Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
An American View of Soviet Contributions WILLIAM K. ESTES AND CHARLES TILLY We are searching for promising convergences and fruitful differences between the approaches of Soviet and American scholars to the study of interdependence. Convergences signal areas in which it should be relatively easy to pool results and establish collaboration. Differences identify opportunities for comparison and constructive debate. We have found both: current Soviet work on political leaders' concepts, models of interdependence, strategic stability, crises in Soviet-American relations, awareness of ecological issues, and other topics intersects in challenging ways with American work on related subjects. Although our task is chiefly to review Soviet contributions to the 1988 and 1989 workshops for possible connections with current American work let us begin by classifying the workshop papers from both national groups. The two workshops took up four principal objects of study: con- cepts, interaction, cognition, and public opinion. A crude but serviceable taxonomy results from crossing the four objects of study with a conven- tional division into conceptual, methodological, modeling, and empirical approaches, for a total of 16 possible categories. The breakdown of papers appears in Able 1. The tabulation brings out the significant weight given to conceptual work and the study of concepts. Semantic and historically based analyses of concepts that enter into the thinking of political and military leaders are a highly favored topic among Soviet investigators but almost unrepresented among American investigators. Americans, on the other hand, presented more than their share of empirical studies of interaction and cognition. Formal work on method and application of models to yield predictions about empirical phenomena are also largely the province of 7
Table 1 Subjects of Papers and TaLks Presented at the 1988 and 1989 Soviet-Amencan Workshops (full titles are In Appendix) Object of Study Approach 1988 Workshop 1989 Workshop CONCEPTS Conceptual Keohane, interdependence Averthev, interdependence Kokoshin, strategic stability Kremenyuk, negotiation Method Stein, measurement of interdependence Modeling Empirical INTERACTION Conceptual Averchev, Method Stein, measurement of interdependence Modeling Axelrod, evolution of cooperation Empirical George, U.S.-Soviet cooperation COGNITION Conceptual Method Modeling Sergeev et al., Caribbean missile crisis Empirical Tetlock, cognitive thresholds PUBLIC OPINION Conceptual Method Modeling Empirical- Korobeinikov, opinion polls Averchev and Kochetkov, structural comparability Axelrod, stability Baranov, casus belli Kremenyuk, balance of interest Stem, legitimate interest Kokoshin et al., concept of victory Blagovolin, military choice at. Elm, commumcatlon In interdependence Petrov, semantics of interdependence Estes, cooperation Tsyganov, multistage negotiation Baryshnikov and Berezovskiy, game theory and interdependence Putnam, diplomacy and domestic .. . pOlltICS Tilly, Third World conflict Sergeev and Parshin, Caribbean missile crisis Tetlock, cognitive-rhetorical styles Gamson, framing political issues Nistratov and Velichkovsky, mass consciousness Korobeinikov, shared political values Mitchell, environmental consciousness Sokolov and Abalkina, politics and ecology . 8
OVERVIEWS 9 American investigators, as are empirical studies in which events are classi- fied and their properties (notably probability of occurrence) systematically related to the historical contexts in which they occurred. The table al-so identifies several changes from the first meeting to the second. Empirical studies of public opinion gained prominence; the second meeting gave more attention to comparative studies of public opinion and connected survey research on nuclear issues with investigations of ecological and environmental issues. Finally, some subjects did not come up in either meeting, except as secondary themes: modeling of concepts and public opinion, and methods for the study of cognition and of public opinion. CONCEPTS Soviet work on concepts includes a mixture of historical analysis, case studies, philosophy, and rhetonc. The currently favored approach eluci- dates concepts that appear in the thinking of political and military leaders, as revealed in speeches, Political writings, and reports in mass media. The primary method is semantic analysis of texts, guided to some degree by concepts borrowed from cognitive science. Reports generally lack method- ological details. The most critical lack is of methods for sampling texts in such a way as to avoid bias in selection of particular cases for analysis. A useful Soviets erican collaboration could be focused on sampling prob- lems in textual analysis. The results of this line of investigation would not yield empirical generalizations, but they would elucidate the thinking of So- viet social scientists on concepts related to nuclear issues and presumably the thinking of political and military