リアリズムが国際政治学をダメにした: ハンス・J・モーゲンソウの場合 2-3

Pocket

第3章: リサーチデザイン─リアリスト・パラダイムの定義化と操作化

第3章でVasquezはリアリスト・パラダイムの基本的仮定を明らかにし、1950年代以降の国際政治学がこのパラダイムに基づいて研究をしているかどうか確かめるために、リアリスト・パラダイムの「操作化」(operationalization)を行います。

リアリズムが国際政治学におけるパラダイムとすれば、リアリスト・パラダイムが国際政治学の①理論構築、②データ作成、③リサーチを導いているということ、が確認できるはずです。

Kuhn (1970a: ch. 3) points out that there are three major activities of any discipline theory construction, data collections, and research. Therefore, if the major proposition of this analysis is true, one would expect the following three propositions to be true:
1. The realist paradigm guided theory construction in the field of international relations in the 1950s and 1960s.
2. The realist paradigm guided data making in the field of international relations in the 1950s and 1960s.
3. The realist paradigm guided research in the field of international relations in the 1950s and 1960s. (p.46)

リアリズムの「模範例」はハンス・J・モーゲンソウが1948年に出版したPolitics Among Nationsです。本書は3つの基本的仮定をします。第1が国際政治における中心的アクターは国家であるという事。

Since a paradigm involves a set of fundamental assumptions made in the exemplar, the realist paradigm can be defined by delineating the fundamental assumptions in Politics Among Nations, as was done in the previous chapter. This is a valid procedure if Politics Among Nations was the most influential of all realist writings.

In order to define the fundamental assumptions of the realist paradigm, it is necessary to delineate the phenomena it focuses upon. Morgenthau focuses on two phenomena: nation-states and the struggle for power and peace. In doing so, he makes three fundamental assumptions delineated in chapter 2. The first assumption Morgenthau makes is that nation-states are the most important actors for understanding international relations. Why Morgenthau makes this assumption can be demonstrated by a simple syllogism:
1. Politics consists of a struggle for power, and in order to be a political actor a person or group must wield significant political power (true by definition).
2. In international politics, during the modern state system, only nations wield significant power (empirical statement).
3. Therefore, in international politics, during the modern state system, only nations are actors (conclusion).

Given the first two premises, the conclusion follows logically. (p.48)

第2の仮定は、国際政治は国内政治と根本的に異なるという事。

Morgenthau’s second assumption is that there is a sharp distinction between domestic politics and international politics. The use of the concept international politics as a way of demarcating the field assumes by its definition that there is something about politics that occurs outside nations that makes it different from politics that occurs inside nations. (p. 48)

Morgenthau points out in the same work that it is specifically the decentralized or anarchic system of international society that makes domestic politics different from international politics. Domestic politics is played in an arena where the government can legitimately and effectively regulate the actions of the actors, but in the world arena no such regulation occurs (Morgenthau 1960: 501-509 and ch. 19; 1973: 481-489 and ch. 19). In international politics, only nations have power, and their power can only be limited by the power of other nations. The sovereignty of nations, therefore, has an important effect on the way politics is played; hence a theory of international politics cannot be the same as a theory of domestic politics. It can be concluded that Morgenthau does in fact make what was delineated as the second assumption. (p. 49)

第3の仮定は、国際政治は権力と平和の闘争であるという事。

Morgenthau’s third assumption is that international relations is the struggle for power and peace. Morgenthau (1960: 23; 1973: 24) clearly states that the two concepts around which Politics Among Nations is planned are power and peace. It is evident from the following quotation that to Morgenthau (1960: 38; 1973: 40) international relations is a struggle for power and peace: “All history shows that nations active in international politics are continuously preparing for, actively involved in, or recovering from organized violence in the form of war.” (p.49)

リアリスト・パラダイムはこの3つの仮定をすることで、国際政治学者が何を研究し(何が重要か)、何を研究しないか(何が重要でないか)の暗黙の了解事項を生み出します。

By defining the purpose of his work in this manner, Morgenthau is doing what Kuhn has stated is the prerequisite of all scientific inquiry, that is, focusing on and magnifying certain phenomena while allowing other phenomena to disappear from the picture. There is nothing wrong with this procedure, and, as a number of philosophers of science point out (Kuhn 1970a: ch. 5; Hanson 1965: ch. 1; Popper 1970: 51-52), it would be logically impossible for science to proceed in any other manner. By providing a definition of international politics, Morgenthau states what he is going to study, what he is not going to study, and by implication what is important and not important (or less important) to study. (p.50)

