EIGHT
Domestic Political Explanations in the Analysis of Foreign Policy
Joe D. Hagan, West Virginia University
Editors' Introduction
In this chapter Joe Hagan provides a comprehensive overview of the linkages between domestic political dynamics and foreign policy behavior. The reader will see that this chapter is considerably longer than other chapters in this book, a necessity given the varied literatures being brought together here. Hagan shares a similar focus with Haney in that both consider the role of institutions in foreign policy behavior. Hagan explicitly considers different institutional responses to domestic political opposition. For example, do authoritarian and democratic regimes respond differently to their opponents? What are the foreign policy implications of those differences? In this way Hagan links the governmental and societal levels of analysis. Hagan not only provides a thorough review of the literature on these questions, but also develops an analytical framework to account for domestic political influences on foreign policy behavior.
In reading this chapter, consider the widely varied influences of domestic political factors on foreign policy behavior. What contingencies are important in explaining these variations? Which domestic variables outweigh others when they compete? Hagan uses many examples to illustrate the effects of domestic political factors on foreign policy; can you think of others that illustrate the concepts introduced here? ■
At the heart of domestic political explanations of foreign policy is the idea that leaders engage in what Putnam (1988) calls "two-level games" or what Tsebelis (1990; also Starr 1991) refers to as "nested games." That is, in explaining government choices in foreign affairs, leaders are viewed as coping simultaneously with the pressures and constraints of their own domestic political systems as well as with those of the international environment. These games are not simply the decision-making dynamics discussed in previous chapters; of concern here is the broader array of autonomous political actors (e.g., factions, parties, and institutions) that influence the regime's daily governing authority and ultimately its long-term hold on office. This balancing
117
118        Second-Generation Foreign Policy Analysis
of domestic and international concerns in the foreign policy process is not an isolated, aberrant game. Thus, much theoretical research has long argued that the interna] political configurations of national governments broadly influence, or modify, their involvement in international affairs. The importance of this insight goes well beyond foreign policy analysis. The implication is that international politics is driven not solely by systemic structures (as posited by realism) but also by the domestic political patterns of at least the major powers.1
The purpose of this chapter is to explicate the general logic by which domestic political phenomena are linked to foreign policy. In doing so it draws upon diverse areas of research on this topic that typically have been treated separately. These various core concepts, explanations, and demonstrated foreign policy effects provide the basis for arguing that the linkage between domestic politics and foreign policy is far more complex than portrayed by researchers pursuing their own agendas and interests separately. As is discussed in the major portion of this chapter, the relationship is complex because leaders pursue dual domestic political games involving multiple arenas of opposition and, then, respond to that opposition with alternative strategies with divergent foreign policy effects. Furthermore, these games and strategies are pervasive across different types of political systems, although their impact on foreign policy is actually subtle in modifying how leaders respond to international pressures. The chapter closes with an overview of research strategies for incorporating various levels of opposition and contingencies into empirical, particularly cross-national, research on domestic politics and foreign policy.
■ Theoretical Research on Domestic Politics and Foreign Policy
Even though it is often assumed that foreign policy in a dangerous world is, or at least should be, above a nation's internal squabbles, the general significance of domestic politics is a major theme in a wide variety of literatures. These areas of research include U.S. foreign policy, comparative foreign policy analysis, foreign policy making in various specific non-U.S. settings, and recently the development of what Müller and Risse-Kappen (1993) call "complex models of international politics." One goal of this chapter is to bring together the insights of these largely separate areas of research. In doing so, however, it should be kept in mind that none of the four is entirely cohesive. Rather, within each there are separate—if not competing— strands of theoretical logic and research. Juxtaposition of research themes within these four bodies of research, as well as among them, provides the basis for explicating the theoretical logic linking domestic politics to foreign policy.
U.S. Foreign Policy
Political explanations have a long tradition in the study of U.S. foreign policy. The broad significance of domestic political constraints was acknowledged early on by prominent cold war realists such as George Kennan (1951) and Hans Morgenthau (1951), who worried that resurgent isolationism would undercut the country's commitment to an active global role in countering the communist threat.2 Although a "cold war consensus" giving the president a relatively free hand in countering
Domestic Political Explanations in the Analysis of Foreign Policy        119
communism emerged by the early 1950s, domestic politics remained a dominant concern in "national security" studies of the army-navy-air force rivalries over roles, missions, and budgets and related congressional-executive wrangling over military preparedness.3 Since the Vietnam War, however, domestic politics has been brought into U.S. foreign policy studies in more fundamental ways. First, with the collapse of the cold war consensus after Vietnam and the rise of global economic interdependence, interest was revived in constraints emanating from broader political arenas. Presidents were now viewed as ensnared in a web of constraints imposed by an institutionally jealous Congress, by elite and mass public opinion that was both divided and skeptical, and by a "weak" state apparatus unable to manage the emerging pressures of complex interdependence.4 Second, the theoretical rigor of the political logic within foreign policy analysis was enhanced by works on "bureaucratic politics" by Allison (1969, 1971) and Halperin (1974), as well as by their critiques and the development of alternative models of the decision-making process.5 Finally, primarily among historians such as John Lewis Gaddis (1972, 1982), there has emerged a large body of "postrevisionist" research arguing that domestic politics intensified—not diminished—the hard-line, militant cold war policies of the administrations from Truman to Reagan.6
Comparative Foreign Policy
Political explanations in this area have been emphasized since the field's inception in the mid-1960s with its call for, among other things, systematic and cross-national analyses of foreign policy. Political phenomena are a major component of Rosenau's (1966) long influential "pre-theory" framework. "Governmental" factors are one of his five general "sources" of foreign policy, and political system properties (accountability and development) are two of the three national characteristics determining the relative importance of the source variables. Conceptualizations of a wider variety of political phenomena are more fully specified in the other major theoretical frameworks proposed by Brecher, Steinberg, and Stein (1969), the Interstate Behavioral Analysis Project (Wilkenfeld et al. 1980), and the Comparative Research on the Events of Nations Project (East, Salmore, and Hermann 1978). Each of these frameworks in its own way focuses on three kinds of phenomena: political system structure (mainly "democratization"), various patterns of political opposition (e.g., regime constraints, linkage mechanisms), and certain shared attributes of the political leadership (e.g., shared images, elite profiles). However, empirical cross-national research has fallen far short of the ambitions of these theoretical frameworks. Most early cross-national studies were limited mainly to political system structure, comparing the foreign policy behavior of "open" and "closed" systems.7 Far fewer studies attempt to capture the more subtle domestic political phenomena of opposition patterns and leader attributes. The one major exception is the extensive cross-national research into the most extreme manifestation of domestic political instability in the form of mass unrest: the "linkage" between domestic conflict and foreign conflict.8 Only recent cross-national analyses examine the effects of more routine, organized political opposition within the "regime" as well as from other institutional arenas such as the military, legislature, regional actors, and ruling party factions.9
120        Second-Generation Foreign Policy Analysis
The Politics of Foreign Policy in Non-U.S. Settings
A third area of research consists largely of case studies by country specialists on foreign policy making in various non-U.S. settings. Although initially dismissed by comparative foreign policy researchers as atheoretical and unsystematic, by the 1980s this literature had become more analytic and now offers detailed insights into the precise nature of domestic politics and its influence on foreign policy.10 Emerging from this scattered literature is a common conclusion that fluid political constraints may or may not exist in any type of political system. Domestic political influences are central to analyses of the foreign policy of the former Soviet Union. Empirical "kremlinological" accounts of politics in the Khrushchev and Brezhnev regimes, and especially case studies of Soviet military intervention, carefully document the actors and debates within the Kremlin as well as their precise impact on foreign policy action." Political explanations also receive widespread attention in the Third World foreign policy literature, as an attempt to depart from idiosyncratic, "great man" explanations of foreign policies (Koraný 1986a) and to capture the emerging array of actors and institutions in many of the Third World's political systems.12 Finally, political influences are emphasized in studies of the Western democracies, most extensively in the case of Japan's factionalized Liberal Democratic Party as well as in coalition governments in a variety of European democracies.13 That democracies should face opposition is not surprising; what is interesting here is that the level of political constraints, as well as the magnitude of their foreign policy effects, varies substantially among democracies. In fact, taken together, these studies of authoritarian, Third World, and Western democratic systems suggest that it is difficult to generalize about any type of political system and suggest that political effects vary across different issues, situations, and leaders (Hagan 1993, chap. 2).
