (Berkeley: University of California Press, 1987), 135-52.
18. Kenneth N. Waltz, Man, the State and War (New York: Columbia University Press,
1954).
19. Paul Edwards, ed. The Enryd1Jpedia of Phikisopl?J, vol. 8 (New York: Macmillan, 1967),
s.v. "Thucydides" by William T. Bluhm.

Political Realism: The Long March to Scientific Theory

13

Thucydides as a perceptive historian who shares both the modem empiricist spirit and perhaps even its scientific method,20 is commonplace among
social scientists, and it is particularly congenial to contemporary realists because it allows a fairly straightforward appropriation of his insights for purposes of theorizing about international politics. Thus, Morgenthau explicitly uses the Thucydidean formulation that "identity of interests is the surest of bonds whether between states or individuals," to justify his conception of the primacy of the national interest. 21 Similarly, Waltz credits
Thucydides for being among the earliest theorists preoccupied with the
problem of "the use of force and the possibility of controlling it." 22 Robert
Gilpin, perhaps his most systematic redactor, not only describes Thucydides as "the first scientific student of international relations"23 but also
interprets his work as setting forth a "theory of hegemonic war" where "the
uneven growth of power among states is the driving force of international
relations."24 Operating within the horizon of positivist history, several generations of realist scholars have thus read Thucydides as a historian concerned with documenting the dynamic patterns of conflict and strife as they
relate to the growth and decline of hegemonic states in world politics.25
While admitting that such readings of Thucydides are both legitimate
and, in the case of Gilpin, especially instructive, a rational reconstruction of
the Thucydidean achievement cannot rest content with viewing his work as
simply a magisterial history of the rise and decline of great powers. It can20. For a classic reading of Thucydides as a practitioner of modem scientific history see
G. F. Abbott, Th11rydides: A St11t!J in Historical &ality (New York: Russell & Russell, 1970).
21. Hans). Morgenthau, Politics among Nations, 4th ed. (New York: Knopf, 1967), 8.
22. Kenneth N. Waltz, Theory of International Politics (New York: Random House, 1979),
186.
23. Robert Gilpin, "The Richness of the Tradition of Political Realism," in Neortalism and
its Critics, ed. Robert 0. Keohane (New York: Columbia University Press, 1986), 306.
24. Robert Gilpin, "The Theory of Hegemonic War," jo11'71al of Interdisciplinary History 18,
no. 4 (spring 1988): 591.
25. Such empiricist readings of Thucydides have increasingly come in for criticism, usually
from critical theorists who argue that the Thucydidean opus is less about power politics in a
scientific sense than it is about how discourse interacts with material constraints to produce
specific political outcomes. See, by way of example, Daniel Garst, "Thucydides and Neorealism," International Studies Q11arter!J 33 (1989): 3-27; and Laurie M. Johnson Bagby, "The Use
and Abuse of Thucydides in International Relations," International 01!,aniz.ation 48 (1994):
131-53. Criticism of this sort, however, may be overstated from a methodological point of
view because none of the realists who interpret Thucydides pretend to provide authoritative
exegeses of the text. Rather, they are content to read it for its insights into power politics.
From a substantive point of view such criticism-while useful and cautionary-may be
overstated as well because a "scientific" reading of Thucydides, especially if conducted in
accordance with the critical rationalist notion of "situational logic," can readily accommodate
the argument that ideas influence individuals in their decision calculus. The relative degree of
influence, however, can be determined only by reconstructing the actual situation that any
given individual finds himself or herself in.