実際に、モーゲンソウ以降の国際政治学はこの3つの仮定を受け入れているかどうか確かめるためにVasquezは実証分析をします。そのために彼は「操作化」の手続きをします。以下、国際政治分析におけるアクターと探究トピックのコーディングです。

ACTOR:
1. Intergovernmental organization (IGO)2. International nongovernmental organization (NGO)
3. Nation-state ←リアリスト・パラダイム
4. Subnational group or individuals
5. No actor
6. Any combination of 1, 2, 4
7. The nation and any other combination of actors

次に探究トピック
10. Conflict-cooperation ←リアリスト・パラダイム
11. Non-conflict-cooperation and non-power perceptions of decision makers
12. Non-(war/peace or power issues), issue positions of actors, and issue salience
13. Alignment and alliances ←リアリスト・パラダイム
14. Integration and regionalism
15. Magnitude of transactions
16. National power and/or weakness – including social, cultural, economic, political, and geographic characteristics; penetration, dependence, prestige, success, and failure ←リアリスト・パラダイム
17. Isolationism-involvement
18. Miscellaneous
19. Sociological characteristics of actors – age, party, education, religion, etc.
20. Propaganda
21. Supranationalism – support and participation in United Nations, League, or International Courts
(p. 53)

リアリスト・パラダイムが重視するトピックは国力(national power)、同盟(alliances)、国家間の対立/協力の3つです。

These three topics – national power, alliances, and inter-nation conflict-cooperation – constitute the central core of the realist paradigm. This conclusion is supported by a rank order of the common nouns in the index of Politics Among Nations, which showed that the three most frequently used nouns were balance of power (86 pages), national power (69 pages), and war (62 pages). (See Vasquez 1974a: 89-90, 92 for the data and evidence.) (p.54)

モーゲンソウは主に反証することを目的に理想主義者が唱える超国家主義(supranationalism)も分析対象にします。また、国益追求の手段としての宣伝(propaganda)も分析対象に入れます。

While inter-nation conflict-cooperation, national power, and alliances provided the central core, Morgenthau saw other topics as important for a complete understanding of the field even if they were not central. Because Morgenthau was writing in opposition to the idealists and at a time when many Americans hoped the United Nations could become, at some point, the foundation for a world government, supranationalism was an interesting topic to him. He had two main concerns with it. The first was to debunk illusions about the United Nations and argue that the United Nations simply reflected existing power relationships and the struggle for power. The second was to stipulate the conditions that create a stable supranational entity. He dwelt on what forces created a nation and how a world community (and from there a world state) might be created (Morgenthau 1960, 1973: chs. 27, 29, 30). It is because of this last concern that inter-state integration is labeled as part of the realist research program, even though Morgenthau’s own purpose was to show that the proper conditions for supranationalism did not exist. Finally, Morgenthau (1960: 338-345; 1973: 332-339) used the concept of propaganda to explain some of the verbal acts of states, although his concern with it was marginal. (p.55)

リアリストが研究対象にしないのは以下の通り。国際政治学者がこれらの研究をしていれば、「国際政治は権力と平和の闘争である」という第3の仮定の反証事例となります。

The following topics of inquiry were taken as indicators of work outside the realist paradigm: non-conflict-cooperation and nonpower perceptions of decision makers; non-war /peace issues, issue positions of actors, and issue salience; magnitude of transactions; sociological variables of actors – age, size, party, education, etc.; and finally a miscellaneous category to make the classification logically exhaustive. A scholar who studied one of these topics would be said to have rejected the third assumption. (p.55)

国家リーダー以外の国内アクターを分析対象にしたり、政策決定者の心理分析をする研究は「国際政治における中心的アクターは国家である」という第1の仮定の反証事例となります。

