リアリズムが国際政治学をダメにした: ネオ・リアリズムの場合


今回はJohn A. VasquezThe Power of Power Politics: From Classical Realism to Neotraditionalism (Cambridge University Press, 1999)Part IIについて。Part IIは以下の構成になっています。

Part II: Neorealism and Neotraditionalism: International Relations Theory at the Millennium
9. Retrospective: neorealism and the power of power politics
10. The promise and potential pitfalls of post-modernism: the need for theory appraisal
11. The realist paradigm as a degenerating research program: neotraditionalism and Waltz’s balancing proposition
12. Mearsheimer’s multipolar myths and the false promise of realist policy prescriptions: the empirical inaccuracy of the realist paradigm
13. Challenging the relevance and explanatory power of the realist paradigm: the debate on the end of the Cold War
14. Conclusion: the continuing inadequacy of the realist paradigm

本書は国際関係論におけるパラダイムであるリアリズムを分析し、リアリズムが国際事象の説明・予測にことごとく失敗している のは、リアリスト自体の分析枠組みに問題があることを明らかにしています。今のリアリストは第三世代です。主要なリアリストは以下の通り。
1世代: Hans J. Morgenthau, E. H. Carr, Reinhold Niebuhr, and George Kennanなど
2世代: Kenneth Waltz and Robert Gilpin
3世代: John Mearsheimer, Stephen Walt, Joseph Grieco, Randall Schweller, Michael Mastanduno, Barry Posen, Jack Snyder, Kenneth Oye, and Stephen Van Everaなど

Part IIのメインターゲットはウォルツ、ギルピン、ミアシャイマーの3人、とくにケネス・J・ウォルツです。彼が国際政治学をダメにしたと批判しています。ウォルツは昨年5月に亡くなっています。享年88歳です。
Kenneth Waltz, Foreign-Relations Expert, Dies at 88

タイトル副題のFrom Classical Realism to NeotraditionalismClassical Realism (古典的リアリスト)は第1世代を指します。第2世代はNeorealism (ネオリアリズム)と呼ばれますが、Vasquezはウォルツ以降の質的な実証研究をするリアリストのことをNeotraditionalism(ネオ伝統主義)と命名しています(通常は第2世代と第3世代はともにネオリアリズムと呼ばれます)。イギリスの国際政治学者Hedley Bullが計量分析に批判的な質的分析のことを伝統主義(traditionalism)と呼びましたが、そのニュー・バージョンということです。しかし、本書を読んでもなぜneoなのかよくわかりません。彼が新著でneotraditionalismという言葉にこだわった一番の理由は1983年に出版された前著(本書のPart Iに収録)でおこなった計量分析によるリアリズム批判が酷評されたためです。大半のリアリストは統計に興味をもっていないのに統計分析で批判するとは何たることか、ということです。だから新著では質的分析を用いたリアリストに対して質的な観点から批判を試みます。その批判の対象となるのがneotraditionalistということです。

The best way to address the objection that the conclusion of the original text could not be accepted (because it focused on quantitative findings and quantitative scholars) was to look at non-quantitative research. Examining this research would be a logically compelling way of demonstrating that the anomalies the realist paradigm needs to explain away are not exclusively associated with the use of a particular method. Empirical research that is done well should not produce different results depending on the research techniques employed; statistical, historical, and comparative case studies should produce convergent findings. (p.3)

Part IIが焦点を当てるのはネオリアリズム、ネオ伝統主義の勢力均衡研究、J・ミアシャイマーの一連の研究、冷戦終焉に関する論争、の4点です。まずはネオリアリズムについて。

ネオリズムの模範テキストはKennth N. Waltz1979年に出版したTheory of International Politicsです。『国際政治の理論』というタイトルで翻訳もされています。次がRobert GilpinWar and Change in World Politics (1981)です。

Part Iではトマス・クーンの科学論が援用されましたが、Part IIではイムレ・ラカトシュのリサーチ・プログラム論を用い、とくにリアリストの唱える勢力均衡論が「前進的リサーチプログラム」(progressive research program)と「退行的リサーチプログラム」(degenerating research program)のどちらか分析されます(pp.7-8)


9: 前著を回顧して: ネオリアリズムとthe power of power politics

9章では、1983年に出版されたThe Power of Power Politics1980年以降に国際政治学で展開されたネオリズム論争と照らし合わせて、今もその議論が妥当であったかどうか再考します。