leaders who produce the materials under study. The paper by Kokoshin et al. [1989; dates cited in brackets signify papers at the workshops, all of which are listed in the Appendix] on the concept of victory typifies Soviet historical and semantic analyses of concepts pertaining to interdependence. The research group examined texts of speeches, reports to party congresses, and the like for frequency and pattern of use of terms related to the goals of Soviet military and nuclear policy. In data from the first decade following World War II, the investigators found evidence that the outcome of that war was being taken as a prototype for the outcome of any future conflict essentially, victory by one of the powers involved and unconditional surrender by the other. Many statements referred to the inevitability of victory by Soviet forces in any conflict as a consequence of the perceived strengths of socialism versus capitalism. By the 1960s, however, the earlier development of nuclear forces had made the notion of "natural victoriousness of socialism" meaningless and led to a change of emphasis from immutable historical laws to the analysis of the
10 SOVIET-AMERICAN DL4LOGUE IN THE SOCL4L SCIENCES technical problems of achieving victory in a nuclear conflict. Texts from the 21st and 22nd Communist Park Congresses exhibited much concern with assurances that Soviet strength would enable the USSR to be victorious over a nuclear aggressor. Beginning in the early 1970s, references to the "promise of victory" began to be replaced by a new emphasis' namely, that the goal of Soviet nuclear forces was not to establish supremacy, but to deliver a "decisive rebuff" to any nuclear aggressor. Themes of the moral superiority of the socialist cause continued to appear, but the goal of establishing its supremacy by force gave way to that of employing deterrent threats to prevent any outbreak of a nuclear convict, in which victory for either side would be impossible. The conclusions of Kokoshin et al. are potentially significant, but im- portant questions of methodology remain. Because the issues under study are politically charged, the analyses require stringent controls against bi- ased selection or interpretation. A major question for American analysts is to what degree the results of the historical analysis reflect changes in the thinking of political leaders as distinguished from changes in the interpre- tations of Soviet social scientists. This study points up the potential value of a collaborative effort in which American investigators would examine the methodology of a Soviet study closely and replicate its findings, while Soviet investigators conducted similar analyses of texts in which the same terms were used by American political leaders. Some Soviet conceptual analyses are scholarly essays with no specified data bases. They tend to yield hypotheses for future investigation rather than to test existing assumptions. Examples are papers by Kremenyuk [1988, 1989] on negotiation and balance of interests. In his first paper, Kremenyuk perceives a historical trend from emphasis on formal negotiations over nuclear issues between the superpowers to continuing diplomatic activity that is not closely tied to formal negotiation. He points up the need for more formal research involving the analysis of specific data bases and for work toward improvement of the less formal negotiation process. The 1989 paper is a historical sketch of perceived changes in the concept of balance of interests, especially with reference to the views of U.S. leaders. In Kremenyuk's opinion, the American picture of national interest has always depended on business considerations, with the addi- tion of the motif of anticommunism during the Stalin period. Jawaharlal Nehru's challenge of the early l950s to the growing emphasis on unilateral security and military strength was echoed by John F. Kennedy in 1963. Nikita Khrushchev then renewed the challenge in a speech maintaining that another world war was not inevitable and that peaceful coexistence between the superpowers should be possible. In Kremenyuk's view, the risk of war, whether advertent or inadvertent, has declined as a consequence
OVERV7EWS 11 of the growing recognition that the Soviet Union and the United States share basic values, despite their opposed interests in specific arenas and despite the need to balance national political and economic interests with the needs of mutual security. INTERACT ION Soviet work on interaction includes a variety of conceptual analyses, together with attempts at formal modeling of particular kinds of interde- pendence. ~Oim's report on principles of natural reasoning [1989] illustrates the conceptual approach. It could easily appear under the heading of cognition instead of interaction, since it stresses the reasoning by which individuals arrive at judgments about the desirability of different kinds of social situations. The paper presents concepts underlying a simple hedonic calculus, then adds to them a set of principles such as Principle 