Studying aspects of decision making other than inter-nation power relationships suggests a topic of inquiry that sees decision making itself as the primary dependent variable. Such a perspective implies a rejection of all three assumptions. If individuals other than leaders of nation-states are studied, then the first two assumptions are violated, particularly if the explanations are psychological and/or social psychological, since this implies a single theory at the individual, group, and state level. The third assumption is also violated because the unique nature of global anarchy is not seen as affecting behavior. (pp.55-6)

安全保障以外のトピックを分析対象にすることは「国際政治は権力と平和の闘争である」という第3の仮定の反証事例となります。

Studying issues other than war/peace is seen as a rejection of the second and third assumptions, because Morgenthau (1960, 1973: 27) explicitly states that all substantive goals can be reduced to the struggle for power. To study different kinds of issues is to imply that the realist paradigm is applicable only to one aspect of the global system, and hence incomplete. (p.56)

Vasquezはアクターとトピック双方においてリアリスト・パラダイムの仮定に準じるときにのみその研究はリアリスト的という事ができるとします。つまり、片方でも準じていない場合は、リアリズムが国際政治学におけるパラダイムであるという主張の反証例になります。

A scholar’s work was coded as realist only if all the actor and topic categories were realist. For example, if a scholar studied nation-states but did not study them in the context of a realist topic of inquiry, then the entire work was coded as nonrealist, because the third assumption was rejected. In other words, all three assumptions had to be employed before a work was coded as realist. (p.56)

第4章: パラダイム指向の活動としての理論構築

理論は何もないところからは生まれない。基本的仮定を共有しなければ理論構築の営みを集団で行うことはできない。Vasquezは国際関係論でこの基本的仮定を提供するのがリアリスト・パラダイムであると主張します。

Kuhn’s notion that theory construction is a paradigm-directed activity has both an analytical and an empirical meaning. Analytically the proposition means that it is logically impossible to construct theories without the prior existence of a paradigm (Kuhn 1970a: 15-17). This aspect of Kuhn’s notion is substantiated by definition, since it is impossible to have a theory that does not make certain fundamental assumptions. Empirically, the proposition means that a single specific paradigm guides theory construction (Kuhn 1970a: 10-11). It is this empirical aspect that is embodied in the proposition that will be tested in this chapter: The realist paradigm guided theory construction in the field of international relations during the 1950s and 1960s. (p.60)

クーンによれば、通常科学(normal science)においてパラダイムが理論構築の方向性を指し示します。

According to Kuhn, theory construction in normal science involves clarifying the concepts presented in the dominant paradigm and employing them in light of research to elaborate theories. Kuhn calls theory construction paradigm articulation because the process is conducted by a division of labor, with different scholars working in specialized problem areas suggested by the research agenda of the paradigm. In a sense, the paradigm provides an outline, and theory construction articulates the paradigm by filling in the details. The paradigm provides guidance in that it focuses scholars’ attention on certain problems and provides them with a set of fundamental assumptions that the new theoretical work never violates. (p.60)

The need for paradigm articulation presupposes that the work that originally presented the paradigm did not provide all the answers. Kuhn (1970a: 23-24) states that a paradigm often does not provide any answers at all, only the promise of answers. How much and what type of articulation is necessary depends on the specific paradigm and the state of the science or field. In some fields, very little new conceptual formulation is needed. In other cases, particularly when the science is in qualitative stage, conceptual reformulation may dominate efforts at paradigm articulation (Kuhn 1970a: 29, 33). (pp.60-1)

リアリスト・パラダイムに基づく代表的研究は以下の通りです(pp.67-8)。
Carr, Edward Hallett (1939, 1964) The Twenty Years’ Crisis, New York: Harper and Row.
Claude, Inis L., Jr. (1956, 1964) Swords into Plowshares, New York: Random House.

Claude, Inis L., Jr.(1962) Power and International Relations, New York: Random House.
Waltz, Kenneth N. (1959) Man, the State, and War, New York: Columbia University Press.
Aron, Raymond (1966) Peace and War, New York: Doubleday.