The Power of Power Politics (1983)が明らかにしたのは以下の4点。

Vasquezが前著で批判したのは1960年代以前に活躍したリアリストでした。彼は前著の出版をきっかけにリアリスト批判が強まることを期待しましたが、リアリズムはKenneth N. WaltzTheory of International Politics (1979)Robert GilpinWar and Change in World Politics (1981)の出版後、ネオリアリズムとして生まれ変わります。(p.190)


Waltz’s main contribution to the field was to bring in structuralism. Theory of International Politics can be seen as basically a systematization of Morgenthau’s thought cast in a logically rigorous and parsimonious frame that subordinates all other levels of analysis. In doing this, Waltz raises in importance three factors that played a more marginal role in Morgenthau’s theory of power politics. First, anarchy is now given a pre-eminent role in the realist paradigm, a role that it did not have in Politics Among Nations. This emphasis later played an important role in the debates between neorealism and neoliberalism. Second, the balance of power becomes a key law (behavioral regularity) in international relations that theory must explain. Morgenthau never saw the balance of power as automatic (see Claude’s “1962: 25-37” analysis) but as a foreign policy that leaders had to follow in order for it to be implemented. Furthermore, Morgenthau (1960: ch. 14) was skeptical about the merits of the balance of power as a foreign policy. Third, the number of actors in the “great power” system (bipolar or multipolar) was seen as critical, something which Morgenthau again did not see as salient for explaining international politics. More important than these specific original contributions, however, was the elegance and deductive rigor with which Waltz rewrote classical realism to create a structural realism. Even his critics cannot deny the intellectual achievement of melding structuralism with realism. This, when coupled with language from economics, at a time when political economy was grabbing the attention of the field, accounts for much of the influence Waltz has commanded during the last two decades. (p.191)

ウォルツのTheory of International Politicsは様々な論者から批判されます。Ruggie, Keohane, Ashley, Coxらによる批判はRobert O. Keohane, ed., Neorealism and Its Critics (1986)に収録されています。(pp.192-3)


In many ways, the work of the 1980s and 1990s has extended and deepened the criticisms made of the first and third assumptions in the original text (see ch. 8, above). The work of Putnam (1988) and of Rosecrance and Stein (1993), among many others, demonstrates the need to open up the black box, this time by bringing in the effect of domestic politics. Likewise, recent quantitative work, for instance Bueno de Mesquita and Lalman (1992), has shown that domestic politics is critical in explaining the onset of war and has more of a direct impact than other levels. The findings that democracies do not fight each other (Russett 1993; Ray 1995), as will be discussed more extensively below, have been particularly damaging for the systemic perspective and for the realist paradigm generally, because they are so unexpected and anomalous for the paradigm. Also, recent work on World War II has shown that one of the main factors leading to war was a shift in foreign policy toward territorial expansion that was brought about by regime changes in Germany and Japan (see Vasquez 1996a, 1998), a conclusion consistent with current findings in the comparative study of foreign policy (Hagan 1993, 1994) that domestic leadership changes can bring about major changes in international relations (see also Stein 1994). Even those sympathetic to neorealism have seen the need to bring in second-image variables, as does Snyder (1991), in his account of overexpansion, and as Van Evera (1984, forthcoming) does by looking at the cult of the offensive and the role of offensive and defensive weapons (see also Jervis 1978; Snyder 1984). (pp.195-6)

