TÜBINGER ARBEITSPAPIERE ZUR INTERNATIONALEN POLITIK UND FRIEDENSFORSCHUNG

Nr. 24

Gabriele Kittel, Volker Rittberger, Frank Schimmelfennig

Between Loyalty And Exit

Explaining the Foreign Policies of Industrialized Countries in the UNESCO Crisis (1978-87)

© G. Kittel, V. Rittberger, F. Schimmelfennig, Tübingen 1995

ISBN 3-927604-17-8

Redaktion: Andreas Hasenclever, Thomas Nielebock, Volker Rittberger, Frank Schimmelfennig

Table of Contents

1. Why Study Foreign Policy in the UNESCO Crisis?

2. Analytical Model

3. Conflict Behavior

4. Conflict Positions and Behavior: Interest Orientation vs. Policy Styles

5. Explanation of Conflict Positions in the NWICO debate

6. Summary and Interpretation of Findings

Notes

Bibliography

1. Why Study Foreign Policy in the UNESCO Crisis? [Note 1]

The UNESCO crisis was probably one of the most severe crises ever to have threatened the survival of an international organization within the United Nations system. It began in the mid-1970s, in the aftermath of decolonization and growing Third World self-assertiveness, and it developed from three major fields of conflict (Beigbeder 1987: 29-35; Imber 1989: 96-120):
(1) By promoting a New World Information and Communication Order (NWICO) a majority of developing countries sought to restructure the global flow of information. They denounced Northern control of the world media, the "one-way flow" of information to the South, and the negative bias in Western news coverage of the Third World. Instead of abiding by the liberal principle of "free flow", they demanded a "balanced flow", an improved access to communication technology, and a right to exercise a certain control over the information reaching their countries from abroad.
(2) Conflict also developed about the organizational efficacy and efficiency of the organization. Western states complained about excessive budget growth, inadequate financial control, an inflated and ineffective bureaucracy, and nepotism in the recruitment of staff members. Charges of mismanagement focused on the Director-General, Amadou Mahtar M'Bow.
(3) Finally, the positions on the admissible degree of general political activities were in conflict. Whereas Third World and Socialist countries used UNESCO as a forum for their anticolonial, antiracist and peace campaigns, Western countries criticized this as excessive politicization and insisted on the functionally limited nature of UNESCO's mandate.

Conflicts of this sort were not confined to UNESCO but developed in a great number of universal international organizations at that time. However, only in UNESCO did two major member states, the United States and Great Britain, withdraw from the organization in 1983 and 1984, respectively, thereby reducing its budgetary resources by 30% and endangering its viability. Meanwhile, this crisis has largely been overcome: In 1987, the Spaniard Federico Mayor followed M'Bow as Director-General and, more importantly, superpower competition concerning influence in the Third World no longer exists. On top of the world-wide wave of democratization (Huntington 1991) UNESCO, by and large, discarded NWICO as a live issue and has turned to an unequivocal defense of the freedom of the press (e.g. in its 1991 Windhoek Declaration). Under the Clinton administration the U.S. now even considers returning to the organization. [Note 2]

Whereas the UNESCO crisis has now passed into history, it still presents a very useful subject for comparative foreign policy analysis:
(1) The response of the industrialized states to the Third World claims in UNESCO varied widely. Whereas the United States and Great Britain eventually left the organization, the Soviet Union and other Socialist countries strongly supported the Third World majority. Between these poles, the Nordic states and France sought to partly accommodate the demands of the developing countries. Other Western states like the Federal Republic of Germany, the Netherlands or Switzerland strongly opposed any efforts to restrict the "free flow of information" but chose not to leave the organization. Why they behaved so differently is the puzzle which constituted the starting point of our research: All of these countries were industrialized and possessed a developed communications infrastructure as well as the capacity for active participation in the international media market. So why did they refrain from putting up a "united front" against the South? Except for the Eastern industrialized countries all of them were major liberal-democratic countries which share fundamental values. So why did they not collaborate against the attempt to curb the freedom of information? High variation in the dependent variable "foreign policy" in spite of some basic common characteristics of the countries investigated promises potentially interesting answers to the question how differences in foreign policy behavior can be explained.

(2) The UNESCO crisis concerns an issue area of international relations which has so far been widely neglected as a distinct field of research. Whereas a rich body of literature on the security and foreign economic policies of Western industrialized countries has emerged in the last decade [Note 3], the comparative analysis of foreign cultural policy broadly defined has remained a research desideratum. Political scientists studying international relations do well to take the battles for the "minds of men" mentioned in the Preamble to UNESCO's Constitution at least as seriously as those for the distribution of the world material product or for the control over polities and their territories.

The article starts out with the presentation of an analytical model for foreign policy analysis which integrates not only a great variety of causal factors proposed by different strands of International Relations research but also several competing causal pathways to the explanation of foreign policy: interest-oriented foreign policy behavior vs. foreign policy styles, structure vs. agency, and subsystemic vs. systemic causation. Furthermore, the model contains a dynamic element which allows to take into account the effects that international interaction outcomes have on foreign policy formulation. The comparative analysis of the UNESCO policies of four major industrialized countries, the United States, the Soviet Union, France and the Federal Republic of Germany, produces the following results:
(1) The UNESCO policy of these states can satisfactorily be explained as interest-oriented if one controls for the distorting effect that an unfolding crisis has on the behavior of states.
(2) The UNESCO policy of the four states can, to a very large extent, be attributed to structural factors operating at the domestic level (their media system) or at the international level (their position in the overall international power structure): However, in order to account for foreign policy change over time, a dispositional variable, the ideological orientation of the government, had to be added.
We conclude that there are two consistent and empirically tenable yet competing explanations, one related to the Liberal theory of international relations, the other to Realism. However, since "parties matter" in the explanation of UNESCO policy, Liberalism possesses a certain advantage.

2. Analytical Model

The field of foreign policy analysis is not only characterized by a great variety of proposed explanatory factors and hypotheses. It also abounds in highly different analytical approaches and causal pathways. In order to be able to take into account as many of them as possible and to test their relative explanatory strength our first task was to integrate them into one analytical model. However, we disregarded those explanatory variables and approaches that are not amenable to the formulation and testing of hypotheses, not suited for comparative analysis or do not use "foreign policy (behavior)" as their dependent variable. For example, we had to exclude most of what can be subsumed under the "decision-making approach" because it only offers a checklist of relevant factors in foreign policy decisionmaking or explanations of the policy process but no testable hypotheses about substantive foreign policy output .[Note 4] For lack of comparability idiosyncratic factors pertaining to specific characteristics of one country or the personality of a decision-maker could not be taken into account either.

The analytical model (Figure 1) starts from the fact that the industrialized countries had to react to a common challenge which was presented to them by UNESCO's Third World majority and consisted in its claims for a new international information and communication order, its control over the organization's resources and its attempt to "politicize" the organization in their interest. The model depicts various causal factors and pathways (symbolized by the arrows) which can be assumed to have produced this reaction ("conflict behavior"). Since the analysis covers a time period of roughly ten years during which the foreign policy of the industrialized countries can be assumed to have changed, the model also contains a dynamic component (indicated by the broken line at the bottom).

The different causal pathways reflect three general debates in International Relations theory.
(1) Interest-oriented foreign policy behavior vs. foreign policy styles : The juxtaposition of interests and styles has a methodological and a theoretical aspect. The methodological question is: Is foreign policy best analyzed as intentional action guided by interests? Or should we regard it as mere behavior without making assumptions about the motivation of the actors? The American "Comparative Foreign Policy Analysis" of the 1970s was oriented towards behavioralism because it viewed the empirical verification of actors' goals and interests as highly problematic (Brady 1982b: 138; Callahan 1982: 305; Hermann 1978: 32). This methodological orientation did not fail to invoke criticism.

"[I]n the attempt to study foreign policy scientifically, the notion that one of the purposes of the discipline was to explain the substantive foreign policy of states was lost." (Vasquez 1986: 207)

Carlsnaes asserted in the same year that foreign policy was purposive action, it should be defined with reference to substance alone and explained by the goals, interests and beliefs of the actors (1986: 26-38, 56). However, he discards too lightly the problem of an independent verification of goals and interests, which inevitably arises if an interest-based explanation is to be testable and falsifiable:

Figure 1: Analytical Model
Starting point: reaction to the challenge presented by UNESCO's Third World majority

"I submit [...] that the analysis of motives and intentions is essential to the causal explanation of foreign policy actions furthermore, that in principle these are observable when and if contained in policies ; but that we should not try to seek for motives or intentions - 'true' or not - when they are not posited or recognizable in policy actions." (1986: 62)

Carlsnaes rightly points out that we are not able to observe the actors' minds directly. However, the inference from the actors' behavior to their goals is likely to be tautological or indeterminate - the same behavior can result from different goals (cf. Brady 1982b: 139; Hermann 1978: 31). We therefore need independent evidence of the substantial interests of the actors, on the one hand, and their concomitant behavior, on the other, in order to ascertain whether this behavior really is interest-oriented.

This is also the only way to decide between the theoretical alternatives of foreign policy as interest-oriented behavior and foreign policy styles. According to the concept of interest- oriented behavior foreign policy is understood as the rational or, at least, intentional pursuit of state or "national" interests.[Note 5] Usually, rational-actor models are either not concerned with the nature of the goals pursued and regard them as exogenously given variables (e.g. in game theory). Or they consider states to be monistic actors, i.e. they posit an invariable "national interest" like the "eternal quest" for power and independence.[Note 6] By contrast, we assume that foreign policy interests have to be empirically determined and theoretically explained. If one assumes interest-oriented behavior, it is sufficient to account for the interests of a state expressed in its conflict position in a given issue area (A arrow), for it is assumed that behavior is but a direct and undistorted consequence of interest. The implicit correlation between conflict position and conflict behavior (B arrow) here is: "The greater the conflict intensity in an issue area (i.e. the distance of one actor's position from that of another), the more uncooperative the behavior toward this other actor."