Domestic Politics and "Complex Models of International Politics"14
Most recently, there has been a surge of interest in domestic politics as a supplement to "systemic" explanations in the broader international relations literature—for example, neorealism (Waltz 1979), complex interdependence and cooperation (e.g., Keohane and Nye 1977; Keohane 1984), and global change (Gilpin 1981).15 Across these topics is a common conclusion: that nations with different domestic political arrangements respond to the constraints of international systemic arrangements with different foreign policy strategies. Research concerned with managing complex interdependence shows that different nations respond to common international economic crises in different ways, which can be traced to domestic coalitions, institutional arrangements, and particularly the strength of the state vis ä vis societal actors (e.g., Gourevitch 1986; Katzenstein 1976).16 Theoretical frameworks also emphasize domestic politics in explaining foreign policy "restructuring" as a response to global change, recognizing that political factors may constrain, prevent, change, or even facilitate foreign policy change (K. Holsti 1982; Goldmann 1988; Hermann 1990).17 It is, however, the research into the origins of war that has most extensively incorporated domestic political phenomena. Much of this literature examines the "war proneness" of democratic and authoritarian systems, arguing that democracies are less likely to initiate war and to fight each other because they select more moder -
Domestic Political Explanations in the Analysis of Foreign Policy        121
ate leaders and/or because their leaders face greater domestic constraints (e.g., Bueno de Mesquita and Lalman 1992; Doyle 1986; Lake 1992; Russett 1993b).18 Equally important, although seemingly inconsistent, is research explaining war proneness in terms of more fluid (i.e., nonstructural) political dynamics that may occur in any type of political regime. It shows that domestic political pressures for aggressive and militant foreign policies are found in both democratic and authoritarian systems (e.g., Lamborn 1991; Levy 1988, 1989; Rosecrance and Stein 1993; Snyder 1991). Authoritarian regimes are not the only ones capable of producing leaders who have a hard-line, bellicose orientation to world affairs (Vasquez 1992).19
■ The Logic of Domestic Political Explanations of Foreign Policy
The task of this section is to explicate how domestic politics influences foreign policy, and to do so in a way that is well grounded in the theoretical and empirical research. When one examines the literature mentioned in the previous section, it quicldy becomes apparent that this is not a simple task. There is no single political perspective used to explain foreign policy. Rather, explanations vary not only in terms of what aspect of foreign policy is to be explained but also with respect to the relevant political arenas and tasks, the alternative political strategies used by leaders to cope with domestic constraints, and the significance of political system structure on foreign policy making. The existing literature offers not a comprehensive theory but rather different pieces of the linkage between domestic politics and foreign policy. The intent here is not to argue in favor of one particular perspective. Rather, my strategy is to attempt a broad synthesis by treating the various themes in the literature as comp lem entary within a broader framework.
All that follows is rooted in the basic notion that foreign policy makers simultaneously cope with the pressures of domestic and international affairs (Putnam 1988; Tsebelis 1990). I elaborate upon the character of these "nested" and "two-level" games by arguing that they are inherently complex in the following ways. First, political leaders as foreign policy makers cope with dual domestic political imperatives involving opposition in multiple arenas. Second, the effect of domestic politics is contingent upon leaders' chokes of alternative political strategies that, in turn, have sharply divergent effects on foreign policy. Third, these dual games and alternative strategies are pervasive across the different types of political systems and are not limited to established democracies. Finally, these games and strategies have subtle effects on foreign policy; they condition how leaders respond to international pressures but, except in extreme circumstances, are not the sole, or even primary, determinant of foreign policy. This exercise provides the reader with an image of the overall logic of domestic political explanations of foreign policy. Its implication is that any genera] conception of that relationship must incorporate the complexity inherent in the games and strategies in various national settings.
Dual Domestic Political Games and Multiple Arenas of Opposition
As noted, domestic political explanations of foreign policy rest on the idea that decision makers must simultaneously contend with the pressures of international affairs
122        Second-Generation Foreign Policy Analysis
and domestic politics. What is striking is that different literatures characterize these games in different ways. At one extreme, domestic politics is pictured as the clash of particularistic interests within the well-structured institutional environment of national governments. Take as two examples the intense battles after World War II among the U.S. Navy, Army, and Air Force over defense budgets or certain weapons systems and similar struggles within the "iron triangle" of interest groups, congressional committees, and executive agencies over trade tariffs. At the other extreme are accounts of severe domestic crises in which national leaders aggressively manipulate foreign policy in order to save themselves from being overthrown by domestic opponents. As in the cases of Austria-Hungary, Germany, and Russia before World War II, leaders facing widespread unrest at home used foreign policy as a device for ensuring their personal survival and for preserving the domestic order. While other conceptions fall between these extremes, the point here is that there is not a single view of the domestic political setting as it influences foreign policy. Rather, leaders deal with not one, but two, domestic political games: building policy coalitions and retaining political power.
Building policy coalitions. This imperative concerns the leadership's need to build domestic support for any proposed policy initiative. At issue is the task of achieving agreement among at least a subset, or "coalition," of the various actors who formally or informally share the authority to commit the nation's resources and/or implement policy in a sustained manner. Although policymaking often occurs within a single group (e.g., a president and his or her advisers), two conditions may politicize an issue and move it into the broader political arena. The first of these is when decision-making authority is fragmented among autonomous and powerful actors, either because of a power-sharing arrangement between the executive and other institutions (e.g., an autonomous legislature or politicized military) or because of the division of executive authority among separate party factions or multiple parties. Either way, no single group controls the policy process, and a decision requires agreement among multiple actors. The second politicizing condition is the degree to which the actors involved are polarized over the issue itself. In other words, to what extent do the actors involved disagree with each other over the policy's merits? When policy authority is dispersed across politically powerful and contentious actors, the policymaking process becomes a task of building a coalition of supporters necessary to gain passage of an initiative. Foreign policy becomes a "political resultant" in the sense that the final decision outcome reflects the political strategies (discussed subsequently) necessary to build agreement to support implementation of the policy.20
Coalition building, not surprisingly, is pervasive in the open and pluralist systems of established Western democracies. Although extensive bureaucratic and interest-group activity is common to these complex systems, it differs with respect to the arrangements of national political authority in the American presidential system and the Western European and Japanese parliamentary systems. In the United States, coalition building on major issues ultimately revolves around relations between the Congress and the executive even though the "separation of powers" system provides Congress with the authority (and political autonomy) to veto or restrict major foreign commitments such as waging war, regulating trade, entering into treaties, and spending in foreign and defense policy. When Congress and the presidency are polarized
Domestic Political Explanations in the Analysis of Foreign Policy        123
over foreign policy, the impact can be dramatic, indeed, much of the pre-World War II "underreaction" in U.S. foreign policy can be traced to the failure of internationalist presidents (Wilson and Roosevelt) to gain the support of an isolationist Congress (e.g., Dallek 1977; Stein 1993). In parliamentary systems, in contrast, central authority is lodged entirely in a single institution—the cabinet.21 Where a single party controls the cabinet with a solid parliamentary majority, political constraints are quite minimal and limited largely to bureaucratic politics. However, parliamentary situations become quite problematic when the cabinet is itself politically fragmented by the well-established factions of a single ruling party (as in Japan under the Liberal Democratic Party) and/or by multiple political parties in a coalition cabinet, none of which commands a majority in parliament (e.g., Fourth Republic France, Italian coalitions since World War II, and Israel under Labour and Likud). The deadlocks underlying Fourth Republic France "immobilism" on the Algerian War (Andrews 1962), Japan's reactive posture in foreign policy negotiations (Destler, Fukui, and Sato 1979; Hellmann 1969), and Israel's difficulty in dealing with the occupied territories (Yaniv and Yishai 1981) illustrate the inability of cabinets to agree on even the most pressing policy matters.22
Coalition building is not limited to established democracies. Although power in authoritarian systems is conventionally assumed to be controlled by a cohesive political elite, historical and area study research indicates that this is not always the case. Authoritarian regimes such as Nazi Germany and Stalinist Russia had minimal domestic constraints, but others clearly fell short of the monolithic character of "totalitarian" systems (Friedrich and Brzezinski 1956). Political authority in the former Soviet Union, although concentrated in the Kremlin, was significantly dispersed across contending political actors in the two intertwined tiers of that Communist oligarchy: the Central Committee and the Politburo (Roeder 3 988). At one level were entrenched bureaucratic and institutional interests represented in the Central Committee, which were polarized over foreign policy into hard-line "security-producer-ideological" and more moderate "consumer-agriculture-public service" groupings (Aspaturian 1966). At another level, ultimate political authority rested in the Politburo, whose authority was, to varying degrees, dispersed among its fifteen or so members that constituted the body. Under both Khrushchev and Brezhnev, achieving a consensus across the tiers of policymaking posed broad constraints on Soviet foreign policy.23 Other prominent twentieth-century authoritarian regimes— Wilhelmine Germany prior to World War 1 and militarist Japan in the 1930s—were even more extremely fragmented, or "cartelized," with authority dispersed across separate institutions and groups. Kaiser William II was but one player in an authoritarian coalition that also included the military, agricultural elites, heavy industry, and democratic elements in the German parliament (Reichstag).24 Similarly, in 1930s Japan, real control over foreign policy was spread among the imperial court, the army, and the navy, while parties in the Diet were suppressed.25 In both regimes, despite a hard-line, militant consensus on foreign affairs among these actors, there was still significant debate about the pace and direction of future military expansion (Snyder 1991).