Putnam, Robert D. (1988) “Diplomacy and Domestic Politics: The Logic of Two-Level Games,” International Organization 42 (Summer): 427-460.Rosecrance, Richard, and Arthur A. Stein (eds.) (1993) The Domestic Bases of Grand Strategy, Ithaca: Cornell University Press.
Bueno de Mesquita, Bruce, and David Lalman (1992) War and Reason, New Haven: Yale University Press.
Russett, Bruce M. (1993)
Grasping the Democratic Peace, Princeton: Princeton University Press.
Ray, James Lee (1995)
Democracy and International Conflict, Columbia: University of South Carolina Press.
Vasquez, John A. (1996a) “The Causes of the Second World War in Europe: A New Scientific Explanation,”
International Political Science Review 17 (April): 161-178.
(1998) “The Evolution of Multiple Rivalries Prior to the Second World War in the Pacific” in P. Diehl (ed.)
The Dynamics of Enduring Rivalries, Urbana/Champaign: University of Illinois Press, pp. 191-224.
Hagan, Joe D. (1993)
Political Opposition and Foreign Policy in Comparative Perspective, Boulder: Lynne Rienner.
(1994) “Domestic Political Regime Change and Foreign Policy Restructuring” in J. Rosati, J. Hagan, and M. Sampson, III (eds.)
Foreign Policy Restructuring: How Governments Respond to Change, Columbia: University of South Carolina Press, pp. 138-163.
Snyder, Jack (1991)
Myths of Empire: Domestic Politics and International Ambition, Ithaca: Cornell University Press.
Van Evera, Stephen (forthcoming)
The Causes of War, Ithaca: Cornell University Press.
Jervis, Robert (1978) “Cooperation under the Security Dilemma,”
World Politics 30 (January): 167-214.
Snyder, Jack (1984)
The Ideology of the Offensive, Ithaca: Cornell University Press.


Given these intellectual currents, it is not surprising that a third criticism of neorealism is of Waltz’s concept of anarchy. For Ashley (1988), anarchy is not something that is given in nature, but a social conception that constructs reality. Wendt (1992) extends this line of reasoning by analyzing how anarchy is constructed and how power politics arises. Power politics and the self-help system that Waltz identifies are not caused by the structure of anarchy (i.e. the absence of hierarchical government). For Wendt (1992: 394) there is no logical reason why power politics and self-help grow out of anarchy. Wendt maintains that power politics and self-help are not automatic system effects, but grow out of how actors treat each other, in other words, the pattern of their interactions. If actors are predatory and threatening, then you get power politics. Power politics is a function not of structure, but of a process (of learning) (Wendt 1992: 391) that gives rise to identity and interests (of oneself and of others). This set of meanings (whether one is an enemy or friend) determines whether the structure of anarchy will be associated with power politics, indifference, or cooperative security. Behavior in anarchy is not predetermined; “anarchy is what states make of it” (Wendt 1992: 394-395). (pp.197-8)

Ashley, Richard K. (1988) “Geopolitics, Supplementary Criticism: A Reply to Professors Roy and Walker,” Alternatives 13 (January): 88-102.
Wendt, Alexander L. (1992) “Anarchy is What States Make of It: The Social Construction of Power Politics,”
International Organization 46 (Spring): 391-425.


Whether the absence of government produces the violent chaos Hobbes supposed needs to be determined by empirical research and not by the meaning of words or “anthropological” thought experiments. The history of the modern state system does not seem to reflect the constant state of war Hobbes envisioned. Many states have been at peace for very long periods, and even states that have been involved in war have not been involved in all the wars they logically could have been, given the extent of their military reach. Empirically, Wallensteen (1984) has shown that there are clear periods of peace among major states, and Small and Singer (1982: 59-60) show that war is much rarer than is commonly thought, once a precise definition of war is used as a basis of measurement. In fact, more civil wars and revolutions occur in many periods than do interstate wars (see Small and Singer 1982: 233). The mere presence of hierarchical government cannot insure the absence of violence, nor does the absence of government insure the presence of violence. At the basic empirical level, Waltz (1959, 1979) may simply be wrong about war and anarchy. (p.200)

Wallensteen, Peter (1984) “Universalism vs. Particularism: On the Limits of Major Power Order,” Journal of Peace Research 21 (3): 243-257.
Small, Melvin, and J. David Singer
Resort to Arms: International and Civil Wars, 1816-1980, Beverly Hills: Sage.


While Waltz (1979: 114) is prepared to deny the world can be “reliably peaceful,” he does not say that there is no order whatsoever. In fact, one of his queries is to try to explain how an order can emerge “without an orderer” (Waltz 1979: 89). For the answer, he turns to microeconomic theory, and ends up making the balance of power a law that brings order out of potentially pervasive chaos. Because such great emphasis is placed on power, as would be expected in any preeminent theory produced by the realist paradigm, other possible sources of order, specifically rules, norms, and institutions, are seen as impotent (almost by definition). What is significant about Waltz, then, is not that he will deny that order can emerge, but that the order he sees is so narrow, and the possibilities for more and/or different kinds of order so limited. If one kind of order can emerge from this anarchy, why cannot another (Vasquez 1995: 133)? (pp.200-1)