By contrast, the concept of foreign policy as a behavioral style refers to a state's tendency to prefer certain foreign policy instruments and strategies over others across issue areas. It is seen as at least partially independent of the goals at stake: States may well have identical interests and yet pursue them by very different means. Policy styles can, for instance, result from instutionalized behavioral routines or bureaucratic politics in the state apparatus.[Note 7] According to this approach, a state's foreign policy can, e.g., be described as generally "aggressive", "risk-averse", "multilateral", or "low profile" (cf. Brady 1982a: 28). We do not observe foreign policy styles directly but assume that they intervene in the relationship between conflict position and conflict behavior (C arrow) systematically distorting the outcome which would have been expected as a result of interest-oriented behavior. A cooperative style distorts the expected behavior in the cooperative direction; a uncooperative style makes the behavior more uncooperative than it would have been expected according to the hypothesis of interest-oriented conflict behavior. Consequently, foreign policy styles come into play when the assumption of interest-oriented behavior fails, and they can only be discovered if the variable "conflict position" is controlled for.

(2) Structure vs. agency : The agency-structure debate takes up the old meta-theoretical and theoretical issue of structuralism vs. individualism in the social sciences.[Note 8] Whereas structuralism accords social structures ontological priority over actors and actions and demands that all social phenomena be attributed to structural causes, individualism claims that social systems and their structures are reducible to human agency and demands that all explanations in the social sciences be rooted in the dispositions and behaviors of individuals. However, as a result of this meta-theoretical debate, it meanwhile seems to be "widely accepted that any social theory must be able to generate explanations that take account of both structure and agency" (Buzan/ Little/Jones 1993: 104).

We take this widespread consensus to mean that structure and agency should be equally represented in any analytical model and must be indispensible components of any causal pathway. We therefore take into account structural and dispositional factors for both "interest" and "style" explanations. Furthermore, we include intermediating situational factors which are at work only in the specific conflict interaction and possess neither structural nor dispositional quality. They are less durable than structural factors but, in contrast to dispostional factors, of an objective nature. The costs of a foreign policy decision and the media coverage of a foreign policy issue are examples of situational influences on foreign policy. As indicated by the arrows in Figure 1, it is assumed that structures create a more or less extended scope for foreign policy action. This scope is subsequently narrowed down by situational characteristics. Finally, the state actors choose their conflict positions and conflict behaviors according to their dispositions within in the limits of the remaining scope of action.

However, a consideration of structure and agency which puts them on an equal heuristic footing does not exclude a theoretical imbalance. Within a structurationist ontology it is still possible to formulate either highly deterministic explanations which claim a predominant influence of social structures on human behavior or highly voluntaristic explanations which presume a large degree of structural indeterminacy and freedom of action. The aim of establishing the relative explanatory weight of structural vs. dispositional causal factors remains unchanged (Archer 1985: 62).

(3) Second image vs. third image : The "level-of-analysis" problem has traditionally received high attention in International Relations research (cf. Singer 1969). One distinguishes a systemic level, at which international interactions and structures are located, and a sub- systemic or unit level, where foreign policy actions and domestic structures can be found. This distinction also helps to clarify the difference between the agency-structure and the level-of- analysis problems. Whereas dispostional factors are of an exclusively sub-systemic nature, structural and situational conditions of foreign policy, can be located both at the systemic and at the sub-systemic level.

Like the two other debates, the level-of-analysis problem has a methodological and a theoretical aspect. In meta-theory, it concerns the fallacies of reductionism, i.e. of explaining system-level phenomena by unit-level variables, and of holism, i.e. of attributing unit-level phenomena to system-level characteristics. For a valid explanation, all variables have to be specified at the same level which sometimes demands a transformation of the independent variables. Our dependent variable "conflict behavior", as "foreign policy" in general, is situated at the unit level of analysis so that explanations are liable to the fallacy of holism. In order to avoid this problem, systemic variables have to be reconceptualized as or transformed into positional variables. For instance, whereas it is logically inadmissible to explain the foreign policy of an individual country by the power structure of the international system, it is possible, in principle, to attribute its foreign policy to the country's position in this structure: in this case both variables describe characteristics of the units.

If this heuristic rule is observed, it is up to theory and empirical investigation whether policy choices are mainly influenced by systemic or sub-systemic structural and situational conditions (dispositional factors are, by definition, always sub-systemic). For these different causal pathways, Waltz (1965) has coined the terms "second image" and "third image". "Second image" approaches, which are typical for the Liberal school of International Relations theory, focus on unit-level factors, i.e. the characteristics of states or their societes in explaining foreign policy behavior. "Third image" approaches are prominent in Realism. They assume foreign policy behavior to be mainly determined by the states' position in relation to each other and in the (power) structure of the international system.

These causal pathways are also useful for classifying individual variables proposed in the literature as conditions and causes of foreign policy (cf. Fig. 1). We drew them from various strands of IR-theory, foreign policy analysis and peace research and adapted them to our specific field of investigation, the UNESCO crisis.[Note 9] The variables listed in Figure 1 only serve as illustrations at this point. Their theoretical background and the corresponding hypotheses will be described in section 5.

Lastly, the analytical model is designed to take into account the foreign policy change which has occurred within the period of investigation from 1978-87. Change can be generated in two different ways. Exogenously it can be produced by a change in one or several of the presumed independent variables. However, it can also come about endogenously through the feedback effects of the outcome resulting from the interaction of foreign policies. The interaction outcome at a certain point in time possibly affects the dispositions of the actors, the situation and, in extreme cases, even the structural setting and thereby may alter the foreign policy at a later point in time.

In contrast to the quantitative, data-based analysis predominant in the American "Compa rative Foreign Policy" movement of the 1970s [Note 10], we follow the more recent scholarly work on the foreign economic policy and the security policy of industrialized states in using a qualitative case study design. It is based on the method of "structured, foused comparison" (George 1979) which seeks to achieve the necessary control by an intentional selection of cases. They have to fulfill three conditions: (1) a high variation in the dependent variable, (2) a high variation in the test variable(s), and (3) a minimal variation in the other independent variables. [Note 11]

The first condition is fulfilled: Starting from the observation of highly divergent foreign policies during the UNESCO crisis we selected our cases on the dependent variable "conflict behavior". In order to maximize variation we chose the - by traditional standards - most important country from each of the different behaviorial groups: the Soviet Union as the major representative of Third World supporters, France for the accommodationists, the Federal Republic of Germany as one of the opponents of Third World demands which did not leave the organisation, and the United States as the most important country which withdrew from UNESCO. This selection also contains three liberal-democratic countries which, however, differ with respect to their domestic structures (conditions 2 and 3). The third condition is met insofar as all selected countries are industrialzed and possess a highly developed media infrastructure. Moreover, none of the selected countries is a "small country".

We measured the conflict behavior of these four countries and the values of the independent variables at the time of each of the five UNESCO General Conferences within our period of investigation: 1978, 1980, 1983, 1985 und 1987, thus obtaining 20 data points or cases. From a methodological standpoint, our study should not be regarded as producing strong causal inferences or explanations which are generalizeable for the whole universe of cases. For that purpose, the number of our explanatory variables and causal pathways is too high. The cases are too few, and they cannot be assumed to be sufficiently independent of each other (in particular the five data points for each country are probably highly interrelated). Finally, our method of selection is liable to bias, although we attached great importance to a high variation of the dependent variable (King/Keohane/Verba 1994: 129-137, 141). All we can hope to achieve is a tenable explanation for the variation in the selected and investigated cases and a general clue as to which hypotheses merit further and more rigorous testing.

3. Conflict Behavior

The dependent variable "conflict behavior" was determined according to a scale of coo peration based on Hirschman's (1970) typology of exit, voice and loyalty which he had originally developed to describe different forms of reaction to dissatisfaction with the performance of an enterprise or an organization. Following Keohane, we consider conflict behavior to be the more cooperative the more it makes it possible for a party involved in a conflict to realize its objectives (1984: 51). "Loyalty" to the organization, i.e. the adaptation of one's policy to the behavior decided and demanded by the organization's "legislative" bodies can be regarded as the most cooperative, "exit" from the organziation as the most uncooperative behavior. The in-between option of "voice", i.e. the open articulation of dissatisfaction and opposition within the organization, was further differentiated into a "constructive", a "conditional", and an "confrontational" variant. We speak of
(1) constructive behavior if the state is ready to negotiate, to compromise and to make prior concessions unconditionally and independently of the other party's behavior;
conditional behavior if the state makes its readiness to negotiate or to make concessions dependent on the behavior of the other party (e.g. through linkages, reciprocal bargaining, a tit-for-tat strategy);
(3) confrontational behavior if the state is not ready to negotiate or to compromise and, instead, ignores or rejects the other party's offers.

From 1978 through 1985, the Soviet behavior in the UNESCO can be characterized as basically " loyal " toward the organization and its Third World majority. Generally, the Soviet Union subordinated its special interests to the creation of a "common front" with the South against the North and tried to prevent compromises and cooperation between West and South. Relations between the Soviet Union and the majority of states in UNESCO were by no means completely harmonious. However, even though the Soviet delegates regularly criticized the recruitment policies of UNESCO and its budget growth and complained that the "cause of peace" and the "fight against the danger of nuclear war" were not taken into due account in the UNESCO decisions about budget allocations and program activities, the Soviet Union never voted against resolutions sponsored by a Third World majority or openly dissociated itself from the South in any other way.

In order to achieve majoritarian support for its UNESCO policies, the Soviet Union em ployed adaptive or, at the most, "constructive" behavior. The usual procedure of Soviet policy was to first introduce resolutions which reflected the narrow interests of the Socialist countries but usually did not convince a majority of states in UNESCO. Therefore, the Soviet delegates agreed to add Third World concerns to their resolutions and accepted compromise formulas until enough support was secured. Although they publicly criticized these compromises, they finally voted in favor of them.

By trying to preserve a "common East-South front" the Soviet Union also tried to prevent an understanding between the Western industrialized countries and the developing countries. Soviet delegates did not tire of warning the Third World not to succumb to attractive Western technological offers and not to lose sight of the anti-imperialist thrust of the NWICO campaign. Western proposals for the reform of UNESCO or for the preservation of the existing international information and communication order were rejected, American and British threats to leave, and their decision to leave, were condemned as "blackmail" to which UNESCO ought not to give in (Uranov/Kanaev 1986: 56).

By 1987, Soviet conflict behavior had noticeably changed. The behavioral pattern "loyalty towards the Southern majority, confrontation with the West" was replaced by "constructive behavior" , oriented towards finding compromises agreeable for all UNESCO members and towards leaving aside the most controversial ideological and political issues. As a result, the Soviet Union dissociated itself from its own past practices and from those Southern countries which insisted on a confrontational course.