以下の研究はリアリスト以外の概念的枠組みを使っていますが、リアリスト・パラダイムの3つの仮定は守っている(ホント?)とVasquezは主張します(p.68)。
Deutsch, Karl W. (1953) Nationalism and Social Communication, Cambridge, Mass.: MIT Press. ─ cybernetics or communication theory
Deutsch, Karl W., Sidney Burrell, Robert Kann, Maurice Lee, Jr., Martin Lichterman, Raymond Lindgren, Francis Lowenheim, and Richard Van Wagenen (1957) Political Community and the North Atlantic Area, Princeton: Princeton University Press.
Snyder, Richard C, H. W. Bruck, and Burton Sapin (1954) Decision-Making as an Approach to the Study of International Politics, Princeton: Foreign Policy Analysis Project, Princeton University. ─ decision-making
Wright, Quincy (1955) The Study of International Relations, New York: Appleton-CenturyCrofts.
Kaplan, Morton A. (1957) System and Process in International Politics, New York: John Wiley. ─ systems language
Schelling, Thomas C. (1960) The Strategy of Conflict, New York: Oxford University Press. ─ game theory

リアリスト・パラダイム外の研究として
Wright, Quincy (1942, 1965a) A Study of War, 1st and 2nd edns. Chicago: University of Chicago Press.
Deutsch, Karl W. (1964)
The Nerves of Government, New York: Free Press.
の2冊をあげています。

Finally, two books, Wright (1942) and K. Deutsch (1964), are seen as outside the paradigm. Wright employed both idealist and realist assumptions in his study of war, focusing on individuals as well as states, and attempting to develop a single theory of conflict and violence rather than one unique to the international system. Deutsch presented a general theory of government that violated the second assumption of the paradigm and wrote on a topic not directly relevant to the research agenda of the field. Indeed, most people in the field would probably say that it was a book about comparative politics, not international politics. (p.69)

14冊の主要著作を検討した結果、6冊が権力政治の概念的枠組みを用い、さらに6冊がリアリズムの3つの基本的仮定を守っており、1950年代から1960年代にかけてリアリストが国際政治学のパラダイムであったという仮説の反証例は2冊にすぎなかったとVasquezは主張します。

On the basis of the above classifications, it can be stated that six of the fourteen leading non-edited books (including Morgenthau 1948) employ a power politics conceptual framework and that an additional six do not violate the three basic assumptions and do study a topic of inquiry central in the realist paradigm. The first six can be seen as reflecting realism in the narrow sense as well as being traditional in methodology. The second six are behavioral and attempt to rework power politics propositions in a more scientific manner and to introduce new concepts to aid this task. Since only two books out of fourteen do not reflect work guided by the realist paradigm, it can be tentatively concluded that hypothesis 1c has failed to be falsified. (p.69)

Although preliminary, the tests in this section have shown that the scholar seen as contributing more to international relations inquiry than any other scholar is Hans J. Morgenthau; the work most frequently nominated by scholars as the leading work in the field is Morgenthau’s Politics Among Nations; and twelve of the fourteen books nominated by scholars in a sample survey as leading works in the field do not violate the three fundamental assumptions of the realist paradigm. All these findings provide support for the proposition that the realist paradigm guided and directed theory construction in the field in the 1950s and 1960s. (p.69)

第5章: パラダイム指向の活動としてのデータ作成

トマス・クーンによれば通常科学における「事実の収集」(fact gathering)、つまり「データ作成」(data making)はパラダイムによって規定される。パラダイムに基づく概念の規定化によって、無尽蔵の現象の中から何が重要な事実で何が重要でないかが振り分けられるのだ。

Kuhn (1970a: ch. 3) explicitly states that fact gathering (i.e., data making) in normal science is guided by the dominant paradigm in the field; such guidance is necessary because the world consists of numerous phenomena, and phenomena only take on meaning to the extent that they are conceptualized. Conceptualization, as pointed out earlier, is a function of theory construction or paradigm articulation. Facts, then, presuppose a paradigm that sifts through the welter of phenomena to focus on what is important. In the pre-paradigm stage of science, fact gathering tends to be random because there is no single paradigm to distinguish the chaff from the wheat (Kuhn 1970a: 16-17). In normal science, however, fact gathering becomes highly directed, not only because the paradigm focuses on certain phenomena, but because fact gathering usually “consumes much time, equipment, and money” (Kuhn 1970a: 25). Consequently, the gathering of facts becomes a highly selective activity. (p.77)