14

ROOTS OF REALISM

not treat his work as yet another empiricist "history" because both the architectural structure of the text and the philosophical content suffusing its
historical narrative preclude reading The Peloponnesian War as a descriptive
chronicle of facts. 26 1bis reconstruction, therefore, comprehends The Peloponnesian War as a classic form of "epic"27 writing in which two levels of
reality are described and reflected upon in an intertwined fashion. On one
hand, the work describes events occurring at a world-historical level, where
different diplomatic and military initiatives within and between states are
narrated and analyzed in a linear fashion. At this level, it may be described
as an analysis of the causes and consequences of Athenian imperialism insofar as it affected both the domestic body politic and its external relations
with the other Greek city-states. On the other hand, it is a deep philosophical reflection on the nature of political decay and moral dissolution as exhibited both in the cosmic realm of order and in the human realm of politics. At this level, it may be described as an attempt to delineate the archetypal form (eidos) of a violent convulsion (kinesis)-a convulsion which
Thucydides judged to be without parallel or precedent (1.1) and which, if
correctly understood, would provide unique insight into all convulsions yet
to come. Both these levels of analysis constantly interact such that even
simple narratives at the world-historical level always convey an ineffable
sense of preternatural drama and tragedy, while the profound philosophical
reflections about nature and artifice, order and decay, as conveyed through
the various speeches in the text, always speak-at some level-to simple,
and sometimes trite, issues of political, military, and strategic choice. By
recognizing that both these levels inextricably pervade the work, it is possible to read Thucydides, as Leo Strauss correctly interpreted him, as a
"philosopher-historian" who sought to do justice to both the demands of
philosophy as well as of history.28
From a methodological point of view, Thucydides' technique appears
"inductive" in character in that it attempts-following the Hippocratic
method-to give a detailed observational account of the "symptoms" or
the "form" of the "disease" or "convulsion" in question.29 The positive
26. All quotes from Thucydides' The Peloponnesian War are cited in accordance with convention within the text by Book and Paragraph numbers. All quotes are drawn from Thucydides, The Peloponnesian War, trans. Richard Crawley (New York: Random House, 1951).
27. For a succinct statement about the character of epic writing see Luciano Canfora,
"Epic and Historical Claim to Totality," PoZ!'an Studies in the Phi/osop1!J of the Sdences and the
Humanities 5 (1979): 37-54.
28. Leo Strauss, "On Thucydides' War of the Peloponnesians and the Athenians," chap.
in The Ci!) and Man (Chicago: University of Chicago Press, 1978), 139-45.
29. Eric V oeglin, Order and History, vol. 2, The World of the Polis (Baton Rouge: Louisiana
State University Press, 1957), 353-58.

Political Realism: The Ling March to Scientific Theory

15

description of these symptoms at the world-historical level is remarkable:


first, for its detail; second, because its selection of facts is not determined
by a mythological view of history; and, third, because it attempts, even if
only implicitly, to induce from the described "symptoms" a deeper understanding of the universal essence of disorder. 30 Hence, it is no surprise to
find Eric V oeglin describing Thucydides "as the first craftsman who tried
to transform the empirical knowledge of politics into a science, using the
science of medicine as his model for this purpose."31
The attempt to derive an archetypal model of disorder from a single observation-event implies that Thucydides saw the Peloponnesian war as
more than just a random political incident. Viewing it, instead, as an exemplar which furnishes universally valid insights-and doing so in a way that
cannot be duplicated by any iteration of other apparently similar eventshe confidently offers this single case "as an aid to the interpretation of the
future" (1.24). Such confidence stems, in the first instance, from the inductivist belief that the future will resemble the past, even if it does not repeat
it entirely (1.24). In the final instance, however, it derives from the apparently simple claim that this was "a great war, and [hence] more worthy of
relation than any that had preceded it" (1.1). Consistent with this claim,
Thucydides goes to great lengths in the Archaeology to establish that this
kinesis is without parallel or magnitude in history. Its exemplary importance-its "greatness"-is validated by three crucial reasons which are
mentioned only innocuously.
First, the warring parties are Greek and all the Hellenic races are bracing
in one way or another for the quarrel (1.1). Second, the Peloponnesian
kinesis seems to envelop both the human order pertaining to political relations within and between the cities and the cosmic order as manifested in
the disorders of nature and the universe (1.24). Third, "the preparations of
both the combatants were in every department in their last stage of perfection" (1.1). The implications of these reasons are consequential to Thucydides. To begin: if a polity, like the Greek, so advantaged by all its achievements and so committed to attaining good political order, could be on the
verge of a devastating political and moral breakdown, then, Thucydides
reasons, such a kinesis could not be just another random world-historical
event; rather, it had to be an upheaval of truly paradigmatic proportions,
whose truth would be worthy "as a possession for all time" (1.23).32 Fur30. Ibid.
31. Ibid., 356-57.
32. Werner Jaeger, Paideia: The Ideals of Greek C11/t11re, vol. 1, trans. Gilbert Highet (New
York: Oxford University Press, 1939), 410-11.

16

ROOTS OF REALISM

ther, the paradigmatic nature of this upheaval is confirmed by the fact that
the cataclysm in the earthly city finds reflection in the fractures of nature
itself; these fractures, then, came to represent cosmic reminders of the significance of this extraordinary political event. 33 Finally, because the kinesis
occurs when the combatants are at the peak of their skill and perfection,
they bequeath more of enduring value "for all time" than if examined when
truly inconsequential.
Thucydides' attention, therefore, centers on the epoch when the bipolarity of the system is at its tightest, when both the Athenian and Spartan cities
are at the peak of their achievements, and when their citizens exhibit peculiarities of character in sharpest relief.3 4 Now, he could capture them, in a
manner of speaking, as entities writ large, when both the strengths and the
weaknesses of the state system, the cities and their citizens are magnifiedthus making those easier to observe, understand and reflect upon.
This attempt at reflecting upon the "human things" when they are at
their apotheosis, then, compelled Thucydides not merely to describe a set
of significant experiences but to create a new unit of understanding which
is now recognized as the Peloponnesian War. It is important to acknowledge that whereas today most readers would readily identify-thanks to
Thucydides-the Greek convulsion between 431 B.C. and 404 B.C. as the
Peloponnesian War, it is unlikely that Thucydides' contemporaries were
aware of any such fact. They had witnessed over a century of conflict, beginning with the Median expeditions against the Hellenes in 490 B.C. and
culminating with the defeat of Sparta and the rise of Theban hegemony at
Leuctra in 371 B.c. 35 Until 404 B.C., when the Thucydidean opus was not yet
published, the history of the Greek wars was perceived as nothing other
than a loosely connected series of events. This perception is reflected in
Thucydides' work itself when, during the Decelean War, the Lacedaemonians are portrayed as considering the fault to be their own "in the former
war" (21.19). It was Thucydides' genius that, from the large number of sequential conflicts, he selected carefully and unified together just four conflicts occurring between 431-404 B.C.-the Archidamian and Decelean
wars, the Peace of Nicias, and the Sicilian Expedition (the last did not even
involve the Lacedaemonians directly)-to create a new "event" for poster-

33. This issue is discussed in Elizabeth Young-Bruehl, "What Thucydides Saw," History
and Theory 25 (fall 1986): 1-16.
34. Peter J. Fliess, Th11rydides and the Politics of Bipolarity (Baton Rouge: Louisiana State University Press, 1966), 13-79.
35. The history of this century of conflict has been usefully summarized in ibid., 3-11.

Political Realism: The Long March to Scientijic Theory

17

ity.36 In a real sense, therefore, "The Peloponnesian War" is a creation of


Thucydides. It has been created, never having occurred as such. Hence, he
does not describe his opus-in contrast to his modern translators-as a
"history" (historia), despite the fact that the word was commonly known and
accepted in Greece at the time. Instead, he uses the peculiar locution
"::irynegraphe ton polemon'' which, literally translated, implies that he "wrote
up" the Peloponnesian War.37
The use of the term .ryngraphe conclusively suggests that the work, far
from being a mere historical narrative, was in fact a construction of a great
"deed worthy of speech"-a construction which, while describing "the
harsh grandeur, ruggedness, and even squalor"38 of power politics in the
earthly city, would open itself to limning the eternal eidos of dissolution and
decay. Given this fact, any rational reconstruction of the realist tradition
must admit that Thucydidean "science," no matter how modern it appears
in intention and how sophisticatedly inductivist it appears in technique,
represents ultimately a "metaphysical research program."39 That is, the scientific component of Thucydides, which deals with cause-and-effect at the
world-historical level of politics, is suffused by a set of larger philosophical
ideas about the inevitability of decay afflicting even the most admirable order as that represented by Periclean Athens. This overarching philosophy
not only conditioned Thucydides' selection of the facts, his creation of the
narrative, and his redaction of the numerous speeches made by the various
participants, but it also--most fundamentally-limited the freedom of action he could accord to those individuals faced with the situations he describes in ostensibly "positivist" detail. This metaphysics of necessitywhich, as Voeglin put it, resists "the temptation to obscure the dilemmatic
structure of political existence by any attempt at rationalization" 40pervades the entire .ryngraphe, and from it derives the ethic of tragedy and
despair which has since been appropriated fully by political realism. The
presence of such a metaphysical thematic right at the very founding of the
realist tradition, then, serves to confirm the Popperian insight that all science ultimately is rooted in and born out of metaphysics and that
"metaphysical ideas belong to scientific research as crucially important

36. Voeglin, The World of the Polis, 349-51.


37. Ibid.
38. Strauss, "On Thucydides' War," 139.
39. The phrase is Popper's, and a succinct summary of the concept can be found in Paul
A. Schilpp, ed., The Philosopl?J of Karl Popper, vol. 1 (La Salle: Open Court, 197 4), 133-43.
40. Voeglin, The World of the Polis, 364.

18

ROOTS OF REALISM

regulative ideas [just as] scientific [discourse] belongs to the rational debate
concerning metaphysical ideas."41
Given this background, the Thucydidean .ryngraphe--when viewed in
terms of the realist program at the world-historical level-begins with a
conventional third-image explanation of the causes of conflict. Distinguishing between real and apparent causes of the war, Thucydides declared that
"the real cause I consider to be the one which was formally most kept out
of sight. The growth of the power of Athens, and the alarm which this inspired in Lacedaemon, made war inevitable" (1.24). War, in this sense, was
engendered by an anarchic system, in which the absence of superordinate
authority resulted in states becoming sensitive to the security threats posed
by the uneven growth of other states. The primacy of coercive military
power in this context is abundantly emphasized early on by the Corcyraeans
who, possessing a navy second only to that of Athens, seek Athenian assistance for their cause, despite the recognition that such assistance would
violate Athenian obligations to Sparta:
For your first endeavor [the Corcyraeans admonish the Athenians]
should be to prevent, if possible, the existence of any naval power except your own; failing this, to secure the friendship of the strongest that
does exist. And if any of you believe that what we urge is expedient, but
fear to act on this belief, lest it lead to a breach of [your] treaty [with
Sparta], you must remember that on one hand, whatever your fears,
your strength will be formidable to your antagonists; on the other, whatever confidence you derive from refusing to receive us, your weakness
will have no terrors for a strong enemy. (2.36)

Thus, military power alone is salient in security competition among egoists


and it cannot be replaced by confidence derived from other instruments
such as legal treaties or friendship. Failure to recognize this, while dangerous at the best of times, may be fatal when "the breaking out of war .. .is all
but upon you" (2.36). This arfirmation of the value of military power over
any mitigating principle, though provoked by the demands of security
competition, is exacerbated by the bipolarity of the system in which each
hegemon seeks to supplant the other. As a result, the two hegemons are
not only drawn into other third-party conflicts on opposing sides, but their
own bilateral struggle also provides opportunities for the smaller states to