"We do not regard UNESCO as a field of propagandistic and, even worse, confrontational exercises. Less slogans, more practical work." [Note 12]

The most obvious indication of the new Soviet behavior was the fact that the Soviet delega tion at the 1987 General Conference in Paris supported the Western candidate for the post of UNESCO's Director General, Federico Mayor, in the third ballot and thus tilted the balance against M'Bow.

French behavior in UNESCO can be characterized as " voice " according to Hirschman. The French readiness to find negotiated solutions and France's efforts to prevent the organization from breaking-up showed that this dissatisfaction was expressed in a "constructive" way. France attempted to bridge the gap between Third World and Western countries. Playing a mediator´s role, France hoped to be able to avoid the risk of alienating friends by taking sides. Therefore, France never voted against its Western allies but abstained several times when its position differed from the one of the other Western partners. Wherever possible, France sought to initiate compromise solutions either by trying to establish a study group or by searching for a compromise formula in the Drafting and Negotiating Group where it usually played an eminent role. Another example for this "constructive" foreign policy is the French support for the International Program for the Development of Communication (IPDC). Already in 1982 France contributed an amount of FF 500.000 to the special account.[Note 13]

Moreover, France ostentatively supported UNESCO and its Senegalese Director-General by regularly voting in favor of the draft budget and by defending M´Bow against attacks. In 1983 France supported the compromise budget proposal which had been introduced by the Nordic States. This draft budget intended a lower increase of the budget than the one proposed by M´Bow. But there was always a limit beyond which France did not yield. France never succumbed to a condemnation of Israel or to an abandoning of basic Western values in texts on information and communication issues.

When the USA announced its withdrawal, French reactions were still mixed. The British withdrawal, however, and the prospect of other major contributors turning their backs on UNESCO led to a hardening of French behavior towards M´Bow and his management style. Together with Britain, France initiated a 12-member Temporary Committee in the Executive Board which produced more than 100 reform proposals. Therefore, beginning with 1985 the French "voice" consisted of both " constructive " and " conditional " elements. France could no longer resist the pressure exerted by the other Western states to pursue a policy of financial restrictions and to behave more uncooperatively towards UNESCO in general. The French government began to emphasize the necessity of thorough reforms in order to assure the survival of the organization and to make Director-General M´Bow responsible for their success or failure.[Note 14]

German foreign policy behavior was never as cooperative as the French behavior. Whereas the Federal Republic of Germany´s voice contained " constructive " as well as " conditional " features during the initial stages of the UNESCO crisis, it turned predominantly " conditional " after the height of the crisis marked by the announcement of U.S. withdrawal. German behavior was characterized by the attempt to find a solution for UNESCO´s problems without yielding to the pressure to withdraw which was exerted by the CSU, the most conservative faction of the West German governing coalition, and by the U.S. government. Unlike France with its close relations to former colonies and its special status as host country of UNESCO, Germany always sought to act in a framework of strong European concertation.

In close cooperation with France, the Federal Republic was searching for acceptable com promises in the media debate. This "constructive" element in German behavior was, for example, essential for agreeing on the compromise formula "free flow and a wider and better dissemination of information" (cf. Petzsch 1987: 22). However, West Germany opposed all potential decisions of UNESCO which would have contradicted the Western value of freedom of information. The "conditional" element in Germany´s behavior is revealed by its attitude towards the IPDC: The Federal Republic linked its comparatively generous support to the program's budget with the expectation to deradicalize Third World demands.

Mismanagement in UNESCO was heavily criticized. In 1980 and 1983 the Federal Republic of Germany abstained in the vote on the budget. But only in 1984 did it temporarily withhold financial contributions in order to step up reform pressure. For the same reason, the budget committee of the German Parliament freezed DM 8 million of the German contribution for 1985 which should have been paid to compensate for U.S. withdrawal. At the same time, the Federal Republic stopped its "constructive" policy of toleration with regard to General- Director M'Bow. In a move towards "conditional" behavior, it actively sought an alternative candidate acceptable to the Third World arguing that it might not to be able to resist the pres sure to withdraw from UNESCO if M'Bow was re-elected in 1987. Concerning the issue of "politicization" German foreign policy behavior is to be classified as "conditional" for the whole period of the UNESCO crisis. While constantly voting against general resolutions on the Palestinian question or on disarmament, Germany was ready to propose compromise texts which limited these issues to questions of education, science and culture covered by the UNESCO mandate.

Among the four countries investigated here the United States behaved most uncooperatively throughout the UNESCO crisis. From the beginning the United States "voiced" strong reservations concerning the Third World claims in UNESCO. However, during the first years of the crisis United States behavior can still be classified as " conditional-confrontational ". The "conditional" strategy was most visible in the issue area of information and communication policy. The U.S. voted against any definite restriction of the free flow but went along with the consensus found in the 1978 Media Declaration. The American objective was to split the fragile coalition of the Third World countries. Therefore, the U.S. pursued a policy of selective aid to, and collaboration with, non-Socialist members of the NWICO coalition (cf. Schiller/Herman/Preston 1989: 298). It hoped that those developing countries which the United States helped building up national information and communication infrastructures would in return accommodate the American ideas of an international information and communication order without state control. In connection with the adopting of the Media Declaration the United States promised, for example, "[...] to initiate, through its international exchange programs, a plan of cooperation and assistance which would help develop and improve regional training centers in mass communication, broadcasting, and journalism in the developing world".[Note 15] The 1978 General Conference unanimously adopted the American proposal that UNESCO convene a conference "to develop a proposal for institutional arrangements to systematize collaborative consultation on communication development activities, needs and plans" [Note 16] which resulted in the establishment of the (IPDC).

Concerning organizational issues and the debate about "politicization" American behavior was already "confrontational" before the decision to withdraw. In neither case was the United States ready to seek or support compromise formulas. Confronted with draft resolutions condemning Israel or demanding disarmament, the United States harshly denounced them as anti-Western propaganda and voted against them.[Note 17] It also regularly voted against the budget and was the only country to refuse the compromise found when the Nordic states proposed a budget ceiling of US-$ 374 million for the biennium 1983/84.

Table 1: Conflict Behavior During the UNESCO Crisis

Obviously, neither the consensus in the media debate nor the reduced budget ceiling or the promise of reforms prevented the U.S. from announcing its December 1983 decision to withdraw from UNESCO and thus to choose the " exit " option. In the last year of American membership U.S. policy was geared to leaving UNESCO. The U.S. administration continued to behave "confrontationally" by constantly refusing to negotiate and to compromise. The U.S. General Accounting Office, which was granted access to UNESCO's internal files in 1984 after heavy political pressure from the U.S. Congress, produced a report which served as a factual basis of U.S. demands for a thorough restructuring of the UNESCO management. However, when UNESCO reacted with detailed reform plans they were rejected by the State Department as half-hearted and not far-reaching enough. The United States withdrawal from UNESCO took effect on 31 December 1984.

Table 1 gives a summary overview of the conflict behavior for the four selected countries. It shows the described high variation in behavior reaching from a mostly "loyal" Soviet policy to U.S. "exit". French behavior was predominantly "constructive", German policy mainly "conditional". It can be seen as well that the behavior of all countries changed during our period of investigation and at different points in time became more uncooperative. How did this behavioral variation and its change over time come about?

4. Conflict Positions and Behavior: Interest Orientation vs. Policy Styles

According to our analytical model, the first and most fundamental issue to be resolved is whether the observed differences in behavior can be attributed to differences in interests or are best explained as the effects of varying policy styles. Therefore, we will now have to look at the conflict positions of the four countries for the same 20 data points at which we observed their behavior. In order to establish a possible causal effect of conflict positions on conflict behavior and to avoid giving mere rationalizations of behavior the status of motivating interests we had to be particularly careful assessing the interests of the governments independently of, and prior to, their concomitant actions. We based our data mainly on internal position papers of the governments which were written ahead of the General Conferences of UNESCO and on interviews with decision-makers and experts. The values for the conflict positions were determined according to a matrix which lists the major objects of contention in the UNESCO crisis together with the position of the Third World majority on them (0) and deviating positions in three degrees of conflict intensity. This matrix is reproduced as Table 2.

Table 2: Issues and Conflict Positions in the UNESCO Crisis

Concerning most of the issues debated during the UNESCO crisis, the Soviet positions were in harmony with Third World demands (median of 0). The Soviet Union regarded the NWICO discussion as an opportunity to further the restructuring of the international information and communication order along lines it had developed long before the 1970s. The Soviet position was based on the principle of national sovereignty and state control, and it was opposed to any form of market regulation in the field of information. "Free flow" was critized as an "imperialist" doctrine designed to promote the ideological and commercial expansion of the Western monopolies, to interfere with the domestic affairs of foreign countries, and to disseminate militarist propaganda.[Note 18] Instead, the Soviet Union demanded the "prior consent" of every state to the influx of information across its borders and the responsibility of every state for the information that was disseminated from its territory and by its citizens. Information should only be distributed if it was "progressive", i.e. it served "disarmament and peace", the "fight against militarism and apartheid" and other goals of which the Soviet Union considered itself to be a champion. This anti-liberal project of an international information and communication order had not yet been given up by 1987 although it was proposed in a more moderate tone.

With regard to the organizational issues there was a weak conflict between the Soviet Union and the Third World, but then the Soviet Union attached little importance to them in comparison with "politicization" and "NWICO". The Soviet Union opposed the high rates of budget growth and felt underrepresented in the UNESCO staff. It demanded an improved management but was in favor of centralization as well as long-term planning. The Soviet Union supported M'Bow against Western criticism and did not openly dissociate itself from him even before the 1987 election. Soviet representatives were permanently demanding that UNESCO deal with the "key problems of the present" and rejected Western accusations of "politicization". However, when it came to determining what those "key problems" were, the USSR was not always in agreement with the Third World. Whereas the condemnation of Israel was fully supported, the Soviet Union wanted the issue of "peace and disarmament" to be given priority over the complex of "development, decolonization, racism, Apartheid" in UNESCO programming.