「データ作成」とは何か。

Data making is the process by which facts are measured and quantified so that they can be used for hypothesis testing (see J. D. Singer 1965, 1968: 2). This definition is similar to that of Kuhn (1970a: 25-28), who defines fact gathering as not only the observation and recording of facts, but their transformation by measurement techniques into a form that allows them to be used to test hypotheses. (p.78)

1960年代の行動論革命以降、質的研究が中心であった国際政治学でもデータ作成プロジェクトが進み始める。Vasquezはこのデータ作成はリアリスト・パラダイムの基本的仮定に即した形で行われると予想します。

The appearance of two distinct periods of international relations data making, before the 1950s and during the 1960s, provides an opportunity to examine the validity of the proposition that data making in the field has been guided by the realist paradigm in the 1950s and 1960s. It would be expected that data making in the 1960s would be guided by the realist paradigm while data making prior to the 1950s would not. (p.80)

1960年代にリスリストが重視する「国家属性」(national attributes)、「対外政策行動」、「戦争」を中心にデータが作成されます。国家属性は独立変数で、後の2つは従属変数としてリアリスト仮説を検証するのに利用されます。「対外政策行動」はとくに安全保障分野での国家行動に焦点が当てられます。

The second occasion of data making in the field occurred in the early 1960s and continues through the present. In the 1960s, data making was centered on three areas – national attributes, foreign policy behavior, and war. As a result of the articulation of the realist paradigm that had been conducted in the 1950s, these three areas were the most obvious ones in which to collect data to test hypotheses related to the realist paradigm. (p.82)

National attributes, the first area, was highly relevant because Morgenthau and the other power politics theorists had maintained that knowledge of national power was a particularly revealing aspect of the conduct of international relations. (p.82)

The second and third areas, foreign policy behavior and war, were relevant to the paradigm because they provided the major dependent variables. They were the topics that the paradigm wanted to explain. If the proposition being examined in this chapter is accurate, then it would be expected that data collected in these two areas would reflect the characterizations of foreign policy behavior and wars made by the paradigm; that, is, foreign policy behavior would be viewed in terms of a struggle for power and peace (in other words, conflict and cooperation among nations), and war would be viewed as something occurring among nations and related to the balance of power. Since it was demonstrated in chapter 3 that these three areas were of central concern to the realist paradigm, it is only necessary to show that data were collected in these areas. (pp.82-3)

国家属性に関するデータについて。

The initial data published and made available to scholars in the 1960s were products more of comparative politics than international relations (see J. D. Singer 1968: 11-12). The collections consisted basically of attribute data on nations, and although not collected specifically with theories of international relations in mind, they provided a set of relevant variables on what Morgenthau had called the elements of national power. The major collections were A Cross Polity Survey (1963) by Banks and Textor and the World Handbook of Political and Social Indicators (1964) by Russett, Alker, Deutsch, and Lasswell of Yale. These two projects were initiated not only to collect data for specific research projects but to provide general data sets that could be used by scholars working on a variety of projects. Consequently, not only was the data published, it also was made available on computer tapes stored at the Inter-University Consortium for Political Research, which has become the data library for the entire field of political science. A Cross Polity Survey provided data from widely scattered sources on demographic, economic, cultural, and social characteristics of 115 nations. In addition, the authors provided new data on political characteristics through the use of content analysis. The World Handbook also provided data on the demographic, economic, cultural, social, and political characteristics of nations; however, none of the data was derived by coding. The World Handbook differed from the A Cross Polity Survey in that it provided more variables (75 versus 57), but most of these were not as a “political” as the coded data of Banks and Textor. (p.83)

戦争に関するデータについて。「戦争の相関関係」(COW)プロジェクトは戦争だけでなく、同盟、外交関係、政府間組織、国力のデータも作成・収集します。

War was the third major area in which data were collected in the sixties. The most extensive project in this area and the successor to the efforts of Richardson and Wright was the Correlates of War project of J. David Singer and Melvin Small (1972). (p.85)