8: When a man learns of a relevant event in his physical environment, he first becomes interested in the consequences of the event; when he learns of an event in his social environment, he first becomes interested in the reasons (motives) for the event. Oim's brief report does not do any of the three things one might expect from an American author of such an essay: (1) summarize the observa- tional and experimental work that substantiates the proposed principles; (2) formalize the implied relationships in one or more models; (3) deduce empirical consequences, including puzzling or controversial phenomena the principles might explain. Other papers, such as Averchev [1988] on interde- pendence and Blagovolin [1989] on military choices,- deal with interaction among states rather than individuals but leave some of the same uncertainty about empirical bases and implications. Soviet papers presenting models of interdependence came closer to American scholarly custom. Tsyganov [1988, 1989], a member of the In- stitute for Control Problems, is investigating the possibilities of adapting control theory developed for the management of relatively limited systems (for example, health care) to international relations. The Institute's models typically deal with the interaction of multiple units over whose inputs and outputs a single center exerts considerable influence. A special problem in the international case is the lack of a single control center to impose stability. With two control centers (for example, governments), stability can evidently be obtained only through an adaptive learning process. Other special problems are the need for developing a common political language and objective means of evaluating national aims and interests. States make decisions in terms of their own interests, but the constraints of interde- pendence in the nuclear age require that each evaluate the interests and objectives of the other parties to an interaction. Although it is unlikely that
12 SOVIET-AMERIC~N DIALOGUE IN THE SOCIAL SCIENCES control theory can be extended directly to international interdependence in the near future, it does seem that using control theory to guide comparisons between problems of control in relatively well understood, limited systems and those faced in the international arena may be useful and instructive. It might also be possible to reverse the analytic procedure that is, to exam- ine what characteristics would be required of a unitary international system of control for such problems as arms flows and environmental pollution, given substantial conflicts of interest among oresumablv subordinate units such as states and corporations. O . In a study that was planned for inclusion in the 1989 workshop but was not actually reported, Baryshnikov and Berezovskiy present an anal- ysis indicating that the axioms of classical choice and game theory are inadequate to handle the interdependence of utilities that characterizes interactions among states. They consider possible revisions of game theory and conclude that the concept of a Nash equilibrium (Nash, 1951) provides a better starting point than that of a Pareto optimum (von Neumann and Morgenstern, 1953) but that the Nash system needs to be extended to allow for gradual change over time in utilities of the parties to an interaction.) Their model implies that, in interdependent situations, one party to an interaction cannot attain security by reducing the security of the other, or by imposing his system of values on the other. A problem for research is to identify the empirical conditions under which this model would be applicable. All in all, Soviet efforts in the modeling of interaction seem ready for more extensive Soviet-American collaboration. A number of Soviet contributions to both workshops do not report research results but rather analyze problems of interdependence, usually in historical context, and point up areas needing future research. A prime example is the paper by Kokoshin [1988] on strategic stability. It argues that adjusting to increasing interdependence among nations in the present state of the world requires increased reciprocity in international interactions. Reciprocity is double-edged, however, for it may create conditions for conflict as well as for cooperation. Kokoshin notes that problems involved in furthering reciprocity have been examined in several different contexts. The problems have been best solved, perhaps, in the economic realm, where management of international interactions is fairly effective. But the traditional focus on East-West relations needs to yield to more attention to North-South relations. In the ecological realm, reciprocity is found to be sorely lacking. Individual nations have impressive technological developments that pose iA Pareto optimum is a situation in which no player can improve his position without worsening that of the other participants. In a Nash equilibrium, each participant is assumed to make his decision on the premise that the strategies of the others remain unchanged.