The overall degree of conflict intensity between France and the Third World majority can be characterized as medium (median of 2) in 1978, 1980 and again in 1987 but was only weak (median of 1) in 1983 and 1985. As to the French position toward the international information and communication order, we have to distinguish two aspects. From the ideological perspective, the French position was liberal insofar as French political actors demanded that UNESCO must adhere to its constitutional values. Any attempt to slip a legitimation of censorship into UNESCO´s documents, for instance, provoked fierce French opposition. At the same time, however, there was a strong undercurrent in French policy opposing an alleged American cultural hegemony. This tendency intensified after 1980 and was most prominently expressed by Jack Lang, the new Socialist minister of culture. The emphasis on cultural identity was paralleled by the protection of the domestic and foreign markets in favor of the French-based communication industry. Especially in its former African colonies France tried to defend its economic and political interests. At the same time, however, it - at least rhetorically - supported the claims of the developing nations to create their own communications infrastructure.

In the issue-area of "organizational efficacy and efficiency" the French position was until 1985 characterized by only weak differences of position. France accepted moderate budgetary growth rates and did not oppose UNESCO´s centralized management techniques which were - at least partly - inspired by French traditions and were elaborated by M´Bow's predecessor, the Frenchman René Maheu. France favored a slightly better representation of developing countries in the Secretariat but cautiously avoided picking out that issue as a central theme because of its own overrepresentation in the Paris-based organization. Only when the French government realized that "business as usual" could eventually lead to a break-up of UNESCO did it demand a thorough restructuring of the management, dissociate itself from M'Bow, and support the nomination of a Third World moderate. At the beginning France generally opposed discussing anti-Israeli or disarmament draft resolutions out of fear that "politicization" might paralyze the organization. Moreover, the French were careful not to irritate their Arab friends or Israel and the West by being forced to cast a vote on controversial resolutions. After 1980 the categorical opposition to political debates decreased: The new French ambassador to UNESCO, Gisèle Halimi, regarded the fight against racism as a constitutional task of the organization.

Between Germany and the majority of Third World countries in UNESCO the overall degree of conflict intensity was medium (median of 2). German politicians readily acknowledged the asymmetry of the international news flow and the necessity to assist developing countries in establishing their own media structures. They accepted a formula describing "NWICO" as an evolutionary process. A substantial number of communication projects in Third World countries was funded by the Federal Ministry of Economic Cooperation. Germany´s stance in respect to communication contents can be characterized as rigidly liberal. It was not prepared to accept any restriction of the freedom of the press.

Less importance but still high priority was assigned to the organizational problems of UNESCO. As a major financial contributor Germany demanded a reduced growth and from 1983 onwards even a freeze of the organization's budget. As the crisis escalated with the American announcement to withdraw, the Federal Republic insisted upon thorough organiza tional reforms. M´Bow´s obvious favoritism towards personnel of African origin did not pose a substantial problem for Germany although it often criticized the fact that the number of Germans working in the Secretariat did not match Germany´s role as a major financial contributor. The question of M´Bow´s succession was not openly discussed until 1986 when Germany decided to support the candidacy of a Third World moderate. "Politicization" did not play an exceedingly important role for the German government. In contrast to Anglo-Saxon positions, Germany never denied UNESCO the right to discuss political topics -- provided, however, they fell within the scope of the organization´s mandate.

The United States positions in the UNESCO crisis and the interests of the Third World majority were in conflict on all issues. During the height of the UNESCO crisis the United States showed the highest degree of conflict intensity among the four countries investigated here (median of 3). Even in the years before, the differences of positions are to be classified as medium (median of 2). The American position toward the international information and communication order can be characterized as strictly liberal. The United States opposed any attempts to restrict the free flow or impose state control on the media. The Carter ad ministration acknowledged at least some concerns of the Third World about the imbalance of the news flow and supported technical help to the developing countries for the establishment of a local communications infrastructure. In the Reagan years, however, it was no longer willing to make such concessions.

As far as "organizational efficacy and efficiency" is concerned, the USA adopted a zero- growth position on the budget-issue combined with the demand for some reform measures in the area of management and programming. In 1983 a hardening of the American conflict position became visible in this area, too. The United States began to call for a thorough reform of the management. Confronted with the Third World claims for a better representation of their nationals in the Secretariat, the Americans reacted by criticizing the fact that many staff members from the Third World had been employed by M´Bow for purely political reasons. This can be interpreted as a position slightly favoring the major contributors because their reservoir of skilled personnel is larger. The Carter administration still supported the Director- General M´Bow during the first years of the crisis. Although the USA did not openly push for his resignation, it is clear that it would have preferred another Director-General at the height of the crisis. "Politicization" was a major issue for the USA. While it accepted the UNESCO issue-specific discussion of topics like Palestine, Apartheid and disarmament during the first years of the crisis, even a discussion of issue-specific topics such as the educational situation in the occupied territories was rejected after 1980 as serving the propagandistic aims of the Soviet bloc and its radical Third World allies.

Table 3: The Conflict Positions of the Soviet Union, France, West Germany, and the U.S. during the UNESCO crisis

The values for each country are summarized in Table 3. For the purpose of comparison it is sufficient to look at the figures set in bold that represent the median conflict intensity for each country, first for the NWICO debate only, and at the bottom for all issues. As in the case of conflict behavior we see that the Soviet Union and the United States take the extreme positions with almost no Soviet-Third World conflict and a strong U.S. conflict with the majority in UNESCO. France and West Germany mainly come under the medium category of conflict intensity with French positions being more moderate on the whole. The mostly parallel distribution of values for conflict behavior and conflict positions creates the impression that the hypothesis of interest-oriented behavior will most probably be confirmed: "The greater the intensity of conflict between an industrialized country and the Third World majority (in UNESCO), the more uncooperative is its behavior."

However, in order to obtain more than impressionistic evidence we subjected our data to statistical analysis. As a tool we chose Kendall's tau (b), a measure of ordinal association based on the comparison of each possible pair of values - a choice determined by the fact that both our data and our hypotheses are based on ordinal relations.[Note 20] Furthermore, Kendall's tau (b) is easier to interpret than the other versions of Kendall's tau . In contrast to Kendall's tau (a) and Goodmann/Kruskall's gamma it also takes into account tied pairs and thus calculates the coefficient on the basis of all pairs. Finally, it is a rather prudent measure of association the values of which are usually smaller than the corresponding values of Goodmann/Kruskall's gamma .[Note 21] Tau (b) varies between -1 and 1. A value of -1 indicates a perfect negative, a value of 1 a perfect positive correlation; a value of 0 indicates no correlation at all.

Since our hypotheses concern partly all issues of contention and partly only the conflict about NWICO, we associated first the median of all conflict positions of one country for each year of investigation with its concomitant behavior and then only the median of its positions on the NWICO debate. The results are a tau (b) of .80 for all issues and a tau (b) of .84 for the NWICO debate only. The unexplained variance can mainly be attributed to "ties", i.e. to the fact that some of the compared value pairs are connected in either the independent or the dependent variable. Only in the case of France do we find discordant pairs which openly contradict the hypothesis of interest-oriented behavior: Whereas in 1978 and 1980 the conflict intensity between France and the UNESCO majority was higher than in 1985 and 1987, its conflict behavior was more cooperative at the beginning of the crisis. Although the figures are mean ingful enough to justify the conclusion that the different UNESCO policies of the four countries mainly reflected a difference in their substantive interests concerning the issues of the UNESCO crisis, it would be desirable to have an explanation for at least some of the residual variance.

The obvious thing to do would be to turn policy styles. One could assume that different policy styles of the selected countries are responsible for the deviation of the foreign policy behavior from the ideal line suggested by the foreign policy goals and interests. According to our defintion of policy styles - different conflict behavior despite equal or similar differences of interests - one would expect to find among the residuals a particularly high number of pairs which are tied in the independent variable. This is the case for 43 out of 54 residuals with regard to all conflict issues, and with 24 out of 39 residuals with regard to the NWICO issues (cf. Table 4).

However, the table also shows that this kind of tie not only appears in the paired comparison of values for different countries but also for the same country at different points in time. For the conflict about the NWICO these intra-country ties even exceed the number of inter- country ties, and there is no country without intra-country ties (see the left column of paragraphs 1.2 and 2.2 in Table 4). This means that the residual behavior cannot result from a durable, country-specific behaviorial style if that was the case, the same conflict position of one country would always be followed by the same behavior. These findings discourage us from viewing the residuals as indications of policy styles.

The data itself suggest a different explanation. The intra-country residual pairs which are tied in the independent variable follow a clear pattern: All of them are in chronological order. Whenever a state changed its behavior but not its conflict position, the later behavior was less cooperative than the earlier behavior.[Note 22] In our test of ordinal association we have so far disregarded the dynamics of foreign policy reflected in our analytical model, i.e. we did not analyze our data points in chronological order. That may have been an oversimplification. To improve on this deficit, we assume that the UNESCO crisis falls into distinct time periods created by the dynamics of an unfolding crisis and separated by important interaction outcomes. One such temporal break immediately comes to mind: the withdrawal of the United States and Great Britain marking the height of the crisis in 1983/84.

Table 4: Residuals of the Correlation Between Conflict Intensity and Conflict Behavior

1. All Issues
1.1. Discordant Pairs: F78-F85, F80-F85
1.2. Tied in Conflict Intensity (conflict behavior is less cooperative for second value of each pair)
SU78-SU87, SU80-SU87, SU83-SU87       F78-D78, F78-D80, F78-D83
SU85-SU87                             F78-D85, F78-D87, F80-D78
F83-F85, F78-F87, F80-F87             F80-D80, F80-D83, F80-D85
D78-D85, D78-D87, D80-D85             F80-D87, F87-D85, F87-D87
D80-D87, D83-D85, D83-D87             F78-US78, F78-US80, F80-US78
US83-US85, US83-US87                  F80-US80, F87-US78, F87-US80
                                      D78-US78, D78-US80, D80-US78
                                      D80-US80, D83-US78, D83-US80
                                      D85-US78, D85-US80, D87-US78
                                      D87-US80
1.3. Tied in Conflict Behavior (conflict intensity is higher for second value of each pair) 
SU87-F83, SU87-F78, SU87-F80
F83-F78, F83-F80, F85-F87
F85-D78, F85-D80, F85-D83

2. NWICO issues
2.1. Discordant Pairs: F78-F85, F78-F87, F80-F85, F80-F87
2.2. Tied in Conflict Intensity
SU78-SU87, SU80-SU87, SU83-SU87       F78-D78, F78-D80, F78-D83
SU85-SU87                             F78-D85, F78-D87, F80-D78
F83-F85, F83-F87                      F80-D80, F80-D83, F80-D85
D78-D85, D78-D87, D80-D85             F80-D87
D80-D87, D83-D85, D83-D87
US83-US85, US83-US87
2.3. Tied in Conflict Behavior
SU87-F83, SU87-F78, SU87-F80
F83-F78, F83-F80,
F85-D78, F85-D80, F85-D83, F87-D78, F87-D80, F87-D83

It is a plausible hypothesis that the withdrawal of these two important states moved the industrialized countries to push more firmly for reforms and to risk a stronger confrontation with the Third World majority in order to prevent a further disintegration of the organization and, eventually, to induce the United States and Great Britain to return. Accordingly, it can be expected that the behavior of the selected countries will be more cooperative in the first than in the second time period.