In addition to collecting data on war, the Correlates of War project collected data on several important independent variables. The first effort was focused on alliance data, which clearly reflects the realist concern with the balance of power and with the polarity debate (see J. D. Singer and Small 1966a, 1968). The alliance data were collected first for the period 1815-1940 (J. D. Singer and Small 1966b) and then updated to 1965 (Small and Singer 1969). The second major area of data making for the project was on diplomatic ties from 1815 on (J. D. Singer and Small 1966b; see also Small and Singer 1973). These data initially served as a way of determining membership in the central system (see J. D. Singer and Small 1968) and were used by the end of the 1960s to test propositions on status inconsistency (Wallace 1970, 1971). Data were also collected on the number of intergovernmental organizations in the system from 1815 to 1964 (J. D. Singer and Wallace 1970). Each of these data sets was updated periodically, Finally, data on national capability began to be collected in the mid-sixties (J. D. Singer 1976: 27; J. D. Singer, Bremer, and Stuckey 1972; Ray and Singer 1979; Bremer 1980). (pp.85-6)

COWはリアリスト・パラダイムの影響を強く受けています。

The selection of these three independent variables and the order in which they received priority reflect the strong influence of the realist paradigm and its priorities. Alliances and national capability were thought by the early power politics theorists to be the two most important determinants of peace and war. The status-ordering data were taken as an indirect indicator of power, and the data on IGO’s were employed to test realist propositions (see J. D. Singer and Wallace 1970; Wallace 1972). (p.86)

Each of the major data sets, then, reflects the realist paradigm’s fundamental assumptions that nations are the most important actors and research should be focused on the struggle for power and peace. In addition, each data set was used to test specific hypotheses relevant to the realist paradigm. Consequently, it is not surprising that most of the data consist of national attributes and inter-state conflict and cooperation. The proposition being examined in this chapter maintains that such a result is not an accident but the product of the power of the realist paradigm to guide and direct scholarly activity within the field. Now that the proposition has been elaborated and its plausibility demonstrated, it is necessary to specify how it will be empirically tested. (p.87)

Vasquezは1950年代から1960年代の国際関係論でリアリスト・パラダイムに基づいてデータ作成がおこなわれているかどうか実証分析します。仮説2aでデータが国家およびリアリストが関心をもつトピックかどうか、仮説2bで変数が国力と国家間対立/協力の概念かどうか、テストします。

In order to test the proposition that the realist paradigm guided data making in the field of international relations in the 1950s and 1960s, it is necessary to operationalize data making and the realist paradigm. Since data making is the transformation of facts into variables for the purpose of hypothesis testing, data making can be operationalized as variables available to international relations scholars in a form that permits hypothesis testing. (p.87)

Two hypotheses can be derived from the proposition to test its adequacy:

2a. Variables available for use by international relations scholars will tend to provide information on nation-states and topics of inquiry that are labeled realist in the coding scheme.

2b. More variables will be available for use by international relations scholars on the two most central concepts in the realist paradigm – national power and inter-nation conflict-cooperation – than on other concepts. (p.89)

サンプル収集について(←unbiased sampleがちゃんと集められたかどうかはなはだ疑問)

Since the Inter-University Consortium for Political Research classifies and lists variables it has on file, the consortium’s list of all the variables in its international relations archive was taken as the sample. This provides a non-biased sample of data available in the field. It also includes the universe of data readily and routinely available to all scholars by the end of the 1960s. (p.90)

そして仮説検証。仮説2aをテストするために1,650の変数をリスリスト指標(アクターが国家でリアリストトピックを取り扱っている変数)と非リアリスト指標に分類化します。結果は75.9%の変数がリアリスト指標でした。よって仮説2aは反証されませんでした。

Hypothesis 2a predicted that variables produced and available in the field would tend to provide information on nation-states and on topics of inquiry that were labeled realist in the coding scheme. In order to test this hypothesis, the 1,650 variables were classified into two categories – realist indicators or nonrealist indicators. In order to be classified as a realist indicator, a variable had to have the nation as its actor and a realist topic; any mixed cases (nation as actor with a nonrealist topic or vice versa) were classified as nonrealist. If the hypothesis were true, it would be expected that a large proportion of the variables would be usable as realist indicators. The findings are presented in figure 5.1. The figure clearly shows that just about three-fourths (74.9 percent) of the variables can be used as realist indicators. This would hardly appear to be a random distribution, and the calculation of a binomial distribution (p<.01) for interpretive purposes#supports this assumption. On the basis of the finding, it can be concluded that hypothesis 2a has failed to be falsified. (pp.91-2)

仮説2bの妥当性もテストしましすが、結果は66.4%の変数が国力と国家間対立/協力を取り扱っていました。よって仮説2bは反証されませんでした。

Hypothesis 2b predicted that variables produced and available in the field would tend to provide more indicators of national power and inter-nation conflict-cooperation than of any other concept. Table 5.1 tests this hypothesis by rank-ordering the concepts. It can be seen from this table that the concepts of national power and inter-nation conflict-cooperation compose 66.4 percent of the total variables produced (49.9 percent and 16.5 percent respectively). In addition, none of the other concepts constitutes more than 9.1 percent of the data available. On the basis of these findings, it can be concluded that hypothesis 2b has failed to be falsified. These findings, provide considerable support for the proposition. (p.92)

よってVasquezは、1950年代から1960年代にかけての国際政治学におけるデータ作成はリアリスト・パラダイムに強く影響を受けたと結論づけます。ただし、COWプロジェクトで戦争だけでなく内戦のデータも集めているのはリアリスト・パラダイムにはそぐわないことも指摘しています(p.94)。

The findings of this chapter provide considerable evidence to support the proposition that data making in international relations was guided by the realist paradigm in the 1950s and 1960s. A review of the major data projects conducted during this time period shows that they have collected data primarily on nations and realist topics of inquiry. It has also been found that the initial use of these data has been to test realist hypotheses. (p.102)

第6章: パラダイム指向の活動しての研究

クーンの科学論での重要ポイントは、通常科学においてはパラダイムがパズルをうまく説明できず、パラダイムに基づく仮説がことごとく反証されても科学者がパラダイムを捨て去ることはないということだ。実証研究で使われるデータ自体がパラダイム依存的なので、実証テストに必要なデータはパラダイム内部でしか使用できず、客観的なデータを使って複数のパラダイムのどれが有用かテストすることはできません。クーンはたしかこれをincommensurable(通約不能な)という用語で説明したかと思います。

Kuhn maintains that the chief characteristics of normal science are that research is guided by the dominant paradigm and that research is seen as a puzzle-solving activity (Kuhn 1970a: ch. 4). In normal science, the scientist’s primary role is to develop hypotheses to explain puzzles that the paradigm has focused upon. One of the significant characteristics of this research, according to Kuhn (1970a: 146-148), is that the paradigm’s failure to resolve puzzles does not lead to the falsification of the entire paradigm, but to incremental changes known as paradigm articulation. Persistent failure to resolve puzzles is not seen as a flaw in the paradigm but as a flaw in the individual scientist (Kuhn 1970a: 35-36). Thus, while Karl Popper’s (1959) notion of falsification may be applied to individual hypotheses and even to theories, it is never applied to the most fundamental assumptions of the field, that is, the paradigm (Kuhn 1970a: 146-148). Hypothesis testing in normal science tends to be a process of testing competing hypotheses “derived” from the same paradigm rather than testing hypotheses derived from competing paradigms (Kuhn 1970a: 24). The latter, if it occurs at all in science, occurs during periods of scientific revolution and is then viewed as more of a change of world view than of testing hypotheses from competing paradigms. (Kuhn 1970a: ch. 10). The notion of a crucial experiment is only established with the aid of historical hindsight and is an indicator that the new paradigm has gained dominance in the field (Kuhn 1970a: ch. 11). (p.104)

本章では国際関係論における理論的仮説がリアリスト・パラダイムに準じているかどうか確かめます。

3a. 記述的研究における変数(アクターと探索トピック)がリアリスト的か。
3b. 相関的・説明的研究における独立変数がリアリスト的か。
3c. 相関的・説明的研究における従属変数がリアリスト的か。
3d. 相関的・説明的仮説における独立変数と従属変数がリアリスト的か。
3e. 国力が独立変数として最もよく使われているか。
3f. 国家間対立/協力が従属変数として最もよく使われているか。
3g. 最もよくテストされる命題では、国力が国家間対立/協力を説明・予測するものであるか。

Seven hypotheses can be derived from the proposition to test its adequacy:
3a Variables used in descriptive research will tend to have actors and topics of inquiry that are labeled realist in the coding scheme.
3b Independent variables used in correlational/explanatory research will tend to have actors and topics of inquiry that are labeled realist in the coding scheme.
3c Dependent variables used in correlational/explanatory research will tend to have actors and topics of inquiry that are labeled realist in the coding scheme.
3d Correlational/explanatory hypotheses tested will tend to relate independent and dependent variables whose actors and topics of inquiry are labeled realist in the coding scheme.
3e National power will tend to be the most frequently used independent variable.
3f Inter-nation conflict-cooperation will tend to be the most frequently used dependent variable.
3g The most frequently tested proposition will be the one that employs national power to predict or explain inter-nation conflict-cooperation. (p.106)

実証分析をすると、H3aについては377の変数のうち280 (74.3%)、H3bでは385の独立変数のうち262 (68.1%)、H3cでは233の独立変数のうち184 (78.9%)、H3dでは7,827の仮説のうち7,275 (92.9%)、H3eでは4,650 (59.4%)、H3fでは4,734 (60.5%)、H3gでは7,241の仮説のうち3.018 (41.7%)がリアリストの基本的仮定に反しないという結果になりました。

The findings of the nine tests conducted in this chapter provide considerable evidence to support the proposition. Employing the Jones and Singer (1972) abstracts, it was found that: (1) about three-fourths (74.3 percent) of the variables employed in descriptive research were realist; and (2) 68.1 percent of the independent variables and 78.9 percent of the dependent variables employed in correlational/explanatory research articles were realist. An examination of how these variables were combined to form hypotheses, using the second sample, revealed that the realist variables are used much more frequently than is suggested by the abstracts in Jones and Singer. It was found, for example, that 94.0 percent of the independent variable units and 94.2 percent of the dependent variables employed in actual hypotheses were realist. A review of how these independent and dependent variables were combined showed that 92.9 percent of the 7,827 hypotheses tested in the field were realist. (pp.119-120)

In addition to these tests, a number of predictions were made about the specific variables and propositions used in research. Employing the second sample, it was found that the chief independent variable of the realist paradigm, national power, was the most frequently employed independent variable in research (59.4 percent of all independent variables). It was also found that the chief dependent variable of the realist paradigm, inter-nation conflict-cooperation, was the most frequently employed dependent variable in research (60.5 percent of all dependent variables). Finally, it was found that the central proposition of the realist paradigm, relating national power to inter-nation conflict-cooperation, was the most frequently tested proposition in the field (41.7 percent of the 7,241 tested hypotheses). On the basis of these findings, it can be concluded that research in the field has been guided by the realist paradigm. (p.120)

ということで、1950年代以降、国際関係論ではリアリスト・パラダイムに基づいてデータが作られ、リアリストが関心を持つトピックに関する実証分析ばかり行われたとVasquezは結論付けます。

The findings of this chapter, when combined with the findings of the two preceding chapters, demonstrate that international relations inquiry has had an underlying coherence since the early 1950s. The realist paradigm has been used by scholars to focus on certain phenomena and develop concepts and propositions about them. This theory construction, or paradigm articulation, has directed scholars to collect data on realist indicators. It has been shown in this chapter that the data collected in the field have been used primarily to test realist hypotheses. The tests of the three propositions on theory construction, data making, and research in the field have all been supported. Therefore, the claim that the realist paradigm has dominated international relations inquiry in the 1950s and 1960s has been given credence. (p.120)

このあたりについての議論は本書の中で一番説得力が低いように思われます。1950年代から1960年代にかけて安全保障に関する分析が中心になったのは、リアリスト・パラダイムがあったからというよりも、当時は米ソ冷戦という非常に大きな問題を抱えていたからかと思います。安全保障優先だったからこそ当時、安全保障のことをHigh Politics、経済その他をLow Politicsと言っていました(今はあまりこの用語は使われていないと思います)。国家をアクターと見なす研究ばかりだったというのも、Vasquezが計量データしか分析していないからだと思います。質的データを用いた分析では国家を単一アクターとして扱わない研究もそれなりにあったであろうことが予想されます。第5章と第6章はあんまり読む価値がなさそうです。

12月 11, 2014 · Pukuro · No Comments
Posted in: ☆社会科学

Leave a Reply