OVERVIEWS 13 enormous threats to the global environment for example, the production and testing of nuclear weapons, insufficient provision for prevention of accidents and disposal of wastes from nuclear reactors, and the discharge of chemicals into the atmosphere. Massive efforts are needed to generate cooperative responses to such threats. In the military arena, Kokoshin finds wide agreement among arms control experts that stability is based on mutual deterrence, that is, on preventing the use of nuclear weapons. It is important to distinguish stability from parity, for parity can be maintained while stability is threatened and vice versa. Stability requires that a nation decide on its actions in the context of their consequences for the overall system and with attention to long-term as well as more immediate objectives. Progress toward international stability has been hindered by the super- powers' failure to recognize developing nationalism in other parts of the world. Thus part of the task of avoiding nuclear war is the expansion of attention beyond superpower relationships and the development of deeper understanding of the motives, perceptions, cultural values, and aspirations of ethnic and national groups around the world. These motifs are echoed in the papers by Averchev [1988, 1989] dealing with the awareness and understanding of issues of interdependence by individuals and groups responsible for decision making on the international level. They also resound in Blagovolin's study [1988] of military aspects of global interdependence. Clearly, the subject of international interaction invites further Soviet-U.S. collaboration. COGNITION The chief Soviet contributions under the general heading of cognition are papers on the Cuban missile crisis of 1962 by Sergeev et al. [1988] and Sergeev and Parshin [1989~. The purpose of these studies is to produce a model of the thinking of political leaders; the 1988 paper treats Kennedy; the 1989 paper, Khrushchev. The analysis of the Caribbean crisis is initiated with a novel and instructive variation on the classical approach of game theory. Up to a point, the procedure follows the standard routine. It is assumed that each of the two participants in the interaction has a choice between cooperation and conflict at each step. The values of the outcomes of joint choices to the participants are given by a payoff matrix. The rows of the matrix correspond to choices of cooperation or conflict by one participant, the columns to choices by the other. The cell entries are the relative values of outcomes to the two participants on an ordinal scale. The standard approach is to assume a particular structure for the interaction, which determines the cell entries in the payoff matrix.
14 SOVIET-AMERICAN DIALOGUE IN THE SOCIAL SCIENCES For a situation like that at the start of the Caribbean crisis, the usual assumption is that the participants are engaged in an interaction with the structure of the familiar Prisoner's Dilemma. This has the payoff matrix: Cooperation Conflict Cooperation 3,3 Conflict 4,1 1,4 2,2 where iJ in any cell denotes values of the outcome to the row and column participants, respectively. The theoretical prediction from game theory is that the participants will fail to find the optimal outcome and will settle for the payoff combination 2,2 in the lower right-hand cell, representing choices of conflict by both participants. Since neither the Caribbean crisis nor other, lesser crises involving the United States and the Soviet Union since the beginning of the nuclear era have actually resulted in conflict, there is reason to suspect that the assumption of the Prisoner's Dilemma might be incorrect. The approach of Sergeev et al. differs from the traditional one in that it does not assume any payoff structure in advance but rather infers the structure from data. In this instance, the data are the protocols of a sequence of news reports of notable events during the crisis, beginning with Kennedy's decision to escalate what the Soviet investigators refer to as "subversive activities" against Cuba; continuing with the agreement between the USSR and Cuba about deploying nuclear weapons, the relocation of the American air defense forces to the southeast of the country, the announcement of the quarantine of Cuba by the United States, and so on; and concluding at the 11th and i2th steps with the removal of the U.S. quarantine from Cuba and the joint request of the United States and the USSR to the United Nations to cancel the issue of the Caribbean crisis from the agenda. The first step in the analysis was to categorize each of a sequence of statements into the class of cooperation or conflict. The frequencies with which statements