For the test of this hypothesis we subdivided our data into two samples, one for the 1978-83 time period, the other for the period encompassing 1985 and 1987.
(1) In order to establish whether these two samples really represent distinct behavioral patterns, we first calculated the ratio of conflict positions to conflict behavior for each. In the case of the totality of issue-areas the median of all conflict positions is 2 in the first as well as in the second period. The median behavior, however, changed from "constructive"- "conditional" to "conditional"-"confrontational" at the height of the crisis indicating that other elements than just conflict positions impacted on foreign policy behavior. In the case of the NWICO debate, the positions were even slightly less conflictual after 1983 (between 2 and 2- as compared to 2- before) whereas the behavior became more uncooperative. We can thus conclude that the sample subdivision was justified because the aggregate foreign policy behavior of the four countries has indeed turned generally more uncooperative at the height of the crisis, despite an aggregate constancy of conflict positions.

(2) In order to find out whether the temporal subdivision improves the ordinal association of conflict positions and conflict behaviors, we then calculated the tau (b) for each sample individually. We obtained .81 for the first period and .91 for the second regarding all issue- areas. This is only a slight improvement over the previous .80 value for the years of 1985 and 1987. However, for the NWICO debate, the tau (b) increased from .84 to .93 and .98 respectively.

Thus, the relationship between conflict positions and conflict behavior can be explained almost entirely if one assumes that
(1) the industrialized countries made their behavior conditional upon the intensity of their conflict with the Third World majority in the UNESCO;
(2) the conflict about the New World Information and Communication Order exerted a particularly high influence on the choice of behavior;
(3) in addition, the worsening of the UNESCO crisis made the behavior of the industrialized countries less cooperative.
It is therefore a tenable proposition that the foreign policy of the four industrialized countries toward and within UNESCO was mainly determined by their interests concerning the international information and communication order but was not unaffected by the unfolding of the UNESCO crisis itself. As a result of the American and British withdrawal from the organization, which marked the height of the crisis, the UNESCO policies of the remaining three tended to become comparatively more uncooperative. With a certain time lag this even applies to the Soviet Union: Dissatisfied with the results of its policy of confrontation with the West, it chose a more accommodating behavior in 1987 although its conflict positions had not changed. This is a sound basis for disregarding the potential existence of different foreign policy styles and for proceeding with the explanation of foreign policy behavior by exclusively testing those interest-oriented hypotheses which attempt to account for the position of a country in the NWICO debate.

5. Explanation of Conflict Positions in the NWICO debate

In our test of interest-explaining hypotheses we begin with structural factors because they are assumed to set the most general and fundamental conditions for foreign policy. Only where they lack explanatory power or do not discriminate enough between cases do we shift the focus of explanation to situational and, finally, to dispositional factors. In a first step we will briefly discuss the theoretical background of the hypotheses.

(H1) The more liberal the media system of a state is, the more conflictual is its position on the NWICO and the more uncooperative is its behavior in the UNESCO crisis.
This hypothesis takes ideology seriously. It is based on the idea of a "domestic analogy", i.e. the assumption that states want their international environment to be ordered by the same values and principles as their domestic systems (cf., e.g. Burley 1993; Suganami 1989). In the area of information and communication these values and principles are institutionalized in the structure of the media system.

(H2) Trading states adopt a more conflictual position on the NWICO and act more uncoo peratively in the UNESCO crisis than military-political states.
Rosecrance (1986) distinguishes between "trading states" and "military-political states" with respect to their foreign policy orientation. However, since we intend to use this approach as an independent variable, we direct our attention exclusively to the internal differences of both types of states . According to Rosecrance, all military-political states seek to be self-sufficient in order to achieve full independence. Therefore, we can assume that a military-political state strives for control over news flows and supports the NWICO. In contrast, the trading world is composed of functionally differentiated nations. Each may seek to improve its position but, because the states depend upon each other, we can assume that they are interested in a liberal international order which is the most compatible with functional differentiation.

(H3) Given a bipolar overall power structure of the international system both great powers take opposite conflict positions and behave the opposite during the UNESCO crisis. The more a state is dependent on or allied with one of the great powers, the more this state supports the position and follows the behavior of the great power.
It is a basic Realist tenet that the overall international power structure exerts the most important influence on foreign policy. Given the bipolar structure of the international system at the time of the UNESCO crisis, it can be expected that each superpower tries to maximize its influence in international politics or at least tries to prevent the other from shifting the balance. This would explain their antagonistic positions and behaviors in what appears to be a struggle for control over the world information flow and for influence in the Third World. The lesser powers are constrained to align themselves with one of the superpowers and to support its policy (cf. Waltz 1979: 157, 171).

(H4) The more powerful a state is in international communication, the more conflictual is its position on the NWICO and the more uncooperatively it behaves in the UNESCO crisis.
The hypothesis about the effects of issue area power on world order preferences was developed in the context of neo-Realist international political economy. According to the theory of hegemonic stability, a hegemonic power is both interested in a liberal world order and strong enough to guarantee international compliance with liberal rules. When its relative power declines, its policy becomes more self-centered: the "benevolent hegemon" turns into a "predatory hegemon" (Gilpin 1987: 88-92). However, states, like the majority of Third World states, which are economically so weak that they cannot withstand global market pressures, favor authoritative control over market regulation (Krasner 1985: 3-13).

Table 5 displays the tau (b) values for the structural factors. As the last column shows, neither the "state type" nor the states' positions in the "issue-area power structure" successfully explain the variation of conflict positions. By contrast, the variables "media system" and "position in the overall power structure" yield satisfactory explanatory results without any discordant pairs in ordinal association.

(H2) Rosecrance´s typology of trading and military-political states does not seem to be helpful in explaining interests in the field of international communications because it does not discriminate enough between the states. The Soviet Union, the United States amd France had to be classified as military-political states for two reasons: Both their import and export rates were below the average of the industrialized states, whereas the military expenditures and the participation of the three countries in military conflicts were above the average of all industrialized countries. Yet, in spite of this similarity, the foreign policy behavior of the Soviet Union and the United States differed to the greatest imaginable extent from each other.

Table 5: Values for the Structural Factors Assumed to Explain Interests

(H4) Positions in the issue-area power structure were determined with regard to each country's share of the world news and TV markets and its possession of key technologies in the sphere of communication.[Note 23] This hypothesis fails mainly because one would have expected France to take a more liberal stand and to pursue a less cooperative policy in UNESCO than West Germany given France's stronger position in both media markets and its edge in space technology.

(H1) The Soviet Union not only exercised strict state control over its national media but also favored sovereign state control over the international information exchange, i.e. a position close to mainstream Third World demands. Whereas all Western states possessed a commer cial press system, the state-media relationship was structured differently with regard to the audiovisual media which are considered to be most influential for mass public opinion. The United States with its almost exlusively commercial radio and TV system also took the most liberal, pro-market position on the international information flow. West Germany with its "public" (öffentlich-rechtlich) audiovisual media and a pluralistic party and societal control over them was more concerned about a just distribution of opportunities in the international media market. And France with its tradition of high governmental influence on television and radio more or less openly favored protection of the national media market while strongly opposing any ideological censorship of the international news flow. (The partial commercialisation of German and French TV and radio was only in its very beginnings during our period of investigation.)

(H3) The Realist conception of interests based on a country´s position in the overall power structure very well explains why the antagonistic superpowers United States and Soviet Union took widely different positions on the NWICO issue. And it also accounts well for the fact that West Germany, highly dependent on the United States in security matters, was closer in its position to the United States than France which, in its Gaullist tradition, was more loosely attached to NATO and sought to play an independent role in world politics.

However, since the structural factors "media system" and "overall power structure" remained constant during our period of investigation, they cannot account for changes in the actors' positions. In particular, they do not explain why the French position in the NWICO debate became less liberal after 1980 and why the U.S. administration took a more liberal stand between 1980 and 1983 than before. Finally, they fail to explain why France and Germany had the same position in 1978 and 1980 although their media systems and their placement in the international power structure differed. Let us now turn to the situational and dispositional factors to find out whether they are able to supplement the explanation provided by these two hypotheses.

(H5) The higher the costs are which a state expects in case of a change of the liberal interna tional information and communication order, the more conflictual is its position on the NWICO and the more uncooperative is its behavior in the UNESCO crisis.
Hypotheses about situational factors in the utilitaristic mode of foreign policy explanation focus on the concept of costs and benefits .[Note 24] In this vein the most basic assumption would be that a state which expects high costs from a change in the liberal international information and communication order, opposes NWICO more strongly than a state which expects lower costs or even benefits.

(H6) The higher the media political problem pressure is for a state, the less conflictual is its position on the NWICO and the more cooperative is its behavior in the UNESCO crisis.
Problem pressure is a related situational factor.[Note 25] The presumed relevance of this variable is based on the fact that transboundary communication is continually increasing. This creates an especially high problem pressure for states which are forced to maintain an information monopoly because of their system of rule. Societies attaching high importance to the preservation of their cultural identity can also be expected to react adversely to a liberal international information and communication order.

(H7) The more negative the media coverage of the Third World positions is in a state, the more conflictual is its position and the more uncooperative is its behavior in the UNESCO crisis.
Public opinion is supposed to co-determine the scope of foreign policy options for the foreign policy decision-makers (cf. Müller/Risse-Kappen 1993: 38-41; Russett 1990; Wittkämper 1986). UNESCO has never been a matter of great public interest among the overall population, and indeed there is a general lack of survey data on this issue. Therefore we assumed that any potential influence must have been exerted by the media and limited the analysis to the so-called "opinion leaders" among the "elite press". It can be expected that negative media coverage of UNESCO would have strengthened critical attitudes towards the organization whereas a more positive treatment of UNESCO would have induced a more cooperative foreign policy.