What then does the neorealist conception of anarchy supply, if not an accurate description or explanation of the system? Let me venture to say, realist discourse has made global anarchy a constructed condition that institutionalizes how actors (in a jungle) should treat each other in their relationships (in order to survive) (Vasquez 1993: 282). Reflecting on one set of experiences, namely those associated with the most devastating wars within the war-torn Euro-centric context, it has generalized one set of traumatic experiences to all experience. Realist discourse thereby helps construct a normative reality to which states are then prodded and advised to conform by realist intellectuals from Machiavelli to Kissinger. Since those prescriptions are not always followed and fail to guide in certain realms, realist theory often falls short as a description and predictor of actual practice. It hangs on, however, in part because of its familiarity and its institutionalization in certain critical governmental and academic circles, and in part because certain of its practices are followed, even though they do not always work out the way they were intended. (p.202)


Waltz, however, cannot even say this. He can either predict constant warfare (which he does not do presumably because he knows it is untrue), or he can do what he does – say that an anarchic system permits war, but fail to tell us how or why it comes about in any one specific instance. It is the latter that constitutes the great failure of Waltz’s work, albeit a failure of omission. For a paradigm and theory whose main concern has been security and survival in the face of the threat of war, it still does not have any precise idea as to what makes war come about! In more technical language, Waltz’s permissive cause, by definition, leaves unspecified either the sufficient or necessary conditions of war. (p.204)


This review has shown that despite neorealism’s ability to articulate the realist paradigm in new directions, it has still failed to produce accurate explanations of international politics that are able to pass empirical tests. In addition, at the center of the paradigm’s concern, neorealism, like classical realism before it, has failed to provide a theoretically complete explanation of war. (p.212)


More importantly, the structure of the system does not exist as he depicts it. It is not as anarchic as his stipulative definition of anarchy makes it out to be. In modern times, it has certainly not been the kind of Hobbesian state of nature that has characterized realist discussions of the system (see Bull 1977; Alker 1996). Its level of governance and order varies, both over time and by issue area (Milner 1991). Anarchy/hierarchical government or anarchy(chaos)/order are much better conceived as a continuum than as a dichotomy (Milner 1991; Vasquez 1992: 854; 1993: 268). The paradigm’s conception of the system as anarchic has hidden two of the real fundamentals of the system that have profoundly shaped its order and nature; namely, that it has been an international capitalist system and that it has been an international legal system constructed around the rule of state sovereignty. Focusing on the anarchy of the system has hidden these other structural characteristics that are probably more important. (p.212)


In addition, focusing on structure to the exclusion of other levels of analysis has proven to be too simple to account for the complexities of world politics (Snyder and Jervis 1993). The role of domestic politics cannot be left out (Putnam 1988; Rosecrance and Stein 1993; Hagan 1994); neither of course can the foreign policy interactions of states (Posen 1984; Vasquez 1993). Parsimony and elegance are not a substitute for empirical accuracy. Theories need not include all possible variables (Vasquez 192: 230-231), but they should not exclude critical variables just because they may operate at a different level of analysis, especially if these have been shown to have an important impact on the behavior in question. (pp.212-3)

Hagan, Joe D. (1993) Political Opposition and Foreign Policy in Comparative Perspective, Boulder: Lynne Rienner.
Posen, Barry R. (1984) The Sources of Military Doctrine: France, Britain, and Germany Between the World Wars, Ithaca: Cornell University Press.


10: ポストモダニズムの約束と潜在的な落とし穴 ─ 理論評価の必要

1. The arbitrary nature of modernity
2. Choice posing as Truth
3. Reality is a social construction
4. Language and conceptual frameworks are prone to self-fulfilling prophecies
5. The process of identification and the construction of identity is a form of power and an act of violation (pp.215-220)

potential pitfallsは飛ばして、理論評価の基準についてだけ見てみます。以下の6つの基準を満たすほど「良い理論」ということになります。

The criteria of adequacy for empirical theory presented here are based on the assumption that a good theory must be true. The criteria are justified on the basis of the argument that following and using them increases the probability that an empirical theory, research program, or paradigm that satisfies the criteria is less likely to be false than one that does. If one prefers not to accept a philosophically realist view of theories (see Nagel 1961: 117-118, 141-152, 196), then in more instrumental terms, a theory that satisfies these criteria can be said to be more promising for achieving and making progress toward the ultimate goal of science, which is the acquisition of knowledge. There are six criteria (all of them standard in philosophy of science) relevant to international relations inquiry. “Good” empirical theories should be:
2. falsifiable
3. capable of evincing great explanatory power
4. progressive as opposed to degenerating in terms of their research program(s)
consistent with what is known in other areas
6. appropriately
parsimonious and elegant.
I label these, respectively, the criteria of accuracy, falsifiability, explanatory power, progressivity, consistency, and parsimony. (pp.229-30)