Coalition building is often neglected in analyses of Third World foreign policies, mainly because established institutions and powerful bureaucratic, political,
124        Second-Generation Foreign Policy Analysis
and social groups are assumed to be absent. Although it recognizes many cases where authority is concentrated in the hands of a single "predominant leader," accumulating research points to two general exceptions where power is quite dispersed.26 First, after several decades of independence, many Third World governments are quite institutionally complex, with entrenched bureaucratic and party interests. This is the case in most larger countries in the Middle East (Koraný and Dessouki 1984), Latin America (Lincoln and Ferris 1984), and East Asia, including China (e.g., Barnett 1985; Vertzberger 1984b). Political authority in the Third World is also dispersed in ways that range from well-established collective rule in the royal family in Saudi Arabia (Quandt 1981) to fragmented and unstable regimes in South Korea (Kim 1971), pre-1967 Syria (Bar-Siman-Tov 1983), and Argentina (Levy and Vakili 1990). The second set of exceptions are associated with regimes that appear to be dominated by a single predominant leader but are actually racked by political infighting among contending factions. In these cases the leader is unable or unwilling to act on foreign policy issues until a consensus forms among the actors. Well-documented examples of this situation are the intensely confrontational foreign policies of China under Mao Zedong during the Cultural Revolution (Hinton 1972), Indonesia under Sukarno (Weinstein 1976), and Iran under Khomeini (Stempel 1981). Even though Mao, Sukarno, and Khomeini were uniquely prominent as revolutionary leaders, closer research demonstrates that none of these leaders was politically willing or able to take complete control of foreign policy and, specifically, resolve conflicts among moderate and hard-line leadership factions. Despite each regime's constant flow of intense anti-Western rhetoric in these cases, none was able to take substantively meaningful action to resolve international crises; major foreign policy actions were taken only after power was stabilized and consolidated by one of the government factions.
Retaining political power. Whereas coalition building concerns authority over an immediate policy issue, the longer-term political survival of the ruling group is at the heart of the political imperative of retaining power. One of these goals is political survival. Foreign policy decision makers, acting in their concurrent role as national and political leaders, simultaneously work to maintain and enhance the political base necessary for staying in office. Therefore, when domestic political pressures threaten to evict the leadership from power, foreign policy must be adjusted so that it imposes fewer domestic costs. Certainly not all foreign policy issues become politicized in this manner, but two dynamics combine to suggest that policymakers are likely to be sensitive to the long-term domestic consequences of their foreign policy decisions. One is simply the occurrence of "significant" opposition in the wider environment, that is, competing political groups whose opposition extends beyond specific policy disagreements to a direct challenge for government control. This is most significant when these groups have (or are acquiring) sufficient political resources that indicate they may succeed in the not-too-distant future. The other is a mounting public perception of foreign policy issues as being linked to the overall credibility of the current leadership. For example, confrontation and accommodation with foreign adversaries, respectively, are easily perceived by the public as indicating the government's overall willingness to risk war or its weakness in world affairs. The pressures of political survival are most dramatically illustrated by the crisis leading to World War I, in which
Domestic Political Explanations in the Analysis of Foreign Policy        125
the highly vulnerable leaders in most countries feared that not going to war would undercut their nationalist credentials and show them to be at their weakest.27
The theme of political survival, though not as prominent as coalition policymaking, is found in the various strands of literature on U.S. foreign policy, including core theoretical concepts such as Halperin's (1974) "presidential interest," Mueller's (1971) "presidential popularity," and Hampson's (1988) "divided decision maker." Particularly with the rise of a divided and skeptical public after the failed Vietnam policy, all U.S. presidents appear to have been closely attentive to shifting public opinion on the two broad—and seemingly contradictory—concerns of (1) risking involvement in another war and (2) projecting international weakness suggestive of a decline in U.S. world status (Schneider 1983). Analyses of the logic of Nixon/Ford, Carter, and Reagan/Bush foreign policies explain their alternative "strategies of containment" partly in terms of their alternative ways of managing the media and public opinion. Nor are such interpretations limited to the post-Vietnam period. Presidential survival is a core political theme in postrevisionist analyses of the "origins of the cold war," beginning with the argument that Roosevelt and Truman tailored the emerging cold war confrontation to domestic audiences (Gaddis 1972; Yergin 1977). Even at the height of the cold war, when there was a very wide reservoir of public support for containment, presidents still feared public opinion. For example, research of major episodes such as the Cuban missile crisis (Hampson 1988) and the Vietnam War (Berman 1982; Gelb and Betts 1979) now suggests that Presidents Kennedy and Johnson were preoccupied with the memory of how the "loss of China" destroyed the credibility of the Truman administration and feared conciliation with the communists would lead to the collapse of their own administrations.
Retaining political power is, of course, an important political dynamic in the foreign policies of parliamentary democracies, but as with coalition policymaking the situation reflects their particular institutional arrangements (Waltz 1967). Whereas U.S. presidents are directly elected, prime ministers and the cabinet are selected more indirectly via elections in the Parliament. Although a disciplined and cohesive ruling party can insulate a prime minister from public opinion, a fragmented cabinet severely constrains his or her ability to conduct foreign policy because the defection of any one faction or party can precipitate the government's collapse. If cabinet factions or coalitions are intensely competitive, foreign policy issues tend to be viewed less on their substantive merits than according to their perceived effect on political balances in the cabinet. Indeed, some would argue that the immobilism in the foreign policies of Japan's Liberal Democratic Party (Destier, Fukui, and Sato 1976; Fukui 1977a) or the coalitions of France's Fourth Republic (Andrews 1962) stemmed mainly from the intensely competitive "ministerial merry-go-round" as well as fears of cabinet members' defection and ensuing government collapse.
Like coalition building, the unelected leaders in authoritarian regimes would seem less affected by the task of political survival. Various research suggests that this is not necessarily the case. As portrayed in the historical and war-proneness literature, the leaders of authoritarian political systems in Germany, Russia, and Austria-Hungary were very sensitive to broad domestic political crises (Kennedy
126        Second-Generation Foreign Policy Analysis
1980; Lamborn 1991; Snyder 1991). During the crisis of 1914, these leaders were probably more preoccupied with the domestic political consequences of their decisions than with the international repercussions (e.g., Lebow 1981). Even in the post-Stalinist Soviet Union, no leader could ignore the task of maintaining his nationalist and ideological credentials within the Party and even the broader political system. Constant uncertainty over power sharing in the Politburo meant that any leader could "suddenly find himself removed from office by a political conspiracy of his former political associates" (Schwartz 1975, 177). Not only did this happen to Khrushchev in October 1964, but kremlinological accounts of bis leadership as well as those of Brezhnev and Gorbachev reveal a constant game in which each leader sought to consolidate and retain power (Gelman 1984; Linden 1978). Indeed, Valenta (1979) asserts that throughout the 1968 Czechoslovak crisis, Brezhnev remained uncommitted to any particular option and came to support military intervention only to avoid the consequences of being in the Politburo minority on that politically critical issue.
Political survival is a more dominant theme than coalition building in the literature on the politics of foreign policy in Third World countries (e.g., David 1991; Good 1962; Weinstein 1972). A distinctive feature of Third World foreign policy making is the extreme political instability of many regimes, reflective of a persistent crisis of domestic legitimacy and the lack of an institutionalized elite political process. Thus, political survival is pressing because of the extreme distrust among contending players. For example, most accounts of the U.S.-Iranian hostage crisis contend that the policy in Iran was driven mainly by the "fight-to-the-death" attitudes of factions in the revolutionary coalition (Stempel 1981).28 Even in regimes with an unchallenged predominant leader like Cuba's Castro, Iraq's Hussein, or Syria's Assad, leaders still engage in a continual game of maintaining that centralized power through dealings with support groups as well as the broader public (Hermann 1982; Lawson 1984). It would be a mistake, though, to overgeneralize about Third World political instability. In more established Third World systems, leadership politics largely parallels the processes found in both authoritarian states and advanced industrial democracies. Conflicts among Communist Party factions in Chinese foreign policy since the Cultural Revolution are not terribly different from those of the former Soviet Union (e.g., Barnett 1985). Similarly, competition for power over India's foreign policy does not diverge fundamentally from that in a number of European democracies (Vertzberger 1984b).
Recognition of the twin political games of building policy coalitions and retaining power is important for several reasons. First, both reinforce the point that domestic politics is a widespread and nonaberrant influence on foreign policy. Even if a government is either internally very cohesive or entirely secure in power, it is logically rare that a government would be entirely free of both. Second, recognition of these dual political games suggests that conceptualizations of domestic opposition need to be cast in a way that taps actors in multiple arenas of the political system (Hagan 1993). A broad assessment of domestic political influences must consider the constraints of both divisions within the leadership (and its support groups) and those of groups and movements in the wider political environment that threaten its
Domestic Political Explanations in the Analysis of Foreign Policy        127
hold on power. Third, and perhaps least obvious, these dual imperatives provide insight into the intensity of constraints faced by foreign policy makers. At one level the constraints of foreign policy choices are particularly severe when strong pressures associated with both games come together and reinforce each other.29 Leaders confronting losses in both games may be unable to act on the issue, even when facing unrelenting international pressures. Post-Vietnam constraints in U.S. foreign policy often bring together the combined pressures of public opinion and congressional opposition. Similarly, extreme deadlock in fragmented parliamentary and authoritarian coalitions often is due to the combined impact of political competition and policy disagreements. However, as shown by Snyder (1991), the strong interaction of these games does not always lead to an inability to act. The combination of coalition building and fears of losing power create "logrolling" dynamics that, in turn, produce an intensely aggressive foreign policy that satisfies the interests of its component groups and their key support groups. It also permits leaders to manipulate foreign policy to fend off outside challenges to their hold on power. These dynamics occurred across a wide range of political settings, including Victorian England, Wilheimine Germany, 1930s militarist Japan, and both superpowers during much of the cold war, and in all cases contributed to foreign policy aggressiveness and ultimately an overextension of national power.30
Alternative Political Strategies and Their Divergent Foreign Policy Effects
This section concerns the dynamics by which the games of building policy coalitions and retaining political power influence foreign policy. As with the games themselves, the literature on domestic politics and these foreign policy dynamics is not of a single mind. A substantial body of literature views domestic politics as undercutting a government's ability to take meaningful action in international affairs. Classic illustrations of this are the political constraints on British, French, and U.S. foreign policy in dealing with Fascist aggression in the 1930s. Leaders in these systems had considerable political incentive to avoid another world war. Even when leaders saw war as inevitable, as did U.S. President Roosevelt by 1940, they were precluded from taking forceful action by opposing elements within the government—the isolationist Congress, in FDR's case. Yet, equally often, other literature views domestic politics as propelling governments toward sharply intensified foreign policy confrontations and commitments. Indeed, in the twentieth-century's other world war, domestic politics is widely argued to have contributed to the major powers' failure to stop the drift to war by 1914. By then the governments of almost all the major powers, except Great Britain, faced mounting political crises tied to fragmenting governing authority and the legitimacy of the domestic political order. For the leaders of Germany, France, Russia, and Austria-Hungary going to war was clearly an attractive means for shoring up the domestic political position of each. In fact, these vulnerable leaders feared that backing down in July 1914 would bring about the collapse of their governments, if not the existing domestic order.