Values for the situational factors are displayed in Table 6. They show a very good overall fit both for the "cost-benefit" and the "problem pressure" variables. Only the factor "media coverage" does not seem to be helpful in explaining the interests and the foreign policy behavior of the four industrialized countries.

Table 6: Values for the Situational Variables Assumed to Explain Interests

(H7) For lack of an independent press this hypothesis was not applicable in the case of the Soviet Union. The relatively low value of tau (b) mainly results from two discordant pairs in the case of France: The position of the French government on the international information and communication order was more in conflict with the NWICO proposals in the years 1978 and 1980 than in 1985 and 1987 although the media coverage of the UNESCO crisis (in Le Monde and Le Figaro) was more negative after the withdrawal of the United States and Great Britain from UNESCO than before. Moreover, it cannot be ascertained whether media coverage influences government positions as an independent factor or rather follows government policy. For both reasons, it appears to be safer to discard this hypothesis altogether.

(H5/6) Looking at the two other factors together, it is obvious that problem pressure was virtually non-existent for the United States since there was no danger whatsoever of foreign media dominating the American market. On the other hand, a successful change in the international information and communication order would have caused high costs for the United States. That the United States´position was the most market-oriented and its behavior the most uncooperative is exactly what our hypotheses would predict under these circum stances. At the other end of the spectrum, for a system of rule based on the state monopoly of information, the uncontrolled flow of information constitutive for the liberal international information and communication order resulted in a high problem pressure for the Soviet Union. The benefits of a change in the international order would have clearly outweighed any possible costs. So the Soviet support for Third world demands for a NWICO is highly understandable. In France the dominance of American TV and movie productions was widely regarded as a threat to national culture and therefore created a marked problem pressure and a call for protectionism. Therefore, France might have benefitted from a more protectionist international order. This led France to partially support the NWICO. Germany took a more liberal stand than France because the German weakness in the international media market was no issue of public concern and therefore created no problem pressure. At the same time, because of this weakness a change in the international information and communication order would have caused less costs to West Germany than to the U.S. That the German behavior in UNESCO was much more cooperative than U.S. behavior although both supported the liberal international information and communication order can thus also be explained.

However, these variables do not improve on the structural explanation presented before because they did not vary significantly over time and therefore cannot account for changes in the conflict positions of the four countries. Rather, it seems that the values for the situational variables "expected costs" and "problem pressure" are not independent of the structure of the media systems: It is quite plausible to assume that a state worries the less about the influx of foreign information and the more about political regulation of international information and communication the more liberal its media system is. The liberal international media order therefore created the highest problems for the communist countries because the stability of their political systems depended on the information monopoly of the one-party-state. And a restructuring of this order would have implied the highest costs for the United States whose commercial media industries benefited most from a market-type order. The situational factors "costs" and "problem pressures" illustrate well how structural conditions translate into situational constraints and options. However, for the sake of parsimony, they could be omitted from the explanation of conflict positions.

Although we did not formulate any specific hypotheses, the Realist explanation based on the overall power structure can also be translated into a cost-benefit assessment of the actors. One could argue that an illiberal change of the international information and communication order would have strengthened the Soviet Union in its global power struggle with the United States. Because of their different military dependence on the United States, it was more costly for Germany than for France to deviate from the American position.

The second step in improving on the structural explanation is to take into account dispositional factors .

(H8) The more a government is on the right of the ideological spectrum, the more conflictual is its position and the more uncooperatively it behaves in the UNESCO crisis.
It has been hypothesized in research on domestic policy that "parties matter", particu-larly in issue areas of high ideological import (Schmidt 1982). According to this view, centrist or conservative parties are rather status-quo oriented and emphasize private initiative. In con trast, left-wing parties support government intervention in the markets, frequently criticize the existing North-South relationship, and emphasize international solidarity. Because of that, one may suppose that left-of-centre governments are more open to the claims for a NWICO.

(H9) The more negative a state defines the situation in the UNESCO crisis, the more con flictual is its position and the more uncooperative is its behavior.
"Cognitive mapping" is a dispositional approach closely linked to the rationalist mode of foreign policy explanation (Axelrod 1976). It takes the causal beliefs of foreign policy decision-makers as its point of departure and reconstructs the causal connections the actors establish between concepts and according to which they define their situation. Since it is diffi cult to derive general statements from this approach we stick to a simplified application of "cognitive mapping" and assume that the degree of conflict intensity between a state and the Third World majority in UNESCO depends on how negative the decision-makers evaluate their situation in the UNESCO crisis and the prospects of the organization.

The values for the dispositional variables are listed in Table 7. As could be expected the variation in the dispositional variables is particularly high because governments and disposi tions change more readily than structures and situations. Both hypotheses have a fairly good fit but the factor "governing parties" does not only attain a higher tau (b) value but also fills the gap left by the structural explanation.

(H9) The general trend of the values for the definition of the situation is in accordance with the hypothesis. For example, the United States after 1980 considered the UNESCO crisis to be the tip of the iceberg in the general crisis of the whole UN system and the debate over a NWICO to be instrumentalized by the Soviet Union. By contrast, the Soviet Union was extremely content with the situation until 1987 when the global situation was re-interpreted in the light of the "New Thinking". Before that, the Soviets were convinced that UNESCO's im portance in international relations was increasing and that the demands of the Third World countries were justified and in line with their own goals. That the country with the most negative definition of the situation behaved most uncooperatively during the whole UNESCO crisis is exactly what we would have expected. However, the hypothesis cannot explain why the negative definition of the situation in France in 1987, West Germany from 1983 through 1987 and the United States in 1978 and 1980 did not induce these countries to pursue a similar foreign policy. Concerning the insufficiencies of the structural explanation, the definition of the situation can explain the shift in the U.S. position but fails to account for the changes in the French position and the similarity of the French and German positions in 1978 and 1980.

Table 7: Values for the Dispositional Variables Assumed to Explain Interests

(H8) Hypothesis 8 possesses more explanatory power with regard to the residuals of the structural explanation than Hypothesis 9. Concerning the "media system", it explains that the change in the French position from a moderately liberal to a more protectionist stance occurred after President Giscard d'Estaing and the center-right RPR-UDF government were replaced by President Mitterrand and the left-wing PS-PCF government in 1981. The change in the American position from a moderately liberal position to liberal orthodoxy coincided with the change from the Democratic Carter to the Republican Reagan administration. Finally, this hypothesis helps to account for the similarity of West German and French positions in 1978 and 1980 in spite of differently structured media systems: France was governed by a center- right coalition at the time which pushed the position of the French government to the liberal extreme of what was structurally possible in a country with high state influence on the audiovisual media. By contrast, Germany was governed by a center-left coalition which took a moderate position within the structurally determined range of possible interests.

The same logic can be applied to improve the Realist explanation of foreign policy interests by the overall power structure: As left-wing governements are in general, President Mitterrand and the PS-PC-government were more critical of the United States than their predecessors. The Reagan administration's highly negative view of East-West relations can explain the shift in American interests. And the parallel incumbency of a center-right coalition government in France and a center-left coalition in Germany accounts for the French-German proximity in their positions despite different placements in the international power structure. Thus, by adding the dispositional variable "ideological orientation of the government" to both structural variables, we arrive at a complete explanation of the conflict positions of the four industrialized countries selected for comparative analysis.

6. Summary and Interpretation of Findings

The findings that our research yielded can now be summarized:
(1) The assumption that foreign policy behavior is determined by the substantive interests of a state in a given issue area, as expressed in its conflict positions, is supported to a very high degree by the empirical evidence. The findings were sufficiently convincing to discard the alternative causal pathway of "foreign policy styles" working at least partly independently of interests and to proceed with testing interest-based hypotheses only.

(2) The test results for the causal pathway of interest-oriented behavior were strongest when we subdivided the sample into two time periods separated by the withdrawal of the United States and Great Britain from UNESCO, i.e. the most decisive single events during our period of investigation. This implies an independent effect of the unfolding crisis on the foreign policy of the four countries and underlines the usefulness of a comparative research design taking into account the dynamics of foreign policy formulation.

(3) The interests of the states can be explained, to a very large extent, as structurally determined either by their societal context or by their international position. However, structures cannot explain changes in the conflict positions of the United States and France, and do not discriminate sufficiently between France and Germany at certain points in time. This slight underdetermination of conflict positions and thus foreign policy behavior cannot be corrected by adding situational variables. However, a dispositional variable, the ideological orientation of the government, as indicated by its party composition, proved successful in accounting for the change in conflict positions and in fine-tuning the structural explanation.

(4) For our research, the Realist-Liberal debate about the international-positional or the societal causation of foreign policy ends in a draw with a certain Liberal advantage. We found two consistent and empirically tenable explanations which account for almost the entire variation in foreign policy behavior, one based on the media systems of the four countries and the other based on their position in the international overall power structure. The advantage of the Liberal explanation is two-fold: First, for the residual variance we had to add the dispositions of the governing parties, i.e. a variable which fits well into a Liberal framework but would usually be regarded as irrelevant by Structural Realism. Although Waltz generally concedes that unit level variables can be causally relevant, he does not integrate them into his systemic theory (1979: 65ff.). According to the Realist model of international politics, which is strongly shaped by international power relations and the notion of a "national interest", one would not expect "parties to matter" in alliance politics.

Second, whereas Realism accounts for the inevitably strong controversy over international order issues between the United States and the Soviet Union, only a Liberal theory of societal interest formation explains why the United States supported an economically and ideologically liberal information and communication order and why the Soviet Union aimed at a maximum of state control. [Note 26]

A generalized Liberal account would stress that issue area-specific and societal factors are capable of explaining the variation in governmental positions and in foreign policy behavior. One can observe a "domestic analogy": States want their international environment ordered essentially according to the same values and principles as their domestic societies. The structure of the international information and communication order should therefore be compatible with the structure of the national media system. However, as changing gov ernments are able to pursue different media policies emphasizing different aspects of the national media system without changing its structure, so does the domestic media system offer some scope for varying positions on the international information and communication order depending on the political-programmatic orientation of the government of the day. Whereas right-of-center governments emphasize commercialization and privatization, left-of-center governments are more open to state intervention.