fall into the two classes provide the data for the theoretical analysis. In the absence of either a priori theory or data, it would be assumed that all arrangements of the payoff values 1 to 4 in the cells of the payoff matrix are equally probable. Starting with that null assumption, the procedure of Sergeev et al. was to revise the probabilities of the cell entries on the basis of the sequence of statements. At each step, they determined whether the observed statements on both sides pointed toward cooperation or conflict, observed which arrangements of the payoff values in the cells of the matrix were compatible with the observed joint choice, and then updated the probabilities of the cell entries accordingly. This procedure enables the analysts to determine at each step the likelihood,
OVERVIEWS 15 given the data, that the participants in the interaction were governed by any one of the possible payoff matrices. A number of results of the analysis are of particular interest. Perhaps most noteworthy, in view of the tradition of concentrating on the Prisoner's Dilemma structure for international interactions of this kind, is the fact that this structure never appeared as one of the most probable at any of the 12 steps of the interaction. The estimated probabilities indicated that in the early stage of the crisis the United States placed the highest value on anticipated outcomes associated with a choice of conflict but that the USSR placed the highest value on outcomes resulting from a choice of cooperation. By the time of Kennedy's second message to Khrushchev, the United States had shifted to higher anticipated value for cooperation, while the USSR had not changed. At the final step of the analysis, corresponding to a point just after the end of the crisis (the January 7, 1963, joint message of the United States and the USSR to the United Nationsy, the highest probability matrix was the one associated with the familiar game of Chicken, represented by the payoff matrix: Cooperation Conflict Cooperation 3,3 Conflict 4,2 2,4 1,1 Each participant places the highest value on the outcome of a choice of conflict, depending on the other to "weaken" and choose cooperation. An important question, assuming the validity of the result, is whether this final state of affairs represents the actual thinking of the leaders of the two countries or simply the kinds of statements that each felt it necessary to make in order to undercut criticism from hawkish elements within their respective countries. This analysis is a welcome change from the traditional, purely rational- istic approach of game theory, but the new, empirically based procedure is fraught with hazards of interpretation. A major one in this instance is the small size of the data base. A sequence of 12 events, each classified into the cooperation or conflict category, provides an extremely small number of cases for estimating a posterior) probabilities of theoretical matrices from observed data. Further, there is necessarily some arbitrariness In choosing particular events to represent the stages of the interaction and some sub- jectivity, or at least ambiguity, in categorizing the actions of the powers into the categories of cooperation and conflict. This approach deserves much more extensive testing, with other (and one would hope appreciably larger) data bases and with independent but parallel analyses being carried out by investigators from the two countries. In view of the difficulties of interpretation, the Soviet investigators decided not to rely very heavily on the outcomes of this analysis but rather
16 SOV7ET-AMERICAN DL4LOGUE IN THE SOCIAL SCIENCES to complement them with the results of a quite different approach involving cognitive mapping (Axelrod, 1976~. Sergeev et al. concluded that the game-theoretical approach suffered because the leaders' perceptions of the situation changed radically over the period of the crisis; in effect, different games were being played at different times. Consequently, Sergeev et al. put their major effort into modeling these changing perceptions of the political and strategic situation. The data base for the cognitive mapping analysis in the 1988 paper was a set of speeches and records of press conferences by Kennedy in September and October 1962. The procedure was to carry out semantic analyses of these texts in order to construct a cognitive map by identifying uses of conceptual terms related to the individual's goals, interests, and perceptions of the situation. The cognitive map is an oriented graph in which nodes correspond to states of the world (for example, nuclear war or peace) or actual or possible constituents of the political situation (for example, elimination of