A set of propositions is accepted as satisfying the criterion of empirical accuracy if they consistently pass a set of reasonable and valid tests. Although theories are never proven and science is openended, theories whose propositions have passed tests can be tentatively accepted as accurate (and true), or at least not inaccurate and false. Conversely, theories that consistently do not pass tests can be regard as false or dismissed as no longer being useful guides to research. This is because if the purpose of scientific inquiry is to produce knowledge, then failure to produce strong and statistically significant associations is an indicator of the failure to produce knowledge. (pp.230-1)

explanatory powerとは

When two theories have passed tests and are vying for the allegiance of the scientific community, the criterion of explanatory power maintains that the theory that resolves puzzles and anomalies that could not be explained before, and predicts or explains new phenomena, is superior. (p.231)

accuracyexplanatory powerの区別がわかりにくいですねえ。実証テストにパスしたものはaccurateであり、パズルや例外事例を解決し、新しい現象を予測・説明できたものはexplanatory powerがあるという説明にいまいち納得できません。

この6つの基準の中で12の基準が特に重要だそうです。国際政治学者はparsimonious (少ない変数で多くの現象を説明できる)を重視し過ぎと述べています。

This is particularly the case, since some criteria are more important than others. Thus, the first two are essential; if a theory is not accurate or falsifiable (in at least the broad sense of specifying at some point what evidence would lead the theorist to say the theory was inaccurate), it cannot be accepted regardless of how well it satisfies the other criteria. Having great explanatory power is of little use if the explanation turns out to be inaccurate or is non-falsifiable. Likewise, the case for parsimony is often given too much weight in international relations. Theories, as Craig Murphy (personal communication 1993) argues, should have an appropriate degree of complexity. They should not include all possible variables without regard for their relative potency; nor should they leave out important factors to keep the explanation simple. What is crucial is that theories be able to pass tests – first in principle and then in fact. (p.232)


Finally, good theories should not contradict what is known in other fields of knowledge. Assumptions about motivation or cognition in international relations should be consistent with what is known (as opposed to theorized) in psychology. (p.233)

本書はpractical theoryを評価する基準も示しています。

While postmodernistm and post-positivism has made space within international relations inquiry for normative analysis, such work has not been very rigorous, and if it is to gain more respect, it too must have criteria for appraisal.Since the purpose of normative theory in international relations is to guide practice, it can be assessed in terms of the extent to which it provides an enabling function; that is, how well it guides practitioners. Throughout history most international relations theorizing has been devoted to this kind of practical theory. Practical theory can be appraised directly in terms of whether the theory actually provides information practitioners need to know and can use. A philosophy and theory of practice can also be tested indirectly by the policies and actions to which it gives rise. Practical theory, therefore, can be appraised both by looking at some of its intrinsic characteristics (e.g., the kinds of information it provides) and by the quality of policy prescriptions it produces. There are seven criteria of adequacy that can be applied to make such an appraisal. A “good” guide to practice must:
1. have a
good purpose and consequences
2. be able to be implemented in
3. provide comparatively complete and precise advice as to what should be done##
4. be
relevant to the most difficult policy problems of the day
5. have
anticipated costs (including moral costs) that are worth anticipated benefits
6. achieve success and avoid failure.
I label these, respectively, the criteria of goodness, practicability, completeness, relevance, anticipated utility, and success-failure. To this we can add a seventh, which is that:
7. the latent
empirical theory of a practical theory must be scientifically sound.

Part IIの第9章と第10章はとくに面白くはありません。Vasquezの議論が面白くなるのは次章からの、リアリズムの勢力均衡論、ミアシャイマーの二極化安定論、リアリズムの冷戦の終焉原因論に対する批判的分析です。

12月 21, 2014 · Pukuro · No Comments
Posted in: ☆社会科学

Leave a Reply