In contrast, a generalized Realist account would underline the explanatory strength of a state's position in the international system in general and its power structure in particular. For the superpowers, the UNESCO was one arena in which their worldwide competition for influence took place, and a new world information and communciation order could have prepared the way for a major redistribution of cultural power between them. Therefore, the superpower interests in UNESCO differed widely. The other industrialized states supported one or the other side more or less strongly depending on the degree of dependency on one of the superpowers. When the U.S. realized that a majority of the Third World states had formed a coalition with the Soviet Union in UNESCO, the American government first tried to split this alliance through material incentives to the Third World and then put pressure on it by announcing its intention to leave the organization. Variations within the range of possible behavior defined by the structural placement of a country can be explained by the ideological orientation of the governing parties in the liberal-democratic industrialized countries. Whereas center-right governments tend to emphasize ideological confrontation, preservation of the status quo, and alliance cohesion, more left-wing leaders and coalitions lean more strongly towards reform of the international order. Both explanations are illustrated in Figure 2 and Table 8.

Compared with the relevant scholarly literature on the foreign policy of industrialized states towards UNESCO our research design differs in several respects: First, the comparative analysis of the foreign policies of four industrialized countries towards one specialized agency of the United Nations, UNESCO, contrasts with the writings of those scholars who either deal with the foreign policy of a single state towards UNESCO or the foreign policy of one state towards several specialized agencies of the United Nations (Karns/Mingst 1990) or the whole UN system (Gregg 1993). Second, our analysis goes beyond these studies with regard to the scope of theoretical approaches and explanatory variables integrated into one analytical model. Third, our study is based on a dynamic model of foreign policy that is able to take into account changes over time and their causes.[Note 27]

Figure 2: Explanatory model

Table 8: Values for the Explanatory Model

By pursuing a comparative, multivariate, and dynamic analysis we found support for some of the results in the literature and can suggest modifications of others. In particular, we can show that explanations which appear to be sufficient in inductive single-case studies are neither the only possible nor the most convincing explanations. Sometimes single explanatory factors turn out to be only partially successful or reducible to other factors in a broader comparative analysis. Finally, single-case studies tend to advance (apparently) idiosyncratic explanations which can be avoided when the variables are conceptualized differently. Some examples from the literature may help to substantiate this claim:

(1) Coate (1988: 119-149) as well as Imber (1989: 118-120) emphasize the ideological factor in their explanations of the United States foreign policy at the height of the UNESCO crisis. Coate points out that a small ideologically motivated group of government officials determined the U.S. policy towards UNESCO. He argues that this was possible because of an administration in office for which the ideological orientation of newly appointed officials was crucial in the recruitment process. Imber explains the withdrawal of the United States among other things with the neoconservative outlook of President Reagan. The findings of our analysis confirm the importance of this ideological factor for the re-orientation in U.S. policy. However, whereas Coate and Imber emphasize the orientation of the government of the day we would argue that the fundamental values of the society (as institutionalized e.g. in the media system) account for much more of the variance in UNESCO policies across states.

(2) Our explanation of American UNESCO policy differs somewhat from the one put forward by Karns and Mingst (1990: 300-317). On the one hand, they emphasize the explanatory power of the characteristics of international organizations. This is an independent variable we could not take into account. On the other hand, they do not consider a state's position in the international power structure to be as important as we do. We agree with Karns/Mingst that the majority's policy goals in UNESCO did not correspond to the American societal values and were therefore opposed by the United States. Apart from this Karns/Mingst mention the explanatory power of "domestic factors", among others the role Gregory Newell, Assistant Secretary of State for International Organization Affairs, played in the decision-making process concerning the UNESCO crisis. Gregory Newell was appointed to his office by the Reagan administration so that this seemingly idiosyncratic influence is covered by our general factor "ideological orientation of the governing parties".

(3) According to Gregg (1993: 143-146) the United States foreign policy towards the United Nations can be explained by the position of the United States in the international system and by the American desire to see American values and interests mirrored in the policies of the United Nations. Our analysis shows that these two explanations hold in a comparative foreign policy analysis, too.

(4) Roth (1982; 1983) explains the position of the Soviet Union in the conflict over the NWICO with its concern for the preservation of its system of rule and the corresponding goal to prevent a free flow from the West to the socialist countries. This single case explanation could be substantiated through the comparative analysis but we showed that an explanation emphasizing the international power structure was tenable as well.

This again demonstrates the superiority of comparative analysis to single case studies: The systematic comparison of cases raises merely plausible explanations to the status of more tenable ones. In addition, testing hypotheses systematically protects against accepting the first plausible explanatory factor as the necessary and the only possible one. However, our findings cannot claim to be unqualifiedly valid beyond the policies of the four selected countries towards UNESCO either: The selection of cases and variables limits the extent to which the results are generalizeable; potential causal effects of UNESCO's organizational structure (compared with other international organization) were not taken into account; and the findings do not allow to decide between the Realist and the Liberal explanation of foreign policy. There are three conceivable ways to improve on our results:
(1) by analyzing the UNESCO policies of other industrialized countries - preferably those whose international power position and media system lead to contradictory predictions about their conflict behavior;
(2) by a new research design in which cases are selected on those independent variables which have proved to be of high explanatory value in this study and in which other influences are more carefully controlled;
(3) by comparing the UNESCO policies of industrialized countries with their policies in other specialized organizations of the UN system in order to test the causal influence of organizational structures on foreign policy.

Notes

1. This paper summarizes the results of a research project on the "Policies of Industrialized Countries towards UNESCO 1978-1987". It was conducted between 1991 and 1995 at the Center for International Relations/Peace and Conflict Research, University of Tübingen, and funded by the German Research Association (DFG). The complete results of the project, including the case studies, are published in German as Rittberger, Volker (Hrsg.): Anpassung oder Austritt: Industrieländer in der UNESCO-Krise. Ein Beitrag zur vergleichenden Außenpolitikforschung, Berlin: edition sigma (Rainer Bohn Verlag), 1995. Preliminary results of the project were published as issue 19 of the TAP series and in Cooperation and Conflict 28: 2, 1993, 143-180.
2. See Americans for the Universality of UNESCO , X (May 1994), 1-5.
3. For an overview see Müller/Risse-Kappen (1993).
4. Cf. the criticism of Haftendorn (1990: 410f.) and Vasquez (1986: 207). Well-known works of the decision-making approach are Snyder/Bruck/Sapin (1962) and Allison (1971). For an overview of cognitive factors shaping foreign policy decisions cf. Vertzberger (1990).
5. See Allison's rational actor model in Allison (1971: 10-38).
6. Cf. the discussion of monism in Krasner (1978: 35-54) and Zürn (1992: 86-89).
7. See the models of "organizational process" and "bureaucratic politics" which Allison (1971) contrasts with the "rational actor".
8. In International Relations, the debate has been relaunched by Wendt (1987). For foreign policy analysis in particular cf. Carlsnaes (1992).
9. Some factors, such as the overall power structure or the ideological orientation of the governing parties in a state, can be plausibly assumed to cause "interests" as well as "policy styles". Some components of the analytical model, such as systemic situational factors influencing foreign policy interests, could not be filled with any variables.
10. Cf. as representative compilations Rosenau (1974) and East/Salmore/Hermann (1978).
11. Cf. also Lijphart (1975: 158ff.) and King/Keohane/Verba (1994: ch. 4, 6). The first condition, variation in the dependent variable, can be dropped, when cases are selected on the independent variables.
12. Speech by the Soviet Vice-minister of Foreign Affairs Adamishin at the 24th General Conference in Paris as cited in Komissija 1989: I, 390. Cf. also the interview with the First Secretary of the Soviet UNESCO Commission Karlov in Bjulleten' Komissii SSSR po delam JUNESKO (January 1988), 5.
13. Cf. UNESCO: Information Sheet 1 (COM/DCS, March 1982).
14. Cf. Le Monde , 18 April 1985.
15. See U.S. State Department, United States Participation in the United Nations 1978 (Washington, D.C.: U.S. Government Printing Office, 1980), 171.
16. See U.S. State Department, United States Participation in the United Nations 1978 .
17. See U.S. State Department, United States Participation in the United Nations 1980 (Washington, D.C.: U.S. Government Printing Office, 1981), 241f.
18. For an overview of the Soviet position Kolosov/Cepov (1983) and Kaslev (1981).
19. Since medians based on three values do not discriminate enough between cases, plus and minus was added to indicate the upward or downward tendency of the value. "Conflict behavior" is also expressed in ordinal numbers: loyalty equals "0", constructive behavior "1", conditional behavior "2", confrontative behavior "3", exit "4".
20. One should keep in mind that the statistical tests performed here only apply to the selected cases and that we do not purport to make any statements about the universe of cases. Significance tests which also exist for tau (b) cannot be meaningfully applied to a small number of cases which have not been selected randomly and do not fully meet the criteria of independence.
21. Cf. Benninghaus (1985: ch. 6) and Dreier (1994: 224ff.) for the characteristics of measures of ordinal association and their calculation.
22. For the mixed pairs (data from two countries), this correlation is only significant in the case of the NWICO issue-area (7 out of 10 pairs in contrast to only 9 out of 28 pairs when all conflicts are taken into account).
23. Figures were taken from Höhne (1984) with regard to the amount of words sold by national news agencies and from "International Flow of Television Programmes," UNESCO Reports and Papers on Mass Communication 100 (1985), with regard to the import and export of TV programs. Variation was low, however, with regard to technology.
24. See Müller/Risse-Kappen (1993: 42-46); in the context of foreign environmental policy e.g. Prittwitz (1984).
25. This factor is also used in the analysis of foreign environmental policy (cf. e.g. Schwarzer 1990).
26. Cf. Ruggie's similar argument with regard to hegemonic stability theory (1983: 198): "Power may predict the form of the international order, but not its content ."
27. Cf. e.g. John Fobes, former chairman of the U.S. National Commission for UNESCO (1979-1981), who suggested to study the impact which the United States withdrawal from UNESCO had on the UNESCO policy of other states, in Coate (1988: vii).