41. Joseph Agassi, "The Nature of Scientific Problems and Their Roots in Metaphysics,"
chap. in Science in Flux, 212.

Political Realism: The Long March to S cient!ftc Theory

19

entrap them into alliances aimed at resolving the local security competition
between these small states.42
These twin processes recur endlessly throughout the text and they provide Thucydides with the opportunity to underscore the concept of necessity, where political entities are forced to engage in unpalatable behavior for
reasons beyond their control. Thus, the Athenian Empire, once in existence, is viewed as "a major cause of the murder and devastation of the
war-but the compulsion of historical necessity [is seen to] enforce the
Empire."43 This leitmotif, which has since been appropriated by the realist
program in its explanations of politics, is never asserted by Thucydides in a
proposition emanating from himself. It appears, however, as a pervasive
motif in the speeches made by the various participants.
Having initiated the third-image explanation of why conflict arises and
what the role of coercive military power is in such circumstances, Thucydides promptly proceeds to decompose this structural explanation by offering a detailed second-image account of why the Peloponnesian war became
inevitable. This explanation, framed magnificently in the Corinthian speech
at the Congress at Lacedaemon, is a specific and pointed description of
how the character of a particular state-in this case, the spiritedness of
Athens-was responsible for the creation of an empire without parallel in
Hellenic history:
They are swift to follow up a success, and slow to recoil from a reverse.
Their bodies they spend ungrudgingly in their country's cause; their intellect they jealously husband to be employed in her service. A scheme
unexecuted is with them a positive loss, a successful enterprise a comparative failure. The deficiency created by the miscarriage of an undertalcing is soon filled up by fresh hopes; for they alone are enabled to call
a thing hoped for a thing got, by the speed with which they act upon
their resolutions. Thus they toil on in trouble and danger all the days of
their life, with little opportunity for enjoying, being ever engaged in getting: their only idea of a holiday is to do what the occasion demands,
and to them laborious occupation is less of a misfortune than the peace
of a quiet life. To describe their character in a word, one might truly say
that they were born into the world to take no rest themselves and to
give none to others. Such is Athens, your antagonist. (3. 70)

If third-image explanations initially framed the context of events, there is


little doubt that second-image explanations now suddenly acquire new
42. Fliess, Th11rydides and the Politics of Bipolarity, 85-106.
43. David Grene, Man in His Pride: A St11tfy in the Political Philosopl?J
(Chicago: University of Chicago Press, 1950), 64.

of Th11rydides and Plato

20

ROOTS OF REALISM

prominence as a causal explanation of the Peloponnesian War. Thus, the


spiritedness of the Athenians takes center stage, and it is justified both as
that virtue which saved Hellas from the Mede and as a consequence of the
universal laws of politics.44 The Athenians, therefore, are depicted as arguing that the empire actually began when the other states inspired by Athenian spiritedness joined it to fight the Medes, and it was transformed into a
domination based on superior coercive power only when "fear be[came
their] principal motive, though honor and interest afterwards came in"
(3.75). In other words, state character-spiritedness-produced a systemic
outcome-empire-which brought in its trail the logic of inescapable necessity-continued domination out of fear, interest, and honor-and this
logic is ordained by an ubiquitous natural law:
It follows that it was not a very wonderful action, or contrary to the
common practice of mankind, if we did accept an empire that was offered to us, and refused to give it up under pressure of three of the
strongest motives, fear, honor and interest. And, it is not we who set the
example, for it has always been the law that the weaker should be subject to the stronger. (3.76)

Fear, honor, and interest thus conspire-as necessity-to create a pragmatic, amoral, rationality of political action which seeks to preserve the
empire at all costs. 45 The natural law undergirding this logic reaches its
most systematic enunciation in the Melian Dialogue, where it is reasoned
that "the strong do what they can and the weak suffer what they must"
(17.90). This argument is then used by Thucydides to describe how the
Athenian drive to imperial domination, and its effort to maintain that
domination through additional conquests, results in the triggering of various opposing balances. Because this dynamic is part of the logic of domination resulting from egoist competition, neither hegemon nor challenger
can escape it. It must be accepted for what it is: a natural process (3.77).4<>
While attention is thus focused on Athenian spiritedness as the cause of
empire and the depredations flowing from it, such is but half the story. The
relative shift in the balance of power between Athens and Lacedaemon is
caused equally by the latter's inaction and passivity, an argument articulated
clearly in the Corinthian speech at the Second Congress at Lacedaemon:

44. Jacqueline de Romilly, Thucydides and Athenian Imperialism, trans. Philip Thody (Oxford:
Blackwell, 1963), 312-13.
45. de Romilly, Thucydides and Athenian Imperialism, 270-72.
46. Grene, Man in His Pride, 61.