The Political Economy

1. Regimes involve mutual policy adjustments through a joint process of coordination and collaboration leading to the establishment of binding principles, rules, and decision-making procedures. See Stephen D. Krasner, International Regimes (Ithaca, N.Y.: Cornell University Press, 1983); Friedrich Kratochwil and John G. Ruggie, "International Organization: A State of the Art on an Art of the State," International Organization, Vol. 40, No. 4 (Autumn 1986), pp. 753-776; Stephan Haggard and Beth A. Simmons, "Theories of International Regimes," International Organization, Vol. 41, No. 3 (Summer 1987), pp. 491-517; Oran R. Young, International Cooperation (Ithaca, N.Y.: Cornell University Press, 1989); and John G. Ruggie, "Multilateralism: The Anatomy of an Institution," International Organization, Vol. 46, No. 3 (Summer 1992), pp. 561-598. of Nuclear Restraint


Etel Solingen
The choice between establishing a regional nuclear regime and maintaining an ambiguous nuclear status among the second tier or would-be nuclear powers is at the heart of debates about global security in the aftermath of the Cold War era. 1 The study of nuclear postures of regional powers (beyond the original five nuclear states) in the last three decades has traditionally emphasized their external security concerns. Such emphasis provided a powerful tool to explain the pursuit of a nuclear deterrent by countries like South Korea, Israel, and Taiwan, on the basis of legitimate existential fears. However, while their security concerns have been more or less constant for over thirty years, the nuclear postures of some of these countries have shifted over time. The external security context in and of itself is not enough, therefore, to advance our knowledge about why these states embraced different instruments, at different times, for coping with such fears.
More recently, the notion that the democratic nature of states explains their reluctance to wage wars against their democratic brethren (but not against others) has become central to theoretical endeavors in international relations theory. The explosion of studies on the relationship between liberal democ-racy and peace has not yet included a systematic extension to the study of nonproliferation, but it is often asserted that democratization will have a benign effect on denuclearization. However, this apparent connection may be less solid than we might like to expect: I argue that examining the eco nomic component of domestic liberalization in the different regional contexts may bring us closer to identifying an important engine of regime creation. In particular, ruling coalitions pursuing economic liberalization seem more likely to embrace regional nuclear regimes than their inward-looking, nation alist, and radical-confessional counterparts. 2 I do not suggest that security considerations are irrelevant to nuclear postures. Rather, I suggest an interpretation for why different states choose over time-different portfolios to cope with their respective security con cerns. My emphasis is more on explaining a favorable disposition to enter regime-like arrangements in nuclear matters than on listing incentives to procure nuclear weapons, as in the classical tradition of nonproliferation scholarship. Moreover, my argument is only relevant to would-be or second tier nuclear powers-"fence-sitters"-whose choices have taken place at a different world-time than that of the great powers. 3 Finally, although I refer to nuclear-weapons-free-zones (NWFZ) as the ultimate form of a regional nuclear regime, there may be other points "along the Pareto frontier" that might help avoid the dangers of nuclearization. 4 The next section summarizes the strengths and weaknesses of two alter . native ways of conceptualizing the choices of "fence-sitters": neorealism and liberal-democratic theories of peace. I then explore the link between the nature of domestic political coalitions (liberalizing versus nationalist-confes-2. I am aggregating under the "nationalist" rubric an eclectic group that often colludes in challenging different aspects of liberalizing agendas. Not all elements are present everywhere and their relative strength varies across states and regions. The confessional category includes radical ethnic or religious groups that are commonly labelled "fundamentalist." Because of some uneasiness with this last term-among some scholars of Islam, for instance-I use the terms radical or extreme confessionalism instead. 3. "Fence-sitters" are undecided states reluctant to commit themselves fully and effectively to a denuclearizing regime (it is important to differentiate between a formal commitment, such as sional) to regional nuclear postures. The following section examines the argument in light of evidence from South Asia, the Korean peninsula, the Middle East, and the Southern Cone of Latin America. It is followed by a section exploring the impact of international institutional forces on the sta bility of domestic coalitions and on their effectiveness in pursuing nuclear cooperation. The last section lays out the implications of this perspective for conceptualizing international-domestic linkages, and suggests ways in which the framework may be applied to further our understanding of how regional security regimes emerge.

Competing Perspectives on the Sources of Fence-Sitting
Neorealism and liberal-democratic theories of peace provide two alternate ways of viewing the choices made by would-be nuclear powers.

NEOREALISM
The point of departure of neorealist perspectives is that in an anarchic world states strive through self-help to increase their power relative to that of other states, in a zero-sum context, to secure their survival. This structure compels states to secure a balance-of-power equilibrium, and nuclear weapons can do the job by increasing security for all, and by generating caution, rough equality, and a clarity of relative power. 5 Regimes, in this view, are anomalies of international life and their occurrence ought not to be expected anyway. Where they emerge, they are no more than an epiphenomenon of deeper forces in world politics, that is, of power distribution. Prima facie, this per spective provides an intuitive entry into the kind of thinking that might have attracted many of these regional powers to the nuclear fence. 6 Yet, several logical and empirical problems afflict this theoretical approach. The Evolution of Nuclear Ambiguity in the Middle East," International Studies Quarterly, Vol. 38, No. 4 (June 1994). 6. Moreover, competition in the realm of security explains almost singlehandedly the decisions of the original five nuclear weapon states in the decade following World War II. As the intro duction makes clear, the focus in this article is on second-tier states that weighed their nuclear postures against a different "world-time," characterized by a highly integrated global economy and an integrating multilateral institutional foundation.
A general weakness of neorealism in explaining nuclear choices is its inconclusiveness. States indeed hope to reduce their external vulnerability yet, under a given structure, such an objective leaves room for a wide range of means. A regional power with fears for its survival may opt for any of a number of different solutions to alleviate them, from a full-fledged declara tion of nuclear capabilities to their total renunciation (to avoid escalation and instability, or to induce the other side to accept regional denuclearization). Consider, as an example of this variability of responses, India's test of a nuclear device and rejection of a regional nuclear regime; Israel's abstention from testing (but its warning that it would never be "second" in a regional nuclear race), while developing greater receptivity over time to a nuclear weapons-free-zone (NWFZ); South Korea's, South Africa's, Egypt's, and Tai wan's unilateral ratification and implementation of the Nuclear Non-prolif eration Treaty (NPT), after tinkering with a weapons program; and Pakistan's new openness to NPT and NWFZ solutions after dedicated efforts to acquire a nuclear deterrent. Moreover, not only have different states chosen contrast ing portfolios, but almost all have shifted their postures throughout the years. Taiwan, South Korea, Egypt and, more recently, South Africa, Brazil, and Argentina each overturned an ambiguous nuclear status by entering inter national commitments for effective denuclearization. The pursuit of security simply does not tell us enough about differences across space nor changes over time.
In the extensive, mostly descriptive, literature on the choices of fence sitters, one finds three main structural explanations of such differences. 7 The first explanation points to variance in vulnerability to massive conventional attacks. Thus, the more acutely vulnerable, such as Israel, Taiwan, South Korea, and (some have argued) South Africa, could be expected to be less likely to renounce a nuclear deterrent. 8 However, the last three of these four countries have done precisely that. A second explanation addresses the impact on its regional adversaries of one regional power's acquisition, or pursuit of, a nuclear option. However, the activities of a nuclear "pioneer" did not invariably lead to matching capabilities among adversaries. Egypt, South Korea, Taiwan (and even some African states) were suspected of harboring such designs to one degree or another but, in the end, all re nounced that path. A third explanation traces these differences or, more particularly, the decision to forgo a nuclear deterrent, to the willingness of a hegemon to provide fence-sitters with protection. Many analysts interpret the behavior of South Korea and Taiwan-under U.S. tutelage-in this light. 9 But hegemonic protection does not seem to be either necessary or sufficient for a turnabout in nuclear postures. Such guarantees of protection played no role in the decisions of Egypt, Argentina, Brazil, or South Africa to reverse ambiguous nuclear stances, while the security commitments of superpowers were insufficient to persuade North Korea, Iraq, Pakistan, or Israel to effec tively abandon nuclear weapons programs. Moreover, the U.S. commitment to Taiwan, said to have convinced it not to go nuclear, became questionable following the normalization of relations with the P.R.C. and the abrogation of the Washington-Taipei mutual security treaty.
In sum, the valuable insights we gain from a structural perspective are offset by its deficiencies. A state's structural context helps identify potential sources of nuclear postures, but does not account parsimoniously for the great variation encountered across countries and throughout time. Diverse responses came about under comparable and quite stable regional security threats, and against a common global-historical background characterized by three constants: a bipolar world, unrelenting pressure from the respective hegemons (the United States and the Soviet Union) for eschewing nuclear weapons, and a global normative structure squarely opposed to the horizon tal proliferation of such weapons. Shifts in overt substantive postures and signals of internal differences over the virtues and costs of alternative nuclear paths suggest that, at the very least, the consequences of, and solutions to, similar security predicaments are not universally consistent. Exclusively structural analyses offer limited ground for inter-regional comparisons be yond general truisms, such as arguing that the security context is more fragile in the Middle East than in Latin America. But that reality did not prevent Brazil and Argentina from nurturing weapons capabilities for over two de cades. Indeed, two contrasting security contexts-the Middle East and the Southern Cone-had similar outcomes: regional powers embracing ambigu ous nuclear postures and unwilling to commit to safeguarded denucleariza tion. In the end, studies that focus exclusively on structural explanations either cast the argument in nonfalsifiable terms, or explain variation away through brief references to domestic considerations or to a rough-and-tumble bureaucratic-politics account.10 One cannot, therefore, understand differences in hegemonic effectiveness without studying the domestic political conditions that make certain states more receptive to external persuasion than others. The lack of a rigorous examination of the domestic sources of nuclear postures among fence-sitters is particularly puzzling in light of the lessons from the U.S.-Soviet experience, where domestic processes acquired paramount importance in explaining arms control negotiations and eventual steps toward dramatic, if incomplete, nuclear reductions. Yet, the nonproliferation literature has largely resisted analytical inroads such as those that Matthew Evangelista and Jack Snyder applied to the study of major powers. Neither has it explored the links between security and trade strategies.11 There seems to be an emerging concern with domestic considerations, but mostly in the context of democ ratization and its likely impact on the nuclear evolution of regions.

LIBERAL-DEMOCRATIC THEORIES OF COOPERATION
Different hypotheses seek to explain the "democratic peace" or why liberal democracies are not likely to wage wars amongst themselves, and why they are at least as likely as others to engage non-democratic partners in armed conflict. 1 2 However, there has been no systematic attempt to extend these hypotheses to account for cooperative or non-cooperative behavior in the nuclear realm. What follows is an effort to (1) summarize, from the extant literature, a list of complementary institutional, perceptual, and normative 10. A related problem is evident in the widespread disagreement among experts about actual or perceived "levels of vulnerability" (the most common formulation of the independent varia ble). There may be more agreement on the Israeli and South Korean situation than on India, South Africa, or North Korea. The bottom line is that an elusive independent variable under mines efforts to operationalize it without resorting to internal factors. explanations of why democracies cooperate, and (2) to expand these hy potheses to specify potential nuclear outcomes.
A first set of explanations relates the "democratic peace" to domestic legitimacy and accountability, to democracy's built-in institutional checks and balances, and to the basic norm of peaceful resolution of disputes. Following a Kantian conception of citizens' consent, the assumption is that the legiti macy granted by the domestic public of one liberal democracy to the elected representatives of another has a moderating effect away from violent solu tions. 1 3 Moreover, free speech, electoral cycles, and the public policy process act as restraints on the ability of democratic leaders to pursue extreme policies toward fellow democracies. 14 Similarly, the normative rejection of violent behavior at home is extended to cover citizens of other democracies.
How do we apply this reasoning to gauge the potential behavior of dem ocratic dyads regarding nuclear weapons? One might extrapolate it to suggest that democratic dyads would be likely to shy away from basing their mutual security on nuclear weapons, which entail the most violent and extreme form of protection. Nuclear deterrence cannot preclude some measure of risk; even a low probability of deterrence failure leaves the door open to complete devastation. Such risk would be particularly intolerable in the context of a relationship between fellow democracies, which is rarely characterized by the kind of hostility apparent in mixed dyads. Dyads in which a democracy faces a non-democratic rival can be expected to behave differently than a democratic dyad. Abhorrence of authoritarianism and its lack of popular accountability could encourage a democracy to deter non-democratic adver saries through nuclear weapons, as in Western deterrence strategy against the Warsaw Pact. 15 The authoritarianism and praetorianism of an adversary can cancel the moderating effects of institutional checks and balances in a democracy that faces it. 16 The fear that a non-democratic rival with a concen trated monopoly of political power and with praetorian domestic structures could hold a democracy hostage through nuclear threats might raise the threshold of tolerance for risk among citizens and leaders of that democracy. Where a democracy suspects an asymmetry in risk-aversion or propensity to go to war, it will arguably be more willing to contemplate extreme solutions, such as nuclear weapons, for its security predicament.
A second, related, set of explanations for the "democratic peace" points to the credibility, transparency, and predictability of democratic systems. De mocracies are respectful of the rule of law and appear to undertake more credible and durable commitments, which strengthens their reputation as predictable partners. 17 Moreover, democracies are information-rich societies where knowledge about the internal evaluations of a policy or of the intensity of the preferences reinforces mutual credibility, enhancing the propensity to cooperate. 18 Maximizing information is of particular importance in the cre ation of security regimes, where the risks of error and deception can be catastrophic. 19 How would these conditions influence the nuclear behavior of democratic dyads? One might expect that by minimizing fears of deception (which underlie classical prisoners' dilemmas), mutual credibility and trans parency would strengthen their propensity to accept mutual nuclear dis armament. In contrast, problems of uncertainty over ratification and imple mentation would be exacerbated for mixed or non-democratic dyads, because procedures cannot easily be followed. Asymmetric levels of transparency could lower the incentives of democracies to join a nuclear regime, because Korea does not appear to be responding to North Korean nuclear behavior with a nuclear deterrent of its own (discussed below). 16. On practorianism, see Jack Snyder, "Averting Anarchy in the New Europe," in Lynn-Jones,  potential assaults on its stipulations by non-democratic would-be partners are harder to foresee (given more secretive and informal institutional proce dures). 2 0 Stable democracies bind successive governments to international agreements, whereas non-democratic regimes might be replaced by chal lengers capable of reneging on international commitments to maintain legit imacy at home. 21 Thus, under the conditions of lack of transparency, severe uncertainty, and low credibility characteristic of mixed or non-democratic dyads, one can expect greater reluctance to embrace nuclear disarmament.
This overview enables us to extract some preliminary conclusions and suggest some obvious limitations. 22 The expectation is that (1) nuclear re gimes based on the principle of disarmament might be more likely where all potential partners share like-minded democratic political systems; and that (2) asymmetric and non-democratic dyads might be more prone to maintain ambiguity or outright deterrence postures. 23 In other words, both democra cies and non-democracies discount the value of future nuclear cooperation with non-democracies.
This seemingly reasonable, albeit speculative, extension of democratic the ories of cooperation to explain the behavior of fence-sitters exposes some problems of logical inference and empirical fit. First, more research is required to assess whether or not the same mechanisms explaining the "absence of war" among liberal democracies are useful in explaining nuclear cooperation. After all, democracies do engage in conflict with other democracies, short of war. 24 Second, the expectation that non-democracies would be less transpar-20. Keohane, After Hegemony, pp. 93-95. 21. In fact, authoritarian rulers may enter an agreement and defect soon after, as Saddam Hussein's recanting of his earlier acceptance of some cease-fire stipulations demonstrates. 22. Additional propositions linking the democratic nature of states to cooperation might be logically extended to account for nuclear behavior. For instance, states sharing an open political system develop high levels of formal and informal communication, which lowers the cost of forming a regime (Keohane, After Hegemony, pp. 95-97). Openness allows the "trans-govern mental networks" of democratic dyads to share information on their respective domestic con ditions, thus facilitating transnational logrolling of support for a regime. Mixed or authoritarian dyads, in contrast, are likely to engage in more contained communication patterns in efforts to protect the autonomy of decision-making from outside interference. The efforts necessary to improve mutual communication raise the costs of creating a regime, particularly one involving mutual disarmament. ent and credible in their commitments and less reliable on ratification is often postulated, but very seldom explained or tested: it is more of a premise than a hypothesis. Third, the empirical evidence contradicts the notion that re gimes, including nuclear ones, are less likely to emerge between mixed or non-democratic dyads. In fact, most international regimes in every issue-area emerged out of compromises among a wide variety of political systems. The global nuclear nonproliferation regime is a case in point. 25 At the regional level, democratic administrations in Pakistan and in India failed to agree to mutual denuclearization, while military dictatorships in Brazil and Argentina set in motion bilateral nuclear cooperation in 1980. A highly mixed and politically quite unstable group of Latin American states signed the Treaty of Tlatelolco in 1968, establishing a NWFZ. On the Korean peninsula, a relic of Cold War authoritarianism and a democratizing (but far from genuinely democratic) South began negotiations over a system of reciprocal inspections in 1991, a process that is now derailed. In the Middle East a mixed lot is negotiating regional arms control, including weapons of mass destruction, in the context of the multilateral peace process. Moreover, the diffusion of democracy might be expected to mitigate Israeli concerns with the credibility and reliability of its non-democratic neighbors; however, where radical Is lamic forces are its bearers, democratization may not bode well for a security regime, given these forces' advocacy of total and final wars against infidels generally, and against Israel in particular. 26 Political freedom thus seems neither necessary nor sufficient for the emergence of a nuclear regime.
Perhaps the most severe problem lies in inferring the potential nuclear behavior of fence-sitters, without qualifications, from the experience of ad vanced industrialized democracies, 27 from whose history the democratic pre disposition to avoid wars and build regimes with fellow democracies is overwhelmingly drawn. First is the question of the extent to which demo cratic stability, far more abundant among industrialized states, plays a critical role in explaining the "zone of peace" these states have created. 28  democratic stability might thus anticipate why the emergence of "zones of peace" in Third World regions would be more tentative and piecemeal, particularly on nuclear matters. Second is the possibility that economic lib eralism, rather than democracy, may be more useful in explaining nuclear cooperation, a hypothesis I explore in the remainder of this paper.

Economic Liberalization, Political Coalitions, and Nuclear Preferences
The type of domestic political system does not explain why fence-sitters shift their nuclear postures and join international and regional nuclear regimes; therefore a more disaggregated analysis of domestic determinants of nuclear cooperation, and particularly of the role of economic liberalization and of the political coalitions that sustain it, are required. 29 I suggest that ruling coali tions pursuing economic liberalization are more likely to embrace regional nuclear regimes than their inward-looking, nationalist, and radical-confes sional counterparts.
This hypothesis is based on two main assumptions: first, the kinds of ties binding actors (groups, sectors, parties, institutions) to economic and other international processes affect their conceptions of interests. 30 Such ties influ ence actors' definitions of what trade-offs are desirable or tolerable. For example, a state's decision to maintain ambiguity in nuclear intentions (e.g., by refusing full-scope safeguards or by cheating on NPT commitments) has involved, since the 1970s, a series of trade-offs: access to international mar kets, capital, investments, and technology has been curtailed, directly and indirectly. 31 Such tradeoffs create domestic coalitions favoring or rejecting such linkages: groups that might otherwise pay little attention to their coun try's nuclear posture become more attentive to the elements of the interna tional bargain. Groups interested in importing highly sophisticated comput ers, for example, might not be concerned with their country's refusal to commit unambiguously to renounce nuclear weapons, if such refusal had no direct implications for their access to external resources. Interested actors both respond to, and anticipate, international constraints. 3 2 Second, nuclear postures are not merely a response to international con straints: the domestic consequences of alternative nuclear paths are no less important to political actors and coalitions. For instance, the political effects of doing away with nuclear ambiguity often includes the weakening of state bureaucracies and industrial complexes that constitute an impediment to economic rationalization. 33 Conversely, denuclearization tends to be part of a broader program of domestic reform that strengthens market-oriented forces and the political entrepreneurs and central economic institutions pro moting their development. This has clearly been the case in Argentina and Brazil, where multibillion dollar nuclear investments undertaken in the 1970s became primary casualties of the contraction of state activities in the 1980s and 1990s.
Thus, two basic types of coalition-one advocating economic liberalization, the other opposing it-may develop contrasting perspectives on both the domestic and international consequences of alternative nuclear paths.

LIBERALIZING COALITIONS
The interests of political coalitions favoring economic liberalization are gen erally internationalist; that is, they require openness to global market and institutional forces. 34 A policy of economic liberalization implies a reduction of state control over markets and of barriers to trade, an expansion of private economic transactions and foreign investment, and the privatization of public sector enterprises. Different liberalizing coalitions emphasize different as pects of economic liberalization, depending on their interest in specific issues, such as expanding exports, deregulating financial flows, opening the do- mestic market to foreign goods and investment, and reducing state entre preneurial activities. Their selective and gradual agenda of economic liber alization explains why the adjective "liberalizing" is more appropriate than "liberal"; the state often plays a powerful role in steering this process. Political liberalization-reduction of state monopoly over political life-is not an im mediate requirement for economic liberalization, at least during its initial phase. 35 For the most part, the pillars of liberalizing coalitions are liquid-asset holders and export-oriented firms-including large banking and industrial complexes capable of surviving without state protection-and state monetary agencies. 36 They tend to be more receptive to structural adjustment policies, and opposed to external confrontations with the international financial and investing community. 37 The ability of big business (locally and foreign owned) to influence domestic investment patterns and to move capital abroad gives them an important voice in shaping domestic and external adjustment policy. 38 Smaller firms engaged in exports or supplying internationalized enterprises can also be part of these coalitions, along with the highly-skilled labor force associated with these firms. Public and private managerial, tech nical, scientific, educational, information, and service-oriented professional groups, which might be called "symbolic analysts," similarly tend to be oriented towards an open global economic and knowledge system. 39 The economic orientation of such coalitions, which strive to maximize their gains from international economic exchange, makes them more likely to be receptive to compromise nuclear postures that do not endanger their inter- ests. These coalitions rely extensively on the global economy and on the political support of major powers within regimes and institutions involved in managing international economic relations. A policy of nuclear disarma ment enhances their status with international institutions and powerful states, who associate these coalitions with the promise of democracy, ration alization, and regional cooperation. Related domestic considerations reinforce these coalitions' opposition to large-scale, ambiguous and unbounded nu clear programs: such programs often contribute to the ailments afflicting these countries' domestic political economy, such as the expansion of state power, the maintenance of unproductive and inflation-inducing military in vestments, and the perpetuation of rent-seeking patterns. 40 In other words, liberalizing coalitions do not merely trade away the right to have "the bomb" for the right to make money; they perceive little inherent benefit in a policy of nuclear ambiguity for both domestic and international reasons. For such coalitions, denuclearization is quite compatible with an agenda of liberalizing the economy and reining in adversarial political forces at home. A restrained nuclear posture can secure certain international economic, financial, and political benefits-such as debt relief, export markets, technology transfer, food imports, aid, and investments-that can be used to maintain or broaden domestic political support and to strengthen the domestic institutional frame work underpinning economic liberalization. In sum, cooperative regimes in the economic and security realms are mutually reinforcing; they spell trans parency, predictability, a good reputation, and the blessing of the interna tional community. They also help carry out domestic policies largely in line with these coalitions' political and economic preferences. financial institutions, and favor a more expansionist course. 41 These inward looking coalitions include popular sectors comprising unskilled blue-collar workers, white-collar and other state employees, and small businesses; firms that compete with imports and that have close ties to the state and domestic markets; underemployed intelligentsia, and politicians who fear the disman tling of state enterprises and the consequent erosion of their basis of political patronage. 4 2 They may also include arms-importing military establishments, which are often adversely affected by adjustment programs. 4 3 In the Middle East and South Asia, nationalist coalitions often attract extremist religious movements. Such movements thrive on popular resent ment over adjustment policies they regard as externally-imposed, reliance on foreign investment, and the "Western" principles and norms embodied in most international regimes. 44 The material basis for opposing internation alization and liberalization may be particularly strong where import-substi tuting and state-based industrial interests are powerful. In other cases reli gious or ideological components may be the driving force. Very often these two tend to reinforce each other. Leaders of these logrolled coalitions of rent seeking economic interests and militant religious, ethnic, or cultural groups rely heavily on what Jack Snyder labels mythmaking; that is, they sell myths that justify the allocation of state resources to the wide array of interests backing an inward-looking strategy. 45 Their rejection of global markets and in stitutions is echoed in their adversarial regional postures. In the extreme form of nationalist coalitions, such as the one that supported Saddam Hussein, nuclear weapons often play a central (and more open) role in the call for final, "redeeming" solutions. Many such inward-looking, state-oriented (rather than market-oriented) economic leaderships, whether democratic or otherwise-such as North Korea, India, Iraq, Libya, and Cuba, and even Brazil and Argentina-long resisted external pressures for joining or comply ing with the global nonproliferation regime or any regional alternative to it.
Summing up the argument so far: the political impact (at home and abroad) of transcending nuclear ambiguity is positive for liberalizing coalitions, which pursue an export-oriented and internationalist grand strategy of industriali zation, and negative for nationalist coalitions, which pursue an inward looking industrialization strategy. 46 The foregoing analysis of the relationship between political-economic strat egies and nuclear postures suggests a pattern, not an infallible rule. Thus, domestic coalitions in an industrializing state may be strongly supportive of their country's integration within the international economy, while resisting other (political, security, environmental, human rights) global regimes; China is an example. However, international tolerance for attempts by ruling coali tions to disaggregate a state's allegiance to emerging global arrangements ("we will trade as freely as we repress and pollute") may be rapidly declining. The commitment to international regimes is becoming increasingly indivis ible. Nuclear postures have become nested in a broader context of global (primarily economic) relations that create certain mutual expectations. The international community expects adherence to nonproliferation principles, while fence-sitters expect to share in the benefits of international economic interdependence.
Coalitions of one type or the other come to power through electoral means, where democratic institutions are in place, or through more or less coercive methods, often in alliance with the military. At times, a rough parity in the competition between these two basic kinds of coalition precludes the adop-46. Such strategies not only determine a country's relation to the global political economy but also the internal allocation of resources among groups and institutions. On grand strategy as an economic, political, and military means-ends chain designed to achieve security, see Barry For a more detailed elaboration of the connections between coalition type and security postures, see Etel Solingen, "Democracy, Economic Reform, and Emerging Regional Orders," in David Lake and Patrick Morgan, eds., Reconceptualizing Regional Relations (unpublished ms., University of California, 1994).
tion of a clear-cut strategy concerning international economic, nuclear, and other regimes, as in Argentina during much of the postwar era. At other times, one coalition is able to impose a relatively unchallenged path, as with South Korea, Taiwan, and other East Asian countries. Most often, however, industrialization strategies incorporate different mixes of inward and out ward-looking instruments, and vary in the extent to which it is state officials or rather powerful societal forces that dominate the definition and imple mentation of strategy. This point underscores the need to understand differ ences between the ideal-types of coalitions outlined above and the more nuanced versions in the real world.
Political institutions affect when and how certain coalitions of interests can prevail. 47 There is considerable variation in the way in which the preferences of different coalitions are aggregated. For instance, incipient democratization and electoral trial-runs provided new opportunities for nationalist-confes sional coalitions in the Middle East. These opportunities were quickly shut down where the military perceived them to weaken its institutional viability and strength, as in Algeria in 1993. In Israel, proportional representation has precluded the emergence of a single dominant coalition. Understanding the domestic determinants of different coalitions' success in gaining or maintain ing power is an important question in itself, but not a task that can be undertaken here.

The Empirical Record Across Nuclear Regions
This section examines the impact of coalition type on the nature of nuclear postures in four regions. Reasons of space preclude a more detailed historical account.

THE KOREAN PENINSULA
South Korea provides a classic example of a liberalizing coalition imple menting a nuclear policy compatible with its fundamental interests and grand strategy of industrialization. Following a relatively brief import-substitution phase, state bureaucrats guided the country's integration in the global econ- omy, leaning on foreign markets, loans, technology, and investments. 4 8 To embrace a nuclear deterrent, which the leadership considered doing in 1971, would have threatened that strategy and, consequently, the regime's viabil ity. 4 9 Thus, the coalition backing the Park regime responded to U.S. threats of a major break in bilateral economic relations by cancelling the incipient weapons program and ratifying the NPT in 1975.50 For a regime taking cues from a strong military establishment, this shift in nuclear posture makes the triumph of political-economic considerations in defining the country's sur vival strategy even more remarkable. 51 By the 1980s South Korea's export oriented coalition included virtually all segments of business and labor. 52 At the end of the decade this coalition was actively pursuing a NWFZ, despite South Korea's unquestionable technical and industrial capacity to "beat" North Korea in a conventional or nuclear arms race.53 Despite a North Korean threat to turn Seoul into a "sea of fire," there seems to be little popular and governmental support for a South Korean nuclear deterrent, and few in the South believe the North would ever use an atomic weapon.54 No doubt hegemonic protection (that is, U.S. tactical nuclear weapons) was an important consideration in weighing denuclearization, but one should not exaggerate its impact. In the aftermath of Vietnam, the Nixon Doctrine, and the normalization of relations with Beijing, the South Korean leadership 49. Export-oriented firms were critically dependent on primarily U.S. and Japanese investors, loans, and markets. Moreover, in an oil-dependent economy, the promise of plentiful nuclear energy from Western-supplied power plants to fuel heavy industry and intermediary sectors was at risk if nonproliferation commitments were not maintained. Reiss had reason to question the robustness of U.S. security guarantees.55 The prospects of U.S. economic sanctions were even more troubling for a lead ership that was acutely aware of U.S. contributions to the country's internal political stability and, in essence, to the regime's survival. The potential for disrupting an average annual rate of growth of over 10 percent and for domestic political turmoil was a powerful threat to reckon with.
The North Korean case approaches the inward-looking, nationalist coali tion ideal-type as closely as any real case can. Self-reliance and the cult of the leadership became central political instruments of mythmaking, some what like a secular version of radical confessionalism. Nuclear weapons (or ambiguity about their possession) became the ultimate expression of national independence (juche) and technical achievement that the regime could wield as evidence of its own viability; this was particularly critical once the South's economic vigor became apparent. 56 An independent and ambiguous stand on nuclear matters was thus an important ingredient in the North's political economic grand strategy of self-reliance, and one with high payoffs for soothing a restive military and nuclear establishment and its nationalist allies in the bureaucracy.5 7 The United States and North Korea's former Soviet protector coerced it into ratifying the NPT in 1985, an event that, tellingly, followed a feeble North Korean attempt at economic liberalization. However, giving new meaning to the difference between formal and effective commit ments to nonproliferation principles, North Korea continued to reject full scope IAEA inspections de facto, even after the United States had removed tactical nuclear weapons from South Korea in 1991. This outcome highlights the difficulty of reducing North Korea's behavior, as paranoiac as it may seem, exclusively to security concerns. 58 What compelled the North Korean leadership to drag its feet on the nuclear issue was the fact that U.S., South Korean, Japanese, and multilateral promises of improved economic ties did not materialize. 59 With this, the regime's expectation of preventing its own collapse, by relying on economic reforms, evaporated. In the absence of tangible international commitments to provide economic aid to North Korea, the incipient liberalizing forces in North Korea lost out to the hard-liners in the military and nuclear establishment. By March 1993 North Korea an nounced its intention to withdraw from the NPT, unleashing an international crisis. Despite the efforts of very powerful strategic allies and very powerful strategic adversaries, North Korea was not persuaded to relinquish nuclear aspirations for an extended period of time. Domestic receptivity is an im portant intervening factor in the relationship between hegemonic assertion and fence-sitters' responses.

INDIA-PAKISTAN
The coalition shaping economic and industrial policy in India's post-inde pendence era embraced a classical import-substitution model of industriali zation, aimed at avoiding vulnerability to international markets and economic institutions. 60 It strongly criticized international regimes as constructs of Western powers, opposed the nonproliferation regime as the crowning ex ample of neocolonialism, and conspicuously exploded a nuclear device in 1974. Advocates of an Indian nuclear deterrent pointed to China's nuclear status as a convenient justification for their position, despite China's clear "no first use" policy. A group of Congress Party officials favored nuclear weapons for domestic political reasons (polls indicated majority support for such weapons), to counteract the party's responsibility for India's defeat by China in 1962. 61 A similar overall inward-looking strategy at first character ized Pakistan under democratically-elected Zulfiqar Ali Bhutto, a populist who pursued nuclear weapons and discussed the merits of an "Islamic bomb." 62 President-General Zia ul-Haq maintained a patronage system and used Islam as a source of national identity, but sought greater reliance on the West; political survival required straddling antagonistic nuclear postures, or the maintenance of ambiguity. 63 India took initial steps at economic liberalization under the Janata govern ment (1977-80), one far less friendly than any of its predecessors to state centered inward-looking policies in general and to the nuclear-industrial complex in particular. Prime Minister Morarji Desai vehemently opposed India's nuclear weapons program. 64 Subsequent incipient Indian and Paki stani efforts at liberalizing their domestic markets and foreign trade during the 1980s coincided with a modest attempt by Indian Prime Minister Rajiv Gandhi and Pakistani Prime Minister Benazir Bhutto to initiate nuclear co operation. 65 But in India, liberalization (including lowering of some import barriers) encountered heavy opposition from beneficiaries of the old model, bent on keeping international economic institutions and foreign investment at arms' length. Business groups, state central financial agencies, and the professional middle class supported the relaxation of state controls, but many opposed lowering trade barriers. The rank and file in the Congress party, public sector employees, the intelligentsia, which is mostly state-employed, and some rural sectors (known as "middle peasants") rejected all efforts at liberalization, internal and external. Nationalist, inward-looking constituen cies, including the technical and entrepreneurial military-industrial complex, were increasingly under attack, but not yet retreating. 66 India's government under P.V. Narasimha Rao represents a stronger at tempt at economic liberalization; it has shaken subsidized sectors and state owned industries, actively pursued European Community, Japanese, and U.S. investments and World Bank and IMF loans, and allowed Finance Minister Manmohan Singh to rechart not only India's economic course, but its foreign policy as well. 67 Very early in its tenure, and following a strong electoral showing by the opposition Bharatiya Janata Party (BJP), Rao's mi nority government did not embrace a 1991 Pakistani overture for a NWFZ, but neither did it reject the offer completely, and it agreed to exchange information with Pakistan on the location of their respective nuclear facili ties. 68 The Pakistani proposals notably followed the November 1990 ascen dancy of Prime Minister Nawaz Sharif, whose trademark was an emphasis on free markets, economic liberalization, foreign investment, and interna tional financial aid; urban business, commercial, and professional groups backed this strategy. 69 A representative of industrial interests, Sharif publicly rejected the label "fundamentalist" and lamented the political energy in vested in debates over Islamization "while the world is marching fast to meet the challenges [of] the twenty-first century." 7 0 Yet Sharif, like his predecessor Benazir Bhutto, proved to be too willing to undertake public projects with high political payoffs, which helped bankrupt Pakistan.71 With the ousting of Prime Minister Sharif, former World Bank vice president Moeen Qureshi challenged the power of entrenched elites during his brief transitional ad ministration, giving the Central Bank new powers to control government deficits while he attempted to freeze nuclear activities. 7 2 The policies of Sharif and Qureshi of attracting foreign loans and investments required a dramatic reduction of defense spending, which antagonized segments of the Pakistani military. 73 The second administration of Benazir Bhutto, re-elected in October 1993, rejected Qureshi's reforms and remained responsive to feudal land owners, the bureaucracy, and the army; it also pledged to protect Pakistan's nuclear program. 74 The platform of India's main opposition party, the radical-confessional Hindu BJP, combines calls to ban foreign loans, investments, and imports with a call to build and deploy nuclear weapons. 75 Cashing in on widespread popular resentment against the West, both for its economic success and for imposing a nuclear cartel, the BJP also enjoys increasing support from import competing industries such as food processing, automobile manufacturing, banking, and communications. The party thus expressly rejects a wide range of "Western" regimes, from the NPT, to GATT, the World Bank, and IMF imposed restructuring plans, to the policies of international development agencies that favor population control and the eradication of illiteracy. Many of these positions were echoed by Pakistan's militant Islamic party Jamaat-i Islami, which challenged what it regards as the Westernized policies of the Sharif coalition. 76 President Ghulam Isaq Khan, representing the nationalist political camp, was thought to support efforts to produce a nuclear weapon, and to have refused to allow elected politicians to negotiate over nuclear matters. 77 The relative strength of competing radical nationalist-confessional coalitions is bound to play a key role in shaping the Kashmir crisis and South Asia's nuclear future.

THE MIDDLE EAST
Current ruling coalitions in the Middle East comprise oil-exporting industries in the Gulf and the Arabian peninsula; tourist-based, commercial-agriculture, and munfatihun ("openers") economies in Egypt and Jordan; high-tech export oriented industrialists in Israel; and influential Sunni merchants in Damas cus. 78 These coalitions advocate openness to international markets, invest ments, and tourism; cooperative relations with international financial insti tutions; and support for the Arab-Israeli peace process. Most have ratified the NPT and have consistently advanced NWFZ proposals at UN fora. 79 Leading exemplars of such coalitions-Iran under the shah and Egypt under President Sadat-played entrepreneurial roles in organizing support for a NWFZ. This behavior does not merely imply a passive acceptance of security concessions in exchange for economic advantages, as opponents of liberal izing coalitions often argue. The domestic consequences of cooperative re gional postures have positive attractions for advocates of reform. They help the coalition politically, in coping with the socioeconomic havoc left by de clining oil prices, bloated bureaucracies and economic mismanagement, over population, militarization, and foreign-policy adventurism. 80 Infitah ( eco nomic liberalization) was at the heart of a grand strategy for a new, triumphant Egypt, introduced at a time of scarce resources and dwindling state revenues. 81 It required foreign capital, financial assistance, and Western technology, as well as a commitment to private capitalist accumulation, all of which secured the backing of a powerful coalition of business interests and technocrats. Infitah played an important role in persuading President Sadat to negotiate an unprecedented peace treaty with Israel. 8 2 The Camp David agreements, in turn, marginalized the domestic opposition to Sadat's regional politics, even if it radicalized the Islamic fringes. It is quite suggestive that infitah was launched in 1974, the same year Egypt first proposed, with Iran, a NWFZ. 8 3 Different strains of radical Islamic challenger offer themselves as an alter native to liberalizing coalitions. Islamic blocs include "bourgeois fractions, some rural agrarian capitalists, notables and estate-owners, and the virtually proletarianized members of the state-employed petit-bourgeoisie, the under employed intelligentsia, and the large student population. " 8 4 These blocs propose a new political economy that, for the most part, appears incompat ible with a regional nuclear regime. Opposition to liberalizing coalitions often involves a rejection of ties to the international economy and its perceived associated scourges: inequality, corruption, unemployment, and enslaving indebtedness. 8 5 In addition, militant Islamic political movements promote a new social order that is not receptive to the idea of a comprehensive peace settlement, let alone a regional nuclear regime. 8 6 The domestic political appeal of these movements increases primarily with their ability to satisfy popular socioeconomic and educational needs, and in some cases, with calls for extreme "redeeming" solutions. 8 7 However, the ability of such radical coali-  tions to "deliver" on the warfare (jihad) side, by investing in military infra structure, is limited by their need to secure welfare and redistribution. These tasks are much harder to fulfill without preserving the export-oriented rentier state and, therefore, preserving extensive ties to the world economy. 88 Radical Islamic movements are not an ideological monolith; their political economy themes vary, as do their approaches to the West. 89 Moderate Islamic movements do exist, even if their influence in Middle East politics has so far been somewhat limited. 90 Some of these movements are less concerned with military buildups and exotic weapons than with the socioeconomic transfor mation of their societies. Such differences are neglected if one focuses only on the behavior of the two Islamic Republics in existence, Iran and Sudan, and on the political platforms of militant movements. These do not bode well for compromising solutions to regional security predicaments. Likewise, the most forceful message of the radical Islamic bloc in the 1993 Jordanian elections was opposition to an Arab-Israeli peace settlement, relegating the rejection of IMF-induced economic reform to a secondary theme.
IRAN. The shah is credited with embarking on a large-scale nuclear energy program and an interest in nuclear weapons has even been imputed to him; yet his regime pioneered a Middle East NWFZ. 91 Following the Islamic rev olution in 1979, Iran discontinued its active role in promoting a NWFZ. The emerging Islamic Republic of Iran acquired many of the characteristics of the radical-confessional inward-looking ideal-type, expressing contempt for prin ciples of diplomatic extraterritoriality, individual freedoms, and anti-terror ism, and executing a national redistribution of wealth from the private to the public sector. 92 Although an NPT signatory, Iran is suspected of pursuing a weapons program, an accusation President Hashemi Rafsanjani has denied. 93 However, Vice-president Sayed Ayatollah Mohajerani argued in 1992 that "we, the Muslims, must cooperate to produce an atomic bomb, regardless of UN efforts to prevent proliferation." 94 Although mention of Iran's "mod erate" wing often raises incredulity, outward-looking undercurrents do ap pear to be alive, if not well. Teheran's Bazari (bazaar) merchants and money lenders have backed President Rafsanjani' s attempts at reform and at reduc ing state control over the economy. Their gradually increasing control of the 270-seat Parliament could consolidate their struggle for domestic political dominance. 95 They call for privatization of an extensive network of state-run factories and power plants, and are stepping up pressures for increased trade with Europe and Asia and for a utilitarian-as opposed to an ideological approach to foreign policy. The unresolved contest between liberalizing and radical Islamic blocs may help explain the unclear and unstable nature of Iran's nuc�_,rr_postures in the past decade.
IRAQ. Iraq's Ba'athist regime approximates the inward-looking, state based, nationalizing, militarized ideal-type rather well. Saddam Hussein's more recent use of Islam is an ideological ornamentation with occasional political payoffs; it does not alter-in fact, reinforces-the basic rejection of economic liberalism that Ba'athism embraced in the first place. His regime's survival can be traced to a combination of repressive controls, involving abominable human rights abuses and successful redistributive policies. 9 6 The entrenched combined power of state enterprise bureaucracies, import-sub stituting interests, and their respective beneficiaries among the professional, construction, and working class may have been responsible for preventing change beyond a limited, tentative effort at economic liberalization. external expression of this domestic model was a formal rejection of the international economic order and of the Camp David peace agreements, a reliance on the Soviet Union, advocacy of nonalignment, and a "hawkish" position within the Organization of Petroleum Exporting Countries (OPEC). Iraq's nuclear program represented a crowning symbol of the interests and wherewithal of this ruling coalition emphasizing self-reliance. The activities of the UN Special Commission on Iraqi disarmament have neutralized a substantial portion of that program, in an unusual case of forced denuclear ization. Genuine Iraqi openness to any regional cooperation-nuclear or otherwise-may have to await a new leadership that reflects the interests of a coalition oriented to global trade, investments, and to the emerging inter national institutional order.
ISRAEL. Israel's developmental strategy in the immediate post-indepen dence era combined statism and import-substitution with some dependence on foreign capital and agricultural exports. 98 Despite a general political and economic orientation toward the West, Israel developed high mistrust of international institutions, where Third World majorities could automatically endorse Arab positions regardless of substantive merit. It is assumed that a small group around Prime Minister Ben-Gurion, who embodied statism (Mamlachtiut), developed the foundations of a nuclear weapons program in the 1950s and 1960s, never acknowledging its existence. 99 The ambiguous nature of this program relieved a succession of precarious, unwieldy ruling coalitions from the need to debate a program about which little agreement could be found. 100 External and especially alliance considerations and their impact on domes tic constituencies played an important role in the way in which different coalition members defined their positions on nuclear weapons. For instance, in the 1950s the leadership of the Mapam and Ahdut Avoda parties (both politically powerful within Israel's General Federation of Labor, Histadrut) forcefully rejected nuclear deterrence; the policy would have threatened not only relations with the Soviet Union, but also the anti-nuclear feelings of the pro-Soviet constituencies on which these two parties relied.101 Leading mem bers of the largest bloc in the Labor coalition (Mapai) similarly opposed reliance on nuclear deterrence. In the late 1960s, Prime Minister (and previ ously Finance Minister) Levy Eshkol and Finance Minister Pinhas Sapir con ceived an incipient policy of economic liberalization aimed at promoting export-led and high-tech industrialization. By that time, not only did depen dence on foreign capital leap, but donors (including the United States) ap plied much more restrictive conditions than those characterizing earlier dis cretionary capital inflows. As Michael Barnett argues, the Israeli state at that time became "more beholden to foreign actors to stabilize its financial life."10 2 Mapai' s leaders were not about to endanger the nascent domestic strategy or its external underpinnings (international markets and a political alliance with the West) by being unresponsive to U.S. concerns over Israel's nuclear designs. Eshkol is credited with attempts to rein in Israel's nuclear pro gram. 103 The political windfalls of such steps for Eshkol' s Mapai faction included weakening a program that, since Ben-Gurion, had access to un supervised budgetary sources.
In the last decade two basic modalities have come to characterize Israeli coalition politics. On the one hand, Labor-led coalitions tend to attract sup port from the urban professional and middle class, from high-tech industri alists, highly-skilled labor, and export-oriented cooperative agriculture. It was Labor's Shimon Peres who designed and implemented the economic reform and currency stabilization plan of the mid-1980s, albeit in the context of a government of national unity with Likud. Labor coalitions (which include smaller secular left-of-center parties) are more receptive to international in stitutions, territorial compromise, regional regimes, and arms control agree-ments. With the exception of Ben Gurion and his group (which eventually formed a new party, Rafi) during the 1950s, most of the Labor party has traditionally opposed reliance on nuclear deterrence. Many influential Labor leaders have favored a NWFZ. 1 04 Prime Minister Yitzhak Rabin declared in 1974, in response to Defense Minister Moshe Dayan's call for nuclear weap ons: "Attempts to rely on mystical weapons are negative trends." 1 0 5 In 1975 Foreign Minister Yigal Allon (of the Ahdut Avoda party) proposed direct negotiations over a regional NWFZ at the UN General Assembly. By 1980 Israel was voting in favor of the Egyptian NWFZ proposal, submitted regu larly to the UN since 1974. Yet, in the context (until very recently) of a rigid Arab refusal to recognize the existence of the state of Israel, let alone its security concerns, any attempt by Labor to endorse a unilateral denucleari zation was politically prohibitive.
On the other hand, populist Likud-led coalitions, a bloc that in Israel's early years represented free-enterprise liberalism, are now more susceptible to demands from nationalist economic interests (including small business, blue-collar and underemployed workers, development towns, and West Bank and Gaza settlers) and from the orthodox, including some radical religious and nationalist constituencies. They tend to be more dismissive of interna tional institutions and untrusting of their objectives, and to use the myth of self-reliance to gather political support from an array of economic and ideo logical groups, most of which reject a territorial compromise with Palestin ians. 1 0 6 Menachem Begin's Likud government bombed Iraq's Osirak reactor in 1981 (in a pre-elections period), against the opposition of important Labor leaders who, sensitive to the response of allies and institutions like the International Atomic Energy Agency, sought more time for diplomacy. 1 0 7 Although prominent Likud leaders are associated with opposition to a NWFZ, the party has no declared policy on this issue.10 8 Overall, Likud and its more natural coalition partners on the right are more responsive to na tionalist policies and economic constituencies, and distrustful of international institutions and regional cooperation.
Labor's political comeback in 1992 foreshadows greater Israeli willingness to embrace a regime, and eventually a NWFZ. First, the Labor coalition has led, rather than followed, public opinion in matters of national security; it presented the public with a fait accompli in the form of the September 1993 Declaration of Principles recognizing the Palestine Liberation Organization. A 3:2 margin of voter approval followed. Moreover, the Labor coalition is more sensitive to Israel's international standing, and to the domestic political and economic consequences of such status.1 0 9 A recent statement by Deputy Foreign Minister Yossi Beilin summarizes the aims of Labor diplomacy: "to use the new situation in order to become a more welcome member of the international club."11 0 The coalition emphasizes the exigencies of economic survival, privatization, and international competitiveness as well as the fu tility of technological fixes as the solution to Israel's security dilemma.111 This approach threatens the rents and political influence of military-industrial groups and state bureaucracies, while expanding the opportunities for civil ian-oriented private entrepreneurship and the power and autonomy of eco nomic agencies such as the Central Bank.112 108. On the position of Ariel Sharon, Yuval Ne'eman, and Rafael Eitan, see Y. Nimrod, "Arms Control or Arms Race?" New Outlook, September/October 1991, pp. 15-18; and Uri Bar Joseph, "The Hidden Debate: The Formation of Nuclear Doctrines in the Middle East," The Journal of Strategic Studies, Vol. 5, No. 2 (June 1982), pp. 205-227. Supporters of Likud and its rightist allies justify the use of nuclear weapons by a larger margin (46 and 57 percent respectively) than the electorate of Labor and its partners (43 and 36 respectively); Arian, et al., National Security, p. 72. 109. As a reward for the Rabin government's more flexible positions in the Middle East peace talks, even prior to September 1993, the United States approved $10 billion in loan guarantees for investments in infrastructure and jobs, a program that may help solidify Labor's position as the core of future coalitions. Likud had chosen to retain an intractable position in the peace negotiations at the expense of economic gains. The issue of the loan guarantees reaffirms the need to transcend neorealist interpretations of state behavior. Moreover, Labor's flexible posi tions were far from externally imposed; the loan approval was a windfall for a policy that Labor supported anyway. 110. Quoted in Eric Silver, Financial Times, December 7, 1992. 111. The inability to prevent Iraqi Scud missiles from landing on Tel Aviv (or on the Negev desert, where the Dimona nuclear center is located), and the internal threat posed by the intifada increased popular receptivity to political solutions to Israel's security predicament.

On pressures for conversion to civilian industries, see Aharon Klieman and Reuven Pe-
The future of Israel's alleged nuclear arsenal is on the agenda of the Arms Control Group in the multilateral peace negotiations, although there is dis agreement whether non-conventional weapons should be discussed at the outset or at the end of the process.113 It is doubtful that representatives from the Labor government can politically afford to agree to any arrangements on the nuclear issue prior to the achievement of a comprehensive peace settle ment and to the resolution of outstanding problems with Iran and Iraq. There is a strong proclivity among Labor leaders and their core supporters to do away with most barriers to effective regional cooperation.11 4 However, the need to secure swing votes (potentially attracted t� Labor and its political allies in all but their security postures) may require caution on the question of tradeoffs for peace, and in the rate at which they are delivered. On the other hand, Egypt and other incumbent coalitions in the region may be unlikely to obtain popular ratification of a comprehensive peace settlement that omits the curtailment of Israel's nuclear capabilities, at least at some point in the future. The successful completion and implementation of such a settlement will deflate the mythmaking potential of radical-confessional groups on both sides. Despite promising developments, the concluding chap ter of the Middle East's NWFZ has yet to be written.

ARGENTINA AND BRAZIL
The presidency of Juan D. Peron in Argentina epitomized the national pop ulist economic model that vied for control of the state for half a century. 11 5 It involved a coalition of national small and medium-sized firms involved in import-substitution industrialization, state firms producing the required infrastructural inputs, and popular sectors represented in the central trade union organization (the General Labor Confederation). The external expres sion of national populism was a challenge to free trade and the unpredicta bility of the international market, and also a rejection of foreign borrowing and investment as well as membership in the IMF and the World Bank. Peron actively pursued nuclear capabilities in the early 1950s, and announced the country's mastery of fusion technology in 1953, on the basis of a false claim by expatriate Austrian physicist Ronald Richter, who managed Argen tina's nuclear program at the time.11 6 The origins of a well-funded nuclear program in Argentina are thus deeply rooted in the national populism of the Peron era; such origins endowed the program with the myth of self-reliance.
After the military coup of 1955 deposed Peron, a tripartite division of state industrial assets among Argentina's armed services allowed the navy to shelter the nuclear program during an unstable succession of mostly military regimes. This succession was notable for alternating stop-go economic and industrial policy cycles, reflecting the inability of either coalition-the one supporting liberalization, the other opposing it-to prevail politically for a sustained period of time. Attempts at liberalization, as with President Arturo Frondizi's acceptance of an IMF stabilization plan and of foreign exploitation of Argentina's oil reserves, coincided with attempts in the 1960s to curtail the nuclear program and reduce its autonomy. The military administration of Videla in 1976, strongly influenced by Economic Minister Martinez de Hoz and his orthodox policies, challenged the costly nationalist-mercantilist ori entation of the nuclear program, the bloated and inefficient state sector and non-competitive national private industry. Although privatization and dwin dling governmental expenditures threatened the nuclear program, the navy was able to defend it throughout the 1970s and 1980s. Martinez de Hoz was ousted, and no coalition supporting widespread liberal economic reforms was strong enough to implement them until the early 1990s.
In Brazil, the administration of Getulio Vargas in the early 1950s evoked many of the same elements of national populism as in Argentina under Peron. Restrictions on foreign investment led to a refusal by the World Bank to finance the Vargas strategy of industrialization, or that of his successors, until 1964. In the nuclear realm this policy was expressed in the regime's attempt to develop independent national nuclear capabilities as early as the 1950s.117 In 1952 President Vargas approved directives to the National Secu rity Council demanding "specific compensations" in the form of transfer of technical know-how on plant construction, and delivery of equipment and Unlike the nationalist inward-looking coalitions backing Argentina's Pres ident Peron and Brazil's Presidents Quadros and Goulart, the military re gimes that intervened in Brazil and Argentina since the 1960s shifted without abandoning import-substitution altogether-to greater reliance on foreign direct investment, industrial exports, and indebted industrializa tion. 11 8 This strategy required stronger economic ties with the United States, Western Europe, Japan, and international financial institutions. Although formally headed by repressive military rulers with some constituencies that favored nuclear weapons, these coalitions of technocrats, industrialists, and bankers maintained considerable control over economic policy. They were thus able to resist pressures (such as that of Peronist legislator Quiroz) for a nuclear posture that could trigger international penalties or other restrictions on technology transfer. Throughout these years, characterized by a hybrid model of secondary import-substitution and closer trade and financial links to the international system, nuclear policy revealed two main features. On the one hand, both Brazil and Argentina continued their longstanding rejec tion of the NPT as a discriminatory tool of nuclear powers, and their resis tance to making the regional Tlatelolco NWFZ effective on their territory. On the other hand, they refrained from developing nuclear weapons and from threatening to do so.11 9 The beginnings of moderate nuclear cooperation can be traced to the late 1970s and, more specifically, to the 1980 agreements signed in Foz do Igua�u by Presidents-General Videla of Argentina and Figueiredo of Brazil.
In the mid-1980s, Brazil's President Jose Sarney implemented a nationalist populist mixture of domestic heterodoxy and anti-IMF policy that led even tually to Brazil's 1987 debt moratorium. In an attempt to maintain both business and popular support, President Raul Alfonsin defined a heterodox adjustment policy in Argentina, relying on neither the old radical Peronist populism nor the radical orthodox rhetoric of the military' s policy under Videla, while preserving a cooperative stance with international creditors. The new democratic administrations, neither of which was prepared to adopt orthodox liberal medicine for their countries' economic ailments, proceeded with a moderate pace, but no real breakthrough, in both economic liberali zation and nuclear cooperation. In Brazil, sections of the military continued to develop a "parallel program" with weapons applications, even after at tempts, through the Constitution drafted in 1988, to place all nuclear activities under democratic control.1 20 By the late 1980s, drops in real wages, price freezes, and tax reforms alienated Sarney's popular constituencies, forcing Brazil to turn to an IMF style orthodox stabilization package. The Argentine government was at this time particularly careful to provide strong reassurance to its banking and industrial firms, in light of populist challenges to the state during the late 1980s. 121 At the end of the decade, both President Fernando Collor de Mello of Brazil and President Carlos S. Menem of Argentina supported shock ec onomic programs unambiguously c ommitted to effec tive ec onomic liber alization and struc tural adjustment. The liberalizing Menem revolution re duc ed a Weimar-style inflation level to single digits, balanc ed the budget, privatized many public servic es, and attrac ted an avalanc he of foreign in vestment. The external dimension of these polic ies inc luded not only an unprecedented embrace of liberal trade rules but also of other international regimes, including missile control. 122 Following his election, President Collor was equally c ommitted to liberalizing Brazil's ec onomy, a polic y that won him an approval rating of close to 90 percent in early 1990. By November of 1990 Brazil and Argentina agreed explic itly, for the first time, to renounce nuclear weapons and to establish mutual verification and inspection proce dures, which were ultimately approved in December 1991. 12 3 The two coun tries also expressed their intention to put into effect an updated version of the regional Tlatelolc o Treaty, and Brazil's President Collor c losed down presumed nuclear weapons test sites. After over 35 years of unassailed navy c ontrol, Argentina's nuc lear program was now at the merc y of President Menem's direc tor of planning, whose major goal was the privatization of nuclear ac tivities. Aided by advisors from large Argentine corporations and joint ventures, the Menem administration neutralized the program's sensitive nuclear facilities. 124 Menem has gone as far as expressing unilateral readiness to ratify the NPT by the time of the 1995 Extension and Review Conference. 12 5 In Brazil, following Collor's resignation in 1992 over a corruption scandal, his successor ltamar Franco began wooing a nationalist and military consti tuenc y, attac king international financial institutions and their domestic "al lies," and endorsing statements on Brazil's sovereignty in nuclear matters. 126 However, nationalist forc es failed to prevent Brazil's House of Deputies, under heavy pressure from the Foreign and Ec onomic Ministries, from ap-proving the mutual inspection agreements with Argentina in late 199 3 . 127 The Senate is expected to follow suit. Brazil's opposition to NPT ratification may well be explained as a side-payment to the nationalist camp, including portions of the military.

Lessons and Implications
This cross-regional analysis suggests that the political-economic nature of domestic coalitions and the choice of nuclear postures are related. I suggest that this relationship can be traced to the type of industrialization strategy these coalitions embrace. Liberalizing coalitions strive to maximize their gains from international economic exchange. Their receptivity to compromising regional nuclear postures secures them access to international economic re gimes and the political support of major powers; denuclearization is also quite compatible with the domestic agenda of liberalizing the economy and reining in adversarial political forces and institutions. In contrast, nationalist and radical-confessional coalitions logroll economic interests and militant groups that regard nuclear weapons as a useful political tool, to rally oppo sition to global markets and regimes and to the settlement of regional con flicts. As with most propositions about the sources of state behavior, how ever, the argument suggests no more than a probabilistic relationship.
These assumptions help explain the behavior of states operating a) in different regional security contexts; b) with different associations with he gemonic powers; and c) over time, throughout a historical succession of alternating coalition-types. The Middle East, the Korean peninsula, and South Asia offered a natural quasi-experimental ground to examine the im pact of different political regimes, controlling for the intensity of the security dilemma and the presence of hegemons providing protection. Under com parable regional structural contexts we would have expected to find similar responses, but we do not. And under disparate security-related conditionscompare the Southern Cone with the Middle East, the Korean peninsula, or South Asia-some states embraced similar policies of nuclear ambiguity for lengthy periods of time. A wide variety of domestic political regime types (democratic and otherwise) converged in cooperative practices, despite ex pectations from the theories connecting democracy and cooperation.
The cross-regional and longitudinal analysis suggests that where liberal izing coalitions had the upper hand, nuclear policy shifted towards more cooperative nuclear postures. Nationalist-confessional coalitions, in contrast, shied away from any commitments for effective denuclearization. Moreover, where the domestic interests potentially affected by external sanctions were most concentrated and coherent, and less challenged domestically, as in South Korea and Taiwan, the shift in nuclear policy was relatively swift. The stronger the coalition supporting economic liberalization grew, the more clear-cut the departure from nuclear ambiguity was (even where the security context deteriorated, as in the Korean peninsula). This is illustrated by Ar gentina's commitment to the full-scope safeguards regime in the early 1990s, following the consolidation of political forces supporting economic liberali zation. It is also clear from the example of South Africa's acceptance of NPT arrangements in 1991, even as it disclosed past attempts to produce a bomb. In another example, Spain endorsed the NPT when a liberalizing coalition eager to join the European Community was able to put the inward-looking, nationalist policies of the Franco era behind it.
In contrast, the weaker the liberalizing coalitions-as was the case histor ically in India and Israel, Argentina until the early 1990s, and Iran today the more politically constrained they were in curbing nuclear programs. Weak liberalizing coalitions are often less able to defend themselves from the ac cusation of selling out; their very weakness also renders them more depen dent on additional domestic partners. Liberalizing coalitions walk a tightrope to sustain their legitimacy: they must not only deliver on their promises but also preserve fluid external ties while avoiding the appearance of foreign subordination. 128 Such conditions may help explain the hesitation of the Rao government in India to promise effective denuclearization, or Brazil's initial wariness under ltamar Franco to implement it.
Of all states beyond the original five that have considered a nuclear option in the last three decades, not one endorsed a NWFZ under a nationalist coalition. Only liberalizing coalitions undertook effective commitments to denuclearization. The North Korean case may offer fresh insights into the process by which economic liberalization, coalition survival, and nuclear postures become entangled. There are indications that the same political forces staunchly opposed to economic liberalization are using the nuclear issue to stave off the end of the present regime. 129 What are the more general implications for theories of international co operation and regimes? First, the approach suggested here points to a more precise link between economic liberalism and the probability of cooperation than general theories of interdependence have postulated. 130 Rather than assuming that the expanded domestic welfare resulting from free trade fos ters cooperative preferences, it suggests that where free-trade coalitions pre vail, their interests at home and abroad dictate compatible security regimes. The gains from trade need not be highly concentrated nor contribute to widespread societal welfare, at least in the short term, to have this effect. Moreover, cooperation does not depend on the extent of economic interde pendence between or among the regional participants in a regime. Finally, my argument is more specific about whose absolute gains matter in the analysis of cooperation: the gains that matter are those of particular coalitions.
Second, this last point places more constraints on purely neorealist for mulations beyond those discussed above. The preferences, domestic and international, of domestic ruling coalitions matter a great deal. These coali tions evaluate costs and benefits with an eye to strengthening their political standing at home, and they define the balance between the costs (if any) of nuclear cooperation and the overall gains from participating in global re gimes. To say that once these coalitions embrace an internationalist strategy of industrialization they become more sensitive to pressures from powerful states and international institutions is not the same as arguing that foreign pressures or inducements singlehandedly, and invariably, account for the outcome. Such pressures were a constant for most of the Cold War era, yet they cannot settle the puzzle of why they triggered a regime-oriented behav ior at certain times and not others, and among certain states, and not others. The accession of contending domestic coalitions provides a more powerful predictor of such dynamics and variability. Identifying the domestic condi tions underlying behavioral shifts takes us several steps beyond structural explanations in understanding how external and internal factors interact to produce changes in nuclear postures. The next section explores this inter- action by an alyzin g how in tern ation al in stitution s can stren gthen different coalitions over others.

CONDITIONALITY"
How do international institutions affect the domestic balance of power be tween coalitions, and thus, as a second-order effect, their respective nuclear postures? 131 As allies of liberalizing coalitions, the international institutions that provide credit (World Bank, IMF, private banks) and define the terms of trade and investment (GATT, regional common markets) can play an important role in the political longevity of these coalitions. 132 Imposing harsh and widespread structural adjustments can undermine these coalitions' le gitimacy and survival, and weaken their capacity to gather support for re gional security regimes. The failures of some liberalizing coalitions in the Arab world (and of the international institutions within which their interests are embedded) to bring about a genuine socio-economic transformation in the region has provided fertile soil for the rise of radical Islamic challengers. To prevent further erosion of popular support for liberalizing coalitions, international economic regimes must encourage domestic redistribution. 133 Tight conditionality arrangements have been ineffective anyway, whereas securing a stable political environment improves the borrowers' ability to attract investments, repay debts, and stem authoritarian challenges. 134 The IMF and the World Bank could return to their true call by lending for eco nomic development, stabilization, and recovery, rather than helping debtors pay their debts to big banks. 135 In other words, the survival of liberalizing coalitions requires that the benefits from a cooperative nuclear posture-in trade, investments, removal from export control lists, debt-relief, and aid-be distributed more broadly, beyond just the concentrated interests which sustain these coalitions. Pro viding resources, compensatory payments, and relief from the pressures of international competition can weaken domestic opposition to liberalization and pragmatism. Such efforts may build on a wave of growing popular awareness of the opportunity costs of arms races. 13 6 Furthermore, a shift in the style of foreign institutional intervention toward effective consultation over domestic political needs, and more active participation of developing countries in the decisionmaking process within international institutions, can deflate nationalist resentment. 137 Such an approach may help these institu tions tame extreme views and foster a form of liberalism, even one attentive to moderate confessional aspirations, that would view regional and interna tional regimes positively. 138 The other side of this coin, of course, is the power of liberalizing coalitions to "use" the threat from nationalists and radical-confessional movements to extract concessions from their interna tional partners and to alleviate the conditions for continued credit and in vestment. This "reverse conditionality" will continue to be part of the bar gaining strategies of struggling liberalizing coalitions in the future.
International institutions can strengthen the hands of certain domestic institutions at the expense of others. For example, externally-induced struc tural adjustment efforts often threaten military-industrial complexes while strengthening those in charge of reform (particularly finance ministries, cen tral banks, and export-promotion bureaus). Similarly, international pressures for human rights standards empower the domestic groups responsible for monitoring compliance, at the expense of repressive agencies. Environmental regimes endow local institutional networks with the ability, backed by un precedented legal powers, to challenge certain industrial activities such as nuclear energy production. The resulting coalitional balances are more likely to reinforce openness and receptivity to nuclear regimes than the coalitions and institutions they are replacing. 139 Interests opposed to nuclear weapons could become more concentrated (and therefore, more attractive partners for logrolling) than those that favor them.
Finally, the fact that international regimes strengthen the influence of the most powerful countries that created them is not lost on developing coun tries, particularly fence-sitters, or on those in transition to market-oriented economies. If such regimes continue to be regarded as an instrument for control of the less powerful, their legitimacy could be eroded. But this can be ameliorated if their injunctions are universally binding, especially where they require the elimination of nuclear arsenals, as required by Article VI of the Nonproliferation Treaty.

Conclusions
In this article I offer a framework for understanding the domestic sources of regime creation by outlining how contending coalitions affect nuclear pos tures. 1 4 0 The growing attention to domestic factors has mostly been directed at understanding the structure of interests within a specific issue area to explain cooperation ( or its absence) in that same area, 1 4 1 but understanding outcomes in the security arena requires a broader consideration of how political-economic strategies affect security choices. 142 Such an approach helps specify what early neoliberal-institutionalism left unexplained: where the preference to cooperate comes from. By relying on a single analytical category, this approach transcends the practice of nonproliferation studies of explaining each country or region through a list of individualized peculiari ties.
The evidence points to an association between strategies of industrializa tion and nuclear postures that is worthy of both theoretical and policy-making consideration. Understanding this association may prove an effective means of moving beyond extant scholarship and conventional wisdoms that have become more convention than truth. The findings suggest that the credibility of commitments by fence-sitters may be more affected by what kind of domestic political-economic coalition underwrites them than by the institu tional constraints of democracy. Where these coalitions rely for their domestic political survival on an open economic system, they will not only be more susceptible to international inducements to cooperate but will favor denu clearization for its domestic effects as well. State structures influence the fate of different coalitions and, in turn, are changed by them; states are both the agents of liberalization and the victims of it. The performance of coalitions varies with the nature and strength of technocratic agencies on the one hand, and of rent-seeking actors and their challengers on the other. Exploring how this variation accounts for different paths to regional denuclearization may be a logical next step. Additional research may also enable us to understand thresholds, lags, and sequences in the process by which developmental grand strategies and nuclear postures become linked.
Because international institutions bankroll free-trade coalitions, they are a great source of strength for such coalitions, as repositories of side-payment "currency," and at the same time a potential Achilles heel, a symbol of curtailed sovereignty. Thus, these institutions must calibrate their perfor mance to enable cooperative coalitions to mobilize societal resources in sup port of nuclear regimes. 143 Imposing heavy burdens on such coalitions may result in their "involuntary defection," or in their inability to deliver because of low prospects for domestic ratification. 144 Understanding the impact of international processes on the strength of domestic coalitions is not equiva lent to reducing the politics of these countries to external forces. As the international political economy literature suggests, different coalitions have chosen contrasting grand strategies of industrialization (integrative or inward-looking) under similar international circumstances.
Finally, economic liberalization appears to require democratization if it is to be sustained over the long term. 145 In that sense, it may well be that many 143. This might be relevant for the emerging Eurasian nuclear regime. Ukraine, Belarus, and Kazakhstan, unlike the cases reviewed here, acquired overt nuclear status at birth, rather than contemplating their acquisition. 145. The Middle East and China appear to require longer lags than previously thought. For a more nuanced interpretation of the political consequences of economic adjustment, see Haggard and Kaufman, "Economic Adjustments." regional partners negotiating nuclear regimes, now and in the future, are and will be democratic. Yet both democracy and nuclear cooperation could still be an outcome of economic liberalism. Exploring further the extent to which political freedom will be necessary or sufficient for the emergence and maintenance of regional nuclear regimes is a compelling task for a social science theory sensitive to the construction of a more peaceful global order.