The significance of these alternative "pull" and "push" dynamics is that leaders may respond to domestic opposition (at any level) in fundamentally different ways, each of which has sharply different effects on foreign policy. In other words, there is
128        Second-Generation Foreign Policy Analysis
no single dynamic by which political opposition affects foreign policy. Instead, leaders use alternative political strategies to respond to opposition in the foreign policy making process. Literatures treating this issue, if taken collectively, suggest the causal linkage between domestic politics and foreign policy reflects one of three general political strategies: "bargaining and controversy avoidance," "legitimization of the regime and its policies," or "insulating" foreign policy from domestic political pressures. As is argued in the following, these alternative responses are key to linking domestic politics to foreign policy.
Accommodation: bargaining and controversy avoidance. In accommodation, decision makers respond to opposition with restraint in foreign policy, as democratic leaders did prior to World War II. In the game of building policy coalitions, accommodation involves bargaining among the players necessary for the ratification and sustained implementation of the decision (Putnam 1988).31 Compromise among the players is the typical "resultant" of the bargaining process as decision outcomes reflect a choice somewhere between the preferences of the original contenders on the policy (Allison 1971).32 Where the opposition challenges the leaders' hold on power, accommodation centers on leaders' avoidance of domestically controversial actions that could discredit the overall leadership or upset internal balances within the ruling party (Salmore and Salmore 1978). Leaders seek to contain opposition, and thus retain political power, by avoiding publicly disputed policies and actions that make the country appear weak in international affairs or are closely associated with a widely acknowledged adversary. Whether in response to divisions within the government or opposition in the broader environment, the resulting foreign policy actions are similar. Politics constrains initiatives that are strong in intensity and commitment; in other words, the government engages in low-risk behavior and often avoids changes in policy. In the most extreme scenario, the leadership is deadlocked and unable to take any sort of meaningful action on a foreign policy issue, as occurred between FDR and the Congress right up to Japan's attack on Pearl Harbor.
Accommodation through bargaining and controversy avoidance is a pervasive phenomenon across various political settings but is most typically ascribed to democratic politics.33 Cold war realists were interested in domestic politics largely because they feared democratic leaders would, as they did in the 1930s, have to accommodate resurgent isolationist or pacifist opinion in a way that would undercut the containment of communist power (Kennan 1951; Morgenthau 1951). Post-Vietnam public opinion research indicates that declining public support does lead U.S. presidents to avoid the use of force (e.g., Mueller 1971) and that polarized opinion precludes stable public support over the long term (e.g., Holsti and Rosenau 1984; Schneider 1983). Even some postrevisionists (e.g., Gaddis 1982), who usually focus on domestic pressures favoring hard-line policies, are careful to acknowledge that certain postwar administrations responded to domestic politics with a "strategy of containment" that, for example, reduced military spending by emphasizing cheaper forms of statecraft (Eisenhower's "New Look") and decreased use of U.S. forces abroad (the "Nixon Doctrine"). Notions of bargaining and controversy avoidance are most often the dominant theme in political analyses of foreign policies of parliamentary democracies with coalitions and/or factionalized ruling parties. Cases of coalition deadlock, such as the inability of Fourth Republic France to extricate itself from Algeria
Domestic Political Explanations in the Analysis of Foreign Policy        129
(Andrews 1962), West German hesitation on early detente initiatives (Hanrieder 1970), and the Netherlands' refusal to deploy cruise missiles (Everts 1985), are traced to very broad compromises within the cabinet and fears that any controversial policy could provoke parliamentary defections. Comparable constraints are found in the dynamics of factional politics in Japanese foreign policy, although these processes are reinforced by strong norms of consensus building in that country's culture (Destler et al. 1976; Fukui 1978). Finally, these dynamics extend to models of foreign policy making outside the senior political leadership and are emphasized for countries with less politicized settings centering around bureaucratic and interest-group politics, such as Great Britain (Wallace 1976; Smith, Smith, and White 1988) and the Scandinavian countries (Sundelius 1982).
Accommodation is not limited to Western democratic politics. Discussions of bargaining and controversy avoidance, albeit within centralized and/or fluid settings, can be found in the theoretical and case study research on the foreign policy of the former Soviet Union and various Third World countries. Analyses of Soviet foreign policy on certain issues or episodes suggest that accommodation was often a central dynamic, particularly on issues provoking the clash of moderate and hard-line institutional and Party actors in the Central Committee (e.g., Aspaturian 1966). Politburo decision making often took the form of political accommodation, most notably the hesitation and delay in that body's reactive and vacillating handling of the 1968 Czechoslovak crisis (Valenta 1979). In fact, one could argue that bargaining and compromise prevailed during the Brezhnev era in Soviet foreign policy, due to the well-institutionalized character of leadership politics (Ross 1980) and Brezhnev's "consensual" style of decision making (Breslauer 1982). Among Third World analyses, accommodation is often similarly downplayed, but there are at least two exceptional situations. The first is literature that points to the institutional complexity of certain well-established political systems, such as the People's Republic of China (Barnett 1985), India (Vertzberger 1984b), and many in Latin America (Lincoln and Ferris 1984). The other Third World exception concerns situations of considerable political instability, which usually lead to legitimization strategies (as noted in the next section), and in which leaders are unable to manipulate the issue aggressively. This is an understudied phenomenon, although Kim's (1971) examination of South Korea's restrained negotiations on the post-World War II peace treaty with Japan is an excellent illustration of a leader's having to accommodate domestic opposition when acting on a volatile foreign policy issue.
Mobilization: legitimization of the regime and its policies. Under mobilization, leaders confront the opposition by asserting their own legitimacy, a strategy taken to the extreme by governments in the July 1914 crisis. Although hoping to deter that opposition, the intended effect is to mobilize new support for the regime and its policies—or, at least, prevent the defection of supporters. This strategy is most often associated with the game of retaining power in which a leadership manipulates foreign policy issues. Leaders seek to enhance the domestic political position of the regime in a number of ways: (1) appealing to nationalism and imperialist themes, or "scapegoating" or "bashing" foreign elements; (2) showing that the leaders have a special capacity and wisdom for maintaining the nation's security and international status; and/or (3) diverting attention away from divisive domestic problems (Levy
130        Second-Generation Foreign Policy Analysis
1988, 1989; Russett 1990; Snyder 1991). Political legitimization is also a strategy in building policy coalitions. Here, instead of accommodating the positions of adversaries, leaders build coalitions by aggressively selling their policy, often to audiences outside the regime, and thereby increase support for their initiative while discrediting their opponents (George 1980a; Trout 1975). Whatever the political imperative, the effect on foreign policy can be significant: strong, forceful (and typically conflictual) foreign policy actions that cast in the clearest possible light the position and ability of the political leadership. Although often limited to political theater (e.g., threats and promises to act), in the most extreme situations legitimization can drive a nation to commit itself to the use of force or to going to war, as occurred on the road to World Warl.
If coalition building is mostly associated with democratic politics, then mobilization via regime and policy legitimization is often linked to authoritarian and especially Third World political systems. In ways that can be traced mainly to the lack of political institutionalization, politics via legitimization strategies is widely emphasized in literature on African, Asian, and especially Middle Eastern foreign policies (e.g., Dawisha 1990). An atmosphere of political instability usually precludes bargaining with opposition that is to be distrusted and unlikely to be accommodated though mutual compromise. Foreign policy is a correspondingly viable means for unifying the public and discrediting domestic adversaries. Although other issues might be divisive, assertions of the nation's independence and status (as well as threats to it) in world affairs play upon one of the few issues that a government can control and about which there is a domestic consensus (Weinstein 1972). Even in highly fragmented regimes, the result can be an extremely hostile and active pattern of foreign policy.34 Among the more established authoritarian systems, legitimization in the foreign policy process is also a central theme. The political pressures for going to war in Germany, Russia, and Austria-Hungary all reflect the habits of leaders there to manipulate foreign threats and nationalism to channel growing mass participation (Levy 1988, 1989; Snyder 1991). Legitimization dynamics were also prevalent in the former Soviet Union, particularly as a pressure for hard-line, anti-Western policies dating back to Stalin and decisions to intervene in the Third World. In part this reflects the personality of certain leaders. Following Breslauer (1982), political legitimization is the primary strategy for "confrontational" leaders such as Khrushchev (and Yeltsin), whereas accommodation is preferred by the more "consensual" leaders like Brezhnev (and Gorbachev). Yet decisions for intervention in Africa (Valenta 1980b) and the Middle East (Spechler 1987) under the cautious Brezhnev show that he could lean toward assertive action when it provided a means for demonstrating his ideological credentials to critics of detente and other moderate policies toward the West (Gelman 1984).
Though not as widely emphasized as the strategy of accommodation in the U.S. foreign policy literature, political mobilization is a central theme in "postrevi-sionist" research into the political roots of hard-line, militant cold war policies (e.g., Paterson 1988; Dallek 1983). Trout (1975; Lowi 1967) argues that cold war presidents, including Truman at the advent of the cold war, were predisposed toward more simple and hard-line anti-communist policies because they were attractive to the voting public. Similarly, it is generally accepted that two decades of U.S. interven-
Domestic Political Explanations in the Analysis of Foreign Policy        131
tion in Vietnam was driven, in part, by the desire of successive presidents to demonstrate their foreign policy leadership skills and anti-communist credentials (e.g., Berman 1982; Gelb with Betts 1979). That this represents a broader pattern has been supported by recent empirical studies'which show that U.S. presidents are generally more likely to use military force in international crises when they are politically vulnerable, particularly if their approval ratings are declining among their supporters (Ostrom and Job 1986; James and Oneal 1991; Morgan and Bickers 1992). But this is not limited to the United States. Fifth Republic French foreign policy starting with President de Gaulle is a classic case of an active and independent (from NATO) foreign policy designed to unify an otherwise divided public around a new domestic order and policy agenda (Morse 1973). Similarly, one analysis of British foreign policy argues that its military response to the Argentine invasion of the Falkland Islands was motivated, in part, by Prime Minister Thatcher's worries about low opinion ratings (Farrands 1988). Finally, legitimization strategies are not entirely absent from highly constrained settings such as Japan under the Liberal Democratic Party. There is evidence suggesting that "in some cases, [factional politics] can bring new policy initiatives and commitments to the surface" (Destier et al. 1976, 24), namely, when a prime minister believes that pushing a particular issue will enhance his or her faction's political standing.
Insulation: deflecting, suppressing, and overriding opposition. While both of the preceding strategies indicate a strong (though divergent) domestic political impact on foreign policy, it is important to keep in mind another political dynamic: the ability of leaders to insulate a foreign policy issue from domestic politics, even in a situation where significant opposition exists. The essence of this dynamic is that political leaders deflect or reduce domestic constraints on their foreign policy choices. Containing political pressures may involve a number of actions: ignoring opposition challenges, suppressing opponents entirely, or co-opting them with political favors or concessions on other policy issues. Alternatively, when leaders have strong preferences on the foreign policy issue, they are willing to override even strong opponents and hope that domestic controversies may be restricted to acceptable levels.35 Indeed, if the two world wars correspond to these strategies, the cold war can be treated as a parallel case of avoiding war, partly because Soviet and U.S. leaders were able to keep hard-line and nationalistic domestic political pressures under control.36
Much of the U.S. foreign policy literature indicates that leaders can avoid domestic constraints. Some of these stem from critiques of the bureaucratic politics approach by pointing out when political divisions or leaders' sensitivity to them varies. Some of these studies suggest that conflict among contending political players might be contained by the nature of the situation. In crisis situations with immediate, visible threats, political constraints are contained by the contraction of government authority as well as the pressure for leaders to achieve consensus (Paige 1968; Hermann 1972). Other critiques stress that presidents are not passive, constrained actors in the political process but instead can assert their influence to overcome bureaucratic constraints (Perlmutter 1974; Krasner 1971). Analysts argue that a president can set up decision-making groups in a way that permits dissent, or "multiple advocacy," and allows the president to retain autonomy in making the final choice (George 1980a; Hermann 1993). Another strategy is engaging in trade-offs (or
132        Second-Generation Foreign Policy Analysis
payoffs) across separate issues, in which the president and other executive actors gain bureaucratic or Congressional support by making concessions on separate issues, as, for example, in gaining support for arms control agreements (e.g., Johansen 1980) as well as the recent North American Free Trade Agreement. Whatever the strategy, it is important to keep in mind that, even in the face of strong opposition, leaders may insulate foreign policy from domestic pressures.
Non-U.S. cases also provide insight into conditions in which leaders are able to insulate foreign policy from domestic opposition. The research on parliamentary democracies shows that even a highly fragmented cabinet sometimes can get around internal constraints. In analyses of economic issues such as trade negotiations and exchange rate adjustments, Fukui (1978; n.d.) points out that Japan's Liberal Democratic Party factional leaders are willing to insulate certain issues from factional politics.37 Studies of coalition decision making in Western Europe go even further; although there are numerous cases of these divided cabinets, there also are numerous examples of coalition governments whose members were able to reach meaningful agreements for foreign policy action. Particularly notable here are analyses of the highly "consensual" style of coalition decision making in the Scandinavian countries (Goídmann, Berglund, and Sjostedt 1986; Sundelius n.d.). Comparative studies of Dutch foreign policy show that the deadlock on NATO cruise missiles was not typical of the coalition government's actions on a variety of other issues (see Everts 1985). Additional episodes of Soviet military intervention demonstrate that, at times, Politburo members shared a strong consensus and did not develop policy within a context of political competition. In his analysis of the 1979 invasion of Afghanistan, for example, Valenta (1980b) found few political pressures in the Brezhnev Politburo that he observed in the Czechoslovak and Angola interventions.
These three strategies have considerable implications for research linking domestic politics to foreign policy. Instead of emphasizing one dynamic or the other (as is done in most of the literature), a general theoretical treatment of domestic politics and foreign policy must recognize the possibility of both the "push" and the "pull" of domestic opposition, as well as the fact that domestic pressure might be effectively contained (Hagan 1993). There is a need to recognize that leaders have choices in dealing with political opposition and that the strategies they employ ultimately tell us much about the effect of opposition on foreign policy. In other words, the linkage between opposition and foreign policy is an inherently contingent relationship. Ascertaining the foreign policy effects of opposition thus involves a sequence of two general questions: (1) Does opposition have an impact on foreign policy behavior? (2) If so, how and reflective of which strategy (accommodation or mobilization)?
The Subtle Effects of Domestic Politics on Foreign Policy
This chapter has argued that foreign policy making is an inherently political process, but it is equally important to recognize that domestic political effects on foreign policy are typically rather subtle. These domestic political games and strategies do not alone drive foreign policies but instead usually act as a supplementary influence on how leaders cope with more fundamental constraints from the international environment. A main contribution of research on "complex models of international politics" is the placement of domestic political influences into the context of pres-
Domestic Political Explanations in the Analysis of Foreign Policy        133
sures determined largely by the nation's position in the international system. For example, as Snyder (1991) carefully details, realist analyses are correct in stating that Wilhelmine Germany did face a real security problem of encirclement by powerful adversaries and that 1930s Japan was indeed seriously dependent on foreign sources of natural resources critical to its economy. Domestic politics did not lead leaders to imagine these situations but instead contributed to these leaders' inclinations to overreact to foreign threats and overextend their nation's power. Similarly, the logic of Putnam's (1988) "two-level bargaining" is not that domestic politics defines the economic interests but that domestic constraints affect the extent to which a government can respond to the constraints (and opportunities) of complex interdependence. Change in world politics is dictated mainly by the differential rates of growth in military and economic power (Gilpin 1981; Kennedy 1980). The contribution of foreign policy analyses is to show that domestic politics affects the ability of governments to adapt to change (Skidmore 1994) and the ways they respond to challenges of rising powers (Schweller 1992).
The subtle role of domestic political influences is further illustrated by case analyses of foreign policy decision making. Some of the key studies discussed above are insightful in placing political considerations (debates, public image, etc.) into the broader context of the leadership's consensus about the overall orientation of the nation's foreign relations. Political analyses of the U.S. cold war policies do not dispute an exceptionally strong national consensus on the militant containment of anti-Communism; political pressures instead involved questions of competing military strategies (Gaddis 1982; Gelb and Betts 1979) and the geographic range U.S. commitments (Snyder 1991). The Japanese Liberal Democratic Party factional disputes over how to respond to U.S. economic pressure did not question the party's consensus favoring close U.S.-Japan ties, but rather raised the issue of whether or not concessions should be made to the United States on that issue (Destler, Fukui, and Sato 1979). Ail Soviet leaders in 1968 saw the need to curtail Czechoslovak reforms; the Brezhnev Politburo debated the narrower question of how to pressure the Dubček government, that is, through military force or by economic and political sanctions (Valenta 1979). Even in intensely fragmented Third World regimes, as in Iran during Khomeini's early days and Indonesia prior to Sukarno's overthrow, there was broad agreement that the superpowers were the threat; debates concerned how much dependence or contact with the West could be tolerated (Stempel 1981; Weinstein 1976). The collective insight here is that the magnitude of political influences is limited, even from a political decision-making perspective. The leadership's core shared beliefs and interests are the primary motivational basis of the overall direction of foreign policy. Domestic political debates and grandstanding typically involve the relatively narrow matters of how policy is to be implemented.
Seen in this context, one should not expect domestic political explanations to account for the general foreign policy goals and the identification of allies and adversaries. Rather, the effects of the domestic political games and strategics discussed throughout this chapter are tied to two more specific dimensions of foreign policy: commitment and risk taking. Commitment in foreign policy concerns the allocation of resources internationally or the entering into foreign agreements with other actors—or promises or threats to do either.38 The foreign policies of the United States
134        Second-Generation Foreign Policy Analysis
and Japan in the twentieth century illustrate the political bases of commitment. In the ! 930s isolationism led to the "underextention" of U.S. power in countering recognized threats to interests in East Asia and Europe (Stein 1993). In Japan, army-navy logrolling propelled the expansionist leadership to overextend the country's resources in military actions in China and the Pacific Rim (Snyder 1991). The situation after World War II and throughout the cold war was the reverse: U.S. domestic politics ultimately contributed to an overextension of its power to Vietnam (e.g., Snyder 1991; Paterson 1988), while factional politics within Japan's ruling Liberal Democratic Party created the basis for an unusually reactive and passive foreign policy for a major power (e.g., Calder 1988; Hellmann 1969). The precise effect of domestic politics in these cases centered around the range of international commitments taken.
Risk taking concerns the propensity to initiate action that escalates a confrontation and invites costly foreign retaliation.39 Analyses cited throughout this chapter consistently suggest that domestic politics affects the likelihood that leaders will use military force in ongoing confrontations. Debate over the level of necessary (and acceptable) risk is at the core of differences between "moderates" and "hard-liners," who must be accommodated in order to authorize the use of force. As illustrated by Soviet military interventions (e.g., Valenta 1979; Anderson 1982), Politburo debates led to compromises invoking economic and diplomatic pressures but not military force (as originally tried in Czechoslovakia) or resulted in decisions not to intervene at all (as in Poland in 1980). Similarly, unstable multiparty coalitions in parliamentary systems avoid risky behavior for fear of parliamentary defections leading to government's collapse. Part of the deadlock in Fourth Republic France's Algerian policy was lhat any intensification of its war effort would bring about the collapse of the government. Risk-taking propensities are also a primary manifestation of domestic political pressures, which leaders seek to contain by legitimizing the regime and its policies. In the July 1914 crisis the failure of major power leaders to resolve the crisis diplomatically was partially due to their fear that backing down in the crisis would undercut their governments' legitimacy. Expectation of domestic unrest was one factor predisposing German, French, Russian, and Austrian leaders toward finally going to war (Lebow 1981; Snyder 1991). More recently, the escalation of the long Argentine-British dispute over the Falkland Islands can be traced to the fact that both the Galtieri and Thatcher governments expected that a military victory could reverse their political decline at home (Levy and Vakili 1990; Farrands 1988). In these cases, domestic politics did not create the problem or crisis but did contribute to the regimes' propensity to escalate existing tensions through the use of military force and thus risk war.
The theoretical implication here is that the foreign policy effects of domestic political processes should to be viewed in the context of broader international dynamics. The intrinsic significance of domestic political processes is that they "diminish" or "amplify" the propensity to incur commitments and take risks internationally. Where there are strong, sustained pressures for domestic political accommodation, a government is unlikely to respond to systemic dictates. Leaders then reduce commitments and underreact to foreign threats. In contrast, efforts at political legitimization by mobilizing support amplify a leader's predisposition to act. If done in a sustained manner, the result is an overreaction to foreign threats and/or
Domestic Political Explanations in the Analysis of Foreign Policy        135
an overextension of the nation's capabilities. In sum, while not determining patterns of conflicts and issues in international affairs, domestic politics is an important influence determining how leaders manage those disputes.
■ Approaches Accounting for Domestic Political Impacts on Foreign Policy
This chapter's purpose is to explicate the core theoretical logic by which domestic politics can be linked to foreign policy. Drawing upon various areas of research on the topic, my argument is that this linkage is inherently complex, with dual domestic political imperatives and contingent upon leaders' choices of alternative political strategies with divergent foreign policy outcomes. Although pervasive across different types of political systems, their effects are subtle in that they modify government commitments and risk-taking propensities in the face of international threats and pressures. While it is hoped that this chapter improves upon earlier depictions of the relationship, it clearly complicates empirical research. Simple research designs comparing basic political systems and asserting a singular effect on foreign policy appear to be inadequate. This is not to say that broad, even cross-national, studies are impossible, but only that they are more complicated. Let me close by briefly identifying three basic approaches for gauging the magnitude and direction of the effects of domestic politics on foreign policy.
The first approach focuses on the characteristics of opposition, based on the assumption that leaders are more sensitive to some types of opposition than to others and that their political strategies depend on the opposition's location, strength, and intensity.40 The "location" of opposition refers to its proximity to the process in which policy and leadership change decisions are made. Three arenas of opposition are (1) divisions within the leadership itself, (2) organized groups operating within (or having access to) state and party institutions, and (3) mass-level activity in the wider political environment.41 The "strength" of opposition concerns the political resources (votes, military force, etc.) an opposition can use to block policy initiatives or threaten the leadership's hold on power. Opposition "intensity" is the extent to which it challenges the government leadership, and ranges from challenges limited to government policies to those seeking to evict the current leadership, and can even include demands for the entire political system's overthrow. Leaders are more likely to respond to opposition that is close to the centers of power, that controls the greatest amounts of political resources, and that poses the greatest challenge to the domestic political order. It might also be argued that leaders' choice of political strategies depends in part on the character of the opposition they face. For example, accommodation tactics might be employed more often with intraregime opposition that has moderate resources, and whose demands are not extreme. Mobilization strategies are typically used to cope with opposition outside the regime that is extreme in its demands and so strong that it rejects any accommodation with the
regime.
A second approach considers political system structure based on the assumption that it broadly defines the context in which leaders cope with domestic political opposition. While I argue the imperative of looking at opposition in all types of
136        Second-Generation Foreign Policy Analysis
political systems, structural arrangements remain important. Even though political system structure does not directly account for the presence or absence of opposition, it does shape games of building policy coalitions and competing for power as well as leaders' choice of political strategies. Democratization and institutionalization are two of the most important political system properties. Democratization concerns the extent to which opposition is able to challenge the government's leadership and policies (Dahl 1971). Conventional logic suggests that democratic leaders are more sensitive to opposition they cannot control and are forced to accommodate it in the foreign policy process, while their authoritarian counterparts are more prone to control or manipulate opposition by aggressively legitimizing the regime and its policies.42 Equally important is political institutionalization, that is, the extent to which political norms and procedures (democratic or authoritarian) are established and accepted as legitimate and constrain the political behavior of the regime and its opponents (Huntington 1968). Leaders and opposition in highly institutionalized systems are willing to accommodate each other because of well-established norms and mutual trust. In contrast, political suppression and legitimacy strategies are more likely in the foreign policy processes of less institutionalized systems, because of distrust among opponents and the absence of political constraints on government coercion.43
A third approach considers the characteristics of the decision setting as a set of factors mediating the impact of domestic politics on foreign policy. This long-established approach asserts that situational (e.g., Hermann 1972; Snyder, Brück, and Sapin 1954) and issue area (Art 1973; Zimmerman 1987) factors, along with the political configurations within the regime, influence what type of decision body manages a particular problem.44 This research tradition is extended by Hermann, Hermann, and Hagan (1987; Hagan, Hermann, and Hermann n.d.) through the concept of the "ultimate decision unit." This concept refers to the decision-making body whose members have the authority and the power to commit national resources to a particular course of action in foreign policy. There are three types of decision units: predominant leader, single group, and a coalition of autonomous groups.45 The significance of the decision unit scheme is that some of each type of unit are relatively open to outside pressures, while others are closed. Closed decision units are driven by their own internal dynamics and thus are generally unresponsive to and intolerant of broader opposition. These decision units include the highly principled predominant leader who is likely to ignore or suppress opposition, the single group with strong internal loyalty whose decision making is driven by strong group norms creating an atmosphere of "group-think," and a coalition of autonomous actors without well-established decision rules that creates an atmosphere of extreme internal competition for power within the decision unit. Open decision units are, in contrast, more responsive to the positions of outside opposition, even to the point of attempting to accommodate them. They include pragmatic predominant leaders highly sensitive to actors who might challenge their regime, single groups without strong internal loyalty whose members represent their own constituencies, and coalitions with well-established rules that permit and require incorporation of outside actors and interests.
This concluding sketch hopefully points to the overall logic by which systematic research could take into account the complex, contingent, and pervasive nature of
Domestic Political Explanations in the Analysis of foreign Policy       137
Divergent Effects on Foreign Policy
Amplified ^Risk
Taking and
Commitment
Propensities
Original Propensity to Act Unaffected
Diminished Risk _»- Taking and Commitment Propensities
Note; This figure builds on Hagan (1993).
political influences on foreign policy, as outlined in this chapter. This explanatory logic is diagrammed in Figure 8.1. This scheme begins with the idea of dual domestic political imperatives of building policy coalitions and retaining political power that confront leaders coping with the international environment. The pressure of these games stems from the strength and intensity of political constraints, ranging from divisions within the political regime, organized groups in party and state institutions, and mass-level activity such as societal unrest and public opinion. Different levels of opposition are central (but not limited to) the games of coalition building and retaining political power, reflective of their proximity to foreign policy decision makers.
Next, the figure links these political games to foreign policy by way of leaders' choices of political strategies: (1) accommodation through bargaining and controversy avoidance, (2) insulating foreign policy from domestic politics by suppressing, buying off, or overriding opponents, and (3) mobilizing support (and isolating adversaries) by legitimizing the regime and its policies. Two kinds of contingencies likely affect the strategy, or mix of strategies, leaders employ to cope with opposition. The first are "contextual" and concern the conditioning role of political system characteristics, namely, the extent of democratization and institutionalization. The other contingency concerns the "mediating" effects of the relevant type of decision unit and, in particular, the degree to which its internal dynamics predispose its members to be "open" or "closed" to domestic politics. How these contingencies combine to shape leaders' responses to opposition ultimately determines how domestic politics affect foreign policy.
Figure 8.1
Elements in the Logic of Politicial Explanations of Foreign Policy
-------- Contextual Effects oj Political System Structure: -------
Democratization and Institutionalization
Domestic
Political
Imperatives
Building Policy Coalitions
divisions within the regime
organized groups in party and state institutions
mass-level,
non-institutional
opposition
Retaining IPolitical Power
Mediating
effects
of Decision
Unit's
Internal
Political
Dynamics
Alternative
Political
Strategies
Mobilization: Legitimization   — of Regime and/or Its Policies
Insulation of Foreign Policy
''Accomodation: Bargaining and Controversy Avoidance
138        Second-Generation Foreign Policy Analysis,
Finally, the political strategies of foreign policy makers, and the contingencies that predispose them to choose one or the other, may ultimately propel foreign policy actions in sharply divergent directions. Accommodation diminishes the government's ability to respond to international pressures or, in the extreme case of political deadlock, immobilizes the government entirely, causing it to grossly under-react to international pressures. In contrast, legitimization strategies amplify the predisposition to act on foreign threats and crises; the result is overreaction to international pressures. Domestic political perspectives that incorporate the complexity and pervasiveness of such effects provide important insights into the ongoing overreaction and underreaction of national governments to the international dynamics of balance of power, interdependence, and change.
■ Acknowledgments
This chapter draws upon several summer seminars at Ohio State University's Mershon Center and especially the contributions of Charles Hermann, Margaret Hermann, Valerie Hudson, and Eric Singer. This debt includes a reading of Charles Hermann's "Political Opposition as Potential Agents of Foreign Policy Change; Developing a Theory." The chapter has also benefited from the substantive ideas and editorial assistance of my research assistant, Deborah Wituski, and the book's editors, particularly Jeanne Hey.
■ Notes
1. The argument presented here should not be seen as posing an alternative to systemic explanations, but conforms with Haggard's (1991) "hegemony" of systemic theories. Except in the most extreme domestic situations (i.e., revolutions or other political convulsions), the role of domestic constraints would be to modify responses to international .systemic structures. The primary challenge of research on domestic politics and foreign policy is to show the regularized interconnection between external pressures and internal politics, not to demonstrate that the latter is more important than the former.
2. Similar kinds of concerns have been raised more recently in Destier, Gelb, and Lake (1984). See Almond's (1950) American People and Foreign Policy (or the most important systematic attempt by a political scientist to gauge isolationist sentiment in U.S. politics at the beginning of the cold war.
3. The literature on interservice rivalries richly describes the contending organizational positions and the compromises over budgets and weapons systems development. See, in particular, Caraley (1966), Davis (1967), Huntington (1961), Schilling, Hammond, and Snyder (1962), and Hammond (1963).
4. Overviews of the political setting of U.S. foreign policy since Vietnam are Nathan and Oliver (1987) and Rosati (3993). On Congress, in particular, see Frank and Weisband (1979), Spanier and Nogee (1981), and Destler (1986). The current major works on public opinion since Vietnam include Holsti and Rosenau (1984), Wittkopf (1990), and Schnieder (1983). Political economy analyses of the U.S. as a "weak" state include Krasner (1978,1988), Katzenstein (1976), and Ikenberry, Lake, and Mastanduno (1988).
Domestic Political Explanations in the Analysis of Foreign Policy        139
5. The most important of the many critiques of the bureaucratic politics approach are Art (1973), Krasner (1971), Perlmutter (1974), and Bendor and Hammond (1992). Also important are alternative conceptions of the decision-making process concerning "groupthink" (Janis 1982), cognitive processes (Steinbruner 1974), and presidential decision-making styles (George 1980b). For a synthesis of these perspectives, which are typically treated as competing alternatives, see Hermann (1993).
6. The postrevisionist literature represents an attempt to find a middle ground on the question of who was to blame for the cold war, and thus falls between orthodox perspectives (arguing the United States reacted to Soviet aggression) and revisionist perspectives (holding the capitalist U.S. system threatened the Communist bloc). Along with Gaddis's work, key general postrevisionist works include Dallek (1977), Paterson (1988), and Yergin (1977). On U.S. intervention in Vietnam, works by political scientists that employ a parallel logic include Gelb and Betts (1978) and Berman (1982).
7. This was partly because early efforts sought to assess the empirical validity of Rosenau's "genotypes" (e.g., East and Hermann 1974; Moore 1974b), but also because most researchers assumed political constraints on foreign policy could be traced to democratization (e.g., Moore 1974a; Salmore and Salmore 1970; Russettand Monsen 1975). Other studies employ democratization as an intervening variable and find that it conditions the effects of leader personality (Hermann 1980), political opposition (Hagan 1993), and domestic conflict (Wilkenfeld et al. 1980)
8. A very extensive empirical literature examines the cross-national, statistical association between domestic and foreign conflict. See, for example, Wilkenfeld (1973), Hazelwood (1975), and lames (1988).
9. See cross-national analyses by Geller (1985) and Hagan (1987,1993), both of which build on the early empirical work of Salmore and Salmore (1972,1978).
10. Rosenau (1966) raised this criticism most forcefully as part of his call tor a field of comparative foreign policy, but certainly by the 1980s he recognized the theoretical and methodological importance of more recent case study work (Rosenau 1987b).
11. General theoretical frameworks on Soviet foreign policy that stress domestic political factors include Aspaturian (1966), Bialer (1981),Dallin (1981),Simes (1986), and Valenta (1979, chapter 1). Good kremlinological overviews detailing the general political dynamics of the Khrushchev and Brezhnev regimes are, respectively, Linden (1966) and Gelman (1984). Juri Valenta's (1979, 1980a, 1980b, 1984) case studies of Soviet decisions to use military force abroad are the most innovative in detailing the substance of debates and their impact on foreign policy.
12. In addition to Koraný (1983,1986a), important assessments of Third World foreign policy analyses include Weinstein (1972), Clapham (1977), and David (1991). Much of the theoretical work on Third World decision making is, expectedly, regionally specific and covers the Middle East (Koniny and Dessouki 1984; Ismael and Ismael 1986," Dawisha 1990), sub-Saharan Africa (Good 1962; Shaw and Aluko 1984), Latin America (Lincoln and Ferris 1984; Muňoz and Tulchin 1984), and Asia (Vertzberger 1984a; Barnett 1985; and Chan 1979). Three excellent analytic cases studies ofthe politics of foreign policy are Vertzberger's (1984b) analysis of India's 1962 border conflict with China, Weinstein's (1976) examination of Indonesia's confrontation with the West under Sukarno, and Kim's (1971) study of the making of Korea's postwar peace treaty with Japan.
13.  Among the Western democracies, the research on the politics of Japanese foreign policy decision making is particularly impressive, led by the research of Haruhiro Fukui. It includes both theoretical overviews of the domestic political actors and processes (Calder 1988; Destler et al. 1976; Fukui 1977a; Hellmann 1969;Hosoya 1976; On 1976;
140        Second-Generation Foreign Policy Analysis
Pempe! 1977; Scalapino 1977) and thorough case studies of various types of both security issues (e.g., Fukui 1970, 1977a; Hellmann 1969) and economic issues (e.g.,Destler, Fukui, and Sato 1979; Fukui 1978). The work on other advanced democracies is a bit more dispersed but includes analyses of the Netherlands (Everts 1985), the Scandinavian countries (Goldmann, Berglund, and Sjostedt 1986; Sundelius 1982), Israel (Brecher 1972,1975), and the European powers of Germany, France, and Britain (e.g., Hanrieder and Auton 1980; Hanrieder 1967; Morse 1973; Wallace 1976; Smith, Smith, and White 1988).
14. As noted earlier, the phrase is taken from Müller and Risse-Kappen (1993). Their essay provides- a very useful survey of how domestic political phenomena are incorporated into international relations perspectives such as realism and complex interdependence.
15- Careful and precise discussions of how political perspectives can supplement systemic explanations of foreign policy can be found in Snyder (1991), Mastanduno, Lake, and Ikenberry (1989), Putnam (1988), and Siverson and Starr (1994).
16. See Katzenstein (1978) for comparative analyses of statist arrangements in the United States, Japan, and various Western European nations.
17. Cross-national, empirical assessments showing that foreign policy patterns (e.g., alignments) can change as a result of changes in domestic political regimes are Hagan
(1989), Moon (1985). and Siverson and Starr (1994).
18. This has spawned a large amount of empirical research demonstrating the absence of war among democracies, even if also finding that they are otherwise as war prone as authoritarian systems (e.g., Chan 1984; Maoz and Abdolali 1989; Maoz and Russett 1993; Schweller 1992; Small and Singer 1976). Dixon (1993) extends the argument to propensity for greater cooperation, while Maoz (1989) links warproneness to the political development with the conclusion that states undergoing radical political change are most likely to enter into foreign conflicts. See also Neack's essay (chapter 13) in this volume.
19. Snyder (1991), for example, argues that the fragmentation of political authority (i.e., cartelization) and resulting bargaining, or "logrolling," among contending hard-line groups contributed to aggressive and overextended foreign policies in Victorian Britain, pre-World War I Germany, 1930s Japan, as well as the Soviet Union and the United States at the height of the cold war. Lamborn (1991) examines political dynamics in Britain, France, and Germany across the two world wars, and his detailed case studies also illustrate general political dynamics common to both authoritarian and democratic systems. Compared with the democracy/war-proneness studies, there have been far fewer studies of these kinds of nonstructural political phenomena, except for Donike (1988), Morgan and Campbell (1991), and Morgan and Schwebach (1992). These literatures are brought together and analyzed in Hagan (1994).
20- Coalition-building processes are also at the core of foreign policy models of domestic structure (e.g., Mastanduno, Lake, and Ikenberry 1989; Lamborn 1991; Risse-Kappen 1991) and of models of single group and coalitional decision units (Allison 1971; Hagan, Hermann, and Hermann n.d.; also Snyder and Diesing 1977). The crisis of authority in the policy process underlies some of the arguments that democracies are less war prone than authoritarian systems with highly centralized governments.
21. Rich comparisons of U.S. and British foreign policy decision-making arrangements are found m Neustadt (1970) and Waltz (1967), and differences between the United States and Japan are overviewed in Destleret at. (1976) and Destler, Fukui, and Sato (1979).
22. Furthermore, the power of bureaucracies and interest groups is enhanced by this extreme fragmentation of leadership authority. Not only does it provide for more
Domestic Political Explanations in the Analysis of Foreign Policy       141
points of access for these particularistic interests, but these actors outside the cabinet also may become important political allies for squabbling central authorities.
23. Thus, for example, even on the critical issue of containing Czechoslovak reforms in the late 1960s, the Brezhnev Politburo was both pressured by the Central Committee and internally divided over the relative merits of military force as opposed to economic and political sanctions. It acted only after eight months of debate and internal wavering (Valenta 1979).
24. For discussions of the fragmentation of political authority in Wilhelmine Germany see Kennedy (1980), Lamborn (1991), and Snyder (1991).
25. Compared to pre-World War ! Germany, cartelization appears to have been more fluid, with factionalism within major actors as well as ties with various civilian players. On Japanese foreign policy making in the 1930s, see Fukui (1977b), Hosoya (1976), and Snyder (-1991).
26. The concept of the predominant leader is developed by Margaret Hermann in Hermann, Hermann, and Hagan (1987) and in Hagan, Hermann, and Hermann (n.d.). Historical examples are Germany's Bismarck and Hitler and the Soviet Union's Stalin, while contemporary cases are Cuba's Castro, North Korea's Kim Il-Sung, Iraq's Sadaam Hussein, Syria's Assad, as well as the first generation of many African postcolonial rulers.
27. Political survival is pervasive in foreign policy explanations of war (e.g., Levy 1989; Lebow 1981). The task of retaining power is central to the logic of various general foreign policy models (e.g., Hudson, Sims, and Thomas 1993;Salmoreand Satmore 1978; Van Belle 1993), particularly the extensive empirical research on the linkage between domestic conflict and foreign conflict (e.g., James 1988; Wilkenfeld 1973).
28. Other well-documented cases of foreign policy making by similarly unstable coalitions include Syria prior to the June 1967 War (Bar-Siman-Tov 1983), Indonesia's confrontation with the West under Sukarno (Weinstein 1976), Argentina's invasion of the Falkland Islands (Levy andVakili 1990), and China during the Cultural Revolution (Hinton 1972).
29. Risse-Kappen's (1991) model of public opinion and foreign policy effectively combines the dynamics of political survival and coalition building, showing that the latter is constrained in cohesive governments, which are then able to contain the pressures of public opposition.
30. In still other situations the tasks of building policy coalitions and retaining political power may be different, with the result that leaders may be able to act in ways that either game might separately suggest. Leaders, for example, might pursue a publicly unpopular policy because the regime is sufficiently cohesive to carry out its strong preferences, or because one of the critical members of the regime threatens to defect and bring down the regime immediately if it does not act. Another situation is where a severely fragmented regime engages in unexpectedly strong (though often erratic) foreign policy activity. This occurs when deadlocked actors within the regime seek to alter the situation by appealing to political groups outside the regime, often with virulent appeals to public nationalism and other unifying nationalist and ideological themes.
31. Accommodation via bargaining underlies statist approaches in the political economy literature (e.g., Ikenberry, Lake, and Mastanduno 1988; Katzenstein 1976; Krasner 1978) and models of economic policymaking (Destler 1986; Vernon, Spar, andTobin 1991).
32. Compromise is a central theme in the bureaucratic politics case research on U.S. national security policy, as cited in note 3.
142        Second-Generation Foreign Policy Analysis
33. Much of the international relations theory on pacific democracies rests upon the idea that domestic politics inhibits foreign policy makers (e.g., Doyle 1986). Lamborn (1991) treats domestic politics as a constraint on the risk propensities of both democratic and authoritarian belligerents.
34. This is true of the additional cases of unstable coalitions, as cited in note 29.
35. This is a critical point in any theory that incorporates domestic political phenomena into a broader theory of foreign policy. Any assessment of domestic political constraints must take into account that these pressures are balanced international imperatives and risks. Frameworks in this direction are developed by Lamborn (1991) and by Hudson, Sims, and Thomas (1993).
36. Nincic (1992), Snyder (1991), and Russen (1990) suggest that these dynamics can be contained by democratic systems before they provoke major wars or irreversible overextension of power.
37. These issues were then handled by senior ministerial bureaucrats, although it is important to note that these actors too were divided on the issue and unable to act much more effectively.
38. This definition is broadly based on Callahan (1982). Discussions of overextended commitments in international affairs are found in Snyder (1991) and Kennedy (1987), while Stein (1993) develops the theme of underextension.
39. The concept of risk taking in foreign policy is developed in Bueno de Mesquita (1981) and Lamborn (1991).
40. The articulation of the logic of this approach draws upon Hermann (1987b).
41. See Hagan (1993) for a complete discussion of the conceptualization and operational-rzation of organized opposition in terms of the following three properties. Mass-level opposition is the focus of cross-national analyses of domestic and foreign conflict (e.g., Wilkenfeld 1973; James 1988).
42. Of course, the importance of democratic constraints is demonstrated by the extensive literature that shows that democracies rarely fight wars with each other, although they are as war prone as authoritarian systems when fighting nondemocracies. See the literature cited in the first section of this chapter as well as the more extended discussion in Hagan (1994).
43. Several studies point to the importance of institutionalization as a broad foreign policy influence. Snyder's (1991) analysis of five cases of cartelized regimes shows that these pressures were contained in relatively institutionalized systems (mid-eighteenth-century Britain and the U.S. and U.S.S.R. during the cold war) but were not contained in less institutionalized systems (pre-Wbrld War I Germany and Japan in the 1930s). Maoz (1989) finds that newly established polities (both democratic and authoritarian) created through revolutionary processes are more prone to international disputes, as compared with more established systems undergoing evolutionary change. Hagan s analysis (1993) finds political influences on foreign policy are most pronounced for regimes in moderately institutionalized political systems, whose leaders have neither the power to suppress opposition nor the norms of consensus building to work with them. Finally, across the case study literature of the politics of foreign policy in non-U.S. settings, the contextual effects of institutionalization would seem as important as those of democratization (Hagan 1993, chapter 2).
44. It is widely accepted, for example, that in "crisis" situations policymaking becomes concentrated at the highest levels of government and debate is minimized by the pressures of high threat, short decision-making time, and surprise. Others argue that "redistribu-tive"(Lowi 1967) and "institutionally grounded" (Art 1973) issues involve a greater
Domestic Political Explanations in the Analysis of Foreign Policy        143
dispersion of authority and more intense conflict than do others of a less "zero-sum" nature.
45. Formal definitions of the three decision units are as follows. A predominant leader is a single individual with the authority to commit, irreversibly, national resources in response to a problem. A single group is an entity of two or more people who interact directly with all other members and who collectively are able to reach a decision without consulting outside entities (e.g., cabinets, juntas, and politburos). A coalition of autonomous groups has two or more politically autonomous groups (e.g., parties, institutions/bureaucracies), none of which alone has the ability to commit national resources without the support of all or some of the other actors. See Hagan (1993, 218-23) for a full discussion of adapting the decision-units approach to understanding how leaders respond to domestic opposition in foreign policy.