missiles from the western hemisphere, stability of Castro's regime in Cuba, stagnation of Cuban industries, Cuba's support by the USSR). These nodes are connected by links representing perceived causal relations (actual or possible). The results of this procedure are presented in two graphs, one representing President Kennedy's perception of the situation based on his news conference of September 13, 1962 Just before the crisis), and the second based on his television address to the American people on October 22, 1962 (when U.S. air defense forces began their relocation to southeastern states and the United States announced a quarantine on Cuba). The cognitive maps constructed for points in time shortly before the crisis show Kennedy believing that the international situation was control- lable by resources at the disposal of the United States and the Caribbean not being perceived as comparable to the Middle East in terms of potential danger. By late September 1962, the picture of the President's perceptions changed radically. The nuclear threat from the USSR and political insta- bility in the Caribbean fused to generate a major perceived threat to U.S. security, with the ratio of factors under U.S. control to those not under U.S. control having shifted strongly in the latter direction. Then, in the investigators' interpretation, the idea of interdependence "made a powerful breakthrough" into the President's cognitive map of the world, manifest in a perception of the situation as one in which neither superpower could achieve its desired ends and safeguard its security by unilateral measures. The perception of the presence or absence of "sincerity" and "restraint" on the part of the USSR emerged as a major factor in the situation. In the absence of these traits, escalation of the crisis might be the only possible U.S. response (the simulation showing that unilateral concessions by the United States would not secure its goals), whereas in the presence of these
OVERVIEWS 17 factors nearly all U.S. goals could be achieved without further escalation of the crisis. The paper lacks a further detailed analysis of shifting perceptions, but evidently the message from Khrushchev to Kennedy in late October 1962 lent enough credibility to warrant an effort to implement the second scenario. This effort took the form of acceptance by the United States of conditions contained in the second of Khrushchev's messages (October 27, 1962), following which the USSR indicated its readiness to withdraw the missiles from Cuba, and the crisis receded. This analysis gives only half the picture, of course. What of Soviet thinking? The complementary analysis of Khrushchev's perceptions by Sergeev and Parshin is based on documents published in the Soviet press in 1961 and 1962. The principal cognitive maps are based on the report of the Communist Party Central Committee to the 22nd Party Congress in October 1961, a year before the crisis, and Khrushchev's report to the Supreme Soviet and Pravda in December 1962, just following the height of the crisis. The cognitive map for Khrushchev constructed prior to the crisis shows a major goal of "curbing imperialism," much emphasis on metaphors expressing the power of the socialist bloc but little evaluation of the power of the United States and NATO, and a striking lack of analysis of possible U.S. responses to deployment of missiles in Cuba. The later cognitive map shows changes very similar to Kennedy's, with the notion of interdependence emerging in the form of recognition of the need for cooperative action to preserve nuclear peace and of the impossibility of either power's achieving all of its goals unilaterally. These analyses may be highly revealing of the modes of thought that first allowed the Caribbean crisis to develop, and then at the last moment prevented it from escalating to a fatal outcome. The validity and objectivity of the analyses deserve serious attention; they require new research involving much broader data bases. In particular, Soviet and American researchers could develop mutually acceptable criteria for identifying thematic units in the texts analyzed and creating from them onto cognitive representations. Without such criteria, investigators run a great risk of having their perceptions shape the conclusions. PUBLIC OPINION Opinion research and sociology in general are enjoying a vigorous rebirth in Soviet social science, in part because of increased freedom to carry out and report the results of such studies and in part because of the sharply increasing interest of Soviet political leaders in the results of survey research- evidenced, for example, by the recent development of a Soviet facility for nationwide public opinion sampling.
18 SOVIET-AMERICAN DIALOGUE IN THE SOCKS SCIENCES American participants in the workshops did not find evidence of new methodologies or theoretical developments in this line of research. Yet, as far as they go, the results of polls comparing Soviet and American samples on issues related to nuclear- conflict and environmental threats are relevant and instructive. This area of research has so far been one of the most fertile for the development of concrete collaborative efforts between Soviet and American participants in the workshops. Each group has much to gain from comparisons with the population to which the other has access, and the joint research poses methodological problems from which each can learn. Several workshop papers describe studies of public opinion by Soviet social scientists. Korobeinikov [1988] reported on the use of opinion surveys to assess the public understanding of interdependence, chiefly that of the Soviet and U.S. publics. One result taken to indicate that Soviet and Amer- ican publics increasingly recognize the need for reciprocity came from a poll conducted in Moscow concerning the results of the Gorbachev-Reagan summit: 90 percent of the respondents expressed satisfaction. That finding provides an interesting comparison with an NBC-Washington Post poll in the United States in which 70 percent of the respondents approved of a summit agreement on the elimination of intermediate nuclear forces in Eu- rope. Another aspect of public awareness of international interdependence appears in the results of a poll conducted in Japan: 38 percent of respon- dents expressed their belief that Japan would be devastated, and 56 percent that there would be serious destruction, in the event of a Soviet-American nuclear conflict. A follow-up paper by Korobeinikov [1989] cited results supporting the thesis that increasing awareness of shared values on the part of the peo- ples of the Soviet Union and the United States-is leading to a shift from a philosophy of mutual hostility to a philosophy of mutual dependence. Polls conducted by the Institute of Sociology of the Academy of Sciences, USSR and the Gallup organization in the United States at the end of 1987 addressed the question of what relationship should hold between the armed forces of the two countries. Seventy-one percent of Soviet respon- dents favored equality and only 15 percent Soviet superiority, whereas 50 percent of American respondents favored equality and 43 percent U.S. superiority. Another question asked respondents in each country whether they agreed that further increase or improvement in nuclear arms would yield no advantage for their country; 83 percent of Soviet and 69 percent of American respondents agreed. On the question of whether the total elimination of nuclear weapons should be possible, 71 percent of Soviet and 37 percent of American respondents answered in the affirmative. We need, however, much more information about the samples and methods
OVERVIEWS 19 of Soviet surveys before we can judge the comparability of the Arneri- can and Soviet results; in general, Soviet reports include little information concerning methodological problems. This kind of'information about values and attitudes held by the citizenry of the mro superpowers is an important addition to the knowledge base available to social scientists, and presumably also to political decision makers, who deal with interdependence and international reciprocity. We see an urgent need to expand public opinion research to obtain stratified national samples of both countries in order to systematically trace opinions and attitudes on crucial issues 'from year to year. Although the practical difficulties would be much greater, such research also needs to be extended to peoples in other parts of the world, particularly in developing countries. Research on public awareness of and involvement in issues need not be confined 'to opinion research. Another important approach is exemplified in the report by Sokolov 'and Abalkina [1989] on the development of awareness of ecological issues in the USSR. These investigators traced the development of such awareness from the time of the revolution (1917) to the present, in terms of political speeches, decrees, legislation in response to public concern and demands, and activities such as demonstrations and protests by groups of citizens. Their analysis begins with Lenin's decrees concerning the use of land, forests, and other resources in the USSR, then comments on the near absence of attempts to control industrial pollution between 1920 and 1940, the major period of increased industrialization in the USSR. The authors then examine instances of public outcry and O protest in the 1960s that stopped the construction of pulp and paper plants and hydroelectric plants that would have adversely affected the ecology of various regions. Between the 1960s and 1980, legislation on water and land use and on wildlife protection) proliferated, to be implemented in the 1977 constitution. During the current period of perestroika, ecological reform has been driven by the Chernobyl disaster and the attempts to divert the courses of Siberian rivers to meet the needs of industrialization. Recent speeches by Mikhail Gorbachev urge radical environmental reform and the initiation of large-scale projects to protect the environment. The results of this historical study, which needs to be paralleled in other countries, suggest that the aversion to nuclear arms buildup evidenced in public opinion polls may be simply one aspect of a widespread increase in public awareness of and concern about threats to the environment. MISSING ELEMENTS What was missing from our workshops? Research involving mathemat- ical and computer models or other formal methods had little representation. This deficiency is due not to any lack of up-to-date work on mathematical