Bibliography

• Allison, Graham T. 1971: Essence of Decision. Explaining the Cuban Missile Crisis, Boston: Little, Brown.
• Archer, Margaret S. 1985: Structuration versus Morphogenesis, in: Eisenstadt, S. N./ Helle, H. J. (Hrsg.): Macro-Sociological Theory (Perspectives on Sociological Theory 1), London: Sage, 58-88.
• Axelrod, Robert 1976: The Cognitive Mapping Approach to Decision Making, in: Axelrod, Robert (Hrsg.): Structure of Decision. The Cognitive Maps of Political Elites, Princeton: Princeton University Press, 3-17.
• Becker, Jörg 1984: Vom Vorurteil zur Abhängigkeit. Konflikte auf dem Weg zu einer Neuen Internationalen Informationsordnung. Eine Einleitung, in: Friedensanalysen (Medienmacht im Nord-Süd-Konflikt: Die Neue Internationale Informationsordnung) 18, 16-44.
• Beigbeder, Yves 1987: Management Problems in the United Nations Organizations : Reform or Decline?, London: Pinter.
• Benninghaus, Hans 1985: Deskriptive Statistik, 5. Aufl., (Statistik für Soziologen 1), Stuttgart: Teubner.
• Brady, Linda P. 1982a: Goal Properties of Foreign Policy Activities, in: Callahan, Patrick/Brady, Linda P./Hermann, Margaret G. (Hrsg.): Describing Foreign Policy Behavior, Beverly Hills/London/New Delhi: Sage, 137-152.
• Brady, Linda P. 1982b: A Proposal, in: Callahan, Patrick/Brady, Linda P./Hermann, Margaret G. (Hrsg.): Describing Foreign Policy Behavior, Beverly Hills, London, New Delhi: Sage, 17-30.
• Burley, Anne-Marie 1993: Regulating the World: Multilateralism, International Law, and the Protection of the New Deal Regulatory State, in: Ruggie, John Gerard (Hrsg.): Multilateralism Matters. The Theory and Praxis of an Institutional Form, New York etc.: Columbia University Press, 125-156.
• Buzan, Barry/Jones, Charles/Little, Richard 1993: The Logic of Anarchy. Neorealism to Structural Realism, New York: Columbia University Press.
• Callahan, Patrick 1982: Event Data and the Study of Policy, in: Callahan, Patrick/Brady, Linda P./Hermann, Margaret G. (Hrsg.): Describing Foreign Policy Beha-vior, Beverly Hills, London, New Delhi: Sage, 293-306.
• Carlsnaes, Walter 1986: Ideology and Foreign Policy. Problems of Comparative Conceptualization, Oxford: Basil Blackwell.
• Carlsnaes, Walter 1992: The Agency-Structure Problem in Foreign Policy Analysis, in: International Studies Quarterly 36:3, 245-270.
• Coate, Roger A. 1988: Unilateralism, Ideology & U.S. Foreign Policy: The United States In and Out of UNESCO, Boulder, Colo.: L. Rienner.
• Dreier, Volker 1994: Datenanalyse für Sozialwissenschaftler, München/Wien: Oldenbourg.
• East, Maurice A./Salmore, Stephen A./Hermann, Charles F. (Hrsg.) 1978: Why Nations Act. Theoretical Perspectives for Comparative Foreign Policy Studies, Beverly Hills etc.: Sage.
• George, Alexander L. 1979: Case Studies and Theory Development: The Method of Structured, Focused Comparison, in: Lauren, Paul Gordon (Hrsg.): Diplomacy: New Approaches in History, Theory and Policy, New York: Free Press, 43-68.
• Gilpin, Robert 1987: The Political Economy of International Relations, Princeton: Princeton University Press.
• Gregg, Robert W. 1993: About Face? The United States and the United Nations, Boulder, Colo.: Lynne Rienner.
• Haftendorn, Helga 1990: Zur Theorie außenpolitischer Entscheidungsprozesse, in: Politische Vierteljahresschrift, Sonderheft 21, 401-423.
• Hermann, Charles F. 1978: Foreign Policy Behavior: That Which is to be Explained, in: East, Maurice A./Salmore, Stephen A./Hermann, Charles F. (Hrsg.): Why Nations Act. Theoretical Perspectives for Comparative Foreign Policy Studies, Beverly Hills, London: Sage, 25-47.
• Hirschman, Albert O. 1970: Exit, Voice, and Loyalty. Responses to Decline in Firms, Organizations, and States, Cambridge, MA: Harvard University Press.
• Höhne, Hans Joachim 1984: Report über Nachrichtenagenturen, Baden-Baden: Nomos.
• Huntington 1991: The Third Wave. Democratization in the Late Twentieth Century, Norman/London: University of Oklahoma Press.
• Imber, Mark 1989: The USA, ILO, UNESCO and IAEA: Politicization and Withdrawal in the Specialized Agencies, New York: St Martin's Press.
• Karns, Margaret P.; Mingst, Karen A. (Hrsg.) 1990: The United States and Multilateral Institutions : Patterns of Changing Instrumentality and Influence, Boston etc.: Unwin Hyman.
• Kaslev, Ju B. 1981: Massovaja informacija i mezdunarodnye otnosenija, Moskva: Mezdunarodnye Otnosenija.
• Keohane, Robert O. 1984: After Hegemony. Cooperation and Discord in the World Political Economy, Princeton, N.J.: Princeton University Press.
• King, Gary/Keohane, Robert O./Verba, Sidney 1994: Designing Social Inquiry. Scientific Inference in Qualitative Research, Princeton: Princeton University Press.
• Kolosov, Ju M./Cepov, B. A. 1983: Novyj mezdunarodnyj informacionnyj porjadok i problema podderzanija mira, Moskva: Nauka.
• Komissija SSSR po delam JUNESKO (Hrsg.) 1989: Dokumenty i materialy 1954-1987, Bd. 2 Bde, Moskva: Mezdunarodnye Otnosenija.
• Krasner, Stephen D. 1978: Defending the National Interest. Raw Materials Investments and U.S. Foreign Policy, Princeton, NJ: Princeton Univ. Press.
• Krasner, Stephen D. 1985: Structural Conflict The Third World Against Global Liberalism, Berkeley, Cal.: University of California Press.
• Lijphart, Arend 1975: The Comparable-Cases Strategy in Comparative Research, in: Comparative Political Studies 8, 158-177.
• Müller, Harald/Risse-Kappen, Thomas 1993: From the Outside In and from the Inside Out: International Relations, Domestic Politics, and Foreign Policy, in: Skidmore, David/Hudson, Valerie M. (Hrsg.): The Limits of State Autonomy. Societal Groups and Foreign Policy Formulation, Boulder etc.: Westview Press, 25-48.
• Petzsch, Regine 1987: Die Bundesrepublik Deutschland in der UNESCO. Reaktion und Aktion angesichts der UNESCO-Krise in den achtziger Jahren, Bonn: Deutsche UNESCO-Kommission.
• Prittwitz, Volker 1984: Umweltaußenpolitik: Grenzüberschreitende Luftverschmutzung in Europa, Frankfurt/New York: Campus.
• Rosecrance, Richard 1986: The Rise of the Trading State. Commerce and Conquest in the Modern World, New York: Basic Books.
• Rosenau, James N. (Hrsg.) 1974: Comparing Foreign Policies. Theories, Findings and Methods, New York [u.a.]: Wiley.
• Roth, Paul 1982: Die neue Weltinformationsordnung. Argumentation, Zielvorstellung und Vorgehen der UdSSR (I, Berichte des BioSt 44), Köln: Bundesinstitut für internationale und ostwissenschaftliche Studien.
• Roth, Paul 1983: Die neue Weltinformationsordnung. Argumentation, Zielvorstellung und Vorgehen der UdSSR (II, Berichte des BioSt 17), Köln: Bundesinstitut für internationale und ostwissenschaftliche Studien.
• Ruggie, John Gerard 1983: International Regimes, Transactions, and Change: Embedded Liberalism in the Postwar Economic Order, in: Krasner, Stephen D. (Hrsg.): International Regimes, Ithaca etc.: Cornell University Press, 195-231.
• Russett, Bruce 1990: Controlling the Sword. The Democratic Governance of National Security, Cambridge, Mass. etc.: Harvard Univ. Press.
• Schiller, Herbert I./Herman, Edward S./Preston, William 1989: Hope and Folly: The United States and UNESCO, Minneapolis: University of Minneapolis Press.
• Schmidt, Manfred 1982: Wohlfahrtstaatliche Politik unter bürgerlichen und sozialdemokratischen Regierungen. Ein internationaler Vergleich, Frankfurt etc.: Campus.
• Schwarzer, Gudrun 1990: Weiträumige grenzüberschreitende Luftverschmutzung: Konfliktanalyse eines internationalen Umweltproblems, (Tübinger Arbeitspapiere zur internationalen Politik und Friedensforschung 15), Tübingen: Attempto.
• Snyder, Richard C./Bruck, H. W./Sapin, Burton 1962: Foreign Policy Decision-Making. An Approach to the Study of International Politics, New York: Free Press of Glencoe.
• Suganami, Hidemi 1989: The Domestic Analogy and World Order Proposals, (Cambridge Studies in International Relations), Cambridge etc.: Cambridge Univ. Press.
• Uranov, G. V./Kanaev, N. M. 1986: JUNESKO - 40 let, Moskva: Znanie.
• Vasquez, John A. 1986: Explaining and Evaluating Foreign Policy: A New Agenda for Comparative Foreign Policy, in: Vasquez, John A. (Hrsg.): Evaluating U.S. Foreign Policy, New York: Praeger, 205-229.
• Vertzberger, Yaacov Y. I. 1990: The World in Their Minds. Information Processing, Cognition, and Perception in Foreign Policy Decisionmaking, Stanford: Stanford University Press.
• Waltz, Kenneth N. 1965: Man, the State and War. A Theoretical Analysis, New York etc.: Columbia University Press.
• Waltz, Kenneth N. 1979: Theory of International Politics, New York: Random House.
• Wendt, Alexander E. 1987: The Agent-Structure Problem in International Relations Theory, in: International Organization 41:3, 335-370.
• Wittkämper, Gerhard W. 1986: Medienwirkungen in der internationalen Politik, Münster: Lit.
• Zürn, Michael 1992: Interessen und Institutionen in der internationalen Politik. Grundlegung und Anwendung des situationsstrukturellen Ansatzes, Opladen: Leske + Budrich.

Address of the Authors: