Domestic Coalitional Analysis and the Democratic Peace

DEMOCRATIC PEACE THEORY - which postulates that democratic states do not wage wars against one another - has developed into a thriving research programme in international relations. The programme includes both supporters and detractors who apply statistically oriented, formal-modelling, and case-study methods.1 The last has particu­lar affinity with historical analysis. Spencer R. Weart, in testing the dieory against evidence running from antiquity to the twentieth century, finds unqualified support for it.2 This article scrutinizes the democratic peace from the perspective of the relationship between domestic political pro­cesses and international behaviour. It explores an alternative, and in some cases complementary, explanation for war and peace that derives from a domestic coalitional approach to politics. The framework pays special attention to the effects of internationalization, in its economic and political dimensions. Some scholars, like Weart, have granted the democratic peace theory the status of a historical law; others only endorse its validity for the con­temporary era. The coalitional argument oudined here offers more limited claims, taking the form of a set of general and falsifiable expectations, and acknowledges systematic world-time effects on die behaviour of states. For example, global world-time at the dawn of the twenty-first century is characterized by tension between a globalizing political economy and local or national challenges to it; resistance to an institutionalized global political order; conflicts between a pluralist (multicultural) political approach to human diversity and exclusivist ethno-religious forms; and similar conflicts between regionally differentiated versus globally homogeneous solutions to the opportunities and predicaments posited by the first three. Acknow­ledging die notion of a world-time could imply that our ability to discover historical laws of international behaviour is compromised, though not necessarily negated. In a landmark analysis of die impact of commerce on war and peace, for example, Richard Rosecrance mapped the longue durée stages in the evolution of states from territorial states to trading states and, more recendy, virtual states.


ETEL SOLINGEN
Democratic Peace Theory -which postulates that democratic states do not wage wars against one another -has developed into a thriving research programme in international relations. The programme includes both supporters and detractors who apply statistically oriented, formal-modelling, and case-study methods. 1 The last has particu-lar affinity with historical analysis. Spencer R. Weart, in testing the theory against evidence running from antiquity to the twentieth century, finds unqualified support for it. 2 This article scrutinizes the democratic peace from the perspective of the relationship between domestic political processes and international behaviour. It explores an alternative, and in some cases complementary, explanation for war and peace that derives from a domestic coalitional approach to politics. The framework pays special attention to the effects of internationalization, in its economic and political dimensions.
Some scholars, like Weart, have granted the democratic peace theory the status of a historical law; others only endorse its validity for the contemporary era. The coalitional argument outlined here offers more limited claims, taking the form of a set of general and falsifiable expectations, and acknowledges systematic world-time effects on the behaviour of states. For example, global world-time at the dawn of the twenty-first century is characterized by tension between a globalizing political economy and local or national challenges to it; resistance to an institutionalized global political order; conflicts between a pluralist (multicultural) political approach to human diversity and exclusivist ethno-religious forms; and similar conflicts between regionally differentiated versus globally homogeneous solutions to the opportunities and predicaments posited by the first three. Acknow-ledging the notion of a world-time could imply that our ability to discover historical laws of international behaviour is compromised, though not necessarily negated. In a landmark analysis of the impact of commerce on war and peace, for example, Richard Rosecrance mapped the long;ue duree stages in the evolution of states from territorial states to trading states and, more recently, virtual states. 1 Elsewhere I have devised a framework responsive both to the global polit icaleconomic context and to the regional context of which states must take account in their decisions over war and peace. 2 The approach shares with Weart's the inclusion of various regime types, not just democracies, and various outcomes, not just the absence of war. Given the wider range in both the explanandum and explanans, both lines of argument are more demanding and not as easily tractable as the simpler claim that democracies do not wage wars against one another. However, unlike Weart's, the coalitional approach offers no more than probabilistic statements about the impact of global forces on the definition of domestic political coalitions which, in turn, advance competing regional agendas in war and peace.
Three idealtypical coalitions form in response to worldtime consider ations: internationalizing, backlash, and hybrid. Politicians organize these coalitions according to the hypothesized position of different groups with respect to internationalization, in material terms, or on the basis of ideo logical or normative proclivities. Thus, coalitions rest on both converging interests and norms. It is easier to identify the most likely partners log rolled in internationalizing and backlash coalitions than in hybrid variants. In any event, these are ideal types never to be found in the real world. The suffix in internationalizing indicates a process, a path, an empirical example that rarely approximates the ideal.
Internationalizing coalitions aggregate primarily constituencies from the most dynamic sectors favouring openness to the global economy and its associated political institutions, including international financial, service, and labour groups, as well as consumers of imported products, state agen cies entrusted with economic reform (such as independent central banks and managers of exportprocessing zones), and competitive agricultural exporters. Politicians organizing backlash coalitions attract sectors threat ened by external liberalization, such as private and stateowned enterprises and banks, uncompetitive bluecollar labour, state employees, and agencies vulnerable to reform (such as those dealing with capital controls or import IR. Rosecrance, The Rise ofthr TradingStatr (New York, 1986) and Tht Rise oftht Virtual Stat, (New York, 1999). licensing). The armed forces and their associated industrial complex often join, as internationalization threatens them financially and institutionally by restricting budgets and curtailing national sovereignty. Nationalist, eth nic, and religious groups concerned with the homogenizing effects of inter nationalization also gravitate to back.lash coalitions, often appropriating longstanding critiques of international capitalism as wasteful and cor rupting. The composition of hybrid coalitions is less clearcut. They can bring together strange bedfellows.
Each coalition endorses a grand strategy with synergistic effects across the domestic, regional, and global arenas. Domestically, the grand strategy of internationalizing politicians includes the pursuit of economic policies compatible with global access. Externally, the strategy is designed to main tain secure access to foreign markets, capital, investments, and technology. Regionally, a cooperative (nonviolent and stable) neighbourhood serves the grand strategy in all its aspects, providing a secure investment environ ment and appropriate macroeconomic conditions, while avoiding expen sive arms races. 1 Resource mobilization for military conflict is counter productive for internationalizers, potentially inducing inflation, budget deficits, higher interest rates, and the continued protection of stateowned enterprises under a mantle of national security. Such developments are dis ruptive of the kind of domestic political economy internationalizers pro pose to develop.
The grand strategy of back.lash politicians seeks to preserve allocations to military and other protected industrial complexes. Externally, it resists pressures for internationalization while challenging an array of inter national regimes. Regionally, insecurity and competition help to sustain the strategy whereas cooperation threatens it, as cooperation compels downsized allocations to the military and associated industrial enterprises, diminishes opportunities for nationalist mythmaking, and endangers the extraction and allocation of fiscal resources to the coalition's members.
Grand strategies can be more pristine or watereddown versions of the ideal according to a ruling coalition's strength relative to its opponents at home and in the region. Hybrids are expected to span the grand strategies of the ideal types but rarely forcefully or coherently. The regional context helps explain a particular coalition's behaviour. For example, strong internationalizing coalitions interacting with each other in a region dyads or clusters like Western Europe or the southern cone of Latin America in the early 1990s are expected to create more cooperative and peaceful regional orders than dyads or clusters of strong backlash coalitions like the Middle East in the 1960s. Co-operation declines whenever weaker versions of internationalizers are threatened by backlash forces at home and in the region, and even more so in relations between internationalizing and backlash coalitions across the border, for example North and South Korea since the late 1960s. Most conflictive, however, are relations between/ among backlash dyads and clusters; a conclusion that challenges Weart's rule that heterogeneity among domestic regimes breeds conflict. For every combination, regional dynamics affect both the nature and regional proclivities ofdomestic coalitions.
The results expected from this framework will be under severest challenge from cases of strong internationalizing coalitions that go to war with one another, or from strong backlash clusters that create zones of durable, intensive, and extensive co-operation. The framework outlines behavioural expectations on the basis of the composition and strength of both domestic and neighbouring coalitions, leading to a wide menu of falsifiable propositions linking coalitional make-up, strength, and regional behaviour. Both coalitional strength at home and the regional coalitional landscape are conditional or intervening conditions (variables), changing the relationship between coalitions and outcomes. Both case-study and statistical evidence are marshalled to test h yp otheses derived from the framework. 1 This form of coalitional analysis has several implications for democratic peace theory.2 The following areas of convergence and divergence can be demarcated. First, democratic peace theory has only recently, in its second or third generation, become attentive to economic variables. 3 It now incorporates classical economic interdependence arguments, but rarely outlines a domestic political logic connecting internationalization with a state's regional or international behaviour. 4 Furthermore, the new turn builds on conditions of bilateral interdependence among the states whose behaviour is being explained rather than on their relations to the global political economy. In addition, statistical findings on the democratic peace have given rise to an array ofh yp otheses about the possible connection between democracy and peace, but not as yet to a unified theory, some of them relying on normative considerations, others on institutional ones, and yet others on the self-interest of politicians. A coalitional perspective integrates insights from comparative politics, international relations, international political economy, and international security into the unifying concept of coalitional grand strategy.
Second, only in recent years has democratic peace theory moved beyond its focal explanans -the absence of war within democratic dyadsand there the expectations are unclear. 1 Coalitional analysis provides propositions about both conflict and co-operation. For instance, it predicts a generally positive relationship between strong internationalizing coalitions and the propensity to co-operate regionally -particularly with similarly oriented neighbours -but the opposite relationship between backlash coalitions, regardless of whether or not they are democratic. Both democratic and authoritarian (including democratizing) regimes have launched and nourished backlash and internationalizing strategies. When in power, both internationalizing and backlash coalitions can flourish or flounder in uncompetitive political systems and both t yp es can also benefit from democratization, particularly when in opposition. The presence or absence of democracy does not correlate with success in implementing an internationalizing strategy. However, a weak internationalizing ruling coalition in a non-democratic context is burdened with a concerted challenge from an alliance between backlash forces and pro-democracy groups, as in many Middle East states of which E gyp t and Jordan offer good examples. These conditions do not bode well for regional co-operation. Weak internationalizing coalitions in South Asia and the Middle East have been subjected to frequent popular, legislative, or electoral challenges, as have their cooperative regional policies.
Third, democratic peace theory is relevant only insofar as stable democratic dyads and clusters exist. That they are the exception rather than the norm in most of the industrializing world in the post-Second World War era renders the theory irrelevant before 1990 for most states outside the Organization for Economic Co-operation and Development (OECD). India and Pakistan, which arguably passed the test briefly, then tested nuclear weapons in 1998 and fought one another in 1999 at the battle of Kargil.
Fourth, some preliminary, tentative expectations from both democratic peace and coalitional theories may be combined, as in Figure 1, to guide further testing. Despite both democracy and internationalization being recent in much ofthe industrializing world, one might posit that, in theory, the confluence of democracy and internationalization makes the relationship between internationalizing coalitions and co-operative behaviour more robust (Cell I), particularly where internationalizing coalitions are strong throughout a region as well as domestically. Strong internationalizing coalitions with wide support for their economic programmes can afford to reinvent themselves through the democratic process. In the 1990s, democracy invested internationalizing grand strategies -and regional co-operation -with greater legitimacy, from South Korea to the southern cone of Latin America. At the same time, non-democratic internationalizing coalitions also face strong incentives to maintain a peaceful regional environment, as among ASEAN (Association ofSouth-East Asian Nations) members (Cell II). This expectation remains compatible with democratic peace theory which does not claim to be a necessary condition for co-operation. Next, only shallow co-operation, at best, might be expected in backlash democratic contexts (Cell III), with a higher probability of conflict than in Cell 1. Finally, a non-democratic backlash cluster (Cell IV) might be expected to yield the most conflictive regional orders of all, combining the most pernicious effects of both approaches: deficient democracy and negative incentives from internationalization.

Co-operative Most Conflictive
Fifth, when moving beyond the absence of war to examine the conditions for co-operation, neither democracy nor internationalizing coalitions appear to be necessary: other things can lead to co-operation. 1 On the one hand, the democratic peace implies sufficient conditions for the emergence of co-operation. On the other hand, the claim that strong internationalizing coalitions systematically beget regional co-operation that is intensive (in depth) and extensive (in scope) suggests that they tend to create nearly sufficient conditions for the emergence of zones of stable peace.
Preliminary evidence from the industrializing world suggests that democracy has been neither necessary nor sufficient for the emergence of co-operation in the southern cone, South Asia, the Middle East, the Korean peninsula, and ASEAN. The behaviour of democratic dyads in the southern cone and South Asia does not support the view of democracies as necessarily more co-operative; and the history of the relationship between regime-type and conflictive or co-operative behaviour in both regions has been mixed. Democratic regimes in the southern cone in the 1980s did not create a radically different regional order from the one they inherited from their non-democratic predecessors. lnternationalizers created one, in the 1990s. An overview of war and peace in the Korean peninsula and the Middle East since 1950 highlights the irrelevance of democratic peace theory, given the absence of democratic dyads. Indeed, the moderate cooperative breakthroughs in the Arab-Israeli arena, for example the Oslo agreement of 1993, and the Korean peninsula unfolded despite a lagging democratic impulse. Finally, co-operation among ASEAN members can be traced to anything but a meeting of democratic minds.
The study of the links between democratization and war may benefit from incorporating the dimension of economic reform and internationalization more systematically. As Miriam Fendius Elman argues, economic and political liberalism may be complementary, providing compatible sufficient and interdependent explanations for peace along Kantian lines, but they can also provide competing expectations. 2 Weart describes the affinity between republican principles and commerce among members of the late medieval Hanseatic League, self-governing city states that maintained the longest peace among any group of states in history, from the thirteenth to the seventeenth century. Whether or not commerce begat republicanism or vice versa is unclear, but Weart acknowledges that 'the League's founda tion was commerce,' and the peace was sustained not by ideology but by 'busy merchants [who] found no time to develop doctrines'. 1 Looking at more recent times, Bruce Russett and John R. Oneal find support for a virtuous Kantian triangle connecting economic inter dependence, democracy, and international organizations. Mark Peceny and William Stanley suggest that internationalizing coalitions have a proclivity to embrace democracy. Conversely, Fareed Zakaria foretells the rise of illiberal democracies, from Russia to Latin America, under the pre text that economic reform requires extraordinary executive powers. In his view, new democracies, lacking constitutional liberalism, have fomented nationalism, ethnic conflict, and even war. Edward D. Mansfield and Jack Snyder are similarly conscious of the h yp ernationalism and warmongering accompanying the process of democratization. 2 Clearly, expectations about sustained synergies between democracy and economic reform vary.
The worldtime at the end of the twentieth century places different mixes of constraints and opportunities on different regime t yp es and do mestic coalitions. The twin processes of democratization and integration in the global economy are likely to provide a wider range of cases for testing expectations from both democratic peace and coalitional approaches to war and peace. Neither process is irreversible or inevitable, as the case of Venezuela, and the Hanseatic League itself, suggest. Even if fewer and fewer democratic states are now ruled by coalitions with entrenched backlash agendas, there is no guarantee that the future will replicate the recent past. Clearly, political closure and repression, as in Iraq, Iran, and Syria, can be a major barrier to internationalization and, as a result, to regional peace. Furthermore, the more conflictive and threatening the regional environment, the easier it is for a coalition to broaden the sphere of the state and restrict the sphere of civil society on the grounds of national security. A conflictprone regional environment thus strengthens backlash coalitions and antidemocratic forces at the same time.