20 SOVIET-AMERICAN DIALOGUE IN THE SOCIAL SCIENCES and computer modeling in the Soviet Union; indeed some of the Soviet workshop participants do such modeling. During visits to Soviet research institutes, members of the American delegation encountered substantial research concerned with computer modeling of large-scale ecological sys- tems, including treatments of ecological effects of nuclear explosions. In the Institute for Control Problems, they discovered a serious interest in formal choice theory and game theory; however, that interest does not yet seem to have been directed toward the prevention of nuclear conflict. In the area of political science, Soviet investigators at the workshops evinced interest in American research on the evolution of international cooperation and the modeling of cooperation and competition, but they reported no similar research efforts in the Soviet Union. In the area of cog- nitive science, the simulation models reported by Sergeev and his associates differ from American approaches in being based primarily on analyses of texts rather than causal or time-series modeling of actual events; that dif- ference reflects the Soviet preoccupation with semantic analysis. Although not represented in the workshops, recent work by these investigators is beginning to make use of connectionist networks, moving away from se- mantic analysis and toward analysis of actual phenomena of competition and cooperation. Some of the omissions are surprising. We thought that some Soviet social scientists would be doing experimental studies of communication, interaction, cooperation, and conflict, but we found no trace of counter- parts to American work along these lines. Similarly, formal modeling and measurement of international relations seem to have found less favor in the USSR than might have been expected. Historical and contemporary research on war and political conflict are so separate from one another in the Soviet Union that there seems to be almost no collaboration of the sort that has American historians and political scientists comparing models and descriptions of past and present wars:. Seen from the Soviet Union, American research no doubt has similar lacunae. RESUME Interdependence among nations can be studied at different levels; and to understand the research presented at these two workshops, we need to distinguish two principal approaches, which we term systemic and behavioral. Systemic denotes the study of the structures and processes characterizing political, economic, and military organizations; behavioral denotes the study of behavior and mental processes of individuals usually national leaders. Contrary to our expectations, we found that Soviet research presented in the workshops concentrated heavily on the behavioral approach. Outside the workshops, we have encountered more research dealing with large-scale
OVERVIEWS 21 systems. Systemic studies included formal modeling of control systems and collective choice, but with no application to problems of international interdependence or nuclear arms control. We found an interest in game theory, but the only related research reported was the innovative adaptation of game-theoretic methods by Sergeev et al. to the interpretation of the Caribbean crisis. We are aware of potentially relevant Soviet research on large economic systems and coalitions, and we hope that it will be represented in future workshops. The relatively extensive Soviet contributions in behavioral research included studies of public opinion and attitudes, but they focused more sharply on the mental processes of political and military leaders. One strand of this work deals with concepts and traces changes in the meaning of important concepts by means of both historical studies and semantic analysis of texts of speeches and other communications from national lead- ers. Another strand is based in cognitive science and seeks to construct accounts of the mental processes of leaders by means of semantic analysis and cognitive mapping. A short time ago it would not have seemed possi- ble that such research could be reported and discussed in an international setting. Now, however, the possibility is being realized. No doubt our explorations missed other areas of Soviet research that are becoming ripe for international cooperation. Vigorous pursuit of the kinds of research already reported, with mutual criticism, exchanges of source materials, and even collaboration between social scientists from the two countries (perhaps ultimately expanding to include others), might contribute to elim- ination of misconceptions and misperceptions that pose special hazards to international security in the nuclear age. REFERENCES Axelrod, R. 1g76 Structure of Decision. Pnnceton, NJ: Princeton University Press. Nash, J. 1951 Noncooperative games. Annals of Mathematics 54:239-295. van Neumann, J., and 0. Morgenstern 1953 Theory of Games and Economic Behavior. Princeton, NJ: Princeton University Press.
